15 Freezing J. M. Fletcher, Unilever R & D Colworth 15.1 Introduction The modern frozen food industry was started by Clarence Birdseye in America in 1925. As a fur trader in Labrador Birdseye had noticed that fillets of fish left by the natives to freeze rapidly in arctic winters retained the taste and texture attrib- utes of fresh fish better than fillets frozen in milder temperatures at other times of the year. Frozen foods were available before Birdseye’s pioneering innovations, but they were of poor and uncertain quality. Birdseye’s insight was that speed of freezing is crucial to retain quality and he was the first to develop machinery that could freeze foods rapidly on an industrial scale. Quick freezing allowed the trans- port of produce over long distances and the year-round consumption of seasonal produce that was of very superior quality compared with alternative preservation methods such as canning and drying. Although Birdseye was probably unaware of this particular advantage, quick freezing, if combined with appropriate treatments prior to freezing, also has the potential to ensure excellent preservation of nutri- tional value for a wide range of foods. In the context of the nutritional value of vegetables and fruits, the US Food and Drug Administration has recently (1998) approved frozen produce to be labelled as healthy. Based on presented data the Food and Drug Administration concluded that ‘. . . because frozen fruits or veg- etable products are nutritionally comparable to the raw versions, they would likely have the same inherent beneficial effects as the raw version’. In the years since 1925 the application of freezing has become globally an important aspect of food processing technology. In the year 2000 total world wide sales of frozen foods (excluding ice cream) was estimated as 13.6 million tonnes with a retail value of US$ 58.5 billion (Euromonitor). The process of quick freez- ing was first applied to a limited range of fish, meat, fruits and vegetables; today, in addition to these still important basics there is a very wide range of processed foods, meal components and whole meals available in the frozen format. Cur- rently, the sectors of red meat, poultry, fish/seafood and vegetables make up approximately 10% each of the total tonnage of the total frozen food market as well as frozen potatoes at 15% and ready meals at 20% (Euromonitor). As might be expected, there are considerable geographical differences between regions and countries in usage of frozen food. Whereas in the US and in Europe approxi- mately 13 kg and 10 kg of frozen food are consumed per capita per year, in Africa and Asia the amounts consumed are only 0.3 kg and 0.9 kg respectively. In the future it is anticipated that freezing as a processing option will take an increas- ing share of the food market. In both developed and undeveloped nations the increased demand for frozen foods will come from consumers’ wishes for high convenience, high organoleptic quality and high nutritional value. Although freezing on its own has a negligible impact on nutrient levels in food, the associated pre-freezing processing, storage in the frozen state and structural damage evident in some thawed frozen foods may have significant detrimental effects. The early literature describing the effects of freezing and associated pro- cessing on nutrient content and nutritional value has been reviewed by Bender in 1978, and more recently in 1993. This chapter will summarise the key principles, review newer findings and highlight areas of continuing uncertainty in assessing the nutritional impact of freezing. 15.2 Change and stability in frozen foods The defining step in freezing is the removal of heat. This lowers the temperature of foods so that microbial and chemical changes are prevented or minimised. By storing in the frozen state it is possible to prolong greatly the length of time that many foods may be maintained with an excellent sensory and nutritional value. It is, however, important to realise that at the typical temperatures used for indus- trial and domestic storage of frozen foods (typically -24°C and -18°C respec- tively), chemical reactions that can lead to a reduction of quality and nutrient loss may continue to occur. Many of these reactions take place in solution and even at -24°C, natural foods such as fruit, vegetables and meats may still contain 2–5% of their total water content in the liquid phase. As the temperature of natural foods is reduced below 0°C ice crystals begin to form and the solutes present in intra- and extra-cellular fluids become more concentrated in the remaining liquid water, thereby lowering the freezing point of this water. Therefore, although the rates of most reactions will be substantially reduced by the lower temperature of frozen foods, the increased solute concentration may to some extent counteract this effect. Another effect of the increased solute concentration is to move water by osmosis between compartments. The formation of ice may also rupture cell struc- tures causing mixing and reactions between components previously held apart. The complex nature of the changes that take place when foods are frozen makes it difficult to predict effects on quality and stability. 