application of vacuum drying technique in dehydrofreezing...
TRANSCRIPT
（千葉大学学位申請論文）
Application of Vacuum Drying Technique
in Dehydrofreezing EggpRant Pulp and
Quality Evaluation of the
Vacuum－dehydrofrozen Product
2008年1月
千葉大学大学院自然科学研究科
生物資源応用科学専攻環境園芸システム学
武 龍
Table of Contents
Abstract ofthe Dissertation．．＿．．＿．．．＿．．．＿＿＿＿＿．，．．．＿，＿．＿．＿．＿．＿．．．．．．．＿一＿一．．＿．．＿一＿＿．．iii
List of Figures…＿＿＿．＿．＿＿＿＿．．＿＿＿＿＿＿．．＿＿＿．．．．＿＿＿＿＿＿．＿＿＿＿．＿＿＿＿＿＿、．＿，v
List of Tables・＿．＿．＿．＿＿．＿＿．．．．＿．．．．＿．．．．＿．＿＿．＿＿＿＿．＿．＿．＿．．．．．．．＿．．．．，．．＿＿＿．＿．．．．．＿．viii
CHAPTER 1：Introduction．．．＿．．＿＿＿．．．．．＿＿．．．＿＿．．．．．．．．，＿＿．，．．．＿，．＿＿．．．．．．，1
CHAPTER 2：Mathematical M・deling・fVacuum Dゆg Pr・cess・fEg即1ant Pu正P．＿．＿7
2・1・Introduction．．．．－－．．．，響．．．．．．，．．引．．．．．．．．．．．．．．響．．凸．．．．．．．，，．．．．．．甲，，＿引．．．．．．．．．．．．．，‘．．．．．＿9
2．2．Material and Methods， ．．，，．＿＿．，＿．，＿．．＿．．．一．．．＿．一．一＿．．．．＿．＿＿＿＿＿g
2．3．Results and Discussion＿．＿．＿．．．1＿．＿．．＿．．．．＿．．．＿．．＿．．＿．．＿．．．．＿15
2．4Summary＿，＿＿＿．＿．．．．．．＿．．．．＿．．＿．＿＿．＿＿．＿．＿．＿．．．＿，＿＿30
CHAPTER 3：E茄ects of vacuum Drying Process on Quality Attributes of EggPlant
Pulp．．．．．．．．．．．．『凸．．．．．一『．．．、．．．．．．．引．．．．．．．．．．．．．．．．．．、．『．層層＿凸，．．．←．．．．，．．．引．．．引．．．．．．．．『，引欄一．．．，．．’＿．31
3。1．Introduction．．．．．．．．，．T－．．邑．の．．．，．－．．－－．．．．．．．噛．．．引．．．．．．．．．．耀．．．．．．．．．．．．．．邑，．．．．‘．、．．．．の，，．32
3．2。Matedal and Methods．．＿＿＿＿＿．＿．．＿＿＿．．．．＿．＿一．．＿＿＿．．。＿．．＿32
3．3．Results and Discussion．．＿．＿＿．．＿＿，．＿．．＿．．＿＿，＿．．．＿＿．．．＿＿＿＿35
3．4Summary．．．＿．．幽＿．．、、＿．．．．．．．．＿＿．＿．．．＿．．．．＿．．、．＿＿．＿．＿．＿＿．．＿．．50
CHAPTER 4：Electrical Irnpedance Spectroscopy Analysis of Eggplant Pu正p Tissue．．＿52
4．1．Introduction『．．．．『欄．．．．，．，．．．．．．『．．邑．．．．，，．．．，．．．，，．層．．腎．．．．．，，。．．、．．．．．．1，．．引．．．．．．，，申．層．．＿53
4．2．Material and Methods＿＿＿．＿．．．．．．．＿．＿．．，．．＿．＿＿．＿．．＿．．＿＿．．，．55
43．Results and Discussion．，．＿．．．．＿＿＿＿．．．、＿．＿．＿．．＿．．＿．．．＿．一．．．，＿＿．61
4．4．Summary＿＿．＿、．＿．．，＿．，．＿＿．．．＿，＿．，．＿，＿＿．．＿．＿．＿．＿．＿．，＿．．，，＿75
CHAPTER 5：Effects・f Dehydr・freezing Pr・cess。n S・me Physical Pr・pe丘ies。f
Eggplant Pulp＿＿＿．．＿＿．＿．．．．＿、．．．＿．．．．．＿．．．．．．，＿．．．．＿．。＿＿＿．．．．．．．．．＿．．．．77
5・1・Introduction・・、＿．．．＿，．．．．．．．，画，．引引．．．．．．．．．．．．，．．．，．、．．．．．．．．＿．．。．．引．．．．，．邑．．．．．．．．．．．．．．．78
52．Material and Methods．．．．＿，＿，．．＿．．．．．．一．＿＿＿一＿．．．．．．＿．＿一＿＿．．．－79
53．ResuIts and Discussion＿＿．．，．＿．＿＿＿．．＿．＿．＿．．．＿．．＿＿．＿＿．＿．＿．83
5．4．Summary．、＿、．＿．＿．．．．．＿．．．．＿．＿、．＿＿．．．．＿＿．．．＿．．＿．．．．。．＿＿．．．＿＿．97
CHAPTER 6：Conclusions．．＿．．＿＿＿＿．．．．．．．，＿＿．．．．．．．＿．＿．．．．．＿＿．，＿．＿．．＿98
References＿＿＿一．＿＿＿＿＿．＿＿＿＿＿．．＿＿．．＿．＿．＿．＿＿＿101
Acknowledgements．．．．＿．．＿．＿．．＿，．＿．．＿＿．．．．．．＿＿．．．．＿．．．．＿．．．．．．＿．＿＿＿．．．．．109
ABSTRACT OF THE DISSERTATION
ApPlicati。n。fv・・uum・Drying・T・。hniqu・in・D。hyd・。丘eez並g EggPl・nt・Pulp・nd Qu・lity・Ev・1u、ti。n。f
the Vacuum－dehydro丘ozen Product
by
Wu Long
Doctor of Philosophy in Advanced Bioresources Science
（i｝raduate School of Science and Teclmology， Chiba Universi｛y，2008
Professor Akio Tagawa， Chah・
In order to evaluate the practicability of dehydrofreezing technology fbr the long－term
preservation of perishable fbits and vegetables wlth high water content， experimental and
theoretical studies were carried out using eggplant iiuit as the research obj ect．
The vacuuln drying characteristics of eggp正ant pulp was discussed in detail using a
selfLdeveloped vacuum／convective drying system；amathematical mode蓋fbr describing
and predicting the whole vacuum drying process was proposed；the effects of vacuum
drying conditions on some quality attributes of the eggplant pulp samples were evaluated．
Electrical impedance spectroscopy（EIS）analysis was adopted in order to investigate the
changes in physiolo9孟cal status of eggPlant pulp tissue during dehydrofreezing Process．
Appropriate EIS analytical methodology fbr the present research was developed；several
commonly reported equivalent circuit models l量nking impedance spectmm to the
physiological status ofbiological tissue were introduced and compared；the effeet of drying
arld freeze－thaw processes on the eggpIant pulp tissue was discussed using eIectrical
impedance spectroscopy method． Effe cts of dehydrof士eezing process including par重ial
dehydration， f士eeze－thaw and rehydration treatments on the surface color， texture and
electrical impedance characteristics of the eggPlant pulp samples were investigated．
Experime打tal results showed that under the suggested process conditions， the
dehydrofreezing treatment effectively prevented the post－thawing drip loss；the
vacuum－dehydrof｝ozen eggplant samples showed a remarkable improvement in the
iii
quality－indicating physical properties including surface color and firmness over the
conventionally frozen ones． The changes in the quality attributes of the samples in
dehydrofreezing process were explained in terms of drying and freezing mechanisms
associated with the EIS analysis． Exper重mental results also indic航ed that the damage to the
cell structures caused by ice formation during freezing was still inevitab正e despite adopting
vacuum dehydration pretreatment and varying the f士eezing－thawing conditions．
κ砂woアぬ Dehydrofヒeezing；Drying mode1；Eggplant；E正ectrical impedance spectroscopy；
Freezing－thawing；Quality；Vacuum drying
iv
正ist of F量gures
Figure 2－1・Schematic diagram of vacuum drying system：1－vacuum pump，2－cold trap，
3－vacuum control un｛t，4－fbrced convection drying oven，5－glass desiccator，6－load
cell，7－supporting f士ame，8－wire netting sample hoIder，9－instrumentation amplifier，
10－data lo99er．．，．騨層＿．．．．．ゆ．＿．，．．．．．．，，．．．．『，．，．．．．．．．．．冒．．．．．』．．．．．騨．．．．引』．．．．．『．．．．』』．引．，．．．．．13
Figure 2－2． Changes in average moisture content of samples during vacuurn drying at 2．5
kPa and 30to 50°C．．．．．，．．．．．．．．．，．．．．一．．．．．．．．、．．．．．．．．．．．．，．．．．．＋．。．＿．．．．．．．．．．．．．．．．．．．．．．．．16
Figure 2－3． Changes in average moisture content of samples during vacuum drying at
different drying chamber pressures（a－30°C， b－50°C）＿＿．．．．＿．．．．．．＿．．＿．．．．．＿．．．．」7
Figure 2－4． Variations in drying rate during drying at 2．5 kPa and different temp eratures，，19
Figure 2－5， Compar五son of dryi且g rate between 2．5 kPa vacuum drying and hot－air drying
at the same drying temperature（a－30°C， b－50°C）．．＿．，．．．＿．．＿＿＿．．．．．．＿．＿＿．．．＿20
Figure 2－6． cilN4／dt against drying time（a）and dry basis moisture content（b）at 25kPa and
diffbrent temperatures＿．＿＿＿．＿．＿、．．＿＿．．．＿＿＿．．．．．．．．．＿＿＿、．．．．．．＿．、．．＿23
Figure 2－7． Comp arison of exp erimenta1 and predicted moisture ratio by drying model Eq．
（2－lI）at 2．5 kPa and various drying temperatures＿．＿．＿．．，．．．＿．．．．．．．．＿＿＿．．．．25
Figure 2－8． Cornparison of experilnental and predicted moisture ratio using diffbsion
equation in the MR． range of l to O．35．＿．．．．．．．，．．．．．．．＿．，，．，，．，．．，．．．．．．、，．．．．．．．．．．，．．．．＿28
V
Figure 2－9． Temperature dependence of effbctive moisture diffUsivity＿，．．＿＿．．．＿．．．29
F量gure 3－1． Changes in moisture content of samples during drying at 2．5 kPa and 30 to
50°．．．＿．．帽．欄引＿＿．．．一引引．＿．．＿．，．．．．．．．＿．．．．．．．．．．．．．．．．．、，虚．．．。帽．．．．．．．．．．．．．．．，．．．．＿＿．＿．、．．．36
Figure 3－2． Relationship between volumetric sh血kage and volume of removed water
during drying of eggP正ant．．．．．．．．．．．．．．，．．．，。．．，．．．．．．．．，．．，，。．，，．．．．．．．，．．．，，．．．．．．．．，．．．．．．．．38
Figure 3－3． Changes in volume shrinkage ratio（VIVo）with moisture content of samples
when drying at 2．5 kPa and different drying temperatures，＿．．．．＿．．＿．．．＿＿．．．．＿．．40
Figure 3－4． Relationship between volume shrinkage ratio（VIV。）and moisture content of
samples when drying at 50°C and different drying chamber pressures＿．．，＿＿＿41
Figure 3－5． Change in surface area with moi sture content of sample when drying at 2．5 kPa
and 30°C．．．．．．．引帽＿引．．匿．．引，層．．，．．．．．，．．，，．．．．．．．．．．，．t、－．．．－、．．．．．．．－．．響．、．．．．．＿，．．．．．．．凸凸．．．、．45
Figure 3－6． Total surface coior difference of the samples dried under different condition．．47
Figure 3－7． Changes in L零with the moisture content of the samples dried under different
conditions．．．．．．．．．．．．．．‘，．幽．．．．．幽．．．．引帽．．．．．，．．．．．．．．．．層欄．．．、．．．．．．．，．．．．．．．－．．、層．．．．＿．．＿＿＿．48
Figure 3－8． Force－Defomiation curves of the samples with different moisture content．＿．51
Figure 4－1． Schematic diagram of EIS measuring system，1－eggplant pu工p sample；2－clip
test probe；3－LCR tester；4－computer＿．、＿＿＿．＿＿＿＿＿＿、＿＿．．＿＿．．＿56
Figure 4－2． Equivalent circuit models used to analyze impedance characteristics of
eggplant pulp tissue，（a）－lumped model proposed by Hayden et al，（1969）；（b）－1umped
rnode正proposed by Zhang et a1．（1990）；（c）＆（d）－distributed models．．．．．．．．＿．，．＿．．58
Figure 4－3． Electrical impedance spectra of f士esh and processed eggplant pulp tissue．＿63
F量gure 4－4． Electrical impedance spectra of f士esh and processed eggplant pulp tissue
（Cole－Cole plot）。，．．．帽．．．．．．、．．．．．．．．．．．．，．＿、』．．．．．，．．．．、．．．．．＿，．．．．．．．．．．．，，．．．．。．．．．，．．．．．．．．64
Figure 4－5． Comparison between calcu重ated impedance data according to the equivalent
circui重models and experimental results丘om f｝esh eggplant pulp tissue． a：model（a）；
『b：model（b）；c：model（c）．．．＿＿，＿．．＿．，．＿＿＿．．．．．．．．．＿．．．．．．．．，＿．＿，＿＿＿＿．68
Figure 4－6． Relationship between IZl and moisture content of partia11y vacuum dried
eggplant pulp tissue（30°C，25kPa）．．．．．．．．．＿．．．＿．．．＿．．．．，，＿＿．．．＿．＿．＿．．．＿．．．71
Figure 5－1．Freezing curves of samples processed under different conditions．．．．．＿．．．．．＿85
Figure 5－2． Total color difference of samples processed under different conditions＿．．．．．．89
Figure 5－3， Comparison ofゴof samples processed under various conditions。．＿＿＿＿90
Figure 5－4， Force－Deformation pro丘les of fresh and processed samples．．．．．．．．．．＿＿＿＿91
Figure 5－5． Compar且soll of firmness of samples processed under various conditions．．＿＿92
Figure 5－6． Electrica正impedance spectrum of fresh and processed eggplant pulp tissue＿95
Figu、re 5－7． Cole－Cole plot of fヒesh and processed eggplant pulp tissue．．，．．．。．．＿＿．．，．．．、96
vii
List of Tables
Table 2－1．Five drylng models used to fit experimental data、，．，＿＿＿．＿．．．．＿，＿．．．．．．14
Table 2－2・Model coef猛cients and goodness of且t of different drying models fitted to
experimenta正data of 2．5kPa and 30°C．＿．，．．．．．．．．．．＿＿．．．＿．＿＿．．＿，．．．＿．．．＿＿24
Table 3－1． Results of linear regression fbr descr量bing drying shrinkage of eggPlant pulp
samples．．＿＿．．．．．．．＿．，．＿＿，＿．．＿、．，．＿．．．唱．＿＿＿＿＿＿，．．．＿．．＿．，＿＿44
Table 4－1． Model parameters and goodness of趾indexes of the circuit models丘tted to
impedance data of f士esh eggplant pulp tissue．．．，＿．．．．．＿．．．一＋＿、、．＿＿．．．．．．．＿＿．＿67
Table 4－2． Model parameters and goodness ’盾?@fit indexes of the circuit models fitted to
impedanoe data of partially dried eggplant pulp tissue＿．＿．＿＿．．＿．．＿．．．＿．．＿．73
Viii
CHAPTER l
Introduction
Freezing is one of the most popular means of long－term foods storage． By transforming
most ofthe 1孟quid water ex1sting in fbod materials into ice， fセeezing greatly slows down the
physical and biochemical processes involved in the deterioration of fbod and the growth
and reproduction of spoilage microorganisms． Generally， freezing preserves the taste，
texture and nutritional value of 」阯esh products better than any other preservation
technology． However， when using ffeezing technology to preserve peri shable fUits and
vegetables having high water contents， large ice crystals fbrmed in the tissue during
freezing process may cause irreversible damage to tender cell stru（rtures， and eventually
lead to drip loss and deterioration of overall quality ofthe products after thawing（George，
正993；R．eid，1983＆1990）． Theref（）re， in order to obtain h1gh－quality frozen fbod， besides
high－quality raw materials， processing， distribution and storage conditions must also be
carefUlly controlled．
Until now， a lot of studies have been carried out to improve the quality of f士ozen fbods
through changing process conditions． Dehydrefセeezing is a newly developed fbod f｝eezing
tec㎞ique in which high water content plant products are dehydrated to an appropriate
moisture level prior to f｝eez｛ng． Theoretically， the partial dehydration process befbre
丘eezing not only reduces the amount of water to be frozen but also makes cell structures
less susceptible to breakdown， consequ．ently， whell under appropriate f｝eezing／thawing
conditions the damage to cell structures caused by ice crystals and the post－thawing quality
1
degradation， such as deteriorative changes in apPearance and texture and loss of fi avor and
nutritional value， may be alleviated． In addition， by removi且g part of water in fbod
materials befbre ffeezing， the ref士igeration load can be lightened and the丘eezing cost as
wel董as packaging and distribution costs could be lowered（Li＆Su恥2002）． Presently，
some research on dehydrofreezing teohnoiogy fbr preserving various fUits and vegetables，
e．g， strawberry， green bean， pear， apple， tomato， muskmelon aIld kiwifruit etc． has been
reported；the dehydro丘ozen products were proved to be organoleptically acceptable，
（Agnelli et al．，2005；Biswal et aL，1991；Bolin＆Huxsoll，1993；Dermesonlouoglou et al、，
2007；Lazarides＆Mavroudis，1995；Maestrelli et a1，，2001；R．obbers et al．，1997；
Tregunno＆Goff，1996）．
Two mε酉or procedures， 孟．e． partial dehydration and ffeezing are invo玉ved in
dehydrofteezing technology， either ofthem greatly affects the quality ofthe final products・
According to the literature， osmotic dehydration， hot air dry萱ng and the combination of
them are co拠monly used water remova重techniques fbr dehydro丘eezing process．
Nevertheless， these rnethods might cause negative impacts on the quality of dehydroffozen
products；正eaching of natural acid， solute uptake and severe loss of firmness usually occur
during osmotic deh、ydration（B童swal et al．，1991；Bolin＆Huxsoll，1993；Lazarides＆
Mavroudis，1995；Raoult－Wack，1994；Robbers et aL，1997）；hot－air drylng at high
tempera加re may lead to color and texture degradation and Ioss of flavor and nutrient
sub stance as well （Lewicki＆Jakubczyk，2004；Orikasa et al・，2005（b）；Watson＆Harp er・
1988）．
2
Cornpared with the above mentioned dehydration techniques， vacuum drying possesses
some distinctive features， such as lower processing temperature and Iow－oxygen
processing environment etc． Presently， vacuum drying technology has been applied to food
processing， especially to handling those high－value products containing heat－sensitive
substances．（Alevalo－Pinedo＆Murr，2006＆2007；Bazyma et a1．，2006；Cui et a1．，2004；
Jaya＆Das，2003；Methakhup et al．，2005；Nakamura et al．，2005）． The application of
vacuum drying to the dehydro丘ozen fbod processing could minimize the impacts of the
present dehydration techniques on the product quality， and offer a new option for obtaining
high quality frozen ftUits and vegetables． Accordingly， it is very necessary to carry out
research on the vacuum drying characteristics of agricuitural products and to evaluate the
effects ofvacuum drying process on the their quality attributes。
Generally， the qua工ity of f士ozen fbods is largely influenced by freezing cond三tions． The
interaction between nucIeation and ice crystal growth during f士eezing， whloh are
determined by the rate of freezing， has a fatal effect on the size of the resulting ice crystals
and consequently the quality of the丘ozen products（AgnelIi，2002；Boonsumrej et al．，
2007；Heldman，1982＆1983；Zhu et al．，2004）． Therefbre， the other major process
involved in dehydrofrozen fbods processing一丘eezillg， must also be studied in depth
Until now， little study has been dene on the effects of f｝eezing－thawing oonditions on the
quality of dehydroftozen products， The research into various freezing－thawing methods
and conditions may help fUrther improve the practioability of dehydrof｝eezing techno正ogy
Electrical impedance spectroscopy（EIS）measures the dielectric properties of a medium as
3
afしmction of ffequency， It is based on the iIlteraction of an external electric f1eld with the
electric dipole moment of materlals（Grimnes＆Martinsen，2000）EIS has been
extensively used to characterize properties of solid materials（Pan et aL，2003；Prabakar＆
