3 ms·
No. They're metabolically equivalent. "High fructose" corn syrup is "high" relative to normal corn syrup, not relative to sugar, some of the HFCS formulae in pa
by tptacek 3mo ago
No. They're metabolically equivalent. "High fructose" corn syrup is "high" relative to normal corn syrup, not relative to sugar, some of the HFCS formulae in packaged food has less fructose than table sugar, and, most importantly, your body is a well-oiled machine at breaking down sucrose, which is cleaved enzymatically immediately upon it entering your gut; there's no material slowdown in digestion from eating sucrose.
(The key thing to remember is that sugar is bad for you; it's not that HFCS is good, but rather that cane sugar is just as harmful.)
Two fun additional notes:
* There's a video somewhere showing how cane sugar breaks down into fructose and glucose in bottled sodas anyways, and
* Perhaps unsurprisingly, taste testers can't reliably distinguish between HFCS and cane sugar formula Coke.
- vel0city 3mo agoYeah, a big part of the flavor difference between Mexican Coke and American Coke really comes down to the extra sodium in the Mexican Coke variety. You can get the US formula coke with cane sugar during Passover in markets with a high Jewish population. Taste it, it'll be practically the same as HFCS US Coke and not like Mexican Coke. Because, it has the same amount of salt as the US variety.
- TylerE 3mo agoThe bigger difference, imo, is that Coke just tastes better out of a glass bottle. Orifice size and shape makes a big difference in taste and mouthfeel.
- vel0city 3mo agoI definitely agree that the container can make a big difference on how we can perceive beverages, and I don't doubt it's a part of the experience fans of Mexican coke enjoy even if they don't realize that's playing a role.
- selectodude 3mo agoIronically, the coca-cola in Mexico uses high fructose corn syrup. It's the Mexican coke they export to the US that has sugar - which breaks down to glucose and fructose in the presence of phosphoric acid anyway. Anyway, the only thing Mexican about Mexican Coke is the "Medio Litro" on the bottle.
- dfe 3mo agoI have a theory on this. Yes, sugar breaks down to the same fructose and glucose proportion. That takes months. Even longer in concentrated syrup form. Even longer in the one of two components that bottlers get to make the syrup. A freshly bottled coke in a glass bottle still has the undivided sucrose and that is why it tastes more crisp. After a while, it separates and it doesn’t taste any better because it physically isn’t different. A freshly bottled HFCS coke still tastes fresh in all other respects, but is essentially stale on day one as far as the sweetener goes. But by far the best is a properly calibrated soda machine with fresh syrup— but it’s only really fresh if it is sucrose. Otherwise it’s basically stale before you even make it.
- tptacek 3mo ago(You mean that it takes months in the bottle; it's pretty much instant in the gut.)
- jandrewrogers 3mo agoUnder typical conditions the half-life of sucrose in coke is hours to a couple weeks via acid catalysis. In the body the half-life is seconds to minutes via enzymatic catalysis. The half-life of sucrose is greatly reduced by acidity and to a lesser extent temperature. It is stable for centuries in a neutral pH when kept cool but that describes almost nothing we do with sugar in foods.
- jandrewrogers 3mo agoChemistry doesn't support your hypothesis. Sucrose rapidly decomposes into glucose and fructose via acid-catalysis. Coke is very acidic. The half-life of sucrose in coke under good storage conditions is ~2 weeks[0]. Modestly higher temperatures, such as during transport, reduces that to hours. The internet suggests that that coke is typically consumed 4-12 weeks after manufacturing, depending on the supply chain. Even in the most optimistic scenario, no more than 25% of the sucrose remains when you drink it. Realistically, it will round to 0%. Almost no one has drank a coke that still had sucrose in it. The process is slower than enzymatic catalysis in the body (seconds to minutes) but still quite fast. [0] I used an LLM to run the math but the result passes a cursory eyeball test.