chemical analysis of typical beverages and açaí …epubs.surrey.ac.uk/852962/1/thesis - fernanda...

Chemical Analysis of Typical Beverages and Açaí Berry from South America

by

Fernanda Vanoni Matta

A thesis submitted to the Department of Chemistry in conformity with the

requirements for the Degree of Doctor of Philosophy

Faculty of Engineering and Physical Sciences

University of Surrey, Guildford, GU2 7XH

2019

i

Declaration of Originality

This thesis and the work to which it refers are the results of my own

efforts. Any ideas, data, images or text resulting from the work of others (whether

published or unpublished) are fully identified as such within the work and

attributed to their originator in the text, bibliography or in footnotes. This thesis

has not been submitted in whole or in part for any other academic degree or

professional qualification. I agree that the University has the right to submit my

work to the plagiarism detection service TurnitinUK for originality checks.

Whether or not drafts have been so-assessed, the University reserves the right to

require an electronic version of the final document (as submitted) for assessment

as above.

______________________________________

Fernanda Vanoni Matta

ii

Abstract

Brazil is a major producer of special natural foods and beverages that are commercialised and sold, locally and globally, as natural and processed products. Many are marketed as good sources of elements (minerals) and polyphenols, that play an important role in human health. At present, very few scientific studies have reported the chemical composition of these natural foods or beverages obtained in Brazil. The aim of this research was to determine the levels of elements and polyphenols in yerba mate, roasted coffee and açaí berries. The chemical composition was determined for the elemental content by inductively coupled plasma mass spectrometry and polyphenols by ultra-high performance liquid chromatography and ultraviolet–visible spectrophotometry. The elemental levels of non-commercial yerba mate leaves from the Barão de Cotegipe plantation (southern Brazil) had higher levels in the old leaves. New leaves grown on trees from an organic plantation had higher elemental levels, especially when compared with other plantations treated with NPK fertilisers. Moreover, higher elemental levels were found in plants grown in traditional organic plantations than in natural forests. The elemental levels of commercial yerba mate products from Brazil and Argentina were found to be similar. All levels were higher for commercial tea bag products than for green loose material. In Brazil, yerba mate is also sold as a roasted product (loose and tea bag) which had higher elemental levels than that for the green loose material. Infusions prepared using tea bag samples had higher elemental, polyphenol and xanthine levels than that for green loose regular infusions. Moreover, regular infusions made with green loose yerba mate had significantly higher levels of trace elements, polyphenols and xanthines in comparison with the roasted samples. All infusion methods (regular, Brazilian and bombilla) represented 0.1 to 5.0 % of the recommended daily allowance (RDA) of the trace elements measured. A regular infusion serving (1 cup of 200 mL) would provide 23.7 to 106.0 % for males and 30.3 to 135.5 % for females of the manganese RDA, depending on the type of yerba mate product. In terms of the total polyphenol intake, a regular infusion serving (200 mL) could contribute 4.0 to 14.5 % of the daily intake. The effect of roasting different coffee varieties (Obatã, Catuaí, Bourbon Amarelo and blend) collected from the Fazenda Palmares and Flor plantations (Amparo, São Paulo State) resulted in a slight increase of the elemental content of the beans during the roasting process. The total polyphenol content of coffee infusions, produced from beans collected at different times of the roasting process, showed a variation of 7.0 to 52.0 % higher levels in the dark roast (10 min) when compared to the green bean infusions (0 min). The chlorogenic acids and caffeine data showed a similar trend with an increase in the levels of the infusions prepared using the medium roast coffee. A cup of coffee (92 mL) can contribute up to 7.0 % of the estimated daily intake of polyphenols. Açaí berries obtained from the Amazon region are a major nutritional source for the local population and the processed pulp is becoming a major national and global ‘super-fruit’ product. The non-commercial purple mature pulp had a significantly higher concentration of

iii

total polyphenols and anthocyanins in comparison with the white samples (different variety). These samples were found to have high antioxidant activity due to the higher levels of total polyphenols and total anthocyanins when compared to the commercial purple and non-commercial white pulp samples. The strong antioxidant effect of açaí pulp was confirmed on mouse cells through the inhibition of producing radical oxygen species (ROS). A wound healing experiment performed using human fibroblast cells confirmed a migration effect on cells subjected to açaí pulp extracts. These results are very important, as such an experiment has never been reported, and implies that processed açaí pulp may have potential as a wound healing agent. There were no statistically significant differences in the elemental content between purple and white pulp samples. Processed açaí pulp, with less water added, had higher elemental levels (based on a fresh weight). Based on a regular consumption of purple açaí (500g), the dietary intake of total polyphenols would be more than 100% of the RDA. The consumption of açaí represents a good source of Mn (average of 1500% of the RDA), Cu (90%), Mg (30%), Ca (20%) and Zn (15%). In summary, this research provides a unique database of chemical values using analytically robust methods that can be used to evaluate the nutritional quality of Brazilian natural and commercial products and the impact of consumption on dietary intake and human health.

iv

“Eu sou a chuva que lança a areia do Saara

Sobre os automóveis de Roma,

Eu sou a sereia que dança, a destemida Iara,

Água e folha da Amazônia”

Caetano Veloso

v

List of Contents

Declaration of Originality .................................................................................. i

Abstract .............................................................................................................. ii

List of Figures ................................................................................................... x

List of Tables ................................................................................................... xv

Abbreviations ................................................................................................. xxi

Glossary ........................................................................................................ xxiv

Acknowledgements ..................................................................................... xxvi

Chapter 1. General Introduction ............................................................. 1

1.1. Overview of Brazil ................................................................................. 2

1.2. Trace Elements in the Human Diet and Health ................................. 5 1.2.1. Dose response curve and homeostasis of chemical elements .................... 7 1.2.2. Dietary intake – World Health Organisation (WHO) guidelines .................... 8 1.2.3. Deficiency and toxicity effects of chemical elements .................................... 9 1.2.4. Chemical elemental content of typical foodstuffs and beverages from Brazil

10 1.2.5. Manganese chemistry ................................................................................. 11

1.3. Polyphenols and Xanthines ............................................................... 14 1.3.1. Polyphenol chemistry .................................................................................. 15 1.3.2. Health effects of polyphenols ...................................................................... 17 1.3.3. Food total polyphenol range ....................................................................... 18

1.4. Polyphenol and elemental relationship............................................ 18

1.5. Analytical Methods and Challenges ................................................. 19

1.6. Aim and Objectives............................................................................. 24

Chapter 2. Methodology ........................................................................ 27

2.1. Introduction ......................................................................................... 28

2.2. Sample Collection ............................................................................... 29 Sample preparation ..................................................................................... 30

vi

2.3. Inductively Coupled Plasma Mass Spectrometry (ICP-MS) for Elemental Analysis ..................................................................................................... 31

Instrumentation – ICP-MS ........................................................................... 34 Internal standards – ICP-MS ....................................................................... 35 Limit of detection (LoD) and linear dynamic range (LDR) – ICP-MS .......... 36 Validation (accuracy and precision) – ICP-MS ........................................... 39

2.4. UV-Vis Spectroscopy for the Total Polyphenol Content Analysis 43

Instrumentation – UV-Vis ............................................................................ 44 Total polyphenol content by Folin-Ciocalteu analysis ................................. 45

2.5. High Performance Liquid Chromatography (HPLC) for Polyphenol Profile Analysis ........................................................................................................... 46

Instrumentation - HPLC............................................................................... 48

2.6. Statistical Analysis ............................................................................. 50 D'Agostino and Pearson normality test ....................................................... 51 Significance tests ........................................................................................ 51 Correlation coefficients................................................................................ 54

2.7. Summary .............................................................................................. 55

Chapter 3. Yerba Mate ........................................................................... 57

3.1. Introduction ......................................................................................... 58

3.2. General Introduction to Yerba Mate ................................................. 58 3.2.1. Natural occurrence ...................................................................................... 59 3.2.2. Production (plantation to processing plant) and products .......................... 60 3.2.3. Methods of consumption ............................................................................. 64

3.3. Health Effects of Yerba Mate Consumption .................................... 65 3.3.1. Chemical composition of yerba mate .......................................................... 66

3.4. Aim and Objectives............................................................................. 68

3.5. Non-Commercial Studies on Yerba Mate ......................................... 69 3.5.1. Description of the samples .......................................................................... 70 3.5.2. Materials and method.................................................................................. 71 3.5.3. Production by traditional plantations ........................................................... 71 3.5.4. Production by natural forest plantations ..................................................... 81

vii

3.5.5. Commercial processing plant ...................................................................... 84

3.6. Studies on Commercial Yerba Mate ................................................. 86 3.6.1. Description of the samples .......................................................................... 86 3.6.2. Materials and method.................................................................................. 87 3.6.3. Total elemental composition of commercial yerba mate ............................ 89 3.6.4. Elemental composition of yerba mate infusions ......................................... 94 3.6.5. Polyphenolic composition of yerba mate .................................................. 104

3.6.5.1 Total polyphenol of infusions ......................................................... 104 3.6.5.2 Chlorogenic acids, caffeine and theobromine levels in yerba mate

infusions 110 3.6.6. Link to dietary intake through consumption of yerba mate ....................... 114

3.6.6.1 Dietary intake of trace elements ..................................................... 114 3.6.6.2 Dietary intake of polyphenols ......................................................... 116

3.7. Conclusions....................................................................................... 119

Chapter 4. Chemical Composition of Roasting Brazilian Coffee ... 123

4.1. Introduction ....................................................................................... 124 4.1.1. Coffee production in Brazil ........................................................................ 124 4.1.2. Roasting of coffee beans .......................................................................... 125

4.2. Review of Roasting Coffee in Brazil (Elemental and Polyphenols) 126

4.3. Coffee Beans and the Roasting Process at Amparo, São Paulo State, Brazil 127

4.3.1. Sample collection and preparation of coffee beans .................................. 128 4.3.2. Roasting process ...................................................................................... 128 4.3.3. Grinding roasted coffee beans and particle size....................................... 129 4.3.4. Coffee infusions ........................................................................................ 129

4.4. Elemental Levels of Roasted Coffee .............................................. 130

4.5. Total Polyphenol and Chlorogenic Acid Levels in Roasted Coffee (infusions) 134

4.6. Effect of Pore Size of Ground Roasted Coffee .............................. 141

4.7. Chemical Composition of Roasted Coffee Infusions and Human Dietary Intake ............................................................................................................ 145

viii

4.8. Summary ............................................................................................ 146

Chapter 5. Brazilian Açaí ..................................................................... 149

5.1. Introduction ....................................................................................... 150

5.2. General Introduction to Açaí Berries from the Amazon Region, Brazil 150

Natural occurrence .................................................................................... 150 Brazilian açaí production and products ..................................................... 152 Health effects of açaí consumption ........................................................... 155

5.3. Chemical Composition of Açaí........................................................ 157

5.4. Aim and Objectives........................................................................... 161

5.5. Investigation of the Relationship between the Chemical Composition and Biological Activity of Açaí Samples ........................................ 162

Introduction................................................................................................ 162 Description of the samples ........................................................................ 162 Sample identification based on colour ...................................................... 163 Method development for the açaí extractions for organic analysis .......... 166 Total polyphenol and flavonoid content: Materials and method ............... 167 Total polyphenol and flavonoid content: Results and discussion ............. 168 Total anthocyanin content: Materials and method .................................... 174 Total anthocyanin content: Results and discussion .................................. 176 Total proanthocyanidin content: Materials and method ............................ 178

Total proanthocyanidin content: Results and discussion ........................ 178 Chemical antioxidant activity: Materials and method .............................. 180 Chemical antioxidant activity: Results and discussion ........................... 182 Elemental composition: Materials and method ....................................... 183 Elemental composition: Results and discussion ..................................... 183 Biological toxicity (cell viability assay): Materials and method................ 190 Biological toxicity (cell viability assay): Results and discussion ............. 190 Biological effect of açaí on radical inhibition assays: Materials and method

191 Biological effect of açaí on radical inhibition assays: Results and

discussion 193 Biological effect of wound healing in human cells: Materials and method

195

ix

Biological effect of wound healing in human cells: Results and discussion

197

5.6. Evaluation of the Amazon Geographical Variability and Industrial Processing on the Chemical Composition of Açaí ............................................... 200

Introduction................................................................................................ 200 Description of the samples ........................................................................ 200 Total polyphenol content: Materials and method ...................................... 205 Total polyphenol content: Results and discussion .................................... 205 Elemental composition: Materials and method ......................................... 208 Elemental composition: Results and discussion ....................................... 208

5.7. Link to Dietary Intake of Total Polyphenols and Minerals of Açaí 211

5.8. Conclusion......................................................................................... 217

Chapter 6. Conclusions and Future Work ........................................ 222

6.1. Overview ............................................................................................ 223

6.2. Yerba Mate ......................................................................................... 224

6.3. Chemical Composition of Roasting Brazilian Coffee ................... 227

6.4. Açaí ..................................................................................................... 228

6.5. Limitations ......................................................................................... 231

6.6. Future Work ....................................................................................... 232

Bibliography .................................................................................................. 233

x

List of Figures

Page

Figure 1.1 Map of South America with the highlighted production sites of yerba

mate, coffee and açaí (maps, 2019) 5

Figure 1.2 Biological dose response curve for elements (Underwood and Mertz,

1979) 8

Figure 1.3 Schematic of the Mn-forms as a function of Eh/pH in aqueous matrices

and aerobic conditions (Dorronsoro et al., 2006). 12

Figure 1.4 Schematic of the chemical formulae of xanthine, caffeine and

theobromine, important compounds present in yerba mate and coffee

(Merck, 2018).

15

Figure 1.5 Schematic of the chemical formulae of typical polyphenols found yerba

mate, coffee and açaí: (A) phenolic acids; (B) hydroxybenzoic acids; and

(C) flavonoids (Merck, 2018).

17

Figure 2.1 Analytical sequence adopted for the study of the chemical analysis of

typical beverages and açaí berries from South America. 29

Figure 2.2 Instrumentation of an inductively coupled plasma mass spectrometer

(ICP-MS) Agilent Series (Agilent, 2017). 32

Figure 2.3 Calibration curve for manganese (55Mn) using 115In as internal standard

for the Agilent 7800 ICP-MS in helium collision cell mode. 38

Figure 2.4 Gallic acid calibration curve obtained by the Folin-Ciocalteu assay on a

UV-Vis instrument (refer to section 2.4.2) 46

Figure 2.5 5-caffeoylquinic acid calibration curve obtained by the HPLC analysis

(refer to section 2.5.1). 50

Figure 3.1 Natural occurrence of yerba mate in South America. Adapted from

Maccari Junior (2005). 59

Figure 3.2 Main state producers of yerba mate in Brazil with their contribution (%) to

Brazilian production. Adapted from IBGE (2017). 61

Figure 3.3 Scheme of production of yerba mate. Adapted from Maccari Junior

(2005). 62

Figure 3.4 (A) Yerba mate tree; (B) Sapeco stage; (C) Yerba Mate cancheada for

the Brazilian and Argentine markets. Adapted from UFRS (2012). 63

Figure 3.5 Argentine (left) and Brazilian (right) green yerba mate commercial

samples. 64

Figure 3.6 Typical Brazilian Chimarrão (mate) consumption (Forma, 2016). 65

Figure 3.7 Schematic of the proposed bombilla method. Adapted from NGC (2015). 89

Figure 3.8 Manganese concentration (mg/200 mL) and rate of accumulation* (i.e.

potential intake) between 5 fractions (successive additions of hot water)

based on using the bombilla method (refer to section 3.6.2) for green

103

xi

loose yerba mate purchased in Brazil (n= 16) and Argentina (n=7). n is

the number of samples. The WHO set the upper limit for Mn in 11

mg/day (IOM, 2002). The analyses were determined using ICP-MS (refer

to section 2.3). *Calculated as the sum of the previous fractions.

Figure 4.1 Different maturation stages of the coffee cherry, showing green and the

mature red berries. 125

Figure 4.2 Roasting of Brazilian coffee beans from green (time = 0 minutes) and at

2 minutes intervals until the production of the dark roasted product (t =

10 minutes).

129

Figure 4.3 Typical Brazilian coffee infusion method. 130

Figure 4.4

Concentration of chlorogenic acids and caffeine (mg/L) of Brazilian

coffee infusions produced from beans sampled during the roasting

process (green t = 0 minutes, medium t = 6 min and dark roasted t = 10

min). The samples were the blend of the coffee varieties collected from

the Fazenda Palmares plantation (Amparo, São Paulo State) and

analysed by UHPLC (refer to section 2.5). n = 1.

135

Figure 4.5

The concentration of chlorogenic acids and caffeine (mg/L) of Brazilian


process (t = 0 to 10 min). The samples were of the Obatã coffee variety

collected from Fazenda Flor plantation (Amparo, São Paulo State) and

analysed by UHPLC (refer to section 2.5). n = 1.

137

Figure 4.6



process (t = 0 to 10 min). The samples were of the Catuaí coffee variety

collected from the Fazenda Palmares plantation (Amparo, São Paulo

State) and analysed by UHPLC (refer to section 2.5). n = 1.

137

Figure 4.7



process (t = 0 to 10 min). The samples were of the Bourbon Amarelo

coffee variety collected from the Fazenda Palmares plantation (Amparo,

São Paulo State) and analysed by UHPLC (refer to section 2.5). n = 1.

138

Figure 4.8


coffee infusions as a function of the different bean particle sizes

according to the method of infusion: (1) coarse for French press; (2)

regular for siphon; (3) electric perk; (4) drip; (5) fine for Brazilian

infusions; and (6) espresso. The samples were collected at the medium

roast time of the process (t = 6 minutes), being a blend of the coffee

varieties sampled from the Fazenda Palmares plantation (Amparo- São

Paulo) and analysed by UHPLC (refer to section 2.5). n = 1.

140

Figure 4.9


coffee infusions as a function of the different bean particle sizes

according to the method of infusion: (1) coarse for French press; (2)

regular for siphon; (3) electric perk; (4) drip; (5) fine for Brazilian

infusions; and (6) espresso. The samples were collected at the dark

roast time of the process (t = 10 minutes), being a blend of the coffee

141

xii

varieties from the Fazenda Palmares plantation (Amparo- São Paulo)

and analysed by UHPLC (refer to section 2.5). n = 1.

Figure 4.10 (A) to (F)

Scanning electron microscope images of Brazilian coffee beans sampled

during the roasting process (t = 0 to 10 min). The samples were the

blend of the coffee varieties collected from the Fazenda Palmares

plantation (Amparo, São Paulo State).

145

Figure 5.1 (A) Map of South America with Amazon forest in green (Fao, 2015); and

(B) Natural botanical distribution of two different species of açaí, namely,

Euterpe precatoria and Euterpe oleracea. Adapted from Yamaguchi et al.

(2015).

151

Figure 5.2 (A) The natural occurrence of açaí (E. oleracea) in the flooded forest

near Belém, Para State, Brazil; (B) the purple açaí fruit (pulp and seed

separated); and (C) the white açaí (‘greenish’) berries. Adapted from Potsch (2010).

152

Figure 5.3 (A) Harvesting of açaí by locals in the Amazon; and (B) a

‘despolpadeira’, machine used to mechanically extract the pulp of the

açaí berries. Adapted from Vida (2010).

155

Figure 5.4 Chemical structure of most predominant anthocyanin compounds found

in açaí berries being (A) cyanidin-3-glucoside and (B) cyanidin-3-

rutinoside (Yamaguchi et al., 2015).

158

Figure 5.5

Picture of the açaí berry samples where: (A) is the purple açaí whole

used as reference; (B) is the white açaí whole; (C) is the purple de-fatted

sample; (D) is the white de-fatted; (E) is the freeze-dried frozen pulp from

São Paulo; (F) is the commercial sample from São Paulo; and 7 is the

commercial sample bought in United Kingdom.

164

Figure 5.6 Illustration of the CIELAB colour space international parameters.

Adapted from Molino et al. (2013). 164

Figure 5.7 Molecular structures of the complexation between quercetin and

aluminium chloride used to determine the levels of total flavonoids.

Adapted from Frederice et al. (2010).

168

Figure 5.8 Box plots of the total polyphenol content (gallic acid equivalent mg/g) of

the açaí extracts determined using the Folin-Ciocalteu assay (refer to

section 5.5.5). The values relate to the type of sample (purple, n= 6;

white and commercial n= 4; n is the number of samples).

171

Figure 5.9 Standard curve of Cy-3-Glu concentration (mg/L) and the areas of the

peaks (mAU) used as the calibration curve for the determination of the

anthocyanin content of açaí extracts using a HPLC-DAD chromatogram

(at 520 nm).

175

Figure 5.10

The determination of the anthocyanin content of a purple non-

commercial açaí sample following methanolic extraction and using a

HPLC-DAD chromatograph (at 520 nm). The anthocyanin peaks are

cyandin-3-glucoside (retention time, tR = 17.458 minutes), cyandin-3-

rutinoside (tR = 21.069 min) and peonidin-3-rutinoside (tR = 26.237 min).

177

Figure 5.11 Total proanthocyanidin content (PAC) of açaí extractions presented as 179

xiii

B1 equivalents (B1E) via DMAC assay (refer to section 5.5.8) and

compared between the methanolic (70%MeOH) and aqueous (0.5%HAc)

extraction methods (n= 4, n, number of instrument replicates). Sample 1:

Purple açaí whole; 2: Purple açaí de-fatted; 3: white açaí whole; 4: white

açaí de-fatted; 5: oil extracted from white acai ; 6: oil extracted from

purple acai ; 7: pulp SP; 8: powder SP; 9: powder UK.

Figure 5.12 Antioxidant activity of açaí extracts determined by the ABTS assay, data

reported as Trolox equivalents (TE) (n=3; n, number of instrumental

replicates).

182

Figure 5.13

Box plots of the total elemental content of minor elements (mg/kg d.w.) of

açaí pulp samples using ICP-MS (refer to section 2.1) relating to the type

of sample (non-commercial: purple n= 6; and white n= 4; and

commercial: purple n= 4; n is the number of samples). The commercial

sample is a combination of pulp SP and powders SP and UK.

186

Figure 5.14

Box plots of the total elemental content of trace elements (mg/kg d.w.) of

açaí pulp samples using ICP-MS (refer to section 2.1) relating to the type

of sample (non-commercial : purple n= 6; and white n= 4; and

commercial: purple n= 4; n is the number of samples). The commercial

sample is a combination of pulp SP and powders SP and UK.

187

Figure 5.15 Formazan production levels in RAW 264.7 macrophage cells treated with

açaí extract solutions. Results expressed as mean ± st dev, n=3; n,

number of instrumental replicates).

191

Figure 5.16 Nitric oxide (NO) production in RAW 264.7 macrophage cells stimulated

with lipopolysaccharide (LPS). The cells were treated with 50 µg/mL açaí

extracts and dexamethasone (DEX). The results are expressed as the

mean ± st dev, n=3; n, number of instrumental replicates.

194

Figure 5.17

Radical oxygen species (ROS) production in RAW 264.7 macrophage

cells stimulated with lipopolysaccharide (LPS). The cells were treated

with 50 µg/mL açaí extracts and dexamethasone (DEX). Results are

expressed as the mean ± st dev, n=3; n, number of instrumental

replicates.

195

Figure 5.18 Cell migration determined using the OrisTM Cell 2-D migration of

adherent cells assay kit (Oris, 2017). 196

Figure 5.19

Florescence absorption of the radical oxygen species (ROS) production

in human dermal fibroblast cells (adult). The cells were treated with 50

µg/mL açaí extracts or 10% FBS (refer to section). Results are

expressed as the mean ± st dev, n=3; n, number of instrumental

replicates.

198

Figure 5.20 Fluorescence images of ‘wound healing’ of human dermal fibroblast cells (adult) between time 0 (A) and after 48 hours of incubation after

treatment with non-commercial white açaí whole sample (B).

199

Figure 5.21 (A) Picture of the open-air açaí market and (B) purple açaí fruits in Belém

(PA), Brazil. 201

Figure 5.22 Map of the sources where the açaí samples were harvested by the native 202

xiv

people. Adapted from Google Maps (2019).

Figure 5.23 Summary of the açaí processing steps and the differences between

companies, being Company I: Açaí Amazonas, Company II: Point do

açaí; and Company III: Açaí Santa Helena.

204

Figure 5.24

Box plots representing the total polyphenol (TP) content (gallic acid

equivalent mg/kg f.w.) of açaí extractions using the Folin-Ciocalteu assay

(refer to section 2.2.). The samples relate to the type of sample (non-

commercial, purple/whole: fruit, n= 7; seed, n=5; and processed freeze-

dried pulp, n= 8; where n is the number of samples).

206

Figure 5.25

Box plots representing the percentage intake (%) of minor elements

based on the consumption of a 500 g serving of açaí pulp (purple/whole)

(fresh weight). The data is compared with the World Health Organisation

(WHO) recommended daily allowance (RDA) for males (M) and females

(F). Note: WHO provide no gender data for Ca.

216

Figure 5.26 Percentage intake (%) of total polyphenol (TP) and trace elements of 500

g serving of açaí (fresh weight) in relationship to the recommended daily

allowance (RDA) for males (M) and females (F).

217

xv

List of Tables

Page Table 1.1

Symptoms of deficiency and toxicity of calcium, magnesium, iron, zinc

and copper (Strachan, 2010, Combs, 2013). 10

Table 1.2

Major sources and concentrations of calcium, magnesium, iron, zinc

and copper in the Brazilian diet (Unicamp, 2011). 11

Table 1.3

Manganese levels (mg/kg) in traditional beverages of Brazil analysed

by inductively coupled plasma optical emission spectrometry (ICP-

OES) (Unicamp, 2011).

14

Table 1.4 Analytical review of published studies on the elemental levels of

Brazilian yerba mate. 20

Table 1.5 Analytical review of published studies on the elemental levels of

Brazilian açaí 22

Table 1.6 Analytical steps involved in the analysis of polyphenols in plants and

food (Plaza et al., 2018). 24

Table 2.1 Typical operating conditions for the Agilent ICP-MS instruments. 35

Table 2.2 Investigated isotopes; limits of detection (LoD) in 1% HNO3 and

double-distilled deionised water (DDW) using 115In as internal standard

for the Agilent 7800 ICP-MS.

37

Table 2.3 Investigated isotopes and linear dynamic range (µg/L) used for each of

the type of samples (yerba mate, coffee and açaí) used in this study.

115In was used as internal standard for the Agilent 7800 ICP-MS

analysis.

39

Table 2.4 Evaluation of accuracy (comparison of measured and certified

elemental concentrations) and precision (relative standard deviation

(RSD %)) of for NIST SRM 1640a and CRM 3 (for Na, Mg, K and Ca).

41

Table 2.5 (a) Comparison of the certified reference values (CRM) with the calculated

concentration for the Tea Leaves INCT-TL-1 certified reference

materials.

42

Table 2.5 (b) Comparison of the certified reference values (CRM) with the calculated

concentration for the Peach Leaves SRM 1547 certified reference

materials (95% confidence level).

43

Table 2.6 Gradient programme for UHPLC analysis of yerba mate and coffee

infusions. 49

Table 3.1 Element content of yerba mate leaves for selected elements reported

in literature (weight basis not reported). 68

Table 3.2 Total elemental levels are reported as the mean and range (min –

max) of yerba mate leaves (based on age – new and old) for non-

commercial samples (mg/kg, dry weight) collected from traditional

plantations cultivated either using NPK fertilisers or non-chemical

73

xvi

(organic). The digested samples were analysed by ICP-MS (refer to

section 2.3).

Table 3.3

Statistical analysis using a two-tailed t-test (Miller et al., 2018) to

evaluate the relationship between the elemental levels of yerba mate

leaves (based on age – new and old) for non-commercial samples

collected from traditional plantations cultivated either using NPK

fertilisers or non-chemical (organic).

74

Table 3.4 Statistical analysis using a two-tailed t-test (Miller et al., 2018) to


leaves based on the use or non-use (organic) of NPK fertilsers during

traditional cultivation.

77

Table 3.5

Total elemental levels (mean ± standard deviation; mg/kg, dry weight)

of yerba mate leaves collected at different heights from the same tree

grown in a traditional plantation (cultivated as organic). The digested

samples analyses were analysed using ICP-MS (refer to section 2.3).

The number of samples, n = 2 replicates.

80

Table 3.6

Total elemental levels are reported as the mean and range (min –



cultivated organic plantations and grown between and beneath trees of

a native forest. The digested samples were analysed by ICP-MS (refer

to section 2.3)

82

Table 3.7 Statistical analysis using a two-tailed t-test (Miller et al., 2018) to


leaves (new and old) grown in traditional organic or native forest

plantations (refer to Table 3.6).

83

Table 3.8 Total elemental levels (mean ± standard deviation) of non-commercial

yerba mate samples (mg/kg, dry weight) during the commercial

processing of the harvested tree material. The analyses were

determined using ICP-MS (refer to section 2.3).

86

Table 3.9 Total elemental levels reported as mean and range (min – max) of

different types of commercial yerba mate samples (mg/kg, dry weight)

obtained from Brazil and Argentina. The analyses were determined

using ICP-MS (refer to section 2.3); n is the number of samples.

90

Table 3.10 Statistical analysis using a two-tailed Student t-test (Miller et al., 2018)

to evaluate the relationship between the origin (Brazil and Argentina);

packaging (loose and tea bags) and roasting (green and roasted) of

commercial yerba mate samples (refer to Table 3.9).

93

Table 3.11 Elemental levels (µg/200 mL), reported as mean and range (min –

max), of regular infusions of commercial yerba mate samples from

Brazil and Argentina. The analyses were determined using ICP-MS

(refer to section 2.3).

97

Table 3.12 Statistical analysis using a two-tailed Student t-test (Miller et al., 2018)

to evaluate the relationship between the origin (Brazil and Argentina); 98

xvii

packaging (loose and tea bags) and roasting process (green loose and

roasted) of regular infusions of commercial yerba mate (refer to Table

3.11).

Table 3.13 Percentage extraction (%) of regular infusions of commercial yerba

mate samples from Brazil and Argentina. Values reported as a mean

and range (min – max).

98

Table 3.14

Elemental levels (µg/200 mL), reported as the mean and range (min –

max), of Brazilian iced tea infusions prepared using commercial yerba

mate products from Brazil and Argentina. The analyses were

determined using ICP-MS (refer to section 2.3). n is the number of

infusion samples.

101

Table 3.15

Manganese levels (µg/200 mL) of bombilla infusions of commercial

green loose yerba mate samples from Brazil (n= 16) and Argentina (n=

7). n is the number of samples. The total manganese content refers to

a sum of the five fractions. The analyses were determined using ICP-

MS (refer to section 2.3).

102

Table 3.16

Total polyphenol content (mg/200 mL), reported as the mean and

range (min – max), of regular infusions of commercial yerba mate

samples from Brazil and Argentina. The samples were analysed by the

Folin-Ciocalteu method using a UV-Vis spectrometer (refer to section

2.4). n is the number of samples.

107

Table 3.17

Total polyphenol content (mg/200 mL), reported as the mean and

range (min – max), of Brazilian iced tea infusions of commercial yerba

mate samples from Brazil and Argentina. The analyses were

determined by the Folin-Ciocalteu method using a UV-Vis

spectrometer (refer to section 2.4). n is the number of samples.

108

Table 3.18

Total polyphenol content (mg/200 mL) of bombilla infusions of

commercial green loose yerba mate samples from Brazil (n= 16) and

Argentina (n= 7). The total content refers to the sum of the five

fractions. The samples were analysed by the Folin-Ciocalteu method

using a UV-Vis spectrometer (refer to section 2.4). n is the number of

samples.

109

Table 3.19

Chlorogenic acid, theobromine and caffeine content (mg/200 mL) of

regular infusions of commercial yerba mate samples from Brazil. The

samples were analysed by UHPLC (refer to section 2.5). For green

loose samples the results are presented as mean and range (min –

max). n is the number of samples.

111

Table 3.20

Chlorogenic acid content (mg/200 mL) of bombilla infusion fractions of

green loose commercial yerba mate products from Brazil. The samples

were analysed by UHPLC (refer to section 2.5). n = 4, n is the number

of samples. The percentage (%) refers to the contribution of the

fraction to the total (sum of the fractions).

113

Table 3.21 Theobromine and caffeine content of bombilla infusion fractions of

green loose yerba mate commercial products (mg/200 mL) from Brazil.

The samples were analysed by UHPLC (refer to section 2.5). n = 4, n

113

xviii

is the number of samples. The percentage (%) refers to the

contribution of the fraction to the total.

Table 3.22

Percentage intake (%) of manganese based on a serving (200 mL for

regular and Brazilian iced tea infusions; 1L for bombilla method) of

non-commercial yerba mate samples. The data is compared with the

World Health Organisation recommended daily allowance (RDA) of

manganese for males (M) and females (F).

116

Table 3.23

Percentage intake (%) of total polyphenol based on a serving (200 mL

for regular and Brazilian iced tea infusions; 1L for bombilla method) of

commercial yerba mate samples. The data is compared with the

values reported by (Fukushima et al., 2009) for the daily intake of

polyphenols.

118

Table 4.1 Element content of roasted coffee beans for selected elements

reported in the literature. Adapted from Pohl et al., (2013). 127

Table 4.2

Elemental levels (mg/kg, dry weight) of Brazilian coffee beans sampled

at different roasting times (minutes). The samples were of the Obatã

coffee variety collected from the Fazenda Flor plantation (Amparo, São

Paulo State) and analysed by ICP-MS (refer to section 2.3). Analysis in

duplicate.

131

Table 4.3


at different roasting times (minutes). The samples were of the Catuaí coffee variety collected from the Fazenda Palmares plantation

(Amparo, São Paulo State) and analysed by ICP-MS (refer to section

2.3). Analysis in duplicate.

131

Table 4.4


at different roasting times (minutes). The samples were of the Bourbon Amarelo coffee variety collected from the Fazenda Palmares plantation



132

Table 4.5

Elemental concentration (mg/kg, dry weight) of roasted Brazilian coffee

beans. The different coffee varieties include those selected for their

quality (section 4.3.2) or as defected beans and were collected from

the Fazenda Palmares and Flor plantations (Amparo, São Paulo State)

and analysed by ICP-MS (refer to section 2.3). Analysis in duplicate.

133

Table 4.6 Total polyphenol content (mg/L) of Brazilian coffee infusions during

roasting time (minutes). The samples were from the different coffee

varieties from Fazenda Palmares and Flor (Amparo, São Paulo State)

and analysed by UV-Vis (refer to section 2.4). n = 1.

134

Table 4.7

The concentration of chlorogenic acids, caffeine and total polyphenol

(mg/L) of roasted Brazilian coffee infusions. The samples were

prepared from selected and defected beans (section 4.3.2) from

different coffee varieties collected from the Fazenda Palmares and Flor

plantations (Amparo- São Paulo) and analysed by UHPLC (refer to

section 2.5). n = 1.

139

xix

Table 4.8 Percentage intake (%) of total polyphenol based on the Brazilian daily

consumption (92 mL) of coffee infusion. The data is compared with the

values reported by Fukushima et al. (2009) for the daily intake of total

polyphenols.

146

Table 5.1 Literature review of the elemental content of açaí according to weight

basis, dry weight (d.w.) or fresh weight (f.w.) and sample type. 160

Table 5.2

Colour parameters of açaí samples where L* indicates lightness, a* the

red/green coordinate, b* the yellow/blue coordinate and ΔE the total colour difference (refer to Equation 5.1) determined by a reflectance

spectrophotometer (CR-400, Konica, Minolta, Japan). The data relates

to a pooled freeze-dried sample.

166

Table 5.3 Total polyphenol, flavonoid, anthocyanin (ANC) and proanthocyanidin

content (PAC); and chemical antioxidant activity (ABTS and DPPH) of

açaí pulp samples. The values are expressed as mean ± standard

deviation and dry weight. 169

Table 5.4 Literature values for the total polyphenol content (mg GAE/ 100g) and

antioxidant activities DPPH (g/g DPPH) and ABTS μmol Trolox/g of typical tropical berries from Brazil (dry weight). Table adapted from

Rufino et al. (2010).

172

Table 5.5 Literature review of the total polyphenol and total anthocyanin content

of other berries obtained by two different methods (HPLC and pH); and

the total proanthocyanidin (mg/100g) (fresh weight). Table adapted

from Rothwell et al. (2013).

174

Table 5.6 Gradient programme for the determination of the anthrocyanin content

of açaí extracts using an Agilent 1200 HPLC instrument. 175

Table 5.7 (a)

Total elemental levels (mean ± standard deviation) of essential trace

elements (mg/kg fresh weight) of açaí pulp samples determined using

ICP-MS (refer to section 2.1): data relates to the type of sample (non-

commercial : purple n= 6; and white n= 4; and commercial: purple n=

4; n is the number of samples).

184

Table 5.7 (b)

Total elemental levels (mean ± standard deviation) of non-

essential/toxic trace elements (mg/kg fresh weight) of açaí pulp

samples determined using ICP-MS (refer to section 2.1): data relates

to the type of sample (non-commercial : purple n= 6; and white n= 4;

and commercial: purple n= 4; n is the number of samples).

185

Table 5.8

Literature review of typical Brazilian fruits the total elemental content of

calcium, magnesium, manganese, iron, zinc and copper in (mg/kg)

(fresh weight). Data for açaí is reported as a commercial processed

material and all the others as raw natural typical Brazilian fruits. Table

adapted from Unicamp (2011).

189

Table 5.9 Total polyphenol (TP) content of non-commercial açaí (purple/whole)

samples, (mg GAE / g) determined by Folin-Ciocalteu analysis (refer to

section 5.6.3). Fruit and seeds refer to non-processed berries and the

pulp is processed material (fluid, medium and thick relates to the

207

xx

moisture content) (refer to section 5.6.2 for sample information).

Results are expressed as mean ± st dev in fresh weight, n is the

number of replicates, n = 3. Refer to Appendix 5.3 for code information.

Table 5.10

Total elemental concentration of açaí (non-commercial, purple/whole)

samples (mg/kg, fresh weight) analysed by ICP-MS (refer to section

5.6.4). Fruit and seeds refer to non-processed berries (section 5.6.2).

Results are expressed as a mean, n is the number of replicates, n = 3.

Refer to Appendix 5.3 for code information.

209

Table 5.11 Total elemental concentration in commercially processed açaí samples

(mg/kg, fresh weight) analysed by ICP-MS (refer to section 5.6.4).

Results expressed as a mean, n is the number of replicates, n = 3.

Refer to Appendix 5.3 for code information.

211

Table 5.12

Percentage intake (%) of total polyphenol and minor elements based

on the consumption of a 500 g serving (Heinrich et al., 2011) of the

commercial and non-commercial açaí pulp (reported on a fresh weight

basis). The data is compared with the World Health Organisation

recommended daily allowance (RDA) for males (M) and females (F).

213

Table 5.13

Percentage intake (%) of selected trace elements based on a 500 g

serving (Heinrich et al., 2011) of non-commercial or commercial açaí

pulp or powder (fresh weight). The data is compared with the World

Health Organisation recommended daily allowance (RDA) for males

(M) and females (F).

215

xxi

Abbreviations

% percentage m/z mass-to-charge ratio

℃ Celsius MAE microwave-assisted extraction

< less than mAU milli absorbance unit

Å angstroms max maximum

AA Açaí Amazonas MHz mega-hertz

ABIC Associação Brasileira da Industria

do Café min minutes

ABTS azino-bis(3-ethylbenzothiazoline-6-

sulphonic) acid min minimum

AC alternating current mm millimeters

ACN acetonitrile MS mass spectroscopy

AI adequate intake MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-

diphenyl-tetrazolium bromide

ANC anthocyanin mΩ milli-ohm

ANOVA analysis of variance n number

ARG Argentina n.r. not reported

BRA Brazil NCSU North Carolina State University

CAD charged aerosol detection NIR near-infrared

calc calculated NIST National Institute of Standards

and Technology,

CIELAB Commission internationale de

l'éclairage L*a*b* nm nanometer

CPS counts per second NO Nitric oxide

crit critic NPK nitrogen, phosphorous and

potassium

CRM certified reference materials org organic

CXLE carbon dioxide-expanded liquid

extraction; ORS Octopole Reaction System

d.w. dry weight PA Para

DAD diode array detection PAC proanthocyanidin

DC direct current PAM Produção Agricola Municipal

DCF 2′,7′-dichlorofluorescein PBS phosphate-buffered saline

DDW double-distilled deionised water PCA principal component analysis

DEX dexamethasone PEF pulse electric field

DEX dexamethasone pH potential of Hydrogen

DMEM Dulbecco’s modified Eagle’s medium

PIXE particle-induced X-ray emission

DMSO dimethyl sulfoxide PLE pressurized liquid extraction

xxii

DNA deoxyribonucleic acid ppb parts per billion

DOU Diário Oficial da União ppm parts per million

DPPH diphenyl-1-picrylhydrazyl psi pound force per square inch,

EC50 concentration required to obtain a

50% antioxidant effect RDA recommended daily allowance

eH Redox potential redox reduction-oxidation

EMBRAPA Empresa Brasileira de Pesquisa

Agropecuária RF radio-frequency

EVM Expert group on Vitamins and

Minerals, ROS reactive oxygen species

F female RP reverse phase

f.w. fresh weight rpm Revolutions per Minute

FAAS flame atomic absorption

spectroscopy rps revolution per second

FAES flame atomic emission spectroscopy RSD relative standard deviation

FAO Food and Agriculture Organization s second

FBS foetal bovine serum SCAA Specialty Coffee Association of

America

FD freeze-dried SFE supercritical fluid extraction

fert fertilsers SH Açaí Santa Helena

GAE gallic acid equivalents SLE solid-liquid extraction

GDP gross domestic product SP São Paulo

GF-AAS graphite furnace atomic absorption

spectroscopy SPE solid phase extraction

HDFa Human Dermal Fibroblast cells

(adult) SRM Standard Reference Material

HHPE high hydrostatic pressure extraction St dev standard deviation;

HPLC high performance liquid

chromatography SUS Sistema Único de Saúde

hr(s) hour(s) TACO Tabela Brasileira de

Composição de Alimentos

HSCC high-speed countercurrent

chromatography TE Trolox equivalents

IBGE Instituto Brasileiro de Geografia e

Estatística TP total polyphenol

ICP-MS inductively coupled plasma mass

spectrometry; tR retention time

ICP-OES inductively coupled plasma optical

emission spectrometry u atomic mass unit

INYM Instituto Nacional de la Yerba Mate UAE ultrasound assisted extraction

IOM Institute of Medicine UFPA Universidade Federal do Pará

IQ intelligence quotient UFRS Universidade Federal do Rio

xxiii

Grande do Sul

ISO International Organization for

Standardization UFRS

Universidade Federal do Rio

Grande do Sul

ISTD/IS internal standard UHPLC ultra-high performance liquid

chromatography

IU international unit UK United Kingdom

K kelvin UL upper level

Km kilometer UNICAMP Universidade de Campinas

LC liquid chromatography US(A) United States (of America)

LDL low-density lipoprotein UV ultra-violet

LDR linear dynamic range UV-Vis ultra-violet visible spectroscopy

LLE liquid-liquid extraction V volts

LoD limit of detection W watts

LoQ limit of quantification WHO World Health Organisation

LPS lipopolysaccharide α probability

LSGS Low Serum Growth Supplement ΔE total colour difference

M male 𝜇g/L ppb

xxiv

Glossary

accuracy Degree of closeness of measurements from

a true value (Ward, 2000).

analytical figures of merit Performance characteristics of an analytical

determination, such as limits of detection

and quantification (Skoog et al., 2017).

anti-proliferate Substance capable of avoiding the

accelerated grow of bacteria.

antioxidant Molecules that can prevent the oxidation of

biomolecules in biological systems (Bastos

et al., 2007).

bioavailability Fraction of the analyte which can be

absorbed and utilised for physiological

functions (Fairweather-Tait and Hurrell,

1996).

hydrophilicity Chemical molecules that are capable of

forming ionic or hydrogen bonds with water

molecules.

hydrophobicity Chemical molecules that repel water

molecules.

isomerisation Process where a molecule is transformed

into another maintaining the same number

of atoms, but in a different arrangement.

macronutrients Chemical substances that humans consume

in the large quantities, such as, fat, protein

and carbohydrate.

micronutrients Essential chemical nutrients required in

xxv

small quantities for the maintenance of

human health, such as, vitamins and

minerals.

minerals Chemical elements required as an essential

nutrient for human health.

nutraceuticals Products derived from food sources that can

provide health benefits.

precision Degree to which repeated measurements

present similar results (Miller et al., 2018).

repeatability Level of agreement between replicate

analysis of the same sample within the

same instrument conditions (Skoog et al.,

2017).

reproducibility Precision of replicate analysis under

different conditions (between-run) (Skoog et

al., 2017).

super-fruit Marketing term for fruits that promote health

benefits due to an outstanding nutritional

composition.

trace elements Elemental concentrations in biological or

environmental systems ranging from 0.01 –

100 mg/kg (Ward, 2000).

ultra-trace elements Elemental concentrations in biological or

environmental systems that are less than

0.01 mg/kg for ultra-trace elements (Ward,

2000).

xxvi

Acknowledgements

Initially, I would like to thank my sponsor, Science without Borders

(CAPES), for providing me the opportunity and necessary financial support for

my PhD. I would also gratefully acknowledge the funding received from the

University Global Partnership Network (UGPN) programme that allowed the

collaborative work for the Açaí project.

To my supervisors, Prof Neil I Ward and Dr Mónica Felipe-Sotelo, I would

like to show my appreciation for their enthusiasm, motivation and exceptionally

support throughout this PhD. Thank you for allowing me to grow as a scientist

and supporting me as a person.

To the members of the ICP-MS group and the University of Surrey, thank

you for your assistance and guidance with the practical activities. Special thanks

to Dr Catherine Donnelly, Dr Andrea Pedronda and Dr Jonathan Brown for their

helpful advice and support. My special appreciation to the Plants for Human

Health Institute group for their warmly welcome and support, in special Dr Mary

Ann Lila for all her valuable advice. I would also like to mention Dr Debora

Esposito for her incredibly helpful advice on both research and career wise. I

would like to acknowledge Dr Cassiana Nomura and Alexandrina Carvalho for

their appreciated collaboration. This research was only possible due to the

collaboration of all the local producers that I worked in Brazil (Barão de Cotegipe,

Cooperativa de cafeicultores de Amparo and açaí producers of Pará State) who

generously shared their wide knowledge and history with me.

Finally, my special appreciation and extreme thanks to my family, who

have always been my pillar and source of strength throughout this PhD. I would

like to dedicate this achievement to my parents and brother, whose through their

affection and tireless encouragement, supported me in these challenging years

so far away from home. Also, a huge thank to my all my special friends, who

were always there for me, thank you for all your patience, love and dedication to

our friendship.

1

Chapter 1. General Introduction

2

1.1. Overview of Brazil

Brazil is a federal republic, the largest country of Latin America, the fifth

largest in the World covering an area of 8511965 km2 and has a population of

212 million (IBGE, 2019). The gross domestic product (GDP) of the country is

US$ 2056 billion, which is linked to services (55%), industry (36%) and

agriculture (9%). This makes Brazil an advanced emerging economy (Reynolds

et al., 2019). Brazil is a rich supplier of a variety of natural resources, such as

minerals, water and natural foodstuffs. The availability of an adequate climate

and rainfall, besides the fertile nature of its soil, makes Brazil ideal for agriculture.

The major agricultural products in Brazil include sugar cane, corn, cassava,

soybean, oranges, coffee, cotton, tobacco, cocoa and fruit juices (FAO, 2018).

Brazil is the third largest agricultural exporter in the World, with trading activities

in products being worth more than US$ 100 billion (2018). The main trade

products are soybeans, green coffee, corn and fruit juices (IBGE, 2019).

The exportation and globalisation of the agricultural market is bringing

benefits to the economic expansion of the region by developing autonomous and

small businesses. However, the negative aspects of the globalisation process are

not yet fully understood, including for example, land use (deforestation) and the

impact on local health (Costa et al., 2019). In rural communities of Brazil, the

availability of natural resources extracted from the forests, small farming or

fishing have been able to guarantee a healthy source of calories for low-income

families (Mansur et al., 2016). The opening of the region to globalisation and

exportation of natural products has started a change in the eating habits of the

Brazilian population. This may be due to the increased cost of traditional

foodstuffs for the locals and an increasing trend (in emerging countries)

associated with the introduction of food products that are not part of the

traditional diet. This has resulted in the consumption of foreign imported and/or

highly processed foodstuffs (Brasil, 2006). A change in the consumption of food

associated with a traditional diet has already been observed in Brazil (and other

emerging countries), especially amongst the younger population (< 25 years)

3

(Costa et al., 2019). Moreover, an increased incidence of chronic diseases, such

as diabetes and hypertension, has also been reported, which may be linked to

the changes in diet (SUS, 2018).

A recent study about the intake of antioxidant nutrients (refer to glossary)

by the Brazilian population has reported an insufficient intake of antioxidant

nutrients, especially vitamins E, A and C. It was also found that the intake of

antioxidant nutrients varied based on nutritional status, gender and life stage

(Tureck et al., 2017). The Dietary Guidelines for the Brazilian Population (2017)

recommends that the population should return to the consumption of foodstuffs

linked with the traditional Brazilian diet. In summary, this means a diet based on

the preparation of cereals and legumes (rice and beans), fruits and vegetables,

as such foodstuffs provide the healthiest intake of nutrients, antioxidants and

minerals (refer to glossary) (Brasil, 2006).

Brazil is the World's third largest fruit producer, behind China and India.

Between 1990 and 2004 exports grew by 183% in value, 277% in quantity and

915% as a net value. The commercial production of Brazilian fruits has two

designations; unprocessed fruit, that represents 47% of the country's production,

and processed fruit, representing 53%. In terms of the total amount of

unprocessed fruit produced in Brazil, 31% is exported (EMBRAPA, 2017). The

European Union is the largest market for Brazilian fruits; almost 70% of the fruit

exported by Brazil is consumed by this economic bloc. Moreover, 34% of the

exports are destined to the Netherlands, however the country works not only as a

consumer, but also as a distributor to other countries of the European Union

(Cunha Filho and Carvalho, 2005).

Among all of the natural products that Brazil produces and exports, three

have gained special attention, namely, yerba mate, coffee and açaí, due to their

popularity, high quality and health claims (Heck and De Mejia, 2007, Abrahão et

al., 2010, Yamaguchi et al., 2015). Yerba mate (Ilex paraguariensis) is found in

South America and has been consumed by native peoples since pre-Columbian

times (Bracesco et al., 2011). It is consumed as a hot or cold infusion and is one

4

of the most popular beverages in South America, with an estimated 1 million

people consuming it due to its high levels of caffeine and antioxidants (de Morais

et al., 2009). The production of yerba mate is restricted to southern Brazil, as

shown in Figure 1.1, with an estimated exportation value of US$ 85 million in

2017 (IBGE, 2017). Coffee production in Brazil is responsible for about a third of

all global coffee, making Brazil the World's largest producer and exporter. The

estimated exportation value for 2017 was US$4 billion (IBGE, 2017). The crop

(Coffea arabica L.) first arrived in Brazil in the 18th Century and the country had

become a dominant producer by the 1840s, due to the local environmental and

climate being ideal growing conditions (Pendergrast, 2010). Coffee plantations

are mainly located in the south-eastern states of Minas Gerais, São Paulo and

Paraná, as shown in Figure 1.1. Coffee is one of the most commonly consumed

beverages in Brazil and one of the major sources of antioxidant intake in the daily

diet of Brazil (Tureck et al., 2017). Finally, the açaí berry (Euterpe oleracea) is a

native fruit from the Amazonian region of Brazil and has recently become very

popular due to its status as a ‘super-fruit’ (refer to glossary) related to the high

antioxidant activity of the fruit (Yamaguchi et al., 2015). The production of açaí is

still mainly an extractive activity from the Amazon forest, as shown in Figure 1.1

(Homma et al., 2006, Maciel et al., 2018, Tagore et al., 2018). Brazil is the main

producer and exporter of açaí, generating an estimated monetary source of

US$ 9 billion in 2017 (IBGE, 2017), exceeding the price per ton of common

exportation products, such as soybeans and Brazil nuts (Yamaguchi et al., 2015).

These important Brazilian products, namely, yerba mate, coffee and açaí, will be

discussed and investigated in chapters 3 (yerba mate); 4 (Brazilian coffee) and 5

(açaí), respectively.

5

Figure 1.1: Map of South America with the highlighted production sites of yerba

mate, coffee and açaí (Maps, 2019).

1.2. Trace Elements in the Human Diet and Health

Micronutrients (refer to glossary) play an important role in human

metabolism and health (Goldhaber, 2003, Strachan, 2010). There is a wide range

of elemental concentrations in biological systems, ranging from 100 mg/kg (or

parts per million) for minor elements, 0.01 – 100 mg/kg for trace elements, and <

0.01 mg/kg for ultra-trace elements (refer to glossary) (Versieck and Cornelis,

1980). The elements within biological systems can be further classified into

essential, non-essential and toxic elements (Strachan, 2010). Food and

beverages are the primary sources that provide the levels of the elements

(sometimes referred to as ‘metals’ or minerals) required to regulate various

chemical and biochemical reactions (Fraga, 2005). Frieden (1985) proposed a

biological classification of trace elements based on the amount in tissues. This

biological classification included: (i) essential trace elements (B, Co, Cu, I, Fe,

6

Mn, Mo and Zn); (ii) probably essential trace elements (Cr, F, Ni and V); and (iii)

physically promotive trace elements (Br, Li, Si, Sn and Ti) (Frieden, 1985). There

have been various definitions of what is an essential trace element, and many

include that they are ‘chemical micronutrients which are required in minute

quantities but play a vital role in maintaining the integrity of various physiological

and metabolic processes occurring within living tissues. The deficiency of any of

these trace elements may be apparent as a combination of various clinical

manifestations rather than a specific presentation as each trace element is

related to many enzyme systems’ (Bhattacharya et al., 2016). A recent report

stated that essential elements are defined in terms of their chemical state (as a

free ion, or as a variety of chemical compounds) that are included in all cells and

tissues of the human body (Skalnaya and Skalny, 2018). Physiological effects of

these elements depend on the dose. For each element there is an optimum

range of concentrations to perform vital functions (refer to section 1.2.1). At

deficiency or excessive accumulation levels of these elements, there is a

disturbance in the physiological activity associated with the element, which is

reversable with the addition of physiological amounts of the specific element – in

which chemical species is important (Mertz, 1981, Skalnaya and Skalny, 2018).

Essential ‘macro’ elements, whose concentration in the body exceeds 0.01%

(100 mg/L or ppm) include O, C, H, N, Ca, P, K, Na, S, Cl and Mg. Essential

‘trace’ elements, found at concentration ranges from 0.01% to 0.00001% (100 to

0.1 mg/L) includes: Fe, Zn, F, Mo, Cu, I and Mn. Finally, essential ‘ultra-trace’

elements are found at concentrations lower than 0.000001% (< 100 µg/L or ppb)

are Se, Co, V, Cr, As, Ni and Sn (Skalnaya and Skalny, 2018). There is some

debate about the essential roles of some trace elements, especially B and As

(Prashanth et al., 2015).

Essential trace elements have a number of roles in the maintenance of

human health. These elements have a key role within enzyme systems, as

metallo-enzyme complexes (Donnelly, 2015). Essential elements can also

participate in reduction-oxidation (redox) reactions by donating or accepting

electrons, or are involved in the transport of oxygen (iron) (Nielsen, 2007). Whilst

7

the diet is a major source of essential elements it can also be an exposure or

uptake route for non-essential and toxic elements. Non-essential or ‘possibly

essential’ elements are those for which no evidence has been published on an

essential role in the human body. This group includes: B, Al, Ba, Br, Bi, Li, Sr,

Rb, Sb, Ge, Be and Cs (Prashanth et al., 2015, Skalnaya and Skalny, 2018).

Toxic elements are those that have no nutritional value even at trace amounts

(Skalnaya and Skalny, 2018). They can be present in the regular human diet,

and include Cd, Pb and Hg, and are considered toxic even at low concentrations

(WHO, 1996).

Most plants obtain elements (minor, trace and ultra-trace) from the soil via

the soil solution into the root, but sometimes salts and minerals from fertilisers

can be taken up through the leaves (Kabata-Pendias, 2010). The formation of

soil chemistry is a result of the eroding of rock where its structure is broken down

into soluble compounds by physical processes (Garrett, 2000). These are

washed by rain and rivers where various reactions can occur between the

minerals (or elements/metals) and the organic matter from decomposing remains

of plants, animals and microorganisms. This increases the solubility of trace

elements (including the effect of pH and Eh – redox potential) and changes the

translocation to different parts of a plant (Garrett, 2000). The elemental content of

the soil can also change with the addition of fertiliser or compost, and from

human activities, such as, pollution, discharge of wastes or mining (Kabata-

Pendias, 2010).

1.2.1. Dose response curve and homeostasis of chemical elements

The main source of chemical elements for humans is through the diet. A

balanced diet will provide a normal person with an adequate supply of chemical

elements to maintain optimal health (Donnelly, 2015). Moreover, the human body

is capable of maintaining the content of these essential chemicals due to

homeostasis, which is a self-regulating mechanism that involves absorption,

8

storage and excretion processes (Nielsen and Hunt, 1989, Nielsen, 2007). Figure

1.2 represents the biological dose response curve (Underwood and Mertz, 1979).

If there is an inadequate supply of an element via the diet or the uptake is

compromised through disease, the physiological processes required by that

element are also reduced causing suboptimal health, with the manifestations of a

deficiency disease. Furthermore, if the essential elemental levels are too low

then death can occur (Nielsen and Hunt, 1989, Nielsen, 2007). Equally, a high

intake of an element that exceeds the homeostatic regulation leads to toxic

levels, and (if high enough) can also result in death (Nielsen and Hunt, 1989,

Nielsen, 2007). It is important to highlight that every element can produce signs

of toxicity at certain concentrations in the human body (Underwood and Mertz,

1979, Donnelly, 2015).

Figure 1.2: Biological dose response curve for elements (Underwood and Mertz,

1979)

1.2.2. Dietary intake – World Health Organisation (WHO) guidelines

In order to protect the health of the population, The World Health

Organisation (WHO) have undertaken several risk assessments of nutrient

intake, including for the chemical elements. The WHO have recommended levels

9

to prevent the population from inadequate and excessive intakes of the elements

(including major/ minor and trace – refer to section 1.2). In order to provide

guidance, a Recommended Dietary Allowance (RDA) has been set for different

population groups (Schumann, 2006), which is defined as the amount of a

nutrient sufficient to ensure the needs of nearly all of the population (97.5%)

(Schumann, 2006). If there is a lack of data to establish an RDA, an adequate

intake (AI) is proposed. Conversely, to protect against the toxic effects of over-

intake, a Tolerable Upper Level Intake (UL) has also been defined as the highest

average daily intake that is unlikely to provide a risk of toxic effects in almost all

of the general population (EFSA, 2006). It is important to highlight that the

bioavailability (refer to glossary) of these chemical elements is also critical for

considering the human dietary intake.

1.2.3. Deficiency and toxicity effects of chemical elements

In this study, calcium, magnesium, manganese, iron, zinc and copper are

highlighted because the concentration in the analysed products may provide a

significant contribution to the nutritional intake of these elements. Each one of

these essential elements follows a dose response curve (refer to Figure 1.2) and

has a nutritional recommended daily allowance (RDA) in order to prevent

deficiency (Goldhaber, 2003), or a tolerable upper level intake (UL) to prevent

toxicity effects (refer to section 1.2.2). Manganese, iron, zinc and copper are all

important micro-nutrients and are constituents of key proteins or enzymes in the

human body that provide a variety of functions (Combs, 2013). Moreover,

calcium and magnesium are essential macro-nutrients for structure and electro

regulation (Combs, 2013). Table 1.1 presents some of the symptoms of

deficiency or toxicity of the essential elements (calcium, magnesium, iron, zinc

and copper). Manganese will be further discussed in section 1.2.5.

10

Table 1.1: Symptoms of deficiency and toxicity of calcium, magnesium, iron, zinc

and copper (Strachan, 2010, Combs, 2013).

Element Deficiency symptoms Toxicity symptoms

Calcium Compromised bone structure and

nervous transduction

Bone and muscle weakness, kidney

stones, fatigue, cardiac arrhythmia

Magnesium

Compromised bone structure,

electrochemical regulation and

enzyme catalysis

Lethargy, nausea, muscle weakness,

urine retention, cardiac arrhythmia

Iron Anaemia, compromised immune

functions, fatigue

Hepatic cirrhosis, diabetes, heart

failure, arthritis

Zinc Poor growth, reduced testicular

development, osteoporosis risk Anaemia, copper depletion

Copper Anaemia, reproductive failure, bone

abnormalities, poor growth

Nausea, cirrhosis (hepatic

accumulation), gastroenteritis

1.2.4. Chemical elemental content of typical foodstuffs and beverages from Brazil

The TACO project (Tabela Brasileira de Composição de Alimentos,

Brazilian Table of Food Composition) was concluded in 2011 with the objective of

creating reliable analytical data for a national database on the composition of

typical Brazilian foods (Unicamp, 2011). TACO provided reliable information on

the food consumption of the Brazilian population that serves as a basis for

formulating policies and action plans for food and nutritional security (Galeazzi,

2001). Table 1.2 summarises the major sources of calcium, magnesium, iron,

zinc and copper in the Brazilian diet, using TACO as the reference. Manganese

will be discussed in section 1.2.5.

11

Table 1.2: Major sources and concentrations of calcium, magnesium, iron, zinc

and copper in the Brazilian diet (Unicamp, 2011).

Element Dietary source Concentration (mg/100 g)

Ca

Coriander (dried) 784

Fish (Lambari, raw) 1181

Milk (powder) 1363

Mg

Beans (Carioca, raw) 210

Soya (powder) 242

Brazilian nuts (raw) 365

Fe

Porridge (powder) 42.0

Beans (Rajado, raw) 18.6

Babaçu nut (raw) 18.3

Zn

Porridge (powder) 15.2

Beef (cooked) 8.1

Caruru leaves (raw) 6.0

Cu

Beef liver (raw) 9.01

Brazilian nuts (raw) 1.79

Papaya (Formosa, raw) 1.36

1.2.5. Manganese chemistry

The manganese levels in yerba mate and açaí berries have been reported

to be higher than any other common plant used for infusions (Bragança et al.,

2011, Wróbel et al., 2000) and other typical fruits (Unicamp, 2011). Therefore, in

this study, this particular element will be the focus of the research.

Manganese (Mn) has an atomic number of 25 and a molecular weight of

approximately 55 g/mol. It is a transition element located in group 7 and period 4

of the periodic table. It is not found in its elemental state in nature but is often a

component of a mineral, in combination with iron (ATSDR, 2000). Anthropogenic

(or man-made) contamination of Mn is usually associated with mining activities,

iron and steel production or via pollution associated with the combustion of fossil

fuels (Grygo-Szymanko et al., 2016). Manganese is an essential trace element

required for human health as it is a constituent of enzymes with many functions,

12

including immunity, regulation of blood sugar and cellular energy, blood clotting,

reproduction, digestion and bone growth (Roth et al., 2013). However, at high Mn

concentrations, this element is neurotoxic and chronic exposure may lead to a

condition known as manganism, a disorder that has symptoms similar to

Parkinson's disease (Michalke et al., 2007).

The most common oxidation states are from +2 to +7; with +2 being the

most stable and bioavailable; +3 is only present in complexes; +4 is insoluble

and found in particulates and colloids; + 5 and + 6 are instable in neutral

solutions; and + 7 is found in the permanganate ion (MnO4−). The behaviour of

manganese in water is affected by oxidation and reduction processes which have

a major influence on the species present (Dorronsoro et al., 2006). In aqueous

matrices and under aerobic conditions, Mn is more soluble in acidic (pH <6) and

insoluble in alkaline (pH >8) conditions, as shown in Figure 1.3.

Figure 1.3: Schematic of the Mn-forms as a function of Eh/pH in aqueous

matrices and aerobic conditions (Dorronsoro et al., 2006).

The guideline limit of manganese in drinking water is 400 µg/L according

to the World Health Organisation (WHO, 1996), although the limit in the UK is set

to 50 µg/L (EVM, 2003). This is mainly to avoid water colouration and deposition

in pipes rather than preventing negative effects on human health. Manganese is

absorbed in the human gut and excreted by bile, and because this system is not

13

fully developed in children, they are more susceptible to Mn toxicity (Neal and

Guilarte, 2012). There are several studies about toxic manganese exposure

through inhalation, which is common in mine workers, leading to problems in

intellectual and cognitive development (Rumsby et al., 2014). Although some

studies try to associate the same problems with Mn exposure in food and water,

there is a lack of conclusive correlations due to other cofactors in cognitive

development. A Canadian study has shown that exposure of Mn at levels in

drinking water, below that recommended by the WHO and above the UK limit,

can caused a decrease in the intelligence quotient (IQ) in children (Bouchard et

al., 2010). A study on rats found that exposure to Mn in drinking water results in

accumulation in the same tissues as that associated with inhalation and could

cause behavioural and locomotor effects (Reichel et al., 2006). Furthermore, Mn

is a key element in photosynthesis. The elemental content of a plant material

usually reflects the conditions where it was grown and could be influenced by

many factors, such as soil chemistry, environment and age (Saidelles et al.,

2010). In the case of manganese, it is rapidly taken up from the soil by plants and

distributed to the leaves and chloroplasts, where it plays an important role in

oxygen evolution and electron transport (Kabata-Pendias, 2010).

The recommended dietary intake levels for manganese are set at 2.3 and

1.8 mg/day for men and women respectively; and the tolerable upper intake level

is at 11 mg/day (IOM, 2002). Table 1.3 summarises the reported levels of

manganese in popular beverages from Brazil. It must be noted that fruit juices in

Brazil are made from frozen fruits (pulp) blended with water. The Brazilian

beverages that could contribute to the daily intake of manganese are açaí and

yerba mate.

14

Table 1.3: Manganese levels (mg/kg) in traditional beverages of Brazil analysed

by inductively coupled plasma optical emission spectrometry (ICP-

OES) (Unicamp, 2011).

Beverages Mn (mg/kg) Açaí, pulp, guaraná and glucose 32.86

Coffee, roasted, powder 25.79

Pineapple, pulp, frozen 10.21

Coconut water 2.53

Sugarcane, juice 2.06

Cappuccino, powder 1.71

Cupuaçu, pulp, frozen 1.70

Mango, pulp, frozen 1.19

Black infusion, 5% infusion 0.89

Cajá, pulp, frozen 0.70

Passion fruit, pulp, frozen 0.70

Graviola, pulp, frozen 0.56

Caju, pulp, frozen 0.54

Pitanga, pulp, frozen 0.53

Umbu, pulp, frozen 0.49

Coffee, 10% infusion 0.38

Acerola, pulp, frozen 0.34

Orange, juice 0.17

Lime, juice 0.10

Milk, whole < 0.05

Cachaça < 0.05

Beer 0.05

Soda < 0.05

1.3. Polyphenols and Xanthines

Polyphenols are plant metabolites characterised by the presence of at

least one aromatic ring with one or more hydroxyl groups (phenol structural units)

attached (Campos‐Vega and Oomah, 2013). In plants, both natural phenols and

15

the larger polyphenols play important roles in the ecology of most plants. Their

functions in plant tissues can include giving colour, acting as an insect or as a

mammal feeding deterrent. The end result is to help protect the plant against

stress from ultra-violet (UV) radiation damage, temperature, oxidative activity,

and tolerance to heavy metals through chelation of metal ions, or as a defense

against pathogens (Gould and Lister, 2005, Stevenson and Hurst, 2007,

Donnelly, 2015).

Xanthine or 3,7-dihydropurine-2,6-dione, is a purine base found in most

biological systems. A number of stimulants are derived from xanthine, including

caffeine and theobromine, as shown in Figure 1.4 (Voet et al., 2008). In plants,

caffeine is found in the seeds, nuts, or leaves of a number of plants and helps to

protect them against predator insects and to prevent germination of nearby

seeds (Saxena et al., 2013). Theobromine is a bitter alkaloid found in chocolate,

leaves of the tea plant and in coffee beans (Martínez-López et al., 2014).

Caffeine differs from theobromine in having an extra methyl group, as shown in

Figure 1.4.

Figure 1.4: Schematic of the chemical formulae of xanthine, caffeine and

theobromine, important compounds present in yerba mate and

coffee (Merck, 2018).

1.3.1. Polyphenol chemistry

Polyphenols can be divided into classes, namely, phenolic acids,

hydroxycinnamic acids and flavonoids (Crozier, 2003). Phenolic acids or

16

hydroxybenzoic acids have a C6-C1 structure, such as gallic acid and salicylic

acid, as shown in Figure 1.5. During the different stages of plant maturation and

growing conditions the concentration and type of phenolic acids change. The

functions of phenolic acids in plants range from nutrient uptake to protein

synthesis, enzyme activity, photosynthesis and structural components (Donnelly,

2015). Hydroxycinnamic acids have a general C6-C3 structure, as shown in

Figure 1.5. One of the most important derivative groups of hydroxycinnamic acids

is the chlorogenic acids or quinic acid conjugates, which are present in fruits and

yerba mate leaves (Manach et al., 2004, Bravo et al., 2007). Finally, flavonoids

have a C6-C3-C6 structure (Figure 1.5) and are the most common family of

polyphenols present in plants. Flavonoids can be further divided in 6 main sub-

classes, namely, flavones, isoflavones, flavonols, flavan-3-ols, flavanones and

anthocyanidins (Manach et al., 2004). The most widely distributed and diverse

group of flavonoids in nature is the flavonols, such as quercetin, kaempferol,

myricetin and isorhamnetin (Crozier, 2003). In fruits and vegetables quercetin

compounds are the most commonly occurring (Saltmarsh and Goldberg, 2003,

Kyle and Duthie, 2005, Donnelly, 2015). Flavan-3-ols are the most complex

group of flavonoids which occur as simple monomers or in complex polymeric

forms, such as proanthocyanidins. They are found in red wine, cocoa and berries

(Skates et al., 2018, Crozier et al., 2009). Anthocyanidins are present mostly in

fruits and flowers and are responsible for the red, blue and purple colouration.

The main source of anthocyanidins in the human diet is from fruits, especially

berries (Ovaskainen et al., 2008, Crozier et al., 2009). They play a major function

in plants mainly attracting pollinating insects and protecting plants from damaging

light (Gould and Lister, 2005).

17

Figure 1.5: Schematic of the chemical formulae of typical polyphenols found in

yerba mate, coffee and açaí: (A) phenolic acids; (B) hydroxybenzoic

acids; and (C) flavonoids (Merck, 2018).

1.3.2. Health effects of polyphenols

The polyphenols have an increasing recognition as an emerging field of

interest in nutrition in recent decades (Cory et al., 2018). The polyphenol

consumption may play a major role in the maintenance of human health, mainly

due to their antioxidant activity, by removing free radicals and reactive oxygen

species of the biological system (Donnelly, 2015). Moreover, polyphenols help

regulate the metabolism, weight, susceptibility to chronic disease and cell

proliferation in humans (Cory et al., 2018). Studies on biological systems have

shown that numerous polyphenols also have anti-inflammatory and antioxidant

activities, that could prevent and/or have therapeutic effects for

neurodegenerative disorders, cancer, cardiovascular disease and obesity (Pérez-

Jiménez et al., 2010, Singh et al., 2011). There is also evidence that the long-

term consumption of polyphenols helps protect humans against type-2 diabetes,

osteoporosis, pancreatitis, gastrointestinal problems and lung damage (Fraga et

al., 2010, Martín‐Peláez et al., 2013, Xiao and Hogger, 2015).

A number of studies have acknowledged cellular targets that can be

involved in the health claims of polyphenols. However, the mechanism of the

molecular interactions of polyphenols with cellular targets remains mostly

18

speculative (Fraga et al., 2010, Cory et al., 2018). Some of the mechanisms

proposed are free radical scavenging, metal sequestration and the interactions of

polyphenols with membranes, enzymes, transcription factors or receptors (Fraga

et al., 2010). The health benefits alleged to polyphenols are derived from the

polyphenol or the phenolic metabolites resulting from the transformation of the

compounds in the gut microbiota (Selma et al., 2009).

1.3.3. Food total polyphenol range

It is not easy to estimate the daily intake of total polyphenols, due to the

variation in the structural diversity of phenolic compounds for a particular

foodstuff (Scalbert and Williamson, 2000). Moreover, there is a variation in the

dietary intake of polyphenolic compounds between geographical regions and

consumption age groups. Several studies have agreed on a proposed range of 1

g of total polyphenols per day (Kühnau, 1976, Faller and Fialho, 2009, Landete,

2013, Fukushima et al., 2009). Beverages, such as coffee, wine, fruit juices and

tea, are the largest contributors to the dietary sources of polyphenols (Saura-

Calixto and Goñi, 2006). The total polyphenol content of foods is usually

determined by the Folin-Ciocalteu assay, described in section 2.4.2. In relation to

typical beverages, the total polyphenol content of green tea ranges from 8.7 to

25.8 g/100 g; black tea 8.0 to 26.3 g/100 g, and brewed coffee reported that the

total polyphenol content of coffee prepared in a commercial brewer was 0.96 to

2.27 g/L (Lakenbrink et al., 2000, Obuchowicz et al., 2011, Stodt and Engelhardt,

2013). The literature values for the total polyphenol content of typical tropical

berries from Brazil is presented in Chapter 5, Table 5.5.

1.4. Polyphenol and elemental relationship

The elements in solution can exist as free ions or complexes with naturally

occurring bioligands. Polyphenols are one of these natural ligands, which can

complex elements (or metals) through hydroxyl, carboxylate and phenolate

19

groups (Pohl and Prusisz, 2007, Khokhar and Owusu Apenten, 2003). There is

only a limited number of studies investigating the relationship between the

chemical interactions of elements and organic compounds in beverages. These

interactions could interfere on the bioavailability of not only the elemental

species, but also the polyphenols. Pohl and Prusisz (2007) reported that this

chelation between elements and flavonoids considerably reduced the

bioavailability of the metal for the body or completely impaired the absorption of

these chemicals. Recent studies have shown that iron is not bioavailable in

beverages and foodstuffs that are rich in polyphenols, tannins or fibers (Yuyama

et al., 2002, Toaiari et al., 2005; Perron and Brumaghim, 2009). However, there

are other compounds, such as vitamin C, that could enhance the bioavailability of

iron (Silva et al., 2004).

Therefore, the bioavailability ofelements and polyphenols may impact on

the human intake of these chemicals through the consumption of food products,

thereby increasing or decreasing the potential bioinorganic effect on the human

body (Fairweather-Tait and Hurrell, 1996).

1.5. Analytical Methods and Challenges

Having reviewed the chemistry of the Brazilian foodstuffs and beverages

under investigation, it is now important to evaluate the analytical methods that

are traditionally used in previously published studies looking at the levels of

elements and polyphenols in foodstuffs and infusions of products from Brazil.

Moreover, one of the important parts of analytical chemistry is the analytical

sequence which defines the methodology from establishing a project hypothesis

through to the critical analysis of the data and what actions should be taken to

address the hypothesis (refer to Figure 2.1) (Ward, 2000). The main stages of

the sequence involve the hypothesis, sample selection, sample preparation,

chemical analysis (including optimisation, calibration and validation of

instruments and methods) and the statistical analysis and presentation of data

(Ward, 2000).

20

Table 1.4: Analytical review of published studies on the elemental levels of

Brazilian yerba mate.

Author Year Elements Sample selection

Sample preparation

Chemical analysis Data analysis

Barbosa et

al. 2015

C, K, N, Mg, Ca,

P, Al, Na, Zn,

Mn, Fe, Ba, Cu,

Ni, Mo, Pb, Cr,

As, Co, Ag, V

and Cd.

non-

commercial

sample

acid digestion

and dry weight

element

analyser;

FAES; UV-Vis;

ICP-OES

mean and st

dev

Barbosa et

al. 2018

C, N, P, K, Ca,

Mg, Fe, Mn, Zn,

Cu, Ni, B, Mo,

Co, As, Cd, Pb,

Ba, Cr and V

non-

commercial

sample

acid digestion

and dry weight

element

analyser and

ICP-OES

mean and ratio

Bastos et

al. 2014

P, K, Ca, Mg,

Na, Fe, Mn, Cu

and Zn

non-

commercial

sample and

soluble extracts

dry ashing and

water

extraction

FAES

UV-Vis and

FAAS

mean and %

solubility

Giulian et

al. 2007

Mg, Al, Si, P, S,

Cl, K, Ca, Ti, Mn,

Fe, Cu, Zn and

Rb

commercial

and infusion

pellets;

bombilla method and

water

extraction

PIXE (X-ray)

metal

extraction

values

Heinrichs

et al. 2001

N,

P, K, Ca, Mg, S,

B, Cu, Fe, Mn,

Ni, Zn, Al, Cd,

Co, Cr, Na and

Pb

commercial

and infusion

acid digestion

and tea-based

method

ICP-OES

min, max,

mean, RSD

and %

solubility

Jacques et

al. 2007

K, Ca, Na, Mg,

Mn, Fe, Zn, and

Cu

non-

commercial dry ashing

FAAS and

FAES

mean, st dev

and ANOVA

Malik et al. 2008

Al, B, Cu, Fe,

Mn, P, Zn, Ca, K

and Mg commercial dry ashing

FAAS and

ICP-OES

mean and st

dev

Magri et

al. 2019

Mn, Al, Fe, Zn,

Cu, Ni, Cd and

Pb

non-

commercial microwave

ICP-OES and

GF-AAS

mean, RSD

and ANOVA

Milani et

al. 2019

Al, As, Ba, Cd,

Cr, Cu, Fe, Mn,

Ni, Pb, Se and

Zn

commercial

infusions

tea-bad

method ICP-MS

mean, min,

max and

ANOVA

Pozebon

et al. 2015

Al, Ba, Ca, Cu,

Fe, K, Mg, Mn,

P, Sr, Zn, Li, Be,

Ti, V, Cr, Ni, Co,

commercial acid digestion

ICP-MS, ICP-

OES

LOD, LOQ,

spike recovery

mean and st

dev

21


Sample preparation

Chemical analysis Data analysis

As, Se, Rb, Mo,

Ag, Cd, Sb, La,

Ce, Pb, Bi and U

and CRM

Rossa et

al. 2015

C, N, K, Ca, Mg,

Na, Fe, Mn, Cu

and Zn

non-

commercial dry ashing

NIR and UV-

Vis

min, max, st

dev and

ANOVA

FAES: flame atomic emission spectroscopy; FAAS: flame atomic absorption spectroscopy; UV-Vis: ultraviolet–visible spectroscopy; ICP-OES: inductively coupled plasma optical emission spectrometry; ICP-MS: inductively coupled plasma mass spectrometry; GF-AAS: graphite furnace atomic absorption spectroscopy; PIXE: particle-induced X-ray emission; NIR: near-infrared spectroscopy; st dev: standard deviation; min: minimum; max: maximum; RSD: relative standard deviation; ANOVA: analysis of variance; LOD: limit of detection; LOQ: limit of quantification; CRM: certified reference material.

Table 1.4 reports a review of published studies on the elemental levels of

Brazilian yerba mate and açaí (Table 1.5), with a critical assessment of the

important stages of the analytical sequence (see above). In summary, there has

only been a limited number of published studies (n = 11 for yerba mate and n = 7

for açaí), with the focus being on major/minor elements and for non-commercial

and commercial products.

In relation to the yerba mate infusions the method is normally based on

water extraction, a simulation of the bombilla (traditional method) or the regular

tea-based methodology (refer to Table 1.4). There is a lack of details provided on

the mass or volume of commercial yerba mate products used to prepare

samples, whether the solution was filtered and the method of pre-analysis

storage.

22

Table 1.5: Analytical review of published studies on the elemental levels of

Brazilian açaí


Sample preparation

Chemical analysis

Data analysis

Menezes

et al.

2008

a

Na, Mg, Al, Mn,

Co, Ni, Cu, Al,

As, Rb, Mo, O,

Ca, Se, Ag, Cd,

Ba, Hg, Pb, Th,

U, K, Sr, Sb and

Fe

commercial

processed pulp n.r. ICP-MS value

Unicamp 2011

Ca, Fe, Mg, Mn,

P, Na, K, Cu and

Zn

commercial

processed pulp dry ashing ICP-OES mean

Yuyama 2011 Na, Ca, K, Fe,

Zn, B, Co and Cr

non-

commercial

processed pulp

n.r. NAA and

validation

mean and st

dev

Llorent-

Martínez

et al.

2013

Ag, Al, As, Ba,

Be, Ca, Cd, Co,

Cr, Cu, Fe, K,

Mg, Mn, Mo, Na,

Ni, Pb, Sb, Se,

Tl, V, Hg and Zn

commercial

processed juice microwave

ICP-MS

Calibration,

validation,

spike recovery,

interferences

and LOD

min and max

Santos et

al.

2014

a

Mn, Ca, Cu, Fe,

Mg and Zn

non-

commercial

processed pulp

microwave

ICP-MS

Validation and

CRM

distribution

and ANOVA

Moreda-

Piñeiro et

al.

2018 Ca, Co, Cu, K,

Mg, Ni, P and Rb

commercial

supplement

enzymatic

hydrolysis -

microwave

ICP-MS

Recovery,

LOD

mean and st

dev

Santos et

al.

2014

b

Sm, Tb, Th, La,

Eu, Dy, Pr, Yb

and Tm

non-

commercial

processed pulp

microwave

ICP-MS

Validation and

CRM

PCA

ICP-OES: inductively coupled plasma optical emission spectrometry; ICP-MS: inductively coupled plasma mass spectrometry; NAA: neutron activation analysis; st dev: standard deviation; min: minimum; max: maximum; ANOVA: analysis of variance; PCA: principal component analysis; LOD: limit of detection; CRM: certified reference material; n.r.: not reported.