Mallika巾n Rao，2007；Takashima＆Schwan，1965）as well as estimate the physiologicaI
state of various biological tissues（Cole，1932；Damez et a1．，2007；Zywica et al．，2005）．
Compared with other techniques of physiologieal investigation， EIS measurement is very
simple and easy to be conducted． Until now， most EIS studies of plant tissue have been
fbcused on the natural physiological properties of the tissue as well as related issues suoh
as rip ening， aging and fro st hardiness etc，（Harker＆Dunlop，1994；Harker＆Maindonald，
1994；Zhang et al．，1990；Zhang＆Willison，1991）， there are only few reports available
detailing the impact of artifioial processes on the electr量cal impedance characteristics of
ftUits and vegetables． The use of EIS analysis could provide a new approach to the
evaluation of the quality of processed agrioultural products due to its simplic量ty and
effectiv童ty．
Eggplant（Solanum melonge刀αvα1：eseulenta）is an important market vegetable of Asian
and Mediterranean countries having a very limited shelf lifb fbr f㌃eshness． Storage of
eggplant is generally less than 14 days（at optimum storage temperature 10－12°C and
reIative humidity 90－95％）as visual and sensory qua1三ties deteriorate rapidly The use of
conventiona1丘eezing technique is obviously inapPropriate in this case because the very
high water content（about 94％， w．b，）of eggplant ftUit may cause severe quality loss after
thawing． Since the vacuum－dehydrofreezing method possesses potential advantages in
4
improvement．of the quality of fセozen products， and so far the application of
dehydrofヒeezing technique to preserving eggPlant fhユit has not been reported yet，
theoretical and experimental studies on the dehydrofreezing processing of f士esh－cut
eggplant pulp were canied out to estilnate the practicability of vacuum－dehydrofヒeezing
fbr the long－term preservation of eggplant and similar perishable plant products．
The objectives ofthis research are as fbllows：
Firstly， to investigate the vacuum drying characteristics of eggplant pulp and develop
suitable mathematical model fbr describing and predicting the vacuum drying process．
Secondly， to evaluate the impacts of vacuum drying process on the quality attributes of
eggplant pulp， and accord1ngly choose an optimal vacuum drying condition fbr the
dehydrofreezing Process．
Thirdly， to establish appropriate electrical impedance（EIS）analysis inethod and select an
appropriate electrical circuit model describing the physiological stmcture of eggplant tissue
in order that the changes in physioIogical status of the tissue subjected to drying and
丘eezing－thawing treatments can『be correctly evaluated at celIular IeveI according to EIS
analytiCal reSUltS．
Finally， to investigate the effects of the whole vacuum－dehydrof士eezing process including
different f卜eezing－thawing methods and conditions on the quality and impedance
charaoteristics of eggPlant pulp．
This d萱ssertation consists of six chapters． In Chapter l， the research background about
dehydrof㌃eezing technology， progress of present research and the obj ectives of the present
5
w・rk are b「iefly int・・duced・In・rder t。 integ・ate vacuum drying technique int。 the
dehydrofreezing Process， the vacuum drying characteristics and the efflects of vac1」uln
drying on the quality of eggPIant pulp are discussed in Chapters 2 and 3． Chapter 4 reviews
the use of electrical impedance spectroscopy technology fbr b童010gical tissue， and the
changes in the physiological status of eggPlant pulp tissue afヒer the partial dehydration and
freeze－thaw treatments were discussed by means of the EIS analysis． In Chapter 5，
ass・ciated with the previ・us studies， the effects・f dehydr・freezing pr・cess including
partial dehydrati・n・ freezing－thawing and rehydrati・n pr。cesses・n the quality・f eggplant
pulp are discussed． In Chapter 6， the research works on the dehydroffeezing technology are
briefly summarized and the major conclusions drawn from the present studies are
presented．
6
CHAPTER 2
Mathematical Modeling of Vacuum】）rylng Process of
Eggplant Pulp
Absかact’The vacuum drying characteristics of eggplant were investigated． Drying
experiments were carried out at vacuum ohamber pressures of 2．5，5and 10 kP礼and
drying temperature ranging f士om 30 to 50°C． The effects of drying pressure and
temperature on the rate of vacuum dry萱ng were evaluated． A mathematical mode正fbr
describing the vacuum drying process of eggplant sample was developed and compared
with five commonly used drying models． the effective moisture diffUsivity and activation
energy were calculated using an infinite series solution of Fick’s diffUsion equation．
Results showed that increasing drying temperature accelerated the vacuum drying process，
while drying chamber pressure did not show significant effect on the drying process within
the temperature range investigated． The goodness of fit tests indicated that the proposed
drying model gave the best fit to the experimental results among all tested models． The
temperature dependence of the effective moisture dif比sivity fbr the vacuum drying of the
eggplant samples was satisfactorily described by an Arrhenius－type relationship．
Keyivords’Drying characteristics；Drying model；Effective moisture diffbsivity；Eggplant；
Vacuum dτying
7
A恥瓦LM輪喚楓
P
PR〆
RMSE
t
T
Wd
θ
2
NOMENCLATURE
area（m2）
effective moisture dif壬Usivity（m2 s－1）
activation energy（kJ kgi）
sampIe thickness（m）
moisture content（d．b，， decima1）
initial moisture content（d．b．， decimal）
equilibrium moisture content（d．b．， decimal）
molsture rat10
Pressure（kPa）
mean relative deviation（％）
gas constant（0．462 kJ kg－l K1）
coeffricient of determination
drying rate（kg m－2 h帽1）
rOOt mean SqUare errOr
time（h）
temp erature（K）
weight of dry matter（kg）
time（S）
reduced chi－square
8
2．1．1皿trod皿ction
Drying is one of the most important methods of long－tem fbod preservation． The removal
of moisture from the fbod materials slows down the action of enzymes and m量nimizes
many of the moisture mediated deteriorative reactions， Compared with convent圭onal
convective drying， vacuum drying has some distinctive characteristics such as higher
drying rate， lower drying temperature and lower oxygen concentration in drying chamber
etc， these characteristics may help to improve the quality and nutritive value of dried
products． Presently， vacuum dry董ng has been widely applied to dry various fbod materia重s，
the vacuum drying kinetics of many fruits and vegetables has been investigated and the
effects of vacuum drying conditions on the drying Process and the qualities of dried
products has been evaluated（Alevalo－P1nedo＆Murr，2006＆2007；Bazyma et al．，2006；
Cui et al．，2004；Jaya＆Das，2003；Methakhup et al．，2005）．
Eggplant（Solanum melongena var esculentのis a very perishable popular vegetab重e， In
order to evaluate the practicability of vacuum drying fbr improving the quality of dried
eggplant， it is necessa ry to carry out research on the vacuum drying characteristics of
eggplant fUit． The obj ectives of this study were to量nvestigate the vacuurn drying
characteristics of the eggplant samples， to evaluate the effect of drying conditions on the
vacuumL drying Process， and to deve正op a suitabIe drying model fbr describing the drying
P「ocess．
2．2．Material and methods
Experim ental setup
9
Aschematic diagr＆m of the experimental vacuum drylng system is shown in Fig．2－1．The
system primarily consists of an oil rotary vacuum pump（TSW－300， SArO VAC， Japan）， a
vacuum control unit（NVC－2000L， Tokyo Rikakikai， Japan）to obtain various processing
pressures in the vacuum drying chamber（a glass desiccator）and a fbrced convection
drying oven（1）0600FへAS ONE， Japan）to maintain desired drying temperatures． A data
acquisition system composed of a load cell（］LTS－50GA， KYOWA， Japan）which was fixed
on a supporting f士ame， a wh’e netting sample holder suspended f士om the load ce11， an
instrumentation amplifier（WGA－710A， KYOWA， Japan）and a data logger（KEYENCE，
NR－1000， Japan）was used to on－line monitor and record the changes in sample weight
during drying． Hot－air drying mns at 30 to 50°C and atmospheric pressure were also
conduoted in the same glass desiccator with the top lid removed fbr oomparison．
Sa〃Ptepreparation
Fresh eggplants（cultivated in K。chi PrefectUre Japan， cultivar：unl（no㎜）were purchased
丘om a Iocal market and stored at 10°C befbre drying experiments started， the storage time
was not more than 12h in this study The central part of each eggplant fiuit was cut into one
rectangular－shaped pulp block of45×25x20mm fbr the experiments． The moisture content
of the f士esh eggplant pulp was measu．red according to the vacuum oven method（AOAC，
1995），in which about l Og fresh pulp was cut into very small pieces， weighed， and dried at
70°Cand 2．5kPa fbr 12h in the vacuum drying system， the dried pulp was weighted again
and its moisture content was calculated． The average value of 20 repeated measurements
was regarded as the initial moisture content of the pulp samples and used f｛〕r the
10
calculation of molsture content ofthe dried samples．
Experimentalprocedure
The vacuum drying chamber was preheated fbr 12h befbre the experiments started to
obtain stable drying temperature． Drying experiments were conducted in the drying
ohamber at temperatures ranging丘om 30 to 50°C， and pressures of 2．5，5，10 kPa as we重l
as atmospheric pressure respective正y． One pulp sample was placed on the wire netting
basket and dried in each run， its weight was continuously recorded at intervals of 5min
using the data acquisition system throughout the drying process． It was considered that the
sample reached the equiIibrium m．oisture content（EMC）of drying when the reading of
weight remained the same fbr lh．
Data analysis
The average mo三sture colltent of each sample during drying was calculated using the initial
moisture content value and its weight recorded by the data acquisition system（Moisture
distribution in the sample was considered to be unifbrm in this study）． Moisture ratio（MR）
ofthe sample was detemlined by the fbllowing equation：
（M－M，） （2－1） MR＝
（M。－M，）
where M is the average moisture content of a sample at any time of drying， Mo and Me
stand fbr the average initial and equilibriumエnoisture content respectively．
drying curves（MR vs． time）were plotted and fitted by five commonly reported empiricaI
or theoretical drying models presented in Table 2－1 and a new mathematicai model
developed in the present study；model coefficients were calculated using OriginPro 75
11
software（Origin上ab CorP）・The goodness of fit was evaluated by the coefficient of
deterrn｛nation（R2）， the root mean square error（RMSE）， the reduced chi－square（Z2）and
mean relative dev量ation modulus（P’）defined by the equation below：
P』1響守 （2－2）
where Y已xp，i is the experimental result of the investigated variable， Ypre，i is the predicted
value from various mathematical models， N is the number of observations（Chen＆Morey，
1989；Jena＆Das，2007；Madamba et al，，1996；Sac玉lik＆Elicin，2006）．
The best model describing the vacuum drying process of the eggplant samples was chosen
as the one with the highest R2 and the least RMSE， X2 and P’．
Comparisons between meanLs were perfbrmed in SPSS l2．O software（SPSS Inc．）using
Duncan’s multiple range tests at a significance Ievel of O．05．
12
Fig2－1． Schematic diagram of vacuum drying system：1 －vacuum pump，2－cold trap，
3－vacuum control unit，4－fbrced convection drying oven，5－glass desiccator，6－load
cell，7－supperting frame，8－wire netting sample holder，9－instrumentation amplifier，
10－data lo99er
13
三
（卜8べ
@
@
@
@
@
@
@
@
@
@
@
@
@
@
@
@
@
@
（年）曾器些 丁お＋㌔凸Σ
i曳↓受。↓自
@（M）量。闘告≧
一N箇寸し〔
2．3．ReSlllts and discussion
The average initial moisture content of the pulp samples was determined as 94．00％（wet
basis， Nニ20， stand＆rd devlation：0．52％）according to the vacuum oven method．
碓c∫げvσcπ配撹吻’”8伽4’ガoπ50π吻’π8Pアo‘8∬
Afヒer vacuum drying， the moisture content of the eggplant pulp samples was reduced from
ainitial value of l5．67 to less than O，2 kg water／kg dry rnatter． The effects of drying
temperature and pressure on the vacuum drying process are shown in Fig2－2 alld Fig2－3
（a，b）， in which drying curves（moisture content in dry basis vs． time）under different
drying conditions were plotted． From Fig．2－2， the drying time needed to reach the EMC
was shortened notablely with an increase in drying temperature due to a larger driving
fbrce fbr heat and mass transfer at higher drying temperature． Fig2－3 showed that drying
chamber pressure ranging f㌃om 2．5 to 10 1〈Pa did not affect the drying process as strongly
as the drying temperature did． For the present vacuum drying conditions， the effect of
drying chamber pressure on the drying process was not significant． According to the
reports of ArevaIo－Plnedo＆Murr（2006＆2007）fbr carrot and pumpkin， Cui et al．（2004）
fbr carrot， Gri＆Prasad（2007）fbr mushroom， and Methakhup et a1．（2005）fbr Indian
gooseberry， drying pressure had sorne effe ct on the drying process， the drying time was
reduced by decreasing drying Pressure． The differentiation betWeen the resu王ts ofth圭s study
and the literature couid be attributed to the different processing cond量tions as well as
differe nt degrees of boiling Point elevation caused by various plasma concentrations of the
tested materials，
15
（罵量9書．O．巳
0月JnU＝JnU211ーエ
嘱紮 諮甑、
P 2．5 kPa
x300C ◇40°C
・50°C
0 5 10 15 20 25
1）rying time（h）
Fig．2－2． Changes in average moisture content of samples du血g vacuum drying at 2．5 kPa
and 30 to 50°C
16
20 T＝30°C り 8⇔ ×2．5kPa
0
0 5 10 15 20 25
DIy血g time（h）
a
T＝50°C ” 20
1葺・5 謂a
葵・・ …ld｝a
麗5 匿 0
0 5 10 15 20
町画gtime（h）
b
Fig2－3． Changes in average moisture content of samples dur三ng vacuum drying at different
drying chaml）er pressures（a－30°C， b－50°C）
17
Rate oj「yσc距脚吻’π8
According to Toei（1975）， drying rate（Rd）is defined as：
Wd dM Rd＝一 （2－3）
A dt
where Rd is the drying rate（kg・m’2・h－1）， Wd is the weight of dry matter of the sample（kg），
Ais the drying area ofthe sample（m2）， M is the volume－averaged moisture content， t is the
drying time（h）． Nakamura et al．（2005）and Or董kasa et al．（2005， a＆b）reported that the
relationship between surface area and dry basis moisture content of eggplant sample could
be approximately expressed by the linear equation：
A＝αM＋β （2－4）
Substituting Eq（2－4）into Eq（2－3）gives；
IR，，i＝一謙βd詳 （2－5）
The drying rate of the samples under various drying conditions was calculated using Eq，
（2－5）and plotted against the dry basis moisture content． Fig．2－4 shows the changes in
drying rate as a fUnction of rnoisture content at 2．5 kPa and various dry童ng temperatures，
simi正ar trends were observed at other drying chamber pressures． From the figure， the
drying temperature significantly affected the drying rate， drying at higher temperature was
apparently faster than at lower temperature． The results also indicated that the drying rates
of the samples decreased with decreasing moisture content throughout the drying Processes，
that is to say， vacuum drying of the eggPlant samples under the investigated drying
conditions took p1ace in the fhlling rate period only．
18
（聡ぺ目身）
8聾bΩε昏自
β話420nUnUnUnU
0
P 2．5kPa
×30°C ◇400C ・500C
・口
5 10 15 20
Moisture content（d．b． decimal）
Fig．2－4． Variat ions i n dryi血g rate during drying at 2．5 kPa and different temperatures
19
0．8
3肖へ0．6巽1‘
．ぎ㌔o．4
言豊。．2
①
8
0（聡．㌔．⇔の巴
ε雪。。εあ占
0
0 5 10 15 20
MoiStUre content（曲． decimal）
a
△2．5kPa
50°C ・Atomosphe盛c p騰ssure
鰯麟繍識鈴∴△
0 5 10 15 20
Moisture con亡ent（d．b． de cimal）
b
Fig．2－5． Comparison of drying rate between 2．5 kPa vacuum drying and hot－air dry｛ng at
