- Email: [email protected]
Moisture Sorption Isotherms of Red Chillies
ARTICLE IN PRESS Biosystems Engineering (2004) 88 (1), 95–104 doi:10.1016/j.biosystemseng.2004.01.003 PH}Postharvest Technology
Available online at www.sciencedirect.com
Moisture Sorption Isotherms of Red Chillies S. Kaleemullah1; R. Kailappan2 1
Department of Agricultural Engineering, SV Agricultural College, Tirupati 517 502 India; e-mail of corresponding author: [email protected] 2 Department of Agricultural Processing, Tamil Nadu Agricultural University, Coimbatore 641 003 India (Received 23 June 2003; accepted in revised form 27 January 2004; published online 24 March 2004)
The equilibrium moisture contents were determined for red chillies using the static method at 25, 35 and 458C over a range of relative humidities from 0115 to 0865. The sorption capacity of chillies decreased with an increase in temperature at constant relative humidity. The sorption isotherms exhibited the phenomenon of hysteresis, in which the equilibrium moisture content was higher at a particular equilibrium relative humidity for desorption curve than for adsorption. Ten models were applied for analysing the experimental data. The equilibrium relative humidity of chillies can be predicted by using the modified Halsey and Kaleemullah models for adsorption process whereas the modified Oswin, Kaleemullah and Oswin models can be used for desorption process of chillies. As there was no common model to predict the equilibrium relative humidity of chillies during both the adsorption and desorption processes, the Kaleemullah model that was ranked second in both the adsorption and desorption processes for chillies can be considered and used with good predictive accuracy. # 2004 Published by Elsevier Ltd on behalf of Silsoe Research Institute
1. Introduction
of grains. Labuza (1975) pointed out that no sorption isotherm model could fit data over the entire range of relative humidity because water is associated with the food matrix by different mechanisms in different activity regions. Boquet et al. (1978) compared various twoparameter equations for their accuracy in describing the isotherms of various foods. They concluded that the Henderson equation could adequately represent the isotherms of fruits and starchy foods whereas the Halsey equation is more suitable for meats, milk products and vegetables (Table 1). Chirife and Iglesias (1978) reviewed 23 isotherm equations, both theoretical and experimental and their use for fitting sorption isotherms of foods and food products. None of these equations described accurately the sorption isotherm over the whole range of relative humidity and for different types of food materials. Chen and Morey (1989a) evaluated four EMC–ERH equations for their ability to fit data for 18 cereal grains and seeds and found that no universal equation could be established to fit all isotherms. Pandey and Aich (1989) stated that Smiths’ equation (Table 1) gave the best fit over the entire range of relative humidity and temperature for dehydrated
In studying the behaviour of food or agricultural commodities during storage and processing, it is important to have information, which relates environmental conditions to the material properties. Equilibrium moisture content (EMC) is one such property and is defined as the moisture content of a hygroscopic material in equilibrium with a particular environment (temperature and relative humidity). The EMC has already been used in determining storage ability at various conditions (Labuza, 1968), prediction of drying time (King, 1968) and in general for the efficient design and operation of drying systems (Young, 1976). The hysteresis effect is the difference between desorption and adsorption EMC values for the same equilibrium relative humidity (ERH). Sorption isotherms for a food product over the full range of water activity form a hysteresis loop in which the desorption curve lies above the adsorption curve (Shatadal & Jayas, 1992). A number of theoretical, semi-theoretical and empirical isotherm equations have been developed to model the relationship between EMC or ERH and temperature 1537-5110/$30.00
95
# 2004 Published by Elsevier Ltd on behalf of Silsoe Research Institute
ARTICLE IN PRESS 96
S. KALEEMULLAH, R. KAILAPPAN
Notation a; b; c; d df ei Em Es HR M
dimensionless coefficients of the models degrees of freedom of regression model randomness of residual mean relative deviation, % standard error of estimate equilibrium relative humidity, decimal equilibrium moisture content, % d.b.
N R2 Ro Rp t
number of data points coefficient of determination observed equilibrium relative humidity, decimal predicted equilibrium relative humidity, decimal temperature, 8C
Table 1 Isotherm models for fitting experimental data Name of the model
Equation
Oswin (1946) HR ¼ Smith (1947) Halsey (1948) Henderson (1952) Day and Nelson (1965)
Modified Henderson (Thomson et al., 1968) Modified Halsey (Iglesias & Chirife, 1976) Modified Chung and Pfost (Pfost et al., 1976)
1 þ ðM=aÞ1=b
HR ¼ 1 expfða MÞ=bg HR ¼ expfa=ðtM b Þg HR ¼ 1 expðatM b Þ d
HR ¼ 1 expðatb M ct Þ HR ¼ 1 expfaðt þ bÞM c g HR ¼ exp½fexpða þ btÞg=M c
aexpðbMÞ HR ¼ exp ðt þ cÞ
Modified Oswin (Chen, 1988) HR ¼ Kaleemullah (Kaleemullah, 2002)
ðM=aÞ1=b
1 ½fða þ btÞ=Mgc þ 1
HR ¼ a b expðct M d Þ
HR , equilibrium relative humidity as a decimal; M, equilibrium moisture content as a % d.b.; t, temperature in 8C; a,b,c,d, dimensionless coefficients specific to individual equation.
mushroom. Oswin, Smith, Halsey and Henderson equations (Table 1) described well the moisture adsorption isotherms of ground turmeric at low temperatures but became less satisfactory at higher temperatures. The relative mean square root value of EMC of turmeric powder with the above equations was recorded as 516–
983% at 15–258C and 724–1520% at 35–358C (Pawar et al., 1992). Menkov (2000) reported that the modified Oswin model (Table 1) was the most suitable for describing the relationships between equilibrium moisture content, relative humidity and temperature. Soysal and Oztekin
ARTICLE IN PRESS ISOTHERMS OF RED CHILLIES
(2001) evaluated seven equilibrium moisture content equations for their ability to fit data for some medicinal and aromatic plants. They found the modified Halsey and the modified Oswin equations (Table 1) to be as the most versatile models for medicinal and aromatic plants; the modified Henderson equation (Table 1) is good for fennel and cinnamon. The modified Chung–Pfost equation (Table 1) was considered as the best model to explain the EMC–ERH relationship of rough rice (San Martin et al., 2001). Red chillies are the ripe fruits of the species of plants in the genus Capsicum. The most important quality characteristics of chillies are the colour and pungency. The chillies produced for condiment and culinary supplements are subjected to long–term storage. During that time, important physiochemical and biological changes take place with a strong impact on the colour and pungency. Therefore, it is necessary to investigate the equilibrium moisture content of chillies for various relative humidities and temperatures to enable the storage conditions for the chillies to be correctly specified. The automatic control of these conditions requires a reliable mathematical description of the EMC–ERH using suitable models. The objective of this work was to obtain the equilibrium moisture isotherms for red chillies at 25, 35 and 458C and to fit suitable models for describing the sorption characteristics.
