OODSIM'2010 BRAGANÇA, PORTUGAL· JUNE 24-26, 2010

JUNE 24-26, 2010

CIMO RESEARCH CENTRE BRAGANÇA, PORTUGAL

ORGANIZED BY

EU'Dpron r«h".I~I1I"HI.""

lN COOPERATION WITH

.AiL

1TiiIT UNIVERSITEIT

GENT

6TH INTERNATIONAL CONFERENCE

ON

SIMULATION ANO MOOELLlNG

lN THE

FOOO ANO BIO-INOUSTRY

2010

FOODSIM'2010

EDITED BY

Vasco Cadavez

and

Daniel Thiel

JUNE 24-26, 2010

BRAGANÇA,PORTUGAL

A Publication of EUROSIS-ETI

Printed in Ghent, Belgium

II

6th International Conference on Simulation and Modelling

ln the Food and Bio-Industry

BRAGANÇA, PORTUGAL

JUNE 24-26, 2010

Organised by

ETI The European Technology Institute

CIMO Mountain Research Centre, Bragança

Sponsored by

Cámara Municipal Bragança

CoraNE

CVRTM

EUROSIS The European Simulation Society

Ghent University

INATEL Fundação

IPB Instituto Politécnico de Bragança

Escola Superior Agraria

Mirandela Municipio

Hosted by

CIMO

Bragança, Portugal

III

EXECUTIVE EDITOR

PHILIPPE GERIL (BELGIUM)

EDITORS

General Conference Chair

Vasco Cadavez Mountain Research Centre (CIMO)

ESA - Instituto Politécnico de Bragança, Campus de Santa Apolónia, Apartado 1172, 5301-855 Bragança,

Portugal

General Program Chair Daniel Thiel , University of Paris 13, Sciences Economiques

et Gestion , Paris, France

Local Programme Committee Joana Amaral, ESTiG-IPB, Bragança, Portugal Vasco Cadavez, ESA-IPB, Bragança, Portugal Elsa Ramalhosa, ESA-IPB, Bragança, Portugal

International Programme Committee

Simulation in Food Engineering and Processing Lionel Boillereaux, ONIRIS, Nantes, France

Cedric Brandam, INPT-ENSIACET, Toulouse, France Matthew de Roode, NIZO, The Netherlands

Jean-Yves Monteau, ONIRIS, Nantes, France Alcina Morais, ESBUCP, Porto, Portugal Cristina Silva, ESBUCP, Porto, Portugal

Olivier Vitrac, INRA-UMR FARE, Reims, France

Simulation in Food Sciences and Biotechnology Francis Buller, UCD, Dublin , Ireland

Enda Cummins, UCD, Dublin, Ireland Ricardo Dias, ESTiG-IPB , Bragança, Portugal

Carla Fernandes, ESTiG-IPB , Bragança, Portugal António Peres, ESA-IPB, Bragança , Portugal

José Teixeira, UM, Braga, Portugal

Methods and Tools Applied to Food and Bio-Industries Kristel Bernaerts, Katholieke Universiteit, Leuven, Belgium

Paulo Cortez, UM, Guimarães, Portugal Pierre-Sylvain Mirade, INRA-Theix, St Genes Champanel le, France

LeQn Rothkrantz, Delft University of Technology, Delft, The Netherlands Xavier Serra, IRTA, Girona, Spain

IV

International Programme Committee

Methods and tools applied to Food Quality and Safety Evaluation Manuel Rui Alves, IPV, Viana Castelo, Portugal Joana Amara!, ESTiG-IPB, Bragança, Portugal

Ursula Gonzales Barron, University College Dublin, Food Science and Veterinary Medicine, Ireland

Vasco Cadavez, ESA-IPB, Bragança, Portugal Luis Dias, ESA-IPB, Bragança, Portugal

Isabel Ferreira, ESA-IPB, Bragança, Portugal Isabel Mafra, REQUIMTE, Porto, Portugal Beatriz Olivei ra, FF-UP, Porto, Portugal

José Alberto Pereira, ESA-IPB, Bragança, Portugal Isabel M. Vica riao, University of Sevilla, Spain

Miguel Vilas-Boas, ESA-IPB, Bragança, Portugal

Simulation in Functional Foods Lourdes Amigo Garrido, CSIC, Madrid, Spain

Gianpaolo Ruocco, University of Basilicata, Potenza, Italy

Simulation in Food Production Management, Economics and Traceability Martin Cloutier, UQAM, Montreal, Canada

Magda Aguiar Fontes, Faculdade de Medicina Veterinária , UTL, Lisboa, Portugal Martin Grunow, Technical University of Denmark, Lyngby, Denmark

Vincent Hovelaque, INRA, Rennes, France

Sustainable Food Production Albino Bento, ESA-IPB, Bragança, Portugal

Cesar de Prada, University of Valladolid, Spain Federico Ferreres, Cebas-CSIC, Murcia, Spain

Bjórn Johansson, Chalmers University of Technology, Sweden Jaime Pires, ESA-IPB, Bragança, Portugal

Manuel Ângelo Rodrigues, ESA-IPB, Bragança , Portugal

Innovation in Traditional Food Products Maria J. Alcaide , University of Sevilla, Spain

Leticia Estevinho, ESA-IPB, Bragança, Portugal Dietrich Knorr, Technische Universitat, Berlin , Germany

Toomas Paalme, Tall inn University of Technology, Tallinn, Estonia Elsa Ramalhosa, ESA-IPB, Bragança, Portugal

v

© 2010 EUROSIS-ETI

Responsibility for the accuracy of ali statements in each peer-referenced paper rests solely with the author(s). Statements are not necessarily representative of nor endorsed by the European Simulation Society. Permission is granted to photocopy portions of the publication for personal use and for the use of students providing credit is given to the conference and publication. Permission does not extend to other types of reproduction nor to copying for incorporation into commercial advertising nor for any other profi t-making purpose. Other publications are encouraged to include 300- to 500-word abstracts ar excerpts from any paper contained in thi s book, provided credits are given to the author and the conference.

