A comparison of dynamic mechanical properties of processing-tomato peel as affected by hot lye and infrared radiation heating for peeling

A comparison of dynamic mechanical properties of processing-tomato peel as affected by hot lye and infrared radiation heating for peeling

Journal of Food Engineering 126 (2014) 27–34 Contents lists available at ScienceDirect Journal of Food Engineering journal homepage: www.elsevier.co...

1MB Sizes 0 Downloads 23 Views

Journal of Food Engineering 126 (2014) 27–34

Contents lists available at ScienceDirect

Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng

A comparison of dynamic mechanical properties of processing-tomato peel as affected by hot lye and infrared radiation heating for peeling Yong Wang a,b,1, Xuan Li a,1, Gang Sun c, Dong Li d, Zhongli Pan a,e,⇑ a

Department of Biological and Agricultural Engineering, University of California at Davis, Davis, CA 95616, USA COFCO Nutrition & Health Research Institute, COFCO, Beijing 100020, China c Division of Textiles and Clothing, University of California at Davis, Davis, CA 95616, USA d College of Engineering, China Agricultural University, Beijing 100083, China e Western Regional Research Center (WRRC), United States Department of Agriculture (USDA), Agricultural Research Service (ARS), Albany, CA 94710, USA b

a r t i c l e

i n f o

Article history: Received 31 March 2013 Received in revised form 26 July 2013 Accepted 21 October 2013 Available online 29 October 2013 Keywords: Dynamic mechanical analysis Frequency sweep Temperature scan Creep behavior Infrared radiation Tomato peel

a b s t r a c t This study investigated the viscoelastic characteristics of tomato skins subjected to conventional hot lye peeling and emerging infrared-dry peeling by using dynamic mechanical analysis (DMA). Three DMA testing modes, including temperature ramp, frequency sweep, and creep behavior test, were conducted to evaluate the transition temperatures and dynamic moduli of tomato peels heated by infrared radiation and hot lye at four heating durations (30, 45, 60, and 75 s). Fresh tomato peels were used as a control. Results showed that dynamic moduli of tomato peels were sensitive to temperature ramp and frequency sweep tests. Over a temperature range from 20 °C to 100 °C, transition temperatures of infrared treated peels (63–72 °C) and lye treated peels (43–75 °C) were significantly lower than those of fresh control (88 °C). Values of both storage and loss moduli of infrared heated peels were considerably higher than those of the fresh control, whereas values of the storage and loss moduli from the lye peeled samples were lower than those of fresh peels. DMA tests effectively differentiated the viscoelastic behaviors of tomato peels and indicated mechanistic differences between the lye peeling and infrared dry-peeling. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Tomatoes are usually peeled prior to further processing into canned products since tomato skins are very tough and undesirable to consumers (Barringer, 2003; Shao et al., 2013). The conventional peeling process applies hot lye to separate skins from tomato flesh and then utilizes a mechanical peeler to remove the loosened skins. The use of lye (sodium hydroxide or potassium hydroxide) for peeling tomatoes results in a significant amount of peeling effluent discharges containing high salinity and organic solids, causing considerable negative environmental impacts (Rock et al., 2011). Sustainable and non-chemical peeling alternatives have long been desired by tomato processors to eliminate the reliance on lye and water. For the first time, we have investigated an infrared (IR) dry-peeling method which can reduce water usage and wastewater while producing high quality peeled products without using lye and water (Pan et al., 2009, 2011; Li, 2012). IR heating is non-ionizing radiation with surface heating characteristics. The IR radiation heats only a shallow layer of tomato surface ⇑ Corresponding author. Address: Processed Foods Research Unit, USDA-ARSWRRC, Albany, CA 94710, USA. Tel.: +1 510 559 5861; fax: +1 510 559 5851. E-mail addresses: [email protected], [email protected] (Z. Pan). 1 These authors contributed equally to this work. 0260-8774/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jfoodeng.2013.10.032

and leaves the edible inner part of the tomato with minimum changes in texture and nutritional quality (Li and Pan, 2013a,b). In our efforts in developing this IR dry-peeling method, the dynamic nature of viscoelastic properties of tomato skin is of particular interest and importance because these properties are closely related to several pivotal engineering parameters, such as minimal IR heating time for successful peel removal and critical temperature for peel separation. Thus, in-depth studies are needed to better understand how different peeling methods and conditions affect the dynamic mechanical properties of tomato skins. Tomato skin, also known as exocarp, consists of a thin cuticle layer, a single layer of epidermal cells, and two to four layers of hypodermal cells (Fig. 1). Epidermal and hypodermal cells are tablet shaped collenchymas which normally feature with unevenly thickened cell walls with the greatest thickness of the cell wall located in the cell corner (Evert, 2006). A hydrophobic cuticle, which mainly consists of solvent-soluble and polymerized lipids (Matas et al., 2005), overlies the epidermal cells as a continuous extracellular membrane (Bargel and Neinhuis, 2005; López-Casado et al., 2007). In addition to physiologically defined skin, mechanically removed skins after peeling also have attached small portions of soft pulp (Garcia and Barret, 2006; Li et al., 2009) consisting of round shaped outer pericarp cells. All these cellular tissues as a unique and complex mix of biopolymers need to be considered

28

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

Epidermal cells

Cuticle

Hypodermal cells

Pericarp cells

1 mm

Fig. 1. Anatomical features of outermost pericarp tissue of tomato in the cross section. (The sample in the image of scanning electron microscopy was prepared by chemical fixation and critical-point drying in our preliminary study.)

