Effects of drying methods on sweetpotato starch properties and substitution for mung bean noodles

Overall, this research entitled “effects of drying methods on sweet potato starch properties and its substitution for mung bean starch in starch noodles” aims to evaluate the influences

EFFECTS OF DRYING METHODS ON SWEETPOTATO STARCH

PROPERTIES AND ITS SUBSTITUTION FOR

MUNG BEAN NOODLES

by

Ho Minh Thao

A thesis submitted in partial fulfillment of the requirements for the

degree of Master of Engineering in Food Engineering and Bioprocess Technology

Examination Committee: Prof Athapol Noormhorm (Chairperson)

Prof Sudip K Rakshit

Dr Anil Kumar Anal

Nationality: Vietnamese

Previous Degree: Bachelor of Food Technology

An Giang University

An Giang, Vietnam Scholarship Donor: UNEP RRCAP Organization

Asian Institute of Technology School of Environment, Resources and Development

Thailand December 2010

ACKNOWLEDGMENTS

Firstly, I wish to express my profound appreciation to Prof Athapol Noomhorm for his valuable advices, suggestions, guidance and encouragements during this research as well as the course of this study I also want to express my sincere gratitude to Prof Sudip K Rakshit and Dr Anil Kumar Anal for their helpful comments and discussions on my thesis

as well as useful knowledge that they transferred during this course

Specials thanks are extended to all secretaries and laboratory technicians in Food Engineering and Bioprocess Technology field for their helps in all document preparation and technical assistances during my research Next, thanks to all my classmates and friends

at Asian Institute of Technology (AIT) who shared useful information, experiences, and knowledge as well as served as evaluating panelists for my products

I would like to thanks to UNEP (United Nations Environment Program) Organization for fully financial supporting my Master study at AIT Without this supporting I would have never dreamt of this academic achievement

Finally, I am very indebted to my parents and relatives for their love and understanding, especially my little daughter and dearest wife for their love, help, encouragement and metal supporting during this study course

ABSTRACTS

Mung bean starch (MBS) is traditionally considered as ideal material for noodle production due to its unique properties However, to meet continuously increasing demand of starch noodles in recent years, finding other starch materials that are abundant output and cheap price to replace partially or totally for MBS which is limited yield and expensive price will

be valuable Therefore, the overall objective of this study was to investigate partial substitutiton of sweet potato starch (SPS) for MBS in noodle production Physiochemical properties of commercial MBS and starches isolated from four popular types of SP varieties, namely white skin and yellow-red flesh color (SP1_W_YR), purple skin and purple flesh color (SP2_P_P), purple skin and yellow flesh color (SP3_P_Y), and orange-purple skin and orange flesh color (SP4_OP_O) variety were determined to select suitable variety for noodle making The effects of various drying conditions namely tray, infrared and fluidized bed drying at 45, 55 and 65oC on starch properties, and mathematical modeling of drying process as well as determination of effective diffusivity (Deff) and energy of activation (Ea) under these drying methods were also included in this research Comparing to SPSs, MBS was superior for noodle production due to high in amylose content (40.69%); low in protein, lipid and ash content; high gel hardness, less gel stickiness; and good hot paste stability and high cold paste viscosities The SP4_OP_O starch has the highest amylose content (34.86%), following by SP1_W_YR starch (31.50%) in comparison to remaining SPSs (28.50%) High amylose starches resulted in high hardness and less stickiness of gels, and high cold paste viscosities The highest starch yield (17%) and purity were found in SP1_W_YR starch, significantly higher than starch yield of the others (12-15%) Due to containing high pigments, the colors of SP2_P_P and SP4_OP_O starch was inferior to that of SP1_W_YR and SP3_P_Y starches In general, SP1_W_YR starch was the most suitable for noodle production

Those drying conditions just slightly affected color, gel texture, swelling power, solubility and pasting properties of starches For tray drying, it took about 15, 8.5 and 5.5 hr to get about 10% final MC at 45, 55 and 65oC respectively while to obtain the same MC at the same drying temperature, the drying time for infrared drying was 12, 6.5 and 4.5 hr, and for fluidized bed drying was 0.2, 0.28 and 0.42 hr respectively The high R2 (> 0.93), and low in RMSE (0.002739 to 0.085240) and χ2 (0.000003 to 0.007160) were found for all eleven models in which Midilli model was found to be the best for explaining the starch

drying behavior for all drying conditions A generalized Midilli model also was developed

for each drying method The Deff for fluidized bed drying was 4.92.10-7 - 7.26.10-7 (m2/s), significant higher than those in tray and infrared drying ranging 2.049.10-9 - 5.674.10-9 (m2/s) The Ea in tray and infrared drying were 35.88 and 33.21 (KJ/mol) respectively, and nearly double that in fluidized bed drying (17.33 KJ/mol)

The quality of noodle made from 20% SPS and 80% MBS was not significant difference to that of pure MBS noodles The increasing of solid content of starch slurry resulted in considerable increasing in cooking time, cooking loss, rehydration and tensile of noodles while aging time only markedly affected to cooking loss and tensile Overall, the starch noodle derived from 20% SPS and 80% MBS, 35% of initial solid content and aging at 4oC for 10-20 hr can be comparable to PMBSN as considering in both noodle quality and cost

Keyword: sweet potato starch, mung bean starch, drying method, moisture ratio, effective

diffusivity, energy of activation and starch noodles

TABLE OF CONTENTS

Chapter Title Page

TITLE PAGE……… ………….………i

ACKNOWLEDGMENTS ii

ABSTRACTS iii

TABLE OF CONTENTS iv

LIST OF TABLES v

LIST OF FIGURES v

Chapter 1 INTRODUCTION 1

1.1 General information 1

1.2 Objectives 3

Chapter 2 LITERATURE REVIEW …4

2.1 Sweet potato roots and its chemical compositions 4

2.2 The extraction of sweet potato starches 4

2.3 The physiochemical properties of sweet potato and mung bean starches 5

2.4 Drying technology 12

2.5 Starch noodles 16

Chapter 3 MATERIALS AND METHODS 20

3.1 Materials and equipments 20

3.2 Methods 20

Chapter 4 RESULTS AND DISCUSSIONS 32

4.1 Selection of sweet potato cultivar for starch noodle production 32

4.2 Effect of drying conditions on the sweet potato starch properties 44

4.3 Modeling drying process and determination of effective diffusivity and energy of activation for sweet potato starch under various drying conditions 49

4.4 The substitution (SPS) for (MBS) in noodle production and effects of processing conditions on starch noodle quality 68

Chapter 5 CONCLUSIONS AND RECOMMENDATIONS 78

5.1 Conclusions……….……… 78

5.2 Recommendations……….……… 80

REFERENCES ………….81

APPENDICES 90

LIST OF TABLES

Table No Title Page Table 2.1 Chemical compositions of Chinese sweet potato (SP) cultivars (Chen, 2003) 4

Table 2.2 Thin layer drying models 15

Table 3.1 The models used for modeling drying process (Menges & Ertekin, 2006) 25

Table 4.1 Chemical compositions of sweet potato and mung bean starches (*) 33

Table 4.2 Mean size of sweet potato and mung bean starch granules 36

Table 4.3 Gel texture of sweet potato and mung bean starches 38

Table 4.4 Pasting properties of four types of sweet potato and mung bean starches 40

Table 4.5 Correlation among property parameters of sweet potato starches 43

Table 4.6 The pasting properties of sweet potato starches at various drying condition 48

Table 4.7 The gel texture, swelling power and solubility of sweet potato starches at various drying conditions 49

Table 4.8 Equilibrium moisture content at different drying conditions 50

Table 4.9 Statistical results of modeling criteria (R-square, RNSE and χ2) of tray drying at different temperatures 54

Table 4.10 Statistical results of modeling criteria (R-square, RNSE and χ2) of infrared drying at different temperatures 55

Table 4.11 Statistical results of modeling criteria (R-square, RNSE and χ2) of fluidized bed drying at different temperature 56

Table 4.12 Statistical results of constants and coefficients for tray drying at different temperature 57

Table 4.13 Statistical results of constants and coefficients for infrared drying at different temperature 58

Table 4.14 Statistical results of constants and coefficients for fluidized bed drying at different temperature 59

Table 4.15 The effective diffusivity (m2/s)in tray, infrared and fluidized drying 67

Table 4.16 The Arrhenius constant and energy of activation in different drying methods 68

Table 4.17 The color of dry and cooked noodles at different ratios of SPS and MBS 69

Table 4.18 The cooking quality of noodles made from different ratios of SPS and MBS 70

Table 4.19 Tensile characteristics of cooked noodles at different ratios of SPS and MBS 71

Table 4.20 Average scores of sensory evaluation for noodles made from various SPS ratios 73

Table 4.21 The color of dry and cooked noodles at different solid content 74

Table 4.22 The cooking quality of noodles at different solid content 74

Table 4.23 Tensile characteristics of cooked noodles at various solid content 75

Table 4.24 The color of dry and cooked noodles at different aging time 76

Table 4.25 The cooking quality of noodles at different aging time 76

Table 4.26 Tensile characteristics of cooked noodles at different aging time 77

LIST OF FIGURES

Figure No Title Page

Figure 2.1 Procedure for sweet potato starch extraction by wet-milling method

(Duc, 1994) 5

Figure 2.2 Morphological property of sweet potato (a) and mung bean (b) starches 6

Figure 2.3 Starch particle size distribution from sweet potato varieties (a, b and c) as compared with potato (e) and mung bean starches (d) (Chen, 2003) 7

Figure 2.4 Amylogram of potato starch (4%, w/v) (Chen, 2003) 9

Figure 2.5 The function of amylose in starch gel system (Ott & Hester, 1965) 10

Figure 2.6 Transformation process of starch structure (Tomasik, 2004) 12

Figure 2.7 Typical drying rate curve of for food solids(Singh & Heldman, 1993) 14