Ali author contact information provided in this Proceedings falis under the European Privacy Law and may not be used in any form , written or electronic, without the written permission of the author and th e publisher. Infringements of any of the above rights will be liable for prosecution under Belgian civil or criminal law.

Ali articles published in these Proceedings have been peer reviewed

EUROSIS-ETI Publications are ISI-Thomson and INSPEC referenced

For permission to publish a complete pape r write EUROS IS, elo Philippe Geri!, ETI Executive Director, Greenbridge NV, Wetenschapspark 1, Plassendale 1, B-8400 Ostend Belgium

EUROSIS is a Division of ETI Bvba, The European Technology Institute , Torhoutsesteenweg 162, Box 4,8-8400 Ostend , Belgium

Printed in Belgium by Reproduct NV, Ghent, Belgium Cover Design by Grafisch Bedrijf Lammaing , Ostend, Belgium

EUROSIS-ETI Publication

ISBN: 978-90-77381-56-1 EAN: 978-90-77381-56-1

VI

r(s). ·n is it is ling are ded

t be 1er.

00

PREFACE

Dear colleagues,

Welcome to the 6th Inlernational Conference on Sirnulation and Modelling in Food and Bio Industries (FOODSIM'2010), which is held in Bragança, Portugal from 24 to 26 June 2010.

The FOODSIM'2010 brings together researchers, food experts and industrial users to present the state-of-art sirnu lation research in the food industry, new research results and lo exchange ideas and experiences about the modeling and simulation tools used in the food industry.

The main theme of FOODSIM'201 O is: "Sirnulation applied to food processes, quality, safety, and suslainability", and the success of the conference is already assured, as can be witnessed by the quality and scientific rigor of the 47 published papers. We also take Ihis opportunity to challenge the researchers altending lhe FOODSIM'20 10 to produce a seed for a FP7 project to be submilted at the nexl call for pro posais, which is expected to open nexl July.

We present our recognition for the inestimable collaboration that we had in the FOODSIM'2010 organisation by Prof. Joana Amaral and Prof. Elsa Ramalhosa , and io ali the reviewers for their professional work in the papers evaluation. We also presenl our recognition to ali Institutions Ihat contributed to prepare a pleasant social programme for FOODSIM'2010.

Finally, we wish you ali a pleasant staying in Bragança and we are sure thal you will have the opportunity to be delighted by the Portuguese hospitality.

Prof. Vasco Cada vez, Moutain Research Centre (CIMO), ESA - Instituto Politécnico de Bragança , Bragança, Portugal

Prof. Daniel Thiel , Universite de Paris 13, UFR Sciences Economiques et Gestion, Paris, France

VII

PHYSICAL CHARACTERiSTlCS AND DRYI NG KlNETiCS OF PORTUCUES E 'Longal' CHESTNUT

Elsa Ramalhosa Hugo Lamas

Jose Albe rto Pereira

C IM O, Escola Superior Agrária, Instituto Politécnico de Bragança,

Campus de Santa Apolónia, Apartado 1172, 5301-855 Bragança, Portugal

E-mail : e lsa@ipb.pt; aI9759@alunos.ipb.pt; jpereira@ ipb.pt

KEYWORDS Chestnut, Physical charactcristics, Drying kinetics, Fick's second law, Apparent diffusivity.

ABSTRACT

Chestl1ut is a fruit of great impOItance in Portugal, being generally sold fresh or frozcn. Alternative products may be obtained by hot air drying. The presem work is 00 lhe dehydrati on behavior af Portuguese <Longal' chestnul, lhat is the 1110s1 used in industry, Diffcrent models for representing the variations af \Valer content anel drying rate along time were tested successfully. As expectecl, higher temperalures corrcspond to faster drying processes. lhe apparcnt diffusivity was predicted by Fick ' s second law equalion, and ii ranged from 1.25x I O-II n,';s, at 20"C, to 8.42x I O-III m' /s, at 100'C.

INTROO UCT ION

Chestnut (Casfal1ea sativa) has always becn used in human

feeding. Nowadays, chcstlluts have high eCOnOl11lC

importance in Portugal because this country is a11 important

exporter of th is fmit , contribllting significant ly to lhe

cqui librium of Ponugucsc commercia l balance. CheSllll1t

production is more significant in the North of Portugal,

namely in Trás-os-Montes and Beiras regions, reaching the

highest commercial values and representing 86% of the

national product ion (INE 2008). ln 2007, for example, 7

774 ton ofchestnut wcre exported, originating € 14 844 000.

Portuguese variet ies of chestnut are of excel!ent quality.

Three Protected Origin Dcnominations have been

identified, namcly ChestnuI of lcrm Fria, Chestllut of

Soutos da Lapa and C hestnu t of Padre ia. Bcing a seasonal

product, lhe chestnut needs to be prescrved in order to be

sold as a fresh produce, 01' to be further processed.

