The first principle is that you must not fool yourself, and you are the easiest person to fool. — Richard P. Feynman

Friday, July 1, 2011

How does boiling work?
This post is a joint outcome from a couple classes I took in the spring term, one of which was on two-phase fluid flow and the other of which was on scientific communication. The science communication class included a project in which we were to translate a bit of technical literature to a popular science level, and I selected for this purpose the discussion of boiling processes presented in my two-phase flow course. It turns out that boiling processes are a lot more interesting than I'd expected (at least, to me), so on the off chance that anyone else might also find this interesting I've posted my writing project below. Let me know what you think!

Most of us, even those who don't cook, are probably familiar with the way that water in a heated pot on the stove changes as it boils. First it begins to steam slightly, then it begins to show a scattering of tiny bubbles on the bottom surface. Eventually a few larger bubbles form and start to drift up towards the top, and finally the whole pot becomes consumed by rapidly rising columns of bubbles, roiling the surface, steaming madly, and sometimes even overflowing the pot if we're not careful.

Engineers who deal with boiling in industrial contexts need to know far more detail than this. If you're using the heat produced by a nuclear reactor to boil water to produce steam which turns a turbine to provide electrical power, you need to know exactly how much heat the water can absorb as it boils, in order to make sure that your reactor doesn't overheat and melt down. The rate of heat absorption depends strongly on the details of the boiling process, including the type of surface the water touches as it absorbs the heat, the size of the bubbles, the rate of bubble formation, and many other subtle parameters of the system.

In order to sort it all out, scientists and engineers studying the boiling process have spent a lot of time watching the laboratory equivalent of heated pots of water. What have they learned? Well, for one thing, it turns out that a watched pot does indeed boil, eventually. It's just that most of us lack these researchers' patience in attending to the slow and subtle processes involved. If you do decide to sit down someday and pay close attention, here's what their work tells us you'll see:
  1. Convection: Convection is actually the process that takes place before the real boiling starts. When a pot of water is heated from the bottom, the water on the bottom gets hot more quickly than the water on top. Since hot water is less dense than cool water, the hot water begins to rise toward the top of the pot, while the cool water begins to sink. When the hot water reaches the top, the most energetic molecules will steam away, causing the hot water to cool. At the same time, the cool water at the bottom is in contact with the hot surface of the pot and begins to heat up. Eventually the water on top has cooled enough and the water on the bottom has heated enough that the water on the bottom again becomes less dense and the cycle continues.

    This process of convection, wherein heat is moved from the bottom of the pan to the top by the motion of the water, tends to organize itself into a multitude of tiny little hexagonal regions, each of which has cool water sinking at the edges and hot water upwelling in the center. These cells organize into a honeycomb pattern, which can sometimes be seen if there are small particles suspended in the water, such as tea leaves. This phenomenon is referred to by scientists as Rayleigh-Bénard convection, and there are dozens of cool YouTube videos you can watch to see it for yourself. Here's one example:

  2. Onset of boiling: As the bottom surface of the pan gets hotter, the water in contact with it doesn't simply heat up, it actually begins to vaporize. This forms tiny bubbles, which are usually initially trapped by surface tension in the little rough spots of the surface. These rough spots are called ``nucleation sites'', and it turns out that the chance that a bubble will form at a particular site depends on the size of the site. Nucleation sites that are too big or too small will not be able to form bubbles.

    If the surface is sufficiently smooth, such as the interior of a glass or Pyrex container, there are no nucleation sites which are the right size for bubbles to form, and the water can actually be heated significantly above its boiling point without being able to pass into the bubble formation stage. This is called superheating. When a liquid is superheated, even the slightest disturbance, such as putting in a spoon to stir it, can cause a sudden explosive rush of bubble formation, which in turn can burn you if you're splashed by the hot liquid or steam. This is why it's important to be careful if you microwave water in a glass container --- superheating of the water is not an uncommon result!

    Here's the Mythbusters' demonstration of superheating water in a microwave:

    Note that despite what they say, it's definitely possible to superheat tap water (probably depending on your local water source). I've had it happen myself.