Figure 3.1 The framework of study 21

Figure 3.2 Four sweet potato varieties 22

Figure 3.3 Tray dryer equipment 23

Figure 3.4 Infrared drying equipment 24

Figure 3.5 Fluidized bed drying equipment 24

Figure 3.6 Description of extension measurements 31

Figure 4.1 Apparent amylose content (%) in sweet potato and mung bean starches 32

Figure 4.2 The starch yield (%) in all sweet potato roots 34

Figure 4.3 The color values of sweet potatoes and mung bean starches(*) 35

Figure 4.4 Light microscopy (40X) of starch granules of sweet potatoes and mung bean 37

Figure 4.5 RVA pasting profiles of four types of sweet potato and mung bean starches 39

Figure 4.6 Lightness (L value) and whiteness of sweet potato starches at different drying conditions 45

Figure 4.7 Greenness (a value) and yellowness (b value) of sweet potato starches at different drying conditions 46

Figure 4.8 The drying rate at different temperature in tray and infrared drying 51

Figure 4.9 The drying rate at different temperature in fluidized bed drying 51

Figure 4.10 MC (%, db) changes with drying time at different drying temperature in tray and infared drying 52

Figure 4.11 MC (%, db) changes with drying time at different drying temperature in fluidized bed drying 52

Figure 4.12 The changes of moisture ratio with drying time in tray and infrared drying 53

Figure 4.13 The changes of moisture ratio with drying time in fluidized bed drying 53

Figure 4.14 Relationship between drying constant k of Midilli model and temperature in tray and infrared drying 60

Figure 4.15 Relationship between drying constant k of Midilli model and temperature in fluidized bed drying 60

Figure 4.16 Relationship between drying coefficient a of Midilli model and temperature in tray and infrared drying 60

Figure 4.17 Relationship between drying coefficient a of Midilli model and

temperature in fluidized bed drying 61

Figure 4.18 Relationship between drying coefficient b of Midilli model and temperature in tray and infrared drying 61

Figure 4.19 Relationship between drying coefficient b of Midilli model and temperature in fluidized bed drying 61

Figure 4.20 Relationship between drying coefficient n of Midilli model and temperature in tray and infrared drying 62

Figure 4.21 Relationship between drying coefficient n of Midilli model and temperature in fluidized bed drying 62

Figure 4.22 Comparison of experimental and predicted moisture ratio by the generalized Midilli model (4.1) for tray drying 64

Figure 4.23 Comparison of experimental and predicted moisture ratio by the generalized Midilli model (4.2) for infrared drying 64

Figure 4.24 Comparison of experimental and predicted moisture ratio by the generalized Midilli model (4.3) for infrared drying 65

Figure 4.25 Relationship between logarithmic moisture ratio and time at different drying temperatures in tray and infrared drying 66

Figure 4.26 Relationship between logarithmic moisture ratio and time at different drying temperatures in fluidized bed drying 66

Figure 4.27 Arrhenius relationship between effective diffusivity and drying temperature in tray and infrared drying 67

Figure 4.28 Arrhenius relationship between effective diffusivity and drying temperature in fluidized bed drying 67

Figure 4.29 Starch noodles made from various ratios of SPS and MBS (*) 71

Figure 4.30 Starch noodles made from various solid content (*) 75

Figure 4.31 Starch noodles made from various aging time (*) 77

CHAPTER 1 INTRODUCTION

1.1 General information

Sweet potato (Ipomoea batatas) is one of the most produced food crops in the world,

especially in developing countries, ranking after cereal grain in providing nutritional and daily calorie needs for human in tropical countries Among the reasons making sweet potato is great crop are relatively to grow, relatively high productivity, high energy, and contains many biochemical active compounds used as functional food constituents, depending on the varieties Sweet potatoes are rich in starch (6.9-30.7%, wb) in which amylose content ranges from 8.5 to 38% depending on variety (Chen, 2003) Therefore, it

is considered as a good material for manufacturing starch-based products According to FAO statistics in 2008, total world production quantity of sweet potato was approximately

110 million tons in which Asian countries made up more than 80% of quantity, followed

by Africa countries about 12% total production quantity In many Asian countries such as China, Vietnam, Indonesia, or India, sweet potato is considered as a traditional food crop and was cultivated throughout country However, the profits obtained from sweet potato cultivation is much lower than those achieved from other tuber crops because most of sweet potatoes are used as a fresh, boiled, baked product or feed to animals To raise economic value of sweet potato roots needs researches to create highly valued products from sweet potatoes, and starch is one of these products because it can be used as a main ingredient or substituted component for the other starches that are high price in many food processing industries Nevertheless, currently starch processing from sweet potato just makes up a small part in total of produced starches due to its limited application in food industries Therefore, researching sweet potato starch properties as well as its applications

in food products are urgent issues to diversify products from sweet potato, and enhance its economic value

The process of starch extraction from tubers, which can be different for various materials, usually consists of three basis steps, namely extraction, purification and drying The drying

is considered as the oldest method for preservation agricultural products and enhancing food quality by reducing the moisture content to level that allows safe storage over an extended period (Akpinar et al., 2006) In starch manufacture, drying process can affect to physiochemical properties of starch (Oduro et al., 2008) The various types of drying methods can be used to dehydrate in starch cake such as tray drying, flash drying, spray drying, fluidized bed drying, and microware-vacuum or infrared drying in which the heat can be added from outside objects by conduction, convection and radiation or generated within solid objects by electric resistance (Sahin et al., 2002) Among these methods, tray drying is more favored due to its cheap cost, relatively easy to operate and control But this method also poses many drawbacks such as lengthy drying time and tremendous energy consumption and low drying efficiency because of rapid evaporation of surface moisture leading to surface hardness resulted in reduction of heat and moisture transfer and degradation of some quality attributes (Maskan, 2001) Under the similar conditions, the infrared and fluidized bed drying offer many advantages comparing to tray drying These may consist of high energy efficiency and heat transfer rate, reduced drying time and uniform temperature during drying (Sharma et al., 2005; Khir et al., 2007 & Madhiyanon

et al., 2009) In drying technology, the drying kinetic mainly depends on drying conditions,

material characteristics and drying equipment design The modeling of drying process is one of the most important aspects in drying technology to study the drying behaviors of different food products and to design effectively dryers Besides modeling drying process, the effects of drying conditions (temperature, RH, air velocity, etc) on product quality need

to be accounted to obtain desired quality products For tray drying, the various models that describe the drying behavior of different materials were proposed by many authors such as for red pepper (Akpinar et al., 2003), for eggplant (Ertekin & Yaldiz, 2004), for stanley plums (Menges & Ertekin, 2006), for red chilies (Kaleemullah & Kailappan, 2006), for parsley leaves (Akpinar et al., 2006), for golden apples (Menges & Ertekin, 2006), for aromatic plants (Akpinar, 2006), for green table olives (Demir et al., 2007), for pistachio

nuts (Kashaninejad et al., 2007), for date palm (Phoenix dactylifera L.) fruits (Falade & Abbo, 2007), for water chestnut (Trapa natans) (Singh et al., 2008), for potato slices

(Aghbashlo et al., 2009), for cocoa (Hii et al., 2009) For infrared drying, drying modeling also were performed on various products, rough rice (Abe & Afzal, 1997), onion (Sharma

et al., 2005), carrot and apple (Togrul, 2005 & 2006), wet olive husk (Celma et al., 2008), shrimp (Supawan et al., 2008), turmeric (Napasiri et al., 2009) For fluidized bed drying, many researchers also have focused on drying kinetic and drying modeling for various products such as chopped coconut (Madhiyanon et al., 2009), granular material (ragi) and green pepper (Srinivasakannan & Balasubramanian, 2008 & 2009), chillies (Tasirin et al., 2006) However, there is little information regarding to effects of tray, infrared and fluidized bed drying on properties of tuber or root starch as well as modeling of drying process under these conditions

The use of sweet potato starch as a component in starch noodle products has caused for economic value of sweet potato root rising significantly Unlike the other type of noodles such as wheat flour noodles or pasta, starch noodles is made from free-gluten starches, so the starch properties are crucial for noodle processing and final product quality Starch noodles are primarily consumed in Asian countries in which China is the most produced and consumed country Traditionally, mung bean starch is the best material for production high quality starch noodle due to high amylose content, its paste with high shear resistance, and limited swelling during gelatinization (Liu & Shen, 2007) However, the price of mung bean starch is higher than the other starches, and output of mung bean cannot meet the starch noodle rising demand of customers (Kasemsuwan et al 1998), many researches are carried out to substitute totally or partially mung bean by other kinds of starch with cheaper price in starch noodle manufacture to reduce the production cost as well as meet increasing customers demands about starch noodles such as from legume starches (Sung & Stone., 2004), from corn and potato starches (Kaur et al., 2005), from sago starches (Purwani et al., 2006), from sweet potato starches (Chen, 2003 & Lee et al., 2005), from sorghum starch (Beta & Corke, 2001), from edible bean and potato starches (Kim et al., 1996) Even although these studies showed some positive results, almost no these starches was applied

on a commercial scale Among tuber and root starches, sweet potato starch, quite cheap and abundantly available, is one of the promising substitutes for mung bean starch in noodle production; but noodle made from purely sweet potato starch is inferior compared

to purely mung bean noodle (Chen, 2003) Therefore, substitution partially sweet potato starch for mung bean starch in starch noodle production needs to be investigated to produce noodles which its quality was similar to that of mung bean starch noodles, but cheap price

Overall, this research entitled “effects of drying methods on sweet potato starch properties

and its substitution for mung bean starch in starch noodles” aims to evaluate the influences

of various drying conditions on properties of sweet potato starch, to find the best fit model

to describe sweet potato starch drying behavior under these drying conditions as well as its partially substitute ability for mung bean starch in noodle production and effects of processing conditions on the quality of starch noodles

1.2 Objectives

The specific objectives of this study include

(i) To select the sweet potato variety for starch noodle production among four types

of sweet potato variety available in Thailand market

(ii) To evaluate the influences of various drying conditions which were tray, infrared and fluidized drying methods at 45, 55 and 65oC on properties of sweet potato starch

