We have a recipe for a salted caramel spread that is basically a big pan of sugar, glucose, milk and butter, cooked until it becomes a thick paste, then mixed with double cream to achieve a smooth consistency and a glossy finish. Our only problem was that we kept having big variations between batches -- some ended up as thick, sticky spreads (like we intended), others were runny and sauce-like.

After some experiments we discovered that the final stage of the process -- pouring the double cream in -- was responsible. Pour the cream cold and the caramel would come out runny; heat the cream before pouring and the final result would be thick and rich.

Although we feel we're in control of the process now, I was wondering if anyone could explain to us why does heating or not heating the cream have such a big impact on the thickness of our spread.

2 Answers 2


You don't tell us neither the ratio of double cream to caramel, nor the time you heat the completed mix, so this is just a guess. But it sounds logical that your problem is evaporation.

Double cream consists mainly of fat and water. I don't remember the exact percentages, but more than half of it is water. So, if you heat cream, part of it evaporates during the heating, and it continues evaporating after being added to the pan with bubbling caramel. But if you use cold cream, its water hasn't evaporated during a heating phase, and if you add a large amount of cream to a hot caramel mixture, the whole mixture cools considerably. If you don't heat it and let it simmer for long enough afterwards, the water content of the cream continues thinning the spread. You end up with a runny sauce instead of a sticky spread.

Instead of pre-boiling, you should be able to just simmer for longer time after adding the cream, if you find it more convenient. But you'll have to try it a few times until you have found the cooking time which gives you optimal consistency.

  • Makes perfect sense! We'll probably experiment with simmering the final mixture (with the cream poured in) to see what other consistencies we can achieve too! Thanks so much for your answer! Commented Jan 18, 2012 at 21:20

Other than evaporation as in the accepted answer, boiling the cream is also denaturing the proteins in the milk which provides a better, more pleasurable consistency.[0]

Just ask any coffee drink connoisseur or barista, the magic of a Cafe Latte or any other milk and coffee drink is in the milk consistency. These coffee folk have don a lot of work and research into milk and milk temperatures. I'll sum it up here: don't go over 158F but do get up to 140F[1]

you want the fats to melt and the whey to denature but you don't want the denatured whey combining with the lactose. I assume, it's ideal to have the denatured proteins combine with your sugary slurry instead.

All in all, the advice 'boils down' to the same: simmer it. don't boil. Although i would argue keeping it under 158F(70C) at least until adding to the spread mixture. And even then maybe not.

[0]: I don't care if it's half and half, whipping cream, heavy cream, double cream, or even butter. if it's not Ghee it's got milk protein in it.

[1]: a good rundown of milk temp as it relates to coffee drinks: https://perfectdailygrind.com/2019/02/what-temperature-should-your-cappuccino-milk-be/

ok, i just double checked myself and technically whey doesn't denature until after 70C. so before 158F(70C) what you're really doing is smoothing the fats and condensing the milk. (evaporation). In reality, you don't want to denature the globulin and albumin because you'll break up the lactose/protein bonds which will make for a yucky flavor, I assume because they recombine in weird ways upon cooling. Burnt meat is yummy, Burnt milk is not. Ask Monsieur Maillard.

  • 1
    I don't think that this is relevant here. First, both raw milk and scalded milk are quite tasty in both flavor and texture, so the effect you describe must be small. Second, I doubt that commercial baristas are allowed to use raw milk, only pasteurized or ESL, which has been heated to above 158 anyway. Third, the OP is unlikely to have access to raw double cream. Fourth, protein denaturation works slightly differently in the presence of large amounts of fat, and double cream is almost 50% fat. Fifth, the OP is dumping the cream into caramel at about 375 Fahrenheit, heating the cream instantly.
    – rumtscho
    Commented Mar 22, 2021 at 8:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.