Freeze/thaw-stability 33733

Freezing is one of the best ways to preserve the quality of a food product. Properly processed, packaged and stored frozen food comes close to fresh in terms of nutrition, color, flavor and – for cooked foods – texture.

However, since the world is an imperfect place, it is difficult to maintain food products in a constant, optimum frozen state. With every change in the mercury comes a little more freezing or a little more thawing. Ultimately this temperature cycling can adversely affect the finished product quality.

The chill drill

Any ingredient or product termed freeze/thaw-stable shows a resistance to deterioration after repeated temperature cycling. Such temperature fluctuations cause the available water to form ice crystals, then melt, then refreeze, and so on. This creates a number of undesirable effects, including moisture migration, dehydration, syneresis, structural breakdown, and the formation of large ice crystals which impart a gritty mouthfeel to frozen foods.

For these effects to occur, the product temperature does not necessarily have to change significantly or even rise above 32°F. In last November's issue of Food Product Design, we discussed the theory behind freezing in "Cold, Hard Facts: The Fundamentals of Freezing." Pure water will crystallize, or freeze, at 32°F (O°C). Water in a food system is anything but pure; it contains varying amounts and types of solutes that depress the freezing point. As the temperature drops and ice crystals form, the solutes become more concentrated. This further depresses the freezing point. Depending on the product, the point at which no unfrozen water exists -- the eutectic point -- is rarely achieved during commercial freezing and storage.

This means that usually a certain amount of water remains in a liquid state. Unless that water is controlled in some manner, it will be up to its old tricks, wreaking havoc with the quality in all of those ugly ways previously mentioned.

Free water is not the only culprit, however. Dropping the temperature causes the solutes to concentrate, and this may change properties such as pH, ionic strength and viscosity, as well as increase the occurrence of undesirable reactions such as oxidation. Concentrating a solution may create insoluble precipitates or lactose crystallization, thus producing grittiness or off-flavors, depending on the compounds formed.

The ice crystals themselves may cause damage. As water freezes it increases in volume. In addition, the size of the crystal formed contributes to the problem -- the larger the crystal, the worse the effect. Large crystals form during slow freezing processes and during recrystallization. Recrystallization can occur not only with changes in temperature, but also with changes in pressure, with molecular ice and water migration, or aggregation of crystals. As the water freezes and expands, particularly if large crystals form, it produces physical damage to the rest of the product matrix. Ice crystals can rupture cell walls or granules, or break molecular networks. The more rigid the structure, the more likely it is to be damaged.

Vegetable tissue tends to be more susceptible to damage during freezing and subsequent refreezing than animal tissue because the plant tissue is much more rigid and the cells are poorly aligned. This causes them to break rather than bend to accommodate the growing ice crystals. Even this can vary with the type of tissue, its stage of maturity, and prior treatments.

In vegetables, enzymatic activity continues under frozen conditions. The rate is a function of both the concentration, increased by water crystallization, and the temperature. As the temperature increases, so does the enzymatic activity. Therefore, to eliminate the defects like enzymatic oxidative browning, enzymes must be deactivated, usually by a heat treatment called blanching. In some cases, browning can be controlled by adding ingredients such as sulfites, citric acid and ascorbic acid. These may affect the reaction by interacting with the enzyme or food itself, by lowering the pH, or by acting as an antioxidant and tying up the oxygen.

Because of their high fat content, meats can undergo oxidative rancidity during frozen storage, particularly when the concentration of prooxidants such as metal catalysts increases with freezing. To prevent this deterioration processors can add antioxidants, or eliminate oxygen from contacting the product by packaging under vacuum or with an inert gas such as nitrogen.

The chilling story

To guarantee freeze/thaw-stability, a product would need to be formulated completely without moisture, packaged in a vacuum, and surrounded by a total moisture barrier -- something like a rock in a can. Unfortunately, that doesn't make for a very palatable product, so food designers must consider the options that maintain finished product quality. These vary with the system, especially the ingredients, the process and the finished product characteristics required.