Moreover, in terms of the chemical analysis, there is a limited amount of

detail provided on the optimisation and calibration of instruments, with the

authors nominally referring to other referenced studies. A major criticism of most

published studies is the lack of validation data, with there being no evaluation of

the accuracy or precision (refer to glossary) of the measurements (refer to

glossary), analysis of certified reference materials or matrix-matched spiked

23

recovery tests (refer to chapter 2, section 2.3.4). There are two exceptions where

the publications provide details on the chemical analyses, namely, Pozebon et

al., 2015 for yerba mate and Llorent-Martínez et al., 2013 for açaí. The authors

present and discuss the limit of detection of the chosen methodology, spike

recovery of the analytes, and validation with the use of certified reference

materials. Also, very few studies report any critical statistical analysis of the

yerba mate data (refer to section 2.6). So, in conclusion the reported studies on

the elemental levels of yerba mate and açaí in Brazil, to date, may be questioned

in terms of the reliability of the data, which will be addressed in chapter 3 (yerba

mate) and chapter 5 (açaí).

In relation to the review of polyphenol analysis, researchers have

focussed on the appropriate extraction, separation, and identification of the

polyphenol compounds. The extraction methods should always be carefully

chosen, taking into consideration aspects, such as low use of organic solvents,

possible automation, effectiveness and selectivity (Plaza et al., 2018). Whist

physical extraction methods, such as filtration and grinding, can be simple and

effective for homogenising samples, chemical digestions can volatilise, change or

result in the loss of analytes. Moreover, the lack of commercially available

standards and the wide range of phenolic structures found in nature make the

identification of phenolic compounds a challenge (Plaza et al., 2018). Table 1.6

shows a scheme of the methodologies involved in the polyphenol analysis of

plants and foods, with a critical assessment of the important stages of the

analytical sequence (see above).

Furthermore, one of the limitations associated with sample selection is

that the research is usually performed on a single sample of plant or fruit

material. This represents a challenge to determine the representativity of the

levels of the polyphenol compounds in the plant species. The variability in the

different polyphenol levels can be due to agronomic and seasonal differences

(Timmers et al., 2017). Moreover, the variety of the different infusion methods,

also introduces a variability of the levels quantified. Finally, the selection of the

24

class of polyphenol analysed may result in an under-reporting and thereby an

under-estimating of the total polyphenol content (Donnelly, 2015).

Table 1.6: Analytical steps involved in the analysis of polyphenols in plants and

food (Plaza et al., 2018).

Sample pre-treatment Extraction Clean-up-

isolation Spectro-photometric

methods Advanced analytical

techniques

Grinding

Filtration

Centrifugation

Milling

Drying

Hydrolysis or

digestion

Conventional (SLE,

LLE, Soxhlet)

Advanced (SFE,

PLE, UAE, MAE,

CXLE, HHPE, PEF)

SPE

LLE

HSCCC

Total phenolics (Folin-

Ciocalteu assay)

Total flavonoids (AlCl3

assay)

Total

proanthocyanidins

(DMAC assay)

Total anthocyanins

(pH differential assay)

Total antioxidant

capacity (ABTS,

DPPH)

HPLC (UV/DAD,

FD, MS, CAD,

ECD)

Others (CE,

SFC, SFC, MS,

GC)

ABTS, 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulphonate; CAD, charged aerosol detection; CE, capillary electrophoresis; CXLE, carbon dioxide-expanded liquid extraction; DAD, diode array detection; DMAC, dimethylaminocinnamaldehyde; DPPH, 2,2-diphenyl-picrylhydrazyl; ECD, electrochemical detection; FD, fluorescence detection; GC, gas chromatography; HHPE, high hydrostatic pressure extraction; HPLC, high performance liquid chromatography; HSCCC, high-speed countercurrent chromatography; LLE, liquid-liquid extraction; MAE, microwave-assisted extraction; MS, mass spectrometry; PEF, pulse electric field; PLE, pressurized liquid extraction; SFC, supercritical fluid chromatography; SFE, supercritical fluid extraction; SLE, solid-liquid extraction; SPE, solid phase extraction; UAE, ultrasound assisted extraction; UV, ultraviolet.

1.6. Aim and Objectives

The overall aim of this research was to generate and provide, especially to

the small producers, a knowledge of the chemical composition and associated

health claims of major natural foodstuffs and beverages of Brazil. Brazil is a

major agricultural producer of specialist foodstuffs for national and global

consumption. The Brazilian economy is US$ 14 billion for the combined

exportation of yerba mate, coffee and açaí. To this end, the aim was to

determine; (i) the levels of chemical elements (major, minor and trace) and

25

polyphenols of yerba mate (from the southern region), coffee (São Paulo state)

and açaí berries (Amazonian region); and (ii) to assess the impact of

consumption in terms of the daily dietary intake on providing an adequate

nutrient supply for humans. Moreover, a review of the literature confirmed that

there is a limited amount of reliable data. It is essential to establish an analytically

robust method(s) that enables an evaluation of: (i) the impact of sample selection

and treatment; (ii) the influence of diluting prepared sample solutions before

instrumental analysis; and (iii) the calculation, statistical analysis and reporting of

date in relation to what is traditionally consumed by individuals in Brazil. Finally, a

key aspect of this research was to establish a database for future yerba mate,

coffee and açaí production and marketing of natural products from Brazil.

The specific objectives of the study were to:

(i) undertake and evaluate an extensive literature review of the

reported elemental and polyphenol levels in typical foodstuffs and

beverages from Brazil;

(ii) source a wide range of non-commercial and commercial samples of

yerba mate, coffee and açaí in Brazil;

(iii) establish and validate a sample preparation method for the

determination of elements (major, minor and trace);

(iv) develop and validate a sample preparation method to analyse the

samples (for elemental and polyphenolic content) as consumed by

the typical consumer;

(v) validate the instrumental analysis by using certified reference

materials and analyte spike recoveries of the products;

(vi) determine the elemental levels of the materials and the infusions by

inductively coupled plasma mass spectrometry (ICP-MS);

26

(vii) determine the polyphenol content of typical infusions of yerba mate,

coffee and açaí extractions by ultra-high performance liquid

chromatography (UHPLC) and ultra-violet visible spectroscopy (UV-

Vis);

(viii) review the reported potential dietary intake of polyphenols from

typical beverages and fruits from Brazil;

(ix) examine the relationship between sample type, origin, processing

and chemical levels; and

(x) evaluate the potential contribution of elements and polyphenols in

infusions of yerba mate (green loose, roasted, and tea-bags)

coffee (roasted) and açaí pulp in terms of human dietary intake.

Chapter 2 provides the analytical sequence for this research (Figure 2.1)

and the various stages relating to the sampling strategies, sample selection and

preparation and instrumental techniques (calibration and validation) for the

chemical (elemental and polyphenolic) analysis of yerba mate, coffee and açaí.

Chapters 3, 4 and 5, respectively, cover the research studies on yerba mate from

Brazil and Argentina, coffee from the São Paulo State of Brazil and açaí berries

and pulp from the Amazonian region of Brazil. Finally, chapter 6 reviews the

findings and presents ideas about future areas of investigation based on the

conclusions of this research.

27

Chapter 2. Methodology

28

2.1. Introduction

This chapter provides the details of the analytical procedures and

techniques used to achieve the Aim and Objectives (as set-out in chapter 1,

section 1.5). The analytical plan presented in Figure 2.1, was designed to outline

the critical analytical steps involved in this research, starting from the sample

collection procedures for the various studies on Brazilian products, namely yerba

mate (chapter 3), coffee (chapter 4) and açaí (chapter 5). This chapter follows

the analytical organisation presented in Figure 2.1, namely, a description of the

sample collection steps (section 2.2), followed by the pre-analysis sample

preparation steps (section 2.2.1). The chemical analysis stages of this study and

used throughout the thesis included: (i) the elemental content of samples

determined by inductively coupled plasma mass spectrometry (ICP-MS), as

described in section 2.3; (ii) the total polyphenol content if samples by ultraviolet-

visible spectroscopy (UV-Vis), outlined in section 2.4; and (iii) the polyphenol

profile and caffeine levels determined by high performance liquid

chromatography (HPLC), reviewed in section 2.5. The instrumental results were

subjected to computational data handling and statistical evaluation before

presentation in this thesis. The instrumentation sections (refer to sections 2.3.1,

2.4.1 and 2.5.1) describes the theory, specific instrument information, and the

optimisation and validation for each instrument and associated data analysis.

Finally, a summary of the data treatment, including the statistical analysis used in

this thesis is presented in section 2.6.

29

Figure 2.1: Analytical sequence adopted for this study of the chemical analysis of

typical beverages and açaí berries from South America.

2.2. Sample Collection

Commercial samples of yerba mate (Ilex paraguariensis), Brazilian coffee

(Coffea arabica L.) and açaí (Euterpe oleracea) were purchased from retail

outlets and commercial suppliers of Brazil and the UK (and Argentina for the

yerba mate). The non-commercial samples were obtained directly from

30

producers in Brazil and brought to the UK. The unopened commercial and trace

element free polypropylene or paper sealed bagged samples were stored in a

fridge (< 4°C) or a closed storage unit (< 15°C) until analysis. A detailed list of the

samples is given in chapters 3 (yerba mate), 4 (coffee beans and products) and 5

(acai berries and products).

Sample preparation

Prior to analysis, the samples needed to be decomposed using suitable

methods of digestion for elemental (or mineral) analysis. The term ‘mineral’ is

included in this description as the study of nutrition uses this term for chemical

elements in relation to human health requirements. A decomposition step is

necessary to reduce the amount of residual carbon in a sample and also to

release the analytes into solution. This decomposition of the sample is a critical

step. Whilst dry ashing methods (i.e. using a muffle furnace) may lead to analyte

loss, wet or acid digestion (i.e. using a water bath or a microwave apparatus)

requires exposing the sample to concentrated acids (nitric, hydrofluoric or aqua

regia (3:1v/v HCl:HNO3) and monitoring the digestion for long periods of time

(typically hours for open vessel methods). For this reason, these steps were

carefully monitored and optimised by using certified reference materials, to

ensure that the processes of ashing or wet acid digestion did not result in an

analyte loss. The following sample digestion procedure was initially optimised for

yerba mate samples and the resultant procedure was also used in the other two

studies (coffee and açaí).

Samples of 0.2500 ± 0.0010 g homogenised sample were weighed on an

analytical balance, transferred to an acid pre-washed ceramic crucible (in

duplicate) and placed in a Carbolite AAF 1100 muffle furnace at 500°C for 12

hours. The resultant ash was homogenised with 1 mL nitric acid (PrimarPlus-

Trace Analysis Grade 68%, Fisher Scientific, Loughborough, UK) in a fume

cupboard. The digest was transferred to a 25 mL Sterilin™ polypropylene tube

and then diluted with double-distilled deionised water (DDW, 18 18 MΩ cm) until

31

25 g (weighed, ± 0.0001 g). Each final solution was filtered using a 0.45 μm

Millex-HA membrane filter (Merck Millipore, Germany) and the resultant solutions

were analysed.

A blank for each digestion was prepared in order to evaluate any possible

sources of elemental contamination or loss during the digestion process. A blank

solution was produced following the same steps as that outlined for the

preparation of the samples, but without any material. All of the standards and

dilutions in this study were corrected by mass using a 4 decimal point analytical

calibrated balance (± 0.0001 g) instead of the traditional procedures using

volumetric flasks, in order to provide a better level of accuracy. This is especially

important for low concentrations, where the error is magnified.

In addition to the determining the total chemical content of the digested

samples, this study also evaluated the total elemental or polyphenol content

available in a regular serving of the foodstuff or beverage. Therefore, a series of

laboratory simulations based on traditional consumption methods were

performed and are detailed in each chapter: (i) regular infusion; iced tea and

bombilla method for yerba mate (refer to chapter 3); (ii) Brazilian brewing for

coffee (chapter 4); and extractions for açaí (chapter 5).

2.3. Inductively Coupled Plasma Mass Spectrometry (ICP-MS) for Elemental Analysis

Inductively coupled plasma mass spectrometry (ICP-MS) was used in this

work to determine the total elemental composition of the samples of yerba mate

(chapter 3), coffee (chapter 4) and açaí (chapter 5). Furthermore, this technique

was also used to determine the elemental levels of the extracted fraction or

infusions of yerba mate and coffee samples.

ICP-MS is capable of measuring the elemental (metals or non-metals) at

very low concentrations, typically sub µg/L (Balcaen et al., 2015). The main

advantages of ICP-MS are: simultaneous multi-element analysis, a wide linear

32

dynamic range (which is the linear part of the calibration curve that is used to

calculate the concentration of an analyte in the solution), very low detection limits

(typically < 0.01 µg/L) and good levels of precision (typically a relative standard

deviation of < ± 10%). Therefore, ICP-MS has been widely used for a variety of

samples matrix, such as environmental, foods and beverages, waters, soils and

geological materials (Baroni et al., 2015, Rousseau et al., 2013). The main

drawback of ICP-MS is the high mantainance costs.

The ICP-MS instrument consists of five main stages: sample introduction,

sample ionisation by the plasma (ICP), interface, mass spectroscopy (MS)

analyser and detector (Agilent, 2017), as shown in Figure 2.2.

The sample (normally as a liquid) is taken up by the instrument through a

peristaltic pump to the nebuliser where it is converted to an aerosol (particles

smaller than 10 µm in the Ar carrier gas) and transported to the inductively

coupled plasma or ICP. Typically, only 1-2% of the sample volume is carried to

the plasma; the remaining larger particles are condensed by the spray chamber

and is drained to waste (Becker, 2007, Thomas, 2013).

Figure 2.2: Instrumentation of an inductively coupled plasma mass spectrometer

(ICP-MS) Agilent Series (Agilent, 2017).

The ICP-MS torch is assembled horizontally and is made of three

concentric quartz tubes where the argon gas flows. The plasma is produced

33

when the carrier gas is passed into the ICP torch and a spark is applied (from a

Tesla coil) which provides electrons that collide with argon atoms causing the

ionisation of the gas (electron collision reaction i.e. X + e- → X+ + 2e-). The

resultant species interact with an intense electromagnetic field generated by the

RF power (typically 750–1500 W) applied to the copper coil, resulting in the

oscillation of an alternating current (AC) within the coil at the frequency of the RF

generator (27 or 40 MHz). When the aerosol sample reaches the plasma, the

high temperature enables the desolvation, atomisation and ionisation of the

elements present in the solution (Dean, 2003, Linge and Jarvis, 2009,

Beauchemin, 2010).

Since the mass spectrometer or MS operates at low pressures and room

temperature; and the plasma is at atmospheric pressure and a high temperature

(6 to 8000 K), the interface between these two parts is critical (Becker, 2007). It

consists of a sampling cone with an orifice of 1.0 mm diameter and a skimmer

cone with a 0.75 mm diameter, and both are cooled by water. They are made of

nickel which gives a high thermal conductivity (Thomas, 2013). When the analyte

ions pass through the first orifice, the pressure is reduced to 1 torr and then

through the second sampler, where the pressure reaches a level that is similar to

that in the MS (Harris, 2010). The resultant ion beam is separated by a negative

potential between electrons and positive ions and focussed by electrostatic

lenses (ion optics), which also removes neutral species and photons (Harris,

2010).

The University of Surrey ICP-MS instruments also include a

collision/reaction cell which is a multipole (octapole, hexapole or quadrupole)

which operates in a radio-frequency (RF) mode only (Thomas, 2013). It is located

between the ion lenses and the mass analyser. The reason is that an added gas

will remove polyatomic spectral interferences before the ions enter into the MS

(Becker, 2007). In this study, an Octopole Reaction System (ORS) or collision

cell with helium (He) gas addition was used in order to eliminate interferences,

such as 35Cl40Ar+, which is replaced with 35Cl4He+ to prevent any overlap with

75As+ (the only isotope for arsenic – 100% natural abundance). Also, helium was

34

used because it is an inert gas and does not react with the sample analytes. The

principle of a collision cell is to reject the interferences by reducing their kinetic

energy when colliding with He gas. Only the analyte ions are transmitted to the

MS, because of the difference on the potential of the octopole and MS (Thomas,

2013, Yamada, 2015).

The next part is the mass analyser, which is a quadrupole mass filter that

separates the positive ions according to their mass-to-charge ratio (m/z). A

quadrupole has four cylindrical rods of the same size (length typically of 15 to 20

cm and diameter of 1 cm). The opposing rods are connected to a direct current

(DC) and the other 2 to an alternating current (AC). This difference in the voltage

(positive and negative) causes the passage of the analyte through the

quadrupole and the specific charges only allow the ions of specific m/z values to

reach the detector (Linge and Jarvis, 2009).

Finally, the detector converts the ions into electrical signals through a

channel electron multiplier. The analytes from the MS hit a surface coated with a

semiconductor material, has a negative potential at the first end and produce one

or more secondary electrons. These electrons move and hit another new surface

and emit more electrons until the generation of a pulse is detected (Thomas,

2013, Skoog et al., 2017).

Instrumentation – ICP-MS

The majority of the analysis performed on this study used an ICP-MS

Agilent 7800 Series (Agilent Technologies, UK) with an SPS 4 series

autosampler controlled through the use of Agilent software (MassHunter). This

Agilent ICP-MS instrument has an Octopole Reaction System (ORS). Sample

introduction is through the quartz, Peltier-cooled, Scott-type double-pass spray

chamber. Part of the studies (commercial samples of yerba mate and coffee;

refer to chapters 4 and 5) was also performed on a similar ICP-MS Agilent 7700x

Series (Agilent Technologies, UK) performed with the same features. The typical

operating conditions for both instruments are detailed in Table 2.1. The isotopes

35

were used to determine the concentration of the analyte elements. Instrument

optimisation was performed before each analysis using a 1 μg/L tuning solution

containing 7Li, 24Mg, 89Y, 140Ce, 204Tl and 59Co (Agilent Technologies, UK). The

sensitivity of the instrument for the selected elements under investigation was

enhanced before each analysis by the optimisation of the nebuliser flow rate and

RF power.

Table 2.1: Typical operating conditions for the Agilent ICP-MS instruments.

Part Parameter Agilent 7700x Agilent 7800

Sample

introduction

Nebuliser MicroMist MicroMist

Carrier gas flow 0.8 L/min 0.9 L/min

Nebuliser pump 0.3 rps 0.1 rps

Spray Chamber

Temperature 2 ºC 2 ºC

Plasma

condition

RF Power 1550 W 1550 W

RF Matching 1.95 V 1.80 V

Sampling Depth 8 mm 8 mm

Collision

cell

[1] ON He mode He mode

He gas flow 4.8 mL/min 4.3 mL/min

Isotopes

23Na, 24Mg, 39K, 40Ca, 51V,

52Cr, 55Mn, 56Fe, 59Co, 60Ni,

63Cu, 66Zn, 75As, 78Se, 95Mo

23Na, 24Mg, 39K, 40Ca, 51V,

52Cr, 55Mn, 56Fe, 59Co, 60Ni,

63Cu, 66Zn, 75As, 78Se, 95Mo

[2] OFF No gas mode No gas mode

Analyte elements 111Cd, 208Pb 111Cd, 208Pb

Detector

parameters

Type of detector Electron multiplier Electron multiplier

Pulse HV 980 V 925 V

Analog HV 1680 V 2163 V

Data

acquisition

Sample uptake time 50 s 40 s

Stabilisation time 30 s 25 s

Sample wash time 120 s 65 s

Internal standards – ICP-MS

Internal standards (ISTDs or IS) are used to correct for any drift in the

signal intensity resulting from an instrument issue, such as blockages, leaks and

36

any change in the instrument conditions during operation (Lord, 2014). The

correction of the data was performed by blank subtraction of the sample counts

per second (CPS), followed by this value being ratioed with the internal standard

CPS signal as shown in Equation 2.1:

𝑅𝑎𝑡𝑖𝑜 =𝐶𝑃𝑆 𝑎𝑛𝑎𝑙𝑦𝑡𝑒 𝑠𝑖𝑔𝑛𝑎𝑙 − 𝐶𝑃𝑆 𝑏𝑙𝑎𝑛𝑘 𝑠𝑖𝑔𝑛𝑎𝑙

𝐶𝑃𝑆 𝐼𝑆𝑇𝐷 𝑠𝑖𝑔𝑛𝑎𝑙

Equation 2.1

A 100 μg/L internal standard (IS) solution of indium (115In) was prepared

from a 1000 μg/mL stock solution (Aristar, UK) in 1% HNO3 (High Purity

Analytical Grade, Fisher Scientific, Loughborough, UK). In this study, indium was

used as the ISTD due to its mid-first ionisation potential and absence in the yerba

mate, coffee and açaí samples.

Limit of detection (LoD) and linear dynamic range (LDR) – ICP-MS

All instruments have a degree of noise associated with the measurements

that limits the precision of the background signal. Therefore, the instrumental limit

of detection (LoD) is the minimum concentration of the analyte that can be

determined to be different from the signal of the blank (Miller et al., 2018), and it

was calculated using Equation 2.2. In this study, the average of 10 replicate

signal values of 1% v/v HNO3 were used as a blank for the digested samples and

DDW for the yerba mate and coffee infusion samples. The instrumental LoDs for

the investigated elements are presented in Table 2.2:

𝐿𝑜𝐷 = 𝑦𝑏 + (3 𝑥 𝑆𝐷)

Equation 2.2

Where:

𝑦𝑏 is the mean blank signal; and

37

𝑆𝐷 is the standard deviation of the blank signal (Miller et al., 2018).

Table 2.2: Investigated isotopes; limits of detection (LoD) in 1% HNO3 and

double-distilled deionised water (DDW) using 115In as the internal

standard for the Agilent 7800 ICP-MS.

Isotope Natural Abundance (%) LoD in 1% HNO3 (µg/L) LoD in DDW* (µg/L) 23Na 100 0.09 0.06

24Mg 79.0 0.08 0.07

39K 93.3 0.07 0.05

40Ca 96.9 0.08 0.06

51V 99.7 0.02 0.05

52Cr 83.8 0.08 0.05

55Mn 26.1 0.03 0.03

56Fe 97.1 1.00 0.50

59Co 100 0.07 0.05

60Ni 26.1 0.06 0.04

63Cu 69.2 0.09 0.05

66Zn 27.9 0.20 0.10

75As 100 0.04 0.02

78Se 23.6 0.04 0.02

95Mo 15.9 0.09 0.09

111Cd* 12.8 0.03 0.03

208Pb* 52.4 0.03 0.03

*collision cell off.

The linear dynamic range (LDR) usually refers to the range of analyte

concentrations in which the detector produces a signal proportional to the

concentration of the analyte, as exemplified in Figure 2.3. A linear regression

equation (y = mx + c) can be calculated within the LDR using the signal (y-axis)

of a series of accurate standards with known concentrations (x-axis); this

equation will allow for the prediction of the concentration of unknown samples

(Miller et al., 2018). In order to prepare the calibration curves for the different

analytes, a range of 1 – 1500 µg/L multi-element solutions were prepared from a

38

1000 mg/L stock solution (TraceCERT®, Sigma-Aldrich, UK) for each element

and diluted in 1% v/v nitric acid.

Figure 2.3: Calibration curve for manganese (55Mn) using 115In as the internal

standard for the Agilent 7800 ICP-MS, in the helium collision cell

mode.

In this study, the range of elemental concentrations found in the samples

(digest solutions and infusions) differs significantly. Therefore, the use of the

appropriate LDR was evaluated to provide an accurate calculation procedure for

determining the analyte concentration of the unknown solution. This is a major

problem in many studies where a calculation curve over an LDR of say 1 to 1500

µg/L is used to calculate the concentration of an unknown solution that has a

digest solution value of 2 µg/L. The higher standards may have a significant

influence on the calculated level of accuracy, especially if the highest standards

are affecting the linearity of the curve. Therefore, the specific linear dynamic

ranges for each type of samples are presented on Table 2.3.

0 200 400 600 8000

500

1000

1500

Mn concentration (mg/L)

Rati

o C

PS

Mn

/ In

Y = 1.6619x + 6.4887R2 = 0.9995

Text

(µg/L)

39

Table 2.3: Investigated isotopes and linear dynamic range (µg/L) used for each of

the type of samples (yerba mate, coffee and açaí) used in this study.

115In was used as the internal standard for the Agilent 7800 ICP-MS

analysis.

Isotope Yerba mate (µg/L) Coffee (µg/L) Açaí (µg/L) 23Na 0 - 1500 0 - 1500 0 - 1500

24Mg 0 - 1500 0 - 1500 0 - 1500

39K 0 - 1500 0 - 1500 0 - 1500

40Ca 0 - 1500 0 - 1500 0 - 1500

51V 0 - 500 0 - 500 0 - 500

52Cr 0 - 750 0 - 500 0 - 500

55Mn 0 - 1500 0 - 1000 0 - 1500

56Fe 0 - 1500 0 - 1500 0 - 1500

59Co 0 - 500 0 - 500 0 - 250

60Ni 0 - 750 0 - 500 0 - 500

63Cu 0 - 1000 0 - 750 0 - 1500

66Zn 0 - 1000 0 - 750 0 - 1500

75As 0 - 250 0 - 250 0 - 250

78Se 0 - 250 0 - 250 0 - 250

95Mo 0 - 250 0 - 250 0 - 250

111Cd* 0 - 250 0 - 250 0 - 250

208Pb* 0 - 250 0 - 250 0 - 250

*collision cell off

Validation (accuracy and precision) – ICP-MS

Validation is an important step for the quality control evaluation of an

analytical technique, to assess if the method is fit-for-purpose. The precision or

repeatability (refer to glossary) describes the level of agreement between

replicate analysis of the same sample within the same instrument conditions. On

the other hand, reproducibility (refer to glossary) reflects the precision of replicate

analysis under different conditions (between-run), such as analysing the same

sample over different days (Miller et al., 2018). In this study, the instrumental

performance was validated by the comparative analysis of a water certified

40

reference material (CRMs), namely, NIST SRM 1640a (National Institute of

Standards and Technology, USA) for trace elements and CRM 3 Multielement

standard solution for ICP (Fluka Analytical, Sigma-Aldrich, Poole, UK) for Ca, Na,

K and Mg as shown on Table 2.4. The study was validated for the level of

accuracy, that is, how close the calculated mean value is from the true certified

value, by comparing the values in Table 2.4 and precision by the use of the

relative standard deviation (%). In general, there is a good agreement between

the certified and calculated values.

It is also important to validate the digestion methods against a matrix-matched

standard reference material. This material has to be analysed under the same

conditions as the samples and compared with the certified value. In this study,

two plant reference materials were used, Tea Leaves INCT-TL-1 (Instytut Chemi

i Techniki Jadrowej, Poland) and Peach Leaves SRM 1547 (National Institute of

Standards and Technology, USA). The accuracy of the elemental measurements

was determined comparing the measured concentration of each element and the

certified value of the CRM. The results are presented in Tables 2.5 (a) and (b). In

general, there is a good agreement between the certified and calculated values.

41

Table 2.4: Evaluation of accuracy (comparison of measured and certified

elemental concentrations) and precision (relative standard deviation

(RSD %)) of for NIST SRM 1640a and CRM 3 (for Na, Mg, K and

Ca).

Element Certified

value (µg/L)

Calculated value (µg/L) RSD (%)

n = 20a

n = 5b

n = 20a

n = 5b

Na 1000 ± 3 1034.58 ± 23.64 998.27 ± 17.84 2.3 1.8

Mg 400 ± 2 393.53 ± 12.52 399.56 ± 18.24 3.2 4.6

K 200 ± 3 188.28 ± 5.67 193.94 ± 4.99 3.0 2.6

Ca 2000 ± 5 1952.94 ± 23.63 1980.26 ± 21.74 1.2 1.1

V 14.93 ± 0.21 15.12 ± 0.43 15.07 ± 0.34 2.8 2.3

Cr 40.22 ± 0.28 38.58 ± 1.78 39.63 ± 0.94 4.6 2.4

Mn 40.07 ± 0.35 38.80 ± 1.30 39.78 ± 1.02 3.4 2.6

Fe 36.5 ± 1.7 40.10 ± 3.00 38.42 ± 1.52 7.5 4.0

Co 20.08 ± 0.24 19.57 ± 0.58 19.85 ± 0.84 3.0 4.2

Ni 25.12 ± 0.12 24.90 ± 1.23 25.01 ± 1.03 4.9 4.1

Cu 85.07 ± 0.48 84.42 ± 3.58 84.74 ± 2.51 4.2 3.0

Zn 55.2 ± 0.32 55.20 ± 0.62 55.01 ± 0.34 1.1 0.6

As 8.010 ± 0.067 7.88 ± 0.07 8.00 ± 0.09 0.9 1.1

Se 19.97 ± 0.16 19.31 ± 1.36 19.12 ± 1.07 7.0 5.6

Mo 45.24 ± 0.59 52.54 ± 2.64 43.92 ± 0.99 5.0 2.3

Cd 3.961 ± 0.072 3.92 ± 0.11 3.96 ± 0.19 2.8 4.8

Pb 12.005 ± 0.040 12.65 ± 0.65 11.93 ± 0.78 5.1 4.9

a : reproducibility, b: repeatability; n is the number of measurements; RSD is relative standard deviation (%).

42

Table 2.5 (a): Comparison of the certified reference values (CRM) with the

calculated concentrations for the Tea Leaves INCT-TL-1 certified

reference material.

Element Certified value (mg/kg)

Calculated value (mg/kg)

Na 24.7 ± 3.2 28.32 ± 5.16

Mg 2240 ± 170 2452.57 ± 183.28

K 17000 ± 1200 16449.19 ± 1374.10

Ca 5820 ± 520 5639.97 ± 630.38

V 1.97 ± 0.37 1.57 ± 0.85

Cr 1.91 ± 0.22 1.82 ± 0.95

Mn 1570 ± 110 1352.85 ± 129.19

Fe 432a 540.28 ± 19.20

Co 0.387 ± 0.042 0.50 ± 0.06

Ni 6.12 ± 0.52 5.82 ± 0.59

Cu 20.4 ± 1.5 21.73 ± 2.01

Zn 34.7 ± 2.7 30.57 ± 3.20

As 0.106 ± 0.021 0.09 ± 0.01

Se 0.076a 0.05 ± 0.02

Mo - -

Cd 0.030 ± 0.004 0.03 ± 0.01

Pb 1.78 ± 0.24 1.38 ± 0.43

ainformation value; all values reported on a dry weight basis.

43

Table 2.5 (b): Comparison of the certified reference values (CRM) with the

calculated concentration for the Peach Leaves SRM 1547

certified reference material.

Element Certified value (mg/kg)

Calculated value (mg/kg)

Na 23.8 ± 1.6 21.93 ± 2.43

Mg 4320 ± 150 4294.75 ± 146.23

K 24330 ± 380 25021.58 ± 353.92

Ca 15590 ± 160 14985.65 ± 156.28

V 0.367 ± 0.038 0.29 ± 0.06

Cr 1a 0.86 ± 0.08

Mn 97.8 ± 1.8 95.73 ± 2.54

Fe 219.8 ± 6.8 209.35 ± 8.30

Co 0.07a 0.03 ± 0.01

Ni 0.689 ± 0.095 0.79 ± 0.08

Cu 3.75 ± 0.37 3.45 ± 0.46

Zn 17.97 ± 0.53 16.36 ± 0.60

As 0.062 ± 0.014 0.05 ± 0.02

Se 0.120 ± 0.017 0.10 ± 0.03

Mo 0.0603 ± 0.0068 0.05 ± 0.02

Cd 0.0261 ± 0.0022 0.02 ± 0.00

Pb 0.869 ± 0.018 0.79 ± 0.04

ainformation value; all values reported on a dry weight basis.

2.4. UV-Vis Spectroscopy for the Total Polyphenol Content Analysis

Molecules can absorb part of the radiation, when exposed to light energy.

Consequently, electrons move from a lower energy state (ground) to a higher

one (excited). Ultraviolet–visible spectroscopy refers to the absorbed energy in

the ultraviolet and adjacent visible spectra and can be translated into an

absorbance spectrum (between the range of wavelengths of 200 to 700 nm)

(Pavia et al., 2014). Also, the extent of light absorbance is expressed by the

Beer-Lambert law, as presented in Equation 2.3:

44

𝐴 = log (𝐼𝑜

I ) = ε c l

Equation 2.3

Where:

𝐴 = absorbance;

𝐼𝑜 = intensity of incident light;

I = transmitted intensity;

ε = molar absorptivity of solute;

c = molar concentration of solute; and

l = path length of sample cell.

Since the path length of the sample cell is usually fixed and the molar

absorptivity is a specific constant, the absorbance of a sample is directly

proportional to the sample concentration. For most of the molecules, this

relationship is linear over a certain concentration range (Harvey, 2006).

Instrumentation – UV-Vis

The UV-Vis instrument has 3 main component parts – a light source, a

monochromator and a detector. As a light source, the UV-Vis spectrophotometer

has a deuterium lamp to emit light in the ultraviolet region and a tungsten

halogen lamp to emit light in the visible region. The light radiation is filtered

through an optical filter before passing through the slit to the monochromator in

order to purify the signal (Skoog et al., 2017).

The monochromator consists of a diffraction prism that disperses the light

radiation to produce a spectrum. If the prism is rotated, the desired wavelength

segment of the spectrum is selected. The radiation then passes through a beam

splitter which permits the radiation to pass through the reference cell or through

the sample cell (1 cm length transparent plastic cells). After the sample and

reference cells, the radiation passes through convex lens into the detector. When

the light radiation arrives at the detector, the light energy is converted into an

45

electrical current (Pavia et al., 2014). The extent of the electrical current is

proportional to the amount of light radiation arriving at the detector. A BioChrom

Libra UV-Vis spectrophotometer was used for the determination of the total

polyphenol content of samples in this study.

Total polyphenol content by Folin-Ciocalteu analysis

The Folin-Ciocalteu assay is used to quantify the total polyphenol content

of a sample throughout a reduction-oxidation (redox) reaction where the Folin-

Ciocalteu reagent reacts with the electrons from a reducing agent (e.g.

polyphenols) (Huang et al., 2005). The Folin-Ciocalteu reagent is a mixture of

phosphomolybdic and phosphotungstic acids. Although the chemistry of the

Folin-Ciocalteu is still unknown, the isopolyphosphotungstates are colourless

when fully oxidised, and the analogous molybdenum compounds are yellow

(Prior et al., 2005). The end-product reduction produces a blue coloured complex

with a maximum absorbance around 765 nm (Singleton et al., 1999b, Ainsworth

and Gillespie, 2007). Although the Folin-Ciocalteu assay is an excepted method

to determine the total polyphenol content, a limitation of the method is that the

reagent can also react with a series of interferences, such as ascorbic acid and

sugars, that could also be present in the samples. This may be especially true for

fruit juice samples (Prior et al., 2005). The samples analysed in this study did not

had a significant concentration of ascorbic acid or sugars.

The total polyphenol content of the yerba mate and coffee infusions and

açaí extracts was determined by the Folin-Ciocalteu assay, as detailed in ISO

14502-1 (ISO, 2005). A 1 mL aliquot of the diluted infusion or extract was placed

in a 15 mL polypropylene plastic test tube and 5 mL of 10% Folin-Ciocalteu

(Fisher Scientific, Loughborough, UK), freshly prepared in deionised water was

added and mixed by a vortex mixer. After 3 to 8 mins, 4 mL of 7.5 % w/v sodium

carbonate solution (Sigma-Aldrich, Poole, UK) was added and the tubes were

mixed using a vortex mixer. The tubes were left for 1 hour at room temperature

before the measurement at 765 nm. All the sample analyses were performed in

46

duplicate due to the limited amount of available sample. The results are

expressed in gallic acid equivalent (Sigma-Aldrich, Poole, UK), which was used

as a calibration over the range of 0 to 50 µg/mL, as shown in Figure 2.4. The

infusion and extract samples were diluted appropriately in deionised water to fit

the absorbance levels within the calibration range of the reference standards.

The limit of detection of the total polyphenol content was calculated for this assay

at 0.16 µg/mL gallic acid equivalent (based on the linear regression of the

calibration curve).

Figure 2.4: Gallic acid calibration curve obtained by the Folin-Ciocalteu assay

using a UV-Vis instrument (refer to section 2.4.2)

2.5. High Performance Liquid Chromatography (HPLC) for Polyphenol Profile Analysis

High performance liquid chromatography (HPLC) is an analytical

technique of separating, quantifying and identifying components in a mixture. In

HPLC, separation is based upon the differential distribution of analyte molecules

between two phases, a mobile and a stationary phase. Each component of the

mixture interacts differently with the two phases, and at different flow rates for

each component, leading to separation as they travel through the column

0 20 40 600.0

0.2

0.4

0.6

0.8

Gallic acid concentration (µg/mL)

Ab

so

rban

ce

y = 0.011x - 0.0008 R² = 0.9999

47

(Niessen, 2006). This technique relies on the pumping of a pressurised liquid

solvent containing the sample mixture (mobile phase) through a column filled

with a adsorbent material (stationary phase) (Snyder et al., 2011, García-

Álvarez-Coque et al., 2017). The resolution of the analytes in chromatography

depends on three parameters: efficiency, retention and selectivity as described in

Equation 2.4. Method development can maximise the resolution by optimising

each of these terms:

𝑅 = (√𝑁4 ) (

𝜅𝜅 + 1) (

𝛼 − 1𝛼 )

Equation 2.4

where:

𝑅 = resolution;

𝑁 = number of theoretical plates;

𝜅 = the retention or capacity factor; and

𝛼 = the separation factor (Dolan and Snyder, 2013).

An increase in the efficiency (refer to Equation 2.4), can be achieved by

increasing the length of the column or the porosity of the stationary phase or

reducing the thickness and particle size of the column (Ali et al., 2012, Cabooter

and Desmet, 2012, Skoog et al., 2017).

The retention factor (Equation 2.4), can be modified by variation of the

mobile phase, in order to have an effect on the hydrophobicity or hydrophilicity

(refer to glossary) interaction between the analytes and the mobile phase

(Fountain and Iraneta, 2012). Also, modification of the elution strength can be

achieved by changing the pH, thereby affecting the retention time (Harvey,

2006).

The selectivity (Equation 2.4) can be changed by modifying the type of

molecular interaction. For example, in the reversed phase (polar mobile phase

Efficiency Retention Selectivity

48

and non-polar stationary phase), the analytes are selected by the hydrophobic

interactions between the stationary and mobile phase. Additional polar functional

groups can change the selectivity of an analyte (Snyder et al., 2011, García-

Álvarez-Coque et al., 2017).

Instrumentation - HPLC

In this study, a reverse phase HPLC was preferred due to the polar nature

of the polyphenol compounds, providing a better separation of the analytes.

Therefore, the column contained a non-polar stationary phase and a polar

compound mobile phase.

The determination of the polyphenolic profile and caffeine levels of the

yerba mate and coffee infusions was performed following the optimised method

proposed by Donnelly (2015) for yerba mate infusions samples. This method was

developed in order to optimise the resolution (refer to Equation 2.4) by selecting

the best column (efficiency and selectivity factor) and mobile phase composition

and gradient (retention factor) (Donnelly, 2015). An ultra-high performance liquid

chromatography or UHPLC instrument was used with almost double the overall

operating pressure (to 15,000 psi) in order to obtain more rapid flow rates and

achieve better resolution separations in shorter time frames (de Souza et al.,

2010). A comparison of using HPLC and UHPLC instruments for chlorogenic acid

analysis of yerba mate was completed by Donnelly (2015), who showed that it

was possible to decrease the run time from 60 to 30 minutes. Therefore, a

Waters Acquity UPLC® (Waters, Milford, USA) instrument was used for the

yerba mate and coffee analysis. It was fitted with a binary solvent manager and

photodiode array detector (PDA) and controlled by Empower 3 chromatography

software (Waters, Milford, USA). The açaí polyphenolic analysis was performed

as described in the respective chapter (refer to section 5.5.7).

The method validation for the polyphenol analysis was performed by

Donnelly (2015). The matrix effects were determined using 3 independent

measurements of the reference materials at 5 different concentrations prepared

49

in water and methanol extracts of green tea, rooibos and hibiscus. The standard

solution curve was plotted against the standards prepared in the sample matrix.

The resultant slopes and intercept were calculated. The levels added were

equivalent to 65 – 87% of the matrix response of the extracts (Donnelly, 2015).

The recovery and precision of the method were also evaluated and ranged from

90 to 103 % and the coefficient of variation of 0.82 % (Donnelly, 2015).

The separation of the compounds was conducted by injecting a 5 μL

sample onto a Phenomenex (Macclesfield, Cheshire, UK) Kinetex© PFP column,

with dimensions 4.6 x 100 mm x 2.6 μm 100 A, held at a constant temperature of

25°C, a flow rate of 0.7 mL/min and controlled by a gradient programme, as

shown in Table 2.6. Data was collected at wavelengths of 280 and 320 nm.

Table 2.6: Gradient programme for UHPLC analysis of polyphenol and caffeine in

yerba mate and coffee infusions.

Time (min) 5% formic acid (%) 80% ACN*, 5 % formic acid (%) 0.1 95 5

8 92.5 7.5

22.3 60 40

24.6 0 100

27 0 100

28 95 5

29.5 95 5

* ACN = acetonitrile.

The polyphenol and caffeine levels were quantified using a 5-point

calibration curve for a blank and reference standards against the peak area (AU)

of a chromatogram. The reference materials of caffeine, 3-caffeoylquinic acid, 4-

caffeoylquinic acid and 5-caffeoylquinic acid were obtained from Sigma (Sigma-

Aldrich, Poole, UK) and prepared over the range of 0-150 mg/L for 5-

caffeoylquinic acid, 0-70 mg/L for 3-caffeoylquinic acid and 4-caffeoylquinic acid

and 0-40 mg/L for caffeine, as shown in Figure 2.5. The mixed standard solutions

were stored in a frozen state until used. The calculated limit of detection (based

on the linear regression of the calibration curve) calculated for caffeine was 0.047

50

mg/L, and the following for the polyphenols: 0.035 mg/L for 3-caffeoylquinic acid,

0.033 mg/L for 4-caffeoylquinic acid and 0.038 mg/L for 5-caffeoylquinic acid.

Figure 2.5: 5-caffeoylquinic acid calibration curve obtained by the UHPLC

analysis (refer to section 2.5.1).

2.6. Statistical Analysis

The statistical analysis applied in this study were carried out using the

statistical software packages GraphPad Prism 6 and IBM® SPSS® Statistics

version 20. After the analysis of the descriptive statistics, such as mean,

standard deviation and relative standard deviation; the D'agostino and Pearson

test was used to study the normality of the data (Miller et al., 2018). Where the

data followed a normal distribution, parametric tests were used, such as two-

tailed Student t-test, paired two-tailed t-test and analysis of variance; and non-

parametric tests for data not normally distributed, such as the Spearman‘s rank

(refer to chapter 5) (Miller et al., 2018).

The arithmetic mean of the measurements was calculated by the sum of

the values divided by the number of measurements (n). The standard deviation

describes the spread of the experimental values around the arithmetic mean and

is defined by Equation 2.5 (Miller et al., 2018):

0 50 100 1500

2

4

6

5-caffeoylquinic acid concentration (mg/L)

Peak A

rea x

10

6 (A

U)

Y = 0.040x + 0.023R2 = 0.9998

51

𝑠 = √∑(𝑥𝑖 − ��)2

𝑛 − 1

Equation 2.5

where:

𝑠 = standard deviation;

𝑥𝑖 = value;

�� = arithmetic mean; and

𝑛 = the number of samples.

The level of precision is then determined by calculating the relative

standard deviation or s/�� x 100 (%). The lower the % rsd value the better the

level of precision (Miller et al., 2018).

D'Agostino and Pearson normality test

In order to check if a set of data is normally distribuited or a quantification

of how far a set of data is from a Gaussian distribution, the D'Agostino and

Pearson test was used. In this test, the Skewness is first calculated relating to

the symmetry of the set of data and then the Kurtosis, which quantifies the peak

shape of the distribution. Finally, the D'Agostino and Pearson test combines both

results into a single value. Assuming that the data are not normally distributed; if

the probability or p value is small (< 0.05 for 95% confidence level) then the null

hypothesis will be rejected (Motulsky, 2014).

Significance tests

In order to compare the means of a certified population with the calculated

one a two-tailed Student t-test was performed, Equation 2.6 shows the

calculation of the parameter t (Miller et al., 2018). The null hypothesis in this test

52

is that the certified population is equal to the calculated one at a probability p =

0.05; if tcalc is lower than the tcrit value, then the null hypothesis is not rejected:

𝑡𝑐𝑎𝑙𝑐 =(�� − 𝜇) √𝑛

𝑠

Equation 2.6

where:

𝑡𝑐𝑎𝑙𝑐 = t calculated;

�� and 𝑠 = sample mean and standard deviation;

𝑛 = number of samples; and

𝜇 = mean of the certified population.

It is also important to compare the standard deviations between two

populations to evaluate the random error of two sets of data (Miller et al., 2018).

In order to test whether the difference between the variances is significant, a F-

test was performed by the ratio of the squares of the standard deviations. The

null hypothesis states the variance are not significant at a probability p = 0.05. If

the calculated value of F is lower than the critical value, then the null hypothesis

is not rejected (Miller et al., 2018).

In terms of evaluating the paired values from the same set of samples (i.e.

comparison of the results obtained for the same cohort of samples using two

different analytical methods), a paired two-tailed t-test was performed using

Equation 2.7 for the calculation of the parameter t (Miller et al., 2018). The paired

t-test was used to compare two extraction methods in the same set of samples in

Chapter 5. The null hypothesis in this test states that the means are not

significantly different at a probability p = 0.05; if tcalc is lower than the tcrit value

then the null hypothesis is not rejected:

53

𝑡𝑐𝑎𝑙𝑐 =�� √𝑛

𝑠𝑑

Equation 2.7

where:

𝑡𝑐𝑎𝑙𝑐 = t calculated;

�� and 𝑠𝑑 = mean and standard deviation of differences between the

paired values; and

𝑛 = number of samples.

In order to compare the means of sets of data with more than two

variables, a Kruskal-Wallis test was performed (refer to chapter 5) (Miller et al.,

2018). The values are arranged in an ascending order (1, 2,... to N; where N is

the number of samples) and given a rank value. The sum of the ranks was used

to calculate the chi-squared, as presented in Equation 2.8. The calculated value

is compared to the critical value. The null hypothesis in this test states that there

is no statistically significant difference between the variables; and the null

hypothesis is not rejected if the calculated value is less than the critical value

(Corder and Foreman, 2011):

𝑋2 = 12

𝑁 + 𝑁2 (𝑅𝑖

2

𝑁𝑖+ ⋯ +

𝑅𝑘2

𝑁𝑘) − 3 (𝑁 − 1)

Equation 2.8

where:

𝑋2= chi-squared;

𝑁 = number of samples; and

𝑅𝑖 = sum of the rank for a particular data set.

54

Correlation coefficients

In order to evaluate the direction and magnitude of a linear relationship

between sets of data (𝑥 and 𝑦); the Pearson product moment correlation

coefficient (𝑟𝑝, refer to Equation 2.9) was used if the data was normally

distributed and the Spearman rank correlation coefficient (refer to Equation 2.11)

if it was not normally distributed (Corder and Foreman, 2011, Miller et al., 2018).

The coefficient values range from -1 to +1, where -1 indicates a strong negative

correlation; 0 no correlation; and +1 a strong positive correlation. If the Pearson

product moment correlation coefficient is lower than 0.8, a two-tailed Student t-

test was used to determine the level of significance (refer to Equation 2.10). The

null hypothesis states that there is no correlation between the two sets of data. If

the calculated value exceeds the critical value, the null hypothesis is rejected and

the two data sets are significantly correlated:

Pearson product moment correlation

𝑟𝑝 =∑ {(𝑥𝑖 − ��)(𝑦𝑖 − ��)}𝑖

{[∑ (𝑥𝑖 − ��)2𝑖 ][∑ (𝑦𝑖 − ��)2

𝑖 ]}12

Equation 2.9

where:

𝑟𝑝 = Pearson product moment correlation coefficient;

𝑥𝑖 and 𝑦𝑖 = values of sets; and

�� and �� = average value of individual sets.

Two-tailed Student t-test

|𝑡| =|𝑟𝑝|√𝑛 − 2

√1 − 𝑟𝑝2

, 𝑤ℎ𝑒𝑟𝑒 𝑑𝑒𝑔𝑟𝑒𝑒𝑠 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑛 − 2

Equation 2.10

55

where:

𝑡 = calculated t-test value;

𝑟𝑝 = Pearson product moment correlation coefficient; and

𝑛 = number of sample pairs.

Spearman rank correlation coefficient

𝑟𝑠 =6 ∑ 𝑑𝑖

2𝑖

𝑛(𝑛2 − 1)

Equation 2.11

where:

𝑟𝑠 = Spearman rank correlation coefficient;

𝑑𝑖 = difference between each ranking pair; and

𝑛 = number of sample pairs.

2.7. Summary

An overall analytical plan to address the aim and objectives of this study

was presented in section 2.1. The type and collection of the samples were

described in section 2.2 and the preparation for each analysis in section 2.2.1.

The total polyphenol content was determined by the Folin-Ciocalteu assay using

a BioChrom Libra UV-Vis spectrophotometer, as described in section 2.3.1. The

assay was described in section 2.3.2. The polyphenol analysis presented in this

study for the yerba mate (chapter 3) and coffee infusions (chapter 4) were

performed using a Waters Acquity UHPLC® instrument. The açaí extract

samples were analysed as outlined in chapter 5. The instrument and HPLC

theory were described in section 2.4. For all of the samples, elemental analysis

56

was performed by an Agilent 7800 inductively coupled plasma mass

spectrometer (ICP-MS). The theory and instrumentation for ICP-MS were

described in section 2.5.1. and the use of internal standards was explained in

section 2.5.2. Also, the analytical figures of merit (refer to glossary), such as, limit

of detection (LoD) and linear dynamic range (LDR) were described for all

elements in section 2.5.3. Furthermore, the validation of the instrument was

undertaken by the analysis of water certified reference materials (CRMs) and the

validation of the method through the analysis of plant certified reference

materials. These measurements were shown to be accurate for the purpose of

providing quality data (refer to section 2.5.4). Finally, the statistical analysis plan

used to evaluate the data presented in this study was summarised in section 2.6

and the normality test was outlined in section 2.6.1. Furthermore, the significance

tests applied in this study was presented in section 2.6.2, along with the

correlation analysis in section 2.6.3.

57

Chapter 3. Yerba Mate

58

3.1. Introduction

Yerba mate (Ilex paraguariensis), a native plant from the southern region

of Latin America, has been gaining attention due to its high levels of caffeine and

antioxidants. This chapter provides an overview of the literature available on

yerba mate (refer to section 3.2) and the relationship of its consumption with

human health (section 3.3). The aim and objectives of this chapter are outlined in

section 3.4. In this research, an evaluation of the non-commercial yerba mate

was carried–out with samples from Barão de Cotegipe, located in the Rio Grande

do Sul State, Brazil, according to the methodology described in sections 3.5.1

and 3.5.2. The elemental results are presented in sections 3.5.3, 3.5.4 and 3.5.5.

Furthermore, an investigation of the chemical composition of the commercial

yerba mate samples obtained from outlets in Brazil and Argentina is described in

section 3.6.1. All samples were analysed for the polyphenolic and elemental

content using the methodologies outlined in section 3.6.2 and the results are

presented in sections 3.6.3, 3.6.4 and 3.5.5. Also, a link to the dietary intake of

these chemicals was evaluated through the preparation of traditional infusions,

as outlined in section 3.6.6. Finally, a summary of the data is presented in

section 3.7.

3.2. General Introduction to Yerba Mate

Yerba mate (Ilex paraguariensis) is a native tree of South America that

has been consumed by indigenous peoples since pre-Colombian times and

adopted by the colonisers of South America (Bracesco et al., 2011). Nowadays, it

is consumed as a hot or cold infusion and is one of the most popular beverages

in South America, with an estimated 1 million people consuming around 1-2 L per

day of mate infusion (de Morais et al., 2009). Traditionally, yerba mate is

consumed using a gourd and a metal straw called a bombilla (Bracesco et al.,

2011). Although, it is gaining popularity in the USA, Europe, Germany and the

59

Middle East where the infusion is prepared using a regular tea bag (Heck and De

Mejia, 2007).

3.2.1. Natural occurrence

Yerba mate is native to the subtropical zone of three countries of South

America: Brazil, Argentina and Paraguay, as shown in Figure 3.1, and is naturally

grown within native forests. Even though the producers have successfully

cultivated yerba mate in plantations, there have been several unsuccessful

attempts at growing this plant in other parts of the World (Ilany et al., 2010).

Although the tree is usually harvest as a bush, it can reach up to 18 meters and

the leaves (evergreen and harvest for the yerba mate products) are up to 15 cm

long (Bracesco et al., 2011). The tree usually flowers from October to November

providing a small, greenish-white flower. Trees produce fruits from March to June

producing a dark red drupe from 4–6 millimeters (Heck and De Mejia, 2007).

Figure 3.1: Natural occurrence of yerba mate in South America. Adapted from

Maccari Junior (2005).

60

3.2.2. Production (plantation to processing plant) and products

The Worldwide production of yerba mate in 2017 was reported to be as

follows: Argentina – 689,196 tons (INYM, 2017); Brazil – 354,398 tons (IBGE,

2017); and Paraguay - 91,640 tons of green yerba mate (Aguinaga, 2017). Yerba

mate is produced in its native or natural state, where the tree grows between

forests, or in cultivated farms. One of the main differences between Argentinian

and Brazilian production is that in the former, all of the yerba is grown in the sun

(because it is cultivated), whilst in the latter, native trees are shaded within local

forests. These different types of cultivation may have an impact on the chemicals

present in yerba mate, especially polyphenols (Donnelly, 2015).

Although production has expanded in recent years, the Brazilian yerba

mate market is still very much restricted to the southern regions of the country.

The main production is undertaken by family units who cultivate the trees in local

forests (Balzon et al., 2004, Vasconcellos, 2012). According to data from PAM -

IBGE (Brazilian Institute of Geography and Statistics) the main region of yerba

mate plantation production is Rio Grande do Sul (IBGE, 2017). In contrast, forest

or native production occurs in Paraná, Santa Catarina and Mato Grosso do Sul

States on a smaller scale, as shown in Figure 3.2.

As the origin of the raw yerba mate material (cultivated or native)

influences the flavour of the final product, the production of yerba mate in Paraná

(mainly native) is more valuable than for other states of Brazil due to the low level

of bitterness - according to consumers (Maccari Junior, 2005). Yerba mate, an

important Brazilian export, is sent mainly to Uruguay, who do not have their own

yerba mate plantations. In 2017, revenue from the commercialisation of yerba

mate was estimated at 107 million US dollars (IBGE, 2017).

61

Figure 3.2: Main state producers of yerba mate in Brazil with their contribution

(%) to Brazilian production. Adapted from IBGE (2017).

There are several stages of yerba mate production, as shown in Figure

3.3. First, the leaves are harvested, which can take place at any time during the

year, however, to minimise damage to the plant, most producers harvest in Brazil

from May to September and in Argentina from April to September. During this

period about 65% of the total amount of yerba mate annual production is

harvested (Valduga et al., 2003). Since one of the most desirable characteristics

of yerba mate for Brazilian consumers is an intense green colour and as the

colour changes during storage, the Brazilian market requires constant production

throughout the year in order to maintain a constant supply of the fresh product

colour (Duarte, 2000).

After harvesting, when the leaves are picked by hand or mechanically, the

yerba mate is transported to the processing facility where it is classified and

stored until the sapeco stage, as shown in Figure 3.4. In this process, the leaves

are rapidly exposed to a flame that ruptures the leaf membranes and denatures

the enzymes thereby preventing further oxidation (Donnelly, 2015). Then, the

yerba is either triturated and dried in continuous rotary metal cylinders

(conventional method) or first dried at low temperatures using hot air and then

triturated (premium method). This process reduces the moisture content of the

leaves to 5 or 6% (Ward and Marcilla, 2003). The dried leaves, now called yerba

62

mate cancheada, are milled and packed for distribution to the market place in a

variety of loose or teabag packs. The teabag packaging found in the Brazilian

market contain the roasted yerba which is a further stage added to the yerba

mate cancheada stage, as shown in Figure 3.4. The differences in temperature,

time and the materials used during the production stages may all have an

influence on the quality and chemical compounds of the final product (Maccari

Junior, 2005).

Figure 3.3: Scheme of production of yerba mate. Adapted from Maccari Junior

(2005).

63

Figure 3.4: (A) Yerba mate tree; (B) Sapeco stage; (C) Yerba Mate cancheada

for the Brazilian and Argentine markets. Adapted from UFRS (2012).

There are various differences between Argentine and Brazilian production

including the different harvesting times. In the Argentinian product, the green

leaves are further aged for about 9 months in chambers, which leads to a change

in the colour of the final product. On the other hand, some of the Brazilian

commercial products are roasted. Moreover, the particle size of the final product

is also different, being around 5 mm for the Argentinian and 300 µm for the

Brazilian product, as show in Figure 3.5.

64

Figure 3.5: Argentinian (left) and Brazilian (right) green yerba mate commercial

samples.

3.2.3. Methods of consumption

Despite the great potential of using yerba mate in soft drinks, sweets,

cosmetics and medicines, mainly because of its antioxidant’s properties, the

dried leaves of the plant are mostly intended for a traditional type of infusion

consumed in South America. Infusions made from green yerba mate leaves are

widely consumed, primarily as chimarrão or mate (hot infusion) and tererê (cold

infusion) in Argentina, southern Brazil, Paraguay and Uruguay. Chimarrão is the

most popular form of yerba mate consumption in Brazil (as presented in Figure

3.6) followed by roasted leaf infusions sold in teabags, which contain only leaf

roasted material and are used to make hot and cold infusions (chá mate).

According to the Brazilian legislation, commercial packages of loose yerba mate

contain about 30% of twigs and 70% of leaves (Heinrichs and Malavolta, 2001).

65

Figure 3.6: Typical Brazilian Chimarrão (mate) consumption (Forma, 2016).

In chimarrão, the cup is made from a dried fruit from the calabash or bottle

gourd tree (Lagenaria siceraria). The “bomba” (Portuguese) or “bombilla”

(Spanish) is the metal straw, which has a filter at the lower end in order to

separate the infusion from the leaves.

The cup is usually filled with around 50 g of the loose green yerba mate

and 200 mL of water at 80 ºC; after consumption (seeping through the “bombilla”)

of the resulting infusion. The cup is then topped up with more hot water. This

cycle of infusion and consumption is repeated between 5 - 10 times, which

represents a unique and social form of consumption of the yerba mate. This is

very different from a regular infusion prepared with teabags, where approximately

3 g of green material (Argentina) and 1.8 g of roasted material (Brazil) are

brewed in a typical cup of 200 mL of water. This difference of consumption will be

evaluated in this study, namely, whether this influences the chemical intake

during the drinking of yerba mate, especially for trace elements and polyphenols

and what is the possible link to the daily dietary intake.

3.3. Health Effects of Yerba Mate Consumption

Yerba mate is regarded as offering various health benefits, it is well known

as a stimulant drink that eliminates fatigue and improves mental and physical

focus, mainly because of its high levels of caffeine (Bracesco et al., 2011). The

66

infusions also have significant levels of antioxidants, which are molecules that

can prevent the oxidation of biomolecules in biological systems (Bastos et al.,

2007).

Besides these important features, yerba mate is alleged to also help in

inflammatory and cardiovascular diseases (Schinella et al., 2005), promote

weight loss (Andersen and Fogh, 2001) and reduce sugar blood, cholesterol and

triglycerides levels (Filip and Ferraro, 2003).

However, negative effects on health have also been reported. Some

studies have indicated an alleged relation between the heavy consumption of

mate and cancer (Vassallo et al., 1985, Pintos et al., 1994, De Stefani et al.,

1996). The causal relationship between the consumption of yerba mate and

cancer has not been fully proven but it may be associated with the high

temperature of the water used for the infusion or other concurrent causes, such

as smoking and nutritional factors, instead of the yerba mate itself (Ramirez-

Mares et al., 2004).

3.3.1. Chemical composition of yerba mate

Some of the alleged health benefits of yerba mate could be explained on

the basis of its chemical composition; however, there is a lack of conclusive

studies. Among the organic compounds, xanthines (such as theobromine and

caffeine) have diuretic properties, can influence the relaxation of smooth muscle

and have been reported to cause myocardial stimulation (Filip et al., 2000,

Leborgne et al., 2002, Schinella et al., 2005).

One of the groups of antioxidants present in yerba mate is polyphenols

(refer to section 1.3), which are able to prevent oxidation of biomolecules (Colpo

et al., 2016). Antioxidant compounds present in yerba mate, such as

polyphenols, have been shown to inhibit lipid peroxidation, especially low-density

lipoprotein (LDL) oxidation (Gugliucci, 1996). In previous studies, the total

content of polyphenols in yerba mate was determined by the Folin-Ciocalteu

assay using gallic acid as a standard. Studies using this method have reported

67

levels of 51.3 - 72.9 mg gallic acid equivalents (GAE)/ 100 mL) in yerba mate

infusions which are similar to that found for green and black tea (Bravo et al.,

2007, Gorjanovic et al., 2012, Zielinski et al., 2014). On the other hand, the total

chlorogenic acid content, which is the main group of polyphenols present in

yerba mate, has been found to be higher than in tea (Camellia sinensis) and

similar to the content of filter coffee (Donnelly, 2015).

The principal polyphenol compounds present in yerba mate is the

chlorogenic acid group, including a range of mono-, di- and tri-acylated

compounds (Donnelly, 2015). The mono-chlorogenic acids found in yerba mate

include 1-, 3-, 4- and 5-caffeoylquinic acid as well 3-, 4- and 5-feruoylquinic acid

and p-coumaroylquinic acid (Jaiswal et al., 2010, Dugo et al., 2009, Bravo et al.,

2007). The predominant compounds present in yerba mate are 3-, 4- and 5-

caffeoylquinic acid and 3,5-, 4,5- and 3,4-caffeoylquinic acid with levels of 0.5 –

3% of the leaf material (Donnelly, 2015). Marques and Farah (2009) and Clifford

and Ramirez-Martinez (1990) also noted that the chlorogenic acid content of

roasted yerba mate (3%) decreases when compared to green yerba mate (9 –

10%).

Besides the polyphenol compounds, yerba mate also have xanthine

compounds, such as, caffeine and theobromine (Donnelly, 2015). A study using

the mass-to-ratio that simulates a traditional South American infusion method

found 1253 ± 72.5 μg/mL of caffeine and 53.1 ± 1.44 μg/mL of theobromine

(Murakami et al., 2013). Moreover, a regular infusion (European tea-based

method) presented 18.8 – 42.5 μg/mg of caffeine and 9.9 – 15.7 μg/mg of

theobromine (de Mejía et al., 2010).

Also found in yerba mate is high levels of aluminum, manganese, iron and

zinc in the leaves and infusions. A review of previous studies is presented in

Table 3.1. All of these studies were performed with different commercial samples

of yerba mate, consequently the elemental levels vary widely, which could be

influenced by different soil conditions, harvest periods, cultivation and processing

methods (Zeiner et al., 2015).

68

Table 3.1: Element content of yerba mate leaves for selected elements reported

in literature (weight basis not reported).

Author Year Concentration (mg/kg)

Mg Ca Mn Fe Cu Zn Barbosa

et al. 2015 2710 - 2730 5770 - 6800 129.0 - 232.6 47.8 – 54.1 7.2 – 8.2 24.9 – 56.6

Barbosa

et al. 2018 1600 - 1900 2900 - 3700 437 – 614 71 - 74 3.8 - 4.6 40 - 73

Bastos et

al. 2014 2700 - 4400 1800 - 2700 137.2 – 325.4 12.1 - 35.4 6.2 - 7.2 20.1 – 30.8

Donnelly

2015 - - 378.3 – 879.9 28.2 168.1 4.31 - 12.12 14.0 – 125.8

Giulian et

al. 2007 5025 ± 186 6785 ± 249 1315 ± 113 254 ± 27 14 ± 2 72 ± 5

Heinrichs

et al. 2001 4300 - 5200 6000 - 6600 665 – 1050 103 - 286 7.6 - 10.7 38 - 43

Jacques

et al. 2007 7447 -7977 3453 - 8520 1168 – 3542 54 - 120 6.20 -11.47 33 - 98

Malik et

al. 2008 3520 - 8115 9562 - 12720 309 – 1114 83.4 – 88.1 7.98 - 12.7 26.1 – 31.8

Magri et

al. 2019 - - 533 – 4865 58 - 173 8.0 - 20.0 13 - 181

Milani et

al. 2019 - - 1115 - 1811 103 - 437 9.5 - 12.2 21 - 26

Pozebon

et al. 2015 4587 - 5574 6947 - 7659 730 – 1368 154 - 226 31.9 - 38.3 44.2 – 79.4

Rossa et

al. 2015 80 - 9550 120 - 5450 236 - 1440 8.0 - 130.0 1.0 - 44.0 34.0 - 146.0

Marcelo et

al. 2014 4591 ± 842 6825 ± 842 1078 ± 377 205 ± 89.1 11.9 ± 2.06 63.6 ± 25.0


The overall aim of this study was to investigate the chemical analysis of

different samples of yerba mate. Moreover, to analyse the elemental levels of

non-commercial samples in order to evaluate the effect of different plantation

methods. Finally, to perform a complete analysis of the commercial yerba mate

samples from the two main producers in the World, Brazil and Argentina. This

enabled an evaluation of the impact of different production methods, pre-

69

treatment of the commercial products and to assess the effect of the mode of

consumption on the uptake of polyphenols and elements.

The objectives were to:

(i) provide a literature review of reported elemental and phenolic levels of

yerba mate;

(ii) investigate the elemental profile of non-commercial and commercial

yerba mate samples;

(iii) assess the impact of different varieties; plantation processing

methods; and ages of tree and leaves on the total elemental analysis;

(iv) determine the total polyphenol and chlorogenic acids of commercial

samples of yerba mate from Brazil and Argentina;

(v) evaluate the chemical analysis of the yerba mate (loose and teabags)

material and infusion methods (regular and bombilla/traditional and

iced); and

(vi) investigate the impact of the consumption of yerba mate in terms of

the polyphenol and elemental dietary intake.

3.5. Non-Commercial Studies on Yerba Mate

This section provides an evaluation of the elemental content of non-

commercial yerba mate leaves collected in April 2017 from the Barão de

Cotegipe plantation. The description of the samples and methods used are

outlined in sections 3.5.1 and 3.5.2. An investigation of the production types

(traditional and natural); use of fertilisers (NPK and organic); age of leaves (new

and old) and height of the leaves in a tree (bottom, middle and top) is presented

in sections 3.5.3 and 3.5.4. Finally, an evaluation of the commercial processing

of the material is proposed in section 3.5.5.

70

3.5.1. Description of the samples

Samples were collected during a field-trip in April 2017 to Barão de

Cotegipe plantation, located in the Rio Grande do Sul State, Brazil. The company

is one of the biggest producers and exporters of yerba mate in Brazil. They

agreed to collaborate on this project to evaluate their different plantations and

processing methods in regard to the elemental composition of the samples. In

terms of the cultivation of the yerba mate, they have 3 different sites: (i)

traditional plantations: cultivated yerba mate planted between trees (Hovenia

dulcis and Araucaria angustifolia) and treated with NPK fertilisers; (ii) traditional

plantations: cultivated yerba mate planted under the sun and treated as an

organic form of farming; and (iii) natural forest: yerba mate trees grown between

and beneath natural forests, without any fertiliser treatments. A list of all samples

is presented in Appendix 3.1.

The yerba mate harvesting is usually performed by manually cutting the

branches of the tree to collect and process the material disregarding the age of

the leaves. An evaluation of the elemental composition of different yerba mate

trees in relation to the age of the leaves (new and old) was undertaken in this

study. Therefore, the small leaves at the end of each branch were collected as

new material and the mature larger leaves collected along the branch were

classified as old material. All of the leaves were collected at a medium height of

the tree (1.5 m).

The yerba mate trees from the Barão de Cotegipe plantation were pruned

to keep the same height (2.5 meters as average). Although, during the harvesting

period the leaves are collected from every height of the tree. An evaluation of the

elemental composition in relation to height where the leaves were collected was

proposed during the field-trip. Consequently, the new leaves from a tree grown

on the organic plantation were collected at the following heights: 0.5 – bottom;

1.5 - middle and 2.5 m- top.

Finally, an evaluation of the processing of the yerba mate leaves was

investigated. Samples were collected at each stage of the commercial

71

processing plant (refer to section 3.2.2): (i) harvested leaves: when the leaves

arrive at the processing plant; (ii) Sapeco: the leaves were collected after the

sapeco stage, where the material is exposed to an open fire for a short period of

time (refer to Figure 3.4); and (iii) dried: sample collected after the drying process

where the leaves were dried until the moisture content was reduced to 5-6%.

Due to the constant processing of yerba mate in the plant; the samples collected

do not refer to the same type of yerba mate production (different origin or

fertiliser treatment).

3.5.2. Materials and method

The determination of the total elemental composition of the yerba samples

was performed as described in section 2.3. The samples were fully digested at

500°C for 12 hrs using a muffle furnace and analysed by inductively coupled

plasma mass spectrometry or ICP-MS (refer to section 2.3).

3.5.3. Production by traditional plantations

This section will report a pilot investigation into the elemental levels of

yerba mate leaves cultivated in the traditional plantations of Barão de Cotegipe,

southern Brazil, with a view to evaluating the impact of different methods of

growing plants (with and without fertiliser; covered with native trees), harvest of

the leaves (height within a tree and if the leaves are new or old). The main

reason behind this investigation was two-fold, most reported studies focus only

on the elemental values of commercial products (refer to section 3.3.1), and to

advise the producers (Barão de Cotegipe) of the potential methods of cultivation

or harvest that may enhance the elemental and/or nutritional quality of the yerba

mate products. An evaluation of the difference between standard deviations

between the sets of data reported in this study was performed using a F-test

(refer to section 2.6.2) and is presented in Appendices 3.25 to 3.27. In general,

there is no statistically significant difference between the standard deviations at

p<0.05, that is the null hypothesis is retined as Fcal<Fcrit (Miller et al., 2018).

72

(i) Effect of cultivated yerba mate leaf age on elemental levels

The total elemental composition of the non-commercial yerba mate

samples cultivated in traditional plantations was evaluated following the method

proposed in section 2.3. The results of important trace elements (i.e. those

relating to human health) are presented in Table 3.2 grouped by the use of

fertilisers or not (organic) and by the age of the leaves (new and old), as

described in section 3.5.1. Magnesium, Ca, Mn, Fe, Cu and Zn were chosen

because the concentration in yerba mate products could have a significant

impact on the nutritional intake of these elements. Moreover, two types of

plantation were also investigated, namely one that uses NPK fertilisers as the

standard method of supporting the growth of yerba mate trees, and the other that

uses no commercial fertilsers (or organic production).

A preliminary inspection of the data (Table 2.3) shows that for both

traditional methods of yerba mate production, there is a wide variation in the

elemental concentrations of both the new and old leaves (refer to section 3.5.1

for the description on classifying the age of the material). In terms of the use of

fertiliser, Mg, Ca, Mn, Fe and Zn all have higher mean levels in the old leaves,

with the exception being for Cu (new > old). Moreover, the elemental ranges do

not overlap for Mg, Ca and Cu, whilst the other elements show some degree of

overlap in elemental levels for the two ages of leaves. As a contrast, organic

production has lower mean levels of Mn in the old leaves. Interestingly, all

elements now have an overlap in the elemental levels of the two leaf ages.

73

Table 3.2: Total elemental levels are reported as the mean and range (min –



plantations cultivated either using NPK fertilisers or non-chemical

(organic). The digested samples were analysed by ICP-MS (refer to

section 2.3).

Fertiliser Organic New leaves Old leaves New leaves Old leaves

n 4 4 9 8

Mg 3172

(2608 – 3724)

7015

(5710 – 8474)

4000

(2779 – 5914)

5180

(2356 – 8218)

Ca 2989

(2257 – 4160)

8631

(6511 – 10298)

4783

(2588 – 10534)

6671

(2536 – 10904)

Mn 623

(425 – 924)

949

(560 – 1394)

1195

(757 – 3395)

712

(349 – 1189)

Fe 47.09

(36.87 – 55.65)

66.13

(49.17 – 76.22)

34.28

(27.83 – 41.39)

41.16

(28.22 – 50.73)

Cu 28.67

(25.08 – 37.19)

14.67

(13.45 – 16.35)

15.10

(8.87 – 26.90)

9.90

(5.53 – 21.29)

Zn 79.50

(67.71 – 104.95)

166.14

(82.49 – 238.50)

90.15

(30.17 – 252.96)

107.47

(15.74 – 251.73)

n is the number of samples.

Statistical analysis of the data in Table 3.2 was undertaken using a two-

tailed Student t-test (Miller et al., 2018), based on the null hypothesis, that there

is no statistically significant difference in the elemental levels of the leaves (new

and old) collected from the same tree and grown in the two different yerba mate

plantations. Table 3.3 reports the statistical data.

74

Table 3.3: Statistical analysis using a two-tailed t-test (Miller et al., 2018) to


leaves (based on age – new and old) for non-commercial samples

collected from traditional plantations cultivated either using NPK

fertilisers or non-chemical (organic).

Fertiliser Organic n 8 12

tcrit 2.36 2.18

tcalc p Direction of

significance tcalc p

Direction of

significance

Mg 5.86 0.0011** new<old 1.76 0.1035 ns

Ca 6.25 0.0008*** new<old 2.49 0.0285* new<old

Mn 1.49 0.1863 ns 2.59 0.0236* new<old

Fe 2.56 0.0458* new<old 1.14 0.1858 ns

Cu 4.80 0.0030** new>old 1.91 0.0801 ns

Zn 2.18 0.0722 ns 0.36 0.7237 ns

n is the number of samples; tcalc is the calculated value (refer to Equation 2.6, section 2.6.2): tcrit is the critical value obtained for n-2 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant at probability p<0.05; ** highly significant at probability p<0.01; and *** very highly significant p<0.001.

Yerba mate leaves grown in traditional plantations, cultivated with the use

of fertilisers, show a significant difference (p<0.05) in the levels of Mg, Ca and Fe

(new < old leaves) and Cu (new > old), confirming the initial inspection of the

data presented above. Moreover, the observed overlap in the elemental ranges

for Mn and Zn are associated with no significant difference (p>0.05). If the leaf

samples were collected from a traditional plantation, cultivated without any

treatment (organic), there was also a significant difference between the age of

the leaves from the same tree, namely, Ca and Mn only (new < old).

Although there is a limited number of leaf samples involved in this

investigation on the effect of leaf age (collected from the same tree), it shows that

for most of the elements (including those reported in Appendix 3.2 to 3.4 for V, Cr

and Se) the general trend is for the element to be at higher levels in the old

leaves.

75

The uptake, distribution and compartmentalisation and/or separation of

particular elements, especially in their soluble ionic form, are required by the

plant for optimal function (Leigh, 1997). Plant organs may have very different

concentrations of elements within tissues which may relate to the xylem vs.

phloem transport mechanisms, sites of complexation (for many heavy metals or

charge-dense ions), or tissue or cell-specific transport (Conn and Gilliham, 2010,

Marschner, 2011, Tester and Leigh, 2001). Therefore, in the yerba mate plant

(Illex paraguariensis), the results of this study confirm that for most of the

elements there is uptake and distribution to the leaves, with there being a

variation in the elemental concentrations for both new and old leaves (refer to

Table 3.2). This could be due to a matrix effect, but because the analysed

samples are from the same plant (different tissues), any possible matrix effects

would be minimal. In terms of explaining why many of these elements are found

at high levels in the older levels (with the exception being copper) various

authors have reported the accumulation of elements in older leaves of different

plants. Immobile or less mobile ions tend to accumulate in the older leaves

simply because the largest total amount of transpiration occurs through these

leaves (Tinker, 1981). There may be a tendency for elements to be deposited at

the leaf margins where transpiration is maximal (Tinker, 1981). De Maria and

Rivelli (2013) also commented on the largest accumulation of Cd, Zn and Cu in

plants being found in the leaves, mainly in the old ones, especially for mature

trees. Kabata-Pendias (2010) commented that zinc is likely to be concentrated in

mature leaves and if there is higher Zn concentrations in the soil, translocation

from the roots to the plant tops is enhanced. The reason why Cu is found at

higher levels in the new leaves (and also Mn for organic cultivation) is not clear,

although there are conflicting reports on the mobility of Cu in plants. Tinker

(1981) stated that Cu is fairly immobile in plants, tending to remain in the older

leaves. However, others have said that many heavy metals, including Cd, have

differential accumulation within the root system as opposed to the shoots (Puig

and Peñarrubia, 2009, Verbruggen et al., 2009). Therefore, the findings for Cu

76

and Mn at this time are not easy to justify. The same trend was found for the

reported levels of Na, K, Co, Ni, Cd and Pb (refer to Appendices 3.2 to 3.4).

(ii) Effect of yerba mate cultivation (with or without the use of

fertilsers) on the elemental levels of leaves (new or old)

The next question to be addressed is to know whether the use of

traditional cultivation methods, with or without (organic) the use of chemical

fertilisers, has a direct impact on the elemental levels of the leaves (for both new

and old). A statistical analysis was undertaken of the data in Table 3.2, using a

two-tailed t-test (Miller et al., 2018) with the null hypothesis being that there is no

significant difference in the elemental levels of new or old leaves grown in

plantations with/without the use of NPK fertiliser/organic, respectively (Table 3.4).

An initial evaluation of the data in Table 3.2 shows some interesting trends

in relation to the cultivation of yerba mate trees, with and without the use of

added fertiliser. At the time of sample collection, no information was available

from the producers about the type of NPK fertiliser or organic additives used on

the plantations. The statistical analysis of this data using a two-tailed Student t-

test is reported in Table 3.4. The new leaves grown on trees from the organic (no

chemical addition) plantation have higher levels of Mg, Ca, Mn (significant

p<0.05), and Zn (with same trend being observed for Co, Se and Cd, as reported

in Appendices 3.2 to 3.4). Only Fe (p<0.01) and Cu (p<0.05) have higher levels

(which are both statistically significant) in the new leaves from the trees grown

with the addition of fertilsers (with the same trend for Na, V, Ni, As and Pb, as

reported in Appendices 3.2 to 3.4). Conversely, for the older leaves from both

plantations, all of the elements reported in Table 3.2 are higher in the fertilser-

addition plantation, with Fe being statistically significant (fert > org; p<0.01). The

same trend was found for V, Cd and Pb; with the exception being for Na, K, Co,

Ni and Se, as reported in Appendices 3.2 to 3.4.

77



leaves based on the use or non-use (organic) of NPK fertilsers during


New leaves Old leaves n 11 11

tcrit 2.26 2.26



Direction of

significance

Mg 0.70 0.5030 ns 1.39 0.1992 ns

Ca 1.08 0.3076 ns 1.03 0.3313 ns

Mn 2.50 0.0374* fert<org 1.85 0.0975 ns

Fe 3.31 0.0092** fert>org 4.54 0.0014** fert>org

Cu 3.20 0.0109* fert>org 1.65 0.1330 ns

Zn 0.50 0.6291 ns 1.01 0.3405 ns

fert – fertilsers addition; org – organic or no chemical addition; n is the number of samples; tcalc is the calculated value (refer to Equation 2.6, section 2.6.2): tcrit is the critical value obtained for n-2 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant at probability p<0.05; ** highly significant at probability p<0.01.

Most NPK fertilisers added to soils increase the levels of N, P and K so as

to stimulate plant growth and hopefully the uptake of these major elements by the

roots and upper plant tissues (Adekiya and Agbede, 2009). Many NPK fertilsers

can contain trace levels of other minerals and minor/trace elements (Ca, Mg, Na,

Fe, Mn, B, Cu and Zn) (Mursito et al., Mursito et al., 2017). The addition of these

other elements is to stimulate plant growth and enhance the bioavailability of

elements to the aerial tissues of the plant. This will also depend on soil chemistry

(pH, organic matter and water content, and the presence of other elements) and

any microbial activity in the soil or soil-root interface (including mycchoriza fungi

and nitrogen-fixation bacteria in the root nodules) (Kabata-Pendias, 2010).

Moreover, the addition of some elements in fertilisers (i.e. Fe) may have an effect

on the nutrition of other elements in plants, including Mo, Cu, Mn and P (Han et

al., 2016). Some studies have reported that the addition of NPK fertilisers

78

decreases soil pH and thereby also influences the exchangeable levels of Ca

and Mg in the plant. Kabata-Pendias (2010) also stated that this may solubilise

Fe leading to large amounts in the plant. This can result in elemental deficiencies

in the younger leaves (Liu et al., 2010). This pH change may be explained by the

leaching of basic cations, such as Ca and Mg (and maybe other trace elements),

from the soil. Moreover, the long-term use of NPK fertilisers has been shown to

result in the deficiency of many essential nutrients in the soil (Bailey et al., 2004)

(Liu et al., 2010) reported that the addition of NPK fertiliser to yellow poplar

(Liriodendron tulipifera Lin.) increased the levels of available nutrients in the soil,

thereby increasing the biomass of the plant, but observed no changes in the

plant nutrient concentrations of the upper plant tissues. In this research, it

appears that for yerba mate plants the addition of NPK fertiliser is having no

effect on the uptake of any essential trace elements by the newer growing

leaves, with the exception of Fe and Cu. This may be because the NPK is

stimulating root growth and the mobilisation of other trace elements to the stems

or shoots (Loneragan, 1981). Moreover, if the NPK fertiliser has high levels of

soluble Fe (which will be influenced by a lowering of the soil pH), this not only

increases the amount available to the plant, but it seems to lead to a significant

increase in the Fe levels of the upper tissues, including the newer leaves.

