Double Peak detecting RDX

[travis6203]

I am detecting RDX mixed with HMX and TNT in acetonitrile samples on a 1100 Agilent HPLC. I tried using a mobile phase of 50:50 water/acetonitrile, but I could not separate the RDX and HMX peaks. I switched to 50:50 water/methanol, which separated all the peaks, but now I have double peaks on all the compounds, even at low concentrations. I am using a brand new Dionex explosives column. I understand this could be a pH issue, but I'm not sure what to add to my water to properly adjust the pH (if that is even the issue). Any ideas on solving the double peak issue, I'd like to keep using a mobile phase of water:methanol.

Thank you!

[Uwe Neue]

I can't see how this could be a pH issue. Either your column is dead, because it did not like the change in solvents, or the double peak is cause by too strong an injection solvent. Dilute the injection 2:1 with water. If the double peak remains, the column is dead.

[travis6203]

I don't think the column is dead because it is new and I conditioned the column with 50:50 methanol/water before running samples and I had the double peak right away. I will try diluting my injections and see what happens.

Thank you very much for your input.

[tom jupille]

IF *all* your peaks are doubling, then the odds are that it's a physical problem at the column inlet (either a partially-plugged frit or a headspace). Think about it this way: if all the peaks show the same problem, it probably happened while they were still together.

The fact that it's a new column is not really relevant. All it takes is one piece of particulate junk lodging on the frit to cause the problem.

Try back-flushing the column and see if the problem goes away.

[travis6203]

I diluted my injections to 50:50 water/acetonitrile which has solved the problem. Thank you for everyone's input.

[bisnettrj2]

Can anyone describe why he would have been seeing "double" peaks in this analysis when injecting acetonitrile extracts into a methanol:water mobile phase? How would a plugged frit cause all 3 peaks to "double"? And why would diluting to 1:1 acetonitrile:water have fixed his issue? I understand that he was injecting a stronger solvent than his mobile phase, but how would that have resulted in a "doubling" of peaks in the first place?

For fun (and because I have the same column and mobile phase he is describing), I'm going to inject this on my 1100 when I get to the lab today, to see if I can duplicate it.

[tom jupille]

A partially-plugged inlet frit or a headspace can distort the flow profile at the column inlet (i.e., mobile phase flows more quickly on one side than on another. The result can be *anything*: tailing, fronting, shoulders, flat-tops, or splitting, depending on the details of the flow dynamics. The usual diagnostic is that all of the peaks in a separation show the same problem.

Using a too-strong diluent causes a variety of peak shape problems because the diluent mixes with the mobile phase in the column and the peak comes to a screeching halt. The distorted peak shape is the chromatographic equivalent of a skid mark. As with a headspace, the resulting distortion can be almost anything: shoulders, tails, fronts, splits, etc. depending on the mixing dynamics of the system.

In both situations, "the devil is in the details".

[bisnettrj2]

I did get to inject a mix of 10 ppm RDX, TNT, and HMX in 100% Acetonitrile today. "The result can be *anything*" was right (as was "skid mark"). Worst looking "peaks" I've ever seen. If I get a chance tomorrow, I'll post a chromatogram of the injection, and an injection of a calibration standard mixed 1:1 with Acetonitrile:Water. Very much a difference in chromatography.

[Consumer Products Guy]

Using a too-strong diluent causes a variety of peak shape problems because the diluent mixes with the mobile phase in the column and the peak comes to a screeching halt. The distorted peak shape is the chromatographic equivalent of a skid mark. As with a headspace, the resulting distortion can be almost anything: shoulders, tails, fronts, splits, etc. depending on the mixing dynamics of the system.

I see this as a common HPLC issue; my co-worker did exactly the same working on a trace pesticide assay a few months ago - standard was in methanol, but sample was in water-methanol, and standard peaks were crappy. As Tom said, use a weaker solvent for the sample/standard such as an organic-water mix, or inject less microliters of the organic-only solution.

[bisnettrj2]

HMX, RDX, and TNT on a Dionex Acclaim E2 column. Solvent is 100% Acetonitrile:




Explosives Mix, same column, but solvent is 1:1:2 Acetonitrile:Methanol:Water

[tom jupille]

Nice example! Thanks!