- Email: [email protected]
Implication of water state on glass transition temperature in hot air-dried carrot slices
LWT - Food Science and Technology 84 (2017) 780e787
Contents lists available at ScienceDirect
LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt
Implication of water state on glass transition temperature in hot airdried carrot slices Congcong Xu a, *, Dekun Liu a, Yunfei Li b, Guanxi Li a, Ju Zhang a, Ruixia Gao a a b
School of Life Science, Qufu Normal University, Qufu, Shandong Province 273165, PR China Department of Food Science and Technology, School of Agriculture and Biology, Shanghai Jiao Tong University, Shanghai 200240, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 October 2016 Received in revised form 23 June 2017 Accepted 23 June 2017 Available online 26 June 2017
This study assessed and correlated the water state and the glass transition temperature (Tg) of carrots treated by hot air drying at 50, 60, and 70 C. Results showed that within carrots with high moisture content (Xw, 4.15 Xw 1.01 g water/g dry matter), free water in vacuoles dominated in tissues and the Tg remained practically constant (48.4 ± 1.04 C). And when the Xw of carrots was 0.41 g water/g dry matter, the Tg was increased by 6.79 and 22.7 C with the relaxation time (T22) of the immobilized water in the cytoplasm/extracellular space decreasing by 12.4 and 11.8 m, respectively. Moreover, correlation equations indicated that the Tg had a negatively linear relationship with the T22 in dried carrots. These results suggested that the Tg were elevated appreciably by reducing the relaxation behavior of immobilized water in hot air-dried carrots. The more pronounced variation of the microstructure (e.g., vacuole disruption, cell wall dissociation, and turgidity loss) of tissues could be responsible for these results. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Carrot Hot air drying Glass transition temperature Water state Correlation
1. Introduction As well known, glass transition temperature (Tg) is a critical reference parameter to predict the quality, stability, and safety of food systems. It is defined as a critical temperature, where an amorphous system shifts from the glassy state to the rubbery one, or the opposite process. At the glassy state (T < Tg), the molecular mobility of materials is extremely slow (about 1012 m/s) (Roos, 2010). If the materials become glassy, this would allow them rigid and physicochemical changes are greatly restricted. In contrast, at the rubbery state (T > Tg), the molecular mobility exponentially increases and the viscosity dramatically decreases (Roos, 2010). These features collectively cause the various time-dependent and viscosity-related structural transformations, like collapse, “stickiness”, shrinkage, etc, during food processing and storage (Ruan et al., 1999; Kurozawa, Hubinger, & Park, 2012). Therefore, foods should be stored at the temperatures below their Tg to avoid loss in quality during the storage. Water acts as a strong plasticizer. High water content and water activity can reduce the Tg of food materials, such as lactose-whey protein system (Maidannyk & Roos, 2017), b-cyclodextrin/water
* Corresponding author. E-mail address: [email protected] (C. Xu). http://dx.doi.org/10.1016/j.lwt.2017.06.052 0023-6438/© 2017 Elsevier Ltd. All rights reserved.
binary system (Zhou et al., 2015), wheat gluten and maltodextrin (Shimazaki, Tashiro, Kumagai, & Kumagai, 2017), etc. Fresh fruits and vegetables have higher water content. This makes their Tg often lower than the common storage temperature (20 C). Thus, fresh tissues are subjected to a rubbery state, easily causing a severe loss of qualities. Hence, vegetables could be dehydrated to shift the Tg to the values close to room temperature (because of the reduction of moisture content) and thus extend their shelf-life during storage. Hot air drying is a conventional and widely applicable drying technology for foods (Moraga, Talens, Moraga, & MartínezNavarrete, 2011). To our knowledge, the detailed information of the changes of Tg in fruits and vegetables during hot air dehydration is still unclear. Additionally, nuclear magnetic resonance (NMR) is a nondestructive technique which is helpful to have more insight about microscopic information regarding with water mobility in biological tissues. Based on the analysis of the spin-spin relaxation times (T2), the water state within raw fruits and vegetables is usually categorized into “free water’’ in vacuole, ‘‘immobilized water’’ in cytoplasm and extracellular space, and ‘‘bound water’’ bounded to cell walls according to the mobility (Shao & Li, 2011; Xin, Zhang, & Adhikari, 2013). They will suffer dramatic changes during the dehydration process. For instance, vacuolar water (the most mobile water) decreases together with the increases of water in cytoplasm and intercellular space in strawberry due to the osmotic treatment
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
(Cheng, Zhang, Adhikari, & Islam, 2014). Tylewicz et al. (2016) have indicated that pulsed electric field pretreatment can cause water redistribution between different compartments in freeze-dried apples. Some water could shift from vacuole into the extracellular/cytoplasm spaces, producing an appreciable increase in its relaxation time. These changes of the water state in tissues can have a dramatic impact on the Tg. Our previous study has found that the Tg approaches a constant value when free water (the most mobile water) dominates in raw carrot tissues, probably owing to the very low viscosity of dilute solution. And the Tg increases with the decreased immobilized water in far-infrared dried carrots with low water content (Xu, Li, & Yu, 2014). Xin et al. (2013) have also showed that changing the states of water using osmotic dehydration can directly affect the Tg of broccoli. Concretely, the values of the Tg in the osmotically dehydrated broccoli are raised compared with the untreated samples. It can be seen that the water state in tissues is also a crucial contributor to influencing the Tg. However, to our knowledge, more works are still concentrated on the simplified water content/temperature state diagrams or the simplified water activity/temperature state diagrams to predict the glass transition phenomena. The reports about the effect of water state on the Tg in dehydrated products of fruits and vegetables is limited to the two ones above. Hence, a fundamental understanding of the relationship between water state and Tg is still greatly necessary for predicting the storage and processing stability of dehydrated products. Carrot, as a root vegetable, contains appreciable amounts of nutritional components (e.g. carotenes, vitamins, fiber content) and is widely consumed worldwide at a relatively low cost. Given these advantages, we took carrot as a model system to assess the changes of the water state (using nuclear magnetic resonance) and the Tg (from differential scanning calorimetry analysis) in hot airdehydrated carrots and establish their relationship using correlation analysis and path analysis. These findings could provide the theoretical basis for enhancing the storage and processing stability of dehydrated products of root vegetables. 2. Materials and methods 2.1. Sample preparation Carrots (Daucus carota L.) were harvested in November in Shanghai (China). Fresh carrots without any physical damage were bought from a local market. Carrots were cut into slices (6 ± 1 mm thickness and 32 ± 2 mm diameter) and randomly placed in a thin layer in an electrical thermostatic drying oven (Shanghai Yaoshi Instrument Factory, Shanghai, China) with 10e15 kg mass loading. Then, samples were dried at 50, 60, and 70 C for 14, 10, and 10 h, respectively, until reaching a final moisture content lower than 0.20 g water/g dry matter. A recirculating air maintains the constant temperature (±1 C) and the uniformity of the temperature at every corner (±2.5%) in the oven. Some samples were taken out every 2 h for analysis.
