Preparing Headspace Vials

[GHGirl]

Hi All,

We are using 10mL headspace vials with gray butyl septa (boiled prior to use to clean and prevent leaking) and capped with aluminum crimps to store 12mL atmospheric gas samples. We will be measuring CH4, N2O, and CO2 on a GC with an FID, ECD, and TCD, respectively. Our GC carrier is He.

As of now (pre-experiment trials), we've been evacuating the vials prior to sampling. Unsurprisingly, even in new vials, our He "blanks" still have detectable N2 and CO2. If we reuse vials that have acetylene in them previously (part of our sample matrix) by reevacuating them, the C2H2 peak shows up in He "blanks".

So, what is the best way to prepare my headspace vials to avoid contamination from the air and how do I clean my vials between uses? Is there a recommended method for prepping vials? Should they be washed between uses? Because I'm already dealing with trace gases, I need to avoid diluting my samples (i.e., injecting 12mL of sample into a vial flushed with 10mL of inert gas (He, N) is not an option).

Thanks in advance,
GHG Girl

[Don Shelly]

How do you evacuate all air from the vials? I would think a perfect vacuum would be difficult to achieve?

[GHGirl]

Well, that's the problem. A perfect vacuum is not really possible, which is why we're seeing contamination from both lab air and any previous sample in the vial.

We avoid evacuating by hand, which can be done by essentially sucking the air out of the vial (using a 30mL B-D syringe with a stopcock) over and over again until the syringe just won't pull any more air out. Since this has to be done repeatedly for each vial, a) we puncture the septa multiple times, which is not optimal, b) it is time consuming, c) with 100's of vials for each experiment we run, we'll all have carpal tunnel in a year.

So, I've built an evacuator, which is a vacuum pump (Edwards Model No. RV12) outfitted with a tygon hose, to which 8 gas-tight valves and gas-tight needles are attached. All the parts used are either vacuum rated or gas-tight. We also have a vacuum gauge, which helps us monitor the pressure to which we evacuate the vials. Besides saving us time and strength, the nice thing with the evacuator is that the vials always reach a constant vacuum (120 millitorrs) and it's really obvious if a vial has a bad crimp or a leaky septa. However, even this strong vacuum is not enough to evacuate the vials completely.

[richiekichi]

Can you not just subtract a blank from from your samples?

Your evacuator sounds epic by the way.

[Peter Apps]

As you have already demonstrated there is no way to get all the air out just by pulling a vacuum on a sealed vial (and with a vacuum inside air will leak in through the hole in the septum anyway.

Can you rig up a system that will pull a vacuum, fill with helium, pull a vacuum, fill with helium over multipe cycles ? A two-way valve to vacuum on one leg, helium on the other and the septum needle on the common connection would do it. Multiple evacuation - fill cylces are the standard way to clean gas cylinders. Because the vial ends up pressurised with helium to 1 bar air leaks are limited to diffusion rather than bulk flow.

You could also consider heating the vials to displace anything adsorbed to the glass, although the main surface contamination will always be on the underside of the septum.

Peter

[Johnny Rod]

Can you not do direct gas injection througha valve? I can't see you'll have great results trying to avoid atmospheric gases with your method to be honest. Where do the samples come from?

[GHGirl]

Thanks for all the responses!

Peter, great idea about rigging the evacuator to a purge system! I've been talking to one of my colleagues/engineering genius, who thinks this is completely possible with just a helium tank and a three-way valve. I'm excited to get it setup, and I'll certainly post with the outcome.

Johnny, my samples are coming from chambers, and unfortunately, my experimental setup prevents me from making direct injections into the GC. Essentially, I have small PVC tubes (40 cm in length) that have septa installed in the walls and o-rings on either end. I use the PVC tube in the field to collect a 10 cm soil core (from wetlands, if you're curious). Once the soil core is inside the tube, I cap both ends of the tube. Air pressure is equalized by inserting a needle into the septa of the caps while capping. The PVC chambers with the intact soil core are transported to the lab, where they undergo various experimental treatments (we add nitrogen to some, incubate them at various temperatures, inject acetylene into the soil to inhibit complete transformation of NO3 to N2, etc.). We begin collecting 12 mL gas samples (from the headspace of the chamber using gas-tight syringes) when the experimental conditions are introduced (Time 0) and incrementally for 36 hours. The 12 mL gas sample from the syringe is injected into the 10 mL headspace vials. We typically collect 8 samples from each chamber. At a minimum, we're dealing with 8 of these chambers at once - sometimes up to 32 if I have enough help. So, I collect between 64 and 256 samples over a 36 hour period, not including field check standards and field blanks. We analyze the samples for CH4, CO2, and N2O on an Agilent 7890A GC. Fortunately, I've graduated from sleeping in front of the GC to inject samples, and I now have a 7697A Headspace Sampler!

Thanks again everyone,
Greenhouse Gas Girl

[Johnny Rod]

Tricky one, I would have thought FTIR was a better technique unless there are other things going on that prevent this. Peter's idea is a good one though, can you also purge several times with the sample gas? I would say CO2 is your biggest problem, methane and nitrous aren't so widespread in air.

[GHGirl]

Hi Johnny,

I'm not very familiar with FTIR, although using an FID, ECD, and TCD is the standard for analyzing samples from this particular chamber/incubation method. I wish we could purge our vials several times with the sample gas, but I'm afraid that pulling more than 12 mL of sample out of the chamber headspace would result in too much headspace turnover over the incubation time - something I've been careful to avoid.

I appreciate the thoughts though!
Best,
GHGirl

[Peter Apps]

If you are filling the vials from a syringe you will need to suck out the helium fill just before you inject the sample, otherwise you end up with 2 bar pressure if you can push the syringe plunger hard enough, and a sample that is diluted 1:1 with helium, and which will leak out through the needle hole in the septum.

I general, having a needle hole through the septum before you load samples to the headspacer will allow leaks when the vials are pressurized by the headspacer. Do you have access to a glove box or other way of capping the vials in an inert gas atmosphere ?

Peter

[Johnny Rod]

Can you leave the vial uncapped in the chamber, or connected to it, until you need to take the sample, then cap/disconnect it?