when investigating the skin viscoelastic characterizes affected by different peeling methods and conditions. Different peeling methods and conditions have substantial effects on the biomechanical and viscoelastic properties of tomato peel, such as skin strength, storage moduli, and loss moduli. However, there is a lack of documented information and fundamental understandings related to characterized viscoelastic response of tomato skins to different peeling methods. The strength of tomato skin primarily comes from the intrinsic strength of primary cell walls, which possess a complicated three dimensional ultrastructure formed by a cellulose–hemicellulose network, pectin matrix, structural proteins, and other non-polysaccharide components such as phenolics (Taiz and Zeiger, 2006). When such an intricate structure is exposed to any dramatic environmental stress, destructive changes in skin tissue occur. For example, diffusion of hot lye solution into tomato skin during the lye peeling process weakens the cellulose–hemicellulose network of cell walls (Barreiro et al., 1995, 2007; Barringer, 2003). As a result, the cuticle melts, pectins in the middle lamella breakdown, and cell wall structures degrade, a sequence leading to skin dissolving (Floros and Chinnan, 1990). Skin separation induced by IR dry-peeling does not involve any chemical diffusion, and thus differs from that of the traditional hot lye peeling. The exact mechanism of skin separation during IR heating still needs to be elucidated. (Pan et al., 2009). When the tomato surface is exposed to a high temperature (>90 °C), thermal effects will dominate the tissue damage and layer separations. Our previous studies have demonstrated that IR heating can substantially affect the microstructure and biomechanical properties of tomato skin (Li, 2012; Li et al., 2013). Characterization of the viscoelastic behavior of tomato skin in response to different processing conditions will add greatly to the understanding of the peeling mechanism of the IR dry-peeling process. Changes in dynamic mechanical properties of polymeric tomato peels under different processing conditions are extremely difficult to measure, especially when biological uncertainty is considerable and the processing conditions cover a range of temperature or frequencies. A refined analytical method that attempts to better characterize the viscoelastic properties of biopolymer materials is known as dynamic mechanical analysis (DMA). During the DMA measurements, a small oscillatory deformation is applied to a tested sample and allows the sample to be studied in response to the temperature, frequency, stress, strain, or other parameters (Menard, 1998). The resultant dynamic moduli are then used to characterize the viscoelastic properties of tested materials. The DMA technique has been widely employed in polymer science to uniquely identify potential transition points of polymeric materials

or changes of viscoelastic behaviors of a material as a function of temperature or frequency (Pothan et al., 2003). In the present work, a comparison of dynamic viscoelastic behavior of polymeric tomato peels as affected by IR and lye peeling was performed using the DMA technique. Three test modes were implemented to characterize the viscoelastic nature of tomato skin. First, a temperature ramp test was used to characterize the dynamic response of tomato skins to temperature. This test leads to identification of the potential phase transitions occurring in tomato skin when the tomato surface temperature was increased from about 20 °C to 100 °C. Second, to expand the understanding of whether tomato peels exhibit frequency dependent viscoelastic properties, a series of oscillatory forces of different magnitude and frequency were applied to the tomato skin by means of the frequency sweep test. Different frequencies were used to simulate a realistic peeling environment with complex vibration and oscillation. Changes in viscoelastic parameters of tomato skin can be characterized by the power law model over the entire range of tested frequency (Özkan et al., 2002). Finally, creep behavior of tomato skin was studied by applying loading and unloading uniaxial tension. The viscoelastic response of tomato skin over time can be determined using Burger’s model. The Burger’s model was developed based upon the Maxwell model and the Voigt–Kelvin model and employs four elements to approximate the creep response (Chuang and Yeh, 2006; Menard, 1998). Analyses using the three DMA test modes were expected to give insight into how tomato skin reacts under different peeling conditions and how its viscoelastic properties vary under different heating methods. Therefore, the main objectives of this study were to (1) characterize the dynamic mechanical properties of tomato peels under three different DMA test modes, and (2) compare the effects of IR and lye peeling methods on the changes in viscoelastic properties of tomato skins.

2. Materials and methods 2.1. Tomatoes Tomatoes of cultivar AB2 grown on a commercial farm (Campbell’s Seeds Co., Woodland, Cal., USA) were used for all the DMA tests. Selected tomatoes were randomly harvested at red-ripening stage (179–183 days after planting) each week over the 2010 peak harvesting season (from August to September). Following industry practice, to ensure homogeneous and consistent ripening of tomatoes, ethylene gas was applied to tomatoes one week before harvest. After harvesting, tomatoes were immediately delivered to the laboratory and stored at 10 °C and 80% relative humidity to avoid chilling injury (Kader, 2002). The DMA tests were completed within four days after each harvest so that the tomatoes were of consistent quality. The tested tomatoes had an average mass of 83 ± 16 g and an average soluble content of 4.9 ± 0.2 °Brix (Li, 2012).

2.2. Peeling procedures Prior to DMA testing, tomatoes were subjected to IR or lye heating. Tomatoes underwent double-sided IR heating for one of the following durations: 30, 45, 60, or 70 s. The distance between two IR emitters was set at 90 mm. The tomatoes were rotated at a speed of 1 rpm to ensure uniform heating. For lye heating, tomatoes were heated in a 10% (w/v) sodium hydroxide solution at 96 °C for the same four time duration as used for IR heating. The detailed experimental set-up and peeling procedure were previously described by Pan et al. (2009) and Li et al. (2013).

