Recommendations on EPA 3665A (sulfuric acid cleanup)

Discussions about sample preparation: extraction, cleanup, derivatization, etc.

17 posts Page 1 of 2
Hi all,

I'm developing an application for PCB analysis of sanitary sewer wastewater by GC-ECD, and as such, there's a lot of clean-up to do before GC analysis to avoid chemical interferants. The first round of extract cleanup after evaporation to 1 mL hexane is EPA 3665A, which starts with reaction of the extract with concentrated (or 1:1 diluted) sulfuric acid. I'm finding that the wastewater samples I'm running require a total of 2-3 additions of 1 mL H2SO4 to yield a clean acid layer, which requires a lot of very careful pipetting of a hazardous reagent, both to avoid injury as well as to avoid discarding part of the sample extract.

Does anyone have any personal experience with this procedure and possibly some recommendations on how to better carry it out? One idea is implementing an automated liquid handling robot, though I think the sulfuric acid will be an issue with its viscosity and tendency to corrode metals. Another idea is supporting the sulfuric acid on a silica gel support, such that the cleanup could be done in a dSPE manner.

I'm unfortunately the only person I know of at my large company (~10K employees total) with domain knowledge on this type of analysis, and haven't been able to get any good suggestions other than "hire a student to help". I would like the method to be robust for others to perform, but as currently implemented is too tedious for the technician. I welcome any comments or suggestions, thank you!
OK - I will jump in first

I emphasise with your problem

One idea is implementing an automated liquid handling robot, though I think the sulfuric acid will be an issue with its viscosity and tendency to corrode metals.


Yes, I would agree from experience with a similar analysis that acid does indeed create problems. I bought a very expensive robotic system to do all the additions / extraction/ agitation etc but found that it couldn't cope with the acid for exactly those reasons. An expensive mistake!

The method requires several steps of addition and mechanical agitation and extraction

Is your real concern that you wish to address just the multiple addition of acid?

"hire a student to help"


If you have to follow the method ( https://www.epa.gov/sites/production/fi ... /3665a.pdf) it states -

1.2 This method is restricted to use by or under the supervision of trained analysts. Each analyst must demonstrate the ability to generate acceptable results with this method.

In fact, a cost benefit analysis showed that it was cheaper, taking everything into account, to pay me or train a competent person to do this compared to the high cost of purchasing/maintaining and developing and demonstrating a modified robust method.

We were all students once and if a student or another competent person is employed then that person will still gain valuable work experience

This was for 20 samples a day - the rate determining step is the GC analysis time itself

"too tedious for the technician."
:-) :-) :-) :roll:

Depending on your budget and throughput it may be worth approaching someone like Gerstal as a challenge for them

Again, if you have to follow the method then any other approach would be a deviation from the method
Regards

Ralph
Thank you for your reply.
Yes, I would agree from experience with a similar analysis that acid does indeed create problems. I bought a very expensive robotic system to do all the additions / extraction/ agitation etc but found that it couldn't cope with the acid for exactly those reasons. An expensive mistake!
I'm sorry to hear the robotic system didn't work out due to the acid. Thank you for sharing your experience, it's helpful. We do have an Agilent prep station that is underutilized. I'm considering immobilizing the acid on silica gel, doing a dSPE style extraction with the acid-silica in an autosampler vial, then having the robot aliquot from that.

If you have to follow the method ( https://www.epa.gov/sites/production/fi ... /3665a.pdf) it states -
1.2 This method is restricted to use by or under the supervision of trained analysts. Each analyst must demonstrate the ability to generate acceptable results with this method.
Fortunately at this point the project is exploratory and is not required to follow the method. That said, I'm trying to get close to the method so that if this work is expanded, adaptations can be made to follow the method directly or show performance equivalence.

