Hydrogen carrier on ECD and increasing responses

[sllimbri65]

I am in the process of validating pest/pcb methods using hydrogen as a carrier gas. i am running a 5890 without epc and an ecd detector. Argon/methane is the makeup gas. The day following a pest or pcb calibration, both of which have %rsds below 10%, the midpoint cal standard quantitates over 20% high. Only stds or clean hexane rinses run after the calibration and the liner is changed. We are using Restek Sky liners to keep breakdown low [about 5%]. Consistent injections volumes of 2ul were verified. We don't have this problem when using helium carrier. I can't figure this out. Any ideas? Thanks.

[jss37992]

Hydrogen will take a higher flow to get the same resolution as helium. If you are using the same flow with the hydrogen as you were with the helium it could be possible that everything injected onto the column could be sticking around and building up. This could explain your high responses. What ID column are you using? Also, are you using 4mmID inlet liners?

[sllimbri65]

At my current carrier setting of 65cm/sec at 120 degrees, the retention times are a bit later with hydrogen than they were with helium. I'm using a .32 diameter Restek CLP column, that hasn't changed. The liners are 4mm. Exactly where could things be sticking around and building up? The injection port? The column? We run hexane blanks that are clean, no peaks to speak of.

[Peter Apps]

At my current carrier setting of 65cm/sec at 120 degrees, the retention times are a bit later with hydrogen than they were with helium. .

That's odd. 65 cm is an optimum flow for hydrogen, but if you were running helium at optimum (35 - 40cm/s) then the peaks should come out much earlier with hydrogen.

Peter

[Stunt]

Let me guess: Your pesticides, specifically, heptachlor, any HCH compounds, endosulfans, and 4,4'-DDT are >20%, but your PCBs are fairly good? It isnt the hydrogen (per se). Inject a couple of dirty samples, or if you dont have any, high standards prior to your first degradation check ( before the ICAL) then run a new ICAL. If your system has been down for a while, or (in your case) you change gases, active sites in the injection port will wreak havoc on your percent diffs. The high concentration samples/standards will fill those active sites. Also, you will want a 0.25 mm column, (and I would switch to a DB-5 and use the CLP-1 for confirmation) not a 0.32 if you are using hydrogen. Also try a 4mm gooseneck liner, and personally, i don't care for the sky or siltek liners. Degradation should be <5% with uncoated liners if inst maint is performed adequately. If deg is climbing, maintain, or drop your inlet temp a few degrees.

I've been using hydrogen since 1990 with my ECDs and I suffer the same issue when the instrument sits idle for some time.

Insofar as the RTs, Peter is correct, the RTs should be significantly earlier with hydrogen if you are using the same column diameter and film thickness.

[sllimbri65]

Stunt,
Thanks for the response. So you think if we run high standards prior to the ical, they will stabilize the active sites?

[sllimbri65]

I was thinking of flushing the split vent line. Do you think that and a good scrub of the injection port would be worth a try? Also, I am using a single taper liner with wool. Do you confirm this is best? This instrument is quite old. I could not get breakdown under 15 with uncoated liners.

[jss37992]

Wool is great for PCB's but not so much with pesticides. Glass wool can contain too many active sites which can lead to poor calibrations and CCV recoveries.

[Stunt]

JSS is correct, get rid of your wool for pesticide analysis. The wool will ruin your percent diff for endrin compounds, especially aldehyde. Lightly scrub the injection port with a stainless wire brush (Restek cat# 20112, lists at $24), then use some Solon q-tips dipped in hexane as a finish. Two Solon q-tips at the same time fit tightly into the port. Do not use brass wire brushes! In regards to your re-post question: Yes. Flood your system with some high concentration standards, then run a cleaner solvent, then your ICAL. Cleaning the split vent is always a good idea as well.

Degradation can be a function of your column as well as thermal breakdown. If you are running a .32 mm column with hydrogen, your breakdown could climb. Try a .25 mm column, and I find DB-5's phase reduces breakdown for a somewhat longer period of time than the CLP-1. Nasty samples can also induce degradation.