What is Temperature?
May 31, 2011 6:56 PM   Subscribe

Science Explanation needed: Trying to get a better understanding of 'temperature'. While discussing whether it's safe to eat from a non-stick pan I pointed out that Teflon doesn't start to break down until it hits around 300 degrees; if you kept the pan temp lower than that it should be fine. I was told that I'm wrong because 'temperature doesn't work that way'.

In a nutshell, the explanation I was given is that atoms are all vibrating at different rates and some are shaking much more than others in a given body. Thus some could be measured as being at 800 degrees and others at 100 and so the temperature of the measured body would be an average of all those temperatures. That would cause the Teflon to lose small bits of itself at the atomically smaller higher temps and get into you.

Does this make any sense? I've been reading about temperature all day and can't find anything that I can understand one way or the other as to this explanation of temperature.
posted by artof.mulata to Science & Nature (24 answers total) 4 users marked this as a favorite
Best answer: Temperature is a measure of the average kinetic energy of the particles in a substance. The higher the temperature of the substance, the higher the average kinetic energy of the particles. Because kinetic energy is related to the speed at which particles move, the higher the temperature, the higher the average speed of the particles in the substance.

Because temperature measures the average energy, some particles can have more energy than others. The actual distribution of energies for all the particles in a substance at a single temperature looks like a bell curve. Most particles have energy near the peak of the curve, but some particles will have far more (or far less) energy than others in the substance.

At temperatures less than 300 degrees, some of the particles in a substance might have more energy (and therefore speed) than the average particle in a 300 degree substance. Some of the higher energy particles at this lower temperature might have enough energy to "break down." As the temperature increases, a greater percentage of the particles will have the "break down" energy you mention.

This is a theoretical discussion- I know nothing about the properties of Teflon and how it breaks down.
posted by jz at 7:17 PM on May 31, 2011 [1 favorite]

Processes like this one can (hand-wavingly) be described as "thermally activated." What that means is that there is some sort of an energy barrier which is keeping the teflon molecules tied together, and if the teflon gets too hot, particles will break free. As your friend noted, even if an object is at a uniform temperature, not all particles will have the same amount of energy and thus some will be more likely to break free from their bonds than others. Still, the temperature of the object generally gives us a way to calculate the average rate at which this sort of process will occur.

For thermally activated processes, the rate can generally be approximated by the Arrhenius equation. The take-home message of the Arrhenius equation is that the dependence of these processes on temperature is very, very strong, in fact, it is exponential. The exact form of the function actually shows that for temperatures much smaller than the activation energy, the rate is essentially zero. It is only as you approach the activation energy that the rate rises significantly.

What does this mean? Well, it means that the amount of teflon that breaks down at 200 C is not 2/3 of the amount of teflon that breaks down at 300 C. In fact, it is almost certainly much less than this. As the pan approaches the breakdown temperature, the amount of stuff that is ejected will rise sharply, so if you stay well below it, you are almost certainly fine.

Hope this helps.
posted by artichoke_enthusiast at 7:26 PM on May 31, 2011 [6 favorites]

Response by poster: Cool. Thank you. In looking for the terms 'Teflon', 'bell curve', and 'temperature', I find a lot of abstracts, but nothing that actually helps me to figure out (as a layperson) if the argument my friend presented is correct. I'd really like to figure this out as if my using the non-stick is going to feed me Teflon then maybe I really should stop. Unless it's only going to feed me a few molecules at a time...

Should I assume that there's a method or a rule that dictates what the normal distribution of the curve will be? It would be interesting to know that to.
posted by artof.mulata at 7:27 PM on May 31, 2011

The breakdown point for Teflon is much higher than 300 ºF. From Wikipedia's well-sourced article: "While PTFE is stable and nontoxic, it begins to deteriorate after the temperature of cookware reaches about 260 °C (500 °F), and decomposes above 350 °C (662 °F). These degradation byproducts can be lethal to birds, and can cause flu-like symptoms in humans." So you have a lot more leeway than you might think.
posted by jedicus at 7:27 PM on May 31, 2011

Response by poster: Thanks science people. But what about the bell curve jz mentions? Is it possible for the temperature of the individual atoms vibrating to achieve 350 °C/662 °F if the flame on the stove is set for say 250°F/121°C?
posted by artof.mulata at 7:32 PM on May 31, 2011

Prvious posters are right on about the theoretical basis of temperature. Practically, I think the more relevant question is how you're measuring the temperature of the pan. Skillets have hot and cold spots, and a 300 degree pan is not super-hot, as pans go. (On preview, the 500-degree figure is much closer to what sauteeing and searing require. But nonstick never gives a good sear.)