(iii) To develop a mathematical model for describing drying behavior of sweet potato starch and determine effective diffusivity and energy of activation under those drying conditions

(iv) To investigate partially substituted potential of sweet potato starch for mung bean starch in starch noodle production, and effects of processing conditions on starch noodle quality

CHAPTER 2 LITERATURE REVIEW

2.1 Sweet potato roots and its chemical compositions

Sweet potato (Ipomoea batatas), or kumara, a dicotyledonous plant that belongs to the family Convolvulacea, is edible tuberous root with a smooth skin whose color ranges

between red, purple, brown and white, and its flesh ranges from white through yellow, orange, and purple It can be used as fresh, processed food, or raw material for manufacturing many industrial products Its flesh contains large amount of starch, high anthocyanins content which has pharmacological properties, especially in purple flesh sweet potatoes (Fan et al., 2008), dietary fiber which is an important bioactive functional component preventing heart disease and hypertension, and small quantity of minerals and protein The variation in chemical compositions of sweet potato largely depends on nutrient level and cultivars (Leonard et al., 1948)

Table 2.1 Chemical compositions of Chinese sweet potato (SP) cultivars (Chen, 2003)

or purple fleshed sweet potato flour had greater tensile force and had higher sensory scores than others due to higher amylose content Chen (2003) used starches isolated from three Chinese sweet potato cultivars for starch noodle making and found that only SuShu8, not all sweet potato varieties was suitable for noodle production Lee et al (2005) also proposed procedure for sweet potato starch noodle production without freezing but its noodle quality cannot compared to mung bean starch noodles Therefore, depend on location in which sweet potatoes were planted, varieties as well as wanted product properties, basing on its chemical compositions was considered as one of criteria for selection suitable cultivar for food production

2.2 The extraction of sweet potato starches

The basic steps that can effect to starch properties in starch extraction process consist of extraction, purification and drying The fleshly harvested roots were washed and ground with or without water by using hammer mill or blender to produce mash After grinding, the mash is washed on a synthetic screen to separate free starch and fibrous residues The starch slurry was allowed to sediment, flowing by decanting and then the wet starch cake is dewatered further by allowing it to drain overnight in a suspended screen The starch cake

after dewatering was divided into small pieces and dried by sun drying or artificial drying until its safe moisture content about 12%, wb (Duc, 1994)

Figure 2.1 Procedure for sweet potato starch extraction by wet-milling method (Duc, 1994) For extraction of sweet potato starch, many researches are carried out with various treatments to enhance the starch yield as well as starch quality Duc (1994) reported that in some communities, the sour liquid used in purification stage in sweet potato starch extraction to promote starch sedimentation, but adding sour liquid must be carefully monitor otherwise will cause negative effects on starch quality Liu & Shen (2007) also used this method for mung bean starch extraction and comparing to centrifuging method, they found that the vermicelli made from mung bean starch obtained from sour liquid processing had a better quality Kun (1998) recommended that the blending and sedimentation time were 1 minute and 6 hours respectively, together with the using cellulase enzymes (Cytolase M102, 1 ml per kg raw material and PCL5, 2.5 ml per kg raw material) in which suitable conditions for enzyme operation were at 55oC, pH 5.4, and incubation time 6.2 hr, can increase the extraction efficiency by 20-30% The use of Ca(OH)2 or lime mainly effects to whiteness of starch, almost no effect on starch yield and its properties (Kun, 1998 & Lan, 1991) Duc (1994) suggested that for starch making, it was not necessary to peel the outer skin of sweet potato before grating stage because there was no significant difference in statistic in whiteness degree of starch between peeled and unpeeled sweet potato Chen (2003) reported that during sweet potato starch isolation, the starch slurry can be browned, effecting to starch color, due to high level of polyphenol oxidase and phenolic compounds, and rinsing five times with tap water will remove brown color of starches better than using 0.2% vitamin C, and 0.2% citric acid The sweet potato after chopping into small pieces can be treated with 0.2% sodium metabisulfite for 5 min to obtain the desired starch whiteness, and this treatment was not effect to starch properties (Mais, 2008) As increasing the times of residue blending getting from filtration resulted in increasing in the starch yield from 40.7 to 164.4%, depending on variety (Rahman, 2000)

2.3 The physiochemical properties of sweet potato and mung bean starches

Chemically, starches are polysaccharides which consist of glucose units, are a mixture of amylose and amylopectin whose ratio and molecular structure, together starch granule size

Fresh roots

Washing

Chopping

Grinding Water

and structure, are play an important role in determining physiochemical and technological starch properties (Hegenbart, 1996)

2.3.1 Color

The desired starch should be high in lightness value and in low chroma value (Galvez and Resurreccion, 1992) Any foreign matters, peelings or chemical other compositions remaining in final product, together with drying methods can change the color of starch to gray Therefore, in the process of starch extraction, these components need to be removed

as much as possible The color of starch would be effect to color and transparent of products produced from that starch such as starch noodles (Lan, 1991) There were many treatments that were performed by many authors to improve sweet potato starch whiteness such as lime or Ca(OH)2, sodium metabisulfite treatment (Mais, 2008; Kun, 1998 & Lan, 1991), peeling before extraction (Duc (1994), using chemicals to inactivate polyphenol oxidase and phenolic compounds (vitamin C and citric acid) or increasing the times of washing starch slurry under tap water (Chen, 2003) However, in some places for example

in Vietnam, people do not like the products were produced from starches treated with chemicals (Duc, 1994)

To determine starch color, various methods have been recommended including visual comparison of color of starch sample with that of standard one The drawbacks of this method did not quantify the degree of difference or closeness to the standard color and was

very arbitrary A colorimetric method consists of comparison of absorbance of starch to be

determined color with that of standard in which Hunterlab and Agton colorimetric can be usually used to determine starch color (Solomon, 2003)

2.3.2 Starch granules

Starch presents in plants in form of partially crystalline granules with different size and shape depending on their botanical sources, growing and harvesting conditions While amylose mainly arranged in the amorphous regions, the amylopectin largely made up crystalline regions Most starches can be identified from their appearance under a light microscope

(a) (b) Figure 2.2 Morphological property of sweet potato (a) and mung bean (b) starches

(http://www.le.ac.uk/has/ps/past/past52/past52.html; Liu & Shen, 2007)

Sweet potato starch granules can be oval round, oval spherical round polygonal or polygonal in shape, and 2-40µm in size depending on genotypes, had a centric distinct hilum (Tan et al., 2009) while mung bean starch granules are small, either spherical or elliptical in shape, and 6.5-32.7µm in granule size and had centered fissure (Liu & Shen, 2007) Starch granule size and particle size distribution significantly affected to the functional properties of starch solubility, swelling ability and digestibility and noodle quality (Moorthy, 2002; Chen, 2003) and that was reason why although there were no different in amylose content of three sweet potato cultivars, namely SuShu2, Sushu8 and Xushu 18, only Sushu8 sweet potato starch gave the best quality noodles which can compare to mung bean starch noodles due to its high portion of small starch granules The sweet potato starch gave better noodle quality if starch size were less than 20µm (Chen, 2003)

Figure 2.3 Starch particle size distribution from sweet potato varieties (a, b and c) as

compared with potato (e) and mung bean starches (d) (Chen, 2003)

2.3.3 Swelling and solubility

When placed in excess of cold water, intact starch granule is freely penetrated by water These changes are reversible if heating temperature is below its gelatinization temperature Over gelatinization temperature, the starch granule will substantial swelling, followed by disruption of granule structure in which the loss of crystalline order and birefringence that

can be observed in the disappearance of the X-ray diffraction (Tomasik, 2004)

During the swelling process amylose tends to leach from the starch granules and the amylopectin becomes fully hydrated The granule swelling ability depending on kinds of starch, amylose content, molecular weight of amylose and amylopectin, chemical composition of starch as well as the degree of polymerization and length of branch in amylopectin (Whistler & Paschall, 1965) The amounts of amylose leaching increase with increasing the temperature and heating period of time, highly associated starch granules will be resistant to amylose leaching and swelling During this process, complexes can be formed between amylose and other components presenting in starch such as lipid and protein resulting in reducing amylose leaching (Tester & Morrison, 1990)

Swelling capacity of starch can be expressed in terms of swelling power and swelling volume which were determined by suspending a weighted quantity of starch with distilled water in a centrifuge tube and heating at particular temperature for 30 minutes The suspension is continuously stirred to obtain homogeneous solution before centrifuging The aqueous supernatant is then removed, and the weight of both wet and dry sediment starch also is determined The swelling power is expressed as the ratio of wet weight of sediment

to its dry weight, and the swelling volume is volume of sediment obtained after centrifugation Solubility is the weight of soluble starch and calculated as ratio of the weight of starch in supernatant to weight of dry matter (Solomon, 2003)

Swelling power and solubility depends largely on starch varieties and temperatures They increase with increasing in temperature The mean solubility of the different genotypes of sweet potato starch was 12.7% and the mean swelling volume was 33.0 ml/g (Collado & Corke, 1997) while the solubility of Peru sweet potato starch was about 10% The swelling power of sweet potato starch is lower than that of potato and cassava starch due to higher degree of intermolecular association At 90oC swelling power of sweet potato starch ranged from 26 to 33% while that of mung bean starch is about 10% There was no significant correlation between amylose content and swelling volume (Tan et al., 2009)

2.3.4 Pasting properties

Another important property of starch is ability creating viscous paste when heated in water The soluble starch and the continued uptake of water by the remnants of the starch granules are responsible for the increase in viscosity With continuous heating, the viscosity can be reduced due to broken down of swollen starch granules by mechanical and thermal forces The dropping of starch viscosity after reaching peak viscosity expresses the stability of starch paste during cooking Upon cooling, the re-association of soluble starches results in increase in viscosity, and indicated by the setback ratio that is the ratio

of the viscosity at the completion of cooling to the viscosity at the onset of cooling and is usually used to predict the retrogradation tendency of starch (Chen, 2003) Pasting behaviors can be determined by Brabender Viscograph and Rapid Visco Analyzer (RVA) through Amylogram (Figure 2.4) in which the latter is more applicable because of its provision of properly result in short time, smaller sample size and ease of using (Radly, 1968; Deffenbaugh & Walker, 1989)

