the yield and biological activity (lc50) of rotenone
TRANSCRIPT
THE YIELD AND BIOLOGICAL ACTIVITY (LC50) OF ROTENONE
EXTRACTED FROM Derris elliptica
SAIFUL IRWAN BIN ZUBAIRI
A thesis submitted in fulfilment of the
requirements for the award of the degree of
Master of Engineering (Bioprocess)
Faculty of Chemical and Natural Resources Engineering
Universiti Teknologi Malaysia
AUGUST 2006
iii
To my most beloved sayang and mak
iv
ACKNOWLEDGEMENTS
‘In the name of Allah, the most gracious and the most merciful. Selawat and
salam to Prophet Muhammad s.a.w.’ I owe first and foremost my profound gratitude to
almighty Allah s.w.t the source of all inspiration and help and without whose assistance;
this study would not have come into existence.
Deep obligation and indebtedness and most sincere gratitude are offered to my
supervisor Professor Dr. Mohammad Roji Sarmidi for his continuous guidance during
all stages of my research work and for his willingness to help. Without his continue
support, interest, wisdom and idea during our discussion, this thesis would not have been
implemented and executed well.
I would also like to acknowledge the Chemical Engineering Pilot Plant (CEPP)
staff, in particular Professor Ramlan Abdul Aziz for his moral support and motivation as
well as Mr. Khairul Annuar Mohd, Ms. Nor Idamalina Ahamad Nordin and Mr. Rafizan
Latip for their assistance on the batch solid-liquid extraction process, qualitative analysis
of Vacuum Liquid Chromatography-Thin Layer Chromatography (VLC-TLC) and
biological activity (LC50) of rotenoids resin.
Last but not least, I would like to express my heartfelt gratitude to my most
beloved parents and wife, Allahyarham Zubairi Abdul Wahid, Zaiton A. Man and
Nurhafzan Anis Ismail respectively, through which the guidance of the spiritual, mental
and physical training that has allowed me to established and continue throughout this
study.
v
ABSTRACT
The objective of this research was to determine the effect of the processing parameters
on the extraction yield of rotenoids resin, rotenone and their biological activities (LC50).
The research was divided into three stages: preliminary, optimization and verification
phases. Preliminary study was carried out to determine the most appropriate processing
parameters for the optimization study. The optimization study was carried out using a
Central Composite Design (CCD) employing the Design-Expert® software version 6.0 to
determine the effects of processing parameters on the three selected response variables
which were the yield of rotenoids resin, yield of rotenone and biological activity (LC50)
of rotenoids resin. The processing parameters studied were the types of solvent
(acetone, chloroform and ethanol), solvent-to-solid ratio (2.0 ml/g to 10.0 ml/g) and raw
material particles size (0.5 mm to 5.0 mm in diameter). The theoretical maximum yield
of rotenoids resin in dried roots obtained from the optimization phase was 12.26 %
(w/w) and 5.99 % (w/w) for the rotenone. The multiple response variables analysis have
consistently verified the theoretical results in the range of 3.62 ml/g to 4.72 ml/g
(solvent-to-solid ratio) and 0.83 mm to 1.41 mm in diameter (raw material particles size)
using the acetone extract. The biological activity (LC50) value of rotenoids resin was
indirectly correlated to the optimum processing parameters due to inconsistency of
rotenone content (mg) and the low value of LC50 which was less than 100 ppm for all
treatments. This is due to the presence of other constituents in the rotenoids resin
(tephrosin, 12αβ-rotenolone and deguelin) which contributed to the low LC50 values.
The optimization of the processing parameters resulted in an increase of yield of
rotenoids resin but reduced yield of rotenone.
vi
ABSTRAK
Objektif kajian ini adalah untuk menilai kesan parameter pemprosesan terhadap
pengekstrakan keberhasilan resin rotenoids, rotenone dan aktiviti biologikalnya (LC50).
Kajian ini dibahagikan kepada tiga peringkat: fasa saringan, pengoptimuman dan
penentusahan. Kajian saringan dijalankan untuk menentukan parameter pemprosesan
yang paling relevan untuk kajian pengoptimuman. Kajian pengoptimuman dijalankan
menggunakan analisis ‘Central Composite Design (CCD)’ menggunakan perisian
‘Design-Expert® version 6.0’ bagi menilai kesan parameter pemprosesan bagi tiga
variabel respon yang dipilih iaitu keberhasilan resin rotenoids, keberhasilan rotenone
dan aktiviti biologikal (LC50) bagi resin rotenoids. Parameter pemprosesan yang dikaji
adalah jenis pelarut (aseton, kloroform dan etanol), nisbah pelarut terhadap pepejal (2.0
ml/g hingga 10.0 ml/g) dan saiz partikel bahan mentah (0.5 mm hingga 5.0 mm dalam
diameter). Keberhasilan maksimum teori resin rotenoids di dalam akar kering yang
diperolehi daripada fasa pengoptimuman adalah 12.26 % (w/w) dan 5.99 % (w/w) untuk
rotenone. Analisis kepelbagaian variabel respon mengesahkan secara konsisten
keputusan teori di dalam julat 3.62 ml/g hingga 4.72 ml/g dan 0.83 mm hingga 1.41 mm
dalam diameter menggunakan pengekstrakan aseton. Nilai aktiviti biologikal (LC50)
resin rotenoids tidak berkaitan secara langsung dengan parameter pemprosesan optimum
disebabkan oleh kandungan rotenone (mg) yang tidak konsisten dan nilai LC50 yang
rendah di mana kurang daripada 100 ppm bagi semua rawatan. Ini disebabkan oleh
kewujudan kandungan lain di dalam resin rotenoids (tephrosin, 12αβ-rotenolone dan
deguelin) di mana turut menyumbang kepada nilai LC50 yang rendah. Pengoptimuman
parameter pemprosesan di dapati telah menyebabkan peningkatan keberhasilan resin
rotenoids tetapi mengurangkan keberhasilan rotenone.
vii
TABLE OF CONTENTS
CHAPTER TITLE PAGE
TITLE PAGE i
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENTS iv
ABSTRACT v
ABSTRAK vi
TABLE OF CONTENTS vii
LIST OF FIGURES xiv
LIST OF TABLES xix
LIST OF ABBREVIATIONS xxii
LIST OF APPENDICES xxiv
1 INTRODUCTION
1.1 Research background 1
1.2 Scopes of research 7
1.3 Contribution of the research 8
viii
2 TUBA, Derris elliptica: OVERVIEW, BIOLOGY,
CULTIVATION AND PHYTOCHEMISTRY
2.1 Overview of the phytochemicals 9
2.1.1 Metabolic pathway of the
phytochemical insecticides 10
2.2 Derris elliptica or ‘Tuba’ 12
2.3 Scientific classification (taxonomy) and species 13
2.3.1 Plant growth, development and ecology 13
2.3.2 The cultivation condition of Derris elliptica 14
2.3.3 Current development on the cultivation
of Derris elliptica 16
2.4 Phytochemistry of Derris species 16
2.4.1 Outline of rotenone as an active chemical
constituents 16
2.4.2 Physico-chemical properties of rotenoids 17
2.4.3 Rotenone stability in water 20
2.4.4 Rotenone stability in soil and groundwater 21
2.4.5 Rotenone stability in vegetation 22
2.4.6 Types of rotenone formulation 22
3 PROCESSING, ANALYSIS AND TOXICOLOGY
3.1 Introduction 24
3.1.1 Extraction method 25
3.2 Extraction mechanism 26
3.2.1 Principles of solid-liquid extraction 28
3.2.1.1 Types of solid-liquid extraction 28
3.2.1.2 Desirable features for the
extracting solvent 28
ix
3.2.1.3 Leaching process
(solid-liquid extraction) 29
3.2.2 Extraction of the rotenone and rotenoids
resin: An overview of pilot and industrial
plant scale production 30
3.3 Analytical methods 34
3.4 Toxicology 34
3.4.1 The use of biological assays to
evaluate botanicals 34
3.4.1.1 Dose-response curves 36
3.4.1.2 Hazard indicator categories 38
3.4.1.3 Toxicity assessment by probit
analysis 40
3.4.2 Brine Shrimp (Artemia salina)
Lethality study 41
3.4.2.1 Artemia life history 41
3.4.2.2 Hatching the Artemia 41
3.4.2.3 Harvesting the nauplii 42
3.4.2.4 Maintenance of brine shrimp 43
3.4.2.5 Optimum Artemia survival
condition 44
3.4.3 Rotenone toxicology data 45
3.4.3.1 Mode of action 45
3.4.3.2 Toxicity 45
(a) Human data 45
(b) Aquatic life data 46
(c) Relevant animal data 46
(c) Relevant in vitro data 47
(d) Workplace standards 47
(e) Acceptable Daily
Intake (ADI) 47
x
(f) Carcinogenicity 47
(g) Mutagenicity 47
(h) Interactions 47
3.4.4 CASE STUDY: Laboratory and field
efficacy studies on the toxicity of the
formulated rotenone 48
3.4.4.1 Laboratory studies (bioassay) 48
3.4.4.2 Field efficacy studies 50
4 METHODOLOGY
4.1 Introduction 51
4.1.1 Preliminary experiments 52
4.1.2 Optimization phase 53
4.1.2.1 Design of Experiments (DOE) 54
4.1.2.2 Factors and experimental matrix 55
4.1.3 Verification phase 58
4.2 Sampling 58
4.3 Process description 58
4.3.1 Pre-processing of Derris roots 60
4.3.2 Extraction of rotenoids resin 61
4.3.3 Analysis of the response variables 63
4.3.3.1 Determination of extraction yield
(rotenoids resin) 63
4.3.3.2 Determination of extraction yield
(yield of rotenone) 64
(a) Qualitative analysis of
rotenone using Thin Layer
Chromatography (TLC) 64
xi
(b) Quantitative analysis of
rotenone using High
Performance Liquid
Chromatography (HPLC):
Measurement of the
rotenone content (mg) 65
(c) Biological activity (LC50)
of rotenoids resin 67
4.4 Statistical analysis 69
4.4.1 Response Surface Methodology (RSM) 70
4.4.2 Model adequacy checking 71
4.4.2.1 F-distribution test 71
4.4.2.2 Coefficient of multiple
determinations (R2) 72
4.4.2.3 Lack of fits test 72
4.4.3 Pearson’s correlation coefficient, r 73
5 RESULT AND DISCUSSION
5.1 Introduction 74
5.2 Preliminary experiment results 75
5.2.1 Effects of the plant parts and types
of solvent on yield 75
5.2.2 Extraction yield model and the effect
of extraction duration on yield 77
5.2.3 Effects of the extraction and concentration
operating temperature on yield 82
5.2.4 Effect of the raw material particles size
and solvent-to-solid ratio on yield 86
xii
5.2.5 Summary of the preliminary experiments 94
5.3 Optimization phase results:
Effect of processing parameters on the response
variables 98
5.3.1 Effect of processing parameters on the
yield of rotenoids resin in dried roots 100
5.3.2 Effect of processing parameters on the
yield of rotenone in dried roots 109
5.3.3 Summary of the optimization phase 118
5.4 Multi response analysis of the yield of rotenone
in dried roots; % (w/w) and rotenone concentration
(mg/ml) 120
5.4.1 Analysis of solvent-to-solid ratio (ml/g)
for the ethanol + oxalic acid solution
extract in relation with the yield of
rotenone in dried roots; % (w/w) and
rotenone concentration; mg/ml 121
5.4.2 Analysis of solvent-to-solid ratio (ml/g)
for the acetone extract in relation with
the yield of rotenone in dried roots;
% (w/w) and rotenone concentration; mg/ml 122
5.4.3 Analysis of raw material particles size
(mm in diameter) for the ethanol +
oxalic acid solution extract in relation
with the yield of rotenone in dried roots;
% (w/w) and rotenone concentration; mg/ml 125
5.4.4 Analysis of raw material particles size
(mm in diameter) for the acetone extract
in relation with the yield of rotenone in
dried roots; % (w/w) and rotenone
concentration; mg/ml 127
xiii
5.5 Biological activity (LC50) of rotenoids resin results 129
5.5.1 The effect of raw material particles size
and types of solvent on the biological
activity (LC50) and yield of rotenone
in dried roots, % (w/w) respectively 132
5.5.2 The effect of solvent-to-solid ratio
and types of solvent on the biological
activity (LC50) and yield of rotenone
in dried roots, % (w/w) respectively 135
5.5.3 Biological activity (LC50) of the
verification phase parameters and
rotenone standard (SIGMA-Aldrich™) 143
5.6 Verification phase results:
Confirmation of the optimization 143
5.7 Comparison of the optimum response variables 144
5.8 Correlation between the yield of rotenoids
resin and yield of rotenone 148
6 CONCLUSIONS AND RECOMMENDATIONS
6.1 Conclusions 151
6.3 Recommendations 154
REFERENCES 155
APPENDICES 173
xiv
LIST OF FIGURES
FIGURE NO. TITLE PAGE
2.1 Biosynthetic origin of some commercially important plant-derived compounds 11 2.2 Derris species that abundantly available in Peninsular of Malaysia: (A) Derris elliptica and (B) Derris malaccensis 13 2.3 Rotenone molecular structures 18 3.1 Schematic diagram of vegetal cell structures 27 3.2 Layout of the pilot plant scale production of the Concentrated Liquid Crude Extract (CLCE) 32 3.3 Dose-response curve 36 3.4 An adult of Artemia salina: (A) male; (B) female 41 3.5 Example of the Brine Shrimp hatchery system 44 3.6 The leaf-dipped method 49 3.7 The larvae of diamondback moth (Plutella xylostella) 49 3.8 Field efficacy of formulated rotenone against Spodotera litura 50 4.1 Phases of the experiment 59 4.2 Flow diagram and overview of the study 59 4.3 Various particles size of Derris roots 60
xv
4.4 Extraction of rotenoids resin from Derris elliptica roots 62 4.5 Evaluation of rotenoids resin 63 4.6 Techniques of spotting the sample on silica plate 65 4.7 External standard method template calculations 66 4.8 Example of the dilution principles to prepare the biological activity concentration 68 4.9 Mortality of Artemia salina when exposed to extracts of the Derris elliptica: (A) dose response curve; (B) probit analysis curve 69 5.1 Yield of rotenone in dried roots, % (w/w) using the Normal Soaking Extraction (NSE) method for different types of solvent 76 5.2 Kinetic equilibrium of the rotenone extraction process (second order polynomial) 80 5.3 Kinetic equilibrium of the rotenone extraction process: Yield of rotenone content in dried roots, % (w/w) 81 5.4 Kinetic equilibrium of the rotenone extraction process: Concentration of rotenone, mg/ml 81 5.5 Kinetic equilibrium of the rotenone extraction process: Yield of rotenone in dried roots, mg 82 5.6 Degradation of rotenone content (mg) during the concentration process at 40 0C and 0.3 mbar of operating temperature and vacuum pressure respectively 83 5.7 Degradation of rotenone content (mg) during the concentration process at 50 0C and 80 mbar of operating temperature and vacuum pressure respectively 84 5.8 Kinetics of the rotenone extraction process from
Derris elliptica - Ethanol + oxalic acid solution: (A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w) 85
5.9 Kinetics of the rotenone extraction process from Derris elliptica - Chloroform: (A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w) 90
xvi
5.10 Kinetics of the rotenone extraction process from Derris elliptica - Acetone: (A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w) 92
5.11 Response surface three-dimensional graphs and contour plot 99 5.12 Normal probability plots of residuals (Yield of rotenoids resin) 104 5.13 The residual versus the predicted response (Yield of rotenoids resin) 104 5.14 Surface plot of the yield of rotenoids resin in dried roots, % (w/w) as a function of raw material particles size and solvent-to-solid ratio: Ethanol + oxalic acid solution extract 105 5.15 Surface plot of the yield of rotenoids resin in dried roots, % (w/w) as a function of raw material particles size and solvent-to-solid ratio: Acetone extract 105 5.16 Normal probability plots of residuals (Yield of rotenone) 113 5.17 The residual versus the predicted response
(Yield of rotenone) 114
5.18 Surface plot of the yield of rotenone in dried roots, % (w/w) as a function of raw material particles size and solvent-to-solid ratio: Ethanol + oxalic acid solution extract 114 5.19 Surface plot of the yield of rotenone in dried roots, % (w/w) as a function of raw material particles size and solvent-to-solid ratio: Acetone extract 115 5.20 Selected processing parameters that obtain maximum yield of rotenoids resin in dried roots, % (w/w) and yield of rotenone in dried roots, % (w/w) based on the desirability values of a given solution 120 5.21 The yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml versus the solvent-to-solid ratio (ml/g) of ethanol + oxalic acid solution extract 123
xvii
5.22 The yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml versus the solvent-to-solid ratio (ml/g) of acetone extract 125 5.23 The yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml versus the raw material particles size (mm in diameter) of ethanol + oxalic acid solution extract 127 5.24 The yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml versus the raw material particles size (mm in diameter) of acetone extract 129 5.25 Relationship between the probit of Artemia salina mortality proportion and log10 dose of rotenoids resin (S1) at 24 hours of treatment 130 5.26 Relationship between the probit of Artemia salina mortality proportion and log10 dose of the rotenoids resin (S1, S7, S20 and S23) at 12 hours of treatment 130 5.27 Relationship between the probit of Artemia salina mortality proportion and log10 dose of rotenoids resin (S7, S11, S13, S19, S23, S24, S25, S28 and S29) at 12 hours of treatment: Continued 131 5.28 Relationship between the probit of Artemia salina mortality proportion and log10 dose of rotenoids resin (S3, S8 and S12) at 6 hours of treatment 132 5.29 Effect of the raw material particles size, mm in diameter
against the biological activity (LC50) of acetone extract (A1) and ethanol + oxalic acid solution extract (A2) respectively 133
5.30 Effect of the raw material particles size, mm in diameter against the yield of rotenone in dried roots, % (w/w) obtained from the extract of acetone (B1) and ethanol + oxalic acid solution (B2) respectively 135
5.31 Effect of the solvent-to-solid ratio, ml/g against the biological activity (LC50) of acetone extract (C1) and ethanol + oxalic acid solution extract (C2) respectively 137 5.32 Effect of the solvent-to-solid ratio, ml/g against the yield of rotenone in dried roots, % (w/w) obtained from the extract of acetone (D1) and ethanol + oxalic acid solution (D2) respectively 139
xviii
5.33 Pearson’s correlation coefficients (r) between the yield of rotenone in dried roots; % (w/w) and yield of rotenoids resin in dried roots; % (w/w) 149
xix
LIST OF TABLES
TABLE NO. TITLE PAGE
2.1 Existence of rotenone in Derris elliptica Benth plant’s organs determined by different analysis methods 15 2.2 Rotenone, methionine and phenylalanine in different organs of Derris plant analyzed by reversed-phase HPLC 15 2.3 The solubility of pure rotenone at 20 0C 19 2.4 Time of rotenone dissipation versus temperature 20 2.5 Detoxification time of varies rotenone concentration 21 3.1 Rotenone extraction methods 33 3.2 Hazard indicator categories 39 3.3 Toxicity of the botanical insecticides against the larvae of DBM collected from Kluang, Johor 48 3.8 Toxicity of the botanical insecticides against the larvae of
DBM collected from Karak, Pahang 49
4.1 Preliminary processing parameters 52 4.2 The preliminary experiment to obtain the rotenoids resin based on the exploratory experiment carried out by Saiful et al. (2003) 52 4.3 Preliminary control processing parameters 53 4.4 Experimental design for the solvent-to-solid ratio of 3.3 ml/g 53
xx
4.5 Experimental design for the solvent-to-solid ratio of 10.0 ml/g 53 4.6 Preliminary response variables 53 4.7 Specification of Central Composite Design (CCD) 54 4.8 Optimization processing parameters 55 4.9 Optimization control processing parameters 56 4.10 Optimization response variables 56 4.11 Experimental matrix for the extraction of rotenoids resin: CCD (23) 57 4.12 Parameters of RP-HPLC recommended by Baron and Freudenthal (1976) 66 5.1 Processing parameters involved in the kinetic of rotenone extraction process 79 5.2 Response variables result in the kinetic of rotenone extraction process 80 5.3 The average yield of rotenone in dried roots, % (w/w) 95 5.4 The preliminary experiments result 95 5.5 The design layout and experimental results
(Yield of rotenoids resin) 100
5.6 ANOVA response surface linear model [responses: Yield of rotenoids resin in dried roots, % (w/w)] 101
5.7 The design layout and experimental results (Yield of rotenone) 109
5.8 ANOVA response surface 2FI model (responses: Yield of rotenone in dried roots) (backward) 110 5.9 Selection criteria of the processing parameters solution 119 5.10 The effects of solvent-to-solid ratio (ml/g) of ethanol + oxalic acid solution extract on the two response variables 122
xxi
5.11 The effects of solvent-to-solid ratio (ml/g) of acetone Extract on the two response variables 124 5.12 The effects of raw material particles size (mm in diameter) of ethanol + oxalic acid solution extract on the two response variables 126 5.13 The effects of raw material particles size (mm in diameter) of acetone extract on the two response variables 128 5.14 Biological activity (LC50) of rotenoids resin at varies time of treatment (6 hours, 12 hours and 24 hours) 140 5.15 Effect of rotenoids resin against Artemia salina at varies time of treatment (6 hours, 12 hours and 24 hours) 141 5.16 Effect of rotenoids resin against Artemia salina on the 24 hours of treatment established by McLaughlin (1991) 142 5.17 List of selected processing parameter that produced theoretical maximum yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone in dried roots, % (w/w) 143 5.18 The verification phase results based on the most appropriate processing parameters 144 5.19 Comparison of the optimum response variables with Different phases of experiment 147 5.20 Pearson’s correlation coefficients (r) of the response variables 148
xxii
LIST OF ABBREVIATIONS
ANOVA - Analysis of Variance
A.i - Active ingredient
CEPP - Chemical Engineering Pilot Plant
CCD - Central Composite Design
CLCE - Concentrated Liquid Crude Extract
CP - Centre point
DAT - Days after treatment
DBM - Diamondback Moth
DF - Dilution factor
DIW - Deionized water
DOE - Design of Experiments
EPA - Environmental Protection Agency
EC - Emulsifiable Concentrates
IS - Internal standard
IPM - Integrated Pest Management
Kg - Kilogram
LC50 - Lethal Concentration of 50 % mortality
LD50 - Lethal Dose of 50 % mortality
LCE - Liquid Crude Extract
L - Litre
Ibm - Pound-mass
NSE - Normal Soaking Extraction
NPK - Nitrogen, Phosphorus and Kalium
ND - Not determined
xxiii
m.p - Melting point
OA - Oxalic acid
ppm - Part per millions
PDA - Photo Diode Array
Rf - Retardation factor
RSM - Response Surface Methodology
RP-HPLC - Reversed-Phase High Performance Liquid Chromatography
RCBD - Randomized Complete Block Design
SFE - Supercritical Fluid Extraction
SST - Total of Sum of Squares
SSR - Sum of Squares due to Regression
SSE - Sum of Squares of Residuals
SF - Sensitivity factor
SF - Safety factor
SD - Standard deviation
SG - Specific gravity
SHD - Safe Human Dose
TLC - Thin Layer Chromatography
ThD0.0 - Threshold Dose
UTM - Universiti Teknologi Malaysia
UPM - Universiti Putra Malaysia
UV - Ultra Violet
VLC - Vacuum Liquid Chromatography
xxiv
LIST OF APPENDICES
APPENDIX TITLE PAGE
A Complete results of the optimization phase experimental design 174 B Mortality of Artemia salina against varies
concentration of the rotenone Concentrated Liquid Crude Extract (CLCE) 179
C Experimental design and results (experimental and predicted values) of Central Composite Design; CCD (Manual calculation) 184 D The upper critical values of the F-distribution for v1 numerator degrees of freedom and v2 denominator degrees of freedom 5 % significance level: F0.05 (v1, v2) 190 E The biological activity (LC50) of S1 (24 hours of treatment) using probit analysis (manual calculation) 192 F Mortality of Artemia salina against varies
concentration (ppm) of rotenone Concentrated Liquid Crude Extract (CLCE) and rotenone standard (SIGMA-Aldrich™) 194
G Complete results of the verification phase and preliminary experiment (yield of rotenoids resin) 196 H Chromatograms of rotenone standard
[SIGMA-Aldrich™, 95 - 98 % (w/w)] and Sample 20 (S20) for the LCE and CLCE 198
xxv
I Purification and identification of rotenone from Derris elliptica using the Vacuum Liquid Chromatography-Thin Layer Chromatography (VLC-TLC) method 201 J Molecular structure of 12αβ-rotenolone, tephrosin and deguelin 206
CHAPTER I
INTRODUCTION
1.1 Research background
One of the important issues facing approximately 6.48 billions world populations
(Anonymous, 2005) is food security. The over population in developing countries and
low food production exacerbated the situation. Low food production productivity is due
to many factors. One of the factors is due to pest and plant diseases.
Crop protections today rely heavily on synthetic pesticides (Coats, 1994). Their
uninterrupted and massive use has led to several side effects such as pesticides resistance
in pests (Stoll, 1988), elimination of naturally occurring bio control agents, insect
resurgence, adverse effects on non-target organisms and environment contaminations
with the potential effect on the entire food chain (Copping, 1998; Harris, 1999). The
growing public alarm about the hazards associated with excessive use of synthetic
pesticides has revived the interest in the use of environmental-friendly crop protection
products or well known as phytochemical pesticides. Phytochemical pesticides are
considered environmentally benign, biodegradable (Devlin and Zettel, 1999), maintain
2
biological diversity of predators (Grainge and Ahmad, 1998) and safer to higher animals
including human beings. Thus, to help meet the food requirements of the 21st century,
scientist throughout the world is looking for ecologically safe plant protection
technologies emphasizing use of the botanical insecticides in the integrated pest
management (IPM) programmes.
A vast number of plant species produce phytochemicals that are not directly
beneficial for the growth and development of the plants. These secondary compounds
are regarded as a part of the plants defence against plant-feeding insects and other
herbivores (Dev and Koul, 1997). The pesticide properties of many plants have been
known for a long time and natural pesticides based on plant extracts such as rotenone,
nicotine and pyrethrum have been commonly used in pest control during the earlier half
of this century. However, after the World War II, they lost their importance with the
introduction of the synthetic organic chemicals (Suraphon and Manthana, 2001). The
synthetic organic chemicals were concentrated products with a high knockdown effect
on pest organisms. These chemicals could be produced in large quantities at relatively
lower cost and they rapidly substituted most of the other pesticides (especially natural
pesticides) in the 1950s. However, with the development of resistant insects, the threat
of contaminated food and high production cost problems, natural pesticides came back
again in 1995 (Suraphon and Manthana, 2001).
Furthermore, Malaysia is one of the most diverse countries to have plants and
animals in the Southeast Asia region. Some of the plants show ability to adapt to and
tolerate herbivores and their environment. The adapting ability derived from the
production of special chemicals is called allelochemicals, which are parts of the
secondary plant substances (Yu and Hsu, 1985). Plant active ingredient that shows
hormonal inhibition in insects is such as azadiracthin from neem seed kernels. The other
that show repellent property is such as citronella from citrus leaves (Visetson and Milne,
2001). In general, plants with pesticidal properties can be used in three ways. Initially,
from the whole plant parts, in powder form or as crude extracts in water or other
solvents. Secondly, as purified extracts, such as rotenone and finally as a chemical
3
template which could be produced industrially such as pyrethrins (chemically
synthesized). Moreover, natural insecticides have been used for centuries to combat
insect’s pests that compete for food and affect the public health (WHO, 1997). As for
that reason, more than 2,000 plant species are catalogued as having insecticide
properties (Heal et al., 1950; Farnsworth, 1966; Sukamar et al., 1991). The most
commonly plant extracts such as pyrethrum, nicotine, saponin and rotenone were among
the first compounds used to control insects of agricultural importance (Coasts, 1994;
Grainage and Ahmed, 1998). Among the ordinary plant extracts that have been studied
and commercialized is pyrethrin (which a complex of esters extracted from flowers of
Chrysanthemum cinerariefolium) which is the one that still be used nowadays to
enhance commercial preparations of the household insecticides (Bell et al., 1990).
Moreover, nicotine extracted from Nicotiana glauca and its nicotinoids derivatives are
also among the choice molecules for the manufacture of new insecticides.
In addition, the one and only plant extracts that in the route of developing its
niche market to be among the important insecticides is rotenone. In general, rotenone
and other toxic constituents in the extract (e.g.: deguelin, tephrosin and 12αβ-
rotenolone) are isoflavonoids occurring in several genera of tropical leguminosae plants
such as Derris (papilionaceae), Antonia (loganiaceae) and Lonchocarpus (fabaceae).
Scientifically, rotenone is a bio-active compound that has a strong paralysis action
(knock-down effect) on cold-blooded animals and used as an insecticide to combat pests
(John, 1944; Andel 2000). Other than that, rotenoids-yielding plants have been also
used for fishing based on their itcthyotoxic properties (Andel, 2000). For that reason,
there have been reports of rotenone-containing plants used by the Indians for fishing due
to its itcthyotoxic effect, as early as the 17th century (Moretti and Grenard, 1982).
Interestingly, rotenone poisoned fish is edible without any risk of food contamination to
humans (Costa et al., 1989). The risk of being poisoned by rotenone on mammalians
can be strongly justify with the lethality tests resulted in LD50 (value is in milligram per
kilogram (mg/kg) of body weight in mammal) range from 50 to 300 (Raws, 1986;
Ellenhorn and Barceloux, 1988) and considered as a moderate hazardous substances
(WHO, 1992).
4
As for it lethal mechanism, rotenone acts by inhibiting respiratory enzyme in the
insects resulting in disruption of cellular metabolism and failure of respiratory functions
(Oberg, 1961; Fukami et al., 1967; Bradbury, 1986). Although rotenone is toxic to the
nervous systems of insects, fish and birds, commercial rotenone products presented little
hazard to humans over many decades (Schmeltz, 1971). Neither fatalities nor systemic
poisonings in humans have been reported in relation to ordinary use. As for that reason,
human or mammals are not highly susceptible or vulnerable to rotenone because they are
protected by effective oxidizing enzyme systems (Schnick, 1974) and inefficient
gastrointestinal absorption (Bradbury, 1986). This extensive research and thorough
evidence on its effect against targeted organisms and non-targeted organisms (especially
human) gives rotenone as one of the botanical insecticide that exceptionally selective
and environmental-friendly (Schnick, 1974; Bradbury, 1986). Other reasons that
accounting safety record of rotenone as a botanical insecticide are low concentration in
commercial products, highly degradable and poor absorption across gut and skin of
humans. Even though rotenone is a naturally occurring chemical with insecticidal,
acaricidal (mite and spider killing) and piscicidal (fish-killing) properties (Extoxnet,
1996), it is a selective, non-specific insecticide and also can be used in home gardens
for insect control, for lice and tick control on pets and for fish eradications as part of
water body management (Weier and Starr, 1950). Because of its advantages, the
extracts material can be formulated into emulsifiable concentrates (EC) and wettable
powders of rotenone and extensively used in lakes, ponds and reservoirs to control
undesirable fish as well as to combat the highly resistant insect pests that still posses a
major threat to farmers all over the world (Kole et al., 1992).
Nowadays, the production of botanical insecticide especially rotenone (from
Derris and Lonchocarpus species) and pyrethrum (from Chrysanthemum
cinerariaefolium) are dominated by the Western country such as Germany, United States
(US), Canada and South America (Murray, 1997). They have the technology to extract,
formulate and purify the bio-active constituents from plant material that have the
insecticidal properties. One of the biggest manufacturers in Europe that produced the
formulated liquid emulsion of rotenone and cube rotenoids resin is SAPHYR S. A. R. L
5
which is based in France. According to Grinda et al. (1986), the used on a batch solid-
liquid of Accelerated Solvent Extraction (ASE) method has made them produced as
much as 14 % (w/w) yield of rotenone in dried roots and 36 % (w/w) yield of rotenoids
resin in dried roots. This achievement is due to the advanced processing techniques they
have implemented and the usage of strong chlorinated organic solvent that extract the
bio-active constituents exhaustively. They have the technological advantages as
compared to the other countries (especially in Asia) that also produced botanical
insecticides product.
In Asia, only several countries are committed on developing the technology and
pursuing to produce large scale of botanical insecticide such as Thailand and Vietnam.
According to Hao et al. (1998), in Vietnam, they have conducted and set up a technical
process to manufacture products from Derris elliptica Benth’s root. They have included
the emulsifiable concentrates (EC), water milk and water-soluble powder preparations.
The technological protocol were established in many ways such as raw material
pre-processing treatment, extraction procedures, types of solvent used, stability of
rotenoids resin, biological activity, preservation and packaging. In fact, they have
successfully extracted rotenone from their native Derris species with the yield of
approximately 1.5 % (w/w) to 5.0 % (w/w) in dried roots using varies organic solvent
such as acetone, chloroform and ethanol on a batch solid-liquid of Normal Soaking
Extraction (NSE) method (Phan-Phuoc-Hien et al., 2003). Other processing parameters
(e.g.: solvent-to-solid ratio, raw material particles size, extraction temperature,
extraction duration, speed of agitation and etc.) used in the process are unknown but
generally ethanol is largely used as a solvent in the extraction process due to its low cost
and simple process (Hien-Phan-Phuoc et al., 1999). In Thailand, they have also
established the technological protocol on manufacturing the botanical insecticides. They
have also implemented a batch solid-liquid of Normal Soaking Extraction (NSE) method
with agitation under room temperature of 26 ± 2 0C and administered for 8 hours
(Suraphon and Manthana, 2001). Hence, they have managed to extract rotenone
approximately 5.2 % (w/w) in dried roots. According to Pitigon and Sangwanit (1997),
the most desirable solvent for the extraction of rotenone are ethanol and although
6
chloroform is also used as a solvent, it is proven to be dangerous to human health.
Therefore, ethanol is more suitable solvent for the extraction of rotenone in favour of
Thailand farmers. Unfortunately, rotenone based bio-pesticide manufacturer is
unavailable in Malaysia despite of its environmental-friendly effect and effectiveness to
treat the persistent insect pests of Diamondback moth (Plutella xylostella Linn.) that
always infested in the leafy vegetables farms. Thus, no rotenone based bio-pesticide
listed in the Pesticide Board of Malaysia registered products until May 2006. As for that
reason, Chemical Engineering Pilot Plant (CEPP) in Johor, Malaysia has initiated a
research on this particular active ingredient (rotenone) since 2001 and also being the
only research institute in Malaysia that undertake the research systematically by
commencing the selection of Derris species, pre-treatment, extraction, formulation,
laboratory bioassay and field trial, toxicity level and risk assessment until product
registration through the Pesticide Board of Malaysia (Saiful et al., 2003)
From all manufacturers and researchers that involved on producing the rotenone
as a potential botanical insecticides, they have one in common which is using a batch
solid-liquid extraction method even though they have implemented different processing
parameters and produced varies yield of rotenoids resin in dried roots; % (w/w) and
yield of rotenone in dried roots; % (w/w). In addition, not one of them has implemented
the biological activity (LC50) of brine shrimp (Artemia salina) to acquire rapid general
toxicity level in which correspond to the effect of processing parameter.
As for that reason, the objective of this research was to investigate the effect of
processing parameters (raw material particles size (mm in diameter), solvent-to-solid
ratio (ml/g) and types of solvent) on the yield and its biological activity (LC50) of
rotenone extracted from Derris elliptica using a batch solid-liquid extraction process.
7
1.2 Scopes of research
In order to achieve the objective, four scopes have been formulated in this
research. The scopes were:
(1) To investigate the effect of processing parameters on the yield of rotenoids resin
in dried roots; % (w/w).
(2) To investigate the effect of processing parameters on the yield of rotenone in
dried roots; % (w/w).
(3) To investigate the effect of processing parameters on the biological activity
(LC50) of brine shrimp (Artemia salina).
(4) To investigate the correlation between the biological activity (LC50) with the
yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone in dried
roots; % (w/w).
The processing parameters studied were solvent-to-solid ratio (ml/g), types of
solvent and raw material particles size (mm in diameter). The other relevant parameters
involved in this research were fixed (control parameter) such as extraction temperature
(0C), weight of raw material (g) and extraction duration (hour). The experiments were
design using experimental design software called Design-Expert® software version 6.0
(Stat-Ease, 2002). Each data obtained from each run of experiments was evaluated for
the yield of rotenoids resin in dried roots; % (w/w), yield of rotenone in dried roots; %
(w/w) and biological activity (LC50) of rotenoids resin. Eventually, the correlation
between these response and independent variables were analyzed and interpreted.
8
1.3 Contribution of the research
This study contributes new knowledge in the area of phytochemical processing
and phytochemical pesticide:
(A) This research help to understand the main and interaction effects of the
processing parameters towards the yield of rotenoids resin, rotenone and their
biological activities (LC50). The correlation determined between independent
variables would further promote and enhance the usage of rotenoids resin as a
phytochemical pesticide or botanical insecticide products. Understanding the
effect of processing parameters against the yield of rotenoids resin, rotenone and
their biological activities (LC50) are essential in designing a better processing
technology to maintain and preserve the bio-active constituents in the extracts
and rotenoids resin effectively.
(B) Although Derris roots have been identified as a potential cash crop due to its
abundance growth in Malaysia, no research work of its own native species has
been conducted locally. The identification of appropriate processing parameters
to acquire maximum yield of rotenoids resin, rotenone and their biological
activities (LC50) against targeted and non-targeted organism are also have not
being studied. According to the local patent and industrial company database
related to the botanical insecticides production, there is no rotenone-based
industry listed in Malaysia until now. Therefore, the opportunities to develop an
option to the synthetic pesticides and environmental-friendly natural bio-
pesticide from local plant species are the main rationale why this extensive study
should be completed and carried out successfully.
CHAPTER II
TUBA, Derris elliptica:
OVERVIEW, BIOLOGY, CULTIVATION AND PHYTOCHEMISTRY
2.1 Overview of the phytochemicals
Phytochemicals are sometimes referred to as phytonutrients; this term is often
used interchangeably (Vickery, 1981; Walton and Brown, 1999). In broad term, they are
said to be any chemical or nutrient derived from a plant source. However, in common
usage, they have more limited definition. They are usually used to refer as compounds
found in plants that are not required for normal functioning of the body but nonetheless
have a beneficial effect on health or an active role in the amelioration of disease
(Vermeulen, 1998). The global trend in the preferences for natural products and away
from synthetics products promotes the use of phytochemicals in various industries. This
is further explained that plants which have a beneficial phytochemicals especially for
human health (Salleh, 1998) usually produced no adverse effect as compared to the
synthetics (Faridah, 1998). Furthermore, phytochemical is one of the recent terms
quoted by scientists and product developers during the 20th century (Love, 1999) to
support the emerging fields of the nutraceuticals, pharmaceuticals, cosmeceuticals and
phytochemical bio-pesticide.
10
2.1.1 Metabolic pathway of the phytochemical insecticides
Phytochemical is defined as any organic substance or chemical constituent
obtained from plants (‘phyto’ is Greek for plants). However in scientific literature,
phytochemical is commonly used to describe the biologically active molecules in plants
that are not classified as vitamins or nutrients (Vermeulen, 1998; HealthComm
International, 1998). Phtyochemicals also refer to chemical active ingredient of plants.
The subject of phytochemistry deals with the chemical structures of the substances, their
biosynthesis, turnover and metabolism, their natural distribution and their biological
function (Harbone, 1998). Natural products can be classified into two major groups
based on the metabolic pathways and the function of the substances themselves
(Fasihuddin and Rahmah, 1993). Plant chemicals or phytochemicals are classified as
either primary or secondary metabolic products depending whether they play an
essential role in plant metabolism and are universally present in all plants (Vickery,
1981; Walton and Brown, 1999). Primary metabolism supplies all the necessary tools
(building blocks and energy) in order to enable the organisms to live. The term
‘secondary products’ or ‘secondary metabolites’ are applied to plant products that have
no apparent role in a growth and development of the plant (Manuel and James, 1985;
Bremness, 1994). Many are important as toxins or feeding deterrents and so contribute
to the plant.
Primary metabolites include common sugars, protein amino acids, purines and
pyrimidines of nucleic acids and chlorophyll. Secondary metabolites include a variety
of phytochemical families or groups and its common sources such as flavonoid
(e.g.: berries, herbs and vegetables); isoflavones or phytoestrogens (e.g.: barley and
soy); isothiocyanates (e.g.: cruciferous vegetables); monoterpenes (e.g.: citrus peels and
essential oils); organosulfur compounds (e.g.: chives, garlic and onions); saponin
(e.g.: beans, cereals and herbs); capsaicinoids (e.g.: chilli peppers) and phytosterol
(e.g.: vegetable oils). The phytochemical families or groups above are originated
respectively from three major compounds which are alkaloids (derived from amino acid,
the building block of proteins); terpenes (a group of lipids) and phenolics (derived from
11
carbohydrates). However, in the light of present day knowledge, this variation is
subjective as there is no sharp division between the two metabolites (Vickery, 1981;
Walton and Brown, 1999). Figure 2.1 shows the basic metabolic pathway in plants.