the same drying temperature（a－30°C， b－50°C）
20
As can be seen in Fig．2－5（a， b）， the warming－up and constant rate period（ffom ini重lal
moisture content to about 5 kg water／kg dry matter）as existed in the hot－air drying process
were not observed in the vacuum drying at 30 to 50°C． For the investigated drying
temperature range， vacuum drying pro cess was apparently faster than hot－air drying at the
same temperature， but the diffbrence between the drying rate of the vacuum and hot－air
drying diminished quiokly with the increasing drying temperature．
Modelling vacuum dryingpアocess ofeggP伽t sample
Aperfe ct linear relationship between dM7dt and drying time t were observed fbr al重
vacuum drying conditions investigated， as shown in Fig．2－6（a）． The relationship between
dM／dt and time（t）can be expressed as：
dM ＝a’t＋b’ （2－6）
dt
By integrating Eq．（2－6）with respect to time（t）using the initial condition M＝Mo at t＝0，
f（）110wing equation can be obtained：
M－9’ t2＋b’t＋IM［c） （2－7）
2
Eq，（2－7）may be fUrther transfbrmed into a more general fbrm as fbIlows：
M－M e ＝at2＋bt＋1 （2－8） MR＝
M。－M。
Where a and b are coefHcients to be dete㎜ined empirically．
Eq，（2－8）could be used to describe the changes in MR of the eggPlant samples with time
during vacuum drying Process． The equation was first proposed by Wang＆Singh（1978）
fbr modelling the drying Process of rough rice（see Table 2－1）．
Generally， empirical drying models as model I to 41isted in Table 2－1 were developed
21
based on the relationship between the dry量ng rate and moisture content of sampIe， thus
Eq．（2－8）， which was derived fセom dM／dt亡f（t）， was considered to be a highly empirical
model with Iimited range of apPlication． When plo廿ing the dM／dt against dry bas三s
moisture content of the samples， as shown in Fig．2－6（b）， it was observed that the curves
can be approximately described by a quadratic fUnction， as fbllows，
dM ＝aM2＋bM＋c （2－9）
dt
By integrating Eq。（2－9）with the initial conditions， fbllowing equations can、 be obtain、ed：
1
M＝ ＋cl （2－10） a’exp（－kt）＋b，
1
0r MR＝ ＋c （2－11）
aexp（－kt）＋b
The appropriateness of Eq．（2－ll）fbr modeling the vacuum drying process was examined
by fitting it to the experimental data and compared with the drying models in Table 2－1．
Drying curves（MR vs． time）under various drying conditions were plotted and fltted with
Eq．（2－11）and other five common重y－used drying models． Table 2－2 shows the calculated
results ofthe model coefficients and goodness of f三t ofthe models f三tted to the drying curve
at 2．5kPa and 30°C． The results indicated that＆mong all drying models tested， the model
described by Eq．（2－11）had the best goodness of fit indexes（i，e．， highest R2 and Iowest
つ
RMSE， X－ ≠獅п@P’， as shown in Table 2－2）． A comparison between the drying curves（MR
vs． t）at 2．5 kPa and various drying temperatures and the data predicted by Eq．（2－11）is
presented in Fig．2－7， the equation was proved to be adequate to model the whole vacuum
drying prooess ofthe eggplant samples，
22
3．0
2．5
も2．①＼Ill 1・5
vLO O．5
0．0
o 5
P 2．5kPa
10
x30°C
。40°C
・50°C
15 20
1）】rying time（h）
a
脚ミ＝唱
4
3
2
1
0
o 2
P 2．5kPa
COO5口COO4φCOO3X Xロ◇x
ロOXロφx
◇X
。叙
ロ X
・鎚
ロ◇X 口φX 口◇X
露
ロ受 ロ◇X ・鍬
@翻
@
4 6 8 10 12 14 16
MoiSture centent（d．b． decima1）
b
Fig．2－6． dM／dt against drying time（a）and dry basis moisture content（b）at 2．5kP a and
different temperatUres
23
寸因
δ．い
トQQ．°Q寸
一eh．い富．ON
寸ひ．ONトひ．トい
卜OひOO．Oい＝OO．O
OOOOO．Oひ一〇〇〇．OひNOOO．O寸200d
卜OO．O舘O．OoonO．0
90dこO．Oひ8．O
Oひひひ．ON卜゜oひ．〇一ひひひdいoQひひdいトひひd800ひ．O
一 国＋お＋㌔11肖Σ
@（轟尋）費。11勇
・一一ぐ刈fヂ1寸い 5
（承）廷
N
口の屋
髭
屋善び。［・℃。芝
dZ
1，2
1
0．8
O．6
0．4
0．2
0
0 4
2．5kPa
300C P貰℃dicted
一一400C P redicted
・… 50 OC P redicted
×30°CExpe血ienta1
◇ 400C Expe】血nenta貰
鰍口5・・CExpe血ental 畿
8 12 16 20 24
Drying time（h）
Fig．2－7． Comparison of experimental and predicted moisture ratio by drying model Eq，
（2－11）at 2．5 kli’a and various drying temperatUres
25
Caleulation Ofbasic diying parameters
Drying models based on the diflMsion theory sometimes failed to accurately predict the
drying Process of some fbodstuff忌due to the complexity offbod drying kinetics as we韮as
the incompliance with the basic assumptions fbr using the modeL however， the diffhsion
equation is still extensively used fbr the ev乱luation of the fUndamental parameters of
drying Process．
Since the drying of the eggP互ant samples took place in the falling rate period or江y， the
fbllowing infinite series solution of Rok’s second law of diffUsio珠which was developed
f（）rparticles with slab geometry assuming unidirectional moisture movement without
volume change， constant diffbsivity and temlperature， unifbrm initial moisture distribution，
was used fbr the calculation of the effe ctive Inoisture diffUsivity of the samples during
drying．
峠書（2nl1）・exp〔一（2n＋1善π2D誼θ〕（2－12）
where MR is the moisture ratio， D己任is the effbctive moisture difiユsivity（m2／s）， L is the
slab thiGkness（m）， the ha夏f sl＆b thickness is used when evaporation occurs on both sides of
the slab， andθis the drying time（s）（Crank，1975；Tutuncu＆Labuza，1996）．
In this study， notable deviations fセom the experimental results were observed whell using
the drying model based on the diffUsien theory to describe the latter part（MRく035）of the
drying process． Consequently， the effective moisture difliisivity of the eggplant samples in
the MR range of l to O．35 was calculated by fitting Eq．（2－12）to the MR data between l
and O．35 f士om var三〇us drying conditions． Fig2－8 showed that the infinite series soiution of
26
Fick’s second law of diffUsion agreed fairly satisfactorily with the experimental results． For
the tested MR range， the values of the Deff of the samples dried at 2．5kPa and 30，40 and
50°Cwere calculated to be 1，653x1σ9，2353x10－g and 3．417x10“g m2／s respectively， which
was significantly higher than those of the hot－air dried samples（varying丘om 1．005x10響9
to 2．086x1σ9）at the same temperature， as shown in Fig．2－9．
Under thLe investigated experimental conditions， the drying chamber pressure showed no
significant effect on the Deff of the samples， whiIe increasing drying temperature互ed to an
apparent increase in the effective moisture diffiJsivity． The temperatUre dependence of Deff
was examined by the following Arrhenius－type equation（Madamba et al．，1996；Pifiaga et
al．，1984；Tagawa et al．，2003）l
D・fiFD・exp〔一翻 （2－13）
where E。 is refbrred to as activation energy fbr moisture diffhsion（kJ）， T is the absolute
temperature（K）、 A typlcal Arrhenius－type relationship between the D¢ff and drying
temperature cou藍d be observed by pIotting Deff with respect to the reciprocal of absolute
temperature in a semi－logarithmic graph， the activation energy fbr moisture dif壬bsion，
which was obtained from the slopes ofthe lines fitted to the data in Fig．2－9， was fbund to
be 1640 and 1652 kJ／kg fbr the vacuum drying and hot－air drying under atmospheric
pressure respectively． The temperatUre dependence of the effeotive moisture diffhsivity f（〕r
the vacuum drying ofthe eggPlant sampIes could be described by the fbIIowing equation：
D 一・㌔xp〔－351°’1〕（2－14）
27
勇
10
1
30°C2．5 kPa Experimental
Pedicte曲y Eq．（11）
0．1
0 5 10 15 20 25 30
Dゆg伽e（x103s）
Fig2－8． Comparison of experirnenta1 and predioted moistUre ratio using difiiision equation
inthe］Y［R range oflto O．35
28
1．OOE－08
iiS 1．00E－09
1．00E－10
0．003
△2．5 kPa
◆Atm o spheric pre ssure
0．0031 0．①032 0，0033 0．0034
1／T（K－1）
Fig．2－9． Temperature dependence of effective moisture dif1fUsivity
2．4．Summary
Vacuum drying of the eggplant samples under the investigated drying conditlens took place
in the falling rate period． Drying chamber pressure had statistically insignificant effect on
the vacuum drying Process， increasing drying temperature shortened the drying Process
notably． A mathematical model was developed to describe the vacuum drying of eggplant
based on the relationship between dM／dt and moisture content， the model showedもetter fit
to the experimental data than five commonly reported drying models for the present
vacuum drying conditions． The effe ctive moisture dif磁sivity of the eggplant samples
within a moisture ratio range f｝om l to O．35 was calculated according to the infinite series
solution of Fick’s second正aw of diffUsion， its temperature dependence was described
satisfactorily by an Aエrhenius－type equation， the activation energy fbr moisture diffhsion
was 1640 kJ／kg．
3⑪
CHAPTER 3
Effects of Vacuum Drying Process on Quality Attributes of
Eggplant Pulp
Absかact’The effects of vacuum drying conditions on the qua蓋ity－indicating indexes
including dimensions， su血ce color and texture of dried eggplant pulp sample were
evaluated， The volumetric shrinkage， darkening and textural change of the samples were
estirnated after vacuum drying at 30 to 50°C and 2．5 to 10 kPa． The relationship between
the quality attributes and the moisture content of the dried samples was investigated．
Experimental results showed that drying shrinkage of the samples was independent of
drying temperature， and affected only by the drying chamber pressure under the present
experimenta正conditions． The relationship between the drying shrinkage rat量o as well as the
surface area of the dried samples and their moistuエe content could be well described by
linear equation． Drying at higher temperature and pressure led to more severe color change，
the browning of the samples during drying could be alleviated under vacuum dryin、g
conditions． Proper measurement and analysis methods fbr evaluating the texture of
eggPlant pulp were established． The drying c。nditi・ns had n・statistically significant effect
on the firmness of the dried samples at the same moisture level・The firmness of the dried
sample was closely related to its moisture content． An optimal vacuum drying condition
was suggested for the eggplant pulp samples in terrns of product qua1ity．
Keyivortts’ Drying shrinkage；EggPlant；Quality attributes；Surface color；Texture；vacuum
drying
31
3、1．1皿troduction
Although dry童ng Processing effectively extends the shelf life of agricultural products，玉oss
of sensory and nutritive qualities is considered inevitable during conventional drying
process due to the undesirable textural and bioohemioal changes（Watson＆Harpeち玉988）．
According to the literature， traditional hot－air drying may lead to severe deterioration of the
quality of dried product， such as shrinkage， darkening and textural change（hardening）as
well as正o ss of flavor and nutrient substance etc．， due to the physical and chemical ohanges
happen at high temperature during drying process．（Lewicki＆Ja㎞bczyk，2004；Orikasa et
al．，2005， b）． In contrast， vacuum drying can be conducted at lower temperature alld
oxygen concentration， therefbre those deteriorative changes in quality existing in
traditional convective drying may be substantially reduced．
The obj ectives of this study were to evaluate the impact of vacuum drying process under
different operating conditions on the quality attributes including dimensions， surface color
and texture of eggplant pulp samples， and select an optimal process condition fbr the
vacuum drying of eggplant accord量ngly．
3．2．Materi段l and methods
The same vacuum drying system as described in Chapter 2 was adopted in this study．
s伽np le」prelワaration
Fresh eggPlants（cultivated in Kochi PrefectUre Japan， cultivar：unknown）were purchased
from a local market and stored at 10°C before the experiments started， the storage tlme was
not more than 12h in this stUdy， The central part of each eggPlant ftuit was cut into one
32
reotangular－shaped block of 45 x25x20mm fbr drying treatment The initial皿oisture
content of the pulp blocks was predetermined as 94．00％（wet basis， N・ ・20， standard
deviationl O，52％）according to the vacuum oven method（see Chapter 2）・
Di ying process
The vacuum drying chamber was preheated fbr 12h before the experiments started in order
to obtain stable dry量ng temperature． Fresh eggplant samples were dried fbr l to 15h
respectively at 30 to 50°C，25，5and lO kPa in the drying chamber． SampIes which were
convective dried（under atmospherio pressure）at the same temperatures fbr the salne
duration were used as control． One eggplant sample was processed in eaoh drying run， its
weight was manually measured using a electronic balance（GX2000， A＆D， Japan）after
drying fbr the calculatiorl ofthe moisture content
euality evaluation
Approximate volume and surface area of each f卜esh and dried sample were calculated from
its dimensions， which were manually measured befbre and after drying process． In
preliminary stUdy， comparisons between the calculated and measured volume of the dried
samples using liquid displacement method（Maskan，2001；Orikasa et aユ．，2005（a）；Zogzas
et a1．，1994）showed that the calculated results were highly reliable， similar results were
also reported by Ratti（1994）and Orikasa et al・（2005， b）・
Experimental measurements fbr the Surface color and texture of the ffesh and dried
samples were performed at room temp erature．
In this study， the L＊ iLightness）， a＊（redness－greenness）and b＊（yell・wness－blueness）
33
indexes of the C工ELAB colorimetric system were used to describe the surface color of
sample（Abbott，1999）． The color measurement was conducted using a chroma meter
（CR－200b， MrNO皿へJapan）with D651ight source at 3 different spots on each sample
befbre and after drying， the change of its surface eolor， which was refbrred to as total color
difference（厘）， was calculated using the measured data・fL零， a＊and bホacc・rding t・the
■ サ
fbllowlng equatlon：
狙一（AIL＊）2＋（△aり2＋（△bり2 （3－1）
Puncture test was carried out to evaluate the changes in the texture of the samples caused
by drying process． The test was perfbrmed using a creep meter（RE2－3305S， YAMADEN，
Japan）with a 200－N load cell and a 3mm diameter flat－ended cy至indrical probe at a
deformation speed of l mm／s． In each test the deformation of the sample（i．e．， the depth the
probe penetrated to）and the stress when the defbrmation occurred were automatically
recorded by a computer cormected to the creep meter． The firmness ofthe eggplant sampies，
which is usually defined as the ratio of mpture fbrce to defbrmation fbr fiuits and
vegetab les（B orwankar， 1992；B。ume，1965；Jackrn an et al．，1990；Steinmetz，1996；Varela
et a1．，2007）， were detemined according to the fbrce－deformation（F－D）profiles obtained
from the tests．
Data anal rsis
Experiments were perfbrmed in triplicate fbr each drying condition・all results were
presented as mean土SE． Comparisons between means were perfbrmed using Dunoan’s
multiple range tests at a significance level ofO．05 in SPSS 12，0 software（SPSS Inc．）、
34
3．3脅Resulits and discussion
In order to investigate the drying shrinkage of the samples， drying experiments with
different drying durations were carried out under various drying conditions、 According to
previous study， inoreasing drying temper＆ture accelerated the vacuum drying process，
whi重e drying chamber pressure did not show significant effect on the drying process under
the present experimental conditions． Simi正ar results were also observed in this study， after
drying fbr l to 15h， the moisture content of the samples decreased to different levels． the
drying curves（dry basis moisture content against drying time）of the sarnples at 2．5 kPa
and 30 to 50°C were presented in Fig，3－1． The changes in the quality attributes inc互uding
volumetric shrinkage， surface color and firmness changes of the samples after drying were
evaluated．
35
・2．5 kP
30°C
40°C
50°C
0 5 血U
2
11 ーユ
＝」 0
（罵巌。。で．ρ．巳
塩㊤鴛80』ξ紹畠≧
10 15 20
Drying tim e（h）
50
Fig．3－1．Chang es in rnoistUre content of samples during drying at 2．5 kl）a and 30 to 50°C