2. Materials and methods 2.1. Experimental procedure Freshly harvested and dried red chillies of the polyster variety (Coimbatore local) were procured from the local market. Equilibrium moisture content was determined both in desorption and adsorption conditions at three temperatures (25, 35 and 458C) and eight levels of relative humidities (0115–0865) using the static method. Various supersaturated salt solutions of different relative humidities viz., lithium chloride (0115–0121), potassium acetate (0184–0227) magnesium chloride (0318–0332), potassium carbonate (0432–0438), sodium nitrite (0605–0643), sodium chloride (0748– 0756), ammonium sulphate (0794–0803) and potassium chromate (0846–0865) were used to provide constant relative humidities at 25, 35 and 458C temperatures. A preliminary experiment was conducted to determine the range of expected values of equilibrium moisture content by storing samples of chillies in the environment corresponding to 458C–0115 relative humidity (RH) and 258C–0865 RH for a period of 30
97
days. The range of equilibrium moisture content was 3–50% dry basis (d.b.). Hence, the samples for adsorption and desorption had to be pre-conditioned. The samples were conditioned to 75% d.b. for desorption purpose and to 25% d.b. for adsorption purpose. The samples were conditioned by drying the samples in a hot air vacuum oven at 708C till the desired moisture contents were obtained. Another preliminary experiment was conducted with conditioned samples to determine the realistic time required for attaining the state of equilibrium. After 10 days, the samples were weighed once in 2 days till a constant weight was observed. It was found that a period of 30 days was required for attaining equilibrium. Hence, a 40-day period was selected to provide a margin for variations in the time to achieve moisture equilibrium. Glass desiccators (16 cm diameter by 16 cm height) containing about 350 ml saturated salt solutions were used to provide constant relative humidities varying from 0115 to 0865. The desiccators were placed in an electric oven at a desired constant temperature and allowed to equilibrate with the environment inside the containers. Samples of pre-conditioned chillies (10 g each for desorption and 5 g each for adsorption) were placed in petri dishes and these dishes were placed on a plastic platform inside the desiccators to avoid direct contact between chillies and salt solution. After a 40-day period, the petri dishes were removed from controlled atmosphere space and stored in activated silica gel environment for 3 h to allow them to attain room temperature (Verma & Gupta, 1988). The moisture content of each sample was determined by keeping the samples in a vacuum oven at 708C and 4100 mmHg pressure (AOAC, 1995) until the weight measurement was constant. All the experiments were replicated three times and the averages of three values were used in the analysis. 2.2. Model fitting Ten isotherm models were selected for fitting the experimental data of adsorption and desorption isotherms of red chillies. These models have two, three or four parameters and the parameters of the models were estimated by using SPSS (Statistical Package for Social Science) 75 software. The models that have been used in this study are listed in Table 1. 2.3. Comparison methods The suitability of the models was evaluated and compared using the mean relative percentage
ARTICLE IN PRESS 98
S. KALEEMULLAH, R. KAILAPPAN
ei ¼ Ro Rp
ð2Þ
40
20
0
0
0.2
ð3Þ
where: Ro is the observed ERH; Rp is the predicted ERH from the model; N is the number of data points; and df is the degree of freedom of regression model (number of data points minus number of constants in model). Data points in a plot of the residual values versus the predicated values should tend to fall in a horizontal band centred on zero and displaying no systematic tendencies towards any clear pattern. If the residual plots indicate a clear pattern, the model should not be accepted (Chen & Morey, 1989a; Menkov, 2000). In this study, the residuals were examined by plotting the residuals with the predicted values of the ERH. A model was considered as good when the residual plots indicate uniformly scattered points, all error terms minimum and the value for coefficient of determination R2 a maximum. If all the residual plots indicate a clear pattern, then that parameter was ignored and the rest of the parameters were considered to predict performance of the models.
3. Results and discussion Figures 1 and 2 show the adsorption and desorption isotherms of chillies at 25, 35 and 458C. All the isotherm curves followed the same sigmoid shape. The isotherms for raisins, currants, figs, prunes and apricots (Tsami et al., 1990), chilli powder (Munde, 2000) and chillies (Wesley et al., 2000) also showed the sigmoid shapes. The experimental values for the EMC of chillies obtained by Wesley et al. (2000) were very low compared to the present study. 3.1. Effect of temperature on equilibrium moisture content The values for the EMC of chillies decreased with increased surrounding air temperature in both adsorption and desorption processes at constant ERH (Figs 1 and 2). The reason is that as the chamber temperature
0.4 0.6 ERH, decimal
0.8
1.0
Figure 1. Moisture adsorption isotherms of chillies at different temperatures: m, 258C; &, 358C; *, 458C; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
60
EMC, % d.b.
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sP 2ffi N R R o p i¼1 Es ¼ df
60
EMC, % d.b.
deviation Em , standard error of estimate Es , randomness of residual ei and coefficient of determination R2 (Chen & Morey, 1989a; Menkov, 2000) which are given below: N Ro Rp 100 X Em ¼ ð1Þ N i¼1 Ro
40
20
0
0
0.2
0.6 0.4 ERH, decimal
0.8
1.0
Figure 2. Moisture desorption isotherms of chillies at different temperatures: m, 258C; &, 358C; *, 458C; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
was increased, the vapour pressure of the moisture within the chillies increased and hastened the transfer of moisture from chillies to the surrounding air. The decrease in the value for the EMC with the increase of air temperature was also observed for mushroom (Pandey & Aich, 1989), okra (Gupta et al., 1999) and chillies (Wesley et al., 2000).
3.2. Effect of equilibrium relative humidity on equilibrium moisture content The values for the EMC of chillies increased with the increase in the ERH in both the adsorption and desorption processes at constant temperature (Figs 1 and 2). Similar findings were also reported in the case of mushroom (Pandey & Aich, 1989), apricot, fig and raisin (Erol et al., 1990), okra (Gupta et al., 1999), chilli powder (Munde, 2000) and chillies (Wesley et al., 2000).
ARTICLE IN PRESS 99
ISOTHERMS OF RED CHILLIES
3.3. Effect of temperature on equilibrium relative humidity EMC, % d.b.
An upward shift in temperature from 25 to 458C, led to a shift of isotherms towards a lower value for the EMC, indicating that at any constant moisture content, the ERH increased with increase in temperature in both the adsorption and desorption processes (Figs 1 and 2). Pawar et al., (1992) observed similar type of results in the case of ground turmeric.
60
40
20
0 0
0.2
0.8
1.0
EMC, % d.b.
60
40
20
0 0
0.2
0.4 0.6 ERH, decimal
0.8
1.0
0
0.2
0.4
0.8
1.0
(b)
60
EMC, % d.b.