Figure 2.1 Biosynthetic origins of some commercially important plant-derived
compounds. Major groups are indicated by boxes (Walton and Brown, 1999)
The less abundant secondary plant metabolites on the other hand, have
apparently no function in the plant metabolism and often derived from the primary
metabolites as a result of the chemical adaptation to environmental stress. These
chemicals serve as chemical defences against pest infestations. Furthermore, the
chemical defensive system is also called as ‘allelochemicals’ in which the adapting
ability derived from the production of special chemicals of secondary plant substances
(Yu and Hsu, 1985). Some of the higher plants are storehouse of extractable secondary
metabolites and usually sufficient to be economically viable as raw materials for the
12
development of botanical pesticides as well as for pharmaceutical and other beneficial
applications. Such pesticidal, natural raw materials also provide stimulus for structural
modification and optimization of the lead molecules to obtain more effective crop
protection chemicals. On top of that, as estimated approximately 250,000 higher plant
species, very few have been surveyed and most remain unexploited and unutilized for
insecticidal active principles (Dev and Koul, 1997).
2.2 Derris elliptica or ‘Tuba’
Derris elliptica, or ‘Tuba’ as it is known locally is an insecticidal plant that has
been known to be used as bio-pesticide. ‘Tuba’ plant is a kind of woody creeper plant
and climber. Derris is a climbing plant of Southeast Asia and its roots contain rotenone,
a strong insecticide (Hutchison Encyclopaedia, 2000). It needs at least 75.0 % soil
moisture content and the surround temperature should be approximately 25 0C to 30 0C
to obtain high yield of the rotenone (mg) during its development. A calm area with low
acidity soil content enhances the production of rotenone (Grinda et al., 1986). In
Malaysia, ‘Tuba’ plants can be found abundantly in the area of palm oil and rubber
plantations. Many Malaysians farmers do not realized the potential of this particular
plant. This plant actually posses a great knockdown effect to the pest especially in the
order of Homoptera (e.g.: Diamondback moth; Plutella xylostella Linn.) (Suraphon and
Manthana, 2001). Figure 2.2 shows the species of Derris elliptica and Derris
malaccensis that can be easily found in the Peninsular of Malaysia. Furthermore, Derris
elliptica or locally known as ‘Tuba Kapur’ can be easily found on laterite or clays soil.
Meanwhile, Derris malaccensis or locally known as ‘Tuba Gading’ can be found on peat
soil (Saiful et al., 2003). Both of the species are extremely different in term of the
amount of fine and coarse roots collected during the pre-processing treatment wherein
Derris malaccensis procured more fine and coarse roots as well as the yield of rotenone
as compared to the Derris elliptica (Saiful et al., 2003).
13
Figure 2.2 Derris species that abundantly available in the Peninsular of Malaysia:
(A) Derris elliptica and (B) Derris malaccensis (Saiful et al., 2003)
2.3 Scientific classification (taxonomy) and species
‘Tuba’ is a member of the leguminosae, fabaceae family which comprises of 200
genera and 68 species including 21 species of Tephrosia, 12 of Derris, 12 of
Lonchocarpus, 10 of Millettia and several of Mundula (John, 1944). Three species are
found in Malaysia, which are Derris elliptica, Derris malaccensis and Derris uliginosa
(Gaby, 1986).
2.3.1 Plant growth, development and ecology
Derris is a small shrub originating in the tropical rainforests of Malaysia. It
grows in lowland areas and does not thrive at higher altitudes. It is the roots, which
contain the active substances, mainly rotenone (Starr et al., 2003). Grown in the shade,
Derris malaccensis requires a period of 1½ years to 2 years for it to produce a
worthwhile content of rotenone. However grown in full sunlight, it needs only nine
(A) (B)
14
months for the roots to develop sufficiently. Meanwhile, Derris elliptica required 26
months for the maximum development of the rotenone. Derris can propagate vegetative
and fully developed after six weeks. The crop is harvested by exposing the shallow
roots and cutting off those with a diameter of 2.0 cm to 6.0 cm wherein this range of
diameter have the highest yield of rotenone (mg). Furthermore, Derris thrives on many
soils but particularly on loams and clays (Gaby, 1986).
2.3.2 The cultivation condition of Derris elliptica
In Malaysia, the suitable areas to grow and develop Derris elliptica is in the area
of palm oil and rubber plantation with a loam and laterite soil (Saiful et al., 2003). In
the province of Soc Trang, Vietnam, Derris elliptica grows at the area of coastal sloppy
with a sandy soil like in Delta Mekong (Phan-Phuoc-Hien et al., 2003). In Vietnam, the
development of Derris elliptica for transplanting and harvesting are prepared in the late
rainy season with the aim to obtain high yield of rotenone. Unfortunately in Malaysia,
this plant is not as important in Vietnam in which they have been implemented and
established the cultivation techniques and technological process (Hao et al., 1998; Phan-
Phuoc-Hien et al., 2003) to produce different preparations of the insecticide products
(e.g.: emulsion concentrates, water milk and water-soluble powder preparations).
Therefore, there is no initiative from the Malaysian farmers to cultivate and use this
plant as an effective botanical insecticide although it has been scientifically proven to
kill insect pests with no major environmental problems. Furthermore, in the cultivation
techniques of Derris elliptica roots, this plant usually yielded approximately 1.8 to 3.4
times in the appropriate NPK application treatment as compared to the control treatment
(Hien et al., 1996). The preliminary researches on the formation, translocation, and
accumulation of rotenone in the Derris elliptica Benth conducted by Hien et al. (1996)
has showed that the yield of rotenone (mg) has the co-variations with their precursors
such as phenylalanine and methionine from the young to the old leaves. In this stage,
15
the biosynthesis of phenylalanine, methionine and rotenone have the co-variations with
the photosynthetic rate. In contrast, from the old leaves to stems and roots are contrary.
In fact, phenylalanine and methionine content decreased about 50 % while the yield of
rotenone increased many times and offered the highest in the roots. The experiments
cutting sieve-tube for blocking the transportation of solutes in the phloem vascular
demonstrated that after biosynthezing in leaf, rotenone translocated downwards to the
lower organs by the phloem route and eventually accumulates in the root. Observing the
structure of the phloem and xylem gave an initial explanation on the mechanism of
‘load’ at the phloem cell source and ‘unload’ at the xylem cell sink of rotenone in the
Derris root. This is the way on how to increase the Derris root yield and yield of
rotenone (mg) at the plantation fields (Hien et al., 2000). Table 2.1 and Table 2.2 show
the existence of rotenone by different analysis methods and the amount of methionine,
phenylalanine and rotenone analyzed by reversed-phase HPLC.
Table 2.1: Existence of rotenone in Derris elliptica Benth plant’s organs determined by
different analysis methods (Hien et al., 1996)
Derris plant’s organs
By Irwin Hornstein’s Titration Method
By Colorimeter UV-VIS Method
By Weigh Measure D. C. Beach Method
Young leaf 0.79 % (w/w) 0.39 % (w/w) 0.54 mg Old leaf 1.36 % (w/w) 0.77 % (w/w) 1.21 mg Branch 1.77 % (w/w) 1.50 % (w/w) 1.69 mg Stem 2.21 % (w/w) 1.74 % (w/w) 1.91 mg Root’s core 9.94 % (w/w) 11.96 % (w/w) 10.32 mg Root’s bark 5.56 % (w/w) 5.80 % (w/w) 4.71 mg
Table 2.2: Rotenone, methionine and phenylalanine in different organs of Derris plant
analyzed by reversed-phase HPLC (Hien et al., 1996)
Derris plant’s organs
Methionine;mg/100 g
Phenylalanine;mg/100 g
Rotenone; % (w/w)
Young leaf 1.98 27.90 0.39 Old leaf 12.97 51.70 0.77 Stem 8.90 21.90 1.74 Root 5.90 26.90 11.60
16
2.3.3 Current development on the cultivation of Derris elliptica
Currently, the intercropping model with other plant species such as Allium
ascalonium (onion shrub plant) was recommended to overcome the shortage of Derris
roots due to Derris monoculture habit by farmers in Vietnam (Phan-Phuoc-Hien et al.,
2003). This technique can also be utilized and implemented in Malaysia due to advance
facilities of the tissue culture laboratory as compared to the other Asia countries. The
intercropping model gained a lot of advantages such as root’s biomass yield and yield of
rotenone (mg) increase 24 % to 27 % as compared to the monoculture model. In
contrast, production expenditure of Derris in the former decreases 15 % to 20 % as
compared to the latter. Eventually, total profit of the intercropping model obtained 3.15
times as compared to the control (Derris monoculture). In addition, the new advanced
model has been applied largely and effectively in Soctrang province of Vietnam and has
proven to acquire high yield of rotenone (Hien et al., 1999).
2.4 Phytochemistry of Derris species
2.4.1 Outline of rotenone as an active chemical constituents
Derris elliptica and Derris malaccensis contain approximately 4.0 % (w/w) to
5.0 % (w/w) rotenone while Lonchocarpus utilis and Lonchocarpus urucu contain
approximately 8.0 % (w/w) to 10.0 % (w/w) rotenone in dried roots (Dev and Koul,
1997). Rotenone comprises of an isoflavone nucleus with an isoprene moiety attached
at C-8 of ring A as shown in Figure 2.3 (Kole et al., 1992). In addition, these plants
contain number of other isoflavonoids compound such as deguelin, 12αβ-rotenolone,
tephrosin, elliptone, sumatrol, toxicarol, malaccol and etc. (Dev and Koul, 1997) which
are toxic or induce behavioural or physiological effects. Low in mammalian toxicity,
17
rotenone is mainly active as a contact or stomach poison. This isoflavonoid is extremely
toxic to cold-blooded animals (especially fish) (Matsumura, 1985) and piercing-sucking
insect such as aphids, red bugs, chewing insects especially caterpillars upon plants,
external parasites such as fleas and lice (John, 1944) but less active in birds and higher
animals (Andel, 2000). In fact, rotenone can enter the insect body through the
alimentary canal, tracheae or integument. It appears to kill insects by specific
inactivation of the respiratory enzyme, glutamic acid oxidase resulting in death through
oxygen (O2) starvation (John, 1944). Due to its low toxicity when ingested, fishes
stupefied by rotenone can be consumed by humans without any adverse reaction
(Acevedo-Rodriquez, 1990). Furthermore, rotenone has three major advantages: (1)
humans can digest it relatively safe (2) they are harmless to plants (non phyto-toxic)
(Gaby, 1986) and (3) it is unstable in light and heat, loosing almost all its toxicity after
two to three days (Matsumura, 1985; Hamid, 1999).
2.4.2 Physico-chemical properties of rotenoids
Five compounds have been isolated and characterized from the chloroform
extract which are deguelin, tephrosin, rotenone, 12αβ-rotenolone, 12α-hydroxyrotenone
(Mourad and Anne, 1986). Therefore, there is a number of toxic constituents that have
been isolated from the roots and seeds of Derris species and the most important of which
is rotenone with a chemical name of 1,2, 12a-tetrahydro-8,9-dimethoxy-2
(1-menthylethenyl-(1) benzopyrano (2,4-b) furo (2,3-h) (1) benzophyran-6 (6H)-one
with melting point (m.p) of 163 0C (Kidd and James, 1991). Rotenone has the following
molecular structure as shown in Figure 2.3.
18
Figure 2.3 Rotenone molecular structures (Kidd and James, 1991; Kole et al., 1992)
Rotenone, with an empirical of C23H22O6, is an isoflavonoids compounds with a
molecular weight of 394.41 g/mol (Schnick, 1974). It consists of 70.04 % carbon, 5.62
% hydrogen and 24.34 % oxygen. It melts at 156 0C to 166 0C. Rotenone is very
soluble in a number of organic solvents like alcohol and acetone, but is practically
insoluble in water (John, 1944). According to Kidd and James (1991), rotenone is
slightly soluble in water with the amount of 15 mg/L at 100 0C. Beside that, the other
naturally occurring rotenoids are elliptone with melting point of 159 0C which has a
furan ring in place of the ring B of rotenone; sumatrol with melting point of 188 0C
which is 15-hydroxyrotenone; malaccol with melting point of 244 0C, which is 15-
hydroxyelliptone; toxicarol with melting point of 101 0C, which has a hydroxyl group at
carbon 15 (*); and deguelin with melting point of 165 0C to 171 0C which has a
hydrogen atom on carbon 15 (*) in place of the hydroxyl group of toxicarol. A related
material, tephrosin with melting point of 197 0C to 198 0C has a hydroxyl group on one
of the carbon atoms between rings A and C. It does not occur naturally in Derris resin
but it is an oxidation product of deguelin. All the naturally occurring rotenoids appear to
exist as levo forms. Furthermore, the toxicity level (LC50) of rotenoids resin against
targeted organism is largely unexplored, but individually rotenone is five to ten times as
effective as compared to the other rotenoids. The molecular structures of other rotenoids
such as 12αβ-rotenolone, deguelin and tephrosin are shown in Appendix J. In addition,
the yield of rotenone and total extractives (rotenoids resin) in the various commercial
plant species are variable. For example, the roots of Derris elliptica consist in
19
approximately 5.0 % (w/w) to 13.0 % (w/w) rotenone with total ether extractives of
approximately 31.0 % (w/w). Derris malaccensis consist in approximately 4.0 % (w/w)
rotenone with total extractives of 27.0 % (w/w) and Lonchocarpus utilis consist in
approximately 8.0 % (w/w) to 11.0 % (w/w) rotenone with total extractives of 25.0 %
(w/w). When exposed to light and air, rotenone decomposes by changing from
colourless through yellow to deep red and resulting in non-insecticidal products. As for
that reason, rotenone preparation should be protected from light and heat during
handling and storage. Additionally, rotenone is readily oxidized in the presence of
alkaline to dihydrorotenone by eliminating two hydrogen atoms to form a double bond
between rings A and C (Schnick, 1974; Branbury, 1986). This material is less toxic than
rotenone, which should therefore be considered incompatible with alkaline dusts such as
lime and soaps and other alkaline wetting and spreading agents (John, 1944). Moreover,
pure crystalline rotenone is prepared by extracting the powdered roots with a solvent
such as ether or carbon tetrachloride and concentrating the solution to produce crystal.
Table 2.3 shows the solubility of pure rotenone in selected organic solvents at 20 0C
(John, 1944).
Table 2.3: The solubility of pure rotenone at 20 0Ca (John, 1944)
Solvent Solubility (g/100 ml)a Water 0.00002 Ethyl alcohol (Ethanol) 0.2 Carbon tetrachloride 0.6 Amyl acetate 1.6 Xylene 3.4 Acetone 6.6 Benzene 8.0 Chlorobenzene 13.5 Ethylene dichloride 33.0 Chloroform 47.2
20
2.4.3 Rotenone stability in water
Rotenone is generally unstable and degrades rapidly in water. It has been shown
to degrade as fast as within fortnight of application (Schnick, 1974) but can also persist
for periods up to five month (Smith, 1941; Leonard, 1939). The length of degradation is
depend on many factors including light, temperature, turbidity, depth, presence of
organic debris and dose (Bradbury, 1986). Despite all the factors that go into rotenone
degradation, Schnick (1974) has reported that waters should still detoxify within five
weeks of the treatment. Rotenone is photochemically unstable and readily breakdown in
the presence of light (Kidd and James 1991). Light oxidatively decompose rotenone
into non-toxic dihydrorotenone and water (Schnick, 1974; Branbury, 1986). This
degradation process occurs at the quicker rate in the presence of high water temperature.
According to Grinda et al. (1986), one of the advantages of rotenone is its rapid
detoxification under natural conditions. In general, high alkalinity (more than pH 8.0 to
9.0), high temperatures (possibly more than 40 0C), abundant light and air and low
concentrations favour rapid detoxification of rotenone. The most apparent chemical or
physical property of the water which affects the breakdown of rotenone is temperature.
Table 2.4 shows variety of temperature condition in water to affect the rotenone
dissipation. Temperature appears to affect the breakdown of rotenone the most. As for
that reason, Dawson et al. (1991) concluded that higher water temperature would readily
facilitate the degradation of rotenone faster than lower water temperature. A variety of
rotenone concentration (ppm) also affected the detoxification of rotenone. Table 2.5
shows variety of rotenone concentration to affect the detoxification process of rotenone.
Table 2.4: Time of rotenone dissipation versus temperature (Grinda et al., 1986)
Temperature (0C) 10 15 20 25 Dissipation time (days) 26 14 7 4
21
Table 2.5: Detoxification time of varies rotenone concentration (Grinda et al., 1986)
Concentrations (ppm) 1 4 8Detoxification time (days) 2 4 7
Turbidity and organic debris in water act by slowing down the decay of rotenone.
It has been shown that rotenone absorb to the sediment and organic particles and persist
for longer periods of time (Dawson et al., 1991). High turbidity also corresponds to the
low light penetration into water, which allows rotenone to be degraded at a slow rate.
Depth of water also plays a role in the breakdown of rotenone. Rotenone tends to
breakdown more readily in the shallow epilimnion of water bodies (Branbury, 1986).
Furthermore, Schnick (1974) has reported that each increase in depth of 1.0 ft (0.31 m)
in a pond increased the length of rotenone toxicity by two days. Not only the epilimnion
has usually warmer than the deeper hypolimnetic waters, but it also got lighter than
hypolimnion. These two factors act to increase the rate at which rotenone degrades in
such waters. As mentioned earlier, the early studies of rotenone degradation has showed
that rotenone break down into two simple products namely as non-toxic dihydrorotenone
and water (Bradbury, 1986). Dihydrorotenone with melting point of 216 0C is about as
toxic as rotenone to many insects and it is more resistant to the decomposition of
sunlight. Further study by Cheng et al. (1972) using photo-degradation, they have
identified that rotenone decomposes to at least 20 degradation products, most of which
are rotenoids. They have reported that only one product is fairly toxic namely as 12αβ-
rotenolone. The fact that other 19 or more degradation product is not toxic is one of the
reasons rotenone can be used safely as an environmental-friendly insecticide.
2.4.4 Rotenone stability in soil and groundwater
Rotenone is rapidly broken down in soil and in water. The half-life in both of
these environments is between one and three days respectively (Augustijn-Beckers et al.,
22
1994). It does not readily leach from soil (Augustijn-Beckers et al., 1994), and it is not
expected to be a groundwater pollutant. Rotenone breaks down readily by exposure to
sunlight (Kidd and James 1991). Nearly all of the toxicity of the compound is lost in
five to six days of spring sunlight or two to three days of summer sunlight.
2.4.5 Rotenone stability in vegetation
Rotenone is a highly active but short-lived photo-sensitizer. This means that an
organism consuming the compound develops a strong sensitivity to the sun for a short
time (Phan-Phuac-Hien et al., 2003). A number of photodecomposition products are
formed when bean leaves are exposed to light. It is also sensitive to heat, with much of
the rotenone quickly lost at high temperatures (Phan-Phuac-Hien et al., 2003)
2.4.6 Types of rotenone formulation
The extract of Derris elliptica which contain rotenone and other toxic
constituents can be formulated and used in many form of insecticide products namely as:
(1) Dusts of ground Derris roots are mixed with 3.0 parts to 7.0 parts of carrier such
as talc, clay, gypsum, sulphur and tobacco or walnut shells ground to pass a 250
mesh to 350 mesh sieves. Furthermore, the impregnated dusts are also used
and produced by mixing the extract of ground Derris roots in a volatile solvent
with an adsorptive carrier. The solvent is then evaporated, leaving each dust
particle coated with the insecticide. This preparation is more uniform in
particle size as compared to the initial preparation of the ground Derris roots
23
dust. Dusts containing 0.5 % (w/w) to 1.0 % (w/w) rotenone and 1.8 % (w/w) to
3.5 % (w/w) total extractives are effective against most insects controlled by
rotenone and should be used at 15.0 lbm (6.8 kg) to 25.0 Ibm (11.3 kg) per acre on
such crops of cabbage and celery. On top of that, alkaline carriers such as lime
should not be used with rotenone (John, 1944).
(2) Dispersible powders may be made from finely ground Derris roots. Two to five
pounds (0.9 kg to 2.3 kg) of Derris powder with 2.0 pounds (0.9 kg) of neutral
soap or the equivalent of the sulfonated oil will make 100 gallons (380 L) of
spray. For the small amount, an ounce of a 4.0 % (w/w) to 5.0 % (w/w) rotenone
dust and a teaspoonful of spreader should be used in 2.0 gallons (7.6 L) of water.
(3) The extract of Derris is also widely used as dried resin which contains 25.0 %
(w/w) to 35.0 % (w/w) rotenone. This dried resin is usually used as emulsifiable
concentrates (EC) or known as spray oils. Because of the limited solubility of
rotenone in spray oils which is approximately 0.05 % (w/w), mutual solvents are
generally employed to increase the solubility to practical limits. Materials which
have been employed for this purpose are dibutyl phthalate, methylated
naphtalenes, alkylated phenols and high boiling ethers. The concentrated
solutions may either used as fly and cattle spray or emulsified in water as
agricultural sprays. Rotenone concentrates containing 1.0 % (w/w) rotenone and
3.5 % (w/w) to 4.0 % (w/w) total extractives may be diluted 1.0 part to 600 or
800 parts of water. For aphids, the rotenone concentrates may be diluted 1.0 part
to 800 parts of water which contain rotenone approximately 0.00125 % (w/w).
CHAPTER III
PROCESSING, ANALYSIS AND TOXICOLOGY
3.1 Introduction
Herbal extraction processes are used to produce herbal extracts from the herbal raw
material in several forms. These are including the extracts which contain the soluble
constituents, oleoresins which contain the volatile and non-volatile constituents and
essential oils which only contain the volatile constituents from the plant material
(Vickery, 1981; Manuel, 1985). Herbal extract could be defined as a compound mixture
obtained from the fresh or dried plant or parts of the plant such as leaves, flowers, seeds,
roots and barks by different extraction procedures. Normally, the active constituents are
obtained together with other materials present in the vegetal mass such as resins, fats,
waxes, chlorophyll and colouring materials. Moreover, the extraction of bio-active
components from the vegetal materials is an essential part of the nutraceuticals,
pharmaceuticals, cosmeceuticals and phytochemical bio-pesticide industry (Rice, 1995;
Mircea, 2001; Pinelo et al., 2006). Therefore, the key objective of this research work
was to determine the appropriate processing parameters with the aim to produce high
yield of rotenoids resin (g) and yield of rotenone (mg). Therefore, it is important to
25
understand the background of the herbal extraction processes and to discover the
correlation between the operating conditions, the yield obtained as well as the toxicity
level of the extract.
3.1.1 Extraction method
To obtain extracts from the vegetal materials, several methods are available:
(Harborne, 1984; Houghton and Raman, 1998; Mircea, 2001).
(1) Distillation:
(a) Direct essential oil distillation.
(b) Water steam distillation.
(c) Water and steam distillation.
(2) Conventional extraction technique or solvent extraction:
(a) Solvent extraction (percolation).
(b) Maceration with solvent or Normal Soaking Extraction (NSE).
(c) Boiling with water (infusion).
(d) Extraction with cold fat (enfleurage).
(e) Extraction with hot fat.
(3) Cold compression, which is the usual method for the natural oil industry.
(4) Non-conventional extraction technique:
(a) Supercritical Fluid Extraction (SFE).
(b) Vortical or turbo extraction.
(c) Extraction by electrical energy.
(d) Ultrasonic Assisted Extraction (UAE).
(e) High-pressure liquid extraction (Accelerated Solvent Extraction).
26
Nowadays, three promising technologies that are expended and applied in the
pharmaceuticals, cosmeceuticals, food industries and agriculture for pest control are
Supercritical Fluid Extraction (SFE), Accelerated Solvent Extraction (ASE) and
Ultrasonic Assisted Extraction (UAE). Supercritical Fluid Extraction (SFE) utilises a
supercritical fluid such as carbon dioxide (CO2) to extract the phytochemical of interest
from the plant matrix (Smith, 1999; Lang and Wai, 2001; Catchpole et al., 2002). By
varying the temperature and pressure, the permeability and solubility of the supercritical
fluid is varied and can be adjusted to extract and precipitate the specific compound of
interest. However, due to the high cost of operation as well as the consumer preference
of whole herbal extract approach to phytochemical processing, this method is rarely
used. The high-pressure liquid extraction method has been developed to use
conventional fluids under higher pressure and temperature conditions. A solvent such as
ethanol is heated under high pressure to enhance the solvent permeation and solute
solubility during extraction process. This method also known as Accelerated Solvent
Extraction (ASE) has reduced the extraction time significantly as well as produced high
yield of the extract (Ollanketo et al., 2002; Choi et al., 2003). Unfortunately, it is
currently utilized only on a small laboratory and analytical scale due to high
maintenance and operating cost. Lastly, the Ultrasonic Assisted Extraction (UAE) has
been utilized as a method to enhance the conventional extraction method such as
percolation and maceration. It is found that the application of ultrasonic waves during
the extraction process does increase the yield as well as reduce the extraction time under
certain condition (Mircea, 2001).
3.2 Extraction mechanism
Vegetal tissue consists of cells surrounded by the walls as shown in Figure 3.1.
The extraction mechanism involves two types of physical phenomena:
27
(A) Diffusion through the cell walls.
(B) Washing out (rinsing) the cell contents once the walls are broken.
Figure 3.1 Schematic diagram of vegetal cell structures (Mircea, 2001)
Currently, the Ultrasonic Assisted Extraction (UAE) is the most preferable
unconventional method used to extract high yield of bio-active constituents either in the
aqueous extract, essential oil or oleoresin. In fact, both phenomena above are
significantly affected by the ultrasonic irradiation of the Ultrasonic Assisted Extraction
(UAE) as compared to the other unconventional method. Theoretically, some cells that
existed in the form of glands (external or internal) are filled with essential oil or
oleoresin (Mircea, 2001). A characteristic of such glands (when external) is that their
skin is very thin and can be very easily destroyed by any method of extraction such as
sonication. Thus, the extraction of essential oil as well as fat oil (oleoresin) is facilitated
by sonication. Moreover, the milling degree of the vegetal material plays an important
role for the internal glands (Mircea, 2001). It is obvious by reducing the size of the
vegetal material particles will increase the number of the cells directly exposed to
extraction by solvent. This effect can be utilized by milling the material before
extraction (Mircea, 2001).
Diffusion
Rising
28
3.2.1 Principles of solid-liquid extraction
Solid-liquid extraction involves a mass transfer from one phase to another or in
other words it is concerned with the extraction of a soluble constituent from a solid by
means of solvent. Solid-liquid extraction is also known as leaching (Harborne, 1984;
Coulson et al., 1991; Houghton and Raman, 1998).
3.2.1.1 Types of solid-liquid extraction
There are several types of solid-liquid extraction such as percolation, infusion or
maceration and countercurrent extraction method (soxhlet). For this particular study, the
maceration method is employed due to its simplicity to handle and collect the samples.
Furthermore, infusion are prepared by leaving the plant material to soak in the solvent
generally at room temperature for a period of time with or without intermittent shaking,
followed by the filtration to separate the plant debris (Houghton and Raman, 1998).
3.2.1.2 Desirable features for the extracting solvent
There is no such thing as ‘universal solvent’. The solvent extraction is unique
for each separation problem. Among the desirable features for the extracting solvents
are high capacities for the species being extracted into it, selective in dissolving the
desired compounds, low mutual solubility with water, easily generated, have suitable
physical properties such as density, viscosity and surface tension and relatively
inexpensive, non-toxic and non-corrosive (Rydberg et al., 1992).
29
3.2.1.3 Leaching process (solid-liquid extraction)
Many biological, inorganic and organic substances occurred in a mixture of
different components in a solid. In order to separate the desired solute constituent or
remove undesirable solute components from the solid phase, the solid is contacted with a
liquid phase. The two phases are in intimate contact and the solute or solutes can diffuse
from the solid to liquid phase, which causes the separation of the components originally
in the solid. This process is known as leaching (Geankoplis, 1995).
Leaching is concerned with the extraction of a soluble constituent from the solid
by means of solvent. The process can be used for the production of a concentrated
solution of a valuable solid material. As far as the amount of soluble constituent present,
its distribution throughout the solid determines the method used for the leaching process.
If the solute is uniformly dispersed in the solid, the material near the surface will
dissolve first, leaving a porous structure in the solid residue (Geankoplis, 1995). The
solvent will then have to penetrate this outer layer before it can reach further solute and
the process will become progressively more difficult and the extraction rate will fall. If
the solute forms a very high amount of the solid, the porous structure may break down
almost immediately to give a fine deposit of insoluble residue and the access to the
solute will difficult and need for further treatment (Geankoplis, 1995; Mircea, 2001).
Generally, the process can be considered in three parts:
(A) The change of phase of solutes as it dissolves in the solvent.
(B) Diffusion through the solvent in the pores of solid to the outside of the particle.
(C) The transfer of the solute from the solution in contact with the particles to the
main bulk of the solution.
These processes may be responsible for limiting the extraction rate, though the
first process usually occurs so rapidly that it has negligible the effect of overall rate of
the extraction process.
30
In the biological and food processing industries, many products are separated
from their original structure by the leaching process. One of the important processes is
the use of organic solvents such as hexane, acetone and ether to extract the oil from
peanut, soybean, flax seeds, castor beans and sunflower seeds (Geankoplis, 1995). In
the pharmaceutical industry, many different pharmaceutical products are obtained by
leaching plant roots, leaves and stems.
Biological materials such as Derris root are cellular in structure and the soluble
constituents are generally found inside the cells. The rate of leaching may be
moderately slow because the cell walls provide another resistance to diffusion. As for
the leaching of bio-active constituents such as rotenone from the leaves, stems and roots,
drying and milling of the plant materials before extraction helps to rupture the cells
walls so that the resistance of diffusion can be minimized (Mircea, 2001).
3.2.2 Extraction of the rotenone and rotenoids resin: An overview of the pilot and
industrial plant scale production
The mass production of high quality and amount of rotenone in rotenoids resin
has been developed since 1980’s by the European company called SAPHYR S. A. R. L.
(France). According to Grinda et al. (1986), they have invented a method that extract
the insecticidal materials contain in the plants by means of liquid which entirely
harmless, both to man and to animals. The method developed successfully extracted
rotenone with the highest yield using the alkyl esters (butyl, hexyl and octyl esters of
fatty acids) up to approximately 36.0 % (w/w) in the cube rotenoids resin. According to
this invention that have been released on the USPTO PATENT 1987, the extraction
process is initially done using 100.0 g of finely crushed Derris powder into a 250 ml of
extraction vessel provided with agitator. The finely crushed Derris powder is soaked
with 25.0 g of octyl stearate and 160.0 g of methylene chloride. The solvent-to-solid
31
ratio of methylene chloride and octyl stearate mixture is approximately 2.0 ml/g. After
agitating for half an hour at 45 0C, the treated powder is separated from the Liquid Crude
Extract (LCE) by filtration and washed on the filter with 50.0 g of methylene chloride.
The Liquid Crude Extract (LCE) is introduced into a round-bottom flask with reflux
condenser, in which the methylene chloride is distilled so as to recover it. Finally, 39.0
g of rotenoids resin is produced with 36.0 % (w/w) rotenone and 64.0 % octyl stearate.
It is observed that the yield of rotenoids resin and rotenone in finely crushed Derris
powder of this product is 39.0 % (w/w) and 14.0 % (w/w) respectively. By diluting this
product with an equal volume of octyl stearate, a 7.0 % (w/w) rotenone solution is
obtained which can be used directly as an insecticide composition. A fraction of the
viscous product is emulsified in the presence of surface active agent with 10 times its
volume of water to serve as spray for plants in order to combat against targeted insects.
Meanwhile, a method of extracting rotenone with fewer chemicals, economically
viable and environmental-friendly is being investigated by Chemical Engineering Pilot
Plant (CEPP) researchers. This new approaches are crucial due to the yield of rotenone
(mg) extracted from the Malaysian species (Derris elliptica and Derris malaccensis) are
lower than market values. Therefore, the main consideration is to avoid as much as
possible thermal degradation that might occurred in the production line and to find the
right solvent that extracts more rotenone and environmental-friendly. Acetone and
ethanol are the appropriate organic solvent that suite with the requirement and also have
the capability to extract large amount of rotenoids resin (mg) as well as the rotenone
content (mg). The rotenone extraction methods that can be found from 1930 to 2003 are
summarized and shown in Table 3.1.
A process flow diagram of pilot plant scale is shown in Figure 3.2 comprises of a
few stages. It started with a selection and harvests the roots conforming to the
specifications to assure high purity of rotenoids resin and total extractives. The roots are
chopped to the correct size before been fed onto the extraction vessel. An appropriate
amount of oxalic acid, deionized water (DIW) and ethanol (acetone is the most
preferable solvent to extract more rotenone) are charged into the extraction vessel. The
32
extraction process is carried out for about 12 hours. After the extraction cycle
completed, the Liquid Crude Extract (LCE) from the extraction vessel is transferred to
the intermediate holding tank using the transfer pump and strainer. Using the same
pump and fine strainer, the extract is transferred to the evaporator unit for the recovery
of solvent. The rule of thumb for this recovery unit is to avoid the thermal degradation
that occurred during the concentration process using high vacuum pressure pump. This
is to assure that the minimum operating temperature of 40 0C have a sufficient heat and
vacuum pressure to evaporate solvent as quicker as possible at maximum rate of the
solvent recovery. Finally, the product is discharged to another storage tank of the
Concentrated Liquid Crude Extract (CLCE). On top of that, the polyolefin containers
are used for storages, protecting the products against excess heat and light.
Figure 3.2 Layout of the pilot plant scale production of the Concentrated Liquid
Crude Extract (CLCE) (Saiful et al., 2003)
33
Table 3.1: Rotenone extraction methods (Saiful et al., 2003)
SOURCES SOLVENT USED REMARKS CLAIM Kilgore (1936); US-Patent
Mesityl oxide oxalate, n-butyl ester
U.S. Patent No.: 2,149,917 METHOD: Unknown
The extract is a red-yellow colour and contains approximately 0.8 % (w/w) rotenone
Whitmino (1941); US-Patent
90 % carbon tetrachloride and 10 % dichloro diethylether
U.S. Patent No.: 2,267,385 METHOD: Unknown
The residuum is in the form of gummy mass which may contain from 2.0 % (w/w) to 12.0 % (w/w) rotenone
Gosselin (1984); Hayes (1982)
Chloroform, benzene
Laboratory research METHOD: Unknown
Not stated
Grinda et al. (1986); US-Patent
Metylene chloride and octyl stearate aliphatic acid ester (preferably C6 to C30)
U.S. Patent No.: 4,698,222 and assignee: SAPHYR S. A. R. L (France) METHOD: Accelated Solvent Extraction (ASE)
The yield of rotenoids resin and rotenone in finely crushed roots of this product are 39.0 % (w/w) and 14.0 % (w/w) respectively
Suraphon and Manthana (2001) (Thailand)
Ethanol Laboratory research. METHOD: Normal Soaking Extraction (NSE) with agitation and soxhlet extraction method
The yield of rotenone in dried roots is 5.20 % (w/w) to 8.60 % (w/w)
Gusmao et al. (2002); Brazil
Ethanol Laboratory research. METHOD: Unknown
Not stated
Saiful et al. (2003); Malaysia
Acetone Laboratory research: Temperature (0C): Ambient (26 ± 2 0C) Solvent-to-solid ratio: 10.0 ml/g Raw material particles size: 0.5 to 2.0 mm in diameter Extraction time: 24 hours METHOD: Normal Soaking Extraction (NSE)
The yield of rotenoids resin and rotenone in dried roots are approximately 9.50 % (w/w) and 1.95 % (w/w) respectively. Product is identical to the commercial product of SAPHYR S. A. R. L (France) and analytical grade (SIGMA-Aldrich™)
34
3.3 Analytical methods
Natural product extracts contain a wide variety of chemical compounds. Derris
elliptica and Derris malaccensis have over six major chemical constituents. To assist in
the quantitative determination as well as the qualitative identification of the extract
bio-active constituents, various analytical techniques have been used. Two major
techniques have been utilized in this study, which include Thin Layer Chromatography
(TLC) and reversed-phase High Performance Liquid Chromatography (RP-HPLC).
3.4 Toxicology
Toxicology is the science that deals with the study of adverse effects chemicals
or physical agents may produce in living organisms under specific conditions of
exposure. It is a science that attempts to qualitatively all the hazards, for example the
organ toxicities that associated with a substance as well as to quantitatively determine
the exposure conditions under which those hazard or toxicities are induced (Gosselin et
al., 1984; Philip, 2000). Toxicology is the science that experimentally investigates the
occurrence, nature, incidence, mechanism and risk factors for the adverse effects of toxic
substances (Philip, 2000).
3.4.1 The use of biological assays to evaluate botanicals
Bioassays offer a special advantage in the standardization and quality control of
heterogeneous botanical products. Products can be ‘heterogeneous’ due to the presence
mixtures of the bio-active compounds either from the same or from purposefully mixed
botanical sources (Gosselin et al., 1984; Philip, 2000). Physical analytical methods such
35
as chromatography are inadequate for this purpose as they are usually insensitive to the
chemical complexities found in crude botanicals extract. Most often the desired
biological response is due to a mixture of bio-active components and the relative
proportions of single bio-active compounds can vary from batch to batch while the
biological activity still remains within tolerable limits. Thus, physical or chemical
analysis of a single component in such mixtures is not completely satisfactory.
Unfortunately, the goal of many phytochemists has been simply to isolate, characterize
and publish botanically derived chemical substances without regard to the bioassay. To
achieve applied meaning and significance, today’s work in natural product chemistry
must incorporate bioassay. The extracts must be screened for biological activity
whereby the ‘active’ extracts will be selected and fractionated for further exploitation.
This is the salvation of the natural product chemist and such work must be performed
with all useful bio-active botanicals if these products are to be accepted and incorporated
into legitimate, long term and health practices. Three readily available technologies
must be combined are:
(A) Separation techniques (Vacuum Liquid Chromatography).
(B) Structural elucidation methods (Spectrophotometers and X-ray crystallography).
(C) Simple bioassays.
Nowadays natural product chemists are very familiar with the first two but
generally they ignore the third. Standardization of the products by biological assays will
then generate reproducible benefits and increase consumer confidence. In addition, in
such specific bioassays, the same extracts have to be analyzed many times over and over
again before detecting activities. It would seem more logical to pre-screen with general
bioassays, throw out the negatives and employ specific bioassays on the activities. The
four pre-screening bioassays that useful are:
(A) Brine Shrimp Lethality: A rapid general bioassay for bio-active compounds.
(B) Crown gall tumours on potato disc: An animal sparing bioassay for anti-tumour
compounds.
36
(C) Frond inhibition of Lemna (duckweed): A bioassay for plant growth stimulants
inhibitors.
(D) Yellow Fever Mosquito (YFM) Test: A bioassay for pesticides.
3.4.1.1 Dose-response curves
The major purpose for performing the biological assay is to establish a cause
effect relationship between exposure to a toxic substance and an observed effect in order
to determine a safe exposure limit (Van, 1991). In general, as the dose increases, so
does the number of individually in each group demonstrating the measure response. By
plotting this information on a graph, with the horizontal axis representing the increasing
of doses and vertical axis representing the increasing of response, a curve can be drawn
in which illustrate the relationship between the dose administered and the observed
response (Taylor, 1985). This curve is referred to as a dose-response curve as shown in
Figure 3.3.
Figure 3.3 Dose-response curve (Van, 1991)
Increasing Dose (ppm or µg/ml)
Max
imum
Eff
ect R
ange
No-
effe
ct R
ange
Incr
easi
ng E
ffec
t (%
mor
talit
y)
RANGE OF INCREASING EFFECT (%) WITH INCREASING DOSE (ppm)
Thr
esho
ld
37
A dose-response curve can be developed for most phytochemicals and chemicals.
From these curves the thresh-old level and the relative toxicity of chemicals can be
obtained to help establish safe levels of phytochemical and chemical exposure
(Raven, 1973). A threshold is a dose below which no effect is detected or above which
an effect is first observed. The threshold information is useful in extrapolating animal
data to humans and calculating what may be considered a safe human dose for a given
substance.
The threshold dose (ThD0.0) is measured as mg/kg/day. It is assumed that
humans are as sensitive as the test animal used. To determine the equivalent dose in
man, the ThD0.0 is multiplied by an average weight of a man, which is considered to be
70.0 kg. The calculation used to determine the safe human dose (SHD) is as follows:
where;
SHD: Safe Human Dose.
