How Does This Candle Suck Water Up Into a Glass?

Photo: Rhett Allain

It’s a classic demo and one that you can try yourself. You light a candle and put it in some water (I used blue dye in my water). Next, cover the candle with a glass. As the flame burns out, the water gets “sucked” up into the glass. It’s pretty cool.

Let me start off with the “sucked” part. No, the water isn’t actually sucked up. Actually, the pressure inside the glass is reduced while the pressure outside the glass (due to the atmosphere) stays constant. Since the outside pressure is greater than the inside pressure, the water gets pushed up into the glass.

Check it out.

But why does the pressure inside the glass decrease? It’s mostly due to the chemical reaction between the wax and oxygen. We often call this type of reaction “burning stuff”.

The key to most fire reactions is the oxygen in the air reacting with some type of carbon. Technically, this is called a combustion reaction — but I like to call it fire. Also, it can get pretty complicated to fully explore combustion of wood and wax, so let’s consider a very simple version — burning methane. The main part of the reaction is the same.

If you look at the reaction of methane and oxygen, you get the following.

This says that one molecule of methane plus 2 molecules of oxygen produce one molecule of carbon dioxide and 2 water molecules.

So, let’s consider the gas in the glass before the combustion. It’s approximately 79 percent nitrogen gas and 21 percent oxygen. Then the fire consumes oxygen and produces carbon dioxide and water (vapor). Eventually, there’s not enough oxygen in the glass and the fire goes out.

Since we are talking pressure, we need to talk about some other stuff — the Ideal Gas Law. This model assumes there are bunch of gas particles that are spaced out enough that you can treat them as particle that just bounce off each other. With that, there are 4 things that can change:

  • The volume of gas (V).
  • The temperature of the gas (T).
  • The gas pressure (P).
  • The amount of gas particles (in moles — n).

Putting this together, you have the Ideal Gas Law:

The R is a constant — perhaps not surprisingly called “the gas constant”. It has a value of 8.314 J/K*mol (but that’s not super important right now).

But what about the total number of gas particles in the container? If you take two oxygen molecules and produce 3 more particles (carbon dioxide plus two waters) then it seems like you would INCREASE the pressure. Right?

Ah ha! But the key here is water. Water produced as a gas (water vapor) can easily condense on surfaces to become liquid water. When this happens, you are just left with the carbon dioxide and an overall fewer number of particles in the glass container to lower the pressure. At least, this is what I think happens.

How about some actual data? Yup, that’s what I’m going to do. Instead of putting a glass over a container with water, I’m just going to seal it up with a stopper. Check it out.

Photo: Rhett Allain

In the top stopper, I actually have two sensors — both connected to a Vernier LabQuest data collection device. With this, I can measure both the pressure and the temperature inside the container with the candle burning.

Here is the data. Check it out.

So, you can see that the temperature increases at first (since there’s a fire in the container). But after the fire goes out, the temperature decreases as the gas cools. The pressure reading increases when the stopper is placed on top of the container and initially increases as carbon dioxide and water vapor are produced.

After a short while, the water vapor condenses and causes the pressure to decrease. But wait! The temperature is also decreasing — so is that why the pressure decreases (according to the Ideal Gas Law)?

How about this? What if I take the ratio of pressure divided by temperature? From the Ideal Gas Law, this would be:

If I plot P/T vs. time, we can see something about the number of molecules in the container (assuming the volume stays constant). Here’s what that looks like.

If the number of gas particles was constant, this should be a horizontal line. But it looks like it decreases right away. I’m going to assume this means the water vapor begins condensation right away. But why does the number of particles increase after about 25 seconds? I’m not sure, but it’s possible the container could be leaking and letting air inside. That’s just my first guess.

How about another experiment? What if I burn iron? Well, in this case I’ll use steel wool. When iron reacts with oxygen, you get the following balanced equation.

So, iron (Fe) plus oxygen produces rust. But in this case, there is no water vapor and not even any other gases produced. If I light some steel wool on fire and put it in the container, I get this.

This also decreases in the number of gas particles — but it seems to be a much quicker decrease. I assume that it’s because there isn’t any water that needs to condense. Oh, just a note: I had a tough time getting the steel wool to stay on fire. I think this was because it was old and already partially oxidized. But still, it’s clearly decreasing the amount of gas in the container.

OK, one more comment. There’s this idea that in the water-candle experiment, the water gets sucked up mostly because of the cooling air after the fire went out. The explanation goes like this:

  • The candle heats the air and the air expands.
  • Some of this expanding air escapes the container and leaves through the water (probably making bubbles).
  • After the fire goes out, the air cools and the volume decreases causing the water to rise in the container.

I’m pretty sure this fire model doesn’t work. If this was true, the closed container would see a large increase in pressure followed by a decrease back to the original pressure.

Physics faculty, science blogger of all things geek. Technical Consultant for CBS MacGyver and MythBusters. Former WIRED blogger.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store