36
Volumem’c shrinkage
Drying shrinkage af壬bcts not only the product quality l）ut also th・e dτying Process and
rehydration capability of the dried food material s（Karathanos et aL，1993；Maskan・2001；
Mcminn＆Magee，1997（a＆b））． Rattl（1994）suggested that changes in the dimensions of
dried sample were independent of drying conditions but dependent on the geometric shape
and type of fbodstuff， Souma et al．（2004）reported that the hot－air drying shrinkage of
eggp正ant was very remarkalble and the reduction iロsample volume was larger than the
volume of removed water due to the high porosity of eggplant pulp， similar tendency was
also observed in this study， as shown in Fig3－2．
37
1．0
50°C Hot a ir drying
ξo・8
N 婁o．6
1・・4 ▲▲▲P
50．2 ▲ ▲
0．0
0 0．2 0．4 0．6 0．8 1
（Vo－V）／Ve
Fig．3－2． Relationship betWeen volumetric shrinkage and volume ofremoved water during
drying of eggP lant
38
After drying under various conditions for different time， the volume and surface area of the
dried samples were calculated． Fig．3－3 presents the relationship between the volume
shrinkage ratio（VIVo）and the moisture content of the samples dried at 2．5 kPa and various
drying temperatures， the results indicated that the drying temperature had insignificant
，effect on the drying shrinkage of the eggplant samples fbr the investigated temperature
range． Fig3－4 shows the effect of drying chamber pressure on the drying shエinkage of the
samples at 50°C， it can be easily seen that the shrinkage became more severe at higher
vacuum chamber pressures． This phenomenon could be explained as fbllows：when water
is removed from the rnaterial during drying， a pressure unbalance is generated between the
interior of the dried material and the eXternal environment， and induces the contracting
stresses that lead to drying shrinkage． The drying shrinkage of eggPlant is particularly
severe because of the collapse of the unconsolidated porous tissue during drying・In
contrast with atmospheric drying， the pressure unbalance during vacuum drying is
substantia藍ly reduced due to the reduction in air pressure， consequent重y the drying
shrinkage could be reduced．
39
O＞＼〉
ーユ0610420
nUnU血UnUP
0 4 8 12 16
Moisture content（d．b． decimal）
Fig．3－3．Changes in v・1ume sh敵age rati・（VN。）With・m・isture c・ntent・f samples when
drying at 2．5 kl｝a and different drying temp eratUres
40
O＞＼〉
－」8’6420 0000
0 0．2
T 50°C
×2．5 kPa
◇5kPa
口10kPa◆Atmospheric pressure
0．4 0．6
M／Mb
0．8 1
Fig34． Relati・nship be櫨een v・lume shrinkage rati・（V！V。）and m・is蝕re c・nten重・f
samples when drying at 50°C and dif壬brent drying chamber pressures
41
Until now， many theoretical and emp三rical models fbr describing drying shrinkage have
been proposed（Mayor＆Sereno，2004）， alnong them， linear equation：
V
－ニaM＋b （3－2）
where V is the volume of a sample at any time of drying（m3）， M is the average mo量sture
content of the sample at the same time， Vo is the sample’s initial volume（2、25x10’5 m3 in
this study）， has been successfUlly used fbr describing the drying shrinkage of a wide range
of foodstuffs under var量ous drying conditions（Bailli＆langrish，2007；Lozano et aL，1980
＆1983；］Ratti，1994；Suzuki et aL，1976；Zogzas et aL，1994）．
11this study， Eq．（3－2）was fitted to the experimental data of sample volume from different
drying c・nditi・ns，9・。dness・鐙t・fthe equati・n was evaluated by R2㎜d mean relative
deviation modulus（P’）as presented in Chapter 1． Table 34 demonstrates the results of
linear regression analytical results on the experimental data at 50°C and various drying
chamber pressures． The results indicated that under present conditions， linear model was
adequate t。 m・del the vacuum drying shrinkage。f the eggPlant samples， the R2・f・the
linear regressi・n reached ab・ut O，99， and P’was less than 7％・The resuエts als。 pr・ved that
the drying shrinkage caused by vacuum drying was・bvi・usly less severe than that caused
by atmospherlc drying at the same drying temperature．
C。nsidering the relati。nship between the surface area and v・lume・fa sample・the su伽
area。f the sample can be described by the f・ll。wi且g equati。n（Orikasa et al・・2005（a）；
Pabis，1999；Pab董s＆Jaros，2002；Suzuki et al・，1976）：
尋2
2
£＝7t．〔一y．Vo〕5㌧細b）1・（3－4）
where A is the surface area of a sample at any time（m2）， Ao is initial surface area of the
sample． Due to the complexity of drying process of fbod materials， Eq．（3－4）may lose its
accuracy in some cases， According to the results reported by Nakamura et aL（2005）and
Orikasa et al．（2005， a＆b）， the relationship between surface area and dry basis moisture
content of the sample could be approximately expressed by a linear equation：
A＝a’M÷b’ （3－5）
Eq．（3－5）was fitted to the measured surface area and moisture content data， the f董tting
result showed that the linear equation described the relationship between them very welL
The experimental data fヒom drying at 30°C and 2．5 kPa and the Iinear regression
analytical results are shown in Fi93－5．
43
寸寸
一箇oQO．O【ooohい．翰等8．n9㊤on．的
卜αQoひ．Oひまひ．OひNぴひ．O寸沿Qqひ．O
（ま）』
髭
（四邑§竈ぎ』昌の0．006
0．OO5
0．004
0．003
0．002
0．001
0
▲30°C2．5 kP劉Expe rimenta l
コ
コ
サ
yRP 0．0001x＋0．0027
0．9802
8％
0 4 8 12 16
Mois加re content（d．b． dec㎞al）
Fl9．3－5． Change in surface area with m。isture c・ntent・fsample when dring at 2・5 kPa and
30°C
45
Surfaee eolor
Afヒer drying under various conditions fbr different durations， the surface color of the dried
samples was measured and surface color change， which was referred to as total color
difference（△E）， was calculated． Resu正ts showed that the value of△E gradually increased
with decreasing moisture con・tent， in other words， the Ionger the dτying time， the more the
color change（Fig．3－6）． Drying temperature and vacu．urn chamber pressure significantly
affected the color change of the dried sample， more sever color degradation was observed
ヰwhen the drying was conducted at higher temperatures and pressures． Since L value
represents lightness of a sample（L＊＝O yields black and L’＝100 indicates white）， it could
ホbe used as a indicator of extent of browning of the dried samples． The】L change of the
samples under vari・us drying c・nditi・ns is presented in Fig3－7． The value L零decreased
。nly slightly thr。ugh・ut the vacuum drying pr。cesses；less changes inガvalue were
observed during drying at Iower temperatures and vacuum chamber pressures． In
c・mparis。n with regular h。t－air drying， the br・wning・fthe samples・which is・ne・fthe
most devastating reactions fbr many EtUits and vegetables， could be inh孟bited during
vacuum d珈g due t・a1・w・xygen－c・ncentrati・n pr・cessing envir・nment and
comparatively Iow drying temperature・
46
口300C 2．5 kPa皿］300C lO kPa圃500C lO kPa
20864201五11
90 80 70 60
Moおture content（w．b．，％）
50
Fig．3－6， Total surface color difference ofthe samples dried under different conditions
47
90
80
嘱貞70
60
50
0
合レ七冥口糠▲口藷
▲300C 2。5 kPa
口300C 10 kPa
×500C 10 kPa
2 4 6 8 10 12 14 16
Moおtu1℃content（d．b．，decimal）
Fig．3－7． Changes inゴwith the m・isture c・ntent。fthe samples dried under different
conditions
48
Texture
The fbrce－defbrmation profiles obtained ffom the puncture tests were used to evaluate the
firmness of eggplant samples． The drying cond量tions had no statistically significant effect
on the firmness of the dried samples at the same moisture level． Fig．3－8 shows the
fbrce－defbrmation curves of the samples dried at 30°C and 2．5kPa fbr diffrerent t量me． The
rupture fbrce（Fm）of the vacuum dried samples kept inoreasing with deoreasing moisture
content until the moisture content reached about 30％（w．b．）， and afしerwards the value of
Fm rapidly reduced． While the FmDm of the dried samples remained almost the same before
the moisture content of the sample dropped to 85％， and then it gradually deoreased with
the decreasillg moisture content， after the moisture content of the sa三nples was reduced to
below 30％the value ofFm／D血showed an obvious increase．
As mentioned above， the shrinl（age stress during vacuum drying could be balanced by the
vacuum environment， besides， the vacuum drying of the eggplant samples under the
investigated conditions took place in the falling rate period， in which the lntemal moisture
can not be transported fセom the interior rapidly enough to keep the surf註ce of the samples
saturated． Accordingly， the turgor pressure of the surface celIs wouId gradually decrease
due to the Ioss of water content， and a sponge一重ike flexible porous layer would be fbrmed
on the surface of the sample with the progress of drying， which led to the increase in the
mpture fbrce and the drop in the Fm／Dm va董ue． The rapid drop in Fm and increase in Fm／1）m
of the samples with very low moisture coritent（＜30％）could be attributed to the surface
ha・dening・ccurred during the fi且al stage・f drying・Firmness refiects the integrity・ftested
49
tissue， both mpture fbrce and the ratio of ru．ptUre fbrce to defbrmation were used to
describe the flrmness of fuits and vegetables（Borwankar，1992；Boume，1965；Harker，
2002；Hertog et al．，2004；Jackman et aL，1990；Vare五a et al．，2005）． According to the
experimental results， the higher Fm did not mean the sample was‘‘f三rmer”， in other words，
the textural property could not be fUlly expressed by only the value of Fm， the effect of Dm
must also be taken into consideration in order to obtain correct textural infbrmat圭on of the
samples． Therefbre， when investigating the texture of dried eggplant pulp sample， it is
suggested the firmness of eggplant pulp should be defined as Fm／Dm due to its special
porOUS struoture，
3、4．Summary
The drying shrinkage of the samples was independent of the drying temperature『but
increased with an increasing in the drying chamber pressure． The re亘ationship betweell the
shrinkage ratio and moisture content of the vacuum dried samples’ モ盾浮撃п@be perfectly
expressed by linear equations． Vacuum drying at lower temperature and pressure had less
impact on the su血ce coIor ofthe sampIes；the change in firmness ofthe dried samples was
not slgnificantly affected by the drying condition． Taking drying shrinkage， color and
textural change into consideratieq vacuum drying at 30°C and 2．5 kP＆could provide the
dried samples with a better qua藍ity under the present experimental conditions．
50
（Z）8』o
20
16
12
8
4
0
T＝300C P＝・2．5kPa
94％（fresh）
90％80％40％20％
】）efomation（mm）
Fig，3－8． Force－Defbrmation curves of the samples withL diffbrent molsture content
51
CIIAPTER 4
Electrical lmpedance Spectroscopy Analysis of Eggplant Pulp
Tissue
Absか躍c”The electr重cal impedance characteristics of eggplant pulp were investigated；the
impact of drying and fアeezing－thawing treatments on pulp sample was evaIuated by means
of electrical impedance spectroscopy analysis． Fresh eggplant pu亘p was subjeoted to
vacuum drying， hot－air drying and ffeezing／thawing treatments respectively． The
impedance spectra of the ffesh， dried and thawed samples were measured over a ffequency
range of 42 Hz to 5 MHz， and analyzed with three equivalent circuit models． The model
parameters representing physiological propelties of the tissue were calculated using
non－1inear regression analysis． Results showed that the impedance loci of the fヒesh as well
as partially dried samples could be described by the models， and the distributed mode玉
fitted the data significantly better than the two Iumped models did， The impedance values
of the dried samples increased dramatically across the frequency range tested due to the
loss of moisture；the increase in the impedance of hot－air dried sample was statistically
higher than that of vacu．um dried sample under the present experimental conditions． A且er
丘eezing－thawing treatment， the impedance spectnum of the sampIe lost its original
character and became almost independent of frequency， which verified that the cell
membrane were severeIy disrupted during the f士eezing－thawing process．
κ砂woアゐ’Cole－CoIe equation；Drying；Eggplant；Electrical impedance speotroscopy；
Equivalent circult model；Freezing－thawing
52
4．1．Introductio皿
Electrical impedance spectroscopy（EIS）measures the dielectric properties of a medium as
a fUnction of fセequency． It is based on the interaction of an extema正electric field with the
electric dipole moment of materials． EIS has been used eXtensively to characterize
properties of solid materials（Pan et aL，2003；Prabakar＆Mallika巾n R．ao，2007；
Takashima＆Schwan，1965）． Compared with other tec㎞iques of physiological
investigation， EIS analysis is fast， reliable and easy to carry out， it has been widely used to
estimate the physiological state of various biological tissues（Cole，1932；Damez et aL，
2007；Harker＆DunIop，1994；Zhang＆WiIlison，1992；Zywica et al．，2005）．
In the EIS study fbr biological tissue， it is of great importance to establish appropriate
equivalent circuit model to relate the measured data to the physical and physiological
properties or their changes of the tissue investigated， and thus explanations fbr the
impedance spectra could be prov豊ded at the cellular level． Recent EIS studies suggest that
the impedance spectra of plan、t tissue can be characterized by the equivalent circuit models
composed of resistors and capacitors representing plant cell stru（加res；several equivalerlt
circuit models including lumped elemelt circuits and distributed element circuits fbr
describing the impedance characteristics of p五ant tissue have been reported． A lumped
mode蓋consists of limited circuit elements， each element should correspond to a cellular
stmcture that affeots the impedance characteristics of tissue． In a distr圭buted model， the
complicated cell structures such as memもranes are simulated by transmission I生ne， whioh
ca．n be seen as a multiple combination of many small circuit segments instead of limited
53
elements． Among the equivalent clrcuit models reported in the literature， a three－element
circuit model proposed by且ayden et a1，（1969）， and its adaptation（a five－element model）
proposed by Zhang et aL（1990）have been widely applied to the EIS analysis of various
plant tissues and have provided a lot of usefUl physiological infbrmation（Bauchot et ai．，
2000；Harker＆Maindonald，1994；Zhang＆Wlllison，1991；Zhang＆Willison，1992（a））．
Since both the lumped element models were proposed based on the assumption that the
shape and size of cells as well as their distribution in the tissue is unifbrm， in those oases
where the lumped models failed to satisfactorily reproduce or interpret the measured
impedance loci of some highly heterogeneous tissue， using distributed e互ement model was
proved to be an effective mealls of EIS analysis． Presently， a distributed circuit model
described by the Cole－Cole empirical equation has been applied to some EIS studies；using
the distributed model associated with electrophysiological consideration also offered
valuable infbrmation about the physiological properties of biological tissue（1）amez e亀a1．，
2007；Rep・＆Zhang・1993；Zhang et al・・19g！）・
Until now， most EIS studies of pIant tissue h＆ve been fbcused on the natural physiological
pr・pe貢ies・fthe tissue as well as relevant issues su。h as ripening・aging鋤d丘・st hardiness
etc． There are few rep・rts available detailing the impact・f煎i且cial pr・cesses such as
c。mm。nly－used f・。d pr。cessing techniques・n the electrical impedance ch訂acteristics・f
敏sand vegetables． The use・f EIS analysis c。uld pr・vide a new apPr・ach t・the
evaluation of the f士eshness or quality of processed agricultural products due to lts
simpli。ity and effectivity． The・bjectives・f the present stUdy were as f・ll・ws：（1）t°
54
analyze the impedance characteristics of f士esh， partially dried and f｝ozen－thawed eggplant