ThD0.0: Threshold Dose at which no effect is observed.
70.0 kg: Average weight of a man.
SF: Safety factor (ranges from 10 to 1,000), which varies according to type of test and
data used to obtain the ThD0.0.
The safety factor chosen is dependent on the slope of the dose-response curve,
type of experimental animal used, and the availability of data from human exposure. In
general, the lower LD50 or LC50, the larger safety factor used. The lower LD50 or LC50
implies a more toxic substance, and a higher safety factor is chosen to ensure that a safe
human dose established. For example, presume the ThD0.0 for substance A has the LD50
of 0.5 mg/kg and the ThD0.0 for substance B has the LD50 of 5.0 mg/kg. If all other test
protocols are the same, substance A is 10 times more toxic than substance B.
SHD = ThD0.0 × 70.0 kg/SF = Amount mg/day of toxic substance (3.1)
38
Therefore, to determine a safe human dose the safety factor chosen for substance
A will be larger (100 or 1000) than the safety factor chosen for substance B, which may
be 10. Performing the calculations using these data will result in a safe human dose for
substance A being smaller that for substance B.
SHDA = 0.5 mg/kg × 70.0 kg/100 = 0.35 mg/day of toxic substance A.
SHDB = 5.0 mg/kg × 70.0 kg/10 = 35.0 mg/day of toxic substance B.
3.4.1.2 Hazard indicator categories
The Environmental Protection Agency (EPA) has established four toxicity
categories based on the LD50 or LC50 as well as an eye and skin effects of the various
pesticides. The user is the key to these toxicity categories wherein there is a signal
words present on the front panel of the pesticide label. Table 3.2 summarizes these
toxicity categories. Toxicity is usually expressed as the acute oral LD50. Acute oral
refers to a single dose taken by mouth or ingested. Acute dermal refers to a single dose
applied directly to the skin (skin absorption). Inhalation refers to exposure through
breathing or inhaling.
39
Table 3.2: Hazard indicator categories (EPA, 1996)
< Most toxic/hazardous > Least
toxic/hazardous I II III IV
Oral aLD50: 0 to 50 bmg/kg
50 to 500 mg/kg 500 to 5,000 mg/kg > 5,000 mg/kg
Inhalation cLC50: 0 to 0.02 dmg/L
0.2 to 2.0 mg/L 2.0 to 20.0 mg/L > 20.0 mg/L
Dermal (skin) aLD50: 0 to 200 mg/kg
200 to 2,000 mg/kg 2,000 to 20,000 mg/kg
> 20,000 mg/kg
Eye effects: Corrosive - corneal opacity not reversible within 7 days.
Corneal opacity reversible within 7 days. Irritation.
No corneal opacity. Irritation reversible within 7 days.
No irritation.
Skin effects: Corrosive.
Severe irritation at 72 hours.
Moderate irritation 72 hours.
Mild or slight irritation at 72 hours.
Signal words - DANGER/POISON: In large boldfaced letters on the label and usually accompanied by skull and crossbones symbol.
WARNING: In large boldfaced letters.
CAUTION: In large boldfaced letters.
CAUTION: In large boldfaced letters.
Acute (single) oral dosage to human adults: Few drops to 1 teaspoon.
1 teaspoon to 2 tablespoons.
1 ounce to pint > 1 pint
aLD50: Abbreviation for the amount toxicant (poison) needed to kill 50 % of a test animal population. It is expressed in terms of weight. LD50 is used to measure the acute oral and dermal toxicity of a chemical. The lower the LD50 value, the more poisonous the chemical. LD50 is not a measure of environmental hazard. bmg/kg: mg of chemical per kg of test animal body weight. cLC50: Abbreviation for the amount of toxicant (poison) present in air or water. It is expressed in terms of parts per millions (ppm = mg/L). The lower the LC50 value, the more poisonous the chemical. LC50 is not a measure of environmental hazard. dmg/L: mg of chemical per litre of air or water.
40
3.4.1.3 Toxicity assessment by probit analysis
The relationship between the concentration of an environmental toxicant and its
lethal effects on living organism is often a sigmoidal curve (Finney, 1964). Low
concentration may cause no mortality among members of the test group while high
concentrations cause 100 % mortality. Intermediate levels cause various degrees of
partial-kill in which some organisms die while others live. There are many ways to
analyze the results of such a quantal bioassay. One popular method is to use probit
analysis to make the sigmoidal response curve into straight line so that the contaminant
concentration (LC50) that is lethal to 50 % of the test group can be calculated (Finney,
1971; Hewlett and Plankett, 1979). The manual calculation of the probit analysis is
shown in the Appendix E. The LC50 value can be determined by two methods (Finney
and Colquhoun, 1971):
(1) Construct a graph of the logarithm-converted concentrations (on the X-axis)
versus the probit values (on the Y-axis). Plot only the concentrations with
corrected mortalities between 1.0 and 99.0 %. Draw a straight line of best fit
through the plotted points. Draw a horizontal line from 5.00 (the probit value of
50.0 %) on the Y-axis across to its intersection with the fitted line and observe
the value of X at the intersection point. Calculate the inverse log10 (antilog) of
this value to find the LC50.
(2) Using a statistical calculator or computer, conduct a least-squares simple linear
regression of logarithm-converted concentrations (X) versus the probit values
(Y). Use only concentrations with corrected mortalities between 1.0 % and 99.0
%. From the fitted regression equation (the correlation coefficient (R2) should
exceed at least 0.5), determine the predicted value of X associated with Y = 5.00
(the probit value of 50.0 %). Calculate the inverse log10 (anti-log) of this value
to find the LC50.
41
3.4.2 Brine Shrimp (Artemia salina) Lethality study
3.4.2.1 Artemia life history
The brine shrimp (Artemia salina) is in the phylum arthropoda, class crustacean
and is closely related to zooplankton like copepods and Daphnia as shown in Figure 3.4.
Artemia life cycle begins by the hatching of dormant cysts which are encased embryos
that are metabolically inactive. The cysts can remain dormant for many years as long as
they are kept dry (McLaughlin, 1991; Frank et al., 1996; McLaughlin and Rogers,
1998).
3.4.2.2 Hatching the Artemia
When the cysts are placed into salt water, they are re-hydrated and resume their
development. After 15 to 20 hours at 77 oF (25 oC), the cyst bursts and the embryo
leaves the shell. For the first few hours, the embryo hangs beneath the cyst shell, still
enclosed in a hatching membrane. This is called the umbrella stage. During this stage,
Figure 3.4 An adult of Artemia salina: (A) male; (B) female (Frank et al., 1996)
(A) (B)
42
the nauplius completes its development and emerges as free swimming nauplii. In the
first larval stage, the nauplii are a brownish orange color because of its yolk reserves and
do not feed because its mouth and anus are not fully developed. Approximately 12
hours after hatch they molt into the second larval stage and they start filter feeding on
various micro algae, bacteria and detritus (McLaughlin, 1991; Frank et al., 1996).
The nauplii will grow and progress through 15 molts before reaching adulthood
in about 8 days. Adult Artemia average about 8.0 mm long, but can reach lengths up to
20.0 mm. An adult is a 20 times increase in length and a 500 times increase in biomass
from the nauplii stage. In low salinity and optimal food levels, fertilized females usually
produce free swimming nauplii at a rate of up to 75 nauplii per day (Ken, 1999). They
will produce 10 to 11 broods over an average life cycle of 50 days. Under super ideal
conditions, adult Artemia can live as long as three months and produce up to 300 nauplii
or cysts every 4 days (McLaughlin, 1991; Frank, 1996).
3.4.2.3 Harvesting the nauplii
Harvest the nauplii by turning off the air or remove the air stone and let the
culture settle for about ten minutes. Hatched, empty shells float to the surface, and
unhatched cysts will sink to the bottom (Frank et al., 1996; Ken, 1999). The newly
hatched nauplii will concentrate just above the unhatched cysts on the bottom. Since the
newly hatched nauplii are attracted to light (phototropic), by shining a flashlight at the
centre of the bottle can concentrate them where it is easy to siphon them off or drain the
cysts off the bottom using the valve then drain the nauplii onto another container (Frank
et al., 1996; McLaughlin and Rogers, 1998). The unhatched cysts should be used in the
next culture and not thrown away since part of them might be hatch with the next batch.
43
3.4.2.4 Maintenance of brine shrimp
Being a low volume operation, water quality can deteriorate rapidly, especially
as biomass increases. The problem usually occurs because of over feeding, which leads
to fouling and low oxygen levels. To help overcome this problem, the tank should be
taken care seriously. Clean up the bottom every couple of days, turning off the air and
let the tank to be settled. Meanwhile siphon the crap off the bottom of the tank and
change the salt water about 20.0 % of total volume used per week is adequate enough for
their life cycle development (Frank et al., 1996; Ken, 1999). Moderate aeration with
coarse or air stones, good water quality and generally clean conditions are all important
for raising high densities of adult brine shrimp. Since the Artemia feed constantly, faster
growth rates and better survival is achieved by multiple or continuous feeding over a 24
hours period. Usually, Artemia are drawn to a strong light. The strong light actually
affects their development by slowing down the growth rates. This is due to the increase
of their swimming activity and energy expenditure. Therefore, a sufficient light is
needed for their normal development. Furthermore, in low light the Artemia will spread
out in the water column, swimming slowly and achieving more efficient food
conservation (McLaughlin, 1991; McLaughlin and Rogers, 1998). A complete Brine
Shrimp hatchery system is shown in Figure 3.5.
44
Figure 3.5 Example of the Brine Shrimp hatchery system (Ken, 1999)
3.4.2.5 Optimum Artemia survival condition
Cyst production is induced by conditions of high salinity and chronic food
shortages with high oxygen fluctuations between day and night (McLaughlin and
Rogers, 1998). Adults can tolerate short exposures to temperatures as extreme as 0 oF to
104 oF (-18 oC to 40 oC). Optimal temperature for cyst hatching and adult grow out is 77 oF to 86 oF (25 oC to 30 oC), but there are differences between other strains. Artemia
prefer a salinity of 30.0 ppt to 35.0 ppt (SG: 1.02 gml-1 to 1.03 gml-1) and can live in
fresh water for about 5 hours before they die (Frank et al., 1996; Ken, 1999).
45
3.4.3 Rotenone toxicology data
3.4.3.1 Mode of action
Rotenone inhibits the oxidation of NADH to NAD, blocking the oxidation by
NAD of substrates such as glutamate, α-ketoglutarate, and pyruvate. Rotenone inhibits
the mitochondrial respiratory chain between diphosphopyridine nucleotide and flavine.
This blockade is overcome by Vitamin K3 (menadione sodium bisulphate), which
apparently activates a bypass of the rotenone sensitive site. Rotenone is a powerful
inhibitor of mitochondrial electron transport. The regulation of fatty acid synthesis in
mitochondria by rotenone may be altered after chronic administration, resulting in fatty
changes in the liver (Hayes, 1982; Gosselin, 1984; Goodman and Gilman, 1985).
3.4.3.2 Toxicity
(a) Human data
Adults - Mean lethal oral dose is about 0.3 g/kg to 0.5 g/kg (Gosselin, 1984).
Mammals are not highly susceptible to rotenone because they are protected by
effective oxidizing enzyme system (Schick, 1974) and inefficient gastrointestinal
absorption (Bradbury, 1986).
Children - Mean lethal oral dose is estimate from 0.3 g/kg to 0.5 g/kg (Gosselin,
1984). In one fatal case, postmortem concentration of rotenone in the stomach
and blood were 1,260 ppm and 2.4 ppm (De-Wilde, 1986).
46
(b) Aquatic life data
Rotenone is highly toxic for aquatic life. Most of the LC50 values (96 hours of
treatment) for different fish species and daphnids (water fleas) lie in the range of
0.02 mg/L to 0.2 mg/L (ppm). If used as a piscicide, it may also cause a
temporary decrease in numbers of other aquatic organisms, such as daphnids
(WHO, 1992). On top of that, the use of rotenone in a large scale eradication of
the troublesome carnivorous piranha fish was done in Brazil. It was found that
piranha more sensitive than other regional species to the rotenone. Tests showed
that powdered rotenone at the rate of 0.2 ppm, eliminated the piranha (egg,
larvae, young and adult) within 20 mins. Therefore, rotenone has a number of
advantages as a fish toxicant including low mammalian toxicity at level of use
and rapid detoxification in treated waters. Lethal Concentration (LC50) is
variously reported to be between 0.01 ppm to 0.10 ppm (Grinda et al., 1986).
There is a demand for specific species toxicants to eliminate undesired species
while leaving desired fish unaffected. Some success has been achieved using
rotenone at levels of concentration and by modes of application that differentiate
between species susceptibility and living habits (Grinda et al., 1986).
(c) Relevant animal data
The Lethal Dose (LD50) values in milligram per kilogram (mg/kg) of body
weight in mammals are ranged from 50 to 300 (Ellenhorn and Barceloux, 1988).
Rat (oral) 60 to 132, rat (intravenous) 0.2 to 0.3, Mouse (intraperitonial) 5.4,
rabbit (oral) 1.5, rabbit (dermal) 100 to 200 and rabbit (intravenous) 0.35 to 0.65
(Hayes, 1982). For rat and dog, experimental inhalation of rotenone dust
produced symptoms within minutes. The onset of poisoning is more rapid than
after oral administration and the fatal dose is lower (Hayes, 1982). In addition,
the effect to other organism such as honeybees is nontoxic and harmless.
However, it is toxic to bees when used in combination with pyrethrum (Shane
and Doug, 2000).
47
(d) Relevant in vitro data
In isolated rat liver mitochondria, the aerobic oxidation of pyruvate is almost
completely inhibited by rotenone (Hayes, 1982).
(e) Workplace standards
The TLV-TWA (Threshold Limit Value-Time Weighted Average) for
commercial rotenone is 5.0 mg/m3. This indicates that an occupational intake of
0.7 mg/kg/day is considered safe (Hayes, 1982).
(f) Acceptable Daily Intake (ADI)
The proposed No-Adverse-Response Level (SNARL) for chronic exposure to
rotenone: 0.014 mg/l (National Research Council, 1983).
(g) Carcinogenicity
The carcinogenicity of rotenone is a controversial issue. It has been suggested
that rotenone may cause tumour only in vitamin-deficient animals (Gosalvez,
1983).
(h) Mutagenicity
No mutagenic effects were reported in mouse bone marrow (Waters et al., 1982).
Rotenone is non-mutagenic in bacteria reversion tests (Moriya et al., 1983).
(i) Interactions
When applied in low concentrations to plant foliage, rotenone catalyses
photoisomerization of dieldrin and other cyclodiene insecticide residues.
48
However, photodecomposition was a predominant effect when residues of
rotenone were combined with those of methylcarbamate and phosphothionate
insecticides (Hayes, 1975).
3.4.4 CASE STUDY: Laboratory and field efficacy studies on the toxicity of the
formulated rotenone
3.4.4.1 Laboratory studies (bioassay)
The toxicity of a simple rotenone formulation was evaluated by bioassay in the
laboratory against the early 3rd instar larvae of the diamondback moth (DBM) by
leaf-dipped method. The leaf-dipped method and the larvae of diamondback moth
(Plutella xylostella) are shown in Figure 3.6 and Figure 3.7 respectively. The
diamondback moth (DBM) was collected from Kluang, Johor and Karak, Selangor. The
Kluang strain is known for its resistance against some insecticides while the Karak strain
is considered as a relatively susceptible strain. Mortality was recorded at 48 and 72
hours after treatment and data were subjected to the probit analysis to obtain LC50.
Table 3.3: Toxicity of the botanical insecticides against the larvae of DBM collected
from Kluang, Johor (Dzolkifli, 2004)
Insecticide b ± S.E. LC50 (µg/ml) LC95 (µg/ml) Azadirachtin 1.64 ± 0.21 13.31 134.9 Rotenone 2.21 ± 0.27 8.28 46.18 Chlorfenapyr 3.39 ± 0.58 43.39 132.6
49
Table 3.4: Toxicity of the botanical insecticides against the larvae of DBM collected
from Karak, Pahang (Dzolkifli, 2004)
Insecticide b ± S.E. LC50 (µg/ml) LC95 (µg/ml) Azadirachtin 2.30 ± 0.24 12.11 63.00 Rotenone 4.10 ± 0.43 6.13 15.45 Chlorfenapyr 2.07 ± 0.23 10.37 63.76
In the toxicity study of rotenone against the diamondback moth collected from
Kluang and Karak, both strains showed LC50 of 8.28 and 6.13 µg/ml respectively as
shown in Table 3.3 and Table 3.4. These values are comparable to the standard
compound used against this insect. The study indicated that the diamondback moth
(DBM) has yet to develop resistance to rotenone as oppose to chlorfenapyr.
Figure 3.6 The leaf-dipped method
Figure 3.7 The larvae of diamondback moth (Plutella xylostella)
50
3.4.4.2 Field efficacy studies
The field trial was conducted in the vegetable farm at University Putra Malaysia
(UPM), Serdang, Selangor from February to April 2004. The treatments were made one
month after transplanting the cabbages to the field. The treatments used a simple
formulation of rotenone at 30.0 g A.i/hectare and 15.0 g A.i/hectare and spinosad as a
standard. They were sprayed at 450 L/hectare. Each treatment consisted of 4 replicates.
The assessments were made at 3, 7 and 10 days after treatment (DAT) by sampling the
number of caterpillars. The experimental design was RCBD (Randomized Complete
Block Design). Data collected were subjected to ANOVA and means were compared by
LSD. Figure 3.8 shows the number of larvae of Spodoptera litura following the
application of chemicals. Figure 3.8 indicated that rotenone at 30.0 g A.i/hectare
showed better control of the larvae as compared to the control and the performance was
comparable to the standard spinosad. The study indicated that a simple laboratory
formulation of rotenone was able to provide a good control of the larvae of Spodotera
litura. Further field study using the conclusive formulation is ongoing extensively.
0.62
5
1.12
50.
625
0.53
1
0.37
50.
156
0.12
50.
5
0.31
30.
219
0.31
250.
218
0.15
60.
410.
060.
25
0.12
50.
340.
090.
25
00.
160.
09 0.15
0.03
0.06 0.
130.
06
00.10.20.30.40.50.60.70.80.9
11.11.2
Num
ber
of la
rvae
/pla
nt
0 1 3 7 1 3 7Days After Treatment (DAT)
Rtn30g/ha Rtn15g/ha Spinosad Control
Figure 3.8 Field efficacy result of the formulated rotenone against
Spodotera litura (Dzolkifli, 2004)
CHAPTER IV
METHODOLOGY
4.1 Introduction
The experimental work was carried out in three phases which are the
preliminary, optimization and verification. The processing parameters or independent
variables studied were the types of solvent, solvent-to-solid (ml/g) and raw material
particles size (mm in diameter). The other relevant processing parameters involved for
instance the extraction temperature (0C), weight of raw material (g) and extraction
duration (hour) were fixed as control parameters based on the literature reviews and
exploratory studies. The response variables or the dependent variables were the yield of
rotenoids resin in dried roots; % (w/w), yield of rotenone in dried roots; % (w/w) and
biological activity (LC50) of rotenoids resin.
52
4.1.1 Preliminary experiments
The preliminary study was implemented to identify the most relevant
independent and dependent variables as well as to determine the appropriate range of the
operating conditions. The preliminary and control processing parameters are shown in
Table 4.1 and Table 4.3 respectively. The preliminary experiment to obtain the
rotenoids resin was based on exploratory experiment carried out by Saiful et al. (2003)
as shown in Table 4.2. Table 4.4, Table 4.5 and Table 4.6 show the preliminary
experimental design and response variables for the solvent-to-solid ratio of 3.3 ml/g and
10.0 ml/g and two types of raw material particles size which are coarse (2.0 mm to 5.0
mm) and fine (0.5 mm to 2.0 mm) in diameter.
Table 4.1: Preliminary processing parameters Factor names Factor levels aTypes of solvent Chloroform, ethanol and acetone bSolvent-to-solid ratio 10.0 ml/g and 3.3 ml/g cRaw material particles size Fine and coarse particles size (mm in diameter) Extraction duration 0 to 1440 mins (2 hours interval observation)
aPurity of the solvents were 95.0 % (v/v). bThe solvent-to-solid ratio of 3.3 ml/g and 10.0 ml/g were selected to evaluate the significant effect on the response variables as compared to the ratio carried out by Grinda et al. (1986) and Saiful et al. (2003). cSource: Pagan and Hageman (1949): (a) Fine; 0.5 mm to 2.0 mm in diameter (b) Coarse; 2.0 mm to 5.0 mm in diameter.
Table 4.2: The preliminary experiment to obtain the rotenoids resin based on the
exploratory experiment carried out by Saiful et al. (2003)
aParameters Parameter values Types of solvent Industrial grade acetone 95.0 % (v/v) Solvent-to-solid ratio 10.0 ml/g bRaw material particles size Fine (2.0 mm to 0.5 mm in diameter) Extraction duration 24 hours Extraction temperature Ambient (26 ± 2 0C)
aA Concentrated Liquid Crude Extract (CLCE) from this method was subjected to the Brine Shrimp Lethality study to obtain the Lethal Concentration (LC50) of rotenoids resin. bSource: Pagan and Hageman (1949).
53
Table 4.3: Preliminary control processing parameters Factor names Factor levels Weight of raw material 30.0 g of dried roots Extraction temperature Ambient (26 ± 2 0C)
Table 4.4: Experimental design for the solvent-to-solid ratio of 3.3 ml/g Solvent/particles size Coarse Fine aAcetone A1 A2 aEthanol + oxalic acid + H2O B1 B2 aChloroform C1 C2
aPurity of the solvents were 95.0 % (v/v).
Table 4.5: Experimental design for the solvent-to-solid ratio of 10.0 ml/g Solvent/particles size Coarse Fine aAcetone A3 A4 aEthanol + oxalic acid + H2O B3 B4 aChloroform C3 C4
aPurity of the solvents were 95.0 % (v/v).
Table 4.6: Preliminary response variables
aResponse variables Response values Yield of rotenone in dried roots % (w/w) Concentration of rotenone mg/ml
aThe yield of rotenoids resin in dried roots was not included due to the sampling procedures for 2 hours interval produced insufficient volume (ml) of LCE for the concentration process (to obtain the rotenoids resin). The rotenoids resin was obtained separately based on the exploratory experiment carried out by Saiful et al. (2003) as shown in Table 4.2.
4.1.2 Optimization phase
Based on the preliminary study, an experimental design was developed based on
Central Composite Design (CCD) using the factorial design of three factors at two
levels. Design-Expert® software version 6.0 (Stat-Ease, 2002) was used to design and
54
interpret the results. Experiments were then carried out based on the experimental
matrix design. The process description is discussed in section 4.3.
4.1.2.1 Design of Experiments (DOE)
The Design of Experiment (DOE) is a critical aspect of the research
methodology as it allows for experiments to be designed such that the minimal amount
of experiments can be carried out while extracting the maximum amount of information
(Kuehl, 2000). The design of experiments used in this study was a Central Composite
Design (CCD) of Response Surface Method (RSM). The factorial design of 2k with
three factors at two levels (23) including three centre points, two replicates and one alpha
point (α) were implemented. Table 4.7 shows the CCD specification for the
optimization phase experiment. Furthermore, the Design of Experiment (DOE) is used
to evaluate the effects of several different factors on response variables. Statistical tool
such as analysis of variance (ANOVA) is used to analyze the data from the experiments
and to make decisions about whether a given factor has a significant impact on the
response variable.
Table 4.7: Specification of Central Composite Design (CCD)
TOTAL OF EXPERIMENT = [2n + 3 CP (Centre Point)] × 2 Replicates + (2n × (α))
= (23 × 2 Replicates) + (3 CP × 2 Replicates)
+ [(23 × 1(α)]
∴ = 16 + 6 + 8 = 30 experiments.
Centre Point (CP) Alpha Point (α) Replicate 3 1 2
55
4.1.2.2 Factors and experimental matrix
Processing parameters or independent variables are synonyms. However, factor
is the most frequently used term in the design of experiments. The factors involved in
this study were the types of solvent, solvent-to-solid ratio (ml/g) and raw material
particles size (mm in diameter). Each factor had two levels.
Meanwhile, response variables or dependent variables are measurable
characteristics of the product or process to be studied. Thus, the objective of this study
was to determine which particular factors mostly affected the response variables. On top
of that, coding schemes are normally used to denote one level of a factor. For instance,
+1/+α is used to denote the high level while -1/-α is used to denote the low level.
These levels are predetermined based on the literature work, exploratory and
preliminary study. Experiments were carried out by changing the levels of each factor
and measuring whether and by how much the response changes. The optimization
processing parameters, control processing parameters, response variables and
experimental matrix of the study are shown in Table 4.8, Table 4.9, Table 4.10 and
Table 4.11 respectively.
Table 4.8: Optimization processing parameters Factors Factor names aFactor levels X1 cTypes of solvent bEthanol (-α) and acetone (+α)
X2 Solvent-to-solid ratio d10.0 ml/g (-α) and e2.0 ml/g (+α) X3 fRaw material particles size 0.5 mm (-α) and 5.0 mm (+α) in diameter
aFactor levels range is denoted as (α): +α = The highest level; -α = The lowest level. bEthanol was added with the H2O and oxalic acid - A ratio of ethanol (9): H2O (1) [Prepare 1.0 mg/ml of oxalic acid solution from the volume (ml) of H2O ratio (1)]. cPurity of the solvents were 95.0 % (v/v). dSource: Saiful et al. (2003). eSource: Grinda et al. (1986). fSource: Pagan and Hageman (1949); Maas (1938) and Moore (1940).
56
Table 4.9: Optimization control processing parameters Factor names Factor levels Extraction duration Exhaustive extraction time (14 hours) Weight of raw material 30.0 g of dried roots Extraction temperature Ambient (26 ± 2 0C)
Table 4.10: Optimization response variables Dependent/response variables Response values Yield of rotenoids resin in dried roots % (w/w) Yield of rotenone in dried roots % (w/w) Biological activity, LC50 ppm
57
Table 4.11: Experimental matrix for the extraction rotenoids resin: CCD (23)
Standard run Replicate CP and α aX1 bX2
cX3 Run 1 (27) α Acetone 6.00 5.00 Run 2 (10) α Ethanol 10.00 2.75 Run 3 (6) 1 Ethanol 3.62 4.09 Run 4 (17) 2 Acetone 3.62 1.41 Run 5 (8) 3 Ethanol 8.38 4.09 Run 6 (24) α Acetone 2.00 2.75 Run 7 (20) 4 Acetone 3.62 4.09 Run 8 (7) 3 Ethanol 8.38 4.09 Run 9 (16) 2 Acetone 3.62 1.41 Run 10 (14) CP Ethanol 6.00 2.75 Run 11 (12) α Ethanol 6.00 5.00 Run 12 (4) 5 Ethanol 8.38 1.41 Run 13 (30) CP Acetone 6.00 2.75 Run 14 (5) 1 Ethanol 3.62 4.09 Run 15 (11) α Ethanol 6.00 0.50 Run 16 (21) 4 Acetone 3.62 4.09 Run 17 (28) CP Acetone 6.00 2.75 Run 18 (13) CP Ethanol 6.00 2.75 Run 19 (19) 6 Acetone 8.38 1.41 Run 20 (23) 7 Acetone 8.38 4.09 Run 21 (26) α Acetone 6.00 0.50 Run 22 (3) 5 Ethanol 8.38 1.41 Run 23 (15) CP Ethanol 6.00 2.75 Run 24 (29) CP Acetone 6.00 2.75 Run 25 (22) 7 Acetone 8.38 4.09 Run 26 (1) 8 Ethanol 3.62 1.41 Run 27 (18) 6 Acetone 8.38 1.41 Run 28 (9) α Ethanol 2.00 2.75 Run 29 (25) α Acetone 10.00 2.75 Run 30 (2) 8 Ethanol 3.62 1.41 TOTAL 16 6 CP & 8 α
aX1 - Types of solvent. bX2 - Solvent-to-solid ratio (ml/g). cX3 - Raw material particles size (mm in diameter). TOTAL OF EXPERIMENT = 16 (including replicates) + 6 CP (including replicates) + 8 α
= 30 experiments of CCD
58
4.1.3 Verification phase
The verification phase was then carried out based on the results obtained from
the optimization phase. The experiments were carried out in two replicates and the
result obtained verified the selection of the most appropriate processing parameters.
4.2 Sampling
Matured Derris elliptica roots (1.0 cm to 1.5 cm in diameter) were obtained from
the rubber and palm oil plantations estate at the Kota Johor lama, Johor. The first batch
(200 g) of dried roots was used in the preliminary phase and the second batch (1000 g)
of dried roots was used in the optimization and verification phase.
4.3 Process description
There were three major steps involved in the preliminary and verification phase
experiment which are the pre-processing of Derris roots, extraction of rotenoids resin
and analysis of response variables. As for the optimizations phase experiment, there
were four major steps involved which are the pre-processing of Derris roots, extraction
of rotenoids resin, analysis of response variables and statistical analysis. The
diagrammatic of the experiment phases are as shown in Figure 4.1 and the flow diagram
of the study as shown in Figure 4.2.
59
Figure 4.2 Flow diagram and overview of the study
PRE-PROCESSING
PROCESSING/EXTRACTION
ANALYSIS OF RESPONSE VARIABLES
*STATISTICAL ANALYSIS
Figure 4.1 Phases of the experiment
PRELIMINARY PHASE
OPTIMIZATION PHASE
VERIFICATION PHASE
(1)
(2)
Preliminary, optimization and verification phase were involved
(2)
(3)
Preliminary, optimization and verification phase were involved
*Only optimization phase was involved
60
4.3.1 Pre-processing of Derris roots
An important aspect of the phytochemical processing is the pre-processing of the
herbal material prior to the extraction. The treatment of the herbal material affects the
viability of the phytochemicals and extraction yield. For the preliminary phase, the first
batch of Derris roots as shown in Figure 4.3 were sifted and separated into two main
particles size which are fine (0.5 mm to 2.0 mm) and coarse (2.0 mm to 5.0 mm) in
diameter. As for the optimization phase, the second batch of Derris roots were sifted
and separated into varies particle size as listed in Table 4.11. Furthermore, raw material
particles size for the optimization phase experiments were sifted and measured manually
using the small measuring tape (mm). The amount of 30.0 g of small pieces of dried
roots was constantly used for each treatment in the preliminary, optimization and
verification phase. The samples were dried separately in a forced air oven 1375 FX
(Sheldon Manufacturing, Inc.) at 30 0C for 3 hours. Derris roots was dried to prevent
fungal infestation and to rupture the cells which lead to faster extraction. Similarly
grinding reduces the size of the particles which increases the yield by increasing the
surface area, reducing diffusion distance and rupturing the cells. The dried Derris roots
were kept in dark prior to the extraction process.
Figure 4.3 Various particles size of Derris roots (Pagan and Hageman, 1949)
(A) Coarse (5.0 mm to 4.0 mm in diameter) (C) Fine (2.0 mm to 0.5 mm in diameter)
(B) Intermediate (4.0 mm to 2.0 mm in diameter)
61
4.3.2 Extraction of rotenoids resin
Normal Soaking Extraction (NSE) method or known as a maceration technique
was used to extract the rotenoids resin. For the preliminary phase, 30.0 g of different
Derris roots sizes were added to the 300 ml of organic solvent (ethanol + oxalic acid
solution, acetone and chloroform respectively) with the solvent-to-solid ratio of 10.0
ml/g in the 500 ml of PYREX® glassware. The whole extraction vessels were insulated
securely with the aluminium foil and covered with plastic to avoid evaporation of the
volatile solvent. Subsequently, the extraction process was carried out in ambient
temperature of 26 ± 2 0C by placing the extraction vessel into a dark cabinet for 24
hours. For a reference, the processing parameters, control parameters and response
variables for the preliminary experiment are shown in Table 4.1 to Table 4.6. While for
the optimization phase, the types of solvent, raw material particles size and solvent-to-
solid ratio were based on the experimental matrix as shown in Table 4.11. The weight
of raw material and extraction temperature were similar as in the preliminary experiment
except for the response variables and extraction duration. The kinetic extraction curves
were constructed in the preliminary experiment to obtain the exhaustive extraction time
by checking the yield of rotenone (mg) and rotenone concentration (mg/ml) for 2 hours
interval using the reversed-phase HPLC. From the result, the extraction duration was set
to be as a control parameter in the optimization experiment which is 14 hours. Then, the
extracts were filtered using the Whatman filter paper no. 4 with the aid of Vacuumbrand
GmBH+CO (2.20 m3/hour) at 0.3 mbar. The Liquid Crude Extract (LCE) was subjected
to the rotenone content (mg) determination prior to the concentration process. Next, the
extracts were introduced into a round-bottom flask with reflux condenser of the rotary
evaporator under reduced pressure of 0.3 mbar at 40 0C to remove approximately 90 %
of the solvent to obtain the Concentrated Liquid Crude Extract (CLCE). This CLCE was
subjected again to the rotenone content (mg) determination prior to the lethality study
using RP-HPLC and introduced again into rotary evaporator to remove 100 % of solvent
to obtain the rotenoids resin. The rotenoids resin was weighed to determine the yield of
extract (g). Finally, the rotenoids resin was stored in a cool room (10 0C). The
extraction flow process is shown in Figure 4.4.
62
Figure 4.4 Extraction of rotenoids resin from Derris elliptica roots
Dried Derris roots in small pieces (based on the particles size of optimization phase experiment)
A volume of solvent (acetone, ethanol + oxalic acid solution and chloroform respectively) was added to 30.0 g of sample based on the solvent-to-solid ratio of optimization phase experiment
Normal Soaking Extraction (NSE) for 24 hours (PRELIMINARY PHASE) and 14 hours (OPTIMIZATION PHASE) with ambient temperature and stored into a dark cabinet
Normal Soaking Extraction (NSE) method
(1)
(2)
Collection of the LCE for 2 hours interval *ONLY FOR PRELIMINARY EXPERIMENT
The Liquid Crude Extract (LCE) was filtered using the Vacuumbrand GmBH+CO (2.20 m3/hour) at 0.3 mbar
Evaporated in the rotary evaporator (Heidolph-Laborata 4001) under reduced pressure of 0.3 mbar [Water bath heater temperature was set to be 40 0C for 15 mins to remove ≅ 90 % solvent]
Concentrated Liquid Crude Extract, CLCE (mg/ml)
Rotenoids resin (g)
RP-HPLC: Rotenone content determination
RP-HPLC: Rotenone content determination
Evaporated in the rotary evaporator (Heidolph-Laborata 4001) under reduced pressure of 0.3 mbar [Water bath heater temperature was set to be 40 0C for 10 mins to remove ≅ 100 % solvent]
(3)
(4)
(5)
(6)
(7)
Brine shrimp lethality study (LC50)
63
4.3.3 Analysis of the response variables
Product analysis was carried out on the response variables namely the yield of
rotenoids resin in dried roots; % (w/w), yield of rotenone in dried roots; % (w/w) and
biological activity (LC50) of rotenoids resin as shown in Figure 4.5.
Figure 4.5 Evaluation of rotenoids resin
4.3.3.1 Determination of extraction yield (rotenoids resin)
Yield of extraction was calculated as percentage of rotenoids resin obtained from
the weight (g) of dried roots. The yield resin was calculated as follows:
Yield of resin, (% w/w) = x (g) of resin obtained/(g) of dried roots × 100 % (4.1)
Yield of rotenone in dried roots; % (w/w)
Yield of rotenoids resin in dried roots; % (w/w)
Biological activity (LC50): ppm [Brine Shrimp Lethality Study]
Reversed Phase High Performance Liquid Chromatography (RP-HPLC):
[External standard method]
Vacuum Liquid Chromatography- Thin Layer Chromatography (VLC-TLC)
[Semi-automatic TLC CAMAG Linomat 5]
PRODUCT ANALYSIS
(1)
(2)
(3)
(a) (b)
64
4.3.3.2 Determination of extraction yield (yield of rotenone)
Determination the yield of rotenone (mg) was carried out qualitatively using the
Thin Layer Chromatography (TLC) and quantitatively using the reversed-phase High
Performance Liquid Chromatography (RP-HPLC).
(a) Qualitative analysis of rotenone using Thin Layer Chromatography (TLC)
Qualitative analysis of rotenone was carried out in the preliminary phase in order
to confirm the presence of four major constituents in the Concentrated Liquid Crude
Extract (CLCE) which are rotenone, 12αβ-rotenolone, tephrosin and deguelin. The
mobile phase comprising of petroleum ether and ethyl acetate; 4:2 (v/v) were prepared
and left to equilibrate for at least 35 mins. Silica plates from Merck, Germany were
used as stationary phase. Standard rotenone (Rotenone PESTANAL®; Analytical grade;
96.2 % (w/w); SIGMA-Aldrich™) and the Concentrated Liquid Crude Extract (CLCE)
that has been purified using the Vacuum Liquid Chromatography (VLC) were spotted on
to the base line cycle drawn on the TLC plate. After the development, the spots were
observed under the Ultra Violet (UV) lamp with wavelength (λ) of 254 nm and 365 nm
(Cole-Parmer 9818 Series, Illinois). Retardation factor (Rf) was calculated from the
plates as shown in the Equation 4.2. The non-conventional techniques to spot the
developed samples and standard solution are shown in Figure 4.6. The complete
experimental work and final results are shown in Appendix I.
Retardation factor, Rf = Distance from the baseline to the centre of the zone (a) (4.2)
Distance from the baseline to the solvent front (b)
65
Figure 4.6 Techniques of spotting the sample on the silica plate
(Harborne, 1984; Hougthon and Raman, 1998)
(b) Quantitative analysis of rotenone using High Performance Liquid
Chromatography (HPLC): Measurement of the rotenone content (mg)
The external standard method was implemented in order to identify the amount
of rotenone in the dried roots of Derris elliptica as shown in Figure 4.7. The settings of
reversed-phase HPLC recommended by Baron and Freudenthal (1976) to separate
rotenone and other toxic constituents are shown in Table 4.12. As far as the analysis is
concern, this method is more reliable, economical, low time consumption and easier to
handle as compared to the internal standard method. There are some essential
requirements should be applied such as the standard solution must contain all eluents to
be quantified, standard eluents should be similar concentration as unknowns, the
standard and sample matrix should be as similar as possible and analysis conditions must
be identical.
66
Table 4.12: Parameters of RP-HPLC recommended by Baron and Freudenthal (1976)
Parameters Setting Column temperature (0C) Ambient Flow rate of separation 0.4 ml/min UV wavelength (λ) 294 nm Injection volume 5.0 µl Amplitude Unit Full Scale (AUFS) 2.0
Rotenone standard [SIGMA-Aldrich™; purity of 95 - 98 % (w/w)]: Rotenone standard concentration (Cstd) = X (mg/ml)
Peak area (A) = Y (mV.s) Sensitivity Factor (SF) = (A)/(Cstd) = J (mV.s. × ml/mg)
Sample concentration (LCE or CLCE):
Sample peak area (Asample) = Y (mV.s) Sample concentration (Csample) = (Asample)/SF = Q mg/ml.
If, the sample involves dilution:
(Dilution factor; DF = flask volume/pipette volume).
Actual concentration (rotenone) = Q mg/ml × DF = G (mg/ml) (10.0 mg/ml = 1.0 %) - How to get 1.0 %: 10/1000 (g/ml) × 100 % = 1.0 %
YIELD (rotenone) = Csample (mg/ml) × volume (ml) of LCE or CLCE
= (G) mg/ml × (ml) = K (mg)
**YIELD OF ROTENONE IN DRIED ROOTS, % (w/w) = Yield of rotenone (mg)/weight of raw material × 100 %
**An example of external standard method calculation are shown in Appendix H
Figure 4.7 External standard method template calculations
67
(c) Biological activity (LC50) of rotenoids resin
Bio-active compounds are almost always toxic in high doses (McLaughlin and
Rogers, 1998). Thus, in vivo lethality in a simple zoological organism can be used as a
convenient monitor to screen and fractionate the bio-active constituents for rapid new
discovery. The eggs of brine shrimp, Artemia salinas were readily available in pet shops
at low cost and remain viable for years in dry condition. Upon being placed in seawater,
the eggs hatch within 48 hours to provide large numbers of larvae (nauplii) for
experimental use. The sea salt was obtained from SIGMA-Aldrich™ (S-9883) and
prepared by dissolving 30.0 g of sea salt into 3000 ml of deionized water (SG: 0.01
g/ml). The small tank (hatching chamber) was used to grow shrimp with dividing dam,
cover and lamp to attract shrimp eggs. Allow three days for the shrimp to hatch and
mature as nauplii. The Concentrated Liquid Crude Extract (CLCE) concentration
(mg/ml) was the initial concentration and the dilution was prepared accordingly to the
test concentration as shown in Figure 4.8. Furthermore, two test vials for each
concentration were prepared with a total of 12 vials plus one control vial. From the
diluted solutions, 100 µl was transferred to the test vials corresponded to 1000, 500, 100,
50, 10, 1.0 µg/ml respectively. Solvent from the test vials were evaporated under dark
place at ambient temperature (26 ± 2 0C). Evaporation is needed so that only rotenoids
resin is remaining in the test vials. After 4 days (when the shrimp larvae were ready),
about 2.0 ml of seawater were added to each test vial. Then, 10 shrimps were added for
each test vial and the volume was adjusted with the seawater to 4.0 ml/vial. The vials
were placed, uncovered and under the lamp. The lamp must be sufficient and does not
overheat the vials. The time of treatment was 6, 12 and 24 hours and the number of
survivors or dying for each time was counted and recorded. Subsequently, the mortality
of Artemia salina was evaluated to determine the LC50 values using the probit analysis
(Finney, 1971) based on the dose-response curves as shown in Figure 4.9 (A). By
plotting the probit values against log10 dose of rotenoids resin (ppm) as shown in Figure
4.9 (B), a least-squares simple linear regression of logarithm can be obtained and the
LC50 can be calculated by anti-log10 the concentration (ppm) [Anti-log10 (ppm)].