The adsorption and desorption isotherms exhibited the phenomenon of hysteresis, in which the EMC was higher at a particular ERH for the desorption curve than for adsorption (Fig. 3). Tsami et al. (1990) reported similar type of results in the case of raisins, currants, figs, prunes and apricots. The hysteresis (desorption moisture content minus adsorption moisture content) existed over the entire ERH range (Fig. 4). Furthermore, the magnitude of the hysteresis for isotherms at 258C exceeds that at 458C. The hysteresis values increased gradually nearly up to 060–065 ERH and afterwards, the values increased sharply up to 074–076 ERH. The hysteresis values started declining after 074–076 ERH. That is, after 074–076 ERH, the magnitude of the hysteresis decreased with an increase in temperature. Alam and Shove (1973), Chen and Morey (1989b) and Chen (2000) observed similar types of results in the case of soyabeans, yellow-dent maize and peanuts, respectively.
0.6
ERH, decimal
(a)
3.4. Sorption hysteresis
0.4
40
20
0
3.5. Models (c)
0.6
ERH, decimal
Different models were fitted for both adsorption and desorption isotherms of chillies and the validity of the models were discussed in the following sections.
Figure 3. Sorption hysteresis in chillies at different temperatures: (a) 258C; (b) 358C; (c) 458C; *, absorption; *, desorption; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
3.5.1. Adsorption isotherms The average adsorption data of chillies at various temperatures viz. 25, 35 and 458C were fitted in the ten models (Table 1) and the estimated values of the parameters of the models are listed in Table 2. The residuals of the ERH for different models are shown in Fig. 5 to examine its pattern. For the Oswin, Smith, Halsey, Henderson, Day and Nelson, modified Henderson, modified Chung and Pfost and modified Oswin models, the residual plots indicated a systematic pattern, which means that these models are unsuitable to define the adsorption isotherms. The residuals obtained by modified Halsey and the
newly proposed models denoted random pattern, which made these models suitable for defining adsorption isotherms. The mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot pattern of various models are presented in Table 3. From Table 3, it is clear that the modified Halsey provided the best fit, followed by the Kaleemullah model in predicting the equilibrium relative humidity of chillies during adsorption process.
ARTICLE IN PRESS 100
S. KALEEMULLAH, R. KAILAPPAN
3.5.2. Desorption isotherms The average desorption data of chillies at various temperatures viz. 25, 35 and 458C were fitted in the ten models (Table 1) and the estimated values of the parameters of the models are listed in Table 4. The residuals of the ERH for different models are depicted in Fig. 6 to examine its pattern. For the Smith, Halsey,
10 9
Difference in moistures, % d.b.
8
Henderson, Day and Nelson, modified Henderson, modified Halsey and modified Chung and Pfost models, the residual plots indicated a systematic pattern, which means that these models are unsuitable to define the desorption isotherms. The residuals obtained by Oswin, modified Oswin and Kaleemullah models followed a random pattern, which made these models suitable for defining desorption isotherms. The mean relative percentage deviation Em , standard error of estimate Es, coefficient of determination, R2 and residual plot pattern of various models are presented in Table 5. From Table 5, it is clear that the modified Oswin, Kaleemullah and Oswin models were ranked first, second and third, respectively, in predicting the equilibrium relative humidity of chillies during desorption process.
7
3.5.3. Adsorption and desorption isotherms The modified Halsey and modified Oswin models are ranked first in predicting the equilibrium relative humidity of chillies during adsorption and desorption processes, respectively. As there was no common model to predict the ERH of chillies during both the absorption and desorption processes, Kaleemullah model that was ranked second in both the adsorption and desorption processes of chillies can be considered and used with good predictive accuracy.
6 5 4 3 2
4. Conclusions
1 0 0
0.2
0.4
0.6
0.8
1.0
ERH, decimal Figure 4. Effect of temperature and relative humidity on derived hysteresis: m, 258C; &, 358C; *, 458C; ERH, equilibrium relative humidity
The sorption capacity of chillies decreased with an increase in temperature at constant relative humidity. The sorption isotherms exhibited the phenomenon of hysteresis, in which the equilibrium moisture content (EMC) was higher at a particular equilibrium relative humidity (ERH) for desorption curve than for adsorption. The ERH of chillies can be predicted by using the
Table 2 Estimated values of different parameters of various adsorption models for chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
Parameter values a
b
c
d
942048 10191 31644 000143 0002822 0000967 2858 15883 15234 08427
06305 12773 1195 11495 08364 21818 00205 0118 01643 08748
} } } } 14964 1103 11526 3118 1639 00012
} } } } 0086 } } } } 14074
ARTICLE IN PRESS 101
ISOTHERMS OF RED CHILLIES
0.15 Residual
Residual
0.15 0.00 −0.15
0
1.0
0.5
(a)
0.00 −0.15
1.0
0
0.5
0
0.5
1.0
0
0.5
1.0
0
0.5
1.0
(b)
0.15 Residual
Residual
0.15 0.00 −0.15
0
(c)
1.0
0.5
0.00 −0.15
(d)
0.15 Residual
Residual
0.15 0.00 −0.15 (e)
0
1.0
0.5
−0.15 (f)
0.15 Residual
Residual
0.15 0.00 −0.15
0.00
0
−0.15
1.0
0.5
(g)
0.00
(h)
0.15 Residual
Residual
0.15 0.00 −0.15 (i)
0
0.5 Predicted ERH, decimal
0.00 −0.15
1.0 ( j)
0
0.5 1.0 Predicted ERH, decimal
Figure 5. Residual plots of ten models for the adsorption data of chillies: (a) Oswin; (b) Smith; (c) Halsey; (d) Henderson; (e) Day and Nelson; (f) modified Henderson; (g) modified Halsey; (h) modified Chung and Pfost; (i) modified Oswin; (j) Kaleemullah; ERH, equilibrium relative humidity Table 3 Estimated values of mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot shapes of different models for adsorption of chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah *
Grade of the model.
Mean relative percentage deviation (Em)
Standard error of estimate value (Es)
38782 42942 17555 60252 71962 72169 086562 * 85352 31470 078951
00505 00576 00297 00519 00471 00459 002091 00631 00246 002342
Coefficient of deResidual termination (R2) plot pattern 09660 09567 09884 09647 09736 09740 099451 09503 09924 099352
Systematic Systematic Systematic Systematic Systematic Systematic Random Systematic Systematic Random
Average grade
Rank
} } } } } } 133 } } 166
} } } } } } 1 } } 2
ARTICLE IN PRESS 102
S. KALEEMULLAH, R. KAILAPPAN
Table 4 Estimated values of different parameters of various desorption models for chillies Model
Parameter values
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
a
b
c
d
12457 061103 337576 000122 0001792 0000699 28204 178065 19299 08406
06711 18282 1092 1082 092416 32927 001785 00809 019449 08189
} } } } 173206 1037 10489 45409 1517 0000739
} } } } 01449 } } } } 14102
0.15 Residual
Residual
0.15 0.00 −0.15 (a)
0
0.5
1.0
0.00 −0.15
Residual
Residual
0.00
0
0.5
−0.15
Residual
Residual
0.5
1.0
0
0.5
1.0
0
0.5
1.0
0.15
0.00 −0.15
0
0.5
0.00 −0.15
1.0
(e)
(f)
0.15
0.15 Residual
Residual
0
(d)
0.15
0.00 −0.15
0
(g)
0.5
0.00 −0.15
1.0 (h)
0.15
0.15 Residual
Residual
1.0
0.00
1.0
(c)
0.00 −0.15
(i)
0.5
0.15
0.15
−0.15
0
(b)
0
0.5 1.0 ( j) Predicted ERH, decimal
0.00 −0.15
0
0.5 1.0 Predicted ERH, decimal
Figure 6. Residual plots of ten models for the desorption data of chillies: (a) Oswin; (b) Smith; (c) Halsey; (d) Henderson; (e) Day and Nelson; (f) midified Henderson; (g) modified Halsey; (h) modified Chung and Pfost; (i) modified Oswin; (j) Kaleemullah; ERH, equilibrium relative humidity
ARTICLE IN PRESS 103
ISOTHERMS OF RED CHILLIES
Table 5 Estimated values of mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot shapes of different models for desorption of chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
Mean relative percentage deviation (Em)
Standard error of estimate value (Es)
Coefficient of determination (R2)
Residual plot pattern
Average grade
Rank
132793 * 29837 36627 29338 47206 46580 27902 76442
004003 00446 00361 00447 00297 00298 00253 00527
097903 09740 09830 09730 09895 09880 09919 09650
Random Systematic Systematic Systematic Systematic Systematic Systematic Systematic
3 } } } } } } }
3 } } } } } } }
097521 106192
001411 002282
099751 099382
Random Random
1 2
1 2
*
Grade of the model.
