- Email: [email protected]
Weight loss in superchilled pork as affected by cooling rate
Journal of Food Engineering 219 (2018) 25e28
Contents lists available at ScienceDirect
Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng
Weight loss in superchilled pork as affected by cooling rate Martin G. Landerslev, Adriana Araya-Morice, Luigi Pomponio, Jorge Ruiz-Carrascal* Department of Food Science, University of Copenhagen, Rolighedsvej 26, 1958 Frederiksberg C, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 March 2017 Received in revised form 14 September 2017 Accepted 15 September 2017 Available online 20 September 2017
This study aimed to identify in which extent cooling rate in pork subjected to superchilling influences the final weight loss after storage. Different cooling systems (brines at 15 C and -9 C, forced air at 18 C and still air 24 C) led to a range of cooling rates in pork model systems. The ice crust on the surface of pork grew faster in those systems with a higher energy flux. Higher cooling rates led to lower weight loss after superchilling storage, highlighting the importance of using fast cooling systems for superchilled pork. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Superchilling Cooling rate Weight loss Meat quality Infinite cylinder
1. Introduction The economic advantage of commercializing fresh compared to frozen pork is evident, due to the better quality, higher price and lower energy costs of the former. However, the shelf-life of chilled pork is not long enough for reaching distant markets. Superchilling (SC) is a process in which foods are kept at temperatures slightly below their initial freezing point (Kaale et al., 2011), which in pork means 1 to -2 C. Microbial spoilage in superchilled meat is delayed compared to conventional chilling, which prolongs the shelf life of pork up to 16 weeks (Duun et al., 2008; Kaale et al., 2011). In SC, during cooling, the temperature of the surface is brought below the freezing point of pork, usually by impingement freezing equipment, and a layer of ice is formed on the surface (Duun et al., 2008). The temperature equalizes within the product during storage, and the ice crust may remain or not depending on the storage conditions (Kaale et al., 2011). Crust freezing during the initial phase of cooling may cause structural damages to muscle cells, leading to a higher drip loss and consequently, to a lower meat quality (Kaale et al., 2013). It seems clear that a higher amount of frozen water might lead to more relevant damage of the muscle structure and lower meat quality. The size and distribution of ice crystals formed upon cooling also have an effect on meat quality:
* Corresponding author. E-mail address: [email protected] (J. Ruiz-Carrascal). https://doi.org/10.1016/j.jfoodeng.2017.09.012 0260-8774/© 2017 Elsevier Ltd. All rights reserved.
faster freezing rates lead to a higher amount of smaller and intracellular crystals, while slower freezing contributes to a higher proportion of bigger extracellular ones (James, 2009). Upon crustfreezing, this could have a subsequent influence on the quality of the superchilled meat once the temperature has equalized during the storage. This study aimed to elucidate in which extent some of the commercially used cooling rates for SC pork could influence weight loss. 2. Material & methods Four different cooling systems were studied: immersion in two different brines (15 C and -9 C), dynamic (forced air) cooling at 18 C and static (still air) cooling at 24 C. Heat transfer coefficients were calculated to ensure that they were representative of the systems used in industrial settings. Their cooling rates were calculated, both by following the drop of temperature at different depths in a controlled pork model, and by following the formation of the ice crust in pork models. Weight loss in the samples cooled under the four different systems after 7 days of superchilling storage (1.5 C) was determined. 2.1. Calculation of heat transfer coefficients Heat transfer coefficients were determined by measuring the thermal response in an aluminium block with the same geometry as the pork samples. The high thermal conductivity of aluminium
26
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
Fig. 1. (A) Scheme of the meat model system used for evaluating temperature drop at different depths with different cooling rates. The same system was used for calculating the ice crust formation. The set of pictures (B) shows the growth of the ice crust in pork cooled down in a brine at 15 C.