pulp in reference to the lumped element models proposed by Hayden et al． and Zhang et aL
as weli as the distributed model based on the Cole－Cole impedanoe equ．ation；（2）to
evaluate the effect of drying and fヒeezing－thawing pro cesses on the eggp1ant pulp tissue in
terms of the EIS analytical results．
4．2．Materials and methods
Samp le prep aration
Eggplants used in the study were purohased fセom a local market and stored at 10°C in a
refrigerator before the experiment started、 The pulp of the ffesh eggplallts had an average
moisture content of 94．00％（w．b，）according to the vacuum oven method（see Chapter 2）．
Four rectangular－shaped pulp samples， each of 45x15x6mm（耳ig．4－1）， were cut from the
center。f each eggPlant伽it Tw。 samples were vacuum dried（30°C，2・5kl｝a）and h・t－alr
dried（80・C）respectively t・a・target・m・is加re c。ntent・f 80％（wb・）・in the vacuum／h・t－air
dゆgsystem presented in detail in Chapter 2， the drying time was predetermined using
the mathematical model developed in the previous study． Another丘esh sample wrapped in
plastic film was fr。zen at－40・C in an air－blast freezer（S齪O・MDF－435・Japan）・鋤d
then thawed at 25°C in a constant temperature oven（DK600， YAMArO， Japan）． During
the丘eezing and thawing， the apPr・ximate central temperature・fthe samples was detected
もytype T therm・c・uples embedded in them， the pr。cesses were n・t st・PPed until the
temperature differences between the samples and the freezing／thawing envir。nment we「e
less than 1°C．
55
1
ヨOO一
／
90ゆ
15c盃＼ 2
一馨／ o騒： 口 ［葦≡≡ヨo
1
3 4
Fig．4－1． Schematic diagram。fEIS measuring system， 1－eggPlant pulp sample；2－clip test
probe；3－LCR． tester；4－computer
56
Electri‘al impedance〃teasure〃lent
The EIS measurement fbr above fヒesh and processed samples was perfbrmed at room
temperature， A schematic diagram of the EIS measuring system used in this study is shown
in Fig，4－1． The impedance data of the samp正es were measured using a LCR tester， which
was controlled by a computer， with two parallel c互ip test probes spaced 10mm apart
（HIOKI，3532－50， Japan）． The impedance， resistance， reactance and conductance va重ues of
the samples were measured at 50 frequency points（10garithmic fセequency intervals）over
the frequency range of 42Hz to sMHz under a measuring voltage of I V， and且utomatically
recorded by the computer fbr analysis． The measurement fbr each sample was triplicated at
3dif壬brent positions along the long axis， colnparisons between means were pe1fbrmed
using Duncan’s muItiple range tests at a significance level of O．05 i且SPSS 12つsoftware
（SPSS Inc．），
Since both test probes and tested m・aterial could contribute to the measured impedance data，
apreliminary study was conducted to separate the impedance of eggplant pulp sample
from the electrode impedance according to the rnethod used by Stout et al．（1987）． Results
showed that us童ng the present experimental setUp， the impedance caused by e重ectrode
polarization was very small（＜5％of the sample impedance at any frequency point tested）
and could be considered negligible compared with the impedance of the sample．
Accordingly， the measured impedance data were directly used as the impedance of the
fresh and treated samp正es in this study．
57
C3
R，
（a）
Zl
R，
C5 R2
R4
（b）
乙 Ro－R。。
（c） （d）
Fig．4－2． Equivalent circuit models used to analyze impedance characteristics of eggplant
pulp tissue，（a）－lumped model proposed by Hayden et al．（1969）；（b）－1umped model
proposed by Zhang et a正．（1990）；（c）＆（d）－distributed models
58
Eg躍’v躍1ε配c漉励models・and・da伽nalysis
The equivalent circuit models fbr eggplant pulp tissue were presented i且Fig．4－2． Mode1（a）
proposed by Hayden et aL（1969）took account of the resistance of cell walls（RI in model
（a）），the cytoplasmic resistance including vacuole（R2 in mode1（a））， and the resistance and
capacitance of cell membrane（C3 in mode1（a）， the membrane resistance was considered
very large and omitted）；Zhang et al．（1990）suggested that the capacitance ofthe tonoplast
（C5 in mode1（b））and the interior resistance of the vacuole（R4 in model（b））oontributed
substantially to the total impedance， and should be added as independent elements in the
equivalent circuit（in parallel with symplastic resistance R2 in mode1（b））． In the distributed
element model（c）， RI and R2 stand fbr the extracellular and intracellular resistance of plant
tissue respectively， the membrane impedance with transmission line properties is
representedもy Z 1．
However， fbr the conven圭ence of using the Cole－Cole impedance equation（Grimnes＆
M＆rtinsen，2000）：
Z＝R・・＋1呈lj諾捻 （4－1）
where Rヵand Ro are the tissue resistance at extremely high and extremely low f｝equency
（Ohm）， u・ is the alternating current frequency，τis the time c・nstant（s）・andαis a
dimensi。nless fact。r（ta逝ng values between O and l）， circuit（c）is transf・rmed int・（d）・
which is the simplest equivalent circuit飴r Eq．（1）， c・nsidering the飼1・wing relati・nships：
瑞一慧，；脇煮；z・一（ Zl R．つ1÷一＝），
Rl
s9
The complex impedance Z or admitta血ce Y of the circuit models as a fUnction of AC
倉equency（ω）was broken down into its real and imaginary parts二
Z＝R＋ノX （4－2）
or Y＝G＋ノB （4－3）
Where：R－resistance（Ω）， X－reactance（Ω）， G－conductance（S）， B－susceptance（S）．
For mode正（a）：
RIR、（Ri＋R、）C，2ω2＋R1 （4－4） R＝ （Rl＋R2）2C32ω2＋1
つ ㍗試慧・÷1（4－5）
For丁nodel（b）：
R、R、（R、＋R、）＋ミ・．
G＝せ＋（R，R，一詩＋i輪ジ6）
（R・＋R・）2＋R22＋1
C3ω C5ω C3Cs2tU3 （4－7） B＝＝ （R・R・－c，ぎ，ω・）ユ＋（9t’t＞i54＋甜
距G，皇B，（牛8）
X－G寺B，（4－9）
Since fbllowing transfbrmation can be made according to de Moivrels fbrmula：
Gωτ）（1－a）一（…・sg＋jc・Tsin ｛i）（エa）一（ωτ）（1－a）［c・s（1ヂ＋jsin（t’）冗］（4－1・）
The impedance Z in． Eq．（4－1）can be broken down lnto：
60
R＝＝R．＋
di。－R．，）［1＋（ωτ）（1一のc・s（煤f）π】
（1一α）冗 ］2［1＋（のτ）（1一のc。s
2
－（R。－R．，）（ωτ）（1一のsin
＋［（ωτ）（1一のsin（ 煤f）π12
（1一α）π
2
（4－ll）
（4－1・2） X＝ ［1＋（ωτ）（1－）c・s（t’）π】・＋［（ωτ）（トのsin（÷’）π】2
The parameters representing different physiological stru（加res in the three equivalent
circuit models were obtained by fitting the above Eq，（4－4），（4－6）and（4－11）to the
measured resistance or conductance data using non－1inear curve fit in OriginPro 75
s。ftware（OriginLab C。rP．）、 The Eq．（4－4），（4－5），（4－8），（4－9），（4－ll），（4－12）and the丘tted
m・del parameters were used f・r the calculati。ns・fR and X values・ The 9・・dness。f fit・f
the models to the experimenta1 data showing the appropriateness of using the rnodels was
evaluated by the c・e価cient・f・determinati・n（R2）， the r・・t mean square e∬・r（RMSE）and
the mean rQlative deviation modulus（P）（see Chapters 2 and 3）．
4．3．Results and discussion
Fresh eg9吻吻吻samp’ε
The。retically， f・r the relatively unif・rm undamaged tissue・the path・f the alternating
current would lie in channels of the cell wall at very low testing f士equenoies due to a very
large membrane 1mpedance， hence the ap・plastic resistance is estimated・ve「1°w
丘equency range． The capacitive rea。tance・f membranes gradu＆互ly decreases with
increasing丘equency；the decrease in the reaCtance will significantly affect the t。tai
impedance and cause a de。rease in the impedance value・fthe tissue when the f「equency
rises ab。ve a ce面n level」fthe丘eque且cy is su雌clently high， the effect・fthe capaclt°「
61
elements in the models may be neglected and the impedance will become independent of
the ffequency， accordingly the approximate symplastic resistanoe could be measured
（Bauchot et a互。，2000；Harker＆Dunlop，1994；Hayden et a1．，1969；Zhang et al，1990＞
The measured impedance spectra of the f士esh， partially dried and丘ozen－thawed samples
are shown in Fig．4－3＆4；the impedance of the samples were plotted against testing
丘equency in Fig．4－3（semi－logarithmic scale）， and the relationships between the reaI
（resistance）and imag孟nary（reactance）part of the complex impedance of the samples aτe
presented in Fig．4－4（also known as Cole－Cole plot）． The results indicated that the
impedance characteris重ics of the丘esh eggplant pulp tissue generally agreed with the above
postuIations of the lumped models． Two obvious limiting impedance values indicating the
apoplasmic and the cytopIasmic resistance respectively at low／high frequencies， and a drop
in total impedance with the increasing丘equency within a specific ffequency range（about
lKHz－－1MHz）can be observed in the irnpedanLce locus；when plotting the measured
reactance against resistance， the l。cus presented a fairly perfect semicircle sh。wing the
dielectric relaxat量on process．
62
（目‘○）胃里
20000
15000
10000
5000
0
LE＋00 1．E＋02 1．E＋04
△Fresh
口Hot－air d】ried
◆Vacuum dried
＋Frozen－thawed
1．E＋06 1．E＋08
Frequency（正1［z）
Fig．4－3． Electrical irnpedance spectra of fresh and processed eggl）藍ant pulp tissue
63
（目‘O）8員5⇔＄
4000
3000
2000
1000
0
△Fresh
◆Vacuum dried
ロロ ロ
ロ 恥r♂◆◆◆◆◆
⑳ △△△△△△ △△△
伝
ロHotair dried
十F「ozen－thawed
ロ ロ
ロ
ロ● ロ ◆◆◆
◇◆
“““．
ロロ
ロ
ロロ
0 5000 10000 15000
ResiStance（Ohm）
Fig．4－4． Electrical impedance spectra of丘esh and processed eggplant pulp tissue
（Cole－Cole plot）
64
By丘tting the models to the experimental results， the model parameters as we玉l as the
goodness offit indexes were evaluated， the results are shown in Fig．4－5 and Tab正e 1．From
the figure， the three models could， to different extent， describe the impedance spectnユm of
the fresh eggplant pulp． The caloulated results using model（a）showed comparatively
obvious deviations f士om the measured data over the whole frequency range（Fig．4－5a），
which indicated that the model could be too“general”to precisely characterize the
impedance－related physiological structures of the tissue． According to the literature
（Bauchot et a1．，2000；Harker＆Maindonald，1994；R．epo＆Zhang，1993；Zhang＆
Willison，1991）， mode1（b）was able to satisfactorily describe the impedance spectra of
some plant tissues suoh as potato， carrot， kiwifruit and nectarine etc． While il this stu．dy，
model（b）did not fit the experimental data at low fヒequencies very well， it only provided a
c・mparatively g・・d fit within the frequency range higher than 3×104Hz（Fig・4－5b）・This
result exhibited that the components in model（b）， possibly capacitive element， were
inaccurate fbr simulating the complex tissue stnエcture． Zhang et a1．（1993）also reported
that the lumped circuits（including model（b））did not fit the measured impedance data
丘om wood of Scots pine well on account of a Wide range of ce豆l size in the tissue． The
poorer fitting results of the lumped models in the present stUdy oould be attributed to the
heterogeneity of the eggplant pulp（e．g． seeds and air bubbles）that the lumped models
could not fhlly account for． Model（c）interpreted the impedance characteristics of the
tissue by c・mbining lumped resist。r elements with a circuit element having transmissi・n
line properties． The goodness of fit indexes in Table l showed that model（c）， having the
65
highest R2 and Iowest RMSE and P values， gave a significantly better fit to the measured
impedance data of the f士esh tissue than the tWo lumped models did． The calculated results
almost overlapped with the measured impedance locus（Fig．4－5c）． According to the
transformation relationship between circuit（o）and（d）， as mentioned in section 2．3， the
elements representing the physiological stnユctures in model（c）could also be characterized
using the fitted parameters of the circuit（d）， the caIculated extracellular and intracellular
resistance ofthe f士esh tissue are presented in Table 4－1．
66
卜O
卜．
･α。
､o
『α［
門卜Oひ寸寸oon
αま．Oいまdいトぴ．O
DQ
（ま）自 山のΣ属
犠
・軌v20已田酪自
一⑪
獅潤nΣ
8．E＋03
6．E＋03
4．E＋03
2．E＋03
0．E＋00
－2．E＋03
4．E＋①3
1．E＋00
。Measured R △Measured X
1．E＋02 1．E＋04 1．E＋06 1．E÷08
Frequency（1｛乞）
a
Fig．4－5． Comparison between calculated irnpedance data according to the equivalent circuit
models and experimental results ffom f士esh eggplant pulp tissue．
a；model（a）；bl model（b）；c：model（c）
68
8．E＋03
6．E＋03
4．E＋03
2．E＋①3
0．E＋①0
－2．E＋①3
－4．E＋03
1．E＋00
。 MeasuredR △ Measured X
1．E＋02 1．E＋04
Fig．4－5b
1．E＋06 1。E＋08
Fre（luency（H2）
8．E＋03
6．E＋03
4．E＋03
2．E＋03
0E＋00
－2．E＋03
4．E＋03
1．E＋00
。 Measured R △ Measured X
1．E＋02 1．E＋04 1．］E十〇6 1．E十〇8
Frequency（耳z）
Fig．4－5c
30000
25000 目 20000‘0 15000と国 10000
5000
0
1．E＋00 1．E＋02
◆MC．94％（fresh）
・MC．85％
▲MC．80％
xMC．75％
1．E＋04 LE＋06 1．E＋08
Frequency（】日【乞）
Fig．4－6． Relationship betWeen IZI and moisture content ofpartially vacuum dried eggplant
pulp tissue（30eC，25kPa）
71
Pania’かdried eggplant」pu lp samp les
According to previous studies， the apparent impedance oぎdried eggplant pulp samples
increased markedly with decreasing average lnoisture content（Fig4－6）． While since case
hardening occurred in the drying of the eggplant pulp samples significantly aff巳cted the
impedance measurement， the samples were dried to a target moisture content of 80％i且
this study．
Fig．4－3 demonstrated that the impedance values of the partially dried samples were
significant正y higher than the f｝esh sample， in addition， the measured va重ues from the hot－air
dried sample was statistically higher than those f虻om the vacuum dried one． The shape of
the impedance speotra of the partially dried samp重es was basicaUy consisten重with that of
the untreated sample despite the increase in the tota1量mpedance val鵬s， which implied that
the physiological stmcture of the ce11s in the tissue re斑alned intact＆衰er重he drying
treatments as far as impedance characteristics were coneerned；the sem圭circle loci fbr the
dried samples in the Cole－Cole plot showed only slight devlat量on at lower f｝equencies，
Fitting results indicated that the resistance values of the resistor ele1nents ln a正khe models
increased while the capacitance values of the capacitor elements重n model（a）and（b）
decreased after drying；model（c）still gave the best fit to the experirnental data among the
models examined， as presented in Table 4－2a＆b．
72
Table 4－2． Model parameters and goodness of fit indexes of the circuit models fitted to
impedance data of panially dried eggplant pulp tissue
a：hot－air dried sampIe
Model Parameters R2 RMSE P（％）a、DC R三＝・15637．R2ニ2747， Cヨ＝4．24
RI＝13201，Rユ＝2273， RA＝1369， C3＝2、⑪1． C5＝0．294
Rl＝17337， R2＝703， Ro＝17337， R。，＝676，τ＝0．OOOO8，0r＝0．40
0．962
0．993
0．999
1201
1972
85
30．9
13．8
1．6
bvacuum dried sample
Model Parameters R2 RMSE P（％）a†DC R1＝10238， R2＝2249， C3＝3．80
Rl＝9256， Rユ＝2164，＆＝ユ605， Cヨ＝2．30， C5＝0．305
Rユ＝11181，R2＝747， Ro＝lll81， R．；700， FO．OOOO5， a＝O．39
o．964
0．993
0．999
738
977
119
20．7
ユ0．5
2．4
Dimension：resistance（R）－ohm；capacitance（C）－x 1 O’9 farad；τ second
73
According to Chapter 2， both the vacuum and the hot－air drying under the present
conditions took pIace in the“falling rate period”， in which the evaporation of water on the
surface of the sample is faster than the moisture transfer from the interior to the surface．
Hence， the remarkable increase in the impedance of dried samples could be explained by a
decrease in the mobility of the electrolytes due to the severe loss of surface moisture
（Zhang＆Willison，1992（b））． Preliminary experirnents showed that the difference in the
impedance speotra between the samples processed at different drying temperatures（hot－air
60to 80°C）was statisticaUy insignificant， therefbre only 80°C hot－air drying with shorter
processing time was discussed in this study． From the calculated results presented in Table
2，the extracellular resistance in model（c）（or the counterparts in model（a）or（b））of the
hot－air dried sample differed greatly f｝om that of vacuum dried sample， whereas the
difference in the量ntracellu工ar resistance（or the counterparts）between them was not as
significant． This implied that the difference in the total impedance between the hot－air and
vacuum drled samples consisted primarily in the difference童n their eXtracellu正ar resistance，
in other words， the hot－air drying treatment caused greater extrace互lular moisture loss，