If you want to be precise and safe about using your pans, an infrared thermometer is the way to go. I don't know how you mean to set your stove for a certain temperature.

All that said, I rarely find nonstick necessary, or even useful. Clad aluminum gets much better color, and if you get it hot enough ("Hot pan, cold oil, food won't stick"), adhesion isn't a problem. If you need non-stick, I really like non-PTFE ceramic.
posted by supercres at 7:34 PM on May 31, 2011

Best answer: But what about the bell curve jz mentions?

A body can be at a certain single temperature everywhere, yet the atoms making up that body have a distribution of kinetic energies. That's the bell curve he's talking about. Temperature thus doesn't really make sense for single or few particles... one certainly means "average (expectation value of) kinetic energy" of a single particle if one says that an atom exists at a certain temperature.

A body can also be at different temperatures at different places... consider a pan on the stove and its handle. This has nothing to do with the bell curve jz mentions. This is a body being non-uniformly heated.
posted by fatllama at 7:38 PM on May 31, 2011

Response by poster: I'm not actually wanting to get in there and measure the temperature of the pan (although now I might for fun); I'm more interested in whether there was a basis in fact for my friend's argument against using Teflon based on the nature of how atoms vibrate. Still not totally clear on that, but I'm getting a much better understanding on how to think about this problem.
posted by artof.mulata at 7:40 PM on May 31, 2011

artof.mulata --

In a nutshell, individual atoms do not have temperatures. Temperature is a measure of an ensemble of particles. As far as the "bell curve" that was mentioned before, if you want to get math-y, it is technically called the Maxwell-Boltzmann Distribution.
posted by artichoke_enthusiast at 7:41 PM on May 31, 2011

Response by poster: Wait. So if the curve doesn't matter because it's a register of all the different vibrating frequencies of the constituent atoms of a body then the thing we think of as temperature in a general sense is the only thing to consider?

So this means that even if some atoms are vibrating at a much higher velocity than the 'break down' point for a material it won't because the average (actual?) temperature is the only one that matters?
posted by artof.mulata at 7:44 PM on May 31, 2011

Best answer: artof.mulata :

Basically, yes. For a body at a given temperature, the maxwell-boltzmann distribution tells us how many particles have exactly how much energy. The average is, of course, proportional to the temperature of the body, but there are always particles with more or less energy than average. The key is that the distribution really only has one parameter in it, which is the average temperature. Once you know that, you know everything you need to know about the entire distribution.

When calculating things like chemical reaction rates (which is what the breakdown of teflon is), this distribution is naturally taken into account -- one can get to the Arrhenius equation from the maxwell-boltzmann distribution fairly simply -- but since the temperature gives us (pretty much) everything we need to know about the distribution, then the temperature is all we need to know in order to calculate chemical reaction rates.

Now, of course, there are going to be different temperatures at different points in the pan, but that is due to macroscopic things -- heat diffusion to the air, heat absorption due to the flame, etc.
posted by artichoke_enthusiast at 7:51 PM on May 31, 2011

Response by poster: Got it. Thanks a lot everyone. I fell very confident at being able to explain this to anyone now. You guys are fantastic.
posted by artof.mulata at 7:55 PM on May 31, 2011

It's somewhat intuitive to think that what ought to matter is not the ensemble average (temperature), but something about the kinetics of the most energetic atoms. But:

a) Temperature completely describes just that: the quantity of atoms having a certain (perhaps critical) kinetic energy. And if at one moment in time atom A has a lower than average kinetic energy, it'll like have a higher than average kinetic energy a moment later.

b) Because of the exponential dependence of many chemical and physical processes on temperature, it remains a great description of "critical" turn on points.

That said, if Teflon really is destroyed at 300C (or whatever, clearly this is going to depend on the environment), you can bet it will be destroyed at 295C... just more slowly.
posted by fatllama at 7:57 PM on May 31, 2011

Response by poster: Thank you for the update fatllama. Maybe I will get rid of those pans.
posted by artof.mulata at 8:04 PM on May 31, 2011

So this means that even if some atoms are vibrating at a much higher velocity than the 'break down' point for a material it won't because the average (actual?) temperature is the only one that matters?

I think that's not quite right— it's more that the breakdown temperature is the one at which the more-energetic end of the bell curve starts being energetic enough to break stuff. If you raised the temperature so that the average energy was enough to make the teflon break down, it'd be 'way higher than the "starts to break down" temperature.
posted by hattifattener at 8:06 PM on May 31, 2011

Maybe I will get rid of those pans.

Don't get rid of the pans! Use them for omelets. Get non-teflon pans for searing meat.