Interestingly, copper tends to be immobile in plants (Kabata-Pendias, 2010), but

if you increase the supply of copper, this leads to an increase in mobility within

the plant, primarily to the shoots, although for yerba mate, it seems there is also

an uptake by the newer leaves. Finally, as stated above, many authors have

reported that immobile or less mobile ions tend to accumulate in the older leaves

simply because the largest total amount of transpiration occurs through these

leaves (Kabata-Pendias, 2010, Tinker, 1981). Clearly, the addition of NPK and

other minerals or elements in the fertiliser has resulted in accumulation within the

older yerba mate leaves. Since there is no information about the organic farming

method being used (i.e. the addition of animal or plant manure), it is not possible

to state whether there are any nutrients or elements being added to the trees in

the organic plantation.

79

(iii) Effect of collecting yerba mate leaves from a tree at different

heights on the elemental levels of new leaves

The harvesting of yerba mate leaves from trees grown in traditional

plantations (organic) involves collecting any leaf material from the tree,

irrespective of height or age. Therefore, an important question is whether the

elemental levels of the collected leaves (only new) are different if collected at

different heights within the tree (namely, at 0.5 - bottom, 1.5 - middle and 2.5 m-

top). As a result of time constraints only one tree was investigated, therefore only

replicate samples were collected at each height and the results in Table 3.5 will

not be subjected to statistical analysis.

Interestingly, the levels of Mg, Ca and Zn are highest in the new leaves

collected from the bottom of the tree (0.5 m height). Iron and Mn show a degree

of translocation of the element to the upper parts (2.5 m), whilst Cu is slightly

elevated in the leaves from the middle of the tree (1.5 m). The other elements

reported in Appendices 3.2 to 3.4 demonstrated the following trend in terms of

tissue accumulation: (i) bottom for Na, Co, and Cd; (ii) middle for Ni and Se and

(iii) top for K, V, Cr and Pb.

80

Table 3.5: Total elemental levels (mean ± standard deviation; mg/kg, dry weight)

of yerba mate leaves collected at different heights from the same tree

grown in a traditional plantation (cultivated as organic). The digested

samples analyses were analysed using ICP-MS (refer to section 2.3).

The number of samples, n = 2 replicates.

Bottom (0.5 m)

Middle (1.5 m)

Top (2.5 m)

Mg 5829 ± 1 3940 ± 30 4054 ± 24

Ca 11383 ± 48 5469 ± 181 5857 ± 12

Mn 1070 ± 4 1009 ± 82 1229 ± 21

Fe 39.10 ± 5.45 32.64 ± 5.05 51.33 ± 1.92

Cu 5.41 ± 0.40 7.68 ± 0.28 5.21 ± 0.17

Zn 81.79 ± 0.59 64.08 ± 1.09 65.29 ± 1.28

An interpretation of the above findings confirms, as reported by many

authors, that the mobilisation and accumulation of specific trace elements in plant

tissues varies from plant species to species, soil type and chemistry, and the

concentration of the elements and other chemicals (including organic matter) in

soils and plant organs. Unfortunately, no soil samples were collected as part of

this study. This makes it difficult to know what effect the soil may have in the

organic plantation on the uptake by roots and subsequent mobilisation of trace

elements throughout the tissues of the yerba mate trees. Moreover, only new

growth leaves were studied, and the accumulation of specific trace elements in

such leaves is dependent on the morphology and surface absorption area of the

leaves (Bu-Olayan and Thomas, 2009, Baker and Brooks, 1989). Olowoyo et al.

(2012) evaluated the uptake of trace elements in two different plant species in

South Africa and found a wide variation in the elemental levels of the roots,

stems and leaves of the plants. Furthermore, elements may be adsorbed or

occluded by the presence of carbonates, organic matter or other minerals (Oliva

and Espinosa, 2007). Therefore, the raised levels of Fe, Mn, K, V, Cr and Pb (top

2.5 m) and Cu, Ni and Se (middle 1.5 m) found in the yerba mate trees may be

81

related to many factors which at this time are not known in terms of the soils (pH,

organic content, chemistry of elements and other species), or factors influencing

the mobility for these particular elements in trees grown under organic conditions.

3.5.4. Production by natural forest plantations

The elemental composition of the non-commercial yerba mate samples

cultivated between natural forests was evaluated following the method proposed

in section 3.5.2. Table 3.6 presents the results of selected trace elements

(relating to human health) grouped by the age of the leaves (new and old), as

described in section 3.5.1.

The data in Table 3.6 confirms the findings reported for the elemental

levels of new and old leaves taken from yerba mate trees grown in NPK fertilser-

added and organic plantations (Table 3.2). In general, older leaves have higher

concentrations of Mg, Ca, Mn, Fe and Zn, with the exception being for Cu (new >

old). Additionally, Na, Ni and Pb presented the same trend, where the elemental

concentrations were higher in the old leaves, as reported in Appendices 3.2 to

3.4. The remaining elements (V, Cr, Co, As, Se, Mo and Cd) had similar levels

for both new and old leaves. Also, the older leaves tend to have higher elemental

levels for the trees grown in the traditional cultivated (organic) plantation (with Zn,

Na, K, V, Cd and Pb showing the opposite trend, as shown in Appendices 3.2 to

3.4). Moreover, for the new leaves, there is a similar pattern with Mg, Ca, Fe, Cu,

K, Co and Ni (traditional > native forest) and Mn, Zn, Na, V and Se (traditional <

native forest), as reported in Table 3.6 and Appendices 3.2.

82

Table 3.6: Total elemental levels are reported as the mean and range (min –



cultivated organic plantations and grown between and beneath trees

of a native forest. The digested samples were analysed by ICP-MS


Traditional Native forests New leaves Old leaves New leaves Old leaves

n 9 8 3 3

Mg 4546

(2873 – 7846)

4928

(3782 – 6046)

4000

(2779 – 5914)

5180

(2356 – 8218)

Ca 5921

(3277 – 11139)

7005

(3867 – 8635)

4783

(2588 – 10534)

6671

(2536 – 10904)

Mn 1076

(982 – 1159)

1064

(805 – 1365)

1195

(757 – 3395)

712

(349 – 1189)

Fe 92.27

(91.15 – 94.07)

104.24

(69.63 – 153.99)

34.28

(27.83 – 41.39)

41.16

(28.22 – 50.73)

Cu 18.10

(9.35 – 22.87)

14.56

(11.62 – 20.18)

15.10

(8.87 – 26.90)

9.90

(5.53 – 21.29)

Zn 58.49

(32.83 – 75.71)

53.61

(39.57 – 71.86)

90.15

(30.17 – 252.96)

107.47

(15.74 – 251.73)

n is the number of samples. Table 3.7 lists the two-tailed Student t-test (Miller et al., 2018) data for the

above values in Table 3.6 and confirms the higher elemental levels for plants

grown in the traditional organic plantations, especially for Ca and Fe levels of

new leaves, with p<0.05 and p<0.001, respectively. Furthermore, the Mn and Fe

levels of old leaves were also statistically higher for plants grown in the traditional

plantations (p<0.01). Appendices 3.2 to 3.4 provides the data for the other trace

element levels of new and old leaves for yerba mate trees grown in traditionally

cultivated organic and native forest plantations.

This pilot study was undertaken at the producers request to evaluate

whether growing yerba mate trees within a native forest (between and beneath

larger trees) would influence the elemental quality of the yerba products. The

83

hypothesis was that such trees would have different soil chemistry (influenced by

the roots and leaf litter from the native trees), tree competition for nutrients from

the soil and the impact of reduced sunlight (and thereby photosynthetic activity).

Copper, magnesium, potassium, iron, manganese, molybdenum and zinc are

necessary for plant health and are associated with enzyme systems and

chlorophyll production relating to photosynthesis (Tränkner et al., 2018, Raven et

al., 1999). The role of many of these elements, especially Mg and K, also are

part of light-dependent photosynthetic processes (Tränkner et al., 2018). Clearly,

the yerba mate trees grown under a canopy of native forest is influenced not only

by reduced sunlight, but by nutrient and water competition with the larger trees.

The elemental leaves, in general, are found to be higher in both the new and

older leaves of the traditionally cultivated organic trees.



leaves (new and old) grown in traditional organic or native forest


New leaves Old leaves n 10 10

tcrit 2.31 2.31



Direction of

significance

Mg 1.50 0.1712 ns 0.70 0.5023 ns

Ca 2.54 0.0346* trad>nat 0.77 0.4614 ns

Mn 2.13 0.0661 ns 4.43 0.0022** trad>nat

Fe 23.18 <0.0001*** trad>nat 4.16 0.0032** trad>nat

Cu 0.37 0.7205 ns 2.14 0.0648 ns

Zn 1.00 0.3453 ns 1.11 0.2978 ns

trad – traditional organic plantation; nat – native forest plantation; n is the number of samples; tcalc is the calculated value (refer to Equation 2.6, section 2.6.2): tcrit is the critical value obtained for n-2 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant, p<0.05; ** highly significant, p<0.01, and *** very highly significant p<0.001.

84

3.5.5. Commercial processing plant

An evaluation of the elemental content of yerba mate leaves during the

commercial processing of the material was proposed, as described in section

3.5.1. The results are presented in Table 3.8. The samples were collected from

different stages of the processing plant on the day of the field-trip to the

company. The harvested samples are a combination of new and old leaves from

cultivated trees within a native forest (organic) from another producer (Santa

Catarina State). Therefore, this study is a simulation of what actually happens

during the harvesting of the yerba mate tree where all the leaves are collected

irrespective of the age of the samples and the mixture used to produce the

commercial product. As a result of constraints in the processing plant which has

different materials being processed at the same time, only one sample was taken

at each phase of the process for chemical investigation. Therefore, the results

will not be subjected to statistical analysis due to the limited number of samples.

A primary evaluation of the data presented in Table 3.8 shows a trend in

relation to the pre-process (or post-harvest) samples and after the sapeco

(subjected to an open fire flame - refer to section 3.2.2). In general, Mg, Ca, Mn,

Fe and Zn all had higher elemental levels after the sapeco stage of the process.

The only exception is copper that had very similar levels as the pre-processing

and post-drying stages. Furthermore, the data confirms the findings reported for

the elemental levels of traditional and native plantations, where leaves from the

traditional cultivated plantations had higher elemental levels than for those

collected from the yerba mate trees grown between native forests (Table 3.6).

Moreover, the post-drying samples (the last stage of the processing plant

involves the yerba mate material being dried until a moisture content of 5 to 6 %)

had lower levels of Mg, Ca, Fe and Zn when compared with the sapeco yerba

mate sample.

During the sapeco stage, the yerba mate leaves are thrown into an open

fire for a short period of time (seconds). This leads to the rupture of the leaf

membranes and the denaturation of the enzymes, preventing oxidation

85

(Donnelly, 2015). Several studies have also reported that post the sapeco stage

there was an increase in the levels of bioactive compounds (Isolabella et al.,

2010, Esmelindro et al., 2002, Schmalko and Alzamora, 2001). One study stated

that the extraction of caffeine and caffeoylquinic acids was more effective in dried

processed leaves due to the cell disruption and mechanical impact during the

processing stages (Bastos et al., 2006). Giulian et al. (2009) also demonstrated a

similar trend for the elemental analysis of mate samples (increasing levels as a

result of the sapeco stage, and then decreasing post the drying stage). The

mechanism of the sapeco is related to a quick loss of mass at high temperatures,

where it is expected that most of the water and ‘light’ (lower atomic weight)

elements are released from the leaves, which would lead to a reduction of the

leaf mass and consequently an increase in the concentration of the remaining

elements (Giulian et al., 2009). During the drying process (which involves the

material being transported along a moving belt) the time involved is ~ 10 times

longer than the sapeco stage, whilst the temperature is five times less. This

changes the dynamics of the elemental balance in the yerba mate material,

which could lead to different outcomes (Giulian et al., 2009). The other elements

reported in Appendices 3.2 to 3.4 (Na, K, V, Cr, Co, Ni and Cd) demonstrated a

similar trend where the sapeco stage results in higher elemental levels than for

the pre-processed samples. There are a couple of exceptions with the As, Se,

Mo and Pb values being similar for both stages of the yerba mate process. In

relation to the sapeco and drying stages the following trends were observed for

the elements reported in Appendices 3.2 to 3.4, namely: (i) higher (Na, Cr and

Cd) and lower (K, Se and Pb) levels in the material post the sapeco stage; and

(ii) similar levels for Co, Ni, As and Mo post the sapeco and drying stages.

86

Table 3.8: Total elemental levels (mean ± standard deviation) of non-commercial

yerba mate samples (mg/kg, dry weight) during the commercial

processing of the harvested tree material. The analyses were

determined using ICP-MS (refer to section 2.3).

Pre-processed Santa Catarina State*

Sapeco**

Drying process Native Traditional

n 4 2 2 2

Mg 6754 ± 577 7206 ± 21 4403 ± 40 5962 ± 54

Ca 5463 ± 616 7997 ± 424 6058 ± 244 6771 ± 63

Mn 368 ± 20 632 ± 22 445 ± 1 729 ± 53

Fe 54.2 ± 3.79 132.95 ± 10.85 68.52 ± 4.72 121.61 ± 6.02

Cu 11.34 ± 0.78 10.20 ± 0.22 11.08 ± 0.33 11.55 ± 0.52

Zn 29.79 ± 4.02 69.30 ± 2.72 34.78 ± 3.00 75.75 5.60

* samples from Santa Catarina State plantations, **refer to section 3.5.1.

3.6. Studies on Commercial Yerba Mate

This section provides an evaluation of the chemical content of commercial

yerba mate samples (loose and tea bags) from the two main producers in the

World, namely, Brazil and Argentina. The description of the samples and

methods used are outlined in sections 3.6.1 and 3.6.2. An investigation of the

total elemental content and the assessment of the effect of the mode of

consumption on the uptake of these elements is proposed in sections 3.6.3 and

3.6.4. Furthermore, the phenolic content of the commercial infusions is outlined

in section 3.6.5. Finally, an evaluation of the link to the dietary intake of the

elemental and phenolic content of yerba mate infusions is presented in section

3.6.6.

3.6.1. Description of the samples

All of the samples, with the exception for Barão de Cotegipe (southern

Brazil), where commercial yerba mate products from Brazil, Argentina and

87

Uruguay (which are from Brazilian production) were bought in local

supermarkets. Barão de Cotegipe, agreed to collaborate with this study and sent

a wide range of their products that are available in the South American market:

(i) Premium, selected healthy leaves of the natural plantation, dried

through a bed dryer using hot air at low temperatures (eliminating the

exposure of the material to smoke);

(ii) Nativa, same drying process as Premium, but using non-selected

leaves from the natural plantation;

(iii) Traditional, yerba mate from the traditional and natural plantations

dried together in a rotating metal cylinder where the leaves are in direct

contact with the smoke produced;

(iv) Moida Grossa, same processing method as outlined for Traditional but

ground at a larger particle sizes in the final product;

(v) Exportação, same processing of Moida Grossa but the yerba is aged

for at least 8 months in chambers, and also has a lower content of

twigs;

(vi) Terere, same traditional drying process but with the specific particle

size suitable for Terere (cold infusion);

(vii) Tostada, same processing of Moida Grossa but the final product is

then roasted, suitable for regular infusions and Brazilian iced tea

infusions;

(viii) Cambona, a genetic selected plant dried by Traditional method.

The samples were divided and compared by country of origin (Brazil or

Argentina) and type (loose material or teabags; green leaves or roasted). A list of

all samples is presented in Appendix 3.5.

3.6.2. Materials and method

The determination of the total elemental composition of the yerba mate

samples was performed as described in section 2.3. The commercial samples

88

were fully digested at 500°C for 12 hrs and analysed by inductively coupled

plasma mass spectrometry or ICP-MS.

An evaluation of the chemical content of different types of yerba mate

infusions was proposed and the infusions were analysed for the elemental (refer

to section 2.3); total polyphenol (section 2.4.2) and chlorogenic acid content

(section 2.5).

(i) Regular infusions

Regular infusions of yerba mate (commercial loose and teabag) were

prepared following a standardised industry protocol that reproduces the average

volume of water, weight of material and brew time given as the pack label

instructions (Donnelly, 2015). Double distilled deionised water (DDW) was

preferred instead of tap water for all infusions in order to evaluate the chemical

contribution from only the yerba mate material. A volume of 200 mL of freshly

boiled DDW was added to 1 teabag or 2.000 ± 0.0010 g of loose material, stirred

for 5 seconds and brewed for 4 minutes. In the last 5 seconds, the teabags were

squeezed against the side of the beaker and removed or the loose leaf infusions

were filtered through a teabag tissue (provided by Tata Global Beverages). Each

final solution was filtered using a 0.45 μm filter and the resultant solutions were

analysed.

(ii) Brazilian iced tea infusions

One of the most traditional ways of consuming yerba mate in Brazil is by

making a strong infusion to be cooled and drunk as an iced infusion. To prepare

this solution, one teabag or 1.30 ± 0.01 g of loose material was infused with 200

mL of freshly boiled DDW for 7 minutes. The loose leaf solutions were filtered

through a teabag tissue and all the infusions were filtered using a 0.45 μm filter

and analysed.

(iii) Simulation of traditional mate infusion – bombilla method

It is important to reproduce the traditional way of consuming yerba mate in

South America because this reflects the real chemical intake by the population.

89

A novel method was also proposed in this study in order to simulate the

bombilla method of consumption. It consisted of weighing 5.00 ± 0.01g of loose

yerba mate onto a filter paper (No. 1, Whatmann™, UK) and placing them onto a

Büchner funnel, as shown in Figure 3.7. Then, 10 ± 2 mL of DDW at 80 ± 2ºC

was added, followed after 30 seconds, by the addition of another 10 ± 2 mL of

hot DDW. The funnel was allowed to stand for 30 seconds before turning on the

vacuum, which was maintained until the yerba mate was dry. The filtrate was

then further filtered through a 0.45 µm membrane filter into a polyethylene tube

(first fraction) prior to analysis. The procedure was repeated using the same

yerba mate material to produce fractions 2, 3, 4 and 5.

Figure 3.7: Schematic of the proposed bombilla method. Adapted from NGC,

(2015).

3.6.3. Total elemental composition of commercial yerba mate

The total elemental composition of the commercial yerba mate samples

from Brazil and Argentina was evaluated following the method proposed in

section 3.6.2. The results for important trace elements (significant impact on the

contribution to the nutritional intake) are presented in Table 3.9; grouped by the

90

type of sample (green or roasted), processing package (loose or tea bags) and

origin (Brazil or Argentina), as described in section 3.6.1.

In order to test if any difference exists between the variances of the sets of

data reported in this study, a F-test (refer to section 2.6.2) was performed and is

presented in Appendices 3.28 to 3.29. In general, there is no statistically

significant difference between the standard deviations at p<0.05, that is the null

hypothesis is retined as Fcal<Fcrit. This states that the variability between the

data is related to random errors (Miller et al., 2018).

During the processing of yerba mate for the Argentina commercial

products, the dried samples are aged for up to 24 months in chambers which

helps to develop the flavour preferred by Argentinian consumers. Conversely, the

Brazilian population prefers a fresh and greener product, packed after the drying

process (Isolabella et al., 2010, Heck et al., 2008).

Table 3.9: Total elemental levels reported as mean and range (min – max) of

different types of commercial yerba mate samples (mg/kg, dry weight)

obtained from Brazil and Argentina. The analyses were determined

using ICP-MS (refer to section 2.3); n is the number of samples.

Green loose Green tea bag Roasted loose Roasted tea bag Origin Brazil Argentina Argentina Brazil Brazil

n 16 7 19 3 4

Mg 4577

(2787 – 5934)

4958

(3466 – 5495)

6197

(5577 – 4614)

6140

(5188 – 6934)

5563

(4557 – 7488)

Ca 7375

(6027 – 8639)

7222

(5972 – 8091)

7674

(6302 – 9073)

9782

(9255 – 10122)

8486

(8180 – 9012)

Mn 646

(486 – 834)

545

(383 – 671)

731

(483 – 1098)

746

(658 – 889)

536

(397 – 879)

Fe 37.31

(10.62 – 83.49)

31.20

(10.58 – 61.18)

80.23

(18.21 – 228.13)

31.65

(20.70 – 53.03)

80.07

(47.77 – 109.63)

Cu 9.23

(7.28 – 11.76)

8.39

(7.15 – 10.81)

9.56

(7.97 – 15.00)

9.33

(9.07 – 9.61)

11.68

(11.06 – 12.32)

Zn 58.89

(29.42 – 98.96)

67.09

(36.83 – 104.92)

117.05

(54.24 – 601.58)

110.76

(65.14 – 164.38)

84.78

(78.04 – 98.31)

91

A preliminary inspection of the data listed in Table 3.9 shows that in

general the elemental (Ca, Mn, Fe and Cu) levels for green loose yerba mate

products obtained from Brazil are slightly higher than that for the Argentinian

products, with the exception of Mg and Zn. Donnelly (2015) also reported that

Brazilian green loose samples had higher elemental levels when compared to

Argentinian products. A two-tailed Student t-test (Miller et al., 2018) confirmed a

significantly higher level of copper (tcal = 3.35 for 20 degrees of freedom;

p<0.0032) in the Brazilian samples when compared to the Argentinian products

(Table 3.10). Conversely, the zinc levels were significantly higher in the

Argentinian products (Table 3.10). The remaining elements, reported in

Appendices 3.6 to 3.8, have higher Na, V, Cr, Co, Ni and Se levels for the

Argentinian green loose products and higher K, As, Cd and Pb levels for the

Brazilian samples. In summary, the elemental levels are basically similar for the

two countries, which is to be expected as yerba mate primarily grows between

the Paraná and Paraguay river basins in South America (refer to Figure 3.1).

This yerba mate production area includes regions of Argentina, Paraguay and

Brazil (Cardozo Jr et al., 2007, Heck and De Mejia, 2007). Bastos et al. (2006)

reported that during the ageing of commercial yerba mate, the concentration of

some elements slightly decreases, indicating potential losses during this process.

Furthermore, the Argentinian samples are cultivated in direct sunshine whilst

Brazilian plantations include trees that are shaded as part of a native forest

environment. Previous studies, such as, Reissmann et al. (1999), Caron et al.

(2014) and Barbosa et al. (2015), have also reported a similar trend as the data

in this study, that is, lower elemental concentrations in the leaves of yerba mate

trees cultivated under the sun (Argentina). This condition may also affect the

physiology, transpiration and elemental uptake by yerba mate trees as modern

practices do not resemble the traditional (former) cultivation methods which used

to be similar to that of Brazil (i.e. native forests) (Barbosa et al., 2015).

The next study involved an evaluation of Argentinian commercial yerba

mate products based on the preparation and consumption of infusions using

green loose material or tea bags. In general, all the elements have higher levels

92

for the tea bag products when compared to the green loose material. Statistical

analysis (two-tailed Student t-test) confirmed that Mg, Fe and Cu had significantly

higher levels (tea bags > green loose; p<0.05) as reported in Table 3.10.

Moreover, the same trend was found for the elements (Na, K, V, Cr, Co, Ni, As,

Se, Mo, Cd and Pb) reported in Appendices 3.6 to 3.8. This is not surprising as

the tea bags contain only leaf material and have been processed to produce finer

particles (of around 1 mm). Interestingly, the Argentinian loose material has

some twigs in the commercial yerba mate samples which has a slight impact in

reducing the elemental level of the product (Donnelly, 2015). It was noted that

the green tea bags contain mostly leaf material with very few stems compared to

the loose yerba mate samples (refer to Figure 3.5). This suggests that the

investigated elements are present at higher levels in the leaf material compared

with the stems or may be wholly contained within the leaf (Raguž et al., 2013,

Kabata-Pendias, 2010).

The Brazilian market also sells roasted yerba mate products (both as

loose and tea bags), mostly used to prepare iced ‘tea’ infusions (Heck and De

Mejia, 2007). In order to prepare roasted yerba mate, the leaf material undergoes

a further roasting process, similar to coffee, typically at 160 °C for 12 mins (de

Godoy et al., 2013, Bastos et al., 2006). The total elemental levels between

commercial green and roasted yerba mate samples from Brazil was evaluated in

this study. The roasted (loose and tea bag) samples have higher elemental levels

when compared with the Brazilian green loose material (Table 3.9). This was

confirmed using a two-tailed Student t-test (Table 3.10), where Mg, Ca, and Zn

have significantly higher levels for the roasted samples (green < roast; 20

degrees of freedom, p<0.05). Furthermore, Na, K, V, Cr, Co, Ni, Se, Mo, Cd and

Pb (refer to Appendices 3.6 to 3.8) also had higher levels for the roasted samples

when compared to the green loose material (with the exception of As). The

roasting process changes and/or degrades a series of organic compounds

present in the material, including chlorogenic acids, flavonoids and polyphenols

(Clifford and Ramirez-Martinez, 1990, Farah et al., 2005, Bastos et al., 2006,

Marques and Farah, 2009, Bragança et al., 2011). Furthermore, during the

93

roasting process of the sample, there is a mass loss (volatiles and moisture),

resulting in an ‘enrichment’ of the elemental content of the roasted material

(Geiger et al., 2005, Franca et al., 2009, Schwartzberg, 2002).

Donnelly (2015) investigated the elemental content of commercial yerba

mate samples following a similar methodology to the one presented in this study.

In general, the data of this study, agrees with the reported values of Donnelly

(2015). Although, the analysed samples were different, the same trends were

found in relation to the origin (Brazil, Argentina) and packaging (loose, tea bags)

of the yerba mate products.

Table 3.10: Statistical analysis using a two-tailed Student t-test (Miller et al.,

2018) to evaluate the relationship between the origin (Brazil and

Argentina); packaging (loose and tea bags) and roasting (green and

roasted) of commercial yerba mate samples (refer to Table 3.9).

Origin

(Brazil and Argentina)

Packaging (Argentina)

Roasting (Brazil)

n 22 25 22

tcrit 2.09 2.07 2.09



Direction of


Direction of

significance

Mg 0.90 0.3750 ns 2.78 0.0105* loose<bag 2.62 0.0161* green<roast

Ca 0.92 0.3672 ns 0.28 0.7818 ns 4.55 0.0002*** green<roast

Mn 0.03 0.9755 ns 1.49 0.1489 ns 0.19 0.8535 ns

Fe 2.08 0.0507 ns 2.61 0.0156* loose<bag 1.74 0.0971 ns

Cu 3.35 0.0032** Bra>Arg 3.45 0.0022** loose<bag 1.95 0.0652 ns

Zn 3.54 0.0020** Bra<Arg 1.14 0.2666 ns 4.24 0.0004*** green<roast

Bra – Brazil; Arg – Argentina; n is the number of samples; tcalc is the calculated value (refer to Equation 2.6, section 2.6.2): tcrit is the critical value obtained for n-2 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant, p<0.05; ** highly significant, p<0.01, and *** very highly significant p<0.001.

Furthermore, Table 3.1 lists the elemental values for yerba mate products

cited by other studies. Overall, there is good agreement with the reported values,

especially for Mg (Bastos et al., 2014, Giulian et al., 2007, Heinrichs and

Malavolta, 2001, Malik et al., 2008, Pozebon et al., 2015, Rossa et al., 2015,

94

Alexandre Marcelo et al., 2014); Ca (Jacques et al., 2007, Pozebon et al., 2015);

Mn (Donnelly, 2015, Heinrichs and Malavolta, 2001, Malik et al., 2008, Magri et

al., Rossa et al., 2015); Fe (Barbosa et al., 2015, Bastos et al., 2007, Donnelly,

2015, Jacques et al., 2007, Malik et al., 2008, Magri et al., Rossa et al., 2015);

Cu (Heinrichs and Malavolta, 2001, Jacques et al., 2007, Milani et al., 2016) and

Zn (Barbosa et al., 2018). Interestingly, very few studies have cited data for Cd

and Pb, which in terms of this research will be looked at in depth in the next

section on yerba mate infusions. Moreover, many of the above studies were

conducted only on Brazilian samples, and none reported the various sub-studies

of this research, namely, the effect of processing (roasting) or type of commercial

product (green loose/tea bag) using the same analytical procedures and

instrumental analysis. This study also provides data for various elements not

reported in the literature for yerba mate products (with the exception of Donnelly

(2015), who conducted a study at the University of Surrey, but using different

commercial samples from Argentina and Brazil (Mn, Fe, Cu, Zn, V, Cr, Co, Ni,

As, Se, Mo Cd and Pb).

3.6.4. Elemental composition of yerba mate infusions

The following studies involved an investigation into the elemental levels of

infusions, which are very important as this is the main form of consumption of

yerba mate. Moreover, this data will then be used to assess the human dietary

intake of specific trace elements through the different types of yerba mate

consumption (regular ‘tea-based’/normal and iced, or bombilla), which will be

reviewed in section 3.6.6.

(i) Regular ‘tea-based’ infusions

The total elemental composition of commercial yerba mate products (from

Brazil and Argentina) prepared as regular ‘tea-based’ infusions was evaluated

following the method proposed in section 3.6.2. It should be stressed that in this

research a standardised ‘tea-brewing’ method was followed, according to that

published by Donnelly (2015). The results of the selected trace elements are

95

shown in Table 3.11, grouped according to the type of sample (green or roasted);

processing package (loose or tea bags) and origin (Brazil or Argentina), as

described in section 3.6.1. The results are presented as µg/200 mL representing

the elemental intake in a typical serving (i.e. a cup of 200 mL of infusion). Only

the trace element composition (not including the major elements) was

investigated in this study involving yerba mate infusions. Statistical analysis of

the data in Table 3.11 was undertaken using a two-tailed Student t-test (Miller et

al., 2018), based on the null hypothesis, that there is no statistically significant

difference in the trace elemental levels of the products grown in different

countries (Brazil and Argentina), type of packaging (loose and tea bags) or

roasting process (green and roasted). Table 3.12 reports the statistical data.

The rate of extraction of the trace elements present in yerba mate material

was determined by calculating the concentration in the infusions per gram of leaf

used to prepare the infusion, and expressing this as a percentage (Donnelly,

2015). The mean value of the calculated rate of extraction for the types of yerba

mate products from Brazil and Argentina are reported in Table 3.13.

In relation to the country of origin of the yerba mate (Brazil and Argentina),

the regular infusions of yerba mate (green loose) from Brazil had slightly higher

elemental levels. This is in agreement with the reported results for the total

elemental content, presented in section 3.6.3. As was discussed, this is probably

due to the ageing process of the commercial Argentinian products (Bastos et al.,

2006). Interestingly, the difference in the copper levels of the commercial yerba

mate products from Brazil and Argentina (with Brazil being significantly higher,

p<0.01, as reported in Table 3.10) was not as apparent in the prepared regular

infusions (p<0.05, refer to Table 3.12). Furthermore, the reverse trend was found

for zinc. These differences may be due to the rate of extraction of the elements,

similar for copper and higher in the Brazilian samples for zinc (refer to Table

3.13). Another possible explanation may be due to the difference in the particle

size of the commercial yerba mate products, where the material from Brazil is

powdered during the processing stages, whereas the samples from Argentina

are milled, as shown in Figure 3.5. In terms of the other trace elements reported

96

in Appendices 3.9 and 3.10, Cr, Co and Ni were found at higher levels in the

Argentinian green loose regular infusions. The rest of the elements (V, As, Se,

Mo, Cd and Pb) had slightly higher levels in the Brazilian regular infusions.

The next study involved an evaluation of the infusion process involving

green loose and tea bag products from Argentina. When comparing the

elemental content of a regular infusion associated with a single serving (cup of

tea) using green loose and tea bags, there was a highly significant difference

between the samples. All the elements (Mn, Fe, Cu, Zn presented in Table 3.11

and for V, Cr, Co, Ni, As, Se, Mo Cd and Pb, reported in the Appendices 3.9 and

3.10) had higher levels for the tea bag samples when compared to the green

loose regular infusions. This finding suggests that the trace element extraction

efficiency (based on the calculated percentage extraction) of the material

packaged in the tea bags is higher when compared to the green loose yerba

mate products. The difference in the particle size of the material (loose: 3 – 5

mm; and tea bags - less than 1 mm) could be responsible for this variance.

Furthermore, the higher elemental content determined in the infusion of the

green tea bag samples may be a reflection of the ease of extraction from the

leafier plant parts found in the tea bags than the mixture of leaf and stems found

in the green loose material (Donnelly, 2015). The higher extraction rates of these

elements for yerba mate sold as green tea bags suggests, for example, that the

Mn, Cu and Zn are more easily extracted from the leafy parts of yerba mate than

the stems (refer to Table 3.13).

The elemental levels of green loose and roasted (from Brazil) regular

infusions from Brazil were evaluated, as presented in Table 3.11. Interestingly,

the opposite trend was found for the digested commercial samples when

compared with the regular infusions. In general, the regular infusions made with

green loose yerba mate had highly significant levels of Mn, Cu and Zn (two-tailed

Student t test, p<0.001) than the infusions prepared with roasted samples. The

same trend was found for Co, Ni, Mo, Cd and Pb, as reported in Appendices 3.9

and 3.10. Iron, V, Cr, As and Se had similar levels for both regular infusions

prepared with commercial green loose and roasted yerba mate. The elemental

97

levels in these infusions could depend on the levels of organic chelating species,

such as, chlorogenic acids, tannins, and other phenolic compounds present in

yerba mate, which may be modified or degraded during the roasting step

(Bragança et al., 2011).

Table 3.11: Elemental levels (µg/200 mL), reported as mean and range (min –

max), of regular infusions of commercial yerba mate samples from

Brazil and Argentina. The analyses were determined using ICP-MS



n 16 7 19 3 4

Mn 1409

(473 – 3612)

1331

(519 – 4256)

2439

(1282 – 3937)

431

(373 – 496)

632

(491 – 861)

Fe 5.38

(2.53 – 9.66)

5.03

(1.91 – 8.21)

15.43

(8.74 – 20.48)

3.32

(2.22 – 4.38)

7.84

(5.85 – 10.65)

Cu 9.11

(4.69 – 13.12)

7.40

(4.15 – 11.53)

12.53

(3.62 – 20.37)

0.54

(0.30 – 0.86)

0.29

(0.08 – 0.67)

Zn 57.56

(33.67 – 82.01)

58.97

(44.85 – 93.13)

143.36

(100.84 – 191.92)

23.73

(9.47 – 35.76)

23.83

(18.97 – 30.05)

n is the number of samples

98

Table 3.12: Statistical analysis using a two-tailed Student t-test (Miller et al.,

2018) to evaluate the relationship between the origin (Brazil and

Argentina); packaging (loose and tea bags) and roasting process

(green loose and roasted) of regular infusions of commercial yerba

mate (refer to Table 3.11).

Origin

(Brazil and Argentina)


Roasting (Brazil)

n 22 26 22

tcrit 2.09 2.07 2.09



Direction of


Direction of

significance

Mn 0.97 0.3441 ns 4.59 0.0001*** loose<bag 4.60 0.0002*** green>roast

Fe 1.46 0.1612 ns 6.42 <0.0001*** loose<bag 0.19 0.8512 ns

Cu 2.30 0.0325* Bra>Arg 3.13 0.0046** loose<bag 9.36 <0.0001*** green>roast

Zn 0.77 0.4497 ns 7.13 <0.001*** loose<bag 5.83 <0.0001*** green>roast

Bra – Brazil; Arg – Argentina; n is the number of samples; tcalc is the calculated value (refer to Equation 2.6, section 2.6.2): tcrit is the critical value obtained for n-2 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant, p<0.05; ** highly significant, p<0.01, and *** very highly significant, p<0.001. Table 3.13: Percentage extraction (%) of regular infusions of commercial yerba

mate samples from Brazil and Argentina. Values reported as a mean

and range (min – max).


n 16 7 19 3 4

Mn 99

(44 – 160)

82

(44 – 116)

113

(73 – 184)

29

(28 – 31)

47

(31 – 93)

Fe 7

(4 – 25)

12

(6 – 25)

10

(3 – 33)

6

(4 – 8)

5

(4 – 9)

Cu 50

(29 – 80)

44

(29 – 80)

45

(15 – 74)

3

(2 – 4)

1

(0 – 3)

Zn 55

(28 – 80)

40

(28 – 51)

53

(9 – 80)

10

(7 – 12)

11

(11 – 17)

99

Comparative data in the literature is limited due to the lack of studies

reporting data on yerba mate infusions and the variability of the methods used for

infusions. However, Donnelly (2015) reported the elemental composition of

commercial yerba mate infusions, prepared using the same methodology as

proposed by this study. In general, the trace element concentrations in this study

are similar to the limited number of reported levels (Donnelly, 2015). The

exceptions include, the lower levels of copper in the Brazilian roasted and lower

levels of zinc in the green loose infusions, found in this study.

South American countries have set a limit for the concentration of toxic

metals in yerba mate products of 0.6 mg/kg for arsenic and lead and 0.4 mg/kg

for cadmium (DOU, 2013). All of the commercial yerba mate products analysed

in this study had As, Pb and Cd concentrations below these limits. Moreover, the

mean ‘toxic’ metal levels determined in this study for the regular infusions of

green loose: 0.025 µg/200 mL As; 0.04 µg/200 mL Pb and 0.07 µg/200 mL Cd;

and tea bags: 0.09 µg/200 mL As; 0.04 µg/200 mL Pb and 0.15 µg/200 mL Cd,

respectively. Therefore, the consumption of commercial yerba mate infusions

does not pose a risk of ‘dietary’ exposure to these specific toxic metals (Barbosa

et al., 2015, Pozebon et al., 2015). The lower levels of cadmium and lead in the

yerba mate infusions may be explained by the low levels in the leaf material

(refer to Appendix 3.10) and the ability of these elements to form complexes and

be retained in the plant material, even after the extraction with hot water (Michie

and Dixon, 1977).

(ii) Brazilian iced tea

In Brazil, one of the most popular methods of consuming yerba mate is by

iced tea infusions (refer to section 3.2.2). In order to evaluate the elemental

extraction for Brazilian yerba mate iced tea infusions, solutions were prepared

according to the method reported in section 3.6.2. The results are presented in

Table 3.14. In terms of comparative analysis, two methods were investigated,

namely, the regular infusion (refer to the section above) and the Brazilian iced

tea method. The two methods have different brewing times (regular infusions – 4

100

minutes; iced tea – 7 minutes) and proportions of yerba mate sample (mass of

solid to volume of water) (refer to section 3.6.2). In general, the Brazilian iced

infusion method resulted in higher elemental levels per serving associated with

the more concentrated infusions in comparison with the regular method (Tables

3.11 and 3.14).

As a result of constraints of time, selected samples (n = 3) of each group

were investigated by this infusion method. Therefore, the results will not be

subjected to statistical analysis due to the limited number of samples. A primary

evaluation of the data presented in Table 3.14 shows a trend where the Mn, Fe

and Cu levels were higher in the iced tea infusions (using Brazilian green loose

yerba mate) when compared to that using Argentinian material. An exception

was found for Zn where the Argentinian material resulted in higher levels in the

infusions; the same trend was found for the analysis of the commercial product

material (refer to section 3.6.3). A similar trend was found for Cr and Ni; but the

rest of the elements reported in Appendices 3.11 and 3.12 did not show a

significant difference between the countries.

Furthermore, the iced tea infusions produced using yerba mate material

packaged as tea bags had higher levels of Mn, Fe, Cu and Zn (Table 3.14), and

V, Cr, Co, Ni, As, Se, Mo Cd and Pb (Appendices 3.11 and 3.12) when

compared to the green loose material. Finally, the Brazilian green loose yerba

mate had higher levels of Mn, Cu and Zn when compared to the roasted

samples. The same trend was found for Co, Ni and Mo, as reported in

Appendices 3.11 and 3.12. Iron, V, Cr, As, Se, Cd and Pb (Appendices 3.11 and

3.12) had similar levels for iced tea infusions prepared with Brazilian green loose

and roasted yerba mate products.

101

Table 3.14: Elemental levels (µg/200 mL), reported as the mean and range (min

– max), of Brazilian iced tea infusions prepared using commercial

yerba mate products from Brazil and Argentina. The analyses were

determined using ICP-MS (refer to section 2.3). n is the number of

infusion samples.

Green loose Green tea bag Roasted loose

Roasted tea bag

Origin Brazil Argentina Argentina Brazil Brazil n 3 3 3 3 3

Mn 2276

(1721 – 2660)

1596

(1295 – 1964)

2716

(2061 – 3227)

820

(672 – 899)

650

(542 – 811)

Fe 10.66

(9.12 - 13.04)

6.28

(6.14 – 6.41)

28.35

(17.77 – 38.93)

6.93

(6.88 – 6.98)

17.51

(14.80 – 20.21)

Cu 20.85

(16.38 – 23.62)

12.12

(10.95 – 14.45)

21.62

(12.47 – 28.44) <LOD <LOD

Zn 79.76

(69.91 – 89.24)

93.04

(78.43 – 109.22)

236.65

(139.64 – 316.00)

38.39

(15.87 – 59.64)

30.41

(26.18 – 34.89)

(iii) Bombilla method

In order to simulate the traditional consumption of yerba mate in South

America, a bombilla method was performed following the procedure described in

section 3.6.2. The results for manganese are shown in Table 3.15, grouped by

country of origin and the concentration for the various fractions (that is the

successive additions of hot water, using DDW, to the same yerba mate material

in the cup). Manganese was selected because of the high levels found in yerba

mate products (refer to Table 3.9) and the importance of this trace element in

human health (refer to section 1.1). The full data for all of the elements are

reported in Appendices 3.13 to 3.19.

102

Table 3.15: Manganese levels (µg/200 mL) of bombilla infusions of commercial

green loose yerba mate samples from Brazil (n= 16) and Argentina

(n=7). n is the number of samples. The total manganese content

refers to a sum of the five fractions. The analyses were determined

using ICP-MS (refer to section 2.3).

Brazil Argentina

Fraction Mean Contribution (%) Range Mean Contribution

(%) Range

F1 13906 40.8 3120 – 30592 5312 22.6 2555 – 7165

F2 9801 28.7 3025 - 16964 5828 24.8 2564 – 9003

F3 5250 15.4 1449 – 7806 4596 19.5 1804- 6904

F4 3089 9.1 741 – 5869 5039 21.4 2537 – 13747

F5 2047 6.0 437 - 3756 2750 11.7 1102 - 4311

Total 34093 100% 23526 100%

Table 3.15 shows that there is a variation in the Mn levels of the fractions

(F1 to F5) based on the successive addition of hot water to the original mass of

yerba mate (green loose). This is not surprising as the elemental analysis of the

commercial material from Brazil and Argentina (Table 3.9) also showed a range

of Mn levels; i.e. for Brazilian green loose (486 to 834 mg/kg, Mn, d.w.) and

Argentinian (383 to 671mg/kg Mn d.w.). Moreover, the regular infusion data for

these commercial products had a similar variance in the Mn levels (Table 3.11).

The Brazilian green loose yerba mate initially provides more manganese in the

infusion fraction (F1) than that for the Argentinian product (based on the addition

of 100 mL of hot water). However, after subsequent additions of hot water

(fractions F2 to F5) the Mn levels are basically the same in the resultant

infusions, as shown in Figure 3.8. On the other hand, the accumulation (potential

intake level) of the element after the first fraction (F1) of yerba mate, purchased

from both countries, results in a similar pattern in terms of the Mn level (refer to

Figure 3.8). Both products result in a higher level of Mn intake through the

consumption of green loose yerba mate using this bombilla method, when

103

compared with the upper limit level for Mn intake of 11 mg/day (IOM, 2002). The

potential health impact of this will be discussed in section 3.6.6.

Figure 3.8: Manganese concentration (mg/200 mL) and rate of accumulation*

(i.e. potential intake) between 5 fractions (successive additions of hot

water) based on using the bombilla method (refer to section 3.6.2) for

green loose yerba mate purchased in Brazil (n= 16) and Argentina

(n=7). n is the number of samples. The WHO set the upper limit for

Mn in 11 mg/day (IOM, 2002). The analyses were determined using

ICP-MS (refer to section 2.3). *Calculated as the sum of the previous

fractions.

In terms of the other trace elements (refer to Appendices 3.13 to 3.19)

there is a similar pattern in terms of the fraction levels and the accumulation (F1

to F5 fractions). In general, all the elemental (Fe, Cu, Zn, Ni, Cr, V, Co, Cd and

Pd) concentrations steadily decrease from F1 to F5. The exception is for As, Se,

Mo where all of the fractions have similar concentrations (each fraction

contributes to approximately 20% of the sum of all fractions). Comparison with

the sum of the elemental intake of the 5 fractions for these trace elements leads

to the following order of intake: Zn > Cu > Ni > Fe > Cr > Co > Cd > Pb > Mo >

0

5

10

15

20

25

30

35

40

1 2 3 4 5 6

Con

cent

ratio

n of

Mn

(mg/

200m

l)

Fraction

ArgentinianFractionsBrazilianfractionsBrazilaccumulationArgentinaaccumulationWHO Upperlimit

104

As > Se > V. This refers to the specific solubility or ‘bioavailability’ of each

element through a hot water sequential extraction. Moreover, for As, Pb and Cd,

although there is an increase in the levels per fraction (which was previously

reported in section 3.6.3), this still did not present a significant impact on the

dietary intake levels). Finally, an assessment of the dietary intake of these

elements is presented in section 3.6.6.

3.6.5. Polyphenolic composition of yerba mate

Polyphenols are plant metabolites that have at least one aromatic ring with

one or more hydroxyl groups attached (Campos‐Vega and Oomah, 2013). These

molecules have been proven to be linked to a series of health benefits relating to

the antioxidant activity (Bastos et al., 2007, Colpo et al., 2016). Many studies

have demonstrated the strong antioxidant ability of these molecules in in vitro

analyses, by removing free radicals and reactive oxygen species that may have

harmful effects (Donnelly, 2015). Beverages, fruits and vegetables, may make a

significant contribution to the dietary intake of polyphenols (Fukushima et al.,

2009, Vanamala et al., 2006), as discussed in section 1.2. The measurement and

health beneficial effects of polyphenols in yerba mate infusions have been

reported, as reviewed in section 3.3.

3.6.5.1 Total polyphenol of infusions

In order to evaluate the potential intake of polyphenols in different infusion

methods of yerba mate (regular ‘tea-based’, Brazilian iced tea, or the traditional

bombilla), the total polyphenol content of the infusions was analysed by the Folin-

Ciocalteu assay involving a UV-Vis spectrometer, as described in section 2.4.

Moreover, this data will then be used to assess the intake of total polyphenol

content through the different types of yerba mate (green loose, tea bags and

roasted) in section 3.6.6. The full data for all of the samples are reported in

Appendices 3.20 to 3.22.

105

(i) Regular infusions

The total polyphenol content of commercial yerba mate products (from

Brazil and Argentina) prepared as regular ‘tea-based’ infusions was evaluated

following the method proposed in section 3.6.2. The regular infusion method was

that reported as standardised ‘tea-brewing’ (Donnelly, 2015). The results are

presented in Table 3.16, grouped according to the type of sample (green or

roasted); processing package (loose or tea bags) and origin (Brazil or Argentina).

The results are once more presented as µg/200 mL representing the total

phenolic intake in a typical serving (i.e. a cup of 200 mL of infusion).

A preliminary study of the data presented in Table 3.16 shows that the

commercial green loose yerba mate samples from Brazil have a slightly higher

mean total polyphenol content when compared to the Argentinian green loose

material. Although, the range of results are similar between the two countries, a

two-tailed Student t-test analysis confirmed that there is not a statistically

significant difference between the Brazilian and Argentinian green loose infusion

levels (n = 22, tcrit = 2.09 > tcalc = 1.54; p = 0.1391). This trend is similar to that

found for the elemental content of the yerba mate material (refer to section 3.6.3)

and infusions (refer to section 3.6.4). One of the main differences between

Argentinian and Brazilian production is that in the former country, the yerba is

grown in the sun (because it is cultivated), whilst in the latter, trees are shaded

within local forests. It has been reported that the different types of cultivation may

have an impact on the chemicals presents in yerba, especially polyphenols

(Donnelly, 2015). Furthermore, the Argentinian products are aged for up to 24

months in chambers, whilst the Brazilian commercial yerba mate is freshly

packed. The polyphenols are molecules that can easily degrade or change

during long storage periods or following exposure to high temperatures (Klimczak

et al., 2007). Therefore, it is interesting to note that whilst there are differences

between the two countries, in terms of yerba mate cultivation and processing, the

samples have similar total polyphenol levels.

106

The Argentinian commercial yerba mate packed as green loose or as tea

bags, have higher levels of total polyphenols in the tea bags, as was previously

reported for the elemental content (refer to section 3.6.4). Statistical analysis

using a two-tailed Student t-test confirmed a very highly significant difference,

that is, the total polyphenol content of Argentinian tea bags > green loose (n =

26, tcrit = 2.07 < tcalc = 5.78 and p < 0.0001). The particle size of the yerba mate

packed in the tea bags is smaller (1 mm) than the product packed as green loose

material (4 - 5 mm). This suggests that the polyphenols are more easily extracted

when there is an increase in the leaf contact area. Furthermore, the amount of

stems present in yerba mate sold as green loose also influences the amount of

polyphenols and methyl xanthine compounds present (Donnelly, 2015). Products

that contained 12.4 to 15% stems results in 20 to 40% lower levels of total

polyphenols than products without stems (Tamasi et al., 2007). In addition, the

caffeine content has also been found to be negatively correlated with the amount

of yerba mate stems present in the product (Mazzafera, 1997).

Finally, an evaluation of the impact of the roasting process on the total

polyphenolic content was performed by a comparison between the Brazilian

green and roasted commercial products. There was a very highly significant

difference between these products (a two-tailed Student t-test, where n = 22, tcrit

= 2.09 < tcalc = 8.01 and p < 0.0001). The Brazilian green loose material

presented a much higher total polyphenolic content when compared to the

roasted material. This confirms that the heating process leads to the loss of some

polyphenols, as has been previously reported (Clifford and Ramirez-Martinez,

1990, Marques and Farah, 2009, Donnelly, 2015). Therefore, the roasting

process has a major impact on the presence and levels of polyphenols in the

yerba mate products, but not on the elemental content, as was discussed in

section 3.6.4.

It is difficult to compare this data with that reported in the literature,

because there is a variability in the methods of preparing an infusion. Different

mass-to-water ratios (from 1:82 to 1:100) and infusion times (from 5 to 10

minutes) have been used by different research groups (Gorjanovic et al., 2012,

107

de Mejía et al., 2010, Bravo et al., 2007, Bastos et al., 2006). In comparison,

Donnelly (2015) used the same method of infusion as this study and the reported

range of total polyphenols was found to be very similar (79.9 – 303.1 mg gallic

acid equivalents (GAE)/200 mL for green samples and 79.9 – 303.1 mg

(GAE)/200 mL for roasted samples).

Table 3.16: Total polyphenol content (mg/200 mL), reported as the mean and

range (min – max), of regular infusions of commercial yerba mate

samples from Brazil and Argentina. The samples were analysed by

the Folin-Ciocalteu method using a UV-Vis spectrometer (refer to

section 2.4). n is the number of samples.

Type Origin n Total polyphenol content

Green loose Brazil 15

142.6

(79.7 - 169.1)

Argentina 7 125.0

(81.9 - 159.0)

Green tea bag Argentina 19 215.7

(134.8 - 261.0)

Roasted loose Brazil 3 45.2

(44.0 - 46.7)

Roasted tea bag Brazil 4 70.2

(56.3 - 93.6)

(ii) Brazilian iced tea

An evaluation of the potential intake of total polyphenols through

consuming yerba mate from Brazil (as an iced tea infusion) was assessed in this

study (refer to section 3.2.2). The total polyphenol content of the Brazilian yerba

mate iced tea infusions was measured, as described in section 2.4, using

solutions prepared according to the method outlined in section 3.6.2. The results

are presented in Table 3.17. In order to compare the intake levels linked to the

drinking of regular infusions, it should be stressed that the methods have

different: (i) brewing times (regular infusions – 4 minutes; iced tea infusions – 7

108

minutes); and (ii) proportions of yerba mate sample (mass of solid) to volume of

water (refer to section 3.6.2). Interestingly, the Brazilian regular and iced tea

infusion methods resulted in similar levels of total polyphenols. This suggests

that the polyphenols present in the yerba mate material are equally extracted by

the two methods.

Table 3.17: Total polyphenol content (mg/200 mL), reported as the mean and

range (min – max), of Brazilian iced tea infusions of commercial

yerba mate samples from Brazil and Argentina. The analyses were

determined by the Folin-Ciocalteu method using a UV-Vis

spectrometer (refer to section 2.4). n is the number of samples.

Type Origin n Total polyphenol Green loose Brazil 1 112.2

Green tea bag Argentina 3 275.5

(216.5 – 312.5)

Roasted loose Brazil 3 35.5

(29.2 – 41.7)

Roasted tea bag Brazil 3 71.5

(64.0 – 83.0)

(iii) Bombilla method

A bombilla method was performed in this study, in order to simulate the

traditional consumption of yerba mate in South America, as described in section

3.6.2. The results for the total polyphenol content are shown in Table 3.18, grouped by country of origin and the concentration for the various fractions

(successive additions of hot water to the same yerba mate material in the cup). In

comparison with the regular infusion method, the concentrations of the total

polyphenols in the first fractions are 10 times higher in the bombilla method. As

previously discussed, there was also a difference in the total polyphenol content

of the commercial material from Brazil and Argentina. Furthermore, the regular

infusion data for these commercial products had a similar variance in the levels

of total polyphenols (Table 3.16). In terms of the elemental data for the bombilla

109

infusions (presented in Table 3.15 and Figure 3.8), the percentage contribution to

the total concentration (the sum of the fractions, F1 to F5) is very similar to the

total polyphenol content (Table 3.18). Data reported for the elemental levels of

plants, and in prepared beverages, concluded that the elements can exist in the

infusions as free ions or complexes with naturally occurring bioligands. This is

particularly true for polyphenols, which can complex elements (or metals) through

hydroxyl, carboxylate and phenolate groups (Pohl and Prusisz, 2007, Khokhar

and Owusu Apenten, 2003). The findings of this study suggests that the

elements in these infusions could be chelated to the phenolic compounds

present in yerba mate (Bragança et al., 2011). Pohl and Prusisz (2007) proposed

that element-binding by flavonoids considerably reduces the bioavailability for the

body and/or impairs the absorption of these chemicals.

Table 3.18: Total polyphenol content (mg/200 mL) of bombilla infusions of

commercial green loose yerba mate samples from Brazil (n= 16)

and Argentina (n=7). The total content refers to the sum of the five

fractions. The samples were analysed by the Folin-Ciocalteu

method using a UV-Vis spectrometer (refer to section 2.4). n is the

number of samples.

Brazil Argentina

Fraction Mean Range Contribution (%) Mean Range Contribution

(%) F1 1559.7 475.4 - 2451.4 41.0 882.5 610.0 - 1011.9 29.4

F2 997.8 421.8 - 1215.6 26.2 828.2 565.5 - 999.7 27.6

F3 591.9 300.4 - 807.1 15.5 578.9 444.3 - 717.1 19.3

F4 383.8 194.4 - 552.8 10.1 412.8 302.2 - 473.3 13.7

F5 274.0 129.8 - 386.8 7.2 303.3 226.7 - 340.9 10.1

Total 3807.2 100% 3005.6 100%

110

3.6.5.2 Chlorogenic acids, caffeine and theobromine levels in yerba mate infusions

The predominant type of polyphenol present in yerba mate is the

chlorogenic acid group, including a range of mono-, di- and tri-acylated

compounds. The main chlorogenic acids present in yerba mate are 3-, 4- and 5-

caffeoylquinic acids (Donnelly, 2015). Furthermore, there are important

xanthines, namely, caffeine and theobromine, that are also present in yerba

mate. They are related to a series of health claims linked with the consumption of

yerba mate infusions (refer to section 3.3).

The amount of these compounds in the regular infusions of commercial

yerba mate products from Brazil was determined by ultra-high performance liquid

chromatography (UHPLC) (refer to section 2.5). The regular ‘tea-based’ infusions

were prepared following the method proposed in section 3.6.2., according to that

published by Donnelly (2015). The results are shown in Table 3.19, grouped

according to the type of sample (green or roasted) and processing package

(loose or tea bags), as described in section 3.6.1. Only the commercial yerba

mate from Brazil was investigated in this study because the data for the

Argentinian material is already available (Donnelly, 2015). The full data for all of

the samples are reported in Appendices 3.23 and 3.24.

This study involved an evaluation of the yerba mate infusions involving

green loose and roasted (loose and tea bags) products from Brazil. It is clear

from the data presented in Table 3.19 that all of the organic compounds

(caffeoylquinic acids, caffeine and theobromine) are changing or suffering

degradation during the roasting process. This was to be expected since it was

proven that the roasting process (high temperatures) can change or degrade

organic compounds, such as, chlorogenic acids and xanthine (Clifford and

Ramirez-Martinez, 1990, Farah et al., 2005, Bastos et al., 2006, Marques and

Farah, 2009, Bragança et al., 2011).

111

It was also noted from Table 3.19 that the roasted loose samples have

lower levels than the roasted tea bags of caffeoylquinic acids, caffeine and

theobromine. The yerba mate packed in the tea bags has more leaf material, as

previously discussed. This indicates that the chlorogenic acids and xanthines are

present at higher levels in the leaf compared with the stems (Donnelly, 2015,

Tamasi et al., 2007).

Table 3.19: Chlorogenic acid, theobromine and caffeine content (mg/200 mL) of

regular infusions of commercial yerba mate samples from Brazil. The

samples were analysed by UHPLC (refer to section 2.5). For green

loose samples the results are presented as mean and range (min –

max). n is the number of samples.

Type Green loose Roasted loose Roasted tea bag n 4 1 1

3-Caffeoylquinic acid 46.84

(41.44- 55.62) 2.86 3.98

Theobromine 4.02

(3.31- 5.03) 1.13 1.53


(12.99- 15.73) 2.03 3.16


(19.43- 29.09) 2.79 3.62

Caffeine 25.49

(18.57- 34.58) 5.14 7.58

Donnelly (2015) also reported the following concentrations (mg/200 mL)

for the commercial green loose yerba mate samples from Argentina: 33.37 of 3-

caffeoylquinic acid; 13.35 of 4- caffeoylquinic acid; 22.10 of 5- caffeoylquinic

acid; 1.67 of theobromine and 11.62 of caffeine. The green loose yerba mate

from Brazil has higher levels of caffeoylquinic acids, caffeine and theobromine

(refer to Table 3.19) than the Argentinian samples. The cultivation practices and

112

processing of yerba mate leaf have an impact on the amount of polyphenol and

xanthine compounds (Donnelly, 2015). Sunlight exposed yerba mate plantations

normally lead to higher amounts of caffeoylquinic acids, caffeine and

theobromine, than for samples collected from shaded plantations (Dartora et al.,

2011). Streit et al. (2007) reported lower levels of chlorogenic acids and caffeine

in yerba mate infusions produced from Brazilian products collected from

reforested rather than native plantations.

(ii) Bombilla method

In order to evaluate the traditional consumption of yerba mate in South

America, a bombilla method was performed following the procedure described in

section 3.6.2. The results for chlorogenic acids are shown in Table 3.20 and for

xanthines (theobromine and caffeine) in Table 3.21; grouped by the

concentration for the various fractions, F1 to F5 (that is the successive addition of

hot water to the same yerba mate material). In comparison with the regular

infusion method (refer to Table 3.19), the bombilla method resulted in much

higher concentrations of the analysed compounds (10 times higher in the first

fractions). The percentage contribution of all three chlorogenic acids follows the

same pattern for F1 to F5 fractions (refer to Table 3.20), suggesting comparable

extraction behavior. The same trend was also found for the xanthines (refer to

Table 3.21). This study provides important data showing that any differences

relating to the various infusion methods (regular infusion and bombilla) must be

taken into consideration when using the data to predict the potential intake of

these chemicals through consuming yerba mate by individuals in South America.

113

Table 3.20: Chlorogenic acid content (mg/200 mL) of bombilla infusion fractions

of green loose commercial yerba mate products from Brazil. The

samples were analysed by UHPLC (refer to section 2.5). n = 4, n is

the number of samples. The percentage (%) refers to the contribution

of the fraction to the total (sum of the fractions).

3-Caffeoylquinic acid 4-Caffeoylquinic acid 5-Caffeoylquinic acid

Mean (%) Range Mean (%) Range Mean (%) Range

F1 516.61 36.8 305.08 - 761.37 145.76 34.1 91.04 - 174.31 239.59 33.5 136.81- 284.63

F2 441.82 31.8 395.68 - 483.18 140.75 32.6 116.46- 160.69 232.56 32.5 184.65- 263.42

F3 242.32 17.6 224.83 - 269.4 79.78 18.5 62.11 - 87.45 134.64 18.7 98.18- 159.57

F4 121.22 8.8 82.47 - 198.35 40.23 9.3 29.25 - 64.17 66.97 9.5 46.75- 96.41

F5 72.12 5.2 47.12 - 133.68 24.28 5.6 17.04 - 42.76 40.69 5.8 27.16- 64.82

Total 1394.08 100% 430.8 100% 714.45 100%

Table 3.21: Theobromine and caffeine content of bombilla infusion fractions of

green loose yerba mate commercial products (mg/200 mL) from

Brazil. The samples were analysed by UHPLC (refer to section 2.5).

n = 4, n is the number of samples. The percentage (%) refers to the

contribution of the fraction to the total.

Theobromine Caffeine

Mean (%) Range Mean (%) Range

F1 41.5 36.5 23.8 - 52.43 258.76 35.8 144.13 - 425.62

F2 34.52 30.8 30.66 - 40.50 211.81 30.3 163.09 - 267.67

F3 19.89 17.7 17.31 - 25.46 124.3 17.9 92.96 - 144.85

F4 10.22 9.2 6.87 - 14.68 67.27 9.7 37.30 - 97.79

F5 6.35 5.8 4.25 - 10.51 43.68 6.3 23.88 - 70.90

Total 112.48 100% 705.82 100%

114

3.6.6. Link to dietary intake through consumption of yerba mate

One of the aims of this study was to evaluate the effect of the mode of

consumption of yerba mate in terms of the nutritional intake from consuming

infusions of the products. This study will focus only on the potential intake of the

total levels of elements and polyphenols.

3.6.6.1 Dietary intake of trace elements

An evaluation of the contribution of different infusions prepared from

commercial samples of yerba mate from Brazil and Argentina, in terms of the

levels of trace elements and polyphenols, was undertaken using the data

presented in this chapter. It is important to highlight the differences in the serving

size of the different infusion methods. The serving size for regular tea-based and

Brazilian iced tea infusions is a cup (200 mL). The bombilla method is

traditionally performed with 50 g of yerba mate material and 1 litre of hot water

(added as successive fractions of 200 mL). Therefore, the daily amount of yerba

mate infusion consumed using a bombilla is close to 1 litre.

The recommended daily intake (RDA) of essential elements is defined by

the World Health Organisation or WHO (WHO, 1996) for males (M) and females

(F). The WHO RDA guidelines are compared with the calculated % intake of the

chemicals for the consumption of yerba mate infusions. All of the trace elements

determined in this study represent from 0.1 to 5% of the RDA, for all the infusion

methods (regular, Brazilian and bombilla) as reported in Appendices 3.9 to 3.19.

The exception is for manganese. Table 3.22 compares the different methods of

yerba mate infusion in relation to the recommended intake and upper limit of

manganese on a daily basis. A regular infusion serving (1 cup of 200 mL) can

provide 23.7 to 106.0 % for males and 30.3 to 135.5 % for females, of the daily

recommended manganese intake; depending on the type of yerba mate product.

115

Even at the highest Mn concentrations in regular infusions the contribution to the

Mn uptake level is still below the upper limit value (WHO, 1996). The Brazilian

iced tea infusions provided a slightly higher contribution to the intake of

manganese. In terms of the bombilla method the daily amount of yerba mate

infusion consumed is close to a volume of 1 litre. This provides the

recommended daily intake of manganese and could provide concentrations

higher than the recommended upper consumption limit for this element per day

(refer to Table 3.22). It must be noted that this is a potential intake study

regarding the determination of the total concentration of manganese for these

different yerba mate infusion methods and further research needs to be

undertaken to evaluate the possible toxicity of manganese in relation to

consuming yerba mate using the bombilla method. In relation to the human

exposure of elements and polyphenols in food, it is important to also consider the

bioavailability of the compound or element. Bioavailability is defined as the

fraction of the analyte which can be absorbed and utilised for physiological

functions (Fairweather-Tait and Hurrell, 1996). Many factors can have an impact

on the absorption of elements, such as, the intake of polyphenols and dietary

fiber which can decrease the level of elemental absorption (Finley et al., 2011).

The manganese data reported in Table 3.11 and the total polyphenol values in

Table 3.16 for regular infusions of yerba mate presents a tendency of the

respective concentrations to increase together (Spearman’s correlation; R= 0.24,

p=0.23 n= 26 and = 0.05). This correlation indicates that the amount of total

polyphenol content could be potentially related to the manganese content of the

yerba mate infusions, influencing the bioavailability of both chemicals. In

particular, it is important to now develop and apply a method to determine the

bioavailability of manganese in these infusions and how such chemical forms

may be influenced by the presence of polyphenols in the infusions. The

concentration of manganese in yerba mate is the highest when compared to

other typical beverages, such as, coffee or green and black tea (Unicamp, 2011).

Moreover, based on the low levels of the toxic elements in the yerba mate

116

infusions (regular, Brazilian and bombilla), the daily intake of As, Cd and Pb were

found to not be significant (refer to Appendices 3.10, 3.12 and 3.17 to 3.19).

Table 3.22: Percentage intake (%) of manganese based on a serving (200 mL for

regular and Brazilian iced tea infusions; 1L for bombilla method) of

non-commercial yerba mate samples. The data is compared with the

World Health Organisation recommended daily allowance (RDA) of

manganese for males (M) and females (F).

WHO RDA* Upper limit* M F

Type Origin 2.3 1.8 11

Regular infusions

Green loose Brazil 67.6 86.4 14.1

Argentina 44.3 56.6 9.3

Tea bags Argentina 106.0 135.5 22.2

Roasted Brazil 23.7 30.3 5.0

Brazilian iced tea


Argentina 86.1 110.0 18.0

Tea bags Argentina 118.1 150.9 24.7

Roasted Brazil 32.0 40.8 6.7

Bombilla method


Argentina 1022.9 1307.0 213.9

* World Health Organisation Recommended Daily Allowance and Upper limit (reported in mg/day) (WHO, 1996).

3.6.6.2 Dietary intake of polyphenols

An estimation of a recommended daily intake for total polyphenols is

difficult due to the variation in the levels of the phenolic compounds in a particular

foodstuff, structural diversity of the phenolic compounds, or lack of standardised

analytical methods (Scalbert and Williamson, 2000). Many studies have agreed

on a range of 1 g of total polyphenols per day (Kühnau, 1976, Faller and Fialho,

117

2009, Landete, 2013). Most authors refer to Fukushima et al. (2009), to evaluate

the intake of polyphenols. The author calculated a daily consumption of 1492 mg

(fresh weight) of polyphenols, based on a balanced Japanese diet. Table 3.23

compares the different methods of yerba mate infusion in relation to the daily

intake of total polyphenols (Fukushima et al., 2009). A regular infusion serving (1

cup of 200 mL) will contribute 4 to 14.5 % of the daily intake of total polyphenols.

The Brazilian iced tea infusions provided a similar contribution to the intake of

total polyphenols. In the bombilla method the daily amount of yerba mate infusion

serving is 1 litre. This traditional method of consuming yerba mate in South

America can provide up to twice the amount of the adequate daily intake of total

polyphenols (refer to Table 3.23).

The range of total polyphenols in yerba mate regular infusions is similar to

that reported for green or black tea and fruit (grape, apple and orange) juices, but

is half of the total polyphenolic content of filter coffee (Donnelly, 2015). Although,

when the traditional method of consuming yerba mate in South America

(bombilla method) is evaluated, a single serving (1L) would make a significant

contribution to the total polyphenol intake for this population.

118

Table 3.23: Percentage intake (%) of total polyphenol based on a serving (200

mL for regular and Brazilian iced tea infusions; 1L for bombilla

method) of commercial yerba mate samples. The data is compared

with the values reported by Fukushima et al. (2009) for the daily

intake of polyphenols.

Total polyphenol daily intake (mg/day)

Type Origin 1492

Regular infusion

Green loose Brazil 9.6

Argentina 8.4

Tea bags Argentina 14.5

Roasted Brazil 4.0

Brazilian iced tea Green loose Brazil 7.5

Tea bags Argentina 18.5

Roasted Brazil 3.6

Bombilla method

Green loose Brazil 255.2

Argentina 201.4

The total polyphenol data for yerba mate infusions was compared with the

European flavonoid database Phenol-Explorer (Pérez-Jiménez et al., 2010). In

this, filter coffee was ranked 37th (534 mg GAE/200 mL), black tea 58th (208 mg

GAE/200 mL) and green tea 67th (124 mg GAE/100 mL) in the top 100

polyphenol-containing foods (Pérez-Jimenez et al., 2010). The mean total

polyphenol content of yerba mate regular infusions of 135.7 mg GAE/200 mL,

Brazilian iced tea of 147.1 mg GAE/200 mL would place yerba mate infusions

above green tea in this ranking. Although, the bombilla method would provide

3406.4 mg GAE/1L and a serving of this method would place yerba mate above

all other beverages.

119

3.7. Conclusions

This chapter presented the chemical analysis of different samples of yerba

mate. This research provided for the first time a comprehensive investigation of

the elemental composition of non-commercial, non-processed and processed

samples of yerba mate. Furthermore, this study provides an evaluation of the

elemental and polyphenolic levels of commercial samples and different infusions

methods of yerba mate from Brazil and Argentina. Finally, this study also

evaluated the potential contribution of the consumption of yerba mate to the

dietary intake of polyphenols and elemental nutrients. Manganese was

highlighted in this study due to its higher concentration in the yerba mate

samples and prepared infusions. As such, no study has assessed what the

potential intake of Mn would be through the consumption of different yerba mate

infusions in terms of the total dietary intake of this element by the South

American population.

The first investigation was to determine the element composition of non-

commercial yerba mate leaves collected from the Barão de Cotegipe plantation

in southern Brazil - April 2017 (refer to section 3.5). This pilot study was

undertaken at the producer’s request to evaluate whether growing yerba mate

trees within a native forest or in traditional plantations (including the use of

fertiliser or the impact of the age of leaves) would influence the elemental quality

of the yerba products. Overall, in terms of the age of the leaves (from trees

grown in both fertiliser and organic areas) collected from traditional plantations

(refer to section 3.5.3), the general trend was for the elements to be at higher

levels than in the old leaves (new < old; refer to Table 3.2). The next objective

was to evaluate the effect of yerba mate cultivation (with or without the use of

fertilsers) on the elemental levels of leaves (new or old). The new leaves grown

on trees from the organic plantation had higher levels of most of the elements,

especially Mn (a two-tailed Student t-test confirmed significantly higher levels at p

< 0.05) in comparison with the plantations treated with NPK fertilisers (new

leaves; fert < org). Conversely, for the older leaves from both plantations, all of

120

the elements reported in Table 3.2 were higher in the fertilser-addition plantation

(old leaves; fert > org). An assessment of the elemental levels of leaves collected

at different heights of a yerba mate tree showed a variation between the bottom

(0.5 m), middle (1.5 m) and upper parts of the tree (2.5 m). Manganese values

confirmed a degree of translocation to the upper parts of the tree. The elemental

composition of non-commercial yerba mate samples cultivated between natural

forests was compared to samples from traditional plantations (refer to section

3.5.4). In general, higher elemental levels were found in plants grown in the

traditional organic plantations (refer to Table 3.6). In terms of the processing of

the yerba mate, the elements had higher elemental levels after the sapeco stage

(material is exposed to an open fire for a short period of time) of the process

(refer to section 3.5.5). Moreover, the post-drying samples (the last stage of the

processing plant involves the yerba mate material being dried until a moisture

content of 5 to 6 %) had lower levels of the elements when compared with the

sapeco yerba mate samples.

The trace elemental composition of commercial yerba mate products from

Brazil and Argentina was evaluated in terms of the type of sample (green or

roasted), processing package (loose or tea bags) and origin (Brazil or Argentina)

(refer to section 3.6.3). In general, the elemental levels for green loose yerba

mate products obtained from Brazil are slightly higher than that for the

Argentinian products. In summary, the elemental levels are basically similar for

the two countries, which is to be expected as yerba mate primarily grows

between the Parana and Paraguay river basins in South America (refer to Figure

3.1). The next study involved an evaluation of Argentinian commercial yerba

mate products based on the preparation and consumption of infusions using

green loose material or tea bags. All of the elements measured had higher levels

for the tea bag products when compared to the green loose material. This was

not surprising as the tea bags contain only leaf material and have been

processed to produce finer particles (of around 1 mm). It was also noted that the

green tea bags contain mostly leaf material with very few stems compared to the

loose yerba mate samples (refer to Figure 3.5). The total elemental levels of

121

commercial green and roasted yerba mate samples from Brazil were also

evaluated in this study. The roasted (loose and tea bag) samples have higher

elemental levels when compared with the Brazilian green loose material (Table

3.9).

The elemental composition, total polyphenol content, chlorogenic acids,

caffeine and theobromine levels of commercial yerba mate products (from Brazil

and Argentina) prepared as infusions was evaluated in sections 3.6.4 and 3.6.5.

In relation to the country of origin of the yerba mate (Brazil and Argentina), the

regular infusions of yerba mate (green loose) from Brazil had slightly higher

chemical levels. This was in agreement with the reported results for the total

elemental content, presented in section 3.6.3. When comparing the chemical

content of a regular infusion associated with a single serving (cup of tea) using

green loose and tea bags from Argentina, there was a highly significant

difference (a two-tailed Student t-test) between the samples. All of the elements

(Table 3.11), polyphenols and xanthines (section 3.6.5) had higher levels for the

tea bag samples when compared to the green loose regular infusions. The

chemical analysis of regular infusions produced using green loose and roasted

samples from Brazil were evaluated, as presented in Table 3.11 and section

3.6.5. Interestingly, the regular infusions made with green loose yerba mate had

significantly higher levels (a two-tailed Student t-test) of trace elements,

polyphenols and xanthines.

Finally, an evaluation of the potential intake of trace elements and

polyphenols through the consumption of yerba mate products was proposed in

section 3.6.6. All the trace elements analysed in this study represent 0.1 to 5% of

the recommended daily allowance (RDA), for all the infusion methods (regular,

Brazilian and bombilla). An exception was found for manganese. A regular

infusion serving (1 cup of 200 mL) can provide 23.7 to 106.0 % for males and

30.3 to 135.5 % for females of the recommended daily intake of manganese,

depending on the type of yerba mate product (refer to Table 3.22). Although, in

terms of the bombilla method, the daily amount of yerba mate infusion consumed

is close to a volume of 1 litre. This results in Mn levels equating to all of the

122

adequate recommended daily intake allowance and could provide concentrations

higher than the recommended upper consumption limit for this element per day

(refer to Table 3.22). In relation to the total polyphenol intake, a regular infusion

serving (1 cup of 200 mL) could contribute 4 to 14.5 % of the daily intake of total

polyphenols. Furthermore, the bombilla method (resulting in the drinking of about

1 litre of the infusion) can provide up to twice the amount of the adequate daily

intake of total polyphenols (refer to Table 3.23).

Finally, this study provides important new data about the chemical quality

of Brazilian and Argentinian yerba mate production and commercial products. It

is proposed that future research should evaluate the possible chemical

interactions between elements and organic compounds in the infusions, which

may influence the bioavailability of not only the elemental species, but also

provide another way of considering what may be the impact on human health

through consuming yerba mate (refer to chapter 6). Moreover, the data on

manganese levels in yerba mate infusions is worthy of further study especially in

relation to the drinking of other beverages, such as, coffee (refer to chapter 4) or

green and black tea.

123

Chapter 4. Chemical Composition of Roasting Brazilian Coffee

124

4.1. Introduction

Coffee is one of the most popular drinks in the World, with global trading

being worth more than 10 billion US dollars; the second most consumed

beverage after water and annual global consumption being approximately 500

billion cups (Clarke and Vitzthum, 2008, Butt and Sultan, 2011). Coffee is the

second largest commodity traded in the World, after petroleum (Butt and Sultan,

2011). The drink is prepared from the roasted seeds of the coffee plant; the

commercially important species being Coffea arabica L. (Arabica) and Coffea

canephora L. (variety Robusta) (Ludwig et al., 2014). It is often consumed, due to

its stimulatory effects, because of the high caffeine content present in the coffee

beans and related beverages. Furthermore, coffee is one of the largest

contributors, Worldwide, to the total dietary intake of polyphenols (Donnelly,

2015). This chapter presents an investigation of the chemical composition of the

roasting process of Brazilian coffee, using samples of small producers from

Amparo, São Paulo State, Brazil, according to the methodology described in

section 4.3.1. The results are presented in sections 4.4 and 4.5. An evaluation of

the physical changes of the beans during the roasting process is presented in

section 4.6. Finally, a study evaluating the link to dietary intake of elements and

total polyphenol derived from coffee infusions was performed and presented in

section 4.7, with the conclusions in section 4.8.

4.1.1. Coffee production in Brazil

The Coffea plant is native to tropical Africa, although more than 70

countries cultivate this plant, with Brazil, Colombia, Ethiopia and India being the

leading producers (Butt and Sultan, 2011). Brazil is the largest coffee producer,

responsible for a third of all coffee consumed Worldwide and the production is

mainly located in Minas Gerais, São Paulo, Espírito Santo and Bahia States

(Caldarelli et al., 2019). The main Coffea species in Brazil is Coffea arabica L.,

but there are several varieties (sub-species) cultivated through selective breeding

125

or natural selection of coffee plants. Such coffee varieties have different traits,

such as, bean size, yield, plant resistance against pathogens and maturation

stage (Clifford, 1985).

Brazilian coffee plantations are harvested during the dry season from June

to September, usually in one annual crop when the coffee berries are ripe, as

shown in Figure 4.1. In Brazil, due to the abundant sunshine, the fermentation of

the bean occurs when the berries are cleaned and dried under the sun for 8 to 10

days (depending on the weather conditions). The beans are further dried in rotary

dryer machines until a constant moisture content. The outer layer is then

removed, and the beans are selected and left to age for at least 6 months to

develop flavour. The green beans are then ready to be roasted and ground in

order to prepare the beverage. Brazil usually exports green coffee to be roasted

in the destination country.

Figure 4.1: Different maturation stages of the coffee cherry, showing green and

the mature red berries.

4.1.2. Roasting of coffee beans

The characteristic colour and aroma of coffee are produced during the

roasting process. Coffee oils, which accounts for 10% of the roasted beans, are

responsible for the aroma (Buffo and Cardelli‐Freire, 2004). The roasting process

126

can be divided into three stages: (i) drying, where most of the moisture is

removed (endothermic); (ii) roasting, where numerous complex reactions take

place, changing the chemistry of the coffee beans, thereby also releasing a large

amount of carbon dioxide and producing hundreds of volatile compounds

related to the aroma and flavour of the coffee; and (iii) the cooling phase in order

to prevent the burning of the beans, using air as a cooling agent (Clifford, 1985,

Illy and Viani, 1995, Buffo and Cardelli‐Freire, 2004).

4.2. Review of Roasting Coffee in Brazil (Elemental and Polyphenols)

The chemistry of roasted coffee was investigated in the literature due to

the high popularity of the beverage all over the World. Furthermore, the

elemental content of roasted coffee in Brazil was investigated not only due to the

nutritional value, but also for food authenticity (Pohl et al., 2013). A review of

previous studies is presented in Table 4.1. All of these studies were performed

with different samples of roasted coffee. As a consequence, the elemental levels

vary, which could be influenced by different soil conditions, harvest periods,

cultivation and processing methods (Zeiner et al., 2015).

Among the organic compounds, xanthines (such as caffeine) and

antioxidant polyphenols may be responsible for the alleged health benefits

related to the consumption of coffee infusions. In previous studies, the total

content of polyphenols in coffee infusions was reported to be 0.96 – 2.27 g/L

(Lakenbrink et al., 2000). The major polyphenol compounds present in coffee are

the chlorogenic acid group, including caffeoylquinic acids, dicaffeoylquinic acids,

feruloylquinic acids and p-coumaroylquinic acids (Wei and Tanokura, 2015). The

roasting process has an impact on the amount of chlorogenic acids in the roasted

coffee beans. A reduction in the chlorogenic acid content has been reported as 8

to 10% for every 1% reduction in dry matter, resulting in losses of 60 to 70% for

medium roast and 90 to 95% for dark roast coffee. This results in the total

chlorogenic acid content of roasted coffee ranging from 1.8 to 80 mg/kg (Farah et

127

al., 2005, Moon et al., 2009, Ferruzzi, 2010, Crozier et al., 2012, Ludwig et al.,

2014).

Table 4.1: Elemental content of roasted coffee beans for selected elements

reported in the literature. Adapted from Pohl et al. (2013).

