Chromatography Forum

Hi all,

I am just starting as an environmental science technician and we are running a brand new dionex 2100 RFIC to run analysis of environmental water samples.

I have run into a bit of a problem that hopefully one of you experts will be able to help me with: During the analysis of the water samples, I am getting a peak that corresponds to nitrite in the sample. However, cross validation against a colorimetric method (EPA 115) indicates that this peak is false.

I have run glassware blanks, reagent blanks, check standards and multiple injections to get to the bottom of what is happening and I am still confused. I have no doubt that the "nitrite" peak is real, as it turns up in samples but not in the blanks, we have cross-validated between the IC and the colorimetric method and determined that the IC and other method are comparable to 0.1 PPM, so we are certain that we are not seeing shadows.

The final thing that we can do is run an internal spike to see what happens, but I suspect that there might be something eluting at about the same time as I would expect to see nitrite. Has anyone got any suggestions or advice to help me determine what is going on? -I know that we have a meeting with the local EPA coming up and I really don't want to be in a position of not knowing why I am getting "nitrite" where there should be none.

The method that we are running is 33mM KOH at 1mL/min with a dionex AS18 column heated to 30degress and conductivity detection.

Thanks for the time.

James

Hmmm.
Is there anything different about the glassware used for the samples? Perhaps acid washed? Anything that makes it unique compared to whatever you have been using to prepare blanks and standards is significant at this point. Even the caps...
You didn't say, but were samples filtered into the vials? If so you need filtration blanks as well.

The conductivity detector is a bulk property measurement, so you don't get much more than "it's a peak" when someone asks about the peaks in your chromatogram. If you possibly have a UV detector available, you should be able to look at low UV (205-215nm depending on background of your KOH mobile phase) and see if the peak also absorbs UV. Nitrate, like nitrite, iodide, iodate, bromide and bromate, is a fair UV absorber and a corresponding and proportional signal in UV and conductivity (compared to your standards) might help you decide if it is a non-nitrate random co-eluter or actually nitrate in the samples for ??? reason. You may have other detectors available that could detect nitrate and be compatible with being inline with the conductivity detector, so be creative. Perhaps electrochemical -- there are references to that being usable.

Hi James,

Try checking for carbonate, under isocratic conditions that you are using carbonate on the AS-18 elutes just prior to nitrite. There is a chance you have carbonate in your samples and not your standards. If this is the case use a CRD-200 or 300 to remove or reduce the effect of carbonate.

Hi James,
We use the same column & I can confirm that the carbonate can be close to the nitrite.
If I recall there is a production method using the As18 column using an initial KOH conc. of 23mM that will show separation of nitrite & carbonate.
I think the details are in the column manual or in the Chromeleon standards methods
Failing that could you have an organic contaminant such as an organic acid - we have seen those species in that region of the chromatogram before.

Kevin

Hi James,

Look up the operators manual for the AS-18 column its on the CD that came with the column. Try running the dual linear gradient, if your mystery peaks moves from the nitrite region and starts to elute just prior to sulphate then its carbonate.

Cheers


Mark

Hi All,

Thanks for the helpful advice.

As for the glassware, we ensure that all glassware that we use for samples is freshly washed with decon 90 and then copiously rinsed. But we do run blanks of everything that we do in the lab before the sample goes into the vial to try to account for any interferences there. Unfortunately, the person doing the sampling is "too good of a scientist to need field and trip banks" so that rules out examining those for contamination.

I will definitely look into the use of other detection methods. As I said, I am just getting started, so learning about other detection methods can only benefit me.

Some research into the chemistry of the samples has show evidence of signficant carbonate, so I will be looking into the AS18 software and the dual linear gradients and a carbonate removal device to eliminate that error.

Thanks so much for your help.

James.

Hi James,

Keep us posted on the outcome

Cheers

Mark

Hi Mark,

The gradient method has separated the nitrite from the carbonate and just needs a little tweaking for resolution. I will probably be calling you and having a chat about installation of the CRD in the near future.

Once again, thanks for the assistance everybody,

James.