69
No.
of m
orta
lity
(%)
Mor
talit
y (p
robi
t)
Figure 4.9 Mortality of Artemia salina when exposed to the extracts of Derris
elliptica: (A) dose response curve (Van, 1990); (B) probit analysis curve (Finney, 1971)
4.4 Statistical analysis
All experiments in this study involved replicates. In the optimization phase
experiments, analysis of variance (ANOVA) and Response Surface Methodology (RSM)
were carried out in order to determine the significance of the results. The RSM was
carried out based on the Design of Experiments (DOE) generated by the Design-Expert®
software version 6.0 (Stat-Ease, 2002).
0 10 50 100 500 1000 Dose treatment (ppm)
50 %
100 %
0 1.00 1.70 2.00 2.70 3.00 Log-dose (ppm)
50 % mortality
LC50 = Anti-log10 (C)
100 % mortality *Linear regression of logarithm-converted concentrations (x) versus the probit values (y); y = mc ± c
(B)
(A)
Sigmoidal curve
70
4.4.1 Response Surface Methodology (RSM)
Response surface methodology (RSM) is a set of techniques designed to find the
best value of the response. RSM is a statistical-mathematical method which uses
quantitative data in an experimental design to determine and simultaneously solve
multivariate equations to optimize processes or products (Mizubuti et al., 2000). If
discovering the best value of the response is beyond the available resources of the
experiment, then response surface methods are used to at least to gain a better
understanding of the overall response system and locate the optimum response region.
RSM could not replace the statistical tool of ANOVA but ANOVA supported the RSM.
A response surface is represented by the mathematical equation or a model called
polynomial. Another way to summarize the information in a response surface is to
create a contour plot or three-dimensional graphs (Cornell, 1990). A Central Composite
Design (CCD) can be used to locate a maximum when the centre (0, 0, 0) is used as the
mid point to allow a second order surface to be fitted (Kuehl, 2000). Main and
interactions effects can be determined by constructing graphs of the average response
versus the factor level. The graph is called effects plots. Effects plots do not take place
of statistical tests. When statistical testing of ANOVA shows that an interaction effect is
significant, then the results of the experiment must be interpreted by examining the
interaction plots, not the main effects plots.
The main effect of a factor is the effect occurred between the levels of the factor
itself. Statistically, the main effect of a factor is defined as the averages response value
for all test runs at the level of the factor minus the average response value for runs at the
low level of the factors (Devore and Farnum, 1999). The interaction effects occurred
when one factor influence the other which results in different response if compared to
one factor alone. Statistically, two factor interaction effects are one-half of the
difference between the main effects of the one factor calculated at the two levels of the
other factor. In multifactor designs, main effects and two factor interactions are
interpreted in the same manner as in two factor designs. With more than two factors, the
71
opportunity arises to incorporate even higher order interactions between three or more
factors. The highest order interactions are the least likely to be important. In this study,
the main and interactions effects are all summarized in the ANOVA table and the
response surface graphical tools.
4.4.2 Model adequacy checking
Several aspects need to be checked to ensure a model developed is adequate.
These aspects are the F-distribution test, coefficient of determination (R2 and adjusted
R2), residual analysis and lack-of-fit test which can be obtained straight away from the
ANOVA table. If all the aspects are not qualified, meaning that the experiment done
may consist errors and have to be repeated in order to obtain the accurate result
(Rahman, 2004).
4.4.2.1 F-distribution test
F-distribution test and the corresponding P-value are used to test the hypothesis
for each source of terms. F-table is set for 5.0 % and 1.0 % significant levels. If 5.0 %
of significant level is choose, value of ‘Prob > F’ less than 0.05 indicate model term is
significant (Rahman, 2004).
72
4.4.2.2 Coefficient of multiple determinations (R2)
Coefficient of determination consists of two values which are the coefficient of
multiple determinations (R2) and adjusted R2. The coefficient of multiple
determinations (R2) is a measure of the amount of variation around the mean explained
by the model while adjusted R2 is a measure of variation in the dependent variables for
which the model accounts (Rahman, 2004). A good model should have R2 > 80 %. The
R2 is an accompanying statistics to the F ratio as in the Equation 4.3.
R2 = 1 - SSR/SST (4.3)
where;
SSR: Sums of Squares of Residuals (unexplained variation).
SST: Sums of Squares of Total Variation.
The R-square (R2) value is an indicator of how well the model fits the data or
indicates a good correlation or relationship between model values and experimental
values. The R-square value is always 0 to 1.0. For example, an R-square (R2) close to
1.0 indicates that almost all of the variability with the variables specified in the model
has been accounted (Devore and Farnum, 1999).
4.4.2.3 Lack-of-fits test
The lack-of-fit test compares the residual error to the pure error from the
replicated design points. If there is a significant lack-of-fits, as shown by a low
probability value ‘Prob > F’, the model is not suitable to be used as a response predictor
(Rahman, 2004).
73
4.4.3 Pearson’s correlation coefficient, r
Correlation is a technique for investigating the relationship between two
quantitative or continuous variables. It is a measure of strength of the association
between the two variables increase or decrease together, whereas negative correlation
indicates that as one variable increases, so the other decreases and vice versa (Cornel,
1990). The formula of Pearson’s correlation coefficient, r is shown by the Equation 4.4.
The correlation coefficient is always -1 and +1. The closer the correlation is to +/- 1, the
closer to a perfect linear relationship or they are uncorrelated (Cornel, 1990).
(4.4)
( ) ( )
( ) ( )⎟⎟⎠
⎞⎜⎜⎝
⎛∑
∑−∗⎟⎟⎠
⎞⎜⎜⎝
⎛∑
∑−
∑∑∗∑−
=
NYY
NXX
NYXXY
r2
22
2
CHAPTER V
RESULT AND DISCUSSION
5.1 Introduction
This chapter presents the results of the effect of processing parameters on the
relevant response variables namely yield of rotenoids resin in dried roots; % (w/w) and
yield of rotenone in dried roots; % (w/w). The processing parameters considered in the
research were the types of solvent, solvent-to-solid ratio (ml/g) and raw material
particles size (mm in diameter). All experimental results were evaluated and discussed
in accordance to the preliminary phase, optimization phase, multi response variables,
verification phase, comparison and correlation analysis. The biological activity (LC50)
of rotenoids resin for the preliminary phase, verification phase and rotenone standard
(SIGMA-Aldrich™) were also evaluated and discussed although the response was
insignificant to optimize the processing parameters.
75
5.2 Preliminary experiment results
The experimental works were carried out as elaborated in the Chapter IV. The
preliminary experiments were carried out prior to the optimization phase experiment to
determine the appropriate factor levels as well as to investigate other related aspects.
5.2.1 Effects of the plant parts and types of solvent on yield
Figure 5.1 shows the extraction yield of the Normal Soaking Extraction (NSE)
method using different types of solvent. The extraction was carried out using Normal
Soaking Extraction method (NSE) by utilizing three types of solvent (chloroform,
ethanol and acetone) and three types of Derris parts which were fine roots (0.5 mm to
2.0 mm), coarse roots (2.0 mm to 5.0 mm) and stems (0.5 mm to 2.0 mm) in diameter.
The extraction process was done for 24 hours. The result indicate that acetone gives the
highest extraction yield of 1.14 % (w/w) as compared to the chloroform; 0.77 % (w/w)
and ethanol; 0.31 % (w/w). For the optimization phase, acetone and ethanol were
selected due to their capability of extracting less colouring matters, waxes and other
plant material as well as their extraction performance to extract rotenone more than the
other organic solvent.
In Normal Soaking Extraction (NSE) using acetone result, fine roots sample
collected from Kangkar Pulai give the highest yield of 1.14 % (w/w) followed by
Kangkar Pulai coarse roots; 0.60 % (w/w), Kangkar Pulai stems; 0.13 % (w/w), Kulai
roots (combination of fine and coarse roots); 0.33 % (w/w), Kangkar Pulai stems; 0.13
% (w/w) and Kulai stems; 0.13 % (w/w). In Normal Soaking Extraction (NSE) using
ethanol and oxalic acid solution result, once more the fine roots sample collected from
Kangkar Pulai give the highest yield of 0.31 % (w/w) followed by Kangkar Pulai coarse
roots; 0.28 % (w/w), Kulai roots; 0.24 % (w/w), Kulai stems; 0.23 % (w/w) and Kangkar
76
Pulai stems; 0.18 % (w/w). In Normal Soaking Extraction (NSE) using chloroform
result, furthermore indicates that fine roots sample collected from Kangkar Pulai give
the highest yield of 0.69 % (w/w) followed by Kangkar Pulai coarse roots; 0.33 %
(w/w), Kangkar Pulai stems; 0.15 % (w/w), Kulai roots; 0.11 % (w/w) and Kulai stems;
0.06 % (w/w).
1.14
0.31
30.
692
0.60
20.
28 0.33
4
0.32
7
0.23
90.
111
0.12
5 0.22
5
0.06
3
0.13
40.
181
0.15
3
00.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
Yiel
d of
rote
none
in d
ried
root
s; %
(w/w
)
Kangkar Pulai(Fine roots)
Kangkar Pulai(Coarse roots)
Kulai (Roots) Kulai (Stems) Kangkar Pulai(Stems)
Sources of sample
Normal Soaking Acetone; 95 % (v/v)Normal Soaking Ethanol; 95 % (v/v) + Oxalic Acid Solution (1 mg/ml)Normal Soaking Chloroform; 95 % (v/v)
From the results obtained, there were a significant effect on the plant parts and
types of solvent used in the extraction process. Acetone with the solvent strength of 5.1,
extracts more rotenone up to 39.5 % to 72.8 % as compared to the chloroform and
ethanol respectively (Kangkar Pulai fine roots). According to John (1944), the solubility
of pure rotenone in acetone is 0.66 g/ml, ethanol; 0.0002 g/ml and chloroform; 4.72
g/ml. In contrast with John (1944) finding, the anomalies between high solubility of
rotenone in acetone as compared to chloroform are due to three other major constituents
(e.g.: tephrosin, 12αβ-rotenolone and deguelin) which consist in the resin are highly
Figure 5.1 Yield of rotenone in dried roots, % (w/w) using the Normal Soaking
Extraction (NSE) method for different types of solvent
77
soluble in acetone thus dissolved collectively with rotenone in a large amount. On top
of that, chloroform was strongly capable to dissolve rotenone in pure form as compared
to the rotenoids resin. Therefore, the solubility of rotenone in pure form (Rotenone
PESTANAL®; Analytical grade, 96.2 % (w/w); SIGMA-Aldrich™) was tremendously
more soluble in chloroform (4.1) as compared to the other solvents which have high
level of solvent strength. On the contrary, Pagan and Loustalot (1949) proclaimed to
have high value of total chloroform extractives whereby almost all insecticidal
constituents of Derris roots are well dissolved in chloroform as compared to the other
solvents. In conclusion, the results were strongly indicated that acetone extracts more
rotenone and other constituents as compared to the high polarity solvent group. While
chloroform (lower polarity group), it has less solubility on rotenoids (rotenone and other
bio-active constituents) but high solubility of rotenone in pure form. Furthermore, the
plant parts also affected the yield of rotenone in dried roots. The fine Derris roots
(below than 1.0 cm in diameter) was appeared to be the major contributor of rotenone
content (mg) in the acetone extract as compared to the coarse roots (1.0 cm to 2.0 cm in
diameter) and stem (2.0 cm to 3.0 cm in diameter). As for that reason, rotenone was
accumulated greatly either in the roots of fine or coarse and the bio-active constituents in
the roots were always superior as compared to the stem. In overall, the fine roots was
found higher in rotenone content (mg) because of the resin cell tissue that contain the
rotenoids (e.g.: rotenone, tephrosin, 12αβ-rotenolone and deguelin) was reasonably
abundant in roots of small (below than 1.0 cm) and medium in diameter (2.0 cm to 5.0
cm) (Francis and Franklin, 1943; Pagan and Hageman, 1949).
5.2.2 Extraction yield model and the effect of extraction duration on yield
Table 5.1 and Figure 5.2 show response variables result in the kinetic of rotenone
extraction process and can model as a second order of polynomial. All results were
analyzed using the external standard method of reversed-phase High Performance
Liquid Chromatography (RP-HPLC). The example calculation of external standard
78
method is shown in the Appendix H. Figure 5.3 and Figure 5.5 show the kinetic
equilibrium of the rotenone extraction process via yield of rotenone in dried roots, %
(w/w) and yield of rotenone, mg respectively and also modelled as a second order of
polynomial. Figure 5.2, Figure 5.3, Figure 5.4 and Figure 5.5 show that by increasing
the extraction time, the yield of rotenone (mg) and rotenone concentration (mg/ml) were
increased proportionally towards the maximum point. No stagnant phase was observed
during the experiment. It appears that only maximum point was observed at 14 hours
and then gradually decreased the yield of rotenone (mg) until 24 hours. As for the
rotenone concentration (mg/ml), it increased until end of the extraction process to
generate a maximum concentration. In addition, approximately 51.25 % to 52.44 % of
extraction was achieved within 30 mins and 90.0 % within 8 hours. Therefore, the steep
rate of extraction at the beginning was possibly due to the washing (List and Schmidt,
1989; Mircea, 2001) of solute from the ruptured cells rather than leaching alone, where
the phytochemicals released from within the cells by crushing or grinding are quickly
dissolved into the bulk solution. Ultimately, it can be concluded that the exhaustive
extraction of Derris elliptica occurred approximately 14 hours. In contrast with
Suraphon and Manthana (2001), the exhaustive extraction time was occurred at 8 hours
of ethanolic extraction process at room temperature using the stirring soaking method.
Figure 5.3 and Figure 5.5 also show the approximate exhaustive extraction time to
acquire the maximum yield of rotenone in dried roots, % (w/w) and yield of rotenone,
mg which was approximately 840 mins (14 hours) and 895.0 mg/50.0 g respectively.
Thus, the maximum yield of rotenone in dried roots and its concentration were
approximately 1.79 % (w/w) and 2.84 mg/ml respectively.
Figure 5.4 shows the maximum concentration that can be obtained from the
Normal Soaking Extraction (NSE) method which is approximately 3.15 mg/ml at the 24
hours of extraction. The mass of rotenone in this period was approximately 760.0 mg
rotenone/50.0 g dried roots which is fewer approximately 15 % than the mass at the
maximum or exhaustive time (895.0 mg/50.0 dried roots). These losses were possibly
due to the over exposure of room lighting during the extraction sampling and RP-HPLC
sample preparation, unstable ambient temperature, inappropriate insulation of extraction
79
vessel and insufficient RP-HPLC detector. On top of that, the volume of solvent (ml)
used in the Liquid Crude Extracts (LCE) were rapidly decreased within 12 to 20 hours
during the extraction process may perhaps be the explanation why rotenone
concentration (mg/ml) increased immensely. Hence, the solution or Liquid Crude
Extract (LCE) became more concentrate with insufficient amount of rotenone and
possibly promoted further dissipation of rotenone during the concentration process. The
most important consideration to construct an appropriate procedures for extracting the
bio-active constituents that susceptible to heat and light are by installing or constructing
the extraction vessel that monitor systematically critical processing parameters (e.g.:
solvent volatility, operating temperature, pressure and time) in order to minimize the
dissipation of valuable constituents in the extract. Moreover, sample preparation for the
analysis should also be in the control of temperature and light as well as periodically
calibrate the RP-HPLC equipment (e.g.: PDA or UV detector, columns, column heater,
isocratic pump and etc.) with the intention that the result is always precise, reasonable
and acceptable. Table 5.1 and Table 5.2 show the processing parameters involved in the
studies and response variables result respectively.
Table 5.1: Processing parameters involved in the kinetic of rotenone extraction process
Parameters involved in the extraction process: INDEPENDENT VARIABLES
Extraction time: 0 min to 1440 mins (24 hours) DEPENDENT/ RESPONSE VARIABLES
Yield of rotenone, mg Yield of rotenone in dried roots, % (w/w) Rotenone concentration, mg/ml
CONTROL VARIABLES Ambient temperature (26 ± 2 0C) Solvent-to-solid ratio (10.0 ml/g) Weight of dried roots (50.0 g) Types of solvent (Industrial grade of acetone 95.0 % (v/v)) Fine particles size (2.0 mm to 0.5 mm in diameter)
80
Table 5.2: Response variables result in the kinetic of rotenone extraction process
aExtraction time (min)
Volume of LCE (ml)
Yield of rotenone (mg)
Rotenone concentration (mg/ml)
Yield of rotenone in dried roots, % (w/w)
30 470 391.96 0.83 0.78 60 460 435.16 0.95 0.87 90 450 469.85 1.04 0.94
150 430 540.51 1.26 1.08 180 420 594.66 1.42 1.19 210 418 683.43 1.64 1.37 240 406 727.64 1.79 1.46 270 390 642.68 1.65 1.29 300 385 675.64 1.76 1.35 330 375 674.45 1.80 1.35 360 370 635.77 1.72 1.27 420 355 732.80 2.06 1.47 450 345 727.07 2.11 1.45 600 320 816.55 2.55 1.63 660 310 789.88 2.55 1.58 720 300 848.60 2.83 1.70
1380 250 751.45 3.01 1.51 1440 240 756.46 3.15 1.51
aNoted that some of the extraction time was eliminated due to unreasonable response variables
y = -0.0005x2 + 0.9827x + 412.76R2 = 0.8901
0
100
200
300
400
500
600
700
800
900
1000
0 150 300 450 600 750 900 1050 1200 1350 1500 1650
Time, min
Yiel
d of
rote
none
, mg
Figure 5.2 Kinetic equilibrium of the rotenone extraction process (second order polynomial)
81
y = -1e-06x2 + 0.002x + 0.8253R2 = 0.8873
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 150 300 450 600 750 900 1050 1200 1350 1500 1650
Time, min
Yiel
d of
rot
enon
e in
drie
d ro
ots,
% w
/w
Figure 5.3 Kinetic equilibrium of the rotenone extraction process: Yield of rotenone
in dried roots, % (w/w)
y = -2E-06x2 + 0.0038x + 0.747R2 = 0.9807
0
0.4
0.8
1.2
1.6
2
2.4
2.8
3.2
3.6
0 150 300 450 600 750 900 1050 1200 1350 1500 1650
Time, min
Rote
none
con
cent
ratio
n, m
g/m
l
Figure 5.4 Kinetic equilibrium of the rotenone extraction process: Concentration of
rotenone, mg/ml
(A) Exhaustive extraction time: ≅ 840 mins (14 hours) (B) Exhaustive yield of rotenone in dried roots, % (w/w):≅ 1.79 % (895.0 mg/50.0 g) *Different response variables equation model produced slightly different response variables result though at the same extraction time
(A) Exhaustive extraction time: ≅ 840 mins (14 hours) (B) Exhaustive rotenone concentration, mg/ml: ≅ 2.84 mg/ml (C) The maximum of rotenone concentration: ≅ 3.15 mg/ml *Different response variables equation model produced slightly different response variables result though at the same extraction time
(A) and (B)
(A) and (B)
(C)
≅ 1440 mins (24 hours)
≅ 840 mins (14 hours)
≅ 840 mins (14 hours)
82
y = -0.0005x2 + 0.9827x + 412.76R2 = 0.8901
0
100
200
300
400
500
600
700
800
900
1000
0 150 300 450 600 750 900 1050 1200 1350 1500 1650
Time, min
Yiel
d of
rote
none
, mg
Figure 5.5 Kinetic equilibrium of the rotenone extraction process: Yield of rotenone
in dried roots, mg
5.2.3 Effects of the extraction and concentration operating temperature on yield
The easiest way to identify the significant effect on the bio-active constituents
against temperature is using the concentration process. Moreover, the concentration
process is used to produce a rotenoids resin from the Liquid Crude Extract (LCE).
Figure 5.6 indicates that there was a significant effect on the extraction and
concentration operating temperature against the yield of the rotenone (mg) in the LCE.
It appears that rotenone was strongly exaggerated by the operating temperature above
40 0C. The yield of rotenone (mg) in the LCE was reduced by 90.0 % (w/w) for the first
15 mins and gradually decreased until end of the 105 mins concentration process. The
finding was in accordance with Saiful et al. (2003) wherein at 40 0C (operating
temperature under reduced pressure of 0.3 mbar), approximately 13 % yield of rotenone
(A) Exhaustive extraction time: ≅ 840 mins (14 hours) (B) Exhaustive yield of rotenone in dried roots, mg: ≅ 885.50 mg/50.0 g *Different response variables equation model produced slightly different response variables result though at the same extraction time
(A) and (B)
≅ 840 m
ins (14 hours)
83
(mg) in the LCE dissipated after 30 mins of the concentration process. Figure 5.7 shows
the amount of rotenone (mg) in Liquid Crude Extract (LCE) and Concentrated Liquid
Crude Extract (CLCE) before and after the concentration process respectively. On the
contrary, Grinda et al. (1986) have claimed to use the extraction and concentration
operating temperature up to 45 0C for half an hour and retained the yield of rotenone as
much as 14.0 % (w/w) in finely crushed Derris powder. Suraphon and Manthana (2001)
reported that they have used high extraction temperature of 70 0C for 8 hours using the
Soxhlet extractor and managed to get 8.2 % (w/w) rotenone in dried roots. Furthermore,
Visetson and Chuchoui (1999) have also reported that temperatures above 70 0C may
cause decomposition of plant active ingredient. The thermal degradation of this study
however was 40 0C, which is lower than 70 0C obtained by Visetson and Chuchoui
(1999). Besides, Grinda et al. (1986) have found that the usage of high vacuum pressure
of reflux condenser was the best method implemented to date in order to minimize heat
exposure to the extracts and simultaneously obtained high yield of rotenoids resin (g)
and yield of rotenone (mg).
Liquid CrudeExtract (LCE) Concentrated
Liquid CrudeExtract (CLCE)
750.75 mg
657.09 mg
0
100
200
300
400
500
600
700
800
Yiel
d of
rote
none
(mg)
Figure 5.6 Degradation of rotenone content (mg) during the concentration process at
40 0C and 0.3 × 10-3 bar of operating temperature and vacuum pressure respectively
(A) Reduction of rotenone content, %: 750.75 mg - 657.09 mg/750.75 mg × 100 % ≅ 13 %
84
0; 35072
15; 3394
105; 110390; 53875; 254360; 300545; 2864
30; 1300
0
3000
6000
9000
12000
15000
18000
21000
24000
27000
30000
33000
36000
39000
0 10 20 30 40 50 60 70 80 90 100 110 120
The length of concentration process, min
Yie
ld o
f rot
enon
e, m
g
As far as the study is concern, the major factors which contribute to the rotenone
dissipation either during the extraction or concentration process was due to the improper
experimental condition such as over exposure of room lighting and insufficient vacuum
pressure pump (Saiful et al., 2003). Perhaps, improper insulation of the extraction
vessel from the external heat and high volatility of solvent as well as insufficient
vacuum pressure during the concentration process were the major factors which
contribute to the dissipation of rotenone. Hence, it would be the best indicator where if
there was a sudden reduction to the volume of extract, there might be an increased in the
ambient temperature. Furthermore, this type of extraction vessel can be only existed or
constructed specifically for the pilot and industrial scale production and uneconomically
for the laboratory scale study. Hence, an assumption using the insulated glassware of
batch solid-liquid extraction vessel (Normal Soaking Extraction (NSE) method) was
(A) Reduction of rotenone content, %: 3394.0 mg - 35,072.0 mg/35,072.0 mg × 100 % ≅ 90.0 %
(A)
15 mins
Initial yield of rotenone (mg) in the Liquid Crude Extract (LCE)
Figure 5.7 Degradation of rotenone content (mg) during the concentration process at
50 0C and 80 × 10-3 bar of operating temperature and vacuum pressure respectively
85
made by controlling some of the processing parameter (e.g.: extraction temperature)
with the intention to avoid as much as possible rotenone dissipation.
While using the conventional method of Normal Soaking Extraction (NSE),
evaporation of the volatile solvent (e.g.: acetone and chloroform) versus time was barely
undeniable and cause some difficulties to acquire samples for each 30 mins interval due
to insufficient volume (ml) of Liquid Crude Extract (LCE) available in the extraction
vessel. Therefore, the only method available to identify the significant effect of the
extraction and concentration operating temperature against the yield of rotenone (mg)
was by concentrating the Liquid Crude Extract (LCE) using the rotary evaporator (under
reduced pressure of 80 × 10-3 bar and operating temperature of 50 0C). At the same
time, the yield of rotenone (mg) was determined via RP-HPLC for each 15 mins interval
as shown in Figure 5.7. Although there was no data or any extensive research has been
acquired on the rotenone deterioration at varies operating temperature either for the
extraction or concentration process, those processes should not be exceeded 40 0C to
preserve the bio-active constituents (Surya and John, 2001). Furthermore, the key issues
to be taken seriously to acquire high value added products are by minimizing thermal
degradation of the extracts during the upstream processing since phytochemicals are
commonly higher value-lower volume products than the primary metabolites (Manuel,
1985; Surya and John, 2001). As far as the bio-processing is concern, to avoid thermal
degradation are mainly concerning on the dissipation of the protein substances especially
from the plant’s primary metabolite compounds. Hence, the secondary metabolite
compounds are less important as compared to the primary metabolite compounds. As
for the case of rotenone and its derivatives, the possibility of degradation on the yield at
higher temperature (Saiful et al., 2003) as well as the effect on the toxicological values
(Pagan and Hageman, 1949) were the main reason these particular compounds should be
considered to avoid surpassing 40 0C. This is to be sure that only high quality of fine
chemical compounds in the end of product development can be successfully obtained
(Surya and John, 2001). According to the engineering point of view, as the extraction
temperature is increased, it increases the rate of extraction by increasing the internal
diffusion as well as the mass transfer coefficient values and reduced the extraction time
86
(Frank et al., 1999). However, it should be noted that increasing the temperature beyond
certain values led to a decrease in isoflavonoid compounds yield due to the high
susceptibility of the isoflavonoid to high temperature (Cacace and Mazza, 2003). The
finding was in accordance with Cacace and Mazza (2002) wherein the critical
temperature was in the range of 40 0C to 50 0C which contribute to major degradation of
the flavonoids compounds. Therefore, the main constraint in this study was to identify
an appropriate extraction temperature to preserve as much as possible rotenone content
(mg) despite of rotenone is a light and heat sensitive compound. Therefore, when
exposed to light and air, rotenone decomposes into non-toxic dihydrorotenone and water
and resulting in non-insecticidal bio-active compounds (Schnick, 1974). On top of that,
as reported by Grinda et al. (1986), rotenone usually decomposed and detoxify within
one or two weeks and it is difficult to predict in any given condition on how long the
toxicity will remain. In general, high alkalinity (more than pH 8.0 to pH 9.0), high
temperature, abundant light and air and lower concentrations favour rapid dissipation of
rotenone and its half life is predicted to be 3½ hours (rotenone concentration of 0.2 ppm
or 0.2 µg/ml) when exposed to bright sunlight (≅ 30 0C to 40 0C) or on the oven drying
(Grinda et al., 1986). In contrast, Pagan and Hageman (1949) have reported that the
rotenone content of Derris roots which exposure with all treatments (direct sunlight and
oven drying) was practically the same, indicating that all treatments had no effect on the
toxic constituents.
5.2.4 Effects of the raw material particles size and solvent-to-solid ratio on yield
Figure 5.8, Figure 5.9 and Figure 5.10 show a kinetic extraction of rotenone from
Derris elliptica namely as (A) rotenone concentration, mg/ml; (B) yield of rotenone, %
(w/w) using ethanol + oxalic acid solution, chloroform and acetone respectively. All
figures show a significant effect on the raw material particles size (mm in diameter) and
solvent-to-solid ratio (ml/g) against the concentration and yield of rotenone respectively.
87
The yield of rotenone in dried roots, % (w/w) and rotenone concentration, mg/ml was
observed for 2 hours interval using the external standard method of reversed-phase High
Performance Liquid Chromatography (RP-HPLC). The example calculation is shown in
Appendix H. In this study, the yield of rotenone, % (w/w) and rotenone concentration,
mg/ml is defined as follows:
Yield of rotenone, % (w/w) = Yield of rotenone (mg)/weight of raw material (mg) × 100
Rotenone concentration, mg/ml = Yield of rotenone in the LCE (mg)/volume of LCE
A Figures 5.8 indicates that using the ethanol + oxalic acid solution with a
solvent-to-solid ratio of 3.3 ml/g (fine particles size), the highest concentration of
rotenone was obtained at 3.81 mg/ml after 50 hours of the extraction process. This was
followed using a solvent-to-solid ratio of 10.0 ml/g (coarse particles size) which
produced a concentration of 1.14 mg/ml. For a solvent-to-solid ratio of 3.3 ml/g (coarse
particles size), a concentration of 0.72 mg/ml was observed whilst the lowest
concentration of 0.67 mg/ml was produced using a solvent-to-solid ratio of 10.0 ml/g
(fine particles size). For the yield of rotenone in dried roots, a solvent-to-solid ratio of
10.0 ml/g (coarse particles size) produced the highest yield of 0.91 % (w/w) while a
solvent-to-solid ratio of 10.0 ml/g (fine particles size) produced a yield of 0.47 % (w/w).
Using a solvent-to-solid ratio of 3.3 ml/g (fine particles size), a yield of 0.14 % (w/w)
was obtained while a solvent-to-solid ratio of 3.3 ml/g (coarse particles size) produced
the lowest yield of 0.06 % (w/w). A sequence of rotenone concentration (mg/ml)
produced using ethanol + oxalic acid as a solvent were B2 (fine particles size, 3.3 ml/g)
> B3 (coarse particles size, 10.0 ml/g) > B1 (coarse particles size, 3.3 ml/g) > B4 (fine
particles size, 10.0 ml/g) and a sequence of rotenone yield in dried roots, % (w/w) were
B3 (coarse particles size, 10.0 ml/g) > B4 (fine particles size, 10.0 ml/g) > B2 (fine
particles size, 3.3 ml/g) > B1 (coarse particles size, 3.3 ml/g).
88
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Time, hours
Rot
enon
e co
ncen
trat
ions
, mg/
ml
B1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)B2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)B3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)B4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)
Figure 5.8 Kinetics of the rotenone extraction process from Derris elliptica - Ethanol +
oxalic acid solution: (A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w)
B1; Coarse particles size; 3.3 ml/g
00.10.2
0.30.40.50.60.7
0.80.9
1
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54
Time, hours
Yiel
d of
rote
none
, % (w
/w)
B3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)B4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)B1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)B2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)
(A)
B2; Fine particles size; 3.3 ml/g
B4; Fine particles size; 10.0 ml/g
B1; Coarse particles size; 3.3 ml/g
(B) B3; Coarse particles size; 10.0 ml/g
B4; Fine particles size; 10.0 ml/g
B2; Fine particles size; 3.3 ml/g B1; Coarse particles size; 3.3 ml/g
89
A Figure 5.9 indicates that using chloroform with a solvent-to-solid ratio of 3.3 ml/g
(coarse particles size), the highest concentration of rotenone was obtained at 0.84 mg/ml
after 50 hours of the extraction process. This was followed using a solvent-to-solid ratio
of 3.3 ml/g (fine particles size) which produced a concentration of 0.29 mg/ml. For a
solvent-to-solid ratio of 10.0 ml/g (fine particles size), a concentration of 0.24 mg/ml
was produced while the lowest concentration of 0.09 mg/ml was produced using a
solvent-to-solid ratio of 10.0 ml/g (coarse particles size). For the yield of rotenone in
dried roots, a solvent-to-solid ratio of 10.0 ml/g (fine particles size) produced the highest
yield of 0.16 % (w/w) while a solvent-to-solid ratio of 10.0 ml/g (coarse particles size)
produced a yield of 0.07 % (w/w). Using a solvent-to-solid ratio of 3.3 ml/g (coarse
particles size), a yield of 0.06 % (w/w) was obtained while a solvent-to-solid ratio of 3.3
ml/g (fine particles size) produced the lowest yield of 0.01 % (w/w).
A sequence of rotenone concentration (mg/ml) produced using chloroform as a
solvent were C1 (coarse particles size, 3.3 ml/g) > C2 (fine particles size, 3.3 ml/g) > C4
(fine particles size, 10.0 ml/g) > C3 (coarse particles size, 10.0 ml/g) and a sequence of
rotenone yield in dried roots, % (w/w) were C4 (fine particles size, 10.0 ml/g) > C3
(coarse particles size, 10.0 ml/g) > C1 (coarse particles size, 3.3 ml/g) > C2 (fine
particles size, 3.3 ml/g).
90
Figure 5.9 Kinetics of the rotenone extraction process from Derris elliptica - Chloroform:
(A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w)
(A)
(B)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Time, hours
Rot
enon
e co
ncen
trat
ions
, mg/
ml
C1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)C2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)C4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)C3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54
Time, hours
Yie
ld o
f rot
enon
e, %
(w/w
)
C1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)C2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)C3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)C4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)
(B)
(A)
C1; Coarse particles size; 3.3 ml/g
C4; Fine particles size; 10.0 ml/g
C2; Fine particles size; 3.3 ml/g C3; Coarse particles size; 10.0 ml/g
C4; Fine particles size; 10.0 ml/g
C3; Coarse particles size; 10.0 ml/g
C1; Coarse particles size; 3.3 ml/g
C2; Fine particles size; 3.3 ml/g
91
Finally, Figures 5.10 signify that using acetone with a solvent-to-solid ratio of
3.3 ml/g (fine particles size), the highest concentration of rotenone was obtained at 8.30
mg/ml after 50 hours of the extraction process. This was followed using a solvent-to-
solid ratio of 10.0 ml/g (fine particles size) which produced a concentration of 2.14
mg/ml. For a solvent-to-solid ratio of 10.0 ml/g (coarse particles size), a concentration
of 1.20 mg/ml was observed while the lowest concentration of 0.77 mg/ml was produced
using a solvent-to-solid ratio of 3.3 ml/g (coarse particles size). For the yield of
rotenone in dried roots, acetone with a solvent-to-solid ratio of 10.0 ml/g (fine particles
size) produced the highest yield of 1.32 % (w/w) whilst a solvent-to-solid ratio of 10.0
ml/g (coarse particles size) produced a yield of 0.86 % (w/w). Using a solvent-to-solid
ratio of 3.3 ml/g (fine particles size), a yield of 0.14 % (w/w) was obtained while a
solvent-to-solid ratio of 3.3 ml/g (coarse particles size) produced the lowest yield of 0.03
% (w/w).
A progression of rotenone concentration (mg/ml) produced using acetone as a
solvent were A2 (fine particles size, 3.3 ml/g) > A4 (fine particles size, 10.0 ml/g) > A3
(coarse particles size, 10.0 ml/g) > A1 (coarse particles size, 3.3 ml/g) and a progression
of rotenone yield in dried roots; % (w/w) were A4 (fine particles size, 10.0 ml/g) > A3
(coarse particles size, 10.0 ml/g) > A2 (fine particles size, 3.3 ml/g) > A1 (coarse
particles, 3.3 ml/g).
92
Figure 5.10 Kinetics of the rotenone extraction process from Derris elliptica - Acetone:
(A) rotenone concentration, mg/ml; (B) yield of rotenone, % (w/w)
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50Time, hours
Rot
enon
e co
ncen
trat
ions
, mg/
ml
A2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)A1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)A4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)A3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54
Time, hours
Yiel
d of
rote
none
, % (w
/w)
A4 (Fine, 0.5 - 2 mm in diameter - 10 ml/g)A2 (Fine, 0.5 - 2 mm in diameter - 3.3 ml/g)A1 (Coarse, 2 - 5 mm in diameter - 3.3 ml/g)A3 (Coarse, 2 - 5 mm in diameter - 10 ml/g)
A2; Coarse particles size; 3.3 ml/g
A4; Fine particles size; 10.0 ml/g
A1; Coarse particles size; 3.3 ml/g
A4; Fine particles size; 10.0 ml/g
A3; Coarse particles size; 10.0 ml/g
A2; Fine particles size; 3.3 ml/g A1; Coarse particles size; 3.3 ml/g
A3; Coarse particles size; 10.0 ml/g
(A)
(B)
93
All results as shown in Figure 5.8, Figure 5.9 and Figure 5.10 indicate that the raw
material particle size (mm in diameter) affects the extraction rate by increasing the total
mass transfer area when the particle size is reduced (Schwartzberg and Chao, 1982).
Theoretically, it was expected that the fine raw material particles size produced the
highest yield of rotenoids resin as well as the yield of rotenone. Nevertheless, if all the
other processing parameters were accounted, a slight differentiation can be expected
depending on how well the parameters were suited to each others. Hence, the fine
Derris roots with particles size of 0.5 mm to 2.0 mm in diameter were the best root size
to extract a sufficient amount of rotenone for further research purposes. Therefore, in
the optimization phase experiments, a range of 0.5 mm to 5.0 mm in diameter was
selected to obtain the optimum processing parameters on the yield of rotenoids resin in
dried roots; % (w/w), yield of rotenone in dried roots; % (w/w) and biological activity
(LC50) of rotenoids resin. Since the previous exploratory experiments were based on the
amount of rotenone or total extractives of the roots contained, it seems desirable to assay
various particles size biologically and to compare these values with various chemical
criteria.
Moreover, Figure 5.8, Figure 5.9 and Figure 5.10 indicate that there were a
significant effect of the solvent-to-solid ratio (ml/g) against the yield of rotenone in dried
roots; % (w/w). There was a significant increased in the yield of rotenone in dried roots;
% (w/w) with the increased of the solvent-to-solid ratio. The acetone, ethanol + oxalic
acid solution and chloroform extract were observed to have increment in the yield of
rotenone in dried roots; % (w/w) as the solvent-to-solid ratio increased. The rotenone
extraction kinetics using the acetone, ethanol + oxalic acid solution and chloroform at
solvent-to-solid ratio of 3.3 ml/g and 10.0 ml/g are presented in Figure 5.8, Figure 5.9
and Figure 5.10. In contrast, Grinda et al. (1986) have reported that the solvent-to-solid
ratio of 1.6 ml/g gives the yield of rotenone in dried roots up to 14 % (w/w) using the
methylene chloride and octyl stearate. Additionally, the increment of rotenone yield; %
(w/w) with the solvent-to-solid ratio (ml/g) is in accordance with the mass transfer
principles (Cacace and Mazza, 2003). These experiments indicated that the driving
force during mass transfer within the solid is considered to be the concentration gradient,
94
which is greater when the higher solvent-to-solid ratio used, resulting in an increase of
the diffusion rate (Cacace and Mazza, 2003). However, the solvent-to-solid ratio (ml/g)
did insignificantly affected diffusivity wherein the extractions are stopped when the
equilibrium is reached (Cacace and Mazza, 2003). Therefore, by modifying on the
solubility and solute-solvent interactions, the yield of rotenone can be affected
significantly (Cacace and Mazza, 2003). A solid’s solubility is affected by changes in
the activity coefficient, which varies with the temperature and composition of the
solution (Frank et al., 1999). Interactions of the compounds with the solvent could have
modified the activity coefficient and thus the solubility of the compounds. In summary,
the main effect of the solvent-to-solid ratio is to modify the solubility thus increase to
the highest yield of rotenone (mg) at the appropriate solvent-to-solid ratio.
