- Email: [email protected]
Influence of process control strategies on drying kinetics, colour and shrinkage of air dried apples
Applied Thermal Engineering 62 (2014) 455e460
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Influence of process control strategies on drying kinetics, colour and shrinkage of air dried apples Barbara Sturm a, b, *, Anna-Maria Nunez Vega b, Werner C. Hofacker b, c a
Department of Agricultural Engineering, Kassel University, Nordbahnhofstrasse 1a, Witzenhausen D-37213, Germany Department of Process and Environmental Engineering, Thermal Process Engineering, University of Applied Sciences Konstanz, Braunegger Straße 55, Konstanz D-78462, Germany c Institute for Applied Thermo- and Fluiddynamics IaTF, University of Applied Sciences Konstanz, Braunegger Straße 55, Konstanz D-78462, Germany b
h i g h l i g h t s Development of quality attributes significantly depends on control strategies. Online measurements show that information can be used for process control optimisation. Product temperature controlled drying is gentler at comparable drying times.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 June 2013 Accepted 30 September 2013 Available online 16 October 2013
The influence of two different control strategies, constant air temperature and constant product temperature, on product quality and drying behaviour of apples was investigated. The interactions of the drying parameters air temperature (35e100 C), product temperature (35e85 C), dew point temperature (5e30 C) and air velocity (2.0e4.8 m/s) with drying time, colour changes and shrinkage were measured continuously and determined for both strategies. Based on these results the two strategies were compared with regards to their effect on drying performance. The results show that the drying strategy has a significant influence not only on the development and duration of the drying process but also on the development of colour changes and shrinkage. Controlling product temperature led to shorter drying times and generally lower colour changes. Furthermore, it was shown that the product temperature develops characteristically; two stages and a clearly visible transition period can be detected. This potentially can be used to control the process and, therefore, improve its performance regarding duration and quality aspects. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Apple Convective drying Drying kinetics Process control Colour Shrinkage Product temperature
1. Introduction Convection drying is one of the oldest and most wide spread means of preserving food. It was identified a long time ago that, as most food products are heat sensitive, the crucial factor in the convection drying of food and other sensitive agricultural products are the product temperature and, in particular, preventing that temperature from exceeding critical levels. Excessive temperatures lead to structural, organoleptic (smell, taste, visual appearance, chewiness etc.) and nutritional changes [1]. The extent of changes of the most important quality characteristics like colour (pigments,
* Corresponding author. Newcastle Institute for Research on Sustainability, Newcastle University, Newcastle upon Tyne NE1 7RU, UK. Tel.: þ44 191 246 4951. E-mail addresses: [email protected], barbara.sturm@ newcastle.ac.uk (B. Sturm). 1359-4311/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.applthermaleng.2013.09.056
enzymatic and non-enzymatic browning) and nutritional value (anti-oxidants, vitamins) usually increases with increasing temperature. The dependency of reaction constants on temperature allows for the assumption that drying at lower temperatures leads to a lower loss of valuable compounds [2]. However, traditionally, the drying process is controlled by keeping the air temperature constant on a predefined level. Often product temperature is unknown or if it is measured it is not included in the process control system [3]. The changes of product characteristics during the drying process are related to reactions of zero-order or 1st order [4] and usually product temperature is assumed to be constant. However, several studies on medicinal herbs, fruits and vegetables [5e7] have shown that for products with high initial product moisture content, product temperature changes almost throughout the process and only comes close to air temperature at the end of the process.
456
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Nomenclature A
DE P R2 S t v x x X y
air total colour difference product coefficient of determination shrinkage time (min) velocity (m/s) arithmetic mean independent variable moisture content (gW/gDM) response variables
Several authors have shown for different sensitive biological products [8,9] that they are not excessively damaged when exposed to high air temperatures during the first period of drying. This can be explained by the initially very high moisture content of the product that leads to a great difference between dry and wet bulb temperature, therefore no thermal damage is risked in this phase. A multitude of studies have investigated the influence of drying conditions in convection drying on drying kinetics and quality aspects of apples [9e16]. The general consent is that increase of air temperature has the most significant influence on drying kinetics, in addition to colour change, whilst high air velocity is regarded as reducing shrinkage [17]. In recent years, the importance of dynamic process control has gained increased recognition in production of dried food stuffs. Research on the influence of periodically changing drying conditions showed that reducing temperature step wise from an initially high temperature results in a reduction of colour changes in guava [18]. Product quality of dried potatoes can be improved by cyclically changing drying conditions [19]. Cho and Chua [20] concluded that intelligent online-control of drying necessitates step wise change of air temperature and observation of product temperature. However, only a very limited number of studies have actively controlled the product temperature. Isothermal drying has been used for control of combined drying processes and for the measurement of effective diffusivity and shrinking of hygroscopic materials [13,21]. Nonetheless, only a few studies [22] have actually obtained and verified isothermal conditions, uniform temperature is usually assumed. According to Srikiatden and Roberts [23] isothermal conditions may be possible for convective hot air drying of a porous material having a very small characteristic dimension. Lengyel [24] showed that for the system in question, in bodies with less than 10 mm thickness, the core warms very quickly and therefore the body temperature is quasi uniform. Colour and shrinkage are two of the most important aspects in determination of product quality. They are both directly related to consumers’ appreciation of a product as they tend to associate product colour and other visual properties with its taste, hygienic security, shelf life, nutritional value and personal satisfaction [25]. Furthermore, colour is directly linked to aroma and taste of the product [26] while shrinkage tends to be linearly related to moisture content [27]. The degree of shrinkage depends directly on the temperature, humidity and velocity levels applied. This paper is concerned with the determination of the influence of drying strategy and conditions on the quality of dried apples and performance of the process represented by the duration, temperature and product quality developments. Colour and twodimensional shrinkage have been used as indicators of product quality as they comprehensively represent stresses during drying.
