552.2 Haloacetic Acids : Matrix interference in samples

[greendrake]

I am running Haloacetic acids on a 6890 micro ECD by method 552.2. I am using Rtx-CLPesticide and Rtx-CLPesticide 2 columns. I also run PCBs, pesticides and herbicides on this instrument so I don't have flexability with the columns. My calibration extracts well, however I frequently run into matrix interference problems that affect my IS from real world samples. I also have matrix co-elutions with my DCAA on one column and my MCAA is a split peak. Any suggestions? Any ideas for temerature programs? Injection temps? Thanks.

[bluejay]

A couple of issues (not specifically matrix related) I remember working through with the HAA method were contamination in the sodium sulfate (bake it in a muffle furnace at 400C for 4 hours) and contamination in the sulfuric acid (some brands were OK, others weren't). Sufficient syringe rinsing during the extraction process was also critical.

We also had to do a pretty high temperature ramp at the end of each sample analysis to bake out the column.

I've also known a couple of labs that had an interferant with their IS peak and had to change to a different compound for the IS.

I'm not sure how running PCBs etc would affect HAA analysis, we only ran HAA5s and THMs on those columns.

What GC conditions are you using?

[greendrake]

We are currently baking our sodium sulfate. The calibration standards are extracted along with the samples using all the same reagents and do not exhibit the same contamination issues so I don't think it is the supplies. The contamination only shows on samples. The MCAA split peak is a problem with the calibration also though.

The instrument conditions are: Inlet temp: 240C, 1uL injection, total flow 31.6ml/min, initial oven temp 45C and hold for 4min. then ramp at 6C/min to 140C then ramp 25C/min to a final temp of 225C.

What was the alternate IS? Does the drinking water method allow for a substituted IS?

[bluejay]

MCAA split peak - I'm not too sure about the potential causes of that, but I would check that your liner volume is sufficient for your injection of 1uL of extract (backflash), maybe perform inlet maintenance and re-install the column. What range are you calibrating?

I don't know if it would affect MCAA analysis, but another thing we occasionally ran into was inadvertantly getting a tiny bit of water into the gc vial when transferring the extract. This really messed with things, and it was easier to see if you put the vials in the freezer for awhile so the water would freeze and you could see the ice crystals.

Alternate IS - whether using a different IS compound is acceptable will depend on what state you're in and how your state auditor feels about it. I'll see if I can find out what an alternate compound might be.

Has this method run OK in the past, or has it always had these issues? Do all your samples have the matrix interference issues?

[greendrake]

I checked and the liner volume is sufficient. The calibration range for MCAA is 0.4ug/L to 120ug/L.

That's a good idea about putting the extracts in the freezer, I will try that.

I have always had the split MCAA problem. The DCAA problem is weird because as I do instrument maintenance and cut column, the matrix co-elution separates a little. The retention time for the DCAA gets smaller but it seems the matrix peak stays where it was and they start to separate. By then my column is getting too short and it is about time to change out columns.

Most of the samples give me some matrix problems around the IS retention time. I will occassionally get some that don't but that's not the rule. We have been running HAA5s for a little over a year and these problems have always plagued us.

Thanks so far for your help!

[bluejay]

The alternate IS that I know one lab uses is Hexachlorobutadiene. I'm not sure what concentration they use.

Calibration range should be fine, we ran 1ug/L to 80ug/L with no problems on the same 6890/uECD combo.

Have you determined if the split MCAA peak is indeed MCAA as a split peak, or if it is a interferant/contaminant coeluting with MCAA?

I'm assuming you run a reagent water blank through the extraction/derivatization procedure and that it does not show the interferance with the IS or DCAA. I suppose there could be something showing up from the sample bottles (cleaning or sampling procedure) or the preservative.

[JReed]

I see this is a fairly old topic but I'm new on here and I run HAAs via 552.2 so I'd like to talk to fellow sufferers!
I have the same problem with the IS co-eluting, but 9 times out of 10 it's only happening on my confirmation column. Sometimes there is enough of a dip between peaks to split it, but mostly not. But still the only time that's a problem for me is if I have a CCV or RLS failure on the first column that I can't report. I'm assuming you're using 1,2,3 trichloropropane? I've looked into different columns and run programs but haven't had any luck. What is your run time?
My MCAA doesn't split exactly, but there is a dip right after it, which sometimes makes it difficult to see where it should be integrated. I also sometimes get a small hump between MBAA and DCAA.
I run on a 5890 II, and have a current problem that I posted on today. If you have a chance to take a look I'd appreciate it. Thanks!

[bluejay]

(It's been about 2 years since I've run HAA's)
For what it's worth,
I ran 552.2 on a DB-1 column, same one used for 551.1 THMs.
I consistently got a dip in the baseline after the MCAA peak. To account for it, I used integration parameters that always integrated it the same way and it worked OK.

[JReed]

Good to know. Thanks for your input!

[greendrake]

It sounds like we are experiencing similar problems with our IS. I often get interference on my confirmation column, but only in my samples not my cal or CCVs. I started adding twice as much IS to compensate for the interference. It seemed to help. The interference is now a smaller percentage of the IS peak and the IS recovery can fall into acceptance ranges. My total run time is 23.23 minutes. I am running on a 6890N. I am using a single gooseneck liner. I tried using a cyclo double gooseneck and it didn't seem to make a difference. My initial temp is 45C and my inlet is 240C. Have you had any luck adjusting your inlet temps? How long is your run and what columns are you using? What I originally thought was a split peak on my MCAA turns out to be a coelution with a matrix bump. I can usually get enough separation to split this when necessary. Also, I get an interference bump on my DCAA resulting in 50% higher results than my confirmation column. Are you performing the extractions or does someone else do them for you? Thanks

[JReed]

My run time is 26.50, but I'm on a 5890. I haven't tried adjusting temperatures yet, but I'm sure that's next. I'm using a 2mm single goose as well. Initial temp is 40 and inlet is 250. My columns are RTXCLP (30m, 0.32mmid, 0.5umdf) and RTXCLP2 (same specs.) I get a sort of bumpy tail on my DCAA that I can usually split off. In this lab we have a prep department for semi-volitiles, but I came from there and this was one of the methods I would extract a ton of so I am very familiar with the process.