modified Halsey and Kaleemullah models for adsorption process whereas modified Oswin, Kaleemullah, Oswin models can be used for desorption process of chillies. Based on the experimental observations, as there was no common existing model to predict the ERH of chillies during both the adsorption and desorption processes, the Kaleemullah model which was ranked second in both the adsorption and desorption processes of chillies can be considered and used with good predictive accuracy.
References Alam A; Shove G C (1973). Hygroscopicity and thermal properties of soybeans. Transactions of the ASAE, 16(4), 707–709 AOAC (1995). Official Methods of Analysis (16th Edn.). Association of Official Analytical Chemists, Arlington, USA Boquet R; Chirife J; Iglesias H A (1978). Equations for fitting water sorption isotherms of foods: evaluation of various two parameter models. Journal of Food Technology, 13(4), 319–327 Chen C (1988). A study of equilibrium relative humidity for yellow-dent corn kernels. PhD Thesis, University of Minnesota, St Paul, USA Chen C; Morey R V (1989a). Comparison of four EMC/ERH equations. Transactions of the ASAE, 32(3), 983–990 Chen C; Morey R V (1989b). Equilibrium relative humidity relationships for yellow-dent corn. Transactions of the ASAE, 32(3), 999–1006 Chen C (2000). A rapid method to determine the sorption isotherms of peanuts. Journal of Agricultural Engineering Research, 75, 401–408 Chirife J; Iglesias H A (1978). Equations for fitting water sorption isotherms of foods. Journal of Food Technology, 13, 159–174 Day D L; Nelson G L (1965). Desorption isotherms for wheat. Transactions of the ASAE, 8(2), 293–297
Erol A; Guler A; Zafer D (1990). Moisture sorption isotherms of dried apricot, fig and raisin at 20oC and 36oC. Journal of Food Science, 55(6), 1591–1593, 1665 Gupta A; Shivhare U S; Bawa A S; Singh S (1999). Equilibrium moisture content of okra. Journal of the Institution of Engineers (India), 80, 9–11 Halsey G (1948). Physical adsorption on non-uniform surfaces. Journal of Chemical Physics, 16(10), 931–937 Henderson S M (1952). A basic concept of equilibrium moisture. Agricultural Engineering, 33(1), 29–32 Iglesias H A; Chirife J (1976). Prediction of effect of temperature on water sorption isotherms of food materials. Journal of Food Technology, 11(2), 109–116 Kaleemullah S (2002). Studies on engineering properties and drying kinetics of chillies. PhD Thesis, Tamil Nadu Agricultural University, Coimbatore, India King C J (1968). Rates of moisture sorption and desorption in porous, dried food stuffs. Food Technology, 22(4), 509–514 Labuza T P (1968). Sorption phenomena in foods. Food Technology, 22(2), 263–268 Labuza T P (1975). Interpretation of sorption data in relation to the state of constituent water. In: Water Relations of Foods (Duckworth R B, ed.), pp 155–172. Academic Press, London, UK Menkov N D (2000). Moisture sorption isotherms of vetch seeds at four temperatures. Journal of Agricultural Engineering Research, 76, 373–380 Munde A V (2000). Moisture adsorption characteristics of chilli powder. Journal of Maharastra Agricultural Universities, 25(1), 55–59 Oswin C R (1946). The kinetics of package life III}isotherm. Journal of the Society of Chemical Industry (London), 65, 419–421 Pandey M C; Aich J C (1989). Equilibrium moisture content of dehydrated mushroom. Journal of Food Science and Technology, 26(2), 108–109 Pawar V S; Dev D K; Pawar V D; Rodge A B; Surve V D; More D R (1992). Moisture adsorption isotherms of ground turmeric at different temperatures. Journal of Food Science and Technology, 29(3), 170–173
ARTICLE IN PRESS 104
S. KALEEMULLAH, R. KAILAPPAN
Pfost H B; Mourer S G; Chung D S; Miliken G A (1976). Summarizing and reporting equilibrium moisture data for grains. ASAE Paper No. 76-3520, St Joseph, MI San Martin M B; Mate J I; Fernandez T; Virseda P (2001). Modelling adsorption equilibrium moisture characteristics of rough rice. Drying Technology, 19(3–4), 681–690 Shatadal P; Jayas D S (1992). Sorption isotherms of foods. In: Drying of Solids (Mujumdar A S, ed.), pp 433–448. Oxford & IBH Publishing Co. Pvt. Ltd, New Delhi Smith S E (1947). Sorption of water vapour by proteins and high polymers. Journal of American Chemical Society, 69, 646–651 Soysal Y; Oztekin S (2001). Comparison of seven equilibrium moisture content equations for some medicinal and aromatic plants. Journal of Agricultural Engineering Research, 78(1), 57–63
Tsami E; Marinos–Kouris D; Maroulis Z B (1990). Water sorption isotherms of raisins, currants, figs, prunes and apricots. Journal of Food Science, 55(6), 1594–1597, 1625 Thomson T L; Peart R M; Foster G H (1968). Mathematical simulation of corn drying } a new model. Transactions of the ASAE, 24(3), 582–586 Verma D R; Gupta D K (1988). Equilibrium moisture content isotherms of sugarcane bagasse. Journal of Agricultural Engineering, 25(3), 49–56 Wesley B J; Chakraverty A; Sukumaran C R (2000). Equilibrium moisture content of chillies. The Andhra Agricultural Journal, 47(3–4), 254–257 Young J H (1976). Evaluation of models to describe sorption and desorption equilibrium moisture content isotherms of Virginia type peanuts. Transactions of the ASAE, 19(1), 146–150
Available online at www.sciencedirect.com
Moisture Sorption Isotherms of Red Chillies S. Kaleemullah1; R. Kailappan2 1
Department of Agricultural Engineering, SV Agricultural College, Tirupati 517 502 India; e-mail of corresponding author: [email protected] 2 Department of Agricultural Processing, Tamil Nadu Agricultural University, Coimbatore 641 003 India (Received 23 June 2003; accepted in revised form 27 January 2004; published online 24 March 2004)
The equilibrium moisture contents were determined for red chillies using the static method at 25, 35 and 458C over a range of relative humidities from 0115 to 0865. The sorption capacity of chillies decreased with an increase in temperature at constant relative humidity. The sorption isotherms exhibited the phenomenon of hysteresis, in which the equilibrium moisture content was higher at a particular equilibrium relative humidity for desorption curve than for adsorption. Ten models were applied for analysing the experimental data. The equilibrium relative humidity of chillies can be predicted by using the modified Halsey and Kaleemullah models for adsorption process whereas the modified Oswin, Kaleemullah and Oswin models can be used for desorption process of chillies. As there was no common model to predict the equilibrium relative humidity of chillies during both the adsorption and desorption processes, the Kaleemullah model that was ranked second in both the adsorption and desorption processes for chillies can be considered and used with good predictive accuracy. # 2004 Published by Elsevier Ltd on behalf of Silsoe Research Institute