Pasting properties of starch depend on the amylose, protein, lipid content and branch chain length of the amylopectin While amylose and lipids retard the swelling, the amylopectin contribute to swelling and pasting of starch (Jane et al, 1999; Juliano, 1985; & Tomasik, 2004) Cereal starches normally consisting of high lipid level display higher pasting temperatures and less peak viscosity and less shear thinning while root and tuber starches consist of a little lipid and thus display lower pasting temperatures and greater peak viscosities (Tomasik, 2004) Eliminating protein will enhance the starch gelatinization expressed by higher peak viscosity and breakdown viscosity (Juliano, 1985) Increasing amylose contents resulted in significantly increase starch pasting temperature and setback viscosity, decrease in peak viscosity and shear thinning (Jane et al, 1999) Collado & Corke (1997) also reported that for sweet potato starches, peak viscosity was significantly negatively correlated with amylose content However, according to Chen (2003) there was

no obvious relationship between the peak viscosity and amylose content of the Chinese 3 sweet potato starches studied They found that the differences of the peak viscosities of the

3 sweet potato starches may partly result from the different phosphorus contents Sandhu et

al (2010) also found that potato starch exhibited higher PV in comparison to rice starch due to its higher phosphorus content Some genotypes of sweet potato starch showed a broad, plateau peak while some sweet potato starches have sharp peak The values of RVA parameters for these sweet potato starches were far lower than those of mung bean starch (Tan et al., 2009) Nevertheless, the gelatinization temperature and pasting temperature of sweet potato starches was obviously higher than those of potato starch and mung bean starches although amylose content in mung bean starch was significant higher than that in sweet potato starches (Chen, 2003)

Figure 2.4 Amylogram of potato starch (4%, w/v) (Chen, 2003)

Schoch & Maywald (1968) classified starch Brabender paste viscosity patterns into four types, namely type A that show high pasting peak and rapidly thinning during cooking, type B that have lower pasting peak and less thinning during cooking, type C which have

no pasting peak but high viscosity will unchanged or further increase during cooking, and type D in which the starch quantity should be increased two or more times to get a considerable hot paste viscosity of type C Ideal starch for noodle production should show type C in Brabender Amylogram and high amylose content Due to owning those properties, mung bean starch is the most suitable raw material for making starch noodle (Tam et al., 2004; Collado & Corke, 1997 and Tan et al, 2009) while most viscosity patterns of sweet potato starch belonged to type B depending on variety (Chen, 2003) The correlation between pasting property of starch and quality of various noodles are reported by many investigators The pasting or viscoamylograph characteristics, especially the peak viscosity and the rate of viscosity breakdown after gelatinization, have been commonly used to predict the quality of Japanese white salted noodles and also applied

to Chinese yellow alkaline noodles (Batey et al., 1997 and Bhattacharya & Cork, 1996) Collado & Corke (1997) reported that increasing in pasting temperature, final viscosity and stability ratio tend to increase the firmness, but decrease the water absorption of sweet potato starches noodles Pasting properties of legume and potato starches were found to be

correlated to cooking loss starch noodles made from these starches (Kim et al., 1996) Pasting profile of starch was used as tool for predicting the quality of wheat noodles (Bhattacharya et al., 1997), and rice noodles (Bhattacharya et al., 1999)

2.3.5 Gel texture

Figure 2.5 The function of amylose in starch gel system (Ott & Hester, 1965)

(A, Amylose gel with water bond or entrapped in a three-dimensional network B, gel with

highly swollen and fragment starch granules High proportion of water held by fragment,

amylose necessary to join them in a continuous network C, gel with highly swollen and

intact starch granules High proportion of water held by granules, only limited amylose

necessary to join the large particles in a continuous network D, gel with intact starch

granule which are not highly swollen Amylose necessary to bind or entrap free water and

to join granules in continuous network)

On cooling the gelatinized starch, gelation occurs by formation of three-dimensional network of segment of linear and branched molecules to bind the swollen granules Structure and strength of gel is determined by many factors such as structure, composition and degree of polymerization of starch granule; chemical modification; the amount and types of amylose and amylopectin leached out from granules; the rate and extend of heating; addition of other substances; particle size and degree of hydration, etc (Lii et al, 1995) The soluble amylose is considered primarily responsible for gel formation, serves as main material for forming the network which binds and entraps the unabsorbed water, or links intact granules or fragments together, thus providing additional strength to network (Ott & Hester, 1965) The function of amylose in starch gel system was well described in Figure 2.5 Starch granules with low amylose content were less firm and likely to disintegrate easily, a high amylose starch a granule were more rigid and not ruptures easily (Lii et al, 1996) For rice starch gel, increasing the amylose content in gel system leads to exponentially increase the storage modules, indicating higher attribute of elastic material The gel texture was not significantly correlated to amylopectin chain length distribution (Vandeputtea et al, 2003) The starch gel hardness greatly is caused its retrogradation, which is syneresis of water and crystallization of amylopectin, leading to harder gels The starch showing harder gel is likely to have higher amylose content and longer amylopectin chain (Sandhu & Singh, 2007) The storage modulus and yield stress of the starch paste were concentrate-dependant, increasing starch concentration from 5% to 30% resulted in

suddenly increase in the storage modulus of intermediate and high amylose rice starch gel because the time to archive a closed packed system was shortened (Lii et al, 1995)

The gel formation is one of primary factors which are in charge of final quality of based products (Lii et al, 1995) The textural parameters of the gels formed in the RVA canister were well correlated with actual noodle textures Bhattacharya et al (1999) found that amylose content was major factor affecting to RVA pasting and gel texture properties RVA pasting properties as well as texture of gel obtained after running RVA were highly positive correlated to cooking and texture of rice noodles Therefore, the use of RVA together with starch gel texture obtained from Texture Analyzer could provide simple, rapid, and accurate tool for predicting the textural quality of noodle And in a previous research Bhattacharya et al (1997) also found a similar correlation between RVA pasting

starch-properties, amylose content and gel texture and wheat noodle quality

2.3.6 Retrogadation

During storage period, starch pastes may become cloudy, more opaque, rigid or rubbery This is caused by dispersed amorphous starches in pastes which gradually develops double helical crystalline structures and loses its water-binding capacity, and known as starch retrogradation process (Tomasik, 2004) Amylose molecules with linear structures are largely responsible for retrogradation process due to its development of double helical crystallites faster than amylopectin molecules with branched structures The rate of crystallization of amylopectin depends on length of branch chain in which the long-branched chains is faster in retrograde rate than those with short-branch chains During cooling storage, the gelatinized starches underwent short-term retrogradation in which largely amylose is involved while the amylopectin will contribute into retrogradation process during freezing storage (Chen, 2003) In addition amylose content, retrogradation process depends on starch concentration, storage temperature, and pH, source of starch and

presence and concentration of other ingredients The mung bean starch showed the higher

retrogradation than sweet potato starch due to higher amylose contents (Chen, 2003), and another reason for slow retrogradation rate in sweet potato starch was its amylose was likely to have more branch per amylose molecules (Tan et al, 2009)

Retrogradation is usually mentioned as an unwanted effect in starch-based products, for example, bread staling or syneresis of gel products Nevertheless, retrogradation can improve the quality of starch-based products by which more desirable textures could be achieved and reduced stickiness for Chinese rice vermicelli and Japanese noodle The starch retrogradation process will be accelerated at low temperature (Tan et al., 2009) and that was reason why sweet potato starch noodles prepared by non-freezing methods were softer and sticker than those produced by freezing method (Lee et al., 2005) The transparency of dried noodle increased with the increasing cooling and freezing treatment (Chen, 2003) Retrogradation was usually indicated by freezing-thawing stability or syneresis which was the force liquid out of the starch gels Generally, higher syneresis indicates higher retrogradation tendency of starches (Karim et al, 2000) Besides retrogradation tendency of starch can be measured by setback ratio of paste viscosity in Brabender Amylogram or RVA pasting profile and this tendency was in good agreement to the values measured by syneresis without freezing-thawing treatment (Chen, 2003) Summarily, starch can be experienced many transformation processes and gives diverse physical structures and properties (Figure 2.6)

Figure 2.6 Transformation process of starch structure (Tomasik, 2004)

2.4 Drying technology

Drying technology normally described process thermal removal of moisture to obtain a solid product When a moist product is subjected to thermal drying, two processes whose rate governs drying rate occur simultaneously, namely transfer of heat from surrounding environment to provide the necessary latent heat for moisture vaporization and transfer of moisture from internal to surface and then its subsequently vaporization due to resulting of heat transfer (Mujumdar, 2006) On basis of the way in which heat is supplied, drying process can be perhaps classified into drying by conduction, convection, radiation, high frequency or microwave and freezing drying (Eberle, 2005) The raw material characteristics, final product quality and economic profit need to be considered whenever final choosing of drying method (Leniger & Beverloo, 1975) In dehydration operation, hot air drying, especially tray drying is the most frequently used to remove moisture from food products because of its advantages such as cheap cost, relatively easy to operate and control However, it also has many drawbacks such as lengthy drying time and tremendous energy consumption and low drying efficiency (Maskan, 2000) These disadvantages can

be overcome by using infrared and fluidized bed drying Both methods can be used for dehydration from particle-like materials as starch cake

2.4.1 Hot air drying

The largest and the most important group of driers are those in which air is used as drying medium after heated to required temperature in a heat exchanger They are called as hot air drying or convection drying because the heat is supplied largely by convection from heating medium into surface before transferring into the interior of material by conduction Therefore, the surface layers of material become dry first and then hinder the dehydration process (Leniger & Beverloo, 1975 & Earle, 2004) In hot air drying, the rate of moisture removal depends on the conditions of the air, the properties of material and the design of the dryer The water which is loosely hold, free-moisture, will be removed easily in drying process while removal of bound water requires longer time and elevated temperature As drying proceeds, the moisture content falls and the access of water from the interior of the food to the surface affects the rate and decreases it (Earle, 2004) The hot air drying can be performed by various drying equipment such as tray driers in which material to be dried is spread on a tray and then placed in an oven; spray driers in which liquid or slurry is sprayed in form of droplets in a chamber by an atomizer and droplets come to contact to hot air that is moved co-current, counter-current or mixed flow to the solids; drum driers consisting of one or more rotating heated drums and slurry or liquid is dropped onto