332 The nutrition handbook for food processors Probably the most important reaction leading to both quality and nutrient losses in frozen foods is oxidation. The consequences of oxidative instability are the key factors that limit the storage life of frozen foods. Just as in foods kept at more normal ambient temperatures, unless they are stored in a vacuum, or in an inert gas, atmospheric oxygen will diffuse through frozen food and may react with many of the soluble and insoluble components. One consequence of oxida- tion on sensory quality is the development of ‘off flavours’ and rancidity, usually caused by oxidative breakdown of membrane and storage lipids (Erickson, 1997). Other adverse consequences of oxidation may include colour loss and/or change, and in fish and meat foods a toughening of muscle structures. Although macro- molecular components such as carbohydrates and protein may undergo limited oxidation, any influence on nutritional value is likely to be small. However, several vitamins such as ascorbate and folates are particularly susceptible to oxidative damage. A feature of the quick freezing of foods is the formation of a large number of relatively small ice crystals that cause minimal damage to cellular and tissue structures but on prolonged frozen storage, and particularly in conditions where temperatures fluctuate, crystals of ice grow in size. Although at any temperature below 0°C, the total amount of ice in a food will remain constant, large crystals grow instead of a larger number of smaller crystals, a process known as Ostwald ripening. The growth of larger ice crystals may break delicate food structures and compress others. On thawing of frozen foods these changes may have serious effects on texture leading to poor sensory quality; vegetables and fruits may lose their characteristic crispness and meat or fish may become tougher and drier. An adverse consequence for nutritional value is the reduced water-holding capacity of structurally damaged foods, leading to increased ‘drip loss’. Significant amounts of water-soluble nutrients may be discarded if this drip loss is not incor- porated into the food to be consumed. 15.3 Vegetables and fruits There are several factors that potentially contribute to differences in nutrient levels between vegetables and fruits in the frozen format and those supplied as fresh or preserved by other processes. Any differences are likely to be in the loss and preservation of vitamins; it has been shown that compared with fresh veg- etables, there are negligible differences between the mineral and fibre contents of equivalent frozen vegetables (Polo et al, 1992; Nyman, 1995). 15.3.1 Selection of cultivar and time of harvesting Particular cultivars and harvest times are chosen to optimise sensory quality and these may differ between those selected for freezing and those that are consumed in fresh, canned or dried formats. The cultivar and harvest time may have some effects on nutritional value (Shewfelt, 1990); for example peas selected for Freezing 333 canning are usually harvested at a more mature stage than those selected for freez- ing and consequently have approximately 10% lower ascorbate concentration. The type of cultivar may also influence the amount of nutrient lost during pro- cessing, reflecting differences between culitvars in morphology and mechanical strength. 15.3.2 Storage after harvest Many vegetables, and to a lesser extent fruits, are relatively unstable after har- vesting and undergo rapid chemical changes that result in significantly reduced levels of some nutrients. For example, concentrations of ascorbate in spinach may fall to 50% of their initial, pre-harvest level, after two days of storage as shown in Fig. 15.1 (Favell, 1998). The magnitude of nutrient losses during storage prior to freezing is highly variable and depends on the crop, the method of harvesting and the duration and conditions of storage. To preserve the nutritional value of fresh vegetables and fruits it is clearly desirable to minimise the time in blanch- ing and freezing and to cause minimal mechanical damage. 