Traditionally, chestnuts are left in warehollses at room

84

Alcina 1VI .IVI.Bernardo 1VI0rai s

CBQF/ Escola Superior de Biotecnologia, Univers idade Católica Portuguesa,

Rua Dr. António Bernardino de Almeida, 4200-072 Porto, Portugal

E-mail: abmorais@esb.ucp.pt

temperature, being cons1I111ed along lhe year by lhc local

families. This chestnut is known as '"castanha pilada"', being

dehydrated anel extreme ly sweet. On the contrary, in

industry chestnuts are stored under refrigeration until lhcir

salc as fresh producc, or they are frozen after pee ling.

ln industrial chestnut prescrvation, two main problems have

been detected. These are related to weight losses (duc to

dehydration of the Iresh ehestnut) and to the dcvelopment

of microrganisllls, namely fungi produccrs af micotoxins,

such as aflatoxins: which present an eminent dangcr for

public health, as the latter are considcred can.:inogenic,

hepatotoxic, and tcratogenic. Besides, there is a need of

finding altcrnatives for chestnuts with low quali ty to the

fina l cons u111ers, such as sll1all fruits, tha t are OftC Il

discarded, and of developing innovative produets based on

chestnut, allowing the diversification of products and the

developrnent of others able to be incorporated imo a wide

var iety of food stuff.'i, sllch as chestnllt fiour. T herefore, ii is

Df great importance to find alternative prcservalio ll and

processing technologies.

laking imo account lhe popular knowledge, the

dehydralion of chestl111t seems to be a promising

technology, as chestnll ts cOllld bc storcd fo r longe r periods

withollt the problem of [ungi development. Rccently, some

illtcrcsting stlldies on drying of chestnut have been

perfomled, involving Ncw Zealand (Cletus anel Carson

2008), Italian (Attanasio ct aI. 2004), Turkish (Koyuncu et

aI. 2004) and evell Portuguese (Guiné and Fernandes 2006)

varieties. ln relation 10 this last work, three Portllguese

varieties were analyzed, namely ' Longal', 'Martainha' and

·Viana·. which were subjected to thl'ce drying temperaturcs;

70, 80 and 90°C. The firs l lwo variet ies showed better

drying features than the last. Nevcrtheless, the use of lower

temperatures may also be promising. Moreover, variability

of the characteristics af chestnuts harvestcd in different

l

years might have an impol1ant role on Iheir drying

behaviour a nd be a factor thm musl be taken into account.

ln fact ~ chestnut prodm:ers refer that the yea r 2009 was not

a good year fo r chestnut production. due to the lack of

waler during the maturat ion stage, orig inating sma!! fru its .

Taking into account these aspects, the aims of thi s work \Ve re : i) lO determine some physical properties of ' Long:1l' chestnuts , which is one af lhe mosl used in industry (due to the eas iness of peeling and to the pu lp not bcing too sectioned). collcered in the year 2009: ii) to s tudy and finei e mpirical 1l10dels lhat \Vi II a llow lhe est imation of lhe drying rates Df these chestnuts when subjected to different drying tcmperatures (20, 40, 50, 65 , X5 and 100°C); iii ) to evaluate the eftect of drying 011 chestnuts dimcns ions; iv) to evaluate the aelequacy of the Fick 's second law to express lhe dry ing process; v) to estimale lhe apparem diffusivity of water in this Portuguese chcslnul variely.

MATERIALS ANO METHOOS

P0I1uguese <Longal' cheslnuts \Vere supplied by a local producer of Trús-os-Montes regian, North of POrlugaJ , being harvestcd in Nove mber 2009. When aniving 10 the laboratory, chestlluts were carefully observed and the ones lha! were rotten were discarded.

Chestnuts were individu"lI y weighcd "mi their elli psoid axis dimensions ((I: length; h: width; c: thickncss) determined. To perfonn lhe drying exper iments, chestnuts \Vere placed in a convection oven (Binder, Germany) at 40, 50 , 65, 85 aml 100°C. To perform the assay aI 20"C, chestnuIS were left ai room tempera ture. ln each experiment, chestnuts \Vere put insidc the oven in several Petri di shes. Each Petri dish cantained 5 c hestnu ts. One Petri dish was removed ai defined time moments, the fru its being ana lyzed aga in (weighl anel e llipsoid aX IS

d imensions) \V ith and w ithout outer pceI, in order to evaluate the shrinkage.

At the beginning of each experiment, the dry matter content o f the chestnuts used in lhe assay was dctermined by drying fi ve e hesl llll ts ai 105°C 1I1lli l lhe samplc weight reached a steady value. The dry-basis moisture conlcnt (W, kg af \Vater! kg of dry matter) was determined for eaeh chestnut. At each time moment , the ave rage , sta ndard deviation anel cocffic ient of variation we re detc rmined for lhe fi ve chestnuls removed.

ln order 10 evaluate lhe sui'tability of approaching the chestnuts to sphercs in lhe diffusion eq uation, the diameter of chestn uts \Vas dc rcrmined a ftcr immersion of twenly ehestnuts in \Vater and measure ment o f the vo lume of \Vate r disp laced. These values \Ve re late r compared with lhe arilhmetic (Da) aml geometric (D.IJ average di amcters detennined by equat ions ( I ) and (2), respectively (M ohseni n 1970; Güner 2007; KJ!lçkan and Gliner 2(08 ):