Greek letters b approximated coefficient ε experimental variance 4 relative humidity (%) s standard deviation w temperature ( C) Subscripts A air dp dew point Dr drying DM dry matter P product W water
2. Materials and methods 2.1. Raw material Apples of the Jonagold variety were used as experimental material. Fruits were obtained from a local farmer (Lake Constance, Germany) and stored in the fridge at 4 C. Before drying, raw material was cut into slices of 3.9 mm 0.2 mm thickness with an outer diameter of 72 mm and an inner diameter of 20 mm. Average weight per slice was 13 g 0.2 g. 2.2. Convection drying Drying tests were carried out using a high precisions laboratory dryer developed in the department of Thermal Process Engineering, University of Applied Sciences Konstanz, Konstanz (Germany). The experimental system essentially comprises of 5 units: (i) an air flow control unit, (ii) an air conditioning unit, with a water bath including heating and chilling units and a packed bed, (iii) a heating control unit with heating elements, (iv) a drying compartment that allows for either over or through flow air stream for convective drying of product (v) a non-invasive online measurement system, including a CCD camera and a pyrometer. A detailed description of the dryer was given by Sturm and Sturm et al. [7,28]. For the air temperature controlled tests, the dryer was pre heated for 20 min to reach the defined set temperatures. The apple slices were then distributed as described by Sturm [28]. The total weight of the samples was approximately 90 g. In the case of surface temperature control the dryer was turned on at the start of the experiment. Air temperature controlled drying was conducted at 35, 45, 60, 75 and 85 C. Product temperature controlled drying was conducted at 35, 40, 47.5, 55 and 60 C. During experiments where product temperature was controlled, maximum air temperature was limited to 100 C. This was due to the fact that the industrial dryers in question are not operating at any temperatures higher than 100 C. Dew point temperature was set to 5.0, 10.0, 17.5, 25 and 30 C and air velocity to 2.0, 2.6, 3.4, 4.2 and 4.8 m/s for both strategies investigated. The samples were dried until they reached a moisture content of approximately 0.13 gw/gDM. This moisture content was chosen because it is typically used in production of dried apple products (aW ¼ 0.5 [29] in desorption and 0.52 to 0.55 [30] in absorption). During drying, the weight, air temperature and product temperature were measured continuously. Colour and two dimensional shape (visible cross section) of samples were measured every five minutes using the integrated CCD camera (The Imagingsource DFK 31BU03.H). The details of the optical systems as well as the calibration were described in Sturm et al. [7]. For the representation of
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
457
changes in product colour the total colour difference DE is used, which is increasingly used in the determination of food quality changes. DE calculated from the L*, a*, b* values as follows:
DE ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 2 L* þ a* þ b*
(1)
The moisture content was measured by the gravimetric method using a laboratory convection oven (Heraeus UT 12) over 48 h at a constant temperature of 70 C. This method was chosen over the standard method of drying at 105 C for 24 h [31] as it had proven to degrade the samples. 2.3. Experimental design and statistical analysis The quality parameters during the process were evaluated using Origin 7.5 (Microcal, Northhampton, USA). Equation (2) gives the basic third order polynomic equation used.
Fig. 1. Moisture ratio, product temperatures and air temperature for both strategies as a function of time, wN ¼ 60 C, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
y ¼ b0 þ b1 x þ b2 x2þ b3 x3 þ ε εwN 0; s2
(2)
Shrinkage was best represented using linear fits. The total colour difference functions of moisture ratio was best represented using third and the drying rates using fifth order polynomial regression. The fits were significant for all parameters described (p < 0.001). Duration of drying (tDr) was evaluated as dependent variable yi (response variables). The two series of 20 tests had a randomized order. Equation (3) provides the basic quadratic regression model used to estimate the influence of the independent parameters xi on the dependent variable yk.
yk ¼ ß0 þ
3 X
ßi xi þ
i¼1
3 X i¼1
ßi x2i þ
3 3 X X
ßi;j xi xj þ ε
(3)
i ¼ 1 j ¼ iþ1
Variance and regression analysis were performed using the software Design-ExpertÓ 7.1 by Stat-Ease. Regression analysis was repeatedly carried out, successively rejecting factors by backward elimination. Although some of the factors are by definition not significant (p > 0.1) they were included as further reduction of the models would have led to a reduction of accuracy of the overall models. Table 1 shows the resulting models for both control strategies. In both cases investigated the development of the process is best represented using quadratic models. 3. Results and discussion Fig. 1 shows the temperature and moisture ratio developments as functions of time for identical nominal drying temperatures (60 C). It can be seen that drying time is significantly reduced when controlling product temperature instead of air temperature. If results for similar duration of the process are compared, a lower
maximum temperature level is required when product temperature is controlled (Fig. 2). While an increase of drying temperature (Fig. 2) and dew point temperature (Fig. 3) have a similar effect on resulting drying times for both strategies applied, the change of air velocity (Fig. 4) has a much more significant influence when air temperature is controlled than when product temperature is controlled. This phenomenon can be explained by the fact that in product temperature controlled drying the lower velocity and, therefore, mass flow is compensated by an increase of air temperature, as the driving force for the control system is to reach the set product temperature [28]. Fig. 5 depicts the drying rates as a function of moisture ratio. When product temperature is controlled, the relative drying rates in the beginning are much higher and decrease much more significantly than when air temperature is controlled. This can be explained with the fundamentally different influences of isothermal and non-isothermal conditions on drying behaviour. When comparing nearly identical drying times (A60, P47) the drying rate for P47 is higher than that of A60 until a moisture ratio of 0.4 is reached. Comparing the temperature developments (Figs. 6 and 7) it can be seen that at this point air temperature in product temperature controlled drying falls below 60 C. At the same time product temperature in air temperature controlled drying exceeds 47 C. Fig. 6 depicts the temperature development of the air temperature controlled samples. The samples are transitioning from the second to the third phase of drying at a moisture ratio of approximately 0.2 which translates to a moisture content of circa 2 gw/gDM. At the end of this transition the product temperature is nearly constant and assimilating to air temperature. The described temperature developments can be explained by the relationships between dry bulb, wet bulb and dew point temperature. At the
Table 1 Equations for prediction of drying time and their statistics. Response
Modela
R2a
Pb
xc
s xd
tA tP
tDr ¼ 1298:08 23:87wA þ 8:15wDp 149:05vA 0:11wA wDp þ 0:86wA vA þ 0:14w2A þ 8:82v2A tDr ¼ 1384:96 51:27wP þ 23:45wDp þ 30:28vA 0:37wP wDp 2:60wDp vA þ 0:51w2P þ 0:14w2Dp
0.989 0.946
<0.0001 <0.0001
195 207
14 21
a b c d
Coefficient of determination. Pr > F. Arithmetic mean. Standard deviation.