The first step to promote stability is to freeze the product quickly, producing numerous, small ice crystals. Smaller ice crystals generally mean less initial damage, especially once the product is thawed. Any structure that retains or controls the water, whether cell wall or protein network, is more likely to remain intact.

"The ice crystals actually crush or squeeze the protein and cause it to destabilize," explains Mark Freeland, director of advanced colloids, Rhône Poulenc Food Ingredients, Cranbury. NJ. "Destabilizing a protein system causes wheying off and deteriorates the texture of the product. That often occurs in cheese products."

Smaller ice crystals also tend to slow the rate of recrystallization into larger crystals. Larger crystals usually grow at the expense of smaller crystals because of the difference in their surface energies: the large crystals bind molecules more tightly than the smaller ones. This effect occurs to some degree at a steady temperature but accelerates as the temperature fluctuates.

The size of the ice crystal influences the eating quality of products consumed frozen, such as ice cream. A large crystal, somewhere in the area of 30 to 40 microns, can be perceived by the tongue as grittiness. As the product cycles through freeze/thaw temperatures, the crystals gradually increase in size.

"Ice cream is initially processed under high shear, creating very fine ice crystals in the product," says Freeland. "After processing, if it melts then refreezes without that shearing action, the crystals tend to get bigger."

Adding hydrocolloids -- such as guar gum, locust bean gum, carboxymethylcellulose (CMC) or, to a lesser degree, carrageenan -- helps control that crystal growth because they bind the water and keep it from coalescing into bigger ice crystals.

"The same effect occurs in lower moisture products," Freeland adds. "You can certainly destroy the texture of a frozen cake by freezing and thawing if you don't trap and hold that water in place."

Quick freezing also immobilizes the water, keeping migration-caused problems to a minimum. This is critical in multiple-phase products with widely differing moisture contents, such as pies or pizzas.

"If there is a great disparity in water activities between two components, there will always be a driving force for the water to migrate," observes Mike Augustine, manager of food ingredient applications at A.E. Staley Manufacturing Co., Decatur, IL. "The best you can do is slow it down."

"The rate of freezing depends on the rate of heat transfer and the method used. If freezing is not done properly, problems may arise because of the cooling rate disparity in commercial operations. Freezing by the pallet generally decreases the overall quality and resistance to deterioration.

"The faster you freeze a product, the better the quality and stability," says Freeland. "If it sits for two or three days before it freezes, the texture and quality will suffer."

When a product depends on its structure to entrap air, repeated freezing and thawing can cause that structure to change. A whipped topping, for example, is a high-overrun, high-fat emulsion. If the ice crystals grow big enough, the product volume can shrink and eventually collapse. Some shrinkage also can be attributed to dehydration.

Freezer burn is caused by dehydration, or sublimation of the ice during storage resulting in discoloration and the formation of off-flavors. It is not a result of cycling temperatures. The main method of prevention is adequate packaging to avoid the moisture loss. However, because some ingredients used to provide freeze/thaw-stability control the movement of water, they may impart some increased resistance to this defect by slowing the rate at which the moisture is released.

Ice versus glass

For most products, the target temperature for Iong-term storage is 0°F. This doesn't mean that the product is completely frozen, nor that it has reached its eutectic point. A significant amount of unfrozen water will still exist and will increase with any temperature increase. In reality, however, this effect may not be as severe as expected. Before freezing completely, most products experience a glass transition state.

The glass transition temperature, Tg, is the point at which the product becomes a noncrystalline solid rather than a rubbery or leathery fluid. The liquid or unfrozen product becomes so viscous that molecular mobility is basically curtailed. Theoretically, this is the temperature at which a frozen system is most stable. The greater disparity between the glass transition temperature and the freezer temperature, the more likely that undesirable reactions can occur. As long as the glass transition temperature is below the storage temperature some molecules will be mobile, increasing the propensity for undesirable reactions.