Heating Cooling Retrogradation Storage

outside of the drum, the dried material is scrapped of by using knife; fluidized bed driers including a vessel in which air is blown through perforated plate above which is a layer of solid particles, the air-flow rate must enough strong to suspend the solids in upward flow

of air; rotary driers in which consist of a rotating cylinder inclined a few degree to horizontal, the material is fed in at the higher end and hot air can move co-current or counter-current to solids; and tunnel driers which a series of trolleys or trays moves slowly through a heated tunnel (Mercer, 1994) In fluidized bed drying, high rate of heat and mass transfer, and uniform drying were obtained through mixing of particles This method was easy to control and overheating can be avoided, resulting in better quality of final product This method can be used for drying wet particulate, granular materials and even paste, slurry which can fluidized in beds (Law & Mujumdar, 2006; Madhiyanon et al., 2009) Yadava et al (2006) compared the effects of drum drying and tray drying on properties of sweet potato flour, and found that viscosity of starch suspension in both drying methods decreased significantly comparing to native starch in which the dropping degree of viscosity in tray drying was greater and in-vitro digestibility was better due to starch breakdown to simple sugars owing to enzyme action during processing However, swelling power and solubility of drum dried flour is higher than tray dried flour Oduro et al (2008) investigated the effects of solar and tray drying on physiochemical properties of sweet potato starches, and concluded that the swelling power and solubility values for tray dried starches were significantly different from solar dried ones, water binding capacity and amylase content were higher in solar dried starches but there was no statistically significant difference Ahmed et al (2010) found that the increasing drying temperature in hot air drying strongly negative effect to color, browning index and acid ascorbic content of sweet potato flours But, increasing drying temperature led to increase water absorption, water solubility index and swelling capacity of them

2.4.2 Infrared drying

Infrared drying is drying with heat radiation which derived from special filament bulbs or special types of gas This radiation has wavelength between 0.7 and 1,000 µm which is usually divided into near-infrared (0.75-1.4 µm), short-wavelength infrared (1.4-3 µm), mid-wavelength infrared (3-8 µm), long-wavelength infrared (8-15 µm), and far infrared (15-1,000 µm) Thermal radiation incident upon a body may be absorbed and its energy converted into heat, reflected from the surface or transmitted through the material (Ratti & Mujumdar, 2007) When a wet object is subjected to infrared radiation, the highest temperature occurs under the irradiated surface layer and then is conducted toward to the center as well as surface of product while the moisture all the time is transported from center to surface Therefore, fluxes of heat and moisture are countercurrent in some part of the material and in layers close to the surface is co-current The penetrated depth of infrared radiation (IR) depends on wavelength of IR, composition, water activity, structure and thickness of the product (Tirawanichakul et al 2008) Infrared radiation with short wavelength is transmitted through material while at long wavelength it is absorbed on the surface Therefore, for thin layer drying, far infrared radiation tend to be more efficient while near infrared radiation is more suitable for thicker objects (Nowak & Lewicki, 2004) Infrared heating presents many advantages with respect to tray drying such as high energy efficiency and heat transfer rate, reduced drying time and necessity for air flow across product, uniform temperature during drying (Khir et al 2007)

2.4.3 Drying kinetics

Drying is a complicated process involving simultaneous heat and mass transfer The removal of moisture from a typical food product usually follows the drying rate curve as illustrated in Figure 2.7 which consists of three consecutive periods, namely heating product to required temperature (AB), constant-rate drying (BC) and falling-rate drying period (CD) (Singh & Heldman, 1993)

Figure 2.7 Typical drying rate curve of for food solids(Singh & Heldman, 1993)

In drying technology, a critically important aspect is mathematic modeling of drying process which permits designers of drying equipments to select the most suitable operating conditions as well as size of equipment accordingly to desired operating conditions (Demir

et al., 2007) To simplify, almost all mathematic models only were developed for thin layer drying that means to dry as one layer of sample particles or slices and materials fully exposed to heated air leading to dry uniformly throughout the drying layer, and air humidity does not change significantly while passing through the products (Akpinar, 2006 and Pabis & Stanislaw, 1998) There are three important types of models for predicting the movement of moisture during the thin layer drying namely theoretical, semi-theoretical and empirical models Among these models, the first approach considered only internal resistance to moisture transfer while the other models accounted for only external resistance to moisture transfer between product and air (Menges & Ertekin, 2006 and Akpinar, 2006) To develop drying models for agricultural products, the moisture content

of the material at any time must be measured and expressed in terms of moisture ratio and correlated to the drying parameters (Akpinar, 2006)

(2.1)

MR = Mt - Me

Mo - Me

Where MR = Moisture ratio (dimensionless)

Mo = Initial moisture content at t = 0 (g water/g dry matter)

Mt = Moisture content of product at any time t (g water/g dry matter)

Me = Equilibrium moisture content (g water/g dry matter)

Recently, many researches on the mathematical modeling have been conducted on the thin layer drying processes of various agricultural products For hot air drying, the various models that describe the drying behavior of different materials were proposed by many authors such as for red pepper (Akpinar et al., 2003), for eggplant (Ertekin & Yaldiz, 2004), for stanley plums (Menges & Ertekin, 2006), for red chilies (Kaleemullah & Kailappan, 2006), for parsley leaves (Akpinar et al., 2006), for golden apples (Menges & Ertekin, 2006), for aromatic plants (Akpinar, 2006), for green table olives (Demir et al.,

2007), for pistachio nuts (Kashaninejad et al., 2007), for date palm (Phoenix dactylifera L.) fruits (Falade & Abbo, 2007), for water chestnut (Trapa natans) (Singh et al., 2008), for

potato slices (Aghbashlo et al., 2009), for cocoa (Hii et al., 2009) For infrared drying, drying modeling also were performed on various products, rough rice (Abe & Afzal, 1997), onion (Sharma et al., 2005), carrot and apple (Togrul, 2005 & 2006), wet olive husk (Celma et al., 2008), shrimp (Tirawanichakul et al., 2008), turmeric (Laosanguanek et al., 2009) For fluidized bed drying, modeling of drying process also was focused on chopped coconut (Madhiyanon et al., 2009), granular materials (ragi) and green pepper (Srinivasakannan & Balasubramanian, 2008 & 2009), chillies (Tasirin et al., 2006) The drying models that these authors used to develop a general equation for various materials were summarized in Table 2.2

Table 2.2 Thin layer drying models

Henderson and Pabis

Wang and Singh

Logarithmic

Two term

Two term exponential

Modified Henderson and Pabis

2.5 Starch noodles

2.5.1 The methods of starch noodles production

Starch noodles or cellophane noodles due to their transparent or translucent appearance of before and after cooking, made primarily from various gluten-free starches, are major type Asian noodles The starch properties play an essential role in determining how easy to make noodles and its final quality Traditionally, mung bean starch is considered as the best material for production high quality starch noodles (Kim et al., 1996) Depending on the starch properties and desirable qualities of final products, the starch noodles can be produced by dropping, cutting, or extruding method The common characteristics of these methods are heat treatment starch dough or slurry which are boiling or steaming to gelatinize the starch and cooling or freezing to accelerate retrogradation process which sets noodle structure (Tan et al., 2009)

a) The dropping method

This method includes gelatinizing about 5% of starch in hot water to produce starch paste, then mixing with 95% of dry starch and water to form dough about 50% moisture content, followed by mixing and stirring in a blender with suitable speed and time to obtain a smooth starch dough which does not stick to the hands The dough was extruded through the holes 0.5-1.5cm in diameter of the stainless steel cylinder directly into hot water for about 1 minute and when noodles floated upon the surface of water, transfer them into cold water After that, the strands are drained, separated, and dried at 40oC in convection dryer until 12% of MC, and then packed in polyethylene bags and stored at room temperature

(Tan et al., 2009) This method was used by many researches for evaluating starch noodle

quality made from various starch sources such as starches of chick peas, peanut beans and pinto beans (Sung & Stone., 2004), corn and potato starches (Kaur et al., 2005 and Singh et al., 2002), sago starches (Purwani et al., 2006), sorghum starches (Beta & Corke, 2001)

edible bean and potato starches (Kim et al., 1996), sweet potato starches (Chen, 2003) b) The cutting method

Cutting method is another traditional method in manufacturing starch noodle, and is largely used in Japan Because of high cost of freezing process, noodle is produced without freezing The gelatinized starch is quickly cooled and removed from conveyer in form of elastic sheets after cooking in steam chamber The starch sheet then is aged at low temperature, cut into thin noodle strands, and dried in hot air oven This method is low cost and simple but the quality of starch noodle in terms of texture and cooking quality are inferior that making from drooping method (Tan et al., 2009) By utilizing this method, Lee et al (2005) found that the sweet potato starch noodles prepared from a slurry of 45% solids, aged for 21 hour, and then dried at 25 or 65°C were most comparable in textural properties and cooking loss to the commercial starch noodles Hormdok & Noomhorm (2007) also use this method to evaluate improvement of quality of rice noodles by using hydrothermal treatments of rice starches

c) The extrusion method

Extrusion cooking has become a popular processing method for starch based foods, gelatinized starches, or paste products (Li & Vasanthan, 2003) The starch noodles made

pre-by this method show highly transparent and less melted pre-by boiling (Tan et al., 2009) Starch dough is prepared by adding about 45-55 parts by weight of hot water into 100 part starch and kneading it If amount of added hot water is insufficient the dough sheets cannot form or easy to break during extrusion because dough particles are small, brittle and hard Starch dough is then extruded into dough sheets in an extruder Starch dough must be passed through the degassing zone to sufficiently remove gases otherwise; the dough sheet will not form due to the voids The unevenly distribution of water in dough also lead to lack of uniform transparent of starch noodles After rolling, the dough sheets are placed into chamber to add water into it, and steamed in steamer to completely gelatinize starch The dough sheets are then retrograded by cooling Retrograded dough sheets are cut linearly into noodle strands by cutting rollers, and dried in hot air oven until 10-15% MC