458
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Fig. 2. Drying time as a function of drying temperature, wDp ¼ 17.5 C, vA ¼ 3.4 m/s. Fig. 5. Drying rate as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
Fig. 3. Drying time as a function of dew point temperature, wA ¼ 60 C, wP ¼ 47.5 C, vA ¼ 3.4 m/s.
beginning of the process the moist body very quickly reaches the wet bulb temperature, which is lower than the effective air temperature but higher than the dew point temperature and depends directly on air temperature and humidity. In the first phase of
Fig. 4. Drying time as a function of air velocity, wA ¼ 60 C, wP ¼ 47.5 C, wDp ¼ 17.5 C.
drying the product temperature will stay constant, in the second and consequent third phase, it rises until it reaches the temperature of the air at the end of the process. When controlling product temperature, air temperature will rise as long as the surface temperature has not reached the set-point. To maintain a constant product temperature at a decreasing difference of wet and dry bulb temperature air temperature then needs to be reduced accordingly. In case of product temperature control (Fig. 7) a transition phase can be observed too and starts with a significant air temperature drop around a moisture ratio of 0.3 and reaches its end at a moisture ratio of roughly 0.2. Independent of the strategy the transition from one phase to the next can clearly be detected by the development of temperature: in case of air temperature controlled drying using the product temperature and vice versa. The experiments described above show that controlling the process using the product temperature as the control variable is significantly different to traditional control of air temperature. The drying kinetics agree with the results of Srikiatden and Roberts [13] who used a combination of convection and microwave for isothermal drying of apples. While the drying kinetics for air
Fig. 6. Air and product temperatures as a function of moisture ratio for air temperature controlled drying, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Fig. 7. Air and product temperature as a function of moisture ratio for product temperature controlled drying, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
temperature control can be described by Fick’s diffusion model, this is not possible for product temperature control. Controlling product temperature initially leads to very high drying rates, which can be explained by the high initial air temperatures. After the product has reached the set temperature, drying rate decreases quickly with reducing air temperatures. Especially after the transition from the second to the third drying phase air temperature is only marginally higher than product temperature, which leads to an unnecessary prolonging of process if product temperatures are set too low. The developments of product temperatures depicted for air temperature controlled drying clearly show that in the case of hot air apple drying the assumption of isothermal conditions is not valid. Both strategies applied significantly affect the colour of the product. Apple tissue underwent extensive non-homogeneous colour changes (Fig. 8). In the case of air temperature control, the application of high temperatures led to a decrease in discolouration at otherwise constant process conditions, which is in agreement with the findings of Contreras et al. [32] for apples and Pott et al.
Fig. 8. Total colour difference DE as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
459
[33] for mangoes. In the case of product temperature controlled drying, however, the situation is different. After the initial significant changes in colour, the DE stayed almost steady for an extended period. At a moisture ratio of about 0.4 it started increasing significantly again. Overall DE is lower for product temperature controlled systems and depends less on the process settings. One possible explanation for this might be the inactivation of polyphenol oxidase PPO at high initial air temperature levels. PPO catalyses the hydroxylation and oxidation of phenolic compounds in fruits and is the major cause for enzymatic browning. Its inactivation reduces browning significantly. This is in accordance to the findings of Yemenicioglu et al. [34]. Shrinkage, displayed in Fig. 9, is significantly influenced by both the strategy and the temperature levels applied which is in agreement with literature [12,14]. When air temperature is kept constant, shrinkage is significantly higher than that of comparison specimen which is initially dried at very high air temperatures. This might be due to the fact that slow drying reduces internal stresses, which in consequence results in increased shrinkage. Applying high temperatures, however, increase internal stresses but at the same time the resulting fast drying leads to a mechanical stabilisation of the surface [12,14]. Drying characteristics and development of quality aspects are significantly different for single stage and non-stationary drying. For both strategies applied, two clearly visible phases were characterized, separated by sizable changes of product temperature (air temperature controlled, Fig. 6), respectively air temperature (product temperature controlled, Fig. 7). Analogously, colour changes are divided into clearly distinguishable phases, see Fig. 8. Shrinkage is significantly reduced by higher temperatures at the beginning of the process. The required drying time varies significantly less for product temperature control than for air temperature control. The factors investigated, temperature, air humidity and air velocity affect the drying results significantly differently to air temperature control. Using product temperature as the control variable results in significantly better optical product characteristics. It can be assumed that applying a lower maximum product temperature at comparable drying times additionally results in a higher retain of valuable compounds such as vitamins and aromas [35e38]. The results obtained give valuable information for the development of optimized control strategies. Keeping the control
Fig. 9. Shrinkage as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
460
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
variable constant throughout the process, either air temperature or product temperature, does not result in an optimum solution. Based on the results obtained a step wise process needs to be developed. In general, both air or product temperature could be used for this; a combination of the two might even be considered. The crucial information necessary for either is the point of transition, which can be obtained from the information of temperature development. Reductions of drying times of up to 40% can be expected for the suggested control strategy. 4. Conclusions A comparative study of single stage and non-stationary convection drying of apples utilizing non-invasive measurement devices showed the advantages of on-line monitoring of important state variables as product temperature, colour and shrinkage. It was possible to determine the influence of both the strategy and process conditions on drying kinetics, temperature and quality development throughout the process. Results presented allow deduction for optimised process conditions. Segmentation of the drying process into phases with different drying conditions and control strategies shows potential. Additionally, an active integration of information derived from pictures taken continuously throughout the process into the control system is conceivable. Based on the results obtained adaptive control systems for existing and newly designed drying devices can be developed. Increasing process temperature during the first phase of drying will reduce processing time and, therefore, increase throughput and reduction of specific energy demand while the integration of optical systems might allow for more targeted control of specific product characteristics. Integration of active product temperature control can prevent overheating of the product and therefore increased degradation. Initial tests on other products, such as onions and leek, showed that the control strategies can be applied to these products with similar results regarding drying time and product quality. Implementation of sensors and control systems in existing tray dryers was successful. However, further research is necessary to determine correct settings, especially concerning the influence of differences in drying rate of particles depending on their relative position to the air inlet. Acknowledgements The authors acknowledge the German Federal Ministry of Economy and Technology for the financial support within the PRO INNO II program. References [1] G.H. Crapiste, Simulation of drying rates and quality changes during the dehydration of foodstuffs, in: J.-E. Lozano, C. Anon, E. Parada-Arais, BarbosaCanovas (Eds.), Trends in Food Engineering, Technomic Publishing Co., Pennsylvania, USA, 2000. [2] C. Ratti (Ed.), Advances in Food Dehydration, CRC Press, Boca Raton, Florida, USA, 2009. [3] A.S. Mujumdar, C.L. Law, Drying technology: trends and applications in postharvest processing, Food and Bioprocess Technol. 3 (6) (2009) 843e852. [4] K. Kröll, W. Kast, Trocknen und Trockner in der Produktion, Bd. 3, Springer Verlag, Berlin, Heidelberg, Germany, 1989. [5] J. Müller, Trocknung von Arzneipflanzen mit Solarenergie, Eugen Ulmer GmbH & Co., Stuttgart, 1992. Dissertation Universität Hohenheim, Germany. [6] E.F.A.A. Bashir, Solar Drying of Sliced Onion and Quality Attributes as Affected by the Drying Process and Storage Conditions. VDI-MEG 328. Dissertation, Universität Hohenheim, Germany, 1998. [7] B. Sturm, W. Hofacker, O. Hensel, Optimizing the drying parameters for hot air dried apples, Drying Technol. 30 (14) (2012) 1570e1582.