which consequently led to a more sign量ficant increase in the impedance values of the
sampIe． Under the present experimental conditions， the physiological status of the vacuum
dried t圭ssue were closer than the hot－air dried one to that of the f士esh sample according to
the EIS analytical results．
」Frozen－thewed eggplantpulp sample
Unlike the fresh and partially dried samp重es， the i五npedance spectra of the sample subj ected
74
to freezing－thawing treatment exhibited little change over the ffequency range tested， the
impedance value of the thawed sample was close to the limiting impedance value of the
fresh tissue within the high（＞1MHz）f士equency range（Fig．4－3）。 The Cole－Cole p重ot also
presented a irregu重ar spectrum instead of semicircle， the Iocus was even hard to identifシin
Fig．4－4 due to very small and nearly constant values（close to the left end ofx－axis， marked
by‘‘＋”）． Th圭s result proved that the physiological structure of the ce1藍s， especially the
membranes with dielectric properties， was disrupted by the ice crystals fbrmed during the
freezing process． A丑er thawing， e互ectrolyte leakage from symplasm led to the substantial
change in the impedance characteristics；10ss of cell turgor pressure and leakage of cell
contents accompanying disruption of membranes caused the severe drip loss and
deterioration of appearance and texture of the sample． ConsequentIy， those new
technologies that could protect tissue structure ffom丘eezing－thawing ir面ury should be
taken into consideration in ffozen fbod processing．
4．4．Summary
The electrical impedance spectroscopy analysis was applied to the evaluation of the effect
of drying and freezing－thawing treatm．ents on the eggpIant pulp． The measured量mpedance
data of the丘esh and treated sample were analyzed with respect to three equivalent
electrical circuit models． Model趾ting was perforrned by means of non．linear curve丘
using a regular statistical analysis software． The results indicated that the distributed model
based on the Cole－Cole impedance equation gave the best fit to the measured impedance
data of the fresh sample among the models examined・Drying treatment sigllificantly
75
increased the total impedance value but did not fUndamentally change the impedance
characteristics of the salnples． The impedance spectra of the partially dried samples
（moisture content：80％w，h）could still be best described by the distributed element modeL
Under the present experimental conditions， the impedance of hot－air dried sample was
apparently higher than that ofthe vacuum dried sample with the same moisture content． An
explanation fbr the variations in the impedance spectra of the partially dried samples was
given based on drying theory． Freezing－thawing treatment had a severe impact on the
impedance characteristics of the sample， the impedance locus became almost
frequency－independent after treatment． The result proved that the rnembranes of the celIs
possessing dielectric properties in the tissue were seriously damaged during
freezing－thawing cycle， which was considered the uppermost reason for the post－thawing
deterioration in quality attributes of f｝ozen products．
76
CHAPTER 5
Effeets of Dehydro血eezing Process o丑Some Physica豆
Properties of Eggplant Pulp
Abstraet’Effbcts of dehydrof士eezing process（including thawing and rehydration）on the
surface color， texture and electrical impedance characteristics of eggplant pulp sample
were evaluated． Fresh eggplant samples were partially deh、ydrated to molsture content of
80and 60％（w．b．）by means of vacuum drying at 30°C and 2．5 kPa． The dried s駐mples
were then su1〕」 ected to freezing（air－blast or immersion at－40°C＞－thawing（a重50r
25°C）－rehydration（at 25°C）treatments， their physical propert量es圭ncluding surface color，
texture and electrical irnpedance were measured and compared w茎th those of the
non－dehydrated control group． Results showed that the f士eezing processes of the partia玉ly
dehydrated samples were greatly shortened in contrast to f士esh sa憩ple；the rate of
三mmersion ffeezing was markedly higher than the air・b正ast f｝eezing． The
Vacuum－dehydrofrozen eggp重ant samples showed a remarkab重e inlprovement in the
quality－indicat壼ng Physica正 properties over the conven行onaHy frozen ones・ The
combination of量mmersion f｝eez圭ng at－40°C and thawing at 5℃had the正east impact on
the texture of the samples． Electricai impedance spectroscopy anaIys圭s indicated重hat the
damage t・the cell stn・ctures caused by ice f・rmati・n during freezing was still fatal despite
ad・pting va。uum dehydrati・n pretreatment and changing the fteezing－thawing c。ndit量。ns・
κ釧脚アゐ’Dehydr・fteezing；EggP至a斌王mpedance char＆cterist量cs；Physical pr・鋼ies；
Surface color；Firmness
77
5．1．Introduction
Freezing is one of the most popular means of long－term fbods storage． By transforming
most ofthe liquid water existing in food materials into ice， f卜eezing greatly slows down the
physical and biochemical processes involved in the deterioration of fbod and the growth
and reproduction of spoilage microorganisms． Genera正ly， freezing preserves the taste，
texture and nutritional value of fbods better than any other preservation technology．
However， when using freezing technology for preserving high water centent food materials
such as丘esh恥its and vegetables， large ice crystals formed in the tissue during freezing
may cause irreversible damage to the tender cell stn」（加res and eventually lead to drip Ioss
and deterioration of overall quality of the produCts after thawing（Reid，1983＆1990）． Up
to no・w， a lot of studies have been carried out to try to improve the quality of f≧ozen foods
through changing Process conditions of pretreatment，丘eezing and thawing operations・
Dehydrofreezing is one of the new正y developed fbod丘eezing techniques in which fresh
plant products with high water content are partially dehydrated prior to f｝eezing．
TheoreticalIy， the dehydration process not only reduces the amount of water to be fヤozen
but also makes ce盈l structures less susceptible to breakdown by changing oell turgor
pressure（Li＆Sun，2002）． Presently， s・me research。n dehydr。丘eezing techn・1・gy飾r
preservmg various fh」its and vegetables such as strawberry， green bean， peaちapple， tomato
and kiwifhlit have been reported．
As the name implies， dehydr・丘eezing is c・mp・sed・f tw。鯛・r pr・cesses・i£・pa蛇ial
dehydrati・n and食eezing， b・th。f them are critical t。 the qua董ity。臨al pr。duct・
78
According to the previous studies， vacuum drying had some advantages that could help
maintain the quality of dried eggplant sample over other dehydration techniques． The
application of vacuum drying to th、e dehydrof㌃ozen fbod processing might contribute to
improving the sensory and nutritive qualities of the product．
Immersion f士eezing is one of the fastest f｝eezing techniques， as the heat transfer coeffricient
is up to 20 times higher in正iquid media than in air， the high heat transfer rate gave a shorter
processing time and孟ncreased yield． Another advantage of immersion freezing is a higher
quality end product． The high heat rate of the immersion丘eezing proGess helps to improve
the apPearance and texture of produGt by reduoing the size of ice crystals fbrrned during
freezing（Lucas＆Raoult－Wack，1998）． Thus， the use of immersion丘eezing technique in
the dehydroffozen foods processing might be of benefit for produ（1 quality improvement．
The objective of this study was to investigate the effects of dehydrofreezing process
involving vacuum partial dehydration pretreatment together with immersiorゾair－blast
廿eezing as well as thawing and rehydration operations on the appearance， texture and
electrical impedance ch．aracteristics of eggplant pulp sample， and thus to evaIuate the
practicabiIity of dehydrof士eezing technology fbr preserving eggPlant and similar pIant
products in terms of product quality．
5．2．Material and methods
Samp le」prelワaration
EggP1ants used in the study were purchased fr・m a l・cal market and st・red at 10°C in a
・efrigerat・r bef・re the experiment started， The central脚。f each eggPlant伽it was cut
79
into one rectangu重ar－shaped sample of 45 x25×20mm fbr processing、 The average init1al
moisture content of the fセesh pulp samples was 94．00％（wet basis， N＝20， standard
deviation：0．52％）according to the vacuum oven method（see Chapter 2）．
EXperin・entalproeedure
（1）Partial dehydration pretreatment
Fresh eggplant pu正p samples were vacuum dried at 30°C and 25 kl｝a to the target moisture
content of 80 and 60％（wet basis）respectively in the vaouum drying system described in
りChapter 2． The drying time was predetemlined using the mathematical model developed in
the previous study． A食er the dehydration pretreatment， three groups of samples at various
moisture content levels， which were referred to as control group（M．C．94％）and partially
dehydrated groups（M℃．80 and 60％）respectively in th宣s study， were obtained fbr
丘eezing，
（2）Freezing－thawing process
Four f士eezing－thawing conditions（i．e． combinations of 40°C air－blast or immersion
丘eezing（refヤigerating medium：995％ethano1）and 5°C low temperature or 25°C room
temperature thawing）were examined in this stUdy． The samples f士om each moisture group
were put into zip top f士eezer bags（in order to prevent the samples from contacting with the
refrigerating mediuln）and subject to air・blast fteezing at－40°C in a freezer（SANYO，
MDF435，」apan）・r immersi。n freezing treatment in a beaker filled with ref「igerating
medium whiGh was placed in the freezer． The fr・zen samples were then thawed at 25°C in
ac。nstant temperature・ven（DK600， YAMArOJapan）・・r at 5°C in a d・mestic
80
reffigerator． During the freezing and thawing processes， the approximate central
temperature of eaoh sample was detected by type T thermocouples embedded in it and
recorded by a data logger（KEYENCE， NR－1000， Japan）， the processes were not stopped
until the temperature differences between the samples alld the f｝eezing／thawing
environment were less than 1°C．
（3）Rehydration treatment
A丘er thawing， the samples subjected to dehydration treatment were rehydrated by being
soaked in distilled water at 25°C． The weight of each sample was manually measured us短g
an electronic balance（GX2000， A＆D，∫apan）at intervals of 1 h during rehydration process．
The samples were considered to reach their equilibrium moisture content when the weight
differences betWeen two measurements were less than O．01g．
Determination ofphysicalpropeM’es ofsample
All experimenta正measurements fbr the physical properties of the fresh and processed
samples were perfbrmed at room temperature・
（1）Surface color
L＊ iLightness）， a＊（redness－greenness）and bホ（yell・wness－blueness）indexes。f the
CIELAB colorimetric system were used to describe the surface color of sample（Abbott，
1999）．The color measurement was made using a chroma meter（CR－200b， MINOLTA，
Japan）with D651孟ght source at 3 di ffere nt spots on each sample． The changes in surface
co正or of the sample， which was refbrred to as total color difference（△E）， was calculated
恥mthe L＊ C a‡ ≠獅с[value measured befbre and after prooess量ng accord董且g to the
81
のequatlon：
△E－（△L’）2＋（△aり2＋（ムゼ）2 （5－1）
（2）Texture
Puncture test was carried out to evaluate the chafiges in the textuτe of the samlples caused
by processing． The tests were perfbrmed uslng a creep meter（RE2－3305S， YAMADEN，
Japan）with a 20－N load celI and a 3 mm diameter flat－ended cylindrlcal punctu∫e probe＆t a
defbrmat圭on speed of l mm／s， During each test the def｛〕rmation of the sa拠ple（i．e憾epth the
probe reached）and the stress when the defbrmatio難occurred were au重o搬a重ically recorded
by a c。rnputer connected t・the creep meter． The firmness・f the eggPlan重sa即les，幡ch
is defi韮ed as the ratio of mpture fbrce£o defbrmation fbr f由三ts and vegetables（see C難a費ter
3），was deteimined according to the ferce－deformation（FD）ρro負1es obtaine赫om the
tests．
（3）Electrical i烈peda1lce oharacteristics
One piece of tissue measuring 40xlsx6mm was cut from each fresh cr processed eggp重ant
pulp sa搬P藍e負〕r e至ectrical impedance spectroscopy analysis・
In Cartesia且fbrm， complex impedance Z can be deξined as：
Z＝R一トノX （5－2）
Where the real脚。f impedance is the resistance（R）and the imaginary脚至s£he
re＆ctance（X）．
The impedance data induding resistan。¢and rea。tance ・f・the t量ssue was measure曲y a
LCR・tester（3532－50， H1◎照，」a脚）at 5。融re継ftequeRcy p磁s・ver撫e食equen。y
82
range of 42H乞to 51＞田z（logarithmic fセequency intervals）under a measuring voltage of IV
with two parallel clip test probes spaced 10mm apart， and automatically recorded by a
oomputer fbr analysis．
Data an alysis
The experiments were pe㎡formed in triplicate for each moisture content level and
freezing／thawing condition；alI results are presented as mean士SE． Comparisons between
means were performed using Duncan’s multiple range tests at a significance Ievel of O．05
in SPSS l2．O software（SPSS Inc．）．
5．3．Results and discussion
The moisture content of the sampIes was reduced丘om a initial va正ue of 94％to 80±1％
and 60士1％respectively after vacuum dehydration treatment． According to previous study，
the drying processes under the present experimental condition took place in the‘‘falling
rate period of drying”in which drying rate declines over time．
The three groups of samples with different moisture content were then subj ected to
freezing－thawiIIg treatment under fbur experimental conditions， their f士eezing curves are
shown in Fig．5－1． The freezing process of the samples shortened clearly with the
decreasing moisture content；the immersion ffeezing were significantly faster than the
ak－blast freezing for the samp正es at the same moisture content． According to the literature，
the rate of f士eezing can be defined as the characteristic time to traverse the temperatUre
range伽m－1 to－5°C， which is also known as “the zone of maximum ice formation” for
pIants（Li＆Sun，2002；Ngapo et a豆．，1999）． Damage to the membrane can be attributed to
83
difi7erent events within the tissue， depending on the rate of fセeezing， a higher freezing rate
is generally consider to be beneficia正to f士ozen products quaIity． At rapid f士eezing rates， the
propagation of the initial ice crystals is insuf丑cient to keep pace with theτate of heat
rernoval， resulting in supercooling and an increased f｝equency of nucleation． The result is
that more nucleation sites become active and there is an increase in the number of ice
crystals－with a corresponding decrease in crystal size． During slow freezing the
propagation of ioe can better keep pace with heat remova1， resulting in fewer nuclei
becoming active and the formation of larger ice crystals（George，1993）． From the figure，
the air－blast丘eezing rate of the non－dehydrated， MC80 and 6q％samples was 35，23 and
Ilmin， while the immersion freezing rate of them was shortened to about 29，17 and 8mill
respectively， obvious supercooling can be observed in the fヒeezing Processes of partially
dehydrated samples． In order to estimate whether the operating conditions including
freezing methods had significant effect on the quality of the frozen samples， some
quality－indicating physical properties of the processed samples were measured、
84
（9°）㊤」ヨ雪巴9。
30
20
10
0
－10
－20
－30
－40
MIC，4％a．ir」blast
磨
X MC、80％aiトblast
◇ MCβ0％immerslon
竃。
、
口▲ MC．60％ai卜blast
lCあ0％immersion、㌔ 、 ▲ ロ ▲
×◆ XO X
▲ 〔コ
｣ ロ?@ロ
@▲ ロ
、 ×A 、
0 20 40 60 80 100 120
TRme（mi1）
Fig．5－1．Freezing curves of samples processed under different conditions
85
In contrast to the non－dehydrated sampIes（M．C．94％）， no drip loss was observed in the
partially dehydrated groups（both M C．80 and 60％）after thawing． By rehydration， the