15.3.3 Washing and blanching The need for washing of vegetables and fruits may cause some loss of water- soluble nutrients, particularly from cut surfaces. As noted above, oxidation is a key factor influencing stability in the frozen state and this is particularly a concern with vegetables and fruits because they contain many enzyme systems that give rise to reactive oxygen species. It is to prevent enzyme-mediated oxidation reac- tions that most vegetables and fruits are blanched before freezing. Another reason 334 The nutrition handbook for food processors 0 50 100 071421 Time since harvest (days) Ascorbate (% retention) Ambient Chill Frozen Fig. 15.1 Effects of storage and freezing on ascorbate retention in spinach: typical values for retention of ascorbate in spinach stored at either ambient or chill temperature (4°C) compared with blanched and frozen spinach. All samples were taken from the same field and time zero levels were obtained from freshly harvested spinach. Blanching and freez- ing were carried out in a commercial factory. (from Favell, 1998) is to ensure microbiological safety but this can be achieved by other means. The advantages of blanching can be illustrated with reference to cauliflower and spinach. If they are frozen without blanching they become unpalatable after only a few months due to the development of ‘off’ flavours and odours caused pri- marily by oxidation of membrane lipids. If these vegetables are blanched before freezing they have a storage life of 18–24 months. Commercial blanching con- ditions typically involve heating in water or steam at 95–100°C for 3–10 minutes, depending on the type and size of material to be blanched. The conditions are chosen so as to ensure inactivation of the enzymes responsible for oxidation. During blanching, nutrients may be lost by leaching and by chemical degrada- tion. A great deal of information has been published on losses of labile nutrients during blanching (for review see Clydesdale et al, 1991). Ascorbate is often used as an indicator of potential nutrient loss because of its high solubility, sensitivity to heat and ease of measurement. Typical losses of ascorbate from vegetables during blanching are of the order 5–40% (Favell, 1998; Bender, 1993). In general, it may be concluded that nutrient losses are minimised if the raw material is as little damaged as possible during handling and if processing conditions are chosen that keep the temperature, duration of heat exposure and product to water ratio as low as is consistent with denaturing the enzymes responsible for oxidative spoilage. 15.3.4 Frozen storage Bender (1993) has summarised the contradictory results of published studies designed to estimate the magnitude of vitamin loss during frozen storage of veg- etables and fruits. Even for a particular vegetable, processed and stored under apparently similar conditions, the extent of ascorbate loss has been reported as negligible, or up to 40% after a year of frozen storage (Bender, 1993). As Bender comments, there are many possible sources of experimental variation that may lead to these different conclusions, most notably incomplete denaturation of oxidative enzymes during blanching. Since the review by Bender no large scale systematic study addressing this issue has been published. It may be concluded that if vegetables and fruits are adequately blanched and stored at conven- tional freezer temperatures without undue temperature fluctuations they will still possess valuable levels of potentially labile nutrients for a period of at least 12–18 months. 15.3.5 Cooking When comparing the nutritional value of different processing methods it is also necessary to consider the ways in which consumers handle these different prod- ucts. Cooking methods may have important effects on the quantity of nutrients within a food. Because frozen vegetables have already been blanched, they require less cooking time than fresh vegetables to reach the same levels of palat- ability. This means that while frozen vegetables may have lost some nutrients during blanching they will probably suffer reduced losses during cooking. Freezing 335 It is increasingly recognised that regular consumption of vegetables and fruits significantly reduces the risk of some cancers and of cardiovascular disease. Although it is by no means certain, it appears likely that these beneficial effects are not just a consequence of consuming the recognised nutrients found in veg- etables and fruits. Alarge number of potentially beneficial compounds, the so called phyto-nutrients, or non-nutrient phyto-chemicals are found in vegetables and fruits. It is not yet clear which particular compounds, or even which group of phy- tochemicals may be responsible for the health benefits, but if and when the pro- tective agents are identified it will be necessary to ascertain the effects of freezing and associated processes on their retention in frozen vegetables and fruits. 15.4 Meat and fish Quick freezing is extensively used to preserve a wide range of raw and cooked meat and fish. Freezing and frozen storage does not significantly affect the nutri- tional value of meat and fish proteins. However, as pointed out above, on thawing frozen meat and fish substantial amounts of intra- and extra-cellular fluids and their associated water-soluble proteins and other nutrients may be lost (‘drip- loss’). The volume of drip-loss on thawing of meat and fish is highly variable, usually of the order of 2–10% of wet weight but in exceptional circumstances up to 15% of the weight of the product may be lost. Many factors influence the amount of drip loss and not all are related to freezing: ? Variables influencing the raw material – The age, the species and the variety of the animal may have important effects. Additional factors may include the diet fed to the animal, the method of slaughter, and the pre- and post- slaughter handling. ? Water-binding chemicals – A variety of chemicals are often used as additives to meat and fish before freezing, e.g. polyphosphates. These chemicals pen- etrate muscle fibres and they associate with proteins where they serve to protect the texture and succulence of meat and fish and to reduce drip loss. ? Freezing and frozen storage – The rate of freezing, the temperature of frozen storage and temperature fluctuations during storage. ? Thawing – The rate of thawing from the frozen state and the holding tem- perature before cooking. In the frozen state meat and fish are generally less susceptible to oxidative spoilage than are vegetables and fruits and they are not subjected to the equiva- lent of blanching. On prolonged storage, however, oxidation may lead to signifi- cant chemical changes and loss of labile vitamins. The poly-unsaturated fatty acids in meat and fish are particularly susceptible to oxidation. As with vegeta- bles and fruits, it is the products of fatty acid oxidation that give rise to charac- teristic ‘off’ and rancid flavours and aromas. The recommended storage lives of frozen meat and fish products are chosen to be within the period before ‘off’ and 336 The nutrition handbook for food processors rancid flavours and aroma are detectable. In general, those meat and fish prod- ucts that contain a larger amount of poly-unsaturated fatty acids are least stable and have shorter storage lives. For example, oily fish have a typical frozen shelf life in the region of 6–9 months at -18°C whereas white fish have a frozen shelf life of 12–24 months. Equivalent cuts of pork and beef have frozen shelf lives of 10–12 months and 18–24 months respectively (International Institute of Refrigeration, 1986). A particular nutritional advantage of fish, and especially of oily fish, is as a dietary source of long chain n-3 poly-unsaturated fatty acids (docosahexanoic acid and eicosapentanoic acid; DHA and EPA respectively). Intake of these fatty acids has been implicated in many health benefits and as noted above they are particularly susceptible to oxidation. Several recent studies have been carried out to determine the effects of freezing and frozen storage on their levels in fish. A significant reduction in the total n-3 PUFA content was reported in saithe (a lean fish) fillets stored at -20°C for six months (Dulavik et al, 1998). Similarly, levels of total n-3 PUFA were reduced in salmon fillets stored at -20°C (Refsgaard et al, 1998) and levels of DHA and EPA were reduced in sardine and mackerel fillets stored for 24 months (Rougerou and Person, 1991). In contrast to these reports of PUFA loss, Polvi et al (1991) found no difference in total n-3 PUFA levels when salmon fillets were stored at the relatively high temperature of -12°C for three months. Xing et al (1993) also failed to see any losses of DHA and EPA in mackerel and cod fillets stored frozen at -20°C for 28 weeks. As with many aspects of nutrient stability, the extent of n-3 PUFA loss from frozen fish by oxidation will depend on several factors, e.g. access of oxygen to the muscle, handling before freezing and the type of muscle (dark fish muscle suffers higher rates of iron-catalysed oxidation than does white muscle). Although loss of nutritionally important n-3 PUFAs from frozen fish may undoubtedly occur on prolonged frozen storage, in practice this is not likely to be a serious cause for concern. The threshold for sensory detection of rancidity is very low and therefore if frozen fish are consumed within the recommended period of storage, significant proportions of their original content of n-3 PUFAs will not have been lost to oxidation. 15.5 Nutritional implications of new developments in freezing In considering the introduction of new developments in the freezing of foods and in associated technologies it is clear that they are unlikely to be driven solely by the motivation to improve nutritional value. If processed according to current good practice and consumed within their recommended storage lives, frozen foods already often have a nutritional value equivalent to foods available as fresh in the retail supply chain. Nevertheless, new developments designed to improve the organoleptic properties of frozen foods or to reduce the costs of production may have significance for nutrient retention. Freezing 337 15.5.1 Developments in blanching of vegetables and fruits More rapid blanching of vegetables and fruits, and alternatives that do not use hot water immersion would be expected to preserve labile nutrients from leach- ing and chemical destruction. Alternative heating systems have been developed, such as those using steam and microwaves. However, as pointed out by Bender (1993), consistent evidence for nutritional benefits from these alternative blanch- ing procedures has not been observed. Part of the reason lies in the inherent vari- ability in plant raw materials. For example, ascorbate levels may differ by as much as two-fold in freshly harvested vegetables and the improved ascorbate retention to be achieved by alternative methods to conventional blanching may be only within the order of 5–10%. 15.5.2 Frozen storage in the glassy state As pointed out above, natural foods stored at -18°C to -24°C contain significant amounts of liquid water in which reactions leading to quality and nutrient loss may occur. If the temperature of foods is further lowered, the remaining liquid eventually enters a so-called ‘glassy state’, i.e. a non-crystalline solid (for reviews, see Levine and Slade, 1989; Goff, 1997). In this state, rates of reaction, including enzyme mediated reactions become insignificant or greatly reduced. This gives rise to the possibility of storing frozen foods for longer periods than currently used without the risk of significant oxidation. There is also the pos- sibility of freezing vegetables and fruits without the need for blanching and suffering the associated nutrient losses. The effects on ascorbate retention of stor- ing unblanched peas at different temperatures compared with conventionally blanched and frozen peas are shown in Fig. 15.2. The temperature at which peas 338 The nutrition handbook for food processors 50 60 70 80 90 100 110 036912 Time of storage (months) Ascorbate (% retention) -55°C -38°C -24°C -20°C Blanched Fig. 15.2 Effects of frozen storage temperature on ascorbate retention of peas: Ascorbate retention in unblanched peas stored frozen at different temperatures compared with commercially blanched and frozen peas stored at -24°C. (from Sharp, unpublished) are estimated to be in the glassy state is approximately -30°C and below this temperature they do not lose significant amounts of ascorbate. The temperature at which foods enter the glassy state varies and depends on the type and con- centration of molecules in solution. Generally, the glassy state transition tempera- tures for foods are well below those used in the commercial supply chain and the costs entailed in modification of freezer operation would delay widespread uptake of this procedure. 15.5.3 Use of anti-freeze peptides Anti-freeze peptides (AFP) are a class of compound that both depress the freez- ing point of water and prevent ice crystal enlargement during frozen storage (Lillford and Holt, 1994; Griffith and Ewart, 1995). If incorporated into frozen foods they may potentially prevent the structural and mechanical damage caused by ice crystal enlargement, thereby improving the sensory properties of food and potentially reducing drip loss from frozen food when it is thawed. This is illus- trated by the finding that fish naturally containing AFPs suffer a lower amount of drip loss on freezing and thawing than those without such peptides (Payne and Wilson, 1994). Widespread applications of AFPs in frozen foods are currently limited by their cost and the need to produce them on any commercially relevant scale by using biotechnology. 15.6 Sources of further information and advice 15.6.1 Literature ? For an extensive review of the effects of freezing on the chemical and physi- cal properties of foods see Low temperature preservation of foods and living matter (1973), edited by Fennema OR, Powrie WD and Marth EH, published by Marcel Dekker, New York. ? For details of industry standards and procedures relating to frozen food see Recommendations for the Processing and Handling of Frozen Foods (1986), published by the International Institute of Refrigeration, Paris. ? For a description of the effects of blanching, freezing and other processing steps on the nutritional value of individual vegetables see Handbook of veg- etable science and technology (1998), edited by Salunkhe DK and Kadam SS, published by Marcel Dekker, New York. ? For a summary of modern frozen food theory and practice see Maximising quality and stability of frozen foods (1999), edited by Kennedy CJ and Archer GP, published by the EU Concerted Action CT96–1180. ? For more comprehensive reviews of frozen food theory and practice (includ- ing a review on nutritional aspects by Bender 1993, see references) see ‘Frozen Foods Technology’ (1993) edited by Mallett CP and published by Blackie Academic and Professional and ‘Quality in Frozen Food’ (1997), edited by Erickson MC and Hung Y-C, published by Chapman & Hall. Freezing 339 15.6.2 Trade organisations Below are listed the trade organisations that are sources of general information on frozen food and the frozen food industry: ? The British Frozen Food Federation at 3rd Floor, Springfield House, Springfield Business Park, Springfield Road, Grantham, Lincolnshire, NG31 7BG. Email on http://www.bfff.co.uk ? The (US based) National Frozen and Refrigerated Foods Association, at 4755 Linglestown Rd., Suite 300, P.O. Box 6069, Harrisburg, PA 17112. Email on http://www.nfraweb.org ? The American Frozen Food Institute at 2000 Corporate Ridge, Suite 1000, McLean, Virginia 22102. Email on http://info@affi.com 15.7 References bender a e (1978), Food Processing and Nutrition, Academic Press bender a e (1993), ‘Nutritional aspects of frozen foods,’ in Frozen Food Technology, ed Mallett CP, Blackie Academic and Professional, 123–40 clydesdale f m, hoct, leecy, mondy n i and shewfelt r l (1991), ‘Effects of post- harvest treatment and chemical interaction on the bioavailability of ascorbic acid, thiamine, vitamin A carotenoids and other minerals,’ Critical reviews in Food Science and Nutrition 30, 599–638 dulavik b, sorensen n k, barstad h, horvli o and olsen r l (1998), ‘Oxidative sta- bility of frozen light and dark muscles of saithe (Pollachius virens),’ Journal of Food Lipids 5, 233–45 erickson m c (1997), ‘Lipid oxidation: flavour and nutritional quality deterioration in frozen foods,’ in Quality in Frozen Food, eds Erickson MC and Hung Y-C. Chapman & Hall, 141–73 favell d j (1998), ‘A comparison of the vitamin C content of fresh and frozen vegeta- bles,’ Food Chemistry 62, 59–64 goff h d (1997), ‘Measurement and interpretation of the glass transition in frozen foods,’ in Quality in Frozen Food, eds Erickson MC and Hung Y-C, Chapman & Hall, 29–50 griffith m and ewart k v (1995), ‘Antifreeze proteins and their potential use in frozen foods,’ Biotechnol. Adv., 13, 375–402 International Institute of Refrigeration (1986), Recommendations for the Processing and Handling of Frozen Foods, Paris. levine h and slade l (1989), ‘A food polymer science approach to the practice of cryosta- bilisation technology: comments,’ Agric. and Food Chemistry, 1, 315–96 lillford p j and holt c b (1994), ‘Antifreeze Proteins, Journal of Food Engineering, 22, 475–82 nyman m (1995), ‘Effects of processing on dietary fibre in vegetables,’ European Journal of Clinical Nutrition, 49, S215–S218 payne s r and wilson p w (1994), ‘Comparison of the freeze/thaw characteristics of Antarctic cod (Dissostichus mawsoni) and black cod (Paranotohenia augustata),’ J. Muscle Foods, 5, 233–44 polo m v, lagarda m j and farre r (1992), ‘The effect of freezing on mineral element content of vegetables,’ Journal of Food Composition and Analysis, 5, 77–83 polvi s m, ackman r g, lall s p and saunders r l (1991), ‘Stability of lipids and omega- 3 fatty acids during frozen storage of Atlantic salmon,’ Journal of Food Processing and Preservation, 15, 167–81 340 The nutrition handbook for food processors refsgaard h h f, brockhoff p b and jensen b (1998), ‘Sensory and chemical changes in farmed Atlantic salmon (Salmo salar) during frozen storage,’ Journal of Agricultural and Food Chemistry, 46, 3473–9 rougerou a and person o (1991), ‘Influence of preservation method on unsaturated fatty acids of nutritional interest in sardines and mackerels,’ Medicine et Nutrition, 27, 353–8 shewfelt r l (1990), ‘Sources of variation in the nutrient content of agricultural com- modities from the farm to the consumer,’ Journal of Food Quality, 13, 37–54 xing y, yoo y, kelleher s d, nawar w w and hultin h o (1993), ‘Lack of changes in fatty acid composition of mackerel and cod during iced and frozen storage,’ Journal of Food Lipids, 1, 1–14 Freezing 341