[8] K.J. Chua, A.S. Mujumdar, M.N.A. Hawlader, S.K. Chou, J.C. Ho, Batch drying of banana pieces e effect on stepwise change in drying air temperature on drying kinetics and product colour, Food Res. Int. 34 (8) (2001) 721e731. [9] E.L. Schultz, M.M. Mazzuco, R.A.F. Machado, A. Bolzan, M.B. Quadri, M.G.N. Quadri, Effect of pre-treatment of drying, density and shrinkage of apple slices, J. Food Eng. 78 (3) (2007) 1103e1110. [10] A. Kaya, O. Aydin, C. Demirtas, Drying kinetics of red delicious apple, Biosyst. Eng. 96 (4) (2007) 517e524. [11] H.O. Menges, C. Ertekin, Mathematical modeling of thin layer drying of golden apples, J. Food Eng. 77 (1) (2006) 119e125. [12] P.P. Lewicki, E. Jakubczyk, Effect of hot air temperature on mechanical properties of dried apples, J. Food Eng. 64 (3) (2004) 307e314. [13] J. Srikiatden, J.S. Roberts, Moisture loss kinetics of apple during convective hot air and isothermal drying, Int. J. Food Prop. 8 (3) (2005) 493e512. [14] Y. Bai, M.S. Rahman, C.O. Perera, B. Smith, L.D. Melton, Structural changes in apple rings during convection air-drying with controlled temperature and humidity, J. Agric. Food Chem. 50 (11) (2002) 3179e3185. [15] L. Fernandez, C. Castillero, J.M. Aquilera, An application of image analysis to dehydration of apple discs, J. Food Eng. 67 (1e2) (2005) 185e193. [16] A. Vega-Galvez, A.H. Kong, M. Chcana, J. Vergara, J. Martinez-Monzo, P. GarciaSegovia, R. Lemus-Mondaca, K. Di Scala, Effect of temperature and air velocity on drying kinetics, antioxidant capacity, total phenolic content, colour, texture and microstructure of apple (var. Granny Smith) slices, Food Chem. 132 (1) (2012) 51e59. [17] C. Ratti, Shrinkage during drying of food stuffs, J. Food Eng. 23 (1) (1994) 91e105. [18] K.J. Chua, S.K. Chou, J.S. Ho, A.S. Mujumdar, M.N.A. Hawlader, Cyclic air temperature during drying of guave pieces: effects on moisture and ascorbic acid contents, Food and Bioprod. Process. 78 (2) (2000) 28e72. [19] J.C. Ho, S.K. Chou, K.J. Chua, A.S. Mujumdar, M.N.A. Hawlader, Analytical study of cyclic temperature drying: effect on drying kinetics and product quality, J. Food Eng. 51 (1) (2003) 65e75. [20] S.K. Chou, K.J. Chua, On the study of the drying behavior of a heat-sensitive biomaterial undergoing stepwise-varying temperature schemes, Ind. Eng. Chem. Res. 42 (20) (2003) 4939e4952. [21] J.S. Roberts, C.H. Tong, The development of an isothermal drying apparatus and the evaluation of the diffusion model on hygroscopic porous material, Int. J. Food Prop. 6 (1) (2003) 165e180. [22] J.F. Nicoleti, J. Telis-Romero, V.R.N. Telis, Air-Drying of fresh and osmotically pre-treated pineapple slices: fixed air temperature versus fixed slice temperature drying kinetics, Drying Technol. 19 (9) (2001) 2179e2191. [23] J. Srikiatden, J.S. Roberts, Predicting moisture profiles in potato and carrot during convective hot air drying using isothermally measured effective diffusivity, J. Food Eng. 84 (4) (2008) 516e525. [24] A. Lengyel, The change of body temperature during convective drying of cubeshaped apple pieces, Drying Technol. 25 (7e8) (2007) 1275e1280. [25] F. Pedreschi, J. Léon, M. Domingo, P. Moyano, Development of a computer vision system to measure the colour of potato chips, Food Res. Int. 39 (10) (2006) 1092e1098. [26] N.J. Morris, I.W. Lohmann, R.T. O’Connor, A.F. Freeman, I.V. Peanut Butter, Determination of color of peanut butter by spectral reflectance method, Food Technol. 7 (1953) 393e396. [27] L. Mayor, A.M. Sereno, Modelling shrinkage during convective drying of food materials: a review, J. Food Eng. 61 (3) (2004) 373e386. [28] B. Sturm, Einfluss der Führung des Trocknungsprozesses auf den Trocknungsverlauf und die Produkteigenschaften empfindlicher Biologischer Güter (Influence of Process Control on Drying Kinetics and Product Attributes of Sensitive Biological Products), in: Forschungsbericht Agrartechnik 491 des Arbeitskreises Forschung und Lehre der Max-Eyth Gesellschaft Agrartechnik VDI (VDI-MEG), Witzenhausen, Germany, 2010. [29] M. Wolf, J.E. Walker Jr., J.G. Kapsalis, Water vapor sorption hysteresis in dehydrated foods, J. Agric. Food Chem. 20 (1972) 1073e1077. [30] P.P. Lewicki, A. Lenart, Wptyw wstepnego odwadniania osmotycznego na wlásciwosci adsorpcyjne jablek suszonych, Przem. Spozyw. 30 (1977) 141. [31] AOAC Official Methods of Analysis, fifteenth ed., Association of Official Analytical Chemists, Washington D.C, 1990. [32] C. Contreras, M.E. Martin-Esparza, A. Chiralt, N. Martinez-Navarrete, Influence of microwave application on convection drying: effects of drying kinetics, and optical and mechanical properties of apples and strawberries, J. Food Eng. 88 (1) (2008) 55e64. [33] I. Pott, S. Neidhard, W. Mühlbauer, R. Carle, Quality improvement of nonsulphited mango slices by drying at high temperatures, Innovative Food Sci. Emerging Technol. 6 (4) (2005) 412e419. [34] A. Yemenicioglu, M. Ozkan, B. Cemeroglu, Heat inactivation kinetic of apples polyphenoloxidase and activation of its latent form, J. Food Sci. 62 (3) (2006) 508e510. [35] O.R. Fennema (Ed.), Food Chemistry, second ed., Marcel Dekker, New York, USA, 1985. [36] M.D.A. Hawlander, C. Perera, M. Tian, K.L. Yeo, Drying of guava and papaya: impact of different drying methods, Drying Technol. 24 (1) (2006) 77e87. [37] M. Miranda, H. Maureira, K. Rodríguez, K. Vega-Gálvez, Influence of temperature on the drying kinetics, physicochemical properties, and antioxidant capacity of Aloe Vera (Aloe Barbadensis Miller) gel, J. Food Eng. 91 (2) (2009) 297e304. [38] N. Boudhrioua, P. Giampaoli, C. Bonazzi, Changes in aromatic components of banana during ripening and air-drying, Lebensm. Wiss. Technol. 36 (6) (2003) 633e642.