29

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

2.3. Dynamic mechanical analysis

2.3.3. Temperature ramp test During the temperature ramp test, a constant strain was applied and the temperature was varied while recording the stress changes within peel specimens (Menard, 1998). The strip segment was held with a constant preload displacement of 0.05 mm and heated from 20 °C to 100 °C at a heating rate of 5 °C/min. The frequency was maintained at 1 Hz through the whole temperature ramp. The storage and loss moduli were calculated from the measured force and dimensional changes with the increasing temperature with PerkinElmer DMA software (PerkinElmer, Waltham, Mass., US).

2.3.1. Sample preparation After peeling, rectangular uniform segments (9 mm  5 mm) were dissected from tomato skins along the stem–blossom axis direction by using a striking die cutter (The Right Image Co., Sacramento, Cal., USA) (Fig. 2A). The die cutter was custom designed for DMA analysis and contains four rectangular-shaped blades mounted on a wood block to assure uniform shape and size of the segment (Fig. 2B). Soft pulp attached to the excised segment was removed gently using a razor blade before each DMA test. The thickness, width, and length of each segment were measured at three locations with a digital micrometer having an accuracy of 0.001 mm. After dimensional measurements, each flat strip segment was inspected for possible damage or micro-cracks. Only segments with no cracks or damage were used for subsequent DMA tests. All sample preparation and DMA tests were conducted in a controlled environment of a textile laboratory in the University of California at Davis, at 20 °C and 75% relative humidity.

2.3.4. Frequency sweep test and parameter evaluation The frequency sweep tests were performed at a constant preload displacement of 0.05 mm over the frequency range from 0.01 to 10 Hz. The testing frequency and displacement were selected based on preliminary experimental results. The frequency measurements of storage modulus (E0 ) and loss modulus (E00 ) show a power law behavior between E0 or E00 and frequency (x) (Ikeda and Nishinari, 2001; Ikeda and Foegeding, 1999; Menard, 1998; Wang et al., 2009).

2.3.2. Dynamic mechanical measurements Dynamic mechanical properties of tomato peel were examined using a DMA 8000 system (PerkinElmer, Waltham, Mass., USA) under three test modes: a temperature ramp test, a frequency sweep test, and a creep test. The DMA 8000 system utilized a driver shaft connected to a dual cantilever jig (Fig. 2C). Two ends of the flat strip segment were carefully mounted directly onto the jig grips under gentle tension with screws to ensure an adequately firm grip (Fig. 2C). The gauge length of each segment between the two clamps was measured with a hand-held caliper prior to tests. To prevent moisture loss during the frequency sweep and the creep testing, the portion of segment between two clamps was loosely wrapped with plastic film. To prevent moisture loss and at the same time allow use of high temperature in the temperature ramp test, aluminum foil was used to cover the tested portion of the segments. After a head cover was installed outside the specimen holder, the specimen was equilibrated for 10 min prior to testing (Fig. 2D). Average and standard deviation for each treatment are reported for three replicates.

A

0

E0 ¼ K 0  xn

E00 ¼ K 00  xn

ð1Þ 00

ð2Þ n

where K0 and K00 are the proportionality constants, MPa s ; x represents the frequency, Hz; n0 and n00 are the dimensionless frequency exponents. The values of all four parameters (i.e., K0 , K00 , n0 , and n00 ) were determined by fitting the Eqs. (1) and (2) to data using a least-squares method. 2.3.5. Creep test and parameters evaluation The creep experiments were carried out by applying a 0.3 N static force to the segment along its longitudinal axis. The changes in total strain in response to the applied force were measured over a period of 2 min. Once the force was removed, the specimen was allowed to recover for another 2 min. Creep behavior of viscoelastic materials was approximated using Burger’s model. The total strain e is given by the following expression (Huang et al., 2006):

B

C

D Dual Cantilever Jig

Head Cover

Driver Shaft Segment Peel Segment

Fig. 2. Equipment of DMA 8000 and experimental setup: (A) tomato peel dissected along the blossom–stem axis direction; (B) die cutter for segments dissection; (C) flat strip peel segment mounting on the cantilever clamps; (D) front view of DMA8000.

30

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

pattern, but the values of the storage modulus were considerably greater in magnitude than those of the loss modulus. This means that the elastic property is the dominant factor in tomato peel. Hence, values of the storage modulus are mainly discussed hereafter. Transition points were identified as the lowest points on the storage moduli curves. The corresponding transition temperature for the four IR heating conditions ranged from 63 °C to 75 °C (Fig. 3A). Prior to the transition zone, a decrease in the storage moduli occurs, indicating a reduction of the elastic property of tomato peel. A similar trend has been reported from a previous study of the temperature effect on the elastic modulus of tomato cuticular membranes (Matas et al., 2005). In that study, a significant decrease in elastic modulus of tomato peels occurred when temperature increased from 10 °C to 45 °C. Test temperature was extended over a broader range in this study, and increased stiffness of tomato skin occurred when the temperature exceeded the transition points. This phenomenon is also known as strain-hardening (Huang et al., 2006). Fundamentally, higher temperature tends to cause melting of tomato cuticular waxes, loss of free water molecules, and further dissociation of biopolymers such as pectin lying in the middle lamella and epidermal cell walls. Similar parabolic curves for both storage and loss moduli were also observed for lye heated samples (Fig. 4). However, the starting magnitude of the storage moduli of all lye heated samples was significantly lower (a < 0.05), whereas that of IR heated samples (Fig. 3) was significantly higher, as compared to the fresh control. The different values of dynamic moduli for lye and IR heated skins indicate that fundamentally different mechanisms caused skin separation. The hot lye solution appears to diffuse into and reacts with

120

100

Storage Modulus (MPa)

A

Fresh control IR-30s IR-45s IR-60s IR-75s

80

60

40

20

0 20

30

40

50

60

70

80

90

100

o

Temperature ( C) 30

Loss Modulus (MPa)

B

Fresh control IR-30s IR-45s IR-60s IR-75s

25

20

15

10

5

0 20

30

40

50

60

70

80

90

100

o

Temperature ( C)

40

A

Fig. 3. Representative dynamic moduli of IR heated peels and fresh control peels with respect to increasing temperature: (A) storage moduli; (B) loss moduli.