781
2.3. Nuclear magnetic resonance (NMR) Based on our previous study (Xu, Li, & Yu, 2014), a NMI 20Analyst system (22.6 MHz; Shanghai Niumag Corporation, Shanghai, China) was used to monitor the water state in carrot samples. Briefly, approximately 2 g of samples was applied to this study. Three duplicate NMR measurements (total of six runs) were carried out and a mean value taken. 2.4. Differential scanning calorimetry (DSC) The thermal transitions of carrot samples were assayed in a differential scanning calorimeter (DSC 204 F1 Phoenix®, Netzsch, Germany) according to our previous report (Xu et al., 2014) with some modifications. The heat flow and temperature of instrument were calibrated using distilled water (melting point (mp) ¼ 0 C, melting enthalpy (DHm) ¼ 333.88 J/g) and indium (mp ¼ 156.60 C, DHm ¼ 28.45 J/g). An empty sealed aluminum pan was employed as a reference. Nitrogen gas (50 mL/min) was used as the purge gas to avoid the condensation of water around samples during running. Samples (about 10 mg) were cooled from the load temperature (room temperature) to 60 C at 10 C/min and equilibrated at 60 C for 6 min. Finally, samples were scanned from 60 to 40 C at 10 C/min to determine the glass transition temperature (Tg). The onset (Tgo), mid-point (Tgm), and end (Tge) of the glass transition were assessed by using the TA universal analysis software provided with the DSC instrument. The midpoint of the transition was taken to be the Tg (Fig. 1). Data were reported as the mean value of three replicates for each treatment. 2.5. Microstructure According to the method reported by our previous study (Xu et al., 2015) with slight modification, specimens of carrot samples were prepared. Briefly, ultrathin sections (70 nm thickness) were obtained using an ultra-microtome (Leica Ultracut UC6, Germany). Specimens were double-stained with uranyl acetate and lead citrate, and examined in a Tecnai G2 Spirit Biotwin transmission electronic microscopy (TEM, FEI, Hillsboro, OR) at an accelerating voltage of 18 kV. 2.6. Statistical analyses Pearson's correlation analysis was performed by combining the data from all the carrot samples to address the relationship
2.2. Moisture content Carrot samples (initial mass, W1, g) were dried in an oven at 105 C to attain a constant mass (W2, g). Moisture content (Xw, g water/g dry matter) was calculated using Eq. (1). Data were reported as the mean value of three replicates for each treatment.
Xw ¼
W1 W2 W2
(1) Fig. 1. Typical DSC thermogram showing the glass transition temperature (Tg) of hot air-dried carrots at 70 C for 10 h.
782
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
between the selected parameters. Regress analysis and path analysis were carried out to assess the linear relationship and the respective effects of different parameters. Data were analyzed using SAS ver9.1 (SAS Institute, Cary, NC, USA). 3. Results and discussion 3.1. Xw The Xw of carrot samples decreased gradually with increasing drying time (Fig. 2). After drying at 50, 60, and 70 C for 14, 10, and 10 h, respectively, the values of Xw were decreased from an initial Xw of 7.82 g water/g dry matter to 0.12, 0.11, and 0.05 g water/g dry matter. 3.2. Water state and Tg 3.2.1. Water at different states The NMR technology, as a non-destructive test, can be applied to monitor the water state in biological materials (Vicente, Nieto, Hodara, Castro, & Alzamora, 2012). In our study, the spin-spin relaxation time (T2) and the relative signal intensity (M) were detected to investigate the distribution of water in carrots during hot air drying. In the T2 inversion spectra, the population groups of the shortest (T21), intermediate (T22), and longest (T23) relaxation times usually correspond to the bound water combined to cell walls, immobilized water in cytoplasm/extracellular space, and free water in vacuoles, respectively, in raw biological tissues (Cheng et al., 2014). Besides, the corresponding relative signal intensities of M21, M22, and M23 are employed to express the relative content of bound water, immobilized water, and free water, respectively, in these tissues. In our study, two main population groups (T22 and T23) were detected in the initial distribution. The respective ranges of T22 and T23 were 3e50 m and 60e1000 m. The T21 was negligible. Therefore, only the T22 and T23 were analyzed further. 3.2.2. Free water in vacuoles and Tg All the samples between raw carrots and carrots dried for 12, 6, and 4 h (at 50, 60, and 70 C, respectively) contained high Xw (7.82 Xw 1.01 g water/g dry matter; Fig. 2). The analysis of NMR results showed that at 7.82 Xw 1.01 g water/g dry matter, free water in vacuoles (the most mobile water) dominated in tissues (Figs. 3 and 4). The T23 decreased gradually from 682 m in raw samples to 123, 171, and 134 m at 50, 60, and 70 C, respectively
Fig. 2. Changes of moisture content (Xw) based on dry matter of carrots during hot air drying. : 50 C; : 60 C; : 70 C.
(Fig. 3A). Likewise, a continuous reduction of the M23 from 2150 in raw samples to 188, 268, and 366 at the respective temperatures above was observed (Fig. 3B). These results suggested that free water in vacuoles were gradually removed during hot air drying. This trend was in accordance with the finding in microwavedehydrated stem lettuces, where the T2 peak value with respect to free water was reduced by 717 m compared with the untreated samples (Wang, Zhang, Mujumdar, & Mothibe, 2012). Those are because that the loss of free water could cause the concentrated solute in vacuoles, the plasmolyzed cytoplasm, and the considerable accumulation of polysaccharides in the space between cell walls and plasmalemma, thereby shortening the T23 (Cheng et al., 2014; Hatakeyama, Tanaka, & Hatakeyama, 2010). Simultaneously, the changes of Tg were analyzed in the range of 7.82 Xw 1.01 g water/g dry matter, where free water in vacuoles dominated in tissues (Fig. 3A). For all the samples at 7.82 Xw 4.32 g water/g dry matter (drying time, 4, 2, and 2 h at 50, 60, and 70 C, respectively), the Tg appeared to be practically negligible in comparison with the latent heat of ice melting. Other samples at 4.15 Xw 1.01 g water/g dry matter, the values of the Tg remained practically constant. The mean values of the Tgo, Tg, and Tge for all these samples were estimated to be 51.7 ± 1.40 C, 48.4 ± 1.04 C, and 44.8 ± 1.39 C, respectively. These findings were similar to the result that the Tg of freeze-dried separated plum skin and pulp approaches a constant around 57.5 C in the range of higher water content (Telis, do Amaral Sobral, & Telis-Romero, 2006). This could be because that under the conditions that free water dominates in tissues, the matrix of cells is a dilute solution system with low viscosity. The effect of the viscosity of cell solution is slight compared with the water content. The Tg was independent of the initial concentration. Hence, the Tg of samples approached a constant value. 3.2.3. Immobilized water in the cytoplasm/extracellular space and Tg The changes of the T22 and M22 of carrot samples were more complex (Fig. 4). The T22 and M22 of samples increased within 6 h at 50 C and 2 h at 60 and 70 C, respectively. These increases could have been owing to the shift of free water to immobilized water because of the increase of the concentration of carbohydrates and the degradation of nutritional ingredients in the cytoplasm. Tylewicz et al. (2016) have indicated that some water from vacuole could move into the extracellular/cytoplasm spaces, producing an appreciable increase in the T22. Subsequently, the T22 and M22 of these samples decreased continuously, possibly due to the shift from immobilized water to the water more tightly bound to the polysaccharides in cell walls (e.g., pectin). Three types of hydrophilic groups of pectin can strongly bind to water (Einhorn-Stoll, Hatakeyama, & Hatakeyama, 2012). Carrots possess lots of pectin, thereby easily reducing the water mobility and elevating the T22. Li, Ma, Tao, Kong, and Li (2012) have showed that immobilized water could be transformed into bound water in beef granules during drying at 40, 50, and 60 C. After a short-term decrease, a sharp increase in the T22 and M22 was observed at the respective treatments of 14, 8, and 6 h at 50, 60, and 70 C, during which the amount of free water decreased sharply. Hence, the sharp increase could have been associated with a sharp shift of free water into immobilized water. Finally, the T22 and M22 showed a gradual downside. Under the conditions that most of the free water was lost, the immobilized water could have become more mobile and cause a considerable amount of loss with increasing drying time. Given the analysis of the results from NMR, all of the free water of samples was removed at the conditions of drying for 14, 8, and 6 h at 50, 60, and 70 C, respectively (Fig. 3). Simultaneously, the immobilized water sharply increased (Fig. 4). These results suggested that the dominate water presenting in all the samples (dried
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
783
Fig. 3. Changes of the relaxation time (T23) and relative signal intensity (M23) of free water in vacuoles and the glass transition temperature (Tg) of carrots during hot-air drying at 50, 60, and 70 C, respectively. (A) , , and : the T23 of samples dried at 50, 60, and 70 C, respectively; (B) , , and : the M23 of samples dried at 50, 60, and 70 C, respectively. , , and : the onset (Tgo), mid (Tg), and end (Tge) temperatures of glass transition of samples dried at 50 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 60 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 70 C, respectively.