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Influence of process control strategies on drying kinetics, colour and shrinkage of air dried apples Barbara Sturm a, b, *, Anna-Maria Nunez Vega b, Werner C. Hofacker b, c a
Department of Agricultural Engineering, Kassel University, Nordbahnhofstrasse 1a, Witzenhausen D-37213, Germany Department of Process and Environmental Engineering, Thermal Process Engineering, University of Applied Sciences Konstanz, Braunegger Straße 55, Konstanz D-78462, Germany c Institute for Applied Thermo- and Fluiddynamics IaTF, University of Applied Sciences Konstanz, Braunegger Straße 55, Konstanz D-78462, Germany b
h i g h l i g h t s Development of quality attributes significantly depends on control strategies. Online measurements show that information can be used for process control optimisation. Product temperature controlled drying is gentler at comparable drying times.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 June 2013 Accepted 30 September 2013 Available online 16 October 2013
The influence of two different control strategies, constant air temperature and constant product temperature, on product quality and drying behaviour of apples was investigated. The interactions of the drying parameters air temperature (35e100 C), product temperature (35e85 C), dew point temperature (5e30 C) and air velocity (2.0e4.8 m/s) with drying time, colour changes and shrinkage were measured continuously and determined for both strategies. Based on these results the two strategies were compared with regards to their effect on drying performance. The results show that the drying strategy has a significant influence not only on the development and duration of the drying process but also on the development of colour changes and shrinkage. Controlling product temperature led to shorter drying times and generally lower colour changes. Furthermore, it was shown that the product temperature develops characteristically; two stages and a clearly visible transition period can be detected. This potentially can be used to control the process and, therefore, improve its performance regarding duration and quality aspects. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Apple Convective drying Drying kinetics Process control Colour Shrinkage Product temperature
1. Introduction Convection drying is one of the oldest and most wide spread means of preserving food. It was identified a long time ago that, as most food products are heat sensitive, the crucial factor in the convection drying of food and other sensitive agricultural products are the product temperature and, in particular, preventing that temperature from exceeding critical levels. Excessive temperatures lead to structural, organoleptic (smell, taste, visual appearance, chewiness etc.) and nutritional changes [1]. The extent of changes of the most important quality characteristics like colour (pigments,
* Corresponding author. Newcastle Institute for Research on Sustainability, Newcastle University, Newcastle upon Tyne NE1 7RU, UK. Tel.: þ44 191 246 4951. E-mail addresses: [email protected], barbara.sturm@ newcastle.ac.uk (B. Sturm). 1359-4311/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.applthermaleng.2013.09.056
enzymatic and non-enzymatic browning) and nutritional value (anti-oxidants, vitamins) usually increases with increasing temperature. The dependency of reaction constants on temperature allows for the assumption that drying at lower temperatures leads to a lower loss of valuable compounds [2]. However, traditionally, the drying process is controlled by keeping the air temperature constant on a predefined level. Often product temperature is unknown or if it is measured it is not included in the process control system [3]. The changes of product characteristics during the drying process are related to reactions of zero-order or 1st order [4] and usually product temperature is assumed to be constant. However, several studies on medicinal herbs, fruits and vegetables [5e7] have shown that for products with high initial product moisture content, product temperature changes almost throughout the process and only comes close to air temperature at the end of the process.
456
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Nomenclature A
DE P R2 S t v x x X y
air total colour difference product coefficient of determination shrinkage time (min) velocity (m/s) arithmetic mean independent variable moisture content (gW/gDM) response variables
Several authors have shown for different sensitive biological products [8,9] that they are not excessively damaged when exposed to high air temperatures during the first period of drying. This can be explained by the initially very high moisture content of the product that leads to a great difference between dry and wet bulb temperature, therefore no thermal damage is risked in this phase. A multitude of studies have investigated the influence of drying conditions in convection drying on drying kinetics and quality aspects of apples [9e16]. The general consent is that increase of air temperature has the most significant influence on drying kinetics, in addition to colour change, whilst high air velocity is regarded as reducing shrinkage [17]. In recent years, the importance of dynamic process control has gained increased recognition in production of dried food stuffs. Research on the influence of periodically changing drying conditions showed that reducing temperature step wise from an initially high temperature results in a reduction of colour changes in guava [18]. Product quality of dried potatoes can be improved by cyclically changing drying conditions [19]. Cho and Chua [20] concluded that intelligent online-control of drying necessitates step wise change of air temperature and observation of product temperature. However, only a very limited number of studies have actively controlled the product temperature. Isothermal drying has been used for control of combined drying processes and for the measurement of effective diffusivity and shrinking of hygroscopic materials [13,21]. Nonetheless, only a few studies [22] have actually obtained and verified isothermal conditions, uniform temperature is usually assumed. According to Srikiatden and Roberts [23] isothermal conditions may be possible for convective hot air drying of a porous material having a very small characteristic dimension. Lengyel [24] showed that for the system in question, in bodies with less than 10 mm thickness, the core warms very quickly and therefore the body temperature is quasi uniform. Colour and shrinkage are two of the most important aspects in determination of product quality. They are both directly related to consumers’ appreciation of a product as they tend to associate product colour and other visual properties with its taste, hygienic security, shelf life, nutritional value and personal satisfaction [25]. Furthermore, colour is directly linked to aroma and taste of the product [26] while shrinkage tends to be linearly related to moisture content [27]. The degree of shrinkage depends directly on the temperature, humidity and velocity levels applied. This paper is concerned with the determination of the influence of drying strategy and conditions on the quality of dried apples and performance of the process represented by the duration, temperature and product quality developments. Colour and twodimensional shrinkage have been used as indicators of product quality as they comprehensively represent stresses during drying.