1. Introduction
of grains. Labuza (1975) pointed out that no sorption isotherm model could fit data over the entire range of relative humidity because water is associated with the food matrix by different mechanisms in different activity regions. Boquet et al. (1978) compared various twoparameter equations for their accuracy in describing the isotherms of various foods. They concluded that the Henderson equation could adequately represent the isotherms of fruits and starchy foods whereas the Halsey equation is more suitable for meats, milk products and vegetables (Table 1). Chirife and Iglesias (1978) reviewed 23 isotherm equations, both theoretical and experimental and their use for fitting sorption isotherms of foods and food products. None of these equations described accurately the sorption isotherm over the whole range of relative humidity and for different types of food materials. Chen and Morey (1989a) evaluated four EMC–ERH equations for their ability to fit data for 18 cereal grains and seeds and found that no universal equation could be established to fit all isotherms. Pandey and Aich (1989) stated that Smiths’ equation (Table 1) gave the best fit over the entire range of relative humidity and temperature for dehydrated
In studying the behaviour of food or agricultural commodities during storage and processing, it is important to have information, which relates environmental conditions to the material properties. Equilibrium moisture content (EMC) is one such property and is defined as the moisture content of a hygroscopic material in equilibrium with a particular environment (temperature and relative humidity). The EMC has already been used in determining storage ability at various conditions (Labuza, 1968), prediction of drying time (King, 1968) and in general for the efficient design and operation of drying systems (Young, 1976). The hysteresis effect is the difference between desorption and adsorption EMC values for the same equilibrium relative humidity (ERH). Sorption isotherms for a food product over the full range of water activity form a hysteresis loop in which the desorption curve lies above the adsorption curve (Shatadal & Jayas, 1992). A number of theoretical, semi-theoretical and empirical isotherm equations have been developed to model the relationship between EMC or ERH and temperature 1537-5110/$30.00
95
# 2004 Published by Elsevier Ltd on behalf of Silsoe Research Institute
ARTICLE IN PRESS 96
S. KALEEMULLAH, R. KAILAPPAN
Notation a; b; c; d df ei Em Es HR M
dimensionless coefficients of the models degrees of freedom of regression model randomness of residual mean relative deviation, % standard error of estimate equilibrium relative humidity, decimal equilibrium moisture content, % d.b.
N R2 Ro Rp t
number of data points coefficient of determination observed equilibrium relative humidity, decimal predicted equilibrium relative humidity, decimal temperature, 8C
Table 1 Isotherm models for fitting experimental data Name of the model
Equation
Oswin (1946) HR ¼ Smith (1947) Halsey (1948) Henderson (1952) Day and Nelson (1965)
Modified Henderson (Thomson et al., 1968) Modified Halsey (Iglesias & Chirife, 1976) Modified Chung and Pfost (Pfost et al., 1976)
1 þ ðM=aÞ1=b
HR ¼ 1 expfða MÞ=bg HR ¼ expfa=ðtM b Þg HR ¼ 1 expðatM b Þ d
HR ¼ 1 expðatb M ct Þ HR ¼ 1 expfaðt þ bÞM c g HR ¼ exp½fexpða þ btÞg=M c
aexpðbMÞ HR ¼ exp ðt þ cÞ
Modified Oswin (Chen, 1988) HR ¼ Kaleemullah (Kaleemullah, 2002)
ðM=aÞ1=b
1 ½fða þ btÞ=Mgc þ 1
HR ¼ a b expðct M d Þ
HR , equilibrium relative humidity as a decimal; M, equilibrium moisture content as a % d.b.; t, temperature in 8C; a,b,c,d, dimensionless coefficients specific to individual equation.
mushroom. Oswin, Smith, Halsey and Henderson equations (Table 1) described well the moisture adsorption isotherms of ground turmeric at low temperatures but became less satisfactory at higher temperatures. The relative mean square root value of EMC of turmeric powder with the above equations was recorded as 516–
983% at 15–258C and 724–1520% at 35–358C (Pawar et al., 1992). Menkov (2000) reported that the modified Oswin model (Table 1) was the most suitable for describing the relationships between equilibrium moisture content, relative humidity and temperature. Soysal and Oztekin
ARTICLE IN PRESS ISOTHERMS OF RED CHILLIES
(2001) evaluated seven equilibrium moisture content equations for their ability to fit data for some medicinal and aromatic plants. They found the modified Halsey and the modified Oswin equations (Table 1) to be as the most versatile models for medicinal and aromatic plants; the modified Henderson equation (Table 1) is good for fennel and cinnamon. The modified Chung–Pfost equation (Table 1) was considered as the best model to explain the EMC–ERH relationship of rough rice (San Martin et al., 2001). Red chillies are the ripe fruits of the species of plants in the genus Capsicum. The most important quality characteristics of chillies are the colour and pungency. The chillies produced for condiment and culinary supplements are subjected to long–term storage. During that time, important physiochemical and biological changes take place with a strong impact on the colour and pungency. Therefore, it is necessary to investigate the equilibrium moisture content of chillies for various relative humidities and temperatures to enable the storage conditions for the chillies to be correctly specified. The automatic control of these conditions requires a reliable mathematical description of the EMC–ERH using suitable models. The objective of this work was to obtain the equilibrium moisture isotherms for red chillies at 25, 35 and 458C and to fit suitable models for describing the sorption characteristics.