Element Concentration (mg/kg) Reference

Mg 750 - 3100

(Martin et al., 1996, Martın et al., 1999, Suseela et al., 2001, Anderson and

Smith, 2002, Anthemidis and Pliatsika, 2005, Amorim Filho et al., 2007,

Grembecka et al., 2007, Santos et al., 2008, Ashu and Chandravanshi, 2011)

Ca 490 - 2200

(Martin et al., 1996, Martın et al., 1999, Anderson and Smith, 2002, Vega-

Carrillo et al., 2002, Anthemidis and Pliatsika, 2005, Amorim Filho et al., 2007,

Grembecka et al., 2007, Tagliaferro et al., 2007, Santos et al., 2008, Suseela

et al., 2001, Ashu and Chandravanshi, 2011)

Mn 6.6 – 320.0

(Martin et al., 1996, Martın et al., 1999, Suseela et al., 2001, Anderson and Smith, 2002, Anthemidis and Pliatsika, 2005, Zaidi et al., 2005, Amorim Filho

et al., 2007, Grembecka et al., 2007, Tagliaferro et al., 2007, Santos et al.,

2008, Ashu and Chandravanshi, 2011)

Fe 12.0 - 617.0

(Martin et al., 1996, Martın et al., 1999, Suseela et al., 2001, Anderson and Smith, 2002, Anthemidis and Pliatsika, 2005, Zaidi et al., 2005, Amorim Filho

et al., 2007, Grembecka et al., 2007, Tagliaferro et al., 2007, Santos et al.,

2008, Ashu and Chandravanshi, 2011)

Cu 0.4 - 30.1

(Martın et al., 1999, Suseela et al., 2001, Anderson and Smith, 2002, Anthemidis and Pliatsika, 2005, Amorim Filho et al., 2007, Grembecka et al.,

2007, Santos et al., 2008, Ashu and Chandravanshi, 2011)

Zn 1.2 – 803.0

(Martın et al., 1999, Suseela et al., 2001, Anderson and Smith, 2002, Vega-

Carrillo et al., 2002, Anthemidis and Pliatsika, 2005, Zaidi et al., 2005, Amorim

Filho et al., 2007, Grembecka et al., 2007, Tagliaferro et al., 2007, Santos et

al., 2008, Ashu and Chandravanshi, 2011)

4.3. Coffee Beans and the Roasting Process at Amparo, São Paulo State, Brazil

An evaluation of the chemical effects of the roasting process for Brazilian

coffee was selected as the main topic of research in this study. Therefore, it was

important to analyse samples from the same location in Brazil and using a

specific roasting method. The samples were collected during two field-trips in

April 2017 and 2018, as outlined in section 4.3.1. The coffee beans were roasted

and ground following the procedure described in sections 4.3.2 and 4.3.3. Finally,

128

coffee infusions were prepared in order to simulate the typical Brazilian

consumption of a cup of coffee, as described in section 4.3.4.

4.3.1. Sample collection and preparation of coffee beans

Green coffee samples were collected from two different plantations at

Amparo, northern São Paulo State, namely, Fazenda Palmares and Fazenda

Flor. The samples collected at Fazenda Palmares were from the varieties and

harvesting dates: (i) Catuaí from the 2017 harvest; (ii) Bourbon Amarelo from the

2017 harvest; and (iii) a blend of the different varieties from the previous harvest

in 2016. Fazenda Flor plantation provided a green coffee sample from the Obatã

variety (2017 harvest). The green coffee beans were produced according to the

Brazilian standard method (refer to section 4.1.1) where the mature coffee

cherries were harvest and fermented under the sun, dried until constant moisture

content in rotary dryers and aged for 6 months. The green coffee samples were

then roasted.

4.3.2. Roasting process

The green coffee bean samples were roasted in a commercial IR-5

Diedrich® Roaster at Fazenda Palmares and the samples were collected every 2

minutes over a total 10 minute period, which is equivalent to a bean colour

spectrum from green to medium and finally dark roast, as shown in Figure 4.2.

The roaster was set at 200 °C and when the green beans are added there is a

slight decrease in temperature. Throughout the roasting process the temperature

quickly returns to the set temperature and the only changes that occur are

related to the air flow through the roaster. After 10 minutes, the coffee was

cooled and the beans were manually selected. This procedure was performed by

a trained plantation worker. The final roasted beans were chosen based on the

commercial value of the beans, that is, the quality of the product based on size,

colour and an absence of any cracked bean defects. Also, the blended bean

129

sample from the Fazenda Palmares plantation was roasted for only 6 minutes to

produce the medium roasted sample and at 10 minutes to produce the dark

roasted sample.

Figure 4.2: Roasting of Brazilian coffee beans from green (time = 0 minutes) and

at 2 minutes intervals until the production of the dark roasted product

(t = 10 minutes).

4.3.3. Grinding roasted coffee beans and particle size

The blended coffee beans from the Fazenda Palmares plantation that

were roasted at 6 and 10 minutes (medium and dark roast), were ground in a

commercial Bunn® Coffee Mill, that grinds the coffee according to the infusion

method, namely: (1) coarse for French press; (2) regular for siphon; (3) electric

perk; (4) drip; (5) fine for Brazilian infusions; and (6) espresso. The roasted bean

particle sizes decrease with the infusion method (1 to 6).

4.3.4. Coffee infusions

The coffee infusions were prepared following the Brazilian traditional

method of consumption which is a percolate method where the hot water (boiled)

is put on the coffee and filtered by a paper filter, as shown in Figure 4.3. The

coffee to water ratio was used as set by the Specialty Coffee Association of

America (SCAA), being a standard cup of coffee. Double distilled deionised water

(DDW) was preferred instead of tap water for all infusions in order to evaluate the

130

chemical contribution from only the coffee. A volume of 215 mL of freshly boiled

DDW was added to 7.00 ± 0.05 g of freshly ground coffee, on a commercial filter

paper supported by a plastic Melitta® percolator, as shown in Figure 4.3. The

samples were left to cool down to room temperature and each final solution was

filtered using a 0.45 μm filter. The resultant solutions were analysed for elemental

and polyphenol levels (refer to sections 4.4 and 4.5).

Figure 4.3: Typical Brazilian coffee infusion method.

4.4. Elemental Levels of Roasted Coffee

The elemental levels found in the roasted coffee beans, collected as

described in section 4.3 and analysed by inductively coupled plasma mass

spectrometry (section 2.3), are reported in Table 4.2 for the Obatã, Table 4.3 for

the Catuaí and Table 4.4 for the Bourbon Amarelo coffee varieties.

131

Table 4.2: Elemental levels (mg/kg, dry weight) of Brazilian coffee beans

sampled at different roasting times (minutes). The samples were of

the Obatã coffee variety collected from the Fazenda Flor plantation



Elemental levels of Obatã (mg/kg, d.w.) Roasting time (min) Ca Mg Mn Fe Cu Zn

0 1025.3 1827.0 30.71 27.88 14.21 10.65

2 1261.9 1925.3 38.77 38.08 13.82 7.04

4 788.5 1733.9 29.46 25.94 13.08 8.00

6 1011.4 1759.9 33.80 24.47 13.21 8.42

8 778.9 1832.2 30.73 30.19 16.36 8.56

10 1016.9 2012.2 30.99 30.41 14.43 6.21



the Catuaí coffee variety collected from the Fazenda Palmares

plantation (Amparo, São Paulo State) and analysed by ICP-MS (refer

to section 2.3). Analysis in duplicate.

Elemental levels of Catuaí (mg/kg, d.w.) Roasting time (min) Ca Mg Mn Fe Cu Zn

0 1776.4 1937.3 27.44 23.04 12.65 6.21

2 1409.3 1971.5 25.00 22.06 13.01 6.88

4 1512.1 1940.5 26.71 22.02 12.17 7.29

6 1324.2 1847.2 24.76 22.11 11.67 11.77

8 1401.9 2009.3 32.34 26.81 13.27 8.83

10 1552.3 2256.9 25.52 27.71 13.61 7.56

132



the Bourbon Amarelo coffee variety collected from the Fazenda

Palmares plantation (Amparo, São Paulo State) and analysed by ICP-

MS (refer to section 2.3). Analysis in duplicate.

Elemental levels of Bourbon Amarelo (mg/kg, d.w.) Roasting time (min) Ca Mg Mn Fe Cu Zn

0 1652.7 1842.3 18.32 24.88 15.23 7.88

2 1428.7 1914.7 25.75 24.69 14.14 7.99

4 1521.8 1924.6 22.93 26.67 17.96 6.59

6 1255.1 1908.0 19.82 27.12 17.87 6.09

8 1517.4 1888.2 23.75 25.45 17.48 5.68

10 1619.9 2141.6 26.60 29.95 16.47 10.6

The elemental content of the different coffee bean varieties (Obatã, Catuaí

and Bourbon Amarelo) fluctuates throughout the roasting period but the final

roasted bean product has either the same or a slightly higher elemental content

(mg/kg, dry weight). This is as would be expected, since all of the green bean

samples (t = 0 minutes) were sun-dried at the plantation and throughout the

roasting process there may be a slight modification in the residual moisture

levels. The final roasted beans (subjected to 200 °C) seem to result in a ‘pre-

concentration’ of the elemental levels, thereby leading to a slightly higher

elemental level for the roasted coffee product. The same observation was found

for the other trace elements, as reported in Appendices 4.2 to 4.4. In summary,

this data clearly shows there is no reduction in the elemental content (in fact a

slight increase) of the coffee beans as a result of the roasting process.

133

Table 4.5: Elemental concentration (mg/kg, dry weight) of roasted Brazilian

coffee beans. The different coffee varieties include those selected for

their quality (section 4.3.2) or as defected beans and were collected

from the Fazenda Palmares and Flor plantations (Amparo, São Paulo

State) and analysed by ICP-MS (refer to section 2.3). Analysis in

duplicate.

Bean variety and method of selection (mg/kg, d.w.) Ca Mg Mn Fe Cu Zn

Obatã selected 1016.9 2012.2 30.99 30.41 14.43 6.21

Obatã defected 1140.2 1698.4 31.14 34.05 14.14 5.95

Catuaí selected 1552.3 2256.9 25.52 27.71 13.61 7.56

Catuaí defected 1627.7 2281.7 23.12 29.00 14.74 7.83

Bourbon Amarelo selected 1619.9 2141.6 26.60 29.95 16.47 10.60

Bourbon Amarelo defected 1580.4 2182.6 24.72 40.26 21.66 9.48

Table 4.5 clearly shows that the selection of the final coffee beans (post

the roasting period of 10 minutes) does not influence the elemental levels of the

coffee beans (selected – based on quality factors as outlined in section 4.3.2).

There are also only small differences in specific elemental levels for the different

coffee bean varieties (which may be related to the fact that beans were collected

from different plantations).

134

4.5. Total Polyphenol and Chlorogenic Acid Levels in Roasted Coffee (infusions)

Table 4.6 reports the total polyphenol levels (mg/L) in infusions produced

from the different varieties of coffee beans (Obatã, Catuaí and Bourbon Amarelo)

collected from the Fazenda Palmares and Flor plantations (Amparo, São Paulo

State). The data are reported as a function of the different roasting times (0 to 10

minutes). The total polyphenol levels were determined using the Folin-Ciocalteu

assay (refer to section 2.4).

Table 4.6: Total polyphenol content (mg/L) of Brazilian coffee infusions during

roasting time (minutes). The samples were from the different coffee

varieties from Fazenda Palmares and Flor (Amparo, São Paulo State)

and analysed by UV-Vis (refer to section 2.4). n = 1.

Roasting time (min) Total polyphenol of the coffee variety (mg/L)

Obatã Catuaí Bourbon Amarelo 0 661 859 712

2 764 967 794

4 1134 947 1173

6 962 1084 973

8 1038 1119 912

10 1004 916 963

In general, the effect of roasting the green coffee beans increases the total

polyphenol content until about 4 to 6 minutes (roasting time). Then there is a

slight reduction (depending on the variety of coffee bean) to a final roasted bean

level that is ~ 7% (Catuaí) to 52 % (Obatã) higher than the levels in the green

beans (t = 0 minutes). A possible explanation for this was investigated using

scanning electron microscopy (SEM), as presented in section 4.6, below.

The chlorogenic acids and caffeine levels were also measured using ultra-

high performance liquid chromatography (UHPLC), in the coffee bean samples

135

collected throughout the roasting process (section 2.5). Figure 4.4 reports the

levels of 3-, 4-, or 5- caffeoylquinic acid and caffeine (mg/L) in infusions prepared

from coffee at the different stages of roasting, namely, green (t = 0 minutes),

medium (t = 6 min) and dark roasted (t = 10 min). The samples were a blend of

the coffee varieties collected from the Fazenda Palmares plantation (Amparo,

São Paulo State). The overall trend is similar to that found for the total

polyphenol content (Table 4.6) where there is an increase in the chlorogenic acid

and caffeine levels of the infusions prepared using the freshly ground medium

roast coffee. Furthermore, the dark roast product showed lower levels than the

green coffee for the chlorogenic acids and caffeine. A further evaluation of the

roasting process with selected coffee varieties and time intervals was also

undertaken.

Figure 4.4: Concentration of chlorogenic acids and caffeine (mg/L) of Brazilian


process (green t = 0 minutes, medium t = 6 min and dark roasted t =

10 min). The samples were the blend of the coffee varieties collected

from the Fazenda Palmares plantation (Amparo, São Paulo State)

and analysed by UHPLC (refer to section 2.5). n = 1 due to the

limited amount of available sample.

Figures 4.5, 4.6 and 4.7 report the levels of chlorogenic acids and caffeine

(mg/L) for the coffee bean samples collected at 2 minute intervals throughout the

Green coffee Light roast Dark roast0

100

200

300

400

500

Roasting colour

Co

ncen

trati

on

(m

g/L

)

3- caffeoylquinic acid



caffeine

136

roasting process. All samples were analysed using UHPLC (refer to section 2.5).

The samples were the Obatã (collected from Fazenda Flor and presented in

Figure 4.5), Catuaí and Bourbon Amarelo coffee varieties (collected from

Fazenda Palmares and presented in Figures 4.6 and 4.7, respectively). The

roasting process did not present a significant effect on the caffeine levels.

Interestingly, the chlorogenic acids follow the same trend as outlined for the total

polyphenol levels (Table 4.6 and Figure 4.4), with the levels being higher in the

samples taken during the middle of the roasting process. This is more

predominant for the levels of 5-caffeoylquinic acid, but 3- and 4- caffeoylquinic

acids also show a similar trend at 8 minutes of the roasting time (but of less

magnitude). This study contradicts what has been reported in the literature which

mainly describes the loss of chlorogenic acids in coffee during the roasting

process (Trugo et al., 1985, Bennat et al., 1994, Schrader et al., 1996). It has

been suggested that this may be due to the breakage of the carbon-carbon

bonds with the high roasting temperatures, causing isomerisation (refer to

glossary) or degradation of the compounds (Farah et al., 2005). The peaks

associated with the 3- and 4 - caffeoylquinic acids have already been reported for

the medium roast conditions and it has been suggested that this is due to a

partial hydrolysis of dicaffeoylquinic acids and the isomerisation of the 5-

caffeoylquinic acid, which implies a decrease in the amount of this compound

(Trugo et al., 1985, Farah et al., 2005). An increase in the levels of the total

caffeoylquinic acids after the beginning of the roasting process has also been

reported as a consequence of ‘pre-concentration’ during the roasting process

(loss of moisture content and volatile compounds) (Fujioka, 2006). The difference

between this study and the literature is that this research involved the analysis of

coffee infusions in order to simulate the chemical intake of a cup of coffee,

whereas other studies analysed the coffee bean extractions undertaken using

organic solvents (Donnelly, 2015). This study also considered the physical effect

and the pore size associated with the coffee roasting process, as discussed in

section 4.6.

137

Figure 4.5: The concentration of chlorogenic acids and caffeine (mg/L) of

Brazilian coffee infusions produced from beans sampled during the

roasting process (t = 0 to 10 min). The samples were of the Obatã

coffee variety collected from Fazenda Flor plantation (Amparo, São

Paulo State) and analysed by UHPLC (refer to section 2.5). n = 1.



roasting process (t = 0 to 10 min). The samples were of the Catuaí

coffee variety collected from the Fazenda Palmares plantation

(Amparo, São Paulo State) and analysed by UHPLC (refer to


0 2 4 6 8 100

500

1000

1500

Roasting time (min)

Co

ncen

trati

on

(m

g/L

)




caffeine

0 2 4 6 8 100

500

1000

1500

Roasting time (min)

Co

ncen

trati

on

(m

g/L

)




caffeine

138



roasting process (t = 0 to 10 min). The samples were of the

Bourbon Amarelo coffee variety collected from the Fazenda

Palmares plantation (Amparo, São Paulo State) and analysed by

UHPLC (refer to section 2.5). n = 1.

The effect of manually selecting the final coffee beans as an estimate of

product quality (roasting time of 10 minutes, refer to section 4.3.2) was also

evaluated for the polyphenol and caffeine content, as shown in Table 4.7.

Interestingly, the total polyphenol levels decreased in the infusions prepared with

the defected beans. In relation to the levels of chlorogenic acids and caffeine,

there is no particular effect associated with the selection of the beans (based on

quality) after the roasting process.

0 2 4 6 8 100

500

1000

1500

Roasting time (min)

Co

ncen

trati

on

(m

g/L

)




caffeine

139

Table 4.7: The concentration of chlorogenic acids, caffeine and total polyphenol

(mg/L) of roasted Brazilian coffee infusions. The samples were

prepared from manually selected and defected beans (section 4.3.2)

from different coffee varieties collected from the Fazenda Palmares

and Flor plantations (Amparo- São Paulo) and analysed by UHPLC

(refer to section 2.5). n = 1.

Coffee varieties (mg/L) Total polyphenol

3-caffeolquinic acid

4-caffeolquinic acid

5-caffeolquinic acid Caffeine

Obatã selected 1004 189 226 424 331

Obatã defected 675 90 111 746 235

Catuaí selected 916 59 58 89 269

Catuaí defected 749 74 74 104 264

Bourbon Amarelo selected 963 142 165 296 265

Bourbon Amarelo defected 765 151 170 277 247

The effect of the roasted coffee particle size on the concentration of

chlorogenic acids and caffeine was investigated, as outlined in section 4.3.3.

Figure 4.8 reports the levels of those compounds measured in samples

associated with the medium roast period (t = 6 minutes) and Figure 4.9 for the

dark roast period (t = 10 minutes). It is clear that the roasted coffee particle size

has an influence on the extraction of the analysed chemical compounds. The

efficiency of the extraction is indirectly proportional to the particle size of the

roasted bean product used in the infusions. Interestingly, the behavior of the

efficiency trend changes with the roasted product, being an exponential curve for

the medium roast infusions. A similar, but more linear trend was found for the

dark roast infusions. This could be explained by the physical differences in the

coffee pores produced in the beans during the roasting process, as described in

section 4.6.

140


Brazilian coffee infusions as a function of the different bean particle

sizes according to the method of infusion: (1) coarse for French

press; (2) regular for siphon; (3) electric perk; (4) drip; (5) fine for

Brazilian infusions; and (6) espresso. The samples were collected

at the medium roast time of the process (t = 6 minutes), being a

blend of the coffee varieties sampled from the Fazenda Palmares

plantation (Amparo- São Paulo) and analysed by UHPLC (refer to


0 1 2 3 4 5 60

200

400

600

Particle size

Co

ncen

trati

on

(m

g/L

)




Caffeine

141


Brazilian coffee infusions as a function of the different bean particle

sizes according to the method of infusion: (1) coarse for French

press; (2) regular for siphon; (3) electric perk; (4) drip; (5) fine for

Brazilian infusions; and (6) espresso. The samples were collected at

the dark roast time of the process (t = 10 minutes), being a blend of

the coffee varieties from the Fazenda Palmares plantation (Amparo-

São Paulo) and analysed by UHPLC (refer to section 2.5). n = 1.

4.6. Effect of Pore Size of Ground Roasted Coffee

It has been reported that during the roasting process of coffee beans there

is a loss of water. Moreover, there is a release of gases associated with a high

internal pressure within the bean that changes the volume and porosity of the cell

walls (Schenker et al., 2000). The structure of the coffee beans collected during

the roasting process was evaluated through a series of scanning electron

microscope (SEM) images. The SEM measurements were performed on a Jeol®

JSM-7100F. Figure 4.10 (A) to (F) shows the changes in the physical structure of

the beans, (a blend of the coffee varieties collected from the Fazenda Palmares

0 1 2 3 4 5 60

100

200

300

Particle size

Co

ncen

trati

on

(m

g/L

)




Caffeine

142

plantation, Amparo, São Paulo State). The samples were collected at 2 minute

roasting intervals and range from: (A) 0 to (F) 10 minutes. It is clear from the

images that the structure of the coffee bean changes during the roasting process

and the pores become larger. During the roasting, there is an expansion of the

coffee bean and micropores, and therefore a decrease in the coffee density

(Jokanovic et al., 2012). The change in the porosity of the beans may have an

impact on the efficiency of the chemical extraction in the coffee infusions, as

reported in sections 4.4 and 4.5. It has been suggested that the presence of fine

micropores associated with roasting-induced changes of the coffee bean, can

allow the mobilised coffee oil to migrate to the bean surface (Schenker et al.,

2000). Moreover, the volume increase of the bean and development of pores

during roasting are known to be highly dependent on the roasting conditions

(Ortolá et al., 1998). What has not been reported until this research study is the

effect this has on the levels of chemicals in the coffee beans that are available in

the associated infusions.

(A) t = 0 min

143

(B) t = 2 min

(C) t = 4 min

144

(D) t = 6 min

(E) t = 8 min

145

(F) t = 10 min

Figure 4.10 (A) to (F): Scanning electron microscope images of Brazilian coffee

beans sampled during the roasting process (t = 0 to 10

min). The samples were the blend of the coffee varieties

collected from the Fazenda Palmares plantation (Amparo,

São Paulo State).

4.7. Chemical Composition of Roasted Coffee Infusions and Human Dietary Intake

This section evaluates the levels of total polyphenols in the prepared

roasted coffee infusions and the effect on human dietary intake. Since there are

no regulatory values for chlorogenic acids, these chemicals were not included in

the study. Moreover, no data is available for the elemental levels of the roasted

coffee infusions as the diluted solutions caused a significant reduction in the ICP-

MS instrument performance during sample analyses. This was associated with

the high dissolved solids content and ‘colour chemicals’ blocking the sample

injector (of the ICP) and the interface cones and skimmers (refer to section 2.3).

146

The average annual consumption of coffee by the Brazilian population

was 839 cups (each of 40 mL) per person in 2018 (ABIC, 2018). This suggests

that the daily Brazilian consumption is an average of 92 mL of coffee per day.

The calculated daily consumption of 1492 mg (fresh weight) of polyphenols,

based on a balanced Japanese diet was used to evaluate the potential intake

from coffee of these compounds to the Brazilian diet (Fukushima et al., 2009).

Table 4.8 compares the different roasting times of the coffee beans used to make

infusions in relation to the daily intake of total polyphenols. An average daily

consumption of coffee (92 mL) could contribute 4 to 7 % of the daily intake of

total polyphenols. This range is a higher contribution when compared to the

corresponding values for the consumption of yerba mate in Brazil (refer to

section 3.6.6.2), green or black tea, and fruit (grape, apple and orange) juices

(Donnelly, 2015).

Table 4.8: Percentage intake (%) of total polyphenol based on the Brazilian daily

consumption (92 mL) of coffee infusion. The data is compared with

the values reported by Fukushima et al. (2009) for the daily intake of

total polyphenols.

Roasting time (min) Daily percentage intake of total polyphenols (%)

Obatã Catuaí Bourbon Amarelo

0 4.1 5.3 4.4

2 4.7 6.0 4.9

4 7.0 5.8 7.2

6 5.9 6.7 6.0

8 6.4 6.9 5.6

10 6.2 5.6 5.9

4.8. Summary

Coffee is one of the most popular beverages that is consumed all over the

World. Brazil is the largest producer and exporter of green coffee beans. At

147

present the roasting of green coffee beans is only done on a small scale in local

regions of Brazil, and other countries import the beans and produce their own

commercial roasted coffee products. In Brazil, little is known about the impact of

the coffee roasting process on the chemical composition of the roasted beans

and related beverages. A series of studies on the roasting process and effect on

the elemental and polyphenol content was undertaken in this chapter. The effect

of roasting different coffee varieties (Obatã, Catuaí, Bourbon Amarelo and a

blend) collected from the Fazenda Palmares and Flor plantations (Amparo, São

Paulo State) resulted in a slightly increase in the elemental content during the

roasting process (Tables 4.2, 4.3 and 4.4). This data suggests that there is a

small decrease in the moisture content (‘pre-concentration’) of the beans and

highlights that no elemental losses were found during the roasting process. Any

measured elemental variation, especially for Ca, Mg, Cu and Zn, may be linked

to the soil chemistry or growing features of a specific plantation site from which

the green coffee beans were collected (section 4.4).

The total polyphenol content of the coffee infusions produced from beans

sampled during the roasting time, was 7 to 52 % higher for dark roast (10

minutes) when compared to the levels in infusions produced from green beans (t

= 0 minutes) (Table 4.6). The chlorogenic acid and caffeine analysis showed a

similar trend where there was an increase in the chemical levels of the infusions

prepared using the medium roast coffee (refer to section 4.5; Figures 4.4 to 4.7).

The effect of the roasted coffee particle sizes on the concentration of chlorogenic

acids and caffeine in the infusions was investigated (section 4.5). A strong

inversely proportional relationship existed between the roasted bean particle

sizes and the chemical concentrations of the resultant infusions (Figures 4.8 and

4.9). A manual selection of the good or defected beans, post the roasting

process of 10 minutes (refer to section 4.3.2), did not result in any major

differences in the elemental or chlorogenic acid levels of the roasted coffee

product (Tables 4.5 and 4.7). Although, the selection of higher quality beans

(refer to section 4.2) did result in higher total polyphenol levels for the resultant

coffee infusions. Any possible physical structure changes of the coffee beans

148

were investigated and the scanning electron microscopy (SEM) images showed

that during the roasting process there is an expansion of the coffee bean

resulting in the pores becoming larger (Figures 4.10 (A) to (F)). Finally, an

evaluation of the potential intake of total polyphenols from these coffee infusions

showed that based on the average daily consumption of roasted coffee, as an

infusion, in Brazil (92 mL), would lead to a 4 to 7 % estimated daily intake of total

polyphenols. This means that the daily consumption of coffee as an infusion in

Brazil would lead to an average of 92.96 mg/92 mL intake of total polyphenols

(medium roast, t = 6 min) that is higher than yerba mate, green and black tea.

The total polyphenol content of the Coffea canephora L. produced in Vietnam

was reported to be slightly higher than the Coffea arabica L. from Brazil and

roasted in Europe or USA (Hečimovic et al., 2011).

In recent years another Brazilian fruit has gained international popularity

as a ‘super-fruit’, namely, açaí (Euterpe oleracea), which will be evaluated in

terms of the chemical composition, so it can be evaluated against yerba mate

(chapter 3) and coffee (chapter 4) as a daily dietary sources of these chemicals

for the Brazilian population.

149

Chapter 5. Brazilian Açaí

150

5.1. Introduction

The açaí berry, a native fruit from the Amazonian region of Brazil, has

recently become very popular due to its status as a ‘super-fruit’ (refer to

glossary). In this research, an investigation of the relationship between the

chemical composition and biological activity of açaí samples was performed in

February 2018, using non-commercial (Amazon) and commercial açaí samples

obtained from outlets in Brazil (São Paulo) and the United Kingdom (UK). All

samples were analysed for the polyphenol and elemental content using the

methodologies outlined in section 5.5. The results are also presented in section

5.7. Furthermore, an evaluation of the impact that Amazonian geographical

variability and commercial processing has on the chemical composition of açaí

was carried–out in April 2018. Samples were collected along the Amazonas river

delta, located in the Pará State, Brazil, according to the methodology described

in section 5.6. The results are presented in section 5.8.

5.2. General Introduction to Açaí Berries from the Amazon Region, Brazil

Natural occurrence

The açaí palm (Euterpe genus) is a native tree from the Amazon region,

northern South America, which has 28 known species, though only two, namely

Euterpe oleracea and Euterpe precatoria. (Yamaguchi et al., 2015) are used for

commercial products. One of the differences between the species, besides the

size and format of the tree and leaves, is where they are grown (refer Figure 5.1).

In general, E. oleracea is a native palm from the Amazonas river estuary, usually

found in flooded forest areas alongside the river (Lee et al., 1998). This species

is the most commercially valuable due to the consumer response regarding

product taste or colour (Schauss et al., 2006a).

151

Figure 5.1: (A) Map of South America with Amazon forest in green (Fao, 2015);

and (B) Natural botanical distribution of two different species of açaí,

namely, Euterpe precatoria and Euterpe oleracea. Adapted from

Yamaguchi et al. (2015).

Alternatively, E. precatoria is prevalent in the Amazonas river basin, which

is closer to the Equatorial line (Pacheco-Palencia et al., 2009). The palm tree of

both species can grow up to 25 meters in height (Lee and Balick, 2008) and its

fruit is a black-purple berry, measuring from 0.9 to 1.3 cm in diameter. The seed

accounts for 80 to 95% (in volume) of the fruit and the edible purple mesocarp is

only 1 to 2 mm in thickness (Pompeu et al., 2009), as shown in Figure 5.2. Even

though the dark purple berry is the most common açaí variety, there are also

other naturally occurring varieties (Oliveira et al., 2002), such as the white açaí

berry, where the ripe fruit is ‘greenish’ in colour after maturation (Rogez, 2000, da

Silveira et al., 2017).

152

Figure 5.2: (A) The natural occurrence of açaí (E. oleracea) in the flooded forest

near Belém, Para State, Brazil; (B) the purple açaí fruit (pulp and

seed separated); and (C) the white açaí (‘greenish’) berries. Adapted

from Potsch (2010).

The açaí berry is a key cultural symbol of the Amazonian region and has

often been referred to as “black gold” since pre-colonisation ages (Andrade,

2014). For many years açaí has been an important dietary source of nutrients for

both urban and rural Amazonian communities, contributing up to 43% of their

dietary dry weight basis and 30% of their energy intake (Heinrich et al., 2011). It

is normally consumed as a side dish for fish, prawns and tapioca (manioc flour).

Interestingly, not only are the berries consumed but other parts of the palm tree

are used for medicinal purposes, namely, the roots, ‘heart of the palm’, leaves

and seeds (Yamaguchi et al., 2015).

Brazilian açaí production and products

The production of açaí, which is an extractive activity from the floodplain of

the Amazon forest, has been constantly growing in the past 20 years, due to its

increased popularity for human consumption, and the manufacture of cosmetic

products (Homma et al., 2006, Maciel et al., 2018, Tagore et al., 2018).

Nowadays, açaí is not only consumed by locals of the northern regions of Brazil,

but it has gained popularity in South America and internationally due to its

153

classification as a ‘super-fruit’ (Schauss, 2013). Brazil is the main producer and

exporter of açaí, generating an estimated monetary source of 9 billion US dollars

per year (IBGE, 2017), exceeding the price per ton of soybeans and Brazil nuts

(Yamaguchi et al., 2015). Açaí is considered to be the most important export

product of the Amazon estuary. It is estimated that 200,000 tonnes of açaí are

extracted per year, and the industry is projected to grow by more than 12% over

the period from 2017 to 2025 (Research, 2018). The demand for açaí has been

increasing in southern Brazil and it is usually linked with the younger population

due to its appeal as an energetic, health and nutritional product (Rogez, 2000).

As a result of its increasing value in the market and the low processing yield (due

to the mass of the seed per berry), the processing of açaí berries can lead to

adultered products, especially for economic gain. The palm is also a source of

‘palm hearts’ (the inner core and growing bud of the tree) for the food industry,

where Brazil is the major exporter of this highly valued product. However, the

harvesting of this food source is also a threat to the production of the açaí berry,

since the extraction of the palm hearts can potentially damage the entire tree

(Jardim, 2002).

The present demand for açaí is mainly for the food industry, but it is

expected that other activities, such as the development of nutraceuticals (refer to

glossary), cosmetics and personal care products, will increase significantly in the

near future (Research, 2018). Recent studies reporting the anti-aging properties

of the berries and açaí extracts or oils have led to the increased use of these

materials in cosmetics, such as, anti-wrinkle and body hydrating creams or

products that prevent cutaneous disorders (Herculano, 2013). According to

published details Córdova-Fraga et al. (2004), açaí has been reported to have

potential as a contrast agent for the magnetic resonance examination of the

gastrointestinal tract (primarily linked to the non-toxicity properties of the berries).

The harvesting of açaí berries is still manually undertaken by local people

that go inside the forest, climb the trees and bring the berries to the city by boat

to sell (as shown in Figure 5.3). This harvesting activity, coupled with the natural

occurrence of açaí being in the flooded areas of the delta, restricts the production

154

of the berry to the dry season of the Amazon region, namely, from June to

November (Rogez, 2000). After harvesting, the açaí berries have to be

processed within 24 hours in order to prevent oxidation of the berries. The fruits

are transferred to a processing plant where they are selected according to the

maturation state, size and then washed twice with water in order to remove any

other material (twigs and leaves). Then, the fruits are usually left in a warm water

tank to soften the pulp. The berries are transferred to a specific machine

(despolpadeira PT) as shown in Figure 5.3. This acts like a mild blender that

removes the softened pulp from the seeds with the assistance of a certain

amount of added water. The final product is then packed, ready to be frozen and

consumed mainly by the beverage industry as frozen pulp, or as a smoothie and

as juice. The seeds are treated as a by-product of the industry, or used as

biomass or as a natural fertiliser in the plantation (Schauss et al., 2006b,

Pacheco-Palencia et al., 2008, Pacheco-Palencia et al., 2009, Schauss, 2013).

Açaí pulp can also be dehydrated using the process of lyophilisation of the

product (Carneiro et al., 2015). The resultant açaí powder is used for many

applications, such as, food supplements or additives, and as a natural colour

agent (Bobbio et al., 2000).

The processing of the açaí berries may also include a decontamination

step, which involves an additional washing step with an ozonated or hypochlorite

solution, or pasteurisation of the final product. The latter is the preferred method

for the international market. This is undertaken by changing the temperature

which leads to the oxidation of the pulp, thereby changing the colour and taste of

the final product. This step became important after some reported cases in the

Amazon region relating to the consumption of açaí berries with Chagas disease,

a tropical parasitic condition. This disease is transmitted by a protozoa parasite

(Trypanosoma cruzi), present in the faeces of the insect barbeiro (Tiatoma

brasiliensis) (Hamilton et al., 2012).

155

Figure 5.3: (A) Harvesting of açaí by locals in the Amazon; and (B) a

‘despolpadeira’, machine used to mechanically extract the pulp of

the açaí berries. Adapted from Vida (2010).

The amount of added water is determined by the market requirements and

is classified according to a Brazilian regulation, based on the solid content of the

final product: (i) thin or popular açaí from 8 to 11% of total solids; (ii) regular or

medium açaí from 11 to 14% of total solids; and (iii) special or thick açaí with

more than 14% of total solids (Rogez, 2000, Homma et al., 2006). Furthermore,

due to the limited regulation control of the açaí processing, each açaí company

has their own method of processing the berries. This may be based on a

separate cleaning stage, where the temperature of the water is changed, or the

acidity of the final product is modified by adding citric acid, or further

pasteurisation is undertaken, as presented in Figure 5.11. Nowadays, with the

increasing demand of açaí, there is a need for quality standards to be used in the

production of the fruit, especially for pulp that is targeted for the international

markets (Pagliarussi, 2010).

Health effects of açaí consumption

Blackcurrants, blueberries and açaí berries are frequently classified as

‘super-fruits’ that can help to improve or maintain health (Esposito et al., 2014,

156

Timmers et al., 2017, Skates et al., 2018). The increasing use of the açaí berry

as an energetic and antioxidant drink has led to an expansion in the amount of

scientific research. Furthermore, the consumption of açaí pulp has been linked to

offering various health benefits, such as anti-inflammatory (Schauss et al.,

2006a, Jensen et al., 2008), cardio protective (Rocha et al., 2007, de Souza et

al., 2010), anti-tumor (Hogan et al., 2010), a weight loss agent (Marcason, 2009),

antioxidant (Mertens-Talcott et al., 2008) and also has anti-proliferate (refer to

glossary) properties (in relation to bacteria) (Ribeiro et al., 2010). As stated

above, other parts of the palm tree are also used as traditional medicines in Latin

America. This has been linked with relieving pains, acting as an anti-diarrheal

(Galotta and Boaventura, 2005) or anti-malarial agent (Ruiz et al., 2011).

Various studies have been reported on the chemical composition of the

bioactive molecules of açaí, and related biological activities (Heinrich et al., 2011,

Yamaguchi et al., 2015). The antioxidant activity of açaí has been studied in

cellular models and in-vivo studies through the use of different assays, such as,

scavenging free radicals (Hassimotto et al., 2005, Lichtenthäler et al., 2005b,

Schauss et al., 2006a, Rufino et al., 2011, Kang et al., 2012) and the inhibition of

oxidation in cell cultures (Matheus et al., 2003, Matheus et al., 2006). Although,

the pulp has been shown to have a wide antioxidant activity, the phenolic

compounds were found to not be correlated with this property. This implies that

other compounds, still to be identified, may also contribute to the antioxidant

properties of the açaí berries (Yamaguchi et al., 2015). Also, there have been

some biological assays that were used to investigate the anti-inflammatory

effects of the berries, even though the mechanisms are still unknown (Poulose et

al., 2012, Xie et al., 2012). When compared to other Amazonian fruits, the açaí

berry has been shown to exhibit the best retention capacity for free radicals

(Canuto et al., 2010). Furthermore, one study on açaí has led to a proposal that

açaí extracts have a positive effect on different cellular models, such as, on the

brain cells of rats. This may be due to chemicals that can lead to protection

against neurodegenerative diseases (Spada et al., 2009, Poulose et al., 2012).

Another study, using the vascular cells of rats, found a reduction in the risk of

157

cardiovascular disease (Rocha et al., 2007). It has also been reported that açaí

juice may have an athero-protective effect, that is, it aids in protecting an

organism against atherosclerosis and regulates inflammation (Xie et al., 2011).

Moreover, a research study found that feeding rats with açaí juice resulted in a

balance of levels of high and non–high density cholesterol (de Souza et al.,

2010). Açaí has also been found to be an anti-tumoral, anti-genotoxical and anti-

proliferate agent due to the protection against deoxyribonucleic acid (DNA)

damage and the prevention of the formation of reactive species (Hogan et al.,

2010, Ribeiro et al., 2010, Fragoso et al., 2012). Human consumption trials have

also confirmed the positive effects of an açaí diet through an increase of the

antioxidant effects on human plasma (Jensen et al., 2008, Mertens-Talcott et al.,

2008).

5.3. Chemical Composition of Açaí

Açaí berries are considered to be a functional food due to being an

important source of fibers, anthocyanins, minerals, energy, fatty acids and

vitamin E (Pacheco-Palencia et al., 2008, Vera de Rosso et al., 2008, Darnet et

al., 2011, Yuyama et al., 2011). The main composition of açaí is 50% lipids, 25%

fibers and 10 % of proteins (on a dry mass basis). This represents an important

source of the nutritional components for the human diet (Unicamp, 2011). The

health benefits and the high biological activity associated with açaí are mainly

due to the chemical composition of the açaí berries and the presence of bioactive

compounds, such as, polyphenols and minerals (including elements)

(Schreckinger et al., 2010). The presence of these biomolecules has already

been correlated to the high biological activity of the açaí berries (Kuskoski et al.,

2006, Menezes et al., 2008b).

Anthocyanins, which is a class of flavonoids, are responsible for the

purple, red and blue pigments of the açaí berries, red wine or blueberries (Grace

et al., 2014). The levels of anthocyanin vary depending on the growing

conditions, origin, temperature exposure, seasonality and maturation (Timmers et

158

al., 2017, Rogez et al., 2011). A study on the kinetics associated with the

accumulation of anthocyanin during the maturation of the açaí berry was

performed and demonstrated a relationship between the anthocyanin content of

the berries and the time of harvest (Rogez et al., 2011). Although the amount of

the anthocyanin may change between samples, the chromatographic profiles

found for açaí berries are similar when determined by high performance liquid

chromatography (HPLC). The most predominant anthocyanin compounds found

in açaí berries are cyanidin-3-glucoside and cyanidin-3-rutinoside and their

chemical structures, are presented in Figure 5.4. Other analytical methods have

been used to identify and quantify the other anthocyanin compounds in açaí

berries, such as, ultra-high performance liquid chromatography – photodiode

array (UHPLC-PDA), which has enabled the separation of peonidin-3-glucoside,

pelargonidin-3-glucoside and peonidin-3-rutinoside, and the subsequent

detection of cyanidin-di-O-gly-cosides (Dias et al., 2012).

Figure 5.4: Chemical structure of most predominant anthocyanin compounds

found in açaí berries being (A) cyanidin-3-glucoside and (B)

cyanidin-3-rutinoside (Yamaguchi et al., 2015).

Several studies have evaluated the polyphenol profile of açaí berries by

HPLC and mass spectrometry, and agreed on the presence of ferulic acid, p-

hydroxybenzoic, gallic, protocatechuic, ellagic, vanillic, p-coumaric acids and

ellagic acid glycoside (Del Pozo-Insfran et al., 2004, Gallori et al., 2004,

Lichtenthäler et al., 2005a, Ribeiro et al., 2010, Rojano et al., 2011, Gordon et

159

al., 2012). The lipid content of açaí pulp accounts for 70 to 90% of the total

calories (fresh weight), where the fatty acids, namely, linoleic, oleic and palmitic

acid, were found to be the major poly- and mono-unsaturated fatty acids present

in the samples (Schauss et al., 2006a, Schauss et al., 2006b). The high

concentration of these fatty acids reinforces the impact of açaí in reducing

cholesterol, preventing cardiovascular diseases and suggests that the fruit is a

rich source of essential fatty acids for the Amazonian diet (de Lima et al., 2000,

Yuyama et al., 2011).

Açaí is also a source of minerals, such as manganese, iron, zinc,

phosphorous, sodium, copper, calcium, magnesium, potassium, nickel, boron

and chromium (Rogez, 2000, de Souza et al., 2010, Maria do Socorro et al.,

2010, Rufino et al., 2010, Costa et al., 2013). The term minerals is used in the

field of nutrition and includes elements (major, minor and trace). Unfortunately,

there have only been a few studies that have reported the elemental composition

of açaí, often focussing on the ‘chemical fingerprinting’ of the açaí samples,

namely, preventing adulteration and tracing the possible origins of the berries

(Santos et al., 2014b). However, there is still a gap in the knowledge on how the

consumption of açaí can affect the human nutritional intake of minerals. A

summary of the reported elemental levels in açaí berries is presented in Table

5.1.

160

Table 5.1: Literature review of the elemental content of açaí according to weight

basis, dry weight (d.w.) or fresh weight (f.w.) and sample type.

Reference

Rogez, 2000

Menezes et al., 2008a

Unicamp, 2011

Yuyama et al., 2011

Llorent-Martínez et

al., 2013

Santos et al., 2014a

Moreda-Piñeiro et al., 2018

Weight basis d.w. d.w. - f.w. f.w. d.w. d.w.

Sample type juice pulp frozen pulp juice juice - supplement

Element Concentration (mg/kg)

Ca 3090 3300 350 159.9 –

578.5 230 4800 80.3

P 1470 545 160 - 180 1400 70.5

Mg 1780 1244 170 - 80 1400 62.5

K 9900 9000 1240 737.8 –

3766.9 1080 7400 460

Na 760 2850 50 2.7 –

139.2 800 - -

Zn 17.3 28.2 3 1634.3 –

5853.7 0.2 10.1 -

B 15.84 - - - - - -

Fe 20.59 45 4 4.6 –

11.16 8 - -

Se 13.21 <0.02 - - - - -

Mn 323 107.1 61.6 - 4 34.3 -

Cu 13.76 21.5 1.8 - 0.1 20.4 <0.58

Ni 2.03 2.8 - - 0.05 - 0.37

Cr 5.31 - - 229.0 –

1485.3 - - -

Cd 0.46 <0.02 - - - - -

Pb 0.408 0.14 - - - - -

Sr 44.66 7.9 - - - - -

161

It has been reported that the açaí berry (freeze-dried pulp) is a good

source of K, Ca, Mg, Fe and Mn (Menezes et al., 2008a) and açaí-based health

supplements (Moreda-Piñeiro et al., 2018). Other studies have also reported the

levels of potential toxic elements (As, Cd, Hg, Pb, Sn and Tl) due to food safety

concerns (Llorent-Martínez et al., 2013). There are many inconsistences in the

reported results possibly due to the genetic variability of the samples analysed

(Menezes et al., 2008a), seasonal variation (Timmers et al., 2017) or the

conditions of industrial processing (Correia et al., 2017). Also, the reported

sampling of açaí berries in these studies did not show any systematic approach

in relation to the source of berries, the species, harvesting time or processing

methods (refer to Table 5.1).


The overall aim of this work was to investigate the chemical

characterisation, antioxidant and biological activities of different samples of açaí.

Due to the Worldwide increase in consumption of açaí, this study was designed

to investigate the potential contribution of açaí pulp to the human dietary intake of

polyphenols and elemental nutrients. This information is especially important for

the Amazonian population that heavily relies on açaí as part of their diet, culture

and economy.

The objectives were to:

(I) provide an extensive literature review of the reported chemical values of açaí;

(II) source a wide range of açaí samples from different geographical locations including the Amazon region of Brazil; varieties; processing methods; and commercial locations;

(III) investigate the anthocyanin profile, quantification of the total anthocyanin and proanthocyanidin content and antioxidant activities in non-commercial and commercial açaí samples;

162

(IV) determine the total polyphenol and elemental content of açaí (non-commercial and commercial) samples;

(V) evaluate the difference in the chemical composition of non-commercial and commercial açaí samples;

(VI) characterise the chemical composition of different varieties of açaí and açaí products;

(VII) determine the cell viability and antioxidant capacities of açaí extractions using macrophage cells;

(VIII) investigate the wound healing capacity of açaí extractions on human fibroblast cells; and to

(IX) assess the impact of the consumption of açaí in terms of the dietary intake of polyphenols and elements.

5.5. Investigation of the Relationship between the Chemical Composition and Biological Activity of Açaí Samples

Introduction

This section provides a description of the methods used in relation to the

chemical analysis of the açaí berries collected for analysis in February 2018. The

preparation of açaí extracts is described in section 5.5.4, the chemical analysis

(sections 5.5.5 to 5.5.14) and the biological assays (sections 5.5.15 to 5.5.20) of

non-commercial and commercial açaí berries obtained from Brazil and the United

Kingdom.

Description of the samples

The initial analyses were performed with the following set of samples:

commercialised frozen pulp or pure açaí which was bought in São Paulo (Pulp

SP) and freeze-dried. In addition, the non-commercial açaí samples, both whole

(purple and white whole açaí) and de-fatted (purple and white de-fatted açaí)

berries were obtained directly from the Amazon region. The samples were

163

freeze-dried and the oil was extracted using the supercritical carbon dioxide

method (Pessoa and Teixeira, 2012). Oil extraction was performed at the

Universidade Federal do Pará (UFPA). Commercial samples were bought as

pure açaí powder at local supermarkets in the United Kingdom (commercial UK)

and Brazil (commercial SP).

Sample identification based on colour

The non-commercial and commercial açaí samples showed a visible

difference in the colour of the material, as shown in Figure 5.5. It is known that

the purple colour present in natural materials, such as red wine and blueberries,

is related to the content of anthocyanin (Del Pozo-Insfran et al., 2004). Therefore,

the berries were identified according to their colour following the CIELAB colour

space international parameters (León et al., 2006).

The colour parameters of the berries were measured using a reflectance

spectrophotometer (CR-400, Konica, Minolta, Japan) calibrated with a regular

white tile according to the CIELAB colour space international parameter, as

shown in Figure 5.6 and calculated using Equation 5.1. This analysis was

performed at the Plants for Human Health Institute, North Carolina State

University (NCSU, USA).

164

Figure 5.5: Picture of the açaí berry samples where: (A) is the purple açaí whole

used as reference; (B) is the white açaí whole; (C) is the purple de-

fatted sample; (D) is the white de-fatted; (E) is the freeze-dried frozen

pulp from São Paulo; (F) is the commercial sample from São Paulo;

and 7 is the commercial sample bought in United Kingdom.

Figure 5.6: Illustration of the CIELAB colour space international parameters.

Adapted from Molino et al. (2013).

165

ΔE* = [ΔL^2 + Δa^2 + Δb^2]^1/2

Equation 5.1

where:

ΔE = total colour difference;

L = lightness (L);

a = greenness (−a) or redness (+a); and

b = blueness (−b) or yellowness (+b).

The non-commercial purple whole berries obtained directly from the

Universidade Federal do Pará (UFPA) were used as a reference sample and the

colour differences were calculated using Equation 5.1 so as to compare with a

particular açaí sample. The differences in the colour between the whole purple

berries and others are reported in Table 5.2. Commercial samples collected from

Brazil (SP) had a similar colour compared to the standard (lower total colour

difference), whilst the commercial samples bought in the UK were found to be

lighter (higher L), redder (higher a) and yellower (higher b). The oil extraction (de-

fatted samples) did not have a significant effect on the colour difference of the

purple samples. However, within the white samples, the de-fatted material was

slightly lighter (higher L) and yellower (higher b). This simple and low-cost

experiment can be used in the future to identify possible cases of fraud in the

açaí industry and might be useful to perform a qualitative analysis on the

anthocyanin content of açaí products.

166

Table 5.2: Colour parameters of açaí samples where L* indicates lightness, a*

the red/green coordinate, b* the yellow/blue coordinate and ΔE the

total colour difference (refer to Equation 5.1) determined by a

reflectance spectrophotometer (CR-400, Konica, Minolta, Japan). The

data relates to a pooled freeze-dried sample.

L* a* b* ΔE Purple açaí whole (reference) 35.44 1.34 - 0.31 -

Purple açaí de-fatted1 35.94 5.17 - 0.52 7.48

White açaí whole 44.96 1.49 13.82 145.15

White açaí de-fatted 55.85 - 0.01 20.06 416.66

Pulp SP 33.33 2.65 0.23 3.23

Commercial SP 36.82 6.17 0.22 12.76

Commercial UK 51.46 10.88 6.66 198.12

1de-fatted (removal of the oil fraction); SP – Sao Paulo, Brazil; UK – United Kingdom.

Method development for the açaí extractions for organic analysis

In accordance with the literature about the extraction of regular berries for

polyphenolic analysis, the most common and efficient method is methanolic

extraction (Kapasakalidis et al., 2006, Castaneda-Ovando et al., 2009). In order

to evaluate what would be available for human intake of the chemicals present,

an extraction with water and mild acid is also proposed in this study. A

comparison of using both extraction solvents (methanolic and aqueous) was

undertaken. Furthermore, the anthocyanin compounds reported to be present in

açaí samples only exist as a stable ring structure under mild acidic solution

conditions (Albarici et al., 2006). Therefore, in order to protect the anthocyanin

structure, and to not further change the chemical structure, 0.5% of a weak acid

solution was added to the solvents. Samples of 0.20 ± 0.01g of ground freeze-

dried açaí were extracted (in duplicate) with 5 mL of solvent: acidified 70% v/v

High Pressure Liquid Chromatography (HPLC) grade methanol (Fisher Scientific,

Pittsburgh, USA) in 0.5% v/v HPLC grade acetic acid (HAc) (Fisher Scientific,

Pittsburgh, USA), or acidified water (0.5% v/v HAc). The mixtures were sonicated

167

for 10 minutes and centrifuged (Sorvall RC-6 plus, Asheville, NC, USA) for 10

minutes at 5000 rpm. The supernatant was transferred to a 25 mL volumetric

flask. The resultant pellet of açaí was further extracted (twice) and the extracts

were combined and diluted with double distilled water to a final volume of 25 mL.

Each final solution was filtered using a 0.2 μm PTFE syringe filter (Fisher

Scientific, Pittsburgh, USA) before analysis.

Total polyphenol and flavonoid content: Materials and method

The total polyphenol content of the açaí extractions (methanolic and

aqueous) was determined using the Folin-Ciocalteu assay, as previously

described in section 2.2. The assay was adapted for a micro-plate following the

Singleton method (Singleton et al., 1999a) and performed at NCSU. The

advantage of using a microplate reader, instead of cuvettes, is to reduce the

amount of reagents used (by the factor of 40) with consequent reduction of time

and cost. This results in an increase in the level of precision, because all of the

samples were prepared and read at the same time as the Folin-Ciocalteu

reaction is time-dependent. In a 96 well-plate, 75 µL of distilled water was added

to each well (in triplicate) along with 25 µL of the diluted sample or standard, and

25 µL of diluted (1:1 v/v volume) Folin-Ciocalteu reagent. After 6 minutes, 100 µL

of 7.5% v/v sodium carbonate solution was added. The plate was then left in the

dark for 90 minutes and read using a UV-Vis plate reader (SpectraMax® M3,

Sunnyvale, USA) at 765 nm. The results are expressed in gallic acid equivalent,

which was used as a standard reference (Singleton et al., 1999a).

The total flavonoid content can be determined using a colourimetric

method based on the complexation of the flavonoids with aluminium chloride

(AlCl3) (da Silva et al., 2015). An example of this coordination for the quercetin

flavonoid is shown in Figure 5.7. The resulting complexes present a different

coloration from the initial solution and can be determined spectrophometrically

(yellow to blue).

168

Figure 5.7: Molecular structures of the complexation between quercetin and

aluminium chloride used to determine the levels of total flavonoids.

Adapted from Frederice et al. (2010).

The assay, previously described by Zhishen et al. (1999), was adapted for

a micro-plate assay. In a 96-well plate, 100 μL of açaí extracts or standard were

added to each well (in triplicate) in addition to 100 μL of 2% AlCl3 solution

(Sigma-Aldrich, St. Louis, USA), freshly prepared in methanol (HPLC grade,

Sigma-Aldrich, St. Louis, USA). The plate was mixed and left at 20 °C for 1h. The

absorbance was read using a UV-Vis plate reader at 415 nm. Results are

expressed as quercetin equivalent, which was used as the standard reference for

the calibration curve ranging from 0 to 100 mg/L quercetin equivalent. The

extracts were also diluted with the extraction solution as required to fit the

calibration curve.

Total polyphenol and flavonoid content: Results and discussion

The açaí extracts, collected as part of this study, were analysed in order to

evaluate the total polyphenol content using the Folin-Ciocalteu assay (refer to

section 5.5.5). The results are presented in Table 5.3, along with the data

obtained from the other assays performed on samples. The data is presented on

a dry weight basis (d.w.). In order to compare these values with reported

literature values, the data were converted to a fresh weight basis (f.w.). This

information does not exist for these samples, so an assumption of a 90%

169

moisture content was adopted, based on a procedure described by Pavan et al.

(2012).

Açaí extraction methods (aqueous and methanolic), as reported in section

5.5.4, were also evaluated using a paired two-tailed t-test (Miller et al., 2018).

The data confirmed that the null hypothesis was retained, and as such, confirmed

that there is no statistically significant difference in the total polyphenol content

(tcalc= 1.11 < tcrit = 2.30, p=0.3146, n=9, α =0.05). Therefore, an aqueous solution

was used for further analysis of the total polyphenol content. From the data

presented, it is clear that the samples that were de-fatted (i.e. removal of the oil

fraction) showed a significantly higher total polyphenol content for the non-

commercial samples (paired two-tailed t-test, tcalc= 5.60 > tcrit = 2.15, p<0.0001,

n=16, α =0.05).

Table 5.3: Total polyphenol, flavonoid, anthocyanin (ANC) and proanthocyanidin

content (PAC); and chemical antioxidant activity (ABTS and DPPH) of

açaí pulp samples. The values are expressed as mean ± standard

deviation and dry weight.

Samples Total

polyphenol (mg/g)*

Total flavonoid (mg/g)*

Total ANC

(mg/g) - 70%

MeOH extraction

DMAC Total PAC

(mg/g) - 0.5% HAc extraction

ABTS (mg/g)*

DPPH (mg/g)*

Number of replicates 6 6 2 3 6 6

Non-commercial

Purple Açaí whole 32.00 ±

1.03

6.39 ±

1.23

10.20 ±

0.24

6.10 ±

2.09

438.0

± 17.5

336.0

± 72.0

Purple Açaí de-fatted 39.40 ±

1.67

8.05 ±

0.81

14.33 ±

0.58

5.06 ±

0.68

529.0

± 57.0

419.0

± 69.5

170

Samples Total

polyphenol (mg/g)*

Total flavonoid (mg/g)*

Total ANC

(mg/g) - 70%

MeOH extraction

DMAC Total PAC

(mg/g) - 0.5% HAc extraction

ABTS (mg/g)*

DPPH (mg/g)*

White Açaí whole 9.40 ± 0.70 2.12 ±

0.27 <0.01

3.96 ±

2.39

83.0 ±

9.7

53.2 ±

15.1

White Açaí de-fatted 11.70 ±

0.24

2.38 ±

0.35 <0.01

2.60 ±

0.49

101.0

± 16.2

67.4 ±

10.7

Oil White 2.74 ± 1.13 1.42 ±

0.96 <0.01

1.54 ±

0.22 <15.3 <7.4

Oil purple 1.68 ± 0.52 0.87 ±

0.38 <0.01

3.40 ±

1.42 <15.3 <7.4

Commercial

Pulp SP 28.30 ±

0.64

5.00 ±

0.68

3.59 ±

0.12

4.75 ±

1.58

310.0

± 41.4

222.0

± 38.6

Powder SP 42.40 ±

1.53

6.07 ±

1.45

4.70 ±

0.01

4.47 ±

0.54

316.0

± 39.1

304.0

± 43.2

Powder UK 5.09 ± 0.42 1.88 ±

0.73 <0.01

5.48 ±

2.20

55.2 ±

31.4 <7.4

*Results presented as a combination of both aqueous and methanolic extracts. GAE: gallic acid equivalent; ANC: anthocyanin content; PAC: proanthocyanidin content; DMAC: Dimethylacetamide; ABTS: 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid); DPPH: 2,2-diphenyl-1-picrylhydrazyl; SP: São Paulo; UK: United Kingdom.

Figure 5.8 presents a comparison of the total polyphenol content of non-

commercial and commercial purple and white açaí freeze-dried berries. The

purple non-commercial samples have a significantly higher amount of total

polyphenol content in comparison with the white samples (two-tailed Student t-

test, tcalc= 5.95 > tcrit = 2.30, p=0.0003, n=9, =0.05). Furthermore, the

commercial purple berries have a higher level of variability (standard error of the

mean of 9.26 GAE mg/g) when compared with the purple pure açaí berries (2.90

GAE mg/g).

171

Figure 5.8: Box plots of the total polyphenol content (gallic acid equivalent mg/g)

of the açaí extracts determined using the Folin-Ciocalteu assay (refer

to section 5.5.5). The values relate to the type of sample (purple, n=

6; white and commercial n= 4; n is the number of samples).

The results for this study show a similar pattern in terms of the de-fatting

process. In general, the removal of the oil fraction results in the pre-concentration

of the chemicals in the de-fatted samples. Moreover, this indicates that the oil

does not have a significant level of the polar chemicals analysed. A previous

study on a range of fatty acids found in açaí oil confirm a similar pattern to the

data obtained here for the levels of total polyphenol, flavonoid and anthocyanin in

de-fatted açaí pulp (Pacheco-Palencia et al., 2008). This pre-concentration of the

chemicals will also have an effect on the antioxidant activity and therefore the

radical inhibition of the berries.

Purple

White

Comm

erci

al p

urple

0

10

20

30

40

50

To

tal p

oly

ph

en

ol co

ncen

trati

on

(G

AE

) m

g/g

172

Table 5.4: Literature values for the total polyphenol content (mg GAE/ 100g) and

antioxidant activities DPPH (g/g DPPH) and ABTS μmol Trolox/g of

typical tropical berries from Brazil (dry weight). Table adapted from

Rufino et al. (2010).

*concentration required to obtain a 50% antioxidant effect

Table 5.4 presents a literature review of the total polyphenol content and

antioxidant activities of typical tropical berries from Brazil. The data is reported on

a dry weight basis and presents a significant correlation between total polyphenol

and antioxidant activities (Spearman’s correlation; R= -0.81, p<0.0001 for DPPH

and R= 0.77, p=0.0002 for ABTS; n=18 and =0.05). This correlation indicates

Fruits Extractable polyphenols DPPH ABTS+ mg GAE/100 g EC50* (g/g DPPH) μmol Trolox/g

Açaí 3268 ± 527 598 ± 164 64.5 ± 19.2

Acerola 10280 ± 77.7 49.2 ± 2.5 953 ± 34.1

Bacuri 1365 ± 43.3 6980 ± 854 18.1 ± 3.7

Cajá, yellow mombim 579 ± 12.9 1064 ± 162 40.7 ± 2.2

Caju, cashew apple 830 ± 26.5 906 ± 78.2 79.4 ± 15.7

Camu-camu 11615 ± 384 42.6 ± 1.4 1237 ± 33.8

Carnaúba 830 ± 28.3 4877 ± 24.3 16.4 ± 0.2

Gurguri 1364 ± 24.8 360 ± 32.7 136 ± 20.1

Jaboticaba 3584 ± 90.9 138 ± 3.1 317 ± 2.7

Jambolão, java plum 1117 ± 67.1 938 ± 46.9 125 ± 10.8

Juçara 5672 ± 55.9 70.1 ± 4.8 606 ± 142

Mangaba 935 ± 37 890 ± 69.1 65.6 ± 7.4

Murici, nance 2380 ± 104 238 ± 17.7 412 ± 13

Murta 2055 ± 75.7 363 ± 27.4 166 ± 4

Puçá-coroa-de-frade 1047 ± 77 316 ± 2 161 ± 3

Puçá-preto 2638 ± 48.9 65.6 ± 2.4 346 ± 21.7

Umbu 742 ± 19 933 ± 109 77 ± 15.4

Uvaia 1930 ± 129 276 ± 22.2 182 ± 14.2

173

that the amount of total polyphenol content of the açaí berries could be

potentially related to the chemical antioxidant activity. The total polyphenol

content of the açaí pulp (Figure 5.8) is in accordance with the range reported in

the literature for açaí purple berries (Gordon et al., 2012, Augusti et al., 2016),

white berries (Lichtenthäler et al., 2005b) and oil (Pacheco-Palencia et al., 2008).

It should be pointed out that a direct comparison between the results presented

in this study and the literature is often difficult due to a lack of information relating

to the weight basis or origin and treatment of the samples (published data). The

non-commercial purple and commercial samples from São Paulo represent an

average level of the total polyphenol content, in comparison with other Brazilian

fruits, based on the classification reported by Rufino et al. (2010). The main

advantage of consuming açaí rather than the other fruits presented in Table 5.4

is associated with the easy access to açaí in all regions of the country (as a

frozen pulp). As such, the consumption of the processed frozen pulp is not

dependent on a seasonality factor (Tonon et al., 2009), in contrast to other

Brazilian fruits and seasonal berries. In comparison with other traditional berries,

presented on Table 5.5, the non-commercial purple açaí and the commercial

powder SP have a similar total polyphenol content to that of cranberries, higher

than raspberries and grapes and lower than blackberries, bilberries and

blackcurrants (Rothwell et al., 2013). The non-commercial white and the other

commercial açaí samples have lower values than the reported values for other

berries (refer to Tables 5.4 and 5.5).

The açaí extracts were also analysed by the AlCl3 assay (refer to section

5.5.6) in order to determine the total flavonoid content. The results for the total

flavonoid concentration are reported in Table 5.3 and have a similar pattern to

the data in Figure 5.8. Furthermore, a two-tailed paired t-test (Miller et al., 2018)

showed that the açaí extraction methods (aqueous and methanolic) do not

present a significant difference in terms of the flavonoid content (tcalc= 0.56 < tcrit

= 2.30, p=0.5907, n=9, =0.05).

174

Table 5.5: Literature review of the total polyphenol and total anthocyanin content

of other berries obtained by two different methods (HPLC and pH);

and the total proanthocyanidin (mg/100 g) (fresh weight). Adapted

from Rothwell et al. (2013).

Total polyphenol

Total anthocyanin

(HPLC)

Total anthocyanin (pH method)

Total proanthcyanidin

(HPLC) cranberry 315 49.89 32 -

blackberry 569.43 172.59 146.8 17.95

bilberry 525 - 299 -

raspberry 154.65 72.47 43.57

blackcurrant 820.64 592.22 225.04 138.21

grape (black) 184.97 62.1 - 61.2

low bush blueberry 471.55 187.24 149.17 333.1

strawberry 289.2 73.01 - 145

The total flavonoid content of açaí pulp extractions have a similar pattern

to that for the total polyphenol content, as would be expected because the

flavonoids are a subclass of polyphenols (Tsao, 2010). The data is also in

agreement with the available literature for commercial açaí samples (Rufino et

al., 2010, Horszwald and Andlauer, 2011).

Total anthocyanin content: Materials and method

The determination of the total anthocyanin content of the açaí extracts

was performed following the Grace et al. (2013) method using a high

performance liquid chromatography or HPLC instrument (Agilent 1200 HPLC)

with photodiode array detector (DAD). The separation was conducted using a RP

Supelcosil-LC-18 column, with dimensions 250 mm × 4.6 mm × 5 μm (Supelco,

Bellefonte, USA) held at a constant temperature of 30°C with a flow rate of 1

mL/min and the gradient programme, as shown in Table 5.6.

175

Table 5.6: Gradient programme for the determination of the anthrocyanin content

of açaí extracts using an Agilent 1200 HPLC instrument.

Time (minutes) 5% v/v formic acid (%) 100% methanol (%) 0 90 10

5 85 15

15 80 20

20 75 25

25 70 30

45 40 60

47 90 10

60 90 10

The quantification of the total anthocyanin content was based on the sum

of the integrated anthocyanin peaks from 0 to 60 minutes, calculated against the

standard curve of Cy-3-Glu (as shown in Figure 5.9).

Figure 5.9: Standard curve of Cy-3-Glu concentration (mg/L) and the areas of the

peaks (mAU) used as the calibration curve for the determination of

the anthocyanin content of açaí extracts using a HPLC-DAD

chromatogram (at 520 nm).

0 1 2 30

2

4

6

Cy-3-Glu concentration (mg/L)

Are

as o

f th

e p

eaks (

mA

U) Y = 1.908x + 0.07211

R2 = 0.9982

176

Total anthocyanin content: Results and discussion

Figure 5.10 reports the HPLC-DAD chromatogram at 520 nm used to

integrate the anthocyanin peaks and determine the total anthocyanin content of

the açaí samples, as described in section 5.5.7. The anthocyanin peaks, found in

the extracts for purple açaí (non-commercial and commercial) samples, are

cyandin-3-glucoside (retention time, tR = 17.458 minutes), cyandin-3-rutinoside

(tR = 21.069 min) and peonidin-3-rutinoside (tR = 26.237 min), respectively. The

identification of the anthocyanin peaks was performed using an cyanidin 3-

glucoside reference standard and the peak identification and elution order by the

method previously proposed Vera de Rosso et al. (2008). Only the purple

commercial SP, pulp SP and non-commercial purple açaí samples had

detectable anthocyanin peaks.

The quantification of the total anthocyanin content was performed, as

described in section 5.5.8 and is presented in Table 5.3. Statistical analysis

confirmed that there is no significance difference between the two extraction

methods (paired two-tailed t-test, tcalc= 2.79 < tcrit = 3.18, p=0.0684, n= 4,

=0.05).

177

Figure 5.10: The determination of the anthocyanin content of a purple non-

commercial açaí sample following methanolic extraction and using a

HPLC-DAD chromatograph (at 520 nm). The anthocyanin peaks are

cyandin-3-glucoside (retention time, tR = 17.458 minutes), cyandin-3-

rutinoside (tR = 21.069 min) and peonidin-3-rutinoside (tR = 26.237

min).

The anthocyanin profile and content of the purple açaí berries are in

agreement with previous studies (Del Pozo-Insfran et al., 2004, Lichtenthäler et

al., 2005b, Pacheco-Palencia et al., 2008, Gordon et al., 2012, Gouvêa et al.,

2012). The purple açaí samples were the only extracts that showed an

anthocyanin profile as it could be expected from the purple colour of the material,

which is due to the anthocyanin molecules (Del Pozo-Insfran et al., 2004). As

such, the white açaí berries were expected to not show significant levels of

anthocyanins, as confirmed by the results. Secondly, the pure açaí powder,

commercially purchased in the UK, did not show any significant anthocyanin

peaks, below the limit of detection. This suggests that the UK sample might have

some other material added to it or that the chemicals might have been lost during

commercialisation. The total anthocyanin content results are in accordance with

the qualitative analysis of the colour of the material, as presented in section

178

5.5.3. In comparison with other typical berries, as presented in Table 5.5, the

total anthocyanin content for the non-commercial purple açaí samples ranges

from 102.0 to 143.3 mg/100 g (f.w.) and for the commercial purple açaí samples

35.9 to 47.0 mg/100 g (f.w.) Therefore, açaí would occupy the fourth position in

ranking of the berries (Table 5.5) in terms of the total anthocyanin content of the

fruit.

Total proanthocyanidin content: Materials and method

The total proanthocyanidin (PAC) content was determined using the 4-

dimethylaminocinnamaldehyde (DMAC) method adapted for a micro-plate assay,

as previously described by Prior et al. (2010), where the DMAC reacts with the

terminal units of the PAC oligomers. In a 96 well-plate, 63 µL of the diluted

sample, standard or blank was added, in triplicate, to each well along with 189 µL

of the DMAC reagent (Sigma-Aldrich, St. Louis, USA). The plate was set on the

plate reader at 640 nm to read the absorbance value of the wells, at time

intervals of a minute for a total period of 30 minutes. The results are expressed in

procyanidin B1 dimer (Sigma-Aldrich, St. Louis, USA) equivalent, that was also

used as a standard reference. The extracts were also diluted with the extraction

solution to fit a calibration curve ranging from 0 - 100 mg/L B1 equivalent.

Total proanthocyanidin content: Results and discussion

The total proanthocyanidin (PAC) content of the açaí samples was

determined using the DMAC assay (refer to section 5.7.4). Interestingly, this was

the only assay that produced a statistically significant difference between the two

sets of extraction samples (i.e. aqueous vs methanolic) (paired two-tailed t-test,

tcalc= 4.16 > tcrit = 2.30; p=0.0031, n=9, =0.05), as shown in Figure 5.11. The

aqueous extraction resulted in significantly higher PAC concentrations, due to the

nature of the PAC compounds. Proanthocyanidin are oligomeric flavonoids (He

179

et al., 2008) and previous studies have shown that the polymers (more than 10

units) are the major PAC in freeze-dried açaí (Schauss et al., 2006b). Therefore,

it was expected that the polymers are more soluble in the aqueous rather than

the methanolic extracts (He et al., 2008). The difference in the PAC levels

between the purple and white samples were lower than that for the other assays

presented in this study. This supports the claim that PAC could be the class of

polyphenols responsible for the antioxidant activity of the white açaí berries.

Figure 5.11: Total proanthocyanidin content (PAC) of açaí extractions presented

as B1 equivalents (B1E) via DMAC assay (refer to section 5.5.8) and

compared between the methanolic (70% MeOH) and aqueous (0.5%

HAc) extraction methods (n= 4, n, number of instrument replicates).

Sample 1: Purple açaí whole; 2: Purple açaí de-fatted; 3: white açaí

whole; 4: white açaí de-fatted; 5: oil extracted from white açaí; 6: oil

extracted from purple açaí; 7: pulp SP; 8: powder SP; 9: powder UK.

The analysis of the oil extracts for both white and purple açaí berries

showed a significant level of PAC (typically 2 – 4 mg/g), confirmed by the fact

that the whole samples have a higher PAC content (6 mg/g for purple and 4 mg/g

for white) than that for the de-fatted samples (5 mg/g for purple and 3 mg/g for

white). The commercial samples also have PAC levels (typically 5 mg/g) similar

to that for the non-commercial purple whole sample (6 mg/g).

1 2 3 4 5 6 7 8 90

2

4

6

8

10

Samples

To

tal P

AC

(B

1E

) m

g/g

70% MeOH

0.5% HAc

180

In comparison with other ‘super-fruits’, the reported values of PAC levels

for açaí obtained in other studies is in agreement with the data in this study for

purple açaí samples. Furthermore, açaí values are similar to the values for

blueberries and cranberries; but higher than that for pomegranate and lower than

cocoa seeds (Crozier et al., 2011). The reported values of PAC for typical berries

are reported in Table 5.5 (based on a HPLC method). These literature values are

compared with the data of this study which were obtained using the DMAC

method, as described in section 5.5.8. Although any critical analysis of the values

would be questionable (based on the variation of techniques), it is interesting to

note that the non-commercial purple açaí samples of this study have similar PAC

levels to those reported for black grapes (Rothwell et al., 2013).

Chemical antioxidant activity: Materials and method

Free radicals and other oxidising molecules have been recently

considered as one of the main reasons linked to the onset of cancer,

Alzheimer’s, Parkinson and cardiovascular diseases (de Souza et al., 2010). The

excess of these free radicals can be balanced by antioxidants produced by the

body or acquired via the diet.

The antioxidant activity of a compound or natural product can be

measured by different mechanisms, such as electron transfer, reducing power,

hydrogen atom transfer or radical scavenging (Shahidi and Zhong, 2015). In this

study, the antioxidant activity of açaí was determined using two different

methods, namely, radical scavenging by the 2,2'-azino-bis(3-

ethylbenzothiazoline-6-sulphonic acid (ABTS) and 2,2-diphenyl-1-picrylhydrazyl

(DPPH) molecules.

Antioxidants can scavenge reactive oxygen species (ROS) or other free

radicals by hydrogen atom transfer or electron transfer. This activity is usually

expressed as the Trolox equivalent, which is a well-known antioxidant analog of

vitamin E (Arts et al., 2004).

181

The ABTS assay involves the conversion of 2,2'-azino-bis(3-

ethylbenzothiazoline-6-sulphonic acid) or the ABTS reagent to a radical through

reacting it with persulfate. The resultant solution is then colourimetrically reacted

with the antioxidant compounds present in the extracts, especially the

polyphenols. The antioxidant activity was measured by ABTS following the Re et

al. (1999) method. The 2.45 mM ABTS+• solution was prepared by reacting the

ABTS solution (Sigma-Aldrich, St. Louis, USA) with potassium persulfate (Sigma-

Aldrich, St. Louis, USA) and left to stand in the dark for 12 to 16 hours. The final

solution was diluted until the absorbance at 734 nm reached 0.70 0.02

absorbance units using a UV-Vis plate. In a 96 well-plate, 9 µL of açaí extract

was added to 271 µL of ABTS+• solution (in triplicate). The absorbance of the

plate was read after 10 minutes at 30C along with 9 µL of distilled water and 271

µL of ABTS+• solution, read as the blank. The results are expressed in Trolox

equivalent (Sigma-Aldrich, St. Louis, USA), which was used as a calibration over

the range of 100 to 500 µM.

The other radical scavenging assay involves the free radical 2,2-diphenyl-

1-picrylhydrazyl or DPPH• which reacts with an antioxidant molecule resulting in

a discoloration of the solution, measured at 515 nm using a micro-plate UV-Vis

reader (Truong et al., 2007). The radical 150 µM DPPH• solution was prepared

by mixing the DPPH reagent (Sigma-Aldrich, St. Louis, USA) in 80% v/v

methanol (HPLC grade, Sigma-Aldrich, St. Louis, USA). In a 96 well-plate, 180

µL of the radical solution was added to 20 µL of the açaí extract, standards or

70% methanol v/v, which was used as a blank. The plate was left in the dark for

40 minutes and read at 515 nm. The results were also quantified using a

calibration curve of Trolox (Sigma-Aldrich, St. Louis, USA) over the range of 100

to 500 µM.

182

Chemical antioxidant activity: Results and discussion

The antioxidant activity results are presented in Figure 5.12 and Table 5.3.

The data shows a significantly higher level of antioxidants for most of the

samples, with the exception being for the oil extraction samples. The extraction

methods did not show a significance difference, based on using a paired two-

tailed t-test (tcalc= 1.56 < tcrit = 2.45, p=0.1705, n=7, =0.05). The white and the

commercial açaí samples bought in UK, have significantly lower levels of

antioxidant activity when compared to the non-commercial purple açaí berries

(One-way ANOVA, r = 0.96, p<0.0001, n=40, =0.05). In addition, the

commercially bought açaí in the UK have a significantly lower activity when

compared to the other commercial purple samples (One-way ANOVA, r = 0.92,

p<0.0001, n=23, =0.05).

Figure 5.12: Antioxidant activity of açaí extracts determined by the ABTS assay,

data reported as Trolox equivalents (TE) (n=3; n, number of

instrumental replicates).

The DPPH assay was performed as described in section 5.5.6 and the

results are presented in Table 5.3. The freeze-dried pulp sample from São Paulo

had a slightly lower activity in this assay when compared to the ABTS assay. On

Purple

White

Comm

ercia

l purp

le0

200

400

600

AB

TS µ

mol

TE

/g

183

the other hand, the commercial sample bought in the UK did not show any

significant antioxidant activity.

The same trend was found for the de-fatted samples analysed by this

assay, with the whole açaí extracts showing a slightly lower antioxidant activity

due to the concentration of the material when removing the oils. These findings

are similar to the ABTS assay confirming the high antioxidant activity of the purple

non-commercial açaí material. The data relating to the antioxidant activity of the

açaí extracts are in agreement with the range of results reported in the literature

for açaí samples (Gordon et al., 2012, Augusti et al., 2016, Garzón et al., 2017).

Furthermore, in comparison with typical Brazilian fruits (refer to Table 5.4), açaí

has an average level of antioxidant activity.

Elemental composition: Materials and method

The determination of the total elemental composition of the açaí samples

was performed as described in section 2.3. The samples were fully digested and

analysed by inductively coupled plasma mass spectrometry or ICP-MS.

Elemental composition: Results and discussion

The total elemental composition of the açaí samples was evaluated

following the method proposed in section 5.5.10. The calcium, magnesium,

manganese, iron, zinc and copper; essential minor and trace elements found in

significant levels in açaí samples are presented in Table 5.7 (a) calculated as

fresh weight and shown on a dry weight basis in Figures 5.13 and 5.14.

Moreover, these elements were chosen because their levels in açaí products

may play a significant contribution to the nutritional intake of these minerals. It

should be stated that the powdered samples were obtained by freeze-drying, so

the same ‘conversion’ factor used to convert dry to fresh weight (based on 90%

of the material being water) was also applied to the data in Table 5.7. A full set of

data for all elements is presented in Appendices 5.1 and 5.2.

184

Table 5.7 (a): Total elemental levels (mean ± standard deviation) of essential

trace elements (mg/kg fresh weight) of açaí pulp samples

determined using ICP-MS (refer to section 2.1): data relates to the

type of sample (non-commercial : purple n= 6; and white n= 4; and

commercial: purple n= 4; n is the number of samples).

Samples Non-commercial Commercial

Purple Açaí

whole

Purple Açaí de-

fatted

White Açaí

whole

White Açaí de-

fatted Pulp SP Powder

SP Powder

UK

Ca 468.82 ±

14.15

527.34 ±

58.07

416.23 ±

2.20

525.11 ±

13.28

165.04 ±

16.12

202.93 ±

17.25

66.17 ±

15.65

Mg 233.23 ±

2.16

272.36 ±

4.15

246.92 ±

12.13

301.77 ±

9.00

202.21 ±

2.75

203.69 ±

2.14

85.92 ±

1.13

Mn 64.06 ±

0.93

80.92 ±

0.77

61.14 ±

0.60

80.89 ±

1.57

26.76 ±

1.51

54.70 ±

1.78

1.62 ±

0.01

Fe 3.01 ±

0.00

4.17 ±

0.12

3.65 ±

0.75

4.30 ±

0.10

3.00 ±

0.04

2.19 ±

0.03

0.23 ±

0.01

Zn 2.49 ±

0.07

3.24 ±

0.17

2.65 ±

0.03

3.62 ±

0.17

2.70 ±

0.65

2.32 ±

0.31

1.27 ±

0.12

Cu 1.81 ±

0.03

2.22 ±

0.01

1.72 ±

0.06

2.50 ±

0.10

1.52 ±

0.05

1.50 ±

0.09

0.35 ±

0.02

The difference between the elemental concentrations of the non-

commercial whole and de-fatted samples (section 5.5.2) follow the same trend as

the previous data for organics and biological analysis (section 5.7). Overall, for

both white and purple açaí pulp, all of the elemental values for the de-fatted

samples are slightly more concentrated than the whole material due to the

removal of the oil fraction. The enhancement of the elemental levels in the

defatted samples is on average 28% for purple and 38% for white samples.

However, there is no statistically significant difference for all the 15 analysed

elements for both purple (paired two-tailed t-test, tcalc= 1.76 < tcrit = 2.15,

p=0.1004, n=15, =0.05) and white samples (paired two-tailed t-test, tcalc= 1.69 <

tcrit = 2.15, p=0.1131, n=15, =0.05). Interestingly, the commercial pulp and

185

powder products purchased in São Paulo have similar elemental levels.

However, the powered sample purchased in the UK has either been subjected to

a different process or has been ‘adulterated’ as the elemental levels are typically

30% of the Brazilian powdered samples.

The commercial samples, especially the açaí pulp (purchased in São

Paulo) has lower (and a larger spread) elemental levels than the non-commercial

samples. For example, the manganese levels of the non-commercial pulp were

64.06 ± 0.93 mg/kg Mn (f.w.) for purple and 61.14 ± 0.60 mg/kg Mn (f.w.) for

white; whilst the commercial pulp (purple) was 26.76 ± 1.51 mg/kg Mn (f.w.).

Clearly, the commercial processing has reduced the Ca and Mn levels by about

50% and the commercial pulp has a significant variance (as shown by the

standard deviation values). This may be due to the origin of the berries used in

the commercial sample or the processing method used (which are both

unknown).

Table 5.7 (b): Total elemental levels (mean ± standard deviation) of non-

essential/toxic trace elements (mg/kg fresh weight) of açaí pulp

samples determined using ICP-MS (refer to section 2.1): data

relates to the type of sample (non-commercial : purple n= 6; and

white n= 4; and commercial: purple n= 4; n is the number of

samples).