for 14, 8, and 6 h at 50, 60, and 70 C, respectively) was the immobilized water. All these samples were with low Xw (0.41, g water/g dry matter; Fig. 2). At the critical times when the immobilized water began to dominate, the Tg of all samples showed an obvious increase (Fig. 4). Subsequently, the Tg increased gradually with decreasing content of immobilized water during dehydration. The Tg of samples dried at 60 and 70 C increased by 6.79 and
22.7 C with the T22 decreased by 12.4 and 11.8 m, respectively. Besides, the Xw decreased by 0.30 and 0.05 g water/g dry matter, respectively. A similar trend has been reported in several studies on various dried fruits and vegetables, such as tomatoes (Goula, Karapantsios, Achilias, & Adamopoulos, 2008), Agaricus bisporus (Shi, Wang, Zhao, & Fang, 2012), and jambolan (de Santana, de Oliveira Neto, Santos, Soares, Lima, & Cardoso, 2015).
784
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
Fig. 4. Changes of the relaxation time (T22) and relative signal intensity (M22) of immobilized water in cytoplasm/extracellular space, and the glass transition temperature (Tg) of carrots during hot-air drying at 50, 60, and 70 C, respectively. (A) , , and : the T22 of samples dried at 50, 60, and 70 C, respectively; (B) , , and : the M22 of samples dried at 50, 60, and 70 C, respectively. , , and : the onset (Tgo), mid (Tg), and end (Tge) temperatures of glass transition of samples dried at 50 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 60 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 70 C, respectively.
3.2.4. Correlation between water state and Tg The curves in Fig. 4 intuitively showed that the Tg increased with decreasing T22, M22, and Xw of immobilized water in samples at Xw 0.41 g water/g dry matter. Furthermore, Table 1 indicated that the Tgo, Tg, and Tge of samples were correlated negatively with the T22 (P < 0.01) and M22 (P < 0.05). Likewise, the report published by Xin et al. (2013) has shown that the Tg is correlated negatively with
the T21 of immobilized water and the relative signal intensity of free water in osmotically treated broccoli. Moreover, the regression equations (Tgo ¼ 5.64e1.23 T22, R2 ¼ 0.81, P < 0.01; Tg ¼ 11.7e1.31 T22, R2 ¼ 0.85; P < 0.01; Tge ¼ 18.7e1.40 T22, R2 ¼ 0.87, P < 0.01) and the equations from path analysis (Tgo ¼ 0.90 T22, R2 ¼ 0.81, P < 0.01; Tg ¼ 0.92 T22, R2 ¼ 0.85, P < 0.01; Tge ¼ 0.94 T22, R2 ¼ 0.87, P < 0.01) were obtained in the
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787 Table 1 Correlation coefficients between water state and glass transition temperature (Tg) of hot air-dried carrots. Coefficient
Tgo
Tg
Tge
T22 M22
0.935** 0.698*
0.922** 0.698*
0.899** 0.689*
T22 and M22: the relaxation time and relative signal intensity of immobilized water in cytoplasm/extracellular space, respectively; Tgo and Tge: the onset and end temperature of glass transition, respectively. *Correlation is significant at the level of 0.05; ** Correlation is significant at the level of 0.01.
785
present study. These equations showed that the Tg had a linear relationship with the T22 and that the changes of the Tg was affected appreciably by the immobilized water. Compared with the effect of immobilized water, the effect of free water of carrots was not as appreciable. Hence, in comparison with free water in vacuoles, the immobilized water in cytoplasm/extracellular space exerted a negative and dominating part in the effect on the Tg in the hot airdried carrots. Additionally, compared with the correlation between M22 and Tg, the one between T22 and Tg was larger. This finding could be
Fig. 5. Changes of cell microstructure in raw and dried tissues. A: raw tissue; BeE: dried tissues at 70 C for 2, 4, 6, and 8 h, respectively. PM: plasmalemma; CW: cell wall; ML: middle lamella; TP: tonoplast; MT: mitochondria.
786
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
because that the relaxation behavior has a direct relationship with water mobility in different parts of the cells as well as the interaction between water and other cell components (Guizani, Al-Saidi, Rahman, Bornaz, & Al-Alawi, 2010; Gussoni et al., 2007). The water redistribution between cellular structures could cause the changes of T2 in carrot slices treated by pulsed electric field treatment -Aguayo et al., 2014). Nevertheless, the M22 reflects the (Aguilo relative content of water at different states. Hence, the relaxation behavior well correlated with the Tg in hot-air dried carrots.
viscosity of the matrix over a wide range content, collectively lowering the mobility of the matrix. Consequently, the Tg of samples was elevated. Another reason for the increased Tg could be that because of the severe degradation of middle pectin lamella, the bound water (linked to the hydrophilic groups of pectin) might shift to more mobile water (immobilized water) and be removed further.
3.3. Cell microstructure
The water state and Tg dramatically changed during hot air dehydration in carrots. The moisture content of carrots at 4.15 Xw 1.01 g water/g dry matter, free water in vacuoles dominated in tissues and the Tg remained practically constant. When the Xw decreased to lower than 0.41 g water/g dry matter, the immobilized water in cytoplasm/extracellular space exerted a dominating and negative impact on the Tg of samples. As expected, the Tg value increased with the decreasing of immobilized water content during hot air dehydration. The regression equations showed the Tg increased appreciably according to the decreased relaxation behavior of the immobilized water in hot air-dried carrots. The more pronounced alteration in microstructure could explain these changes in hot air-dried carrots.
