Drug Identification & Chemical Testing for DWI Defense
Headspace Gas Chromatography
Kevin Leckerman: Why use headspace gas chromatography? Well, it is the best method for testing for volatile organic compounds, such as alcohol. It can be used to test inorganic and organic compounds – like a teacher said to me one time, “If you can smell it, you can test it” using gas chromatography.
Headspace gas chromatography is a little bit different than just regular gas chromatography. What’s the purpose of gas chromatography? Now, there’s one thing you have to do which is the total basis for it: separation of compounds, separation of molecules. When blood is tested, it’s going to have a number of different compounds in it. What you have to be able to do is separate and identify each one of those. If through the testing process they cannot separate, then that’s not good. And you can’t quantitate it. So if you can’t separate two compounds, the machine is going to read them together and potentially identify both of the compounds as one and give you a quantity for ethanol when it could be mixed with another type of compound and if you separate them, it’s going to be much lower than a situation where they’re not separated.
Separation occurs by boiling. Essentially, for each compound, they each have their own boiling point. You can heat each compound warm enough and it causes a vapor, and then the vapor is shot through the gas chromatograph. There should be separation that’s caused and that’s how you started to determine if this is alcohol.
Here’s Henry’s Law, and this is essentially what’s used. Once the vial of blood comes into the lab, what the lab chemist does is actually takes a small portion of that ten milliliter vial, tests that by putting it into another little vial, puts another chemical in there, caps it off, and then puts it into the carousel. The machine heats it up, and what happens is Henry’s Law. I’ll give you the definition: at a constant temperature, the amount that a given gas dissolves and a given type of volume of liquid is directly proportional to partial pressure of that gas in equilibrium with that liquid. All right, what does that mean?
Think of a bottle of soda. When you look at an unopened bottle of soda, you can’t see the bubbles in it. There’s constant pressure that is keeping the gas – the carbon dioxide – at equilibrium, meaning there’s an equal amount in the air above to the molecules in the liquid itself. If you keep the bottle out at room temperature and then you take off the cap, what happens? You see the bubbles, right? They just come up. That’s essentially the process of Henry’s law. If you put a balloon over the bottle, the balloon fills with the carbon dioxide because it’s all escaping. Essentially, the warmer it gets, the more the carbon dioxide and the liquid want to escape into the air above it.
That’s essentially what’s happening with headspace gas chromatography. You’re not testing the actual liquid ethanol; you’re testing the vapor. You’re heating up the molecules in the liquid so those molecules escape into the gas above, you draw out the gas, and you test that gas or the vapor.
So here’s the device and the internal components. You would eject the gas here, and then there’s the gas that pushes it through what’s called a column. It goes out through the machine. There’s a flame ionization detector that burns it up, and then essentially records at what time it’s coming out of the machine and how much of that compound is there. Then it goes to the computer and the computer produces what’s called a chromatogram and you look at that; it gives you the numbers and tells you what the compound is. The carrier gas that’s used is typically helium or nitrogen; it’s inert, so it’s not going to interfere with the testing process.
Kevin Leckerman: Generally in these labs what they do is auto-pipetting. They’re not doing it by hand, but I have a hand pipette right here. The lab chemist puts the pipette into the vial, takes out a certain amount – it’s going t be one milliliter on here – and then puts it into the other vial. Now, as you can see, there are two actions on this. There’s down and then you have to go further down, so if you’re hand pipetting that could be a good source of error, if they’re not putting enough blood into it or the internal standard, which we’ll get into. You have these auto-pipettes where you just push a button and it pumps the stuff out.
Kevin Leckerman: There are lab protocols that need to be followed. The vial itself has a barcode on it and that’s going to be used to identify when it’s being tested. So like I said, there’s going to be this run of a whole bunch of unknown samples. You don’t know what’s in the sample; they’re called unknowns. The vials all have barcodes on them, and that’s typed into the computer. Each little vial is placed into a part on the carousel, given a number, typed into the computer, and then correlated to the actual barcode.
It’s kind of hard to see, but here’s the vial. Here is the auto-pipetter, which is connected to an internal standard and going right into the blood sample. The lab chemist here is pushing the button.
Something that you should know that’s important is cleaning. Look at the standard operating procedures for the laboratories – how often do they clean their vials? Obviously if they’re not disposing them, they should have a method of cleaning each vial that’s used, otherwise there are going to be remnants of somebody else’s blood. The same goes for the pipettes.
Like I said before, usually they’re taking one milliliter of blood and it’s called an aliquot. If you’re taking more than one milliliter, you’re ruining the entire process. If you’re taking less than one milliliter, you’re ruining the entire process. It has to be precise: one milliliter. The same goes for the internal standard; there has to be a particular amount of the internal standard.
Here are these little vials. They’re capped with a septum on top of it. A needle goes down and takes out the headspace – the vapor – and sucks it into the machine. This whole thing could be filled with hundreds of vials; as long as it’s typed into the computer properly there shouldn’t be a problem. You see how the numbers correlate to each vial.
There’s something called an injector. As I showed you before, the vapor gets injected into the machine. It’s a place for the introduction of the sample and it’s at this boiling point.
Here’s the port; it goes right into there and the carrier gas comes in at the same time, pushing the sample through the machine. You can see that there’s this rubber septum, and once you lift it up, below there’s this glass sleeve that the vapor goes into. There’s an O-ring on top of that. This glass sleeve, the injector cap, the septum: they all have to be clean. There could be contamination in that process.
I was explaining this to another attorney. He said something interesting to me that stuck with me: “You have to make sure there’s no carryover or contamination from somebody else’s blood sample, because you don’t want another guy’s booze in your client’s blood sample.” That’s the case. After a while, these get dirty; there are contaminants that get stuck to it and it can carry over into subsequent tests. There’s a way to prevent that to make sure it doesn’t happen.