(200 W m1$K1) ensures a uniform thermal response and the heat transfer coefficients can be calculated from the thermal curve by equation (1):
U¼
Ts T Ts T0
¼e
h$A m$cp
specific heat capacity of aluminium. The determined heat transfer coefficients (h) were 900e1000 W m2 K1 for the brines, 17e18 W m2 K1 for the dynamic air system and 5e6 W m2 K1 for the static air system. 2.2. Cooling rates
$t
(1)
Where Ts is the temperature of the surrounding medium, T is the actual measured temperature of the aluminium block at time t, and T0 is the initial temperature of the aluminium block. A is the exposed surface area of the block, m is the mass and cp is the
In order to address the heat flow in meat under different cooling rates, an infinite cylindrical meat model was used. A total of 4 Longissimus dorsi muscles were cut using a sharpened edge stainless steel cylinder of 73 mm diameter. Cylindrical shaped meat pieces were fitted into aluminium cans of the same diameter,
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
27
Fig. 2. Weight loss (grey bars; average plus standard deviation) and freezing rates (black line with white squares) of meat model systems cooled down in different systems and stored for 7 days at superchilling temperature (1.5 C). Bars with different letters were significantly different (P < 0.05) in the Tukey's test.
avoiding air gaps between the meat and the can. Meat cylinders were thermally insulated with a polystyrene disk in both flat areas, so that the only possible heat transmission was through the side of the aluminium can (Fig. 1a). The temperature was recorded in the surface (between the meat and the aluminium can), and at 10, 20 and 36.5 mm of depth with needle T-type thermocouples connected to a computer through a thermocouple rack (TC-08 Pico Technology, Cambridgeshire, UK). An extra thermocouple was used to measure the temperature of the medium where the cooling was performed (either brine or air). Temperatures were recorded every 5 s until the estimated chilling time was achieved, and then the samples were immediately moved to superchilled storage at 1.5 C (see supplementary material). To determine the cooling rate, the Newton law of cooling was followed (equation (2)), stating that the energy flux (q), is proportional to the exposed surface area, the temperature difference and the heat transfer coefficient (h).
q ¼ h$A$ Ts Tp
(2)
Where Ts is the surrounding media temperature and Tp is the temperature of the product surface. In the conducted freezing experiment, we varied the energy flux, by manipulating both the heat transfer coefficient and the temperature difference. The flux was calculated from the experimental data of heat transfer coefficients and product surface temperatures. The cooling peztime to reach the wanted frozen fraction was estimated (Lo €m, 2003) based on enthalpy diagram for lean Leiva and Hallstro beef meat (Riedel, 1957). The theoretical cooling time prediction was used for estimating when the samples should be moved from the active cooling to the superchilled storage at 1.5 C (see supplementary material).
2.3. Ice crust thickness and weight loss To address the growing speed of the thickness of the ice crust formed during superchilling cooling conditions, meat model systems (7 for each cooling rate) similar to those previously described (Fig. 1) were tested. One of them was used as a dummy sample to
control temperature drop at the pre-established depths. The other 6 meat cylinders were covered with a layer of transparent polyethylene terephthalate (0.13 mm thickness), and the polystyrene insulation was placed on top. At fixed time intervals, the polystyrene insulation cover was briefly removed and the depth of the ice crust was measured with a ruler. The weights of the samples before cooling and after superchilling storage for 7 days were recorded and used to calculate weight loss (%). 2.4. Statistics Data were analysed using a one-way ANOVA, with cooling system as the main factor, using the General Linear Model (GLM) procedure (SPSS 15.0). When the effect was significant (P < 0.05), the Tukey's test was used at the 5% level to make pair wise comparisons between sample means. 3. Results & discussion The initial freezing point of meat has been reported to range from 1.2 C to 2.5 C. In our experiment the initial ice crystallization temperature and the equilibrium freezing point of the pork samples were measured by the cooling curve method (Shafiur Rahman et al., 2009). The temperature/time diagrams at different depths of the pork samples under different conditions showed initial freezing points in the range of-1.25 C to 1.31 C (see supplementary material), which are within those shown in the literature (Shafiur Rahman et al., 2009). The freezing rate of pork in a specific system depends on a number of factors, such as heat transfer coefficient, cooling temperature, thermodynamic properties of the meat or its size (James, 2009). Typical values for freezing rates are 0.2e0.5 cm/h in slow freezing (bulk freezing in cold chambers), 0.5e3 cm/h in quick freezing (air blast and contact plate freezers), 5e10 cm/h in rapid freezing (individual quick freezing of small sized products in fluidized beds), or 10e100 cm/h in ultra-rapid freezing by spraying or immersion in cryogenic fluids (liquid nitrogen, carbon dioxide). In our experiments, we achieved a range of freezing rates corresponding to those typically used in the meat industry (Fig. 2).
28
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
The progression of the advance of the ice crust ring was quite evident (Fig. 1b and supplementary material). The ice crust grew faster in those samples exposed to faster cooling rates (Fig. 2), but the final extension of the ice crust was somehow equalized in all samples during SC storage. After this point, there was not an evident crust and all the samples showed a similar semi-frozen state. Faster freezing rates lead to higher nucleation, and a greater number of crystals of smaller sizes distributed intra- and extracellularly (Nesvadba, 2009). At low cooling rates extracellular water freezes first. The consequent extracellular ice formation brings to an increase in solutes concentration in the unfrozen water of this location. This creates a chemical potential difference across the membrane of the cells producing a flux of water, which dehydrates and shrinks the cells. Slower cooling rates extend the dehydration of the muscle cells and causes irreversible damage. This is in agreement with the differences in weight loss shown in our trials (Fig. 2). In fact, a Pearson's correlation coefficient of R ¼ 0.88 (p < 0.01) between freezing rate and weight loss was found. Cooling at a higher freezing rate clearly led to lower weight loss, most likely due to the formation of smaller ice crystals at faster freezing rates, as it has been described previously in fish (Kaale et al., 2013). 4. Conclusions The conditions used for crust-freezing pork aimed for superchilling influence the weight loss of the product after SC storage, with higher cooling rates leading to lower weight loss. This highlights the importance of using fast cooling systems, such as impingement cooling, for maximizing the quality of pork.
Acknowledgements This research was supported by a GUDP grant from the Ministry of Food, Agriculture and Fisheries (Globalmeat: 34009-13-0699). The cooperation of Linda de Sparra Terkelsen in sampling is acknowledged. Jorge Ruiz-Carrascal thanks Norma & Frode S. Foundation for its support. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jfoodeng.2017.09.012. References Duun, A.S., Hemmingsen, A.K.T., Haugland, A., Rustad, T., 2008. Quality changes during superchilled storage of pork roast. LWT - Food Sci. Technol. 41 (10), 2136e2143. James, S., 2009. Freezing of Meat, Frozen Food Science and Technology. Blackwell Publishing Ltd, pp. 124e150. Kaale, L.D., Eikevik, T.M., Bardal, T., Kjorsvik, E., 2013. A study of the ice crystals in vacuum-packed salmon fillets (Salmon salar) during superchilling process and following storage. J. Food Eng. 115 (1), 20e25. Kaale, L.D., Eikevik, T.M., Rustad, T., Kolsaker, K., 2011. Superchilling of food: a review. J. Food Eng. 107 (2), 141e146. pez-Leiva, M., Hallstro € m, B., 2003. The original Plank equation and its use in the Lo development of food freezing rate predictions. J. Food Eng. 58 (3), 267e275. Nesvadba, P., 2009. Thermal Properties and Ice Crystal Development in Frozen Foods, Frozen Food Science and Technology. Blackwell Publishing Ltd, pp. 1e25. Riedel, L., 1957. Calorimetric investigation of the meat freezing process. €ltetechnik 9 (2), 38e40. Ka Shafiur Rahman, M., Machado-Velasco, K.M., Sosa-Morales, M.E., Velez-Ruiz, J., 2009. Freezing Point, Food Properties Handbook, Second Edition. CRC Press.
Contents lists available at ScienceDirect
Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng
Weight loss in superchilled pork as affected by cooling rate Martin G. Landerslev, Adriana Araya-Morice, Luigi Pomponio, Jorge Ruiz-Carrascal* Department of Food Science, University of Copenhagen, Rolighedsvej 26, 1958 Frederiksberg C, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 March 2017 Received in revised form 14 September 2017 Accepted 15 September 2017 Available online 20 September 2017
This study aimed to identify in which extent cooling rate in pork subjected to superchilling influences the final weight loss after storage. Different cooling systems (brines at 15 C and -9 C, forced air at 18 C and still air 24 C) led to a range of cooling rates in pork model systems. The ice crust on the surface of pork grew faster in those systems with a higher energy flux. Higher cooling rates led to lower weight loss after superchilling storage, highlighting the importance of using fast cooling systems for superchilled pork. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Superchilling Cooling rate Weight loss Meat quality Infinite cylinder
1. Introduction The economic advantage of commercializing fresh compared to frozen pork is evident, due to the better quality, higher price and lower energy costs of the former. However, the shelf-life of chilled pork is not long enough for reaching distant markets. Superchilling (SC) is a process in which foods are kept at temperatures slightly below their initial freezing point (Kaale et al., 2011), which in pork means 1 to -2 C. Microbial spoilage in superchilled meat is delayed compared to conventional chilling, which prolongs the shelf life of pork up to 16 weeks (Duun et al., 2008; Kaale et al., 2011). In SC, during cooling, the temperature of the surface is brought below the freezing point of pork, usually by impingement freezing equipment, and a layer of ice is formed on the surface (Duun et al., 2008). The temperature equalizes within the product during storage, and the ice crust may remain or not depending on the storage conditions (Kaale et al., 2011). Crust freezing during the initial phase of cooling may cause structural damages to muscle cells, leading to a higher drip loss and consequently, to a lower meat quality (Kaale et al., 2013). It seems clear that a higher amount of frozen water might lead to more relevant damage of the muscle structure and lower meat quality. The size and distribution of ice crystals formed upon cooling also have an effect on meat quality:
* Corresponding author. E-mail address: [email protected] (J. Ruiz-Carrascal). https://doi.org/10.1016/j.jfoodeng.2017.09.012 0260-8774/© 2017 Elsevier Ltd. All rights reserved.
faster freezing rates lead to a higher amount of smaller and intracellular crystals, while slower freezing contributes to a higher proportion of bigger extracellular ones (James, 2009). Upon crustfreezing, this could have a subsequent influence on the quality of the superchilled meat once the temperature has equalized during the storage. This study aimed to elucidate in which extent some of the commercially used cooling rates for SC pork could influence weight loss. 2. Material & methods Four different cooling systems were studied: immersion in two different brines (15 C and -9 C), dynamic (forced air) cooling at 18 C and static (still air) cooling at 24 C. Heat transfer coefficients were calculated to ensure that they were representative of the systems used in industrial settings. Their cooling rates were calculated, both by following the drop of temperature at different depths in a controlled pork model, and by following the formation of the ice crust in pork models. Weight loss in the samples cooled under the four different systems after 7 days of superchilling storage (1.5 C) was determined. 2.1. Calculation of heat transfer coefficients Heat transfer coefficients were determined by measuring the thermal response in an aluminium block with the same geometry as the pork samples. The high thermal conductivity of aluminium
26
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
Fig. 1. (A) Scheme of the meat model system used for evaluating temperature drop at different depths with different cooling rates. The same system was used for calculating the ice crust formation. The set of pictures (B) shows the growth of the ice crust in pork cooled down in a brine at 15 C.
(200 W m1$K1) ensures a uniform thermal response and the heat transfer coefficients can be calculated from the thermal curve by equation (1):
U¼
Ts T Ts T0
¼e
h$A m$cp
specific heat capacity of aluminium. The determined heat transfer coefficients (h) were 900e1000 W m2 K1 for the brines, 17e18 W m2 K1 for the dynamic air system and 5e6 W m2 K1 for the static air system. 2.2. Cooling rates
$t
(1)
Where Ts is the temperature of the surrounding medium, T is the actual measured temperature of the aluminium block at time t, and T0 is the initial temperature of the aluminium block. A is the exposed surface area of the block, m is the mass and cp is the
In order to address the heat flow in meat under different cooling rates, an infinite cylindrical meat model was used. A total of 4 Longissimus dorsi muscles were cut using a sharpened edge stainless steel cylinder of 73 mm diameter. Cylindrical shaped meat pieces were fitted into aluminium cans of the same diameter,
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
27
Fig. 2. Weight loss (grey bars; average plus standard deviation) and freezing rates (black line with white squares) of meat model systems cooled down in different systems and stored for 7 days at superchilling temperature (1.5 C). Bars with different letters were significantly different (P < 0.05) in the Tukey's test.
avoiding air gaps between the meat and the can. Meat cylinders were thermally insulated with a polystyrene disk in both flat areas, so that the only possible heat transmission was through the side of the aluminium can (Fig. 1a). The temperature was recorded in the surface (between the meat and the aluminium can), and at 10, 20 and 36.5 mm of depth with needle T-type thermocouples connected to a computer through a thermocouple rack (TC-08 Pico Technology, Cambridgeshire, UK). An extra thermocouple was used to measure the temperature of the medium where the cooling was performed (either brine or air). Temperatures were recorded every 5 s until the estimated chilling time was achieved, and then the samples were immediately moved to superchilled storage at 1.5 C (see supplementary material). To determine the cooling rate, the Newton law of cooling was followed (equation (2)), stating that the energy flux (q), is proportional to the exposed surface area, the temperature difference and the heat transfer coefficient (h).
q ¼ h$A$ Ts Tp
(2)
Where Ts is the surrounding media temperature and Tp is the temperature of the product surface. In the conducted freezing experiment, we varied the energy flux, by manipulating both the heat transfer coefficient and the temperature difference. The flux was calculated from the experimental data of heat transfer coefficients and product surface temperatures. The cooling peztime to reach the wanted frozen fraction was estimated (Lo €m, 2003) based on enthalpy diagram for lean Leiva and Hallstro beef meat (Riedel, 1957). The theoretical cooling time prediction was used for estimating when the samples should be moved from the active cooling to the superchilled storage at 1.5 C (see supplementary material).
2.3. Ice crust thickness and weight loss To address the growing speed of the thickness of the ice crust formed during superchilling cooling conditions, meat model systems (7 for each cooling rate) similar to those previously described (Fig. 1) were tested. One of them was used as a dummy sample to
control temperature drop at the pre-established depths. The other 6 meat cylinders were covered with a layer of transparent polyethylene terephthalate (0.13 mm thickness), and the polystyrene insulation was placed on top. At fixed time intervals, the polystyrene insulation cover was briefly removed and the depth of the ice crust was measured with a ruler. The weights of the samples before cooling and after superchilling storage for 7 days were recorded and used to calculate weight loss (%). 2.4. Statistics Data were analysed using a one-way ANOVA, with cooling system as the main factor, using the General Linear Model (GLM) procedure (SPSS 15.0). When the effect was significant (P < 0.05), the Tukey's test was used at the 5% level to make pair wise comparisons between sample means. 3. Results & discussion The initial freezing point of meat has been reported to range from 1.2 C to 2.5 C. In our experiment the initial ice crystallization temperature and the equilibrium freezing point of the pork samples were measured by the cooling curve method (Shafiur Rahman et al., 2009). The temperature/time diagrams at different depths of the pork samples under different conditions showed initial freezing points in the range of-1.25 C to 1.31 C (see supplementary material), which are within those shown in the literature (Shafiur Rahman et al., 2009). The freezing rate of pork in a specific system depends on a number of factors, such as heat transfer coefficient, cooling temperature, thermodynamic properties of the meat or its size (James, 2009). Typical values for freezing rates are 0.2e0.5 cm/h in slow freezing (bulk freezing in cold chambers), 0.5e3 cm/h in quick freezing (air blast and contact plate freezers), 5e10 cm/h in rapid freezing (individual quick freezing of small sized products in fluidized beds), or 10e100 cm/h in ultra-rapid freezing by spraying or immersion in cryogenic fluids (liquid nitrogen, carbon dioxide). In our experiments, we achieved a range of freezing rates corresponding to those typically used in the meat industry (Fig. 2).