The changes of cell microstructure of raw and treated carrot samples (at 70 C within 8 h) were shown in Fig. 5. In raw samples (Fig. 5A), the mitochondria, tonoplast, and plasmalemma were intact. A large vacuole occupied most of the protoplast. The plasmalemma was tightly against the cell walls, which consisted of tightly packed and longitudinally-organized fibrillar materials. The middle lamella was limited between them. In samples at 2 h (Fig. 5B), organelles, plasmalemma, and tonoplasts were severely disrupted. The vacuoles shrunk clearly. Cell walls became distorted and swollen. Plasmolysis occurred with an obvious retraction of the plasmalemma and tonoplasts to the center of the cells, resulting in a separation of up to 1.1 mm between the plasmalemma and the cell wall in some areas. This result was similar to that observed in rezsliced carrots blanched at 90 C for 10 min (Neri, Hernando, Pe Munuera, Sacchetti, & Pittia, 2011). In samples at 4 h (Fig. 5C), no intact membrane structures (e.g. plasmalemma and organelles) were observed. Vacuoles shrunk and fractured further, only leaving some smaller ones in cells. The reticulate fibrillar pattern of cell walls became striated and loose. The central zone appeared denser, representing the presence of middle lamella and the separation of adjacent cell walls. This generation of swollen and loose cell walls might be due to the loosening of the hemicellulose-cellulose network (Phothiset & Charoenrein, 2014). The deterioration above was aggravated in samples at 6 and 8 h (Fig. 5D and E). In these cells, nearly all of vacuoles were lost. The zone of cell walls became lighter because of the severe degradation of fiber structure components. Besides, the middle lamella trended towards thinner, meaning more solubilisation of pectin components. Collectively, the dynamic changes of the water state as a consequence of the microstructure variation may be the main underlying cause of the changes of Tg of hot air-dried carrots. Hot air drying could cause the vaporization of escaped cytosolic fluid and increase the internal vapor pressure under the membrane structures, which will disrupt when the membrane stress exceeds the critical rupture stress (Li, Pan, Atungulu, Wood, & McHugh, 2014). With increasing time and temperature, these changes become more noticeable. Moreover, high temperatures (60 C) could directly degrade the membrane structures of tissues. These features could result in a substantial loss of free water in vacuoles, accompanied with cytoplasm plasmolysis and concentrated vacuoles. In our study, with the shrinkage of the large vacuole observed in raw samples (Fig. 5A) to small ones in samples (at 70 C within 4 h; Fig. 5B and C), the latent heat of ice (from free water) melting decreased. And the Tg became detectable. However, the matrix of samples is still a dilute solution system. The Tg was independent of the initial concentration. Hence, the Tg of the samples approached a constant value. With further dehydration, all of the vacuoles disappeared, meaning all of the free water was removed (Fig. 5D and E). Some could shift to the less mobile water (immobilized water), where a sharp increase of Tg was detected. Then, the immobilized water was also gradually removed. At these conditions, the immobilized water possesses a well plasticizing activity. Its decrease could reduce the free volume and increase the local
4. Conclusion
Acknowledgements Our authors gratefully acknowledge the financial support of China National Natural Science Foundation (Contract No.31271909) and the research project of experimental technology of Qufu Normal University (Contract No. SJ201503). References -Aguayo, I., Downey, G., Keenan, D. F., Lyng, J. G., Brunton, N., & Rai, D. K. Aguilo (2014). Observations on the water distribution and extractable sugar content in carrot slices after pulsed electric field treatment. Food Research International, 64, 18e24. Cheng, X. F., Zhang, M., Adhikari, B., & Islam, M. N. (2014). Effect of power ultrasound and pulsed vacuum treatments on the dehydration kinetics, distribution, and status of water in osmotically dehydrated strawberry: A combined NMR and DSC study. Food and Bioprocess Technology, 7(10), 2782e2792. Einhorn-Stoll, U., Hatakeyama, H., & Hatakeyama, T. (2012). Influence of pectin modification on water binding properties. Food Hydrocolloids, 27(2), 494e502. Goula, A. M., Karapantsios, T. D., Achilias, D. S., & Adamopoulos, K. G. (2008). Water sorption isotherms and glass transition temperature of spray dried tomato pulp. Journal of Food Engineering, 85(1), 73e83. Guizani, N., Al-Saidi, G. S., Rahman, M. S., Bornaz, S., & Al-Alawi, A. A. (2010). State diagram of dates: Glass transition, freezing curve and maximal-freezeconcentration condition. Journal of Food Engineering, 99(1), 92e97. Gussoni, M., Greco, F., Vezzoli, A., Paleari, M. A., Moretti, V. M., Lanza, B., et al. (2007). Osmotic and aging effects in caviar oocytes throughout water and lipid changes assessed by 1 H NMR T 1 and T 2 relaxation and MRI. Magnetic Resonance Imaging, 25(1), 117e128. Hatakeyama, T., Tanaka, M., & Hatakeyama, H. (2010). Thermal properties of freezing bound water restrained by polysaccharides. Journal of Biomaterials Science, Polymer Edition, 21(14), 1865e1875. Kurozawa, L. E., Hubinger, M. D., & Park, K. J. (2012). Glass transition phenomenon on shrinkage of papaya during convective drying. Journal of Food Engineering, 108(1), 43e50. Li, X., Ma, L. Z., Tao, Y., Kong, B. H., & Li, P. J. (2012). Low field-NMR in measuring water mobility and distribution in beef granules during drying process. Advanced Materials Research, 550, 3406e3410. Li, X., Pan, Z., Atungulu, G. G., Wood, D., & McHugh, T. (2014). Peeling mechanism of tomato under infrared heating: Peel loosening and cracking. Journal of Food Engineering, 128, 79e87. Maidannyk, V. A., & Roos, Y. H. (2017). Water sorption, glass transition and “strength” of lactose-Whey protein systems. Food Hydrocolloids, 70, 76e87. Moraga, G., Talens, P., Moraga, M. J., & Martínez-Navarrete, N. (2011). Implication of water activity and glass transition on the mechanical and optical properties of freeze-dried apple and banana slices. Journal of Food Engineering, 106(3), 212e219. rez-Munuera, I., Sacchetti, G., & Pittia, P. (2011). Effect of Neri, L., Hernando, I. H., Pe blanching in water and sugar solutions on texture and microstructure of sliced carrots. Journal of Food Science, 76(1), E23eE30. Phothiset, S., & Charoenrein, S. (2014). Effects of freezing and thawing on texture,
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787 microstructure and cell wall composition changes in papaya tissues. Journal of the Science of Food and Agriculture, 94(2), 189e196. Roos, Y. H. (2010). Glass transition temperature and its relevance in food processing. Annual Review of Food Science and Technology, 1, 469e496. Ruan, R., Long, Z., Chen, P., Huang, V., Almaer, S., & Taub, I. (1999). Pulse NMR study of glass transition in maltodextrin. Journal of Food Science, 64(1), 6e9. S., et al. de Santana, R. F., de Oliveira Neto, E. R., Santos, A. V., Soares, C. M. F., Lima, A. (2015). Water sorption isotherm and glass transition temperature of freezedried Syzygium cumini fruit (jambolan). Journal of Thermal Analysis and Calorimetry, 120(1), 519e524. Shao, X. L., & Li, Y. F. (2011). Application of low-field NMR to analyze water characteristics and predict unfrozen water in blanched sweet corn. Food and Bioprocess Technology, 1e7. Shimazaki, E., Tashiro, A., Kumagai, H., & Kumagai, H. (2017). Analysis of water sorption isotherms of amorphous food materials by solution thermodynamics with relevance to glass transition: Evaluation of plasticizing effect of water by the thermodynamic parameters. Bioscience, Biotechnology, and Biochemistry, 81(4), 743e749. Shi, Q. L., Wang, X. H., Zhao, Y., & Fang, Z. (2012). Glass transition and state diagram for freeze-dried Agaricus bisporus. Journal of Food Engineering, 111(4), 667e674. Telis, V. N., do Amaral Sobral, P. J., & Telis-Romero, J. (2006). Sorption isotherm, glass transitions and state diagram for freeze-dried plum skin and pulp. Food Science and Technology International, 12(3), 181e187.
787
Tylewicz, U., Aganovic, K., Vannini, M., Toepfl, S., Bortolotti, V., Dalla Rosa, M., et al. (2016). Effect of pulsed electric field treatment on water distribution of freezedried apple tissue evaluated with DSC and TD-NMR techniques. Innovative Food Science & Emerging Technologies, 37, 352e358. Vicente, S., Nieto, A. B., Hodara, K., Castro, M. A., & Alzamora, S. M. (2012). Changes in structure, rheology, and water mobility of apple tissue induced by osmotic dehydration with glucose or trehalose. Food and Bioprocess Technology, 5(8), 3075e3089. Wang, Y., Zhang, M., Mujumdar, A. S., & Mothibe, K. J. (2012). Experimental investigation and mechanism analysis on microwave freeze drying of stem lettuce cubes in a circular conduit. Drying Technology, 30(11e12), 1377e1386. Xin, Y., Zhang, M., & Adhikari, B. (2013). Effect of trehalose and ultrasound-assisted osmotic dehydration on the state of water and glass transition temperature of broccoli (Brassica oleracea L. var. botrytis L.). Journal of Food Engineering, 119(3), 640e647. Xu, C., Li, Y., & Yu, H. (2014). Effect of far-infrared drying on the water state and glass transition temperature in carrots. Journal of Food Engineering, 136, 42e47. Xu, C., Li, Y., & Yu, C. (2014). Effect of blanching pretreatment on carrot texture attribute, rheological behavior, and cell structure during cooking process. LWTFood Science and Technology, 62(1), 48e54. Zhou, G., Zhao, T., Wan, J., Liu, C., Liu, W., & Wang, R. (2015). Predict the glass transition temperature and plasticization of b-cyclodextrin/water binary system by molecular dynamics simulation. Carbohydrate Research, 401, 89e95.