Student: Are they supposed to document when they did that so you can find out if they did it properly?
Kevin Leckerman: They sure are, and that’s what you’re asking for: how often are they maintaining and cleaning these components? The same goes for this gold seal, which is at the bottom of the injection port.
Anatomy of Gas Chromatography
Kevin Leckerman: There’s this column. Vapor goes in, carrying gas pushes it around the column. And here’s the column: it’s a very thin, long wire that’s wrapped around inside of what’s called the oven, and that’s heated. Here’s the injector port. The molecules come in, and they go around and around and around and around and then out. There’s a circulating fan that is keeping this oven at a constant temperature. The column gets hung right there and there’s insulation there as well.
What’s the reason for the heat? Why does it have to remain at a certain temperature? It has to remain at a constant temperature all the time because otherwise the compounds just sit there. The hotter they get, the faster they want to move and they move out of the machine. The temperature is essential and you have to keep that temperature constant.
So here’s the capillary column. There’s the outside of the column and then there’s the stationary phase. The compounds go in, they go through this column, and inside the column there’s this liquid. What it does is it slows down each compound. I know this is getting a little bit complicated, but basically each compound has a certain affinity for specific column phases. The lab has to get the right phase, meaning there are certain columns that are not good for ethanol testing. They’re good for other things, but not for ethanol. You’ve got to get one that’s actually good for ethanol – one that ethanol likes. So the compound goes through, and the ethanol is going to hang out in this stationary phase for a certain period of time. And it would sit there forever, unless the carrier gas finally pushed it through and it was heated up to a certain temperature. For each column, there’s supposed to be a known period of time that this compound stays there. So ethanol is going to be there for, let’s say, one minute and thirty seconds. Isopropanol is going to be there for one minute and fifty seconds. That is how we figure out what the compound is. We’re going to find out when it’s coming out of the capillary column.
As I was saying before, there’s a moving phase and there’s a stationary phase. The compounds go in and they sit there in the stationary phase, and they all come out after a certain period of time. If it comes out at one minute and thirty seconds, then you know it’s ethanol. It’s called elution – it’s eluting at a certain period of time. There’s a known retention time – it’s retained in that column for a certain period of time and then it elutes or comes out.
Kevin Leckerman: Here’s an important part about how the science has to work. It’s just a machine – it’s a stupid machine. It knows nothing until you teach the machine what ethanol is. You have to introduce to the machine what ethanol is by using what’s called a certified reference material. Certified reference materials are something that you get from the National Institute of Science and Technology (NIST), and it’s NIST-traceable. That means there are documents that come with it that say, “We guarantee that this is ethanol. There’s no problem with this. You can put it into the machine, and when it elutes and the detector finds it, we’re guaranteeing that it is ethanol.” If it’s NIST-traceable, that’s the highest standard. You can teach the machine what ethanol is and when it comes out. One minute and thirty seconds, all right, we know this is ethanol. Now, anything that comes out of this machine at one minute and thirty seconds is going to be ethanol (most likely).
And then, you want to create what’s called a calibration curve. How do we know what’s 0.01 and what’s 0.15? We have to teach the machine what it looks like to have a 0.15 BAC, what it looks like to have a 0.05 BAC. It creates a curve. (Well, it doesn’t actually curve; it’s a straight line.) You should create at least three data points, which should be, like, and 0.05, a 0.10, and a 0.20. But you can have five data points, ten data points – as many as you want to make sure it’s as accurate as possible.
There’s something called a separation mix, which means teaching the machine to identify other compounds that are typically coming out in human blood. Maybe there’s isopropanol. Maybe there’s methanol, acetaldehyde. You’ve got to teach the machine to be able to differentiate between all of those compounds. You’re introducing it to the machine as standard and you’re teaching the machine, “Hey, this is what acetaldehyde looks like. This is what isopropanol looks like.” And if there’s proper separation, you’re going to see the difference. Isopropanol is going to come out at one minute and fifty seconds and acetaldehyde’s two minutes, let’s say, or thirty seconds – whatever the column does.
And then you want to do a verifier. So before each one, you want to put in a control – actually put in a known ethanol – make sure the machine is doing its job properly and it can detect ethanol. You put in all the unknowns – all the defendants’ samples – you test them, and at the end you have a verifier. You put in another know sample to see, at the end of this run, can the machine tell where ethanol’s come in? The best way to do this is to run dual-column testing, so you actually have two columns set up in the machine as opposed to one. That’s the best way to do it; most labs do not do it that way.
A simple saying is “garbage in, garbage out.” If the lab can’t prove that they did this properly – if they didn’t set up the calibration properly, if they didn’t do the controls, if they didn’t do the verifiers – it’s not a good test and you can’t say to a degree of scientific reliability that this is actually alcohol.
Here’s an example of calibration curve at 0.10, 0.15, 0.20, 0.30, and 0.35. Can anybody tell me what the problem would be if you didn’t do a calibration curve that went down to, let’s say, a 1.0?
Student: Well, because the limit is 0.08.
Kevin Leckerman: Exactly. They’re going to be able to identify it, but how are they going to reliably tell you the quantity of ethanol of 0.08 if they don’t have a calibration point that’s below an 0.08? And then you have for commercial truck drivers it’s an 0.04, and for minors it’s an 0.01. So you’ve got to go down to at least an 0.01 and up to at least a 1.5, because that’s where ignition interlock devices start to come into play.
The calibration curve should be done before each run. You don’t use an historical curve. You don’t go back to the one you did a week ago to quantitate today. No, you’ve got to do it each time before the run.