Greek letters b approximated coefficient ε experimental variance 4 relative humidity (%) s standard deviation w temperature ( C) Subscripts A air dp dew point Dr drying DM dry matter P product W water
2. Materials and methods 2.1. Raw material Apples of the Jonagold variety were used as experimental material. Fruits were obtained from a local farmer (Lake Constance, Germany) and stored in the fridge at 4 C. Before drying, raw material was cut into slices of 3.9 mm 0.2 mm thickness with an outer diameter of 72 mm and an inner diameter of 20 mm. Average weight per slice was 13 g 0.2 g. 2.2. Convection drying Drying tests were carried out using a high precisions laboratory dryer developed in the department of Thermal Process Engineering, University of Applied Sciences Konstanz, Konstanz (Germany). The experimental system essentially comprises of 5 units: (i) an air flow control unit, (ii) an air conditioning unit, with a water bath including heating and chilling units and a packed bed, (iii) a heating control unit with heating elements, (iv) a drying compartment that allows for either over or through flow air stream for convective drying of product (v) a non-invasive online measurement system, including a CCD camera and a pyrometer. A detailed description of the dryer was given by Sturm and Sturm et al. [7,28]. For the air temperature controlled tests, the dryer was pre heated for 20 min to reach the defined set temperatures. The apple slices were then distributed as described by Sturm [28]. The total weight of the samples was approximately 90 g. In the case of surface temperature control the dryer was turned on at the start of the experiment. Air temperature controlled drying was conducted at 35, 45, 60, 75 and 85 C. Product temperature controlled drying was conducted at 35, 40, 47.5, 55 and 60 C. During experiments where product temperature was controlled, maximum air temperature was limited to 100 C. This was due to the fact that the industrial dryers in question are not operating at any temperatures higher than 100 C. Dew point temperature was set to 5.0, 10.0, 17.5, 25 and 30 C and air velocity to 2.0, 2.6, 3.4, 4.2 and 4.8 m/s for both strategies investigated. The samples were dried until they reached a moisture content of approximately 0.13 gw/gDM. This moisture content was chosen because it is typically used in production of dried apple products (aW ¼ 0.5 [29] in desorption and 0.52 to 0.55 [30] in absorption). During drying, the weight, air temperature and product temperature were measured continuously. Colour and two dimensional shape (visible cross section) of samples were measured every five minutes using the integrated CCD camera (The Imagingsource DFK 31BU03.H). The details of the optical systems as well as the calibration were described in Sturm et al. [7]. For the representation of
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
457
changes in product colour the total colour difference DE is used, which is increasingly used in the determination of food quality changes. DE calculated from the L*, a*, b* values as follows:
DE ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 2 L* þ a* þ b*
(1)
The moisture content was measured by the gravimetric method using a laboratory convection oven (Heraeus UT 12) over 48 h at a constant temperature of 70 C. This method was chosen over the standard method of drying at 105 C for 24 h [31] as it had proven to degrade the samples. 2.3. Experimental design and statistical analysis The quality parameters during the process were evaluated using Origin 7.5 (Microcal, Northhampton, USA). Equation (2) gives the basic third order polynomic equation used.
Fig. 1. Moisture ratio, product temperatures and air temperature for both strategies as a function of time, wN ¼ 60 C, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
y ¼ b0 þ b1 x þ b2 x2þ b3 x3 þ ε εwN 0; s2
(2)
Shrinkage was best represented using linear fits. The total colour difference functions of moisture ratio was best represented using third and the drying rates using fifth order polynomial regression. The fits were significant for all parameters described (p < 0.001). Duration of drying (tDr) was evaluated as dependent variable yi (response variables). The two series of 20 tests had a randomized order. Equation (3) provides the basic quadratic regression model used to estimate the influence of the independent parameters xi on the dependent variable yk.
yk ¼ ß0 þ
3 X
ßi xi þ
i¼1
3 X i¼1
ßi x2i þ
3 3 X X
ßi;j xi xj þ ε
(3)
i ¼ 1 j ¼ iþ1
Variance and regression analysis were performed using the software Design-ExpertÓ 7.1 by Stat-Ease. Regression analysis was repeatedly carried out, successively rejecting factors by backward elimination. Although some of the factors are by definition not significant (p > 0.1) they were included as further reduction of the models would have led to a reduction of accuracy of the overall models. Table 1 shows the resulting models for both control strategies. In both cases investigated the development of the process is best represented using quadratic models. 3. Results and discussion Fig. 1 shows the temperature and moisture ratio developments as functions of time for identical nominal drying temperatures (60 C). It can be seen that drying time is significantly reduced when controlling product temperature instead of air temperature. If results for similar duration of the process are compared, a lower
maximum temperature level is required when product temperature is controlled (Fig. 2). While an increase of drying temperature (Fig. 2) and dew point temperature (Fig. 3) have a similar effect on resulting drying times for both strategies applied, the change of air velocity (Fig. 4) has a much more significant influence when air temperature is controlled than when product temperature is controlled. This phenomenon can be explained by the fact that in product temperature controlled drying the lower velocity and, therefore, mass flow is compensated by an increase of air temperature, as the driving force for the control system is to reach the set product temperature [28]. Fig. 5 depicts the drying rates as a function of moisture ratio. When product temperature is controlled, the relative drying rates in the beginning are much higher and decrease much more significantly than when air temperature is controlled. This can be explained with the fundamentally different influences of isothermal and non-isothermal conditions on drying behaviour. When comparing nearly identical drying times (A60, P47) the drying rate for P47 is higher than that of A60 until a moisture ratio of 0.4 is reached. Comparing the temperature developments (Figs. 6 and 7) it can be seen that at this point air temperature in product temperature controlled drying falls below 60 C. At the same time product temperature in air temperature controlled drying exceeds 47 C. Fig. 6 depicts the temperature development of the air temperature controlled samples. The samples are transitioning from the second to the third phase of drying at a moisture ratio of approximately 0.2 which translates to a moisture content of circa 2 gw/gDM. At the end of this transition the product temperature is nearly constant and assimilating to air temperature. The described temperature developments can be explained by the relationships between dry bulb, wet bulb and dew point temperature. At the
Table 1 Equations for prediction of drying time and their statistics. Response
Modela
R2a
Pb
xc
s xd
tA tP
tDr ¼ 1298:08 23:87wA þ 8:15wDp 149:05vA 0:11wA wDp þ 0:86wA vA þ 0:14w2A þ 8:82v2A tDr ¼ 1384:96 51:27wP þ 23:45wDp þ 30:28vA 0:37wP wDp 2:60wDp vA þ 0:51w2P þ 0:14w2Dp
0.989 0.946
<0.0001 <0.0001
195 207
14 21
a b c d
Coefficient of determination. Pr > F. Arithmetic mean. Standard deviation.
458
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Fig. 2. Drying time as a function of drying temperature, wDp ¼ 17.5 C, vA ¼ 3.4 m/s. Fig. 5. Drying rate as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
Fig. 3. Drying time as a function of dew point temperature, wA ¼ 60 C, wP ¼ 47.5 C, vA ¼ 3.4 m/s.
beginning of the process the moist body very quickly reaches the wet bulb temperature, which is lower than the effective air temperature but higher than the dew point temperature and depends directly on air temperature and humidity. In the first phase of
Fig. 4. Drying time as a function of air velocity, wA ¼ 60 C, wP ¼ 47.5 C, wDp ¼ 17.5 C.
