-
- Posts: 3
- Joined: Mon Mar 30, 2020 3:25 pm
I've been running EPA Method TO-15 for chlorinated VOCs associated with soil vapor intrusion into indoor spaces on a TD-GC-MS without any serious persistent problems for the better part of two years now. However, as of late I've started experiencing a very major problem with ISTD response that I haven't seen before. Most of the tricks I use to fix GC-MS problems and troubleshoot haven't helped so far.
I realize this isn't a typical matrix and method that is seen here (at least judging by the topics from the past two years), but I am open to input/help/suggestions from anybody who has ever had a similar issue in their analytical careers. Below I'll present my problem, followed by my hardware and general settings as well as some context.
I'm seeing the ISTD response vary somewhat wildly, with an overall decreasing trend, over the course of 10 diagnostic CCVs around the midpoint of my usual calibration range. I've tabulated this data and I've seen RSDs around 20% for my ISTD responses, which has proven to be quite troublesome when calibrating the instrument for low RSDs as required by the method (the method requires RSDs to be < 30%). What's even stranger is that analyte response for these CCVs is very stable--the RSD's are around 3%. This leads me to believe that there isn't a problem with the source (which I have cleaned and whose filaments I replaced just a week ago), but rather with the ISTD or maybe somewhere along the ISTD flowpath. Curiously enough, the first sample within a batch of say, 10 diagnostic CCVs, always has a much higher ISTD response than the subsequent samples. I'm talking about starting at 5 million area counts and dropping to 2.5 million area counts (and the ISTD response will then never go above 4 million area counts later in the batch).
I've checked for leaks on all three subsystems using both a conventional leak detector and the manual tune/profile mode with a can of dust-off and cannot find anything suspect. The air/water levels are acceptable (<5% for N2, O2, CO2, and H2O), and as far as I can tell the column is healthy. The TIC is clean and there is no sign of contamination and no peak tailing.
I'm running the method on a TD-GC-MS. Specifically, I'm using a Markes International Unity II/CIA Advantage-HL as my thermal desorption/autosampler system, with a 7890B/5977B as my GC-MS. I inject onto the column from the TD using an inert transfer line that's about 18" (not sure if this is considered a direct injection). The column is a Restek Rxi 624Sil MS column, 30 m long with a 0.25 mm internal diameter and a 1.4um film thickness. The trap used in the TD is a quartz trap, and desorbs from 30 degrees C up to 300 degrees C with a 40 degrees C/min temperature ramp. My GC oven ramps from 35 C to 220 C using successively larger temperature increase rates.
To whomever is reading this, thanks for taking the time. Hopefully somebody out there is running a similar method or instrumentation. Please let me know if I've left out any key details, I really want to paint the full picture here.