85

a+b+ c Do = , -,

( I )

Dg = Va X b X C (2)

ChestnuI sphericily (CP) was a lso evalualcd by equation (3) (Mo hsenin 1970; Güner 2007; Klhçkan and Güncr 2(08):

Vax b xc (/) = (3) b

, considering h lhe biggest dimensiono

ORYING CURVES COMPUTER FITTINC

For a li lhe assays, lhe drying curves (W versus lime)

fa ll owed a s imi lar partem, expressed by í.ln exponemial

function ofthe forl11 :

W = y+ A x eH n (4)

whcrc y , A anel k were parameters of the mode!.

ln o rde r lO determine the d rying rales ( a;~ ). the method

of lhe approximatiol1 of the derivat ive to ti nite differences was used (Guiné and Fernandes 2006) :

For t = I {)

(a; )= (first -order forward finitc

diffe rcnces) (5)

For / = / , (i = I . ... , N-I )

(aw)= W'+I - W,-I aI ' i +I - ! i - 1

(second-o rdc r centred finitc

di fte rences) (6)

For t =t,v

(aw) = W,v - WN _1

aI 'N - t N- I

(first -order backwa rd fin ite

diffe renees) (7)

A lincar and a s igmoid func lioll (Eq . 8) \Ve re used to

ex press lhe relat ionship e ncountered betwee n thc d rying

rate an el lhe l110isture content.

_(a:)=y+ (~l I +e ii

(8)

Ali the modcls presented, with exception of the linear, \Vere

obtained after using the nan-linear regression af the SPSS

software package.

The quality af ali modc1s was evaluated by the conelation

coefficicnt (r), reduced chi-squarc (Z2), mean bias error

(MBE) and root mean square errar (RMSE) (Guiné and

Fernandes 2006), determined by the following equations:

l N , I( ), - - -- V . -V -X - N . exp.1 f1r",i.1 - 11 1=1

(9)

MBE-_If(v-v) - ~ pred.1 cxp.i

N i= 1

(10)

RMSE= (lI) l N

_'\'(V . - V ), N f1 pred./ cxp.i

where Ve'\JJ.I and VI"-L'd,i are the experimental and predicted

values for the observation i; N is the numbcr af

observations; and 11 the number af parameters in the model.

The higher the vaI ue of"r" and the lower the values of/.

MBE and RMSE, the better the model fitted the

experimental results.

RESULTS AND DISCUSSION

Physical Properties of Portuguese 'Longal' Chcstnuts

Table I shows some physical properties Df the 'Longal '

chestnuts used in al1 drying assays. The dimcnsions of

ellipsoiJ axis ar chestnuts (a, b, c) varied within the ranges of 2.62-2.82, 3.10-3.29 anel 1.69-1.9 1 cm, respectively,

corresponding to variatian coeffícients less 1ha11 7.9, 6.4

anel 16.4%. The arithmetic anel geometric mean diameters

of chestnuts varied between 2.47-2.66 anel 2.39-2.58 cm,

respectively, corresponding to variation coefficients less

than 6.3 anel 7.1%, respectively. The low values obtained

for the variation coefficients~ \Vith the exception af

dimension c, that was the most difficult to measure, are

indicative of the existence of homogeneity among the

chestnuts used in ali assays.

ln terms of chestnut sphericity, this parameter varied within

the range of 76.9 to 79.2%, indicating that chestnuts might

be approximated to spheres.

86

When comparing the dia meter (D, results not shown)

detennined by immersion in water (considering chestnuts as

spheres) with D" or Dp" detennined by equations (12) and

( 13), respectively, linear relationships were encountered

between both parameters, namely:

Da = 0.7707 X D + 0.7304 (1=0.922) ( 12)

D, = 0.8139 X D + 0.5451 (1= 0.916) ( 13)

These reslllts indicate that either Dg or Da may be llsed to

estimate the diameter af 'Longal' chestnuts when it is

intended to approach them as spheres.

Drying curves of Portuguesc 'Longal' chestnuts

The experimental points dctcrmined along the drying of

'Longal' ehestnuts at 20, 40, 50, 65, 85 and 100°C are in

Figure 1. ln each graph the prcdicted points obtained in the

fits by lIsing Eq. 4 are also presented, as well JS the model

cqllation and the correlation coefficient.

Figure 1 shows that the fíts to lhe experimental data at 20,

40, 50, 65, 85 and 100°C are good, with eorrelation

coefficients ranging from 0.989 to 0.998. Thc 1..2, MBE and

RMSE values were also low, varying between 3.05x 1 ()"I

aml 9.87xIO·", _6.77xI0·4 and 1.69xI0·2 and 1.59x10·' anel

2.79xI0·',respectively.

Thc drying rate vallles determined by derivation of Eg. 4

are in Figure 2, with correlation coeffícients ranging from

0.863 to 0.969. The quality of lhe model is quite good

taking into account the similarity with the experimental

data, and the 'l, MBE and RMSE values, which were low,

between 5.43xI0·H and 5.28x I 0 .. 1, -2.49xI0·' and 3.71xI0·3,

2.21 x I 0-4 anel 2.07x 1 0.2, respectively.

By observing the dtying rate curves obtaineel, no constant

rate period was observeel (Figure 2). Higher temperatures

corresponded to taster drying processes, which is reflected

on the incrcasing dehydration constants: 0.003 for 20°C,

0.019 for 40c C, 0.029 for 50°C, 0.062 for 65°C, 0.142 for

85°C. aJ1(1 0.189 for 100°C.

ln terms of the relation between drying rate curves anel

moisture contcnt, both linear and sigmoidal (Eg. 8)

functions seem to represent well the experimental data

(Table 2). The cOITelation coeffieients of both models

varied between 0.902-0.956 and 0.927-0.967, respectively.

The i, MBE anel RMSE valucs were low again, varying

between 5.27xl O·H and 3.86x 1 0.4, -2.57xHr' and 1.86x I 0.3

,

l

aml 2.1 7x I0·' and l.77x I O", for the li near model, and

belween 4.09x I 0.6 and 2. 76x I O", -1.30x 1 O·' and 1.50x 1 0.3,

and 1.84x I 0.3 anel 1.50x I O", for sigmoidal model,

respecti ve ly.

Tablc 1: Phys ical properties or'Longal' chestnuts used in the dry ing assays'.

Temperature of a b c t he assay (0C) (cm) (cm) (cm)

20 2.75+0.20 3.20+0.15 1.7 1+0.28 40 2.72+0.20 3.23 +0. 19 1.86+0.27

50 2.62+0. 17 3. 10+0. 16 1.69+0.22 65 2.76+0.20 3.26+0.21 1.9 1 +0.27 H5 2.67+0.21 3. 16+0. 18 1.74+0.24

100 7.82+0. 17 3.29+0. 17 1.88+0.29 + Results me lhe dverages of the detcrmlnatlOll s...:.... Stcl ndard devi3tlon.

1.0

O,S o

1 , o.c, ~.

;: o., ~ ~~ ;: 0.2

:: 0.0

-0.2

"

1.0

I o., o

1 0.6 j ,' ~

OA

~

;: 0-1

:: 0,0

I -0.2 , o

Drying Curve at 20"C

o E ~pt:rimcnt~ l U;IW

• Modcl

---~---~

!

lllU 2I)() 3011 400 suo T imc( hours)

Drying C urves a t SO"C

".'"", ~.I= O.O .. UI+O. 765 *('(·O.Ql~. ' ''''1"'1 , - 0.997

600

r f f ! o E .'r~rimcntal dat;l

. I .... loocl

~ t ~ ~ • ! ! f t , ,

211 60 ' 00 l ::'.tJ

T ime (hollrs)

1.0

O.R

1 o., ,. ~ ;: 0.4 , ; " ;: 0_2

:: 0.0

·0.2 100

1.1)

0'

1 , O.,

,1 ~

" 0.4 ,

i 0.2

:: 0.0

..IJ.2 14ft

I •

O

I • i i

o

D" DK (cm) (cm)

2.55+0. 15 2.46+0. 17 2.60+0.16 2.53 +0. 18 2.47+0. 12 2.3 9+0. 13 2.64+0. 15 2.57+0. 16 2.53 +0.16 2.45+0. 17 2.66+0. 15 2.58+0. 16

Oryi ng C urve at 40"C

t

,o

\V " ,nol. l " U.112+0.620*c-1·0.0 1'" ""'P"I r = IJ .9H9

50 ' 00 15U

T illle (hulIrs)

Dr)' ing Cur'ves at 65UC

\" ••. ,1,1 = 0.024+U.757·,,1·0.1ó1 . , .. ", •• ) r "' U.993

f § •

20 lO 41' T im c- (holl rs)

(P (%)

76.9+6.0 78.5+4.4 77. 1+4. 1 79.2+4.6 77.4+4.0 78.7+4.6

o [ XflCTIIllcnt al dala • Mod c-I

2()()

o E~ pcrilllc nt al dat ;)

• Modcl

• !

'O

250

Drying Curves at 85°C Orying Cnrves at IOO"C 1.0 - --- --- -'-- - - - - - - - - - - ---, 1.0

O.S O

• , 0.6 ! • ,

" ~

OA

• g 0.2

:: (LO

-0.2

O

W """,,1= 0. 11 (o.j.0.687*('1 ·0, I') 'l<mpo)

r = 0.994

f

'O 15 Ti rnc(hou rs)

o Experimental data

• ~l (l dc l

• 'I f

,o 25

1 • O.s -100

~ 0.6

i , , .1

OA

:. 0.2

0 .0

.().2 _

O

' . •

\ " ",,~,.I == U. 03:Õ+O.1I05*el·0.11O • '''''I"') .·= 0.996

• •

" T im"lhuurs)

Figure l: Drying data of ' Longa l' chestnuts at 20, 40,50,65, 85 and 100°C.

87

o [r;pcrimcl1tal data • Mudei

l O 12

0.0025

:.. 0.0020 o ;' • o :. . ~ i 0.00 15

, 0 .00 10 " " ~ 0.0005 .,

0.0000

Drying Rate at ZO"C versus Time

o

•• o

-1I\\'/dl",,,,ld = 0.0018*l,I·I,U.1 ' "o'p")

r - 0.92 1

i.

o Exp~ri !l1c n t;l 1 d:\ln

' MoJcI

o ... °oo~" ••• ,~

o o o o

0.111 5

I • 11.0 ' 0 I • o

IUlOS

0 ,01111 I

-o.uns

Drying Rate at 40"C versus Time

-d\Vlul,,",<ld '" 0.0 I 18*cl·U.U1". '''''I'''' r - O.9 IH

o o

o • • • o o o • o o o

o o o

o EX1,~rimcnt;L l dal:t.

' Modcl

o

o

" III{) :wn 31KI 5l1li 7011 " sn ,ou ,so 2011 250

O.()~S

-=,i:: O.o:w o D

·t ~. 0.01 5

!!: , (1,1110

:t (1 ,005

~ O.IKMI ., ·O.()O~

"

Time(h,,"")

Drying Rate at 50"C versus Time

20

- d\V/dl ~,,"'1 '" O.0222*l~·8.V : ~ , '~'~I""

r = O.9~

• o • o

o

o E~p~nm~ntnl ti nia

, Mod cl

~ • c: o D • ~ e

o o o

<lO CiO !lO Timc lh"urs)

120 140

Drying Rate at 85"C versus Time 0. 12 ----::.......:::-------- - - ------,

:::: O.HI -=: • f, o •

O,O!!

-d\\'/dl,,",d .-l '" o. o~n6* c" '· ' ~:· "" "1"'1

r oo 0.91'19

~ ~ ~.

" ~ '" ~ ~

~ .,

(l ,050

11.0411 o ~ o

(UI)U

(),O ~(I

n.Ol 0

II .OUO

· O.OIU

" 0. 1(,

• ,. ;::: 0. 14 • ,

-=~ o' r 0.12 100 o , o

Timc(huIIN)

Drying Rate at 6SnC versus Time

o

-d\\'ldt .... d,1 '" 0.0-169*,,1 · •.•• 1, "'~I··)

1" - 11 .1) 12

o o

o o

o

o

"' 20 30 ·10 Tim" (hn urs)

o Exp.:ruIlCIII ;11 IIJ ta

• ~'I ()dd

o o

o

,,,

Drying Rate at IUO"C versus Time

o .(I\"/uln,o~.I "" O.l;;Z~~" 'O. ,~" " .. ~) r "" U.1-I63

6"

" O' é o. "

. o

o Experimental dnla

' Modd ~. 0. 10

1," o 1 ·.~ p",nl1lclII:t.1 lI:t.l:l o , (1.114

" ~ 0.02

~ (U)O

-0.02

o

• o • •

o o

lU " 20 25 Timc (lloufli)

~ O.UK , ~ 11.01'1

~ 0.04 ~ "1 0.U2

n.OO II

• o

4

o

o

o

o

6 , Tim" l llours)

Figure 2: Drying rates of 'Longal' chestnllts at 20, 40, 50, 65, 85 and 100' C.

• tI.'l odd

o o

o • lU "

Table 2 : Linear and s igmoidal models for the drying rates in function of moisture contento

Model T("C) 1110l1eJ Pal"U11lel erS Corre/fIliou Coe{ficielll (r)

Linear mo dei 20 A ~ -0.0005; B ~ 0.0032 0.924

(8Wi 40 A - -0.002; B - 0.01 83 O.90X - \& J = A +B x W 50 A - -0.0011: B - 0.0277 0.952

65 A ~ -0.0004; 8 ~ 0.058 0.906 85 A - -0.0 114; 8 - 0.1266 0.956 100 A - 0.0089; B - 0.1447 0.902

Sigmoidal model (Eg. 8) 20 Not adequate --

(8W ) A 40 Not adequate --50 y- -O.OOI ; A - 0.023: 13 - 0.158; C - OA02 0.957 - & = y+ [ C-IV ) 65 Y ~ 0.006; A - 0.03 7; 8 - 0.059; C - 04 00 0.930

I + e H ) 85 y - -32.935 ; A - 33.963 ; 8 7.167; C - -24.724 0.967 100 Y - 0.034; A - 0.086; 8 - 0.059; C - OA27 0.927

88

I·'

Evaluatioll of the effect of drying on chestnuts

dimen sions

Drying at 40, 50. 65, 85 and 100"C did not cause signifi cant

changes in chestnuts dimensions, indieating thar the effect

of shr inkage may be neglected. ln fac t, lhe average

geometric diamcters of the chestnuts \Ve re ident icaJ at tbe

beginning and at the end of a I! assays (Table 3). Even ar high tcmperaturcs no significant reduction on ches tnuts size

was deteeled. Moreover, the fru it wilhout outer shell a fter

dry ing corresponded always to 76.5 - 80 .2% o f lhe frui l

wirh olltcr shell, no differences being detected among

drying expcrimcnts at different temperatures.

Evaluation of the adequac)' of Fick's second law to the

drying process

Assuming that chestnuts may be approximated to sphcres,

and considcring that lhe dry ing rate depeneis solely OIl the

moisture movement by diffusion witl1 in the chestnut, the

process might bc represented by lhe fo llowing Fick's

second law equation fo r non-s leady state:

aW. = D (a ' w+~xaw ) ( 14) aI "P ar' r ar

where D ilI' is lhe apparenl di t'fusivity, r is the radius and t is

time. Assuming uniform initial tlloisture content (Wo) and

that the internal mílSS transfer resiSlance is controlling oveI'

external resistancc, the analyti cal solution of Eq. 14 is the

fo llowing:

w- w ,. ( IS)

w -w o ,

where WL

• is the equili br ium dry-basis moisture contenl.

Considering tl1at only lhe first term is significant and W"

may be estimated by Eq. 4 for t--+", (W,. = y) , taking the

natura l logarithm Df each side of Eq. IS, the fo llowing

linear equat ion will be obtained:

(1 6)

89

Fitti ng lhe da la by Eq. 16, linear corre lations \Vere obtained

(Figure 3), wi th coefficient corre lat ions ranging between

0.947 and 1.000. These resulls indicale Ihat the assumpl ions

assull1cd seem to be adequale.