2. Materials and methods 2.1. Experimental procedure Freshly harvested and dried red chillies of the polyster variety (Coimbatore local) were procured from the local market. Equilibrium moisture content was determined both in desorption and adsorption conditions at three temperatures (25, 35 and 458C) and eight levels of relative humidities (0115–0865) using the static method. Various supersaturated salt solutions of different relative humidities viz., lithium chloride (0115–0121), potassium acetate (0184–0227) magnesium chloride (0318–0332), potassium carbonate (0432–0438), sodium nitrite (0605–0643), sodium chloride (0748– 0756), ammonium sulphate (0794–0803) and potassium chromate (0846–0865) were used to provide constant relative humidities at 25, 35 and 458C temperatures. A preliminary experiment was conducted to determine the range of expected values of equilibrium moisture content by storing samples of chillies in the environment corresponding to 458C–0115 relative humidity (RH) and 258C–0865 RH for a period of 30
97
days. The range of equilibrium moisture content was 3–50% dry basis (d.b.). Hence, the samples for adsorption and desorption had to be pre-conditioned. The samples were conditioned to 75% d.b. for desorption purpose and to 25% d.b. for adsorption purpose. The samples were conditioned by drying the samples in a hot air vacuum oven at 708C till the desired moisture contents were obtained. Another preliminary experiment was conducted with conditioned samples to determine the realistic time required for attaining the state of equilibrium. After 10 days, the samples were weighed once in 2 days till a constant weight was observed. It was found that a period of 30 days was required for attaining equilibrium. Hence, a 40-day period was selected to provide a margin for variations in the time to achieve moisture equilibrium. Glass desiccators (16 cm diameter by 16 cm height) containing about 350 ml saturated salt solutions were used to provide constant relative humidities varying from 0115 to 0865. The desiccators were placed in an electric oven at a desired constant temperature and allowed to equilibrate with the environment inside the containers. Samples of pre-conditioned chillies (10 g each for desorption and 5 g each for adsorption) were placed in petri dishes and these dishes were placed on a plastic platform inside the desiccators to avoid direct contact between chillies and salt solution. After a 40-day period, the petri dishes were removed from controlled atmosphere space and stored in activated silica gel environment for 3 h to allow them to attain room temperature (Verma & Gupta, 1988). The moisture content of each sample was determined by keeping the samples in a vacuum oven at 708C and 4100 mmHg pressure (AOAC, 1995) until the weight measurement was constant. All the experiments were replicated three times and the averages of three values were used in the analysis. 2.2. Model fitting Ten isotherm models were selected for fitting the experimental data of adsorption and desorption isotherms of red chillies. These models have two, three or four parameters and the parameters of the models were estimated by using SPSS (Statistical Package for Social Science) 75 software. The models that have been used in this study are listed in Table 1. 2.3. Comparison methods The suitability of the models was evaluated and compared using the mean relative percentage
ARTICLE IN PRESS 98
S. KALEEMULLAH, R. KAILAPPAN
ei ¼ Ro Rp
ð2Þ
40
20
0
0
0.2
ð3Þ
where: Ro is the observed ERH; Rp is the predicted ERH from the model; N is the number of data points; and df is the degree of freedom of regression model (number of data points minus number of constants in model). Data points in a plot of the residual values versus the predicated values should tend to fall in a horizontal band centred on zero and displaying no systematic tendencies towards any clear pattern. If the residual plots indicate a clear pattern, the model should not be accepted (Chen & Morey, 1989a; Menkov, 2000). In this study, the residuals were examined by plotting the residuals with the predicted values of the ERH. A model was considered as good when the residual plots indicate uniformly scattered points, all error terms minimum and the value for coefficient of determination R2 a maximum. If all the residual plots indicate a clear pattern, then that parameter was ignored and the rest of the parameters were considered to predict performance of the models.
3. Results and discussion Figures 1 and 2 show the adsorption and desorption isotherms of chillies at 25, 35 and 458C. All the isotherm curves followed the same sigmoid shape. The isotherms for raisins, currants, figs, prunes and apricots (Tsami et al., 1990), chilli powder (Munde, 2000) and chillies (Wesley et al., 2000) also showed the sigmoid shapes. The experimental values for the EMC of chillies obtained by Wesley et al. (2000) were very low compared to the present study. 3.1. Effect of temperature on equilibrium moisture content The values for the EMC of chillies decreased with increased surrounding air temperature in both adsorption and desorption processes at constant ERH (Figs 1 and 2). The reason is that as the chamber temperature
0.4 0.6 ERH, decimal
0.8
1.0
Figure 1. Moisture adsorption isotherms of chillies at different temperatures: m, 258C; &, 358C; *, 458C; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
60
EMC, % d.b.
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sP 2ffi N R R o p i¼1 Es ¼ df
60
EMC, % d.b.
deviation Em , standard error of estimate Es , randomness of residual ei and coefficient of determination R2 (Chen & Morey, 1989a; Menkov, 2000) which are given below: N Ro Rp 100 X Em ¼ ð1Þ N i¼1 Ro
40
20
0
0
0.2
0.6 0.4 ERH, decimal
0.8
1.0
Figure 2. Moisture desorption isotherms of chillies at different temperatures: m, 258C; &, 358C; *, 458C; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
was increased, the vapour pressure of the moisture within the chillies increased and hastened the transfer of moisture from chillies to the surrounding air. The decrease in the value for the EMC with the increase of air temperature was also observed for mushroom (Pandey & Aich, 1989), okra (Gupta et al., 1999) and chillies (Wesley et al., 2000).
3.2. Effect of equilibrium relative humidity on equilibrium moisture content The values for the EMC of chillies increased with the increase in the ERH in both the adsorption and desorption processes at constant temperature (Figs 1 and 2). Similar findings were also reported in the case of mushroom (Pandey & Aich, 1989), apricot, fig and raisin (Erol et al., 1990), okra (Gupta et al., 1999), chilli powder (Munde, 2000) and chillies (Wesley et al., 2000).
ARTICLE IN PRESS 99
ISOTHERMS OF RED CHILLIES
3.3. Effect of temperature on equilibrium relative humidity EMC, % d.b.
An upward shift in temperature from 25 to 458C, led to a shift of isotherms towards a lower value for the EMC, indicating that at any constant moisture content, the ERH increased with increase in temperature in both the adsorption and desorption processes (Figs 1 and 2). Pawar et al., (1992) observed similar type of results in the case of ground turmeric.
60
40
20
0 0
0.2
0.8
1.0
EMC, % d.b.
60
40
20
0 0
0.2
0.4 0.6 ERH, decimal
0.8
1.0
0
0.2
0.4
0.8
1.0
(b)
60
EMC, % d.b.
The adsorption and desorption isotherms exhibited the phenomenon of hysteresis, in which the EMC was higher at a particular ERH for the desorption curve than for adsorption (Fig. 3). Tsami et al. (1990) reported similar type of results in the case of raisins, currants, figs, prunes and apricots. The hysteresis (desorption moisture content minus adsorption moisture content) existed over the entire ERH range (Fig. 4). Furthermore, the magnitude of the hysteresis for isotherms at 258C exceeds that at 458C. The hysteresis values increased gradually nearly up to 060–065 ERH and afterwards, the values increased sharply up to 074–076 ERH. The hysteresis values started declining after 074–076 ERH. That is, after 074–076 ERH, the magnitude of the hysteresis decreased with an increase in temperature. Alam and Shove (1973), Chen and Morey (1989b) and Chen (2000) observed similar types of results in the case of soyabeans, yellow-dent maize and peanuts, respectively.