"Glass transition is a a unifying theory but far from perfect. However, the important concept in terms of stability is that the less difference between these temperatures, the more stable a frozen product will be with time," says Augustine. "That can be difficult in some cases because if a product contains a high level of monosaccharides, it may produce a very low glass transition temperature."

The glass transition temperature, as well as the freezing point, is a colligative property -- that is, it depends on the number of molecules. Adding low-molecular-weight compounds, such as sugar, depresses this temperature because there are more molecules in a given weight. The Tg for a given product is basically a weighted average of molecular weights. It can be measured by a number of means -- on a molecular, mechanical or thermal level, all of which may give different figures.

A product designer can change the formulation and bring the glass transition temperature closer to the storage temperature. However, Augustine warns that there is a danger of making the product too hard if the product is consumed frozen.

"Low-molecular-weight compounds plasticize the product, making it soft," he says. "The high-molecular-weight compounds make things firm, so you have to be careful to balance them. One frozen dessert on the market contained a high level of high-fructose corn syrup to provide texture and make the product easily spooned right out of the freezer. It depressed the glass transition temperature to such an extent that at frozen storage temperatures, the ice crystals that formed tended to aggregate and form huge crystals that destroyed the texture of the product."

Frozen formulations

Hydrocolloids can help control unfrozen water or stabilize ice crystals. In products consumed in a frozen state, the crystal size and shape affects the product quality. In products thawed for consumption, the aim is to bind the free water to prevent migration, condensation or syneresis.

Some simple gum solutions, those without the tendency for the molecules to self-associate, go back to the original form after freezing and thawing. However, freezing concentrates a solution, which tends to promote self-association and certain hydrocolloids may form a gel. This gel can be broken by the ice crystals. Ingredients such as starches, gums and locust bean gum will not return to their original structure.

"You can disrupt the gel network by freezing," notes George Sanderson, Ph.D., research fellow, Kelco, San Diego. "With a gel, as opposed to a solution, freezing promotes chain association. When it thaws, the gel has been changed and it releases free water. It requires a thickener, as opposed to a gelling agent, in order to contain that free liquid. You need a gum that does not self-associate, like xanthan gum, which is often used in combination with starches to give added freeze/thaw-stability. Guar and some types of CMC can also work."

Concentrating the hydrocolloids in the unfrozen portion has an additional benefit. As the hydrocolloids become more concentrated in the aqueous phase, the viscosity increases, reducing the mobility of the reactive molecules and the formation of ice crystals. This also can be achieved by adding intermediate polymers such as low DE corn syrup, maltodextrin and starch hydrolysis products. So although the reactivity increases with concentration, the reaction rate may not show comparable increases because of the restricted motion. This can occur before the product reaches a glass transition.

Still, indiscriminately adding hydrocolloids doesn't always improve product quality. "If you over-utilize hydrocolloids you can actually cause problems," warns Freeland. "Too high a level can cause syneresis; you've destabilized the system. That happens in dairy products with the wrong levels of CMC, guar and locust. If the levels are over the critical concentration, the water is actually squeezed out of the product."

In products that are not thawed before consumption, the main goal is controlling the shape and size of the ice crystals. Several factors influence the outcome. While the process typically has the greatest influence, certain ingredients can help. Alginates, for example, can promote flat, plate-like crystals that, although large, are not readily perceived in the mouth in certain products.

"Most stabilizer systems tend to promote the same sort of things, but they differ in how they perform during the process and how they are perceived in the mouth -- for instance, whether they are gummy," points out Sanderson. "You need sufficient control of the viscosity during the process and sufficient control of the ice crystals to maintain the eating quality. If the product is too thick, you can blow the plates on the heat exchanger. You want to maintain the structure, but not so much that the product never melts."

Different base starches have different inherent stability to freeze/thaw conditions, but chemical modification can increase their resistance to breakdown. Common starch, potato starch, and tapioca, to a lesser degree, do not exhibit an inherent stability. They contain high levels of amylose, a straight chain molecule that tends to reassociate with itself, creating a gel.