We were all students once and if a student or another competent person is employed then that person will still gain valuable work experience
The difficulty is that student are typically temp hires (e.g., just the summer) and we operate in a regulation-heavy environment. By the time the student is hired, completed regulatory training, and is trained on the PCB method, they will be about ready to leave! I should mention the samples are waste water from the sanitary sewer, which most students aren't flocking to do :| The technicians here generally have a choice of work to do and will either take shortcuts, waste time, or transfer to other groups if they think something isn't worth the effort. It's my job to make a method as easy as possible to reduce repetitive tasks, reduce occupational hazards, etc., so I can get help running it.

Depending on your budget and throughput it may be worth approaching someone like Gerstal as a challenge for them
Right now the budget is in the negative, but if we can get everything working for now, there's a chance for future funding.
Before using the silica gel be sure to check out EPA 3630C to make sure what the silica gel might do to the PCB targets. I know silica gel works great to clean up PAH extracts but haven't looked into doing it on PCBs.

If it is hydrocarbon or similar type interference you may want to look at EPA 3620 Florisil cleanup, that works really well and you don't have to worry about the acid, and it can be done as dSPE. We I worked for a major hazardous waste disposal company, we would take each PCB sample and do a quick vortex with a scoop of Florsil as the first cleanup step.

Here now in the lab if we have to clean up PCBs for sulfur interference we use EPA 3660B copper cleanup, which is really easy to do also. You do have to clean the copper with sulfuric acid but that is not as bad as trying to separate the acid layer in the extracts.

We also do the sulfuric acid cleanup on all of our waste water PCBs. It is just something the extraction analyst does before the samples are sent over to the GC analyst. It may be easier to stop at 5ml sample extract and do the cleanup, then blow the sample down to 1ml as a final step. The other thing to remember is that if you leave a few 100ul of extract behind before doing a second acid cleaning, it shouldn't matter on your final extract result since you are already at your final concentration the way you are doing it now going to 1ml before cleanup. The tiny loss of extract doesn't alter the concentration of PCBs in the solvent since it should be homogeneous at that point. Sacrifice a little with each washing to be sure you don't kill a column, shooting 700ul of a 1ml extract will give the same result in the end since you only inject 1-2ul of that extract.
The past is there to guide us into the future, not to dwell in.
A little trick I used when running PCBs, was to have the final step as a Florisil cleanup using a 100mg cartridge. I found one that would pass all the PCBs through without extra eluent so I only had to push 1mL of extract through into an a/s vial. This took out stray contaminants that made it through the acid treatment and ensured I didn't have any acid in my extract.
The only downside was that the surrogate TCMX, wouldn't recover from the blanks. I had to drop it and only use DCB.
James_Ball wrote:
Before using the silica gel be sure to check out EPA 3630C to make sure what the silica gel might do to the PCB targets. I know silica gel works great to clean up PAH extracts but haven't looked into doing it on PCBs.
If it is hydrocarbon or similar type interference you may want to look at EPA 3620 Florisil cleanup, that works really well and you don't have to worry about the acid, and it can be done as dSPE. We I worked for a major hazardous waste disposal company, we would take each PCB sample and do a quick vortex with a scoop of Florsil as the first cleanup step.
I'll take a look at that. I was doing Florisil cleanup with Supelco's SPE tubes, but getting negative peaks on the GC-ECD from something in the tubes. The samples are also coming out a lot cleaner with much lower baselines with the sulfuric acid compared to the Florisil SPE. I will try out some Florisil dSPE style and see how that does.

Here now in the lab if we have to clean up PCBs for sulfur interference we use EPA 3660B copper cleanup, which is really easy to do also. You do have to clean the copper with sulfuric acid but that is not as bad as trying to separate the acid layer in the extracts.
Happy to report that copper cleanup is working wonders on the sulfur in the sample extracts. I'm activating it with nitric acid at this point, as listed in EPA 3660B, but may switch to hydrochloric acid if it performs well. Last thing I need is an oxidizing acid waste stream. Strangely, my activated copper turns dark during the solvent rinsing steps prior to use, not sure if it's something in the solvents or my N2 blowdown gas for drying. Still working on figuring that one out.