0.6
ERH, decimal
(a)
3.4. Sorption hysteresis
0.4
40
20
0
3.5. Models (c)
0.6
ERH, decimal
Different models were fitted for both adsorption and desorption isotherms of chillies and the validity of the models were discussed in the following sections.
Figure 3. Sorption hysteresis in chillies at different temperatures: (a) 258C; (b) 358C; (c) 458C; *, absorption; *, desorption; EMC, equilibrium moisture content; ERH, equilibrium relative humidity
3.5.1. Adsorption isotherms The average adsorption data of chillies at various temperatures viz. 25, 35 and 458C were fitted in the ten models (Table 1) and the estimated values of the parameters of the models are listed in Table 2. The residuals of the ERH for different models are shown in Fig. 5 to examine its pattern. For the Oswin, Smith, Halsey, Henderson, Day and Nelson, modified Henderson, modified Chung and Pfost and modified Oswin models, the residual plots indicated a systematic pattern, which means that these models are unsuitable to define the adsorption isotherms. The residuals obtained by modified Halsey and the
newly proposed models denoted random pattern, which made these models suitable for defining adsorption isotherms. The mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot pattern of various models are presented in Table 3. From Table 3, it is clear that the modified Halsey provided the best fit, followed by the Kaleemullah model in predicting the equilibrium relative humidity of chillies during adsorption process.
ARTICLE IN PRESS 100
S. KALEEMULLAH, R. KAILAPPAN
3.5.2. Desorption isotherms The average desorption data of chillies at various temperatures viz. 25, 35 and 458C were fitted in the ten models (Table 1) and the estimated values of the parameters of the models are listed in Table 4. The residuals of the ERH for different models are depicted in Fig. 6 to examine its pattern. For the Smith, Halsey,
10 9
Difference in moistures, % d.b.
8
Henderson, Day and Nelson, modified Henderson, modified Halsey and modified Chung and Pfost models, the residual plots indicated a systematic pattern, which means that these models are unsuitable to define the desorption isotherms. The residuals obtained by Oswin, modified Oswin and Kaleemullah models followed a random pattern, which made these models suitable for defining desorption isotherms. The mean relative percentage deviation Em , standard error of estimate Es, coefficient of determination, R2 and residual plot pattern of various models are presented in Table 5. From Table 5, it is clear that the modified Oswin, Kaleemullah and Oswin models were ranked first, second and third, respectively, in predicting the equilibrium relative humidity of chillies during desorption process.
7
3.5.3. Adsorption and desorption isotherms The modified Halsey and modified Oswin models are ranked first in predicting the equilibrium relative humidity of chillies during adsorption and desorption processes, respectively. As there was no common model to predict the ERH of chillies during both the absorption and desorption processes, Kaleemullah model that was ranked second in both the adsorption and desorption processes of chillies can be considered and used with good predictive accuracy.
6 5 4 3 2
4. Conclusions
1 0 0
0.2
0.4
0.6
0.8
1.0
ERH, decimal Figure 4. Effect of temperature and relative humidity on derived hysteresis: m, 258C; &, 358C; *, 458C; ERH, equilibrium relative humidity
The sorption capacity of chillies decreased with an increase in temperature at constant relative humidity. The sorption isotherms exhibited the phenomenon of hysteresis, in which the equilibrium moisture content (EMC) was higher at a particular equilibrium relative humidity (ERH) for desorption curve than for adsorption. The ERH of chillies can be predicted by using the
Table 2 Estimated values of different parameters of various adsorption models for chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
Parameter values a
b
c
d
942048 10191 31644 000143 0002822 0000967 2858 15883 15234 08427
06305 12773 1195 11495 08364 21818 00205 0118 01643 08748
} } } } 14964 1103 11526 3118 1639 00012
} } } } 0086 } } } } 14074
ARTICLE IN PRESS 101
ISOTHERMS OF RED CHILLIES
0.15 Residual
Residual
0.15 0.00 −0.15
0
1.0
0.5
(a)
0.00 −0.15
1.0
0
0.5
0
0.5
1.0
0
0.5
1.0
0
0.5
1.0
(b)
0.15 Residual
Residual
0.15 0.00 −0.15
0
(c)
1.0
0.5
0.00 −0.15
(d)
0.15 Residual
Residual
0.15 0.00 −0.15 (e)
0
1.0
0.5
−0.15 (f)
0.15 Residual
Residual
0.15 0.00 −0.15
0.00
0
−0.15
1.0
0.5
(g)
0.00
(h)
0.15 Residual
Residual
0.15 0.00 −0.15 (i)
0
0.5 Predicted ERH, decimal
0.00 −0.15
1.0 ( j)
0
0.5 1.0 Predicted ERH, decimal
Figure 5. Residual plots of ten models for the adsorption data of chillies: (a) Oswin; (b) Smith; (c) Halsey; (d) Henderson; (e) Day and Nelson; (f) modified Henderson; (g) modified Halsey; (h) modified Chung and Pfost; (i) modified Oswin; (j) Kaleemullah; ERH, equilibrium relative humidity Table 3 Estimated values of mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot shapes of different models for adsorption of chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah *
Grade of the model.
Mean relative percentage deviation (Em)
Standard error of estimate value (Es)
38782 42942 17555 60252 71962 72169 086562 * 85352 31470 078951
00505 00576 00297 00519 00471 00459 002091 00631 00246 002342
Coefficient of deResidual termination (R2) plot pattern 09660 09567 09884 09647 09736 09740 099451 09503 09924 099352
Systematic Systematic Systematic Systematic Systematic Systematic Random Systematic Systematic Random
Average grade
Rank
} } } } } } 133 } } 166
} } } } } } 1 } } 2
ARTICLE IN PRESS 102
S. KALEEMULLAH, R. KAILAPPAN
Table 4 Estimated values of different parameters of various desorption models for chillies Model
Parameter values
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
a
b
c
d
12457 061103 337576 000122 0001792 0000699 28204 178065 19299 08406
06711 18282 1092 1082 092416 32927 001785 00809 019449 08189
} } } } 173206 1037 10489 45409 1517 0000739
} } } } 01449 } } } } 14102
0.15 Residual
Residual
0.15 0.00 −0.15 (a)
0
0.5
1.0
0.00 −0.15
Residual
Residual
0.00
0
0.5
−0.15
Residual
Residual
0.5
1.0
0
0.5
1.0
0
0.5
1.0
0.15
0.00 −0.15
0
0.5
0.00 −0.15
1.0
(e)
(f)
0.15
0.15 Residual
Residual
0
(d)
0.15
0.00 −0.15
0
(g)
0.5
0.00 −0.15
1.0 (h)
0.15
0.15 Residual
Residual
1.0
0.00
1.0
(c)
0.00 −0.15
(i)
0.5
0.15
0.15
−0.15
0
(b)
0
0.5 1.0 ( j) Predicted ERH, decimal
0.00 −0.15
0
0.5 1.0 Predicted ERH, decimal
Figure 6. Residual plots of ten models for the desorption data of chillies: (a) Oswin; (b) Smith; (c) Halsey; (d) Henderson; (e) Day and Nelson; (f) midified Henderson; (g) modified Halsey; (h) modified Chung and Pfost; (i) modified Oswin; (j) Kaleemullah; ERH, equilibrium relative humidity
ARTICLE IN PRESS 103
ISOTHERMS OF RED CHILLIES
Table 5 Estimated values of mean relative percentage deviation Em, standard error of estimate Es, coefficient of determination R2 and residual plot shapes of different models for desorption of chillies Model
Oswin Smith Halsey Henderson Day and Nelson Modified Henderson Modified Halsey Modified Chung and Pfost Modified Oswin Kaleemullah
Mean relative percentage deviation (Em)
Standard error of estimate value (Es)
Coefficient of determination (R2)
Residual plot pattern
Average grade
Rank
132793 * 29837 36627 29338 47206 46580 27902 76442
004003 00446 00361 00447 00297 00298 00253 00527
097903 09740 09830 09730 09895 09880 09919 09650
Random Systematic Systematic Systematic Systematic Systematic Systematic Systematic
3 } } } } } } }
3 } } } } } } }
097521 106192
001411 002282
099751 099382
Random Random
1 2
1 2
*
Grade of the model.