drying the product temperature will stay constant, in the second and consequent third phase, it rises until it reaches the temperature of the air at the end of the process. When controlling product temperature, air temperature will rise as long as the surface temperature has not reached the set-point. To maintain a constant product temperature at a decreasing difference of wet and dry bulb temperature air temperature then needs to be reduced accordingly. In case of product temperature control (Fig. 7) a transition phase can be observed too and starts with a significant air temperature drop around a moisture ratio of 0.3 and reaches its end at a moisture ratio of roughly 0.2. Independent of the strategy the transition from one phase to the next can clearly be detected by the development of temperature: in case of air temperature controlled drying using the product temperature and vice versa. The experiments described above show that controlling the process using the product temperature as the control variable is significantly different to traditional control of air temperature. The drying kinetics agree with the results of Srikiatden and Roberts [13] who used a combination of convection and microwave for isothermal drying of apples. While the drying kinetics for air
Fig. 6. Air and product temperatures as a function of moisture ratio for air temperature controlled drying, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
Fig. 7. Air and product temperature as a function of moisture ratio for product temperature controlled drying, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
temperature control can be described by Fick’s diffusion model, this is not possible for product temperature control. Controlling product temperature initially leads to very high drying rates, which can be explained by the high initial air temperatures. After the product has reached the set temperature, drying rate decreases quickly with reducing air temperatures. Especially after the transition from the second to the third drying phase air temperature is only marginally higher than product temperature, which leads to an unnecessary prolonging of process if product temperatures are set too low. The developments of product temperatures depicted for air temperature controlled drying clearly show that in the case of hot air apple drying the assumption of isothermal conditions is not valid. Both strategies applied significantly affect the colour of the product. Apple tissue underwent extensive non-homogeneous colour changes (Fig. 8). In the case of air temperature control, the application of high temperatures led to a decrease in discolouration at otherwise constant process conditions, which is in agreement with the findings of Contreras et al. [32] for apples and Pott et al.
Fig. 8. Total colour difference DE as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
459
[33] for mangoes. In the case of product temperature controlled drying, however, the situation is different. After the initial significant changes in colour, the DE stayed almost steady for an extended period. At a moisture ratio of about 0.4 it started increasing significantly again. Overall DE is lower for product temperature controlled systems and depends less on the process settings. One possible explanation for this might be the inactivation of polyphenol oxidase PPO at high initial air temperature levels. PPO catalyses the hydroxylation and oxidation of phenolic compounds in fruits and is the major cause for enzymatic browning. Its inactivation reduces browning significantly. This is in accordance to the findings of Yemenicioglu et al. [34]. Shrinkage, displayed in Fig. 9, is significantly influenced by both the strategy and the temperature levels applied which is in agreement with literature [12,14]. When air temperature is kept constant, shrinkage is significantly higher than that of comparison specimen which is initially dried at very high air temperatures. This might be due to the fact that slow drying reduces internal stresses, which in consequence results in increased shrinkage. Applying high temperatures, however, increase internal stresses but at the same time the resulting fast drying leads to a mechanical stabilisation of the surface [12,14]. Drying characteristics and development of quality aspects are significantly different for single stage and non-stationary drying. For both strategies applied, two clearly visible phases were characterized, separated by sizable changes of product temperature (air temperature controlled, Fig. 6), respectively air temperature (product temperature controlled, Fig. 7). Analogously, colour changes are divided into clearly distinguishable phases, see Fig. 8. Shrinkage is significantly reduced by higher temperatures at the beginning of the process. The required drying time varies significantly less for product temperature control than for air temperature control. The factors investigated, temperature, air humidity and air velocity affect the drying results significantly differently to air temperature control. Using product temperature as the control variable results in significantly better optical product characteristics. It can be assumed that applying a lower maximum product temperature at comparable drying times additionally results in a higher retain of valuable compounds such as vitamins and aromas [35e38]. The results obtained give valuable information for the development of optimized control strategies. Keeping the control
Fig. 9. Shrinkage as a function of moisture ratio, wDp ¼ 17.5 C, vA ¼ 3.4 m/s.
460
B. Sturm et al. / Applied Thermal Engineering 62 (2014) 455e460
variable constant throughout the process, either air temperature or product temperature, does not result in an optimum solution. Based on the results obtained a step wise process needs to be developed. In general, both air or product temperature could be used for this; a combination of the two might even be considered. The crucial information necessary for either is the point of transition, which can be obtained from the information of temperature development. Reductions of drying times of up to 40% can be expected for the suggested control strategy. 4. Conclusions A comparative study of single stage and non-stationary convection drying of apples utilizing non-invasive measurement devices showed the advantages of on-line monitoring of important state variables as product temperature, colour and shrinkage. It was possible to determine the influence of both the strategy and process conditions on drying kinetics, temperature and quality development throughout the process. Results presented allow deduction for optimised process conditions. Segmentation of the drying process into phases with different drying conditions and control strategies shows potential. Additionally, an active integration of information derived from pictures taken continuously throughout the process into the control system is conceivable. Based on the results obtained adaptive control systems for existing and newly designed drying devices can be developed. Increasing process temperature during the first phase of drying will reduce processing time and, therefore, increase throughput and reduction of specific energy demand while the integration of optical systems might allow for more targeted control of specific product characteristics. Integration of active product temperature control can prevent overheating of the product and therefore increased degradation. Initial tests on other products, such as onions and leek, showed that the control strategies can be applied to these products with similar results regarding drying time and product quality. Implementation of sensors and control systems in existing tray dryers was successful. However, further research is necessary to determine correct settings, especially concerning the influence of differences in drying rate of particles depending on their relative position to the air inlet. Acknowledgements The authors acknowledge the German Federal Ministry of Economy and Technology for the financial support within the PRO INNO II program. References [1] G.H. Crapiste, Simulation of drying rates and quality changes during the dehydration of foodstuffs, in: J.-E. Lozano, C. Anon, E. Parada-Arais, BarbosaCanovas (Eds.), Trends in Food Engineering, Technomic Publishing Co., Pennsylvania, USA, 2000. [2] C. Ratti (Ed.), Advances in Food Dehydration, CRC Press, Boca Raton, Florida, USA, 2009. [3] A.S. Mujumdar, C.L. Law, Drying technology: trends and applications in postharvest processing, Food and Bioprocess Technol. 3 (6) (2009) 843e852. [4] K. Kröll, W. Kast, Trocknen und Trockner in der Produktion, Bd. 3, Springer Verlag, Berlin, Heidelberg, Germany, 1989. [5] J. Müller, Trocknung von Arzneipflanzen mit Solarenergie, Eugen Ulmer GmbH & Co., Stuttgart, 1992. Dissertation Universität Hohenheim, Germany. [6] E.F.A.A. Bashir, Solar Drying of Sliced Onion and Quality Attributes as Affected by the Drying Process and Storage Conditions. VDI-MEG 328. Dissertation, Universität Hohenheim, Germany, 1998. [7] B. Sturm, W. Hofacker, O. Hensel, Optimizing the drying parameters for hot air dried apples, Drying Technol. 30 (14) (2012) 1570e1582.