  3. Ordinary boiling: As the pan is heated still further, the initial tiny bubbles begin to grow in size until eventually they are too buoyant for the surface tension to hold them down any longer. Bubbles begin to break free and float to the water surface, where they burst and release their trapped vapor. As the pan continues to warm, bubbles grow and escape more quickly at each nucleation site, and more and more nucleation sites become able to form bubbles. Eventually bubbles begin to form and escape so quickly that successive bubbles from the same site merge together to form amorphous vertical globs called ``slugs'', or even a continuous column of vapor rising from the nucleation site to the surface of the water.

    This is the point where a chef would consider the water to be at a ``rolling boil''. A thick cloud of steam rises from the liquid from the continuous bursting of bubbles at the surface, and the surface itself roils and churns. From a scientific perspective, this is also the point at which heat is being transferred from the stove heating element to the liquid in the pot as quickly as possible. Fortunately, kitchen stoves do not usually heat water beyond this point. Science, however, goes further yet.

  4. Transition and film boiling: If you did have a stove that could get hot enough, the next step in the boiling process would be for even the water between the columns of bubbles to begin to vaporize on the bottom surface of the pot, and the columns themselves to begin to merge. This is a dangerous regime to enter, because at this point the bottom of the pan starts to become completely covered with water vapor, and liquid water is no longer able to reach the surface. A similar phenomenon is displayed in the Leidenfrost effect, which is what allows the guy in this video to safely pour liquid nitrogen over his hand (don't try this at home!):

    This is also the phenomenon which allows water droplets to dance and skate around just above the surface of a hot skillet:

    In the ordinary boiling regime, heat can be transferred from the pot surface to the water relatively quickly by converting liquid water into water vapor, because the water vapor absorbs a great deal of energy as it forms. However, once the vapor has been formed, it absorbs energy at a much slower rate. If only vapor is in contact with the surface of the pot, the rate of heat absorption by the water will be very slow indeed. When the rate of heat transfer from the pot to the water has slowed down, the temperature of the pot itself begins to rise rapidly. If you have a powerful enough source of heat, your pot, or even the heating element itself, can get hot enough to melt. Usually this is not a desirable outcome.
Because of the dangers of transition and film boiling, engineers design systems very carefully to avoid this regime. However, the ordinary boiling regime, which is very close to the transition/film boiling danger zone, is where one finds the optimal rate of heat transfer, and so is the most desirable design target. Determining the exact relationship between system design parameters and boiling behavior is thus crucial to safe and efficient operation of steam-based power generation systems, especially nuclear reactors, as well as other systems which incorporate boiling.

I bet you never knew that a bit of kitchen science could be so important!

Citation info:
Van P. Carey (2008). Chapter 7: Pool Boiling, Section 1: Regimes of Pool Boiling Liquid Vapor Phase Change Phenomena: An Introduction to the Thermophysics of Vaporization and Condensation Processes in Heat Transfer Equipment, Second Edition Other: 978-1591690351


jfwlucy said...

I like it and I understand it but I have questions. You say the water at the bottom of the pot is vaporizing. Does that mean that the atoms of oxygen and hydrogen that make up water are separating? What is inside a bubble formed during boiling if not that? Is it air that was mixed in with the water? Can you not boil some kinds of water as easily, then?

And the honeycomb cells of the Rayleigh-Benard thing? What determines their size? Will they be similar in size whether the pan for boiling is eight inches in diameter or eight feet?

Anne C. Hanna said...

jfwlucy, apologies for my slow comment moderation. I apparently hadn't set up an email notification for moderation (fixed that now), and since this blog is so new and low-traffic I'm not really in the habit of monitoring the moderation queue!

Anyway, I'm going to promote my answers to your questions to a full post, seeing as how I'm finding myself starting to go on at sufficient length about the subject that it deserves more than a comment. It'll be up later today. ;)

Anne C. Hanna said...

Okay, here's the first half:

Post a Comment