E1

þ

r E2

1e

t

s



r þ t g

ð3Þ

where r is the stress, MPa; E1 is the instantaneous elastic modulus, MPa; E2 is the retarded elastic modulus, MPa; s is the relaxation time, s; g is the coefficient of viscosity associated with viscosity flow, MPa s. By fitting the experimental data to Burger’s model, parameters E1, E2, g, and s are obtained to characterize the creep behavior of tomato peels under different peeling conditions. A statistical analysis, when necessary, was carried out by using SAS software (version 9.2, SAS Institute Inc., Cary, N.C., USA). The calculated parameters were subjected to ANOVA with a significance level of 0.05 and followed by a Duncan’s multiple comparison to determine difference among means.

Storage Modulus (MPa)



r

35

20 15

5 20

30

40

50

60

70

80

90

100

o

Temperature ( C) 70

B

Fresh control Lye-30s Lye-45s Lye-60s Lye-75s

60

Loss Modulus (MPa)

Changes of storage and loss moduli of tomato peels in response to the increasing temperature followed parabolic relationships for both IR heated and fresh samples (Fig. 3). Storage and loss moduli first decreased and then increased with the increase in temperature. Despite having a similar pattern, storage and loss moduli were significantly higher a < 0.05) for all IR heated samples when compared to the fresh control. The results suggest that the thermal effects due to IR heating considerably changed the viscoelastic properties of tomato skin. Variations in the values of initial storage modulus among the control and different IR heated samples were mainly due to the heterogeneous nature of skins. For all samples, both storage modulus and loss modulus changed in a similar

25

10

3. Results and discussion 3.1. Temperature ramp measurements

Fresh control Lye-30s Lye-45s Lye-60s Lye-75s

30

50 40 30 20 10 0

20

30

40

50

60

70

80

90

100

Temperature (oC) Fig. 4. Representative dynamic moduli of lye heated peels and fresh control peels with respect to increasing temperature: (A) storage moduli; (B) loss moduli.

31

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

the cuticle and tomato epidermal tissue via complex biochemical reactions. Chemical diffusion and biochemical reactions eventually damaged the epidermal and pericarp tissues in the lye peeling method. Unlike lye peeling, IR heating demonstrated the advantages of the rapid surface heating and high heat delivery for skin loosening. The high heat was delivered through radiation heat transfer and penetrated to a depth less than 1 mm (Ginzburg, 1969). Radiation heat exchange on tomato skin and heat flow within tomato tissues through conduction heat transfer resulted in thermally induced peel loosening. In conclusion, the changes of viscoelastic properties of IR heated peels are primarily attributable to complicated heat transfer mechanisms and the resultant thermal effects. It is noticed that large biological variations existed in our measurements of dynamic modulus values may be due to the complex structure of tomato epidermal tissues. Similar studies have also reported considerable variations in biomechanical properties of tomato skins (Hershko et al., 1994; Hetzroni et al., 2011; Lopez-Casado et al., 2010). Given the fact that skin is a matrix of various constitutive biopolymers, comparison of the dynamic mechanical properties of tomato skins between lye and IR heating in the following study was emphasized on the overall tendency of the skin viscoelastic behaviors under different treatment and heating conditions. To better understand the effect of IR and lye heating on skin viscoelastic properties, the transition temperatures at four different heating times of the two peeling methods were measured. As illustrated in Fig. 5A and B, differences in transition temperatures among different heating conditions for either lye or IR heating were insignificant. However, all transition temperatures of IR and lye heated samples were significantly lower than those of the fresh control. The corresponding storage and loss modulus values at different transition temperatures were further compared in Fig. 5C and D. For IR heated samples, values of storage and loss moduli

were greater than those of fresh control (Fig. 5C). Increased storage modulus was consistent with the findings of the tensile failure stress (Li, 2012), indicating that skin stiffness was enhanced. Higher storage moduli for lye heated samples than for fresh control were found at the transition temperature (Fig. 5D), even though the storage moduli were lower than the control before reaching the transition point (Fig. 4A). The changes of storage modulus could be partially attributed to the loss of moisture and partially due to the further degradation of cuticular membrane and pectin polysaccharides during the temperature ramp. Cuticular membrane on tomato skin and pectin in the middle lamella between fruit cells contributed significantly to the mechanical structure of tomato skin (Anthon and Barrett, 2010; Matas et al., 2004a; Shi et al., 1997; Houben et al., 2011). In addition, the measured storage modulus values of fresh control were comparable to the elastic modulus of tomato skin reported in previous studies (Matas et al., 2005; Hetzroni et al., 2011) but slightly lower than the elastic modulus of the tomato cuticular membrane (Matas et al., 2004b). As expected, the cuticular membrane provides the major strength of tomato skin (Allende et al., 2004; Matas et al., 2004b), although the tested peeled tomato skins were composed of cuticular membranes on top of epidermis, collenchymas epidermal cells, and the parenchyma cell structure beneath the epidermis. Several anatomical differences of a tested specimen may increase the modulus to some extent. 3.2. Frequency sweep measurements Fig. 6 shows that the storage and loss moduli increased with the increase in frequency for IR heated samples and control. All modulus values of IR heated samples were significantly higher than that of the control (a < 0.05), a finding was consistent with the results of the temperature ramp tests. Also similar to the findings of 100