5.2.5 Summary of the preliminary experiments
Based on the preliminary study, the maximum yield of rotenoids resin in dried
roots obtained from the preliminary experiment was 9.50 % (w/w) whilst the maximum
yield of rotenone in dried roots was 1.95 % (w/w) and the biological activity (LC50) of
rotenoids resin was ≤ 1.00 ppm which consider very active constituents against brine
shrimp (Artemia salina). The biological activity (LC50) of rotenoids resin was done
separately due to the sampling procedures for 2 hours interval produced insufficient
volume (ml) of Liquid Crude Extract (LCE) after the concentration process (to acquire
rotenoids resin). Table 5.3 and Table 5.4 show the average yield of rotenone in dried
roots; % (w/w) carried out by Saiful et al. (2003) and the preliminary experiments result
respectively.
95
Table 5.3: The average yield of rotenone in dried roots, % (w/w)
aProcurement places and extraction date bYield of rotenone in dried roots, % (w/w) 15 April 2003, Kangkar Pulai, Johor 1.14 06 July 2003, Felda Taib Andak, Johor 1.60 01 December 2003, Felda Taib Andak, Johor 1.75 11 December 2003, Felda Taib Andak, Johor 2.48 16 March 2004, Kota Johor Lama, Johor 1.86 25 March 2004, Kota Johor Lama, Johor 1.65 26 March 2004, Kota Johor Lama, Johor 1.13 05 January 2005, Kota Johor Lama, Johor 2.60 25 April 2005, Kota Johor Lama, Johor 3.34 Average of rotenone content in dried roots ≅ 1.95
a‘Tuba’ plants were collected in the state of Johor and have been identified as ‘Tuba Kapur’ or Derris elliptica. bAll results were done using the same method of extraction as shown in Table 4.2 (Chapter IV).
Table 5.4: The preliminary experiments result
Dependent/response variables Preliminary experiment results Yield of rotenoids resin in dried roots, % (w/w) ≅ 9.50 (Saiful et al., 2003) Yield of rotenone dried roots, % (w/w) ≅ 1.95 (Saiful et al., 2003) Brine Shrimp Lethality study (McLaughlin and Rogers, 1998) Biological activity, LC50 (ppm) ≤ 1.00
The fine part of Derris roots and types of solvent of acetone and ethanol + oxalic
acid solution were affected the yield of rotenoids resin in dried roots; % (w/w) and yield
of rotenone in dried roots; % (w/w). This suggest that rotenone was accumulated at the
roots part of either in fine or coarse due to the cell tissues that contain the rotenoids resin
were relatively abundant in the small (below than 1.0 cm) and medium (1.0 cm to 2.0
cm) roots part in diameter in the form of milky sap. Meanwhile, the usage of moderate
polarity organic solvent of acetone and ethanol were also significantly affected the
amount of rotenoids resin and rotenone in the extract. Ethanol is added with oxalic acid
solution to facilitate the solubility of rotenone during the extraction process. In addition,
rotenone is an isoflavonoid molecule which contains a sugar molecule tends to be
96
slightly soluble in water. Furthermore, rotenone dissolved in water in the form of liquid
emulsion (milky solution) due to its natural occurring emulsifier in the resin substances.
Hence, the combination of alcoholic solvent with water is usually used to facilitate the
extraction process depending on what particular compounds are extracted (Fasihuddin
and Hasmah, 1993).
Pitigon and Sangwanit (1997) also reported that ethanol is the most suitable
solvent for the rotenone extraction. Besides, these two types of solvent are non-
hazardous solvent, biodegradable, economical and environmental-friendly (less phyto-
toxic). Although the comparison of solvents selectivity towards the yield of rotenoids
resin in dried roots; % (w/w) and yield of rotenone in dried roots; % (w/w) could be
done and selected directly, the results suggested that different solvent would gives
different values. Thus, acetone and ethanol were taken into consideration as a factor
levels in the optimization phase. As for the extraction time, the minimum extraction
time studied in the preliminary phase was 1 hour and the maximum was 24 hours with
2 hours interval. Overall, it shows that the increment of extraction time facilitated the
increment of rotenone content (mg) and rotenone concentration (mg/ml). The
equilibrium time of the exhaustive extraction under ambient temperature was appeared
to be approximately 14 hours and the time was fixed (control) as a time of extraction
process in the optimization phase. Experiments in the preliminary phase also verified
the appropriate temperature for the extraction process in order to avoid any thermal
degradation of the bio-active compounds. The experiments were shown that the
extraction and concentration process should not be exceeded 40 0C to avoid rotenone
dissipation. Hence, temperature for the optimization phase experiments was fixed to an
ambient temperature which is approximately 26 ± 2 0C. There was an additional setup
to the extraction vessel to avoid fluctuation on the ambient temperature during extraction
process. Hence, the whole extraction vessel was insulated with a thick aluminium foil,
covered with plastic and place into dark cabinet to minimize any excess heat from the
surrounding environment. Ambient temperature was monitored using the thermometer
for 1 hour interval. As for the concentration process, the temperature was fixed at 40 0C
under reduced pressure of 0.3 × 10-3 bar.
97
The solvent-to-solid ratio on the preliminary experiments was also shown a
significant effect on the yield of rotenone in dried roots; % (w/w). The experiments
were shown that as the solvent-to-solid ratio (ml/g) increased; the yield of rotenone (mg)
increased significantly until it reached an equilibrium phase. The increment of rotenone
yield (mg) with the increment of the solvent-to-solid ratio (ml/g) is consistent with the
mass transfer principles (Cacace and Mazza, 2003). Theoretically, the yield of rotenone
(mg) is increased in the bulk solution through the increment of solvent volume (ml).
When the yield of rotenone reaches its optimum value (equilibrium), any additional
volume of solvent (ml) would not significantly affect the yield of rotenone in the bulk
solution (Cacace and Mazza, 2003). The final concentration (mg/ml) at that moment
start to decrease as the volume of the solvent keeps increasing. This is the indicator that
all desirable constituents in the roots have been extracted exhaustively. Furthermore, the
solvent-to-solid ratio of 3.3 ml/g was purposely selected to accommodate the significant
effect on the yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone in
dried roots; % (w/w) against the solvent-to-solid ratio carried out by Grinda et al. (1986)
and Saiful et al. (2003) which are 2.0 ml/g and 10.0 ml/g respectively. In order to obtain
the optimum processing parameter, the solvent-to-solid ratio carried out by Grinda et al.
(1986) and Saiful et al. (2003) were selected for the optimization phase experiments so
that the optimum yield of rotenoids resin in dried roots; % (w/w), yield of rotenone in
dried roots; % (w/w) and biological activity (LC50) of rotenoids resin can be achieved.
Meanwhile, the raw material particles size also verified that the fine (below than
2.0 mm) and medium (2.0 mm to 5.0 mm) particles size in diameter promoted the high
yield of rotenone (mg) for the optimization phase experiments. According to Grinda et
al. (1986), they were using the finely crushed Derris roots to produced as much as 14.0
% (w/w) yield of rotenone. The range of raw material particles size to acquire as much
as they have extracted from the finely crushed Derris roots were unavailable and kept
secretly. Moreover, the other Derris species such as Lonchocarpus and Millettia were
mixed during the extraction process to produce high yield of rotenoids resin (g) and
rotenone (mg). In fact, the Derris species that have been used in the exploratory and
preliminary experiments were Derris elliptica (Tuba Kapur) and Derris malaccensis
98
(Tuba Gading). In this study, the plant species were not exactly the same used by
Grinda et al. (1986) due to the unavailability of the species in Malaysia. Therefore, the
yield of rotenoids resin (g) and rotenone (mg) as reported by Grinda et al. (1986) were
noticeably differed with the exploratory and preliminary experiment.
5.3 Optimization phase results: Effect of processing parameters on the
response variables
Effect of processing parameters on the yield of rotenoids resin in dried roots;
% (w/w), yield of rotenone in dried roots; % (w/w) and biological activity (LC50) of
rotenoids resin were discussed in this section. The experiments were carried out based
on the experimental design generated using the Design-Expert® software version 6.0
(Stat-Ease, 2002) to obtain the regression and to analyze graphically the data produced
by the Design of Experiment (DOE). The results were presented in response surface
three-dimensional graphs and contour plot. Since response variables of the biological
activity (LC50) of rotenoids resin appeared to be insignificant, there were only two
response variables studied. Each response variable produced two graphs. All four
graphs must be taken into consideration in order to derive the final result of the effect of
processing parameters on the response variables. The graphs consist of three axes where
Z axes represent the response variable while X and Y axes represent three conditions as
shown in Figure 5.11.
99
Graph 1 = X: Raw material particles size, Y: Solvent-to-solid ratio - Types of solvent A
Graph 2 = X: Raw material particles size, Y: Solvent-to-solid ratio - Types of solvent B
A: Acetone, B: Ethanol + oxalic acid solution
(aNOTE: Response variable of the Brine Shrimp Lethality study was not included due to insignificant result from the ANOVA interpretation)
Figure 5.11 Response surface three-dimensional graphs and contour plot
The contour plot consists of two axes, which represent the processing parameters
as in the three-dimensional graphs. However, the response variable is represented by the
contours. There are many types of contours such as true maximum, stationary ridge,
rising ridge and minimal (Davies, 1954). From the ANOVA table simulated by the
software, the significant effects of each processing parameters (X,Y) towards the
response variables (Z) will be calculated using the F-value and if the effects is
significant, the optimum responses can be interpreted, obtained and the conclusion can
be made conclusively.
aZ: Yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone in dried roots; % (w/w)
100
5.3.1 Effect of processing parameters on the yield of rotenoids resin in dried roots The result of processing parameters (X1, X2 and X3) towards the yield of
rotenoids resin in dried roots was obtained from the experimental data in Table 5.5.
Table 5.5: The design layout and experimental results (Yield of rotenoids resin) Run Factor 1
X1: Solvent-to-solid ratio
Factor 2 X2: Raw material particles size
Factor 3 X3: Types of solvent
Yield of rotenoids resin in dried roots (Experiment)
Yield of rotenoids resin in dried roots (Predicted)
ml/g mm in diameter Treatment % (w/w) % (w/w) 1 6.00 5.00 Acetone 2.48 11.86 2 10.00 2.75 Ethanol 22.77 21.70 3 3.62 4.09 Ethanol 25.83 22.41 4 3.62 1.41 Acetone 2.50 12.37 5 8.38 4.09 Ethanol 28.21 21.82 6 2.00 2.75 Acetone 18.46 12.49 7 3.62 4.09 Acetone 32.69 12.21 8 8.38 4.09 Ethanol 19.34 21.82 9 3.62 1.41 Acetone 3.09 12.37 10 6.00 2.75 Ethanol 23.53 22.19 11 6.00 5.00 Ethanol 23.78 22.06 12 8.38 1.41 Ethanol 26.90 21.98 13 6.00 2.75 Acetone 13.36 12.00 14 3.62 4.09 Ethanol 16.68 22.41 15 6.00 0.50 Ethanol 21.09 22.33 16 3.62 4.09 Acetone 14.45 12.21 17 6.00 2.75 Acetone 11.48 12.00 18 6.00 2.75 Ethanol 31.98 22.19 19 8.38 1.41 Acetone 20.03 11.78 20 8.38 4.09 Acetone 6.15 11.62 21 6.00 0.50 Acetone 26.73 12.13 22 8.38 1.41 Ethanol 10.64 21.98 23 6.00 2.75 Ethanol 27.60 22.19 24 6.00 2.75 Acetone 4.24 12.00 25 8.38 4.09 Acetone 9.49 11.62 26 3.62 1.41 Ethanol 16.97 22.57 27 8.38 1.41 Acetone 8.61 11.78 28 2.00 2.75 Ethanol 7.64 22.69 29 10.00 2.75 Acetone 6.17 11.50 30 3.62 1.41 Ethanol 29.96 22.57
101
These results were analyzed using the Design-Expert® software version 6.0
(Stat-Ease, 2002) for further analysis. Examination of the test summary output revealed
that the linear model was statistically significant for the response and therefore it will be
used for further analysis. The ANOVA performed indicated the model was significant
while the lack of fit was insignificant. Table 5.6 shows the ANOVA final result.
Table 5.6: ANOVA response surface linear model [responses: Yield of rotenoids resin
in dried roots, % (w/w)]
aAnalysis of variance (ANOVA) table for response surface linear model [Partial sum of squares]
The ANOVA table for the effects of processing parameters on the yield of
rotenoids resin in dried roots, % (w/w) is shown in Table 5.6. It shown that the ‘Model
F-value’ of 3.69 implies the model is significant. There was only a 3.0 % (0.03) chance
that a ‘Model F-value’ this large could occur due to noise. Values of ‘Prob > F’ less
than 0.05 (5.0 % confidence limit) indicate model terms were significant. In this case, C
(X3) was a significant model terms. Values greater than 0.10 indicate the model terms
were not significant. If there were many insignificant model terms (not counting those
required to support hierarchy), model reduction may improve the model.
aSource Sum of Squares
DF Mean Square F value Prob > F
Model 782.72 3 260.91 3.69 0.03 A 2.36 1 2.36 0.03 0.86 B 0.16 1 0.16 2.37 × 103 0.96 C 780.20 1 780.20 35.21 0.03 × 10-1 Residual 1837.22 26 70.66 Lack of fit 1220.05 14 87.15 1.69 0.18 Pure error 617.17 12 51.43 Cor Total 2619.94 29 Std. Dev. 8.41 R-Squared 0.30 Mean 17.10 Adj R-Squared 0.22 C.V. 49.17 Pred R-Squared 0.04 PRESS 2505.04 Adeq Precision 3.65
Not significant
Significant
102
The ‘Lack of Fit F-value’ of 1.69 implies the ‘Lack of Fit’ was not significant
relative to the pure error. There was an 18.0 % (0.18) chance that ‘Lack of Fit F-value’
this large could occur due to noise. Non-significant of the ‘Lack of Fit’ represent that
the model is good and fitted well in the experiments.
In this study, the value of the determination coefficient (R2 = 0.30) indicates only
70.0 % (0.7) of the total variations were not explained by the model. Therefore, the
model obtained from the design was fairly well fitted with the experiment data. The
value of predicted determination coefficient (Pred. R-Squared = 0.04) was in reasonable
agreement with the adjusted determination coefficient (Adj. R-Squared = 0.22). An
‘Adeq. Precision’ measures the signal (response) to noise (deviation) ratio wherein a
ratio that greater than 4 is desirable. An ‘Adeq. Precision’ of this model was 3.64 (< 4),
indicated that inadequate signal occurred and therefore the model was significantly fair
for the process. Overall, there was a fair significant effect of the processing parameters
on the yield of rotenoids resin in dried roots, % (w/w).
The factors that most influenced on the yield of rotenoids resin in dried roots, %
(w/w) were ranked as follows: C (X3 - types of solvent) > A (X1 - solvent-to-solid ratio)
> B (X2 - raw material particles size). These factors were ranked based on the value of
‘Prob > F’ greater than 0.10 indicate that the model terms are not significant. It can be
observed that using a specific range of raw material particles size (mm in diameter) in
the extraction process gives the lowest significant processing parameter to affect the
yield of rotenoids resin in dried roots, % (w/w). While using specific types of solvent
with different solvent strength (polarity) between acetone and ethanol gives the most
significant processing parameter to affect the yield of rotenoids resin in dried roots, %
(w/w).
The yield of rotenoids resin in dried roots, % (w/w) was calculated using the
yield of rotenoids resin in dried roots, mg as shown in Appendix A. The final empirical
models in terms of actual factor for the yield of rotenoids resin in dried roots, % (w/w)
value were expressed by the equation (5.1) and (5.2) respectively.
103
(a) Types of solvent - Ethanol + oxalic acid solution:
Yield of rotenoids resin in dried roots, % (w/w) = 23.10 - [0.12 × solvent-to-solid
ratio] - [0.06 × raw material
particles size] (5.1)
(b) Types of solvent - Acetone:
Yield of rotenoids resin in dried roots, % (w/w) = 12.90 - [0.12 × solvent-to-solid
ratio] - [0.06 × raw material
particles size] (5.2)
This model can be used to predict the yield of rotenoids resin in dried roots, %
(w/w). From Figure 5.12, the normal probability plot of residuals revealed that the
residuals generally fall on a straight line that errors are distributed normally. The
difference between experimental value and predicted value from the model of
experiment is defined as residual (Stat-Ease, 2002).
There were no obvious patterns and unusual structure observed in the plot
residual versus the predicted response for the yield of rotenoids resin in dried roots, %
(w/w) value as shown in Figure 5.13. In addition, the size of the studentized residual
should be independent of its predicted value. In other words the spread of the
stundentized residuals should be approximately the same across all levels of the
predicted values (Stat-Ease, 2002).
104
DESIGN-EXPERT PlotYield of rotenoids resin in dried roots
Studentized Residuals
Nor
mal
% P
roba
bilit
y
Normal Plot of Residuals
-1.97 -0.82 0.33 1.48 2.63
1
5
10
20
30
50
70
80
90
95
99
Figure 5.12 Normal probability plots of residuals (Yield of rotenoids resin)
DESIGN-EXPERT PlotYield of rotenoids resin in dried roots
Predicted
Stu
dent
ized
Res
idua
ls
Residuals vs. Predicted
-3.00
-1.50
0.00
1.50
3.00
11.50 14.30 17.10 19.89 22.69
Figure 5.13 The residual versus the predicted response (Yield of rotenoids resin)
105
DESIGN-EXPERT Plot
Yield of rotenoids resin in dried rootsX = A: Solvent-to-Solid RatioY = B: Raw material particles size
Actual FactorC: Types of solvent = Ethanol
21.8233
22.009
22.1947
22.3803
22.566
Yie
ld o
f rot
enoi
ds re
sin
in d
ried
root
s
3.62
4.81
6.00
7.19
8.38 1.41
2.08
2.75
3.42
4.09
A: Solvent-to-Solid Ratio
B: Raw material particles size
Figure 5.14 Surface plot of the yield of rotenoids resin in dried roots, % (w/w) as a
function of raw material particles size and solvent-to-solid ratio: Ethanol + oxalic acid
solution extract
DESIGN-EXPERT Plot
Yield of rotenoids resin in dried rootsX = A: Solvent-to-Solid RatioY = B: Raw material particles size
Actual FactorC: Types of solvent = Acetone
11.624
11.8097
11.9953
12.181
12.3667
Yie
ld o
f rot
enoi
ds re
sin
in d
ried
root
s
3.62
4.81
6.00
7.19
8.38 1.41
2.08
2.75
3.42
4.09
A: Solvent-to-Solid Ratio
B: Raw material particles size
Figure 5.15 Surface plot of the yield of rotenoids resin in dried roots, % (w/w) as a
function of raw material particles size and solvent-to-solid ratio: Acetone extract
106
The surface plots representing the yield of rotenoids resin in dried roots, % (w/w)
versus the effect of solvent-to-solid ratio (ml/g), raw material particles size (mm in
diameter) and types of solvent (acetone and ethanol + oxalic acid solution). The surface
plots are shown in Figure 5.14 and Figure 5.15. Regarding on the interaction effects, it
shows that different types of solvent with different polarity were significantly affected
the yield of rotenoids resin in dried roots, % (w/w). In addition, the solvent-to-solid
ratio (ml/g) was also significantly affected the yield of rotenoids resin in dried roots, %
(w/w). The decrement of solvent-to-solid ratio from 8.38 ml/g to 3.62 ml/g increases
proportionately the yield of rotenoids resin in dried roots, % (w/w) for ethanol + oxalic
acid solution and acetone extract from 21.0 % (w/w) to 22.5 % (w/w) and 11.81 %
(w/w) to 12.37 % (w/w) respectively. Furthermore, the raw material particles size (mm
in diameter) was the lowest parameter to affect the yield of rotenoids resin in dried roots,
% (w/w). The decrement of raw material particles size from 4.09 mm to 1.41 mm in
diameter gives only a small increment of the yield of rotenoids resin in dried roots, %
(w/w) for ethanol + oxalic acid solution and acetone extract from 21.82 % (w/w) to 22.0
% (w/w) and 11.62 % (w/w) to 11.81 % (w/w) respectively. Overall, the interaction
between processing parameters against the yield of rotenoids resin in dried roots, %
(w/w) can be stated as follows:
(1) Ethanol extract: A ↓; B ↓ = ↑ Yield of rotenoids resin in dried roots, % (w/w)
(2) Acetone extract: A ↓; B ↓ = ↑ Yield of rotenoids resin in dried roots, % (w/w)
* A - Solvent-to-solid ratio (ml/g), B - Raw material particles size (mm in diameter)
Moreover, Pagan and Hageman (1949) also reported that the raw material
particles size (mm in diameter) was affected significantly on the yield of rotenoids resin
in dried roots; % (w/w) at different range of sizes. However, the range of raw material
particles size (mm in diameter) considered in the study seems to be insignificant on the
yield of rotenoids resin (g). Based on the optimization phase analysis, 3.62 ml/g
solvent-to-solid ratio, 1.41 mm in diameter raw material particles size for both extracts
of acetone and ethanol + oxalic acid solution were required to achieve maximum yield
of rotenoids resin in dried roots, % (w/w). As referred to Figure 5.14 and Figure 5.15,
107
the maximum yield of rotenoids resin in dried roots [g and % (w/w)] for ethanol + oxalic
acid solution extract and acetone extract were approximately 6.77 g/30.0 g [≅ 0.23 g
rotenoids resin/g dried roots] or 22.57 % (w/w) and 3.71 g/30.0 g [≅ 0.12 g rotenoids
resin/g dried roots] or 12.37 % (w/w) respectively. This is lower about 16.0 % (ethanol
+ oxalic acid solution extract) and 70.0 % (acetone extract) than the yield as reported by
Grinda et al. (1986). According to Grinda et al. (1986), the yield of rotenoids resin in
dried roots; % (w/w) was approximately 39.0 g/100.0 g [≅ 0.39 g rotenoids resin/g dried
roots] or 39.0 % (w/w) in which far beyond the yield obtained from the study. The
results indicated that acetone and ethanol is the most desirable organic solvent to extract
large amount of rotenoids resin. In addition, adjusting the polarity of ethanol by adding
with the other chemical actually gives a significant increment of the yield of rotenoids
resin in dried roots, % (w/w) as compared to the used of single solvent. According to
Gaikar and Dandekar (2001), this could be achieved using the additional aqueous
hydrotope solution where it allows water insoluble organic compounds to be diluted in
the aqueous. However, Derris roots have its own natural emulsifier which available in
the resin to allow water insoluble compounds to be well dissolved in the water. Besides,
Grinda et al. (1986) preferred to use hazardous chlorinated hydrocarbons such as
methylene chloride (polarity of 3.1) and chloroform (polarity of 4.1) to extract good
quality and quantity of rotenoids resin.
Even though Grinda et al. (1986) have noticed the risk to human health
(especially for the operators in the production line) and the environment effects using the
hazardous material, they kept on using this material until recently. They are also facing
major problems such as phyto-toxic, deterioration of soil nutrient and fatality to the non-
targeted or beneficial organism (e.g.: bees, birds and pollinated beetle). According to
Pagan and Loustalot (1949), the chlorinated hydrocarbons such as chloroform was
apparently extracted almost all constituents including waxes, fats, chlorophyll, resin and
other plant materials. As a result, the yield of rotenoids resin in dried roots, % (w/w) as
reported by Grinda et al. (1986) was definitely impure resinous form and comprised
with large amount of unwanted substances that had no appreciable toxicity (Jones and
Pagan, 1949). In addition, the rotenoids resin that has been obtained from this study is
108
in pure form wherein less than 6 unidentified compounds were detected in the resin
using the reversed-phase HPLC (Appendix H). A production of white and milky
solution when dissolved into water would also be the good indicator to identify the
purity of the extracted resin (Andel, 2000). This mechanism can be theoretically
described as the cell wall of Derris roots is broken, the globules of resin and oil are
suspended in the sap, probably by means of saponin that acts as solubilizer or emulsifier
contributed the solubility of rotenoids resin in the water (Francis and Franklin, 1943;
Andel 2000).
109
5.3.2 Effect of processing parameters on the yield of rotenone in dried roots
The result of processing parameters (X1, X2 and X3) towards the yield of
rotenone in dried roots was obtained from the experimental data in Table 5.7.
Table 5.7: The design layout and experimental results (Yield of rotenone) Run Factor 1
X1: Solvent-to-solid ratio
Factor 2 X2: Raw material particles size
Factor 3 X3: Types of solvent
Yield of rotenone in dried roots (Experiment)
Yield of rotenone in dried roots (Predicted)
ml/g mm in diameter Treatment % (w/w) % (w/w) 1 6.00 5.00 Acetone 0.40 0.045 2 10.00 2.75 Ethanol 1.38 0.017 3 3.62 4.09 Ethanol 0.17 -0.085 4 3.62 1.41 Acetone 9.82 5.75 5 8.38 4.09 Ethanol 0.81 1.43 6 2.00 2.75 Acetone 0.19 1.31 7 3.62 4.09 Acetone 0.79 0.35 8 8.38 4.09 Ethanol 0.48 1.43 9 3.62 1.41 Acetone 8.67 5.75 10 6.00 2.75 Ethanol 0.45 0.67 11 6.00 5.00 Ethanol 1.35 0.68 12 8.38 1.41 Ethanol 0.49 -0.87 13 6.00 2.75 Acetone 1.08 2.66 14 3.62 4.09 Ethanol 0.27 -0.085 15 6.00 0.50 Ethanol 0.98 0.65 16 3.62 4.09 Acetone 0.52 0.35 17 6.00 2.75 Acetone 5.86 2.66 18 6.00 2.75 Ethanol 0.73 0.67 19 8.38 1.41 Acetone 0.59 2.69 20 8.38 4.09 Acetone 0.54 1.86 21 6.00 0.50 Acetone 2.77 5.28 22 8.38 1.41 Ethanol 0.44 -0.87 23 6.00 2.75 Ethanol 0.94 0.67 24 6.00 2.75 Acetone 4.86 2.66 25 8.38 4.09 Acetone 0.33 1.86 26 3.62 1.41 Ethanol 0.64 2.19 27 8.38 1.41 Acetone 0.36 2.69 28 2.00 2.75 Ethanol 0.19 3.31 29 10.00 2.75 Acetone 3.14 2.01 30 3.62 1.41 Ethanol 0.66 2.19
110
These results were analyzed using the Design-Expert® software version 6.0
(Stat-Ease, 2002) for further analysis. Examination of the test summary output revealed
that the 2FI (2 factor interaction terms) model was statistically significant for the
response and therefore it will be used for further analysis. The ANOVA performed
indicated the model was significant while the lack of fit was significant. In order to
eliminate the insignificant terms, the backward elimination procedure was selected. The
data was again evaluated. The resulting ANOVA table reveals that the model and
previously significant model terms were still significant. Similarly, the ‘Lack of Fit’
was still significant. Table 5.8 shows the ANOVA final result.
Table 5.8: ANOVA response surface 2FI model (responses: Yield of rotenone in dried
roots) (backward)
Significant
aAnalysis of variance (ANOVA) table for response surface reduced 2FI model [Partial sum of squares]
aSource Sum of Squares
DF Mean Square F value Prob > F
Model 87.97 5 17.59 4.81 0.04 × 10-1 A 4.07 1 4.07 1.11 0.30 B 16.34 1 16.34 4.46 0.05 C 29.88 1 29.88 8.16 0.09 × 10-1 AB 20.98 1 20.98 5.73 0.03 BC 16.72 1 16.72 4.57 0.04 Residual 87.86 24 3.66 Lack of fit 74.22 12 6.19 5.44 0.03 × 10-1 Pure error 13.64 12 1.14 Cor Total 175.84 29 Std. Dev. 1.91 R-Squared 0.50 Mean 1.66 Adj R-Squared 0.40 C.V. 115.03 Pred R-Squared 0.21 PRESS 138.78 Adeq Precision 7.74
Significant
111
The ANOVA table for the effects of processing parameters on the yield of
rotenone in dried roots, % (w/w) is shown in Table 5.8. It shows that the ‘Model F-
value’ of 4.81 implies the model was significant. There was only a 0.4 % (0.04 × 10-1)
chances that a ‘Model F-value’ this large could occur due to noise. Values of ‘Prob > F’
less than 0.05 (5.0 % confidence limit) indicate model terms were significant. In this
case, B (X2), C (X3), AB (X1X2) and BC (X2X3) were a significant model terms. Values
greater than 0.10 indicate the model terms are not significant. If there are many
insignificant model terms (not counting those required to support hierarchy), model
reduction may improve the model.
The ‘Lack of Fit F-value’ of 5.54 implies the ‘Lack of Fit’ was significant
relative to the pure error. There was a 0.3 % (0.03 × 10-1) chances that ‘Lack of Fit
F-value’ this large could occur due to noise. Significant of the ‘Lack of Fit’ represent
that the model is bad and not fitted well in the experiments. This may be the indication
of a large block effect or a possible problem with the model or input data. Things to
consider are model reduction, response transformation and outliers. Outlier T detection
helps to remove outliers from the model in order to build a more effective model.
Therefore, when the outlier is omitted from the model fitting process, the results fit
better to all points accepts the outlying point (Stat-Ease, 2002).
In this study, the value of the determination coefficient (R2 = 0.50) indicates only
50.0 % (0.5) of the total variations were not explained by the model. Therefore, the
model obtained from the design was moderately well fitted with the experiment data.
The value of predicted determination coefficient (Pred. R-Squared = 0.21) was in
reasonable agreement with the adjusted determination coefficient (Adj. R-Squared =
0.40). An ‘Adeq. Precision’ measures the signal (response) to noise (deviation) ratio
wherein a ratio that greater than 4 is desirable. An ‘Adeq. Precision’ of this model was
7.74 (> 4), indicated that adequate signal occurred and therefore the model was
significantly moderate for the process. Overall, there was a moderate significant effect
of the processing parameters on the yield of rotenone in dried roots, % (w/w).
112
The significant factors were ranked based on the value of ‘Prob > F’ ratio. The
factor with the lowest ‘Prob > F’ value was the most influences on the yield of rotenone
in dried roots, % (w/w). Thus, the factors were ranked as follows: C (X3 - types of
solvent) > B (X2 - raw material particles size) > B (X1 - solvent-to-solid ratio). It can be
observed that using a specific range of solvent-to-solid ratio (ml/g) in the extraction
process gives the lowest significant processing parameter to affect the yield of rotenone
in dried roots, % (w/w). While using specific types of solvent with different solvent
strength (polarity) between acetone and ethanol gives the most significant processing
parameter to affect the yield of rotenone in dried roots, % (w/w). The yield of rotenone
in dried roots, % (w/w) was calculated using the yield of rotenone (mg) before the
concentration process as shown in Appendix A. The final empirical models in terms of
actual factor for the yield of rotenone in dried roots, % (w/w) value were expressed by
the equation (5.3) and (5.4) respectively. The yield of rotenone in dried roots after the
concentration process, % (w/w) was not selected for the optimization phase analysis due
to the occurrence of thermal degradation which result a large dissipation of rotenone
content (mg). Consequently, the results were insignificant to obtain the optimum
processing parameter. On top of that, the results represent the actual amount of rotenone
consisted in the rotenoids resin after the concentration process but unfortunately cannot
be used due to the insignificant model achieved.
(a) Types of solvent - Ethanol + oxalic acid solution:
Yield of rotenone in dried roots, % (w/w) = 7.56 - [1.15 × solvent-to-solid ratio] -
[2.15 × raw material particles size] + [0.36
× solvent-to-solid ratio × raw material
particles size] (5.3)
(b) Types of solvent - Acetone:
Yield of rotenone in dried roots, % (w/w) = 12.77 - [1.15 × solvent-to-solid ratio] -
[3.32 × raw material particles size] + [0.36
× solvent-to-solid ratio × raw material
particles size] (5.4)
113
This model can be used to predict the yield of rotenone in dried roots, % (w/w).
As referred to Figure 5.16, the normal probability plot of residuals revealed that the
residuals generally fall on a straight line that errors were distributed normally. The
difference between experimental value and predicted value from the model of
experiment is defined as residual (Stat-Ease, 2002). No obvious patterns and unusual
structure were observed in the plot residual versus the predicted response for the yield of
rotenone in dried roots, % (w/w) value as shown in Figure 5.17.
DESIGN-EXPERT PlotYield of rotenone content in dried roots
Studentized Residuals
Nor
mal
% P
roba
bilit
yNormal Plot of Residuals
-1.79 -0.73 0.32 1.38 2.44
1
5
10
20
30
50
70
80
90
95
99
Figure 5.16 Normal probability plots of residuals (Yield of rotenone)
114
DESIGN-EXPERT PlotYield of rotenone content in dried roots
Predicted
Stu
dent
ized
Res
idua
ls
Residuals vs. Predicted
-3.00
-1.50
0.00
1.50
3.00
-0.87 0.78 2.44 4.09 5.75
Figure 5.17 The residual versus the predicted response (Yield of rotenone)
DESIGN-EXPERT Plot
Yield of rotenone content in dried rootsX = A: Solvent-to-Solid RatioY = B: Raw material particles size
Actual FactorC: Types of solvent = Ethanol
-0.87441
-0.109015
0.656381
1.42178
2.18717
Yie
ld o
f rot
enon
e co
nten
t in
drie
d ro
ots
3.62
4.81
6.00
7.19
8.38
1.41
2.08
2.75
3.42
4.09
A: Solvent-to-Solid Ratio
B: Raw material particles size
Figure 5.18 Surface plot of the yield of rotenone in dried roots, % (w/w) as a function
of raw material particles size and solvent-to-solid ratio: Ethanol + oxalic acid solution
extract
115
DESIGN-EXPERT Plot
Yield of rotenone content in dried rootsX = A: Solvent-to-Solid RatioY = B: Raw material particles size
Actual FactorC: Types of solvent = Acetone
0.346433
1.69678
3.04712
4.39747
5.74782
Yie
ld o
f rot
enon
e co
nten
t in
drie
d ro
ots
3.62
4.81
6.00
7.19
8.38
1.41
2.08
2.75
3.42
4.09
A: Solvent-to-Solid Ratio
B: Raw material particles size
Figure 5.19 Surface plot of the yield of rotenone in dried roots, % (w/w) as a function
of raw material particles size and solvent-to-solid ratio: Acetone extract
The surface plots demonstrate the yield of rotenone in dried roots, % (w/w)
versus the effect of solvent-to-solid ratio (ml/g), raw material particles size (mm in
diameter) and types of solvent (acetone and ethanol + oxalic acid solution) as shown in
Figure 5.18 and Figure 5.19. It was observed that different types of solvent with
different polarity significantly affected the yield of rotenone in dried roots, % (w/w).
Besides, the raw material particles size (mm in diameter) was significantly affected the
yield of rotenone in dried roots, % (w/w). The decrement of raw material particles size
from 4.09 mm to 1.41 mm in diameter increases proportionately the yield of rotenone in
dried roots, % (w/w) for ethanol + oxalic acid solution and acetone extract from -0.87 %
(w/w) to 2.19 % (w/w) and 0.35 % (w/w) to 5.75 % (w/w) respectively. Furthermore,
the solvent-to-solid ratio (ml/g) was the lowest parameter to affect the yield of rotenone
in dried roots, % (w/w). The increment of solvent-to-solid ratio from 3.62 ml/g to 8.38
ml/g increases proportionately the yield of rotenone in dried roots, % (w/w) for ethanol
+ oxalic acid solution and acetone extract from -0.87 % (w/w) to 1.42 % (w/w) and 0.35
116
% (w/w) to 1.70 % (w/w) respectively. Overall, the interaction between processing
parameters against the yield of rotenone in dried roots, % (w/w) can be stated as follows:
(3) Ethanol extract: A ↓; B ↓ = ↑ Yield of rotenone in dried roots, % (w/w)
(4) Acetone extract: A ↓; B ↓ = ↑ Yield of rotenone in dried roots, % (w/w)
* A - Solvent-to-solid ratio (ml/g), B - Raw material particles size (mm in diameter)
The finding was in accordance with Pagan and Hageman (1949) who reported
that the raw material particles size (mm in diameter) was affected significantly to the
yield of rotenone (mg) at different range of sizes. Theoretically, a fine particles size
would lead to a greater interfacial area between parenchymatous cell and the bulk
solution. The parenchymatous cells were found in the xylem, phloem, pericycle and
xylem rays as shown in Figure 3.1 contain a huge amount of rotenone and other toxic
constituents (Francis and Franklin, 1943). As a result, these bio-active constituents can
diffuse easily within the finely rupture cell wall into a bulk solution in a shorter time
until it reaches the equilibrium phase. Additionally, rotenone and other toxic
constituents are diffused along with the resin right after the cell wall has been ruptured
due to the pre-processing treatment. Furthermore, the resin can be found as globules or
particles with 0.8 µm to 3.9 µm in diameter and appeared to be found largely in the
opaque cells (Francis and Franklin, 1943). Thus, the smaller the particles size, the larger
the amount of rotenone-containing resin can be extracted. However, the particles size
should not be too fine for it may wedge in the interstices of the larger particles and
impeded the solvent flow (Pinelo et al., 2006).
Based on the optimization phase analysis, 3.62 ml/g solvent-to-solid ratio, 1.41
mm in diameter raw material particles size for both extract of acetone and ethanol +
oxalic acid solution were required to achieve maximum yield of rotenone in dried roots,
% (w/w). The appropriate processing parameter values for this response variable were
relatively the same with the yield of rotenoids resin in dried roots, % (w/w) response
variable. Although the appropriate processing parameter values were same, the best
processing parameter values to have the best response variables can be calculated based
117
on its desirability values. As referred to Figure 5.18 and Figure 5.19, the maximum
yield of rotenone in dried roots [mg and % (w/w)] for the ethanol + oxalic acid solution
extract and acetone extract were approximately 0.66 g/30.0 g [≅ 0.02 g rotenone/g dried
roots] or 2.19 % (w/w) and 1.73 g/30.0 g [≅ 0.06 g rotenone/g dried roots] or 5.75 %
(w/w) respectively. This is lower about 12.0 % (ethanol + oxalic acid solution extract)
and 8.0 % (acetone extract) than the yield as reported by Grinda et al. (1986) which
applied the solvent extraction method with operating temperature and agitating duration
of 45 0C and half an hour respectively (14.0 g/100.0 g [≅ 0.14 g rotenone/g dried roots]
or 14.0 % (w/w)).
A solvent with polarity between acetone and ethanol + oxalic acid solution were
desired to give a significant effect against the yield of rotenone in dried roots, % (w/w).
Ironically, this acidic isoflavonoid compound (Fasihuddin and Hasmah, 1993) is consists
with hydroxyl group or sugar (glycoside bond) wherein it easily dissolved with the polar
organic solvent (e.g.: alcohol, acetone, H2O and etc.). Since ethanol alone extracts less
amount of rotenone, the usage of H2O and oxalic acid (hydrotope solution) has proven to
extract more rotenone and other toxic constituents. This hydrotope solution actually
facilitates the insoluble rotenone and other toxic constituents to be dissolved in the
extract solution. Therefore, the differential of the yield of rotenone in dried roots; %
(w/w) between acetone and ethanol + oxalic acid solution with the solvent used as
reported by Grinda et al. (1986) were irrelatively too far if the additional chemicals used
in the study. According to Grinda et al. (1986), the chlorinated hydrocarbons such as
methylene chloride (polarity of 3.1) and chloroform (polarity of 4.1) with the assist of
esters of aliphatic acids were the best solvent combination to extract high yield of
rotenone. Therefore, the adjustment polarity of potential solvent such as acetone and
ethanol (by adding the additional potent chemical such as aliphatic acids) should be
investigated and studied further.
Additionally, the main consideration to obtain rotenone is not just the amount
that can be extracted but the other considerations should be taken seriously are for
instance the environmental issues and economically feasible of the solvent used.
118
Methylene chloride is used in many applications. It has been used as a principal active
ingredient in the organic-based paint strippers, cosmetic and consumer products,
industrial paint removers as well as in the chemical processing (EPA, 1994). Due to it
vast usages, the health effect especially to human and other beneficial organism were
barely undeniable (WHO, 1996). Despite of those adverse effects, the usage of acetone
and ethanol were suggested to be the best solvent that easily biodegradable, low in cost,
easy to handle and environmental-friendly. Not to forget that these solvent were
dissolved less coloring matters and the toxic constituents were found to be more stable
than the other solvents (Pagan and Loustalot, 1948). Furthermore, Grinda et al. (1986)
proclaimed to have used variety of species to extract large amount of rotenone by
mixing the species such as Lonchocarpus nicou (known as cube or barbasco; Amazon
forest, Brazil), Lonchocarpus urucu (Amazon forest, Brazil), Milletia ferruginea (French
West Africa) and Mundulea suberosa (Africa) during the extraction process. Using the
combination of these species, Grinda et al. (1986) have successfully extracted 0.14 g of
rotenone in 1.0 gram of the finely crushed roots as compared to the acetone and ethanol
+ oxalic acids solution extract for only 0.06 g and 0.02 g respectively. The results of
this study were incomparable with Grinda et al. (1986) due to different approaches that
have been used. But for the sake of this study, the results obtained for each phases need
to be relatively comparable in order to recognize which parameters or method gives the
most significant effect of all.
