Báo cáo khoa học hạt bí Đỏ

Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (190.94 KB, 8 trang )

Trang 1<div class="page_container" data-page="1">

Eﬀect of drying methods on thin-layer drying characteristicsof hull-less seed pumpkin (Cucurbita pepo L.)

Keywords: Hull-less seed pumpkin; Solar tunnel drying; Hot air drying; Moisture ratio; Eﬀective diﬀusivity; Activation energy

1. Introduction

The drying technique is probably the oldest and themost important method of food preservation practiced byhumans. The removal of moisture prevents the growthand reproduction of microorganisms which cause decay,and minimises many of the moisture-mediated deteriora-tive reactions. It brings about substantial reduction inweight and volume, minimizing packing, storage and trans-portation costs and enables storability of the product underambient temperatures (Mujumbar, 1995). During dryingmany changes take place; structural and physic-chemicalmodiﬁcations aﬀect the ﬁnal product quality, and the qual-ity aspects involved in dry conversation in relation to thequality of fresh products and applied drying techniques(Baysal, Icier, Ersus, & Yıldız, 2003). Currently hot air dry-

ing is the most widely used method in post-harvest ogy of agricultural products. Using this method, a moreuniform, hygienic and attractively coloured dried productcan be produced rapidly (Doymaz, 2004). However, it isan energy consuming operation and low-energy eﬃciency,so more emphasis is given on using solar energy sourcesdue to the high prices and shortage of fossil fuels. Solardryers are now being increasingly used since they are a bet-ter and more energy eﬃcient option. The solar dryers couldbe an alternative to the hot air and open sun drying meth-ods, especially in locations with good sunshine during theharvest season (Pangavhane, Sawhney, & Sarsavadia,2002). However, large-scale production limits the use ofthe open sun drying. Among these are lack of ability tocontrol the drying process properly, weather uncertainties,high labour costs, large area requirement, insect infesta-tion, mixing with dust and other foreign materials and soon (Basunia & Abe, 2001).

technol-Solar drying is essential for preserving agricultural ucts. Using a solar dryer, the drying time can be shortened

Tel.: +90 312 596 15 92; fax: +90 312 318 38 88.E-mail address:

www.elsevier.com/locate/jfoodeng

</div>Trang 2<div class="page_container" data-page="2">

by about 65% compared to sun drying because, inside thedryer, it is warmer than outside; the quality of the driedproducts can be improved in terms of hygiene, cleanliness,safe moisture content, colour and taste; the product is alsocompletely protected from rain, dust, insects; and its pay-back period ranges from 2 to 4 years depending on the rateof utilization. The most important feature of solar dryers isthat the product does not include any kind of preservativesor other added chemical stuﬀs, which allows its use for peo-ple suﬀering from various allergic reactions from chemicalpreservatives and other added stuﬀs. Furthermore, theproduct is not exposed to any kind of harmful electromag-netic radiation or electromagnetic poles (Tiris, Tiris, &Dincer, 1996). Although for agricultural products, solardryers with solar air heater oﬀer better control of requireddrying air conditions, solar tunnel dryers based on plastictunnel greenhouses have a great potential and does notrequire any other energy during operation. Therefore, solartunnel dryer may become a more convenient alternative forrural sector and other areas in which electricity is scarceand in regular supply. Also, it can reduce crop losses,improve the quality of dried product signiﬁcantly and iseconomically beneﬁcial compared to traditional dryingmethods.

Sun shines in Ankara, situated in Middle AnatolianRegion in Turkey, over an average 2466 h/year, deliveringabout 1525 kWh/m2year of solar radiation on the horizon-tal surface. Hours of sunshine and solar radiation betweenJune and September, namely drying period in Ankara,make up about 46.59% and 49.64% of these values, respec-tively (Sacilik, Keskin, & Elicin, 2006). Though othersources of energy may be used for drying of agriculturalproducts, solar energy is preferred more and more sinceit is abundant in Ankara, inexhaustible and non-polluting.It can be tapped at relatively low cost and has no associ-ated environmental dangers (Basunia & Abe, 2001;Mohamed et al., 2005).

