Calibration & Sample prep for P&T VOC analysis

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

15 posts Page 1 of 1
Hello,

I'm checking up on a "we've always done it this way" situation. Namely, in preparing calibrations and samples for 8260 analysis, we've always spiked a 40ml VWR VOA vial directly with our calibration mix, then immediately capped it. The vials are not exactly 40ml but we prepare samples the same way (injecting a few ul of sample, avoiding headspace, mixing the vial well). Mathematically it works out fine... assuming the true volume of the vials used in the cal curve is the same as the true volume of the vials used when running samples.

So I grabbed a handfull of vials and checked the volume... by filling them with water... there was way more difference than I would have expected, sometimes more than a ml. I don't know how much variation was me and how much was the vials, but anyway:

Is prepping solutions/samples in vials rather than volumetric flasks super wrong and podunky? Anyone know of a type of voa vials that are consistent enough in volume to get away with this?
If you want accurate volumes you have to use volumetric glassware - even if you find one brand of vials that have consistent volumes now, they might not have in the future.

The way around the problem, that offers better precision and accuracy than volumetrics is to use weights instead of volumes.

Peter
Peter Apps
I don't do it that way but I have checked my ESS vials for volume and the differences across a batch was well less than 1 mL. (getting a consistent meniscus was the hardest part). Out of the 43 mL total volume that is <2%. For things like MS/MSD etc. that is well within the errors from other sources. For your ICAL... maybe not.
I did your method only for LFM spikes when it was required for 524.2, but always preped cal stds with volumetric glassware. Inject methanolic standard stock, invert 3X, pour contents of flask neck to waste, without righting flask fill 40 ml vial and cap.
This has always worked for me.
I tried doing it gravimetrically and was pretty pleased... lower RSD across my calibration. Thanks!
My pleasure - I find weighing easier, quicker and more accurate than volumetrics nearly every time.

Peter
Peter Apps
my $0.02

I get single-digit cal RSDs on a regular basis preparing stds directly in my vials. Like Steve said, the margin of error is maybe 2%, and I get excellent results with blank spikes and spiked samples. One thing I have found is vital; the sample/standard must be mixed adequately in the vial. The best way I have found is to roll the vial on the countertop several times (sounds silly, but it works) and then invert the vial several times to completely mix the contents. I have neglected to mix standards before, and found the contents were stratified or incompletely mixed, leading to truly bad results - mixing the standard solved that problem.

HTH 8) 8)
For over 20 years I have prepped mine the same way. Make a working standard in methanol and store that in a 5ml micro reaction flask with a Mininert cap so that I can use it for the whole month without it going bad. Store the standard in the freezer at -18C. I never allow it to warm more than a couple minutes simply because if you pull up the volume you need in a syringe, the syringe body will bring the methanol to room temp in a couple seconds, after a wait a couple seconds I remove the syringe from the flask and close the valve then inject into a volumetric flask of Organic free water. 50 or 100ml flask will fill the 40ml vial to a meniscus, add two drops of 1:1 HCL if it is a 524 standard, nothing if it is 8260, then cap and ready to go.

Over 100 analytes in our calibration standard and I can usually get 70% of them calibrated from 1ppb - 200ppb in 8 calibration standards with less than 10% RSD using average response factor. There are always a few that will need to be set to linear calibration (acetone, isobutyl alcohol, ect) and a couple quadratic (methyl iodide, 2-chloroethylvinyl ether).

Plus if using 100ml flasks I can make a duplicate standard for each level and calibrate two instruments at once and still have a little left over I can use to measure out 5ml aliquots into vials to do a Soil calibration following the water calibration. I have not had any problems doing it this way and I can make the entire eight point curve up in about 15 mintues.
The past is there to guide us into the future, not to dwell in.
If you have pentachloroethane in your 8260 standards it is better to acidify them too. Under some conditions it will dehydrohalogenate to form tetrachloroethylene. I have seen 60% degradation in 12 hours.
Steve Reimer wrote:
If you have pentachloroethane in your 8260 standards it is better to acidify them too. Under some conditions it will dehydrohalogenate to form tetrachloroethylene. I have seen 60% degradation in 12 hours.


We now have chilled autosamplers and that helps tremendously with breakdown. If we acidify, we lose 2-Chloroethylvinyl Ether completely along with Acrolein and Acrylonitrile. We did a holding time study once, with full list spikes that were unpreserved, HCl preserved, Sodium Thiosulfate preserved, Ascorbic Acid preserved and Thiosulfate and Ascorbic with HCL. All kept at 4C and ran once a week. The ones with the least breakdown and loss were the unpreserved samples, followed by Ascorbic, Thiosulfate, and HCl. The list for 8260 is so long, you would need to take samples with multiple vials with different preservatives to actually get best performance for each target.

