noisy baseline on purge & trap GC-MS

[kspar]

I’m new to this forum and spent some time looking through archives for a similar problem, but nothing solved my issue. I am looking for advice to fix an elevated baseline/noise problem. Can I clean this up or is this as good as it will get? Here is a lot of information. I came prepared.

I have attached 2 chromatograms(50ppb standard and method blank).

I have a very large CO2 peak at the beginning of my run and a baseline that I believe is causing me problems with early eluting compounds. I inherited this method and have made some alterations. I changed from industrial grade helium to ultra high purity grade and installed a big universal trap leading into the GC and purge and trap. I have installed a new repeller, insulator, liner-septum-gold seal, split vent trap, new column, new Supelco VOCARB 3000 K trap for P&T and cleaned the source with the same results. Also new 85/15 vespel/graphite ferrule. I am having issues determining the cause of this elevated baseline and running out of ideas. I also played with the P&T settings by adding dry purge time. Adding cryofocusing is not an option at this time. I ordered a Supelco carbopack B/carbosieve S-III trap to try as well to see if changing to carbosieve from carboxen would help.

I analyze wastewater samples via 8260B for 74 compounds ranging from Dichlorodifluoromethane to 1,2,4-trichlorobenzene with the usual suspects in between. I run an eight point calibration ranging from 20ppb to 400ppb. I am starting to have some larger problems with the early eluting compounds. Specifically Dichlorodifluoromethane and chloromethane. I think I will solve the chloromethane issue by lowering my upper calibration standard to 200ppb to correct the methanol suppression.

Here is my set-up:

Agilent6890N-GC
Agilent5973-MSD
Restek Rtx-624 (60m, 0.25mm diameter, 1.4um thickness) column

Thermal program
MSD Transfer Line: 280C

Oven: Max temp:250C Equilibration time:0.5min
4 minute solvent delay
Initial temp: 35C for 2.00min
15C/min to 210C, hold 12mins
25.67min run time

Front Inlet: Split mode-with Restek split liner no wool
Initial temp:200C
Pressure: 16.32 psi
Split ratio: 50:1
Split flow: 24.5mL/min
Gas saver: 20mL/min for 2mins

Column flow is 1.0mL/min

Tekmar 2016 purge and trap autosampler
Tekmar 3100 sample concentrator
Supelco VOCARB 3000 K trap

Purge and Trap settings
Line Temp:100C
Valve Temp: 100C
MCS Line Temp: 40C
Purge ready temp: 36C
Purge temp: 36C
PrePurge time: 3mins
Sample Preheat time: 5mins
Sample Preheat Temp: 40C
Purge time: 11mins
Dry purge time: 0mins
GC Start option:Start of Desorb
GC cycle time: 0mins
Desorb Preheat Temp: 220C
Desorb time: 4mins
Desorb Temp: 225C
Sample drain: on
Bake time: 10mins
Bake temp: 225C
Bake Gas bypass: on
Bake Gas bypass delay time: 2mins
MCS bake temp: 300C
20XX Valve temp: 100C
20XX Line temp:100C


Please Help.

Thanks,
Kevin

[carl.nott]

What does a direct injection (injecting into the injection port) look like?

[AICMM]

kspar,

1) You show a 50 ppb standard and you have tons of signal, so why do you think the baseline is causing your issues? 2) The CO2 is probably just from the air in the top of your sparge tube or vial. 400 ppm at 10-20 mL's is a lot of CO2 on trap (relatively speaking.) 3) Can you post a tune file with your PFTBA turned off so we can see what you base ions are? Also, can you post a spectrum from a non-peak segment of your 50 ppb run?

I might have suggested a higher split but you are already running 50:1.

Best regards.

[Bigbear]

You will always get a co2 peak on a purged water sample.
Things look quite similar to what I do for 524. I would suggest you lower your flow to 0.6, and up your split to 75-80:1 . My problem was water buildup in the source. Things I did to reduce water :
Reduce desorb time as much as you can. 0.5-1 min will get most all of your sample to the gc. 524 says "about" 4 min desorb time, and my regulators accept 3 min.
Increase your gas saver to about 150 starting @ 1/2 your desorb time ( 3 min desorb, start gas saver @ 1.5 min). This helps remove water in the transfer line.
Use dry purge 6 min @ 40 cc/min.

[Cmdr Keen]

I also would consider the CO2 peak as normal. With my Thermodesorption system, I have a more or less CO2 peak in every chromatogram but it's far before the substances of interest.

Considering the baseline this looks like there's any source for SVOC or VOC dirt. I would understand if the baseline rises with temperature but in your post it first rises and the decreases again. Very strange. What mass spectrum is averaged in the high area?
Did you analyse perfluorated samples in that GC? Those substances sometimes "host" on your system for several runs...

[kspar]

Sorry it took me so long to reply. I've been covering for a co-worker on top of the GCMS problems.
I am getting good response for most compounds. My concern about the baseline is that it may have something to do with the high results that I’ve been seeing. I now understand the CO2 peak is unavoidable. I’ve altered the purge and trap settings and the GC thermal program to try to reduce the H20 and methanol. I changed my calibration range and now calibrate 25ppb to 200ppb using a 50ppm intermediate. This definitely helped with the methanol suppression. I calibrated and ran a known standard and failed a lot of compounds high. The responses are good, but some of the standards seem to be responding a little low.
Internal STDs
Bromochloromethane response 89,843
1,4-difluorobenzene response 628,652
Chlorobenzene-d5 response 576,197
Surrogate STDs
1,2-dichloroethane-d4 response 40,029
Dibromofluoromethane response 193,433
Toluene-d8 response 705,910
Bromofluorobenzene response 341,849

As a result I changed some of the settings posted earlier to the new setting posted below.

Thermal program
MSD Transfer Line: 280C changed to 260C
Initial temp: 35C for 3.00mins
9C/min to 210, hold10mins
25.67min run time is now 32.44min
Front Inlet:
Split flow: 50.1mL/min
Gas saver: 20mL/min for 0.5min
Purge and Trap changes
Dry purge time: 2mins
Desorb time: 1min
Bake temp: 230C
Changed the purge flow from 40.1mL/min to 38mL/min

Bigbear: I’ll try reducing flow to 0.6 and increasing split to 75:1. My gas saver is at 20mL/min. Are you saying to increase to 150mL/min? Is 2min dry purge long enough or should I set it up to 6?

Here is a recent 50ppb chromatogram and spectrums from the front and tail end of the chromatogram. Seeing column bleed throughout run. I had someone come in for maintenance and I don’t think they conditioned the column correctly. One of the peaks at the end of the chromatogram may be diphenyl sulfide which is not in my standard. I also posted a tune report. I’m not sure how to get a tune report with the PFTBA turned off. Tried it in manual mode, saved tune file, ran tune and it didn’t work.

Going to try another calibration and see what happens. Running out of time.

Thanks for replies.
Kevin

[Bigbear]

You can safely dry purge @40 cc/min for 6 min. That helps with the meoh/h20 supression.
What is your calibration range? I have to calibrate to 0.5ppb and my "noise" looks less. What is your scan rate and threshold? I use 2 and 150 which works fine.
What is your sample size? 25ml? You may benefit by trying larger split rates like 90--120:1. You would have to run some experiments on your lowest std ( or at reporting level whichever is lower). When you start to loose compounds back it off.

As for the " gas saver" values I use , they came from some Agilent folke with the aim of reducing the water buildup in the transfer line and the inlet.