A

o

80

Transition Temperature ( C)

o

Transition Temperature ( C)

100

60

40

20

B 80

60

40

20

0

0 Fresh Control IR-30s

IR-45s

IR-60s

Fresh Control Lye-30s Lye-45s Lye-60s Lye-75s

IR-75s

Treatment Conditions

Treatment Conditions 25

80 70

C

Storage modulus Loss modulus

Storage modulus Loss modulus

D

20

Modulus (MPa)

Modulus (MPa)

60 50 40 30

15

10

20 5 10 0

0 Fresh Control IR-30s

IR-45s

IR-60s

Treatment Conditions

IR-75s

Fresh Control Lye-30s Lye-45s Lye-60s Lye-75s

Treatment Conditions

Fig. 5. Effects of IR and lye heating on transition temperatures of tomato peels during the temperature ramp tests: (A) transition temperature of IR heated samples; (B) transition temperature of lye heated samples; (C) dynamic moduli at the transition point of IR heated samples; (D) dynamic moduli at the transition point of lye heated samples (n = 3).

32

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

Storage Modulus (MPa)

120

100

A

Fresh control IR-30s IR-45s IR-60s IR-75s

80

60

40

20 0.01

0.1

1

10

Frequency (Hz) 35

Loss Modulus (MPa)

30 25

B

Fresh control IR-30s IR-45s IR-60s IR-75s

20 15 10 5 0 0.01

0.1

1

and generate vapor pressure ultimately resulting in a layer separation between tomato peel and flesh. Typically, skins treated by IR can be peeled off in large pieces and maintain the skin integrity. All frequency spectra curves conformed well to the power law model. As shown in Table 1, when fitting storage and loss modulus data to the power law model, the coefficients of determination (r2) ranged from 0.89 to 0.99 for K0 and n0 values, and from 0.75 to 0.98 for K00 and n00 values. The K0 and K00 values represented the storage and loss modulus data of the entire frequency range for each treatment condition. IR heating led to increased K0 and K00 values. After 45 s IR heating, the K0 and K00 values of the IR heated samples were nearly twice those of the values of fresh peels (Table 1), a finding that suggests that sufficient IR heating time was needed in order to cause substantial changes in peel viscoelastic properties. In contrast, chemical diffusion of lye reduces the K0 and K00 values. These values were considerably lower for lye heated than for IR heated samples. Increase of frequency was found to have a broad effect on n0 and n00 values which were insignificant among different heating conditions. Despite biological variations, the viscoelastic properties of the composite tomato skin have been shown to depend not only on the structural relationships of various constitutive biopolymers, but also on the viscoelastic properties of each component (LopezCasado et al., 2010). The changes in the three-dimensional arrangement of biopolymers as affected by IR and lye have important effects on the overall viscoelastic behavior of tomato peels. Fundamental understanding of the skin viscoelastic behaviors affected by heating conditions may provide insights into the IR heating efficiency. Shift in storage modulus values of IR heated skins in

10 50

Frequency (Hz)

Storage Modulus (MPa)

temperature ramp tests, values of the storage modulus were greater in magnitude than the values of loss modulus, a finding that indicates the elastic property of tomato peel is significant. Fig. 7 shows the frequency response of dynamic modulus as affected by lye heating. All the modulus values increased with the increase of frequency. Unlike the IR heated samples, the storage and loss modulus values of lye heated samples were lower than those of the control. This pattern is consistent with the results of lye heated samples in temperature ramp measurements and is similar to previous findings in cherry tomato skins treated by enzymes (Matas et al., 2004c). In addition to the overall tendency of dynamic modulus behaviors, distinct responses of storage modulus under frequency sweep tests were found between IR and lye heated skins. As shown in Fig. 6A, between frequency of 0.8 to 1 Hz, an incremental shift in the storage moduli was observed in the IR heated samples and fresh control. This shift segregated the increase in storage moduli into two linear trends: one linear trend from 0.02 to 0.8 Hz, and another linear trend from 1 to 10 Hz. This shift was no longer observed in the lye heated skins (Fig. 7A). The distinct dynamic modulus behaviors of lye and IR heated skins revealed the mechanistic difference between lye and IR peeling. Due to the dissolving effect in lye peeling, tomato skins after lye treatment tend to lose the integrity and are cleaved into small pieces. Thus, chemical diffusion and a series of enzymatic bioreactions through damage of cuticular membranes and dissociation of pectin polysaccharides are considered to be the main underlying causes of pronounced changes in skin viscoelastic behavior treated by lye. In IR dry-peeling, a high amount of IR thermal energy impinges directly onto tomato surface in a short heating duration to reduce skin strength

40

A

Fresh control Lye-30s Lye-45s Lye-60s Lye-75s

30

20

10

0 0.01

0.1

1

10

Frequency (Hz) 14

12

Loss Modulus (MPa)

Fig. 6. Representative frequency spectra for the storage and loss moduli of IR heated peels and fresh control peels: (A) storage moduli; (B) loss moduli. (Lines represent fitted power law curves.)

B

Fresh control Lye-30s Lye-45s Lye-60s Lye-75s

10

8

6

4

2 0.01

0.1

1

10

Frequency (Hz) Fig. 7. Representative frequency spectra for the storage and loss moduli of hot lye heated peels and fresh control peels: (A) storage moduli; (B) loss moduli. (Lines represented fitted power law curves.)