Pumpkin seed is of considerable nutritional value forhuman consumption due to its 37.8–45.4% oil and 25.2–37.0% protein. It enjoys valuable dietetic and medicinal

advantages besides being a source of edible oils, proteinsand minerals of good quality (Yoshida, Shougaki, Hirak-awa, Tomiyama, & Mizushina, 2004). Hull-less seed pump-kin (var. styriaca) or naked seed pumpkin is grown widelyin the southern regions of Austria (Styria province) and theadjacent regions in Slovenia and Hungary. The pumpkincultivated in Styria has a high content of green seeds with-out husks. The seeds itself can be eaten as a snack andshow good results in curing prostate. Pumpkin seed oilextracted from the seed is used widespread as salad oil(Murkovic & Pfannhauser, 2000). Recently, the hull-lessseed pumpkin have been grown in the some parts of Tur-key, notably Nallıhan province, Ankara. The eﬃcient pro-cessing and long-term storage of pumpkin seed requiresthat the moisture content be reduced to suitable levels byvarious drying methods. To the knowledge of the author,there is no literature speciﬁc to the drying behaviour ofhull-less seed pumpkin found. Therefore, the present studywas conducted with the following objectives:

(1) to study and compare the thin-layer drying istics of hull-less seed pumpkin using the open sun,solar tunnel and hot air drying methods; and(2) to ﬁt the experimental data obtained to semi-theoret-

character-ical models widely used to describe thin-layer dryingbehaviour of agricultural products.

2. Material and methods2.1. Material

The hull-less seed pumpkins used in this study wereobtained from a local grower of Nallıhan, Turkey duringthe summer season of 2003. In order to preserve its originalquality, they were stored in a refrigerator at 4C until dry-ing experiments. The initial moisture content of seed wasdetermined using the vacuum oven method at 70C for24 h (AOAC, 1990). These experiments were replicatedthrice to obtain a reasonable average. After drying, theNomenclature

a, b, c coeﬃcients in modelsDeﬀ eﬀective diﬀusivity, m2/sD0 pre-exponential factor, m2/sEa activation energy, kJ/mol

EMD mean relative percent deviation, %ERMS root mean square error

H half-thickness of the slab in sample, mk, k0 drying rate constants in models, l/hm exponent in drying model

M moisture content at any time, kg [H2O]/kg [DM]Me equilibrium moisture content, kg [H2O]/kg [DM]M0 initial moisture content, kg [H2O]/kg [DM]

MR dimensionless moisture ratio

</div>Trang 3<div class="page_container" data-page="3">

sample was found to have a moisture content of about 67%dry basis (d.b.).

2.2. Experimental set up

The experimental set ups used for determining the ence of various drying methods on the thin-layer dryingbehaviour of hull-less seed pumpkin are presented inFigs.1 and 2. The description of the laboratory scale hot air andsolar tunnel dryer used in present study was described indetail elsewhere in Sacilik et al. (2006) and Sacilik andElicin (2006), respectively.

inﬂu-2.3. Experimental procedure

The hot air drying experiments were conducted at 40, 50and 60C air temperatures and a constant air velocity of0.8 m/s. In each experiment, about 100 g of pumpkin seedsamples were used. After the system was run for at leasthalf an hour to reach steady conditions for the operationtemperatures, the samples were evenly distributed withinthe sample tray as a single layer and dried there. Moisturelosses of samples were recorded at 10 min intervals for ﬁrstone hour and 20 min subsequently thereafter for determi-nation of drying curves. Drying was continued until no fur-ther changes in their mass were observed. The driedsamples were allowed to cool down at an ambient temper-

ature for 15 min and then packed in low-density ene bags.