5.3.3 Summary of the optimization phase
Based on the optimization experiments, the theoretical maximum yield of
rotenoids resin and yield of rotenone in dried roots were 12.26 % (w/w) and 5.99 %
(w/w) respectively. These theoretical maximum responses were produced based on the
desirability values of a given solution. As for the biological activity (LC50) of rotenoids
resin, it was not included due to insignificant response model to obtain the optimum
119
processing parameters. This response variable was notably considered as active
compounds wherein all treatment produced LC50 ≤ 100 ppm after 6, 12 and 24 hours of
the treatment respectively. Hence, only these maximum responses (yield of rotenoids
resin in dried roots; % (w/w) and yield of rotenone in dried roots; % (w/w)) were
obtained at the appropriate processing conditions. The evaluation was done thoroughly
based on the certain setup criteria. The criteria that have been setup are shown in Table
5.9.
Table 5.9: Selection criteria of the processing parameters solution PROCESSING PARAMETER GOAL LIMIT Solvent-to-solid ratio (ml/g) Is in range Lower: 2 ml/g ↔ Upper: 10 ml/g Raw material particles size (mm) Is in range Lower: 0.5 mm ↔ Upper: 5.0 mm Types of solvent Is in range (1) Ethanol or (2) Acetone Yield of rotenoids resin, % (w/w) Is maximum Lower: 2.48 ↔ Upper: 32.69 Yield of rotenone, % (w/w) Is maximum Lower: 0.17 ↔ Upper: 9.82
Since there were three solutions given, the highest desirability of 0.56 was
chosen as the best solution to provide the appropriate processing parameters. The
appropriate processing parameter that produced maximum yield of rotenoids resin, %
(w/w) and yield of rotenone in dried roots; % (w/w) are showed in Figure 5.20. These
appropriate processing parameters were used as processing conditions in the verification
phase.
120
Figure 5.20 Selected processing parameters that obtain maximum yield of rotenoids
resin in dried roots, % (w/w) and yield of rotenone in dried roots, % (w/w) based on the
desirability values of a given solution
5.4 Multi response analysis of the yield of rotenone in dried roots; % (w/w) and
rotenone concentration (mg/ml)
Figure 5.21, Figure 5.22, Figure 5.23 and Figure 5.24 show correlation between
the yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml in
accordance to the solvent-to-solid ratio (ml/g), types of solvent and raw material
particles size based on the optimization phase experimental result. As there is no multi
response analysis offered in the Design-Expert® software version 6.0 (Stat-Ease, 2002),
the multi response analysis were carried out using the Microsoft® Excel 2003. The
results in Table 5.10, Table 5.11, Table 5.12 and Table 5.13 were calculated and
5.99 % (w/w)
12.26 % (w/w)
4.72 ml/g 0.83 mm in diameter
Max Min
Min Max
121
presented in means ± SD based on the optimization phase experimental results
(Appendix A). The assumption has been made for each independent variable used
during the multi-response analysis which are:
(A) Solvent-to-solid ratio (ml/g) - Ethanol + oxalic acid solution extract.
Assumption: Control processing parameter of the raw material particles size
(mm in diameter) with an average of 2.75 mm in diameter.
(B) Solvent-to-solid ratio (ml/g) - Acetone extract.
Assumption: Control processing parameter of the raw material particles size
(mm in diameter) with an average of 2.75 mm in diameter.
(C) Raw material particles size (mm in diameter) - Ethanol + oxalic acid solution
extract. Assumption: Control processing parameter of the solvent-to-solid ratio
(ml/g) with an average of 6.0 ml/g.
(D) Raw material particles size (mm in diameter) - Acetone extract.
Assumption: Control processing parameter of the solvent-to-solid ratio (ml/g)
with an average of 6.0 ml/g.
In overall, only Figure 5.22 and Figure 5.24 show a significant and consistent
relationship between two important response variables in accordance with the types of
solvent, raw material particles size (mm in diameter) and solvent-to-solid ratio (ml/g).
5.4.1 Analysis of solvent-to-solid ratio (ml/g) for the ethanol + oxalic acid solution
extract in relation with the yield of rotenone in dried roots; % (w/w) and rotenone
concentration; mg/ml
Figures 5.21 and Table 5.10 illustrate the fluctuation pattern for both response
variables as the solvent-to-solid ratio (ml/g) was increased from 2.0 ml/g to 10.0 ml/g
using the ethanol + oxalic acid solution. The highest yield of rotenone in dried roots and
122
rotenone concentration were 1.39 % (w/w) and 1.66 mg/ml respectively, at 10.0 ml/g of
solvent-to-solid ratio. Meanwhile, the highest concentration of rotenone and yield of
rotenone in dried roots were 1.88 ± 1.42 mg/ml and 0.44 ± 0.26 % (w/w) respectively, at
3.62 ml/g of solvent-to-solid ratio. Therefore, each of the highest response variables
was inconsistent with each other and consequently indirect correlation can be generated
from these two response variables. In addition, there was indirect correlation observed
between the response variables and independent variables due to incompatibility of
rotenone yield (mg) in the Liquid Crude Extract (LCE) of ethanol + oxalic acid solution
in which simultaneously affected the fluctuation pattern as mentioned in section 5.3.1,
5.3.2 and 5.3.3.
Table 5.10: The effects of solvent-to-solid ratio (ml/g) of ethanol + oxalic acid solution
extract on the two response variables
Solvent-to-solid ratio
Yield of rotenone in dried roots; % (w/w)
Rotenone concentration; mg/ml
2.00 0.20 ± 0.00 1.71 ± 0.00 3.62 0.44 ± 0.26 1.88 ± 1.42 6.00 0.87 ± 0.39 1.69 ± 0.66 8.38 0.64 ± 0.23 1.18 ± 0.95 10.00 1.39 ± 0.00 1.66 ± 0.00
Data represent means ± SD Assumption: Raw material particles size was fixed with an average of 2.75 mm in diameter (0.50 + 1.41 + 2.75 + 4.09 + 5.00/5 = 2.75 mm in diameter)
123
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0 1 2 3 4 5 6 7 8 9 10 11
Solvent-to-solid ratio; ml/g
Res
pons
e va
lue;
mg/
ml o
r % (w
/w)
Yield of rotenone in dried roots; % (w/w) Rotenone concentration; mg/ml
Figure 5.21: The yield of rotenone in dried roots; % (w/w) and rotenone concentration;
mg/ml versus the solvent-to-solid ratio (ml/g) of ethanol + oxalic acid solution extract
5.4.2 Analysis of solvent-to-solid ratio (ml/g) for the acetone extract in relation
with the yield of rotenone in dried roots; % (w/w) and rotenone concentration;
mg/ml
Figures 5.22 and Table 5.11 illustrate the consistent pattern for both response
variables as the solvent-to-solid ratio (ml/g) was increased from 2.0 ml/g to 10.0 ml/g
using the acetone. The highest yield of rotenone in dried roots and rotenone
concentration were 4.98 ± 5.00 % (w/w) and 21.21 ± 21.26 mg/ml respectively, at 3.62
ml/g of solvent-to-solid ratio. Hence, each of the highest response variables was
consistent with each other and consequently a correlation can be generated from these
two response variables. Theoretically, as the solvent-to-solid ratio increased, the yield
of rotenone in dried roots; % (w/w) and its concentration (mg/ml) increased
proportionally until its reached the equilibrium concentration (Cacace and Mazza, 2003).
On the contrary, the yield of rotenone in dried roots; % (w/w) and rotenone
124
concentration; mg/ml plunged noticeably from the solvent-to-solid ratio of 3.62 ml/g to
8.38 ml/g and slightly increased until 10.0 ml/g. There was no equilibrium phases
occurred for both response variables and this indicated that the dissipation of rotenone
was occurred either during the extraction or RP-HPLC analysis process. In conclusion,
the relationship between response variables and independent variables can be generated
wherein as the yield of rotenone in dried roots; % (w/w) increased, the rotenone
concentration; mg/ml increased proportionally with the increment of solvent-to-solid
ratio (ml/g) using the acetone. Hence, the most appropriate solvent-to-solid ratio (ml/g)
to produce the highest yield of rotenone in dried roots; % (w/w) and rotenone
concentration; mg/ml was 3.62 ml/g.
Table 5.11: The effects of solvent-to-solid ratio (ml/g) of acetone extract which on the
two response variables
Solvent-to-solid ratio
Yield of rotenone in dried roots; % (w/w)
Rotenone concentration; mg/ml
2.00 0.19 ± 0.00 2.00 ± 0.00 3.62 4.98 ± 5.00 21.21 ± 21.26 6.00 3.04 ± 2.39 6.26 ± 4.81 8.38 0.46 ± 0.13 0.65 ± 0.21 10.00 3.12 ± 0.00 3.55 ± 0.00
Data represent means ± SD Assumption: Raw material particles size was fixed with an average of 2.75 mm in diameter (0.50 + 1.41 + 2.75 + 4.09 + 5.00/5 = 2.75 mm in diameter)
125
-20
-15
-10
-5
0
5
10
15
20
25
30
0 1 2 3 4 5 6 7 8 9 10 11
Solvent-to-solid ratio; ml/g
Res
pons
e va
lues
; mg/
ml o
r % (w
/w)
Yield of rotenone in dried roots; % (w/w) Rotenone concentration; mg/ml
Figure 5.22: The yield of rotenone in dried roots; % (w/w) and rotenone concentration;
mg/ml versus the solvent-to-solid ratio (ml/g) of acetone extract
5.4.3 Analysis of raw material particles size (mm in diameter) for the ethanol +
oxalic acid solution extract in relation with the yield of rotenone in dried roots; %
(w/w) and rotenone concentration; mg/ml
Figures 5.23 and Table 5.12 illustrate the inconsistent pattern for both response
variables as the raw material particles size (mm in diameter) was increased from 0.5 mm
to 5.0 mm using the ethanol + oxalic acid solution. The highest yield of rotenone in
dried roots and rotenone concentration were 1.35 % (w/w) and 1.66 mg/ml respectively,
at 5.0 mm in diameter of raw material particles size. As for the response variables of
rotenone concentration, the highest concentration of rotenone and yield of rotenone in
dried roots were 2.84 mg/ml and 0.97 % (w/w) respectively, at 0.5 mm in diameter of
raw material particles size. Therefore, each of the highest response variables was
inconsistent with each other and consequently indirect correlation can be generated from
126
these two response variables. Theoretically, raw material particle size (mm in diameter)
affects the extraction rate by increasing the total mass transfer area when the particle
size is reduced (Schwartzberg and Chao, 1982). Additionally, it has been expected that
the smooth raw material particles size results the highest yield of rotenone in dried roots;
% (w/w) as well as the rotenone concentration; mg/ml. Furthermore, if all the other
processing parameters were accounted, it can also be expected a slight variation
depending on how well the parameters were suited to each other. Therefore, there was
indirect correlation observed between the response variables and independent variables
due to incompatibility of rotenone yield (mg) in the Liquid Crude Extract (LCE) of
ethanol + oxalic acid solution in which simultaneously affected the inconsistency pattern
as mentioned in section 5.3.1, 5.3.2 and 5.3.3.
Table 5.12: The effects of raw material particles size (mm in diameter) of ethanol +
oxalic acid solution extract on the two response variables
Raw material particles size (mm in diameter)
Yield of rotenone in dried roots; % (w/w)
Rotenone concentration; mg/ml
0.50 0.97 ± 0.00 2.84 ± 0.00 1.41 0.57 ± 0.11 1.85 ± 1.45 2.75 0.74 ± 0.46 1.52 ± 0.33 4.09 0.44 ± 0.28 1.18 ± 0.95 5.00 1.35 ± 0.00 1.66 ± 0.00
Data represent means ± SD Assumption: Solvent-to-solid ratio was fixed with an average of 6.0 ml/g (2.00 + 3.62 + 6.00 + 8.38 + 10/5 = 6.0 ml/g)
127
0
0.5
1
1.5
2
2.5
3
3.5
0 1 2 3 4 5 6
Raw material particles size; mm in diameter
Res
pons
e va
lues
; mg/
ml o
r % (w
/w)
Yield of rotenone in dried roots; % (w/w) Rotenone concentration; mg/ml
Figure 5.23: The yield of rotenone in dried roots; % (w/w) and rotenone concentration;
mg/ml versus the raw material particles size (mm in diameter) of ethanol + oxalic acid
solution extract
5.4.4 Analysis of raw material particles size (mm in diameter) for the acetone
extract in relation with the yield of rotenone in dried roots; % (w/w) and rotenone
concentration; mg/ml
Figures 5.24 and Table 5.13 illustrate the consistent pattern for both response
variables as the raw material particles size (mm in diameter) was increased from 0.5 mm
to 5.0 mm using the acetone. The highest yield of rotenone in dried roots and rotenone
concentration were 4.89 ± 5.11 (w/w) and 20.28 ± 22.71 mg/ml respectively, at 1.41 mm
in diameter of raw material particles size. Thus, each of the highest response variables
was consistent with each other and therefore a correlation can be generated from these
two response variables. Theoretically, as the raw material particles size (mm in
diameter) decreased, the yield of rotenone in dried roots; % (w/w) and its concentration;
128
mg/ml increased proportionally until the maximum point. But if the raw material
particles size is reduced further, rotenone and other bio-active constituents can be
blocked for diffusing into the bulk solution due to the coagulation of the fine particles or
fine debris into the cell walls (Geankoplis, 1995; Mircea, 2001). According to Pinelo et
al. (2006), smaller particle size reduces the diffusion distance of the solute within the
solid, thus increasing the extraction rate. The solute takes a shorter time to reach the
surface. The finding was also consistent with Schwartzberg and Chao (1982) wherein
the raw material particles size (mm in diameter) influences the extraction rate by
increasing the whole mass transfer area when the particle size is reduced. The finding
from this study was in accordance with the theory wherein the yield of rotenone in dried
roots; % (w/w) and rotenone concentration; mg/ml increased noticeably from the raw
material particles of 5.0 mm to the maximum response of 1.41 mm and decreased until
0.5 mm in diameter. In conclusion, the relationship between response variables and
independent variables can be generated wherein as the yield of rotenone in dried roots;
% (w/w) increased, the rotenone concentration; mg/ml increased proportionally with the
decrement of raw material particles size (mm in diameter) using the acetone. Hence, the
most appropriate raw material particles size (mm in diameter) to produce the highest
yield of rotenone in dried roots; % (w/w) and rotenone concentration; mg/ml was 1.41
mm.
Table 5.13: The effects of raw material particles size (mm in diameter) of acetone
extract on the two response variables
Raw material particles size (mm in diameter)
Yield of rotenone in dried roots; % (w/w)
Rotenone concentration; mg/ml
0.50 2.83 ± 0.00 6.39 ± 0.00 1.41 4.89 ± 5.11 20.28 ± 22.71 2.75 3.05 ± 2.45 5.94 ± 4.68 4.09 0.56 ± 0.19 1.57 ± 1.17 5.00 0.41 ± 0.00 2.53 ± 0.00
Data represent means ± SD Assumption: Solvent-to-solid ratio was fixed with an average of 6.0 ml/g (2.00 + 3.62 + 6.00 + 8.38 + 10/5 = 6.0 ml/g)
129
-5
0
5
10
15
2025
30
35
40
45
50
0 1 2 3 4 5 6
Raw material particles size; mm in diameter
Res
pons
e va
lues
; mg/
ml o
r % (w
/w)
Yield of rotenone in dried roots; % (w/w) Rotenone concentration; mg/ml
Figure 5.24: The yield of rotenone in dried roots; % (w/w) and rotenone concentration;
mg/ml versus the raw material particles size (mm in diameter) of acetone extract
5.5 Biological activity (LC50) of rotenoids resin results
Design of Experiment (DOE) of 30 experiments in this screening are evaluated,
listed and summarized in the Table 5.14, Table 5.15 and Table 5.16. The LC50 results
were obtained using the probit analysis method (Finney, 1971) and presented in Figure
5.25, Figure 5.26, Figure 5.27 and Figure 5.28. Only S1 was eligible for the probit
analysis on 24 hours of treatment due to the other treatment produced 100 % mortality at
the lowest treatment concentration (1.0 ppm). Furthermore, the number of reasonable
point of mortality (%) for constructing a linear proportion graph of mortality (probit)
versus log10 concentration (ppm) was less than one point. The LC50 values of the 24
hours of treatment excluded S1 were obtained from the interpolation method. The
model for this response variable (bioassay) generated from the Design-Expert® software
version 6.0 (Stat-Ease, 2002) was insignificant to obtain the optimum processing
130
parameters. Data interpretation was made accordingly to the protocol established by
McLaughlin (1991), all treatment should be observed at minimum 24 hours and the
extracts with LC50 ≤ 100 ppm were considered very active.
y = 0.57x + 4.56
4.5
4.6
4.7
4.8
4.9
5
5.1
5.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Log10 concentration (ppm)
Mor
talit
y (p
robi
t)
S1
Figure 5.25 Relationship between the probit of Artemia salina mortality proportion
and log10 dose of rotenoids resin (S1) at 24 hours of treatment
0
12
34
5
67
8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Log10 concentration (ppm)
Mor
talit
y (p
robi
t)
S7 S1 S20 S23
Figure 5.26 Relationship between the probit of Artemia salina mortality proportion
and log10 dose of rotenoids resin (S1, S7, S20 and S23) at 12 hours of treatment
Probit 5.0 = 50 % mortality
Probit 5.0 = 50 % mortality
131
0
12
34
5
67
8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Log10 concentration (ppm)
Mor
talit
y (p
robi
t)
S13 S25 S29 S7 S19 S24 S23 S28 S11
Figure 5.27 Relationship between the probit of Artemia salina mortality proportion
and log10 dose of rotenoids resin (S7, S11, S13, S19, S23, S24, S25, S28 and S29) at 12
hours of treatment - Continued
y = 1.619x + 3.4477
y = 4.3062x - 0.0222
y = 4.3062x - 0.0222
-10
12
34
56
78
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
Log10 concentration (ppm)
Mor
talit
y (p
robi
t)
S8 S3 S12
Figure 5.28 Relationship between the probit of Artemia salina mortality proportion
and log10 dose of rotenoids resin (S3, S8 and S12) at 6 hours of treatment
* S12 and S3 were produced the same value of mortality (probit) ** S12 & S3: y = 4.3062x - 0.022
Probit 5.0 = 50 % mortality
Probit 5.0 = 50 % mortality
132
5.5.1 The effect of raw material particles size and types of solvent on the
biological activity (LC50) and yield of rotenone in dried roots, % (w/w) respectively
Based on the results of the biological activity (LC50) at 24 hours of treatment as
shown in Table 5.14, Table 5.15, Table 5.16 and Figure 5.29, the diameter group of
Derris dried roots of 0.5 mm to 4.09 mm in diameter for the ethanol + oxalic acid
solution extract was the most toxic to produce the LC50 ≤ 1.00 ppm. As for the acetone
extract, the toxicity level was also as toxic as the extract of ethanol + oxalic acid solution
to produce LC50 ≤ 1.00 ppm from the raw material particles size of 0.5 mm to 4.09 mm
in diameter. Then, the toxicity level decreased (LC50 = 5.91 ppm > 1.00 ppm) as the raw
material particles size increased to the 5.0 mm in diameter. The capability of acetone
(polarity of 5.1) and ethanol (polarity of 5.2) to diffuse into the thickest diameter of
Derris dried roots to extract exhaustively rotenoids resin were the main reason
biological activity (LC50) values produced ≤ 1.00 ppm and considered as very active
constituents. The results strongly demonstrated that when the cell wall is broken, the
amount of saponin or glycosides are released and solubilized the resin and eventually
suspended into the cell sap of the rupture cell wall (Francis and Franklin, 1943). The
result also indicated that these types of solvent which are acetone and ethanol (altered
polarity) extract more other toxic constituents such as deguelin, tephrosin and 12αβ-
rotenolone as compared to the other polar organic solvent such as chloroform which
selective only to rotenone and not to the other toxic constituents in the resin.
Additionally, Pagan and Loustalot (1948) have quoted that toxicity values would appear
to be one which would measure not only rotenone but the other toxic constituents as
well. According to Pagan and Hageman (1949), the increase in toxicity due to rotenoids
(rotenone and other toxic constituents) is very noticeable for the roots of 2.0 mm to 4.0
mm in diameter. Hence, the biological activity (LC50) of acetone and ethanol + oxalic
acid solution extract was moderately correlated with Pagan and Hageman (1949)
finding.
133
0.77
5.91
0.50.75
0.67 0.50.84
0.5 0.5
0
1
2
3
4
5
6
7
(M: 0.5) (H and D: 1.41) (P, A and K: 2.75) (I and F: 4.09) (Q: 5.0)
Raw material particles size (mm in diameter)
Bio
logi
cal a
ctiv
ity (L
C50
), pp
m
0.5 0.50.520.5 0.50.5
1.5
0.5 0.5
0
0.5
1
1.5
2
2.5
3
3.5
(N: 0.5) (G and J: 1.41) (R, B and O: 2.75) (E and C: 4.09) (L: 5.0)
Raw material particles size (mm in diameter)
Bio
logi
cal a
ctiv
ity (L
C50
), pp
m
Figure 5.29 Effect of the raw material particles size, mm in diameter against the
biological activity (LC50) of acetone extract (A1) and ethanol + oxalic acid solution
extract (A2) respectively
Furthermore, the yield of rotenone in dried roots; % (w/w) of acetone extract
from the finest (0.5 mm in diameter) to the thickest (5.0 mm in diameter) particles size
was more than the ethanol + oxalic acid solution extract. The results as shown in Figure
5.30 revealed that the capability of acetone to extract more rotenone as compared to the
(A1): Acetone extract
(A2): Ethanol + oxalic acid solution extract
134
ethanol + oxalic acid solution were unquestionable in which the toxic constituents in the
roots particularly rotenone tends to have higher solubility and stability in the acetone
than the other organic solvents (Pagan and Loustalot, 1948). Hence, the biological
activity (LC50) values of all extracts were significantly correlated with the yield of
rotenone in dried roots; % (w/w) as well as the other toxic constituents. The yield of
rotenone in dried roots; % (w/w) as shown in Figure 5.30 were in accordance with Pagan
and Loustalot (1948) wherein acetone was able to extract more rotenone and other toxic
constituents. For the meantime, the ethanol + oxalic acid solution extracts smaller
amount of rotenone (mg) than acetone nevertheless more on the other toxic constituents
due to the high biological activity (LC50) values.
0.69
0.390.41
0.49
2.4
0.66
0.520.69
0.09
0
0.5
1
1.5
2
2.5
3
(M: 0.5) (H and D: 1.41) (P, A and K: 2.75) (I and F: 4.09) (Q: 5.0)
Raw material particles size (mm in diameter)
Yiel
d of
rot
enon
e in
dri
ed r
oots
, % (w
/w)
(B1): Acetone extract
135
0.75
0.97
0.46
0.34
0.99
0.39
0.2
0.31
0.05
0
0.2
0.4
0.6
0.8
1
1.2
(N: 0.5) (G and J: 1.41) (R, B and O: 2.75) (E and C: 4.09) (L: 5.0)
Raw material particles size (mm in diameter)
Yiel
d of
rot
enon
e in
dri
ed r
oots
, % (w
/w)
Figure 5.30 Effect of the raw material particles size, mm in diameter against the yield
of rotenone in dried roots, % (w/w) obtained from the extract of acetone (B1) and
ethanol + oxalic acid solution (B2) respectively
5.5.2 The effect of solvent-to-solid ratio and types of solvent on the biological
activity (LC50) and yield of rotenone in dried roots, % (w/w) respectively
The results in Table 5.14, Table 5.15, Table 5.16 and Figure 5.31 show a
significant effect of the solvent-to-solid ratio (ml/g) on the biological activity (LC50)
values. The ethanol + oxalic acid solution extract exhibited as the most toxic to produce
LC50 ≤ 1.00 ppm from 2.0 ml/g to 10.0 ml/g of the solvent-to-solid ratio without major
fluctuation. Only Sample C with a raw material particles size of 4.09 mm in diameter
and 3.62 ml/g produced the highest LC50 value and considered as the lowest toxicity or
moderate active constituents. Meanwhile, the acetone extract also considered as toxic as
the ethanol + oxalic acid solution extract by producing the LC50 ≤ 1.00 ppm at almost all
solvent-to-solid ratio. Sample Q with a raw material particles size of 5.0 mm in
diameter and 6.0 ml/g produced the highest LC50 value and also considered as the lowest
(B2): Ethanol + oxalic acid solution extract
136
toxicity or moderate active constituents. Theoretically, the increase of solvent-to-solid
ratio with the increase of rotenone content (mg) is consistent with the mass transfer
principles (Frank and Downey 1999). Figure 5.32 shows the comparison between the
yield of rotenone in dried roots, % (w/w) for acetone and ethanol + oxalic acid solution
extract in which the results were in accordance with the theory. This figure indicates
that the yield of rotenone in dried roots, % (w/w) increased as the solvent-to-solid ratio
(ml/g) for acetone and ethanol + oxalic acid solution extract increased respectively.
Figure 5.31 (C2) indicates that there was indirect correlation observed between the
solvent-to-solid ratio (ml/g) and the biological activity (LC50) of ethanol + oxalic acid
solution extract as the solvent-to-solid ratio (ml/g) increased wherein almost all the
biological activity (LC50) values were ≤ 1.00 ppm. Only Sample C with a raw material
particles size of 4.09 mm in diameter and 3.62 ml/g produced the highest LC50 value.
As for the acetone extract, there was also indirect correlation observed. However, the
solvent-to-solid ratio of 6.0 ml/g with the raw material particles size of 5.0 mm in
diameter (Sample Q) shown a significant increased of the biological activity (LC50) as
compared to the same ratio of the ethanol + oxalic acid solution extract (Sample L).
These anomalies were possibly due to the inconsistency of the rotenone content (mg) as
a result from the dissipation of rotenone during the extraction and concentration process
(have been discussed earlier in section 5.2.3). Previous study has shown that rotenone
deteriorated rapidly as much as 90 % (w/w) at the initial 15 mins of the 50 0C operating
temperature due to the insufficient vacuum pressure pump during the concentration
process (Saiful et al., 2003). The insufficient vacuum pressure pump needs longer time
to evaporate solvent from the extracts thus longer time the extracts were exposure to the
heat. From the engineering point of view, a good efficiency of the vacuum pressure
pump reduces boiling point of the vaporized material lower than its own boiling point at
1.0 bar. For that reason, the time consumption and utility operating cost of the process
can be reduced thus minimized any thermal degradation of the heat sensitive compounds
from further deterioration.
137
0.5 0.50.77 0.670.5
5.91
0.50.750.84
0
1
2
3
4
5
6
7
K: 2.0 D and F: 3.62 Q, A and M: 6.0 H and F: 8.38 P: 10.0
Solvent-to-solid ratio (ml/g)
Biol
ogic
al a
ctiv
ity (L
C50
), pp
m
0.5 0.50.50.50.5 0.50.5 0.52
1.5
0
0.5
1
1.5
2
2.5
3
3.5
(O: 2.0) (J and C: 3.62) (L and B: 6.0) (G, E and N: 8.38) (R: 10.0)
Solvent-to-solid ratio (ml/g)
Bio
logi
cal a
ctiv
ity (L
C50
), pp
m
Figure 5.31 Effect of the solvent-to-solid ratio, ml/g against the biological activity
(LC50) of acetone extract (C1) and ethanol + oxalic acid solution extract (C2) respectively
As for the yield of rotenone in dried roots, % (w/w), there were strong
correlations against the solvent-to-solid ratio (ml/g). Figure 5.32 show that acetone was
expected to extract more rotenone as compared to the ethanol + oxalic acid solution.
Only sample P (2.75 mm in diameter and 10.0 ml/g) and sample R (2.75 mm in diameter
and 10.0 ml/g) of the acetone and ethanol + oxalic acid solution extract obtained the
(C1): Acetone extract
(C2): Ethanol + oxalic acid solution extract
138
highest yield of rotenone in dried roots which are 2.40 % (w/w) and 0.99 % (w/w)
respectively. But as for the biological activity (LC50), both extracts produced LC50 ≤
1.00 ppm and demonstrated that as the solvent-to-solid ratio increased, it increases the
yield of rotenone in dried roots, % (w/w) and produced LC50 ≤ 1.00 ppm at almost the
entire solvent-to-solid ratio (ml/g). The results as shown in Figure 5.32 also revealed
that the capability of acetone to extract more rotenone as compared to the ethanol +
oxalic acid solution can be predicted due to the toxic constituents particularly rotenone
dissolved much higher in the acetone than the other organic solvents (Pagan and
Loustalot, 1948). Hence, the biological activity (LC50) of all extracts was directly
correlated with the yield of rotenone in dried roots, % (w/w) as well as the other toxic
constituents. A research by Pagan and Hageman (1949) demonstrated that the
toxicological value of Derris roots was imperfectly correlated with the yield of rotenone
(mg). Data from their research indicated that substances other than rotenone such as
12αβ-rotenolone, deguelin and tephrosin were also contributed to the toxicity of the
extracts.
0.09
0.69
2.4
0.69
0.390.410.52
0.49
0.66
0
0.5
1
1.5
2
2.5
3
K: 2.0 D and F: 3.62 Q, A and M: 6.0 H and I: 8.38 P: 10.0
Solvent-to-solid ratio (ml/g)
Yiel
d of
rot
enon
e in
drie
d ro
ots,
% (w
/w)
(D1): Acetone extract
139
0.05
0.99
0.2
0.75
0.23
0.39
0.340.31
0.46
0
0.2
0.4
0.6
0.8
1
1.2
(O: 2.0) (J and C: 3.62) (L and B: 6.0) (G, E and N: 8.38) (R: 10.0)
Solvent-to-solid ratio (ml/g)
Yiel
d of
rote
none
in d
ried
root
s, %
(w/w
)
Figure 5.32 Effect of the solvent-to-solid ratio, ml/g against the yield of rotenone in
dried roots, % (w/w) obtained from the extract of acetone (D1) and ethanol + oxalic acid
solution (D2) respectively
(D2): Ethanol + oxalic acid solution extract
140
Table 5.14: Biological activity (LC50) of rotenoids resin at varies time of treatment
(6 hours, 12 hours and 24 hours)
Sample Particles size
(mm in diameter ) Solvent-to-solid ratio
Types of solvent
LC50 (ppm)
LC50 (ppm)
LC50 (ppm)
6 hrs 12 hrs 24 hrs 1 5.00 6.00 ml/g Acetone 10.00 7.16 5.91 2 2.75 10.00 ml/g Ethanol 1.82 a0.50 a0.50 3 4.09 3.62 ml/g Ethanol 14.66 5.50 2.50 4 1.41 3.62 ml/g Acetone 10.00 0.71 a0.50 5 4.09 8.38 ml/g Ethanol 0.67 0.55 a0.50 6 2.75 2.00 ml/g Acetone 0.56 a0.50 a0.50 7 4.09 3.62 ml/g Acetone 8.40 1.74 1.00 8 4.09 8.38 ml/g Ethanol 9.09 0.67 0.53 9 1.41 3.62 ml/g Acetone 0.83 0.67 a0.50 10 2.75 6.00 ml/g Ethanol a0.50 0.50 a0.50 11 5.00 6.00 ml/g Ethanol 5.62 0.77 a0.50 12 1.41 8.38 ml/g Ethanol 14.92 5.01 a0.50 13 2.75 6.00 ml/g Acetone 2.22 4.38 a0.50 14 4.09 3.62 ml/g Ethanol 1.00 a0.50 a0.50 15 0.50 8.38 ml/g Ethanol 0.59 a0.50 a0.50 16 4.09 3.62 ml/g Acetone 3.08 0.67 0.67 17 2.75 6.00 ml/g Acetone 10.00 2.64 a0.50 18 2.75 6.00 ml/g Ethanol 3.24 0.77 a0.50 19 1.41 8.38 ml/g Acetone 1.77 4.60 a0.50 20 4.09 8.38 ml/g Acetone 5.00 4.00 a0.50 21 0.50 6.00 ml/g Acetone 5.50 5.50 0.77 22 1.41 8.38 ml/g Ethanol 4.38 4.00 a0.50 23 2.75 6.00 ml/g Ethanol 10.74 4.21 a0.50 24 2.75 6.00 ml/g Acetone 3.24 0.83 a0.50 25 4.09 8.38 ml/g Acetone 2.36 1.00 1.00 26 1.41 3.62 ml/g Ethanol 4.00 3.57 a0.50 27 1.41 8.38 ml/g Acetone 0.77 0.77 a0.50 28 2.75 2.00 ml/g Ethanol 6.44 4.21 a0.50 29 2.75 10.00 ml/g Acetone 7.43 3.08 0.67 30 1.41 3.62 ml/g Ethanol 0.71 0.67 a0.50
aMortality of the samples which produced 100 % from the lowest (1.0 ppm) to the highest (1000 ppm) concentration treatment was calculated as LC50 = 0.5 ppm to simplify the mean calculation.
143
5.5.3 Biological activity (LC50) of the verification phase parameters and rotenone
standard (SIGMA-Aldrich™)
The biological activity (LC50) of the verification phase was based on the selected
processing parameters that produced theoretical maximum yield of rotenoids resin in
dried roots; % (w/w) and yield of rotenone in dried roots, % (w/w) as shown in Table
5.17. Data interpretation was made accordingly to the protocol established by
McLaughlin (1991) and comparison was made between the literatures, preliminary and
rotenone standard solution obtained from SIGMA-Aldrich™ with purity of 95 - 98 %
(w/w). The biological activity (LC50) value is shown completely in Appendix F.
Table 5.17: List of appropriate processing parameter that produced theoretical
maximum yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone in dried
roots, % (w/w)
Processing parameters Parameter values Types of solvent Acetone, 95.0 % (v/v) Solvent-to-solid ratio 4.72 ml/g Raw material particles size 0.83 mm in diameter
5.6 Verification phase results: Confirmation of the optimization
The experiment was based on the most appropriate processing parameters
obtained from the optimization experiment as shown in Table 5.17. The experiment was
carried out in two replicates. The solvent-to-solid ratio (ml/g), types of solvent and raw
material particles size (mm in diameter) were determined based on the solutions given at
the highest desirability which provide the appropriate processing parameters for the
verification experiment. As far as this study is concern, acetone was chosen due to its
high capability to extract a large amount of rotenone (mg), dissolving less colouring
144
matters, waxes, and other plant material as well as to be found stable in it (Pagan and
Loustalot, 1948). The extraction duration was set as a control parameter based on the
preliminary experiment in which result approximately 14 hours to achieve the
equilibrium at ambient operating temperature of 26 ± 2 0C. Moreover, the results shown
in section 5.2.4 strongly indicated that different processing parameter produced varies
equilibrium time of the extraction process although the extraction temperature was set as
a control parameter. Therefore, the extraction duration was controlled with the intention
that instability of rotenone content (mg) in the LCE can be minimized at certain time
under ambient condition. In addition, the extraction temperature was designated as
control parameter based on the preliminary experiment in which the extraction condition
should not be surpassing 40 0C (Surya and John, 2001) to avoid rotenone dissipation.
The result of verification phase experiments and its respective optimum processing
parameters are shown in Table 5.18 and Table 5.17 respectively. The results obtained
verified the selection of the most appropriate processing parameters.
Table 5.18: The verification phase results based on the most appropriate processing
parameters
Dependent/response variables Replicate 1 Replicate 2 Yield of rotenoids resin in dried roots, % (w/w) ≅ 11.73 ≅ 11.33 Yield of rotenone in dried roots, % (w/w) ≅ 2.44 ≅ 2.77 Brine Shrimp Lethality study (McLaughlin and Rogers, 1998) Biological activity, LC50 (ppm) 0.83 0.83
5.7 Comparison of the optimum response variables
The optimum response variables obtained from the literature, preliminary,
optimization and verification experiments were differed slightly from each other but as
compared to the literature, there was a huge difference between the other. The
145
comparison is shown in Table 5.19. This research reports for the first time of the
phytochemical and biological activity (LC50) screening on the roots of Derris elliptica.
The results demonstrate final comparison on the literature, preliminary, optimization and
verification experiments. The rotenoids resin or known as a cube resin was obtained
from the SAPHYR S. A. R. L (France). This cube resin was impure and comprised with
other unwanted substances or impurities such as ester of aliphatic acids, waxes, fats,
chlorophyll and colouring matters. That is the reason why the yield of rotenoids resin in
dried roots; % (w/w) as reported by Grinda et al. (1986) was higher than the preliminary,
optimization and verification experiments. As for this study, pure resin was obtained
during the preliminary, optimization and verification experiments. The indicator to
identify pure resin substances were by analyzing the resin using RP-HPLC wherein less
than 6 unidentified compounds were available in the resin as compared to the industrial
cube resin which consist 30 unidentified compounds as shown in Appendix H. The
other indicator was to dissolve the resin with water until a white and milky solution
appeared (Francis and Franklin, 1943; Andel, 2000).
Another reason was due to the species used as reported by Grinda et al. (1986)
have been mixed with other species such as Lonchocarpus nicou (known as cube or
barbasco; Amazon forest, Brazil), Lonchocarpus urucu (Amazon forest, Brazil), Milletia
ferruginea (French West Africa) and Mundulea suberosa (Africa). Meanwhile, the
procurement of Derris roots in Kota Johor Lama, Johor was done by selecting only one
species which is Derris elliptica. This was to ensure that no major differences occurred
in the result of yield, concentration and toxicity level. Although one of the response
variables (biological activity study) shown insignificant due to inconsistency of the yield
of rotenone (mg), other response variables produced a significant relationship and can be
used to predict the yield of rotenoids resin in dried roots; % (w/w) and yield of rotenone
in dried roots; % (w/w). Furthermore, the usage of new co-solvent (ester of aliphatic
acids such as butyl, hexyl and octyl esters) to facilitate the solubility of rotenone into the
bulk solution was also the main reason of yielding high rotenoids resin. Unfortunately,
several weaknesses were noted from the method established by Grinda et al. (1986)
which is the methylene chloride used in the extraction process was costly, unsafe and
146
unenvironmental-friendly as compared to the acetone and ethanol. Moreover, no
literature has been found for the brine shrimp lethality study due to the biological
activity (LC50) of rotenoids resin was usually applicable only for the higher class of
organisms such as mammals and insects. As for that reason, the biological activity
(LC50) of rotenoids resin indicated that the bio-active constituents available in the
rotenoids resin (e.g.: rotenone, deguelin, tephrosin and 12αβ-rotenolone) were
considered very active against brine shrimp (Artemia salina) although insignificant to
obtain the optimum processing parameters.
In addition, the biological activity (LC50) of rotenoids resin in this study showed
higher LC50 level (hazard indicator class II: ≤ 1.00 ppm) as compared to the rotenone
standard (SIGMA-Aldrich™) (hazard indicator class III: 17.45 ppm). The contrary was
due to the other constituent’s such as deguelin, tephrosin and 12αβ-rotenolone which
contribute to the toxicity of the extracts rather than the rotenone only. Furthermore, a
research by Pagan and Hageman (1949) demonstrated that the toxicological value of
Derris roots was indirectly correlated with the yield of rotenone (mg) only but it was
correlated to the total extractives compounds (resin + rotenone + other toxic
constituents). As a result, the biological activity (LC50) of rotenone standard (SIGMA-
Aldrich™) with purity of 95 - 98 % (w/w) produced lower toxicity value as compared to
the rotenoids resin from literature, preliminary and verification phase. Meanwhile,
analysis of the yield of rotenone (mg) for preliminary, optimization and verification
phase were employed a standard method from Baron (1976) and the AOAC official
method (1983).
148
5.7 Correlation between the yield of rotenoids resin and yield of rotenone
Pearson’s correlation coefficients (r) were calculated to determine whether any
of the response variables were interrelated. The correlation coefficients are shown in
Table 5.20.
Table 5.20: Pearson’s correlation coefficients (r) of the response variables
aND - Not determined: The model for this response variable (biological activity) generated from the Design-Expert® software version 6.0 (Stat-Ease, 2002) was insignificant in order to obtain the optimum processing parameters.
Correlation measures the extent of any linear association, but association does
not imply causation. It frequently happens that two variables are highly correlated not
because one is casually related to the other but because they are both strongly related to
the third variable. Nevertheless, scientific experiments can frequently make a strong
case for causality by carefully controlling the values of all variables that might be related
to the ones under study (Devore and Farnum, 1999).