modified Halsey and Kaleemullah models for adsorption process whereas modified Oswin, Kaleemullah, Oswin models can be used for desorption process of chillies. Based on the experimental observations, as there was no common existing model to predict the ERH of chillies during both the adsorption and desorption processes, the Kaleemullah model which was ranked second in both the adsorption and desorption processes of chillies can be considered and used with good predictive accuracy.
References Alam A; Shove G C (1973). Hygroscopicity and thermal properties of soybeans. Transactions of the ASAE, 16(4), 707–709 AOAC (1995). Official Methods of Analysis (16th Edn.). Association of Official Analytical Chemists, Arlington, USA Boquet R; Chirife J; Iglesias H A (1978). Equations for fitting water sorption isotherms of foods: evaluation of various two parameter models. Journal of Food Technology, 13(4), 319–327 Chen C (1988). A study of equilibrium relative humidity for yellow-dent corn kernels. PhD Thesis, University of Minnesota, St Paul, USA Chen C; Morey R V (1989a). Comparison of four EMC/ERH equations. Transactions of the ASAE, 32(3), 983–990 Chen C; Morey R V (1989b). Equilibrium relative humidity relationships for yellow-dent corn. Transactions of the ASAE, 32(3), 999–1006 Chen C (2000). A rapid method to determine the sorption isotherms of peanuts. Journal of Agricultural Engineering Research, 75, 401–408 Chirife J; Iglesias H A (1978). Equations for fitting water sorption isotherms of foods. Journal of Food Technology, 13, 159–174 Day D L; Nelson G L (1965). Desorption isotherms for wheat. Transactions of the ASAE, 8(2), 293–297
Erol A; Guler A; Zafer D (1990). Moisture sorption isotherms of dried apricot, fig and raisin at 20oC and 36oC. Journal of Food Science, 55(6), 1591–1593, 1665 Gupta A; Shivhare U S; Bawa A S; Singh S (1999). Equilibrium moisture content of okra. Journal of the Institution of Engineers (India), 80, 9–11 Halsey G (1948). Physical adsorption on non-uniform surfaces. Journal of Chemical Physics, 16(10), 931–937 Henderson S M (1952). A basic concept of equilibrium moisture. Agricultural Engineering, 33(1), 29–32 Iglesias H A; Chirife J (1976). Prediction of effect of temperature on water sorption isotherms of food materials. Journal of Food Technology, 11(2), 109–116 Kaleemullah S (2002). Studies on engineering properties and drying kinetics of chillies. PhD Thesis, Tamil Nadu Agricultural University, Coimbatore, India King C J (1968). Rates of moisture sorption and desorption in porous, dried food stuffs. Food Technology, 22(4), 509–514 Labuza T P (1968). Sorption phenomena in foods. Food Technology, 22(2), 263–268 Labuza T P (1975). Interpretation of sorption data in relation to the state of constituent water. In: Water Relations of Foods (Duckworth R B, ed.), pp 155–172. Academic Press, London, UK Menkov N D (2000). Moisture sorption isotherms of vetch seeds at four temperatures. Journal of Agricultural Engineering Research, 76, 373–380 Munde A V (2000). Moisture adsorption characteristics of chilli powder. Journal of Maharastra Agricultural Universities, 25(1), 55–59 Oswin C R (1946). The kinetics of package life III}isotherm. Journal of the Society of Chemical Industry (London), 65, 419–421 Pandey M C; Aich J C (1989). Equilibrium moisture content of dehydrated mushroom. Journal of Food Science and Technology, 26(2), 108–109 Pawar V S; Dev D K; Pawar V D; Rodge A B; Surve V D; More D R (1992). Moisture adsorption isotherms of ground turmeric at different temperatures. Journal of Food Science and Technology, 29(3), 170–173
ARTICLE IN PRESS 104
S. KALEEMULLAH, R. KAILAPPAN
Pfost H B; Mourer S G; Chung D S; Miliken G A (1976). Summarizing and reporting equilibrium moisture data for grains. ASAE Paper No. 76-3520, St Joseph, MI San Martin M B; Mate J I; Fernandez T; Virseda P (2001). Modelling adsorption equilibrium moisture characteristics of rough rice. Drying Technology, 19(3–4), 681–690 Shatadal P; Jayas D S (1992). Sorption isotherms of foods. In: Drying of Solids (Mujumdar A S, ed.), pp 433–448. Oxford & IBH Publishing Co. Pvt. Ltd, New Delhi Smith S E (1947). Sorption of water vapour by proteins and high polymers. Journal of American Chemical Society, 69, 646–651 Soysal Y; Oztekin S (2001). Comparison of seven equilibrium moisture content equations for some medicinal and aromatic plants. Journal of Agricultural Engineering Research, 78(1), 57–63
Tsami E; Marinos–Kouris D; Maroulis Z B (1990). Water sorption isotherms of raisins, currants, figs, prunes and apricots. Journal of Food Science, 55(6), 1594–1597, 1625 Thomson T L; Peart R M; Foster G H (1968). Mathematical simulation of corn drying } a new model. Transactions of the ASAE, 24(3), 582–586 Verma D R; Gupta D K (1988). Equilibrium moisture content isotherms of sugarcane bagasse. Journal of Agricultural Engineering, 25(3), 49–56 Wesley B J; Chakraverty A; Sukumaran C R (2000). Equilibrium moisture content of chillies. The Andhra Agricultural Journal, 47(3–4), 254–257 Young J H (1976). Evaluation of models to describe sorption and desorption equilibrium moisture content isotherms of Virginia type peanuts. Transactions of the ASAE, 19(1), 146–150