Samples Non-commercial Commercial

Purple açaí

whole

Purple açaí de-fatted

White açaí whole

White açaí de-fatted Pulp SP Powder

SP Powder

UK

As <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001

Cd 0.007 ±

0.0000

0.010 ±

0.001

0.009 ±

0.000

0.018 ±

0.001

0.004 ±

0.001

0.006 ±

0.000

0.001 ±

0.000

Pb 0.050 ±

0.000

0.070 ±

0.000

0.030 ±

0.000

0.050 ±

0.000

0.010 ±

0.000

0.003 ±

0.000

0.002 ±

0.000

186

The non-essential or toxic elements, arsenic, cadmium and lead (reported

in Table 5.7 (b)) are found at very low levels in both the commercial and non-

commercial açaí products (typically <0.01 mg/kg f.w.). The low concentration for

arsenic and lead are in agreement with what is normally found in typical tropical

fruits (<0.003 mg/kg for arsenic and <0.006 mg/kg for lead) (Avegliano, 2009,

Vannoort and Thomson, 2003). The cadmium levels found in this study are

higher than that for tropical fruits as described in the literature (<0.002 mg/kg).

However, the levels do not make a significant contribution to the daily dietary

intake of these elements (1.31 ± 0.16 μg/day for cadmium) (Colli, 2005,

Avegliano et al., 2011). This is encouraging in terms of the açaí products being

used for human consumption.

Figure 5.13: Box plots of the total elemental content of minor elements (mg/kg

d.w.) of açaí pulp samples using ICP-MS (refer to section 2.1)

relating to the type of sample (non-commercial: purple n= 6; and


samples). The commercial sample is a combination of pulp SP and

powders SP and UK.

Ca Mg

Purple White Commercial 0

2

4

6

8

Co

ncen

trati

on

of

Ca (

g/k

g)


1

2

3

4

Co

ncen

trati

on

of

Mg

(g

/kg

)

187

Figure 5.14: Box plots of the total elemental content of trace elements (mg/kg

d.w.) of açaí pulp samples using ICP-MS (refer to section 2.1)

relating to the type of sample (non-commercial : purple n= 6; and


samples). The commercial sample is a combination of pulp SP and

powders SP and UK.

Figures 5.13 and 5.14 graphically report the data (as box plots) for the

non-commercial purple or white and commercial purple açaí samples (based on

a dry weight status). In the box plots, the pulp SP, powder SP and powder UK

samples were combined as commercial samples, causing a spread of the target

elemental concentrations. The data (now reported as on a dry weight basis)

relates to the literature presented in Table 5.1. For calcium, the reported values

are lower than that presented in this study for non-commercial purple samples,

with the exception of values published by da Silva Santos et al. (2014), who also

analysed a non-commercial purple açaí sample. The levels of manganese are at

Purple White Commercial

0

200

400

600

800

1000


10

20

30

40

Purple White Commercial

0

10

20

30

40

50


10

20

30

Mn Fe

Zn Cu

Co

ncen

trati

on

of

ele

men

t (m

g/k

g)

188

least 3 times higher than that reported in other study (Table 5.1). Iron, copper

and zinc have similar levels to that reported by Menezes et al. (2008a), who

analysed a commercial sample of açaí pulp purchased in Belém (Amazon -

Pará). When compared to the other 47 typical fruits, açaí has the highest content

of manganese, the 3rd highest of calcium, 4th of copper, 6th of magnesium, 11th of

zinc and 13th of iron content (Unicamp, 2011).

The elemental values for açaí and other typical Brazilian fruits are

summarised in Table 5.8 (Unicamp, 2011). The açaí sample used was a

commercial açaí frozen pulp obtained from São Paulo. Therefore, in order to

compare the values with the ones described in Table 5.7 (a), the commercial

pulp SP was selected. There was no significant difference between the published

values and the reported values of this study (two-tailed Student t-test, tcalc= 1.83

< tcrit = 2.57, p=0.3667, n=6, =0.05). In comparison to other fruits, açaí

represents good levels of manganese and high levels for the other elements (Ca,

Mg, Fe, Zn and Cu), with the exception of graviola and pequi (refer to Table 5.8).

The typical daily fruits of a Brazilian diet include mango and papaya, which when

compared with the results for açaí, have lower levels of the essential elements

(Unicamp, 2011).

Potassium, sodium, chromium, nickel, cobalt, vanadium, molybdenum and

selenium were also analysed in this study and are presented in Appendices 5.1

and 5.2. The typical level of potassium in tropical fruits is 10399 mg/kg (d.w.)

which is similar to the non-commercial açaí pulp samples. However, the açaí

pulp samples have higher levels of sodium (Colli, 2005, Avegliano et al., 2011).

In comparison with the typical value of trace elements in fruits presented by

Kabata-Pendias (2010), the açaí pulp showed higher levels of chromium

(average of 0.48 mg/kg f.w. against 0.08 mg/kg presented on the literature),

nickel (0.76 mg/kg f.w. against 0.06 mg/kg), and cobalt (0.008 mg/kg f.w. against

0.0016 mg/kg), and lower levels of vanadium (0.003 mg/kg f.w. against 0.33

mg/kg), molybdenum (0.01 mg/kg f.w. against 0.07 mg/kg), and selenium (below

the limit of detection against 0.04 mg/kg).

189

Table 5.8: Literature review of typical Brazilian fruits the total elemental content

of calcium, magnesium, manganese, iron, zinc and copper in mg/kg

(f.w.). Data for açaí is reported as a commercial processed material

and all the others as raw natural typical Brazilian fruits. Adapted from

Unicamp (2011).

Fruits Ca Mg Mn Fe Zn Cu Açaí (Euterpe oleracea) 351.8 170.4 61.6 4.3 2.7 1.8

Acerola (Crataegus azarolus) 125.5 131.3 0.7 2.2 1.5 0.7

Cocoa (Theobroma cacao) 121.0 246.2 0.4 2.6 5.9 1.5

Cajá (Spondias mombin) 127.4 112.8 0.5 1.5 1.8 0.2

Cashew fruit (Anacardium occidentale) 14.2 101.1 1.2 1.5 0.9 0.7

Carambola (Averrhoa carambola) 47.9 73.6 1.3 2.0 2.4 0.8

Ciriguela (Spondias purpurea) 274.1 179.6 0.6 3.6 5.3 1.2

Cupuaçu (Theobroma grandiflorum) 131.2 181.7 0.7 4.9 3.4 0.7

Guava (Psidium guajava) 44.5 68.9 0.9 1.7 1.3 0.4

Grape (black) (Vitis vinifera) 76.2 58.3 0.7 1.7 Trace 0.5

Graviola (Annona muricata) 401.2 235.0 0.8 1.7 1.3 0.4

Jaboticaba (Plinia cauliflora) 83.5 177.8 3.0 0.9 2.8 0.7

Jamelão (Syzygium cumini) 30.9 21.6 Trace 0.5 0.5 0.3

Mamão Papaya (Carica papaya) 224.2 221.8 0.1 1.9 0.7 0.2

Mango (Mangifera indica) 116.6 78.2 1.7 1.0 0.7 1.0

Passion Fruit (Passiflora edulis) 53.9 279.7 1.2 5.6 3.9 1.9

Pequi (Caryocar brasiliense) 324.4 297.7 6.4 2.7 9.6 2.1

Pitanga (Eugenia uniflora) 178.8 122.3 3.6 4.0 3.5 0.8

Pomegranate (Punica granatum) 47.5 127.0 1.3 2.6 6.7 1.9

Strawberry (Fragaria vesca) 109.0 96.7 3.3 3.2 1.8 0.6

Umbu (Spondias tuberosa) 115.6 113.5 0.3 0.9 4.2 0.4

The purple and white açaí have similar elemental levels. There was no

significant difference between the elemental levels of non-commercial purple and

white açaí (purple and white whole: paired two-tailed t-test, tcalc= 0.18 < tcrit =

2.15, p=0.8013, n=15, =0.05; purple and white de-fatted: paired two-tailed t-

test, tcalc= 1.48 < tcrit = 2.15; p=0.1622, n=15, =0.05). This suggest that the

190

samples might come from the same region, or from a similar soil composition

(Gonzálvez et al., 2009).

Biological toxicity (cell viability assay): Materials and method

The cell line mouse macrophage RAW 264.7 (ATCC TIB-71, American

Type Culture Collection; Livingstone, USA) used in the biological studies was

maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies,

New York, USA), supplemented with 100 IU/mL penicillin/100 μg/mL

streptomycin (Fisher Scientific, Pittsburg, USA) and 10% foetal bovine serum

(Life Technologies, New York, USA) at a density not exceeding 5 × 105 cells/mL.

This was maintained at 37 °C in a humidified incubator with 5% of carbon dioxide

prior to the analysis.

The RAW 264.7 cells were seeded in a 96 well plate for the viability assay.

The cell viability was measured by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-

diphenyl-tetrazolium bromide) assay, as previously described by Esposito et al.

(2014). In summary, the cells were seeded in a 96-well plate and treated with two

different doses of the açaí extracts, 50 and 250 µg/mL. The solvent vehicle or

dimethyl sulfoxide or DMSO (Sigma-Aldrich, St. Louis, USA) was used as a

positive control. After incubation, the media was discarded and 100 µL of DMSO

was added to dissolve the purple crystals. The resultant solution was quantified

spectrophotometrically at 550 nm using a microplate reader SynergyH1 (BioTek,

Winooski, USA).

Biological toxicity (cell viability assay): Results and discussion

The potential toxicity of the açaí pulp extracts (both methanolic and

aqueous) was determined using the cell viability assay, as described in section

5.5.10. The results, shown in Figure 5.15, were normalised for the concentration

of the formazan levels (which is the reduced purple product of 3-(4,5-

191

dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide or MTT measured by UV-

Vis spectrophotometer) of the blank (non-treated) cells. The normalised value of

the blank cells is related to the number of viable cells. The vehicle or solvent

used in the extractions was also used to treat the cells so as to confirm the non-

toxicity of the solvent. Dimethyl sulfoxide (DMSO) was used as a positive control,

due to its known toxicity to the cells at higher concentrations, as shown in Figure

5.15; where the viability of the cells decreases by half when treated with 5 µL of

DMSO. Neither of the açaí samples showed a significant toxicity response, even

at the higher concentration of 250 µg/mL (One-way ANOVA, r = 0.43, p=0.0757,

n=31, =0.05).

Figure 5.15: Formazan production levels in RAW 264.7 macrophage cells treated

with açaí extract solutions. Results expressed as mean ± st dev, n=3;

n, number of instrumental replicates).

Biological effect of açaí on radical inhibition assays: Materials and method

In order to evaluate the cellular antioxidant activities of the açaí extracts,

the amount of nitric oxide produced and released to the media of the cells was

measured. In infections and other inflammatory conditions, the macrophages are

Blank

Vehic

le

DMSO

(5ul)

Purple

White

Comm

erci

al

Purple

White

Comm

erci

al0.0

0.5

1.0

Fo

rmaza

n levels

F

old

in

cre

ase o

ver

bla

nk

50 mg/mL 250 mg/mL

192

activated in order to produce NO, an important mediator of the immunity system,

due to its regulatory and cytotoxic effects (Fang and Vazquez-Torres, 2002).

The cell line mouse macrophage RAW 264.7 (ATCC TIB-71, American

Type Culture Collection; Livingstone, USA) was also used in this study. The

nitrite production, a stable end product of NO production in activated

macrophages, was accessed colourimetrically at 540 nm and read on the

microplate reader. The cells were induced by lipopolysaccharides of bacteria

(LPS) to respond to an inflammatory process (Bannerman and Goldblum, 2003).

As a positive control, the cells were treated with dexamethasone (DEX), a well-

known compound that has an anti-inflammatory activity and inhibits the

production of NO. The ability of açaí pulp extracts to inhibit the nitric oxide radical

formation was determined according to Oliveira et al. (2010). A 100 μL portion of

the cell culture medium was added to 100 μL of the Griess reagent (1%

sulfanilamide and 0.1% naphthylethylenediamine in 5% phosphoric acid,

Promega, Fitchburg, USA), and the mixture was incubated at room temperature

for 10 min. The cells were treated with 50 µg/mL of the açaí extracts. The

absorbance was compared against a set of sodium nitrite standards (Promega,

Fitchburg, USA).

The in vitro radical oxygen species or ROS assay was used to evaluate

the capacity of the açaí extracts to decrease the production of ROS in the

stressed cells (Choi et al., 2007a). To this end, the cells were tagged with a

fluorescence dye and then induced with lipopolysaccharide (LPS) and treated

with the açaí extracts. The cell line mouse macrophage RAW 264.7 (ATCC TIB-

71, American Type Culture Collection; Livingstone, USA) was used in this study.

A known antioxidant compound, namely dexamethasone (DEX), was also used

as a positive control. In order to determine the in vitro reactive oxygen species

(ROS) generation, a fluorescent dye protocol was adapted (Choi et al., 2007b).

The RAW 264.7 macrophage cells were maintained as previously described and

seeded in a 24-well plate and incubated overnight at 37 °C. Cells were then

treated with 1 μL of 50 μM solution of dichlorodihydrofluorescein diacetate

acetylester (H2DCFDA, Molecular Probes, Eugene, USA), freshly prepared in

193

sterile phosphate-buffered saline (PBS, Sigma-Aldrich, St. Louis, USA) for 30

min. The fluorescent medium was aspirated, and the cells were exposed to 1 μL

of extract and 1 μL of lipopolysaccharide (LPS, from Escherichia coli 026:B6,

Sigma-Aldrich, St. Louis, USA), incubated for 24 h and the fluorescence of 2′,7′-

dichlorofluorescein (DCF) was measured at 485 nm (excitation) and 515 nm

(emission) on the microplate reader. A 10 μM aliquot of the antioxidant,

dexamethasone (DEX), was used as a positive control.

Biological effect of açaí on radical inhibition assays: Results and discussion

The radical inhibition levels of the açaí extracts were determined by NO

and ROS assays (refer to section 5.5.17). A paired two-tailed t-test (Miller et al.,

2018) confirmed that the null hypothesis was retained, that is, the different

extraction methods did not show a statistically significant difference in the radical

inhibition levels (tcalc= 0.14 < tcrit = 2.12; p=0.8939, n=17, =0.05). Therefore, the

results shown in Figure 5.16 represent the combination of the different extraction

methods. Comparison with the LPS induction study, confirmed that neither of the

açaí pulp samples (purple or white, and non-commercialised or commercialised)

showed any inhibition of the NO production by the extracts (One-way ANOVA; r =

0.33, p=0.2705, n=3, =0.05). Even though the samples did not show any

inhibition of the NO production it was only possible to evaluate that the presence

of the açaí extracts does not have an effect on the inducible nitric oxide synthase

(iNOS). This result contradicts previous studies (Matheus et al., 2003, Matheus

et al., 2006), although these researchers used a herbarium açaí pulp sample

cultivated in a different region than the natural occurrence of açaí.

194

Figure 5.16: Nitric oxide (NO) production in RAW 264.7 macrophage cells

stimulated with lipopolysaccharide (LPS). The cells were treated

with 50 µg/mL açaí extracts and dexamethasone (DEX). The

results are expressed as the mean ± st dev, n=3; n, number of

instrumental replicates.

The in-vitro reactive oxygen species (ROS) generation was determined

following the method described in section 5.5.11. The results shown in Figure

5.17 were normalised by the fluorescence levels of the cells induced by LPS only

and compared with the treated cells. It is clear that all of the açaí samples have a

positive effect on the inhibition of ROS generation, confirming the antioxidant

activity shown in the chemical assays. The data for the purple and white açaí

pulp samples, shown in Figure 5.17, are a combination of the non-commercial

whole and de-fatted samples, because they did not presented any significant

difference using a paired two-tailed t-test (for purple, tcalc= 0.28 < tcrit = 3.18;

p=0.9903, n=4, =0.05 and white, tcalc= 0.14 < tcrit = 3.18, p=0.4669, n=4,

=0.05). When compared with the LPS induced cells, all of the açaí pulp

samples confirmed a statistically significant difference (One way ANOVA, r =

0.55, p=0.0012, n=30, =0.05).

Control

LPSDEX

Purple

White

Comm

erci

al0.0

0.5

1.0

NO

pro

du

cti

on

F

old

in

cre

ase o

ver

LP

S

195

Figure 5.17: Radical oxygen species (ROS) production in RAW 264.7

macrophage cells stimulated with lipopolysaccharide (LPS). The

cells were treated with 50 µg/mL açaí pulp extracts and

dexamethasone (DEX). Results are expressed as the mean ± st

dev, n=3; n, number of instrumental replicates.

Biological effect of wound healing in human cells: Materials and method

The process of human wound healing is very complex, but it can be

divided into three main phases: inflammatory, proliferative and maturation (Wild

et al., 2010). The first occurs after body injury when many different inflammation

processes can occur, such as pain and swelling. The proliferative phase occurs

when the fibroblasts migrate from the tissue to the wound, so as to close the

injury. Then, finally at the last stage, collagen is deposited into the tissue (Wild et

al., 2010). There is no time frame for the duration of each one of these phases

and the body can go back and forward in this healing process, based on different

factors. Positive factors that help to improve or accelerate the wound healing

process are: vitamins A, C and E, iron, zinc and fats (Sanchez and Watson,

2016).

Control

LPSDEX

Purple

White

Pulp S

P

Comm

erci

al S

P

Comm

erci

al U

K0.0

0.5

1.0

RO

S p

rod

ucti

on

F

old

in

cre

ase o

ver

LP

S

196

In this study, the wound healing process was evaluated by investigating

the effect of the açaí pulp extracts on enhancing the proliferation phase of the

wound. The cell line Human Dermal Fibroblast cells (adult) – HDFa (Invitrogen C-

013-5C, Thermo Fisher Scientific Massachusetts, USA) was maintained in

Medium 106 (Invitrogen M-106-500, Thermo Fisher Scientific Massachusetts,

USA) supplemented with Low Serum Growth Supplement – LSGS (Invitrogen S-

003-10, Thermo Fisher Scientific Massachusetts, USA) and 1% of the antibiotic

penicillin /streptomycin solution; 10,000 IU/10,000 μg/mL (Fisher MT-30-002-CI,

Fisher Scientific, Pittsburgh, USA) at a density minimum of 2.5 x 104 viable

cells/mL and was maintained at 37 °C in a humidified incubator with 5% of

carbon dioxide prior to the analysis.

The cell migration assay was performed using the OrisTM Cell 2-D

migration of adherent cells assay kit (AMSBio, Cambridge, USA), where stoppers

were used to simulate the wound, as shown in Figure 5.18. Furthermore, the

cells were dyed with NucBlue® Live Cell Stain (Thermo Fisher Scientific

Massachusetts, USA), which is a reagent that bounds to DNA and can be excited

by UV light at 360 nm, with an emission maximum at 460 nm.

Figure 5.18: Cell migration determined using the OrisTM Cell 2-D migration of

adherent cells assay kit (Oris, 2017).

The stoppers were applied to a 96-well plate, where 50 μL of suspended

dyed cells and 1 μL of the açaí extracts (10 mg/mL) were added to each well,

with the exception for the controls that were ran at the same time. For the blank

197

readings only, media was added to the wells; 10% of foetal bovine serum (FBS,

Thermo Fisher Scientific Massachusetts, USA) was used as a positive control;

and for the full cell readings, the stoppers were never added to the wells,

allowing the cells to migrate. The plate was incubated for 2 hours and the

fluorescence was read, and blank corrected, on the micro-plate reader

SynergyH1 (BioTek, Winooski, USA) at time zero. Moreover, the images of the

cells were evaluated using the microscope EVOS FL Cell Imaging System

(Thermo Fisher Scientific Massachusetts, USA). The cells were further incubated

for another 48 hours and the migration of the cells were evaluated both via

florescence and imaging.

Biological effect of wound healing in human cells: Results and discussion

The wound healing experiment was performed as described in section

5.5.19 and the results are shown in Figure 5.19 as the difference between the

fluorescence values at t = 48 hours minus t= 0 hour of incubation. The results are

normalised to the values of the positive control, 10% FBS, and combine the two

extraction methods (refer to section 5.5.4), because they did not show a

significant difference, based on a paired two-tailed t-test (tcalc= 1.56 < tcrit = 2.30;

p=0.1565, n=9, =0.05). Furthermore, the purple açaí results (a combination of

non-commercial purple whole and de-fatted samples) showed no significant

difference based on using a paired two-tailed t-test, tcalc= 0.20 < tcrit = 2.36,

p=0.8449, n=8, =0.05; and the commercial the pulp SP, commercial SP and

commercial UK samples also did not show a significant difference using a one-

way ANOVA test (r = 0.19, p=0.1148, n=21, =0.05).

198

Figure 5.19: Florescence absorption of the radical oxygen species (ROS)

production in human dermal fibroblast cells (adult). The cells were

treated with 50 µg/mL açaí pulp extracts or 10% FBS (refer to

section 5.5.19). Results are expressed as the mean ± st dev, n=3;

n, number of instrumental replicates.

In order to investigate the migration of the cells, a picture of each sample

was taken, as described of section 5.5.19. Figure 5.20 shows a visual difference

of the migration of the cells between time 0 and after 48 hours of incubation ,

following the treatment with the non-commercial white açaí whole sample. This

was the sample that showed the highest potential of wound healing, as shown in

Figure 5.19 (0.5 fold increase over 10% FBS).

Blank

10%

FBS

Full ce

lls

Purple

White

whole

White

de-

fatte

d

Oil

White

Oil

purple

Comm

erci

al0.0

0.5

1.0

1.5F

lore

cen

ce A

bso

rpti

on

F

old

in

cre

ase o

ver

10%

FB

S

199

Figure 5.20: Fluorescence images of ‘wound healing’ of human dermal fibroblast

cells (adult) between time 0 (A) and after 48 hours of incubation

after treatment with non-commercial white açaí whole sample (B).

The results for this assay demonstrated the potential of the açaí pulp

extracts as a wound healing agent. In contrast to the other experiments reported

above in this study, the samples where the oil was present presented a positive

effect on the wound healing. This indicates that the oil plays an important role in

the wound healing process. Previous studies that have analysed the fatty acid

composition of the açaí oil, reported levels of 60% oleic acid, 22% of palmitic

acid, 12% of linoleic acid, 6% of palmitoleic acid and traces of other fatty acids

(Pacheco-Palencia et al., 2008). It has already been established that fatty acids

play an important function in the migration of cells (Sanchez and Watson, 2016).

A further study has also reported the influence of fatty acids on the acceleration

of the wound healing process (Arnold and Barbul, 2006).

200

5.6. Evaluation of the Amazon Geographical Variability and Industrial Processing on the Chemical Composition of Açaí

Introduction

It has been reported that the elemental content of a material is related to

the origin of the samples and that it is possible to track the origin or source of the

samples via ‘a fingerprint’ of the elemental content (Santos et al., 2014b). Also,

the antioxidant molecules are unstable and their content can possibly change

after the harvesting and processing stages of a sample (Timmers et al., 2017).

Therefore, it was important to analyse samples with a known record of the

harvesting and processing dates and the origin or location of these activities. The

samples listed in section 5.6.2 were analysed for the total polyphenol content, as

outlined in section 2.2 and elemental content (section 2.1). The results (fresh

weight basis) are reported in sections 5.8.1 and 5.8.2. The full set of data for both

dry and fresh weight and their moisture content are presented in Appendices 5.4

to 5.7.

Description of the samples

Samples were collected from a local market and companies along the

Amazonas river delta (Pará State). The fresh açaí berries were collected by

native residents from different areas of the Amazon forest, and at specific time

periods. The local market, as shown in Figure 5.21, is an open-air area in Belém,

that operates daily at dawn, where the local Amazonian people (from the

Amazon forest areas or islands of Belém or along the Amazonian rivers) unload

the berries from their boats for sale in the market.

201

Figure 5.21: (A) Picture of the open-air açaí market and (B) purple açaí fruits in

Belém (PA), Brazil.

The market supplies the city with the açaí berry in natura, which is

immediately processed during the morning. The resultant pulp is either ready for

consumption or frozen and shipped to the rest of the country or abroad.

Information was available for all of the açaí samples included in this study,

including source (Figure 5.22) and processing methods, as shown in Figure 5.23.

The full sample list is reported in Appendix 5.3.

The processed pulp samples were collected from 3 companies that have

different processing methods, as described in Figure 5.23.

202

Figure 5.22: Map of the sources where the açaí samples were harvested by the

native people. Adapted from Google Maps (2019).

The Açaí Amazonas (Company I) has the only açaí plantation of the

country. They have two different sites, one in Mangau, which is an organic

plantation and the other one in Macupixi, where the crop is treated with fertilisers.

Products from this company are usually a blend of the two plantations. As such,

the açaí palm has aerial roots and it normally grows in very humid and muddy

soils close to rivers. In order to keep the roots protected, Açaí Amazonas uses

Mombaça, a type of grass, to cover the roots and they also use an artificial

dripping irrigation system to keep the soil humid. Açaí Amazonas also cultivate

açaí BRS, a variety genetically modified by the Brazilian Agricultural Research

Corporation - Embrapa (Homma et al., 2006). In general, harvesting is still

manual, although the palms are usually pruned to maintain a medium height to

aid harvesting operations. The fruits from these plantations are transferred to the

Açaí Amazonas processing plant where the berries are selected according to the

maturation state and size. Post-selection, the material is washed with ozoned

water in order to make the fruits safe for human consumption. Then, after the

softening of the pulp (or outer layer of the fruit), the berries are transferred to a

specific machine (despolpadeira) where the pulp is taken off from the seeds with

a certain amount of added water. The amount is determined by the market and

Brazilian legislation. The separated thick pulp juice is finally packed and frozen,

203

and the remaining seeds are used as biomass or thrown back into the plantation

as a natural fertiliser.

The ‘Point do Açaí’ (Company II) is well known in Belém city and their

production focusses on the local market of processed açaí for the citizens of

Belém. Every day, they receive freshly harvested native açaí grown in the small

islands close to city. When the fruits arrive at the processing plant, they are

selected and washed in a tank with water, then with a hypochlorite solution in

order to make it safe for human consumption. The final step is to repeat the

procedure using tap water to remove the remaining hypochlorite. After the

washing stage, the fruits are softened by thermal shock and the pulp is extracted

with the addition of a pre-determined quantity of water. The thick pulp is

extracted, and the seeds are sent to another location to be sold as biomass. The

local people normally consume the açaí pulp as part of their meal; therefore, they

usually buy and consume the fresh pulp.

The last company selected was Açaí Santa Helena (Company III), which

is focussed on the Brazilian and international market, shipping their product for

all of the regions of the country and abroad. They usually process the fruits that

were harvested the day before inside the Amazon forest. When the fruits arrive at

the plant, they are first selected and washed 3 times with filtered water, then a

chlorine solution and are again filtered with water to wash the remaining chlorine

from the pulp. The berries are softened, and the pulp is extracted. The seeds are

also sold as biomass and the product is finally packed and frozen. Furthermore,

the international market values more the bright purple colour of the berries than

the natural taste of açaí, therefore following their clients demand, they also could

add citric acid to ‘brighten’ the purple colour of the pulp.

204

Figure 5.23: Summary of the açaí processing steps and the differences between

companies, being Company I: Açaí Amazonas, Company II: Point

do açaí; and Company III: Açaí Santa Helena.

Açaí processing

Harvest

Transport toprocessing plant

Step 1: ozonedwater

Step 2: ozonedwater

Water tank at 45°Cuntil softening

Mechanical separation of pulpand seed

Seed

FertiliserBiomass

Pulp

Frozen pulp

Freezedrying

PowderFresh pulp

Selection

Step 1: Tap water Step 2: Hypochoride

solution Step 3: Tap water

Step 1: Water tank at80°C for 10 seconds Step 2: Water tank

room temperature untilsoftening

Company (I) Company (II) Company (III)

Step 1: Filtered water Step 2: Hypochoride

solution Step 3: Filtered water

Water tank at roomtemperature until

softening

Washing steps

Pulp softeningsteps

Addition of water

Addition of citric acid

205

Total polyphenol content: Materials and method

Total polyphenol content was analysed following the Folin-Ciocalteu assay

described in section 2.2. The freeze-dried açaí samples were extracted using a

0.5% (v/v) acetic acid solution, as presented in section 5.5.4 and analysed using

the micro-plate reader.

Total polyphenol content: Results and discussion

The total polyphenol (TP) content of non-commercial açaí (purple and

whole) samples collected during the field-trip study to the Amazon (April 2018)

are reported in Figure 5.23. The fruit and seed are related to the non-processed

berries acquired directly from the Amazon forest, whist the pulp is the processed

commercialised açaí, bought as frozen pulp.

It is clear that during the processing of the açaí berries, the total

polyphenol levels decrease suggesting that they are lost. This could be because

of the delay between harvest and the processing of the material, the washing or

pulp softening steps (refer to section 5.6.2), or even during the pasteurisation of

the material. The results for the processed samples (pulp in Figure 5.24) are in

agreement with the non-commercial, purple/whole samples reported in section

5.7.1 (Table 5.3) and the literature (Gordon et al., 2012, Augusti et al., 2016).

Interestingly, the non-commercial açaí seeds have detectable levels of total

polyphenols supporting the suggestion of the samples have an antioxidant

capacity, as previously reported in the literature (Rodrigues et al., 2006, Wycoff

et al., 2015, Melo et al., 2016). This information could be useful for the future

development of the non-commercial açaí seeds being an important by-product of

this ‘super-fruit’ industry.

206

Figure 5.24: Box plots representing the total polyphenol (TP) content (gallic acid

equivalent mg/kg f.w.) of açaí extractions using the Folin-Ciocalteu

assay (refer to section 2.2.). The samples relate to the type of

sample (non-commercial, purple/whole: fruit, n= 7; seed, n= 5; and

processed freeze-dried pulp, n= 8; where n is the number of

samples).

Furthermore, the data shown in Table 5.9 relates to the total polyphenol

(TP) levels of non-commercial purple/whole berries: fruit and seeds are non-

processed and the pulp is processed (with the moisture content indicated as

fluid, medium or thick).

These açaí berries, obtained from islands close to Belém (Ilhas), have

higher levels of total polyphenols (average of 39.90 mg/kg f.w.) compared with

similar berries from Genipauba (average of 13.67 mg/kg f.w.). Interestingly, the

local berries from the islands are sold at a higher price once the açaí has been

commercially processed. It has been suggested that this may be due to such

factors as the age of the palm tree, the maturation or the fruit exposure to

sunlight (Timmers et al., 2017).

Fruit

Seed

Pulp0

10

20

30

40

50

To

tal p

oly

ph

en

ol co

ncen

trati

on

mg

/kg

(w

et

weig

ht)

207

Table 5.9: Total polyphenol (TP) content of non-commercial açaí (purple/whole)

samples, (mg GAE / g) determined by Folin-Ciocalteu analysis (refer

to section 5.6.3). Fruit and seeds refer to non-processed berries and

the pulp is processed material (fluid, medium and thick relates to the

moisture content) (refer to section 5.6.2 for sample information).

Results are expressed as mean st dev in fresh weight, n is the

number of replicates, n = 3. Refer to Appendix 5.3 for code

information.

Code Origin* Type Total polyphenol GE-WB-P Genipauba Seed 5.07 ± 0.66

GE-WB-S Genipauba Fruit 13.64 ± 0.74

GE-PB-P Genipauba Seed 3.32 ± 0.22

GE-PB-S Genipauba Fruit 13.69 ± 1.47

IL-PA-P Ilhas Seed 3.62 ± 0.58

IL-PA-S Ilhas Fruit 40.51 ± 0.84

MA-PA-P Macapa Seed 4.48 ± 0.34

MA-PA-S Macapa Fruit 22.21 ± 0.32

AN-PA-P Anajas Seed 3.61 ± 0.18

AN-PA-S Anajas Fruit 25.05 ± 7.15

IC-PA-S Ilhas Fruit 39.28 ± 1.45

IM-PA-S Igarape-Miri Fruit 18.59 ± 2.83

PA-IC-PM Ilhas Pulp (medium) 2.93 ± 0.39

SH-AB-PF Abaetetuba Pulp (fluid) 1.63 ± 0.31

SH-IM-PF Igarape-Miri Pulp (fluid) 1.37 ± 0.06

SH-IM-PM Igarape-Miri Pulp (medium) 1.72 ± 0.10

SH-PA-PE Paragominas Pulp (thick) 3.56 ± 0.26

AA-OB-PM Obidos Pulp (medium) 2.09 ± 0.22

AA-OB-PE Obidos Pulp (thick) 2.77 ± 0.05

AA-OB-FD Obidos Freeze-dried 3.11 ± 0.29

*refer to Figure 5.22.

208

The processed material (refer to section 5.6.2), described as non-

commercial pulp in Table 5.9, varies in the amount of water added according to

Brazilian legislation, as described in section 5.6.2, or if it was freeze-dried directly

in the processing plant (AA-OB-FD). The total polyphenol content increases in

the processed pulp which had less water added and a higher solid content.

Elemental composition: Materials and method

In order to evaluate the total elemental composition of the açaí pulp, the

samples were digested and analysed by ICP-MS, as described in section 2.3.

Elemental composition: Results and discussion

The elemental content (mean of 3 replicates) of the açaí (non-commercial

purple/whole) samples is summarised for Ca, Mg, Mn, Fe, Zn and Cu, as

previously outlined in section 5.7.9. The full dataset is in Appendices 5.4 to 5.7

(mean and standard deviations; mg/kg, fresh and dry weight). In general, the

data for Mg, Mn, Fe, Zn and Cu reported in Table 5.10, usually have higher

levels in the fruits when compared to the corresponded seed material.

Furthermore, the data in Appendices 5.6 and 5.7 also confirms this relationship

for Na, K, Cr and Ni. The toxic element levels for As, Cd and Pb are typically <

0.1 mg/kg fresh weight, which agrees with the values reported for other fruits

(Pendias-Kaabatas, 2010). Table 5.10 reports the mean value for the fruit and

seeds, whilst it should be also stated that there was a considerable degree of

variability in the mineral content at all sites (with standard deviations presented in

Appendices 5.4 and 5.5). It has already been reported that some plants tend to

accumulate various minerals in the flesh of the fruits (Kabata-Pendias, 2010).

Interestingly, the same pattern of higher polyphenol levels in fruit was reported in

section 5.8.1. It has been suggested that iron is ‘coordinated’ to polyphenols in

the açaí fruit (Yoshino and Murakami, 1998).

Comparison of this data for non-commercial purple/whole açaí samples

with literature values (Table 5.1) cannot be undertaken as only data for

209

processed samples have been published (Rogez, 2000, Menezes et al., 2008a,

Unicamp, 2011, Yuyama et al., 2011, Llorent-Martínez et al., 2013, da Silva

Santos et al., 2014, Moreda-Piñeiro et al., 2018).

Table 5.10: Total elemental concentration of açaí (non-commercial, purple/whole)


5.6.4). Fruit and seeds refer to non-processed berries (section 5.6.2).

Results are expressed as a mean, n is the number of replicates, n =

3. Refer to Appendix 5.3 for code information.

Code Origin* Type Ca Mg Mn Fe Zn Cu GE-WB-P Genipauba Seed 71.66 133.94 7.02 4.81 5.15 4.52

GE-WB-S Genipauba Fruit 135.97 196.84 9.35 7.22 7.53 6.35

GE-PB-P Genipauba Seed 158.93 302.06 5.74 8.96 5.95 5.89

GE-PB-S Genipauba Fruit 158.1 334.07 6.34 8.53 6.33 4.27

IL-PA-P Ilhas Seed 673.38 343.63 55.15 6.32 5.94 4.96

IL-PA-S Ilhas Fruit 389.98 372.67 90.09 7.58 9.1 6.97

MA-PA-P Macapa Seed 848.72 407.38 83.95 5.75 5.5 4.4

MA-PA-S Macapa Fruit 480.75 343.59 156.17 7.2 7.6 5.36

AN-PA-P Anajas Seed 315.07 249.74 76.73 4.23 4.59 3.96

AN-PA-S Anajas Fruit 217.68 286.40 175.26 5.78 7.37 5.18

IC-PA-S Ilhas Fruit 378.57 297.16 105.97 6.5 5.31 4.6

IM-PA-S Igarape-Miri Fruit 395.12 312.00 45.24 45.94 7.49 6.99

*refer to Figure 5.22.

In this study, it was difficult to predict the effect the geographical origin of

the non-processed açaí has on the mineral content of the material. Yuyama et al.

(2011) suggested that several factors may influence the chemical composition,

such as: (i) the impact of tree growth in different environments, namely, aquatic,

terrestrial and floodable; (ii) climatic changes; (iii) soil composition; (iv) periodic

flooding in floodplain areas; or (v) cycles of agricultural production. This can be

confirmed by an inspection of the data in Table 5.10. As such, the fruit and seeds

collected from Genipauba have approximately 10-fold lower Mn (and Ca) levels

when compared with samples from the other locations. Furthermore, inspection

of the other elemental data in Appendices 5.6 and 5.7 shows variable patterns for

210

the other elements. In general, Genipauba is a community located next to the

Amazon river, as is Ilhas, Anajas and Igarape-Miri (refer to the map in Figure

5.22). Only Macapa is sited in the river estuary. Therefore, the factors reported

by Yuyama et al. (2011) support the fact that geographical effects are complex

and do not provide a clear guide on mineral uptake by açaí berries.

Table 5.11 reports the mineral levels for processed purple/whole açaí

(sampled from the same geographical location) and produced by three

commercial processing plants, namely, Point Açaí (code PA), Açaí Amazonas

(code AA) and Açaí Santa Helena (code SH) (refer to section 5.6.2 for details on

the different processing methods). In terms of the processing method the only

difference is the amount of water added during the pulp extraction step.

Inspection of this data shows that the amount of added water seems to have an

effect on the mineral content of the non-commercial açaí samples. The data

showed that materials with less water added are more concentrated (higher

solids content) and therefore have higher mineral or elemental levels (based on

fresh weight).

The manganese content of the açaí (pulp) samples ranged from 5.26 to

126.26 mg/kg (f.w.). These values are higher than the range of values reported in

the literature for açaí pulp (3.43 to 32.3 mg/kg f.w, refer to Table 5.1), although

the samples analysed in the literature are usually commercial samples with

additives (refer to Table 5.8).

The iron levels of the açaí (pulp) samples ranged from 2.29 to 8.96 mg/kg

(f.w.), with the exception of sample IM-PA-S, which had a much higher level of

45.94 0.56 mg/kg Fe. The values (when converted to a dry weight basis using

the conversion moisture values – refer to section 5.8) are similar to the range of

values presented in the literature (refer to Table 5.1), which have been reported

on a dry weight basis. The iron content did not show a significant trend between

the geographical origin or processing method. The same trend is found for

copper (1.19 – 6.99 mg/kg), zinc (1.93 – 9.1 mg/kg) and magnesium (133.94 –

407.38 mg/kg). The calcium concentration had a large variation ranging from

211

71.66 to 848.72 mg/kg (f.w.). Although samples from the same origin had similar

calcium levels. These range of concentrations were also reported in the literature

(refer to Table 5.1). Furthermore, as stated previously, the levels of arsenic,

cadmium and lead do not present a significant contribution to the chemical

composition of the açaí samples (typically <0.1 mg/kg f.w.).

Table 5.11: Total elemental concentration in commercially processed açaí


5.6.4). Results expressed as a mean, n is the number of replicates,

n = 3. Refer to Appendix 5.3 for code information.

Code Origin* Type Water (%)

Ca Mg Mn Fe Zn Cu

PA-IC-PM Ilhas Pulp 89 276.45 186.31 82.67 3.85 2.54 1.51

SH-AB-PF Abaetetuba Pulp 92 178.76 148.10 31.46 2.29 1.94 1.36

SH-IM-PF Igarape-Miri Pulp 92 241.86 143.68 38.39 2.44 2.46 1.19

SH-IM-PM Igarape-Miri Pulp 89 402.81 212.23 93.03 3.33 3.11 1.73

SH-PA-PE Paragominas Pulp 86 402.54 265.47 25.41 4.73 3.47 2.59

AA-OB-PM Obidos Pulp 89 439.36 171.47 5.26 4.35 1.93 1.59

AA-OB-PE Obidos Pulp 86 562.68 216.03 15.01 4.90 2.15 1.67

AA-OB-FD Obidos FD** 86 525.45 246.16 46.74 5.71 3.24 2.44

*refer to Figure 5.22; **freeze-dried (FD) at the processing plant: PA- Point Açaí; SH - Açaí Santa Helena, and AA - Açaí Amazonas.

5.7. Link to Dietary Intake of Total Polyphenols and Minerals of Açaí

One of the aims of this research was to evaluate the contribution that the

consumption of açaí might make to the dietary intake of total polyphenols and

minerals. A previous study suggested that a 500 g (fresh weight) serving of açaí

would equate to the amount of the product that is consumed on a daily basis by

the Brazilian population (Heinrich et al., 2011). The results presented in this

study are related to the dry weight basis of the material which has been

converted to fresh weight using the conversion factor reported during the drying

process, i.e. 90% of the moisture weight is lost (Pavan et al., 2012). Therefore, in

212

order to calculate the daily intake of açaí, the fresh weight was taken into

consideration in the calculation.

The açaí pulp provided 50.9 to 424.3 mg GAE/100 g fresh weight of total

polyphenols. The content of typical berries is presented in Table 5.5. The range

of total polyphenol content of açaí pulp is similar to that reported for cranberries,

raspberries, blueberries and strawberries (Table 5.5). It is difficult to estimate a

recommended daily intake for total polyphenols, due to the variation in the levels

of the phenolic compounds in a particular foodstuff. This is due to the structural

diversity of the phenolic compounds or the lack of standardised analytical

methods (Scalbert and Williamson, 2000). Even though there may be variations

in the dietary intake of phenolic compounds between geographical regions and

consumption age groups, a selection of previous studies have proposed a range

of 1 g of total polyphenols per day (Kühnau, 1976, Faller and Fialho, 2009,

Landete, 2013). Most authors refer to Fukushima et al. (2009), who based on a

balanced Japanese diet, calculated a daily consumption of 1492 mg (fresh

weight) of polyphenols. In terms of the serving values reported by Unicamp

(2011), a 500 g serving of commercial açaí pulp (fresh weight) would contribute

17 to 142 % of the daily intake of total polyphenols.

The recommended daily intake (RDA) is defined by the World Health

Organisation or WHO (WHO, 1996) for males (M) and females (F) and is

presented in Tables 5.9 (total polyphenols and minor elements) and 5.10 (trace

elements). The WHO RDA guidelines are compared with the calculated % intake

of the chemicals for the consumption of açaí pulp; based on using the mg/day

level (reported in Appendix 5.8) calculated from the data in Tables 5.12 (total

polyphenols) and 5.13 (elements).

213

Table 5.12: Percentage intake (%) of total polyphenol and minor elements based

on the consumption of a 500 g serving (Heinrich et al., 2011) of the

commercial and non-commercial açaí pulp (reported on a fresh weight

basis). The data is compared with the World Health Organisation

recommended daily allowance (RDA) for males (M) and females (F).

Total polyphenol Ca Mg M F M F M F

WHO RDA (mg/day)* 1492 1300 400 310

Non-commercial

Purple whole 107.2 18.0 29.3 37.6

Purple de-fatted 132.0 20.3 34.1 43.9

White whole 31.5 16.0 30.9 39.8

White de-fatted 39.2 20.2 37.7 48.8

Commercial

Pulp SP 94.8 6.4 25.3 32.6

Powder SP 142.1 7.8 25.5 32.9

Powder UK 17.1 2.6 10.7 13.9

* World Health Organisation Recommended Daily Allowance (reported in mg/day) (WHO, 1996) and used to calculate the % intake from the data for total polyphenols and elements (Appendix 5.8; mg/day).

The açaí berries represent a potential source of Ca, Mg, Mn, Fe, Zn and

Cu (with the exception of the UK commercial açaí powder product). It has already

been noted that this product seems to have been adulterated, that is, has much

lower chemical levels than local commercial and non-commercial samples,

suggesting the addition of other non-nutrient ‘bulking’ material to the powder

(refer to sections 5.7.3 and 5.7.9).

It is important to highlight that even though açaí could be considered a

source of iron, recent studies have shown that the element is not bioavailable in

açaí. As reported by Toaiari et al. (2005), 40 anemic rats were fed with a non-

commecial sample of purple açaí and commercial feed for 7 days. The rats

developed anemia which was linked to a depletion of dietetic iron. The

concentration of hemoglobin was measured after the experiment and showed

214

that the rats fed with açaí did not show any significant increase when compared

with the commecial feed, since the recovery of hemoglobin concentration from

anemic rats was not observed. A clinical study based on children also proved the

low bioavailability of iron in açaí (Yuyama et al., 2002). In this study, 85 children

aged from 2 to 6 years old were fed for 120 days with açaí and iron aminoacid

chelate as a source of iron. The results demonstrated the impact of açai as an

energy source with a significant weight gain in the children (1.76 kg). However,

the recovery of the anemic children was higher in the group that received iron

aminoacid chelate (reduction of 34%) in comparison with açaí (reduction of 11%).

The low bioavailability of iron in açaí may be explained due to the high

concentration of tannins and fibers in the berries (Yuyama et al., 2002). However,

the consumption of açaí pulp or powder, along with other sources of vitamin C

(such as citrus), could potentially increase the bioavailability of iron (Silva et al.,

2004). The bioavailability of the chemicals present in açaí may be influenced by

the chelation of a metal-polyphenol complex. Moreover, the total polyphenol data

reported in Table 5.3 and the manganese data reported in Table 5.7(a) suggests

a possible correlation, where the respective concentrations potencially increase

together (Spearman’s correlation; R= 0.36, p=0.43 n= 7 and = 0.05). This

correlation could influence the bioavailability of both chemicals.

Also based on the low levels of the toxic elements, the daily intake of As

and Pb were not significant (refer to Appendix 5.2). The toxic element Cd

represented an average of 0.0035 mg/day, which does not provide a significant

contribution to the Brazilian average dietary intake of 1.31 g of Cd (Avegliano et

al., 2011).

As was stated in section 5.7.10, one of the aims of this research was to

evaluate the contribution that the consumption of açaí might make to the dietary

intake of total polyphenols and minerals. As was explained for the processed

samples, the same evaluation is presented for the non-processed material (fruits,

seeds and pulp) using a 500 g (fresh weight) serving of açaí that equates to the

amount of the product that is consumed on a daily basis by the Brazilian

215

population (Heinrich et al., 2011). Appendix 5.9 provides the calculated % dietary

intake of all the elements based on a daily consumption of 500 g of açaí (pulp,

purple/whole, fresh weight basis) relative to the recommended daily intake

defined by the World Health Organisation (WHO). The results, presented as box

plots in Figure 5.25, are for the minor elements, Ca, Mg, Fe and Zn. Figure 5.26

provides the data for Mn and Cu. Similarly, the daily intake of total polyphenols

are also presented in Figure 5.26, based on the recommendations of Fukushima

et al. (2009).

Table 5.13: Percentage intake (%) of selected trace elements based on a 500 g

serving (Heinrich et al., 2011) of non-commercial or commercial açaí

pulp or powder (fresh weight). The data is compared with the World

Health Organisation recommended daily allowance (RDA) for males

(M) and females (F).

Mn Fe Zn Cu M F M F M F M F

WHO RDA (mg/day)* 2.3 1.8 8 18 11 8 0.9

Non-commercial

Purple whole 1392.7 1779.5 18.8 8.4 11.3 15.6 100.6

Purple de-fatted 1759.2 2247.7 26.1 11.6 14.7 20.3 123.4

White whole 1329.2 1698.3 22.8 10.2 12.0 16.6 95.3

White de-fatted 1758.4 2246.8 26.9 12.0 16.4 22.6 138.7

Commercial

Pulp SP 581.8 743.4 18.8 8.3 12.3 16.9 84.6

Powder SP 1189.2 1519.6 13.7 6.1 10.6 14.5 83.2

Powder UK 35.3 45.1 1.4 0.6 5.8 7.9 19.5

* World Health Organisation Recommended Daily Allowance (reported in mg/day) (WHO, 1996) and used to calculate the % intake from the data for total polyphenols and elements (Appendix 5.8; mg/day).

The data in Figure 5.25 (and Table 5.13) clearly shows a variation in

the % intake of the minor elements relative to the RDA. This is true for both

males and females. This is not surprising as the elemental levels for this açaí

product also covered a wide range, as reported in section 5.8. In general, the

216

açaí pulp samples are potentially a good source of minor elements, with the

order of % intake based on the RDA being: magnesium (both male and female) >

iron (male) > zinc (both M and F) > calcium. In addition, the % intake of sodium

(1 to 5 % for both M and F) and potassium (~9% for M and F) are low. In terms of

the essential trace element % intake of Mn and Cu are very high, being > 1000 %

for both gender groups (Table 5.13). The levels for the other trace elements are

very low, with Cr (8% for M and 11% for F), and for Co, Se and Mo (having 0%).

Figure 5.25: Box plots representing the percentage intake (%) of minor elements

based on the consumption of a 500 g serving of açaí pulp

(purple/whole) (fresh weight). The data is compared with the World

Health Organisation (WHO) recommended daily allowance (RDA)

for males (M) and females (F). Note: WHO provide no gender data

for Ca.

Figure 5.26 confirms that the daily consumption of 500 g of fresh weight

açaí pulp (purple/whole) would contribute to about 350 % of the intake for total

polyphenols. When compared to other berries and typical fruits from Brazil the

total polyphenol content of açaí is similar to other well-known ‘super-fruits’, such

as, blueberries and acerola, contributes more than 100% to the recommended

daily intake of total polyphenols (refer to Tables 5.4 and 5.5).

Ca Mg (M) Mg (F) Fe (M) Fe (F) Zn (M) Zn (F)0

20

40

60

80

% In

take o

f R

DA

in

aça

í (5

00 g

serv

ing

)

217

It is important to highlight that even though this study reports the potential

intake of total polyphenols and elements for all açaí samples, only the processed

açaí is available for consumption due to food safety (refer to section 5.2.2).

Figure 5.26: Percentage intake (%) of total polyphenol (TP) and trace elements

of 500 g serving of açaí (fresh weight) in relationship to the

recommended daily allowance (RDA) for males (M) and females

(F).

5.8. Conclusion

This chapter reviewed the chemical characterisation, antioxidant and

biological activities of different samples of açaí. Furthermore, this study also

evaluated the potential contribution of the açaí pulp to the dietary intake of

polyphenols and elemental nutrients. This research provided for the first time a

comprehensive investigation of the polyphenol and trace element composition of

non-commercial and commercial; processed and non-processed; white and

purple; whole and de-fatted samples of açaí obtained from Brazil.

The total polyphenol content of non-commercial and commercial açaí

samples were analysed by Folin-Ciocalteu assay (refer to section 5.5.5) and

were found to range from 32.00 to 39.40 mg/g d.w. for non-commercial purple

TP Mn (M) Mn (F) Cu0

2000

4000

6000

% In

take o

f R

DA

in

aça

í (5

00 g

serv

ing

)

218

samples; 5.09 to 42.40 mg/g d.w. for commercial purple samples and 9.40 to

11.70 mg/g d.w. for non-commercial white samples (refer to section 5.7.1). The

purple non-commercial samples had a significantly higher concentration of total

polyphenols in comparison with the white samples. Moreover, the commercial

purple samples had a higher level of variability (standard error of the mean of

9.26 GAE mg/g) when compared with the non-commercial purple açaí berries

(2.90 GAE mg/g). The removal of the oil fraction also showed a significantly

higher concentration on the total polyphenol content for the non-commercial

samples as demonstrated on section 5.7.1. The total flavonoid content analysed

by the AlCl3 assay (refer to section 5.5.6) of açaí pulp extractions covered a

range from 6.39 to 8.05 mg/g d.w. for non-commercial purple samples; 1.88 to

6.07 mg/g d.w. for commercial purple samples and 2.12 to 2.38 mg/g d.w. for

non-commercial white samples. This flavonoid analysis shows a similar pattern to

the total polyphenol content data, as would be expected because the flavonoids

are a subclass of polyphenols (refer to section 5.7.2)

When comparing the total polyphenol content of non-processed açaí

samples (non-commercial, purple and whole), presented in section 5.8.1; it is

clear that during the processing of the açaí berries, the total polyphenol level

decreases, suggesting that they are lost. The non-processed açaí fruits ranged

from 13.64 to 40.51 mg/kg f.w.; the non-processed seed from 3.32 to 5.07 mg/kg

f.w. and the processed pulp from 1.37 to 3.56 mg/kg f.w.. The açaí samples from

the islands close to Belém (Ilhas), had higher levels (average of 39.90 mg/kg

f.w.) when compared to samples from Genipauba (average of 13.67 mg/kg f.w.).

Also, the total polyphenol content increased in the processed açaí pulp that has a

higher solids content.

The total anthocyanin profile and levels were analysed by HPLC (refer to

section 5.5.8). The results showed that cyandin-3-glucoside and cyandin-3-

rutinoside are the major anthocyanins in the purple samples and were not

present in the white açaí samples and one of the commercial samples (Powder

UK). This suggests that the Powder UK sample might be adulterated (refer to

section 5.7.3). The total anthocyanin content for the non-commercial purple açaí

219

samples ranges from 102.0 to 143.3 mg/100 g f.w. and for the commercial purple

açaí samples 35.9 – 47.0 mg/100 g f.w. The total proanthocyanidin (PAC)

content of the açaí samples were determined using the DMAC assay (refer to

section 5.7.4). This assay was the only one that presented a significant

difference for the extraction method. The aqueous extraction resulted in

significantly higher PAC concentrations in comparison with the methanolic

extraction, due to the oligomeric nature of the PAC compounds. The PAC for the

non-commercial purple açaí samples ranges from 5.06 to 6.10 mg/g d.w.; the

commercial purple samples from 4.47 to 5.48 mg/g d.w. and for non-commercial

white samples 2.60 – 3.96 mg/g d.w. (refer to section 5.7.4).

The antioxidant activity levels were analysed by ABTS and DPPH assays

(refer to section 5.5.6). The non-commercial purple açaí was the most prominent

of all the antioxidant analysis when compared to the commercial purple and non-

commercial white samples, due to its higher levels of total polyphenols and total

anthocyanins. Furthermore, commercially purple sample (Powder UK) had a

significantly lower activity when compared to the other commercial purple

samples (refer to section 5.7.5). The strong antioxidant effect of the açaí samples

were confirmed on cells through the inhibition on the radical oxygen species

production (refer to section 5.7.7) where all the açaí samples confirmed a

statistically significant effect on the inhibition of the ROS generation. Although,

the samples did not show any inhibition of the NO production, contradicting the

literature (refer to section 5.7.7). Furthermore, neither of the açaí samples

demonstrated any toxicity effect on cells, even at the higher concentration of 250

µg/mL as demonstrated on section 5.7.7. The wound healing experiment was

performed on human fibroblast cells (refer to section 5.5.14). The visually

demonstrated migration effect of the açaí extracts on this cells proved açaí as a

potential wound healing agent. In addition, the samples where the oil was

present presented a positive effect on the wound healing.

The total elemental concentrations were determined in the açaí samples

(refer to section 5.6.4). The calcium, magnesium, manganese, iron, zinc and

copper; essential minor and trace elements found in significant levels in açaí

220

samples were highlighted in section 5.7.9. Overall, for both white and purple non-

commercial açaí samples, all of the elemental values for the de-fatted samples

are slightly more concentrated than the whole material due to the removal of the

oil fraction, as demonstrated before. The enhancement of total polyphenols in the

defatted samples is on average 28% for purple and 38% for white samples.

There was no statistically significant difference between purple and white

samples. In contrast, the commercial sample (Powder UK) had elemental levels

typically 30% lower than other commercial samples. The commercial processing

of açaí reduced the concentration of Ca and Mn by about 50% and the

commercial pulp had a significant variance of these elements. The manganese

levels were on average 703.25 mg/kg d.w. for non-commercial samples and

270.5 mg/kg d.w. for commercial purple samples. Similarly, calcium levels were

4.75 g/kg d.w. for non-commercial and 1.42 mg/kg d.w. for commercial purple

samples. Magnesium (2.59 and 1.60 mg/kg d.w.); iron (36.81 and 17.69 mg/kg

d.w.); zinc (29.4 and 20.47 mg/kg d.w.); and copper (20.20 and 10.99 mg/kg

d.w.). The toxic elements, arsenic, cadmium and lead were found to be at very

low levels in the commercial and non-commercial; purple and white açaí products

(typically <0.01 mg/kg f.w.).

This study also provides elemental data for non-processed açaí samples

(non-commercial, purple and whole), as presented in section 5.6. Overall, the

data for Mg, Mn, Fe, Zn, Cu, Na, K, Cr and Ni are higher in the non-processed

fruits when compared to the corresponding seed material. The fruit and seeds

collected from Genipauba presented approximately a 10-fold lower Mn (and Ca)

level when compared with samples from the other locations. The data showed

that materials with less water added (higher solids content) had higher mineral

levels (based on fresh weight). The manganese content of the açaí (pulp)

samples ranged from 5.26 to 126.26 mg/kg f.w., iron from 2.29 to 8.96 mg/kg

f.w.; copper (1.19 – 6.99 mg/kg), zinc (1.93 – 9.1 mg/kg) and magnesium (133.94

– 407.38 mg/kg) and calcium had a large variation ranging from 71.66 to 848.72

mg/kg (f.w.). The toxic element levels for As, Cd and Pb were typically < 0.1

mg/kg f.w (refer to section 5.8.2).

221

One of the aims of this research was also to evaluate the contribution that

regular consumption of açaí (500g) could make to the dietary intake of

polyphenols and minerals. The total polyphenol intake of purple açaí is similar to

other well-known ‘super-fruits’, such as blueberries and acerola, contributing to

the recommended daily intake of total polyphenols by more than 100% in a single

serving. In relation to the mineral levels, açaí may have a great potential as a

source of manganese (average of 1500%) and copper (average of 90%); along

with calcium (20%), magnesium (30%) and zinc (15%). However, it is known that

the form of iron present in açaí may not be bioavailable to humans (Yuyama et

al., 2002). Also, the low levels of Cd, As and Pb did not present a significant

contribution to the elemental daily intake (refer to sections 5.7.10 and 5.8.3).

222

Chapter 6. Conclusions and Future Work

223

6.1. Overview

Brazil is a major producer of special natural foods and beverages that are

commercialised and sold as natural and processed products. Many are marketed

as being good sources of nutrients, namely, elements (or in terms of nutrition are

referred to as minerals) and polyphenols, that play an important role in human

health (Torres and Farah, 2017). Figure 1.1 summerised many of the main

products that are being consumed, especially throughout Brazil. At present, very

few scientific and health-related studies have been reported on the chemical

composition of these natural foods or beverages using locally sourced materials

in Brazil. Moreover, Tables 3.1 (yerba mate), 4.1 (roasted coffee) and 5.1 (açaí)

listed the data that has been published for selected elements (primarily Mg, Ca,

Mn, Fe, Cu and Zn). In terms of a the critical analysis of these studies, it is

difficult to make an estimation of the daily intake of such chemicals when there is

a general lack of reported analytical details (including the sampling protocols,

weight basis of calculated data, lack of validation procedures and instrumental

details – limit of detection, linear dynamic range of the calibration standards and

dilution factors). Furthermore, many of the published studies focus on only a

particular food or beverage and provide data for specific chemicals (Santos et al.,

2014a, Stelmach et al., 2015, Rusinek-Prystupa et al., 2016).

The aim of this research was to determine the levels of chemical elements

(major, minor and trace) and polyphenols of yerba mate (from the southern

region of Brazil), roasted coffee (São Paulo state) and açaí berries (from the

Amazon region). This data was then used to assess the impact of consumption in

terms of the daily dietary intake in providing an adequate nutrient supply for

Brazilians. As was stated in chapter 1, section 1.5, a main feature of this

research was to establish an analytically robust method(s) that enabled an

evaluation of the impact of sample selection and treatment; the influence of

dilution in prepared sample solutions before instrumental analysis; and the

calculation, statistical analysis and reporting of data in relation to what is

traditionally consumed by individuals in Brazil.

224

In summary, the aim and objectives of this research (refer to section 1.5)

have been meet for the scientific evaluation of Brazilian and Argentinian yerba

mate (plant material, commercial products and infusions) grown in southern

Brazil or the Misiones region of Argentina (chapter 3); roasted coffee beans and

infusions from samples grown in Amparo, São Paulo State (chapter 4); and açaí

fruits (pulp and seeds) and processed pulp collected from the Amazonian region,

northern Brazil (chapter 5).

6.2. Yerba Mate

Chapter 3 presented the chemical analysis of different samples of yerba

mate. This research provided for the first time a comprehensive investigation of

the elemental and polyphenol composition of non-commercial, non-processed

and processed samples of yerba mate from Brazil and Argentina. Manganese

was an element highlighted in this study (section 1.2.5) due to its higher

concentration in the yerba mate samples and prepared infusions (Table 3.1). No

previous study has assessed what the potential intake of Mn would be through

the consumption of different yerba mate infusions in terms of the total dietary

intake of this element by the South American population (section 3.6.6).

The first study focussed on the elemental composition of non-commercial

yerba mate leaves from the Barão de Cotegipe plantation in southern Brazil -

with the field-trip undertaken in April 2017 (section 3.5). Overall, in terms of the

age of the leaves collected from traditional plantations (refer to section 3.5.3), the

general trend was for the elements to be at higher levels in the old leaves (new <

old; refer to Table 3.2). Statistical analysis using a two-tailed Student t-test

confirmed that there was a significant difference between the age of the leaves

(p<0.05) collected from the plantations for Mg, Ca, Fe and Cu (Table 3.3).

Moreover, the new yerba mate leaves grown on trees located in an organic

plantation had higher levels of most of the elements, especially for Mn, when

compared with other plantations treated with NPK fertilisers (new leaves; fert <

org, Table 3.2). Once again, statistical analysis using two-tailed Student t-test,

225

confirmed a significant difference (p<0.05) between the age of the leaves from

the same tree, for Ca and Mn. Interestingly, for the older leaves, all of the

elements were higher in plants grown in the fertilser-added plantation (old leaves;

fert > org; Table 3.2). A two-tailed Student t-test confirmed a significant difference

(p<0.05) for the Fe values reported in Table 3.2, which are higher in the fertilsers-

added plantation. An evaluation of the elemental levels of leaves collected at

different heights of a yerba mate tree grown in an organic plantation confirmed a

translocation of manganese to the upper parts of the tree; Mn concentrations in

the tree: bottom 1070 ± 4; middle 1009 ± 82 and 1229 ± 21 top mg/kg, dry weight

(d.w.); Table 3.5. Higher elemental levels were also found in yerba mate plants

grown in the traditional organic plantations rather than in the natural forests

(Table 3.6). In relation to the processing of the yerba mate material, the elements

showed higher levels after the sapeco stage (leaf exposed to an open fire for a

short period of time) of the process (refer to section 3.5.5, Table 3.8).

Secondly, the elemental composition of commercial yerba mate products

from Brazil and Argentina was evaluated in terms of the type of sample (green or

roasted), processing package (loose or tea bags) and origin (Brazil or Argentina)

(refer to section 3.6.3). In summary, the elemental levels were similar for the two

countries (Table 3.9), which is to be expected as yerba mate is primarily grown

between the Paraná and Paraguay river basins in South America (Figure 3.1).

The Mn levels for the green loose yerba mate samples from Brazil ranged from

486 to 834 and the Argentinian samples from 383 to 671 mg/kg (d.w.). In terms

of the method of packaging, all of the elements measured had higher elemental

levels for the tea bag products when compared to the green loose material

(Table 3.9). The Mn values were found to range from 483 to 1098 mg/kg (d.w.)

for the tea bag samples and 383 to 671 mg/kg (d.w.) for the green loose material.

The roasted (loose and tea bag) samples also had higher elemental levels when

compared with the Brazilian green loose material (Table 3.9). The Mn levels of

the roasted samples ranged from 397 to 889 mg/kg (d.w.) where for the green

loose samples the values were 486 to 834 mg/kg (d.w.).

226

In order to evaluate the chemical intake of yerba mate, different infusion

methods were prepared (refer to section 3.6.2). The regular infusions (prepared

using a tea-based method) of yerba mate (green loose) from Brazil had slightly

higher chemical levels (Tables 3.11, 3.16 and 3.19). All of the elements (Table

3.11), polyphenols and xanthines (section 3.6.5) had higher chemical levels for

the infusions prepared using tea bag samples when compared to the green loose

regular infusions. Moreover, regular infusions made with green loose yerba mate

had significantly higher levels of trace elements, polyphenols and xanthines in

comparison with the roasted samples. Statistical analysis using a two-tailed

Student t-test showed a very highly significant difference (p < 0.0001) between

the total polyphenol content of green and roasted yerba mate samples (Table

3.16).

Finally, an evaluation of the potential intake of minor or trace elements

and polyphenols through the consumption of yerba mate products was

undertaken in section 3.6.6. The trace elements analysed in this study

represented 0.1 to 5% of the recommended daily allowance (RDA) (WHO, 1996),

for all of the infusion methods (regular, Brazilian and bombilla). The exception

was found for manganese. A regular infusion serving (1 cup of 200 mL) would

provide 23.7 to 106.0 % for males and 30.3 to 135.5 % for females of the

manganese RDA, depending on the type of yerba mate product (refer to Table

3.22). Although, the bombilla method (traditional method of consumption in South

America – section 3.2.3) involved a serving of 1 litre of drink, this would provide

up to 1482 % for males and 1894 % for females of the manganese RDA. The

impact of consuming such high levels of manganese may be influenced by the

bioavailability of Mn in the various infusions (as discussed in section 6.5). In

terms of the total polyphenol intake, a regular infusion serving (200 mL) could

contribute 4.0 to 14.5 % of the daily intake and the bombilla method can provide

up to twice the amount of the adequate daily intake of total polyphenols (refer to

Table 3.23).

227

6.3. Chemical Composition of Roasting Brazilian Coffee

Chapter 4 investigated for the first time, the roasting process of Brazilian

coffee (based on samples and a process undertaken at a local plantation in

Brazil) and the impact on the chemical composition of the roasted beans and

related beverages. The effect of roasting different coffee varieties (Obatã, Catuaí,

Bourbon Amarelo and blend) collected from the Fazenda Palmares and Flor

plantations (Amparo, São Paulo State) resulted in a slight increase of the

elemental content of the beans during the roasting process (Tables 4.2 to 4.4).

The manganese values ranged from 18.32 (t = 0 minutes) to 26.60 (t = 0

minutes) mg/kg (d.w.) for the roasting process of a Bourbon Amarelo coffee

variety (Table 4.4). This suggests that a small decrease in the moisture content

may result in a ‘pre-concentration’ of the elements in the beans as a result of the

roasting process. The data also highlighted that there are no elemental losses

during the roasting process (Tables 4.2 to 4.4). The total polyphenol content of

coffee infusions, produced from beans collected at different times of the roasting

process, showed 7 to 52 % higher levels in the dark roast (10 minutes) when

compared to the green bean infusions (0 minutes, Table 4.6). The chlorogenic

acid and caffeine data showed a similar trend with an increase in the levels of the

infusions prepared using the medium roast coffee (Figures 4.4 to 4.7).

Furthermore, the effect of the particle size of the ground beans on the levels of

chlorogenic acids and caffeine in the coffee infusions was investigated. The data

showed that the levels were inversely proportional to the particle size, confirming

that the grinding stage influences the extraction of the chemicals during the

preparation of a coffee infusion (Figures 4.8 and 4.9). The manual selection of

coffee beans after the roasting process (10 minutes) was undertaken based on

quality factors (refer to section 4.3.2). This procedure is undertaken by the

manufacturer to enhance the chemical quality of the coffee product. The data

showed that manual selection does not result in any major difference in the

elemental composition or chlorogenic acid levels of the roasted coffee products

(Tables 4.5 and 4.7). However, a different finding was found for the coffee

228

infusions prepared using selected (higher quality) beans, namely, higher total

polyphenol levels (Table 4.7). The physical changes on the structure of coffee

beans during the roasting process were investigated and scanning electron

microscope images showed an expansion of the coffee bean and an increase in

the size of the bean micropores (Figures 4.10 (A) to (F)). Finally, an evaluation of

the potential intake of polyphenols from coffee infusions showed that a cup of

coffee (92 mL) can contribute up to 7 % of the estimated daily intake of

polyphenols. Elemental data was not obtained for this specific study due to the

coffee infusions causing instrument signal instability for the inductively coupled

plasma mass spectrometer (section 4.7).

6.4. Açaí

Chapter 5 reported the chemical characterisation, antioxidant and

biological activities of different samples of açaí (berries, seeds and processed

pulp) and provided for the first time a comprehensive investigation of the

polyphenol and minor or trace element composition of non-commercial and

commercial; processed and non-processed; white and purple; whole and de-

fatted samples of açaí (refer to section 5.5.2 for the classification of these terms).

In this specific study, priority was given to the polyphenol analysis

because of the investigation of the relationship between the chemical

composition and biological activity. This research was undertaken at North

Carolina State University (USA) (section 5.5). The non-commercial purple mature

berries (that differ from white ones which are a different variety) had a

significantly higher (two-tailed Student t-test, p<0.001) concentration of total

polyphenols in comparison with the white samples (Table 5.3 and Figure 5.8).

The non-commercial purple samples also had a higher range of the total

flavonoid content (Table 5.3). The total anthocyanin profile showed that cyandin-

3-glucoside and cyandin-3-rutinoside are the major anthocyanins in the purple

açaí pulp samples, although the white berries had inconsistent values (Figure

5.10). The total anthocyanin content was higher for the non-commercial

229

processed purple açaí pulp when compared to the commercial purple açaí

samples (Table 5.3). The total proanthocyanidin (PAC) analysis resulted in

significantly higher PAC concentrations in the aqueous extraction (paired two-

tailed t-test, p<0.005), due to the oligomeric nature of the PAC compounds

(Figure 5.11). The non-commercial purple açaí pulp was found to be very

important in terms of the antioxidant analysis (ABTS and DPPH assays, refer to

section 5.5.12). This is due to the higher levels of total polyphenols and total

anthocyanins when compared to the commercial purple and non-commercial

white pulp samples (section 5.5.12). The strong antioxidant effect of the açaí

samples was confirmed using mouse cells through the inhibition of the production

of radical oxygen species (refer to section 5.6.3). This study involved all of the

açaí pulp samples and confirmed a statistically significant effect on the inhibition

of radical oxygen species (ROS) generation (Figure 5.17). A wound healing

experiment was performed using human fibroblast cells. The data confirmed a

migration effect on human cells subjected to açaí pulp extracts. These results

are very important, as such an experiment has never been reported, and implies

that processed açaí pulp may have potential as a wound healing agent (Figures

5.19 and 5.20).

The calcium, magnesium, manganese, iron, zinc and copper levels in açaí

samples were found to be higher than that reported in the literature for other

typical Brazilian fruits (based on TACO values) (Table 5.7 (a) and Figures 5.25

and 5.26) (Unicamp, 2011). Overall, the elemental values for the de-fatted açaí

samples were found to be more concentrated than the whole material due to the

removal of the oil fraction. It is suggested that this results in a ‘pre-concentration’

of the elements in the processed pulp (Table 5.7 (a)). Interestingly, there was no

statistically significant difference in the elemental content of the purple and white

samples (Table 5.7 (a)). Conversely, the commercial açaí sample (Powder UK)

had elemental levels that were typically 30% lower than that measured in other

commercial fruit samples. (Table 5.7 (a)). For example, the manganese levels of

the non-commercial pulp were 64.06 ± 0.93 mg/kg Mn (f.w.) for purple and 61.14

± 0.60 mg/kg Mn (f.w.) for white; whilst the commercial pulp (purple) was 26.76 ±

230

1.51 mg/kg Mn (f.w.). This study also investigated the elemental analysis of non-

processed açaí samples (fruit and seeds) from the Amazonian region of Brazil.

Overall, the data for most of the elements had higher elemental levels for the

non-processed fruits, when compared to the corresponding seeds (Table 5.10).

The data showed that processed pulp, with less water added (higher solids

content), also had higher elemental levels (based on a fresh weight) (Table 5.11).

The manganese content of the açaí (pulp) samples ranged from 5.26 to 126.26

mg/kg (f.w.) depending on the origin and water content of the samples. The total

polyphenol content of non-processed açaí samples showed clearly that during

the processing of the açaí berries, the levels decrease, which suggests that they

are ‘lost’ (Table 5.9). Furthermore, the açaí samples collected from the islands

close to Belém (Ilhas) in northern Brazil, had higher total polyphenol levels when

compared to samples from Genipauba (which is inland within the State of Pará).

Also, the total polyphenol content increased in processed açaí pulp that also had

a higher solids content (Table 5.9).

This study was very important as açaí plays a major nutritional role in

terms of the diet of the local population of the Amazonian region, and also now

as the comercialised products are becoming popular throughout Brazil (section

5.2). An evaluation was made using the data produced in this study to determine

the contribution that a regular consumption of açaí (500g) would make to the

dietary intake of polyphenols and minerals by Brazilians. This study showed that

the total polyphenol intake of purple açaí would be similar to that obtained from

the consumption of other well-known ‘super-fruits’, such as, blueberries and

acerola (section 5.7). Moreover, the consumption of processed açaí pulp would

contribute, in a single serving, more than 100% of the recommended daily

allowance (RDA) intake of total polyphenols (Figure 5.26). In relation to the

elemental data, the consumption of açaí represents a good source of manganese

(with an average of 1500% of the RDA), copper (average of 90%), calcium

(20%), magnesium (30%) and zinc (15%) (Tables 5.9 and 5.10). As was

mentioned in chapter 3, Brazilian yerba mate infusions were also found to be

good sources of manganese, with the high % intake levels based on the daily

231

consumption of this beverage (section 3.6.6.1). Clearly, once again there may be

a concern about the high levels of Mn being consumed from these products,

which will be discussed in the next section in terms of the bioavailability of Mn in

foodstuffs and beverages.

6.5. Limitations

The main aim and objectives of this research have been achieved,

although, it is important to highlight the limitations of the presented study. During

the field-trips to the different plantations in Brazil, there were time contraints

imposed by the commercial owners and the costs involved in the collection and

transport of samples. Moreover, trips to Brazil had to be undertaken at specific

university (non-teaching times) and therefore the availability of samples was

strongly influenced by seasonality and the access to specific types and varieties

of yerba mate, coffee and acai berries. Also, in the processing plants, different

materials were being processed at the same time, therefore only a limited

number of samples were taken at each phase of the process for chemical

investigation. The polyphenol studies were carried-out in commercial laboratories

in London (UK) and North Carolina (USA) which enhanced the quality of the

research but became a limiting factor due to the time made available to use

these instruments. In summary, only a limited number of samples were obtained

and analysed in this research. As a result, the power of the statistical analysis

undertaken has focused mainly on parametric methods. If more samples could

have been collected and transported to the UK it would have been possible to do

more advanced analysis. Moreover, another limitation associated with the

sampling strategy is that the analyses were performed on usually a single sample

of plant or berry material. This results in a challenge to determine the

representativity of the reported levels of the compounds in the actual plant

species. There is also a variability in the different chemical levels due to seasonal

differences that should be further investigated (Timmers et al., 2017). Finally, the

232

limited availability of matrix-matched certified reference materials and standards

may have an impact on the method validation of the polyphenol analysis

6.6. Future Work

It is proposed that future research should evaluate the possible chemical

interactions between elements and organic compounds in the natural and related

processed products. These interactions may influence the bioavailability of not

only the elemental species, but also the polyphenols. One study has already

been reported in terms of an interaction between iron and polyphenols in tea

(Perron and Brumaghim, 2009). This study reported that polyphenols in tea

infusions decreased the levels of bioavailable iron due to complexation of the

chemicals. Moreover, the bioavailability of iron (and in this study for manganese

and other trace elements) would impact on the human intake of these chemicals

through the consumption of such products (yerba mate, roasted coffee and açaí

pulp), thereby increasing or decreasing the potential bioinorganic effect on the

human body (Fairweather-Tait and Hurrell, 1996). To this end, a continuation of

this research would be to study the complexation of metal(s)-polyphenol(s) in

these natural and commercial food products and to evaluate the bioavailabity of

these elements (minerals) using in vivo analysis.

Furthermore, an extensive evaluation of the potential human and mouse

biological effects of the chemicals present in the investigated products (yerba

mate, roasted coffee and açaí) could be undertaken (extending the pilot studies

carried-out in the USA on açaí reported in section 5.5). A series of in vitro

analyses on a range of different human or animal cell lines could be used in

order to assess the claimed health benefits of consuming these Brazilian

products.

233

Bibliography

ABIC. 2018. Consumo de café no Brasil cresceu até 3,5% em 2018 [Online]. Available: http://abic.com.br/consumo-de-cafe-no-brasil-cresceu-ate-35-em-2018-diz-euromonitor/ [Accessed 27/07/2019 2019].

ABRAHÃO, S. A., PEREIRA, R. G. F. A., DUARTE, S. M. D. S., LIMA, A. R., ALVARENGA, D. J. & FERREIRA, E. B. 2010. Compostos bioativos e atividade antioxidante do café (Coffea arabica L.). Ciência e Agrotecnologia, 34, 414-420.

ADEKIYA, A. O. & AGBEDE, T. M. 2009. Growth and yield of tomato (Lycopersicon esculentum Mill) as influenced by poultry manure and NPK fertilizer. Emirates Journal of Food and Agriculture, 10-20.

AGILENT 2017. Agilent 7800 series ICP-MS. Hardware Maintenance Manual. In: AGILENT TECHNOLOGIES, I. (ed.). Japan.

AGUINAGA, H. J. 2017. Trayectoria histórica y contexto actual de la yerba mate en Paraguay: destino de la producción campesina em el distrito de Dr. Juan Manuel Frutos, Caaguazú. BS thesis, UNILA, Foz do Iguazú, 62.

AINSWORTH, E. A. & GILLESPIE, K. M. 2007. Estimation of total phenolic content and other oxidation substrates in plant tissues using Folin–Ciocalteu reagent. Nature Protocols, 2, 875.

ALBARICI, T. R., PESSOA, J. D. C. & FORIM, M. R. 2006. Efeito das variações de pH e temperatura sobre as antocianinas na polpa de açaí: estudos espectrofotométricos e cromatográficos. Embrapa Instrumentação-Comunicado Técnico (INFOTECA-E), 78, 1-5.

ALEXANDRE MARCELO, M. C., MARTINS, C. A., POZEBON, D. & FERRÃO, M. F. 2014. Methods of multivariate analysis of NIR reflectance spectra for classification of yerba mate. Analytical Methods, 6, 7621-7627.

ALI, I., AL‐OTHMAN, Z. A., NAGAE, N., GAITONDE, V. D. & DUTTA, K. K. 2012. Recent trends in ultra‐fast HPLC: New generation superficially porous silica columns. Journal of Separation Science, 35, 3235-3249.

AMORIM FILHO, V. R., POLITO, W. L. & GOMES NETO, J. A. 2007. Comparative studies of the sample decomposition of green and roasted coffee for determination of nutrients and data exploratory analysis. Journal of the Brazilian Chemical Society, 18, 47-53.

ANDERSEN, T. & FOGH, J. 2001. Weight loss and delayed gastric emptying following a South American herbal preparation in overweight patients. Journal of Human Nutrition and Dietetics, 14, 243-250.

ANDERSON, K. A. & SMITH, B. W. 2002. Chemical profiling to differentiate geographic growing origins of coffee. Journal of Agricultural and Food Chemistry, 50, 2068-2075.

ANDRADE, S. S. D. 2014. Põe tapioca, põe farinha d’água?: gastronomia do açaí e identidade socioespacial e cultural na Grande Belém. PhD thesis, UFPA, Belém, 52.

234

ANTHEMIDIS, A. N. & PLIATSIKA, V. G. 2005. On-line slurry formation and nebulization for inductively coupled plasma atomic emission spectrometry. Multi-element analysis of cocoa and coffee powder samples. Journal of Analytical Atomic Spectrometry, 20, 1280-1286.

ARNOLD, M. & BARBUL, A. 2006. Nutrition and wound healing. Plastic and Reconstructive Surgery, 117, 42S-58S.

ARTS, M. J. T. J., HAENEN, G. R. M. M., VOSS, H.-P. & BAST, A. 2004. Antioxidant capacity of reaction products limits the applicability of the Trolox Equivalent Antioxidant Capacity (TEAC) assay. Food and Chemical Toxicology, 42, 45-49.

ASHU, R. & CHANDRAVANSHI, B. S. 2011. Concentration levels of metals in commercially available Ethiopian roasted coffee powders and their infusions. Bulletin of the Chemical Society of Ethiopia, 25 (1), 11-24.

ATSDR. 2000. Toxicological Profile for Manganese [Online]. Available: www.atsdr.cdc.gov/toxprofiles/tp151/pdf [Accessed 08/07/2019 2019].

AUGUSTI, P., TORMA, P., CARVALHO, A., FLÔRES, S. & RIOS, A. D. O. 2016. Compostos bioativos e atividade antioxidante de genótipos de açaí (Euterpe oleracea). Embrapa Amazônia Oriental-Artigo em anais de congresso (ALICE), 2016. In: Congresso Brasileiro de Ciência e Tecnologia de Alimentos, Gramado, 25.

AVEGLIANO, R. P. 2009. Total Diet Study in Sao Paulo State: estimation of dietary intakes of toxic (arsenic and cadmium) and essential elements (calcium, chromium, iron, selenium, sodium, potassium and zinc). PhD thesis, IPEN, São Paulo, 87.

AVEGLIANO, R. P., MAIHARA, V. A. & DA SILVA, F. F. 2011. A Brazilian Total Diet Study: Evaluation of essential elements. Journal of Food Composition and Analysis, 24, 1009-1016.

BAILEY, S. W., HORSLEY, S. B., LONG, R. P. & HALLETT, R. A. 2004. Influence of edaphic factors on sugar maple nutrition and health on the Allegheny Plateau. Soil Science Society of America Journal, 68, 243-252.

BAKER, A. J. M. & BROOKS, R. R. 1989. Terrestrial higher plants which hyperaccumulate metallic elements. A review of their distribution, ecology and phytochemistry. Biorecovery, 1, 81-126.

BALCAEN, L., BOLEA-FERNANDEZ, E., RESANO, M. & VANHAECKE, F. 2015. Inductively coupled plasma–Tandem mass spectrometry (ICP-MS/MS): A powerful and universal tool for the interference-free determination of (ultra) trace elements–A tutorial review. Analytica chimica acta, 894, 7-19.