33

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34 Table 1 Evaluated power law parameters of frequency sweep test for tomato peels heated by IR, lye, and fresh control (n = 3). K0 (MPa sn)

n0

r2

Control

30.1 ± 10.4b

0.077 ± 0.001

0.92

IR-30 IR-45 IR-60 IR-75

48.2 ± 20.9a,b 71.4 ± 12.4a 75.1 ± 21.7a 74.3 ± 14.9a

0.08 ± 0.002 0.078 ± 0.012 0.095 ± 0.019 0.08 ± 0.039

0.90 0.97 0.99 0.89

15.3 ± 5.0b 19.3 ± 5.5a,b 16.4 ± 3.0b 25.8 ± 4.4 a,b

0.086 ± 0.015 0.121 ± 0.022 0.093 ± 0.004 0.089 ± 0.003

0.90 0.97 0.99 0.94

4.79 ± 1.54b 7.05 ± 2.01a,b 4.59 ± 0.39b 6.01 ± 0.53a,b

Conditions

s s s s

Lye-30 Lye-45 Lye-60 Lye-75

s s s s

K00 (MPa sn)

n00

r2

8.60 ± 1.63b

0.104 ± 0.007

0.97

13.9 ± 7.9a,b 18.8 ± 3.21a 20.6 ± 0.88a 20.1 ± 1.96a

0.126 ± 0.011 0.092 ± 0.018 0.144 ± 0.01 0.107 ± 0.068

0.93 0.96 0.98 0.75

0.128 ± 0.036 0.122 ± 0.024 0.125 ± 0.016 0.124 ± 0.008

0.93 0.96 0.98 0.95

Notes: Values represent the mean ± standard deviation of triplicate tests. Mean separation was determined via Duncan’s Multiple Range Test (a < 0.05). Means with a different letter in each column of the same peeling method are significantly different at the 0.05 level.

0.10 0.09

Displacement (mm)

Table 2 Estimated parameters of creep test from Burger’s model for tomato peels heated by IR, lye, and fresh control (n = 3).

A

0.08

Conditions

0.07

Control

0.06

IR-30 IR-45 IR-60 IR-75

0.05 0.04 Fresh control IR-30s IR-45s IR-60s IR-75s

0.03 0.02 0.01

s s s s

Lye-30 Lye-45 Lye-60 Lye-75

s s s s

E2 (MPa)

s (s)

g (GPa s)

r2

9.0 ± 1.7

5.2 ± 0.9

10.5 ± 3.3

0.999

10.3 ± 3.3 10.4 ± 2.8 10.9 ± 3.4 9.8 ± 3.0

7.2 ± 2.8 6.5 ± 1.4 6.1 ± 1.8 7.3 ± 2.9

15.9 ± 15.6 11.2 ± 6.9 7.8 ± 2.4 11.7 ± 5.1

0.991 0.998 0.998 0.991

10.5 ± 2.2 8.7 ± 2.8 8.6 ± 1.6 10.9 ± 1.1

7.5 ± 3.3 6.6 ± 0.6 8.2 ± 3.3 4.8 ± 0.1

7.9 ± 3.1 11.6 ± 9.6 9.8 ± 2.8 31.7 ± 8.7

0.997 0.996 0.990 0.999

0.00 0

10

20

30

40

50

60

70

80

90

100 110 120

Time (s) 0.10

B

Displacement (mm)

0.08

0.06

0.04 Fresh control Lye-30s Lye-45s Lye-60s Lye-75s

0.02

0.00 0

10

20

30

40

50

60

70

80

90

100 110 120

Time (s) Fig. 8. Curves of creep behaviors of tomato peels under different peeling conditions: (A) IR heated; (B) lye heated.

frequency sweep tests and the transition temperatures observed in temperature ramp tests revealed altered skin viscoelastic properties, indicating that the these two test modes can be a sensitive tool to determine the minimal heating condition needed to achieve successful peel removal by means of IR heating. 3.3. Creep testing Creep behaviors of tomato peel under different treatment conditions are presented in Fig. 8. When a static force was applied to the tomato peel sample, the creep curves displayed an enormous displacement at the very beginning due to elasticity, and then continued to increase at a declining rate until the end of the test. Similar trends were found for IR and lye heated peels as well as for fresh peels (Fig. 8). These trends were in good agreement with

observed creep behavior of isolated tomato cuticle and skin in previous studies (Petracek and Bukovac, 1995; Thompson, 2001). The differences among different creep curves in Fig. 8 can be represented by Burger’s model. In Burger’s model, creep behavior of tomato skins can be characterized by using three parameters: the retarded elastic deformation (E2), the relaxation time (s), and the viscosity deformation (g). Values of the instantaneous elastic deformation (E1) in Eq. (3) were neglected because changes of E1 were large and could not fit Burger’s model accurately (Wang et al., 2009). The estimated parameters (i.e. E2, s, g) from fitting Burger’s model are compared in Table 2. The E2 values of all IR heated samples were slightly higher than those of the fresh control, a finding indicates a trend toward strain-hardening of tomato peel after IR heating. Strain-hardening of tomato peel was reported in the temperature ramp results and was thought to relate to the biopolymer structural changes in the peel. The relaxation time s indicates the time required for the applied stress to decrease to approximately 36.8% of its initial value under a constant deformation (Jiménez-Avalos et al., 2005). Although no significant difference in relaxation time s was found among different IR and lye treatments (a < 0.05), slightly higher s values existed for IR heated samples than for control. The acquired s values of fresh control were comparable to the values obtained in another study of tomato cuticles (Lopez-Casado et al., 2010). No significant difference between treatments was detected for g values, indicating no differences in the viscous nature of the tomato skin. The fact that the three parameters did not differ significantly among treatments suggests that the creep behavior of tomato peels may not as sensitive as temperature ramp and frequency sweep testing for differentiating treatment effects on the viscoelastic behaviors. 4. Conclusions The effects of IR and lye peeling treatments on the dynamic mechanical properties of tomato peel were investigated with three