moisture content of partially dehydrated samples was restored to about 86％（w．b，）， the
surface color， texture and electrical impedance characteristics of the rehydrated samples
an、d directly thawed samples from control group were evaluated，
Fig．5－2 shows the color change（△E）of the samp互es after processing under different
conditions， larger△E values represents more obvious color changes of the samples． As can
be seen in the figure， the non－dehydrated samples after thawing（M．C，94％）had the most
sever color degradation among all the samples fbr all the process conditions・The color
change afしer thawing was attributed to the enzymatic browning reactions in the damaged
tissue． Since the vacuum drying processes were in the falling rate period， in which the
evaporation of water on the surface of the sample is faster than the moisture transfer from
the interior to the surface， a layer of tissue with very low moisture content was fbmled on
the surface of the samples du血g drying；in the following freezing process， the tissue layer
might be pr・tected・fr・m the har面l br・wning reacti・ns due t・the l。w m・isture c・ntent・f
it． A fter rehydration treatment， the samples fセom M C．80％group maintained their surface
コcolor slightly better than those丘om 60％group， possibly because of a Ionger processmg
time fbr the lower moisture content samples． In this study， th．e color d童fference of the
immersion frozen samples did not signi丘cantly differ f士om the air－blast f士ozen ones・and
the thawing tempera加re did n。t have statistically significant effe。t・n the c・1・r change・f
the samples either， Since the value・fL‡indicates lightness in the。・1・r spaceα㌔・yie重ds
86
ホblack・L＝100 indicates white）， and the c・1・r。f fresh eggPlant pu正P is nea・ly white， Lホ。f
the samples was als・c・mpared in・rder t・evaluate・the eXtent・f br・wning， as sh。wn in
Fi95－3・The value・fガ・fthe fresh pulp samples was ab・ut 85， afte・pr。cessing theゴ。f
the samples decreased t・different levels・H・wever， the L’・f the pa丘ially dehydrated
samples（b・th 80 and 60％）was significant正y higher than the c・ntr・1 9r・up regardless。f
。peratlng c・nditi・ns・ln・ther w・rds， the・surface・c・1・r・fthe dehydr・倉。zen eggPlant pulp
was maintained c・mparatively well in c。ntrast t・c。nvent量・nal freezing．
Fig・5－4 sh・ws the f・rce－def・rmati。n curves・f the fresh and pr・cessed samples derived
from puncture test， a clear bioyield point can be observed on the curve of fセesh sample；重he
frmness（ratio of rupture fbrce（Fm）to defbrmation（Dm）up to the bioyield point）of the
仕esh eggplant pulp could reach about 1500 NIm． Whi1e fbr the processed samp正es， the
bioyie正d point could not eve丑be observed on the curves due to severe softening after
processing． Therefore the values。f force and deformation when the puncture probe
penetrated into the pu正p reached a depth of 50％of the thickness of the samples， up to
which the F－D curve appeared to be appro）dmately a straight Iine， were used fbr the
caIculatlon of firmness， ResuIts showed重hat the dehydrofreezing－thawing－rehydration
process caused a reエnarkable dete】【ioration量n the texture of t｝1e eggPIant pulp， the firmness
of the samples droPPed from about 1500 to even beIow 200 N／m， as shown in Fig．5－5．
However， no matter what f｝eezing－thawing condition was adopted the f量rmness of the
dehydrof｝ozen sampies was s｛gnificantly higher than that efthe conventienally frozen oRes．
The difference in firmness value between the samples ofM．C．80 and 60％groups was not
87
noticeable， It was observes that the thawing temperature had an significant effe ct on the
texture of dehydro丘ozen samples， the samples thawed at 5°C had higher firmness values
than those thawed at 25°C． Statistically， for all moisture groups， those samples subj ected to
immersion fセozen and thawed at 5°C had the highest firmness values． This result can be
interpreted by that the ice crystal groWth during freezing， which is determined by the rate
of f｝eezing， have an effect on the size of the resulting ice crystals and the quality of the
frozen products；compared to the slower air blast freezing process（see Fig．2－a）， the size of
the ice crystals formed during the immersion freezing could be less and smaller， which in
turn caused less damage to the cell structure an、d consequently less change in the firm、ness
ofthe processed samples．
88
国く
40
35
30
25nU5021111
5
0
El air」blast，50C thawh亘9 n air－blast， 25°C thawing
團㎞茸nelsio n，5°C重hawing 圃㎞meIsion，25°C thawh19
隅・1コ⊥LLi1昏 ［
耕「卜、，
L：〔覇：
ii十｝十1Hl ←ト
1 1 m匝卜卜i十←
一1 ■引
‘”F：’
D：li：i：i：，
P；i…i室”
F：：：：・
奄奄奄奄奄奄堰F：…←田十 ｛iii
【一u「
iiii
E：：：i
ii田1＿
：・：・ii・：・
F：：：：：：：：：
F：・：：：：：’：・：・．・，・1・
F；：：：：：i：i
o……iiiiiii
lElI日；「rr7－
u「「「「「｡1－u［「
i；：；：．’：’・：：：：：：・1：：・
塩ﾖi……
ト離旺LLLi⊥卜i－Hi
韓ii 団±賠瞠
［ii
ii■ii，串
MC．94％ MC80％ MC．60％
Mo蛤tu鷹｝conte皿t befb肥freezi皿9
Fig．5－2． Total color difference of samp正es processed under different conditions
89
80
70
60
50
詣40 30
20
10
0
［］・air・・blast，5°C thawi皿9 ％ air－blast，25°C thawi 19
MC．94％ MC．80％ MC．60％
Mo iStU re c o ntent befo re freezi皿9
Fig．5－3． Comparison ofL’of samples processed under various conditions
go
2）8』o
0001
！042血U
①
fresh．
MC．94％immersion－25°C thawlng
MC．80％air」blast－5°C thawing
MC．60％imme rsio n－5°C thawing
2
▲▲▲▲▲▲▲▲
4
△△ △△△△△△△△
6 8 10
Deformation（mm）
Fig．5－4． Force－Deformation profiles of fresh and processed samples
91
（皐乙吻吻霞目盈
口a血㌔blast，5°C thawing rZ air－blast，25°C thaWing
300
250
200
150
100
50
0
MC．94％ MC．80％ MC．6①％
MoiSture content befo re freeZi皿9
Fig．5－5． Comparison of firmness of samples processed under various conditions
国immersion，5°C thawing 國㎞me歴藍ion，25°C thawin
一ヨ「
g 荊 ・
鵡 l l
十HiI 1，iI
轟 」±出
．雛± 皆iH［■目
欄
：’：・．：：；，
p…i……i………l
奄奄演ﾘi
奄奄奄奄奄奄奄堰G…
奄奄奄奄奄奄奄奄奄
引｝、i糊瑠日iIl
iHI荊仙 弓十梓
翌 至ヨiH
1蓑ii
ll …iil 超一逼：i
判耳1
強 、iiiiiii：iil
淵 i、F，
i：i二i：i：i：
Diiiiliiiii蓉i
珊
iiiiiiiii；
奄奄奄奄奄奄奄堰p
奄奄塩ｪ；
：匿工〔 ii揃
欄 出i｛
出紹十ｮ塒
』｝iiii
蝦 螺］
The use of electrical impedance spectroscopy analysis could provide a new approach to the
evaluation of the physioIogical status or quality of processed agricultural products due to
its simplicity and effectivity．
The electrical impedan、ce spectra（impedance against altem．ating current f士equen、cy）of the
丘esh and processed samples are shown in Fig．5－6． Theoretically， fbr the relatively unifbrm
undamaged biological tissue， low f士equency altemating current would only pass through
channels of cel重wali due to a very large impedance of membranes， accordingly the
apoplastic resistance can be estimated at this point；as the ffequency of altemating current
increases， capacitive reactance of the membranes gradually decreases and the current could
partly flow threugh the sympiasm， which in tum causes a decrease in the total impedance
ofthe tissue；when the f｝equency is so high that the contribution of the capacitive reactance
could be neglected， the impedance would become independent of the frequency and equal
approximately the symplastic resistance（Bauchot et al．，2000；Harker＆Dunlop，1994；
Hayden et al．，1969；Zhang et aL，1990）． The results indicated that the impedance
characteristics of the fresh eggplant pulp tissue generally agreed with the above
P・stulati・ns． Tw・・bvi・us impedance limits sh・wing the ap・Plasmic and the c卿1asmic
resistance respectively at l・wthigh frequencies， and an・bvi・us dr・p in t・ta丘impedance
with the increasing丘equency within a speci丘c frequency range（ab・ut 1K出一1MHz）
むwere observed on the im、pedance spectrum；when plot宅ing the measured reactance agalnst
resistance（als・㎞・㎜as C・le－C・le pl。t）， the 1・cus presented a fairly perfect semicircle
sh。wing the dielectric relaxati・n pr・cess， as sh・wn in Fi95－7・ln c。ntrast t・f「esh sample・
93
the lmpedance loci of the samples subjected to丘eezing－thawing treatment exhibited little
change over the testing frequency range；the value of the thawed non－dehydrated samples
（control）was close to the cytoplasmic resistance of the fresh tissue（Fig，5－6）， which
indicated that the cell membranes was damaged and the cell contents were released． Little
diffbrence between the impedance characteristics of the convent量onally fヒozen
（non－dehydrated）and dehydro丘ozen samples was observed， the impedance values of the
dehydrof士ozen samples were only slightly higher than the con、trol group due to the
decreased moisture content． Instead of the semicircle， irregular curves were presented in、
the Cole－Cole plot of the processed samples；the Iocus was even hard to identifシ量n Fig．5－7
due to their very small and nearly constant values． The results proved that the f「eezing
process completely destroyed the cell memlbranes of the samples， electrolyte ieakage ffom
the damaged cells finally led to the fUndamental changes in the impedance spectra． Loss of
ceH turgor pressure and sever structural damage due to the breakdown of cell membranes
were responsible for the degradation in final quality of the processed samples．
94
9000
8000
〔7000罠
‘0 6000と旦50008壽4000う2S．3000菖
2000
1000
0
◆fresh
・MC．94％a並一blast
“eeA ・MC・94％㎞…i・n 、 xMC．80％a㎞一blast
ee xMC。80％immersion ◆ ◆ 一〕M二C．60％aiir－blast ◆ ◆◆ ＋MC．60％immersion
、
◆ ◆
簿幸簿葦塁縫蠣識鞭葦牽葦轍轍綴棄綴品飴占飴凸品飴台凸飴飴合合凸飴飴飴飴凸合飴合飴飴合飴
1．E＋00 1．E＋01 1．E＋02 LE＋03 1．E＋04 LE＋05 LE＋06 1．E＋07
Frequency（耳Z）
Fig5－6． Electrical impedance spectrum・f fresh and pr・cessed eggPlant pulp tissue
95
（目‘O）㊤冒5。＄
2500
2000
1500
1000
500
0
0
△Fresh
＋MC．80％immersion伽ezing，25°C thawing
MC．600／o immersion freezi皿9，5°C thawing
△△△△△△△ △△△ △△△
試△ △瓠
2000 4000 600① 8000
Resistance（Ohm）
Fig．5－7． Cole－Cole plot of丘esh and processed eggplant pulp tissue
96
5・4・Summary
In this study， the effect of dehydrofセeezing process incIuding thawing and rehydration on
the surface color and teXture of eggPlant pulp were estimated；the changes in the electrical
impedance characteristics of the samples caused by processing was examined in order to
expIain the quality deterioration of the ftozen products． According to the experimentaI
results， vacuum dehydration treatment prior to ffeezing did effectively prevent
post－thawing drip loss and inhibit the f七eezing－thawing deterioration of the eggplant pulp
samples to some extent in contrast to conventional freezing． The surface color of the
dehydrofヒozen eggplant pulp was maintained comparatively wel1． The color difference of
the immersion fセozen samples did not significantly differ fセom the air－blast丘ozen ones，
and the thawing temperature did n・t haye a statistically significant effect・n the c・1・r
change of the samples either． The dehydrof卜eezing－thawing－rehydration process caused an
obvious deterioration in the texture of the eggplan、t pulp， however， the firmness of the
dehydrofrozen samples was significant正y higher than that of the conventionally丘ozen ones．
Thawing temp erature significantly affected the texture of dehydrof士ozen sampl es， samples
thawed at lower temperature had higher firmness value． The impedance spectra of the
samples were 且mdamentally changed after processing， leakage of cell contents
accompanying the disnエption of the membranes accounted fbr the changes in impedance
characteristics・f the pr・cessed samples． Severe dainage t・the cells caused by the ice
crystals fbrrned during freezing Process was stiU inevitable despite changing the
freezing－thawing condition．
97
CHAPTER 6
ConclusiOBS
Theoretical and experimental studies on the vacuum－dehydrofreezing technology were
oarried out in order to estimate the practicability of dehydrofヒeezing feohnology fbr
preserving ffesh－cut eggPlant pulp， According to the research・work presented in the
above chapters， the fbllow重ng maj or conclusions were drawn：
Firstly， the vacuum drying characteristics of eggplant pulp were investigated． ThLe
results showed that vacuum drying of the pulp samples took place in the falling rate
period；increasing drying temperature shortened the drying Process notably， and drying
ohamber pressure ranging from 2．5 to 10kPa had statistically insign童ficant effect on the
drying process． A mathematioal model was developed considering the relationship
between the drying rate（dM／dt）and moisture content（M）during the dτying Process， the
model showed better fit to the experimental data than five common、ly reported drying
mode正s examined． The effective moisture diffi．isivity ofthe samples in the moisture ratio
range of l to O．35 was calculated according to the infinite series solution of Fick’s
second law of diffUsion， the temp erature dependence of the effect三ve moisture
diffusivity was describ ed satisfactorily by an Arrhenius－type equation．
Secondly， the impacts of vacuum drying cond量t圭ons on the qu．ality attr圭butes of eggplant
pulp sample were evaluated． The drying shrinkage of the samples during vacuurn drying
was independent of the drying temperature but increased with an inoreasing in the
drying chamber pressure． The re藍ationship between the shrinl（age ratio and moisture
98
content of the vacuum dried samples could be perfeotly expressed by a Iinear equation．
Vaouum drying at lower temperature and pressure had iess impact on the surface color
of the samples；the flrmness of the dried samples was not s｛gn童ficantly affected by the
drying conditions． Taking drying shrinkage， color and tex加ral changes into
Gonsideration， vacuum drying at 30°C and 2．5kPa provided the dried eggplant samples
with a‘‘better”quality under the present conditions．
Thirdly， the electrical impedance spectroscopy analysis was applied to the evaluation of
the effect of drying and fヒeezing－thawing processes on the eggplant pulp． The measured
impedance data of the ffesh and processed samples were analyzed with respect to three
equivalent electrical circuit models． Model fitting was perfbrmed by means of
non－linear ourve fit using a regular statistical analysis software． The results圭且dicated
that the distributed model based on the Cole－Cole impedance equation gave the best fit
to the experimental data of the ffesh samples among the three models examined． DryinLg
process s童gnifioantly illcreased the total ilnpedance value but did not fUndamentally
change the impedance characteristics of the samples． The impedance spectra of the
panially dried samples（m・isture c・ntent：80％w・b・）c。uld still be described by the
distributed element model． Under the present experimental conditions， the圭mpedance of
hot－air・dried sample was apparently higher than that of the vacuum dried sample with
the same molsture content， an explanation fbr this result was given in consideration of
the drying mechanism． Freezing－thavving treatment had a great impact・n the impedance
characteristics of the sample， the impedance loeus became f士equency－independent after
99
af士eeze／thaw cycle． The resuIt proved that the membranes of the celIs with dielectric
properties in the tissue were severely damaged during fヒeezing process， which was also
considered the uppermost reason fbr the post－thawing deterioration in quality attributes
ofthe frozen product．
Finally， the effects of dehydrof士eezing Process inoluding Partia盟dehydration， different
丘eezing－thawing methods and conditions as well as rehydration process on the quality
and impedance characteristics of eggplant pulp sample were investigated． According to
the experimental results， vacuum dehydration treatment prior to freezing did effectively
prevent post－thawing drip loss and inhibit the freezing／thawing deterioratioll of the
eggplant pulp samples to some extent in contrast to conventional ffeezing． The color
and texture of the dehydrofrozen pulp samples was significantly better th．an the
convent玉onally f沁zen ones， despite the fact that the dehydration， fteezing－thawing and
rehydration processes caused a markedly drop in the衝㎜ess。f the samples， It was