BALZON, D. R., DA SILVA, J. C. G. L. & DOS SANTOS, A. J. 2004. Aspectos mercadológicos de produtos florestais não madeireiros análise retrospectiva. Floresta, 34.

BANNERMAN, D. D. & GOLDBLUM, S. E. 2003. Mechanisms of bacterial lipopolysaccharide-induced endothelial apoptosis. American Journal of Physiology-Lung Cellular and Molecular Physiology, 284, L899-L914.

BARBOSA, J. Z., MOTTA, A. C. V., CONSALTER, R., POGGERE, G. C., SANTIN, D. & WENDLING, I. 2018. Plant growth, nutrients and potentially toxic elements in

235

leaves of yerba mate clones in response to phosphorus in acid soils. Anais da Academia Brasileira de Ciências, 90, 557-571.

BARBOSA, J. Z., ZAMBON, L. M., MOTTA, A. C. V. & WENDLING, I. 2015. composition, hot-water solubility of elements and nutritional value of fruits and. Ciência e Agrotecnologia, 39, 593-603.

BARONI, M. V., PODIO, N. S., BADINI, R. L. G., INGA, M., OSTERA, H. C. A., CAGNONI, M., GAUTIER, E. A., GARCÍA, P. P., HOOGEWERFF, J. & WUNDERLIN, D. A. 2015. Linking soil, water, and honey composition to assess the geographical origin of argentinean honey by multielemental and isotopic analyses. Journal of Agricultural and Food Chemistry, 63, 4638-4645.

BASTOS, D. H. M., DE OLIVEIRA, D. M. & MATSUMOTO, R. L. T. 2007. Yerba mate: pharmacological properties, research and biotechnology. Medicinal and Aromatic Plant Science and Biotechnology, 1, 37.

BASTOS, D. H. M., ISHIMOTO, E. Y., MARQUES, M. O. M., FERRI, A. F. & TORRES, E. A. F. S. 2006. Essential oil and antioxidant activity of green mate and mate tea (Ilex paraguariensis) infusions. Journal of Food Composition and Analysis, 19, 538-543.

BASTOS, M. C., REISSMANN, C. B., KESEKER, J. F., PAULETTI, V., GAIAD, S. & STURION, J. A. 2014. Mineral content of young leaves of yerba mate. Embrapa Florestas-Artigo Em Periódico Indexado (ALICE), 34 (77), 63-71.

BEAUCHEMIN, D. 2010. Environmental analysis by inductively coupled plasma mass spectrometry. Mass Spectrometry Reviews, 29, 560-592.

BECKER, J. S. 2007. Inorganic Mass Spectrometry: Principles And Applications, Wiley Online Library. Chichester, UK, John Wiley & Sons, 314.

BENNAT, C., ENGELHARDT, U. H., KIEHNE, A., WIRRIES, F.-M. & MAIER, H. G. 1994. HPLC analysis of chlorogenic acid lactones in roasted coffee. Zeitschrift für Lebensmittel-Untersuchung und Forschung, 199, 17-21.

BHATTACHARYA, P. T., MISRA, S. R. & HUSSAIN, M. 2016. Nutritional aspects of essential trace elements in oral health and disease: an extensive review. Scientifica, 2016, 12.

BOBBIO, F., DRUZIAN, J. & ABRAO, P. 2000. Identificação e quantificação das antocianinas do fruto do açaizeiro. Euterpe oleracea Mart. Ciênc. Tecnol. Aliment. vol.20, n.3, 388-390.

BOUCHARD, M. F., SAUVÉ, S., BARBEAU, B., LEGRAND, M., BRODEUR, M.-È., BOUFFARD, T., LIMOGES, E., BELLINGER, D. C. & MERGLER, D. 2010. Intellectual impairment in school-age children exposed to manganese from drinking water. Environmental Health Perspectives, 119, 138-143.

BRACESCO, N., SANCHEZ, A. G., CONTRERAS, V., MENINI, T. & GUGLIUCCI, A. 2011. Recent advances on Ilex paraguariensis research: Minireview. Journal of Ethnopharmacology, 136, 378-384.

BRAGANÇA, V. L. C., MELNIKOV, P. & ZANONI, L. Z. 2011. Trace elements in different brands of yerba mate tea. Biological Trace Element Research, 144, 1197-1204.

BRASIL 2006. Guia alimentar para a população brasileira: promovendo a alimentação saudável. Ministério da Saúde, Brasília, 5-15.

236

BRAVO, L., GOYA, L. & LECUMBERRI, E. 2007. LC/MS characterization of phenolic constituents of mate (Ilex paraguariensis, St. Hil.) and its antioxidant activity compared to commonly consumed beverages. Food Research International, 40, 393-405.

BU-OLAYAN, A. H. & THOMAS, B. V. 2009. Translocation and bioaccumulation of trace metals in desert plants of Kuwait Governorates. Research Journal of Environmental Sciences, 3, 581-587.

BUFFO, R. A. & CARDELLI‐FREIRE, C. 2004. Coffee flavour: an overview. Flavour and Fragrance Journal, 19, 99-104.

BUTT, M. S. & SULTAN, M. T. 2011. Coffee and its Consumption: Benefits and Risks. Critical Reviews in Food Science and Nutrition, 51, 363-373.

CABOOTER, D. & DESMET, G. 2012. General overview of fast and high-resolution approaches in liquid chromatography. UHPLC in Life Sciences. Royal Society of Chemistry, London, 1-28.

CALDARELLI, C. E., GILIO, L. & ZILBERMAN, D. 2019. The Coffee Market in Brazil: challenges and policy guidelines. Revista de Economia, 39, 1-18.

CAMPOS‐VEGA, R. & OOMAH, B. D. 2013. Chemistry and classification of phytochemicals. Handbook of Plant Food Phytochemicals: Sources, Stability and Extraction, 5-48.

CANUTO, G. A. B., XAVIER, A. A. O., NEVES, L. C. & BENASSI, M. D. T. 2010. Physical and chemical characterization of fruit pulps from Amazonia and their correlation to free radical scavenger activity. Revista Brasileira de Fruticultura, 32, 1196-1205.

CARDOZO JR, E. L., FERRARESE-FILHO, O., CARDOZO FILHO, L., FERRARESE, M. D. L. L., DONADUZZI, C. M. & STURION, J. A. 2007. Methylxanthines and phenolic compounds in mate (Ilex paraguariensis St. Hil.) progenies grown in Brazil. Journal of Food Composition and Analysis, 20, 553-558.

CARNEIRO, A. P. G., DA SILVA, L. M. R., DE FIGUEIREDO, R. W., DE SOUSA, P. H. M. & MAIA, G. A. 2015. Efeito da Temperatura no Comportamento Reológico de Pó de Açaí (Euterpe oleracea) Reconstituído. Journal of Health Sciences, 14, 1-4.

CASTANEDA-OVANDO, A., DE LOURDES PACHECO-HERNÁNDEZ, M., PÁEZ-HERNÁNDEZ, M. E., RODRÍGUEZ, J. A. & GALÁN-VIDAL, C. A. 2009. Chemical studies of anthocyanins: A review. Food Chemistry, 113, 859-871.

CHOI, S.-Y., HWANG, J.-H., KO, H.-C., PARK, J.-G. & KIM, S.-J. 2007a. Nobiletin from citrus fruit peel inhibits the DNA-binding activity of NF-κB and ROS production in LPS-activated RAW 264.7 cells. Journal of Ethnopharmacology, 113, 149-155.

CHOI, S. Y., HWANG, J. H., KO, H. C., PARK, J. G. & KIM, S. J. 2007b. Nobiletin from citrus fruit peel inhibits the DNA-binding activity of NF-κB and ROS production in LPS-activated RAW 264.7 cells. Journal of Ethnopharmacology, 113, 149-155.

CLARKE, R. & VITZTHUM, O. G. 2008. Coffee: Recent Developments, Blackwell Science, Oxford, 246.

CLIFFORD, M. N. 1985. Coffee: botany, biochemistry and production of beans and beverage, Springer Science & Business Media, London, 452.

237

CLIFFORD, M. N. & RAMIREZ-MARTINEZ, J. R. 1990. Chlorogenic acids and purine alkaloids contents of mate (Ilex paraguariensis) leaf and beverage. Food Chemistry, 35, 13-21.

COLLI, C. 2005. Biodisponibilidade de nutrientes. Revista Brasileira de Ciências Farmacêuticas, 41, 120-120.

COLPO, A. C., ROSA, H., LIMA, M. E., PAZZINI, C. E. F., DE CAMARGO, V. B., BASSANTE, F. E. M., PUNTEL, R., ÁVILA, D. S., MENDEZ, A. & FOLMER, V. 2016. Yerba mate (Ilex paraguariensis St. Hill.)-based beverages: How successive extraction influences the extract composition and its capacity to chelate iron and scavenge free radicals. Food Chemistry, 209, 185-195.

COMBS, G. F. 2013. Geological impacts on nutrition. Essentials of Medical Geology. Springer, 2013, 179-194.

CONN, S. & GILLIHAM, M. 2010. Comparative physiology of elemental distributions in plants. Annals of Botany, 105, 1081-1102.

CORDER, G. W. & FOREMAN, D. I. 2011. Nonparametric statistics for non‐statisticians. John Wiley & Sons, Inc, Hoboken, 274.

CÓRDOVA-FRAGA, T., DE ARAUJO, D. B., SANCHEZ, T. A., ELIAS, J., CARNEIRO, A. A. O., BRANDT-OLIVEIRA, R., SOSA, M. & BAFFA, O. 2004. Euterpe olerácea (açaı) as an alternative oral contrast agent in MRI of the gastrointestinal system: preliminary results. Magnetic Resonance Imaging, 22, 389-393.

CORREIA, R., GRACE, M. H., ESPOSITO, D. & LILA, M. A. 2017. Wild blueberry polyphenol-protein food ingredients produced by three drying methods: Comparative physico-chemical properties, phytochemical content, and stability during storage. Food Chemistry, 235, 76-85.

CORY, H., PASSARELLI, S., SZETO, J., TAMEZ, M. & MATTEI, J. 2018. The Role of Polyphenols in Human Health and Food Systems: A Mini-Review. Frontiers In Nutrition, 5, 87-87.

COSTA, A. G. V., GARCIA-DIAZ, D. F., JIMENEZ, P. & SILVA, P. I. 2013. Bioactive compounds and health benefits of exotic tropical red–black berries. Journal Of Functional Foods, 5, 539-549.

COSTA, S. M. F. D., CARMO, M. B. S. D. & BARJA, P. R. 2019. The urban hierarchy at the delta of the Amazon River and the importance of small cities. urbe. Revista Brasileira de Gestão Urbana, 11, 1-17.

CROZIER, A. 2003. Classification and biosynthesis of secondary plant products: an overview. Plants: Diet and Health: The Report of a British Nutrition Foundation Task Force, 27-48.

CROZIER, A., JAGANATH, I. B. & CLIFFORD, M. N. 2009. Dietary phenolics: chemistry, bioavailability and effects on health. Natural Product Reports, 26, 1001-1043.

CROZIER, S. J., PRESTON, A. G., HURST, J. W., PAYNE, M. J., MANN, J., HAINLY, L. & MILLER, D. L. 2011. Cacao seeds are a "Super Fruit": A comparative analysis of various fruit powders and products. Chemistry Central Journal, 5, 5-5.

CROZIER, T. W. M., STALMACH, A., LEAN, M. E. J. & CROZIER, A. 2012. Espresso coffees, caffeine and chlorogenic acid intake: potential health implications. Food & Function, 3, 30-33.

238

CUNHA FILHO, M. H. D. & CARVALHO, R. M. 2005. Exportações brasileiras de frutas: diversificação ou concentração de produtos e destinos, UERN, Fortaleza, 1-14.

DA SILVA, L. A. L., PEZZINI, B. R. & SOARES, L. 2015. Spectrophotometric determination of the total flavonoid content in Ocimum basilicum L. (Lamiaceae) leaves. Pharmacognosy Magazine, 11, 96-101.

DA SILVA SANTOS, V., DE ALMEIDA TEIXEIRA, G. H. & BARBOSA, F. 2014. Açaí (Euterpe oleracea Mart.): A Tropical Fruit with High Levels of Essential Minerals—Especially Manganese—and its Contribution as a Source of Natural Mineral Supplementation. Journal of Toxicology and Environmental Health, Part A, 77, 80-89.

DA SILVEIRA, T. F. F., DE SOUZA, T. C. L., CARVALHO, A. V., RIBEIRO, A. B., KUHNLE, G. G. C. & GODOY, H. T. 2017. White açaí juice (Euterpe oleracea): Phenolic composition by LC-ESI-MS/MS, antioxidant capacity and inhibition effect on the formation of colorectal cancer related compounds. Journal of Functional Foods, 36, 215-223.

DARNET, S., SERRA, J. L., DA CRUZ RODRIGUES, A. M. & MELLER DA SILVA, L. H. 2011. A high-performance liquid chromatography method to measure tocopherols in assai pulp (Euterpe oleracea). Food Research International, 44, 2107-2111.

DARTORA, N., DE SOUZA, L. M., SANTANA-FILHO, A. P., IACOMINI, M., VALDUGA, A. T., GORIN, P. A. J. & SASSAKI, G. L. 2011. UPLC-PDA–MS evaluation of bioactive compounds from leaves of Ilex paraguariensis with different growth conditions, treatments and ageing. Food Chemistry, 129, 1453-1461.

DE GODOY, R. C. B., DELIZA, R., GHENO, L. B., LICODIEDOFF, S., FRIZON, C. N. T., RIBANI, R. H. & DOS SANTOS, G. G. 2013. Consumer perceptions, attitudes and acceptance of new and traditional mate tea products. Food Research International, 53, 801-807.

DE LIMA, F. E. L., DE MENEZES, T. N., TAVARES, M. P., SZARFARC, S. C. & FISBERG, R. M. 2000. Ácidos Graxos e Doenças Cardiovarculares: uma revisão. Rev. nutr, 13, 73-80.

DE MEJÍA, E. G., SONG, Y. S., HECK, C. I. & RAMÍREZ-MARES, M. 2010. Yerba mate tea (Ilex paraguariensis): Phenolics, antioxidant capacity and in vitro inhibition of colon cancer cell proliferation. Journal of Functional Foods, 2, 23-34.

DE MORAIS, E. C., STEFANUTO, A., KLEIN, G. A., BOAVENTURA, B. C. B., DE ANDRADE, F., WAZLAWIK, E., DI PIETRO, P. F., MARASCHIN, M. & DA SILVA, E. L. 2009. Consumption of Yerba Mate (Ilex paraguariensis) Improves Serum Lipid Parameters in Healthy Dyslipidemic Subjects and Provides an Additional LDL-Cholesterol Reduction in Individuals on Statin Therapy. Journal of Agricultural and Food Chemistry, 57, 8316-8324.

DE SOUZA, M. O., SILVA, M., SILVA, M. E., DE PAULA OLIVEIRA, R. & PEDROSA, M. L. 2010. Diet supplementation with acai (Euterpe oleracea Mart.) pulp improves biomarkers of oxidative stress and the serum lipid profile in rats. Nutrition, 26, 804-810.

DE STEFANI, E., FIERRO, L., CORREA, P., FONTHAM, E., RONCO, A., LARRINAGA, M., BALBI, J. & MENDILAHARSU, M. 1996. Mate drinking and risk of lung

239

cancer in males: a case-control study from Uruguay. Cancer Epidemiology and Prevention Biomarkers, 5, 515-519.

DEAN, J. R. 2003. Methods for environmental trace analysis, John Wiley and Sons, Chicester, 257.

DEL POZO-INSFRAN, D., BRENES, C. H. & TALCOTT, S. T. 2004. Phytochemical Composition and Pigment Stability of Açai (Euterpe oleracea Mart.). Journal of Agricultural and Food Chemistry, 52, 1539-1545.

DIAS, A. L. S., ROZET, E., CHATAIGNÉ, G., OLIVEIRA, A. C., RABELO, C. A. S., HUBERT, P., ROGEZ, H. & QUETIN-LECLERCQ, J. 2012. A rapid validated UHPLC–PDA method for anthocyanins quantification from Euterpe oleracea fruits. Journal of Chromatography B, 907, 108-116.

DONNELLY, C. M. 2015. The Characterisation and Delivery of Flavonoids and Other Minor Components in Traditional Food Ingredients. PhD thesis, University of Surrey, Guildford, 551.

DORRONSORO, B., AGUILAR, J., DORRONSORO-DÍAZ, C., STOOPS, G., SIERRA, M., FERNÁNDEZ, J. & DORRONSORO-FDEZ, C. 2006. Hidromorfía en suelos. Curso de Edafología. Universidad de Granada. [Online]. Available: http://edafologia.ugr.es/hidro/concept.htm [Accessed 08/07/2019 2019].

DOU. 2013. Limites máximos de contaminantes [Online]. Available: http://portal.anvisa.gov.br/documents/33916/393845/RDC%2Bn%25C2%25BA%2B42_2013_final.pdf/eec629cf-8d17-422b-a362-366b275c1a00 [Accessed 23/06/2019 2019].

DUARTE, F. 2000. Seleção, treinamento de julgadores e metodologia para análise sensorial de extrato de erva-mate. PhD thesis, UFPA, Curitiba, 43.

DUGO, P., CACCIOLA, F., DONATO, P., JACQUES, R. A., CARAMÃO, E. B. & MONDELLO, L. 2009. High efficiency liquid chromatography techniques coupled to mass spectrometry for the characterization of mate extracts. Journal of Chromatography A, 1216, 7213-7221.

EFSA. 2006. Tolerable upper intake levels for vitamins and minerals [Online]. European Food Safety Authority. [Online]. Available: http://www.efsa.europa.eu/en/topics/topic/drv.htm [Accessed 08/07/2019 2019].

EMBRAPA. 2017. Exportation statistics [Online]. Available: https://www.embrapa.br [Accessed 08/07/2019 2019].

ESMELINDRO, M. C., TONIAZZO, G., WACZUK, A., DARIVA, C. & OLIVEIRA, D. D. 2002. Caracterização físico-química da erva-mate: influência das etapas do processamento industrial. Ciência e Tecnologia de Alimentos, 22, 193-204.

ESPOSITO, D., CHEN, A., GRACE, M. H., KOMARNYTSKY, S. & LILA, M. A. 2014. Inhibitory Effects of Wild Blueberry Anthocyanins and Other Flavonoids on Biomarkers of Acute and Chronic Inflammation in Vitro. Journal of Agricultural and Food Chemistry, 62, 7022-7028.

EVM 2003. Safe Upper Levels for Vitamins and Minerals. Report of the Expert Group on Vitamins and Minerals, Manganese, Food Standards Agency, UK, 213 - 218.

240

FAIRWEATHER-TAIT, S. & HURRELL, R. F. 1996. Bioavailability of Minerals and Trace Elements: Members of EC Flair Concerted Action No. 10: Measurements of micronutrient absorption and status*. Nutrition Research Reviews, 9, 295-324.

FALLER, A. L. K. & FIALHO, E. 2009. Disponibilidade de polifenóis em frutas e hortaliças consumidas no Brasil. Revista de Saúde Pública, 43, 211-218.

FANG, F. C. & VAZQUEZ-TORRES, A. 2002. Nitric oxide production by human macrophages: there's NO doubt about it. American Journal of Physiology-Lung Cellular and Molecular Physiology, 282, L941-L943.

FAO. 2018. FAO Country Profiles: Brazil [Online]. Available: http://www.fao.org/countryprofiles/index/en/?iso3=BRA [Accessed 08/07/2019 2019].

FAO, F. 2015. AquaSTAT website. Food and Agriculture Organization of the United Nations (FAO).

FARAH, A., DE PAULIS, T., TRUGO, L. C. & MARTIN, P. R. 2005. Effect of roasting on the formation of chlorogenic acid lactones in coffee. Journal of Agricultural and Food Chemistry, 53, 1505-1513.

FERRUZZI, M. G. 2010. The influence of beverage composition on delivery of phenolic compounds from coffee and tea. Physiology & behavior, 100, 33-41.

FILIP, R. & FERRARO, G. E. 2003. Researching on new species of “Mate”: Ilex brevicuspis. European Journal of Nutrition, 42, 50-54.

FILIP, R., LOTITO, S. B., FERRARO, G. & FRAGA, C. G. 2000. Antioxidant activity of Ilex paraguariensis and related species. Nutrition Research, 20, 1437-1446.

FINLEY, J. W., KONG, A.-N., HINTZE, K. J., JEFFERY, E. H., JI, L. L. & LEI, X. G. 2011. Antioxidants in foods: state of the science important to the food industry. Journal of Agricultural and Food chemistry, 59, 6837-6846.

FORMA, B. 2016. Chimarrão: Benefícios à saúde [Online]. Available: http://boaformaesaude.com.br/chimarrao-veja-os-beneficios-dessa-bebida [Accessed 23/06/2019 2019].

FOUNTAIN, K. J. & IRANETA, P. C. 2012. Instrumentation and columns for UHPLC separations. UHPLC in Life Sciences, 29-66.

FRAGA, C. G. 2005. Relevance, essentiality and toxicity of trace elements in human health. Molecular Aspects of Medicine, 26, 235-244.

FRAGA, C. G., GALLEANO, M., VERSTRAETEN, S. V. & OTEIZA, P. I. 2010. Basic biochemical mechanisms behind the health benefits of polyphenols. Molecular Aspects of Medicine, 31, 435-445.

FRAGOSO, M. F., PRADO, M. G., BARBOSA, L., ROCHA, N. S. & BARBISAN, L. F. 2012. Inhibition of mouse urinary bladder carcinogenesis by açai fruit (Euterpe oleraceae Martius) intake. Plant Foods for Human Nutrition, 67, 235-241.

FRANCA, A. S., OLIVEIRA, L. S., OLIVEIRA, R. C. S., AGRESTI, P. C. M. & AUGUSTI, R. 2009. A preliminary evaluation of the effect of processing temperature on coffee roasting degree assessment. Journal of Food Engineering, 92, 345-352.

241

FREDERICE, R., FERREIRA, A. P. G. & GEHLEN, M. H. 2010. Molecular fluorescence in silica particles doped with quercetin-Al3+ complexes. Journal of the Brazilian Chemical Society, 21, 1213-1217.

FRIEDEN, E. 1985. New perspectives on the essential trace elements. ACS Publications , Chemical Education, 62, 11, 917.

FUJIOKA 2006. Quantitation of Volatiles and Nonvolatile Acids in an Extract from Coffee Beverages: Correlation with Antioxidant Activity. Journal of Agricultural And Food Chemistry., 54, 6054-6058.

FUKUSHIMA, Y., OHIE, T., YONEKAWA, Y., YONEMOTO, K., AIZAWA, H., MORI, Y., WATANABE, M., TAKEUCHI, M., HASEGAWA, M. & TAGUCHI, C. 2009. Coffee and green tea as a large source of antioxidant polyphenols in the Japanese population. Journal of Agricultural and Food Chemistry, 57, 1253-1259.

GALLORI, S., BILIA, A. R., BERGONZI, M. C., BARBOSA, W. L. R. & VINCIERI, F. F. 2004. Polyphenolic constituents of fruit pulp of Euterpe oleracea Mart.(açai palm). Chromatographia, 59, 739-743.

GALOTTA, A. L. Q. D. A. & BOAVENTURA, M. A. D. 2005. Chemical constituents from roots and leaf stalks of açai (Euterpe precatoria Mart., Arecaceae). Química Nova, 28, 610-613.

GARCÍA-ÁLVAREZ-COQUE, M. C., TORRES-LAPASIÓ, J. R. & NAVARRO-HUERTA, J. A. 2017. Secondary chemical equilibria in reversed-phase liquid chromatography. Liquid Chromatography in Elsevier, 2017, 125-146.

GARRETT, R. G. 2000. Natural sources of metals to the environment. Human and Ecological Risk Assessment, 6, 945-963.

GARZÓN, G. A., NARVÁEZ-CUENCA, C.-E., VINCKEN, J.-P. & GRUPPEN, H. 2017. Polyphenolic composition and antioxidant activity of açai (Euterpe oleracea Mart.) from Colombia. Food Chemistry, 217, 364-372.

GEIGER, R., PERREN, R., KUENZLI, R. & ESCHER, F. 2005. Carbon dioxide evolution and moisture evaporation during roasting of coffee beans. Journal of Food Science, 70, E124-E130.

GIULIAN, R., IOCHIMS DOS SANTOS, C. E., DE MORAES SHUBEITA, S., DA SILVA, L. M., DIAS, J. F. & YONEAMA, M. L. 2007. Elemental Characterization of Commercial Mate Tea Leaves (Ilex paraguariensis A. St.-Hil.) before and after Hot Water Infusion Using Ion Beam Techniques. Journal of Agricultural and Food Chemistry, 55, 741-746.

GIULIAN, R., IOCHIMS DOS SANTOS, C. E., DE MORAES SHUBEITA, S., MANFREDI DA SILVA, L., YONEAMA, M. L. & DIAS, J. F. 2009. The study of the influence of industrial processing on the elemental composition of mate tealeaves (Ilex paraguariensis) using the PIXE technique. LWT - Food Science and Technology, 42, 74-80.

GOLDHABER, S. B. 2003. Trace element risk assessment: essentiality vs. toxicity. Regulatory Toxicology and Pharmacology, 38, 232-242.

GONZÁLVEZ, A., LLORENS, A., CERVERA, M. L., ARMENTA, S. & DE LA GUARDIA, M. 2009. Elemental fingerprint of wines from the protected designation of origin Valencia. Food Chemistry, 112, 26-34.

242

GORDON, A., CRUZ, A. P. G., CABRAL, L. M. C., DE FREITAS, S. C., TAXI, C. M. A. D., DONANGELO, C. M., DE ANDRADE MATTIETTO, R., FRIEDRICH, M., DA MATTA, V. M. & MARX, F. 2012. Chemical characterization and evaluation of antioxidant properties of Açaí fruits (Euterpe oleraceae Mart.) during ripening. Food Chemistry, 133, 256-263.

GORJANOVIC, S., KOMES, D. E., PASTOR, F. T., BELSCAK-CVITANOVIC, A., PEZO, L., HECIMOVIC, I. & SUZNJEVIC, D. 2012. Antioxidant capacity of teas and herbal infusions: polarographic assessment. Journal of Agricultural and Food Chemistry, 60, 9573-9580.

GOULD, K. S. & LISTER, C. 2005. Flavonoid functions in plants. Flavonoids. CRC Press, 2005. 408-452.

GRACE, M. H., ESPOSITO, D., DUNLAP, K. L. & LILA, M. A. 2014. Comparative Analysis of Phenolic Content and Profile, Antioxidant Capacity and Anti-inflammatory Bioactivity in Wild Alaskan and Commercial Vaccinium Berries. Journal of Agricultural and Food Chemistry, 62, 4007-4017.

GRACE, M. H., GUZMAN, I., ROOPCHAND, D. E., MOSKAL, K., CHENG, D. M., POGREBNYAK, N., RASKIN, I., HOWELL, A. & LILA, M. A. 2013. Stable Binding of Alternative Protein-enriched Food Matrices with Concentrated Cranberry Bioflavonoids for Functional Food Applications. Journal of Agricultural and Food Chemistry, 61, 6856-6864.

GREMBECKA, M., MALINOWSKA, E. & SZEFER, P. 2007. Differentiation of market coffee and its infusions in view of their mineral composition. Science of the Total Environment, 383, 59-69.

GRYGO-SZYMANKO, E., TOBIASZ, A. & WALAS, S. 2016. Speciation analysis and fractionation of manganese: A review. TrAC Trends in Analytical Chemistry, 80, 112-124.

GUGLIUCCI, A. 1996. Antioxidant Effects of Ilex Paraguariensis: Induction of Decreased Oxidability of Human LDLin Vivo. Biochemical and Biophysical Research Communications, 224, 338-344.

HAMILTON, P. B., TEIXEIRA, M. M. G. & STEVENS, J. R. 2012. The evolution of Trypanosoma cruzi: the ‘bat seeding’hypothesis. Trends In Parasitology, 28, 136-141.

HAN, S. H., AN, J. Y., HWANG, J., KIM, S. B. & PARK, B. B. 2016. The effects of organic manure and chemical fertilizer on the growth and nutrient concentrations of yellow poplar (Liriodendron tulipifera Lin.) in a nursery system. Forest Science and Technology, 12, 137-143.

HARRIS, D. C. 2010. Quantitative Chemical Analysis. Macmillian, New York, 792.

HARVEY, D. T. 2006. Analytical Chemistry: A Modern Approach to Analytical Science, (R. Kellner, J.—M. Mermet, M. Otto, M. Varcárcel, and HM Widmer, eds.). ACS Publications, Journal Chemical Education, 83, 3, 385.

HASSIMOTTO, N. M. A., GENOVESE, M. I. & LAJOLO, F. M. 2005. Antioxidant activity of dietary fruits, vegetables, and commercial frozen fruit pulps. Journal of Agricultural and Food Chemistry, 53, 2928-2935.

243

HE, F., PAN, Q.-H., SHI, Y. & DUAN, C.-Q. 2008. Chemical Synthesis of Proanthocyanidins in Vitro and Their Reactions in Aging Wines. Molecules, 13 (12), 307-332.

HECIMOVIC, I., BELSCAK-CVITANOVIC, A., HORZIC, D. & KOMES, D. 2011. Comparative study of polyphenols and caffeine in different coffee varieties affected by the degree of roasting. Food Chemistry, 129, 991-1000.

HECK, C. I. & DE MEJIA, E. G. 2007. Yerba Mate Tea (Ilex paraguariensis): A Comprehensive Review on Chemistry, Health Implications, and Technological Considerations. Journal of Food Science, 72, R138-R151.

HECK, C. I., SCHMALKO, M. & GONZALEZ DE MEJIA, E. 2008. Effect of Growing and Drying Conditions on the Phenolic Composition of Mate Teas (Ilex paraguariensis). Journal of Agricultural and Food Chemistry, 56, 8394-8403.

HEINRICH, M., DHANJI, T. & CASSELMAN, I. 2011. Açai (Euterpe oleracea Mart.)—A phytochemical and pharmacological assessment of the species’ health claims. Phytochemistry Letters, 4, 10-21.

HEINRICHS, R. & MALAVOLTA, E. 2001. Mineral composition of a commercial product from mate-herb (Ilex paraguariensis St. Hil.). Ciência Rural, 31, 781-785.

HERCULANO, F. E. B. 2013. Produção industrial de cosméticos: o protagonismo da biodiversidade vegetal da Amazônia. PhD thesis, UFAM, Manaus, 58.

HOGAN, S., CHUNG, H., ZHANG, L., LI, J., LEE, Y., DAI, Y. & ZHOU, K. 2010. Antiproliferative and antioxidant properties of anthocyanin-rich extract from açai. Food Chemistry, 118, 208-214.

HOMMA, A. K. O., NOGUEIRA, O. L., DE MENEZES, A. J. E. A., DE CARVALHO, J. E. U., NICOLI, C. M. L. & DE MATOS, G. B. 2006. Açaí: novos desafios e tendências. Embrapa Amazônia Oriental-Artigo em periódico indexado (ALICE), 1 (2), 7-23.

HORSZWALD, A. & ANDLAUER, W. 2011. Characterisation of bioactive compounds in berry juices by traditional photometric and modern microplate methods. Journal of Berry Research, 1, 189-199.

HUANG, D., OU, B. & PRIOR, R. L. 2005. The chemistry behind antioxidant capacity assays. Journal of Agricultural and Food Chemistry, 53, 1841-1856.

IBGE. 2017. Produção Agrícola Municipal [Online]. Available: https://sidra.ibge.gov.br/pesquisa/pam/tabelas [Accessed 30/10/2018].

IBGE. 2019. Instituto Brasileiro de Geografia e Estatística [Online]. Available: https://ibge.gov.br [Accessed 08/07/2019 2019].

ILANY, T., ASHTON, M. S., MONTAGNINI, F. & MARTINEZ, C. 2010. Using agroforestry to improve soil fertility: effects of intercropping on Ilex paraguariensis (yerba mate) plantations with Araucaria angustifolia. Agroforestry systems, 80, 399-409.

ILLY, A. & VIANI, R. 1995. Espresso coffee: the chemistry of quality. Academic Press, London, 68.

INYM. 2017. Informe del sector yerbateiro [Online]. Available: https://www.inym.org.ar/wp-content/uploads/2018/02/2017-12-Estadisticas.pdf [Accessed 23/06/2019 2019].

244

IOM 2002. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium and zinc. Institute of Medicine, Food and Nutrition Board. Washington, USA, 10-22.

ISO (2005) ISO 14502-1: 2005 Determination of Substances Characteristic of Green and Black Tea - Part 1: Content of Total Polyphenols in Tea - Colorimetric Method Using Folin- Ciocalteu Reagent, Geneva: International Standards Organisation.

ISOLABELLA, S., COGOI, L., LÓPEZ, P., ANESINI, C., FERRARO, G. & FILIP, R. 2010. Study of the bioactive compounds variation during yerba mate (Ilex paraguariensis) processing. Food Chemistry, 122, 695-699.

JACQUES, R. A., ARRUDA, E. J., DE OLIVEIRA, L. C. S., DE OLIVEIRA, A. P., DARIVA, C., DE OLIVEIRA, J. V. & CARAMÃO, E. B. 2007. Influence of Agronomic Variables on the Macronutrient and Micronutrient Contents and Thermal Behavior of Mate Tea Leaves (Ilex paraguariensis). Journal of Agricultural and Food Chemistry, 55, 7510-7516.

JAISWAL, R., SOVDAT, T., VIVAN, F. & KUHNERT, N. 2010. Profiling and characterization by LC-MS n of the chlorogenic acids and hydroxycinnamoylshikimate esters in mate (Ilex paraguariensis). Journal of Agricultural and Food Chemistry, 58, 5471-5484.

JARDIM, M. A. G. 2002. A cadeia produtiva do açaizeiro para frutos e palmito: implicações ecológicas e socioeconômicas no estado do Pará. Museu Paraense Emilio Goeldi, 18 (2), 287-305.

JENSEN, G. S., WU, X., PATTERSON, K. M., BARNES, J., CARTER, S. G., SCHERWITZ, L., BEAMAN, R., ENDRES, J. R. & SCHAUSS, A. G. 2008. In Vitro and in Vivo Antioxidant and Anti-inflammatory Capacities of an Antioxidant-Rich Fruit and Berry Juice Blend. Results of a Pilot and Randomized, Double-Blinded, Placebo-Controlled, Crossover Study. Journal of Agricultural and Food Chemistry, 56, 8326-8333.

JOKANOVIC, M. R., DZINIC, N. R., CVETKOVIC, B. R., GRUJIC, S. & ODZAKOVIC, B. 2012. Changes of physical properties of coffee beans during roasting. Acta Periodica Technologica, 43, 21-31.

KABATA-PENDIAS, A. 2010. Trace elements in soils and plants, CRC press, Boca Raton, 331.

KANG, J., THAKALI, K. M., XIE, C., KONDO, M., TONG, Y., OU, B., JENSEN, G., MEDINA, M. B., SCHAUSS, A. G. & WU, X. 2012. Bioactivities of açaí (Euterpe precatoria Mart.) fruit pulp, superior antioxidant and anti-inflammatory properties to Euterpe oleracea Mart. Food Chemistry, 133, 671-677.

KAPASAKALIDIS, P. G., RASTALL, R. A. & GORDON, M. H. 2006. Extraction of polyphenols from processed black currant (Ribes nigrum L.) residues. Journal Of Agricultural and Food Chemistry, 54, 4016-4021.

KHOKHAR, S. & OWUSU APENTEN, R. K. 2003. Iron binding characteristics of phenolic compounds: some tentative structure–activity relations. Food Chemistry, 81, 133-140.

KLIMCZAK, I., MAŁECKA, M., SZLACHTA, M. & GLISZCZYŃSKA-ŚWIGŁO, A. 2007. Effect of storage on the content of polyphenols, vitamin C and the antioxidant activity of orange juices. Journal of Food Composition and Analysis, 20, 313-322.

245

KÜHNAU, J. 1976. The flavonoids. A class of semi-essential food components: their role in human nutrition. World Review Of Nutrition And Dietetics, 24, 117-191.

KUSKOSKI, E. M., GARCÍA ASUERO, A., MORALES MILLÁN, M. T. & FETT, R. 2006. Frutos tropicais silvestres e polpas de frutas congeladas: atividade antioxidante, polifenóis e antocianinas. Ciência Rural, 36, 1283-1287.

KYLE, J. A. M. & DUTHIE, G. G. 2005. Flavonoids in foods. Flavonoids. CRC Press, 230-273.

LAKENBRINK, C., LAPCZYNSKI, S., MAIWALD, B. & ENGELHARDT, U. H. 2000. Flavonoids and other polyphenols in consumer brews of tea and other caffeinated beverages. Journal of Agricultural and Food Chemistry, 48, 2848-2852.

LANDETE, J. M. 2013. Dietary Intake of Natural Antioxidants: Vitamins and Polyphenols. Critical Reviews in Food Science and Nutrition, 53, 706-721.

LEBORGNE, L., MAZIERE, J. C., MAZIERE, C. & ANDREJAK, M. 2002. Oxidative stress, atherogenesis and cardiovascular risk factors. Archives des Maladies du Coeur et des Vaisseaux, 95, 805-814.

LEE, R. & BALICK, M. J. 2008. Palms, People, and Health. Explore: The Journal of Science and Healing, 4, 59-62.

LEE, S.-E., LEE, H.-S., LEE, Y.-H., PARK, B.-S. & CHOI, W.-S. 1998. Antioxidative activities of methanol extracts of tropical and oriental medicinal plants. Applied Biological Chemistry, 41, 556-559.

LEIGH, R. A. 1997. Solute composition of vacuoles. Advances in Botanical Research, 25, 171-194.

LEÓN, K., MERY, D., PEDRESCHI, F. & LEÓN, J. 2006. Color measurement in L∗a∗b∗ units from RGB digital images. Food Research International, 39, 1084-1091.

LICHTENTHÄLER, R., RODRIGUES, R. B., MAIA, J. G. S., PAPAGIANNOPOULOS, M., FABRICIUS, H. & MARX, F. 2005a. Total oxidant scavenging capacities of Euterpe oleracea Mart. (Açaí) fruits. International Journal of Food Sciences and Nutrition, 56, 53-64.

LICHTENTHÄLER, R., RODRIGUES, R. B., MAIA, J. G. S., PAPAGIANNOPOULOS, M., FABRICIUS, H. & MARX, F. 2005b. Total oxidant scavenging capacities of Euterpe oleracea Mart.(Acai) fruits. International Journal of Food Sciences and Nutrition, 56, 53-64.

LINGE, K. L. & JARVIS, K. E. 2009. Quadrupole ICP‐MS: Introduction to instrumentation, measurement techniques and analytical capabilities. Geostandards and Geoanalytical Research, 33, 445-467.

LIU, E., YAN, C., MEI, X., HE, W., BING, S. H., DING, L., LIU, Q., LIU, S. & FAN, T. 2010. Long-term effect of chemical fertilizer, straw, and manure on soil chemical and biological properties in northwest China. Geoderma, 158, 173-180.

LLORENT-MARTÍNEZ, E. J., FERNÁNDEZ-DE CÓRDOVA, M. L., ORTEGA-BARRALES, P. & RUIZ-MEDINA, A. 2013. Characterization and comparison of the chemical composition of exotic superfoods. Microchemical Journal, 110, 444-451.

246

LONERAGAN, J. F. 1981. Distribution and movement of copper in plants. Copper in soils and plants, 1981, 165-188.

LORD, G. M. 2014. Arsenic and Other Trace Elements in Natural Waters from Argentina and New Zealand : Relationship between Environmental Exposure and Human Health., PhD thesis, University of Surrey, Guildford, 417.

LUDWIG, I. A., CLIFFORD, M. N., LEAN, M. E. J., ASHIHARA, H. & CROZIER, A. 2014. Coffee: biochemistry and potential impact on health. Food and Function, 5, 1695-1717.

MACCARI JUNIOR, A. 2005. Análise do pré-processamento da erva-mate para chimarrão, PhD thesis, UNICAMP, Campinas, 164.

MACIEL, P. S. F., FERREIRA, A. V., SILVA, R. C. A., DE MELO, C. M. R. F. & UCHÔA, S. B. B. 2018. Monitoramento Tecnológico Sobre A Utilização Do Açaí Na Indústria De Alimentos. Cadernos de Prospecção, 11, 103.

MAGRI, E., MOTTA, A. C. V., VALDUGA, A. T., OLIVEIRA, A., MENEZES, I. M. N. R. & DE ANDRADE NASCIMENTO, P. Teores de Mn, Al, Fe, Zn, Cu, Ni, Cd e Pb em solos de ervais e folhas de erva-mate. Reunião Paranaense de Ciência do Solo, 3.

MALIK, J., SZAKOVA, J., DRABEK, O., BALIK, J. & KOKOSKA, L. 2008. Determination of certain micro and macroelements in plant stimulants and their infusions. Food Chemistry, 111, 520-525.

MANACH, C., SCALBERT, A., MORAND, C., RÉMÉSY, C. & JIMÉNEZ, L. 2004. Polyphenols: food sources and bioavailability. The American Journal of Clinical Nutrition, 79, 727-747.

MANSUR, A. V., BRONDÍZIO, E. S., ROY, S., HETRICK, S., VOGT, N. D. & NEWTON, A. 2016. An assessment of urban vulnerability in the Amazon Delta and Estuary: a multi-criterion index of flood exposure, socio-economic conditions and infrastructure. Sustainability Science, 11, 625-643.

MAPS, G. 2019. Available: https://www.google.co.uk/maps [Accessed 08/07/2019 2019].

MARCASON, W. 2009. What Is the Açaí Berry and Are There Health Benefits? Journal of the American Dietetic Association, 109, 1968.

MARIA DO SOCORRO, M. R., ALVES, R. E., DE BRITO, E. S., PÉREZ-JIMÉNEZ, J., SAURA-CALIXTO, F. & MANCINI-FILHO, J. 2010. Bioactive compounds and antioxidant capacities of 18 non-traditional tropical fruits from Brazil. Food Chemistry, 121, 996-1002.

MARQUES, V. & FARAH, A. 2009. Chlorogenic acids and related compounds in medicinal plants and infusions. Food Chemistry, 113, 1370-1376.

MARSCHNER, H. 2011. Marschner's mineral nutrition of higher plants, Academic press.

MARTÍN‐PELÁEZ, S., COVAS, M. I., FITÓ, M., KUSAR, A. & PRAVST, I. 2013. Health effects of olive oil polyphenols: recent advances and possibilities for the use of health claims. Molecular Nutrition and Food Research, 57, 760-771.

MARTIN, M. J., PABLOS, F. & GONZALEZ, A. G. 1996. Application of pattern recognition to the discrimination of roasted coffees. Analytica Chimica Acta, 320, 191-197.

247

MARTıN, M. J., PABLOS, F. & GONZALEZ, A. G. 1999. Characterization of arabica and robusta roasted coffee varieties and mixture resolution according to their metal content. Food Chemistry, 66, 365-370.

MARTÍNEZ-LÓPEZ, S., SARRIÁ, B., GÓMEZ-JUARISTI, M., GOYA, L., MATEOS, R. & BRAVO-CLEMENTE, L. 2014. Theobromine, caffeine, and theophylline metabolites in human plasma and urine after consumption of soluble cocoa products with different methylxanthine contents. Food Research International, 63, 446-455.

MATHEUS, M. E., FERNANDES, S. B. D. O., SILVEIRA, C. S., RODRIGUES, V. P., MENEZES, F. D. S. & FERNANDES, P. D. 2006. Inhibitory effects of Euterpe oleracea Mart. on nitric oxide production and iNOS expression. Journal of Ethnopharmacology, 107, 291-296.

MATHEUS, M. E., MANTOVANI, I. S. B., SANTOS, G. B., FERNANDES, S. B. O., MENEZES, F. S. & FERNANDES, P. D. 2003. Ação de extratos do açaí (Euterpe oleracea Mart.) sobre a produção de óxido nítrico em células RAW 264.7. Revista Brasileira de Farmacognosia, 13, 3-5.

MAZZAFERA, P. 1997. Maté drinking: caffeine and phenolic acid intake. Food chemistry, 60, 67-71.

MELO, P. S., ARRIVETTI, L. D. O. R., ALENCAR, S. M. D. & SKIBSTED, L. H. 2016. Antioxidative and prooxidative effects in food lipids and synergism with α-tocopherol of açaí seed extracts and grape rachis extracts. Food Chemistry, 213, 440-449.

MENEZES, E. M., TORRES, A. T. & SRUR, A. U. S. 2008a. Valor nutricional da polpa de açaí (Euterpe oleracea Mart) liofilizada. Acta Amazonica, 38, 311- 316.

MENEZES, E. M. D. S., ROSENTHAL, A., SABAA-SRUR, A., CAMARGO, L., CALADO, V. & SANTOS, A. 2008b. High hydrostatic pressure effect on enzyme activity of açai pulp. Food Science and Technology, 28, 14-19.

MERTENS-TALCOTT, S. U., RIOS, J., JILMA-STOHLAWETZ, P., PACHECO-PALENCIA, L. A., MEIBOHM, B., TALCOTT, S. T. & DERENDORF, H. 2008. Pharmacokinetics of Anthocyanins and Antioxidant Effects after the Consumption of Anthocyanin-Rich Açai Juice and Pulp (Euterpe oleracea Mart.) in Human Healthy Volunteers. Journal of Agricultural and Food Chemistry, 56, 7796-7802.

MERTZ, W. 1981. The essential trace elements. Science, 213, 1332-1338.

MICHALKE, B., HALBACH, S. & NISCHWITZ, V. 2007. Speciation and toxicological relevance of manganese in humans. Journal of Environmental Monitoring, 9, 650-656.

MICHIE, N. D. & DIXON, J. E. 1977. Distribution of lead and other metals in tea leaves, dust and liquors. Journal of the Science of Food and Agriculture, 28, 215-224.

MILANI, R. F., MORGANO, M. A. & CADORE, S. 2016. Trace elements in Camellia sinensis marketed in southeastern Brazil: Extraction from tea leaves to beverages and dietary exposure. LWT - Food Science and Technology, 68, 491-498.

MILLER, J. N., MILLER, J. C. & MILLER, R. D. 2018. Statistics and chemometrics for analytical chemistry, Pearson Education Canada, 296.

248

MOLINO, J. A., KENNEDY, J. F., BEUSE, P. A., MILLER, C. C., DAVIS, W. & ANDERSEN, C. K. 2013. Daytime Color Appearance of Retroreflective Traffic Control Sign Materials. United States. Federal Highway Administration, 5-7.

MOON, J.-K., YOO, H. S. & SHIBAMOTO, T. 2009. Role of roasting conditions in the level of chlorogenic acid content in coffee beans: correlation with coffee acidity. Journal of Agricultural and Food Chemistry, 57, 5365-5369.

MOREDA-PIÑEIRO, J., SÁNCHEZ-PIÑERO, J., MAÑANA-LÓPEZ, A., TURNES-CAROU, I., ALONSO-RODRÍGUEZ, E., LÓPEZ-MAHÍA, P. & MUNIATEGUI-LORENZO, S. 2018. Multi-element determinations in foods from Amazon region by ICP-MS after enzymatic hydrolysis assisted by pressurisation and microwave energy. Microchemical Journal, 137, 402-409.

MOTULSKY, H. 2014. Intuitive biostatistics: a nonmathematical guide to statistical thinking, Oxford University Press, USA, Oxford University Press, 608.

MURAKAMI, A. N. N., AMBONI, R. D. D. M. C., PRUDÊNCIO, E. S., AMANTE, E. R., FRITZEN-FREIRE, C. B., BOAVENTURA, B. C. B., DE BONA MUÑOZ, I., DOS SANTOS BRANCO, C., SALVADOR, M. & MARASCHIN, M. 2013. Concentration of biologically active compounds extracted from Ilex paraguariensis St. Hil. by nanofiltration. Food Chemistry, 141, 60-65.

MURSITO, A. T., DIDA, E. N. & ERLANGGA, B. D. Potential Application of Silica Mineral from Dieng Mountain in Agriculture Sector to Control the Release Rate of Fertilizer Elements. 2017. IOP Publishing, 214 (1), 012030.

NEAL, A. P. & GUILARTE, T. R. 2012. Mechanisms of heavy metal neurotoxicity: lead and manganese. J Drug Metab Toxicol S, 5, 1-13.

NGC. 2015. NGC Chromatography [Online]. Available: http://eppingcache.net/LQB681ScROLL/681_C7_NGC_operation.htm.

NIELSEN, F. H. 2007. "Trace Mineral Deficiencies." Handbook of nutrition and food. CRC Press, 177-194.

NIELSEN, F. H. & HUNT, J. R. 1989. Trace elements emerging as important in human nutrition. In Proceedings of the fourteenth national databank conference, PJ Stumbo, ed., Univ. Iowa, Iowa City, 19-21.

NIESSEN, W. M. A. 2006. Liquid Chromatography-Mass Spectrometry, CRC press.

OBUCHOWICZ, J., ENGELHARDT, U. H. & DONNELLY, K. 2011. Flavanol database for green and black teas utilising ISO 14502-1 and ISO 14502-2 as analytical tools. Journal of Food Composition and Analysis, 24, 411-417.

OLIVA, S. R. & ESPINOSA, A. J. F. 2007. Monitoring of heavy metals in topsoils, atmospheric particles and plant leaves to identify possible contamination sources. Microchemical Journal, 86, 131-139.

OLIVEIRA, M., DE CARVALHO, J., DO NASCIMENTO, W. & MULLER, C. 2002. Cultivo do açaizeiro para produção de frutos. Embrapa Amazônia Oriental. Circular técnica, 1-17.

ORIS. 2017. ORIS™ CELL MIGRATION ASSAY [Online]. [Accessed 30/01/2019 2019].

ORTOLÁ, M. D., LONDONO, L., GUTIÉRREZ, C. L. & CHIRALT, A. 1998. Influence of roasting temperature on physicochemical properties of different coffees. Food Science And Technology International, 4, 59-66.

249

OVASKAINEN, M.-L., TORRONEN, R., KOPONEN, J. M., SINKKO, H., HELLSTROM, J., REINIVUO, H. & MATTILA, P. 2008. Dietary intake and major food sources of polyphenols in Finnish adults. The Journal of Nutrition, 138, 562-566.

PACHECO-PALENCIA, L. A., DUNCAN, C. E. & TALCOTT, S. T. 2009. Phytochemical composition and thermal stability of two commercial açai species, Euterpe oleracea and Euterpe precatoria. Food Chemistry, 115, 1199-1205.

PACHECO-PALENCIA, L. A., MERTENS-TALCOTT, S. & TALCOTT, S. T. 2008. Chemical composition, antioxidant properties, and thermal stability of a phytochemical enriched oil from Acai (Euterpe oleracea Mart.). Journal of Agricultural and Food Chemistry, 56, 4631-4636.

PAGLIARUSSI, M. S. 2010. A cadeia produtiva agroindustrial do açaí: estudo da cadeia e proposta de um modelo matemático. BS thesis, USP, São Paulo, 37.

PAVAN, M. A., SCHMIDT, S. J. & FENG, H. 2012. Water sorption behavior and thermal analysis of freeze-dried, Refractance Window-dried and hot-air dried açaí (Euterpe oleracea Martius) juice. LWT - Food Science and Technology, 48, 75-81.

PAVIA, D. L., LAMPMAN, G. M., KRIZ, G. S. & VYVYAN, J. A. 2014. Introduction to spectroscopy, Cengage Learning, 376.

PENDERGRAST, M. 2010. Uncommon grounds: The history of coffee and how it transformed our world, Basic Books, 53.

PÉREZ-JIMÉNEZ, J., NEVEU, V., VOS, F. & SCALBERT, A. 2010. Identification of the 100 richest dietary sources of polyphenols: an application of the Phenol-Explorer database. European Journal Of Clinical Nutrition, 64, S112.

PERRON, N. R. & BRUMAGHIM, J. L. 2009. A Review of the Antioxidant Mechanisms of Polyphenol Compounds Related to Iron Binding. Cell Biochemistry and Biophysics, 53, 75-100.

PESSOA, J. D. C. & TEIXEIRA, G. D. A. 2012. Tecnologias para inovação nas cadeias euterpe. Embrapa Instrumentação-Livros científicos (ALICE), 3-9.

PINTOS, J., FRANCO, E. L., OLIVEIRA, B. V., KOWALSKI, L. P., CURADO, M. P. & DEWAR, R. 1994. Mate, coffee, and tea consumption and risk of cancers of the upper aerodigestive tract in southern Brazil. Epidemiology, 583-590.

PLAZA, M., DOMÍNGUEZ-RODRÍGUEZ, G., CASTRO-PUYANA, M. & MARINA, M. L. 2018. 6 - Polyphenols analysis and related challenges. In: GALANAKIS, C. M. (ed.) Polyphenols: Properties, Recovery, and Applications. Woodhead Publishing, 295-323.

POHL, P. & PRUSISZ, B. 2007. Fractionation analysis of manganese and zinc in tea infusions by two-column solid phase extraction and flame atomic absorption spectrometry. Food Chemistry, 102, 1415-1424.

POHL, P., STELMACH, E., WELNA, M. & SZYMCZYCHA-MADEJA, A. 2013. Determination of the Elemental Composition of Coffee Using Instrumental Methods. Food Analytical Methods, 6, 598-613.

POMPEU, D. R., SILVA, E. M. & ROGEZ, H. 2009. Optimisation of the solvent extraction of phenolic antioxidants from fruits of Euterpe oleracea using Response Surface Methodology. Bioresource Technology, 100, 6076-6082.

250

POTSCH, A. 2010. Aromas e Sabores [Online]. Available: http://www.aromasesabores.com/2010/06/acai-do-pe-tijela.html [Accessed 30/10/2018.

POULOSE, S. M., FISHER, D. R., LARSON, J., BIELINSKI, D. F., RIMANDO, A. M., CAREY, A. N., SCHAUSS, A. G. & SHUKITT-HALE, B. 2012. Anthocyanin-rich Açai (Euterpe oleracea Mart.) Fruit Pulp Fractions Attenuate Inflammatory Stress Signaling in Mouse Brain BV-2 Microglial Cells. Journal of Agricultural and Food Chemistry, 60, 1084-1093.

POZEBON, D., DRESSLER, V. L., MARCELO, M. C. A., DE OLIVEIRA, T. C. & FERRÃO, M. F. 2015. Toxic and nutrient elements in yerba mate (Ilex paraguariensis). Food Additives & Contaminants: Part B, 8, 215-220.

PRASHANTH, L., KATTAPAGARI, K. K., CHITTURI, R. T., BADDAM, V. R. R. & PRASAD, L. K. 2015. A review on role of essential trace elements in health and disease. Journal of dr. ntr university of health sciences, 4, 75.

PRIOR, R. L., WU, X. & SCHAICH, K. 2005. Standardized methods for the determination of antioxidant capacity and phenolics in foods and dietary supplements. Journal of Agricultural and Food Chemistry, 53, 4290-4302.

PUIG, S. & PEÑARRUBIA, L. 2009. Placing metal micronutrients in context: transport and distribution in plants. Current Opinion In Plant Biology, 12, 299-306.

RAGUZ, V., JARSJO, J., GROLANDER, S., LINDBORG, R. & AVILA, R. 2013. Plant uptake of elements in soil and pore water: Field observations versus model assumptions. Journal Of Environmental Management, 126, 147-156.

RAMIREZ-MARES, M. V., CHANDRA, S. & DE MEJIA, E. G. 2004. In vitro chemopreventive activity of Camellia sinensis, Ilex paraguariensis and Ardisia compressa tea extracts and selected polyphenols. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 554, 53-65.

RAVEN, J. A., EVANS, M. C. W. & KORB, R. E. 1999. The role of trace metals in photosynthetic electron transport in O 2-evolving organisms. Photosynthesis Research, 60, 111-150.

RE, R., PELLEGRINI, N., PROTEGGENTE, A., PANNALA, A., YANG, M. & RICE-EVANS, C. 1999. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biology and Medicine, 26, 1231-1237.

REICHEL, C. M., WACAN, J. J., FARLEY, C. M., STANLEY, B. J., CRAWFORD, C. A. & MCDOUGALL, S. A. 2006. Postnatal manganese exposure attenuates cocaine-induced locomotor activity and reduces dopamine transporters in adult male rats. Neurotoxicology and teratology, 28, 323-332.

RESEARCH, C. 2018. Acai Berry Market By Form (Dried, Pulp), By Application (Food & Beverages, Nutraceuticals, Cosmetics, Pharmaceuticals) - Growth, Future Prospects & Competitive Analysis, 2017 – 2025 [Online]. Available: https://www.credenceresearch.com/report/acai-berry-market [Accessed 30/10/2018.

REYNOLDS, E. B., SCHNEIDER, B. R. & ZYLBERBERG, E. 2019. Innovation in Brazil: Advancing Development in the 21st Century, Routledge, 376.

RIBEIRO, J. C., ANTUNES, L. M. G., AISSA, A. F., DARIN, J. D. A. C., DE ROSSO, V. V., MERCADANTE, A. Z. & BIANCHI, M. D. L. P. 2010. Evaluation of the

251

genotoxic and antigenotoxic effects after acute and subacute treatments with açai pulp (Euterpe oleracea Mart.) on mice using the erythrocytes micronucleus test and the comet assay. Mutation Research/Genetic Toxicology and Environmental Mutagenesis, 695, 22-28.

ROCHA, A. P. M., CARVALHO, L. C. R. M., SOUSA, M. A. V., MADEIRA, S. V. F., SOUSA, P. J. C., TANO, T., SCHINI-KERTH, V. B., RESENDE, A. C. & SOARES DE MOURA, R. 2007. Endothelium-dependent vasodilator effect of Euterpe oleracea Mart. (Açaí) extracts in mesenteric vascular bed of the rat. Vascular Pharmacology, 46, 97-104.

RODRIGUES, R. B., LICHTENTHÄLER, R., ZIMMERMANN, B. F., PAPAGIANNOPOULOS, M., FABRICIUS, H., MARX, F., MAIA, J. G. S. & ALMEIDA, O. 2006. Total Oxidant Scavenging Capacity of Euterpe oleracea Mart. (Açaí) Seeds and Identification of Their Polyphenolic Compounds. Journal of Agricultural and Food Chemistry, 54, 4162-4167.

ROGEZ, H. 2000. Açaí preparo, composição e melhoramento da conservação, Belém, PA, EDUFPA, 164.

ROGEZ, H., POMPEU, D. R., AKWIE, S. N. T. & LARONDELLE, Y. 2011. Sigmoidal kinetics of anthocyanin accumulation during fruit ripening: A comparison between açai fruits (Euterpe oleracea) and other anthocyanin-rich fruits. Journal of Food Composition and Analysis, 24, 796-800.

ROJANO, B. A., VAHOS, I. C. Z., ARBELÁEZ, A. F. A., MARTÍNEZ, A. J. M., CORREA, F. B. C. & CARVAJAL, L. G. 2011. Polifenoles y actividad antioxidante del fruto liofilizado de palma naidi (açai colombiano)(Euterpe oleracea Mart). Revista Facultad Nacional de Agronomía, 64.

ROSSA, Ü. B., ANGELO, A. C., NISGOSKI, S., WESTPHALEN, D. J., FRIZON, C. N. T. & HOFFMANN-RIBANI, R. 2015. Application of the NIR Method to Determine Nutrients in Yerba Mate (Ilex paraguariensis A. St.-Hill) Leaves. Communications in Soil Science and Plant Analysis, 46, 2323-2331.

ROTH, J. A., LI, Z., SRIDHAR, S. & KHOSHBOUEI, H. 2013. THE EFFECT OF MANGANESE ON DOPAMINE TOXICITY AND DOPAMINE TRANSPORTER (DAT) IN CONTROL AND DAT TRANSFECTED HEK CELLS. Neurotoxicology, 35, 121-128.

ROTHWELL, J. A., PEREZ-JIMENEZ, J., NEVEU, V., MEDINA-REMON, A., M'HIRI, N., GARCÍA-LOBATO, P., MANACH, C., KNOX, C., EISNER, R. & WISHART, D. S. 2013. Phenol-Explorer 3.0: a major update of the Phenol-Explorer database to incorporate data on the effects of food processing on polyphenol content. Database.

ROUSSEAU, T. C. C., SONKE, J. E., CHMELEFF, J., CANDAUDAP, F., LACAN, F., BOAVENTURA, G., SEYLER, P. & JEANDEL, C. 2013. Rare earth element analysis in natural waters by multiple isotope dilution–sector field ICP-MS. Journal of Analytical Atomic Spectrometry, 28, 573-584.

RUFINO, M. D. S. M., ALVES, R. E., DE BRITO, E. S., PÉREZ-JIMÉNEZ, J., SAURA-CALIXTO, F. & MANCINI-FILHO, J. 2010. Bioactive compounds and antioxidant capacities of 18 non-traditional tropical fruits from Brazil. Food Chemistry, 121, 996-1002.

252

RUFINO, M. S., ALVES, R. E., FERNANDES, F. A. & BRITO, E. S. 2011. Free radical scavenging behavior of ten exotic tropical fruits extracts. Food Research International, 44, 2072-2075.

RUIZ, L., RUIZ, L., MACO, M., COBOS, M., GUTIERREZ-CHOQUEVILCA, A.-L. & ROUMY, V. 2011. Plants used by native Amazonian groups from the Nanay River (Peru) for the treatment of malaria. Journal of Ethnopharmacology, 133, 917-921.

RUMSBY, P., CLEGG, H., JONSSON, J., BENSON, V., HARMAN, M., DOYLE, T., RUSHTON, L., WARWICK L. & WILKINSON, D. 2014. Speciation of Manganese in Drinking Water. report for Drinking Water Inspectorate , London: Imperial College, 1-37.

RUSINEK-PRYSTUPA, E., MARZEC, Z., SEMBRATOWICZ, I., SAMOLIŃSKA, W., KICZOROWSKA, B. & KWIECIEŃ, M. 2016. Content of Selected Minerals and Active Ingredients in Teas Containing Yerba Mate and Rooibos. Biological Trace Element Research, 172, 266-275.

SAIDELLES, A. P. F., KIRCHNER, R. M., SANTOS, N. R. Z., FLORES, É. M. & BARTZ, F. R. 2010. Brazilian Journal of Food and Nutrition, 21, 259.

SALTMARSH, M. & GOLDBERG, G. 2003. Miscellaneous foods. in: Plants: Diet and Health, Ed G. Goldberg. Oxford, UK: Blackwell Publishing, 210-225.

SANCHEZ, L. W. & WATSON, R. R. 2016. Fruits, vegetables, and herbs: Bioactive foods promoting wound healing. Fruits, Vegetables, and Herbs. Elsevier.

SANTOS, V. S., NARDINI, V., CUNHA JR, L. C., BARBOSA JR, F. & TEIXEIRA, G. H. D. A. 2014a. Identification of species of the Euterpe genus by rare earth elements using inductively coupled plasma mass spectrometry and linear discriminant analysis. Food Chemistry, 153, 334-339.

SANTOS, V. S., NARDINI, V., CUNHA, L. C., BARBOSA, F. & DE ALMEIDA TEIXEIRA, G. H. 2014b. Identification of species of the Euterpe genus by rare earth elements using inductively coupled plasma mass spectrometry and linear discriminant analysis. Food Chemistry, 153, 334-339.

SANTOS, W. P. C., HATJE, V., LIMA, L. N., TRIGNANO, S. V., BARROS, F., CASTRO, J. T. & KORN, M. G. A. 2008. Evaluation of sample preparation (grinding and sieving) of bivalves, coffee and cowpea beans for multi-element analysis. Microchemical Journal, 89, 123-130.

SAURA-CALIXTO, F. & GOÑI, I. 2006. Antioxidant capacity of the Spanish Mediterranean diet. Food Chemistry, 94, 442-447.

SAXENA, M., SAXENA, J., NEMA, R., SINGH, D. & GUPTA, A. 2013. Phytochemistry of medicinal plants. Journal Of Pharmacognosy And Phytochemistry, 1.

SCALBERT, A. & WILLIAMSON, G. 2000. Dietary intake and bioavailability of polyphenols. The Journal Of Nutrition, 130, 2073S-2085S.

SCHAUSS, A. G. 2013. Açai Fruits: Potent Antioxidant and Anti‐Inflammatory Superfruits with Potential Health Benefits. Dried Fruits, 393-413.

SCHAUSS, A. G., WU, X., PRIOR, R. L., OU, B., HUANG, D., OWENS, J., AGARWAL, A., JENSEN, G. S., HART, A. N. & SHANBROM, E. 2006a. Antioxidant Capacity and Other Bioactivities of the Freeze-Dried Amazonian Palm Berry, Euterpe

253

oleraceae Mart. (Acai). Journal of Agricultural and Food Chemistry, 54, 8604-8610.

SCHAUSS, A. G., WU, X., PRIOR, R. L., OU, B., PATEL, D., HUANG, D. & KABABICK, J. P. 2006b. Phytochemical and Nutrient Composition of the Freeze-Dried Amazonian Palm Berry, Euterpe oleraceae Mart. (Acai). Journal of Agricultural and Food Chemistry, 54, 8598-8603.

SCHENKER, S., HANDSCHIN, S., FREY, B., PERREN, R. & ESCHER, F. 2000. Pore structure of coffee beans affected by roasting conditions. Journal of Food Science, 65, 452-457.

SCHINELLA, G., FANTINELLI, J. C. & MOSCA, S. M. 2005. Cardioprotective effects of Ilex paraguariensis extract: evidence for a nitric oxide-dependent mechanism. Clinical Nutrition, 24, 360-366.

SCHMALKO, M. E. & ALZAMORA, S. M. 2001. Color, chlorophyll, caffeine, and water content variation during yerba mate processing. Drying Technology, 19, 599-610.

SCHRADER, K., KIEHNE, A., ENGELHARDT, U. H. & GERHARD MAIER, H. 1996. Determination of chlorogenic acids with lactones in roasted coffee. Journal of the Science of Food and Agriculture, 71, 392-398.

SCHRECKINGER, M. E., LOTTON, J., LILA, M. A. & DE MEJIA, E. G. 2010. Berries from South America: a comprehensive review on chemistry, health potential, and commercialization. Journal Of Medicinal Food, 13, 233-246.

SCHUMANN, K. 2006. Dietary reference intakes for trace elements revisited. J Trace Elem Med Biol, 20, 59-61.

SCHWARTZBERG, H. G. 2002. Modeling bean heating during batch roasting of coffee beans. Engineering and Food for the 21st Century. CRC Press, 901-920.

SELMA, M. V., ESPIN, J. C. & TOMAS-BARBERAN, F. A. 2009. Interaction between phenolics and gut microbiota: role in human health. Journal Of Agricultural And Food Chemistry, 57, 6485-6501.

SHAHIDI, F. & ZHONG, Y. 2015. Measurement of antioxidant activity. Journal Of Functional Foods, 18, 757-781.

SILVA, M., BARRETTO, W. S. & SERÔDIO, M. H. 2004. Comparação nutricional da polpa dos frutos de juçara e de açaí. Ilhéus: Centro de Pasquisa do Cacau, Ministério da Agricultura, Agropecuária e Abastecimento, 18, 1-13.

SINGH, A., HOLVOET, S. & MERCENIER, A. 2011. Dietary polyphenols in the prevention and treatment of allergic diseases. Clinical & Experimental Allergy, 41, 1346-1359.

SINGLETON, V. L., ORTHOFER, R. & LAMUELA-RAVENTÓS, R. M. 1999a. [14] Analysis of total phenols and other oxidation substrates and antioxidants by means of folin-ciocalteu reagent. Methods in Enzymology. Academic Press, 299, 152-178.

SKALNAYA, M. G. & SKALNY, A. V. 2018. Essential trace elements in human health: a physician’s view. Publishing House of Tomsk State University, Tomsk, 151-199.

SKATES, E., OVERALL, J., DEZEGO, K., WILSON, M., ESPOSITO, D., LILA, M. A. & KOMARNYTSKY, S. 2018. Berries containing anthocyanins with enhanced

254

methylation profiles are more effective at ameliorating high fat diet-induced metabolic damage. Food and Chemical Toxicology, 111, 445-453.

SKOOG, D. A., HOLLER, F. J. & CROUCH, S. R. 2017. Principles Of Instrumental Analysis, Cengage learning, Boston, 927.

SNYDER, L. R., KIRKLAND, J. J. & DOLAN, J. W. 2011. Introduction To Modern Liquid Chromatography, John Wiley & Sons, Chichester, 35-39.

SPADA, P. D., DANI, C., BORTOLINI, G. V., FUNCHAL, C., HENRIQUES, J. A. & SALVADOR, M. 2009. Frozen fruit pulp of Euterpe oleraceae Mart.(Acai) prevents hydrogen peroxide-induced damage in the cerebral cortex, cerebellum, and hippocampus of rats. Journal Of Medicinal Food, 12, 1084-1088.

STELMACH, E., POHL, P. & SZYMCZYCHA-MADEJA, A. 2015. The content of Ca, Cu, Fe, Mg and Mn and antioxidant activity of green coffee brews. Food Chemistry, 182, 302-308.

STEVENSON, D. E. & HURST, R. D. 2007. Polyphenolic phytochemicals–just antioxidants or much more? Cellular and Molecular Life Sciences, 64, 2900-2916.

STODT, U. & ENGELHARDT, U. H. 2013. Progress in the analysis of selected tea constituents over the past 20 years. Food Research International, 53, 636-648.

STRACHAN, S. 2010. Trace elements. Current Anaesthesia & Critical Care, 21, 44-48.

SUS. 2018. Sistema de cadastramento e acompanhamento de hipertensos e diabéticos/Pará [Online]. Available: http://tabnet.datasus.gov.br/cgi/tabcgi.exe?hiperdia/cnv/hdPA.def [Accessed 08/07/2019 2019].

SUSEELA, B., BHALKE, S. & VINOD KUMAR, A. 2001. Daily intake of trace metals through coffee consumption in India. Food Additives & Contaminants, 18, 115-120.

TAGLIAFERRO, F. S., FERNANDES, E. A. D. N., BACCHI, M. A., BODE, P. & DE FRANÇA, E. J. 2007. Can impurities from soil-contaminated coffees reach the cup? Journal Of Radioanalytical And Nuclear Chemistry, 271, 371-375.

TAGORE, M. D. P. B., DO CANTO, O. & SOBRINHO, M. V. 2018. Políticas públicas e riscos ambientais em áreas de várzea na Amazônia: o caso do PRONAF para produção do açaí. Desenvolvimento e Meio Ambiente, 45, 3-13.

TAMASI, O. P., FILIP, R., FERRARO, G. & CALVINO, A. M. 2007. Total polyphenol content and perceived astringency of yerba mate “Ilex paraguariensis” infusions. Journal Of Sensory Studies, 22, 653-664.

TESTER, M. & LEIGH, R. A. 2001. Partitioning of nutrient transport processes in roots. Journal of Experimental Botany, 52, 445-457.

THOMAS, R. 2013. Practical guide to ICP-MS: a tutorial for beginners, CRC press, Boca Raton, 402.

TIMMERS, M. A., GRACE, M. H., YOUSEF, G. G. & LILA, M. A. 2017. Inter- and intra-seasonal changes in anthocyanin accumulation and global metabolite profiling of six blueberry genotypes. Journal of Food Composition and Analysis, 59, 105-110.

255

TINKER, P. B. 1981. Levels, distribution and chemical forms of trace elements in food plants. Philosophical Transactions of the Royal Society of London. B, Biological Sciences, 294, 41-55.

TONON, R. V., BARONI, A. F., BRABET, C., GIBERT, O., PALLET, D. & HUBINGER, M. D. 2009. Water sorption and glass transition temperature of spray dried açai (Euterpe oleracea Mart.) juice. Journal of Food Engineering, 94, 215-221.

TORRES, T. & FARAH, A. 2017. Coffee, maté, açaí and beans are the main contributors to the antioxidant capacity of Brazilian’s diet. European Journal Of Nutrition, 56, 1523-1533.

TRÄNKNER, M., TAVAKOL, E. & JÁKLI, B. 2018. Functioning of potassium and magnesium in photosynthesis, photosynthate translocation and photoprotection. Physiologia Plantarum, 163, 414-431.

TRUGO, L. C., MACRAE, R. & CLARKE, R. J. 1985. Coffee, vol. 1. chemistry. Elsevier applied science Publishers, London, 97.

TRUONG, V. D., MCFEETERS, R. F., THOMPSON, R. T., DEAN, L. L. & SHOFRAN, B. 2007. Phenolic Acid Content and Composition in Leaves and Roots of Common Commercial Sweetpotato (Ipomea batatas L.) Cultivars in the United States. Journal of Food Science, 72, C343-C349.

TSAO, R. 2010. Chemistry and Biochemistry of Dietary Polyphenols. Nutrients, 2, 1231-1246.

TURECK, C., LOCATELI, G., CORRÊA, V. G. & KOEHNLEIN, E. A. 2017. Evaluation of the Brazilian population’s intake of antioxidant nutrients and their relation with the nutritional status. Revista Brasileira de Epidemiologia, 20, 30-42.

UFRS. 2012. Erva mate: processamento [Online]. Available: http://www.ufrgs.br/alimentus1/objetos/erva-mate/#/processamento.html [Accessed 23/06/2019 2019].

UNDERWOOD, E. & MERTZ, W. 1979. Trace elements in human and animal nutrition, Elsevier, London, 317.

UNICAMP, N. 2011. Tabela brasileira de composição de alimentos–TACO. UNICAMP; NEPA, Campinas, 57.

VALDUGA, A. T., FINZER, J. R. D. & MOSELE, S. H. 2003. Processamento de erva-mate. Erechim: Edifapes, 183.

VANAMALA, J., REDDIVARI, L., YOO, K. S., PIKE, L. M. & PATIL, B. S. 2006. Variation in the content of bioactive flavonoids in different brands of orange and grapefruit juices. Journal of Food Composition and Analysis, 19, 157-166.

VANNOORT, R. W. & THOMSON, B. M. 2003. New Zealand total diet survey. Agricultural Compound Residues, Selected Contaminants and Nutrients. New Zealand Food Safety Authority, Wellington, 37.

VASCONCELLOS, F. C. F. D. 2012. Os impactos da criação do MERCOSUL no mercado de erva-mate no Rio Grande do Sul, BS thesis, UFRS, Porto Alegre, 33..

VASSALLO, A., CORREA, P., DE STÉFANI, E., CENDÁN, M., ZAVALA, D., CHEN, V., CARZOGLIO, J. & DENEO-PELLEGRINI, H. 1985. Esophageal cancer in

256

Uruguay: a case-control study. Journal of the National Cancer Institute, 75, 1005-1009.

VEGA-CARRILLO, H., ISKANDER, F. & MANZANARES-ACUNA, E. 2002. Elemental content in ground and soluble/instant coffee. Journal Of Radioanalytical And Nuclear Chemistry, 252, 75-80.

VERA DE ROSSO, V., HILLEBRAND, S., CUEVAS MONTILLA, E., BOBBIO, F. O., WINTERHALTER, P. & MERCADANTE, A. Z. 2008. Determination of anthocyanins from acerola (Malpighia emarginata DC.) and açai (Euterpe oleracea Mart.) by HPLC–PDA–MS/MS. Journal of Food Composition and Analysis, 21, 291-299.

VERBRUGGEN, N., HERMANS, C. & SCHAT, H. 2009. Mechanisms to cope with arsenic or cadmium excess in plants. Current opinion in plant biology, 12, 364-372.

VERSIECK, J. & CORNELIS, R. 1980. Normal levels of trace elements in human blood plasma or serum. Analytica Chimica Acta, 116, 217-254.

VIDA, M. 2010. Colheita do açaí [Online]. https://moradaviva.wordpress.com/2010/11/29/colheita-e-festa-da-jucara-2010/. Available: https://moradaviva.wordpress.com/2010/11/29/colheita-e-festa-da-jucara-2010/ [Accessed 22/01/2019 2019].

VOET, D., VOET, J. & PRATT, C. 2008. The Major Pathways of Purine Catabolism in Animals. Fundamentals of Biochemistry: Life at the Molecular Level, 840.

WARD, N. 2000. Trace elements., Blackwell Science, Oxford, 345.

WARD, N. I. & MARCILLA, A. L. 2003. Trace element levels of soils and yerba mate samples from Amanda Plantation, Misiones, Argentina; . report for Amanda, Apóstoles, Misiones, Argentina. Guildford, Surrey: University of Surrey.

WEI, F. & TANOKURA, M. 2015. Chapter 10 - Chemical Changes in the Components of Coffee Beans during Roasting. In: PREEDY, V. R. (ed.) Coffee in Health and Disease Prevention. San Diego: Academic Press.

WHO 1996. Trace elements in human nutrition and health, World Health Organization, 165.

WILD, T., RAHBARNIA, A., KELLNER, M., SOBOTKA, L. & EBERLEIN, T. 2010. Basics in nutrition and wound healing. Nutrition, 26, 862-866.

WRÓBEL, K., WRÓBEL, K. & URBINA, E. M. C. 2000. Determination of total aluminum, chromium, copper, iron, manganese, and nickel and their fractions leached to the infusions of black tea, green tea, Hibiscus sabdariffa, and Ilex paraguariensis (mate) by ETA-AAS. Biological Trace Element Research, 78, 271-280.

WYCOFF, W., LUO, R., SCHAUSS, A. G., NEAL-KABABICK, J., SABAA-SRUR, A. U. O., MAIA, J. G. S., TRAN, K., RICHARDS, K. M. & SMITH, R. E. 2015. Chemical and nutritional analysis of seeds from purple and white açaí (Euterpe oleracea Mart.). Journal of Food Composition and Analysis, 41, 181-187.

XIAO, J. B. & HOGGER, P. 2015. Dietary polyphenols and type 2 diabetes: current insights and future perspectives. Current Medicinal Chemistry, 22, 23-38.

XIE, C., KANG, J., BURRIS, R., FERGUSON, M. E., SCHAUSS, A. G., NAGARAJAN, S. & WU, X. 2011. Açaí juice attenuates atherosclerosis in ApoE deficient mice

257

through antioxidant and anti-inflammatory activities. Atherosclerosis, 216, 327-333.

XIE, C., KANG, J., LI, Z., SCHAUSS, A. G., BADGER, T. M., NAGARAJAN, S., WU, T. & WU, X. 2012. The açaí flavonoid velutin is a potent anti-inflammatory agent: blockade of LPS-mediated TNF-α and IL-6 production through inhibiting NF-κB activation and MAPK pathway. The Journal Of Nutritional Biochemistry, 23, 1184-1191.

YAMADA, N. 2015. Kinetic energy discrimination in collision/reaction cell ICP-MS: Theoretical review of principles and limitations. Spectrochimica Acta Part B: Atomic Spectroscopy, 110, 31-44.

YAMAGUCHI, K. K. D. L., PEREIRA, L. F. R., LAMARÃO, C. V., LIMA, E. S. & DA VEIGA-JUNIOR, V. F. 2015. Amazon acai: Chemistry and biological activities: A review. Food Chemistry, 179, 137-151.

YOSHINO, M. & MURAKAMI, K. 1998. Interaction of iron with polyphenolic compounds: application to antioxidant characterization. Analytical Biochemistry, 257, 40-44.

YUYAMA, L. K. O., AGUIAR, J. P. L., SILVA FILHO, D. F., YUYAMA, K., VAREJÃO, M. D. J., FÁVARO, D. I. T., VASCONCELLOS, M. B. A., PIMENTEL, S. A. & CARUSO, M. S. F. 2011. Caracterização físico-química do suco de açaí de Euterpe precatoria Mart. oriundo de diferentes ecossistemas amazônicos. Acta Amazonica, 41, 545-552.

YUYAMA, L. K. O., ROSA, R. D., AGUIAR, J. P. L., NAGAHAMA, D., ALENCAR, F. H., YUYAMA, K., CORDEIRO, G. W. D. O. & MARQUES, H. D. O. 2002. Açai (euterpe oleracea mart.) and camu-camu (Myrciaria dubia (HBK) Mc Vaugh), do they possess anti-anemic action? Acta Amazonica, 32, 625-625.

ZAIDI, J. H., FATIMA, I., ARIF, M. & QURESHI, I. H. 2005. Determination of trace elements in coffee beans and instant coffee of various origins by INAA. Journal Of Radioanalytical And Nuclear Chemistry, 267, 109-112.

ZEINER, M., CINDRIC, I. J., POZGAJ, M., PIRKL, R., SILIC, T. & STINGEDER, G. 2015. Influence of soil composition on the major, minor and trace metal content of Velebit biomedical plants. Journal Of Pharmaceutical And Biomedical Analysis, 106, 153-158.

ZHISHEN, J., MENGCHENG, T. & JIANMING, W. 1999. The determination of flavonoid contents in mulberry and their scavenging effects on superoxide radicals. Food Chemistry, 64, 555-559.

ZIELINSKI, A. A. F., HAMINIUK, C. W. I., ALBERTI, A., NOGUEIRA, A., DEMIATE, I. M. & GRANATO, D. 2014. A comparative study of the phenolic compounds and the in vitro antioxidant activity of different Brazilian teas using multivariate statistical techniques. Food Research International, 60, 246-254.