Another thing that will cause conversion of ethanes to ethylenes is contamination in the transfer lines within the instrument. I have had it so bad that even the tetrachloroethanes broke down into trichloroethene and the trichloroethanes broke down to dichloroethenes. That was back when everyone used nickle lines in the purge and traps and came from salts attacking the nickle causing active sites. It is a lot more rare to have it happen with silcosteel lines thankfully :)
The past is there to guide us into the future, not to dwell in.
I wasn't able to isolate the cause but I had one lot of standards from Ultra that had serious breakdown of pentachloroethane to PERC, but other lots from them had no problem under the same conditions. This was noted on 3 different instruments, 2 P&T and 1 Vacuum distiller.
James_Ball wrote:
For over 20 years I have prepped mine the same way. Make a working standard in methanol and store that in a 5ml micro reaction flask with a Mininert cap so that I can use it for the whole month without it going bad. Store the standard in the freezer at -18C. I never allow it to warm more than a couple minutes simply because if you pull up the volume you need in a syringe, the syringe body will bring the methanol to room temp in a couple seconds, after a wait a couple seconds I remove the syringe from the flask and close the valve then inject into a volumetric flask of Organic free water. 50 or 100ml flask will fill the 40ml vial to a meniscus, add two drops of 1:1 HCL if it is a 524 standard, nothing if it is 8260, then cap and ready to go.

Over 100 analytes in our calibration standard and I can usually get 70% of them calibrated from 1ppb - 200ppb in 8 calibration standards with less than 10% RSD using average response factor. There are always a few that will need to be set to linear calibration (acetone, isobutyl alcohol, ect) and a couple quadratic (methyl iodide, 2-chloroethylvinyl ether).

Plus if using 100ml flasks I can make a duplicate standard for each level and calibrate two instruments at once and still have a little left over I can use to measure out 5ml aliquots into vials to do a Soil calibration following the water calibration. I have not had any problems doing it this way and I can make the entire eight point curve up in about 15 mintues.

I admire your experience Sir. Could you please comment further on this set up? We use 2000ug/L stock for 524.2
I feel volumetric is the best way but we use the 40 ml vial but change actual volume to 44mls and it changes the math on spiking amount. What is correct? Also, what volume syringe do you use to spike? Is it a Hamilton gas tight? Do you have needle full or empty?
Like James Ball I use Mininert 1 mL vials kept at -24C. I buy the Accustandard M-465D-SET-PAK Method 465 - Volatiles Set 5 x (3 x 1 mL). Those are all at 200ppm. I transfer a set into 3 vials. One for the 7 gases, one for 8 adds (MTBE, Acetone) and one for the 54 stable compounds. I dilute my internal standard stock to 70 ppm and bottle them up in 13mm 2mL autosampler vials.

I had a machine shop take an aluminum billet 5"Lx2.5"Wx2.5"D and drill in the top surface 3 holes for my minert vials and two more holes sized for 13mm Autosampler vials for my internal standard and GRO spike. I enclosed this in a stacked assembly of 4 styrofoam 50 mL centrifuge tube holders. The two in the middle are hollowed out so the aluminum is sitting on 2 ice packs inside the block and only the top surface with the vials is exposed to the room. I keep this assembly in the -24C fridge when no in use. The whole thing is 8"x6.5"x6" so it fits easily. This really makes it easy to use for a couple hours without the vials getting too warm. Its the gases which go first and I refill that vial at least once but can use the other two vials until empty.

My samples are groundwater wells from leaking storage tank sites and often have soil settled in the bottom. I use a 5 mL gastight syringe for calibration and sample prep. I add DIWater by filling; or for samples pull up directly from the sample vial; adjust to 5mL by squirting out excess; and then add 5uL of internal standard (70 ppb in the vial); and (For a MS/MSD) 1uL of the three CCV standards in the mininert vials; then I transfer the contents into a 40mL vial for analysis. We prepare the 40mL vials by adding cleaned stir bars, and purge them with UHP nitrogen for use in the autosampler.

My Archon autosampler is run in "soil" mode slaved to a Tekmar 3000 P&T system with Vocarb 3000 trap.

The teflon stir bars are cleaned by sonication in 50C alcanox solution for an hour; then rinsed with DI water and dried overnight at 180C.