Figure 5.33 displays the relationship between yield of rotenoids resin in dried
roots; % (w/w) and yield of rotenone in dried roots; % (w/w). The graph was generated
by the Design-Expert® software version 6.0 (Stat-Ease, 2002) in which identified the
ideal combination of parameters that affected the experimental results (response
variables). The results from this study indicated that the yield of rotenone in dried roots;
% (w/w) was negatively (r = -0.46) correlated with the yield of rotenoids resin in the
dried roots; % (w/w) and showed a strong negative correlation amongst these two
responses.
CORRELATION OF RESPONSES
Yield of rotenoids resin
Yield of rotenone
aBiological activity (LC50)
Yield of rotenoids resin -0.46 ND Yield of rotenone -0.46 ND Biological activity (LC50) ND ND
149
DESIGN-EXPERT Plot
Correlation: -0.459
Yield of rotenoids resin in dried roots
Yie
ld o
f rot
enon
e co
nten
t in
drie
d ro
ots
2.48 10.0325 17.585 25.1375 32.69
0.17
2.5825
4.995
7.4075
9.82
Figure 5.33 Pearson’s correlation coefficients (r) between the yield of rotenone in
dried roots; % (w/w) and yield of rotenoids resin in dried roots; % (w/w)
The strong negative correlation between the yield of rotenoids resin in dried
roots; % (w/w) and yield of rotenone in dried roots; % (w/w) indicated that at a large
increased of the yield of rotenoids resin in dried roots; % (w/w), there would be a large
decreased in the yield of rotenone in dried roots; % (w/w) or vice versa. Hence, a higher
yield of rotenoids resin (mg) indicated a lower yield of rotenone (mg) or vice versa.
This can be explained that the yield of rotenone (mg) was strongly affected at high
operating temperature (40 0C) during the concentration process to obtain the rotenoids
resin. Perhaps, this explanation could be strongly related to Dawson et al. (1991)
wherein the temperature appears to affect severely the breakdown of rotenone. The
demotion of rotenone content (mg) may be due to the deterioration process that occurred
during the process of concentrating the Liquid Crude Extract (LCE) under reduced
pressure of 0.3 mbar at 40 0C (Surya and John, 2001). During the process, prolonged
exposure of LCE to the heat may facilitate the dissipation of rotenone into non-toxic
150
dihydrorotenone and H2O (Schnick, 1974; Bradbury, 1986). However, the result was
contradicted with Visetson and Chuchoui (1999); Suraphon and Manthana (2001)
finding wherein the operating temperature above 70 0C may cause decomposition of the
plant active ingredients. Furthermore, the dissipation of rotenone content (mg) during
the concentration process should not be occurred unless there was an instrumentation
failure during the process such as insufficient vacuum pressure pump. In fact, a
sufficient vacuum pressure pump facilitates the solvent evaporation rate below its
normal boiling point at 1.0 bar thus minimize the heat exposure to the extracts.
Correlation for the biological activity (LC50) towards the yield of rotenoids resin
in dried roots; % (w/w) and yield of rotenone in dried roots; % (w/w) were inconsistent
or incompatible. On the contrary, Pagan and Loustalot (1948) have indicated that the
transmittance values (to calculate the yield of rotenone (mg) using light absorption in a
spectrophotometer) of acetone extract of Derris elliptica was moderately close related to
the biological estimation of the toxicity value (determined on houseflies). However, the
finding was in accordance with Jones et al. (1949) wherein indirect relation was
observed between the yield of rotenone (mg) from Derris roots and their insecticidal
value (biological activity). In addition, a research by Pagan and Hageman (1949)
indicated that the toxicological value of acetone extract of Derris roots was imperfectly
correlated with the yield of rotenone (mg) only but substances other than rotenone
(e.g.: deguelin, tephrosin and 12αβ-rotenolone) were also contributed to the toxicity of
the extracts.
CHAPTER VI
CONCLUSIONS AND RECOMMEDATIONS
6.1 Conclusion
The main conclusions from the experiments are summarized as follows:
1. The preliminary experiments showed that the types of solvent, raw material
particles size (mm in diameter) and solvent-to-solid ratio (ml/g) affected the
yield of rotenoids resin in dried roots, % (w/w) and yield of rotenone in dried
roots, % (w/w). The biological activity (LC50) of rotenoids resin in the
preliminary experiments was done separately due to the sampling procedures for
2 hours interval produced insufficient volume (ml) of the Liquid Crude Extract
(LCE) prior to the concentration process. The maximum yield of rotenoids resin
in dried roots obtained from the preliminary experiment was 9.50 % (w/w) whilst
the maximum yield of rotenone in dried roots was 1.95 % (w/w). The biological
activity (LC50) of rotenoids resin was ≤ 1.00 ppm and considered as very active
constituents against brine shrimp (Artemia salina).
152
2. The theoretical maximum yield of rotenoids resin in dried root and yield of
rotenone in dried roots obtained from the optimization experiments were 5.99 %
(w/w) and 12.26 % (w/w) respectively. These theoretical results were selected
based on the desirability values of given solutions. The biological activity (LC50)
of rotenoids resin was insignificant in relation to the optimum processing
parameters due to inconsistency of rotenone content (mg) and the low value of
LC50 which was less than 100 ppm for all treatments. This was as a result of the
presence of other toxic constituents (e.g.: tephrosin, 12αβ-rotenolone and
deguelin) in the rotenoids resin which contribute to the low LC50 values despite
of the lower yield of rotenone (mg).
3. Based on the optimization phase, the most appropriate theoretical processing
parameters for the types of solvent, solvent-to-solid ratio (ml/g) and raw material
particles size (mm in diameter) were acetone, 4.72 and 0.83 respectively. These
theoretical appropriate processing parameters were evaluated thoroughly based
on the desirability values of a given solution. The verification experiment
revealed that these theoretical appropriate processing parameters produced slight
differentiation as compared to the theoretical maximum yield of rotenoids resin
in dried root; % (w/w) and yield of rotenone in dried roots; % (w/w).
4. Based on the multiple response analysis of the yield of rotenone in dried roots; %
(w/w) and rotenone concentration; mg/ml, the consistent and significant
correlation between response variables and independent variables were solvent-
to-solid ratio (ml/g) and raw material particles size (mm in diameter) of the
acetone extract. Therefore, the most appropriate theoretical processing
parameters to achieve high yield of rotenone in dried roots; % (w/w) and
rotenone concentration; % (w/w) were 1.41 mm in diameter of the raw material
particles size, 3.62 ml/g of the solvent-to-solid ratio and acetone. Although the
values predicted contradict slightly to the ANOVA analysis, it is suggested that
the range of appropriate processing parameter for the solvent-to-solid ratio (ml/g)
and raw material particles size (mm in diameter) are:
153
(A) Solvent-to-solid ratio (ml/g): 3.62 ml/g ↔ 4.72 ml/g.
(B) Raw material particles size (mm in diameter): 0.83 mm ↔ 1.41 mm.
5. Overall, the biological activity (LC50) of rotenoids resin in this study showed
higher LC50 level (hazard indicator class II: ≤ 1.00 ppm) as compared to the
rotenone standard (SIGMA-Aldrich™) (hazard indicator class III: 17.45 ppm).
Nevertheless, both rotenoids resin extracted in this study and the rotenone
standard were considered to give moderate effect to mammals since the hazard
indicator class II and III represented moderate toxicity level.
6. Under optimized condition, all processing parameters were significantly
promoted the increment of the yield of rotenoids resin in dried roots, % (w/w)
and yield of rotenone in dried roots, % (w/w) as compared to the method done on
the preliminary phase experiments.
7. The yield of rotenoids resin in dried roots obtained in the verification phase was
11.53 % (w/w) whilst the yield of rotenone in dried roots was 2.61 % (w/w) and
the biological activity (LC50) of rotenoids resin was 0.83 ppm.
8. The results from this study indicated that the yield of rotenone in dried roots; %
(w/w) was negatively (r = -0.46) correlated with the yield of rotenoids resin in
the dried roots; % (w/w) and showed a strong negative correlation amongst these
two responses. The strong negative correlation between the yield of rotenoids
resin in dried roots; % (w/w) and yield of rotenone in dried roots; % (w/w)
indicated that at a large increased of the yield of rotenoids resin (mg), there
would be a large decreased in the yield of rotenone (mg) or vice versa. Hence, a
higher yield of rotenoids resin (mg) indicated a lower yield of rotenone (mg) or
vice versa.
154
6.2 Recommendation
Based on the results and conclusions obtained from this research, the following
recommendations can be drawn for future work:
1. The effect of seasonal climatic changes on the yield of biomass and rotenoids
play a vital factor for ensuring the accurate time to harvest the Derris plant roots
at the highest yield of rotenoids resin (g) and rotenone (mg). It is believed that if
the production of rotenone from Derris species can be studied and understood,
the exact harvest season to produce an optimum yield of rotenone (mg) can be
cultivated so that the production of rotenone at optimum level can be facilitated
at any season.
2. The development of other bio-processing techniques that minimize thermal
degradation of rotenone content (mg) and simultaneously maximize the yield of
rotenoids resin (g) should be studied further.
3. Other biological activity (LC50) method should be implemented in order to
validate and justify the biological activity (LC50) studied. Furthermore, the
bioassay should be investigated further against the major targeted insect pests
such as Spodoptera litura and Plutella xylostella that commonly infested the
cruciferous plant and likely to be persistent to any types of insecticide.
REFERENCES
Andel, T. V. (2000). The diverse uses of fish-poison plants in Northwest Guyana.
Economic Botany. The New York Botanical Garden Press, Bronx, NY, 54 (4) pp.
500-512.
AOAC Official Method (1983). Rotenone in Pesticide Formulations. Liquid
Chromatographic Method. First Action 1983 and Final Action 1991; 983.06 (2000)
AOAC OFFICIAL METHOD OF ANALYSIS.
Anonymous (2005). World population 2005. Population Reference Bureau, Washington,
DC 20009-5728 USA. Http://www.prb.org. Accessed May, 2005.
Augustijn-Beckers, P. W. M., Hornsby, A. G. and Wauchope, R. D. (1994). Pesticide
properties database for environmental decision making II-SCS/ARS/CES.
Additional compounds. Rev. Env. Cont. Tox. 137:1-82, 2-52.
Acevedo-Rodriquez, P. (1990). The occurrence of piscicides and stupefactants in the
plant kingdom. Pages 1-23 in Prance, G. T. and Balick, M. J., eds. New directions in
the study plants and people. Advances in Economic Botany 8. New York Botanical
Garden, New York.
156
Balandrin, M. F., Klocke, J. A., Wurtele, E. S. and Bollinger, W. H. (1985). Natural
plant chemicals: Sources of industrial and medicinal materials. Science. 228: 1154-
1160.
Bradbury, V. K., Gingerich, W. H., Davis, R. A. and Gilderhus, P. A. (1991). Rotenone
persistence in freshwater ponds: Effects of temperature and sediment adsorption.
North American Journal of Fisheries Management. 11, 2: 850-856.
Bremness, L. (1994). Herbs. London: Dorling Kindersly Ltd.
Banthorpe, D. V., Ekundayo, O., Mann, J. and Tunbull, K. W. (1975). Biosynthesis of
monoterpenes from 14C-labelled acetate and CO2. Phytochemistry. 14: 707-715.
Brenner, R. J., Focks, D. A., Arbogast, R. T. and Weaver (1998). Practical use of spatial
analysis in precision targeting for integrated pest management. Am. Ent. 44:
79-101.
Baron, L. R. and Freudenthal, I. R. (1976). Separation of rotenoids by High Pressure
Liquid Chromatography. Journal of Chromatography. 134: 207-209.
Bell, E. A., Fellows, L. E. and Simmonds, M. S. J. (1990). Natural products from the
plants for the control of insect pests. In Hodgson, R. J. Kuhr (eds), Safer
Insecticides, Marcel Dekker, New York, pp. 337-350.
Bob, B. B. and Wilhelm, G. (2004). Metabolic Engineering: Biochemistry and
Molecular Biology of plants. Http://www.uni-hohenheim.de/de/∼schwekdk/
metabolix/. Accessed May, 2004.
Cheng, H. M., Yamamoto and Casida, J. E. (1972). Rotenone photodecomposition.
Journal of Agricultural Food. Chem. 20, 4: 850-856.
157
Coats, J. R. (1994). Risks from the natural versus synthetic insecticides. Ann. Rev. Ent.
39: 489-515.
Cornell, J. A. (1990). How to apply Response Surface Methodology (RSM). 2nd ed.
Wisconsin, US: ASQC. 1-20.
Croteau, R. (1987). Biosynthesis and catabolism of monoterpenoids. Chem. Rev. 87:
929-954.
Coulson, J. M., Richardson, J. F., Backhurst, J. R. and Harker, J. H. (1991). Chemical
Engineering. Vol. 2. 4th Ed. Particle Technology and Separation Processes. Oxford:
Pergamon Press.
Costa, C. N. B., Carvalho, L. O. D. M., Dulta, S. and Pimental, E. S. (1989). “Uso de
timbo urucu (Derris urucu) no controle do piolho (Haematopinus tuberculatus) em
bubalinos.” Boletim de Pesquisa. 78: Ibama, Brasilia.
Catchpole, O. J., Perry, N. B., da Silva, B. M. T., Grey, J. B. and Smallfield, B. M.
(2002). Supercritical extraction of herbs: Saw Palmetto, St. John’s Wort, Kava Root,
and Echinacea. The Journal of Supercritical Fluids. 22, 2: 129-138.
Copping, L. (1998). The bio-pesticides manual. British Crop Protection Council, UK,
pp. 333.
Choi, M. P. K., Chan, K. K. C., Leung, H. W. and Huie, C. W. (2003). Pressurized
liquid extraction of active ingredients (ginsenosides) from medicinal plants using
non-ionic surfactant solutions. Journal of Chromatography A. 983 (1-2): 153-162.
Cacace, J. E. and Mazza, G. (2003). Mass transfer process during extraction of phenolic
compounds form milled berries. Journal of Food Engineering. 59: 379-389.
158
Colquhoun, D. (1971). Lectures on Biostatics. Clarendon Press, Oxford.
Dev, S. and Koul, O. (1997). Insecticides of Natural Origin. Harwood Academic
Publishers GMBH, Amsterdam. The Netherlands, pp. 365.
Dzolkifli, O. (2004). Laboratory and field efficacy studies on the toxicity of
rotenone. Proposal of Industrial Grant Scheme (IGS) 2004 (2nd proposal);
Ministry of Science, Technology and Innovation (MOSTI). Industrial collaboration
project with Heritage Biotech (M) Sdn. Bhd and Chemical Engineering Pilot Plant
(CEPP), UTM Skudai, Johor.
Devore, J. L. and Farhum, N. R. (1999). Applied statistics for engineers and scientists.
Pacific Grove: Duxbury Press.
Dawson, V. K., Gingerich, R. A., Davis, P. A. and Gilderhus (1991). Rotenone
persistence in freshwater ponds: Effects of temperature and sediment adsorption.
North American Journal of Fisheries Management. 11, 2: 226-231.
Dent, D. R. and Lomer, C. J. (1999). Product development and commercialization in
development assistance projects: A case study of LUBILOSA (unpublished report).
CABI/IITA, pp. 25.
De-Wilde, A. R., Heyndrickx, A. and Carton, D. (1986). A case of fatal rotenone
poisoning in a child. Jour. Forensic Sci. 31: 1492-8.
Davies, O. L. (Ed.) (1954). The design and analysis of industrial experiments. London:
Oliver and Boyd.
Devlin, J. F. and Zettel, T. (1999). Ecoagriculture: Initiatives in Eastern and Southern
Africa. Weaver Press, Harare.
159
Ellenhorn, M. J. and Barceloux, D. G. (1988). Medical toxicology: Diagnosis and
treatment of human poisoning. New York, Elsevier.
Extoxnet (1996). Pesticides news: Pesticide information notebook rotenone. No. 54,
December 2001, pages 20-21.
EPA - Environmental Protection Agency (1994). Chemical summary for the methylene
chloride (dichloromethane). Prepared by: Office of Pollution Prevention and Toxics,
USA Environmental Protection Agency (EPA), August, 1994.
EPA - Environmental Protection Agency (1996). US EPA Classification of Pesticides by
Acute Human Health Hazards. Farm Chemical Handbook (1996) and Crop
Protection Reference (1996).
Faridah, A. F. (1998). The commercialization of local medicinal herb in skin care and
toiletries product. In Nair, M. N. B. and Ganapathi, N. (Ed.). Medicinal plants: Cure
for the 21st century. Biodiversity, conservation and utilization of medicinal plants.
Proceeding of the seminar. 15-16 October 1998. Serdang: Universiti Putra Malaysia.
130-132.
Finney, D. J. (1964). Statistical method in biological assay. 2nd Edition. Griffin, London.
Finney, D. J. (1971). Probit analysis. 3rd Edition. Cambrigde University Press,
Cambrigde.
Fukami, J. -L., Yamamoto, L. and Casida, J. E. (1967). Metabolism of rotenone in vitro
by tissue homogenates from mammals and insects. Science. 155: 713-716.
Francis, A. G. and Franklin, M. T. (1943). The location and state of rotenone in the root
of Derris elliptica. Journal of Agricultural Research. No. 524, University of
California, United States of America.
160
Frank, T. C., Downey, J. R. and Gupta, S. K. (1999). Quickly screen solvent for organic
solids. Chemical Engineering Progress (December); 41-61.
Frank, M., Greco, Marianne, P., Fitzpatrick, M. S., Wendy, S. and Graffam. (1996)
Preliminary evaluation of selected nutrient composition of two life stages of Artemia
salina before and after feeding an enriched Torula yeast product. The Journal of
MaquaCulture. Volume 4; Issue 3.
Farnsworth, N. R. (1966). Biological and phytochemical screening of plants. J. Pharm
Sci. 55: 225-276.
Fasihuddin, A. and Hasmah, R. (1993). Kimia hasilan semulajadi dan tumbuhan ubatan.
Kuala Lumpur. Dewan Bahasa dan Pustaka (DBP).
Grainge, M. and Ahmed, S. (1998). Handbook of plants with pest-control properties.
Wiley, New York, pp. 470.
Gaikar, V. G. and Dandekar, D. V. (2001). Process for extraction of curcuminoids from
Curcuma species. USPTO PATENT FULL-TEXT AND IMAGE DATABASE,
(U. S Patent: 6,224,877).
Geankoplis, C. J. (1993). Transport processes and unit operations. 3rd Ed. Prentice Hall.
New Jersey.
Grinda, Francoise, Gueyne and Jean (1986). Extraction of insecticides from plants.
USPTO PATENT FULL-TEXT AND IMAGE DATABASE, SAPHYR S. A. R. L.
(France) (U. S Patent: 4,698,222).
161
Gertenbach, D. D. (2001). Solid-liquid extraction technologies for manucfacturing
nutraceuticals from botanicals. In functional foods: Biochemical and processing
aspects. Shi, J., Mazza, G. and Le-Maguer, M. (eds.) CRC press. Boca Raton,
Florida, Vol 2: 331-366.
Gaby, S. (1986). Natural crop protection in the tropics - DERRIS: Derris elliptica, D.
Malaccensis, D. uliginosa. Copyright Margraf Publishers Scientific Books.
Gosselin, R. E., Smith, R. P. and Hodge, H. C. (1984). Clinical toxicology of
commercial products. 5th (ed.) Baltimore/London: Williams and Wilkins, pp. 111-
366-7.
Gosalvez, M. (1983). Carcinogenesis with the insecticide rotenone. Life Science. 32:
809-816.
Goodman, L. S., Gilman, A. G., Rall, T. W. and Murad, F. (1985). Goodman and
Gilman's: The pharmacological basis of therapeutics. 7th (ed.) New York, Toronto,
London: Macmillan pp. 1839.
HealthComm. International. Inc. (1998). Phytochemicals in functional medicine and
inflammation. Technical Bulletin.
Harborne, J. B. (1984). Phytochemical method. 2nd ed. Chapman and Hall. London.
Heal, R. E., Rogers, E. F., Wallance, R. T. and Starnes, O. (1950). A survey of plants of
insecticidal activity. Llodyia. 13: 89-162.
Houghton, P. J. and Raman, A. (1998). Laboratory handbook for the fractionation of
natural extracts. Chapman and Hall. London.
162
Hamid, A. (1999). Derris Lour. Pages 234-242 in de Padua, L. S., Bunyapraphatsara, N.,
Lemmens, R. H. M. J., eds. Plant Resources of South-East Asia 12 (1): Medicinal
and poisonous plants 1. Backhuys Publishers, Leiden.
Harris, J. (1999). Pesticides in perspective. CAB International.
Hao, N. C., Hien-Phan-Phuoc, et al. (1998). Technology to produce the preparations
from the ‘Tuba’ roots (Derris elliptica Benth). Proceedings of scientific works for 5
years 1993-1998, Institute of Tropical Biology ITB, National Center for Natural
Sciences and Technology of Vietnam-NCST, HCM Publishing house of Agriculture
(274-281).
Hien-Phan-Phuoc, Hao-Nguyen-Cong and Hoa-Nguyen-Chong. (1996). Researchers on
the growing conditions and effect of using rotenone from Derris elliptica Benth in
Mekong River Delta. Journal of Science, Technology and Economic Management.
Agriculture and Food Industry, No 12/96 (522-524).
Hien-Phan-Phuoc, Hao-Nguyen-Cong, Tri-Mai-Dinh, Giang-Nguyen and Cuu-Thi-
Huong. (1999). Relationships between the formation and accumulation of rotenone
in Derris plant. Proceedings of the National Biotechnology Conference 1999:
Hanoi publishing house 1999 (810-816).
Hien-Phan-Phuoc, Hao-Nguyen-Cong, Tri-Mai-Dinh, Giang-Nguyen and Cuu-Thi-
Huong. (2000). Researches on the formation, translocation and accumulation of
rotenone in nature. Vietnamese Journal of Science and Technology. No. 3, Vol.
XXXXVIII (116-127).
Hayes, W. J. Jr. (1982). Pesticides studied in man. Baltimore/London: Williams and
Wilkins, 81-6.
163
Hewlett, P. S. and Plankett R. L. (1979). The interpretation of Quantal Responses in
Biology. Edward Arnold (Publishers) Limited.
Ikan, R. (1991). Natural products: A laboratory guide. 2nd edn, Academic Press,
London.
John, R. A. and Ron, R. K. (1944). INSECT CONTROL. Jour. Econ. Ento. 37: 400-408.
Jeyaratnam, J. (1990). World Health Statistics. Quarterly No. 43. World Health
Organisation (WHO).
Jones, A. M. and Pagan, C. (1949). A futher toxicological comparison of Derris and
Lonchocarpus. Journal of Agricultural Research. Vol. 77, No. 9, 10, Washington
D. C.
Kole, R. K., Satpathi, C. and Chowdhuri (1992). Isolation of amorphalone, a potent
rotenoid insecticide from Tephrosia candida. J. Agri. Food Chem. 40: 1208-1210.
Koolhass, D. R. (1938). Peculiarities of the roots of different Derris varieties.
Bergcultures. 12: 1045-1053.
Kidd, H. and James, D. R. (1991). The Agrochemicals Handbook, Third Edition. Royal
Society of Chemistry Information Services, Cambridge, United Kingdom: 2-13.
Kuehl, R. (2000). Design of Experiments: Statistical principles of research design and
analysis. 2nd ed. Duxbury Press. Pacific Grove.
Ken, R. (1999). How i successfully hatch Brine Shrimp. Artemia FAQ 2.0.
Http://www.angelfire/artemia.html. Accessed January, 2005.
164
Koul, O. and Smirle, M. J. (1993). Allomones in insect-plant interactions. In G. T. gujar
(ed.), Recent advances in insect physiology and toxicology, Agricole Publishing
Academy, New Delhi, pp. 14-40.
Lisansky, S. (1997). Microbial bio-pesticides. In microbial insecticides: Novelty or
necessity? British Crop Protection Council Proceedings/Monograph Series No.
68, pp. 3-10.
List, P. H. and Schmidt, P. C. (1989). Phytopharmaceutical technology. CRC Press,
Boca Raton.
Leonard, J. W. (1939). Notes on the use of Derris as a fish poison. Transactions of the
American Fisheries Society. 68: 269-279.
Lang, Q. and Wai, C. M. (2001). Supercritical fluid extraction in herbal and natural
product studies: A practical review. Talanta. 53: 771-782.
Love, J. (1999). Phytochemicals, functional foods, nutraceuticals: New names for an
emerging field. Http://www.exnet.iastate.edu/pages/xicosett/9feb12.html. Assessed
July, 2004.
Mourad, K. and Anne-Marie, M. (1986). Lonchocarpene and Lonchocarpusone, an
isoflavone: Two pyranopolyphenols from Lonchocarpus Nicou roots. Natural
products. No. 2, pp. 281-285.
McLaughlin, J. L. and Rogers, L. L. (1998). The use of biological assays to evaluate
botanicals. Drug Information Journal. Vol. 32, pp. 513-524.
McLaughlin, J. L. (1991). Crown gall tumours on potato disc and brine shrimp lethality:
Two simple bioassay for higher plant screening and fractionation. In K.
Hostettmann, Assays of Bioactivity, Academic Press, San Diego, pp. 2-32.
165
Moriya, M., Ohta, T. and Watanabe, K. (1983). Further mutagenicity studies on
pesticides in bacterial reversion assay system. Mutat. Res. 116: 185-216.
Mircea, V. (2001). An overview of the ultrasonically assisted extraction of bioactive
principles from herbs. Ultrasonics Sonochemistry. 8: 303-313.
Mizubuti, I. Y., Junior, O. B., de-Oliveira Souza, L. W., de-Silva, R. S. and Ida, E. I.
(2000). Response Surface Methodology (RSM) for extracting optimization of pigeon
pea protein. Food Chemistry. 70: 259-265.
Murray, B. I. (1997). Neem and other botanical insecticides: Barriers to
commercialization. Phtyoparasitica. 25 (4): 339-334.
Matsumura, F. (1985). Toxicity of insecticides. Plenum Press, New York.
Manuel, F. B. and James, A. K. (1985). Natural plant chemicals: Sources of industrial
and medicinal materials. Science. 228: 1154-60.
Mantell, C., Rodriquez, M. and de-la Ossa, E. M. (2002). Semi-batch extraction of
anthocyanins from the red grape pomace in packed beds: Experimental results and
process modelling. Chemical Engineering Science. 57: 3831-3838.
Mazza, G. and Cacace, J. E. (2002). Extraction of anthocyanins and other phenolics
from black currants with sulfured water. Journal of Agricultural and Food
Chemistry. 50: 5939-5946.
Mourad, K., Amegninou, A. and Anne-Marie, M. (1986). Lonchocarphene, a stilbene
and lonchocarpusone, an isoflavone: Two new pyranopolyphenols from
Lonchocarpus nicou roots. Journal of Natural Products. No. 2, pp. 281-285.
Maas, J. G. J. A. (1938). The cultivation of Derris root. Bergcultures. 9: 1059-1258.
166
Moore, R. H. (1940). Insecticidal crops investigations. Puerto Rico (Mayaquez) Fed.
Expt. Sta. Rpt. 1939: 71.
Moretti, C. and Grenard, P. (1982). Les nivrees ou plantes ichthyotoxiques de la Guyane
Francaise. J. Ethnopharm. 6: 139-160.
Matleson, P. (1995). The 50 % pesticide cuts in Europe: A glimpse of our future. Am.
Ent. 41: 210-220.
National Research Council (1983). Drinking water and health. Vol. 5 Washington, D.
C.: National Academy Press pp. 98.
Ollanketo, M., Peltoketo, A., Hartonen, K., Hiltunen, R. and Riekkola, M. (2002).
Extraction of sage (Salvia officinalis L.) by pressurized hot water and conventional
methods: Antioxidant activity of the extracts. European Food Research and
Technology. 215, 2: 158-163.
Oberg, K. -E. (1961). Site of action of rotenone in the respiration chain. Exp. Cell Res.
24: 163-164.
Philip, L., Williams, Robert, C., James, Stephen, M. and Roberts (2000). Principles of
Toxicology: Environmental and industrial applications. 2nd edition, John Wiley and
Sons, Inc.
Pagan, C. and Hageman, R. H. (1949). The effect of sun, shade and oven-drying on the
toxicological and chemical values of Derris root. Journal of Agricultural Research.
Vol. 78, No. 11. Key No. P. R. - 13, Washington D. C.
Pinelo, M., Sineiro, J. and Nunez, M. J. (2006). Mass transfer during continuous solid-
liquid extraction of antioxidants from grape byproducts. Journal of Engineering. 77
57-63.
167
Pagan, C. and Loustalot, A. J. (1948). A simple rapid method for estimating toxicity
(rotenone equaivalent) of Derris root. Journal of Agricultural Research. Vol. 77, No.
9, 10, Washington D. C.
Phan-Phuoc-Hien, Heinz, G. and Raymond, K. (2003). Rotenone-potential and prospect
for sustainable agriculture. Omonrice. 11: 83-92.
Pitigon, W. and Sangwanit, A. (1997). The study of Derris extract in insect control. pp.
42-48. In Agricultural Toxic Substances Conference. Department of Agriculture.
Ministry of Agriculture and Cooperatives. 2-10 July, 1997.
Pagan, C. and Hageman, R. H. (1949). Effect of root diameter on chemical and bio-toxic
constituents of Derris. Journal of Agricultural Research. Vol. 78, No. 11. Key No.
P. R. - 14, Washington D. C.
Pagan, C. and Loustalot, A. J. (1949). Comparison of chemical values with the
toxicological rotenone equivalent of Derris and Lonchocarpus roots. Journal of
Agricultural Research. Vol. 78, No. 7, Washington D. C.
Raws, R. L. (1986). Experts probe issue, chemistry of light-activated pesticides. Chem.
Eng. News Sept. 22: 2124.
Rahman, A. R. (2004). Response Surface Methods (RSM). A two days course using
Design Expert Software version 6.0 (Stat-Ease, 2002). 6-7 February, 2004. Chemical
Engineering Pilot Plant (CEPP), UTM Skudai, Johor Darul Takzim.
Rodney, B. and Alan, H. (1976). Determination of rotenone in pesticide formulations
and the separation of six rotenoids by reversed-phase High Performance Liquid
Chromatography. Journal of Chromatography. 134: 210-215.
168
Reinertaen, H., Jensen, A., Koksvik, J. I. and Olsen, Y. (1987). Effects of fish removal
on the the limnetic ecosystem of an eutrophic lake. J. Fish Aquat. Sci. 47: 166-173.
Rydberg, J., Musikas, C. and Choppin, G. R. (Eds.) (1992). Principles and practises of
solvent extraction. New York: Marcel Dekker, Inc.
Raven, L. (1973). Drug action and reactions. Levine, R. R., Pharmacology,
Philadelphia.
Rice, E. L. (1995). Biological control of weeds and plant diseases. Advances in Applied
Allelopathy, University of Oklahoma Press.
Salisbury, F. B. and Ross, C. (1974). Plant physiology. Prentice-Hall of India Private
Limited, New Delhi.
Saiful, I. Z., Mohammad, R. S., Ramlan, A. A., Mohd, K. A. and Dzolkifli, O. (2003).
Development of Environmental Friendly Natural Bio-pesticide from Local Plant
Species (Tuba). Proposal of Industrial Grant Scheme (IGS) 2003 (1st proposal);
Ministry of Science, Technology and Environment (MOSTE). Industrial collaboration
project with Heritage Biotech (M) Sdn. Bhd and Chemical Engineering Pilot Plant
(CEPP), UTM Skudai, Johor.
Shane, M. and Doug, P. (2000). Backgrounder: Rotenone and Parkinson's Disease.
Nature Neuroscience. Vol. 3, pp. 1301.
Stat-Ease (2002). Design-Expert 6.0 user’s guide. Version 6.0 Design-Expert® software.
Starr, F., Starr, K. and Loope, L. (2003). Derris elliptica, poison vine, Fabaceae. United
States Geological Survey, Biological Resources Division, Haleakala Field Station,
Maui, Hawai'i.
169
Surya, H. and John, B. B. (2001). Initial studies on alkaloids from Lombok Medicinal
Plants. Molecules. 6: 117-129.
Suraphon, V. and Manthana, M. (2001). Effects of root extract from Derris (Derris
elliptica Benth) on mortality and detoxification enyzme levels in the Diamondback
Moth Larvae (Plutella xylostella Linn.). Am. Ent. 8: 100-120.
Schnick, R. A. (1974). A review of the literature on the use of rotenone in fishies. La
Crosse, W. L.: Fish Control Laboratory.
Seikova, I., Simeonov, E. and Inavona, E. (2004). Protein leaching from tomato seed:
Experimental kinetics and prediction of effective diffusivity. Journal of Food
Engineering. 61, 2: 165-171.
Schwatrzberg, H. G. and Chao, R.Y. (1982). Solute diffusivities in leaching processes.
Food Technology. 36, 2: 73-86.
Sukamar, K., Perch, M. J. and Boobar, L. R. (1991). Botanical derivates in mosquito
control: A review. J. Am. Mosq. Control Assoc. 7: 210-237.
Salleh, M. N. (1998). The forest can cure all ills; Myth or reality. In Nair, M. N. B. and
Ganapathi, N. (Ed.). Medicinal plants: Cure for the 21st Century. Biodiversity,
conservation and utilization of medicinal plants. Proceeding of the seminar. 15-16
October 1998. Serdang: Universiti Putra Malaysia (UPM). 1-6.
Smith, R. M. (1999). Supercritical fluids in separation science: The dreams, the reality
and the future. Journal of Chromatography. 856: 83-115.
Stoll, G. (1988). Natural plant protection is the tropics. Magraf Publishers,
Weikersheim.
170
Schmeltz, I. (1971). Nicotine and other tobacco alkaloids. In M. Jacobson and D. G.
Crosby (eds), Naturally Occuring Insecticides, Marcel Dekker, New York, pp.
99-136.
Smith, M. W. (1941). Treatment of Potter’s Lake, New Brunswick, with rotenone.
Transactions of the American Fisheries Society. 70: 347-355.
Tiwari, K. K. (1995). Extraction technologies related to food processing. In Gaonkar, A.
G. (Ed.). Food processing: Recent developments. Amsterdam: Elsevier Science. 269-
301.
Taylor and Francis (1985). Basic toxicology: Fundamentals, target organs and risk
assessment. New York.
Vickery, M. L. and Vickery, B. (1981). Secondary plant metabolism. Macmillan Press
Ltd. Hong Kong.
Van, R. N. (1991). The dose makes the poison. Ottoboni, M. A., New York.
Visetson, S. and Milne, M. (2001). Effects of roots extracts from Derris (Derris elliptica
Benth) on mortality and detoxicification enyzme levels in the Diamondback Moth
Larvae (Plutella xylostella Linn.). Jour. Nat. Sci. 35: 157-163.
Visetson, S. And Chuchoui, L. (1999). The efficiency of lemon grass (Cymbopogon
winterianus Jewitti) and neem (Azadiractha indica var. Siamensis, Valeton) extracts
on the changes in detoxification enyzmes in the dog tick (Rhipicephalus sanguineus
Latreille). pp. 847-851. In the 3rd Annual Conference in Biodiversity in Thailand
(BRT). 11-14 October, 1999. Hatyai, Songkla, Thailand.
Vermeulen, N. (1998). Encyclopedia of herbs. Vancouver: Whitecap Books.
171
Voirin, B., Brun, N. and Bayet, C. (1990). Effects of day length on the monoterpene
composition of leaves of Mentha piperita. Phytochemistry. 29: 749-755.
WHO - World Health Organisations (1997). Vector Control: Methods for the use by
individual and communities. Prepared by Jan A. Rozendaal, pp. 412.
WHO - World Health Organisations (1996). The International Programme on Chemical
Safety (IPCS): Methylene chloride. Environmental Health Criteria 164.
WHO - World Health Organisation (1992). The WHO recommended classification of
pesticides by hazard and guidelines to classification 1992-1993. Geneva, World
Health Organisation, pp. 66 (Unpublished document, WHO/PCS/92.14).
Walton, N. J. and Brown, D. E. (1999). Chemical from plants: Perspectives on plant
secondary products. Imperial College Press. Singapore.
Waage, J. K. (1999). Beyond the realm of conventional biological control: harnessing
bio-resources and developing biologically-based technologies for sustainable pest
management. Proceedings of the Symposium on biological control in the tropics.
Held at MARDI Training Centre, Serdang, Malaysia, 18-19 March 1999, pp. 5-17.
Weier, J. L. and Starr, D. F. (1950). The use of rotenone to remove rough fish for the
purpose of improving migratory waterfowl refuge areas. J. Wildl. Manage. 14, 2:
203-205.
Wilkonson, J. A. (1998). The potential of herbal products for nutraceutical and
pharmaceutical development. International Business Communication (IBC) Fifth
annual conference. Functional Foods. 7-8 September 1998. London.
172
Yu, S. J. And Hsu, E. L. (1985). Induction of hydrolases by allelochemicals and host
plants in fall armyworm (Lepidoptera: Noctuidae) larvae. Environ. Ent. 14: 512-
515.
APPENDICES
174
Appendix A
Table A-1: Complete results of the optimization phase experimental design
Sample Weight of dried roots
Particles size (mm in diameter)
Ratio (ml/g) Solvent
bVol. extract ml (before)
1 29.47 g 5.00 6.00 Acetone 150.0 2 29.91 g 2.75 10.00 aEthanol 250.0 3 29.62 g 4.09 3.62 aEthanol 96.0 4 29.95 g 1.41 3.62 Acetone 70.0 5 30.56 g 4.09 8.38 aEthanol 220.0 6 29.27g 2.75 2.00 Acetone 28.0 7 29.34 g 4.09 3.62 Acetone 80.0 8 29.78 g 4.09 8.38 aEthanol 220.0 9 29.74 g 1.41 3.62 Acetone 69.0 10 31.74 g 2.75 6.00 aEthanol 135.0 11 29.98 g 5.00 6.00 aEthanol 160.0 12 29.44 g 1.41 8.38 aEthanol 217.0 13 29.49 g 2.75 6.00 Acetone 152.0 14 29.55 g 4.09 3.62 aEthanol 90.0 15 30.01 g 0.50 8.38 aEthanol 103.0 16 29.54 g 4.09 3.62 Acetone 76.0 17 29.36 g 2.75 6.00 Acetone 151.0 18 29.71 g 2.75 6.00 aEthanol 157.0 19 29.30 g 1.41 8.38 Acetone 205.0 20 30.06 g 4.09 8.38 Acetone 206.0 21 29.37 g 0.50 6.00 Acetone 130.0 22 29.31 g 1.41 8.38 aEthanol 215.0 23 29.78 g 2.75 6.00 aEthanol 154.0 24 29.98 g 2.75 6.00 Acetone 141.0 25 28.89 g 4.09 8.38 Acetone 225.0 26 29.64 g 1.41 3.62 aEthanol 77.0 27 29.63 g 1.41 8.38 Acetone 213.0 28 29.46 g 2.75 2.00 aEthanol 34.0 29 30.14 g 2.75 10.00 Acetone 265.0 30 29.41 g 1.41 3.62 aEthanol 55.0 aEthanol was added with the H2O and oxalic acid - A ratio of ethanol (9): H2O (1) [Prepare 1.0 mg/ml of oxalic acid solution from the volume (ml) of H2O ratio (1)]. bVolume of Liquid Crude Extract (LCE) before the concentration process.
175
Continued:
Sample cVol. extract ml (after) dmg/ml (before) eMass (mg) f% (w/w) in dried roots 1 10.00 0.80 119.40 0.41 2 103.80 1.66 414.50 1.39 3 35.00 0.52 49.92 0.17 4 5.40 42.09 2946.30 9.84 5 75.00 1.11 244.20 0.80 6 5.00 2.00 56.00 0.19 7 8.40 2.97 237.76 0.81 8 38.00 0.66 145.20 0.49 9 5.20 37.70 2601.30 8.75 10 41.00 1.00 135.00 0.43 11 26.40 2.53 404.80 1.35 12 60.00 0.68 147.56 0.50 13 29.00 2.14 325.28 1.10 14 27.00 0.92 82.35 0.28 15 34.00 2.84 292.52 0.97 16 28.00 2.07 157.32 0.53 17 14.00 11.64 1757.64 5.99 18 55.00 1.40 219.80 0.73 19 48.00 0.86 176.30 0.60 20 21.00 0.79 161.71 0.54 21 5.60 6.39 830.31 2.83 22 59.00 0.62 133.30 0.45 23 49.00 1.84 283.36 0.95 24 12.00 10.35 1459.35 4.87 25 14.00 0.44 99.00 0.34 26 24.00 2.50 192.50 0.65 27 57.00 0.50 106.50 0.36 28 10.90 1.71 58.14 0.20 29 12.00 3.55 941.28 3.12 30 21.80 3.58 196.90 0.67
cVolume of the Concentrated Liquid Crude Extract (CLCE): The solvent was removed approximately 50 % to 90 % depending on how much the volume of Liquid Crude Extract (LCE) can be obtained after the extraction process. dConcentration of LCE (mg/ml) before the concentration process. eMass of rotenone (mg) in the LCE before the concentration process. fYield of rotenone in dried roots before the concentration process, % (w/w) = eYield of rotenone (mg)/weight of dried roots (g) × 100 % [Selected as response variable (THIS MODEL WAS SIGNIFICANT; R2 = 0.50)].
