DCM in water

[push]

I need to determine ca 1 ppm DiChloromethane in water.I think I cannot inject more than 0.5 mcl water solution that is why GC\FID cannot be used.Can I use by GC\ECD? Is GC\ECD compatible with water? Any ideas how to reach 1 ppm LOQ?

[Don_Hilton]

There are ways to inject water in to a GC - but most of us try to avoid it. What kind of column are you trying to use, by the way? If you are trying to use a column with a silane or wax stationary phase, you need to have the column cool enough to partition the DCM - and that means you will partition or condense the water. And even 1/2 a microliter of water is not going to help your peak shapes. On the types of columns I listed DCM would elute befor the water and you have what some people call a reverse solvent effect. I don't think the problem is the detector at ppm levels.

Do you have access to a purge and trap unit?

[chromatographer1]

DCM in water at 1 ppm is no problem with headspace analysis, unless you insist in using split injection. I measured DCM at 1 ppm (1ng/µL) easily.

Anal. Chem. 1997, 69, 2221-2223

Figure 1, second chromatogram shows a 10ng DCM peak from a 25 microliter sample of water where the peak was almost 2 millivolts in height and 0.5ng of DCM was about 0.1 millivolt in height which is about 40 ppb by weight of water.

FID was used.

best wishes,

Rod

[rb6banjo]

I would also think that headspace/SPME (solid-phase microextraction) would work easily for this analysis at this concentration. It's a very simple analysis.

[chromatographer1]

The reproducibility (variability) is larger but another two or three orders of magnitude lower CAN be achieved. Good suggestion.

Rod

[rb6banjo]

I was curious about this one so I ran a couple just to check sensitivity and reproducibility. I added 0.7 µL of CH2Cl2 to 1000 mL of water and mixed it thoroughly. That's about 900 ppbv (ng/mL) of CH2Cl2 in water. I ran it by headspace SPME (60 °C, 30 minutes just because that's a nice time for the cycling of my GC) using 2 different fibers (100% PDMS and Carboxen/PDMS) on a chromatograph and column where the detector is a flame. Here's what I got:



The precision is pretty good without an internal standard and you get much better sensitivity with the CX/PDMS combination than with PDMS alone. For the noise, I integrated the baseline near the peak for the width of the CH2Cl2 peak at the base. I'm taking the detection limit as the signal required to give a S/N ratio of 3 and the quantitation limit as the signal required to give a S/N of 10. Typically, the PDMS fibers are more robust so it may be good enough for your application. But, if you need to do so, you can really get down there with the CX/PDMS.

[chromatographer1]

Nice work,

I am glad you proved I did not exaggerate. I would have suspected the PDMS would have come in second when testing aqueous samples.

I hope PUSH thanks you for your time and effort. I do.

Rod

[AICMM]

push,

The problem here is that an ECD is not very sensitive to DCM. ECD's don't do all that great with "just" two chlorines. You are probably better off with an FID for total sensitivity. Others have presented a number of options, z.b. headspace and SPME, so you would probably be better off with FID and one of these sample prep techniques. In addition, you avoid the issue of water injections.

Best regards,

AICMM