My cal curve is made up of 5uL surrogate mix, and 0.5, 1, 2, 3, 4, 6, 8, 10, and 12uL of each of the three Mininert stock standards. This gives me 20, 40, 80, 120, 160, 240, 320, 400, and 480 ppb levels. I usually get single digit RSD's except for a few double digit like acetone or napthalene. But all below 15% RSD if the whole system is running well. I will drop the highest standards if the curve is too noisy and live with the bottom 5-7 standards. My CCV is 1uL or 40 ppb.

I run a 40:1 split at the inlet.
Oh and I usually add 20uL of each of the 3 cal standards to 940mL of P&T MeOH in a 1 mL Mininert vial to make a 4 ppm standard. 5uL of this in 5mL is 4 ppb and 1uL is 0.8 ppb. I run these on many runs when I have room; to give myself data to use for determining MDL and also to prove that 4ppb of most analytes is typically good to +/- 25%.
CharliesMom wrote:
James_Ball wrote:
For over 20 years I have prepped mine the same way. Make a working standard in methanol and store that in a 5ml micro reaction flask with a Mininert cap so that I can use it for the whole month without it going bad. Store the standard in the freezer at -18C. I never allow it to warm more than a couple minutes simply because if you pull up the volume you need in a syringe, the syringe body will bring the methanol to room temp in a couple seconds, after a wait a couple seconds I remove the syringe from the flask and close the valve then inject into a volumetric flask of Organic free water. 50 or 100ml flask will fill the 40ml vial to a meniscus, add two drops of 1:1 HCL if it is a 524 standard, nothing if it is 8260, then cap and ready to go.

Over 100 analytes in our calibration standard and I can usually get 70% of them calibrated from 1ppb - 200ppb in 8 calibration standards with less than 10% RSD using average response factor. There are always a few that will need to be set to linear calibration (acetone, isobutyl alcohol, ect) and a couple quadratic (methyl iodide, 2-chloroethylvinyl ether).

Plus if using 100ml flasks I can make a duplicate standard for each level and calibrate two instruments at once and still have a little left over I can use to measure out 5ml aliquots into vials to do a Soil calibration following the water calibration. I have not had any problems doing it this way and I can make the entire eight point curve up in about 15 mintues.

I admire your experience Sir. Could you please comment further on this set up? We use 2000ug/L stock for 524.2
I feel volumetric is the best way but we use the 40 ml vial but change actual volume to 44mls and it changes the math on spiking amount. What is correct? Also, what volume syringe do you use to spike? Is it a Hamilton gas tight? Do you have needle full or empty?


Sorry I missed this one before.

If spiking into the vials we estimated 43mls, but for calibration standards we use the 50ml or 100ml class A volumetric flasks for improved accuracy. If you only invert them slowly three times to mix there isn't any noticeable loss of the gasses. Just dispense immediately into the 40ml vials and not leave the flask sitting at room temp for a long time.

Each analyst is a little different and use the syringe they are most comfortable with, but for me I use an extended handle Hamilton 50ul syringe(the one with the long blue aluminum handle with the glass syringe attached). The teflon washer in the handle controls the tension on the plunger so it doesn't slide down after you pull it to volume. I can use it from 1ul to 50ul with no problems, and for the larger volumes just pull it up to 4 times for 200ul. I found that using the same syringe eliminates any deviation from syringe to syringe. I had a co-worker several years ago who was not comfortable using it at such low volumes and would use three syringes to cover the volume range but I could always get better %RSD on my curves. He had good technique so I believe it was small differences in the accuracy of each syringe that was causing the problems.

Another thing to remember when using the syringes, the needle can have a large volume within it. One co-worker would pull up 5-10ul of air then bring up solution to mark using the meniscus. Problem is, when dispensing the air volume clears the needle volume, so you have the measured amount plus what ever the needle volume was. We found one 100ul syringe actually had 15ul of needle volume, so if you were measuring 50ul you could be off by as much at 30%. I always purge the air from the barrel, then measure to the marks with the plunger tip, that way you have accurate delivery and the needle volume stays in the needle. If switching standards, just pull air a few times to remove most of the liquid in the needle then wash a few times with solvent.

You can also be accurate if you leave the air in the needle, use the plunger tip to determine volume and despense, the air remains in the needle. If you don't aspirate any air and the plunger doesn't leak from above, you will still be accurate as the air acts like an extension of the plunger down to the needle tip. We tested it and found it to work, but I still like to purge the air out of habit.
The past is there to guide us into the future, not to dwell in.
15 posts Page 1 of 1

Who is online

In total there are 4 users online :: 0 registered, 0 hidden and 4 guests (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 4 guests

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, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry