7 Freezing of meat Meat for industrial processing is usually frozen in the form of carcasses, quarters or boned out primals in 25kg cartons. Most bulk meat, consumer portions and meat products are frozen in air blast freezers. Some small indi- viduals items, for example beefburgers, may be frozen in cryogenic tunnels and a small amount of offal and other meat is frozen in plate freezers. It is not unusual for meat to be frozen twice before it reaches the consumer. During industrial processing frozen raw material is often thawed or tem- pered before being turned into meat-based products, i.e. pies, convenience meals, burgers, etc or consumer portions, fillets, steaks, and so on. These consumer-sized portions are often refrozen before storage, distribution and sale. 7.1 Freezing rate There are little data in the literature to suggest that, in general, the method of freezing or the rate of freezing has any substantial influence on a meat’s subsequent storage life, its quality characteristics or final eating quality. There is some disagreement in the literature about whether fast or slow freezing is advantageous. Slightly superior chemical and sensory attributes have been found in food cryogenically frozen in a few trials (Sebranek et al., 1978; Dobryschi et al., 1977; Sebranek, 1980) but other trials did not show any appreciable advantage (Lampitt and Moran, 1933) especially during short term storage (Hill and Glew, 1973). Jackobsson and Bengtson (1973) indicated that there is an interaction between freezing rate and cooking method. Meat that had been cooked from frozen was found to show a favourable effect from faster freezing rates. Mittal and Barbut (1991) showed that freezing rate affected the modulus of rigidity of meat after cooking. Similar values to fresh meat were produced in meat frozen in liquid nitrogen. The value of the modulus increased as the rate of freez- ing decreased. In 1980 A?ón and Calvelo reported a relationship between the rate of freezing and drip loss, drip loss reaching a maximum when the freezing time from -1 to -7°C was ca. 17 min. Mascheroni (1985) used this relationship to produce a method for determining the rate at which frozen meat had been frozen. However, attempts to replicate the work at Langford (James et al., 1983) were unsuccessful because of the variability in drip loss from meat before freezing. Studies using differential scanning calorimetry (DSC) on fresh and frozen bovine muscle at different freezing rates show a decrease of denaturation enthalpies; the slower the freezing rate the greater the loss (Wagner and A?ón, 1985). Investigations covered freezing times from 5 to 60 min. Experiments with pork M. longissimus dorsi found no difference in drip loss between samples frozen at -20°C or -80 °C (Sakata et al., 1995). At -20 and -80 °C samples took 6 and 3h, respectively to pass from -1°C to ca. -6 °C. In the -20 °C samples inter- and intracellular ice were seen but only intracellular ice was seen at -80°C. Methods of freezing clearly affect the ultrastructure of muscle. Slow freezing (1–2 mmh -1 for example (Buchmuller, 1986) tends to produce large ice crystals extracellularly, whilst quick freezing (e.g. 50 mmh -1 ) gives smaller crystals in and outside cells (Buchmuller, 1986; Bevilacqua et al., 1979). Obviously a temperature gradient will occur in large pieces of meat and result in a non-uniform ice morphology (Bevilacqua et al., 1979). Petrovic et al. (1993) found that slowly frozen meat, 0.22 and 0.39 cmh -1 , lost more weight during freezing, thawing and cooking than that frozen at 3.95–5.66cmh -1 (Table 7.1). However, higher weight losses during thawing were measured at an intermediate freezing rate of 3.33 cmh -1 . Meat frozen at rates of 3.33 cmh -1 and faster was rated as more tender and juicier after cooking than unfrozen controls and slow frozen samples (Table 7.2). Petro- vic et al. stated that the optimal conditions for freezing portioned meat are those that achieve freezing rates between 2 and 5 cmh -1 to -7 °C. Grujic et al. (1993) suggest even tighter limits, 3.33–3.95 cmh -1 . Slow freezing at up to 0.39 cmh -1 resulted in decreased solubility of myofibrillar proteins, increase in weight loss during freezing, thawing and cooking, lower water- binding capacity and tougher cooked meat. Very quickly frozen meat (>4.9cmh -1 ) had a somewhat lower solubility of myofibrillar proteins, lower water-binding capacity and somewhat tougher and drier meat. The samples were thawed after storage times of 2–3 days at -20 °C so the relationship between freezing rates and storage life was not investigated. Storage times of 48 h and 2.5 months were used during investigations of the effect of different freezing systems and rates on drip production from 138 Meat refrigeration small samples of mutton muscle (Sacks et al., 1993). In all cases, drip loss after 2.5 months was at least double the percentage measured after 48 h (Table 7.3). After 2.5 months, drip loss from samples frozen using cryogen- ics was >2% less than in those using air freezing. The most recent comparison (Sundsten et al., 2001) revealed some com- mercial advantages of fast freezing, but no quality advantages. The studies compared three different freezing methods, spiral freezing (SF), cryogenic freezing (liquid nitrogen, LN) and impingement freezing (IF). The times required to freeze a 10mm thick 80g hamburger from +4°C to -18°C in the SF, LN and IF were 22 min, 5 min 30 s and 2 min 40 s, respectively. The authors state that dehydration was significantly higher for hamburgers frozen in SF (1.2%) compared to LN (0.4%) and IF (0.4%). No significant Freezing of meat 139 Table 7.1 Relationship between freezing rate of beef M. longissimus dorsi and weight loss during freezing, thawing and cooking Freezing rate (cm h -1 ) % Weight loss during Freezing Thawing Cooking Control – – 36.32 0.22 2.83 0.78 38.41 0.39 2.58 0.72 38.00 3.33 1.15 1.21 37.47 3.95 1.05 0.18 37.24 4.92 0.87 0.10 37.15 5.66 0.63 0.03 37.14 Source: Petrovic et al., 1993. Table 7.2 Relationship between freezing rate of beef M. longissimus dorsi and texture Freezing rate Texture Tenderness Juiciness (cm h -1 ) Control 7.0 6.8 7.0 0.22 7.0 6.0 6.7 0.39 7.0 6.5 7.0 3.33 7.0 7.5 7.5 3.95 7.0 8.0 8.0 4.92 7.0 8.5 8.5 5.66 6.0 7.0 7.3 Source: Petrovic et al., 1993. Texture: 1 = extremely tough, 7 = extremely fine. Tenderness: 1 = extremely hard, 9 = extremely tender. Juiciness: 1 = extremely dry, 9 = extremely juicy. difference could be seen in cooking losses, even after storage for 2 months. Ice crystals were significantly larger in hamburgers frozen in SF compared to LN and IF. Sensory analysis revealed no difference in eating quality between the three freezing methods, even after storage for 2 months. Slow freezing from a high initial temperature can provide conditions for microbial growth compared with a very rapid freezing process. Castell- Perez et al. (1989) predicted that slow freezing from an initial product temperature of 30°C could result in an 83% increase in bacterial numbers compared with a 4% increase from 10 °C. 7.2 Freezing systems 7.2.1 Air Air is by far the most widely used method of freezing food as it is eco- nomical, hygienic and relatively non-corrosive to equipment. Systems range from the most basic, in which a fan draws air through a refrigerated coil and blows the cooled air around an insulated room (Fig. 7.1), to purpose- built conveyerised blast freezing tunnels or spirals. Relatively low rates of heat transfer are attained from product surfaces in air systems. The big advantage of air systems is their versatility, especially when there is a requirement to freeze a variety of irregularly shaped products or individ- ual products. In practice, air distribution is a major problem, often overlooked by the system designer and the operator. The freezing time of the product is re- duced as the air speed is increased. An optimum value exists between the decrease in freezing time and the increasing power required to drive the fans to produce higher air speeds. This optimum value can be as low as 1.0ms -1 air speed when freezing beef quarters rising to 15 m s -1 or more for thin products. 140 Meat refrigeration Table 7.3 Drip loss (%) from 77.6 g samples of longissimus lumborum et thoracis frozen under different methods and thawed at 4 °C Freezing conditions Freezing time Freezing rate Storage time to -2.2 °C (cm h -1 ) at -20 °C 48 h 2.5 months Cryogenic, -90 °C 15 min 6.4 3.34 a 9.49 a Cryogenic, -65 °C 22 min 4.4 4.70 ab 9.72 a Blast freezer, -21 °C 1.83 h 0.55 5.53 b 12.74 b Walk-in-freezer, -21 °C 1.88 h 0.53 4.71 ab 13.18 b Domestic freezer, -25 °C 1.96 h 0.51 5.26 b 11.72 b Means in the same column with different superscripts are different at P > 0.05h Source: Sacks et al., 1993. The use of impingement technology to increase the surface heat trans- fer in air and other freezing systems has received attention recently (Newman, 2001; Sundsten et al., 2001; Everington, 2001). Impingement is the process of directing a jet or jets of fluid at a solid surface to effect a change. When the jets of fluid are very cold gas, the change is a dramatic increase in convective surface heat transfer coefficients. The very high velocity (20–30 m s -1 ) impingement gas jets, ‘breakup’ the static surface boundary layer of gas that surrounds a food product. The resulting medium around the product is more turbulent and the heat exchange through this zone becomes much more effective. 7.2.1.1 Batch systems Placing food items in large refrigerated rooms is the most common method of freezing. Fans circulate air through refrigerated coils and around the products in an insulated room. Large individual items such as meat car- casses are hung from overhead rails, smaller products are placed either unwrapped or in cartons on racks, pallets, or large bins. 7.2.1.2 Continuous systems In a continuous system, meat is conveyed through a freezing tunnel or refrigerated room usually by an overhead conveyor or on a belt. This over- comes the problem of uneven air distribution since each item is subjected to the same velocity/time profile. Some meat products are frozen on racks of trays (2 m high), pulled or pushed through a freezing tunnel by me- chanical means. For larger operations, it is more satisfactory to use feed meat on a continuous belt through linear tunnels or spiral freezers. Linear Freezing of meat 141 Evaporator Reversible fan False ceiling Product on trolleys Fig. 7.1. Example of a freezing tunnel with longitudinal air circulation. tunnels are of simpler construction but are restricted by the length of belt necessary to achieve the cooling time required and on the space available in most factories. Spiral freezers are therefore a more viable alternative. 7.3 Contact freezers Contact freezing methods are based on heat transfer by contact between products and metal surfaces, which in turn are cooled by either primary or secondary refrigerants. Contact freezing offers several advantages over air cooling, i.e. there is much better heat transfer and significant energy savings. However, the need for regularly shaped products with large flat surfaces is a major hindrance. Modern plate cooling systems differ little in principle from the first contact freezer patented in 1929 by Clarence Birdseye. Essentially the product is pressed between hollow metal plates containing a circulating refrigerant (Fig. 7.2). A hydraulic cylinder is used to bring the freezing plates into pressure contact with the product. These plates can be either horizontal or vertical. Good heat transfer is dependent on product thickness, good contact and the conductivity of the product. Plate freezers are often limited to a maximum thickness of 50–70 mm. Good contact is a prime requirement.Air spaces in packaging and fouling of the plates can have a significant effect on cooling time, for example a water droplet frozen on the plate can lengthen the freezing time in the concerned tray by as much as 30–60%. General advantages of plate freezers over air-blast carton freezers include: 142 Meat refrigeration Hydraulic ram Freezing plate Product Separated plates Closed plates Fig. 7.2. Example of a horizontal plate freezer. ? Freezing is either faster for the same refrigerant evaporating tempera- ture, or can take place at a higher evaporating temperature for a given freezing time. ? Product temperatures are easier to control, especially for smaller cuts. ? Power consumption is significantly reduced – savings of at least 30%, and possibly 50% or more, may be expected because air-circulating fans are not required and because higher evaporating temperatures can be used for the same effective cooling medium temperature. ? In many cases, less building space is required. ? The product remains uniform and flat after freezing, unlike air-blast frozen cartons which often bulge. The flat cartons result in stable loads, giving up to 30% higher space utilisation in cold stores. For transport, the stable pallets facilitate unitised loading, and some 8–10% more product can be loaded into a container. Disadvantages of plate freezers relate mainly to cost aspects: ? Capital costs are significantly higher than for equivalent air-blast freezers. Manually loaded plate freezers are comparable in cost to auto- matic air-blast tunnel freezers. Fully automatic plate freezers are more expensive. ? High circulation rates of liquid refrigerant are required; this results in additional costs for larger accumulators and higher capacity pumps. ? For manual plate freezers, simultaneous loading and unloading may require higher labour input than for a batch air freezer. ? Each plate must be loaded with product of the same thickness. ? Damp cartons can stick to plates or cause jams when ice forms. ? Air infiltration must be minimised to prevent frost build up on plates. Freezing unpacked meat has significant advantages because of the sub- stantially shorter freezing times (Fleming et al., 1996). Twice as many freez- ing cycles per day can be achieved with the bare product (Table 7.4). Overall costs for plate freezing can be comparable to those for air-blast freezing. De Jong (1994) carried out a cost analysis (Table 7.5) for a beef plant using either plate or air blast freezers in New Zealand which assumed Freezing of meat 143 Table 7.4 Predicted freezing time of meat blocks in a plate freezer operating at -30 °C Thickness (mm) Freezing time (h) Cycles per day Cartoned Bare Cartoned Bare 80 6.3 2.5 3 6 160 16.5 8.5 1 2 Source: Fleming et al., 1996. a net electricity cost of NZ$0.10 per kilowatt hour, a capital recovery over 10 years and an interest rate of 12%. 7.4 Cryogenic freezing Cryogenic freezing uses refrigerants, such as liquid nitrogen or solid carbon dioxide, directly. The method of cooling is essentially similar to water-based evaporative cooling, cooling being brought about by boiling off the refrig- erant, the essential difference being the temperature required for boiling. As well as using the latent heat absorbed by the boiling liquid, sensible heat is absorbed by the resulting cold gas. Owing to very low operating temperatures and high surface heat trans- fer coefficients between product and medium, cooling rates of cryogenic systems are often substantially higher than other refrigeration systems. Most systems use total loss refrigerants, i.e. the refrigerant is released to the atmosphere and not recovered. Alternatively dichlorodifluoromethane (CCl 2 F 2 ) (otherwise known as Freon 12, R.12 or F12) may be used in a recovery and recycle system, however, R.12 is not generally accepted in all countries. Because of environmental and economic factors total loss refrig- erants must be both readily available and harmless, which limits the choice to atmospheric air and its components, liquid nitrogen (LN) and liquid or solid carbon dioxide (CO 2 ). The particular characteristics of total loss refrigerants that may be regarded as advantages or disadvantages are listed in Table 7.6. Cryogenic freezing is mainly used for small products such as burgers, ready meals, and so on. The most common method is by direct spraying of liquid nitrogen onto a food product while it is conveyed through an insu- lated tunnel. 144 Meat refrigeration Table 7.5 Energy and cost requirements for beef plant freezing 3000 cartons per day Four-batch air Automatic air Manual plate freezers blast freezers Energy analysis Freezing time (h) 38 38 17 Freezer load (kW) 573 477 400 Power consumption (kW) 689 570 462 Economic analysis (NZ$000) Capital cost 1500 2200 2000 Annual capital charges 266 389 354 Annual energy costs 463 383 310 Annual labour cost 60 0 90 Total annual cost 789 772 754 Source: De Jong, 1994. Impingement technologies are being used to increase heat transfer further (Newman, 2001). Newman states that when comparing the overall heat transfer coefficients of cryogenic freezing tunnels, impingement heat transfer is typically 3–5 times that of a conventional tunnel utilizing axial flow fans.With the increased overall heat transfer coefficient, one can either increase the freezing temperature to increase overall cryogen efficiency or continue to run at very cold temperatures and dramatically increase the overall production rate. Impingement freezing is best suited for products with high surface area to weight ratios, for example hamburger patties or products with one small dimension. Testing has shown that products with a thickness of less than 20 mm freeze most effectively in an impingement heat transfer environment. When freezing products thicker than 20mm, the benefits of impingement freezing can still be achieved, however, the surface heat transfer coefficients later in the freezing process should be reduced to balance the overall process efficiency. The process is also very attractive for products that require very rapid surface freezing and chilling. 7.5 Freezing of specific products 7.5.1 Meat blocks James et al. (1979) showed that air temperatures below -30 °C and air velocities exceeding 5 m s -1 are required to freeze 15cm thick meat blocks in corrugated cardboard cartons in less than 24 h (Fig. 7.3). Creed and James (1981) carried out a survey which indicated that only 58% of industrial throughput is frozen in times within ±20% of the actual freezing time required. 7.5.2 Beef quarters James and Bailey (1987a) reported that brine spray and liquid nitrogen immersion systems had been used to freeze beef quarters. However, most Freezing of meat 145 Table 7.6 Advantages and disadvantages of total loss refrigerants in comparison with mechanical refrigeration Advantages Disadvantages Low capital investment High operating cost High refrigerating capacity High refrigerating capacity Low weight when out of use High weight at start of use No residual weight (dry ice) Limited duration without filling No noise Poor temperature control Advantageous storage atmosphere (N 2 ) Reduced humidity Bacteriostatic affect (CO 2 ) Suffocation hazard Low maintenance requirements Limited availability Foolproof once installed (dry ice) investigations had used air. Temperature in air systems ranged from -11 to -40 °C and weight loss from 0.3 to 1.19%. In their own investigations beef quarters ranging in weight from 40 to 140kg were frozen in air at -32 °C, 1.5ms -1 . On average, hindquarters below 50 kg and forequarters below 75 kg could be frozen in a 24 h operation (Table 7.7). There was no statistical dif- ference in bacterial counts before and after freezing. 7.5.3 Mutton carcasses Mutton production is seasonal and continuity of supply for processing can be achieved by frozen storage and subsequent thawing and boning (Creed and James, 1984). Data from the investigations of Creed and James were used to verify a predictive program for freezing of mutton carcasses. The predictions indi- cated that any condition more severe than -20°C, 0.5ms -1 , would achieve a 24 h freezing operation for unwrapped carcasses (Table 7.8). To guaran- tee an overnight (15–16h) freezing cycle for wrapped carcasses, conditions more severe than -30°C, 4ms -1 , would be required. 7.5.4 Offal Although edible offal comprises 3–4% of the cold weight of a carcass there is little published data on its refrigeration (Creed and James, 1983). The 146 Meat refrigeration 0 20 40 60 80 100 120 38 25 55 38 112 75 –30 –20 –10 T ime from 4 to – 7 ° C (h) Air temperature (°C) 0.5 m/s 5.0 m/s Fig. 7.3. Freezing time of 15 cm thick boxed blocks (source: James et al., 1979). Table 7.7 Average freezing time from 4 to -7 °C at thermal centre of 50, 75 and 100 kg beef quarters Weight 50 kg 75 kg 100 kg Hindquarter 22 29 33 Forequarter 13 20 25 Source: James et al., 1987. authors found that liver was amenable to plate freezing and the freezing time to -7°C (Y) was related to the initial temperature (I) and R the reci- procal of -1.5 °C, the refrigerant temperature (Table 7.9). The authors extended their studies to examine the effect of different packaging materials on freezing time (Creed and James, 1985). 7.5.5 Small products The rate of freezing of unwrapped small meat products definitely affects the weight loss, with loss increasing as freezing rate decreases. There is also some evidence that it affects losses during cooking and sensory properties. Freezing times of individual meat patties can range from tens of seconds to over an hour in different systems (Hanenian et al., 1989). Everington (2001) has shown that when freezing thin (11mm) burgers high air veloci- ties will substantially reduce freezing times (Fig. 7.4). Freezing in nitrogen and CO 2 can substantially reduce the amount of weight loss from un- wrapped patties when compared with air systems (Table 7.10). However, cooking losses were higher and overall patty quality lower in those frozen in nitrogen. This was mainly due to cracking in the immersion system. When patties are stacked and placed in boxes before freezing, the freez- ing times increase by at least an order of magnitude. Studies carried out on packaged patties looked at freezing rates between 2 and -18 °C ranging from 24 to 96 h (Berry and Leddy, 1989). Before freezing, tenderness scores Freezing of meat 147 Table 7.8 Predicted freezing time from 4 to -7 °C in thermal centre of unwrapped and stockinette wrapped carcasses -30 °C, 4 m s -1 -30 °C, 0.5 m s -1 -20 °C, 0.5 m s -1 Unwrapped 30 kg 5.5 11.0 16.4 40 kg 8.5 15.6 23.4 Wrapped 30 kg 8.0 12.5 19.0 40 kg 12.0 17.8 26.5 Source: Creed & James, 1984. Table 7.9 Freezing time equations for liver in plate freezer Block thickness Equation (cm) 7.6 Y =-0.3547 + 54.4632 R + 0.02138 I 10.2 Y =-0.1917 + 79.9314 R + 0.05203 I 15.2 Y =-0.8020 + 212.119 R + 0.08880 I Source: Creed & James, 1983. measured using a taste panel (8-extremely tender to 1-extremely tough) ranged from 6.6 to 6.1. All the patties were rated as tougher immediately after freezing and after storage for 18 months (Fig. 7.5). Immediately after freezing, patties frozen in 96 h were significantly tougher than those frozen in 24 h, however, the difference was not significant after 18 months storage. Instrumental texture measurements were in general agreement with those from the taste panel (Fig. 7.5). Commercial freezing rates can be very slow. Sausages at the centre of a pallet require 6–7 days to achieve -15 °C from a starting temperature of 7°C (Wanous et al., 1989). However, studies carried out on similar sausages frozen in 9 h, 2.4 and 6.8 days showed no effect of freezing rate on TBA values during frozen storage of 20 weeks. Many small meat products such as cubes and strips of ham and poultry meat, poultry pieces, cooked meat balls, slices of salami and minced meat can be individually quick frozen (IQF) in a rotary cryogenic freezer 148 Meat refrigeration 012345678910112131415 –20 –15 –10 –5 0 5 T emperature ( ° C) Time (min) Air velocity 25 m s –1 Air velocity 15 m s –1 Air velocity 5 m s –1 Fig. 7.4. Freezing 11 mm thick, 80 mm diameter beef burgers in air at -39 °C (source: Everington, 2001). Table 7.10 Freezing of beef patties from 2 to -10 °C in different systems Method Temperature Thickness Freezing Freezing Weight Cooking (°C) (cm) time rate loss loss (s) (cm h -1 ) (%) (%) Air (23 Wm -2 K -1 ) -14 1.10 3939 0.5 2.0 34.4 Air (31 Wm -2 K -1 ) -25 1.12 2047 1.0 1.7 34.6 CO 2 snow -78 1.13 129 16.5 0.1 35.9 N 2 immersion -196 1.13 22 97.3 0.1 37.7 Source: Hanenian et al. 1989. (Thumel and Gamm, 1994).The product is sprayed with a fine mist of nitro- gen to freeze the surface as it enters the drum. As the tilting drum rotates, it transports the food through a contracurrent flow of cold gas that com- pletes the freezing process. 7.6 Tempering and crust freezing There is no exact definition for the word ‘tempering’ in the meat industry. In practice ‘tempering’ can be a process in which the temperature of the product is either raised or lowered to a value that is optimal for the next processing stage. Tempering systems where the temperature of frozen product is raised in temperature are covered in the thawing and tempering chapter. Tempering and crust freezing operations are used to produce the optimum texture in a chilled product so that it is suitable for mechanical processing. In this case, the product is semi-frozen so that it is stiff enough to be sliced, cubed and so on. 7.6.1 Pork loin chopping Loins from lamb and pork are often processed by chopping with a high- speed chopper. Because of the deformation in this process the yield can be reduced. The yield can be increased by first tempering the meat, providing a stiff outer crust, by freezing with liquid nitrogen or a blast of very cold air. However, the process must be carefully controlled; if too much meat is frozen the subsequent chopped meat will have a large increase in the amount of drip formed, resulting in a loss in yield of some 4–5%. Hence, loin freezing processes must always be carefully controlled. Freezing of meat 149 0 1 2 3 4 5 6 7 8 9 10 Panel 0 m Panel 18 m Instrom 0 m Instrom 18 m 24 48 72 96 Freezing time, 2 to –18 °C (h) T exture Fig. 7.5. Texture of patties measured by taste panel (8 = extremely tender, 1 = extremely tough) and Instrom (kg) after 0 and 18 months storage (source: Berry and Leddy, 1989). 7.6.2 High speed ham slicing Traditional production of ham slices consists in cooling formed ham logs in cold rooms to a core temperature of 2 °C, a process that takes between 2 and 7 days (Lammertz and Brixy, 2001). The logs are cut in 1.5 mm thick slices using standard slicers at rates up to 500 slices per minute. New high rate slicers operate at rates up to 1000 slices per minute. To produce high quality slices at this rate the ham logs have to be crust frozen to a temperature of -5 °C at a depth of 7 mm. A number of different cryo- genic freezers have been developed to perform the crust freezing process (Lammertz and Brixy, 2001). 7.6.3 High speed bacon slicing An increasing proportion of bacon is being presliced and packed before it is delivered to wholesalers and retailers. To achieve the throughput required, slicers have to be operated at very high speeds. To maximise the yield of high quality slices from high-speed slicers the bacon has to be sliced in a semi-frozen tempered state. The optimum tempering temperature is a function of the bacon and the slicer. Most bacon tempering has been tra- ditionally carried out in a long single stage process. However, more efficient two-stage processing systems are now common. The correct design and operation of such systems is critical to the cost effectiveness of the slicing process. Bacon temperature is the critical parameter in a high-speed (typically 800–1400 slices per minute) slicing operation. This operation has more in common with the guillotining of metal than the slow speed slicing normally carried out in a butcher’s shop. The bacon must be presented to the blade in a rigid semi-frozen state to minimise distortion and break-up on cutting. Obtaining the correct temperature throughout the bacon middle is crucial for a high yield of undamaged slices (James and Bailey, 1987b). High-speed photography has been used to demonstrate clearly the effect of incorrect slicing temperature. When the temperature was too low the hard bacon shattered and blade wear was excessive; when too high the soft bacon stuck to the blade and the fat was torn away from the lean. The optimum tem- perature for a particular operation depends on the salt content of the bacon, the maturation time, the type of slicer being used and the slicing speed. Experiments carried out in the 1920s showed that there was a near linear relationship between the initial freezing point of lean pork and its salt content (expressed in grams of salt per 100 g of water in the meat). Conse- quently the initial freezing point, ice content at any temperature and the related slicing temperature depend upon both the salt and water content of the bacon. Work has shown that even using very carefully controlled curing methods there are still considerable variations in salt content within individual slices, between slices and between bacon sides in the same batch. This makes it difficult, and in a commercial situation impossible, to carry 150 Meat refrigeration out analytical tests that will define the optimum slicing temperature for a particular operation. This temperature must therefore be determined experimentally for each slicing operation, and tempering systems devel- oped that will produce a uniform temperature throughout the product in the most efficient way. 7.6.3.1 Determination of slicing temperature There is likely to be substantial variability in the bacon input to a slicing operation. A survey (James and Bailey, 1987b) found that mean percentage salt and water content of bacon supplied to a large slicing plant over a two- year period varied by 1.9 and 2.6%, respectively (Table 7.11). The effect on slicing of the variation in salt content, where the overall mean was 4.34%, was much greater than water, which had an overall mean of 70.8%. Maximum and minimum values from different suppliers were 7.1 and 2.1% for salt, and 74.8 and 63.9% for water. Initial freezing points, and conse- quently the optimum slicing temperatures could therefore vary by 5 °C or more. Examination of freezing curves of bacon from three of the suppliers showed initial freezing points of -3, -3.5 and -6°C. Currently the only method available for determining the optimum bacon temperature for a slicing operation is to carry out slicing trials at different bacon temperatures. Results from such a trial where the yield in each quality grade was determined are shown in Table 7.12. In the specific trial taking into account the relative quantities from each supplier, the best slicing temperature was found to be -9.5°C. 7.6.3.2 Tempering systems A small number of operations use plate freezer, liquid immersion systems and cryogenic tunnels to temper bacon for high speed slicing. However, the majority of industrial systems employ air in a single or two-stage process. Freezing of meat 151 Table 7.11 Salt and moisture contents in bacon middles from different suppliers Supplier Salt (g/100 g water) Moisture (%) no. Maximum Minimum Mean Maximum Minimum Mean 1 7.1 2.9 5.1 73.5 65.3 69.6 2 6.5 2.7 4.1 74.0 68.0 71.1 3 4.7 2.9 4.0 72.5 68.6 71.0 4 5.9 2.1 4.1 74.6 65.0 71.8 5 6.3 2.2 4.2 74.2 63.9 70.9 6 5.9 2.5 5.0 72.6 64.3 70.3 7 6.5 3.1 4.9 74.4 67.5 71.1 8 4.6 2.2 3.2 74.2 70.0 72.2 9 6.3 2.6 4.7 74.7 66.3 69.7 10 5.2 2.7 4.1 74.8 65.3 70.3 Source: James and Bailey, 1987b. 7.6.3.2.1 Single-stage tempering Single-stage tempering is a very simple process.The bacon middle, back and streaky joints are placed on the shelves of trolleys. The trolleys are then wheeled into a room operating at the desired slicing temperature. The bacon remains in the room until its temperature equalises to that of the room. It is then pressed and sliced. For example, in a commercial single stage tempering system, backs were held for 18 h in rooms operating at -9 to -14°C, 0.2–1.2ms -1 with an average product weight loss of 1.18%. At weekends the backs remained in the rooms for a total of 64h and the average weight loss increased to 1.88%. After 18 h the surface temperature had reached -10 °C but the centre was still above -7°C. A number of problems are inherent in a single-stage tempering opera- tion. Equalisation times are long, after 18 h there can still be a 3°C differ- ential across the backs. Using a two-stage system for only half this time resulted in differentials of less than 1 °C. The most obvious drawback of single-stage tempering is that to obtain the same throughput systems have to be far larger, probably by at least three-fold. It is also more difficult to obtain even air distribution and good temperature control in a large room. This problem is exacerbated in that the single-stage system has to fulfil con- flicting roles. To remove heat from the bacon a reasonable air/product tem- perature difference and reasonable air movement are required. In contrast, towards the end of the process when all the required heat has been extracted, a very small temperature differential and minimum air move- ment are desirable to attain an even temperature and a reduced rate of weight loss. 7.6.3.2.2 Two-stage tempering In a two-stage tempering process, an initial blast freezing operation is fol- lowed by a separate period of temperature equalisation. It is critical that 152 Meat refrigeration Table 7.12 Yield (%) of slices in quality classes from bacon backs from different suppliers sliced at an equalised bacon temperature of -6.5, -7.5 or -9.5 °C Quality Prime/second Thrifty/catering Bits and pieces Slicing -6.5 -7.5 -9.5 -6.5 -7.5 -9.5 -6.5 -7.5 -9.5 temperature (°C) Supplier A – 93.6 94.7 – 3.7 3.0 – 2.7 3.3 Supplier B 79.1 82.0 81.7 12.5 8.3 8.8 8.11 9.7 9.5 Supplier C – 79.9 75.7 – 10.2 9.7 – 9.9 14.6 Supplier D 84.1 90.8 93.8 10.8 6.0 2.8 5.1 3.2 3.4 Supplier E – 81.5 78.8 – 11.8 9.6 – 6.7 11.6 Supplier F – 76.4 78.5 – 10.4 10.3 – 13.2 11.2 Source: James and Bailey, 1987b. class the desired amount of heat is extracted in the initial blast freezing operation. Examples of conditions and final equalised temperatures are given in Table 7.13.Typical temperature histories at the surface and thermal centre of backs tempered at -30°C,3.0ms -1 and -35°C,1.0ms -1 are shown in Fig.7.6.Surface temperatures tended to be a few degrees lower after 3 h at -35 °C than at -30 °C, whilst centre temperatures were very similar, ca, -6 °C. In each case the maximum temperature difference across the backs was less than 1°C after 3 h in the equalisation room and the temperature within any part of the back was within 1 °C of the room temperature after 7 h. It is critical that the refrigeration system is sized to extract the required amount of heat from the bacon. The energy released per kilogram of bacon in each half-hour period during one experiment varied by a factor of 2.7 from 0.0139 to 0.0051 kWh kg -1 (Fig. 7.7). The mean total energy extracted per kilogram of bacon in the 3 h operation was 0.0535 kWh kg -1 . In a two-stage system there are several practical considerations. In one study a 3 h blast freeze operation at -35°C, 1.0ms -1 obtained the desired Freezing of meat 153 Table 7.13 Conditions in blast freezer, mean initial weights of bacon backs, % weight losses after 3 h and equilibrium temperatures Air conditions Weight Equilibrium No. samples Temperature Speed Initial 3h loss temperature (°C) (m s -1 ) (kg) (%) (°C) -30 1.0 5.640 (0.51) 0.71 (0.06) -8.6 (0.3) 6 -30 3.0 5.150 (0.32) 0.76 (0.08) -11.6 (0.4) 4 -35 0.5 5.288 (0.13) 0.66 (0.06) -8.5 (0.4) 6 -35 1.0 5.375 (0.26) 0.55 (0.07) -9.9 (0.8) 4 () = standard deviation. Source: James and Bailey, 1987b. 10 5 0 –5 –10 –15 –20 –25 –30 T e mperature ( ° C) Time (h) 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6.5 7.5678 Centre –30 °C, 3m/s Surface –30 °C, 3m/s Surface –35 °C, 1m/s Centre –35 °C, 1m/s Fig. 7.6. Surface and centre temperature in bacon backs during two-stage tem- pering operation (source: James and Bailey, 1987b). equilibrium temperature of -9.5 °C and achieved the lowest weight loss of 0.55%. However, these conditions were very critical and would not allow for a pull down period after loading, or the use of slightly thicker backs or bacon with a higher salt content. Adding heaters to the equalisation room to provide exact temperature control with slightly negative product load was considered more viable than trying to control the refrigeration against the positive load likely in practice. Air distribution and control would also be less exact at 3 m s -1 , and a variation of ±0.5ms -1 in air velocity over the bacon backs would have far less effect on the final equalised temperature than a similar variation about a mean of 1.0 m s -1 . Operating conditions of -30°C, 3.0ms -1 were therefore chosen for this particular industrial plant, because they were less critical and provided a degree of flexibility. Investigations have shown clearly the need to size the refrigeration system to meet the initial rate of heat release from the warm bacon backs (James, 1997). In trials an experimental freezer was unable to maintain the desired set point of -30 °C but rose to -27 °C immediately after loading and took 1.5 h to recover fully. The average equalised temperature in this trial was 0.6 °C higher than in two successive trials where less bacon was used and the freezer reached -30 °C within minutes of loading. Although the average rate of heat release from the bacon backs during the freezing op- eration was 0.0175 kWkg -1 the refrigeration plant had to have twice this capacity to meet the rate of release during the first half hour. In the indus- trial situation heat ingress through the open door during loading and the considerable cooling requirement of the supporting racks also has to be taken into consideration. One practical solution is a central refrigeration 154 Meat refrigeration 0.016 0.014 0.012 0.01 0.008 0.006 0.004 0.002 0 0–0.5 1–1.5 2–2.51.5–22.5–30.5–1 Time (h) Energy (kW h/kg – 1 ) Fig. 7.7. Energy released (kW h kg -1 ) by bacon backs over 0.5 h intervals during blast freezing at -30 °C, 3.0 m s -1 (source: James and Bailey, 1987b). plant serving a number of separate freezing chambers. These can be loaded and unloaded in a sequence to provide the plant with a nearly constant refrigeration load enabling it to operate at optimum efficiency. 7.7 Conclusions 1 Under commercial conditions differences in freezing rates are unlikely to produce noticeable changes in the organoleptic quality of the meat produced. However, current legislation requires a minimum meat temperature of -12 °C to be achieved before meat is moved from the freezing system. Freezing time is therefore of considerable economic importance. 2 Most unprocessed meat is either frozen in batch air systems as bone in carcasses, sides or quarters, or boned out in cartons. Freezing times in such systems are typically 25–72 h. Some offal is frozen in plate freezers. 3 Small processed items are typically frozen in continuous belt freezers or in cryogenic tunnels. 4 Crust freezing and tempering are increasingly being used to allow high speed mechanical portioning or slicing of meat and meat products. The final temperature distribution produced by the freezing system is critical in such operations. 7.8 References a?ón m c and calvelo a (1980), Freezing rate effects on the drip loss of frozen beef, Meat Sci, 4 1–14. berry b w and leddy k f (1989), Effects of freezing rate, frozen storage tempera- ture and storage time on tenderness values of beef patties, J Food Sci, 54(2) 291–296. bevilacqua a, zaritzky n e and cavelo a (1979), Histological measurements of ice in frozen beef, J Food Technol, 14 237–251. buchmuller j (1986), Chilling and freezing meat products with liquid nitrogen, Fleischwirtschaft, 66(4) 568–570. castell-perez m e, heldman d r and steffe j f (1989), Computer simulation of microbial growth during freezing and frozen food storage, J Food Process Eng, 10 249–268. creed p g and james s j (1981), A survey of commercial meat block freezing in the United Kingdom and Eire, Internat J Refrigeration, 4 348–354. creed p g and james s j (1983), The freezing times of liver in a vertical plate freezer, Progress in Refrigeration Science and Technology, Proceedings of the XVIth Inter- national Congress of Refrigeration, Tome III, Paris (France), 4 145–151. creed p g and james s j (1984), The prediction of freezing and thawing times of mutton carcasses, Proceedings of the 30th European Meeting of Meat Research Workers, Bristol, Paper 2 : 5, 59–60. creed p g and james s j (1985), Heat transfer during the freezing of liver in a plate freezer, J Food Sci, 50 285–288, 294. Freezing of meat 155 de jong r s (1994), Plate freezing technology and economics applied to the meat industry, Proceedings IRHACE Technology Conference, Paper 11. dobryschi j, pietrzak e and hoser a (1977), Rheological characteristics of fresh and frozen chicken muscles and their relation to tenderness measured by sensory methods, Acta Aliment, 6(2) 107–111. everington d w (2001), Development of equipment for rapid freezing, Proceedings of the International Institute of Refrigeration Rapid Cooling – above and below zero, Bristol. fleming a k, chadderton t, amos n d and cleland a c (1996), Non-traditional refrigeration technologies for the New Zealand meat industry, MIRINZ Publica- tion 961. grujic r, petrovic l, pikula b and amidzic l (1993), Definition of the optimum freez- ing rate – 1, investigation of structure and ultrastructure of beef M. longissimus dorsi frozen at different freezing rates, Meat Sci, 33 301–318. hanenian r, mittal g s and usborne w r (1989), Effects of pre-chilling, freezing rate, and storage time on beef patty quality, J Food Sci, 54(3) 532–535. hill m a and glew g (1973), Organoleptic assessment of products frozen by two methods: Liquid nitrogen and blast freezing, J Food Technol 8 205–210. jackobsson b and bengtson n (1973), Freezing of raw beef: Influence of ageing, freezing rate and cooking method on quality and yield, J Food Sci, 38 560– 565. james s j (1997), Bacon tempering for high speed slicing, in Proceedings of the Meat Refrigeration – Why and How? EU concerted action programme CT94 1881, Langford, UK. james s j and bailey c (1987a), The freezing of beef quarters, Proceedings XVIIth International Congress of Refrigeration C, Vienna (Austria), C2–4. james s j and bailey c (1987b), Bacon tempering for high speed slicing, Proceedings XVIIth International Congress of Refrigeration C, Vienna (Austria), C2–1. james s j, creed p g and bailey c (1979), The determination of the freezing time of boxed meat blocks, The Proceedings of the Institute of Refrigeration, Volume 75 74–83. james s j, nair c and bailey c (1983), The effects of freezing and thawing rates on the drip loss from frozen beef, in Zeuthen P et al., Thermal Processing and the Quality of Foods, first edition, London and New York, Elsevier Applied Science, 807–813. lammertz m and brixy n (2001), Continuous process and production improvements by application of refrigeration with cryogenic gases, Proceedings of the Interna- tional Institute of Refrigeration Rapid Cooling – above and below zero, Bristol. lampitt l h and moran t (1933),The palatability of rapidly frozen meat, J Soc Chem Ind, L11.21, 143t–146t. mascheroni r h (1985), Generalisation of a method for the characterization of quick frozen beef, Meat Sci, 13 81–89. mittal g s and barbut s (1991), Effects of freezing rate and storage time on the structural properties of minced meat, Lebensmittel-Wissenschaft und-Technologie, 24 226–230. newman m (2001), Cryogenic impingement freezing utilizing atomized liquid nitro- gen for the rapid freezing of food products, Proceedings of the International Insti- tute of Refrigeration Rapid Cooling – above and below zero, Bristol. petrovic l, grujic r and petrovic m (1993), Definition of the optimal freezing rate – 2 Investigations of the physico-chemical properties of beef M. longissimus dorsi frozen at different freezing rates, Meat Sci, 33 319–331. sacks b, casey n h, boshof e and van zyl h (1993), Influence of freezing method on thaw drip and protein loss of low-voltage electrically stimulated and non- stimulated sheep’s muscle, Meat Sci, 34 235–243. 156 Meat refrigeration sakata r, oshida t, morita h and nagata y (1995), Physico-chemical and processing quality of porcine M. longissimus dorsi frozen at different temperatures, Meat Sci, 39 277–284. sebranek j g (1980), Cryogenic freezing of ground beef patties shows superior organoleptic effects, Quick Freezing, August, 50–53. sebranek j g, sang p n, rust r e, topei d g and kraft a a (1978), Influence of liquid nitrogen, liquid carbon dioxide and mechanical freezing on sensory properties of ground beef patties, J Food Sci, 43 842–844. sundsten s, andersson a and tornberg e (2001), The effect of the freezing rate on the quality of hamburgers, Proceedings of the International Institute of Refrigera- tion Rapid Cooling – above and below zero, Bristol. thumel h and gamm d (1994), Loose rotary freezing, Fleischwirtschaft, 74(1) 64–65. wagner j r and a?ón m c (1985), Effect of freezing rate on the denaturation of myofibrillar proteins, J Food Technol, 20 735–744. wanous m p, olson d g and kraft a a (1989), Pallet location and freezing rate effect on the oxidation of lipids and myoglobin in commercial fresh pork sausage, J Food Sci, 54(3) 549–552. Freezing of meat 157