polyethyl-Three sets of the solar tunnel drying experiments werecarried out during the periods of August–September 2003under the climatic conditions of Ankara. Each experimentstarted at 08:00 am and continued till 06:00 pm. The testsamples were uniformly spread on wire mesh tray the load-ing density of which was about 1.5 kg/m2. To determine themoisture loss of drying samples during experiments, pump-kin samples were taken from three points, namely inlet,middle and outlet of the solar tunnel dryer and weighedat various time intervals, ranging from 30 min at the begin-ning of the drying to 1 h during the last stage of the pro-cess. The moisture loss of samples was determined withthe help of a digital electronic balance having an accuracyof 0.01 g. After 06:00 pm, the pumpkin seeds in the solartunnel dryer were collected and placed in plastic boxes inorder to induce fermentation and diﬀusion of moisturewithin the drying samples. These were again spread in thedryer in the next morning and the drying process was con-tinued until no further changes in their mass wereobserved. Also, to compare the performance of the solartunnel dryer with that of open sun drying, control samplesof pumpkin seeds were distributed on a tray at the sameloading density near the solar tunnel dryer. Both experi-mental and control samples were dried simultaneouslyunder the same weather conditions.

2.4. Analysis of drying data

The experimental drying data obtained were ﬁtted to thefour well-known drying models given inTable 1. The mois-ture ratio is given as follows:

MRẳ M MeM0 Me

1ịwhere MRis the dimensionless moisture ratio, M, MeandM0are the moisture content at any time, the equilibriummoisture content and the initial moisture content in kg[H2O]/kg [DM], respectively.

However, MR is the simpliﬁed to M/M0 instead ofEq. (1) due to the continuous ﬂuctuation of the relativehumidity of the drying air during their drying processes(Diamente & Munro, 1993). The drying rate constantsand coeﬃcients of models were estimated using a non-linear regression procedure. The estimation method was

Fig. 1. Experimental set up of laboratory dryer: (1) centrifugal blower, (2)air heating chamber, (3) drying chamber, (4) perforated ﬂoor, (5)electronic balance, (6) holding wire, (7) sample tray, (8) sensors, (9) Pc,(10) door.

Fig. 2. Schematic representation of the solar tunnel dryer.

Table 1

Thin-layer drying models given by various workers for drying curves

PageMR= exp(ktm

Munro (1993)Henderson

</div>Trang 4<div class="page_container" data-page="4">

Levenberg–Marguardt and the statistical validity of modelswas evaluated and compared by means of the coeﬃcient ofdetermination R2, mean relative percent deviation EMD,root mean square error ERMS and reduced chi-square v2.These comparison criteria methods can be calculated asfollows:

ERMS¼ 1N

where: MR,ex,i is the ith experimental dimensionless ture ratio; MR,pre,iis the ith predicted dimensionless mois-ture ratio; N is the number of observations; and z is thenumber of constants.

mois-R2 was used as the primary comparison criteria forselecting the best model to ﬁt the four models to the exper-imental data. Also, a model is considered better thananother if it has a lower value of the EMD, ERMS, v2.3. Results and discussion

3.1. Hot air drying of hull-less seed pumpkin

The moisture content versus drying time curves for hotair drying of hull-less seed pumpkin as aﬀected by variousair temperatures are shown inFig. 3. The pumpkin samplesof average initial moisture content of around 0.67 kg[H2O]/kg [DM] were dried to the ﬁnal moisture contentof about 0.04 kg [H2O]/kg [DM] until no further changesin their mass were observed. It is evident from these curvesthat the moisture content decreases continuously with thedrying time. As expected, the air temperature had a signif-icant eﬀect on the moisture content of samples. Duringthe hot air drying experiments, the time to reach the ﬁnal

moisture content for samples were found to be 9.0, 7.5and 6.0 h at the air temperatures of 40, 50 and 60C,respectively. The increase in the air temperature resultedin a decrease in the drying time. Other researchers havereported similar trend (Akpinar, Bicer, & Yildiz, 2003a;Ertekin & Yaldiz, 2004).

3.2. Solar drying of hull-less seed pumpkin

Fig. 4shows the variations of the ambient air ture, relative humidity and solar radiation during the solartunnel and open sun drying of hull-less seed pumpkin for atypical day of September 2003 in Ankara. During the dry-ing experiments, the weather was generally sunny and norain appeared. The daily mean values of the ambient airtemperature, relative humidity and solar radiation changedfrom 21.6 to 39.7C, 12.1% to 51.5% and 205.1 to796.2 W/m2, respectively. The drying air temperature andrelative humidity in solar tunnel dryer varied continuouslyfrom morning to evening. The ambient air temperature andsolar radiation were reached the highest ﬁgures between12:00 and 15:00, whereas the relative humidity was reachedthe lowest ﬁgures during this time. The diﬀerence betweenthe drying air temperature and ambient temperature wasobserved to be the highest during this time. In other words,inside the solar dryer, it is warmer than outside. Thisclearly indicates that the drying rate in the solar tunnel dry-ing would be higher than open sun drying.

tempera-Fig. 5suggests drying curves for hull-less seed pumpkindried by solar tunnel and open sun drying methods. Theinterruptions of the lines in this ﬁgure represent the nightperiods of the drying process. The pumpkin samples ofaverage initial moisture content of around 0.67 kg [H2O]/kg [DM] were reduced to the ﬁnal moisture content whichchanged between 0.05 and 0.07 kg [H2O]/kg [DM]. It isclear fromFig. 5that the moisture content decreases con-tinuously with the drying time. During the experiments,the time to reach the ﬁnal moisture content of samples

Fig. 3. Drying curves for hull-less seed pumpkin under hot air dryingcondition at indicated air temperatures.

08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00Drying hours (h)

Fig. 4. Variations of ambient air temperature (j), relative humidity (d)and solar radiation (m) with the drying hours for a typical day ofDecember 2003.

</div>Trang 5<div class="page_container" data-page="5">

for solar tunnel were found to be between 27 and 30 h,while the drying time for the open sun drying changedbetween 32 and 34 h. Solar tunnel dryer had a shorter dry-ing time than the open sun drying. In other words, dryingtime was reduced to about 17.9% by the solar tunnel dryeraccording to the open sun drying. Depending on weatherconditions, the solar tunnel dryer developed shortened halfday the drying time of hull-less seed pumpkin samples. Thedecrease in the drying time could be attributed to the valuesof higher temperature and lower relative humidity obtainedin dryer. Similar results have been reported by Schirmer,Janjai, Esper, Smitabhindu, and Muăhlbauer (1996); Balaand Mondol (2001); Bala, Mondol, Biswas, Das Chowd-ury, and Janjai (2003).

3.3. Determination of eﬀective diﬀusivity and activationenergy

The eﬀective diﬀusivity of the samples is estimated byusing the simpliﬁed mathematical Fick’s second diﬀusionmodel. The solution of Fick’s second law in slab geometry,with the assumptions of moisture migration being by diﬀu-sion, negligible shrinkage, constant diﬀusion coeﬃcientsand temperature was as follows (Crank, 1975):

MR¼ M MeM0 Me¼ 8

2n ỵ 1ị2 exp

2n ỵ 1ị2p2Defft4H2

5ịFor long drying periods, Eq.(5)can be further simpliﬁed toonly the ﬁrst term of the series and the moisture ratio MRwas reduced to M/M0 because Me was relatively smallcompared to M and M0. Then, Eq. (5) can be written inlogarithmic form:

ẳ ln 8p2 p

6ịwhere H is the half-thickness of the slab in sample in m, n isa positive integer and D is the eﬀective diﬀusivity in m2/s.

The eﬀective diﬀusivity is typically calculated by plottingexperimental drying data in terms of ln(MR) versus dryingtime. From Eq.(6), a plot of ln(MR) versus the drying timegives a straight line with a slope of

ð8Þwhere D0 is the pre-exponential factor of the Arrheniusequation in m2/s, Eais the activation energy in kJ/mol, Ris the universal gas constant in kJ/mol K and Ta is theabsolute air temperature in K.