Estimation of the apparent diffusivitics of ' Longal'

chestnut variety

Taking into accoullt Eq. 16, lhe apparenl diffusivities of

Longal chestnu t may be determined by the following

equation:

s/opex R' Ir '

( 17)

Considering the slope for each temperature (Figure 3) and

the average radi us 01' lhe chestnuts, based e ither on DI! or

Da, lIsed in each assay, lhe apparenl diffusivitics obtained

for ' Longal' cheSlnut are in Table 4.

Table 4: Apparent diffusivi ties of water in 'Longa l'

chestnut.

Temperature D a p

("C) (m'/s) 20 1.25x10·' 40 6.53x I O'" 50 1.08x I O~ '"

65 2.89x I (r'" 85 5.24x I0·'" 100 8.42xI0·' "

W hen comparing thesc va lues with lhe ones repol1ed by

Gui né anel Fernandes (2006) for the same chestllut va ricty

dri ed at 70-90'C (4.45xHr" to 6.87x Hr' m' /8), lhe prescnt

results are one order of magni tude lower. T his might be

related to the \Vo rse quality o f lhe chestnuls uscd in the

assays, as the year 2009 was not adequate for a good qua li ty

of the nuts. So, in the future it woul d be interest ing to study

the ro le of the harvest year on the drying charactc ri stics of

chestnuts. Howevcr, lhe results determincd at 20-401JC

(1.25xIO· 1I to 6.53x IO· 1I Ill~/s) are in accordance to the one

determineel for lhe New Zealand chestnut vari ety "" 10 15" at

30'C (Cletus and Curson 2008), numely 5.l x IO·" m'/s.

Table 3: A verage geometric diameters of chestnuts detennined before and aftel" the drying assays.

Temperature Dg initial Dg final Dg jina] Dg final (without outer

" "

O.DO

-0.50

~ -1.00

~~ , -1.50

S ~

-2.00

o

("C)

40 50 65 85 100

'00

(with outer peel)

20D

2.50+0.22 2.40 +0.13 2.50 +0.19 2.39 + 0.22 2.54 : 0.10

20"C Time(hunrs) 300 400

(with outer peel)

2.45 + 0.19 2.34 + 0.11 2.40 + 0.18 2.32 + 0.23 2.43 : 0.09

500 flUO 700 o 0.00 +-

(without outer peel) peel) I Dg final (with outer peel) ("lo)

I. 93 + 0.22 78 .8 1.79 + 0.14 76.5 1.91 +0.15 79.6 1.86 + 0.2 1 80.2 1.89 + 0.10 77.8

40"C T iml'(hours)

50 100 150 200 250

. + +

- 1.00 I In(IW.W,. ~'(WwlV. l ) = -0,0032 t· 0.022 1 r " () ,,)9]

+ +

++ +.

+ ••••

+

-2.00 j • +

-3.00

-4.00

+

IIl ((W-Wd(W,,-IV,_J) = ·0.0156 I - 0.211 r = O.947

+

+ +

+ +

+

+

-2.50 '---------- ----------------" -5.00 _~ _ __________________ ..J

o 0.00

- I,O()

~ 6 -2.0U ::: g

-J.OO :: !

--"t.DO

-5.DO

o 0.00

-0.50

~~ -1 .00

i -1.50

" [ ·2.00

~ ·2.50

-3.00

-3.50

lO

+

50"C Time (huurs)

(,0 !lO 100 120

In( IV_\V,)/ IW" W, I) =- -U. (21)() 1+ 0.0251 r = O.'lX6

+ +

+ +

+ +

+

85"C Time (hours)

10 15 20

In((W-W,)I(W,, -W,)) = ·0.136 t - 0.0113 r = 1.000

140

+

25

o 0.00 ..

-1.00

~ -2.00

~ ; ·3.00

-4.(J()

65"C T illll'(hours)

10 20 30 40 50 (,0

-~------~--~--,

+ I,, \(W.W,).'(\\',, _W.)I = -O.()(i67 t + O,07ú4

.. r = 0.%2

+ + + +

+

+ +

+

-5.00 ------------------------"

0.00

-0.50

~ t -1.00

~ :::;; -1.50

~ -= -2.00 ~

-2.50

-3.00

o 4

100"C Time (hours)

III

II1{(W-W,}' I\~·.,_W. 1l = -0. 1')1·1 t ·! 0.0191 r · 0,990

+ +

+ +

+ +

" 14

+

Figure 3: In((W-WJ/(W,,-WJJ versus lime for the drying assays at 20, 40, 50,65,85 anel 100°C.

90

l

CONCLUS IONS

Thcre was homogenei ty al11Qng the chcstnuts used in the

dryi ng experiments. An exponc ntial empirical model fittcd

we ll the experimental drying data for severa I temperatures

between 20 0 e and iOO°C. Linear and sigmoid fun ctions

al50 fiU cd well lhe drying experimental rates.

Approximating chestnurs (sphericity being highcr than

77% ) to spheres, the apparcnt diffusivity predicted by

Fick's sccond law was bctwecn 1.25x I 0.11 m2/s. ar 20°C, to 8.42x 10'11> 01' /5, at 100"C

REFE REN CES

Atwnas io, G.; l. Cinquanta; D. Albanese; and M.D. Mattco 2004. "Effects of drying tC01peralurcs 011 physico-chemical propcrt ies af dried and rchydrated chestlllJls {Castanea salivar', Food Chemisuy. 88, 583-590.

