As long as it's not gazpacho...

January 1, 2006 6:07 PM

ViscosityFilter: Ceteris paribus, will thicker soup (clam chowder) take longer to reach room temperature than chicken noodle, assuming they were both heated to the same core temperature?

I took "honors" chemistry several years ago in high school, and no physics to speak of, so if I'm completely off the mark on this question, please feel free to slap me around.

My friend and I were debating if an equal quantity of "thicker" soup, kept in a sealed thermos would take longer to cool down than a "thinner" soup. My reasoning for such behavior involves the effect higher viscosity might have on the kinetic activity of the fluid's molecules—it's thicker, and therefore, things move more slowly?

Let's ignore, for the purpose of this question, any "skin" that might develop on thicker soups. I doubt such skin would have any effect in a thermos anyway.

Is there a name for this concept? Am I totally incorrect? Would they cool in a roughly equal amount of time? I've read some that viscosity of a fluid can actually be affected by the fluid's temperature, which confuses me even more.

The core question here is about fluids in general. I understand certain properties of soup may introduce different variables, so if the soup factor affects the answer, please make note of it. Thanks in advance!

posted by disillusioned to science & nature (10 comments total)

I think the major effect of viscosity will be to slow the rate of cooling by free convection. The key dimensionless parameter for free convection is the Grashof number, which has the square of kinematic viscosity in its denominator. At high viscosities, viscous forces will dominate buoyant and inertial forces, preventing free convection. There's probably some critical value above which convection won't occur, but even below that value, increased viscosity will slow convection.

As for your explanation that viscosity has an effect on "the kinetic activity of the fluid's molecules", I'm not sure that's a great way to think about it. It's true that a lot of molecular properties that lead to a fluid having a high heat capacity could also lead to it having a low viscosity, but heat capacity and viscosity are both bulk properties, and they don't really apply on the single-molecule level.
posted by mr_roboto at 6:56 PM on January 1, 2006

mr_roboto writes "There's probably some critical value above which convection won't occur, but even below that value, increased viscosity will slow convection."

Whoops! That's backwards. It should say, "There's probably some critical value below which convection won't occur, but even above that value, increased viscosity will slow convection."
posted by mr_roboto at 7:33 PM on January 1, 2006

I'm not great at physics, so this is all I'm going to say. In this problem, shouldn't you be looking at the liquids' comparitive density as well as viscosity? At the very least, you'd have to isolate the two factors, because denser liquid will cool slower even if the viscosity is the same.
posted by Miko at 7:53 PM on January 1, 2006

This seems like a relatively easy question to resolve empirically.

But just for fun: the soup can lose heat by convection, conduction and radiation, right? Radiation seems unlikely to be affected by viscosity or density (disclaimer: I went into biology because physics always made me queasy), leaving convection (mr_roboto seems to have that right: higher viscosity will mean lower convection and so slower cooling) and conduction. I think Miko may have it backwards: denser substances are typically better conductors of heat, so would denser liquids (thicker soups) not cool faster?

Since the rate of heat loss is likely to be fastest by far at the surface of the container, I'd guess that the rate of heat transfer from the soupy depths to the outer layers would be the primary limiting factor. In that case, there's a tradeoff between viscosity (increased viscosity --> slower transfer/cooling) and density (higher density --> faster cooling).

Added to that is the style factor, or chunkiness. Assuming the average chunk to be denser than the soup, more chunks would mean a denser soup but not necessarily a more viscous one, so chunkier soups might be expected to cool faster in this model.
posted by sennoma at 10:04 PM on January 1, 2006

Fun times...

Thanks for your answers thus far.

Now I'm hungry for soup...
posted by disillusioned at 12:49 AM on January 2, 2006

Miko might have a good point. you could imagine that different soups have different specific heat capacities (perhaps depending on the amount of fat). this might correlate with viscousity.

although i agree that convection is the answer you were looking for.
posted by andrew cooke at 6:24 AM on January 2, 2006

Miko writes "In this problem, shouldn't you be looking at the liquids' comparative density as well as viscosity? At the very least, you'd have to isolate the two factors, because denser liquid will cool slower even if the viscosity is the same."

sennoma writes "In that case, there's a tradeoff between viscosity (increased viscosity --> slower transfer/cooling) and density (higher density --> faster cooling)."

It's worth pointing out that there is no general correlation between viscosity and density. So if you're looking at the two systems with different viscosities, ceteris parabus, there's no reason to assume a difference in density.

andrew cooke writes "you could imagine that different soups have different specific heat capacities (perhaps depending on the amount of fat). this might correlate with viscousity."

I don't think that such a correlation generally exists, though I can't find any nice tables of data like the one for viscosity and density linked above.
posted by mr_roboto at 11:00 AM on January 2, 2006

i wasn't suggesting that there is a general correlation that says that viscousity and heat capacity are "always" linked. but, as i said, it's quite possible for soup if two constitutents differ in quantity and also in viscosity/heat capacity.

so fat would give you a mechanism. say, for example, fat increases viscousity but has a lower specific heat than water, then there would be a negative correlation. for there to be no correlation then fat and water would have to have identical viscosity or specific heat.

i think i'm just repeating what i said. bleagh.
posted by andrew cooke at 11:56 AM on January 2, 2006

It's worth pointing out that there is no general correlation between viscosity and density.

I didn't actually mean to indicate such a correlation, but I can see why you might read my comment that way. I meant only that viscosity and density will have opposing effects on cooling rate, and so Miko had a good point about knowing both.

Regarding fat content, most soups are mostly water, so their specific heat capacities will be essentially identical, but a really high fat content might skew that. If AC is right and fat has lower SHC than water, then it will have taken less energy to get the fattier soup to temperature than the less fatty soup. It seems intuitively plausible that the soup which has less energy to lose over a given temperature gradient would cool over that gradient faster.

So denser, fattier, chunkier and less viscous soups might be expected to cool faster than their less dense, leaner, etc fellows.
posted by sennoma at 4:58 PM on January 2, 2006

The viscous soups cool slower. Harold McGee treated this in The Curious Cook, which I recommend!
posted by oxonium at 3:26 PM on April 14, 2006