If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Gas chromatography

Understand how to separate and purify chemicals through gas chromatography and how to interpret a gas chromatogram. By Angela Guerrero. . Created by Angela Guerrero.

Want to join the conversation?

Video transcript

Have you ever watched a TV show where to catch the criminal they take a sample of the liquid found at the crime scene, run it through this big fancy-looking box, and find out that that liquid was actually some gasoline and are able to suddenly trace the criminal back to the nearest gas station? That fancy looking box you saw is probably something that they were trying to use for gas chromatography, but in real life, gas chromatography doesn't really work like that. It's a slower process for separating out compounds that have different boiling points and a few other properties. But let's take a step back and figure out how does the gas chromatograph work. First, what you need to have is a place to inject your sample. Even though you'll be injecting it as a liquid, what happens is it gets to this box, and it gets vaporized into a gas. When it's in a gas, let's say that this particular mixture was made up of two different kinds of gas. I'll show that as some green dots and some orange gas particles. You can't really see these though, because usually the amount you're injecting is so small, on the order of microliters, in fact. And in gas chromatography, we've talked about how the mobile phase is a gas, which means that you need to have an inert carrier gas to push these through. And it's important that this is inert, because you don't want it to react with whatever it is that you're trying to separate. Once it's passed through that, it'll get heated up and then go through a long tube. In order to make it fit into the box, they usually just coil a long length of tube, and the longer the tube, the better separation you'll get. And once it's finished passing through the tube, there needs to be some kind of detector that picks up how many particles of the green compound were found versus how many particles of the orange compound reached it. And they'll be reaching the detector at different rates, which I'll explain shortly. From there, the detector will be able to take these signals and display them in a way that you can analyze on your computer. Often what you'll get is something that looks like this. This is known as a chromatogram, which is just a way of saying, a graph for gas chromatography, and we'll also be explaining this later on. So to recap, we injected our liquid sample, which was vaporized into gas, then it joined up with the stream of inert gas that was already flowing and was pushed onto the long column. But what's going on inside that coiled column? Let's take a closer look. Pretend that this is stretched out, just into a straight column that's horizontal. It has some liquid coating on the side, because the liquid is serving as the stationary phase as the gas is rushing through it. And what you would observe, perhaps, is something that resembles this. You might see the green dots kind of hanging out on the sides, while the orange dots are clustering more in the middle and maybe even traveling a little bit further than the green dots have. What does that mean? Well, we can't really imply too much from that yet, so let's watch it for a little bit longer. At the next time point, what you might see is that, again, the green dot, or the green compound, kind of staying more to the sides. It's traveled a little bit farther now, but this orange compound has gotten pushed all the way over here, the point that it's almost at the detector already. You can already tell that the orange one is going to reach the detector first, meaning it will produce the first peak. This would probably correspond to this on the graph. But wait, what's that tiny peak next to it? Usually, that represents the solvent that you dissolved your compound in. That solvent is usually something with a pretty low boiling point, so it gets pushed through first. But the second peak that's bigger is usually the first peak that actually represents a compound in your mixture. So that last peak you see probably represents this green compound. But why are they coming out at such different rates? What's the reason for this? And one of the reasons is that in chromatography, it's always an interaction between the two phases. Here it's the vapor phase, or the gas phase, with the liquid phase, also known as the stationary phase. So compounds like this orange one that move really fast, really, really like to interact with the gas. And this is because they probably have pretty low boiling points and are vaporized really readily. Whereas compounds like the green one might have higher boiling points, and prefer to spend their time in the liquid phase, and are not quite as ready to go into the gas phase as the compounds like the orange compound that have lower boiling points. So separation by boiling points is a big part of how gas chromatography works. But wait, there's actually a few other things. What if the green and orange compound had more similar boiling points? Could you still distinguish them? Actually, you could. Let's take another example. If instead originally what you had was something that looks like this, where you had these tiny pink dots, those represent tiny pink particles, along with large purple particles. Again, these have the same boiling point, but why is it that it looks like the pink one is getting carried farther by the gas? That's because it's really small. So just based on its size, what would happen next is you'd see something very similar to what we saw in the second image before, where the purple dot hasn't traveled very much, but the small pink ones are just going so fast, they're almost at the detector. The way to visualize this is, imagine the gas pushing through. Now picture that as a really strong wind. If you have a tiny child in a meadow where there's a strong wind, the kid will feel like they're getting pushed around pretty hard. But if you had a big Sumo wrestler instead, they probably wouldn't move too much no matter how hard the wind blew. So in this case, the pink dot's like the child, and the purple one's like the Sumo wrestler. So we've talked about the size of the particles, or the molecular weight of the compound, along with the boiling points as being ways to discern between compounds in gas chromatography. But let's take a closer look at that chromatogram you see on the computer screen. That chromatogram is actually a plot of intensity on the y-axis, representing how many particles are hitting the detector at a time, versus time on the x-axis shown here. So again, we saw something that looked like this. We said that the very first peak that comes out is probably just the solvent that your sample was originally dissolved in. The next one represented our first actual peak, and it represented again the compound that traveled faster and further. Let's call this compound A. The second peak was the one that was a little bit slower, so compound B. Just by looking at this chromatograph, we can already know a little bit about the relative properties of A versus B. Again, compound A was probably smaller and had a lower boiling point, whereas compound B was probably bigger and had a higher boiling point. But that still doesn't tell us anything about the identities of these exact compounds. What you would really need to do in lab is first run a reference, meaning that earlier you could have run a graph that looked like this and got two peaks. And if you knew that your reference sample was a sample of hexane, and it looked like they came out at about the same time as compound A, you could probably infer that compound A is hexane. Although, it's not quite definitive, which is why gas chromatography is usually coupled with other analytical techniques that can give you even more information about the compound. For example, techniques like mass spectrometry can tell you about the molecular weight, so that makes it even easier to narrow down what the exact compound is. And I know that this can be a pretty tricky process to figure out what the compound is, but for analyzing these GC graphs, what you'll mostly want to look at is the relative difference between the peaks and try to compare compounds qualitatively. Quantitatively, you can also note that the area of each peak is directly proportional to the amount of compound in the mixture. So next time you see on TV that they're trying to use GC, you'll really know what actually goes into it and that you really can't catch a criminal quite that quickly using only this. You'd need to use a lot of other lab techniques.