28
M.G. Landerslev et al. / Journal of Food Engineering 219 (2018) 25e28
The progression of the advance of the ice crust ring was quite evident (Fig. 1b and supplementary material). The ice crust grew faster in those samples exposed to faster cooling rates (Fig. 2), but the final extension of the ice crust was somehow equalized in all samples during SC storage. After this point, there was not an evident crust and all the samples showed a similar semi-frozen state. Faster freezing rates lead to higher nucleation, and a greater number of crystals of smaller sizes distributed intra- and extracellularly (Nesvadba, 2009). At low cooling rates extracellular water freezes first. The consequent extracellular ice formation brings to an increase in solutes concentration in the unfrozen water of this location. This creates a chemical potential difference across the membrane of the cells producing a flux of water, which dehydrates and shrinks the cells. Slower cooling rates extend the dehydration of the muscle cells and causes irreversible damage. This is in agreement with the differences in weight loss shown in our trials (Fig. 2). In fact, a Pearson's correlation coefficient of R ¼ 0.88 (p < 0.01) between freezing rate and weight loss was found. Cooling at a higher freezing rate clearly led to lower weight loss, most likely due to the formation of smaller ice crystals at faster freezing rates, as it has been described previously in fish (Kaale et al., 2013). 4. Conclusions The conditions used for crust-freezing pork aimed for superchilling influence the weight loss of the product after SC storage, with higher cooling rates leading to lower weight loss. This highlights the importance of using fast cooling systems, such as impingement cooling, for maximizing the quality of pork.
Acknowledgements This research was supported by a GUDP grant from the Ministry of Food, Agriculture and Fisheries (Globalmeat: 34009-13-0699). The cooperation of Linda de Sparra Terkelsen in sampling is acknowledged. Jorge Ruiz-Carrascal thanks Norma & Frode S. Foundation for its support. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jfoodeng.2017.09.012. References Duun, A.S., Hemmingsen, A.K.T., Haugland, A., Rustad, T., 2008. Quality changes during superchilled storage of pork roast. LWT - Food Sci. Technol. 41 (10), 2136e2143. James, S., 2009. Freezing of Meat, Frozen Food Science and Technology. Blackwell Publishing Ltd, pp. 124e150. Kaale, L.D., Eikevik, T.M., Bardal, T., Kjorsvik, E., 2013. A study of the ice crystals in vacuum-packed salmon fillets (Salmon salar) during superchilling process and following storage. J. Food Eng. 115 (1), 20e25. Kaale, L.D., Eikevik, T.M., Rustad, T., Kolsaker, K., 2011. Superchilling of food: a review. J. Food Eng. 107 (2), 141e146. pez-Leiva, M., Hallstro € m, B., 2003. The original Plank equation and its use in the Lo development of food freezing rate predictions. J. Food Eng. 58 (3), 267e275. Nesvadba, P., 2009. Thermal Properties and Ice Crystal Development in Frozen Foods, Frozen Food Science and Technology. Blackwell Publishing Ltd, pp. 1e25. Riedel, L., 1957. Calorimetric investigation of the meat freezing process. €ltetechnik 9 (2), 38e40. Ka Shafiur Rahman, M., Machado-Velasco, K.M., Sosa-Morales, M.E., Velez-Ruiz, J., 2009. Freezing Point, Food Properties Handbook, Second Edition. CRC Press.