[8] K.J. Chua, A.S. Mujumdar, M.N.A. Hawlader, S.K. Chou, J.C. Ho, Batch drying of banana pieces e effect on stepwise change in drying air temperature on drying kinetics and product colour, Food Res. Int. 34 (8) (2001) 721e731. [9] E.L. Schultz, M.M. Mazzuco, R.A.F. Machado, A. Bolzan, M.B. Quadri, M.G.N. Quadri, Effect of pre-treatment of drying, density and shrinkage of apple slices, J. Food Eng. 78 (3) (2007) 1103e1110. [10] A. Kaya, O. Aydin, C. Demirtas, Drying kinetics of red delicious apple, Biosyst. Eng. 96 (4) (2007) 517e524. [11] H.O. Menges, C. Ertekin, Mathematical modeling of thin layer drying of golden apples, J. Food Eng. 77 (1) (2006) 119e125. [12] P.P. Lewicki, E. Jakubczyk, Effect of hot air temperature on mechanical properties of dried apples, J. Food Eng. 64 (3) (2004) 307e314. [13] J. Srikiatden, J.S. Roberts, Moisture loss kinetics of apple during convective hot air and isothermal drying, Int. J. Food Prop. 8 (3) (2005) 493e512. [14] Y. Bai, M.S. Rahman, C.O. Perera, B. Smith, L.D. Melton, Structural changes in apple rings during convection air-drying with controlled temperature and humidity, J. Agric. Food Chem. 50 (11) (2002) 3179e3185. [15] L. Fernandez, C. Castillero, J.M. Aquilera, An application of image analysis to dehydration of apple discs, J. Food Eng. 67 (1e2) (2005) 185e193. [16] A. Vega-Galvez, A.H. Kong, M. Chcana, J. Vergara, J. Martinez-Monzo, P. GarciaSegovia, R. Lemus-Mondaca, K. Di Scala, Effect of temperature and air velocity on drying kinetics, antioxidant capacity, total phenolic content, colour, texture and microstructure of apple (var. Granny Smith) slices, Food Chem. 132 (1) (2012) 51e59. [17] C. Ratti, Shrinkage during drying of food stuffs, J. Food Eng. 23 (1) (1994) 91e105. [18] K.J. Chua, S.K. Chou, J.S. Ho, A.S. Mujumdar, M.N.A. Hawlader, Cyclic air temperature during drying of guave pieces: effects on moisture and ascorbic acid contents, Food and Bioprod. Process. 78 (2) (2000) 28e72. [19] J.C. Ho, S.K. Chou, K.J. Chua, A.S. Mujumdar, M.N.A. Hawlader, Analytical study of cyclic temperature drying: effect on drying kinetics and product quality, J. Food Eng. 51 (1) (2003) 65e75. [20] S.K. Chou, K.J. Chua, On the study of the drying behavior of a heat-sensitive biomaterial undergoing stepwise-varying temperature schemes, Ind. Eng. Chem. Res. 42 (20) (2003) 4939e4952. [21] J.S. Roberts, C.H. Tong, The development of an isothermal drying apparatus and the evaluation of the diffusion model on hygroscopic porous material, Int. J. Food Prop. 6 (1) (2003) 165e180. [22] J.F. Nicoleti, J. Telis-Romero, V.R.N. Telis, Air-Drying of fresh and osmotically pre-treated pineapple slices: fixed air temperature versus fixed slice temperature drying kinetics, Drying Technol. 19 (9) (2001) 2179e2191. [23] J. Srikiatden, J.S. Roberts, Predicting moisture profiles in potato and carrot during convective hot air drying using isothermally measured effective diffusivity, J. Food Eng. 84 (4) (2008) 516e525. [24] A. Lengyel, The change of body temperature during convective drying of cubeshaped apple pieces, Drying Technol. 25 (7e8) (2007) 1275e1280. [25] F. Pedreschi, J. Léon, M. Domingo, P. Moyano, Development of a computer vision system to measure the colour of potato chips, Food Res. Int. 39 (10) (2006) 1092e1098. [26] N.J. Morris, I.W. Lohmann, R.T. O’Connor, A.F. Freeman, I.V. Peanut Butter, Determination of color of peanut butter by spectral reflectance method, Food Technol. 7 (1953) 393e396. [27] L. Mayor, A.M. Sereno, Modelling shrinkage during convective drying of food materials: a review, J. Food Eng. 61 (3) (2004) 373e386. [28] B. Sturm, Einfluss der Führung des Trocknungsprozesses auf den Trocknungsverlauf und die Produkteigenschaften empfindlicher Biologischer Güter (Influence of Process Control on Drying Kinetics and Product Attributes of Sensitive Biological Products), in: Forschungsbericht Agrartechnik 491 des Arbeitskreises Forschung und Lehre der Max-Eyth Gesellschaft Agrartechnik VDI (VDI-MEG), Witzenhausen, Germany, 2010. [29] M. Wolf, J.E. Walker Jr., J.G. Kapsalis, Water vapor sorption hysteresis in dehydrated foods, J. Agric. Food Chem. 20 (1972) 1073e1077. [30] P.P. Lewicki, A. Lenart, Wptyw wstepnego odwadniania osmotycznego na wlásciwosci adsorpcyjne jablek suszonych, Przem. Spozyw. 30 (1977) 141. [31] AOAC Official Methods of Analysis, fifteenth ed., Association of Official Analytical Chemists, Washington D.C, 1990. [32] C. Contreras, M.E. Martin-Esparza, A. Chiralt, N. Martinez-Navarrete, Influence of microwave application on convection drying: effects of drying kinetics, and optical and mechanical properties of apples and strawberries, J. Food Eng. 88 (1) (2008) 55e64. [33] I. Pott, S. Neidhard, W. Mühlbauer, R. Carle, Quality improvement of nonsulphited mango slices by drying at high temperatures, Innovative Food Sci. Emerging Technol. 6 (4) (2005) 412e419. [34] A. Yemenicioglu, M. Ozkan, B. Cemeroglu, Heat inactivation kinetic of apples polyphenoloxidase and activation of its latent form, J. Food Sci. 62 (3) (2006) 508e510. [35] O.R. Fennema (Ed.), Food Chemistry, second ed., Marcel Dekker, New York, USA, 1985. [36] M.D.A. Hawlander, C. Perera, M. Tian, K.L. Yeo, Drying of guava and papaya: impact of different drying methods, Drying Technol. 24 (1) (2006) 77e87. [37] M. Miranda, H. Maureira, K. Rodríguez, K. Vega-Gálvez, Influence of temperature on the drying kinetics, physicochemical properties, and antioxidant capacity of Aloe Vera (Aloe Barbadensis Miller) gel, J. Food Eng. 91 (2) (2009) 297e304. [38] N. Boudhrioua, P. Giampaoli, C. Bonazzi, Changes in aromatic components of banana during ripening and air-drying, Lebensm. Wiss. Technol. 36 (6) (2003) 633e642.