Chemical Analysis of Typical Beverages and Açaí Berry

from South America

by Fernanda Vanoni Matta

APPENDICES

Faculty of Engineering and Physical Sciences

University of Surrey, Guildford, GU2 7XH

2019

i

List of Appendices

Page Appendix 3.1 Sample list of non-commercial samples from Barão de Cotegipe (Rio Grande do Sul State – Brazil). 1 Appendix 3.2 Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of non-commercial samples yerba mate (mg/kg, dry

weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates. 3

Appendix 3.3 Mn, Fe, Co, Ni, Cu and Zn levels (mean ± standard deviation) of non-commercial yerba mate samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

6

Appendix 3.4 As, Se, Mo, Cd ans Pb levels (mean ± standard deviation) of non-commercial yerba mate samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

9

Appendix 3.5 Sample list of commercial yerba mate samples from Brazil and Argentina. 12 Appendix 3.6 Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg dry weight)

determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates. 15

Appendix 3.7 Mn, Fe, Co, Ni, Cu ans Zn levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates

18

Appendix 3.8 As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates

21

Appendix 3.9 V, Cr, Mn, Fe, Co and Ni levels (mean ± standard deviation) of commercial yerba mate regular infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

24

Appendix 3.10 Cu, Zn, As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of commercial yerba mate regular infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates

27

Appendix 3.11 V, Cr, Mn, Fe, Co and Ni levels (mean ± standard deviation) of commercial yerba mate Brazilian iced tea infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of

30

Appendix 3.12 Cu, Zn, As, Se, Mo Cd and Pb levels (mean ± standard deviation) of commercial yerba mate Brazilian iced tea infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

31

Appendix 3.13 Mn and Fe levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) 32

ii

determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates. Appendix 3.14 Cu and Zn levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL)

determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates. 34

Appendix 3.15 Ni and Cr levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

35

Appendix 3.16 V and Co levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

37

Appendix 3.17 As and Se levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

38

Appendix 3.18 Mo and Cd levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

40

Appendix 3.19 Pb levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates

41

Appendix 3.20 Total polyphenol content of commercial green loose yerba mate regular and bombilla infusions (mg GAE/200 mL) determined using Folin-Ciocalteu assay (refer to section 2.4).

43

Appendix 3.21 Total polyphenol content of commercial green/roasted loose or tea bag yerba mate regular infusions (mg GAE/200mL) determined using Folin-Ciocalteu assay (refer to section 2.4

45

Appendix 3.22 Total polyphenol content of commercial green/roasted loose or tea bag yerba mate Brazilian iced tea infusions (mg GAE/200mL) determined using Folin-Ciocalteu assay (refer to section 2.4).

47

Appendix 3.23 Chlorogenic acid, theobromine and caffeine content of commercial green/roasted loose or tea bag yerba mate regular infusions (mg/200 mL) determined by UHPLC (refer to section 2.5).

48

Appendix 3.24 Chlorogenic acid, theobromine and caffeine content of commercial green/roasted loose or tea bag yerba mate bombilla infusions (mg/200 mL) determined by UHPLC (refer to section 2.5).

49

Appendix 3.25 Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between the elemental levels of yerba mate leaves (based on age – new and old) for non-commercial samples collected from traditional plantations cultivated either using NPK fertilisers or non-chemical (organic).

51

iii

Appendix 3.26 Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between the elemental levels of yerba mate leaves based on the use or non-use (organic) of NPK fertilsers during traditional cultivation.

52

Appendix 3.27 Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between the elemental levels of yerba mate leaves (new and old) grown in traditional organic or native forest plantations (refer to Table 3.6).

53

Appendix 3.28 Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between the origin (Brazil and Argentina); packaging (loose and tea bags) and roasting (green and roasted) of commercial yerba mate samples (refer to Table 3.9).

54

Appendix 3.29 Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between the origin (Brazil and Argentina); packaging (loose and tea bags) and roasting process (green loose and roasted) of regular infusions of commercial yerba mate (refer to Table 3.11).

55

Appendix 4.1 Sample list of Brazilian coffee samples from Amparo, São Paulo State. 56 Appendix 4.2 Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry weight) determined

using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates. 57

Appendix 4.3 Mn, Fe, Co, Ni, Cu and Zn levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

59

Appendix 4.4 As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

61

Appendix 5.1 Total elemental levels (mean ± standard deviation) of açaí pulp samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample (non-commercial : purple n= 6; and white n= 4; and commercial: purple n= 4; n is the number of samples).

63

Appendix 5.2 Total elemental levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample (non-commercial : purple n= 6; and white n= 4; and commercial: purple n= 4; n is the number of samples).

65

Appendix 5.3 Sample list for the evaluation of the Amazon geographical variability and industrial processing on açaí. 67

iv

Appendix 5.4 Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg, dry weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the number of replicates).

68

Appendix 5.5 Total elemental (mean ± standard deviation) of açaí pulp samples (mg/kg dry weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the number of replicates).

70

Appendix 5.6 Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the number of replicates).

71

Appendix 5.7 Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is replicates).

73

Appendix 5.8 Total polyphenol (TP) and minor elements daily intake (mg/day) based on the consumpsion of a 500 g serving of the commercial and non-commercial açaí pulp (fresh weight).

74

Appendix 5.9 Percentage intake (%) of total polyphenol and minor elements based on the consumption of a 500 g serving of the commercial and non-commercial açaí pulp (fresh weight) when compared to the recommended daily allowance (RDA) for males (M) and females (F).

75

Appendix 5.10 Total polyphenol and minor elements daily intake (mg/day) based on a 500 g serving of açaí pulp. 77

1

Appendix 3.1: Sample list of non-commercial samples from Barão de Cotegipe (Rio Grande do Sul State – Brazil).

Code Plantation type Pesticides use Age (years) Sample type Cultivated (fertiliser)

EMC - 01 new Conventional cultivation non organic (NPK) 30 Plant - new leaves EMC - 01 old Conventional cultivation non organic (NPK) 30 Plant - old leaves EMC - 02 new Conventional cultivation non organic (NPK) 30 Plant - new leaves EMC - 02 old Conventional cultivation non organic (NPK) 30 Plant - old leaves EMC - 03 new Conventional cultivation non organic (NPK) 30 Plant - new leaves EMC - 03 old Conventional cultivation non organic (NPK) 30 Plant - old leaves EMC - 04 new Conventional cultivation non organic (NPK) 30 Plant - new leaves EMC - 04 old Conventional cultivation non organic (NPK) 30 Plant - old leaves Natural forests EMN - 01 new Natural forests organic 20 Plant - new leaves EMN - 01 old Natural forests organic 20 Plant - old leaves EMN - 02 new Natural forests organic 20 Plant - new leaves EMN - 02 old Natural forests organic 20 Plant - old leaves EMN - 03 new Natural forests organic 20 Plant - new leaves EMN - 03 old Natural forests organic 20 Plant - old leaves Cultivated (organic) EMO - 01 new Cultivated under sun organic 25 Plant - new leaves EMO - 01 old Cultivated under sun organic 25 Plant - old leaves EMO - 02 new Cultivated under sun organic 25 Plant - new leaves EMO - 02 old Cultivated under sun organic 25 Plant - old leaves

2

Code Plantation type Pesticides use Age (years) Sample type EMO - 03 new Cultivated under sun organic 25 Plant - new leaves EMO - 03 old Cultivated under sun organic 25 Plant - old leaves EMO - 04 new Cultivated under sun organic 25 Plant - new leaves EMO - 04 old Cultivated under sun organic 25 Plant - old leaves EMO - 05 bottom Cultivated under sun organic 25 Plant - bottom height EMO - 05 middle Cultivated under sun organic 25 Plant - middle height EMO - 05 top Cultivated under sun organic 25 Plant - top height EMO - 06 Cultivated under sun organic 25 Plant – leaves EMO - 07 new Cultivated under sun organic 25 Plant - new leaves EMO - 07 old Cultivated under sun organic 25 Plant - old leaves EMO - 08 leaves Cultivated under sun organic 2 Plant – old leaves EMO - 09 leaves Cultivated under sun organic 2 Plant – old leaves EMO - 10 new Cultivated under sun organic 25 Plant - new leaves EMO - 10 old Cultivated under sun organic 25 Plant - old leaves EMO - 11 new Cultivated under sun organic 25 Plant - new leaves EMO - 11 old Cultivated under sun organic 25 Plant - old leaves Processed* NA new Native organic n.r. Plant - new leaves NA old Native organic n.r. Plant - old leaves YM - Sap n.r. n.r. n.r. Plant – Sapeco* YM- CA n.r. n.r. n.r. Plant – dried*

*refer to section 3.5.1; n.r.: not reported.

3

Appendix 3.2: Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of non-commercial samples yerba mate

(mg/kg, dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Code Na Mg K Ca V Cr

Cultivated (fertiliser)

EMC - 01 new 241 ± 11 2738 ± 64 8761 ± 150 2504 ± 46 0.15 ± 0.01 0.96 ± 0.02

EMC - 01 old 159 ± 22 5710 ± 75 5509 ± 168 6511 ± 158 0.20 ± 0.01 1.04 ± 0.01

EMC - 02 new 581 ± 15 3618 ± 1 8845 ± 98 3035 ± 86 0.07 ± 0.003 1.04 ± 0.05

EMC - 02 old 132 ± 5 8474 ± 35 4323 ± 13 9209 ± 9 0.11 ± 0.01 0.93 ± 0.17

EMC - 03 new 406 ± 18 2608 ± 14 8206 ± 37 2257 ± 83 0.09 ± 0.01 0.66 ± 0.10

EMC - 03 old 99 ± 19 7331 ± 171 3563 ± 79 10298 ± 488 0.24 ± 0.002 0.85 ± 0.09

EMC - 04 new 274 ± 26 3724 ± 129 8051 ± 154 4160 ± 305 0.08 ± 0.004 0.66 ± 0.002

EMC - 04 old 225 ± 19 6547 ± 51 5729 ± 64 8504 ± 12 0.19 ± 0.02 1.33 ± 0.06

Natural forest

EMN - 01 old 442 ± 26 3782 ± 48 9299 ± 66 3867 ± 3 0.22 ± 0.003 1.08 ± 0.16

EMN - 01 new 107 ± 6 7846 ± 155 3232 ± 60 11139 ± 547 0.21 ± 0.02 0.68 ± 0.11

EMN - 02 new 635 ± 30 2873 ± 85 9714 ± 373 3277 ± 60 0.32 ± 0.02 0.65 ± 0.10

EMN - 02 old 150 ± 26 6046 ± 49 5199 ± 50 8513 ± 96 0.49 ± 0.005 0.93 ± 0.06

EMN - 03 new 655 ± 53 2918 ± 21 10843 ± 224 3346 ± 7 0.33 ± 0.001 0.39 ± 0.10

EMN - 03 old 201 ± 54 4957 ± 15 4582 ± 45 8635 ± 91 0.22 ± 0.005 0.81 ± 0.01

4


Cultivated (organic)

EMO - 01 new 181 ± 25 3574 ± 10 6391 ± 29 5319 ± 344 0.07 ± 0.01 0.73 ± 0.06

EMO - 01 old 223 ± 1 2665 ± 46 7545 ± 193 2536 ± 172 0.04 ± 0.003 0.76 ± 0.16

EMO - 02 new 379 ± 23 2792 ± 60 8536 ± 182 2884 ± 77 0.05 ± 0.006 0.68 ± 0.08

EMO - 02 old 209 ± 4 4215 ± 38 5609 ± 85 6193 ± 22 0.06 ± 0.01 1.19 ± 0.25

EMO - 03 new 220 ± 8 2779 ± 7 7707 ± 46 3464 ± 6 0.06 ± 0.002 1.43 ± 0.03

EMO - 03 old 92 ± 41 2356 ± 1582 4031 ± 2673 3215 ± 2121 0.05 ± 0.03 1.32 ± 0.82

EMO - 04 new 358 ± 41 2807 ± 15 9156 ± 17 2588 ± 179 0.06 ± 0.006 1.00 ± 0.09

EMO - 04 old 115 ± 1 5604 ± 66 5780 ± 109 8193 ± 114 0.06 ± 0.003 1.32 ± 0.16

EMO - 05 bottom 259 ± 22 5830 ± 0.5 5854 ± 2 11383 ± 48 0.07 ± 0.004 1.15 ± 0.07

EMO - 05 middle 156 ± 14 3940 ± 30 5539 ± 99 5469 ± 181 0.05 ± 0.001 0.85 ± 0.08

EMO - 05 top 188 ± 22 4054 ± 24 6788 ± 15 5857 ± 12 0.12 ± 0.01 1.16 ± 0.24

EMO - 06 214 ± 16 4431 ± 40 6913 ± 58 5217 ± 5 0.10 ± 0.008 1.05 ± 0.07

EMO - 07 new 198 ± 53 5914 ± 40 6709 ± 31 5467 ± 117 0.07 ± 0.004 0.86 ± 0.06

EMO - 07 old 255 ± 22 6485 ± 80 5309 ± 55 7390 ± 197 0.07 ± 0.007 1.16 ± 0.22

EMO - 08 leaves 287 ± 129 5582 ± 80 6157 ± 9 10534 ± 3327 0.05 ± 0.003 0.77 ± 0.05

EMO - 09 leaves 287 ± 4 5270 ± 107 6045 ± 35 6437 ± 228 0.04 ± 0.002 0.34 ± 0.01

EMO - 10 new 652 ± 13 3566 ± 55 10246 ± 161 2839 ± 29 0.04 ± 0.005 0.31 ± 0.04

EMO - 10 old 190 ± 0.4 7466 ± 35 3508 ± 18 9723 ± 192 0.07 ± 0.01 0.68 ± 0.02

5


EMO - 11 new 224 ± 11 3719 ± 294 10536 ± 199 3515 ± 571 0.04 ± 0.002 0.25 ± 0.02

EMO - 11 old 298 ± 80 8218 ± 142 5595 ± 19 10904 ± 189 0.10 ± 0.02 0.79 ± 0.16

Processed*

NA new 164 ± 18 6193 ± 87 3233 ± 27 4853 ± 3 0.09 ± 0.005 0.31 ± 0.003

NA old 213 ± 7 7315 ± 166 3258 ± 103 6073 ± 118 0.07 ± 0.01 0.32 ± 0.02

YM - Sap 227 ± 38 7206 ± 21 4555 ± 173 7996 ± 424 0.34 ± 0.03 3.32 ± 0.10

YM- CA - NA 181 ± 19 4403 ± 40 4511 ± 58 6058 ± 244 0.12 ± 0.002 0.46 ± 0.15

YM- CA 156 ± 6 5962 ± 54 4644 ± 96 6771 ± 63 0.40 ± 0.03 0.83 ± 0.03 *refer to section 3.5.1

6

Appendix 3.3: Mn, Fe, Co, Ni, Cu and Zn levels (mean ± standard deviation) of non-commercial yerba mate samples

(mg/kg dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Code Mn Fe Co Ni Cu Zn

Cultivated (fertiliser) EMC - 01 new 610 ± 18 55.65 ± 3.11 0.39 ± 0.01 6.59 ± 0.05 37.19 ± 0.15 68.41 ± 0.13

EMC - 01 old 1394 ± 42 76.22 ± 3.91 0.07 ± 0.004 2.20 ± 0.12 14.68 ± 0.25 118.05 ± 3.89

EMC - 02 new 425 ± 11 45.08 ± 0.69 0.39 ± 0.01 4.83 ± 0.06 26.17 ± 0.43 76.92 ± 1.72

EMC - 02 old 560 ± 13 49.17 ± 0.57 0.10 ± 0.003 2.52 ± 0.002 14.18 ± 0.04 238.50 ± 11.42

EMC - 03 new 531 ± 9 36.87 ± 2.21 0.18 ± 0.002 3.67 ± 0.11 25.08 ± 0.39 67.71 ± 0.10

EMC - 03 old 715 ± 89 74.82 ± 0.09 0.03 ± 0.005 1.89 ± 0.15 16.35 ± 0.60 82.49 ± 1.99

EMC - 04 new 924 ± 3 50.75 ± 2.88 0.77 ± 0.02 6.45 ± 0.30 26.22 ± 0.68 104.95 ± 1.16

EMC - 04 old 1127 ± 13 64.32 ± 5.24 0.16 ± 0.003 3.13 ± 0.09 13.45 ± 0.04 225.51 ± 2.60

Natural forest

EMN - 01 old 1022 ± 16 69.63 ± 0.56 0.59 ± 0.05 7.93 ± 0.27 20.18 ± 0.19 71.86 ± 0.75

EMN - 01 new 982 ± 31 91.15 ± 2.80 0.37 ± 0.006 2.71 ± 0.06 9.35 ± 0.34 32.83 ± 0.60

EMN - 02 new 1085 ± 25 94.07 ± 4.48 0.71 ± 0.02 5.40 ± 0.02 22.09 ± 0.43 75.71 ± 0.13

EMN - 02 old 1365 ± 5 153.99 ± 0.92 0.26 ± 0.002 2.20 ± 0.33 11.62 ± 0.18 39.57 ± 0.30

EMN - 03 new 1159 ± 51 91.60 ± 2.09 0.27 ± 0.02 5.66 ± 0.05 22.87 ± 0.57 66.93 ± 0.85

EMN - 03 old 805 ± 1 89.10 ± 1.67 0.12 ± 0.005 2.97 ± 0.10 11.87 ± 0.41 49.39 ± 0.63


7


EMO - 01 new 871 ± 54 39.37 ± 0.14 0.25 ± 0.05 2.87 ± 0.003 8.87 ± 0.72 252.96 ± 14.68

EMO - 01 old 914 ± 17 43.22 ± 2.32 1.12 ± 0.03 4.62 ± 0.22 21.29 ± 0.59 80.18 ± 2.33

EMO - 02 new 924 ± 9 36.27 ± 0.10 0.62 ± 0.002 6.64 ± 0.51 17.93 ± 0.19 63.93 ± 0.71

EMO - 02 old 592 ± 3 37.13 ± 4.20 0.27 ± 0.01 2.95 ± 0.11 7.65 ± 0.46 48.82 ± 1.79

EMO - 03 new 775 ± 21 35.16 ± 3.85 0.24 ± 0.005 3.97 ± 0.004 12.00 ± 0.04 30.17 ± 0.34

EMO - 03 old 349 ± 192 28.22 ± 19.12 0.10 ± 0.07 3.52 ± 2.31 5.53 ± 3.66 15.74 ± 7.86

EMO - 04 new 757 ± 1 41.39 ± 1.73 1.27 ± 0.03 4.07 ± 0.12 26.90 ± 0.29 73.39 ± 0.22

EMO - 04 old 792 ± 29 37.87 ± 1.63 0.84 ± 0.002 5.44 ± 0.07 12.68 ± 0.43 119.23 ± 0.34

EMO - 05 bottom 1070 ± 41 39.10 ± 5.45 0.78 ± 0.01 1.37 ± 0.02 5.41 ± 0.40 81.79 ± 0.59

EMO - 05 middle 1009 ± 82 32.64 ± 5.05 0.46 ± 0.006 2.04 ± 0.01 7.68 ± 0.28 64.08 ± 1.09

EMO - 05 top 1229 ± 21 51.33 ± 1.92 0.43 ± 0.01 1.36 ± 0.05 5.21 ± 0.17 65.29 ± 1.28

EMO - 06 1189 ± 8 50.73 ± 1.22 0.22 ± 0.003 1.93 ± 0.17 9.54 ± 0.68 54.98 ± 2.36

EMO - 07 new 818 ± 0.8 31.58 ± 1.91 0.48 ± 0.01 3.92 ± 0.07 9.92 ± 0.84 148.31 ± 12.24

EMO - 07 old 609 ± 16 36.19 ± 4.13 0.37 ± 0.006 2.70 ± 0.26 7.08 ± 1.09 195.76 ± 5.52

EMO - 08 leaves 3395 ± 68 32.00 ± 2.88 0.34 ± 0.004 1.67 ± 0.23 11.62 ± 0.92 73.06 ± 0.15

EMO - 09 leaves 1569 ± 52 27.83 ± 0.12 1.45 ± 0.11 1.16 ± 0.09 11.45 ± 0.02 41.86 ± 1.38

EMO - 10 new 865 ± 25 33.15 ± 1.52 0.12 ± 0.003 5.29 ± 0.03 21.35 ± 0.24 48.19 ± 0.02

EMO - 10 old 613 ± 2 47.64 ± 1.26 0.24 ± 0.01 3.43 ± 0.05 8.90 ± 0.28 93.35 ± 2.67

EMO - 11 new 781 ± 16 31.76 ± 2.63 0.86 ± 0.05 4.68 ± 0.22 15.86 ± 0.21 79.48 ± 4.89

8


EMO - 11 old 641 ± 10 48.29 ± 8.02 0.72 ± 0.04 2.30 ± 0.05 6.55 ± 0.22 251.73 ± 5.53

Processed*

NA new 349 ± 2 51.26 ± 1.01 0.19 ± 0.01 1.95 ± 0.06 12.12 ± 0.07 33.73 ± 1.16

NA old 388 ± 4 57.14 ± 3.22 0.09 ± 0.01 2.46 ± 0.01 10.57 ± 0.05 25.86 ± 0.29

YM - Sap 632 ± 22 132.95 ± 10.85 0.27 ± 0.01 3.19 ± 0.08 10.20 ± 0.22 69.30 ± 2.72

YM- CA - NA 445 ± 1 68.52 ± 4.72 0.07 ± 0.002 2.13 ± 0.18 11.08 ± 0.33 34.78 ± 3.00

YM- CA 729 ± 53 121.61 ± 6.02 0.37 ± 0.005 4.10 ± 0.03 11.55 ± 0.52 75.75 ± 5.60 *refer to section 3.5.1

9

Appendix 3.4: As, Se, Mo, Cd ans Pb levels (mean ± standard deviation) of non-commercial yerba mate samples (mg/kg

dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Code As Se Mo Cd Pb

Cultivated (fertiliser)

EMC - 01 new 0.05 ± 0.003 0.02 ± 0.01 <LOD* 0.53 ± 0.02 0.34 ± 0.02

EMC - 01 old 0.03 ± 0.001 0.03 ± 0.02 <LOD* 0.44 ± 0.03 0.18 ± 0.02

EMC - 02 new 0.01 ± 0.0003 0.01 ± 0.001 0.01 ± 0.001 0.47 ± 0.002 0.12 ± 0.01

EMC - 02 old 0.02 ± 0.0004 0.13 ± 0.01 <LOD* 0.45 ± 0.01 0.09 ± 0.01

EMC - 03 new 0.02 ± 0.002 <LOD* <LOD* 0.35 ± 0.01 0.07 ± 0.002

EMC - 03 old 0.02 ± 0.001 0.05 ± 0.002 <LOD* 0.22 ± 0.03 0.03 ± 0.01

EMC - 04 new 0.02 ± 0.0003 0.01 ± 0.001 <LOD* 0.59 ± 0.02 0.08 ± 0.02

EMC - 04 old 0.02 ± 0.001 0.04 ± 0.01 <LOD* 0.35 ± 0.01 0.06 ± 0.001

Natural forest

EMN - 01 old 0.02 ± 0.001 0.03 ± 0.003 0.01 ± 0.002 0.51 ± 0.02 0.05 ± 0.002

EMN - 01 new 0.01 ± 0.001 0.08 ± 0.01 <LOD* 0.38 ± 0.02 0.03 ± 0.01

EMN - 02 new 0.02 ± 0.0001 0.02 ± 0.01 <LOD* 0.53 ± 0.04 0.02 ± 0.002

EMN - 02 old 0.02 ± 0.0001 0.12 ± 0.002 <LOD* 0.41 ± 0.00 0.01 ± 0.002

EMN - 03 new 0.02 ± 0.0001 0.01 ± 0.01 <LOD* 0.93 ± 0.04 0.02 ± 0.01

EMN - 03 old 0.01 ± 0.004 0.05 ± 0.01 <LOD* 0.78 ± 0.02 <LOD*

10

Code As Se Mo Cd Pb


EMO - 01 new 0.03 ± 0.01 0.04 ± 0.002 0.02 ± 0.002 0.25 ± 0.05 0.01 ± 0.001

EMO - 01 old 0.02 ± 0.003 0.01 ± 0.001 0.72 ± 0.70 0.33 ± 0.01 0.02 ± 0.002

EMO - 02 new 0.03 ± 0.0002 0.01 ± 0.01 0.03 ± 0.002 0.27 ± 0.01 0.01 ± 0.001

EMO - 02 old 0.01 ± 0.001 0.03 ± 0.002 0.01 ± 0.001 0.30 ± 0.01 0.01 ± 0.003

EMO - 03 new 0.02 ± 0.001 0.01 ± 0.001 0.01 ± 0.001 0.28 ± 0.01 0.01 ± 0.001

EMO - 03 old 0.01 ± 0.01 0.03 ± 0.03 0.01 ± 0.01 0.07 ± 0.05 <LOD*

EMO - 04 new 0.01 ± 0.007 0.01 ± 0.01 0.02 ± 0.001 0.53 ± 0.02 0.05 ± 0.01

EMO - 04 old 0.02 ± 0.004 0.09 ± 0.01 <LOD* 0.40 ± 0.02 0.02 ± 0.001

EMO - 05 bottom 0.02 ± 0.002 0.08 ± 0.02 0.01 ± 0.002 0.42 ± 0.01 0.01 ± 0.001

EMO - 05 middle 0.02 ± 0.001 0.09 ± 0.02 <LOD* 0.34 ± 0.02 0.02 ± 0.003

EMO - 05 top 0.02 ± 0.001 0.05 ± 0.001 0.02 ± 0.002 0.23 ± 0.02 0.05 ± 0.002

EMO - 06 0.02 ± 0.005 0.04 ± 0.002 0.02 ± 0.001 0.25 ± 0.01 0.08 ± 0.001

EMO - 07 new 0.02 ± 0.002 0.03 ± 0.002 0.03 ± 0.001 0.44 ± 0.04 0.02 ± 0.002

EMO - 07 old 0.01 ± 0.001 0.04 ± 0.01 0.01 ± 0.001 0.48 ± 0.06 0.07 ± 0.06

EMO - 08 leaves 0.01 ± 0.004 0.04 ± 0.001 0.01 ± 0.002 0.41 ± 0.01 0.05 ± 0.01

EMO - 09 leaves 0.01 ± 0.002 0.04 ± 0.002 <LOD* 0.45 ± 0.02 0.01 ± 0.002

EMO - 10 new 0.01 ± 0.001 0.03 ± 0.003 0.02 ± 0.002 1.17 ± 0.02 0.05 ± 0.02

EMO - 10 old 0.01 ± 0.003 0.14 ± 0.01 <LOD* 0.47 ± 0.01 0.04 ± 0.001

11

Code As Se Mo Cd Pb

EMO - 11 new 0.01 ± 0.002 0.02 ± 0.01 0.02 ± 0.001 0.99 ± 0.05 0.05 ± 0.01

EMO - 11 old 0.02 ± 0.001 0.20 ± 0.02 <LOD* 0.45 ± 0.002 0.04 ± 0.002

Processed**

NA new 0.05 ± 0.004 0.04 ± 0.02 0.05 ± 0.01 0.15 ± 0.01 0.08 ± 0.002

NA old 0.02 ± 0.005 0.05 ± 0.01 <LOD* 0.08 ± 0.01 0.08 ± 0.02

YM - Sap 0.02 ± 0.002 0.04 ± 0.002 0.01 ± 0.001 0.35 ± 0.03 0.06 ± 0.001

YM- CA - NA 0.02 ± 0.003 0.06 ± 0.01 0.01 ± 0.001 0.18 ± 0.04 0.04 ± 0.01

YM- CA 0.03 ± 0.001 0.03 ± 0.02 0.01 ± 0.002 0.48 ± 0.02 0.17 ± 0.11 *refer to section 3.5.1; **refer to Table 2.2

12

Appendix 3.5: Sample list of commercial yerba mate samples from Brazil and Argentina.

Brand Origin Code

Green loose Baldo Uruguay/Brazil* BA Barão de Cotegipe - Cambona Brazil BC- Ca Barão de Cotegipe - Moída Grossa Brazil BC-Mg Barão de Cotegipe - Native Brazil BC- Na Barão de Cotegipe - Premium Brazil BC-PR Barão de Cotegipe - Terere Brazil BC-Te Barão de Cotegipe- tipo Uruguay Uruguay/Brazil* BC-Ex Canarias Uruguay/Brazil* CAN Sara Uruguay/Brazil* SAR Amanda Argentina AM Barão de Cotegipe - Traditional Brazil BC - Try Flor Verde Brazil FV Foller Brazil FO Jerper Argentina JE Ka-a Argentina KA Kraus Organic Argentina KR Pipore Argentina PI Porto Vitoria Brazil PV Roapipo Organic Argentina RO

13

Brand Origin Code Rosamonte Argentina RM Santo Antonio Brazil SA Yerba Uruguaia Uruguay/Brazil* YU

Green tea bag Amanda – Mate cocido Argentina AT Asis Organic Argentina AO Cachamate Argentina CA Carrefour Argentina CR Cruz de Malta – mate cocido Argentina CM Don Lucas Argentina DL Jumala Argentina JU La Anonima Argentina LA La Hoja Argentina LH La Posadena Argentina LP La Tranquera Argentina LT Litoral (La Virginia) Argentina LI Mate Tucangua Argentina MT Playadito Argentina PL Suave Union Argentina SU Taragui Argentina TA Taragui Ninos Argentina TN Vea Argentina VE

14

Brand Origin Code Yer-vita Argentina YVI

Roasted loose Mate Leao Natural Brazil MN Barão de Cotegipe - Mate Tostado Brazil BC-Cm Mate Leao Organic Roasted Brazil MO

Roasted tea bag Dr. Oetker Brazil DO Lin Tea Brazil LIT Qualita Brazil QU Leao – Cha Mate Tostado Brazil ML

*refer to section 3.6.1

15

Appendix 3.6: Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg dry

weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.


Green loose BA 134 ± 16 5503 ± 23 3913 ± 120 7747 ± 65 0.35 ± 0.08 0.53 ± 0.05 BC- Ca 135 ± 11 4199 ± 317 4072 ± 17 6855 ± 730 0.18 ± 0.001 0.61 ± 0.01 BC-Mg 192 ± 18 5933 ± 166 4174 ± 272 8639 ± 280 0.14 ± 0.02 0.50 ± 0.03 BC- Na 176 ± 27 3987 ± 105 4539 ± 173 6306 ± 481 0.21 ± 0.06 0.54 ± 0.002 BC-Pr 132 ± 21 4181 ± 251 4503 ± 95 8280 ± 251 0.15 ± 0.002 0.37 ± 0.03 BC-Te 148 ± 4 4042 ± 98 4868 ± 82 6517 ± 98 0.09 ± 0.002 0.52 ± 0.07 BC-Ex 92 ± 2 3493 ± 266 4010 ± 259 6026 ± 15 0.34 ± 0.02 0.67 ± 0.04 CAN 141 ± 14 4958 ± 372 4288 ± 172 6874 ± 471 0.30 ± 0.03 0.62 ± 0.12 SAR 171 ± 2 5872 ± 13 4221 ± 229 7684 ± 14 0.23 ± 0.01 0.57 ± 0.02 AM 177 ± 34 5417 ± 161 4090 ± 99 7216 ± 217 0.40 ± 0.09 0.76 ± 0.002 BC - Tr 202 ± 42 5894 ± 121 4258 ± 104 8633 ± 96 0.36 ± 0.05 0.69 ± 0.10 FV 162 ± 5 2786 ± 41 3895 ± 10 6348 ± 38 0.35 ± 0.02 1.18 ± 0.04 FO 154 ± 12 3741 ± 168 3935 ± 19 7111 ± 375 0.27 ± 0.01 0.36 ± 0.08 JE 160 ± 16 5449 ± 299 3271 ± 52 8418 ± 651 0.42 ± 0.01 0.89 ± 0.08 KA 137 ± 20 5177 ± 534 3729 ± 249 7960 ± 760 0.17 ± 0.01 1.18 ± 0.02 KR 201 ± 26 4649 ± 65 3720 ± 228 7894 ± 551 0.26 ± 0.03 1.42 ± 0.03 PI 239 ± 39 5053 ± 306 3605 ± 27 8091 ± 193 0.20 ± 0.01 0.93 ± 0.001 PV 128 ± 9 3725 ± 256 3520 ± 67 5971 ± 560 0.21 ± 0.01 0.30 ± 0.03

16

Code Na Mg K Ca V Cr RO 170 ± 28 5494 ± 424 2888 ± 105 6926 ± 1551 0.23 ± 0.01 0.81 ± 0.02 RM 164 ± 9 3466 ± 110 3200 ± 57 6501 ± 58 0.22 ± 0.03 0.98 ± 0.02 SA 243 ± 27 5243 ± 118 3763 ± 182 8090 ± 59 0.35 ± 0.01 0.61 ± 0.05 YU 158 ± 12 5095 ± 268 3686 ± 133 7080 ± 110 0.22 ± 0.02 0.52 ± 0.04

Green tea bag AT 211 ± 30 5502 ± 135 4542 ± 174 7518 ± 421 0.86 ± 0.06 0.98 ± 0.01 AO 209 ± 9 5585 ± 52 4169 ± 165 6739 ± 100 0.25 ± 0.01 1.44 ± 0.16 CA 151 ± 3 4614 ± 36 4552 ± 40 7279 ± 194 0.26 ± 0.002 0.95 ± 0.002 CR 228 ± 27 5582 ± 66 4052 ± 36 7978 ± 171 1.21 ± 0.05 0.98 ± 0.001 CM 162 ± 25 5160 ± 154 4052 ± 3 8734 ± 50 0.70 ± 0.02 1.01 ± 0.04 DL 116 ± 12 5227 ± 26 5005 ± 50 7030 ± 35 1.01 ± 0.20 1.05 ± 0.04 JU 158 ± 0.4 6100 ± 231 3380 ± 364 8743 ± 520 0.41 ± 0.04 1.16 ± 0.02 LA 146 ± 3 5705 ± 114 4453 ± 229 8650 ± 295 1.61 ± 0.02 1.36 ± 0.01 LH 195 ± 34 5277 ± 445 5211 ± 559 7537 ± 733 0.37 ± 0.01 0.90 ± 0.02 LP 211 ± 39 5259 ± 99 4438 ± 49 7720 ± 102 1.45 ± 0.23 1.25 ± 0.01 LT 190 ± 21 6131 ± 309 4913 ± 181 8971 ± 504 0.42 ± 0.07 1.23 ± 0.03 LI 152 ± 6 5631 ± 117 4968 ± 27 8422 ± 13 1.31 ± 0.23 1.13 ± 0.09 MT 175 ± 5 5570 ± 11 4286 ± 39 7218 ± 34 0.26 ± 0.02 1.26 ± 0.09 PL 216 ± 6 6196 ± 78 3681 ± 72 7007 ± 377 0.28 ± 0.04 1.16 ± 0.04 SU 112 ± 6 5778 ± 133 4513 ± 54 6801 ± 99 0.44 ± 0.01 1.14 ± 0.02 TA 124 ± 12 5781 ± 92 4285 ± 48 6924 ± 141 0.37 ± 0.01 0.96 ± 0.08 TN 156 ± 13 5259 ± 16 4207 ± 16 6301 ± 85 0.31 ± 0.03 1.12 ± 0.16

17

Code Na Mg K Ca V Cr VE 137 ± 6 5414 ± 103 4084 ± 226 9072 ± 531 1.20 ± 0.37 1.15 ± 0.04 YVI 142 ± 2 6190 ± 66 3624 ± 24 7154 ± 16 0.94 ± 0.08 1.15 ± 0.07

Roasted loose MN 129 ± 12 5188 ± 311 4837 ± 49 9254 ± 127 0.17 ± 0.02 0.67 ± 0.002 BC-Cm 127 ± 2 6933 ± 171 4065 ± 160 9970 ± 334 0.13 ± 0.003 0.61 ± 0.09 MO 151 ± 20 6297 ± 80 4804 ± 42 10121 ± 490 0.30 ± 0.05 1.00 ± 0.20

Roasted tea bag DO 134 ± 26 4557 ± 19 3412 ± 60 8461 ± 44 0.37 ± 0.01 1.05 ± 0.08 LIT 194 ± 36 5176 ± 44 3637 ± 32 9011 ± 126 0.47 ± 0.01 0.81 ± 0.05 QU 165 ± 13 5028 ± 49 3394 ± 94 8179 ± 335 0.35 ± 0.04 0.91 ± 0.11 ML 215 ± 14 7488 ± 18 4263 ± 97 8290 ± 95 0.50 ± 0.09 0.93 ± 0.02

18

Appendix 3.7: Mn, Fe, Co, Ni, Cu ans Zn levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg

dry weight) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.


Green loose BA 565 ± 44 42.34 ± 1.06 0.12 ± 0.002 3.22 ± 0.13 9.63 ± 0.22 50.76 ± 2.77

BC- Ca 781 ± 54 35.94 ± 0.88 0.13 ± 0.003 1.94 ± 0.01 8.37 ± 0.38 49.62 ± 1.96 BC-Mg 652 ± 16 14.85 ± 9.47 0.14 ± 0.002 1.84 ± 0.26 9.89 ± 0.37 67.16 ± 6.76 BC- Na 538 ± 33 23.17 ± 4.23 0.08 ± 0.01 2.28 ± 0.50 11.20 ± 2.20 43.31 ± 2.24 BC-Pr 663 ± 58 27.76 ± 1.69 0.07 ± 0.01 2.74 ± 0.09 11.76 ± 0.49 43.67 ± 4.33 BC-Te 485 ± 25 30.92 ± 1.16 0.16 ± 0.02 1.00 ± 0.10 7.67 ± 0.47 49.13 ± 6.67 BC-Ex 632 ± 2 64.76 ± 4.74 0.13 ± 0.002 2.81 ± 0.04 9.21 ± 0.01 66.52 ± 7.68 CAN 597 ± 40 45.65 ± 2.11 0.10 ± 0.001 3.58 ± 0.08 11.20 ± 1.39 44.57 ± 0.47 SAR 543 ± 1 44.49 ± 2.84 0.10 ± 0.001 3.33 ± 0.04 9.50 ± 0.31 51.58 ± 1.72 AM 800 ± 96 12.16 ± 4.80 0.25 ± 0.002 3.06 ± 0.23 7.33 ± 0.29 98.96 ± 1.61

BC - Tr 834 ± 0.3 36.82 ± 10.27 0.16 ± 0.02 2.44 ± 0.29 9.39 ± 0.39 62.43 ± 0.41 FV 755 ± 22 83.49 ± 5.32 0.19 ± 0.002 2.22 ± 0.10 9.33 ± 0.23 86.74 ± 2.53 FO 644 ± 33 34.30 ± 4.34 0.10 ± 0.01 3.29 ± 0.07 8.31 ± 0.01 29.42 ± 0.82 JE 661 ± 4 52.33 ± 2.52 0.25 ± 0.01 2.96 ± 0.13 8.41 ± 0.03 68.25 ± 1.55 KA 532 ± 4 10.62 ± 4.05 0.23 ± 0.01 3.2 ± 0.10 7.28 ± 0.13 71.19 ± 4.26 KR 612 ± 58 20.90 ± 4.58 0.49 ± 0.02 4.02 ± 0.13 7.33 ± 0.31 67.44 ± 1.73 PI 671 ± 54 10.58 ± 0.41 0.32 ± 0.01 3.66 ± 0.22 7.15 ± 0.03 80.88 ± 3.53 PV 631 ± 40 34.04 ± 4.85 0.11 ± 0.01 3.84 ± 0.11 10.81 ± 0.16 36.83 ± 2.28

19

Code Mn Fe Co Ni Cu Zn RO 419 ± 9 44.01 ± 25.86 0.23 ± 0.01 3.40 ± 0.004 9.02 ± 1.62 104.92 ± 7.79 RM 606 ± 88 17.63 ± 5.80 0.52 ± 0.08 4.38 ± 0.51 7.92 ± 0.88 60.01 ± 2.38 SA 383 ± 40 61.18 ± 4.22 0.21 ± 0.01 1.62 ± 0.01 7.65 ± 0.03 73.04 ± 5.02 YU 489 ± 31 30.05 ± 9.95 0.09 ± 0.001 3.52 ± 0.17 8.87 ± 0.53 46.48 ± 4.62

Green tea bag AT 675 ± 22 73.39 ± 3.46 0.35 ± 0.03 5.41 ± 0.002 10.70 ± 1.44 97.31 ± 4.81 AO 603 ± 18 28.95 ± 1.50 0.32 ± 0.02 4.54 ± 0.08 8.45 ± 0.14 123.79 ± 2.28 CA 662 ± 2 31.51 ± 7.43 0.18 ± 0.002 3.93 ± 0.10 8.79 ± 0.05 85.90 ± 2.14 CR 634 ± 20 154.51 ± 21.96 0.42 ± 0.001 4.93 ± 0.02 10.82 ± 0.89 121.04 ± 16.66 CM 530 ± 29 95.20 ± 5.67 0.34 ± 0.01 4.65 ± 0.20 15.00 ± 5.15 115.77 ± 2.05 DL 739 ± 10 108.95 ± 15.2 0.34 ± 0.001 4.84 ± 0.08 9.33 ± 0.07 80.98 ± 1.16 JU 741 ± 72 43.79 ± 6.62 0.51 ± 0.01 4.80 ± 0.01 7.97 ± 0.12 54.24 ± 0.55 LA 807 ± 204 164.31 ± 4.25 0.43 ± 0.02 6.06 ± 1.28 9.39 ± 0.10 102.88 ± 2.20 LH 834 ± 15 50.60 ± 0.75 0.23 ± 0.002 4.28 ± 0.06 8.59 ± 0.03 80.86 ± 2.13 LP 539 ± 75 228.13 ± 9.63 0.41 ± 0.01 4.54 ± 0.29 11.60 ± 0.55 117.48 ± 8.32 LT 646 ± 0.5 57.26 ± 2.07 0.28 ± 0.002 4.20 ± 0.09 8.08 ± 0.01 105.38 ± 2.53 LI 642 ± 4 151.70 ± 6.29 0.35 ± 0.002 5.09 ± 0.19 10.15 ± 0.21 113.42 ± 6.01

MT 514 ± 12 22.76 ± 0.03 0.26 ± 0.001 4.65 ± 0.17 8.32 ± 0.17 100.60 ± 2.04 PL 989 ± 11 42.6 ± 7.00 0.46 ± 0.01 5.80 ± 0.22 9.26 ± 0.05 63.30 ± 1.75 SU 1065 ± 6 50.04 ± 1.19 0.44 ± 0.01 5.74 ± 0.05 8.66 ± 0.02 59.27 ± 1.84 TA 1029 ± 19 18.21 ± 2.98 0.45 ± 0.01 6.03 ± 0.16 8.89 ± 0.02 65.03 ± 0.73 TN 1098 ± 35 33.06 ± 1.01 0.55 ± 0.01 5.32 ± 0.03 8.26 ± 0.10 601.58 ± 88.38

20

Code Mn Fe Co Ni Cu Zn VE 482 ± 8 72.82 ± 2.33 0.37 ± 0.002 4.11 ± 0.06 10.14 ± 1.08 80.28 ± 0.41 YVI 651 ± 10 96.65 ± 0.11 0.44 ± 0.01 6.60 ± 0.13 9.26 ± 0.12 54.78 ± 0.50

Roasted loose MN 692 ± 25 20.70 ± 3.32 0.24 ± 0.002 2.41 ± 0.21 9.07 ± 0.08 102.77 ± 23.81

BC-Cm 657 ± 27 21.23 ± 1.40 0.16 ± 0.001 1.54 ± 0.20 9.30 ± 0.08 65.14 ± 2.75 MO 888 ± 22 53.03 ± 11.07 0.35 ± 0.02 2.96 ± 0.07 9.61 ± 0.04 164.38 ± 4.94

Roasted tea bag DO 413 ± 1 83.55 ± 2.34 0.18 ± 0.01 3.37 ± 0.01 11.90 ± 0.18 98.31 ± 18.69 LIT 454 ± 14 109.63 ± 0.31 0.18 ± 0.002 3.82 ± 0.03 12.32 ± 0.73 81.12 ± 4.83 QU 397 ± 8 79.31 ± 11.20 0.18 ± 0.03 3.06 ± 0.05 11.44 ± 0.33 78.04 ± 8.82 ML 879 ± 8 47.77 ± 5.61 0.36 ± 0.07 4.41 ± 0.04 11.06 ± 0.52 81.65 ± 1.15

21

Appendix 3.8: As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of commercial yerba mate samples (mg/kg dry


Code As Se Mo Cd Pb

Green loose BA 0.08 ± 0.01 0.02 ± 0.002 <LOD* 0.25 ± 0.001 0.12 ± 0.02

BC- Ca 0.03 ± 0.001 <LOD* <LOD* 0.21 ± 0.002 0.15 ± 0.01 BC-Mg 0.03 ± 0.002 0.02 ± 0.001 <LOD* 0.35 ± 0.04 0.05 ± 0.001 BC- Na 0.02 ± 0.001 <LOD* <LOD* 0.23 ± 0.05 0.03 ± 0.001 BC-Pr 0.02 ± 0.001 <LOD* <LOD* 0.17 ± 0.002 0.05 ± 0.01 BC-Te 0.01 ± 0.001 0.02 ± 0.001 0.09 ± 0.01 0.25 ± 0.02 0.13 ± 0.03 BC-Ex 0.03 ± 0.01 0.02 ± 0.001 <LOD* 0.29 ± 0.01 0.10 ± 0.01 CAN 0.03 ± 0.002 0.02 ± 0.001 <LOD* 0.24 ± 0.002 0.20 ± 0.03 SAR 0.03 ± 0.001 <LOD* <LOD* 0.22 ± 0.01 0.14 ± 0.002 AM 0.04 ± 0.01 0.03 ± 0.002 <LOD* 0.35 ± 0.01 0.06 ± 0.02

BC - Tr 0.50 ± 0.46 0.02 ± 0.001 <LOD* 0.26 ± 0.06 0.07 ± 0.01 FV 0.03 ± 0.002 0.02 ± 0.001 <LOD* 0.43 ± 0.002 0.12 ± 0.01 FO 0.08 ± 0.05 <LOD* <LOD* 0.13 ± 0.01 0.04 ± 0.01 JE 0.06 ± 0.02 0.03 ± 0.002 <LOD* 0.21 ± 0.001 0.05 ± 0.002 KA 0.11 ± 0.08 0.04 ± 0.001 <LOD* 0.20 ± 0.03 0.07 ± 0.01 KR 0.32 ± 0.28 0.03 ± 0.001 <LOD* 0.12 ± 0.01 0.07 ± 0.01 PI 0.04 ± 0.001 0.04 ± 0.002 <LOD* 0.18 ± 0.01 0.11 ± 0.01 PV 0.41 ± 0.38 <LOD* <LOD* 0.17 ± 0.002 0.08 ± 0.002

22

Code As Se Mo Cd Pb RO 0.05 ± 0.01 0.06 ± 0.02 <LOD* 0.22 ± 0.03 0.10 ± 0.01 RM 0.04 ± 0.01 0.03 ± 0.002 <LOD* 0.18 ± 0.01 0.08 ± 0.01 SA 0.35 ± 0.32 0.02 ± 0.001 <LOD* 0.34 ± 0.02 0.14 ± 0.01 YU 0.11 ± 0.06 0.02 ± 0.003 <LOD* 0.20 ± 0.02 0.11 ± 0.02

Green tea bag AT 0.09 ± 0.01 0.04 ± 0.002 <LOD* 0.31 ± 0.01 0.12 ± 0.04 AO 0.05 ± 0.01 0.04 ± 0.003 <LOD* 0.23 ± 0.002 0.03 ± 0.003 CA 0.06 ± 0.01 0.04 ± 0.003 <LOD* 0.25 ± 0.01 0.08 ± 0.01 CR 0.11 ± 0.03 0.05 ± 0.002 <LOD* 0.29 ± 0.02 0.13 ± 0.02 CM 0.14 ± 0.07 0.04 ± 0.001 <LOD* 0.34 ± 0.02 <LOD* DL 0.10 ± 0.01 0.05 ± 0.003 <LOD* 0.24 ± 0.001 0.20. ± 0.07 JU 0.06 ± 0.01 0.03 ± 0.002 <LOD* 0.18 ± 0.002 0.06 ± 0.003 LA 0.14 ± 0.03 0.04 ± 0.01 <LOD* 0.31 ± 0.01 0.12 ± 0.02 LH 0.09 ± 0.03 0.03 ± 0.002 <LOD* 0.23 ± 0.01 0.06 ± 0.003 LP 0.08 ± 0.01 0.05 ± 0.001 <LOD* 0.43 ± 0.02 0.07 ± 0.01 LT 0.05 ± 0.002 0.03 ± 0.001 <LOD* 0.31 ± 0.01 0.06 ± 0.002 LI 0.12 ± 0.002 0.04 ± 0.001 <LOD* 0.35 ± 0.01 0.10 ± 0.01

MT 0.05 ± 0.01 0.05 ± 0.003 <LOD* 0.25 ± 0.01 0.04 ± 0.002 PL 0.04 ± 0.001 0.03 ± 0.002 <LOD* 0.19 ± 0.002 0.04 ± 0.002 SU 0.06 ± 0.001 0.03 ± 0.003 <LOD* 0.16 ± 0.003 0.07 ± 0.003 TA 0.04 ± 0.002 0.03 ± 0.002 <LOD* 0.18 ± 0.01 0.04 ± 0.002 TN 0.04 ± 0.002 0.03 ± 0.002 <LOD* 0.13 ± 0.002 0.05 ± 0.002

23

Code As Se Mo Cd Pb VE 0.10 ± 0.002 0.03 ± 0.003 <LOD* 0.31 ± 0.001 0.09 ± 0.02 YVI 0.07 ± 0.001 0.04 ± 0.002 <LOD* 0.15 ± 0.003 0.10 ± 0.04

Roasted loose MN 0.03 ± 0.01 0.02 ± 0.003 <LOD* 0.41 ± 0.06 0.36 ± 0.21

BC-Cm 0.02 ± 0.001 0.02 ± 0.002 0.11 ± 0.06 0.25 ± 0.02 0.13 ± 0.05 MO 0.02 ± 0.001 0.02 ± 0.002 <LOD* 0.83 ± 0.10 0.08 ± 0.01

Roasted tea bag DO 0.05 ± 0.002 0.03 ± 0.001 <LOD* 0.39 ± 0.003 0.18 ± 0.02 LIT 0.05 ± 0.01 0.03 ± 0.002 <LOD* 0.39 ± 0.002 0.20 ± 0.02 QU 0.05 ± 0.01 0.02 ± 0.001 <LOD* 0.34 ± 0.01 0.28 ± 0.01 ML 0.03 ± 0.002 0.03 ± 0.001 <LOD* 0.30 ± 0.02 0.10 ± 0.01

*refer to Table 2.2; <LOD less than the limit of detection.

24

Appendix 3.9: V, Cr, Mn, Fe, Co and Ni levels (mean ± standard deviation) of commercial yerba mate regular infusion

samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of

replicates.

Code V Cr Mn Fe Co Ni

Green loose BA 0.07 ± 0.002 0.25 ± 0.02 1322 ± 97 4.58 ± 0.44 0.18 ± 0.01 5.43 ± 0.43

BC- Ca 0.07 ± 0.002 0.37 ± 0.01 1408 ± 56 4.73 ± 0.17 0.24 ± 0.002 3.54 ± 0.16 BC-Mg 0.07 ± 0.001 0.18 ± 0.01 1063 ± 44 3.99 ± 0.23 0.22 ± 0.02 3.50 ± 0.13 BC- Na 0.07 ± 0.001 0.10 ± 0.002 1718 ± 16 3.58 ± 0.06 0.12 ± 0.001 4.01 ± 0.03 BC-Pr 0.07 ± 0.002 <LOD* 1178 ± 71 4.15 ± 0.36 0.11 ± 0.001 4.92 ± 0.26 BC-Te 0.07 ± 0.001 0.09 ± 0.002 473 ± 24 2.53 ± 0.27 0.18 ± 0.01 1.40 ± 0.09 BC-Ex 0.03 ± 0.003 0.49 ± 0.02 1308 ± 101 9.66 ± 0.24 0.29 ± 0.02 5.82 ± 0.26 CAN 0.02 ± 0.003 0.28 ± 0.01 1499 ± 41 6.36 ± 0.07 0.26 ± 0.04 4.60 ± 0.06 SAR 0.07 ± 0.001 0.19 ± 0.001 1239 ± 18 5.31 ± 0.11 0.18 ± 0.003 6.47 ± 0.01 AM 0.08 ± 0.001 0.25 ± 0.01 1647 ± 104 5.98 ± 0.13 0.19 ± 0.01 6.17 ± 0.18

BC - Tr 0.03 ± 0.002 0.43 ± 0.02 1258 ± 17 4.28 ± 0.14 0.42 ± 0.01 5.45 ± 0.22 FV 0.02 ± 0.003 1.39 ± 0.10 951 ± 85 9.41 ± 0.52 0.34 ± 0.02 3.88 ± 0.31 FO 0.02 ± 0.001 0.13 ± 0.002 3612 ± 10 7.68 ± 0.07 0.25 ± 0.002 8.32 ± 0.14 JE 0.03 ± 0.01 0.42 ± 0.07 1218 ± 189 5.77 ± 1.28 0.41 ± 0.08 5.32 ± 1.03 KA 0.08 ± 0.003 0.63 ± 0.01 1239 ± 18 2.72 ± 0.22 0.29 ± 0.01 3.76 ± 0.14 KR 0.05 ± 0.02 1.09 ± 0.05 681 ± 43 4.63 ± 0.02 0.70 ± 0.04 5.88 ± 0.27 PI 0.07 ± 0.002 0.43 ± 0.04 585 ± 57 1.91 ± 0.28 0.28 ± 0.03 3.85 ± 0.38

25

Code V Cr Mn Fe Co Ni PV 0.01 ± 0.003 0.16 ± 0.003 4255 ± 83 5.98 ± 0.14 0.24 ± 0.01 7.72 ± 0.25 RO 0.03 ± 0.001 0.41 ± 0.06 518 ± 14 5.40 ± 0.17 0.29 ± 0.01 5.16 ± 0.10 RM 0.03 ± 0.003 0.76 ± 0.06 1239 ± 112 4.15 ± 0.42 0.77 ± 0.04 6.81 ± 0.55 SA 0.02 ± 0.004 0.27 ± 0.01 614 ± 12 8.21 ± 0.54 0.30 ± 0.002 2.48 ± 0.13 YU 0.08 ± 0.002 0.23 ± 0.002 1421 ± 10 4.95 ± 0.19 0.16 ± 0.001 6.64 ± 0.27

Green tea bag AT 0.09 ± 0.01 1.32 ± 0.14 2734 ± 300 19.70 ±1.67 1.03 ± 0.12 18.05 ± 1.99 AO 0.02 ± 0.002 1.32 ± 0.09 1504 ± 223 9.55 ± 0.28 0.65 ± 0.05 8.97 ± 0.66 CA 0.03 ± 0.002 1.09 ± 0.21 1508 ± 167 8.74 ± 1.60 0.31 ± 0.01 8.04 ± 1.54 CR 0.16 ± 0.01 1.15 ± 0.08 2298 ± 209 18.61 ± 1.38 0.36 ± 0.07 14.18 ± 1.69 CM 0.13 ± 0.003 0.89 ± 0.003 1682 ± 14 11.96 ± 0.09 0.77 ± 0.01 9.50 ± 0.04 DL 0.14 ± 0.002 0.97 ± 0.01 1626 ± 30 20.30 ± 0.53 0.97 ± 0.01 16.33 ± 0.13 JU 0.05 ± 0.002 1.45 ± 0.01 2834 ± 97 10.43 ± 0.21 1.32 ± 0.02 11.74 ± 0.17 LA 0.14 ± 0.001 1.37 ± 0.01 2285 ± 68 20.44 ± 0.13 1.11 ± 0.01 14.44 ± 0.05 LH 0.06 ± 0.002 1.16 ± 0.03 2354 ± 34 20.31 ± 0.28 0.82 ± 0.01 16.29 ± 0.003 LP 0.13 ± 0.002 0.86 ± 0.002 1886 ± 4 17.450 ± 0.49 0.72 ± 0.001 6.91 ± 0.04 LT 0.05 ± 0.01 1.78 ± 0.20 2289 ± 370 12.97 ± 1.52 0.81 ± 0.09 12.44 ± 1.93 LI 0.14 ± 0.01 1.26 ± 0.01 2235 ± 31 20.48 ± 1.30 1.03 ± 0.002 16.04 ± 0.13

MT 0.05 ± 0.002 1.98 ± 0.002 2837 ± 0.1 16.25 ± 0.16 0.83 ± 0.01 14.33 ± 0.05 PL 0.09 ± 0.003 1.69 ± 0.04 3937 ± 15 10.54 ± 0.56 1.29 ± 0.03 14.91 ± 0.01 SU 0.05 ± 0.003 1.71 ± 0.08 3025 ± 147 17.28 ± 0.08 1.42 ± 0.07 19.63 ± 0.79 TA 0.05 ± 0.003 2.01 ± 0.12 3687 ± 53 17.90 ± 0.13 1.93 ± 0.11 20.88 ± 1.10

26

Code V Cr Mn Fe Co Ni TN 0.03 ± 0.003 1.23 ± 0.01 2733 ± 53 10.83 ± 0.01 1.35 ± 0.02 15.54 ± 0.25 VE 0.06 ± 0.003 0.77 ± 0.001 1282 ± 9 10.18 ± 0.19 0.69 ± 0.002 7.89 ± 0.02 YVI 0.12 ± 0.003 1.77 ± 0.03 3587 ± 3 19.22 ± 0.12 1.53 ± 0.03 22.48 ± 0.44

Roasted loose MN 0.02 ± 0.002 0.23 ± 0.002 425 ± 71 3.37 ± 0.19 0.11 ± 0.001 2.58 ± 0.16

BC-Cm 0.07 ± 0.003 0.23 ± 0.01 372 ± 7 2.22 ± 0.13 0.09 ± 0.001 1.14 ± 0.02 MO 0.03 ± 0.003 0.32 ± 0.02 495 ± 35 4.38 ± 0.15 0.15 ± 0.01 2.67 ± 0.34

Roasted tea bag DO 0.03 ± 0.003 0.29 ± 0.01 513 ± 5 7.49 ± 0.30 0.10 ± 0.002 2.87 ± 0.11 LIT 0.05 ± 0.003 0.30 ± 0.01 663 ± 18 10.65 ± 0.20 0.12 ± 0.01 3.66 ± 0.16 QU 0.04 ± 0.002 0.25 ± 0.002 490 ± 20 7.35 ± 0.42 0.09 ± 0.001 2.55 ± 0.17 ML 0.04 ± 0.001 0.56 ± 0.03 861 ± 25 5.85 ± 0.21 0.26 ± 0.01 4.51 ± 0.21


27

Appendix 3.10: Cu, Zn, As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of commercial yerba mate regular

infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of

replicates.

Code Cu Zn As Se Mo Cd Pb

Green loose

BA 10.84 ± 0.59 50.55 ± 4.85 <LOD* <LOD* <LOD* 0.09 ± 0.01 0.05 ± 0.001 BC- Ca 6.68 ± 0.14 54.92 ± 2.53 <LOD* <LOD* 0.14 ± 0.002 0.06 ± 0.002 0.05 ± 0.001 BC-Mg 8.69 ± 0.13 58.81 ± 4.13 <LOD* 0.03 ± 0.002 0.05 ± 0.001 0.09 ± 0.003 0.03 ± 0.002 BC- Na 8.68 ± 0.06 45.51 ± 0.99 <LOD* <LOD* <LOD* 0.06 ± 0.003 <LOD* BC-Pr 10.81 ± 0.34 45.38 ± 2.67 <LOD* <LOD* <LOD* 0.04 ± 0.01 <LOD* BC-Te 4.69 ± 0.07 33.67 ± 1.70 <LOD* <LOD* <LOD* 0.10 ± 0.04 <LOD* BC-Ex 13.12 ± 0.81 76.24 ± 5.23 0.03 ± 0.002 0.08 ± 0.002 <LOD* 0.13 ± 0.01 <LOD* CAN 9.16 ± 0.11 61.90 ± 0.13 0.03 ± 0.002 0.06 ± 0.001 0.13 ± 0.01 0.08 ± 0.003 <LOD* SAR 12.90 ± 0.30 52.13 ± 1.98 <LOD* <LOD* <LOD* 0.08 ± 0.003 0.07 ± 0.002 AM 11.71 ± 0.15 56.32 ± 3.68 <LOD* <LOD* <LOD* 0.08 ± 0.01 0.08 ± 0.002

BC - Tr 6.01 ± 0.11 78.24 ± 4.32 0.03 ± 0.001 0.06 ± 0.01 <LOD* 0.12 ± 0.003 <LOD* FV 8.21 ± 0.23 82.01 ± 7.43 0.03 ± 0.002 0.07 ± 0.002 <LOD* 0.15 ± 0.01 <LOD* FO 11.06 ± 0.23 46.83 ± 1.65 0.02 ± 0.001 0.03 ± 0.001 0.35 ± 0.14 0.06 ± 0.001 <LOD* JE 8.41 ± 1.49 69.58 ± 13.30 0.03 ± 0.01 0.08 ± 0.01 <LOD* 0.07 ± 0.01 <LOD* KA 5.65 ± 0.02 51.24 ± 2.24 0.02 ± 0.002 0.03 ± 0.001 <LOD* 0.06 ± 0.001 0.04 ± 0.01 KR 6.47 ± 0.30 61.34 ± 1.43 0.03 ± 0.001 0.07 ± 0.01 <LOD* 0.04 ± 0.002 <LOD* PI 4.15 ± 0.29 44.85 ± 3.66 <LOD* 0.03 ± 0.002 <LOD* 0.04 ± 0.002 <LOD*

28

Code Cu Zn As Se Mo Cd Pb PV 11.23 ± 0.05 52.96 ± 3.30 0.03 ± 0.001 <LOD* 0.62 ± 0.13 0.07 ± 0.001 <LOD* RO 5.32 ± 0.18 93.13 ± 1.39 0.05 ± 0.002 0.08 ± 0.003 <LOD* 0.07 ± 0.002 <LOD* RM 5.61 ± 0.33 53.72 ± 7.00 0.05 ± 0.002 0.08 ± 0.003 <LOD* 0.01 ± 0.002 <LOD* SA 7.50 ± 0.60 59.70 ± 3.92 0.02 ± 0.001 0.04 ± 0.002 <LOD* 0.11 ± 0.01 <LOD* YU 11.53 ± 0.30 47.06 ± 3.33 <LOD* <LOD* <LOD* 0.13 ± 0.07 0.07 ± 0.01

Green tea bag AT 12.47 ± 0.94 184.91 ± 17.29 0.13 ± 0.01 0.15 ± 0.02 <LOD* 0.21 ± 0.01 <LOD* AO 8.61 ± 0.40 136.18 ± 8.97 0.04 ± 0.001 0.12 ± 0.01 <LOD* 0.12 ± 0.01 <LOD* CA 7.60 ± 1.13 109.59 ± 22.55 0.07 ± 0.01 0.08 ± 0.01 <LOD* 0.14 ± 0.03 <LOD* CR 14.00 ± 0.31 157.68 ± 14.40 0.11 ± 0.01 0.15 ± 0.02 <LOD* 0.18 ± 0.01 <LOD* CM 11.70 ± 0.57 123.32 ± 0.31 0.07 ± 0.002 0.07 ± 0.003 <LOD* 0.20 ± 0.002 0.05 ± 0.002 DL 15.33 ± 0.06 161.73 ± 3.51 0.10 ± 0.002 0.18 ± 0.01 <LOD* 0.13 ± 0.002 <LOD* JU 8.46 ± 0.48 100.84 ± 1.79 0.09 ± 0.001 0.10 ± 0.002 <LOD* 0.10 ± 0.001 <LOD* LA 14.72 ± 0.86 167.69 ± 0.50 0.13 ± 0.01 0.15 ± 0.002 <LOD* 0.20 ± 0.001 <LOD* LH 19.06 ± 0.60 160.00 ± 3.81 0.11 ± 0.002 0.17 ± 0.001 <LOD* 0.14 ± 0.002 <LOD* LP 5.96 ± 0.55 122.11 ± 0.17 0.06 ± 0.001 0.07 ± 0.001 <LOD* 0.19 ± 0.01 <LOD* LT 10.98 ± 2.54 165.00 ± 19.31 0.09 ± 0.01 0.13 ± 0.01 <LOD* 0.18 ± 0.03 <LOD* LI 15.65 ± 0.68 191.92 ± 2.97 0.16 ± 0.01 0.15 ± 0.003 <LOD* 0.20 ± 0.002 0.05 ± 0.01

MT 14.90 ± 1.15 174.67 ± 2.52 0.06 ± 0.001 0.19 ± 0.003 0.12 ± 0.01 0.18 ± 0.01 0.04 ± 0.02 PL 11.84 ± 1.06 102.29 ± 3.69 0.06 ± 0.002 0.07 ± 0.01 0.10 ± 0.003 0.12 ± 0.01 0.06 ± 0.002 SU 17.28 ± 0.09 133.95 ± 6.07 0.08 ± 0.002 0.14 ± 0.002 <LOD* 0.10 ± 0.003 <LOD* TA 17.26 ± 0.12 133.79 ± 7.69 0.09 ± 0.001 0.15 ± 0.001 <LOD* 0.13 ± 0.003 0.04 ± 0.001

29

Code Cu Zn As Se Mo Cd Pb TN 3.62 ± 0.59 162.53 ± 16.96 0.07 ± 0.001 0.12 ± 0.001 <LOD* 0.08 ± 0.002 <LOD* VE 8.27 ± 0.43 104.92 ± 2.84 0.12 ± 0.002 0.10 ± 0.002 <LOD* 0.12 ± 0.01 <LOD* YVI 20.37 ± 0.06 130.68 ± 5.31 0.11 ± 0.003 0.16 ± 0.003 0.10 ± 0.002 0.11 ± 0.002 0.04 ± 0.001

Roasted loose MN 0.47 ± 0.03 25.69 ± 0.20 0.02 ± 0.002 0.07 ± 0.003 <LOD* 0.07 ± 0.002 <LOD*

BC-Cm 0.30 ± 0.01 9.74 ± 0.56 <LOD* <LOD* <LOD* 0.02 ± 0.01 <LOD* MO 0.86 ± 0.04 35.76 ± 1.68 0.03 ± 0.001 0.07 ± 0.002 <LOD* 0.13 ± 0.01 <LOD*

Roasted tea bag DO 0.67 ± 0.03 22.15 ± 1.22 0.04 ± 0.002 0.08 ± 0.01 <LOD* 0.04 ± 0.002 <LOD* LIT <LOD* 24.14 ± 2.16 0.04 ± 0.001 0.07 ± 0.002 <LOD* 0.05 ± 0.001 <LOD* QU 0.16 ± 0.48 18.97 ± 1.14 0.04 ± 0.001 0.07 ± 0.01 <LOD* 0.04 ± 0.001 0.06 ± 0.02 ML 0.24 ± 0.00 30.05 ± 2.10 0.03 ± 0.003 0.08 ± 0.003 <LOD* 0.07 ± 0.002 <LOD*


30

Appendix 3.11: V, Cr, Mn, Fe, Co and Ni levels (mean ± standard deviation) of commercial yerba mate Brazilian iced tea

infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of

replicates.

Code V Cr Mn Fe Co Ni

Green loose

AM 0.03 ± 0.002 0.58 ± 0.02 1295 ± 5 6.14 ± 0.08 0.56 ± 0.002 8.21 ± 0.26 BC - Tr 0.02 ± 0.001 0.38 ± 0.002 1721 ± 37 9.12 ± 0.15 0.44 ± 0.003 6.76 ± 0.02

FO 0.03 ± 0.001 0.18 ± 0.01 2659 ± 63 13.04 ± 1.33 0.43 ± 0.03 16.30 ± 1.22 JE 0.04 ± 0.002 0.56 ± 0.02 1527 ± 23 8.11 ± 0.14 0.54 ± 0.01 7.14 ± 0.07 PV 0.02 ± 0.002 0.24 ± 0.01 2446 ± 55 9.82 ± 0.38 0.42 ± 0.03 14.03 ± 1.95 RM 0.04 ± 0.001 1.29 ± 0.03 1964 ± 4 6.41 ± 0.01 1.19 ± 0.04 10.56 ± 0.33

Green tea bag AT 0.16 ± 0.01 2.40 ± 0.11 3226 ± 364 40.88 ± 1.73 1.85 ± 0.10 32.07 ± 1.21 DL 0.23 ± 0.01 1.65 ± 0.03 2061 ± 18 38.93 ± 0.34 1.63 ± 0.01 27.76 ± 0.12 JU 0.07 ± 0.02 2.25 ± 0.59 2860 ± 883 17.77 ± 6.20 1.99 ± 0.53 19.69 ± 4.85

Roasted loose MN 0.03 ± 0.003 0.47 ± 0.02 899 ± 106 6.94 ± 0.41 0.22 ± 0.01 4.48 ± 0.41

BC-Cm 0.02 ± 0.003 0.53 ± 0.01 671 ± 3 6.88 ± 0.63 0.17 ± 0.02 2.19 ± 0.10 MO 0.05 ± 0.01 0.54 ± 0.08 890 ± 165 6.98 ± 0.94 0.26 ± 0.03 3.78 ± 0.60

Roasted tea bag DO 0.05 ± 0.002 0.42 ± 0.01 541 ± 35 13.67 ± 0.18 0.17 ± 0.001 4.91 ± 0.34 LIT 0.08 ± 0.01 0.47 ± 0.002 811 ± 42 20.21 ± 0.41 0.20 ± 0.001 6.16 ± 0.15

31

Code V Cr Mn Fe Co Ni QU 0.06 ± 0.001 0.37 ± 0.01 596 ± 17 14.80 ± 0.07 0.16 ± 0.002 4.41 ± 0.02

Appendix 3.12: Cu, Zn, As, Se, Mo Cd and Pb levels (mean ± standard deviation) of commercial yerba mate Brazilian iced

tea infusion samples (µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number

of replicates.

Code Cu Zn As Se Mo Cd Pb

Green loose AM 10.97 ± 0.09 109.22 ± 2.86 0.04 ± 0.003 0.11 ± 0.003 <LOD* 0.13 ± 0.004 0.01 ± 0.001

BC - Tr 16.38 ± 0.71 89.24 ± 0.24 0.04 ± 0.002 0.10 ± 0.01 0.07 ± 0.002 0.10 ± 0.004 <LOD* FO 23.62 ± 0.91 69.91 ± 6.04 0.04 ± 0.002 0.09 ± 0.002 <LOD* 0.08 ± 0.01 0.01 ± 0.002 JE 14.45 ± 0.01 91.47 ± 1.98 0.03 ± 0.002 0.13 ± 0.002 <LOD* 0.08 ± 0.003 0.01 ± 0.002 PV 22.55 ± 0.40 80.13 ± 10.32 0.05 ± 0.001 0.09 ± 0.002 0.04 ± 0.01 0.08 ± 0.001 0.01 ± 0.002 RM 10.95 ± 0.74 78.43 ± 0.75 0.05 ± 0.001 0.11 ± 0.002 <LOD* 0.08 ± 0.002 0.02 ± 0.002

Green tea bag AT 23.96 ± 1.60 316.00 ± 11.34 0.23 ± 0.01 0.28 ± 0.01 0.16 ± 0.01 0.37 ± 0.01 0.05 ± 0.01 DL 28.44 ± 1.52 254.30 ± 1.55 0.19 ± 0.01 0.28 ± 0.01 0.15 ± 0.002 0.20 ± 0.002 0.03 ± 0.001 JU 12.47 ± 3.97 139.64 ± 33.23 0.14 ± 0.04 0.16 ± 0.02 0.04 ± 0.03 0.14 ± 0.04 0.04 ± 0.01

Roasted loose MN <LOD* 39.66 ± 1.45 0.03 ± 0.002 0.09 ± 0.001 0.02 ± 0.003 0.08 ± 0.002 0.02 ± 0.002

BC-Cm <LOD* 15.87 ± 0.46 0.03 ± 0.002 0.08 ± 0.001 0.03 ± 0.01 0.03 ± 0.001 <LOD*

32

Code Cu Zn As Se Mo Cd Pb MO 0.05 ± 0.03 59.64 ± 6.05 0.04 ± 0.01 0.09 ± 0.002 0.00 ± 0.01 0.13 ± 0.01 <LOD*

Roasted tea bag DO 0.17 ± 0.13 30.17 ± 2.22 0.07 ± 0.01 0.10 ± 0.003 0.01 ± 0.003 0.06 ± 0.002 0.02 ± 0.002 LIT 0.29 ± 0.05 34.89 ± 0.58 0.06 ± 0.001 0.10 ± 0.002 0.02 ± 0.002 0.09 ± 0.01 0.04 ± 0.001 QU <LOD* 26.18 ± 1.62 0.07 ± 0.001 0.10 ± 0.002 0.02 ± 0.003 0.06 ± 0.002 0.07 ± 0.001

*refer to Table 2.2; <LOD lower than the limit of detection.