34

Y. Wang et al. / Journal of Food Engineering 126 (2014) 27–34

DMA test modes: a temperature ramp, a frequency sweep, and creep measurement. For all samples, the temperature ramp test showed that storage and loss moduli decreased below the transition temperatures and then increased at temperatures above the transition points. Over an entire range of temperature from 20 °C to 100 °C, transition temperatures were identified as 63–72 °C for IR heated samples and 43–75 °C for lye heated samples, ranges that were significantly lower than the transition temperature of fresh control (88 °C). Values of both storage (E0 ) and loss (E00 ) moduli of IR heated tomato peels were considerably higher than those of fresh control, whereas values of storage and loss moduli from lye peeled samples were lower than those of fresh control. Similar trends also existed in frequency sweep tests. Strain-hardening of IR heated peels contrasted with strain-softening of lye heated peels. Dynamic moduli increased with the increase of frequency. K0 and K00 values of IR heated samples calculated from the power law model were higher than the K0 and K00 values of fresh control, whereas K0 and K00 values of lye heated samples were lower than those of control. No significant differences in creep behavior among treatments were detected by three estimated parameters from Burger’s model—the estimated retarded elastic modulus (E2), viscosity deformation (g), and relaxation time (s). However, dynamic moduli between the different treatments obtained from the temperature ramp and frequency sweep tests effectively differentiated the viscoelastic behaviors of tomato peel. The changes in dynamic moduli of tomato peel in response to lye and IR peeling may be attributed to structural changes in composite biopolymers and modification of cell walls anatomy. Transition temperature results can be used to determine the minimal temperature needed to achieve successful peel removal by means of IR heating. Optimization of the IR dry-peeling process by using DMA warrants further study.

References Allende, A., Desmet, M., Vanstreels, E., Verlinden, B.E., Nicolai, B.M., 2004. Micromechanical and geometrical properties of tomato skin related to differences in puncture injury susceptibility. Postharvest Biology and Technology 34 (2), 131–141. Anthon, G.E., Barrett, D.M., 2010. Changes in pectin methylesterification and accumulation of methanol during production of diced tomatoes. Journal of Food Engineering 97 (3), 367–372. Bargel, H., Neinhuis, C., 2005. Tomato (Lycopersicon esculentum Mill.) fruit growth and ripening as related to the biomechanical properties of fruit skin and isolated cuticle. Journal of Experimental Botany 56 (413), 1049–1060. Barreiro, J.A., Caraballo, V., Sandoval, A.J., 1995. Mathematical model for the chemical peeling of spherical foods. Journal of Food Engineering 25 (4), 483– 496. Barreiro, J.A., Sandoval, A.J., Rivas, D., Rinaldi, R., 2007. Application of a mathematical model for chemical peeling of peaches (Prunus persica l.) variety Amarillo Jarillo. LWT – Food Science and Technology 40 (4), 574–578. Barringer, S., 2003. Canned tomatoes: production and storage, in: Hui, Y.H., Ghazala, S., Graham, D. M., Murrel, K. D., & Nip, W. K. (Ed.), Handbook of vegetable preservation and processing. CRC press, pp. 109-120. Chuang, G.C.C., Yeh, A.I., 2006. Rheological characteristics and texture attributes of glutinous rice cakes (mochi). Journal of Food Engineering 74 (3), 314–323. Evert, R.F., 2006. Epidermis, in: Evert, R.F. (Ed.), Esau’s plant anatomy (third edition), pp. 211-253. Floros, J.D., Chinnan, M.S., 1990. Diffusion phenomena during chemical (NaOH) peeling of tomatoes. Journal of Food Science 55 (2), 552–553. Garcia, E., Barret, D.M., 2006. Evaluation of processing tomatoes from two consecutive growing seasons: quality attributes, peelability and yield. Journal of Food Processing and Preservation 30 (1), 20–36. Ginzburg, A.S., 1969. Application of Infra-Red Radiation in Food Processing. Leonard Hill Books, London. Hershko, V., Rabinowitch, H., Nussinovitch, A., 1994. Tensile characteristics of ripe tomato skin. LWT – Food Science and Technology 27 (4), 386–389. Hetzroni, A., Vana, A., Mizrach, A., 2011. Biomechanical characteristics of tomato fruit peels. Postharvest Biology and Technology 59 (1), 80–84. Houben, K., Jolie, R.P., Fraeye, I., Van Loey, A.M., Hendrickx, M.E., 2011. Comparative study of the cell wall composition of broccoli, carrot, and tomato: Structural