Also, be aware of the two temperature unit flying around in this thread. 300 C is about 570 F. Most home ovens won't even achieve 570 F.

Dupont has a lot to say about the temperature limits:

260 C / 500 F -> discoloration in Teflon cookware.
204 C / 400 F -> oil smokes, warning you before the teflon discolors
348 C / 660 F -> Teflon deteriorates, giving you and all your loved ones cancer. (/sarcasm)
posted by fatllama at 8:13 PM on May 31, 2011 [1 favorite]

Maybe some of the MeFi foodies will tell me my cookware sucks (oh please do), but I love my Scanpan nonstick cookware that is apparently free of the worrying compound (PFOA) that is released from overheated teflon. I'm not trendy and I don't care much about these things, however.
posted by fatllama at 8:20 PM on May 31, 2011

Yeah, just nthing the idea that a pan is going to get to 300' on a home based hotplate, be it gas, electric, or induction is really optimistic. Like, crazy optimistic. If you left a crappy teflon pan on the stove - empty, for >15 minutes, it might get there, but it would basically _never_ ever get there if you were cooking something in it. Pans just don't get that hot, especially teflon pans which are typically thin, cheap, aluminium.
posted by smoke at 9:07 PM on May 31, 2011

Fatllama, there doesn't seem to be anything specific on the Scanpan website about this, but I was talking to a salesperson at Sur La Table about this brand of cookware and she said the nonstick coating was titanium nitride. Clearly there's something else there too since the pans were black and TiN is gold colored, but it's been used for years for wear-resistance and chemical inertness. Anyway, take that for what it's worth.

Artof.mulata, you can get an IR thermometer from Radio Shack for about $21 which would be good enough to measure the temperature of your pots and pans. I have a similar cheapo device which I once used to see if I could get my pots to 350°F for searing meat and I ran out of patience before I got to that temperature. They were stainless steel with a heavy aluminum base, not nonstick, but after 3 or 4 minutes over the highest heat my gas range could put out, they still weren't that hot. (I forget what temperature they actually reached.) Anyway, if you can shell out a few bucks you could have some fun with this, although you should probably use uncoated pans at first just in case!
posted by Quietgal at 9:28 PM on May 31, 2011

For context, deep frying of chips, for instance, is usually done at around 200C or less. Most baking is done between 140C and 210C. Stirfries, searing, etc is done at around 150-180C.

Personally, I've been around very few home cooking devices - stovetops, ovens, grills - that can maintain a temp of greater than 200C, especially not if there's a significant mass (eg, food) involved.

More worrying for teflon consumption is physcial damage to the surface - scratches and dings and such, especially since it doesn't stand up to much - cleaning technique and cooking/serving utensils will have a far greater impact (ha ha) than temperature during cooking.
posted by ysabet at 10:13 PM on May 31, 2011

Response by poster: Yeah, I prefer non-stick and stainless cookwear to cast iron. I don't like the seasoned flavor or the weight of cast iron.

I was definitely of the belief that I could never get my pans hot enough to melt the Teflon. Thanks for the breakdown on stove temps!
posted by artof.mulata at 12:17 AM on June 1, 2011

but I was talking to a salesperson at Sur La Table about this brand of cookware and she said the nonstick coating was titanium nitride

See, I heard something like that too. But it is maketing BS. The TiN is impregnated into the ceramic on the Scanpan, but then a normal PTFE (teflon based) nonstick coating is applied. Maybe the only novel thing is that they use a process that is free from PFOA, whatever that is.

See "...specifically formulated PTFE..."

I plan on getting my cancer from cellphones, so...

posted by fatllama at 6:38 AM on June 1, 2011

I'm not sure where posters in this thread are getting their data about stove temperatures, but they are certainly not realistic. It is possible to achieve 300C or 350C on a stovetop. I'm sure I'm not the only person who has melted the aluminum core of a pot or pan on an electric or gas burner (and aluminum melts at 660C). More commonly, I cook pizzas on the bottom of a cast iron pan heated to roughly 325C. I do this by leaving it on a normal electric burner, with the dial at 7, for ten minutes or so. While you aren't likely to heat your pan to 350C during normal cooking, it is certainly possible.
posted by ssg at 9:43 PM on June 1, 2011

ssg: I have a crappy stove, and I've never had a sufficiently noncrappy stove to melt a pot (even when accidentally left on for a few hours on highest setting) in 30 years. YMMV.
posted by ysabet at 6:22 PM on June 2, 2011

« Older Seeking the state of the art in location-based...   |   I'm at a loss for words, darling... Newer »
This thread is closed to new comments.