Appendix 3.13: Mn and Fe levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples

(µg/200 mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Mn Fe

Code F1 F2 F3 F4 F5 F1 F2 F3 F4 F5

BA 9421 ± 1106 10376 ± 152 6910 ± 11 4407 ± 75 2781 ± 347 26.32 ± 3.63 49.08 ± 3.82 35.72 ± 0.92 20.57 ± 1.32 11.38 ± 1.45

BC- Ca 18579 ± 1080 9863 ± 1043 4614 ± 453 2583 ± 289 1905 ± 436 72.07 ± 7.34 18.14 ± 12.98 23.83 ± 3.02 13.97 ± 0.92 8.41 ± 1.48

BC-Mg 11650 ± 319 9443 ± 394 4267 ± 88 2694 ± 68 1711 ± 123 48.68 ± 0.86 29.21 ± 4.44 18.42 ± 1.17 16.04 ± 0.73 9.95 ± 0.25

BC- Na 18988 ± 256 12538 ± 229 6274 ± 501 3586 ± 461 2315 ± 498 36.70 ± 0.16 16.15 ± 1.49 22.27 ± 3.01 12.89 ± 1.74 7.44 ± 1.04

BC-Pr 14610 ± 171 7908 ± 70 3309 ± 137 1034 ± 856 1099 ± 34 49.23 ± 8.00 17.36 ± 1.82 18.6 ± 0.87 11.40 ± 0.26 6.61 ± 0.16

BC-Te 3119 ± 252 3025 ± 138 2430 ± 510 1491 ± 44 1175 ± 15 16.25 ± 0.90 11.88 ± 0.88 14.42 ± 0.27 11.15 ± 0.15 8.53 ± 0.17

BC-Ex 12081 ± 671 10606 ± 385 5813 ± 474 3478 ± 263 2280 ± 189 57.48 ± 5.89 64.73 ± 12.52 45.22 ± 5.09 23.96 ± 1.92 14.29 ± 1.21

CAN 7756 ± 418 9618 ± 94 6359 ± 2251 3954 ± 115 2578 ± 35 21.45 ± 4.21 26.54 ± 5.60 22.43 ± 1.07 22.00 ± 1.33 12.94 ± 0.46

SAR 13205 ± 1229 12121 ± 136 7448 ± 165 4599 ± 225 2957 ± 263 42.81 ± 5.26 53.80 ± 4.58 41.04 ± 2.32 22.54 ± 1.71 12.61 ± 0.62

AM 6808 ± 827 5466 ± 24 4398 ± 559 2822 ± 84 4310 ± 68 13.71 ± 1.75 21.61 ± 3.26 24.35 ± 2.38 13.98 ± 0.04 8.59 ± 0.02

BC - Tr 19459 ± 827 9965 ± 134 5091 ± 977 2643 ± 286 1889 ± 61 67.85 ± 11.76 41.8 ± 11.64 27.44 ± 6.73 13.33 ± 2.50 8.40 ± 0.73

33

Mn Fe


FV 7647 ± 7 6909 ± 419 3737 ± 203 2341 ± 84 1596 ± 164 59.41 ± 17.02 32.95 ± 8.66 21.41 ± 1.08 21.68 ± 0.94 13.10 ± 0.59

FO 30591 ± 2142 16964 ± 1180 7275 ± 649 4073 ± 19 2418 ± 0.9 74.13 ± 8.58 35.78 ± 0.78 25.47 ± 2.65 14.17 ± 0.61 8.87 ± 0.33

JE 6236 ± 732 7105 ± 885 5482 ± 563 3606 ± 165 2566 ± 76 22.68 ± 5.75 39.12 ± 0.55 33.01 ± 2.41 20.04 ± 2.11 12.11 ± 1.53

KA 6232 ± 369 6898 ± 432 6903 ± 275 4952 ± 285 3723 ± 143 10.82 ± 0.96 17.80 ± 1.87 21.09 ± 1.92 14.13 ± 1.04 9.73 ± 0.44

KR 4508 ± 179 4831 ± 613 3480 ± 339 2537 ± 5 1931 ± 106 13.27 ± 0.66 14.64 ± 0.10 23.48 ± 1.41 13.79 ± 1.20 11.20 ± 0.11

PI 3678 ± 394 4930 ± 115 3753 ± 109 2893 ± 83 2065 ± 24 9.38 ± 0.50 12.86 ± 0.22 20.45 ± 0.13 13.77 ± 0.60 9.84 ± 0.27

PV 28403 ± 1606 13986 ± 834 5953 ± 175 2835 ± 3 1808 ± 42 61.59 ± 2.04 30.48 ± 4.37 20.68 ± 0.47 9.68 ± 0.12 6.07 ± 0.11

RO 2554 ± 26 2563 ± 126 1803 ± 78 13747 ± 30 1102 ± 38 13.58 ± 0.43 13.89 ± 3.88 16.30 ± 0.51 11.69 ± 0.38 9.21 ± 0.50

RM 7165 ± 738 9003 ± 708 6353 ± 302 4714 ± 37 3550 ± 43 12.20 ± 1.80 23.77 ± 3.32 22.95 ± 2.01 14.88 ± 0.59 9.18 ± 0.03

SA 5975 ± 820 3094 ± 211 1449 ± 186 741 ± 71 436 ± 47 66.79 ± 18.73 43.55 ± 8.75 28.54 ± 4.33 15.05 ± 1.01 9.94 ± 0.69

YU 7092 ± 522 10596 ± 1241 7806 ± 462 5869 ± 98 3755 ± 45 13.46 ± 0.06 23.42 ± 8.84 31.09 ± 5.61 27.92 ± 1.39 15.88 ± 0.28

34

Appendix 3.14: Cu and Zn levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples


Cu Zn


BA 92.98 ± 4.31 134.90 ± 5.79 76.31 ± 1.48 49.37 ± 2.40 29.47 ± 2.97 360.53 ± 52.72 494.66 ± 20.56 315.55 ± 3.67 198.54 ± 9.44 112.16 ± 13.89

BC- Ca 114.06 ± 5.37 44.27 ± 43.20 38.86 ± 3.12 25.47 ± 3.01 16.99 ± 4.13 646.91 ± 32.29 230.33 ± 228.19 183.68 ± 15.23 101.26 ± 11.58 56.24 ± 11.19

BC-Mg 131.38 ± 2.00 108.67 ± 6.81 58.98 ± 4.79 37.39 ± 2.02 23.84 ± 0.82 673.53 ± 17.11 564.90 ± 0.92 282.45 ± 54.09 176.85 ± 9.86 102.04 ± 6.09

BC- Na 115.83 ± 0.09 87.50 ± 3.56 48.75 ± 2.80 32.12 ± 3.42 20.98 ± 4.32 508.36 ± 2.62 388.72 ± 14.22 195.26 ± 16.07 115.37 ± 15.94 59.22 ± 13.31

BC-Pr 157.09 ± 0.66 97.75 ± 4.33 48.32 ± 2.67 29.72 ± 1.51 16.83 ± 0.13 549.44 ± 21.49 318.09 ± 31.12 130.46 ± 2.99 66.16 ± 3.41 31.37 ± 0.74

BC-Te 48.46 ± 4.13 43.92 ± 1.81 31.36 ± 0.99 24.19 ± 0.18 19.26 ± 0.11 188.16 ± 6.10 196.26 ± 9.73 118.41 ± 0.07 113.32 ± 4.44 80.31 ± 4.14

BC-Ex 142.81 ± 2.21 153.30 ± 3.25 74.42 ± 7.25 44.72 ± 3.77 28.28 ± 2.35 593.09 ± 33.86 652.37 ± 9.14 336.17 ± 25.19 201.13 ± 7.10 124.23 ± 4.33

CAN 103.26 ± 11.87 127.08 ± 8.41 111.68 ± 10.22 52.60 ± 2.34 34.36 ± 0.14 305.47 ± 51.88 391.33 ± 1.55 268.69 ± 2.93 186.99 ± 11.16 116.26 ± 3.70

SAR 113.95 ± 12.04 141.12 ± 8.95 75.31 ± 2.98 46.68 ± 2.33 28.69 ± 2.05 472.53 ± 53.81 546.32 ± 67.06 314.66 ± 17.96 187.71 ± 13.27 111.79 ± 4.70

AM 45.20 ± 6.43 61.62 ± 0.23 42.48 ± 3.00 27.95 ± 0.16 18.28 ± 0.50 478.34 ± 70.28 504.38 ± 2.94 347.13 ± 27.22 211.77 ± 2.93 131.50 ± 3.35

BC - Tr 133.34 ± 11.45 96.62 ± 1.01 46.00 ± 6.79 26.65 ± 2.50 16.95 ± 0.73 690.77 ± 70.38 428.24 ± 5.66 206.65 ± 38.58 97.80 ± 18.42 54.77 ± 4.29

FV 94.00 ± 0.49 76.73 ± 4.76 54.04 ± 0.18 33.08 ± 1.00 21.67 ± 1.61 570.02 ± 2.89 477.61 ± 20.20 288.20 ± 0.56 195.87 ± 6.65 117.87 ± 8.85

FO 183.22 ± 16.31 137.62 ± 9.62 54.73 ± 3.97 30.76 ± 0.25 17.54 ± 0.08 495.81 ± 30.67 361.88 ± 17.61 139.41 ± 12.27 65.00 ± 2.33 32.69 ± 0.02

JE 67.73 ± 20.12 88.57 ± 14.21 54.22 ± 6.04 36.03 ± 1.70 24.02 ± 0.01 306.25 ± 53.86 411.84 ± 24.90 277.16 ± 25.37 190.76 ± 9.00 122.10 ± 1.06

KA 43.54 ± 1.61 54.52 ± 4.70 45.81 ± 2.42 32.14 ± 1.67 23.08 ± 0.55 265.03 ± 13.46 347.11 ± 39.03 308.53 ± 21.15 214.27 ± 15.82 146.37 ± 5.26

KR 52.61 ± 0.29 53.48 ± 5.67 48.99 ± 6.11 30.64 ± 0.43 23.11 ± 1.50 303.49 ± 11.53 338.23 ± 50.64 287.69 ± 32.73 186.13 ± 2.16 132.01 ± 4.61

PI 32.60 ± 2.92 53.98 ± 1.46 44.40 ± 0.43 29.61 ± 1.08 22.21 ± 0.35 250.79 ± 10.63 395.04 ± 11.63 311.90 ± 5.50 218.73 ± 14.67 152.52 ± 8.78

PV 174.98 ± 24.77 110.29 ± 7.40 48.38 ± 0.04 23.24 ± 0.12 12.46 ± 0.07 511.90 ± 43.32 261.28 ± 16.83 130.01 ± 1.92 47.46 ± 0.06 24.25 ± 0.83

RO 35.12 ± 0.06 34.90 ± 3.16 30.80 ± 0.52 21.92 ± 1.07 18.29 ± 1.70 362.57 ± 18.09 347.48 ± 38.09 286.83 ± 5.04 199.91 ± 10.11 150.67 ± 14.61

35

Cu Zn


RM 39.97 ± 0.66 65.13 ± 3.45 39.40 ± 2.54 29.21 ± 0.71 19.64 ± 0.14 253.92 ± 10.25 348.39 ± 2.07 232.15 ± 20.70 164.58 ± 7.34 103.49 ± 1.08

SA 121.35 ± 19.89 76.96 ± 2.28 36.45 ± 3.03 20.33 ± 1.13 11.78 ± 0.42 695.14 ± 101.79 426.60 ± 4.84 197.84 ± 17.45 107.02 ± 4.71 59.06 ± 2.36

YU 77.04 ± 2.08 145.81 ± 43.41 97.22 ± 6.86 64.94 ± 0.49 41.39 ± 0.35 241.18 ± 23.46 410.75 ± 58.79 306.40 ± 13.05 232.73 ± 3.61 142.03 ± 1.51

Appendix 3.15: Ni and Cr levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200

mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Ni Cr


BA 41.78 ± 5.03 58.42 ± 0.10 33.41 ± 0.16 21.45 ± 0.87 11.71 ± 1.31 1.75 ± 0.18 2.55 ± 0.03 1.40 ± 0.02 0.97 ± 0.05 0.58 ± 0.06

BC- Ca 55.36 ± 2.81 19.55 ± 18.29 10.88 ± 0.87 6.16 ± 0.67 3.59 ± 0.78 5.23 ± 0.26 2.34 ± 1.50 1.30 ± 0.09 0.84 ± 0.09 0.54 ± 0.11

BC-Mg 43.59 ± 0.51 38.56 ± 2.23 20.55 ± 1.76 9.46 ± 0.61 5.62 ± 0.17 2.08 ± 0.01 2.05 ± 0.25 1.33 ± 0.06 0.56 ± 0.03 0.35 ± 0.01

BC- Na 48.51 ± 0.74 34.48 ± 1.72 13.71 ± 1.08 8.00 ± 1.15 4.34 ± 0.88 1.10 ± 0.03 1.03 ± 0.05 0.34 ± 0.03 0.22 ± 0.03 0.13 ± 0.02

BC-Pr 72.80 ± 2.40 41.91 ± 1.03 13.00 ± 0.47 6.94 ± 0.52 3.34 ± 0.17 0.81 ± 0.003 0.63 ± 0.02 0.17 ± 0.01 0.11 ± 0.003 0.07 ± 0.002

BC-Te 10.61 ± 0.18 10.50 ± 0.42 5.97 ± 0.35 4.93 ± 0.08 3.95 ± 0.01 0.66 ± 0.07 0.71 ± 0.02 0.37 ± 0.002 0.32 ± 0.01 0.26 ± 0.01

BC-Ex 49.98 ± 1.51 57.54 ± 2.14 27.96 ± 2.96 16.27 ± 1.94 9.93 ± 0.81 3.16 ± 0.10 3.95 ± 0.16 1.92 ± 0.22 1.23 ± 0.10 0.79 ± 0.06

CAN 46.01 ± 5.60 59.18 ± 1.67 47.34 ± 2.05 24.26 ± 1.03 14.51 ± 0.13 1.27 ± 0.15 1.74 ± 0.05 1.43 ± 0.11 0.68 ± 0.04 0.45 ± 0.01

SAR 55.43 ± 6.15 65.50 ± 3.51 34.99 ± 1.08 21.56 ± 0.75 12.36 ± 0.93 2.04 ± 0.21 2.60 ± 0.06 1.34 ± 0.03 0.88 ± 0.05 0.54 ± 0.03

AM 33.66 ± 3.28 42.10 ± 0.03 27.05 ± 2.25 16.68 ± 0.51 10.47 ± 0.23 2.14 ± 0.27 2.88 ± 0.02 1.73 ± 0.12 1.18 ± 0.02 0.77 ± 0.03

BC - Tr 65.88 ± 5.50 41.07 ± 0.40 15.88 ± 4.00 7.36 ± 1.18 4.15 ± 0.43 2.79 ± 0.13 1.96 ± 0.02 0.80 ± 0.14 0.46 ± 0.06 0.29 ± 0.02

FV 29.90 ± 0.24 27.13 ± 0.62 18.75 ± 0.16 9.35 ± 0.31 6.08 ± 0.44 9.29 ± 0.11 9.01 ± 0.16 6.59 ± 0.14 3.59 ± 0.13 2.40 ± 0.15

36

Ni Cr


FO 117.75 ± 9.17 81.20 ± 6.68 27.56 ± 2.11 12.20 ± 0.10 6.05 ± 0.09 0.78 ± 0.05 0.62 ± 0.03 0.18 ± 0.01 0.11 ± 0.002 0.06 ± 0.002

JE 26.78 ± 4.54 36.68 ± 3.64 23.20 ± 2.81 14.73 ± 0.42 9.74 ± 0.12 1.76 ± 0.27 2.43 ± 0.26 1.52 ± 0.18 1.10 ± 0.06 0.74 ± 0.002

KA 22.09 ± 0.26 29.54 ± 1.82 25.18 ± 1.54 18.03 ± 1.32 11.99 ± 0.4 3.69 ± 0.05 5.14 ± 0.34 4.25 ± 0.23 3.17 ± 0.20 2.27 ± 0.06

KR 31.54 ± 2.16 35.41 ± 3.56 32.74 ± 3.45 20.54 ± 0.56 14.31 ± 0.70 4.96 ± 0.21 5.84 ± 0.50 5.53 ± 0.65 3.51 ± 0.08 2.70 ± 0.16

PI 24.64 ± 1.69 37.98 ± 0.21 33.20 ± 0.14 21.74 ± 0.55 14.89 ± 0.20 2.88 ± 0.26 4.46 ± 0.06 4.07 ± 0.06 2.66 ± 0.06 2.05 ± 0.01

PV 103.2 ± 8.69 58.66 ± 2.86 20.25 ± 0.37 7.40 ± 0.11 3.65 ± 0.11 1.48 ± 0.22 0.82 ± 0.02 0.29 ± 0.01 0.15 ± 0.00 0.09 ± 0.002

RO 21.02 ± 0.05 22.29 ± 1.21 19.96 ± 0.25 12.11 ± 0.18 10.04 ± 0.46 1.49 ± 0.05 1.59 ± 0.06 1.48 ± 0.01 0.89 ± 0.00 0.74 ± 0.02

RM 35.77 ± 2.12 53.99 ± 2.06 32.23 ± 2.16 23.68 ± 0.66 14.91 ± 0.14 3.77 ± 0.41 5.82 ± 0.34 3.47 ± 0.38 2.68 ± 0.19 1.78 ± 0.08

SA 31.36 ± 4.49 20.43 ± 1.09 7.76 ± 0.78 4.41 ± 0.31 2.61 ± 0.10 2.46 ± 0.30 1.78 ± 0.06 0.75 ± 0.07 0.45 ± 0.03 0.28 ± 0.02

YU 35.18 ± 1.43 58.26 ± 5.93 49.82 ± 3.11 32.56 ± 0.57 19.96 ± 0.07 1.16 ± 0.03 1.84 ± 0.24 1.61 ± 0.13 0.99 ± 0.03 0.64 ± 0.01

37

Appendix 3.16: V and Co levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200

mL) determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

V Co


BA 0.09 ± 0.02 0.09 ± 0.01 0.04 ± 0.003 0.03 ± 0.001 0.02 ± 0.002 1.48 ± 0.19 1.89 ± 0.02 1.04 ± 0.02 0.67 ± 0.002 0.40 ± 0.03 BC- Ca 0.17 ± 0.02 0.19 ± 0.06 0.43 ± 0.28 0.34 ± 0.11 0.14 ± 0.02 3.23 ± 0.08 1.14 ± 1.24 0.87 ± 0.06 0.55 ± 0.05 0.37 ± 0.07 BC-Mg 0.08 ± 0.01 0.19 ± 0.11 0.30 ± 0.01 0.03 ± 0.002 0.04 ± 0.01 2.57 ± 0.03 2.28 ± 0.20 1.21 ± 0.04 0.64 ± 0.03 0.40 ± 0.02 BC- Na 0.10 ± 0.01 0.22 ± 0.002 0.07 ± 0.04 0.11 ± 0.06 0.12 ± 0.08 1.41 ± 0.03 1.10 ± 0.03 0.49 ± 0.03 0.31 ± 0.03 0.18 ± 0.03 BC-Pr 0.10 ± 0.01 0.11 ± 0.01 0.04 ± 0.02 0.05 ± 0.02 0.04 ± 0.01 1.37 ± 0.08 0.85 ± 0.003 0.33 ± 0.002 0.20 ± 0.02 0.11 ± 0.01 BC-Te 0.06 ± 0.01 0.06 ± 0.01 0.01 ± 0.003 0.01 ± 0.002 0.01 ± 0.002 1.34 ± 0.08 1.49 ± 0.11 0.84 ± 0.02 0.69 ± 0.01 0.56 ± 0.01 BC-Ex 0.24 ± 0.03 0.25 ± 0.08 0.24 ± 0.18 0.34 ± 0.30 0.32 ± 0.29 2.69 ± 0.15 2.62 ± 0.03 1.22 ± 0.10 0.76 ± 0.02 0.48 ± 0.02 CAN 0.11 ± 0.05 0.09 ± 0.05 0.13 ± 0.07 0.03 ± 0.01 0.02 ± 0.01 1.32 ± 0.08 1.63 ± 0.01 1.26 ± 0.09 0.62 ± 0.05 0.41 ± 0.01 SAR 0.14 ± 0.03 0.15 ± 0.01 0.06 ± 0.01 0.04 ± 0.003 0.02 ± 0.001 1.77 ± 0.22 1.97 ± 0.12 0.99 ± 0.05 0.61 ± 0.04 0.37 ± 0.02 AM 0.51 ± 0.25 0.28 ± 0.09 0.18 ± 0.04 0.09 ± 0.002 0.07 ± 0.002 2.94 ± 0.50 3.36 ± 0.01 1.90 ± 0.10 1.24 ± 0.04 0.81 ± 0.03

BC - Tr 0.18 ± 0.03 0.09 ± 0.01 0.05 ± 0.002 0.05 ± 0.01 0.03 ± 0.002 4.34 ± 0.39 2.63 ± 0.15 1.09 ± 0.19 0.58 ± 0.09 0.35 ± 0.03 FV 0.11 ± 0.01 0.11 ± 0.003 0.16 ± 0.002 0.04 ± 0.01 0.03 ± 0.001 2.77 ± 0.01 2.48 ± 0.10 1.60 ± 0.04 0.82 ± 0.02 0.53 ± 0.04 FO 0.14 ± 0.01 0.08 ± 0.01 0.02 ± 0.001 0.02 ± 0.002 0.02 ± 0.001 3.21 ± 0.21 2.26 ± 0.11 0.82 ± 0.04 0.44 ± 0.001 0.24 ± 0.01 JE 0.15 ± 0.03 0.16 ± 0.03 0.07 ± 0.003 0.04 ± 0.001 0.03 ± 0.002 2.15 ± 0.36 2.65 ± 0.18 1.65 ± 0.16 1.16 ± 0.04 0.78 ± 0.02 KA 0.06 ± 0.002 0.06 ± 0.002 0.03 ± 0.01 0.02 ± 0.002 0.02 ± 0.002 1.59 ± 0.02 1.96 ± 0.10 1.71 ± 0.11 1.26 ± 0.08 0.92 ± 0.02 KR 0.10 ± 0.004 0.11 ± 0.04 0.10 ± 0.02 0.02 ± 0.001 0.02 ± 0.002 3.69 ± 0.27 4.08 ± 0.30 3.62 ± 0.50 2.26 ± 0.08 1.72 ± 0.09 PI 0.05 ± 0.01 0.12 ± 0.01 0.11 ± 0.01 0.02 ± 0.002 0.02 ± 0.001 1.83 ± 0.14 2.67 ± 0.08 2.39 ± 0.04 1.47 ± 0.06 1.13 ± 0.01 PV 0.15 ± 0.01 0.07 ± 0.001 0.03 ± 0.002 0.02 ± 0.003 0.02 ± 0.002 3.12 ± 0.13 1.84 ± 0.03 0.68 ± 0.003 0.32 ± 0.01 0.18 ± 0.01 RO 0.08 ± 0.006 0.14 ± 0.02 0.09 ± 0.03 0.03 ± 0.001 0.03 ± 0.01 1.29 ± 0.03 1.30 ± 0.01 0.89 ± 0.01 0.66 ± 0.02 0.54 ± 0.04

38

V Co


RM 0.29 ± 0.11 0.32 ± 0.13 0.22 ± 0.12 0.16 ± 0.07 0.09 ± 0.04 4.21 ± 0.28 6.00 ± 0.32 3.35 ± 0.25 2.50 ± 0.07 1.66 ± 0.01 SA 0.15 ± 0.02 0.09 ± 0.02 0.04 ± 0.01 0.03 ± 0.002 0.02 ± 0.002 3.79 ± 0.47 2.52 ± 0.16 1.06 ± 0.09 0.61 ± 0.03 0.37 ± 0.01 YU 0.08 ± 0.01 0.12 ± 0.04 0.10 ± 0.002 0.03 ± 0.001 0.02 ± 0.001 0.81 ± 0.07 1.24 ± 0.14 1.01 ± 0.10 0.67 ± 0.02 0.43 ± 0.002

Appendix 3.17: As and Se levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples


As Se



BC - Tr 0.28 ± 0.06 0.16 ± 0.03 0.24 ± 0.02 0.18 ± 0.02 0.15 ± 0.03 0.28 ± 0.06 0.04 ± 0.002 0.15 ± 0.02 0.10 ± 0.01 0.04 ± 0.001 FV 0.14 ± 0.06 0.29 ± 0.06 0.16 ± 0.01 0.11 ± 0.002 0.09 ± 0.01 0.14 ± 0.06 0.29 ± 0.06 0.04 ± 0.003 0.12 ± 0.002 0.09 ± 0.01

39

As Se


FO 0.14 ± 0.01 0.07 ± 0.02 0.15 ± 0.01 0.13 ± 0.01 0.11 ± 0.01 0.14 ± 0.01 0.04 ± 0.003 0.10 ± 0.002 0.09 ± 0.001 0.06 ± 0.002 JE 0.07 ± 0.04 0.05 ± 0.001 0.08 ± 0.001 0.06 ± 0.01 0.05 ± 0.01 0.07 ± 0.04 0.03 ± 0.003 0.27 ± 0.03 0.21 ± 0.02 0.13 ± 0.002 KA 0.11 ± 0.003 0.04 ± 0.02 0.16 ± 0.02 0.13 ± 0.02 0.11 ± 0.01 0.11 ± 0.002 0.03 ± 0.002 0.24 ± 0.02 0.20 ± 0.01 0.13 ± 0.003 KR 0.03 ± 0.03 0.11 ± 0.01 0.08 ± 0.01 0.11 ± 0.01 0.09 ± 0.01 0.03 ± 0.03 0.11 ± 0.01 0.03 ± 0.001 0.15 ± 0.001 0.12 ± 0.01 PI 0.05 ± 0.003 0.15 ± 0.05 0.11 ± 0.002 0.16 ± 0.01 0.14 ± 0.01 0.05 ± 0.001 0.15 ± 0.05 0.03 ± 0.002 0.22 ± 0.02 0.18 ± 0.002 PV 0.28 ± 0.002 0.20 ± 0.05 0.22 ± 0.01 0.16 ± 0.01 0.13 ± 0.01 0.28 ± 0.001 0.03 ± 0.002 0.11 ± 0.002 0.08 ± 0.002 0.03 ± 0.004 RO 0.17 ± 0.03 0.19 ± 0.002 0.14 ± 0.01 0.17 ± 0.01 0.16 ± 0.01 0.17 ± 0.03 0.19 ± 0.002 0.04 ± 0.002 0.12 ± 0.003 0.11 ± 0.003 RM 0.28 ± 0.03 0.41 ± 0.12 0.30 ± 0.04 0.24 ± 0.02 0.18 ± 0.01 0.28 ± 0.03 0.04 ± 0.003 0.24 ± 0.02 0.17 ± 0.01 0.08 ± 0.01 SA 0.04 ± 0.002 0.02 ± 0.02 0.06 ± 0.01 0.05 ± 0.01 0.05 ± 0.01 0.04 ± 0.002 0.04 ± 0.003 0.13 ± 0.01 0.10 ± 0.001 0.08 ± 0.002 YU 0.08 ± 0.02 0.23 ± 0.01 0.13 ± 0.01 0.21 ± 0.01 0.16 ± 0.01 0.08 ± 0.02 0.23 ± 0.01 0.03 ± 0.002 0.16 ± 0.001 0.12 ± 0.01

40

Appendix 3.18: Mo and Cd levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples


Mo Cd



BC - Tr 0.41 ± 0.07 0.24 ± 0.07 0.34 ± 0.07 0.19 ± 0.02 0.13 ± 0.01 1.26 ± 0.03 0.68 ± 0.02 0.28 ± 0.05 0.17 ± 0.02 0.12 ± 0.01 FV 0.12 ± 0.08 0.21 ± 0.08 0.17 ± 0.04 0.23 ± 0.002 0.16 ± 0.01 1.72 ± 0.11 1.35 ± 0.11 0.80 ± 0.003 0.41 ± 0.01 0.27 ± 0.02 FO 0.08 ± 0.06 0.04 ± 0.01 0.09 ± 0.01 0.05 ± 0.002 0.03 ± 0.002 0.92 ± 0.05 0.59 ± 0.05 0.20 ± 0.01 0.13 ± 0.001 0.07 ± 0.002 JE 0.05 ± 0.05 0.03 ± 0.01 0.10 ± 0.01 0.07 ± 0.001 0.04 ± 0.003 0.54 ± 0.10 0.59 ± 0.06 0.35 ± 0.02 0.27 ± 0.003 0.18 ± 0.003 KA 0.02 ± 0.03 0.01 ± 0.01 0.14 ± 0.01 0.10 ± 0.01 0.07 ± 0.001 0.47 ± 0.04 0.55 ± 0.04 0.40 ± 0.02 0.28 ± 0.02 0.20 ± 0.002 KR 0.003 ± 0.001 0.05 ± 0.04 0.05 ± 0.03 0.07 ± 0.001 0.05 ± 0.01 0.50 ± 0.06 0.49 ± 0.002 0.34 ± 0.02 0.18 ± 0.002 0.13 ± 0.01 PI 0.02 ± 0.01 0.03 ± 0.01 0.02 ± 0.002 0.09 ± 0.002 0.06 ± 0.001 0.55 ± 0.13 0.65 ± 0.10 0.47 ± 0.02 0.27 ± 0.001 0.19 ± 0.001 PV 0.39 ± 0.01 0.10 ± 0.04 0.18 ± 0.001 0.09 ± 0.001 0.07 ± 0.002 0.93 ± 0.02 0.52 ± 0.02 0.19 ± 0.006 0.11 ± 0.003 0.08 ± 0.005 RO 0.01 ± 0.005 0.04 ± 0.02 0.02 ± 0.01 0.08 ± 0.002 0.06 ± 0.002 0.51 ± 0.01 0.49 ± 0.06 0.37 ± 0.01 0.21 ± 0.01 0.16 ± 0.02

41

Mo Cd


RM 0.05 ± 0.03 0.01 ± 0.03 0.09 ± 0.01 0.06 ± 0.002 0.06 ± 0.001 0.93 ± 0.01 0.78 ± 0.02 0.45 ± 0.01 0.31 ± 0.004 0.22 ± 0.00 SA 0.05 ± 0.02 0.06 ± 0.05 0.09 ± 0.01 0.06 ± 0.01 0.05 ± 0.002 1.76 ± 0.25 1.05 ± 0.002 0.43 ± 0.03 0.25 ± 0.005 0.14 ± 0.003 YU 0.02 ± 0.01 0.12 ± 0.07 0.20 ± 0.06 0.21 ± 0.01 0.12 ± 0.001 0.72 ± 0.07 1.07 ± 0.11 0.74 ± 0.01 0.44 ± 0.01 0.27 ± 0.002

Appendix 3.19: Pb levels (mean ± standard deviation) of commercial yerba mate bombilla infusion samples (µg/200 mL)

determined using ICP-MS (refer to section 2.3) and n= 2; n is the number of replicates.

Pb Code F1 F2 F3 F4 F5

BA 0.54 ± 0.04 0.42 ± 0.04 0.19 ± 0.01 0.12 ± 0.002 0.09 ± 0.01 BC- Ca 0.66 ± 0.01 0.22 ± 0.11 0.13 ± 0.01 0.09 ± 0.01 0.07 ± 0.01 BC-Mg 0.43 ± 0.03 0.36 ± 0.05 0.26 ± 0.02 0.06 ± 0.002 0.05 ± 0.003 BC- Na 0.43 ± 0.002 0.36 ± 0.002 0.08 ± 0.01 0.06 ± 0.01 0.04 ± 0.01 BC-Pr 0.51 ± 0.05 0.39 ± 0.05 0.05 ± 0.01 0.05 ± 0.01 0.04 ± 0.01 BC-Te 0.23 ± 0.01 0.22 ± 0.01 0.14 ± 0.01 0.12 ± 0.002 0.09 ± 0.002 BC-Ex 0.77 ± 0.07 0.45 ± 0.03 0.17 ± 0.01 0.10 ± 0.01 0.08 ± 0.001 CAN 1.01 ± 0.35 0.70 ± 0.06 0.88 ± 0.43 0.13 ± 0.04 0.09 ± 0.002 SAR 0.72 ± 0.01 0.54 ± 0.01 0.23 ± 0.01 0.15 ± 0.01 0.11 ± 0.001 AM 0.70 ± 0.46 0.27 ± 0.08 0.10 ± 0.03 0.07 ± 0.002 0.06 ± 0.001

BC - Tr 0.32 ± 0.01 0.16 ± 0.01 0.05 ± 0.01 0.03 ± 0.003 0.03 ± 0.002 FV 0.59 ± 0.01 0.38 ± 0.04 0.18 ± 0.01 0.08 ± 0.001 0.06 ± 0.01

42

Pb Code F1 F2 F3 F4 F5 FO 0.52 ± 0.03 0.23 ± 0.04 0.07 ± 0.001 0.04 ± 0.002 0.03 ± 0.003 JE 0.05 ± 0.05 0.03 ± 0.01 0.10 ± 0.01 0.07 ± 0.00 0.04 ± 0.003 KA 0.02 ± 0.03 0.01 ± 0.01 0.14 ± 0.01 0.10 ± 0.01 0.07 ± 0.002 KR 0.003 ± 0.001 0.05 ± 0.04 0.05 ± 0.03 0.07 ± 0.002 0.05 ± 0.01 PI 0.02 ± 0.01 0.03 ± 0.01 0.02 ± 0.001 0.09 ± 0.002 0.06 ± 0.001 PV 0.39 ± 0.01 0.10 ± 0.04 0.18 ± 0.002 0.09 ± 0.001 0.07 ± 0.001 RO 0.01 ± 0.005 0.04 ± 0.02 0.02 ± 0.01 0.08 ± 0.001 0.06 ± 0.002 RM 0.05 ± 0.03 0.01 ± 0.03 0.09 ± 0.01 0.06 ± 0.002 0.06 ± 0.001 SA 0.05 ± 0.02 0.06 ± 0.05 0.09 ± 0.01 0.06 ± 0.01 0.05 ± 0.003 YU 0.02 ± 0.01 0.12 ± 0.07 0.20 ± 0.06 0.21 ± 0.01 0.12 ± 0.001

43

Appendix 3.20: Total polyphenol content of commercial green loose yerba mate regular and bombilla infusions (mg

GAE/200 mL) determined using Folin-Ciocalteu assay (refer to section 2.4).

Code Regular infusions Bombilla F1 F2 F3 F4 F5

BA 151.74 1370.27 1062.52 731.10 499.10 338.92 BC- Ca 163.61 2088.29 1090.70 584.18 343.50 230.34 BC-Mg 120.53 1204.56 904.17 548.82 373.90 239.49 BC- Na 156.69 1674.07 968.62 583.80 419.93 291.05 BC-Pr 135.36 2236.81 1020.65 487.19 314.76 228.54 BC-Te 79.73 475.43 421.77 320.51 265.27 232.13 BC-Ex 161.78 1679.59 1105.13 646.40 396.00 305.78 CAN 148.62 911.26 1162.11 807.14 552.80 386.77 SAR 154.48 1087.11 1088.90 711.71 487.19 329.13 AM 159.04 886.37 875.67 652.85 454.98 321.29

BC - Tr 152.20 2259.58 1215.61 554.34 366.54 318.66 FV 155.76 1818.74 966.76 639.86 427.91 282.43 FO 169.09 2055.28 1044.00 544.67 286.02 255.48 JE 149.46 939.82 894.92 548.26 440.49 312.96 KA 106.54 990.74 969.63 717.06 459.18 340.85 KR 125.26 1011.94 999.66 676.43 473.31 319.65 PI 135.76 868.87 701.16 551.03 445.05 319.65 PV 149.91 2451.35 1083.52 593.16 302.18 172.86 RO 81.89 869.77 790.74 462.04 302.18 226.74

44

Code Regular infusions Bombilla F1 F2 F3 F4 F5

RM 117.04 610.07 565.50 444.29 314.16 282.07 SA 103.80 1108.66 638.07 300.39 194.41 129.75 YU 135.76 974.85 1193.91 824.80 528.07 369.11

45

Appendix 3.21: Total polyphenol content of commercial green/roasted loose or tea bag yerba mate regular infusions (mg

GAE/200mL) determined using Folin-Ciocalteu assay (refer to section 2.4).

Code Total polyphenol Green tea bag

AT 196.48 AO 147.17 CA 174.11 CR 235.87 CM 231.33 DL 256.49 JU 181.01 LA 250.35 LH 261.03 LP 166.57 LT 246.17 LI 216.89

MT 252.78 PL 228.03 SU 228.85 TA 237.10 TN 213.59 VE 134.81 YVI 239.58

46

Code Total polyphenol

Roasted loose MN 44.91

BC-Cm 44.03 MO 46.74

Roasted tea bag DO 56.33 LIT 71.39 QU 59.52 ML 93.64

47

Appendix 3.22: Total polyphenol content of commercial green/roasted loose or tea bag yerba mate Brazilian iced tea

infusions (mg GAE/200mL) determined using Folin-Ciocalteu assay (refer to section 2.4).

Code Total polyphenol

Green loose BC - Tr 112.18

AT 312.46 Green tea bag

DL 297.63 JU 216.50

Roasted loose MN 41.71

BC-Cm 29.20 MO 35.69

Roasted tea bag DO 63.97 LIT 82.98 QU 67.68

48

Appendix 3.23: Chlorogenic acid, theobromine and caffeine content of commercial green/roasted loose or tea bag yerba

mate regular infusions (mg/200 mL) determined by UHPLC (refer to section 2.5).

Code 3-Caffeoylquinic acid Theobromine 4-Caffeoylquinic acid 5-Caffeoylquinic acid Caffeine

Green loose BC- Ca 42.95 3.46 15.00 26.95 18.57 BC-Pr 55.62 4.27 13.44 20.56 34.58

PV 47.36 5.03 15.73 29.09 26.94 Green tea bag

CA 41.44 3.31 12.99 19.43 21.87 Roasted loose

BC-Cm 2.86 1.13 2.03 2.79 5.14 Roasted tea bag

QU 3.98 1.53 3.16 3.62 7.58

49

Appendix 3.24: Chlorogenic acid, theobromine and caffeine content of commercial green/roasted loose or tea bag yerba

mate bombilla infusions (mg/200 mL) determined by UHPLC (refer to section 2.5).

Compound Fraction BC-Ca BC Pr PV CAN

3-Caffeoylquinic acid

F1 506.60 761.37 493.38 305.08

F2 415.41 473.00 395.68 483.18

F3 224.83 235.04 240.00 269.40

F4 82.47 103.77 100.27 198.35

F5 47.12 57.84 49.85 133.68

Theobromine

F1 38.49 52.43 51.29 23.80

F2 30.66 33.81 40.50 33.10

F3 17.31 17.38 25.46 19.39

F4 6.87 8.42 10.92 14.68

F5 4.25 5.11 5.52 10.51

4-Caffeoylquinic acid

F1 166.85 174.31 150.84 91.04

F2 150.77 116.46 135.08 160.69

F3 83.72 62.11 85.84 87.45

F4 31.25 29.25 36.24 64.17

F5 17.90 17.04 19.42 42.76

5-Caffeoylquinic acid F1 284.63 261.07 275.84 136.81

F2 263.42 184.65 246.48 235.69

50

Compound Fraction BC-Ca BC Pr PV CAN

F3 150.05 98.18 159.57 130.75

F4 56.75 46.75 67.97 96.41

F5 34.18 27.16 36.61 64.82

Caffeine

F1 203.95 425.62 261.35 144.13

F2 163.09 267.67 209.76 206.73

F3 92.96 144.85 135.23 124.15

F4 37.30 72.28 61.70 97.79

F5 23.88 45.77 34.18 70.90

51

Appendix 3.25: Comparison of standard deviations using a F-test (Miller et al., 2018) to evaluate the relationship between

the elemental levels of yerba mate leaves (based on age – new and old) for non-commercial samples

collected from traditional plantations cultivated either using NPK fertilisers or non-chemical (organic).

Fertiliser Organic DFn, DFd 3,3 6,6

Fcrit 9.28 4.28

Fcalc Level of significance Fcalc Level of

significance Mg 4.11 ns 4.28 ns Ca 3.56 ns 6.97 * Mn 3.14 ns 8.29 * Fe 2.40 ns 3.56 ns Cu 21.4 * 1.40 ns Zn 19.7 * 1.15 ns

n is the number of samples; Fcalc is the calculated value (refer to section 2.6.2): Fcrit is the critical value obtained for n-1 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant at probability p<0.05.

52


the elemental levels of yerba mate leaves based on the use or non-use (organic) of NPK fertilsers during


New leaves Old leaves DFn, DFd 3,6 6,3

Fcrit 4.76 8.94


significance Mg 6.63 ns 3.80 ns Ca 1.98 ns 3.87 ns Mn 12.7 * 4.65 ns Fe 4.58 ns 3.09 ns Cu 1.30 ns 20.0 * Zn 19.5 * 1.14 ns


53


the elemental levels of yerba mate leaves (new and old) grown in traditional organic or native forest


New leaves Old leaves DFn, DFd 2,6 6,2

Fcrit 5.14 19.3


significance Mg 5.11 ns 2.01 ns Ca 11.4 * 1.21 ns Mn 5.08 ns 1.05 ns Fe 2.28 ns 37.9 *** Cu 1.07 ns 1.14 ns Zn 1.27 ns 22.7 *

n is the number of samples; Fcalc is the calculated value (refer to section 2.6.2): Fcrit is the critical value obtained for n-1 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant at probability p<0.05; and *** very highly significant p<0.001.

54


the origin (Brazil and Argentina); packaging (loose and tea bags) and roasting (green and roasted) of

commercial yerba mate samples (refer to Table 3.9).

Origin (Brazil and Argentina)


Roasting (Brazil)

DFn, DFd 14,6 18,6 14,2 Fcrit 2.88 2.66 3.74


significance Fcalc Level of significance

Mg 1.83 ns 2.13 ns 1.26 ns Ca 1.71 ns 1.49 ns 3.71 ns Mn 1.01 ns 2.60 ns 1.08 ns Fe 1.06 ns 7.52 * 1.04 ns Cu 2.24 ns 2.15 ns 2.23 ns Zn 1.29 ns 1.97 ns 11.1 **

n is the number of samples; Fcalc is the calculated value (refer to section 2.6.2): Fcrit is the critical value obtained for n-1 degrees of freedom, p is the level of probability; ns is that there is no statistically significant difference at p>0.05; * statistically significant at probability p<0.05; ** highly significant at probability p<0.01.

55


the origin (Brazil and Argentina); packaging (loose and tea bags) and roasting process (green loose and

roasted) of regular infusions of commercial yerba mate (refer to Table 3.11).

Origin (Brazil and Argentina)


Roasting (Brazil)

DFn, DFd 14,6 18,6 14,2 Fcrit 2.88 2.66 3.74


significance Fcalc Level of significance

Mn 1.51 ns 2.28 ns 3.15 ns Fe 1.92 ns 8.05 * 3.97 ns Cu 1.03 ns 2.26 ns 7.62 * Zn 1.39 ns 2.18 ns 1.07 ns


56

Appendix 4.1: Sample list of Brazilian coffee samples from Amparo, São Paulo State.

Code Variety Roasting time (min) Origin OB-FL-t0 Obatã 0 Fazenda Flor OB-FL-t2 Obatã 2 Fazenda Flor OB-FL-t4 Obatã 4 Fazenda Flor OB-FL-t6 Obatã 6 Fazenda Flor OB-FL-t8 Obatã 8 Fazenda Flor OB-FL-t10 Obatã 10 Fazenda Flor OB-FL-DE Obatã 10 (defected bean) Fazenda Flor CA-FL-DE Catuaí 10 (defected bean) Fazenda Flor CA-PA-t0 Catuaí 0 Fazenda Palmares CA-PA-t2 Catuaí 2 Fazenda Palmares CA-PA-t4 Catuaí 4 Fazenda Palmares CA-PA-t6 Catuaí 6 Fazenda Palmares CA-PA-t8 Catuaí 8 Fazenda Palmares CA-PA-t10 Catuaí 10 Fazenda Palmares CA-PA-t10def Catuaí 10 (defected bean) Fazenda Palmares BA-PA-t0 Bourbon Amarelo 0 Fazenda Palmares BA-PA-t2 Bourbon Amarelo 2 Fazenda Palmares BA-PA-t4 Bourbon Amarelo 4 Fazenda Palmares BA-PA-t6 Bourbon Amarelo 6 Fazenda Palmares BA-PA-t8 Bourbon Amarelo 8 Fazenda Palmares BA-PA-t10 Bourbon Amarelo 10 Fazenda Palmares

57

Code Variety Roasting time (min) Origin BA-PA-t10def Bourbon Amarelo 10 (defected bean) Fazenda Palmares

* def is defected beans.

Appendix 4.2: Na, Mg, K, Ca, V and Cr levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry weight)



OB-FL-t0 35 ± 5 1827 ± 38 14937 ± 278 1025 ± 65 0.03 ± 0.01 0.59 ± 0.02

OB-FL-t2 94 ± 27 1925 ± 18 14145 ± 61 1262 ± 179 0.05 ± 0.01 0.96 ± 0.17

OB-FL-t4 62 ± 22 1734 ± 20 15232 ± 923 789 ± 66 0.04 ± 0.02 0.54 ± 0.09

OB-FL-t6 92 ± 30 1760 ± 5 14937 ± 38 1011 ± 24 0.04 ± 0.02 0.57 ± 0.12

OB-FL-t8 90 ± 1 1832 ± 13 15886 ± 192 779 ± 47 0.01 ± 0.002 0.46 ± 0.11

OB-FL-t10 96 ± 21 2012 ± 94 16761 ± 884 1017 ± 11 0.02 ± 0.01 0.59 ± 0.02

OB-FL-DE 118 ± 36 1698 ± 15 14193 ± 100 1140 ± 45 0.02 ± 0.01 0.51 ± 0.04

CA-FL-DE 100 ± 7 1791 ± 64 15097 ± 416 1310 ± 88 0.01 ± 0.003 0.60 ± 0.02

CA-PA-t0 85 ± 3 1715 ± 10 16655 ± 137 1262 ± 41 0.01 ± 0.004 0.56 ± 0.25

CA-PA-t2 99 ± 31 1937 ± 24 17141 ± 47 1776 ± 113 0.01 ± 0.003 0.50 ± 0.01

CA-PA-t4 133 ± 48 1972 ± 34 18099 ± 26 1409 ± 121 0.03 ± 0.01 0.41 ± 0.02

CA-PA-t6 106 ± 23 1940 ± 53 18597 ± 260 1512 ± 59 0.04 ± 0.01 0.33 ± 0.02

CA-PA-t8 105 ± 43 1847 ± 18 17874 ± 164 1324 ± 10 0.05 ± 0.01 0.27 ± 0.02

58


CA-PA-t10 177 ± 117 2009 ± 0 18366 ± 272 1402 ± 9 0.04 ± 0.01 0.19 ± 0.03

CA-PA-t10def 182 ± 69 2257 ± 33 20877 ± 276 1552 ± 64 0.05 ± 0.04 0.35 ± 0.19

BA-PA-t0 213 ± 119 2282 ± 68 21415 ± 93 1628 ± 11 0.05 ± 0.03 0.30 ± 0.01

BA-PA-t2 135 ± 87 1842 ± 23 18562 ± 460 1653 ± 18 0.06 ± 0.03 0.31 ± 0.03

BA-PA-t4 155 ± 80 1915 ± 4 18130 ± 511 1429 ± 164 0.09 ± 0.04 0.35 ± 0.03

BA-PA-t6 243 ± 30 1925 ± 29 16538 ± 426 1522 ± 17 0.05 ± 0.005 0.31 ± 0.05

BA-PA-t8 246 ± 17 1908 ± 79 19791 ± 415 1255 ± 78 0.06 ± 0.01 0.28 ± 0.04

BA-PA-t10 236 ± 11 1888 ± 10 18670 ± 267 1517 ± 18 0.06 ± 0.01 0.39 ± 0.03

BA-PA-t10def 235 ± 204 2142 ± 22 19548 ± 320 1620 ± 56 0.07 ± 0.07 0.54 ± 0.22 * def is defected beans.

59

Appendix 4.3: Mn, Fe, Co, Ni, Cu and Zn levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry



OB-FL-t0 31 ± 2 27.88 ± 0.79 0.46 ± 0.02 1.11 ± 0.13 14.21 ± 0.07 10.65 ± 5.3

OB-FL-t2 39 ± 5 38.08 ± 5.34 1.87 ± 0.54 1.52 ± 0.24 13.82 ± 0.07 7.04 ± 1.97

OB-FL-t4 29 ± 2 25.94 ± 1.17 0.43 ± 0.07 0.92 ± 0.1 13.08 ± 0.25 8 ± 2.59

OB-FL-t6 34 ± 2 24.47 ± 0.79 0.31 ± 0.04 0.89 ± 0.1 13.21 ± 0.16 8.42 ± 1.51

OB-FL-t8 31 ± 1 30.19 ± 0.1 0.33 ± 0.01 1.71 ± 0.32 16.36 ± 0.12 8.56 ± 0.05

OB-FL-t10 31 ± 1 30.41 ± 1.16 0.37 ± 0.03 0.91 ± 0.14 14.43 ± 0.82 6.21 ± 0.24

OB-FL-DE 31 ± 2 34.05 ± 6.43 0.19 ± 0 0.92 ± 0.37 14.14 ± 1.43 5.95 ± 0.25

CA-FL-DE 36 ± 0.8 25.34 ± 0.68 0.22 ± 0.01 0.81 ± 0.06 14.74 ± 1.9 6.61 ± 0.07

CA-PA-t0 44 ± 2 23.5 ± 1.09 0.3 ± 0.01 0.78 ± 0.21 15.16 ± 0.79 6.42 ± 1.17

CA-PA-t2 27 ± 1 23.04 ± 0.15 0.37 ± 0.02 0.97 ± 0.2 12.65 ± 0.06 6.21 ± 0.44

CA-PA-t4 25 ± 1 22.06 ± 0.52 0.39 ± 0.01 1.26 ± 0.21 13.01 ± 0.05 6.88 ± 1.35

CA-PA-t6 27 ± 1 22.02 ± 0.19 0.27 ± 0.03 1.13 ± 0.23 12.17 ± 0.55 7.29 ± 2

CA-PA-t8 25 ± 0.6 22.11 ± 0.6 0.42 ± 0.01 0.96 ± 0.03 11.67 ± 0.04 11.77 ± 2.84

CA-PA-t10 32 ± 0.2 26.81 ± 6.96 0.38 ± 0 1.04 ± 0.05 13.27 ± 0.76 8.83 ± 0.31

CA-PA-t10def 26 ± 0.7 27.71 ± 0.46 0.44 ± 0.01 0.99 ± 0.09 13.61 ± 0.1 7.56 ± 0.41

BA-PA-t0 23 ± 1 29 ± 6.22 0.44 ± 0.01 0.98 ± 0.2 14.74 ± 1.14 7.83 ± 0.19

BA-PA-t2 18 ± 0.4 24.88 ± 4.86 0.28 ± 0.01 0.74 ± 0.19 15.23 ± 2.36 7.88 ± 0.29

60


BA-PA-t4 26 ± 0.3 24.69 ± 4.3 0.23 ± 0 0.61 ± 0.25 14.14 ± 3.11 7.99 ± 1.87

BA-PA-t6 23 ± 0.4 26.67 ± 0.57 0.41 ± 0.02 0.76 ± 0.05 17.96 ± 0.86 6.59 ± 0.56

BA-PA-t8 20 ± 1 27.12 ± 1.02 0.24 ± 0.01 0.37 ± 0.05 17.87 ± 0.15 6.09 ± 0.48

BA-PA-t10 24 ± 0.02 25.45 ± 0.46 0.44 ± 0.01 0.54 ± 0.01 17.48 ± 0.48 5.68 ± 0.33

BA-PA-t10def 27 ± 1 29.95 ± 7.15 0.32 ± 0.01 0.96 ± 0.19 16.47 ± 3.78 10.56 ± 2.3 * def is defected beans.

61

Appendix 4.4: As, Se, Mo, Cd and Pb levels (mean ± standard deviation) of Brazilian coffee samples (mg/kg dry weight)


Code As Se Mo Cd Pb

OB-FL-t0 <LOD* <LOD* 0.09 ± 0.03 <LOD* 0.05 ± 0.01

OB-FL-t2 0.01 ± 0.002 <LOD* 0.09 ± 0.02 <LOD* 0.04 ± 0.01

OB-FL-t4 <LOD* <LOD* 0.06 ± 0.01 0.01 ± 0.001 0.01 ± 0.01

OB-FL-t6 <LOD* <LOD* 0.04 ± 0.002 <LOD* <LOD*

OB-FL-t8 <LOD* <LOD* 0.04 ± 0.02 <LOD* <LOD*

OB-FL-t10 0.01 ± 0.001 <LOD* 0.29 ± 0.07 <LOD* 0.06 ± 0.01

OB-FL-DE <LOD* 0.01 ± 0.001 0.07 ± 0.002 <LOD* 0.06 ± 0.01

CA-FL-DE 0.01 ± 0.002 <LOD* 0.1 ± 0.002 0.03 ± 0.001 0.01 ± 0.001

CA-PA-t0 <LOD* <LOD* 0.11 ± 0.01 0.04 ± 0.01 0.07 ± 0.02

CA-PA-t2 0.01 ± 0.002 <LOD* 0.16 ± 0.04 0.06 ± 0.001 0.04 ± 0.01

CA-PA-t4 <LOD* <LOD* 0.11 ± 0.02 <LOD* 0.08 ± 0.02

CA-PA-t6 0.01 ± 0.002 <LOD* 0.05 ± 0.001 <LOD* 0.06 ± 0.001

CA-PA-t8 0.01 ± 0.003 <LOD* 0.09 ± 0.02 <LOD* 0.06 ± 0.002

CA-PA-t10 0.01 ± 0.001 <LOD* 0.05 ± 0.01 <LOD* 0.01 ± 0.001

CA-PA-t10def 0.01 ± 0.002 <LOD* 0.07 ± 0.02 0.03 ± 0.001 0.04 ± 0.01

BA-PA-t0 0.01 ± 0.001 0.01 ± 0.001 0.12 ± 0.02 <LOD* 0.05 ± 0.009

BA-PA-t2 0.01 ± 0.002 <LOD* 0.14 ± 0.02 0.01 ± 0.001 0.07 ± 0.005

62

Code As Se Mo Cd Pb

BA-PA-t4 0.02 ± 0.02 <LOD* 0.09 ± 0.02 <LOD* 0.03 ± 0.006

BA-PA-t6 0.02 ± 0.001 <LOD* 0.09 ± 0.01 <LOD* 0.08 ± 0.002

BA-PA-t8 0.02 ± 0.001 <LOD* 0.09 ± 0.01 <LOD* 0.01 ± 0.002

BA-PA-t10 0.02 ± 0.01 <LOD* 0.09 ± 0.02 <LOD* 0.04 ± 0.008

BA-PA-t10def 0.03 ± 0.002 <LOD* 0.15 ± 0.04 0.01 ± 0.001 0.03 ± 0.006 *refer to Table 2.2; def is defected beans; <LOD is below the limit of detection.

63

Appendix 5.1: Total elemental levels (mean ± standard deviation) of açaí pulp samples (mg/kg dry weight) determined

using ICP-MS (refer to section 2.3). Data relates to the type of sample (non-commercial : purple n= 6; and

white n= 4; and commercial: purple n= 4; n is the number of samples).

Non-commercial Commercial

Purple Açaí whole Purple Açaí de-fatted White Açaí whole White Açaí de-fatted Pulp SP Powder SP Powder UK

Na 238.74 ± 12.24 295.14 ± 9.43 493.38 ± 35.04 725.26 ± 38.25 894.91 ± 11.7 449.25 ± 43.56 107.53 ± 4.85

Mg 2332.29 ± 21.62 2723.64 ± 41.5 2469.16 ± 121.25 3017.68 ± 90.03 2022.1 ± 27.53 2036.86 ± 21.41 859.21 ± 11.3

K 12118.41 ± 1548.83 13529.26 ± 1024.94 11941.46 ± 78.7 14737.9 ± 443.56 8219.66 ± 335.61 8321.86 ± 297.01 2382.14 ± 0.86

Ca 4688.24 ± 141.5 5273.38 ± 580.66 4162.31 ± 22.03 5251.08 ± 132.77 1650.41 ± 161.18 2029.33 ± 172.47 661.73 ± 156.47

V 0.02 ± 0.002 0.03 ± 0.01 0.03 ± 0.005 0.04 ± 0.003 0.05 ± 0.02 0.02 ± 0.004 0.01 ± 0.002

Cr 4.84 ± 0.35 5.22 ± 0.09 5.26 ± 0.002 7.15 ± 0.94 4.36 ± 0.15 4.08 ± 0.61 2.55 ± 0.13

Mn 640.63 ± 9.33 809.18 ± 7.67 611.38 ± 5.98 808.86 ± 15.69 267.62 ± 15.07 547.04 ± 17.78 16.23 ± 0.06

Fe 30.09 ± 0.04 41.74 ± 1.18 36.54 ± 7.48 43.04 ± 0.96 30.01 ± 0.44 21.92 ± 0.28 2.29 ± 0.14

64



Co 0.08 ± 0.003 0.10 ± 0.004 0.08 ± 0.002 0.11 ± 0.002 0.10 ± 0.004 0.08 ± 0.005 0.01 ± 0.002

Ni 1.71 ± 0.22 1.89 ± 0.09 1.76 ± 0.02 2.63 ± 0.17 1.96 ± 0.02 1.69 ± 0.05 0.70 ± 0.04

Cu 18.11 ± 0.26 22.22 ± 0.09 17.16 ± 0.65 24.96 ± 0.97 15.23 ± 0.52 14.97 ± 0.90 3.51 ± 0.17

Zn 24.93 ± 0.74 32.40 ± 1.68 26.49 ± 0.26 36.18 ± 1.67 26.98 ± 6.47 23.23 ± 3.10 12.70 ± 1.18

As <LOD* <LOD* <LOD* <LOD* <LOD* <LOD* <LOD*

Se <LOD* <LOD* <LOD* 0.05 ± 0.02 0.06 ± 0.004 <LOD* <LOD*

Mo 0.09 ± 0.00 0.13 ± 0.01 0.07 ± 0.003 0.10 ± 0.005 0.22 ± 0.003 0.04 ± 0.005 0.18 ± 0.007

Cd 0.07 ± 0.003 0.10 ± 0.01 0.09 ± 0.005 0.18 ± 0.01 0.04 ± 0.01 0.06 ± 0.002 0.01 ± 0.005

Pb 0.52 ± 0.04 0.72 ± 0.01 0.31 ± 0.02 0.46 ± 0.02 0.07 ± 0.01 0.03 ± 0.003 0.02 ± 0.003


65

Appendix 5.2: Total elemental levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh weight) determined

using ICP-MS (refer to section 2.3). Data relates to the type of sample (non-commercial : purple n= 6; and

white n= 4; and commercial: purple n= 4; n is the number of samples).



Na 23.87 ± 1.22 29.51 ± 0.94 49.34 ± 3.50 72.53 ± 3.82 89.49 ± 1.17 44.93 ± 4.36 10.75 ± 0.48

Mg 233.23 ± 2.16 272.36 ± 4.15 246.92 ± 12.13 301.77 ± 9.00 202.21 ± 2.75 203.69 ± 2.14 85.92 ± 1.13

K 1211.84 ± 154.88 1352.93 ± 102.49 1194.15 ± 7.87 1473.79 ± 44.36 821.97 ± 33.56 832.19 ± 29.70 238.21 ± 0.09

Ca 468.82 ± 14.15 527.34 ± 58.07 416.23 ± 2.20 525.11 ± 13.28 165.04 ± 16.12 202.93 ± 17.25 66.17 ± 15.65

V 0.002 ± 0.002 0.003 ± 0.001 0.003 ± 0.0001 0.004 ± 0.0002 0.005 ± 0.020 0.002 ± 0.0002 0.001 ± 0.0001

Cr 0.48 ± 0.04 0.52 ± 0.01 0.53 ± 0.002 0.71 ± 0.09 0.44 ± 0.01 0.41 ± 0.06 0.25 ± 0.01

Mn 64.06 ± 0.93 80.92 ± 0.77 61.14 ± 0.60 80.89 ± 1.57 26.76 ± 1.51 54.70 ± 1.78 1.62 ± 0.01

Fe 3.01 ± 0.002 4.17 ± 0.12 3.65 ± 0.75 4.30 ± 0.10 3.00 ± 0.04 2.19 ± 0.03 0.23 ± 0.01

Co 0.01 ± 0.003 0.01 ± 0.002 0.01 ± 0.003 0.01 ± 0.004 0.01 ± 0.002 0.01 ± 0.005 0.001 ± 0.0002

66

Non-commercial Commercial Purple Açaí whole Purple Açaí de-fatted White Açaí whole White Açaí de-fatted Pulp SP Powder SP Powder UK

Ni 0.17 ± 0.02 0.19 ± 0.01 0.18 ± 0.004 0.26 ± 0.02 0.20 ± 0.003 0.17 ± 0.002 0.07 ± 0.006

Cu 1.81 ± 0.03 2.22 ± 0.01 1.72 ± 0.06 2.50 ± 0.10 1.52 ± 0.05 1.50 ± 0.09 0.35 ± 0.02

Zn 2.49 ± 0.07 3.24 ± 0.17 2.65 ± 0.03 3.62 ± 0.17 2.70 ± 0.65 2.32 ± 0.31 1.27 ± 0.12

As <LOD* <LOD* <LOD* <LOD* <LOD* <LOD* <LOD*

Se <LOD* <LOD* <LOD* 0.005 ± 0.002 0.006 ± 0.0003 <LOD* <LOD*

Mo 0.009 ± 0.0002 0.013 ± 0.001 0.007 ± 0.0002 0.010 ± 0.0001 0.022 ± 0.0003 0.004 ± 0.0003 0.018 ± 0.0002

Cd 0.007 ± 0.0003 0.010 ± 0.001 0.009 ± 0.0001 0.018 ± 0.001 0.004 ± 0.001 0.006 ± 0.0002 0.001 ± 0.0003

Pb 0.05 ± 0.001 0.07 ± 0.002 0.03 ± 0.003 0.05 ± 0.004 0.01 ± 0.001 0.003 ± 0.000 0.002 ± 0.0003


67

Appendix 5.3: Sample list for the evaluation of the Amazon geographical variability and industrial processing on açaí.

Code Origin Variety Type Moisture content (%) Processing place GE-WB-P Genipauba White bacapa Seed 46.4 Lab GE-WB-S Genipauba White bacapa Fruit 27.9 Lab GE-PB-P Genipauba Purple bacapa Seed 39.4 Lab GE-PB-S Genipauba Purple bacapa Fruit 32.2 Lab IL-PA-P Ilhas Purple Acai Seed 47.5 Lab IL-PA-S Ilhas Purple Acai Fruit 29.6 Lab

MA-PA-P Macapa Purple Acai Seed 54.9 Lab MA-PA-S Macapa Purple Acai Fruit 30.6 Lab AN-PA-P Anajas Purple Acai Seed 50.2 Lab AN-PA-S Anajas Purple Acai Fruit 31.7 Lab IC-PA-S Ilhas Purple Acai Fruit 33.2 Point do Açaí IM-PA-S Igarape-Miri Purple Acai Fruit 35.5 Açaí Santa Helena

PA-IC-PM Ilhas Purple Acai Pulp (medium) 89.0 Point do Açaí SH-AB-PF Abaetetuba Purple Acai Pulp (fluid) 92.0 Açaí Santa Helena SH-IM-PF Igarape-Miri Purple Acai Pulp (fluid) 92.0 Açaí Santa Helena SH-IM-PM Igarape-Miri Purple Acai Pulp (medium) 89.0 Açaí Santa Helena SH-PA-PE Paragominas Purple Acai Pulp (thick) 86.0 Açaí Santa Helena AA-OB-PM Obidos Purple Acai Pulp (medium) 89.0 Açaí Amazona AA-OB-PE Obidos Purple Acai Pulp (thick) 86.0 Açaí Amazona AA-OB-FD Obidos Purple Acai Freeze-dryed 86.0 Açaí Amazona

68

Appendix 5.4: Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg, dry

weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the

number of replicates).

Code Total polyphenol Na Mg K Ca V Mn Fe Cu Zn

GE-WB-P 9.45 ± 1.23 63.41 ± 32.88

249.84 ± 47.31

3719.68 ± 796.28

133.66 ± 12.12

0.04 ± 0.002

13.10 ± 1.72 8.97 ± 0.57 8.44 ± 1.38 9.61 ± 1.66

GE-WB-S 18.91 ± 1.02

210.13 ± 42.18

272.86 ± 69.82

4001.92 ± 288.76

188.48 ± 77.70 0.06 ± 0.01 12.96 ±

1.04 10.01 ±

0.28 8.80 ± 1.04 10.44 ± 2.26

GE-PB-P 5.47 ± 0.36 353.68 ± 30.94

498.29 ± 12.26

2509.03 ± 38.51

262.17 ± 4.73 0.06 ± 0.02 9.46 ± 0.28 14.78 ±

0.27 9.72 ± 0.15 9.82 ± 0.87

GE-PB-S 20.2 ± 2.17 530.63 ± 42.21

493.08 ± 14.75

2359.96 ± 113.75

233.36 ± 33.83 0.04 ± 0.02 9.36 ± 0.02 12.58 ±

0.40 6.30 ± 0.07 9.34 ± 0.85

IL-PA-P 6.88 ± 1.11 624.63 ± 44.49

654.22 ± 39.89

3528.97 ± 179.01

1282.02 ± 85.48

0.02 ± 0.003

104.99 ± 5.52

12.04 ± 0.72 9.45 ± 0.64 11.30 ±

1.45

IL-PA-S 57.57 ± 1.19

779.00 ± 27.17

529.64 ± 4.75

3284.90 ± 43.01

554.23 ± 58.71 0.05 ± 0.01 128.03 ±

5.05 10.78 ±

0.38 9.91 ± 0.02 12.93 ± 0.27

MA-PA-P 9.94 ± 0.76 891.39 ± 17.66

904.23 ± 74.00

4298.91 ± 315.64

1883.82 ± 307.40

0.01 ± 0.002

186.33 ± 12.15

12.76 ± 0.39 9.78 ± 0.53 12.20 ±

0.88

MA-PA-S 31.99 ± 0.47

1075.45 ± 21.88

494.98 ± 13.56

2573.05 ± 20.49

692.58 ± 39.44 0.02 ± 0.01 224.98 ±

21.55 10.37 ±

0.01 7.73 ± 0.19 10.96 ± 0.69

AN-PA-P 7.25 ± 0.36 1191.57 ± 37.24

501.26 ± 7.35

2630.92 ± 55.04

632.37 ± 124.02 0.01 ± 0.01 154.01 ±

26.27 8.49 ± 0.43 7.95 ± 0.09 9.20 ± 0.45

AN-PA-S 36.66 ± 10.47

1355.07 ± 77.08

419.17 ± 2.96

2966.18 ± 89.33

318.60 ± 22.55 0.02 ± 0.01 256.51 ±

5.89 8.47 ± 0.08 7.58 ± 0.19 10.79 ± 1.33

69


IC-PA-S 58.83 ± 2.18

1456.94 ± 72.83

445.08 ± 12.38

3016.99 ± 9.24

567.01 ± 204.00 0.01 ± 0.01 158.72 ±

14.45 9.74 ± 0.78 6.89 ± 0.42 7.95 ± 0.79

IM-PA-S 28.80 ± 4.39

1588.14 ± 18.34

483.49 ± 10.57

3209.86 ± 48.38

612.29 ± 42.31 0.04 ± 0.02 70.11 ±

2.37 71.19 ±

0.87 10.83 ±

0.23 11.61 ±

0.26

PA-IC-PM 26.60 ± 3.53

1611.51 ± 71.58

1693.77 ± 33.30

10870.64 ± 23.55

2513.16 ± 195.77 0.05 ± 0.01 751.58 ±

11.07 35.04 ±

0.52 13.77 ±

0.51 23.05 ±

0.10

SH-AB-PF 20.40 ± 3.92

1878.26 ± 24.56

1851.27 ± 42.43

10714.62 ± 252.60

2234.51 ± 43.97 0.06 ± 0.02 393.27 ±

13.01 28.56 ±

0.20 17.05 ±

0.10 24.27 ±

0.80

SH-IM-PF 17.15 ± 0.81

2031.44 ± 43.47

1795.99 ± 39.52

11688.08 ± 1541.24

3023.21 ± 131.22 0.05 ± 0.02 479.92 ±

303.87 30.56 ±

0.54 14.88 ±

0.24 30.78 ±

2.47

SH-IM-PM 15.63 ± 0.93

2128.45 ± 50.69

1929.36 ± 7.67

12275.75 ± 189.99

3661.91 ± 192.34 0.06 ± 0.02 845.72 ±

4.58 30.24 ±

0.90 15.71 ±

0.73 28.31 ±

0.22

SH-PA-PE 25.43 ± 1.83

2376.19 ± 14.22

1896.2 ± 34.91

10186.08 ± 208.59

2875.25 ± 44.28

0.04 ± 0.002

181.49 ± 0.62

33.78 ± 2.23

18.49 ± 0.47

24.78 ± 0.31

AA-OB-PM

18.97 ± 2.03

2471.43 ± 44.63

1558.81 ± 45.29

13554.15 ± 123.86

3994.21 ± 48.99 0.09 ± 0.02 47.84 ±

2.36 39.59 ±

4.62 14.41 ±

0.66 17.54 ±

0.71

AA-OB-PE 19.78 ± 0.39

2624.86 ± 57.95

1543.10 ± 5.00

14804.64 ± 16.44

4019.14 ± 243.10

0.04 ± 0.003

107.22 ± 0.23

34.98 ± 0.22

11.96 ± 0.08

15.38 ± 0.14

AA-OB-FD 22.19 ± 2.08

2673.07 ± 52.16

1758.29 ± 48.36

13756.60 ± 370.32

3753.20 ± 201.92

0.07 ± 0.002

333.82 ± 7.95

40.82 ± 0.62

17.42 ± 0.43

23.17 ± 1.06

70

Appendix 5.5: Total elemental (mean ± standard deviation) of açaí pulp samples (mg/kg dry weight) determined using

ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the number of replicates).

Code Cr Co Ni As Se Mo Cd Pb GE-WB-P 1.07 ± 0.47 <LOD* 0.95 ± 0.39 <0.01 0.03 ± 0.01 <LOD* 0.03 ± 0.01 0.07 ± 0.01 GE-WB-S 1.49 ± 0.47 <LOD* 0.89 ± 0.26 <0.01 0.03 ± 0.003 <LOD* 0.20 ± 0.06 0.07 ± 0.01 GE-PB-P 0.73 ± 0.14 <LOD* 0.66 ± 0.09 <0.01 0.02 ± 0.01 <LOD* 0.06 ± 0.04 <LOD* GE-PB-S 1.32 ± 0.13 <LOD* 0.62 ± 0.04 <LOD* 0.01 ± 0.01 <LOD* 0.09 ± 0.05 <LOD* IL-PA-P 0.84 ± 0.10 <LOD* 1.07 ± 0.24 <LOD* 0.02 ± 0.002 <LOD* 0.17 ± 0.002 0.13 ± 0.01 IL-PA-S 1.61 ± 0.05 <LOD* 1.36 ± 0.31 <LOD* 0.01 ± 0.01 0.23 ± 0.08 0.12 ± 0.01 0.11 ± 0.02 MA-PA-P 1.07 ± 0.01 <LOD* 1.38 ± 0.14 0.01 ± 0.003 0.02 ± 0.01 <LOD* 0.05 ± 0.01 <LOD* MA-PA-S 2.05 ± 0.32 <LOD* 1.62 ± 0.29 <LOD* 0.01 ± 0.003 <LOD* 0.03 ± 0.01 <LOD* AN-PA-P 0.97 ± 0.23 <LOD* 1.32 ± 0.11 <LOD* 0.02 ± 0.002 <LOD* 0.03 ± 0.004 <LOD* AN-PA-S 1.31 ± 0.07 <LOD* 1.79 ± 0.26 <LOD* 0.02 ± 0.002 <LOD* 0.02 ± 0.002 0.04 ± 0.01 IC-PA-S 1.36 ± 0.01 <LOD* 1.61 ± 0.33 0.01 ± 0.003 0.01 ± 0.002 <LOD* 0.04 ± 0.01 0.21 ± 0.05 IM-PA-S 1.65 ± 0.03 <LOD* 1.42 ± 0.04 <LOD* 0.07 ± 0.01 <LOD* 0.01 ± 0.003 <LOD* PA-IC-PM 3.91 ± 0.11 0.19 ± 0.003 2.83 ± 0.04 0.02 ± 0.002 0.02 ± 0.02 <LOD* 0.12 ± 0.04 <LOD* SH-AB-PF 2.12 ± 0.01 0.12 ± 0.002 1.83 ± 0.12 0.01 ± 0.003 0.02 ± 0.01 <LOD* 0.05 ± 0.02 <LOD* SH-IM-PF 2.43 ± 0.10 0.10 ± 0.003 2.15 ± 0.34 0.01 ± 0.001 0.02 ± 0.005 <LOD* 0.10 ± 0.01 <LOD* SH-IM-PM 2.17 ± 0.01 0.10 ± 0.001 1.92 ± 0.14 0.01 ± 0.005 0.02 ± 0.003 <LOD* 0.14 ± 0.07 <LOD* SH-PA-PE 2.18 ± 0.03 <LOD* 1.44 ± 0.12 <LOD* 0.07 ± 0.01 0.10 ± 0.00 0.01 ± 0.002 <LOD* AA-OB-PM 6.17 ± 0.14 <LOD* 1.80 ± 0.14 0.01 ± 0.004 0.02 ± 0.01 0.38 ± 0.05 0.03 ± 0.001 0.07 ± 0.01 AA-OB-PE 6.29 ± 0.002 <LOD* 1.89 ± 0.10 0.01 ± 0.003 0.02 ± 0.001 0.40 ± 0.01 0.03 ± 0.002 <LOD* AA-OB-FD 4.80 ± 0.13 <LOD* 2.58 ± 0.15 0.01 ± 0.005 0.02 ± 0.002 0.28 ± 0.03 0.07 ± 0.003 <LOD*


71

Appendix 5.6: Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh

weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is the

number of replicates).


GE-WB-P 5.07 ± 0.66 34 00± 17.63

133.94 ± 25.36

1994.19 ± 426.90

71.66 ± 6.50

0.02 ± 0.002 7.02 ± 0.92 4.81 ± 0.30 4.52 ± 0.74 5.15 ± 0.89

GE-WB-S 13.64 ± 0.74

151.59 ± 30.43

196.84 ± 50.37

2886.96 ± 208.31

135.97 ± 56.05 0.04 ± 0.01 9.35 ± 0.75 7.22 ± 0.20 6.35 ± 0.75 7.53 ± 1.63

GE-PB-P 3.32 ± 0.22 214.40 ± 18.76

302.06 ± 7.43

1520.99 ± 23.34

158.93 ± 2.87 0.03 ± 0.01 5.74 ± 0.17 8.96 ± 0.16 5.89 ± 0.09 5.95 ± 0.53

GE-PB-S 13.69 ± 1.47

359.51 ± 28.60

334.07 ± 9.99

1598.90 ± 77.06

158.10 ± 22.92 0.03 ± 0.01 6.34 ± 0.01 8.53 ± 0.27 4.27 ± 0.04 6.33 ± 0.58

IL-PA-P 3.62 ± 0.58 328.09 ± 23.37

343.63 ± 20.95

1853.59 ± 94.03

673.38 ± 44.90

0.01 ± 0.002

55.15 ± 2.90 6.32 ± 0.38 4.96 ± 0.34 5.94 ± 0.76

IL-PA-S 40.51 ± 0.84

548.12 ± 19.12

372.67 ± 3.34

2311.35 ± 30.26

389.98 ± 41.31 0.03 ± 0.01 90.09 ±

3.55 7.58 ± 0.27 6.97 ± 0.01 9.10 ± 0.19

MA-PA-P 4.48 ± 0.34 401.60± 7.96

407.38 ± 33.34

1936.80 ± 142.2

848.72 ± 138.49

0.01 ± 0.003

83.95 ± 5.47 5.75 ± 0.17 4.40 ± 0.24 5.50 ± 0.4

MA-PA-S 22.21 ± 0.32

746.51 ± 15.19

343.59 ± 9.41

1786.06 ± 14.23

480.75 ± 27.37 0.02 ± 0.01 156.17 ±

14.96 7.20 ± 0.00 5.36 ± 0.13 7.60 ± 0.48

AN-PA-P 3.61 ± 0.18 593.67 ± 18.56

249.74 ± 3.66

1310.80 ± 27.42

315.07 ± 61.79

0.01 ± 0.004

76.73 ± 13.09 4.23 ± 0.21 3.96 ± 0.05 4.59 ± 0.22

AN-PA-S 25.05 ± 7.15

925.85 ± 52.66

286.40 ± 2.02

2026.63 ± 61.03

217.68 ± 15.41 0.01 ± 0.01 175.26 ±

4.02 5.78 ± 0.06 5.18 ± 0.13 7.37 ± 0.91

72


IC-PA-S 39.28 ± 1.45

972.75 ± 48.62

297.16 ± 8.26

2014.33 ± 6.17

378.57 ± 136.20

0.01 ± 0.002

105.97 ± 9.64 6.50 ± 0.52 4.60 ± 0.28 5.31 ± 0.53

IM-PA-S 18.59 ± 2.83

1024.85 ± 11.83

312.00 ± 6.82

2071.39 ± 31.22

395.12 ± 27.30 0.03 ± 0.01 45.24 ±

1.53 45.94 ±

0.56 6.99 ± 0.15 7.49 ± 0.16

PA-IC-PM 2.93 ± 0.39 177.27 ± 7.87

186.31 ± 3.66

1195.77 ± 2.59

276.45 ± 21.54

0.01 ± 0.002

82.67 ± 1.22 3.85 ± 0.06 1.51 ± 0.06 2.54 ± 0.01

SH-AB-PF 1.63 ± 0.31 150.26 ± 1.96

148.10 ± 3.39

857.17 ± 20.21

178.76 ± 3.52

0.005 ± 0.001

31.46 ± 1.04 2.29 ± 0.02 1.36 ± 0.01 1.94 ± 0.06

SH-IM-PF 1.37 ± 0.06 162.52 ± 3.48

143.68 ± 3.16

935.05 ± 123.30

241.86 ± 10.50

0.004 ± 0.001

38.39 ± 24.31 2.44 ± 0.04 1.19 ± 0.02 2.46 ± 0.20

SH-IM-PM 1.72 ± 0.10 234.13 ± 5.58

212.23 ± 0.84

1350.33 ± 20.90

402.81 ± 21.16

0.01 ± 0.003

93.03 ± 0.50 3.33 ± 0.10 1.73 ± 0.08 3.11 ± 0.02

SH-PA-PE 3.56 ± 0.26 332.67 ± 1.99

265.47 ± 4.89

1426.05 ± 29.20

402.54 ± 6.20

0.005 ± 0.0001

25.41 ± 0.09 4.73 ± 0.31 2.59 ± 0.07 3.47 ± 0.04

AA-OB-PM 2.09 ± 0.22 271.86 ±

4.91 171.47 ±

4.98 1490.96 ±

13.62 439.36 ±

5.39 0.01 ± 0.004 5.26 ± 0.26 4.35 ± 0.51 1.59 ± 0.07 1.93 ± 0.08

AA-OB-PE 2.77 ± 0.05 367.48 ± 8.11

216.03 ± 0.70

2072.65 ± 2.30

562.68 ± 34.03

0.01 ± 0.003

15.01 ± 0.03 4.90 ± 0.03 1.67 ± 0.01 2.15 ± 0.02

AA-OB-FD 3.11 ± 0.29 374.23 ± 7.30

246.16 ± 6.77

1925.92 ± 51.85

525.45 ± 28.27

0.01 ± 0.003

46.74 ± 1.11 5.71 ± 0.09 2.44 ± 0.06 3.24 ± 0.15

73

Appendix 5.7: Total elemental and polyphenol levels (mean ± standard deviation) of açaí pulp samples (mg/kg fresh

weight) determined using ICP-MS (refer to section 2.3). Data relates to the type of sample; n= 3; n is

replicates).

Code Cr Co Ni As Se Mo Cd Pb

GE-WB-P 0.57 ± 0.25 <LOD* 0.51 ± 0.21 <LOD* 0.02 ± 0.002 <LOD* 0.01 ± 0.002 0.04 ± 0.001 GE-WB-S 1.07 ± 0.34 <LOD* 0.64 ± 0.19 <LOD* 0.02 ± 0.002 <LOD* 0.14 ± 0.05 0.05 ± 0.01 GE-PB-P 0.45 ± 0.09 <LOD* 0.40 ± 0.06 <LOD* 0.01 ± 0.01 <LOD* 0.03 ± 0.02 <LOD* GE-PB-S 0.89 ± 0.09 <LOD* 0.42 ± 0.03 <LOD* 0.01 ± 0.001 <LOD* 0.06 ± 0.03 <LOD* IL-PA-P 0.44 ± 0.05 <LOD* 0.56 ± 0.12 <LOD* 0.01 ± 0.001 <LOD* 0.09 ± 0.003 0.07 ± 0.01 IL-PA-S 1.13 ± 0.03 <LOD* 0.96 ± 0.21 <LOD* 0.01 ± 0.002 0.16 ± 0.06 0.09 ± 0.01 0.08 ± 0.01

MA-PA-P 0.48 ± 0.01 <LOD* 0.62 ± 0.06 0.003 ± 0.0002 0.01 ± 0.001 <LOD* 0.02 ± 0.005 <LOD* MA-PA-S 1.42 ± 0.22 <LOD* 1.13 ± 0.20 <LOD* 0.01 ± 0.002 <LOD* 0.02 ± 0.003 <LOD* AN-PA-P 0.48 ± 0.12 <LOD* 0.66 ± 0.05 <LOD* 0.01 ± 0.003 <LOD* 0.01 ± 0.006 <LOD* AN-PA-S 0.89 ± 0.05 <LOD* 1.23 ± 0.17 <LOD* 0.02 ± 0.001 <LOD* 0.01 ± 0.01 0.03 ± 0.01 IC-PA-S 0.91 ± 0.01 <LOD* 1.07 ± 0.22 0.004 ± 0.0004 0.01 ± 0.001 <LOD* 0.02 ± 0.01 0.14 ± 0.03 IM-PA-S 1.06 ± 0.02 <LOD* 0.92 ± 0.02 <LOD* 0.05 ± 0.01 <LOD* 0.01 ± 0.006 <LOD*

PA-IC-PM 0.43 ± 0.01 0.02 ± 0.004 0.31 ± 0.00 0.002 ± 0.0003 0.003 ± 0.002 <LOD* 0.01 ± 0.003 <LOD* SH-AB-PF 0.17 ± 0.003 0.01 ± 0.002 0.15 ± 0.01 0.001 ± 0.0002 0.002 ± 0.0002 <LOD* 0.004 ± 0.002 <LOD* SH-IM-PF 0.19 ± 0.01 0.01 ± 0.004 0.17 ± 0.03 0.001 ± 0.0005 0.001 ± 0.0001 <LOD* 0.01 ± 0.005 <LOD* SH-IM-PM 0.24 ± 0.004 0.01 ± 0.002 0.21 ± 0.02 0.002 ± 0.0002 0.002 ± 0.001 <LOD* 0.02 ± 0.01 <LOD* SH-PA-PE 0.31 ± 0.002 <LOD* 0.20 ± 0.02 <LOD* 0.01 ± 0.001 0.01 ± 0.003 0.002 ± 0.0003 <LOD* AA-OB-PM 0.68 ± 0.02 <LOD* 0.20 ± 0.02 0.001 ± 0.0005 0.003 ± 0.001 0.04 ± 0.01 0.003 ± 0.0002 0.01 ± 0.003

74

Code Cr Co Ni As Se Mo Cd Pb AA-OB-PE 0.88 ± 0.005 <LOD* 0.27 ± 0.01 0.001 ± 0.001 0.003 ± 0.0001 0.06 ± 0.006 0.004 ± 0.0002 <LOD* AA-OB-FD 0.67 ± 0.02 <LOD* 0.36 ± 0.02 0.001 ± 0.002 0.003 ± 0.00002 0.04 ± 0.002 0.01 ± 0.004 <LOD*


Appendix 5.8: Total polyphenol (TP) and minor elements daily intake (mg/day) based on the consumpsion of a 500 g

serving of the commercial and non-commercial açaí pulp (fresh weight).

Purple Açaí

whole Purple Açaí de-fatted White Açaí whole White Açaí de-fatted Pulp SP Commercial SP

Commercial UK

TP 1600.0 1970.0 470.0 585.0 1415.0 2120.0 254.5 Ca 234.4 263.7 208.1 262.6 82.5 101.5 33.1 Mg 116.6 136.2 123.5 150.9 101.1 101.8 43.0 Mn 32.0 40.5 30.6 40.4 13.4 27.4 0.8 Fe 1.5 2.1 1.8 2.2 1.5 1.1 0.1 Zn 1.2 1.6 1.3 1.8 1.3 1.2 0.6 Cu 0.9 1.1 0.9 1.2 0.8 0.7 0.2

75

Appendix 5.9: Percentage intake (%) of total polyphenol and minor elements based on the consumption of a 500 g serving

of the commercial and non-commercial açaí pulp (fresh weight) when compared to the recommended daily

allowance (RDA) for males (M) and females (F).

Total polyphenol Ca Mg Mn Fe Zn Cu Male Female Male Female Male Female Male Female Male Female Male Female Male Female

RDA 1492 1492 1300 1300 400 310 2.3 1.8 8 18 11 8 0.9

GE-WB-P 169.9 169.9 2.8 2.8 16.7 21.6 152.6 195.0 30.1 13.4 23.4 32.2 251.1 GE-WB-S 457.1 457.1 5.2 5.2 24.6 31.7 203.3 259.7 45.1 20.1 34.2 47.1 352.8 GE-PB-P 111.3 111.3 6.1 6.1 37.8 48.7 124.8 159.4 56.0 24.9 27.0 37.2 327.2 GE-PB-S 458.8 458.8 6.1 6.1 41.8 53.9 137.8 176.1 53.3 23.7 28.8 39.6 237.2 IL-PA-P 121.3 121.3 25.9 25.9 43.0 55.4 1198.9 1531.9 39.5 17.6 27.0 37.1 275.6 IL-PA-S 1357.6 1357.6 15.0 15.0 46.6 60.1 1958.5 2502.5 47.4 21.1 41.4 56.9 387.2

MA-PA-P 150.1 150.1 32.6 32.6 50.9 65.7 1825.0 2331.9 35.9 16.0 25.0 34.4 244.4 MA-PA-S 744.3 744.3 18.5 18.5 42.9 55.4 3395.0 4338.1 45.0 20.0 34.5 47.5 297.8 AN-PA-P 121.0 121.0 12.1 12.1 31.2 40.3 1668.0 2131.4 26.4 11.8 20.9 28.7 220.0 AN-PA-S 839.5 839.5 8.4 8.4 35.8 46.2 3810.0 4868.3 36.1 16.1 33.5 46.1 287.8 IC-PA-S 1316.4 1316.4 14.6 14.6 37.1 47.9 2303.7 2943.6 40.6 18.1 24.1 33.2 255.6 IM-PA-S 623.0 623.0 15.2 15.2 39.0 50.3 983.5 1256.7 287.1 127.6 34.0 46.8 388.3

PA-IC-PM 98.2 98.2 10.6 10.6 23.3 30.1 1797.2 2296.4 24.1 10.7 11.5 15.9 83.9 SH-AB-PF 54.6 54.6 6.9 6.9 18.5 23.9 683.9 873.9 14.3 6.4 8.8 12.1 75.6 SH-IM-PF 45.9 45.9 9.3 9.3 18.0 23.2 834.6 1066.4 15.3 6.8 11.2 15.4 66.1

76

Total polyphenol Ca Mg Mn Fe Zn Cu Male Female Male Female Male Female Male Female Male Female Male Female Male Female

RDA 1492 1492 1300 1300 400 310 2.3 1.8 8 18 11 8 0.9

SH-IM-PM 57.6 57.6 15.5 15.5 26.5 34.2 2022.4 2584.2 20.8 9.3 14.1 19.4 96.1 SH-PA-PE 119.3 119.3 15.5 15.5 33.2 42.8 552.4 705.8 29.6 13.1 15.8 21.7 143.9 AA-OB-PM 70.0 70.0 16.9 16.9 21.4 27.7 114.3 146.1 27.2 12.1 8.8 12.1 88.3 AA-OB-PE 92.8 92.8 21.6 21.6 27.0 34.8 326.3 416.9 30.6 13.6 9.8 13.4 92.8 AA-OB-FD 104.2 104.2 20.2 20.2 30.8 39.7 1016.1 1298.3 35.7 15.9 14.7 20.3 135.6

77

Appendix 5.10: Total polyphenol and minor elements daily intake (mg/day) based on a 500 g serving of açaí pulp.

Total polyphenol Ca Mg Mn Fe Zn Cu

GE-WB-P 2535.0 35.8 67.0 3.5 2.4 2.6 2.3 GE-WB-S 6820.0 68.0 98.4 4.7 3.6 3.8 3.2 GE-PB-P 1660.0 79.5 151.0 2.9 4.5 3.0 2.9 GE-PB-S 6845.0 79.1 167.0 3.2 4.3 3.2 2.1 IL-PA-P 1810.0 336.7 171.8 27.6 3.2 3.0 2.5 IL-PA-S 20255.0 195.0 186.3 45.0 3.8 4.6 3.5

MA-PA-P 2240.0 424.4 203.7 42.0 2.9 2.8 2.2 MA-PA-S 11105.0 240.4 171.8 78.1 3.6 3.8 2.7 AN-PA-P 1805.0 157.5 124.9 38.4 2.1 2.3 2.0 AN-PA-S 12525.0 108.8 143.2 87.6 2.9 3.7 2.6 IC-PA-S 19640.0 189.3 148.6 53.0 3.3 2.7 2.3 IM-PA-S 9295.0 197.6 156.0 22.6 23.0 3.7 3.5

PA-IC-PM 1465.0 138.2 93.2 41.3 1.9 1.3 0.8 SH-AB-PF 815.0 89.4 74.1 15.7 1.1 1.0 0.7 SH-IM-PF 685.0 120.9 71.8 19.2 1.2 1.2 0.6 SH-IM-PM 860.0 201.4 106.1 46.5 1.7 1.6 0.9 SH-PA-PE 1780.0 201.3 132.7 12.7 2.4 1.7 1.3 AA-OB-PM 1045.0 219.7 85.7 2.6 2.2 1.0 0.8 AA-OB-PE 1385.0 281.3 108.0 7.5 2.5 1.1 0.8 AA-OB-FD 1555.0 262.7 123.1 23.4 2.9 1.6 1.2

chemical analysis of typical beverages and açaí …epubs.surrey.ac.uk/852962/1/thesis - fernanda...

Documents