Going from SW8260C to EPA524 or dual operation?

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

13 posts Page 1 of 1
I've never run EPA524.3 before. I need to set up an EPA524 procedure to get drinking water certified. I've been doing 8260C routinely for 17 years.

Archon autosampler, Run in soil mode for both water and soil (purging in the sample vial instead of moving water over to the purge unit on the Tekmar 3000) tray can be chilled, but is not at present.

Tekmar 3000 P&T, VOCARB 3000 trap
5973inert/6890N GC-MS, standard turbo pump
DB-VRX 20m x 180um x 1um
split/splitless inlet, 1 min desorb, running 40:1 split
Crosslab ultra inert liner Direct, Straight, 1.2mm
18 minute run, helium carrier, typical cycle time 29 minutes

Does anybody have hints on how to get reporting detection limits for the VOC's down from ~0.5-1 ppb down to 0.002 ppb listed in EPA524.3? Is it just a matter of increasing sample size from 5mL to 25mL and being super clean?

Is there any hope of being able to use this Archon run in drinking water mode part of the time for drinking water testing; and then run waters and soils in soil mode for 8260 work? I believe the only shared paths would be the Tekmar trap, valve, and Tekmar-to-GC-inlet transfer line. Or does the difference in cleanliness of valves and transfer line make this just out of the question?

The auditor said vinyl chloride was very important and difficult. Its well behaved at 8260 calibration levels. What makes it that much harder for EPA524 work.

From searching the forum, it looks like the older EPA524.2 lets you use the same IS and surrogates as 8260. But EPA524.3 mandates a different set.
We have ours set up on separate instruments but back in the day I did run both on the same instrument.

We only have to hit 0.5ppb MRL on our list except for EDB and DBCP where it is around 0.01ppb. For those two I use SIM to get the lower detection limits. Anything below 0.5ppb or at least below 0.1ppb usually requires SIM for good stable calibrations. Also I believe for 524.3 you are limited to only using 5ml sparges. 25ml was allowed with 524.2, but with the update they removed that, and the 5x increase in sample size really only netted you about a 2.5x increase in sensitivity because of the lower purge efficiency.

I also do heated sparges at 40c to help improve efficiency and by keeping the temperature elevated just above room temperature assures that each sample is run under the exact same conditions which reduced variability in responses. For the different internal standards I use one side of the IS valve for 8260 IS and the other for 524.3 IS when doing dual methods on the same system. The chilled autosamplers also helps when you have a long run and the CCV has to set on the instrument all night.

Check the local regulations to make sure you need to go down so low on the MRL. I did limits like that when we ran the UCMR3 volatiles but most had to be run in SIM mode for that project even though the method suggest you only use it for things like EDB/DBCP.

Depending on the software and firmware version your 5973 may be able to do simultaneous SIM and Scan so you could set up a SIM set for the ones you need the lowest detect limits on. Also run the lowest detector gain you can, it gives smaller peaks but improves the signal to noise to actually give a lower detection limit, that was a tip I found on the Agilent website when looking for ways to hit the limits we needed for UCMR analysis. (0.03ug/l 1,1-dichloroethane for example) Full scan at those levels was almost impossible.

Otherwise 8260 and 524 are pretty much the same test.
The past is there to guide us into the future, not to dwell in.
I wouldn't even begin to try to run 524 and 8260 on the same system. 524 is quite different, especially when it comes to the calibration and detection limits. It really is designed exclusively for finished (read that really clean) waters. The only way to get to the UCMR limits is by having a really clean system and, in your case with the '73, using SIM. You run conventional 8260 samples on your system and you absolutely will lose VC at low concentrations.
Mark Krause
Laboratory Director
Krause Analytical
Austin, TX USA
mckrause wrote:
I wouldn't even begin to try to run 524 and 8260 on the same system. 524 is quite different, especially when it comes to the calibration and detection limits. It really is designed exclusively for finished (read that really clean) waters. The only way to get to the UCMR limits is by having a really clean system and, in your case with the '73, using SIM. You run conventional 8260 samples on your system and you absolutely will lose VC at low concentrations.


The sad thing here is that VC is regulated at 0.5ppb in drinking water using 524, but when we do landfill wells using 8260 the regulation limit is 0.2ppb. I actually had to do a special project once where the state wanted limits for Appendix 9 volatiles at less than 0.1ppb, I set up an entirely SIM 8260 method that I ran on our 7000QQQ, that was a bear keeping all the SIM windows in sync.
The past is there to guide us into the future, not to dwell in.
This is bad news for me then. My state is willing to certify me to run 8260 VOC's if I get 524 drinking water certified. They can only offer 524 certification because they don't have a state NELAC program. Its about $1000 to get 524 certified and around $12,000 to get NELAC certified.