Li & Vasanthan (2003) used extrusion method to produce starch noodles with hypochlorite oxidation of field pea starch and suggested that mixing and extrusion should be maintained under pressure to avoid chalky in appearance due to presence of small bubbles in final The noodle should be kept under 4oC to accelerate retrogradation which enhances noodle quality Unlike dropping method, this method do not need processes of pre-gelatinized starch, cooking in boiling water and cooling in cold water after extrusion (Tan et al., 2009)

2.5.2 The effects of processing conditions on starch noodle quality

a) Dough properties

The quality and formation of starch dough play an essential role in producing of starch noodles The final moisture content of starch dough greatly effect to all noodle physical properties and cooking temperature and cooking time Increase dough moisture content results in decreasing of starch dough viscosity, water absorption index and cooked noodle hardness, increasing of solid loss (Tan et al., 2009) The final dough moisture content strongly influences to transparency and cooking loss of mung bean starch noodles, (Galvez

et al., 1994) Yoenyongbuddhagal (2002) reported that dough moisture content had influenced to extrusion process and rice vermicelli quality, dough with high moisture content caused vermicelli strands stick together, soft texture while dough with low moisture content leads to difficult for extrusion and lower noodle quality Dough with 35-40% MC was suitable for producing high quality rice vermicelli Besides dough moisture content, other critical factors that affect to noodle quality are the extent of gelatinization before and after extrusion Unlike to wheat noodles in which gluten protein is responsible for forming the network to integrate other components to form visco-elastic dough and gluten content, especially soluble glutenin, was considered the most desired component for wheat noodles (Zhong et al., 2007), in starch noodles the dough was formed by mixing partially pre-gelatinized starch paste, native starch and water in which pre-gelatinized starch paste functions as gluten protein in wheat noodles to integrate native starch to form dough The amount of pre-gelatinized starch paste in dough determined the noodle quality (Chen, 2003) Normally, in starch noodles, the pre-gelatinization starch paste accounted for 5% of total starches and prepared by heating mixture of dry starch and water (1:7 w/v) at

95oC in 5 minutes in water bath (Kim et al., 1996) If the amount of pre-gelatinization starch paste is small, the starch dough will be too large fluidizedity to form starch noodle due to lack of glutinosity (Tan et al., 2009) In spite of reducing the extrude output, dough

with high degree of gelatinization result in enhancing rice vermicelli quality by decreasing

of cooking loss, water absorption index and adhesiveness; and increasing in springiness, cohesiveness and firmness (Yoenyongbuddhagal, 2002) Water uptake during cooking after extrusion effected starch noodle texture and cooking quality, excess water uptake leading

to very soft and sticky texture while insufficient water uptake resulting in coarse and hard texture (Tan et al., 2009) Cooking condition need to be carefully considered to assure sufficient water content for starch to fully gelatinize, depending on solid content in noodle strands The noodles made from sweet potato starch by non-freezing method which contained 38-45% solid content provided a uniform and translucent appearance (Lee et al., 2005)

b) Cooling process

The purpose of this process is to enhance gelatinized starches in noodle strands experience retrogradation process during which cooled gelatinized starches reformed to an ordered system Retrogadation of starch is in charge for stability of starch noodles and ability to withstand boiling temperature This process is carried out by washing noodles in cool water or freezing and thawing treatments (Tan et al., 2009) Lee et al (2005) found that the rate of retrogradation rate is greatly dependent on the starch content in a gel The maximum retrogradation rate was observed at a solid content of 50-55% Aging conditions had significant effect on transparency and cooking loss The cooking loss of noodles decreased as aging time increased but increased with increasing the solid content The firmness of a starch gel increased linearly with aging time and solid content of noodles

c) Drying process

Starch noodles after experiencing retrogradation in cooling step are dried at 40oC in a convective dryer, and cooling before packaging The cooking loss of noodles was not affected by drying temperature, and effect of drying temperature on noodle texture was less significant than those of solid content and aging time (Lee et al., 2005) Lan (1991) reported that drying temperatures did not affect to quality of dry sweet potato noodles in terms of transparent, diameter, water absorption index and hardness but greatly inverse affect to cooked noodle quality such as hardness, water absorption index and sensory evaluation

The various legume starches such as cowpea, pea, cassava, potato, sweet potato, etc are used for noodle manufacture, and many treatments are applied to assure the noodle quality can compare with that of mung bean starch noodles Sung & Stone (2004) reported that the cooking quality of noodle making from pinto bean and peanut bean starches was inferior to that of noodles produced by chick pea and mung bean starches, and the using surfactants did not improve cooking qualities of starch noodles Kaur et al (2005) suggested that using glycerol monostearate enhanced the quality of corn and potato starch noodles Sago starch also can be used for producing noodle after treating the 25% MC starch at 110oC for 16 hours at resulting in improve noodle quality, having higher firmness and elasticity, but lower stickiness and cooking loss comparing to starch noodle making from non-heat moisture treatment (Purwani et al., 2006) Kasemsuwan et al (1998) reported that the quality of the noodles made from a mixture of cross-linked tapioca starch and a high amylose maize starch was similar to the clear noodles produce from mung bean starch Singh et al (2002) found that potato starch was more suitable for making noodle than corn starch, the corn starch noodles showed lower cooked weight, hardness and

cohesiveness than potato starch Besides, starch noodles can be made from sweet potato starches (Lee et al., 2005 and Chen, 2003), from sorghum starch (Beta & Corke, 2001), from navy, pinto, and potato starches (Kim et al., 1996) The stickiness of mung bean starch noodles is much lower than that of noodles made from the other starch sources; it is easy to separate during drying The stickiness of starch noodles can be decreased significantly by freezing treatment Therefore, freezing treatment is a very important step

in starch noodle making (Tan et al., 2009)

2.5.3 Indices for evaluate the starch noodle quality

Noodles qualities are evaluated by visual attributes of the fresh and cooked noodles The dry starch noodles should be high transparent, high glossiness, inexistence of discoloration and straightly fine threads The most important characteristics for cooked starch noodles are texture and mouthfeel, they should remain firmness, not sticky after cooking, high tensile strength, short cooking time and low cooking loss (Kim et al., 1996, Tan et al., 2009) In recently researches, the qualities of starch noodles usually are measured by three different aspects, namely sensory evaluation, cooking quality, and textures of starch noodles (Chen, 2003; Lee et al., 2005; Tan et al., 2009; Kaur et al., 2005; Singh et al., 2002; Lan, 1991; Yoenyongbuddhagal, 2002) Although sensory evaluation is difficult to obtain reliable result due to variability from person to person and distinguishing capability

of test panels, it express acceptance of the sensory attributes of a product by consumers who are the regular users of the product

CHAPTER 3 MATERIALS AND METHODS

3.1 Materials and equipments

3 Fluidized bed drier

4 Rapid Visco Analyzer

(*) : SP1_W_YR - The white skin and yellow-red flesh color; SP2_P_P - The purple skin and purple flesh

color; SP3_P_Y - The purple skin and yellow flesh color; SP4_OP_O - The orange-purple skin and orange flesh color

Figure 3.1 The framework of study

Evaluation noodle quality

1 Using 11 models (Table 3.1)

2 Propose a general model and determination Deff and Ea for each drying method

Selection of sweet potato (SP) variety (Objective 1)

 Starches isolated from 4 SP varieties (*):

SP variety

Drying with tray, infared and fluidized bed drying at the same temperature range (45, 55, 650C) until 10% of MC, wb

Effects of drying methods on

starch properties (Objective 2)

1 Color

2 Pasting properties

3 Gel texture

4 Swelling power and solubility

The effects of substituted ratio of SP starch for MB starch and processing conditions on starch noodle quality (Objective 4)

Substituted ratio of SP starch

3.2.1 Selection of sweet potato cultivar for starch noodle production

Four types of sweet potato cultivar available in the market (Figure 3.2), namely white skin and yellow-red flesh color (SP1_W_YR), purple skin and purple flesh color (SP2_P_P), purple skin and yellow flesh color (SP3_P_Y), and orange-purple skin and orange flesh color (SP4_OP_O) sweet potatoes, were used in this experiment to find the most suitable variety for noodle making on the basis of starch yield (%), proximate analysis (moisture, protein, lipid and ash content), amylose content, granule characteristics, color, gel hardness and pasting properties (RVA) in starches, which are illustrated in 3.2.5 part

Figure 3.2 Four sweet potato varieties Sweet potato starches were isolated according to the method of Mais (2008) with a slight modification The sweet potato roots were washed thoroughly, peeled, cut into small pieces (4x4x4mm) that were then soaked in tap water 30 minutes (2 liters of water for 1kg sweet potato roots) and ground in a blender (Sharp, Model EM-11, Japan) for 2 min The slurry was filtrated through a fabric filter to remove fibers and other components before passing through the 100-mesh sieve The residue was mixed with water (2 liters of water for 1kg initial sweet potato root) for the second filtration to enhance starch yield The filtrate was allowed to undisturbed stand 3 hr The collected starch re-suspended with tap water, settled down and removed water This process was repeated in three times to remove pigment residue The obtained starches were dried in hot air oven at 50oC for 11 hrs until about 10% MC, wb It then was finely ground and packaged in polypropylene bags and kept at

cold room (4oC) until further analysis The starch yield (%) could be calculated by following formula