The activation energy was calculated by plotting the ural logarithm of Deﬀversus reciprocal of the absolute tem-perature as presented inFig. 6. The plot was found to be a

nat-0.00.20.30.50.60.8

</div>Trang 6<div class="page_container" data-page="6">

straight line in the range of air temperatures studied, cating Arrhenius dependence. Then, the dependence of theeﬀective diﬀusivity of hull-less seed pumpkin on the tem-perature can be represented by the following equation:Deff¼ 1:95 105exp 3987:51

ð9ÞThe activation energy for hull-less seed pumpkin wasfound to be 33.15 kJ/mol, which is within the range of15–40 kJ/mol for various foods reported byRizvi (1986).3.4. Fitting of the drying curves

Tables 3 and 4list the estimated parameter and ison statistics of four drying models for the hot air andsolar tunnel drying of hull-less seed pumpkin, respectively.All the four models for hot air drying gave an excellent ﬁtto the experimental data with a value for R2of greater than0.9931. Of all the models tested, the two-term modeloﬀered the highest value for R2, followed by the logarith-mic model. However, the values of EMDfor the logarithmicmodel were less than 10% in all cases, which is in theacceptable range. Also, the values for ERMS and v2

compar-obtained from this model were less than those attainedfrom other models. Therefore, the logarithmic model wasconsidered the best model in present study to representthe hot air drying behaviour of hull-less seed pumpkinwithin the experimental range of study.

For solar tunnel drying, all models other than theHenderson and Pabis model provided an adequate ﬁt tothe experimental data with a value for R2of greater than0.9888, indicating a good ﬁt. The values of EMDfor the log-arithmic and two-term model were less than 10%, which isin the acceptable range. However, the two-term model wasrejected due to its higher value for ERMSand v2despite itshigh value for R2. Hence, the logarithmic model may beassumed to represent the thin-layer solar tunnel dryingbehaviour of hull-less seed pumpkin.

Figs. 7 and 8 suggest comparisons of the experimentaland predicted moisture ratio obtained using the logarith-mic model for the hot air and solar tunnel drying, respec-tively. It can be seen from these there was a goodconformity between experimental and predicted moistureratios. This indicates the suitability of the logarithmicmodel in describing the drying behaviour of hull-less seedpumpkin in the hot air and solar tunnel drying process.

Table 3

Parameter estimation, R2, EMD, ERMSand v2of the four drying models for the hot air drying at an air temperature of 40C

Estimated parameters and comparison criteria of the four drying models for solar tunnel drying

</div>Trang 7<div class="page_container" data-page="7">

than the open sun drying and resulted in saving toextent of about 17.9% of drying time. In addition,the samples of solar tunnel dryer were completelyprotected from insects, birds, rain and dusts.

(3) The eﬀective diﬀusivity varied from 8.53 to17.52· 1011m2/s in the air temperature range of40–60C. The activation energy was found to be33.15 kJ/mol.

(4) The value of eﬀective diﬀusivity for the solar tunneland open sun drying process were found to be1.94· 1011and 1.66· 1011m2/s, respectively.

(5) Of all the four models tested, the logarithmic modelgave an excellent ﬁt to the experimental dataobtained with a value for R2 of greater than 0.99for the hot air and solar tunnel drying process.

I am grateful to Mr. Rıfat Orkan from Orkan tural Products Company in Nallıhan, Ankara, for his assis-tance related to providing hull-less seed pumpkin.

AOAC (1990). Oﬃcial method of analysis. Washington, DC: Associationof Oﬃcial Analytical Chemists (No. 934.06).

Bala, B. K., & Mondol, M. R. A. (2001). Experimental investigation onsolar drying of ﬁsh using solar tunnel dryer. Drying Technology, 19(2),427–436.

Bala, B. K., Mondol, M. R. A., Biswas, B. K., Das Chowdury, B. L., &Janjai, S. (2003). Solar drying of pineapple using solar tunnel drier.Renewable Energy, 28(2), 183–190.

Basunia, M. A., & Abe, T. (2001). Thin-layer solar drying characteristicsof rough rice under natural convection. Journal of Food Engineering,47(4), 295–301.