I do have another GC-MS system. Its a 5890/5971 GC-MS. I can perhaps get that system clean enough but not sure the old chemstation software and MS can achieve 524 limits. Also, I would still need to be using the same Tekmar 3000 P&T system.
mckrause wrote:
I wouldn't even begin to try to run 524 and 8260 on the same system. 524 is quite different, especially when it comes to the calibration and detection limits. It really is designed exclusively for finished (read that really clean) waters. The only way to get to the UCMR limits is by having a really clean system and, in your case with the '73, using SIM. You run conventional 8260 samples on your system and you absolutely will lose VC at low concentrations.
I'll see if I can get clarification about required detection limits. They seem to be OK with me doing 524.2. They require MDL's be 1/5th of the MCL. Except for vinyl chloride (MCL=2) and DBCP (MCL=0.2); the lowest MCL is 5 ppb. That would mean an MDL of 1 ppb which I can probably hit. Those UCRM limits really are brutally low. I agree, I could not reach those with my system and ever run routine storage tank soil or groundwater samples.
LALman wrote:
This is bad news for me then. My state is willing to certify me to run 8260 VOC's if I get 524 drinking water certified. They can only offer 524 certification because they don't have a state NELAC program. Its about $1000 to get 524 certified and around $12,000 to get NELAC certified.

I do have another GC-MS system. Its a 5890/5971 GC-MS. I can perhaps get that system clean enough but not sure the old chemstation software and MS can achieve 524 limits. Also, I would still need to be using the same Tekmar 3000 P&T system.


I ran 524.2 on 5970 and 5971 using Tekmar 2000 and 2016 sampler, so you should be able to hit that without much problem on the 5971 and Tekmar 3000.

You may want to question the state certification about requiring the low MDL since in 524.3 they say MDL is optional but MRL is required. On the other hand, if you can pass the MRL procedure, then spiked at that level the MDL should calculate out below 1/5 the MRL. The DBCP/EDB you will need to run SIM though to hit that MRL.
The past is there to guide us into the future, not to dwell in.
I've asked and I only need to hit 0.5 ppb for all the analytes. So, this just may work.
It should as long as you are not slamming it with nasty oily samples. If you are using the soil position for wastewater/ground water samples and use the sparger for drinking water you will probably be set to go.
The past is there to guide us into the future, not to dwell in.
James_Ball wrote:
LALman wrote:
This is bad news for me then. My state is willing to certify me to run 8260 VOC's if I get 524 drinking water certified. They can only offer 524 certification because they don't have a state NELAC program. Its about $1000 to get 524 certified and around $12,000 to get NELAC certified.

I do have another GC-MS system. Its a 5890/5971 GC-MS. I can perhaps get that system clean enough but not sure the old chemstation software and MS can achieve 524 limits. Also, I would still need to be using the same Tekmar 3000 P&T system.


I ran 524.2 on 5970 and 5971 using Tekmar 2000 and 2016 sampler, so you should be able to hit that without much problem on the 5971 and Tekmar 3000.

You may want to question the state certification about requiring the low MDL since in 524.3 they say MDL is optional but MRL is required. On the other hand, if you can pass the MRL procedure, then spiked at that level the MDL should calculate out below 1/5 the MRL. The DBCP/EDB you will need to run SIM though to hit that MRL.
I need to read that method more closely. They want me to use it. It says use of internal standard calibration is best but also that you must use the lowest level you want to report in the calibration. That seems nuts since the noise in too low a standard would ruin a response factor calibration.

Correct me if I'm wrong. But it appears my 5973 running Chemstation version D.01.02.16 requires an all or nothing approach for the whole run with respect to SIM vs Scan.
I don't now if MSDChemstation D supports running them simultaneously but E versions does. It also depends on the electronics in the MSD, I think you need the Inert version with the fast scanning electronics to run that way.

We make a method full SIM just for EDB, DBCP and TCP to achieve the same MRLs as method 504 and run those samples separately.
The past is there to guide us into the future, not to dwell in.
James_Ball wrote:
I don't now if MSDChemstation D supports running them simultaneously but E versions does. It also depends on the electronics in the MSD, I think you need the Inert version with the fast scanning electronics to run that way.

We make a method full SIM just for EDB, DBCP and TCP to achieve the same MRLs as method 504 and run those samples separately.
I have the 5973 Inert with the fast electronics. Although, about 10 years ago, I had to get a refurbished sideboard when mine failed and the MSD always gives a couple of errors on startup.

Expected 5.01.73 firmware... MSD = 5.02.09 firmware

and the other message... Sideboard log amplifier mismatch

Edit: so, I suppose I don't actually have the fast scanning electronics anymore. I'll check with my support vendor to see if I can do an exchange for not too much money.
James_Ball wrote:
I don't now if MSDChemstation D supports running them simultaneously but E versions does. It also depends on the electronics in the MSD, I think you need the Inert version with the fast scanning electronics to run that way.

We make a method full SIM just for EDB, DBCP and TCP to achieve the same MRLs as method 504 and run those samples separately.
I'm excited, I just bought a complete license and source copy of G1701EA GCMS CHemstation E.02.01. My support guys told me I can make do with a bump up to D.02.02 (which I could not find) but that going all the way to E.02 would be better. I am going to use the SIM/Scan mode on my fastscan 5973inert set up my 524 method. Of course, I will also have to transition my OS from WinXP-Pro to Win7-Pro. I think I will set up a new box for that and keep the WinXP G1701DA GCMS Chemstation D.02.00 system as an emergency backup.
13 posts Page 1 of 1

Who is online

In total there are 3 users online :: 0 registered, 0 hidden and 3 guests (based on users active over the past 5 minutes)
Most users ever online was 234 on Tue Feb 06, 2018 7:33 am

Users browsing this forum: No registered users and 3 guests