176
Continued:
Sample gmg/ml (after) hMass (mg) i% Losses jWeight of CLCE (g) + 10% to 50% solvent1 11.61 116.12 2.75 7.06 2 2.86 296.87 28.38 86.09 3 1.65 57.75 -15.69 27.79 4 36.40 196.56 93.33 4.75 5 2.27 170.55 30.16 61.77 6 5.12 25.60 54.29 2.50 7 19.98 167.83 29.41 6.79 8 2.73 103.74 28.55 32.12 9 40.47 210.44 91.91 4.33 10 2.86 117.26 13.14 33.86 11 2.63 69.43 82.85 23.00 12 1.49 89.40 39.41 50.67 13 10.77 312.33 3.98 21.85 14 2.32 62.64 23.93 22.20 15 6.64 225.59 22.88 26.22 16 4.97 139.16 11.54 20.07 17 12.08 169.12 90.38 11.26 18 1.40 77.00 64.97 45.82 19 3.20 153.60 12.88 38.29 20 9.78 205.38 -27.01 14.46 21 36.03 201.77 75.70 4.84 22 1.88 110.92 16.79 50.00 23 1.74 85.26 69.91 40.77 24 8.56 102.72 92.96 13.12 25 6.12 85.68 13.45 10.42 26 3.22 77.28 59.85 18.92 27 1.58 90.06 15.44 45.86 28 1.31 14.28 75.44 9.37 29 60.20 722.40 23.25 9.07 30 6.84 149.13 24.26 17.91
gConcentration of CLCE (mg/ml) after the concentration process. hMass of rotenone (mg) in the CLCE after the concentration process. iLosses of rotenone during the concentration process (%): mg (before) - mg (after)/mg (before) × 100 %. jWeight of CLCE + 10 % to 50 % of the remaining solvent in CLCE after the concentration process. The solvent was kept at certain volume (ml) in order to prepare an initial concentration (C0) for the bioassay prior to the dilution of 1000, 500, 100, 50, 10, 1.0 ppm respectively (Figure 4.8).
177
Continued:
Sample kWeight of
rotenoids resin lAdditional weight of resin
from CLCE & LCE (g) mTotal weight of
rotenoids resin (g) 1 0.59 0.14 0.73 2 6.59 0.22 6.81 3 6.79 0.86 7.65 4 0.54 0.21 0.75 5 8.21 0.41 8.62 6 3.51 1.95 5.46 7 7.60 1.99 9.59 8 5.40 0.36 5.76 9 0.66 0.26 0.92 10 6.90 0.57 7.47 11 6.29 0.84 7.13 12 7.35 0.57 7.92 13 3.64 0.30 3.94 14 4.33 0.60 4.93 15 5.89 0.44 6.33 16 3.88 0.39 4.27 17 2.89 0.48 3.37 18 8.73 0.77 9.50 19 5.50 0.37 5.87 20 1.66 0.19 1.85 21 5.70 2.15 7.85 22 2.92 0.20 3.12 23 7.56 0.66 8.22 24 1.05 0.22 1.27 25 2.30 0.44 2.74 26 4.43 0.60 5.03 27 2.36 0.19 2.55 28 1.53 0.72 2.25 29 1.59 0.27 1.86 30 7.88 0.93 8.81
kWeight of rotenoids resin (g) after evaporated 100 % of solvent. mTotal weight of rotenoids resin = Weight of resin from the extract (pure resin) + weight of LCE (1.0 ml) for the RP-HPLC analysis + weight of CLCE (1.0 ml) for the RP-HPLC analysis + weight of CLCE for the bioassay (calculate using M1V1 = M2V2). (All results were calculated using the proportional method) Example calculation for Sample 1 (S1): 10.0 ml CLCE = 0.59 g resin, if 1.0 ml CLCE = 0.06 g resin (CLCE ratio) For the bioassay, the volume of CLCE to produce 1.0 mg/ml (initial conc.) was using the M1V1 = M2V2: (11.61 mg/ml)(?) = (1.0 mg/ml)(4.0 ml of diluted solution) = ∴V1 = 0.34 ml CLCE (Figure 4.8). Therefore, TOTAL WEIGHT OF ROTENOIDS RESIN = [1.0 ml CLCE = 0.06 g resin]: l[0.34 ml CLCE (bioassay) + 1.0 ml CLCE (HPLC) + 1.0 ml of LCE for the RP-HPLC analysis (using the proportion calculation from the CLCE ratio as ststed above and assuming that the yield of roteone before
178
and after the concentration process were remain unchanged)] = 0.06 g/1.0 ml × 2.34 ml = 0.14 g + 0.59 g =∴0.73 g (Weight of pure resin from the extract) Continued:
Sample nYield of rotenoids resin in dried roots, % (w/w)
oYield of rotenone in dried roots, % (w/w)
pYield of rotenone in rotenoids resin, % (w/w)
1 2.48 0.39 15.91 2 22.77 0.99 4.36 3 25.83 0.19 0.75 4 2.50 0.66 26.21 5 28.21 0.56 1.98 6 18.46 0.09 0.47 7 32.69 0.57 1.75 8 19.34 0.35 1.80 9 3.09 0.71 22.87 10 23.53 0.37 1.57 11 23.78 0.23 0.97 12 26.90 0.30 1.13 13 13.36 1.06 7.93 14 16.68 0.21 1.27 15 21.09 0.75 3.56 16 14.45 0.47 3.26 17 11.48 0.58 5.02 18 31.98 0.26 0.81 19 20.03 0.52 2.62 20 6.15 0.68 11.10 21 26.73 0.69 2.57 22 10.64 0.38 3.56 23 27.60 0.29 1.04 24 4.24 0.34 8.09 25 9.49 0.30 3.13 26 16.97 0.26 1.54 27 8.61 0.30 3.53 28 7.64 0.05 0.63 29 6.17 2.40 38.84 30 29.96 0.51 1.69
nYield of rotenoids resin in dried roots, % (w/w) = mTotal weight of rotenoids resin (mg)/weight of dried roots (g) × 100 %. [Selected as response variable (THIS MODEL WAS SIGNIFICANT; R2 = 0.30)]. oYield of rotenone in dried roots after the concentration process, % (w/w) = hYield of rotenone after the concentration process (mg)/weight of dried roots (g) × 100 % [THIS RESPONSE VARIABLE WAS INSIGNIFICANT FOR IDENTIFICATION OF THE OPTIMUM PROCESSING PARAMETERS]. pYield of rotenone in rotenoids resin after the concentration process, % (w/w) = hYield of rotenone after the concentration process (mg)/ mtotal weight of rotenoids resin (g) × 100 % [THIS RESPONSE VARIABLE WAS INSIGNIFICANT FOR IDENTIFICTION OF THE OPTIMUM PROCESSING PARAMETERS].
184
Appendix C
Table C-1: Experimental design and results (experimental and predicted values) of
Central Composite Design; CCD with response surface linear model
(Manual calculation)
Run Factor 1 X1: Solvent -to-solid ratio
Factor 2 X2: Raw material particles size
Factor 3 X3: Types of solvent
Yield of rotenoids resin in dried roots (Experiment)
Yield of rotenoids resin in dried roots (Predicted)
ml/g mm in diameter Treatment % (w/w) % (w/w) 1 6.00 5.00 Acetone 2.48 11.86 2 10.00 2.75 aEthanol 22.77 21.70 3 3.62 4.09 aEthanol 25.83 22.41 4 3.62 1.41 Acetone 2.50 12.37 5 8.38 4.09 aEthanol 28.21 21.82 6 2.00 2.75 Acetone 18.46 12.49 7 3.62 4.09 Acetone 32.69 12.21 8 8.38 4.09 aEthanol 19.34 21.82 9 3.62 1.41 Acetone 3.09 12.37 10 6.00 2.75 aEthanol 23.53 22.19 11 6.00 5.00 aEthanol 23.78 22.06 12 8.38 1.41 aEthanol 26.90 21.98 13 6.00 2.75 Acetone 13.36 12.00 14 3.62 4.09 aEthanol 16.68 22.41 15 6.00 0.50 aEthanol 21.09 22.33 16 3.62 4.09 Acetone 14.45 12.21 17 6.00 2.75 Acetone 11.48 12.00 18 6.00 2.75 aEthanol 31.98 22.19 19 8.38 1.41 Acetone 20.03 11.78 20 8.38 4.09 Acetone 6.15 11.62 21 6.00 0.50 Acetone 26.73 12.13 22 8.38 1.41 aEthanol 10.64 21.98 23 6.00 2.75 aEthanol 27.60 22.19 24 6.00 2.75 Acetone 4.24 12.00 25 8.38 4.09 Acetone 9.49 11.62 26 3.62 1.41 aEthanol 16.97 22.57 27 8.38 1.41 Acetone 8.61 11.78 28 2.00 2.75 aEthanol 7.64 22.69 29 10.00 2.75 Acetone 6.17 11.50 30 3.62 1.41 aEthanol 29.96 22.57
aEthanol was added with the H2O and oxalic acid - A ratio of ethanol (9): H2O (1) [Prepare 1.0 mg/ml of oxalic acid solution from the volume (ml) of H2O ratio (1)].
185
Run (N) Y u Ŷ u Y u - Ÿ (MEAN) Ŷ u - Ÿ (MEAN) Y u - Ŷ u 1 2.48 11.86 -14.62 -5.24 -9.38 2 22.77 21.70 5.67 4.60 1.07 3 25.83 22.41 8.73 5.31 3.42 4 2.50 12.37 -14.60 -4.73 -9.87 5 28.21 21.82 11.11 4.72 6.39 6 18.46 12.49 1.36 -4.61 5.97 7 32.69 12.21 15.59 -4.89 20.48 8 19.34 21.82 2.24 4.72 -2.48 9 3.09 12.37 -14.01 -4.73 -9.28 10 23.53 22.19 6.43 5.09 1.34 11 23.78 22.06 6.68 4.96 1.72 12 26.90 21.98 9.80 4.88 4.92 13 13.36 12.00 -3.74 -5.10 1.36 14 16.68 22.41 -0.42 5.31 -5.73 15 21.09 22.33 3.99 5.23 -1.24 16 14.45 12.21 -2.65 -4.89 2.24 17 11.48 12.00 -5.62 -5.10 -0.52 18 31.98 22.19 14.88 5.09 9.79 19 20.03 11.78 2.93 -5.32 8.25 20 6.15 11.62 -10.95 -5.48 -5.47 21 26.73 12.13 9.63 -4.97 14.60 22 10.64 21.98 -6.46 4.88 -11.34 23 27.60 22.19 10.50 5.09 5.41 24 4.24 12.00 -12.86 -5.10 -7.76 25 9.49 11.62 -7.61 -5.48 -2.13 26 16.97 22.57 -0.13 5.47 -5.60 27 8.61 11.78 -8.49 -5.32 -3.17 28 7.64 22.69 -9.46 5.59 -15.05 29 6.17 11.50 -10.93 -5.60 -5.33 30 29.96 22.57 12.86 5.47 7.39
TOTAL 512.85
Ÿ (MEAN) = ∑Y u/N = 512.85/30 = 17.10; SST = ∑ (Yu - Ÿ)2 = 2619.93
SSR = ∑ (Ŷu - Ÿ)2 = 782.54; SSE = ∑ (Yu - Ŷu)2 = 1837.42
p - 1 = 4 - 1 = 3 (Numbers of coded factor in the equation);
N - p = 30 - 4 = 26 (Residual); N - 1 = 30 - 1 = 29
N
U = 1 N
U = 1
N
U = 1
186
Source df Sum of Squares (SS) Mean Squares F R2
Regression 3 782.54 (SSR) 260.85 3.69 Residual 26 1837.42 (SSE) 70.67 Total 29 2619.96 ≅ (calculated SST) 0.30
The final model equation in terms of coded and actual factor for the yield of rotenoids resin in dried roots, % (w/w) was presented as follows: (A) Yield of rotenoids resin in dried roots, % (w/w) = 17.10 - 0. 29X1 - 0.078X2 - 5.10X3 (Coded factors) (B) Types of solvent - Ethanol + oxalic acid solution: (Actual factors) Yield of rotenoids resin in dried roots, % (w/w) = 23.10 - 0.12*Solvent-to-solid ratio - 0.06* Raw material particles size (C) Types of solvent - Acetone: (Actual factors) Yield of rotenoids resin in dried roots, % (w/w) = 12.90 - 0.12*Solvent-to-solid ratio - 0.06* Raw material particles size The variance Fisher analysis: (A) From the F distribution table with confidence limit 5 % (0.05): (F3, 26, 0.05) = 2.98 (B) F ratio (3.69) > (F3, 26, 0.05) = 3.69 > 2.98 = The null hypothesis is rejected at the fixed α level (0.05) of significance or confidence limit and it was inferred that the coefficient estimates were not all zero or in other words, the variance differences between the studied processes were significant
The null hypothesis (H0): µ1 = µ2 The alternative hypothesis (H1): µ1 < µ2
µ1 = Yield of rotenoids resin with original processing parameters (Acetone; 10.0 ml/g; Fine particles size)
µ2 = Yield of rotenoids resin with optimized processing parameters (Acetone; 4.72 ml/g; 0.83 mm in diameter) CONCLUSION: Yield of rotenoids resin in dried roots, % (w/w) that has been evaluated using the optimization processing parameters result a significant increases of the yield of rotenoids resin (g) as compared to the processing parameters carried out by Saiful et al. (2003).
The null hypothesis (H0) is rejected The null hypothesis (H0) is accepted
F(v1,v2,α)
0.95 0.05
F2.98
187
Table C-2: Experimental design and results (experimental and predicted values) of
Central Composite Design; CCD with response surface reduced surface 2FI model
(Manual calculation)
Run Factor 1
X1: Solvent- to-solid ratio
Factor 2 X2: Raw material particles size
Factor 3 X3: Types of solvent
Yield of rotenone in dried roots (Experiment)
Yield of rotenone in dried roots (Predicted)
ml/g mm in diameter Treatment % (w/w) % (w/w) 1 6.00 5.00 Acetone 0.40 0.05 2 10.00 2.75 aEthanol 1.38 0.02 3 3.62 4.09 aEthanol 0.17 -0.09 4 3.62 1.41 Acetone 9.82 5.75 5 8.38 4.09 aEthanol 0.81 1.43 6 2.00 2.75 Acetone 0.19 1.31 7 3.62 4.09 Acetone 0.79 0.35 8 8.38 4.09 aEthanol 0.48 1.43 9 3.62 1.41 Acetone 8.67 5.75 10 6.00 2.75 aEthanol 0.45 0.67 11 6.00 5.00 aEthanol 1.35 0.68 12 8.38 1.41 aEthanol 0.49 -0.87 13 6.00 2.75 Acetone 1.08 2.66 14 3.62 4.09 aEthanol 0.27 -0.09 15 6.00 0.50 Ethanol 0.98 0.65 16 3.62 4.09 Acetone 0.52 0.35 17 6.00 2.75 Acetone 5.86 2.66 18 6.00 2.75 aEthanol 0.73 0.67 19 8.38 1.41 Acetone 0.59 2.69 20 8.38 4.09 Acetone 0.54 1.86 21 6.00 0.50 Acetone 2.77 5.28 22 8.38 1.41 aEthanol 0.44 -0.87 23 6.00 2.75 aEthanol 0.94 0.67 24 6.00 2.75 Acetone 4.86 2.66 25 8.38 4.09 Acetone 0.33 1.86 26 3.62 1.41 aEthanol 0.64 2.19 27 8.38 1.41 Acetone 0.36 2.69 28 2.00 2.75 aEthanol 0.19 3.31 29 10.00 2.75 Acetone 3.14 2.01 30 3.62 1.41 aEthanol 0.66 2.19
aEthanol was added with the H2O and oxalic acid - A ratio of ethanol (9): H2O (1) [Prepare 1.0 mg/ml of oxalic acid solution from the volume (ml) of H2O ratio (1)].
188
Run Y u Ŷ u Y u - Ÿ (MEAN) Ŷ u - Ÿ (MEAN) Y u - Ŷ u 1 0.40 0.045 -1.26 -1.62 0.36 2 1.38 0.017 -0.28 -1.64 1.36 3 0.17 -0.09 -1.49 1.75 0.26 4 9.82 5.75 8.16 4.09 4.07 5 0.81 1.43 -0.85 -0.23 -0.62 6 0.19 1.31 -1.47 -0.35 -1.12 7 0.79 0.35 -0.87 -1.31 0.44 8 0.48 1.43 -1.18 -0.23 -0.95 9 8.67 5.75 7.01 4.09 2.92 10 0.45 0.67 -1.21 -0.99 -0.22 11 1.35 0.68 -0.31 -0.98 0.67 12 0.49 -0.87 -1.17 2.53 1.36 13 1.08 2.66 -0.58 1.00 -1.58 14 0.27 -0.09 -1.39 1.75 0.36 15 0.98 0.65 -0.68 -1.01 0.33 16 0.52 0.35 -1.14 -1.31 0.17 17 5.86 2.66 4.20 1.00 3.20 18 0.73 0.67 -0.93 -0.99 0.06 19 0.59 2.69 -1.07 1.03 -2.10 20 0.54 1.86 -1.12 0.20 -1.32 21 2.77 5.28 1.11 3.62 -2.51 22 0.44 -0.87 -1.22 2.53 1.31 23 0.94 0.67 -0.72 -0.99 0.27 24 4.86 2.66 3.20 1.00 2.20 25 0.33 1.86 -1.33 0.20 -1.53 26 0.64 2.19 -1.02 0.53 -1.55 27 0.36 2.69 -1.30 1.03 -2.33 28 0.19 3.31 -1.47 1.65 -3.12 29 3.14 2.01 1.48 0.35 1.13 30 0.66 2.19 -1.00 0.53 -1.53
TOTAL 49.90
Ÿ (MEAN) = ∑Y u/N = 49.90/30 = 1.66; SST = ∑ (Yu - Ÿ) 2 = 175.84
SSR = ∑ (Ŷu - Ÿ) 2 = 87.99; SSE = ∑ (Yu - Ŷu) 2 = 87.83
p - 1 = 6 - 1 = 5 (Numbers of coded factor in the equation);
N - p = 30 - 6 = 24 (Residual); N - 1 = 30 - 1 = 29
N
U = 1 N
U = 1
N
U = 1
189
Source df Sum of Squares (SS) Mean Squares F R2
Regression 5 87.99 (SSR) 17.60 4.81 Residual 24 87.83 (SSE) 3.66 Total 29 175.82 ≅ (calculated SST) 0.50
The final model equation in terms of coded and actual factor for the yield of rotenone in dried roots, % (w/w) was presented as follows: (A) Yield of rotenone in dried roots, % (w/w) = 1.66 - 0.39X1 - 0.77X2 + X3 + 1.15X1X2 - 0.78X2X3 (Coded factors) (B) Types of solvent - Ethanol + oxalic acid solution: (Actual factors) Yield of rotenone in dried roots, % (w/w) = 7.56 - 1.15*Solvent-to-solid ratio - 2.15*Raw material particles size + 0.36*Solvent-to-solid ratio*Raw material particles size (C) Types of solvent - Acetone: (Actual factors) Yield of rotenone in dried roots, % (w/w) = 12.77 - 1.15*Solvent-to-solid ratio - 3.32*Raw material particles size + 0.36*Solvent-to-solid ratio*Raw material particles size The variance Fisher analysis: (A) From the F distribution table with confidence limit 5 % (0.05): (F5, 24, 0.05) = 2.62 (B) F ratio (4.81) > (F5, 24, 0.05) = 4.81 > 2.42 = The null hypothesis is rejected at the fixed α level (0.05) of significance or confidence limit and it was inferred that the coefficient estimates were not all zero or in other words, the variance differences between the studied processes were significant
The null hypothesis (H0): µ1 = µ2 The alternative hypothesis (H1): µ1 < µ2
µ1 = Yield of rotenone with original processing parameters (Acetone; 10.0 ml/g; Fine particles size)
µ2 = Yield of rotenone with optimized processing parameters (Acetone; 4.72 ml/g; 0.83 mm in diameter) CONCLUSION: Yield of rotenone in dried roots, % (w/w) that has been evaluated using the optimization processing parameters result a significant increases of the rotenone content (mg) as compared to the processing parameters carried out by Saiful et al. (2003).
The null hypothesis (H0) is rejected The null hypothesis (H0) is accepted
F(v1,v2,α)
0.95 0.05
F2.62
192
Appendix E
Table E-1: The biological activity (LC50) of S1 (24 hours of treatment) using probit
analysis (manual calculation) (Finney, 1971)
C (Conc.) Log C a% Mortality b% Mortality (Corrected) cProbit response
1 0.00 30 30 4.56 10 1.00 50 50 5.13 50 1.70 100 100 - 100 2.00 100 100 - 500 2.70 100 100 - 1000 3.00 100 100 -
dCONTROL - 0 - - aExpress the number of dead as percentage of the total number of organism tested at that concentration. bFor the concentrations in which the observed % mortality is greater than the mortality observed among the organisms in the control tank, correct the % mortality using Abbott’s formula: [Corrected mortality (%) = (Mobs - Mcontrol/100 - Mcontrol) × 100 %] where Mobs is the observed mortality calculated into percentage and Mcontrol is the control organisms at that test exposure duration. Do not correct mortality values which are less than the control mortality. cProbit responses were obtained from Table E-2. dDeionized water was used as control medium. Table E-2: Transformation of percentages to probits (Finney, 1971)
EXAMPLE OF DIRECT INTERPOLATION OF INTERMEDIATE VALUE: (If control value is 0) % corrected mortality: 20 % mortality: 2 - Probit value: 4.23 % corrected mortality: 22.41 Probit value x: ? % corrected mortality: 30 % mortality: 3 - Probit value: 4.56 22.41 - 20/30 - 22.41 = x - 4.23/4.56 - x ∴x = 4.31
% 0 1 2 3 4 5 6 7 8 9 0 - 2.67 2.95 3.12 3.25 3.36 3.45 3.52 3.59 3.66 10 3.72 3.77 3.82 3.87 3.92 3.96 4.01 4.05 4.08 4.12 20 4.16 4.19 4.23 4.26 4.29 4.33 4.36 4.39 4.42 4.45 30 4.48 4.50 4.53 4.56 4.59 4.61 4.64 4.67 4.69 4.72 40 4.75 4.77 4.80 4.82 4.85 4.87 4.90 4.92 4.95 4.97 50 5.00 5.03 5.05 5.08 5.10 5.13 5.15 5.18 5.20 5.23 60 5.25 5.28 5.31 5.33 5.36 5.39 5.41 5.44 5.47 5.50 70 5.52 5.55 5.58 5.61 5.64 5.67 5.71 5.74 5.77 5.81 80 5.84 5.88 5.92 5.95 5.99 6.04 6.08 6.13 6.18 6.23 90 6.28 6.34 6.41 6.48 6.55 6.64 6.75 6.88 7.05 7.33 - 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 99 7.33 7.37 7.41 7.46 7.51 7.58 7.65 7.75 7.88 8.09
% corrected mortality % mortality
193
y = 0.57x + 4.56
4.5
4.6
4.7
4.8
4.9
5
5.1
5.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Log10 concentration (ppm)
Mor
talit
y (p
robi
t)
S1 S1
Figure E-1 The biological activity (LC50) of rotenoids resin: An experimental of (S1)
linear regression of logarithm-converted concentrations (x) versus the probit values (y)
From the fitted regression equation (the correlation coefficient should exceed ≅ 0.5): y = 0.57x + 4.56; Determine the predicted value of x associated with y = 5.00 (the probit value of 50 %) 5.00 = 0.57x + 4.56 5.00 - 4.56/0.57 = x 0.44/0.57 = x ∴ x = 0.77 Calculate the inverse log10 (anti-log) to find the LC50: Antilog10 (0.77) = 5.91 ppm So, LC50 (S1) = 5.91 ppm.
198
Appendix H
Figure H-1 Chromatogram of rotenone standard [SIGMA-Aldrich™; 95 - 98 % (w/w)]
Rotenone standard [SIGMA-Aldrich™; purity of 95 - 98 % (w/w)]: Rotenone standard concentration (Cstd) [13.91 mins] = 0.256 mg.ml-1 Peak area (A) = 43,055.34 [mV.s] Sensitivity factor (SF) = (A)/(Cstd) = 43,055.34/0.256 = 168,184.92 [mV.s.ml.mg-1]
199
Figure H-2 Chromatogram of Sample 20 (S20) of Liquid Crude Extract (LCE)
Concentration of rotenone in the Liquid Crude Extract (LCE): [Dilution of 1/50] Rotenone peak area (Arotenone) [13.95 mins] = 2640.73 [mV.s] Concentration of rotenone (Crotenone) = (Arotenone)/SF = 2640.73/168,184.92 = 0.016 mg.ml-1 (Dilution factor; DF = Flask volume/pipette volume): DF = 50/1 ∴Actual concentration of rotenone = 0.016 mg.ml-1 × 50/1 = 0.80 mg.ml-1 Yield of rotenone in dried roots = 0.80 mg.ml-1 × 206 ml (volume of LCE) = 164.80 mg Yield of rotenone in dried roots, % (w/w) = 164.80 mg/30.06 g × 100 % = 0.55 % (w/w)
200
Figure H-3 Chromatogram of Sample 20 (S20) of Concentrated Liquid Crude Extract
(CLCE)
Concentration of rotenone in the Concentrated Liquid Crude Extract (CLCE): [Dilution of 1/50] Rotenone peak area (Arotenone) [14.04 mins] = 32,913.28 [mV.s] Concentration of rotenone (Crotenone) = (Arotenone)/SF = 32,913.28/168,184.92 = 0.195 mg.ml-1 ∴Actual concentration of rotenone = 0.195 mg.ml-1 × 50/1 = 9.78 mg.ml-1 Yield of rotenone in dried roots = 9.78 mg.ml-1 × 21 ml (volume of CLCE) = 205.38 mg Yield of rotenone in dried roots, % (w/w) = 205.38 mg/30.06 g × 100 % = 0.68 % (w/w)
Proceeding of ‘Simposium Kimia Analisis Kebangsaan’ (SKAM-17), 26-28 August 2004, Swiss Garden Resort, Pahang
201
Appendix I
PURIFICATION AND IDENTIFICATION OF ROTENONE FROM Derris elliptica USING THE VACUUM LIQUID CHROMATOGRAPHY-THIN LAYER CHROMATOGRAPHY
(VLC-TLC) METHOD
1S. I., Zubairi*, 1M. R., Sarmidi, 1R. A., Aziz, 2M. K. A., Ramli, 1R., Latip and 1N. I. A., Nordin
1Chemical Engineering Pilot Plant (CEPP), Faculty of Chemical and Natural Resources Engineering,
Universiti Teknologi Malaysia (UTM), 81310 UTM Skudai, Johor Darul Takzim.
2Heritage Biotech (M) Sdn. Bhd, c/o Chemical Engineering Pilot Plant, Faculty of Chemical and Natural Resources Engineering,
Universiti Teknologi Malaysia (UTM), 81310 UTM Skudai, Johor Darul Takzim.
Tel.: +607-5532595 Fax: +607-5569706
e-mail: [email protected], [email protected]
ABSTRACT
History has recorded the use of rotenone as bio-pesticide by plant and vegetables growers worldwide. The re-introduction of this bio-active compound as an environmental-friendly bio-pesticide in organic fruits and vegetables plantations is now an emergent trend. Our laboratory is focused on the development of bio-pesticide using rotenone extracted from local plant species which is Derris elliptica. A liquid chromatographic method with high vacuum pressure was developed for the analysis of rotenone in the Liquid Crude Extract (LCE) of Derris elliptica. The treated root and stem were cut into small pieces prior to the extraction process of the Normal Soaking Extraction (NSE) method. The Liquid Crude Extract (LCE) was concentrated further using the rotary evaporator at 40 0C under reduced pressure of 800 × 10-3 bar. A high-pressure vacuum liquid chromatographic was then carried out for the purification of rotenone using the silica gel (230 mesh to 400 mesh sieve) with variety of eluents polarity (e.g.: hexane, chloroform, acetone and ultra pure water). All eluents that have been collected after the purification process were subjected to the semi-automatic TLC (CAMAG Linomat 5) for the identification of rotenone and other constituents using the isocratic elution of petroleum ether and ethyl acetate with a ratio of 4:2 (v/v). The markers for each bio-active constituent were visualized and recorded using the Ultra Violet (UV) lamp (wavelength (λ) of 254 nm and 365 nm). The Rf value for the desirable constituent (rotenone) was also calculated. The employed method shows significant improvement on reducing the impurities constituent in the LCE as well as the visibility of rotenone identification as compared to the filtration of PTFE filter cartridge (0.45µm) prior to the conventional TLC method. Keywords: Derris elliptica; rotenone; environmental-friendly bio-pesticide; TLC; VLC.
INTRODUCTION Malaysia is embarking on developing its agro-business and to provide food security for the country. The pest control in the country at present relies mainly on toxic chemical pesticides. The conventional pesticides are often toxic to mammals, non-target pests and persist in the environment as recalcitrant. Therefore, the search for bio-active substance which have satisfactory properties that is effective on the target pest is economic viable and biodegradable is great importance. For example, rotenone from a plant belonging to the Derris elliptica family known locally as ‘Tuba Kapur’ has been proven as potent as many conventional synthetic pesticides. Rotenone can be extracted from many tropical Leguminosae such as Derris spp., Lonchocarpus spp. and Tephrosia spp. Derris elliptica is a widely available local plant and contains 4.0 % (w/w) to 5.0 % (w/w) of the active ingredient, rotenone. Rotenone is extremely active as contact and stomach poisons against many crop pests such as Mexican bean beetle, apple and pea aphids,
Proceeding of ‘Simposium Kimia Analisis Kebangsaan’ (SKAM-17), 26-28 August 2004, Swiss Garden Resort, Pahang
202
corn borer and household pests. Besides having low mammalian toxicity, they are reasonably safe to honeybees (Opender, 2001).
METHODOLOGY Plant collection - Derris elliptica roots was collected at Kota Johor Lama in the state of Johor, Malaysia. Raw material - The collected raw materials immediately undergo cleaning process to remove dirt and soil. They were kept and dried into oven for overnight at room temperature (26 ± 2 0C). The cleaned raw materials were sorted to collect the root and stem. Only root and stem were utilized. The root and stem were cut into small pieces prior to the grinding process. Extraction apparatus and procedure - The extraction was carried out by soaking 50.0 g of dried root and stem in 500 ml solution of methanol 95.0 % (v/v) for 24 hours at room temperature of 26 ± 2 0C. The Liquid Crude Extract (LCE) was filtered through 15.0 cm Whatman no. 4 filter paper directly into 500 ml of the PYREX® glassware after 24 hours of the extraction process. The Liquid Crude Extract (LCE) was concentrated further using the rotary evaporator; Laborata 4001 (Heidolph) at 40 0C, purified using the Vacuum Liquid Chromatography (VLC) under reduced pressure of 800 × 10-3 bar and analyzed qualitatively using semi-automatic TLC (CAMAG Linomat 5). Purification of the Concentrated Liquid Crude Extract (CLCE) - The VLC system consist of silica gel powder (230 mesh to 400 mesh sieve) with silica gel-to-CLCE ratio of 40.0 g/g, composition of different eluents as shown in Table 1.1 and vacuum pressure pump (P = 800 × 10-3 bar). Four units of VLC system were prepared for each eluent. Silica gel solution (silica gel powder + H2O) was poured into the unit until it dried as the vacuum pump drain off the air to facilitate the flow of H2O in the reservoir. Eventually, the dried silica gel was transformed into a solid structure with the intention that a sufficient absorption and separation of the desirable bio-active constituents can be occurred during the purification process. The Concentrated Liquid Crude Extract (CLCE) was poured onto the surface of the silica gel until 1/3 of the solution absorbed into the dried silica gel. The 15.0 cm of Whatman filter paper no. 4 was located onto the surface of dried silica gel to stabilize the dried silica gel while pouring the eluents. The amount of 300 ml of eluents were poured onto the dried silica gel until it penetrated and absorbed all over the silica gel. During the flow of eluents occurred thoroughly through dried silica gel, the constituents in the CLCE already been absorbed in the silica gel where the desired constituents were eluted from the CLCE into the eluents accordingly to its similarity of the eluents polarity. The eluted or purified solution was subjected to convective concentration process in the fume cupboard and analyzed qualitatively using semi-automatic TLC (CAMAG Linomat 5).
Table 1.1: Composition (%) of different eluents.
No. Eluents/solvent mixtures Composition (%) 1. aHexane + chloroform 1) 100-0 2) 90-10 3) 80-20 4) 70-30 5) 60-40 6) 50-50 7) 40-60
8) 30-70 9) 20-80 10) 10-90 11) 0-100 2. aChloroform + acetone 12) 100-0 13) 90-10 14) 80-20 15) 70-30 16) 60-40 17) 50-50
18) 40-60 19) 30-70 20) 20-80 21) 10-90 22) 0-100 3. aAcetone + ultra pure water 23) 100-0 24) 90-10 25) 80-20 26) 70-30 27) 60-40 28) 50-50
29) 40-60 30) 30-70 31) 20-80 32) 10-90 33) 0-100 4. bUltra pure water 34) 100
aPurity of the solvents were 95.0 % (v/v). bE-pure, Barnstead: 0.49MΩ-cm.
Proceeding of ‘Simposium Kimia Analisis Kebangsaan’ (SKAM-17), 26-28 August 2004, Swiss Garden Resort, Pahang
203
Analysis of purified solution - The purified solutions were subjected to the qualitative analysis using semi-automatic TLC (CAMAG Linomat 5). The operating conditions for CAMAG Linomat 5 were listed as follows: Plate size (TLC aluminium sheets; Silica gel; 60 F254; 10.0 cm × 10.0 cm), band of spot: 4.0 mm, x and y-axis: 11.0 mm and 10.0 mm respectively, track distance: 10.0 mm and spray volume dosage: 4.0 µl. The plates that have been sprayed were subjected into a development chamber using isocratic elution with a mixture of petroleum eter and ethyl acetate 4:2 (v/v). The markers for each bio-active constituent found were visualized and recorded using UV lamp (wavelength of 254 nm and 365 nm) and the retardation factor, Rf value of desirable constituent was calculated as shown in Table 1.2.
RESULT AND DISCUSSION
Figure 1.2: The presence of rotenone under UV light of wavelength 254 nm and 365 nm respectively.
CLCE [Methanol extract; 95.0 % (v/v)] Rotenone standard; 96.2 % (w/w); Rf ≅ 0.63 Rotenone; no. 8; Rf ≅ 0.61
Rotenone; 96.2 % (w/w) + other constituents; 3.8 % (w/w)
CLCE; UV light; λ: 254 nm
ds
dc
8
7
6
STD
9
10
Rotenone
CLCE; UV light; λ: 365 nm
ds
dc
8
7
6
STD
9
10
Table 1.2: Rf for eluent no. (8): Hexane + chloroform [30-70]
Concentrated Liquid Crude Extract (CLCE)
Vacuum pressure (800 mbar)
Eluents/solvent mixture
Silica gel (230 mesh to 400 mesh)
The eluted/purified solution (eluent + bio-active constituents)
Figure 1.1: Vacuum Liquid Chromatography (VLC) apparatus
Rotenone
Rotenone; 96.2 % (w/w) + other constituents; 3.8 % (w/w)
(dc) (ds)
Proceeding of ‘Simposium Kimia Analisis Kebangsaan’ (SKAM-17), 26-28 August 2004, Swiss Garden Resort, Pahang
204
Analysis of the purified rotenone solution - The markers of each eluent composition for hexane + chloroform from no. 6 to 10 using the semi-automatic TLC (CAMAG Linomat 5) are shown in Figure 1.2 and Rf for the rotenone standard (Rotenone PESTANAL®; analytical grade, 96.2 % (w/w); SIGMA-Aldrich™) and the desirable constituent (rotenone) are listed in Table 1.2. The retardation factor, Rf of rotenone was fairly similar as rotenone standard with approximately 0.61 and 0.63 respectively. Although rotenone standard with purity of 96.2 % (w/w) can be seen easily on the silica plats, actually there were small markers encircling the rotenone marker which is represented as other constituents (e.g.: tephrosin, 12αβ-rotenolone and deguelin). The other encircle markers were insignificant for the Rf determination due to small amount of constituents comprise in the standard (≅ 3.8 % (w/w)). As referred to Figure 1.2, the desirable constituents and rotenone standard (STD) markers were highlighted to show that the purification process using hexane + chloroform [30-70] as an eluent were the best composition which produce small amount of impurities marker as compared to the other eluents composition. The desirable constituents were eluted and absorbed along with the composition of hexane + chloroform [30-70] eluents and gradually disappeared after increasing the concentration of chloroform and decreasing the concentration of hexane by 90 % and 10 % respectively. Rotenone and other constituents were started to elute from composition no. 6 to no. 8 and gradually disappeared (exhaustively flush out) or still being absorbed by the remaining constituents in silica gel along with the changes of the composition. Using a mixture of two different eluents polarity [hexane (0) and chloroform (4.1)], rotenone appeared to be existed in the composition between these values and there was a small amount of unknown markers as compared to the non-conventional TLC method. Non-conventional TLC method showed 13 to 15 markers as compared to the VLC-TLC method which show only 8 markers. The other eluents composition as shown in Table 1.1 demonstrated hard visibility for observing the desired markers after the semi-automatic TLC (CAMAG Linomat 5) development process. As for the ultra pure water, this eluent produced a milky solution in which dissolved and flush out almost all CLCE solution absorbed in the silica gel. In fact, the TLC plat revealed that there was no separation occurred (no markers) either during the VLC or TLC development process. This was due to all constituents in the eluent (ultra pure water) were possibly high in polarity and the separation on the non-polar material (silica gel) either in the VLC or TLC system would be impossible to solubilize into the eluent (absorption to silica gel > solubility in the eluent) and separate accordingly into their own polarity. Furthermore, pure rotenone has the solubility in gram per 100 cubic centimeters of solution at 20 0C as follows: water ≅ 0.00002; ethyl alcohol = 0.2; carbon tetrachloride = 0.6; amyl acetate = 1.6; xylene = 3.4; acetone = 6.6; benzene = 8.0; chlorobenzene = 13.5; ethylene dichloride = 33.0 and chloroform = 47.0. The solubility of pure rotenone in chloroform (47 grams per 100 cubic centimeters of solution at 20 0C) indicated that the elution of rotenone has been suspected in a mixture of hexane + chloroform or chloroform + acetone. The most suitable mixture composition (%) itself was to be unknown until this study has been done extensively. After rigorous study done, it can be concluded that the higher polarity eluents used as an eluent, the less constituents can be eluted or absorbed into the eluents and produced less impurities in the purified solution. Moreover, the techniques to obtain and achieve the bio-active constituents of Derris species as an USP standard (U.S. Pharmacopeia) should be further investigated.
Sequel of eluents based on the best separation and identification of rotenone constituent Hexane (0) + Chloroform (4.1) > Chloroform + Acetone (5.1) > Acetone (5.1) + H20 (9.0) > H20 (9.0)
ACKNOWLEDGEMENT This research was supported by an IRPA grant 09-02-06-0083 EA261 under the Ministry of Science, Technology and Environment (MOSTE), Malaysia.
REFERENCES
Opender, K. and Dhaliwal, G. S. (2001). Prospects and problems of phytochemical bio-pesticide.
Phytochemical Bio-pesticides, pp. 133-134.
Proceeding of ‘Simposium Kimia Analisis Kebangsaan’ (SKAM-17), 26-28 August 2004, Swiss Garden Resort, Pahang
205
Schmeltz, I. (1971). Nicotine and other tobacco alkaloids. In Jacobsan, M. and Crosby, D. G. (eds.), Naturally Occurring Insecticides, Marcel Dekker, New York, pp. 99-136.
Waterman, D. G. and Khalid, A. S. (1980). The major flavonoids of the seed of Tephrosia. Phytochemistry. 19, 909-915.
206
Appendix J
Figure J-1 Molecular structure of deguelin
Figure J-2 Molecular structure of 12αβ-rotenolone
Figure J-3 Molecular structure of tephrosin (an oxidation product of deguelin)