Baysal, T., Icier, F., Ersus, S., & Yıldız, H. (2003). Eﬀects of microwaveand infrared drying on the quality of carrot and garlic. European FoodResearch and Technology, 218(1), 68–73.

Crank, J. (1975). Mathematics of diﬀusions (2nd ed.). London: OxfordUniversity Press.

Diamente, L. M., & Munro, P. A. (1993). Mathematical modelling of thethin layer solar drying of sweet potato slices. Solar Energy, 51(4),271–276.

Doymaz, I. (2004). Drying kinetics of white mulberry. Journal of FoodEngineering, 61(3), 341–346.

Ertekin, C., & Yaldiz, O. (2004). Drying of eggplant and selection of asuitable thin layer drying model. Journal of Food Engineering, 63(3),349–359.

Henderson, S. M. (1974). Progress in developing the thin-layer dryingequation. Transactions of the ASAE, 17, 1167–1168/1172.

Maskan, A., Kaya, S., & Maskan, M. (2002). Hot air and sun drying ofgrape leather (pestil). Journal of Food Engineering, 54(1), 81–88.Mohamed, L. A., Kouhila, M., Jamali, A., Lahsasni, S., Kechaou, N., &

Mahrouz, M. (2005). Single layer solar drying behaviour of Citrusaurantium leaves under forced convection. Energy Conversion andManagement, 46(9–10), 1473–1483.

Mujumbar, A. S. (1995). Handbook of industrial drying (second ed.). NewYork: Marcel Dekker.

Murkovic, M., & Pfannhauser, W. (2000). Stability of pumpkin seed oil.European Journal of Lipid Science and Technology, 102(10), 607–611.

Pangavhane, D. R., Sawhney, R. L., & Sarsavadia, P. N. (2002). Design,development and performance testing of a new natural convectionsolar dryer. Energy, 27(6), 579–590.

Rizvi, S. S. H. (1986). Thermodynamic properties of foods in dehydration.In M. A. Rao & S. S. H. Rizvi (Eds.), Engineering properties of foods.New York: Marcel Dekker.

Sacilik, K., & Elicin, A. K. (2006). The thin layer drying characteristics oforganic apple slices. Journal of Food Engineering, 73(3), 281–289.0.0

Drying time (h)

T=60

˚

C, experimentalT=50

˚

C, experimentalT=40

˚

C, experimentalLogarithmic model

Fig. 7. Comparison of the experimental and predicted moisture ratioobtained using the logarithmic model for hot air drying in the airtemperature range of 40–60C.

Fig. 8. Comparison of the experimental and predicted moisture ratioobtained using the logarithmic model for solar tunnel and open sundrying.

</div>Trang 8<div class="page_container" data-page="8">

Sacilik, K., Keskin, R., & Elicin, A. K. (2006). Mathematical modelling ofsolar tunnel drying of thin layer organic tomato. Journal of FoodEngineering, 73(3), 231–238.

Schirmer, P., Janjai, S., Esper, A., Smitabhindu, R., & Muăhlbauer, W.(1996). Experimental investigation of the performance of the solartunnel dryer for drying bananas. Renewable Energy, 7(2), 119–129.

Tiris, C., Tiris, M., & Dincer, I. (1996). Energy eﬃciency of a solar dryingsystem. International Journal of Energy Research, 20(9), 767–770.

Tog˘rul, _I. T., & Pehlivan, D. (2003). Modelling of drying kinetics of singleapricot. Journal of Food Engineering, 58(1), 23–32.

Westerman, P. W., White, G. M., & Ross, I. J. (1973). Relative humidityeﬀect on the high temperature drying of shelled corn. Transactions ofthe ASAE, 16, 1136–1139.

Yoshida, H., Shougaki, Y., Hirakawa, Y., Tomiyama, Y., & Mizushina,Y. (2004). Lipid classes, fatty acid composition and triacyglycerolmolecular species in the kernels of pumpkin (Cucurbita spp) seeds.Journal of the Science Food and Agriculture, 84(2), 158–163.

</div>