observed that thawing temperature significantly affected the texture of the processed
samples， the dehydroffozen samples thawed at lower temperature had higher v＆lues of
firmness． The results also showed that both conventional f士eezing and dehydrof士eezing
pr。cesses fUndamentally changed the impedance characteristics・f the samples・which
implied that the fatal damage t・the cells caused by the ice crystals f・rmed during
freezing was still inevitable under the present experimental conditions．
100
REFE］RENCES
Abbott， J．A．（1999）． Quality measurement of fltuits＆nd vegetables， Postharvest
Biology and Technology，15（3）：207－225．
Agne1正i， ME．＆Mascheroni， R．H，（2002）． Quality evaluation of fbodstuff諭ozen in
a cryomechanical f士eezer． Journal ofFood Engineering，52：257－263，
AgnelIi， M，E．， Marani， C．M，＆Mascheroni， R．．H．（2005）． Modelling of heat and mass
transfer during（osmo）dehydrofreezing of fruits． Joumal of Food Engineer孟ng，
69（4）二415－424，
AOAC．（1995）． Official methods of analysis（16th ed．）． Washington， DC：Association
of official analytical chemists．
Arevalo－Pinedo， A，＆Murr， F．E．M．（2006）． Kinetics of vacuum drying of pumpkin
（Cucurbita maxima）：modeling with shrinkage． Tournal of Food Engineering，
76（4）：562－567．
Arevalo－Pinedo， A．＆Murr， FEM．（2007）． Influence of pre－treatments on the drylng
kinetics during vacuum drying of carrot and pumpkin． Joumal of Food
Engineering，80（1）：152－156．
Ba董ni， R．＆Langrish， TAG（2007）． Choosing an appropriate drying model fbr
intermittent and continuous drying of bananas． Joumal of Food Engineering，
79（1）：330－343．
Bauchot， AD．， Harker， F R＆Arnold， W M（2000）． The use of electrical impedance
spectr・sc・py t・assess the physi・1・gical c・nditi・n・f kiwifruit・P・stharvest
Biology and Technology，18（1），9－18．
Bazyma， LA．， Gu sko鷲VP， B asteev， A．V， Lyashenko， A．M．， Lyakhno， V＆KutUvoy，
VA．（2006）． The investigati。n・f l・w temperatUre vacuum drying pr・cesses。f
agricultural materials． Journal ef Food Engineering，74（3）：410－415．
Biswal， RN， B・z・rgmehr， K， T・mpkins， FD・＆Liu・X（1991＞Osm。tic
concentration of green beans prior to freezing・Joumal of Food Science・56（4）：
1008－1012．
Bol量n HR．＆Huxsol1，C．C．（1993）． Patial drying of cut pears to improve f㌃eeze／thaw ，
101
texture． Joumal ofFood Science，58（2）：357－360．
Boonsumrej， S．， Chaiwanichsiri， S．， Tant耐ian， S．， Suzuki， T．＆Takai， R．（2007）．
Effects of freezing and thawing on the quality changes of tiger shrimp（Penaeus
monodon）frozen by air－blast and cryogen量c f士eezing． Journal of Food
Engineering，80：292－299．
Borwankar， R．P．（1992）． Food texture and rheologyl a tutorial review． Journal of Food
Engineering，16（1－2）：1－16．
Boume， MC．（1965）． Studies on punch testing of apples． Food Technology，19：
413－415．
Bmce， D．M．（1985）． Exposed－layer barley drying；Three models趾ted to new data up
to l 50°C． Journal ofAgricultura夏Engineering Research，32（4）：337－348．
Chen， C，＆Morey， R．V（1989）， Comparison of fbur ERHIEMC equations，
Transaction ofthe ASAE，32（3）：983－990．
Cole， K．S．（1932）． Electric phase angle of cell membranes． Journal of General
Physiology，15，641－649．
Crank，」．（1975）． Mathematics of di飾sion（2nd ed．）． Oxfbrd：Clarendon Press、
Cui， Z W．， Xu， S．Y＆Sun， D．W（2004）． Microwave－vacuum drying kinetics of carrot
s互ices． Joumal ofFood Engineer量ng，65（2）：157－164，
Damez， J－L．， Clerj on， S，， Abouelkaram， S．＆Lepetit， j．（2007）． Dielectric behavior of
beef meat in the l kHz to 1500 kHz range． Simulation with the Fricke／Cole－Cole
Model． Meat Science， doi二10．1016なmeatsci．2007．04、028、
Dermes・n1・u・gl・u， EK，αannak・ur・u， M．C．＆Ta・ukis， PS・（2007）・Kinetic
modelling of the degradation of quality of osmo－dehydro丘ozen tomatoes dur至ng
storage． Food Chemistry，103：985－993．
Ge。rge， RM．（1993）， Freezing Pr。cesses・used in・the f・・d industry・Trends in F・。d
Science and Technology，4：134－138．
Giri， SK，＆Prasad， S、（2007）． Dゆg kinetics and rehydrati・n cぬaracteristics・f
microwave．vacuum and convective hot－air dried mu．shrooms． Joumal of Food
Eng董neerillg，78（2）：512－521．
Grimnes， S．＆Ma面nsen， O． G（2000＞Bi。impedance＆bi・electricity basics・
102
Academic Press．
Harker， FR＆Dunlop， J．（1994）． Electrical impedance studies of nectarines during
cool storage and firuit ripening． Postharvest Biology and Technology，4（1－2），
125－134．
Harker， ER．＆Maindonald，」，H．（1994）． Ripening of nectarine丘uit． Plant
Physiology，106，165－171，
Harker， F． R．， Maindonald J．， Murray， S．H．， Gunson， E A．， Hallett，1．C．＆Walker， S．B．
（2002）．Sensory interpretation of instrumental measurements l：teXture of apple
Et・tユit． Postharvest Biology and Technology，24：225－239．
Hayden， RJ．， Moyse， CA， Calder， F．W．， Crawford， D．P＆Fensom， D．S，（1969）．
Electrical impedance studies on potato and a正falfa tissue． Joumal of
Experimental Botany， 20（63），177－200．
Heldman， DR，（1982）． Food properties during丘eezing， Food tec㎞ology，36（2）：
92－96．
Heldman， DR．（1983）． Factors influencing fbod f士eezing rates． Food technology，
37（4），103－109．
Henderson， S． M．＆Pabis， S．（1961）． Grain drying theory ll：Temperature effects on
drying coefficients． Journal ofAgricukural Engineering Research，6（3）：169－174．
Hertog， ML．AT．M．， Ben－Arie， R．， R6th， E．＆Nicolal， B．M（2004）． Humidity and
temperature effects on invasive and non一量nvasive firmness measures． Postharvest
Biology and Technology，33（1）：79－9 L
Jaokman， R．L．， Marangoni， A．G＆Stanley， DW．（1990）． Measurement of tomato
fU量t firmness， HortScience，25（7）：781－783．
Jaya， S．＆Das， H．（2003）． A vacuum dτying model fbr mango pulp． Drying
Technology，21（7）：1215－i23 4．
Jena， S．＆Das， H．（2007）． ModeHing fbr vacuum drying characteristics of coconut
pressGake， Journal ofFood Engineering，79（1）：92－99．
Karathanos， V， Anglea， S．＆Karel， M．（1993）． Coll＆pse of structure du加g drying of
ce正ery． Drying Technology，1工（5）11005－1023．
Lazarides， HN．＆Mavroudis， N．E．（1995）． Freeze／thaw effects on mass transfer rates
103
during osmotic dehydration． Journal ofFood Science，60（4）二826－857．
Lewicki， P． P＆Ja㎞bczyk， E．（2004）． Effect of hot air temperature on rnechanicaI
properties of dried app正es． Joumal ofFood Engineering，64（3）：307－314．
Li， B，＆Sun， D，W（2002）． Novel methods for rapid fteezing and thawing of foods－
a review． Journal ofFood Engineering，54（3）：175－182．
Lozano， JE．， Rotstein， E．＆Urbicain， M．工（1980）． Total porosity and open－pore
porosity in the drying of fruits， Journal of Food S eience．45：1403－1407．
Lozano， J．E， R．otstein， E＆Urbicain， MJ．（1983）． Shrinkage， porosity a皿d bulk
density of foodstUffs at changing m。isture contents． Joumal of Fo。d S cience．48：
1497－1502．
Lucas， T．＆Raoult－Wack， A．L．（1998）． Immersion chilling and f｝eezing in aqueous
refhgerating medial review and fUture trends． Intemational Joumal of
Ref士igeration，21（6）；419－429．
Madamba， P S．， Driscoll， R，D．＆Buckle， K，A．（1996）． The thin－1ayer drying
characteristics ofgarlic slices． Joumal of Food Engineering，29（1）：75－97．
Maestrelli， A．， S calzo， RL．， Lupi， D．， B ertolo， G＆Torreggiani， D．（2001）． Partial
removal of water before ffeezing：cultivar and pre－treatments as quality factors
of frozen muskmelon（Cucumis meio， ev reticulatus Naud．）． Journal of Food
Eng董neering，49（2－3）：255－260．
Maskan， M．（2001）． Drying， shrinkage and rehydration characteristics of kiwifhユits
during hot air and microwave drying． Joumal of Food Engineering，48（2）1
177－182．
Mayor， L．＆Sereno， A．M．（2004）． Modelling shrinkage during convective drying of
fbod materiais：a review． Jeurnal of Food Engineering，61（3）：373－386，
Mcminn， WAM．＆Magee， T．R、A（1997， a）． Physical characteristics of dehydrated
potatoes－Part工． Joumal of Food Engineering，33（1－2）：37－48．
Mcminn， W． A．M．＆Magee， T．R， A．（1997， b）． Physical charaoteristics of dehyd耐ed
potatoes－Part II． Journal ofFood Engineering，33（1－2）：49－55・
Methakhup， S．， Chiewchan， N．＆Devahastin， S．（2005）． Effects ef drying methods
and conditions on drying kinetics and quality of Indian gooseberry flake・ LWT －
104
Food Science and TechnoIogy，38（6）579－587，
Nakamura， T・， Tagawa， A・， Orikasa， T＆Iimoto， M．（2005）． Vacuum drying of
cooking tomatoJournal of the Japanese Society of Agricultural Machinery，
67（6）：105－112．
Ngapo， T．M， Babare， r H．， Reynolds， J．＆Mawson， RE（1999）． Freez至ng and
thawing rate effects on drip loss from samples of pork、 Meat Soience，53（3）：
149－158．
Orikasa， T、， Tagawa， A．， Nakamura， T．＆Iimoto， M．（2005， a）． Drying characteristios
of cooking tomato、 Joumal of the Japanese Society of Agricultural Machinery，
67（2）：46－52．
Orikasa， T．， Tagawa， A．， Soma， S．， Iimoto， M．＆Ogawa， Y（2005， b）． Hot air drying
characteristics of finits and vegetables and surface hardening of samples during
dryingJoumal of the Japanese Soclety of Agricultural Machinery，67（6）：62－70．
Pabi s， S．（1999）． The initial phase of convection drying of vegetables and mushrooms
and the ef壬bct of shrinkage． Journal ofAgricultUral research，72：187－195．
Pabis， S．＆Jaros， M．（2002）． The first period of convection drying of vegetables and
the effect of shape－dependent shri鉦kage． Biosystem Engineering，81（2）：201－211．
Pan， L．K，， Huang， H．T．＆Suq C， Q．（2003）． Dielectr量c reIaxation and transition of
porous silicon． Journal ofApp正ied Ph、ysics，94（4），2695－2700．
Pi五aga， F．， Carbonell， J．V， Pefia， J．L．＆Miquel，工J，（1984）． Experimenta正simulation
of solar drying of garlic using an adsorbent energy storage bed、 Jeumal of Food
Engineering，3（3）：187－203
Prabakar， K．＆Mallika巾n Rao， SP．（2007＞． Complex impedance speotroscopy
studies on fatigued soft and hard PZT ceramics． Jouma蓋of A重loys and
Compounds，437（1－2），302－310．
Ratti， C．（1994）． Shrinkage during drying of foodstuffs． Joumal of Food Engineering，
23（1）：91－105．
Raoult－Wack， A．L．（1994）． Recent advances in the osmotic dehydration of fbods．
Trends in Food Sclence and Technology，5：255－260。
Reid， D．S．（1983）． Fundamental physicochemlcal aspects of freezing． Food
lO5
Technology，37（4）：110－115，
Reid・D・S・（1990）・Optimizing the quality・f・fr・zen f・・ds． F・・d Tec㎞。1・gy， 44（7）：
78－82．
Repo， T．＆Zhang， MIN（1993）． Modelling woody plant tissue using a distributed
electrical circuit． Journal of Experimental B otany， 44（262），977－982．
Robbers， M， Singh， RP＆Cunha， L m．（1997）Osmotic－Convective dehydrof卜eezing
process for drying kiwifhユitJoumal of Food Science，62（5）：1039－1042．
Sacilik， K＆Elicin， A．K（2006）． The thin正ayer drying characteristics of organic
apple slices． Joumal ofFood Engineering，73（3）：281－289．
Souma， S．， Tagawa， A＆Iimoto， M（2004）． Structural properties fbr iiuits and
veget＆bles during drying Joumal of the Japanese Society fbr Food Science and
Techno亘09】y．51（11）：577－584．
Steinmetz， V， Crochon， M， Bellon Maure1， V， Garcia Femandez，兀．， B arreiro
Elorza， P．＆Verstreke恥L．（1996）． Sensors fbr fii．iit firmness assessment二
con tparison and fUsion． Journal ofAgricultural Engineering Research，64115－28．
Stout， D．G， Hall， J，W＆McLaughlin， N．B．（1987）． In vivo p正ant impedance
measurements and oharacterization of melnbrane electrical properties：the
inf1uen｛：e of col（三acclimation． Cryobiology，24（2），148－162．
Suzuki， K．， Kubota， K．， Tsutemu， H，，＆Hosaka， H．（1976）． Sh．rinkage in dehydration
of root vegetables．」邑ournal ofFood Science．41：1189－1193．
Tagawa， A．， Kitamura， Y＆Murata， S．1996． Thin layer drying charaoteristics of
Adzkiki beans， Transaction of the ASAE，39（2）：605－609．
Tagaw＆， A．， Muramatsu， Y， Nagasuna， T．， Yano， A．， Iimoto， M．＆Murat亀S．（2003）．
Water ab sorption characteTistics ofwheat and barley during soaking． Transaction
ofthe ASAE，46（2）：361－366，
Takashima， S．＆Schwan， H．P（1965）dielectr｛c dispers1on of crystalline powders of
amino acids， Peptldes， and Proteins． Joumal of Physical Chemistry，69（12），
4176－4182．
Toei， R、．（1975）， Theory and caiculation of chemical machine （2nd ed．）． Tokyo：
Sangyo boo三（s．
106
Tregunno， NB，＆Goff， HD．（1996）． Osmodehydrofreezing of apples：structural and
teXtural effects、 Food Research Intemational，29（5－6）：471－479．
Tutuncu， MA＆Labuza， TP（1996）． Effect of geometry on the eff7ective moisture
transfer diilfbsion coefHcientJournal of Food Engineering，30（3－4）：433－447．
Vare正a， P， Salvador， A．＆Fiszman， S．（2007＞Cha且ges in apple tissue with storage
time：rheological， textural and miGrostnK加ral analyses． Journal of Food
Engineer童ng，78（2）：622－629．
Wang， C Y＆Singh， R．P（1978）． A single layer drying equation for rough／rice． ASAE
Paper No：78－3001， ASAE， St． Joseph， M［．
Wang， Z， SunJ・， Chen， E， Liao， X＆Hu， X．（2007）． Mathematical modelling on
thin layer microwave drying of apple pomace with and without hot air pre－drying．
Journal of Food Engineering，80（2）：536－544．
Watson， E L．＆Harper， J．C（1988）． Elements of Food Engineering（2nd ed．）． New
Yb rk：AVI．
Zhang， MI．N＆Willison， J且M（1991）． EIectrical impedance analysis in pIant
tissues：a double shell model． Journal ofExperimental Botany， 42（244），1465－75．
Zhang， MI．N．＆Wiilison，工H．M（1992， a）． Electrical impedance analysis in plant
tissues：加vivo detection of丘eezing injury． Canadian Journal of Botany，70，
2254－58，
Zhang， MIN．＆Willison， J．H．M．（1992， b）． Electrical impedance analysis in plant
tissuesl the effect of freeze－thaw injury on the electriGal properties of potato
tuber and carrot root tissues． Canadian Journal of Plant Science，72，545－553．
Zbang， MI．N．， Repo， T．， WilIison， J．H．M．＆Sutinen， S．（1995）． Electrical impedance
analysis in plant tissues：on the biological meaning of Cole－Coleαin Scots pine
needles． European Biophysics Joumal，24，99－106、
Zhang， M．1．N，， Stout， D．G＆Willison， J．H．M，（1990）． Electrical impedance analysis
in plant tissues：symplasmic resistance and membrane capacitance in the Hayden
model． Joumal ofExperimental Botany，41（224），371－380．
Zhu， S．， Le Bail， A．， Ramaswamy， H．S，＆Chapleau， N（2004）． Characterization of
ice crystals in pork muscle forrned by pressure－shift freezing as compared with
lO7
㌔
classical freezing methods． Journal of food Science，69（4）：190－197．
Zogzas， N．P．， Maroulis， Z，B．＆Marinos－Kouris， D．（1994）． Density， shrinkage and
porosity of some vegetables during air drying． Drying TechnLology，12（7）l
l653－1666．
Zywica， R， Pierzynowska－Komiak， G＆W61cik，」．（2005）． Application of fbod
products electrical model parameters fbr evaluation of＆pple pur6e dilution．
Joumal of Food Engineeri血g，67（4），413－418．
108
ACKNO’VVLEDGMENTS
IwouId li1（e to express my deepest appreciation to my advisor and committee chair，
Professor Akio Tagawa， fbr his encouragement， patience， guidance and expertise during
this research． Without him， I could not have completed my studies in Chiba University． His
technical and moral supports are deeply acknowledged．
Iam very gratefhl fbr the assistance， generosity， and advice I received f士om Dr． Yukiharu
Ogawa． Throughout my doctoral work he continually stimulated my analytical thinking
and greatly assisted me with scientific writing．1 am al so gratefU1 for having an exceptionaI
doctoral committee and wish to give my gratitude to Professor Hiroki Nakagawa，
Professor Eiji Goto and Dr． Nobuhiro Matsu。ka fbr their helpfUl suggestions and
assistance with this dissertation．
1 would like to acknowledge the Japanese Ministry ofEducation， Culture， Sports， Science
and Technology（MEXT）fbr providing me with the opportunity to study in Japan． I am
also gratefヒl to Dr． Nobuhiro Matsuoka and Professor Li Lite of China Agricultural
Un董versity for their efforts during the application of MEXT scholarship．
Iextend ma且y thanks to the faculty and friends who have helped me to complete my
dissertation．工would like to express my gratitude to these individuals， especially to Dr．
のTakahiro Orikasa and Mr． Taizo Nagasuna， fbr their support and asslstance．
Finally， 1 am deeply indebted t・my parents f・r their。。ntinu・us and unc・nditi。nal physical
and em。ti。nal supP・rt． l w。uld never have c。mpleted this dissertati。n with。ut the supP°「t
and love of my family over the years．
109