The tiny loss of extract doesn't alter the concentration of PCBs in the solvent since it should be homogeneous at that point. Sacrifice a little with each washing to be sure you don't kill a column, shooting 700ul of a 1ml extract will give the same result in the end since you only inject 1-2ul of that extract.
Good point. I had been concerned about losses at this stage since I was previously doing sulfuric acid cleanup then Florisil SPE, which requires evaporation to the endpoint after elution from the SPE tube. However, if the Florisil SPE is dropped, losing a 100 uL here or there isn't a big deal.
Steve Reimer wrote:
A little trick I used when running PCBs, was to have the final step as a Florisil cleanup using a 100mg cartridge. I found one that would pass all the PCBs through without extra eluent so I only had to push 1mL of extract through into an a/s vial. This took out stray contaminants that made it through the acid treatment and ensured I didn't have any acid in my extract.
The only downside was that the surrogate TCMX, wouldn't recover from the blanks. I had to drop it and only use DCB.
I might see if this works for us. What supplier did you use? Did you go with glass or plastic? The Supelco cleanup SPE tubes (54116-U) seem to have some contaminants that cause negative peaks on the GC-ECD, causing issues with peak integration. We are using the polypropylene ones right now, as the pre-packed glass tubes would break the project's budget.
The supplier, like so many, has been sold / changed names to:
Grace, part number 204500
Steve Reimer wrote:
The supplier, like so many, has been sold / changed names to:
Grace, part number 204500
Thank you, Steve. I will look into those. Did you precondition the tubes with hexane before pulling the sample through? Also, if you did precondition, did you pull the bed to dryness before loading the sample? Thank you!
When I did it I just quick pre-rinsed with hexane, gravity fed into A/S vial and then blew out the last bit with a dropper bulb.
Caveat here: I handed off most of my PCB analyses 12 years ago.
Steve Reimer wrote:
When I did it I just quick pre-rinsed with hexane, gravity fed into A/S vial and then blew out the last bit with a dropper bulb.
Caveat here: I handed off most of my PCB analyses 12 years ago.
Good to know, at least gives me a starting point. We're looking at about 20 samples per day peak, with full clean-up to get rid of ECD interferants, with mostly just me working long days. So any ideas that will help simplify and/or speed up the workup are worth investigating.
Back in 1990, (my first environmental lab job) we used to pipet 0.1mL to 1mL of transformer oil into 10 mL of isooctane in a largeish (15mm x 125 mm)? test tube, add a couple mL of sulfuric acid, cap it and vortex it for ~one minute. It took all the nastys right out of the mixture. Use a transfer pipet to get some of the cleaned up isooctane layer and there you go. Protective gear and lab coat, a must.

Edit: Maybe do the cleanup while the volume is larger, then blow down the volume after pulling off the acid. Oh, and back then I experimented with adding a little potassium permaganate to the acid for extra cleaning zing. But that was too vigorous. Did that part-time for a couple years. That's the only place I did PCB's.
high desert chemist wrote:
James_Ball wrote:
Before using the silica gel be sure to check out EPA 3630C to make sure what the silica gel might do to the PCB targets. I know silica gel works great to clean up PAH extracts but haven't looked into doing it on PCBs.
If it is hydrocarbon or similar type interference you may want to look at EPA 3620 Florisil cleanup, that works really well and you don't have to worry about the acid, and it can be done as dSPE. We I worked for a major hazardous waste disposal company, we would take each PCB sample and do a quick vortex with a scoop of Florsil as the first cleanup step.
I'll take a look at that. I was doing Florisil cleanup with Supelco's SPE tubes, but getting negative peaks on the GC-ECD from something in the tubes. The samples are also coming out a lot cleaner with much lower baselines with the sulfuric acid compared to the Florisil SPE. I will try out some Florisil dSPE style and see how that does.

Here now in the lab if we have to clean up PCBs for sulfur interference we use EPA 3660B copper cleanup, which is really easy to do also. You do have to clean the copper with sulfuric acid but that is not as bad as trying to separate the acid layer in the extracts.
Happy to report that copper cleanup is working wonders on the sulfur in the sample extracts. I'm activating it with nitric acid at this point, as listed in EPA 3660B, but may switch to hydrochloric acid if it performs well. Last thing I need is an oxidizing acid waste stream. Strangely, my activated copper turns dark during the solvent rinsing steps prior to use, not sure if it's something in the solvents or my N2 blowdown gas for drying. Still working on figuring that one out.