Waxy starches are more stable since they consist of branched molecules, amylopectin. This structure resists retrogradation, or recrystallization. As the mobility of the molecules decreases with freezing, the structural stability decreases. This creates unacceptable textures and syneresis once the product is thawed. This is the major problem that occurs with freezing. In addition, the formation of ice crystals inflicts physical damage on the starch granule, which can result in a loss in viscosity.

Staling of baked products can be attributed to starch retrogradation or protein dehydration. Both can occur in frozen foods.

The primary modification used to combat freeze/thaw problems in starch is substitution. Cross-linking provides some protection against mechanical stress, such as that created by ice crystals, by tying the starch granule together.

"There are two primary types of substitution that can be used under these conditions: acetylation or a hydroxypropylated starch," notes Augustine. "Of the two, the hydroxypropylated starch tends to be better for a couple of reasons. One, you can go up to higher levels with the hydroxypropylated substitution and the molecule is a little bigger, which helps to some extent. When the product wants to reassociate, it becomes more like trying to stack tree branches rather than tree trunks. It inhibits the linear portions of the molecule from reassociating."

Changing the freezing point can affect the ice crystals' size and stability. This is a function of the molar concentration, or the molecular weight of the solute. The lower the molecular weight, the more the freezing point drops. Using a high molecular weight ingredient such as a hydrocolloid will have a minimal effect on the freezing point. Typically sugars and maltodextrins are used for this purpose.

"Freezing point depression can provide a benefit in some cases to product stability, but then it also tends to melt at a lower temperature," points out Freeland. "If you have good control of your distribution, you can use that kind of effect. If not, a lower melt point may cause more problems."

Other ingredients provide a measure of freeze/thaw-stability. Because freezing causes protein denaturation, reduced protein solubility and dehydration result after repeated freeze/thaw cycles. Phosphates are often added to muscle foods to reduce drip loss by strengthening the protein network.

One substance that appears to be able to inhibit recrystallization of ice is known as an antifreeze glycoprotein. These were originally isolated from fish blood, but they can be synthesized. They appear to lower the freezing point significantly, as well as affect the shape of the crystals formed. These proteins are still in the experimental stage.

The best test

"The physics of freezing is a relatively complicated subject, even when dealing with the basic concepts," says Sanderson. "Once you get into a real-life system with a number of ingredients, it just gets that much more complicated. In reality you can only make an educated guess based on what your product requirements are throughout the process and the life of the product."

Because of this, it's crucial to test the formulation to see if it meets the shelf life and storage requirements. But those waters can be pretty murky, too. Determining the appropriate temperatures and cycling procedure to ensure that a product truly is "freeze/thaw-stable" is a matter of making assumptions of the number and degree of thaw and refreezing cycles a product will undergo before it is consumed. Typically this is based on the conditions the product is likely to encounter after it leaves the manufacturer.

"By and large, the manufacturers understand the problem areas and take care of them. It's really the subsequent distribution and storage where you'll find most of the problems," says Sanderson.

Food companies employ from three to as many as 10 freeze/thaw cycles for testing purposes, with five to seven being the norm. The temperatures used vary widely.

"There are two schools of thought," says Freeland. "I'm not sure one is any more valid than the other. I prefer to thaw the product and refreeze it. That may over-stress the system, and people may argue those conditions are too severe. You could potentially eliminate products for not being freeze/thaw-stable when they are in actual practice. Some companies use elevated temperatures, around 10° to 15°, for a week or two and never thaw the product completely.

"It's a matter of how conservative you want to be," he continues. "But it is certainly possible that the product will be thawed completely. If you buy a frozen pizza and take two hours to get it home and in the freezer, most likely the product has completely defrosted. It may be 45° to 50° at that point."

It is virtually impossible to predict the conditions any one package will encounter throughout the distribution chain. Therefore, it is important to design foods with as much protection as possible before sending them on their way.