Contents lists available at ScienceDirect
LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt
Implication of water state on glass transition temperature in hot airdried carrot slices Congcong Xu a, *, Dekun Liu a, Yunfei Li b, Guanxi Li a, Ju Zhang a, Ruixia Gao a a b
School of Life Science, Qufu Normal University, Qufu, Shandong Province 273165, PR China Department of Food Science and Technology, School of Agriculture and Biology, Shanghai Jiao Tong University, Shanghai 200240, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 October 2016 Received in revised form 23 June 2017 Accepted 23 June 2017 Available online 26 June 2017
This study assessed and correlated the water state and the glass transition temperature (Tg) of carrots treated by hot air drying at 50, 60, and 70 C. Results showed that within carrots with high moisture content (Xw, 4.15 Xw 1.01 g water/g dry matter), free water in vacuoles dominated in tissues and the Tg remained practically constant (48.4 ± 1.04 C). And when the Xw of carrots was 0.41 g water/g dry matter, the Tg was increased by 6.79 and 22.7 C with the relaxation time (T22) of the immobilized water in the cytoplasm/extracellular space decreasing by 12.4 and 11.8 m, respectively. Moreover, correlation equations indicated that the Tg had a negatively linear relationship with the T22 in dried carrots. These results suggested that the Tg were elevated appreciably by reducing the relaxation behavior of immobilized water in hot air-dried carrots. The more pronounced variation of the microstructure (e.g., vacuole disruption, cell wall dissociation, and turgidity loss) of tissues could be responsible for these results. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Carrot Hot air drying Glass transition temperature Water state Correlation
1. Introduction As well known, glass transition temperature (Tg) is a critical reference parameter to predict the quality, stability, and safety of food systems. It is defined as a critical temperature, where an amorphous system shifts from the glassy state to the rubbery one, or the opposite process. At the glassy state (T < Tg), the molecular mobility of materials is extremely slow (about 1012 m/s) (Roos, 2010). If the materials become glassy, this would allow them rigid and physicochemical changes are greatly restricted. In contrast, at the rubbery state (T > Tg), the molecular mobility exponentially increases and the viscosity dramatically decreases (Roos, 2010). These features collectively cause the various time-dependent and viscosity-related structural transformations, like collapse, “stickiness”, shrinkage, etc, during food processing and storage (Ruan et al., 1999; Kurozawa, Hubinger, & Park, 2012). Therefore, foods should be stored at the temperatures below their Tg to avoid loss in quality during the storage. Water acts as a strong plasticizer. High water content and water activity can reduce the Tg of food materials, such as lactose-whey protein system (Maidannyk & Roos, 2017), b-cyclodextrin/water
* Corresponding author. E-mail address: [email protected] (C. Xu). http://dx.doi.org/10.1016/j.lwt.2017.06.052 0023-6438/© 2017 Elsevier Ltd. All rights reserved.
binary system (Zhou et al., 2015), wheat gluten and maltodextrin (Shimazaki, Tashiro, Kumagai, & Kumagai, 2017), etc. Fresh fruits and vegetables have higher water content. This makes their Tg often lower than the common storage temperature (20 C). Thus, fresh tissues are subjected to a rubbery state, easily causing a severe loss of qualities. Hence, vegetables could be dehydrated to shift the Tg to the values close to room temperature (because of the reduction of moisture content) and thus extend their shelf-life during storage. Hot air drying is a conventional and widely applicable drying technology for foods (Moraga, Talens, Moraga, & MartínezNavarrete, 2011). To our knowledge, the detailed information of the changes of Tg in fruits and vegetables during hot air dehydration is still unclear. Additionally, nuclear magnetic resonance (NMR) is a nondestructive technique which is helpful to have more insight about microscopic information regarding with water mobility in biological tissues. Based on the analysis of the spin-spin relaxation times (T2), the water state within raw fruits and vegetables is usually categorized into “free water’’ in vacuole, ‘‘immobilized water’’ in cytoplasm and extracellular space, and ‘‘bound water’’ bounded to cell walls according to the mobility (Shao & Li, 2011; Xin, Zhang, & Adhikari, 2013). They will suffer dramatic changes during the dehydration process. For instance, vacuolar water (the most mobile water) decreases together with the increases of water in cytoplasm and intercellular space in strawberry due to the osmotic treatment
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
(Cheng, Zhang, Adhikari, & Islam, 2014). Tylewicz et al. (2016) have indicated that pulsed electric field pretreatment can cause water redistribution between different compartments in freeze-dried apples. Some water could shift from vacuole into the extracellular/cytoplasm spaces, producing an appreciable increase in its relaxation time. These changes of the water state in tissues can have a dramatic impact on the Tg. Our previous study has found that the Tg approaches a constant value when free water (the most mobile water) dominates in raw carrot tissues, probably owing to the very low viscosity of dilute solution. And the Tg increases with the decreased immobilized water in far-infrared dried carrots with low water content (Xu, Li, & Yu, 2014). Xin et al. (2013) have also showed that changing the states of water using osmotic dehydration can directly affect the Tg of broccoli. Concretely, the values of the Tg in the osmotically dehydrated broccoli are raised compared with the untreated samples. It can be seen that the water state in tissues is also a crucial contributor to influencing the Tg. However, to our knowledge, more works are still concentrated on the simplified water content/temperature state diagrams or the simplified water activity/temperature state diagrams to predict the glass transition phenomena. The reports about the effect of water state on the Tg in dehydrated products of fruits and vegetables is limited to the two ones above. Hence, a fundamental understanding of the relationship between water state and Tg is still greatly necessary for predicting the storage and processing stability of dehydrated products. Carrot, as a root vegetable, contains appreciable amounts of nutritional components (e.g. carotenes, vitamins, fiber content) and is widely consumed worldwide at a relatively low cost. Given these advantages, we took carrot as a model system to assess the changes of the water state (using nuclear magnetic resonance) and the Tg (from differential scanning calorimetry analysis) in hot airdehydrated carrots and establish their relationship using correlation analysis and path analysis. These findings could provide the theoretical basis for enhancing the storage and processing stability of dehydrated products of root vegetables. 2. Materials and methods 2.1. Sample preparation Carrots (Daucus carota L.) were harvested in November in Shanghai (China). Fresh carrots without any physical damage were bought from a local market. Carrots were cut into slices (6 ± 1 mm thickness and 32 ± 2 mm diameter) and randomly placed in a thin layer in an electrical thermostatic drying oven (Shanghai Yaoshi Instrument Factory, Shanghai, China) with 10e15 kg mass loading. Then, samples were dried at 50, 60, and 70 C for 14, 10, and 10 h, respectively, until reaching a final moisture content lower than 0.20 g water/g dry matter. A recirculating air maintains the constant temperature (±1 C) and the uniformity of the temperature at every corner (±2.5%) in the oven. Some samples were taken out every 2 h for analysis.
