Frozen ready meals, gravies, fillings and cook-in sauces face a test that a chilled product never does. They are frozen, often thawed and refrozen somewhere in distribution, and finally reheated. A sauce that fails the test releases water on thawing, a defect called syneresis, and its texture turns spongy or grainy.
What happens to the starch
Cooked starch is not stable. As a starch paste stands, its chains gradually reassociate into more ordered structures, a process called retrogradation, and the reorganised starch holds less water. Freezing and thawing accelerate this. A study of sweet potato starch gels in Food Hydrocolloids measured both the water released after repeated freeze-thaw cycles and the retrogradation induced by them.
Which gums help
That study compared nine gums in starch gels taken through five freeze-thaw cycles. Alginate, guar and xanthan were highly effective at reducing water release. At the higher level tested, guar left the least water, less than half that of the plain starch gel. At the lower level, xanthan was more effective than guar. Interestingly, there was little relation between how much a gum changed the paste's viscosity and how well it prevented syneresis, and alginate slowed retrogradation itself more than xanthan or guar did.
The best gum depends on the starch. In tapioca starch gels, xanthan was more effective than locust bean gum and konjac glucomannan, while guar had no effect at all.
| Starch | What reduced water release | Source |
|---|---|---|
| Sweet potato | Alginate, guar and xanthan were highly effective | Lee and others, 2002 |
| Tapioca | Xanthan most effective; guar had no effect | Muadklay and Charoenrein, 2008 |
| White sauces with corn, waxy corn, potato or rice starch | Xanthan and locust bean gum both helped, xanthan more so | Arocas and others, 2009 |
pH changes the answer
A study of tapioca starch pastes tested xanthan at pH 3, 7 and 9. Xanthan was most effective at pH 7. At pH 3 it was much less effective, and the pastes released the most water within two cycles. The authors put this down to acid breaking down both the starch and the xanthan. Acidic sauces, from tomato-based gravies to tamarind chutneys, need to be tested at their own pH.
How fast it freezes
In the tapioca work, a fast freezing rate prevented more of the changes caused by retrogradation than medium or slow freezing. Formulation and process work together: a formula that passes in a blast freezer may fail in a slower domestic or retail freezer.
In a real sauce
White sauces made with four native starches were the most affected when made with corn and potato starch. Adding xanthan or locust bean gum at low levels reduced both the water released and the structural changes after thawing, with xanthan the more effective of the two.
A faster way to screen
Freeze-thaw trials take days. A study of tapioca starch with xanthan found that setback in a pasting test, a measure of short-term retrogradation, correlated with the water separated in repeated freeze-thaw testing. A quick pasting run can narrow the options before the long trial begins.
The other lever: the starch itself
Gums are one route. The other is to choose a starch that resists retrogradation in the first place, which is one of the main reasons modified starches exist. Our article on why modified starches exist explains how stabilised and cross-linked starches differ.
Before changing the formula
- Name the starch and check the final pH.
- Record how the product is frozen, and how many freeze-thaw cycles it must survive.
- Measure water released after a set number of cycles, against a control made the same day.
- Test gums and starches as separate trial arms, so you know which change worked.