3.2.2 Determination of effects of drying conditions on properties of sweet potato starch

The white skin and yellow-red flesh color (SP1_W_YR) variety was chosen for further experiment (4.1 part) The sweet potato starches with 45% initial MC were dried at different conditions which were tray, infrared and fluidized bed drying at 45, 55 and 65oC for each method until 10% wb final MC (more detail in 3.2.3 part) The starch properties in terms of color, gel texture, swelling power, solubility and pasting properties were analyzed

to evaluate the degree of influences of these drying conditions

3.2.3 Development of modeling drying process of sweet potato starches

The isolated starches were dried under tray, infrared and fluidized bed drying at the same temperature 45oC, 55oC and 65oC for all drying methods Each experiment was carried out

in three replications and average values were used for further analysis

For both tray and infrared drying, about 280g of the isolated starch cake after passing through a 1.6 mm-diameter sieve was spread into a 1-cm layer on a metal tray whose dimension was 25x15x1.5cm (length x width x height) The drying process was carried out until constant sample weight The drying air velocity was also fixed the same for both equipments at 1.2 m/s which was measured by an anemometer (Lutron, AM-4201, Taiwan) The dryers were operated for 1 hr to achieve a steady state before running with sample The reduced sample weight was recorded at 1 minute interval for first 20 minutes, afterward weight intervals increased to 10 minutes during the drying process by using the digital balance (NJW-3000, Japan) The laboratory scale of tray and infrared convective dryers were developed by Food Engineering and Bioprocess Technology workshop, Asian Institute of Technology

Figure 3.3 Tray dryer equipment The tray dryer consists of the drying chamber constructed by stainless steel sheets as a rectangular tunnel whose dimensions were 45x30x30cm The temperature in drying chamber was adjusted by the heater power control The air velocity was controlled by adjusting openings at the end of drying chamber The heating system consisted of an electric 3000W heater placed inside the duct which was used to heat up the air to the

%, Starch yield = Extracted starch

Total amount of raw sweet potato root * 100 (3.1)

Air inlet

Air outlet Fan Heater

desired drying temperature The airflow was passed over the sample layer during drying process (Figure 3.3)

The basic design of the infrared dryer consists of a stainless steel drying chamber with inner dimensions of 35x25x25cm and ceramic infrared heater (122x60mm, 230V, 500W, Thailand) was installed at the top of the drying chamber (Figure 3.4)

Figure 3.4 Infrared drying equipment For fluidized bed drying, fluidized bed drier (Model F10A, Sherwood Scientific Lit UK)

as shown in Figure 3.5 was used for this experiment The fluid bed is glass cylindrical with

28 cm in height, 14.5 cm in inside diameter, and voltage of 230V, 50Hz, 3kW The fluidized bed drier was operated about a half hour to get setting conditions before putting about 280g of sample in form of 1.6mm-diameter particles The flow rate of hot air was also fixed at 1 m/s The weight loss was recorded at 1 minutes interval during the drying process by using digital balance (Sartorius, LC 62008, sensitivity 0.01 g)

Figure 3.5 Fluidized bed drying equipment

3.2.3.1 Modeling of drying process

The initial moisture content (Mo) was determined according to AACC (2000) method Because the equilibrium moisture content of sweet potato starch at various drying conditions was not available in literature review, it would be determined by following dynamic method The moisture content at which sample weight was unchanged with drying time was the equilibrium moisture content (Me) (Kashaninejad et al., 2007, Natthaporn, 2007) These values, together with moisture content at interval drying time (t) were used to calculate the MR which was subsequently regressed with 11 mathematical models (Table 3.1)

Table 3.1 The models used for modeling drying process (Menges & Ertekin, 2006)

Henderson and Pabis

Wang and Singh

Logarithmic

Two term

Two term exponential

Modified Henderson and Pabis

(3.2)

(3.3)

3.2.3.2 Determination of moisture diffusivity and activation energy

The Fick’s second law of diffusion is mathematic equation which is commonly used for describing drying process with assumptions moisture migration being only by diffusion; uniform initial moisture distribution; effective moisture diffusivity and temperature being constant; and sample shrinkage being negligible (Khraisheh et al., 1997) (where, M = the local moisture content (kg water/kg dry solids), r = the diffusion path (m), t = the time (s) and Deff = the moisture dependent diffusivity (m2/s))

dM

d M dr

Crank (1975) solved The Fick’s second law of diffusion for an infinitive slab (Equation 3.6) and spherical object (Equation 3.7) of unsteady state diffusion used to determine the moisture ratio with initial and boundary conditions, firstly the moisture is initially uniform distributed throughout the sample, secondly the mass transfer is symmetric with respect to the centre of the particle layer, thirdly the surface moisture content of the samples instantaneously reaches equilibrium with the conditions of surroundings air (M = moisture content at t time interval (g water/g dry matter), Mo = initial moisture content, Me = equilibrium moisture content, R = radius of particle (m), L = the thickness of slab (m)) For tray and infared drying because the sample was spread into layer, it can be considered as infinitive slab while for fluidized bed drying the sample was subjected into hot air flow in form of separated particles, so Fick’s second law of diffusion for spherical object was used

to find effective diffusivity and activation of energy

MR = M − M

M − M

For infinitive slab

MR = 8

π

1(2n + 1)exp −

(2n + 1) ∗ π ∗ D ∗ t

4L For spherical particles

For long drying periods, MR < 0.6, the equation can be simplified to the first term (n=0 in Equation 3.6 and n = 1 in Equation 3.7) of the series with small error and taking natural logarithm in both members the result is the following equation (Equation 3.8 and 3.9) For infinitive slab

ln(MR) = ln 8

π −

π ∗ D ∗ t4L

For spherical particles

ln(MR) = ln 6

π ∗ D ∗ tR

The effective diffusivity was determined from slope obtained by plotting straight line of

ln(MR) versus drying time (t)

For infinitive slab

Slope = π ∗ D

4L

For spherical particles

Slope = π ∗ D

RThe energy of activation was calculated from an Arrhenius equation (3.12) by taking

natural logarithmic and plotting the graph between ln(Deff) versus (1/T) (3.13)

∗ ) (3.12)

ln (D ) = ln (D ) −

∗ (3.13)

Where D0 = the pre-exponential factor of the Arrhenius equation (m2/s)

Ea = the activation energy of the moisture diffusion (KJ/mol)

T = the air absolute temperature (K)

R = the gas constant (8.3143 kJ/kmolK)

3.2.4 Evaluation of the substituted ratios of sweet potato starch (SPS) for mung bean

starch (MBS) and effects of processing conditions in starch noodle production

The procedure for starch noodle making was followed according to Lee et al (2005) and

Hormdok & Noomhorm (2007) with a slight modification The SPS were well mixed to

MBS to form starch mixture in which SPS accounted for 0, 10, 20, 30, 40 and 100% The

water at was added into starch mixture to obtain starch slurry with solid content of 35, 40

and 45% The starch slurries were equilibrated at room temperature for 1 hr before pouring

into stainless plate whose dimensions were 150x160x1mm in length, width and height

respectively, and then spread to form sheet in 1 mm thickness After steaming at 92.5oC

until complete gelatinization, the samples were cooled at room temperature for 10 minutes

And then covered with aluminum foil and aged in refrigerator (4oC) for 1, 10 and 20 hr

After that, the starch gel was cut into thin-noodle strands (2 mm in width) by cutting knife

The noodle strands were dried in an air oven at 40oC until about 12% of final MC To

simplify the experiment, as studying effects of a processing condition on starch noodle

quality, other processing condition was kept constant The best processing condition from

previous step was chosen for the next step The quality of both dried and cooked noodles

was evaluated by following methods given in 3.2.6 part

3.2.5 Determination of sweet potato starch physiochemical properties

a Moisture content

The moisture content of sweet potato starch was determined by using procedure of AACC

method (2000) including drying 5 gram sample at 105oC in hot air oven about 24 hrs The

(3.11) (3.10)

moisture content was expressed as the percentage of ratio of weight moisture loss to weight

of initial material

b Protein content

Protein content is determined by Micro-Kjeldahl Method that was as described in AACC method (2000) The sample of 0.5 gram was placed into digestion flask, and 20 ml of concentration H2SO4 with amount of catalyst (K2SO4:CuSO4: Se = 50:10:1) was added The sample was digested with digestion system until solution is clear After that, the sample was distilled in distillation unit into receiving flask containing 20 ml of boric acid and about 10 drops of methyl red as indicator (0.1 gram methyl red and 0.1 gram bromcresol green in 100 ethyl alcohol solution) The distillate was then titrated by standard 0.1N HCl solution and protein content was calculated by following formula

(3.14) % Protein = % Nitrogen * 5.95

Where S = ml of titration of sample

d Lipid content

Lipid content was determined by AOAC method (1999) with Soxtec System Wrapping 4 gram sample by free fat filter paper and putting into the thimble with a cup in the extractor system and extract with petrolenum ether The petrolenum ether was boiled to extract oil from sample in 1 hour Before drying the sample in oven at 100oC in 30 minute, the petrolenum ether needs to be evaporated from sample The fat content was calculated by following formula

e Light microscopy observation

The shape and size of starch granule were determined and photographed by using BX 40 microscope (MBL2100, Germany) interfaced to digital camera (DCM 300, resolution 3M pixels) following the method used by Chen (2003) Placing a drop of starch suspension (0.1 g starch in 10 ml water and stirring thoroughly) onto the glass plate and cover with cover slit and observing under microscopic Granule size was measured using a microscope fitted with a calibrated eyepiece to calculate the average and range of the granular size

% Lipid = (Weight of fat extracted – blank) * 100

Sample weight

% Nitrogen = (S – B) * N * 14.007 * 100

Sample weight (mg)

f Amylose content

Apparent amylose content was determined by colorimetric iodine assay index method, following Juliano (1985) A 100 mg of sample was taken in a 100 ml volumetric flask Then 1 ml of 95% ethanol was added and followed by 9 ml of 1M NaOH for washing down the entire sampler adhering to the side of the flask The contents of flasks were boiled in a water bath for 10 min to gelatinize the starch The samples were allowed to be cooled Then distilled water was added to the flask to increase the volume to exactly 100

ml and mix well The prepared solutions were stored over night at room temp (25˚C) Above steps except taking flour to the 100 ml flask were also followed to prepare a blank treatment solution and kept overnight After that, the flask was thoroughly mixed and 5 ml solution was pipetted out into a 100 ml volumetric flask Then about 70 ml of distilled water was added followed by 1 ml of 1M acetic acid and 2 ml of iodine-solution The volume of the solution in the flask was adjusted to 100 ml by distilled water The sample was thoroughly mixed and left for 20 min to develop dark purple color