characterization of the extractable pectins and hemicelluloses. Carbohydrate Research 346 (9), 1105–1111. Huang, Y., Yu, H., Xiao, C., 2006. Effects of Ca2+ crosslinking on structure and properties of waterborne polyurethane-carboxymethylated guar gum films. Carbohydrate Polymers 66 (4), 500–513. Ikeda, S., Foegeding, E.A., 1999. Dynamic viscoelastic properties of thermally induced whey protein isolate gels with added lecithin. Food Hydrocolloids 13 (3), 245–254. Ikeda, S., Nishinari, K., 2001. On solid-like rheological behaviors of globular protein solutions. Food Hydrocolloids 15 (4-6), 401–406. Jiménez-Avalos, H., Ramos-Ramírez, E., Salazar-Montoya, J., 2005. Viscoelastic characterization of gum arabic and maize starch mixture using the Maxwell model. Carbohydrate Polymers 62 (1), 11–18. Kader, A.A., 2002. Postharvest Technology of Horticultural Crops, vol. 3311. Agriculture & Natural Resources. Li, X., 2012. A Study of Infrared Heating Technology for Tomato Peeling: Process Characterization and Modeling. Ph.D. Dissertation, Department of Biological and Agricultural Engineering, University of California at Davis, Davis. Li, X., Pan, Z. (2013a). Dry-peeling of Tomato by Infrared Radiative Heating: Part I. Model Development. Food and Bioprocess Technology, http://dx.doi.org/ 10.1007/s11947-013-1203-8. Li, X., Pan, Z. (2013b). Dry peeling of Tomato by Infrared Radiative Heating: Part II. Model Validation and and Sensitivity Analysis. Food and Bioprocess Technology, http://dx.doi.org/10.1007/s11947-013-1188-3. Li, X., Pan, Z., Bingol, G., McHugh, T.H., Atungulu, G.G., 2009. Feasibility study of using infrared radiation heating as a sustainable tomato peeling method. In: ASABE (Ed.), Proceedings of the American Society of Agricultural and Biological Engineers International. Paper No. 095689. ASABE, Reno, Nevada, St. Joseph, Michigan. Li, X., Pan, Z., Atungulu, G. G., Zheng, X., Wood, D., Delwiche, M., & McHugh, T. H. (2013). Peeling of Tomatoes using Novel Infrared Radiation Heating Technology. Innovative Food Science & Emerging Technologies. http://dx.doi.org/10.1016/ j.ifset.2013.10.011. López-Casado, G., Matas, A.J., Domínguez, E., Cuartero, J., Heredia, A., 2007. Biomechanics of isolated tomato (Solanum lycopersicum L.) fruit cuticles: the role of the cutin matrix and polysaccharides. Journal of Experimental Botany 58 (14), 3875–3883. Lopez-Casado, G., Salamanca, A., Heredia, A., 2010. Viscoelastic nature of isolated tomato (Solanum lycopersicum) fruit cuticles: a mathematical model. Physiologia Plantarum 140 (1), 79–88. Matas, A.J., Cobb, E.D., Bartsch, J.A., Paolillo Jr., D.J., Niklas, K.J., 2004a. Biomechanics and anatomy of Lycopersicon esculentum fruit peels and enzyme-treated samples. American Journal of Botany 91 (3), 352–360. Matas, A.J., Cobb, E.D., Paolillo Jr., D.J., Niklas, K.J., 2004b. Crack resistance in cherry tomato fruit correlates with cuticular membrane thickness. HortScience 39 (6), 1354–1358. Matas, A.J., Cuartero, J., Heredia, A., 2004c. Phase transitions in the biopolyester cutin isolated from tomato fruit cuticles. Thermochimica Acta 409 (2), 165–168. Matas, A.J., López-Casado, G., Cuartero, J., Heredia, A., 2005. Relative humidity and temperature modify the mechanical properties of isolated tomato fruit cuticles. American Journal of Botany 92 (3), 462–468. Menard, K.P., 1998. Dynamic Mechanical Analysis: A Practical Introduction. CRC Press. Özkan, N., Xin, H., Chen, X.D., 2002. Application of a depth sensing indentation hardness test to evaluate the mechanical properties of food materials. Journal of Food Science 67 (5), 1814–1820. Pan, Z., Li, X., Bingol, G., McHugh, T.H., Atungulu, G., 2009. Development of infrared radiation heating method for sustainable tomato peeling. Applied Engineering in Agriculture 25 (6), 935–941. Pan, Z., Li, X., Yong, W., Atungulu, G., McHugh, T.H., Delwiche, M., 2011. Development of infrared heating technology for tomato peeling. In: International Congress on Engineering and Food. pp. 795–796. Petracek, P.D., Bukovac, M.J., 1995. Rheological properties of enzymatically isolated tomato fruit cuticle. Plant Physiology 109 (2), 675–679. Pothan, L.A., Oommen, Z., Thomas, S., 2003. Dynamic mechanical analysis of banana fiber reinforced polyester composites. Composites Science and Technology 63 (2), 283–293. Rock, C., Yang, W., Goodrich-Schneider, R., Feng, H., 2011. Conventional and alternative methods for tomato peeling. Food Engineering Reviews, 1–15. Shao, D., Atungulu, G.G., Pan, Z., Yue, T., Zhang, A., Fan, Z., 2013. Characteristics of isolation and functionality of protein from tomato pomace produced with different industrial processing methods. Food and Bioprocess Technology, 1–10. Shi, J.X., Le Maguer, M., Wang, S.L., Liptay, A., 1997. Application of osmotic treatment in tomato processing – effect of skin treatments on mass transfer in osmotic dehydration of tomatoes. Food Research International 30 (9), 669–674. Taiz, L., Zeiger, E., 2006. Cell walls: structure, biogenesis, and expansion. In: Plant Physiology, fourth ed. Sinauer Associates, pp. 349–375. Thompson, D.S., 2001. Extensiometric determination of the rheological properties of the epidermis of growing tomato fruit. Journal of Experimental Botany 52 (359), 1291–1301. Wang, Y., Wang, L.J., Li, D., Xue, J., Mao, Z.H., 2009. Effects of drying methods on rheological properties of flaxseed gum. Carbohydrate Polymers 78 (2), 213–219.