Cletus, A.S.; and J.K. Carson 2008. "DI)'illg curves and apparcnl diffusivity of Ncw Zealand chcs tnut variety " tOI5"". .1olll"l/al of Foud Engil1eering. 85, 38 1·386.

Guiné. R.P.F.; aml R.M.C. Fel11undcs 2006. "Analys is 01' lhe drying kinctics 01' chestnuls" . JOl/mal of Food Engilleerillg, 76. 460-467,

Giiner. M. 2007. " Pneumatic convcying characteristics of some agricultural seeds" . JVI/mal af Food Engillcering, 80, 904-9t3 ,

INE 2008, "Estatisticas Agricolas 2007" (ISSN 0079-4 139), Klllçkan. A.; and M. Güncr 2008. "Physical propcrties and

mcchanicul behavior of olivc frllits (Dica cl/ropaca L.) under compress ion loading". Jmmwi of Fooel Eng;nccrillg , 87. 222- 228.

Koyuncu , T.; U. Serdar; anel I. ToslIn 2004. " Oryi ng charactcristics and energy requircment for dehydration of cheSlnuts (Cas/aneo SlU; I'C/ Mill )", )0/(1"1101 t?f Food

Engill(!cring, 62. 165-168. Mohscnin, N.N. 1970. Physical Propcr/ies oI Piam al/d Anill/al

Mafericlis. Gordon i.!nd Brcach Sciencc Publishcrs, Inc .. Nc\V York.

BIOGRAPHY

ELSA RAMALHOSA was bom in Oporto, Pot1ugaL She studied in College of Biotechno logy - Pot1uguese Catho lic Uni versity in Oparto, whcrc in 1995 she obtained hcr B.Se. dcgree in Food Engineering. Afterwards , she took her Ph.D. in Analytica l C hemistry in University af Aveiro, Portuga l. Since 2000, she is professor a f Transporl Phenomena at Instituto Politécnico de Bragança, Portugal.

HUGO LAMAS \Vas bom in Paços de Ferreira, Pot1ugaL l-Io studied at the Secondary School of Paços de Ferreira, where in 2007 he obtained his diploma. C urrcntl y he <lttcnds lhe 2nd Semeste r a f the 3rd yea r Df Faoel Engineering with success at 1nstituto Politécnico de Bragança, Portugal.

91

JOSÉ ALBERTO PEREIRA obtained his B,Sc, degree in Agricultural Engineering fram Universi dade de Trás-os­Montes e Alto Douro, his M.Sc. in Qua lity Contrai fmm Faculdade de Farmácia of Uni vers idade do Porto and Ph.D. in Agricultural Sc iences fram Univers idade de Trás·os· Montes e Alto Douro. C urrently he is Professor at the Instituto Politécnico de Bragança, Portuga l. Hi s \Vo rk area compri ses the effec t of agricultural pract ices on qualily am! chemical composition of vcgetablc fooel proclucts. Hc is CD­

author of 80 papers in re fereed international j ourna ls, integrate the edi torial board of fÚtlr refereed journals and has participated in several projects anel supervised different Ph.D ., M.Sc. and undcrgraduate students.

ALCINA M. M. BERNARDO MORAIS currently holds an appoi ntment as Coordinato r of Faod Engineering of the College of Biotechnology - Portuguese Cathol ic U niversily (ES B - UCP), She got her B, Sc, degrec in Chemical Engineering from University of Porto, Portugal , in 1985; her M, Sc" in 1986, aml her Ph ,D" in 1990, in FooJ Process ing fmm Ecole Nat ionalc Supérieu re des Industries AgricoJ es et Alimentaires o f Massy - France. ln 1990. she became Assista nl Professor at ESB . ln 1991 , she attc nded a Postharvest Technology course at Univers ity of California 111 Davi s - Californi a, EUA, and she pursucd a postdoctoral programme at Uni versity of Florida 111

Gainesville - Florida, USA, From 1999 to 200 1. she held an appaintment as Associate Sub-director af lhe Extension of the College of Bioteehnology in Caldas da Raínha (EESBC R - UC P), She is cO-lI mhor 01' 38 papers 111

rcfereed internati onal joumals. She is authorl cO·í:lUthor of 16 papers in non-peered in ternational journals, and 12 papers in national journals and technica l publicat ions. She has served as National Del egate for on EC COST Action, as member af corresponding Managing COlTIm ittee, and as Nationa l Coordinator for a EU CYTEO Act ion. She is responsible for the module Fruits and Vegetables af the EraslTIus Mundus Maste r Course 111 Fooel Scicnce. Technology and Nutri tion (SEFOTECR NUT), She hus been involved as a responsible paTineI' a r coordinator of national and inte rnation,,1 R&D projects (total of 5), Furtherl11ore, she supervised 8 externally funded research fe llowships, including 3 successfully cOllcluded Ph.O. theses and 2 M. Sc . theses. She a lso supcrvisec1 2 Post­Ooctoral programmes. She delivered 6 invited lectuTcs and 62 volunteered presentatio ns in professio nal meet ings \Vorldwide, has served in the a rganizati on or sc iemifíc committces of 5 national ~lI1d international symposia, aml has acted as invited revlewer of several refereed international journals.