The tiny loss of extract doesn't alter the concentration of PCBs in the solvent since it should be homogeneous at that point. Sacrifice a little with each washing to be sure you don't kill a column, shooting 700ul of a 1ml extract will give the same result in the end since you only inject 1-2ul of that extract.
Good point. I had been concerned about losses at this stage since I was previously doing sulfuric acid cleanup then Florisil SPE, which requires evaporation to the endpoint after elution from the SPE tube. However, if the Florisil SPE is dropped, losing a 100 uL here or there isn't a big deal.
If you neutralize that acid stream with bicarbonate, then its just a nitrate stream. Call your local water treatment and ask them if thats OK.
LALman wrote:
high desert chemist wrote:
James_Ball wrote:
Before using the silica gel be sure to check out EPA 3630C to make sure what the silica gel might do to the PCB targets. I know silica gel works great to clean up PAH extracts but haven't looked into doing it on PCBs.
If it is hydrocarbon or similar type interference you may want to look at EPA 3620 Florisil cleanup, that works really well and you don't have to worry about the acid, and it can be done as dSPE. We I worked for a major hazardous waste disposal company, we would take each PCB sample and do a quick vortex with a scoop of Florsil as the first cleanup step.
I'll take a look at that. I was doing Florisil cleanup with Supelco's SPE tubes, but getting negative peaks on the GC-ECD from something in the tubes. The samples are also coming out a lot cleaner with much lower baselines with the sulfuric acid compared to the Florisil SPE. I will try out some Florisil dSPE style and see how that does.

Here now in the lab if we have to clean up PCBs for sulfur interference we use EPA 3660B copper cleanup, which is really easy to do also. You do have to clean the copper with sulfuric acid but that is not as bad as trying to separate the acid layer in the extracts.
Happy to report that copper cleanup is working wonders on the sulfur in the sample extracts. I'm activating it with nitric acid at this point, as listed in EPA 3660B, but may switch to hydrochloric acid if it performs well. Last thing I need is an oxidizing acid waste stream. Strangely, my activated copper turns dark during the solvent rinsing steps prior to use, not sure if it's something in the solvents or my N2 blowdown gas for drying. Still working on figuring that one out.

The tiny loss of extract doesn't alter the concentration of PCBs in the solvent since it should be homogeneous at that point. Sacrifice a little with each washing to be sure you don't kill a column, shooting 700ul of a 1ml extract will give the same result in the end since you only inject 1-2ul of that extract.
Good point. I had been concerned about losses at this stage since I was previously doing sulfuric acid cleanup then Florisil SPE, which requires evaporation to the endpoint after elution from the SPE tube. However, if the Florisil SPE is dropped, losing a 100 uL here or there isn't a big deal.
If you neutralize that acid stream with bicarbonate, then its just a nitrate stream. Call your local water treatment and ask them if thats OK.


For small volumes of acid or base, we normally just neutralize then flush down the drain. Our building also has another neutralization holding tank in the waste water line before it leaves out to the sewer system.
The past is there to guide us into the future, not to dwell in.
LALman wrote:
Edit: Maybe do the cleanup while the volume is larger, then blow down the volume after pulling off the acid. Oh, and back then I experimented with adding a little potassium permaganate to the acid for extra cleaning zing. But that was too vigorous. Did that part-time for a couple years. That's the only place I did PCB's.
I can definitely say the sulfuric acid method works wonders on sewer samples, dramatically improves the baseline and reduces coelutions with PCB congeners that would cause false positives. I'm considering going with a larger volume for this step to reduce losses, difficulty is that all my evaporation glassware has a 1 mL endpoint or lower. Next thing I'll be experimenting with is using a membrane to separate the phases to reduce the pipetting task; it might have some potential.
17 posts Page 1 of 2

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: No registered users and 1 guest

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food & Beverage, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry