development of a beverage based on sweet potato peels ...€¦ · •sweet potato peels and leaves...

Development of a beverage based on sweet potato peels:

formula optimization

Ana Anastácio

DATA SCIENTISTS, ACADEMICS & EDUCATION

ANTIOXIDANT AND PHENOLIC COMPOUNDS IN

SWEET POTATO PEELS AND LEAVES:

FOOD APPLICATIONS AND HEALTH BENEFITS

Ana Isabel Mimoso Tomás Coelho Anastácio

Doutoramento em Ciências Biotecnológicas/Biotecnologia Alimentar

Orientadora: Professora Doutora Isabel Maria Marques Saraiva de Carvalho

2

Production 2014 104 Mtonnes

3http://www.talkafrica.co.ke/sweet-potato-scientists-awarded-the-prestigious-world-food-prize/

0,022

1,3

70,7

Portugal

USA

China

Africa; 20,2%

Americas; 3,7%

Asia; 75,3%

Europe; 0,1% Oceania; 0,8%

(Mtonnes)

FAOSTAT

http://blogs.usda.gov/2016/07/14/crop-insurance-continues-to-strengthen-rural-communities/

Sweet potato wastes3 ton/ha

17

29

Manual Abrasion

Peeling (% loss)

4

Objectives

Food prototype

Waste

1

2

PrototypeANTIOX

New food developmentprocess

5

New food development process

Design of experiments

Review Tests Key factors Optimisation

Phenolics

Antioxidant activity in vitro

6

77

New food development process

Tests

8

Process

Extraction conditions

Product

Beverage formulation

Optimisation

Ana Anastácio & Isabel Saraiva de Carvalho (2016): Development of a beveragebenchtop prototype based on sweet potato peels: optimization of antioxidant activityby a mixture design, International Journal of Food Sciences and Nutrition, DOI:10.1080/09637486.2016.1174984.

Product optimisation

Objective

• Optimisation of antioxidant beverageformula

9

9

10

• Sweet potato peels and leaves can be related to health benefits such as oxidative stress reduction.

• Optimized process for antioxidants extraction.

• No patent application on antioxidant beverage with sweet potato peels and leaves.

Previous knowledge

Components

SPPE 50 – 100 %

SPLE 0 – 50 %

HonS 0 – 50 %

• TPC, FRAP, DPPH assays and SS

Extraction

RSM

ANN

Product optimisation

11

• Mixture design

Mixture designProduct optimisation

SPPE

SPLE

Ho

nS

I-optimal randomized

mixture design

24 runs

12

13

Product optimisation Mixture profiler

14

Product optimisation Profiler

RSM OptimalProduct optimisation

15

SPPE

SPLE

Ho

nS

50.0 %

22.1 %

27.9 %

ANNNeuron

Product optimisation

16

inputs output

ANNProduct optimisation

17

Activation: H1 = w1 x1 + w2 x2

Transformation: Y = ao + a1 Tanh(0.5*H1) + a2 Tanh(0.5*H2)

inputs output

x1

x2

YH1

Artifical neuron

w1

w2

Product optimisation

18

Hi =

+ wi1 X1

+ wi2 X2

+ wi3 X3

Y =

+ ao

+ a1*Tanh(0.5*H1)

+ a2*Tanh(0.5*H2)

+ a3*Tanh(0.5*H3)

ANN

SPPE

SPLE

HonS

Product optimisation

19

SPPE

SPLE

Ho

nS

50.0 %

21.5 %

28.5 %

ANN Optimal

Comparison

FJ: fruit juice

VJ: vegetable juice

HI: herbal infusion

T: tea

PPE: plum peel enriched

nectar

CP: cardio protective beverage

NJ: nutraceutical juice

SPPE Bev: sweet potato peel

beverage

Commercial Laboratory

Product optimisation

20

Formula optimisation

JMP

21

• Design

• Data visualisation

• Modelling

• Optimisation

22

21

(in press)

Thank you.

Ana Anastácio

ana.c.anastacio@gmail.com

23

DATA SCIENTISTS, ACADEMICS & EDUCATION