781
2.3. Nuclear magnetic resonance (NMR) Based on our previous study (Xu, Li, & Yu, 2014), a NMI 20Analyst system (22.6 MHz; Shanghai Niumag Corporation, Shanghai, China) was used to monitor the water state in carrot samples. Briefly, approximately 2 g of samples was applied to this study. Three duplicate NMR measurements (total of six runs) were carried out and a mean value taken. 2.4. Differential scanning calorimetry (DSC) The thermal transitions of carrot samples were assayed in a differential scanning calorimeter (DSC 204 F1 Phoenix®, Netzsch, Germany) according to our previous report (Xu et al., 2014) with some modifications. The heat flow and temperature of instrument were calibrated using distilled water (melting point (mp) ¼ 0 C, melting enthalpy (DHm) ¼ 333.88 J/g) and indium (mp ¼ 156.60 C, DHm ¼ 28.45 J/g). An empty sealed aluminum pan was employed as a reference. Nitrogen gas (50 mL/min) was used as the purge gas to avoid the condensation of water around samples during running. Samples (about 10 mg) were cooled from the load temperature (room temperature) to 60 C at 10 C/min and equilibrated at 60 C for 6 min. Finally, samples were scanned from 60 to 40 C at 10 C/min to determine the glass transition temperature (Tg). The onset (Tgo), mid-point (Tgm), and end (Tge) of the glass transition were assessed by using the TA universal analysis software provided with the DSC instrument. The midpoint of the transition was taken to be the Tg (Fig. 1). Data were reported as the mean value of three replicates for each treatment. 2.5. Microstructure According to the method reported by our previous study (Xu et al., 2015) with slight modification, specimens of carrot samples were prepared. Briefly, ultrathin sections (70 nm thickness) were obtained using an ultra-microtome (Leica Ultracut UC6, Germany). Specimens were double-stained with uranyl acetate and lead citrate, and examined in a Tecnai G2 Spirit Biotwin transmission electronic microscopy (TEM, FEI, Hillsboro, OR) at an accelerating voltage of 18 kV. 2.6. Statistical analyses Pearson's correlation analysis was performed by combining the data from all the carrot samples to address the relationship
2.2. Moisture content Carrot samples (initial mass, W1, g) were dried in an oven at 105 C to attain a constant mass (W2, g). Moisture content (Xw, g water/g dry matter) was calculated using Eq. (1). Data were reported as the mean value of three replicates for each treatment.
Xw ¼
W1 W2 W2
(1) Fig. 1. Typical DSC thermogram showing the glass transition temperature (Tg) of hot air-dried carrots at 70 C for 10 h.
782
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
between the selected parameters. Regress analysis and path analysis were carried out to assess the linear relationship and the respective effects of different parameters. Data were analyzed using SAS ver9.1 (SAS Institute, Cary, NC, USA). 3. Results and discussion 3.1. Xw The Xw of carrot samples decreased gradually with increasing drying time (Fig. 2). After drying at 50, 60, and 70 C for 14, 10, and 10 h, respectively, the values of Xw were decreased from an initial Xw of 7.82 g water/g dry matter to 0.12, 0.11, and 0.05 g water/g dry matter. 3.2. Water state and Tg 3.2.1. Water at different states The NMR technology, as a non-destructive test, can be applied to monitor the water state in biological materials (Vicente, Nieto, Hodara, Castro, & Alzamora, 2012). In our study, the spin-spin relaxation time (T2) and the relative signal intensity (M) were detected to investigate the distribution of water in carrots during hot air drying. In the T2 inversion spectra, the population groups of the shortest (T21), intermediate (T22), and longest (T23) relaxation times usually correspond to the bound water combined to cell walls, immobilized water in cytoplasm/extracellular space, and free water in vacuoles, respectively, in raw biological tissues (Cheng et al., 2014). Besides, the corresponding relative signal intensities of M21, M22, and M23 are employed to express the relative content of bound water, immobilized water, and free water, respectively, in these tissues. In our study, two main population groups (T22 and T23) were detected in the initial distribution. The respective ranges of T22 and T23 were 3e50 m and 60e1000 m. The T21 was negligible. Therefore, only the T22 and T23 were analyzed further. 3.2.2. Free water in vacuoles and Tg All the samples between raw carrots and carrots dried for 12, 6, and 4 h (at 50, 60, and 70 C, respectively) contained high Xw (7.82 Xw 1.01 g water/g dry matter; Fig. 2). The analysis of NMR results showed that at 7.82 Xw 1.01 g water/g dry matter, free water in vacuoles (the most mobile water) dominated in tissues (Figs. 3 and 4). The T23 decreased gradually from 682 m in raw samples to 123, 171, and 134 m at 50, 60, and 70 C, respectively
Fig. 2. Changes of moisture content (Xw) based on dry matter of carrots during hot air drying. : 50 C; : 60 C; : 70 C.
(Fig. 3A). Likewise, a continuous reduction of the M23 from 2150 in raw samples to 188, 268, and 366 at the respective temperatures above was observed (Fig. 3B). These results suggested that free water in vacuoles were gradually removed during hot air drying. This trend was in accordance with the finding in microwavedehydrated stem lettuces, where the T2 peak value with respect to free water was reduced by 717 m compared with the untreated samples (Wang, Zhang, Mujumdar, & Mothibe, 2012). Those are because that the loss of free water could cause the concentrated solute in vacuoles, the plasmolyzed cytoplasm, and the considerable accumulation of polysaccharides in the space between cell walls and plasmalemma, thereby shortening the T23 (Cheng et al., 2014; Hatakeyama, Tanaka, & Hatakeyama, 2010). Simultaneously, the changes of Tg were analyzed in the range of 7.82 Xw 1.01 g water/g dry matter, where free water in vacuoles dominated in tissues (Fig. 3A). For all the samples at 7.82 Xw 4.32 g water/g dry matter (drying time, 4, 2, and 2 h at 50, 60, and 70 C, respectively), the Tg appeared to be practically negligible in comparison with the latent heat of ice melting. Other samples at 4.15 Xw 1.01 g water/g dry matter, the values of the Tg remained practically constant. The mean values of the Tgo, Tg, and Tge for all these samples were estimated to be 51.7 ± 1.40 C, 48.4 ± 1.04 C, and 44.8 ± 1.39 C, respectively. These findings were similar to the result that the Tg of freeze-dried separated plum skin and pulp approaches a constant around 57.5 C in the range of higher water content (Telis, do Amaral Sobral, & Telis-Romero, 2006). This could be because that under the conditions that free water dominates in tissues, the matrix of cells is a dilute solution system with low viscosity. The effect of the viscosity of cell solution is slight compared with the water content. The Tg was independent of the initial concentration. Hence, the Tg of samples approached a constant value. 3.2.3. Immobilized water in the cytoplasm/extracellular space and Tg The changes of the T22 and M22 of carrot samples were more complex (Fig. 4). The T22 and M22 of samples increased within 6 h at 50 C and 2 h at 60 and 70 C, respectively. These increases could have been owing to the shift of free water to immobilized water because of the increase of the concentration of carbohydrates and the degradation of nutritional ingredients in the cytoplasm. Tylewicz et al. (2016) have indicated that some water from vacuole could move into the extracellular/cytoplasm spaces, producing an appreciable increase in the T22. Subsequently, the T22 and M22 of these samples decreased continuously, possibly due to the shift from immobilized water to the water more tightly bound to the polysaccharides in cell walls (e.g., pectin). Three types of hydrophilic groups of pectin can strongly bind to water (Einhorn-Stoll, Hatakeyama, & Hatakeyama, 2012). Carrots possess lots of pectin, thereby easily reducing the water mobility and elevating the T22. Li, Ma, Tao, Kong, and Li (2012) have showed that immobilized water could be transformed into bound water in beef granules during drying at 40, 50, and 60 C. After a short-term decrease, a sharp increase in the T22 and M22 was observed at the respective treatments of 14, 8, and 6 h at 50, 60, and 70 C, during which the amount of free water decreased sharply. Hence, the sharp increase could have been associated with a sharp shift of free water into immobilized water. Finally, the T22 and M22 showed a gradual downside. Under the conditions that most of the free water was lost, the immobilized water could have become more mobile and cause a considerable amount of loss with increasing drying time. Given the analysis of the results from NMR, all of the free water of samples was removed at the conditions of drying for 14, 8, and 6 h at 50, 60, and 70 C, respectively (Fig. 3). Simultaneously, the immobilized water sharply increased (Fig. 4). These results suggested that the dominate water presenting in all the samples (dried
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
783
Fig. 3. Changes of the relaxation time (T23) and relative signal intensity (M23) of free water in vacuoles and the glass transition temperature (Tg) of carrots during hot-air drying at 50, 60, and 70 C, respectively. (A) , , and : the T23 of samples dried at 50, 60, and 70 C, respectively; (B) , , and : the M23 of samples dried at 50, 60, and 70 C, respectively. , , and : the onset (Tgo), mid (Tg), and end (Tge) temperatures of glass transition of samples dried at 50 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 60 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 70 C, respectively.
for 14, 8, and 6 h at 50, 60, and 70 C, respectively) was the immobilized water. All these samples were with low Xw (0.41, g water/g dry matter; Fig. 2). At the critical times when the immobilized water began to dominate, the Tg of all samples showed an obvious increase (Fig. 4). Subsequently, the Tg increased gradually with decreasing content of immobilized water during dehydration. The Tg of samples dried at 60 and 70 C increased by 6.79 and
22.7 C with the T22 decreased by 12.4 and 11.8 m, respectively. Besides, the Xw decreased by 0.30 and 0.05 g water/g dry matter, respectively. A similar trend has been reported in several studies on various dried fruits and vegetables, such as tomatoes (Goula, Karapantsios, Achilias, & Adamopoulos, 2008), Agaricus bisporus (Shi, Wang, Zhao, & Fang, 2012), and jambolan (de Santana, de Oliveira Neto, Santos, Soares, Lima, & Cardoso, 2015).
784
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
Fig. 4. Changes of the relaxation time (T22) and relative signal intensity (M22) of immobilized water in cytoplasm/extracellular space, and the glass transition temperature (Tg) of carrots during hot-air drying at 50, 60, and 70 C, respectively. (A) , , and : the T22 of samples dried at 50, 60, and 70 C, respectively; (B) , , and : the M22 of samples dried at 50, 60, and 70 C, respectively. , , and : the onset (Tgo), mid (Tg), and end (Tge) temperatures of glass transition of samples dried at 50 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 60 C, respectively; , , and : the Tgo, Tg, and Tge of samples dried at 70 C, respectively.
3.2.4. Correlation between water state and Tg The curves in Fig. 4 intuitively showed that the Tg increased with decreasing T22, M22, and Xw of immobilized water in samples at Xw 0.41 g water/g dry matter. Furthermore, Table 1 indicated that the Tgo, Tg, and Tge of samples were correlated negatively with the T22 (P < 0.01) and M22 (P < 0.05). Likewise, the report published by Xin et al. (2013) has shown that the Tg is correlated negatively with
the T21 of immobilized water and the relative signal intensity of free water in osmotically treated broccoli. Moreover, the regression equations (Tgo ¼ 5.64e1.23 T22, R2 ¼ 0.81, P < 0.01; Tg ¼ 11.7e1.31 T22, R2 ¼ 0.85; P < 0.01; Tge ¼ 18.7e1.40 T22, R2 ¼ 0.87, P < 0.01) and the equations from path analysis (Tgo ¼ 0.90 T22, R2 ¼ 0.81, P < 0.01; Tg ¼ 0.92 T22, R2 ¼ 0.85, P < 0.01; Tge ¼ 0.94 T22, R2 ¼ 0.87, P < 0.01) were obtained in the
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787 Table 1 Correlation coefficients between water state and glass transition temperature (Tg) of hot air-dried carrots. Coefficient
Tgo
Tg
Tge
T22 M22
0.935** 0.698*
0.922** 0.698*
0.899** 0.689*
T22 and M22: the relaxation time and relative signal intensity of immobilized water in cytoplasm/extracellular space, respectively; Tgo and Tge: the onset and end temperature of glass transition, respectively. *Correlation is significant at the level of 0.05; ** Correlation is significant at the level of 0.01.
785
present study. These equations showed that the Tg had a linear relationship with the T22 and that the changes of the Tg was affected appreciably by the immobilized water. Compared with the effect of immobilized water, the effect of free water of carrots was not as appreciable. Hence, in comparison with free water in vacuoles, the immobilized water in cytoplasm/extracellular space exerted a negative and dominating part in the effect on the Tg in the hot airdried carrots. Additionally, compared with the correlation between M22 and Tg, the one between T22 and Tg was larger. This finding could be
Fig. 5. Changes of cell microstructure in raw and dried tissues. A: raw tissue; BeE: dried tissues at 70 C for 2, 4, 6, and 8 h, respectively. PM: plasmalemma; CW: cell wall; ML: middle lamella; TP: tonoplast; MT: mitochondria.
786
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787
because that the relaxation behavior has a direct relationship with water mobility in different parts of the cells as well as the interaction between water and other cell components (Guizani, Al-Saidi, Rahman, Bornaz, & Al-Alawi, 2010; Gussoni et al., 2007). The water redistribution between cellular structures could cause the changes of T2 in carrot slices treated by pulsed electric field treatment -Aguayo et al., 2014). Nevertheless, the M22 reflects the (Aguilo relative content of water at different states. Hence, the relaxation behavior well correlated with the Tg in hot-air dried carrots.
viscosity of the matrix over a wide range content, collectively lowering the mobility of the matrix. Consequently, the Tg of samples was elevated. Another reason for the increased Tg could be that because of the severe degradation of middle pectin lamella, the bound water (linked to the hydrophilic groups of pectin) might shift to more mobile water (immobilized water) and be removed further.
3.3. Cell microstructure
The water state and Tg dramatically changed during hot air dehydration in carrots. The moisture content of carrots at 4.15 Xw 1.01 g water/g dry matter, free water in vacuoles dominated in tissues and the Tg remained practically constant. When the Xw decreased to lower than 0.41 g water/g dry matter, the immobilized water in cytoplasm/extracellular space exerted a dominating and negative impact on the Tg of samples. As expected, the Tg value increased with the decreasing of immobilized water content during hot air dehydration. The regression equations showed the Tg increased appreciably according to the decreased relaxation behavior of the immobilized water in hot air-dried carrots. The more pronounced alteration in microstructure could explain these changes in hot air-dried carrots.
The changes of cell microstructure of raw and treated carrot samples (at 70 C within 8 h) were shown in Fig. 5. In raw samples (Fig. 5A), the mitochondria, tonoplast, and plasmalemma were intact. A large vacuole occupied most of the protoplast. The plasmalemma was tightly against the cell walls, which consisted of tightly packed and longitudinally-organized fibrillar materials. The middle lamella was limited between them. In samples at 2 h (Fig. 5B), organelles, plasmalemma, and tonoplasts were severely disrupted. The vacuoles shrunk clearly. Cell walls became distorted and swollen. Plasmolysis occurred with an obvious retraction of the plasmalemma and tonoplasts to the center of the cells, resulting in a separation of up to 1.1 mm between the plasmalemma and the cell wall in some areas. This result was similar to that observed in rezsliced carrots blanched at 90 C for 10 min (Neri, Hernando, Pe Munuera, Sacchetti, & Pittia, 2011). In samples at 4 h (Fig. 5C), no intact membrane structures (e.g. plasmalemma and organelles) were observed. Vacuoles shrunk and fractured further, only leaving some smaller ones in cells. The reticulate fibrillar pattern of cell walls became striated and loose. The central zone appeared denser, representing the presence of middle lamella and the separation of adjacent cell walls. This generation of swollen and loose cell walls might be due to the loosening of the hemicellulose-cellulose network (Phothiset & Charoenrein, 2014). The deterioration above was aggravated in samples at 6 and 8 h (Fig. 5D and E). In these cells, nearly all of vacuoles were lost. The zone of cell walls became lighter because of the severe degradation of fiber structure components. Besides, the middle lamella trended towards thinner, meaning more solubilisation of pectin components. Collectively, the dynamic changes of the water state as a consequence of the microstructure variation may be the main underlying cause of the changes of Tg of hot air-dried carrots. Hot air drying could cause the vaporization of escaped cytosolic fluid and increase the internal vapor pressure under the membrane structures, which will disrupt when the membrane stress exceeds the critical rupture stress (Li, Pan, Atungulu, Wood, & McHugh, 2014). With increasing time and temperature, these changes become more noticeable. Moreover, high temperatures (60 C) could directly degrade the membrane structures of tissues. These features could result in a substantial loss of free water in vacuoles, accompanied with cytoplasm plasmolysis and concentrated vacuoles. In our study, with the shrinkage of the large vacuole observed in raw samples (Fig. 5A) to small ones in samples (at 70 C within 4 h; Fig. 5B and C), the latent heat of ice (from free water) melting decreased. And the Tg became detectable. However, the matrix of samples is still a dilute solution system. The Tg was independent of the initial concentration. Hence, the Tg of the samples approached a constant value. With further dehydration, all of the vacuoles disappeared, meaning all of the free water was removed (Fig. 5D and E). Some could shift to the less mobile water (immobilized water), where a sharp increase of Tg was detected. Then, the immobilized water was also gradually removed. At these conditions, the immobilized water possesses a well plasticizing activity. Its decrease could reduce the free volume and increase the local
4. Conclusion
Acknowledgements Our authors gratefully acknowledge the financial support of China National Natural Science Foundation (Contract No.31271909) and the research project of experimental technology of Qufu Normal University (Contract No. SJ201503). References -Aguayo, I., Downey, G., Keenan, D. F., Lyng, J. G., Brunton, N., & Rai, D. K. Aguilo (2014). Observations on the water distribution and extractable sugar content in carrot slices after pulsed electric field treatment. Food Research International, 64, 18e24. Cheng, X. F., Zhang, M., Adhikari, B., & Islam, M. N. (2014). Effect of power ultrasound and pulsed vacuum treatments on the dehydration kinetics, distribution, and status of water in osmotically dehydrated strawberry: A combined NMR and DSC study. Food and Bioprocess Technology, 7(10), 2782e2792. Einhorn-Stoll, U., Hatakeyama, H., & Hatakeyama, T. (2012). Influence of pectin modification on water binding properties. Food Hydrocolloids, 27(2), 494e502. Goula, A. M., Karapantsios, T. D., Achilias, D. S., & Adamopoulos, K. G. (2008). Water sorption isotherms and glass transition temperature of spray dried tomato pulp. Journal of Food Engineering, 85(1), 73e83. Guizani, N., Al-Saidi, G. S., Rahman, M. S., Bornaz, S., & Al-Alawi, A. A. (2010). State diagram of dates: Glass transition, freezing curve and maximal-freezeconcentration condition. Journal of Food Engineering, 99(1), 92e97. Gussoni, M., Greco, F., Vezzoli, A., Paleari, M. A., Moretti, V. M., Lanza, B., et al. (2007). Osmotic and aging effects in caviar oocytes throughout water and lipid changes assessed by 1 H NMR T 1 and T 2 relaxation and MRI. Magnetic Resonance Imaging, 25(1), 117e128. Hatakeyama, T., Tanaka, M., & Hatakeyama, H. (2010). Thermal properties of freezing bound water restrained by polysaccharides. Journal of Biomaterials Science, Polymer Edition, 21(14), 1865e1875. Kurozawa, L. E., Hubinger, M. D., & Park, K. J. (2012). Glass transition phenomenon on shrinkage of papaya during convective drying. Journal of Food Engineering, 108(1), 43e50. Li, X., Ma, L. Z., Tao, Y., Kong, B. H., & Li, P. J. (2012). Low field-NMR in measuring water mobility and distribution in beef granules during drying process. Advanced Materials Research, 550, 3406e3410. Li, X., Pan, Z., Atungulu, G. G., Wood, D., & McHugh, T. (2014). Peeling mechanism of tomato under infrared heating: Peel loosening and cracking. Journal of Food Engineering, 128, 79e87. Maidannyk, V. A., & Roos, Y. H. (2017). Water sorption, glass transition and “strength” of lactose-Whey protein systems. Food Hydrocolloids, 70, 76e87. Moraga, G., Talens, P., Moraga, M. J., & Martínez-Navarrete, N. (2011). Implication of water activity and glass transition on the mechanical and optical properties of freeze-dried apple and banana slices. Journal of Food Engineering, 106(3), 212e219. rez-Munuera, I., Sacchetti, G., & Pittia, P. (2011). Effect of Neri, L., Hernando, I. H., Pe blanching in water and sugar solutions on texture and microstructure of sliced carrots. Journal of Food Science, 76(1), E23eE30. Phothiset, S., & Charoenrein, S. (2014). Effects of freezing and thawing on texture,
C. Xu et al. / LWT - Food Science and Technology 84 (2017) 780e787 microstructure and cell wall composition changes in papaya tissues. Journal of the Science of Food and Agriculture, 94(2), 189e196. Roos, Y. H. (2010). Glass transition temperature and its relevance in food processing. Annual Review of Food Science and Technology, 1, 469e496. Ruan, R., Long, Z., Chen, P., Huang, V., Almaer, S., & Taub, I. (1999). Pulse NMR study of glass transition in maltodextrin. Journal of Food Science, 64(1), 6e9. S., et al. de Santana, R. F., de Oliveira Neto, E. R., Santos, A. V., Soares, C. M. F., Lima, A. (2015). Water sorption isotherm and glass transition temperature of freezedried Syzygium cumini fruit (jambolan). Journal of Thermal Analysis and Calorimetry, 120(1), 519e524. Shao, X. L., & Li, Y. F. (2011). Application of low-field NMR to analyze water characteristics and predict unfrozen water in blanched sweet corn. Food and Bioprocess Technology, 1e7. Shimazaki, E., Tashiro, A., Kumagai, H., & Kumagai, H. (2017). Analysis of water sorption isotherms of amorphous food materials by solution thermodynamics with relevance to glass transition: Evaluation of plasticizing effect of water by the thermodynamic parameters. Bioscience, Biotechnology, and Biochemistry, 81(4), 743e749. Shi, Q. L., Wang, X. H., Zhao, Y., & Fang, Z. (2012). Glass transition and state diagram for freeze-dried Agaricus bisporus. Journal of Food Engineering, 111(4), 667e674. Telis, V. N., do Amaral Sobral, P. J., & Telis-Romero, J. (2006). Sorption isotherm, glass transitions and state diagram for freeze-dried plum skin and pulp. Food Science and Technology International, 12(3), 181e187.
787
Tylewicz, U., Aganovic, K., Vannini, M., Toepfl, S., Bortolotti, V., Dalla Rosa, M., et al. (2016). Effect of pulsed electric field treatment on water distribution of freezedried apple tissue evaluated with DSC and TD-NMR techniques. Innovative Food Science & Emerging Technologies, 37, 352e358. Vicente, S., Nieto, A. B., Hodara, K., Castro, M. A., & Alzamora, S. M. (2012). Changes in structure, rheology, and water mobility of apple tissue induced by osmotic dehydration with glucose or trehalose. Food and Bioprocess Technology, 5(8), 3075e3089. Wang, Y., Zhang, M., Mujumdar, A. S., & Mothibe, K. J. (2012). Experimental investigation and mechanism analysis on microwave freeze drying of stem lettuce cubes in a circular conduit. Drying Technology, 30(11e12), 1377e1386. Xin, Y., Zhang, M., & Adhikari, B. (2013). Effect of trehalose and ultrasound-assisted osmotic dehydration on the state of water and glass transition temperature of broccoli (Brassica oleracea L. var. botrytis L.). Journal of Food Engineering, 119(3), 640e647. Xu, C., Li, Y., & Yu, H. (2014). Effect of far-infrared drying on the water state and glass transition temperature in carrots. Journal of Food Engineering, 136, 42e47. Xu, C., Li, Y., & Yu, C. (2014). Effect of blanching pretreatment on carrot texture attribute, rheological behavior, and cell structure during cooking process. LWTFood Science and Technology, 62(1), 48e54. Zhou, G., Zhao, T., Wan, J., Liu, C., Liu, W., & Wang, R. (2015). Predict the glass transition temperature and plasticization of b-cyclodextrin/water binary system by molecular dynamics simulation. Carbohydrate Research, 401, 89e95.