The absorbance of the color was measured in the spectrophotometer (Jenway-6405, UK) at 620nm after setting zero of the equipment with the blank treatment solution The value of the absorbance was then converted back to amylose content using standard calibration curve developed for pure potato amylose standard

g Starch color

The color values of starches namely L* (brightness), a* (redness to greenness), and b* (yellowness to blueness) values were measured by Hunterlab Colorimeter (Colorflex, USA) The whiteness values were obtained by following equation

100100

h Pasting properties

Rapid Visco Analyzer (RVA, Mode l4D, Newport Scientific, Australia) was used to determine the starch pasting properties (AACC, 2000) 3 gram of sample was suspended into 25 ml distillated water in a test canister The starch suspensions were processed with a controlled heating and cooling cycle under the condition of constant shearing, in which the sample was held at 50oC in 1 min, heated to 95oC in 7.5 min, and held at 95oC for 5 min before cooling to 50oC in 7.5 min and holing for 2 min The rotating speed was maintained

at 160 rpm along the process Amylogram profile showed measured RVA parameters in terms of pasting temperature, peak viscosity, breakdown, final viscosity, setback, and consistency

i Gel texture

Gel texture was determined by using Instron TA.XT2 Plus, uniaxal compression (Stable Micro System, USA) as reported by Pons et al (1996) Starch gel was made by heating and stirring continuously 10% starch slurry at 85oC for 20 minutes It then was poured into in PVC pipe (3/4 inch in diameter) After sealing with covers at both two ends the tubes were kept overnight at 4ºC The dimension of gel obtained was 25 mm in height The gel was compressed at 1 mm/s to distance of 10 mm by using stainless steel punch probe (P/35,

35.0 mm in diameter) The gel hardness and adhesiveness were obtained from the peaks of force-time curve (Figure A.1)

j Swelling power and solubility

Swelling power and solubility of starch was determined by the method introduced by Shimelis et al (2006) 0.2 gram of sample was suspended in centrifugal tube containing 10

ml of distilled water The suspended sample was heated at 92.5oC for 30 min in water bath with intermittent stirring to keep starch granule suspended The gelatinized starches were rapidly cooled to room temperature and then centrifuged at 3500 rpm for 15 minutes The aqueous supernatant was taken out and placed in a petri dish, dried at 100oC in hot air oven for 4 hrs The weight of swollen starch sediment was determined The swelling power and solubility obtained by following formulas

(3.17)

(3.18)

Where W1 = weight of supernatant after drying

W2 = weight of swollen starch sediment

Ws = weight of sample

MC = moisture content, dry basis (decimal)

Wdm = weight of dry matter = Ws*(1–MC)

3.2.6 Determination of quality of starch noodles

Quality of starch noodles was evaluated for both dry and cooked noodles For dry noodles, color and sensory (appearance, texture and overall acceptability) were determined while quality of cooked noodles was examined in terms of color, tensile, cooking quality and sensory (appearance, elasticity, stickiness and overall acceptability)

a Color of starch noodles

The color of both dry and cooked starch noodles was determined by Hunterlab Colorimeter (Colorflex, USA)

b Cooking quality of starch noodles

The cooking time of the starch noodles measured by cooking 5 g noodle (2-3 cm long) in

200 ml distillated water, every 30 second the noodle strands were removed and pressed between two pieces of watch glass Optimum cooking time was achieved when the center

of the noodles was fully hydrated (Purwani et al., 2006)

Cooking loss and percent of rehydration starch noodles were determined by the method introduced by Hormdok & Noomhorm (2007) A 1 g of dried starch noodles were cut into small pieces (2.0 cm in length) and boiled in 60 ml distilled water for one min more than cooking time Cooked samples were washed with 20 ml distilled water, drained for 5 min

Solubility, % =

W1 * 100 Ws*(1 – MC)

Swelling power, % =

W2 * 100

Wdm * (100 – Solubility)

and weighed immediately The cooking water was collected and transferred to a petri dish and dried at 105oC until constant weight Total cooking loss (CL) was calculated based on the dry weight of noodles The rehydration (RE) was calculated as the percentage increase

in weight of the cooked noodle compared to the weight of dried noodle

c Extension of starch noodles

The extension of cooked starch noodles (a single strand) was measured by using Instron TA.XT2 Plus, uniaxal compression (Stable Micro System, USA) as shown in Figure 3.6 with the test speed was 1.00 mm/s The extension modulus (E, MPa) and the relative extension (re, dimessionless) were calculated from the following equations : E = (F/∆L)(L/A) and re = ∆L/L Where F is the extension force (g force), A is the cross sectional area of starch noodle (mm2), ∆L is the increased length (mm) and L is the original length of starch noodle (mm)

Figure 3.6 Description of extension measurements

d Sensory evaluation of starch noodles

Multiple comparison test was used for sensory evaluation the quality of both dry and cooked starch noodles to minimize the complexity of simultaneous comparion for all samples The cooked noodles were prepared by cooking in boiling water for 1 minute more than cooking time and cooling in cold water for 5 minutes before evaluation The twelve Vietnamese panelists were asked comparing each of all samples to the reference sample made from pure mung bean starch in terms of appearance, texture, flavor and overall acceptability by using 9-point scale (Appendix B) They were informed the method evaluation and terminology of quality attribute before evaluation

3.2.7 Data Analysis

All experiments and analysis were performed in three times The data were subjected to statistical one-way ANOVA test and Fisher’s Least Significant Difference (LSD) test or Duncan Multiple Range Test (DMRT) to compare among treatments at the 5% significant level by using SPSS version 12 Software MS Excel 2007 and MATLAB 7.0 version also were used to analyze all drying data and draw the charts

10 mm

CHAPTER 4 RESULTS AND DISCUSSIONS

4.1 Selection of sweet potato cultivar for starch noodle production

4.1.1 Chemical compositions

The apparent amylose content and chemical compositions of four types of sweet potato starches; white skin and yellow-red flesh color (SP1_W_YR), purple skin and purple flesh color (SP2_P_P), purple skin and yellow flesh color (SP3_P_Y), and orange-purple skin and orange flesh color (SP4_OP_O) variety; and mung bean starch (MBS) are shown in Figure 4.1 and Table 4.1, respectively

Figure 4.1 Apparent amylose content (%) in sweet potato and mung bean starches

It could be seen that starches from all sweet potato varieties were high in apparent amylose content ranging from 28% to 34%, but much lower than that in mung bean starch which was 40% The results are in agreement with the previous reports, amylose content in sweet potato starch ranged from 8.5% to 37.4% depending on variety (Collado & Corke, 1997; Brabet et al, 1998; Hoover, 2001; Moorthy, 2002; Chen, 2003; Osundahunsi et al, 2003; Mais, 2008; Aprianita et al, 2009) Amylose content in starch was one of important factors influencing to starch pasting and the strength of starch gel due to its quick retrogradation, association and interaction to lipids and amylopectin to form helical complex giving strong gel structures (Jane & Chen, 1992 and Tan et al., 2009), and therefore effected on the application of starch in food processing Starches with high amylose content were desired for manufacture of starch noodles (Liu & Shen, 2007) However, Kim et al (1996) found that amylose content was no significant correlation to hardness of noodle and suggested that respecting to cooking quality of noodles, next to a amylose threshold level, other starch properties were more important than amylose content Tam et al (2004) found that

28.06d 29.34d 31.50c 34.52b

high amylose corn starch was not suitable for noodle making because it was not sufficiently gelatinized at atmosphere condition, leading to almost no amylose molecules release to participate into retrogradation process which fixed the noodle structure There was significant difference in amylose content in all types of sweet potato starches in which SP4_OP_O starch had the highest in amylose level (34%), the next rank belongs to SP1_W_YR starch with 31% While amylose content in the starches obtained from SP2_P_P and SP3_P_Y cultivars account for about 29% and 28% respectively

Table 4.1 Chemical compositions of sweet potato and mung bean starches (*)

SP1_W_YR 09.97a(0.50) 0.21b(0.021) 0.031c(0.0073) 0.137b(0.0238) SP2_P_P 10.14a(0.29) 0.15a(0.018) 0.084a(0.0027) 0.282d(0.0480) SP3_P_Y 10.03a(0.10) 0.21b(0.017) 0.061b(0.0026) 0.110b(0.0221) SP4_OP_O 09.79a(0.63) 0.23b(0.028) 0.039c(0.0079) 0.221c(0.0353) MBS 10.13a(0.09) 0.16a(0.019) 0.038c(0.0054) 0.053a(0.0055)

of starch paste and hinder the swelling of starch granule because of high rate formation of amylose-lipid complex (Kasemsuwan et al., 1998 and Chen, 2003) Nevertheless, according to previous researches (Dahle & Muenchow, 1968; Grzybowski & Donnelly,

1979 and Kim et al 1996), protein and lipid played important role in retention of amylose

in starch noodles during cooking, resulting in minimizing cooking loss The amount of protein, lipid and ash can be used as index of starch purity High this content expressed that the elimination of them presenting in sweet potato roots was less completed as for mung bean

4.1.2 Starch yield

The starch yield (%) of all sweet potato cultivars determined by following a method given

in 3.1 part ranged from 12 to 17% depending on cultivars (Figure 4.2) in which the SP1_W_YR variety showed the highest extracted starch with about 17.52%, following by SP2_P_P cultivar with 15.54% The isolated starch content from both SP3_P_Y and SP4_OP_O roots was not significant difference, holding 12.5% This may be explained that starch content in SP1_W_YR and SP2_P_P roots was much higher than that in SP3_P_Y and SP4_OP_O roots However, Rahman et al (2000) found that the roots with