Is there a limit to the volume you can inject into an agilent split/splitless inlet?

Theoretically, the answer is no; the limiting factor of a concurrent solvent recondensation - large volume splitless injection (CSR_LVSI) is the surface area available to support a stable film of recondensed solvent. If you read my previous post, which utilized a 50 µL injection to meet lower detection limits for 1,4-dioxane and various nitrosamines, you may have noticed that I used a 10 m 0.53 mm ID pre-column. This was to insure there was enough surface for the 50 µL injection to recondense and form a stable film, without flooding the phase-coated analytical column.

Theoretical limits aside, if you don't have a 7693 autosampler with the enhanced sample handling syringe carriage, you are effectively limited to 50 µL injections because the standard syringe carriage can't hold larger than a 100 µL autosampler syringe. While you can perform multiple injections for a single analysis, peak shapes will surely suffer without the sub-ambient cooling and analyte trapping associated with a PTV (and we are working with compounds too volatile for this to be effective anyhow).

Earlier this week, I finally installed my enhanced sample handling syringe carriage and 500 µL autosampler syringe, giving  me an effective injection volume limit of 250 µL. The chromatogram above is the result. Notice that I used a 30m 0.53 mm ID pre-column to provide plenty of room for solvent recondensation. I plan to empirically discover the optimal pre-column length for this large injection volume through a series of column trimming experiments, so stay tuned.

I also want to point out the strength of the signal for a 0.1 µg/mL standard. If this was a extract prepared via EPA 521 or 522, it would be equivalent to a 2 ng/L sample (2 ppt). This would allow for trace analysis of drinking water samples using a single quadrupole instrument in scan mode.

As with most things in life, there are tradeoffs. The solvent delay for a 250 µL injection is a little over 30 minutes (if you need to detect THF and 1,4-dioxane for EPA Method 522) or 38 minutes in the chromatogram above, which deals solely with the nitrosamines. Using a simple 11.12 degree per minute ramp after an extended 50 degree hold to evaporate and elute the dichloromethane yields a total analysis time of nearly 60 minutes.

If you look at the chromatogram below, you'll see there is still a good separation between the solvent peak and tetrahydrofuran (THF), the first eluting (highly volatile) peak in the mix. Even with a 250 µL injection, THF, which has a boiling point only 26 degrees higher than dichloromethane, is not lost to "backflash" and shows good symmetric peak shape.