Trying to find a general method to check solvents for behavior in a UV detector (including scattering phenomena due to gases) we found it useful (not anywhere perfect) to extensively evacuate the solvents in pressure vessels followed by applying pressure with N2 to transport the solvents through a 250 x 0.1mm capillary (no column), the UV detector, and a restrictor (34 bar). When evakuated water was compared to water which was helium or air saturated (in similar pressure vessels) we found that not only was the baseline shifted coniderably upward, but noise was much higher as well. Some numbers:
Differences in baseline level: evak. vs He_H2O: ~0.003AU; evac. vs Air-H2O: 0.001-0.005AU
Noise with evacuated H2O: 0.00005-0.00008AU;
with He-H2O: 0.0002-0.0013AU;
with Air-H2O: 0.0006-0.0013AU.

Questions:
Is this what other people have seen? If so, why use He?
Does anybody have comparative figures on porous teflon cap. vakuum systems?

Some other method for purity checks are highly appreciated.
The method given by A.Mouse in "Water HPLC grade quality" on Sept. 27 is useful only for impurities which absorb UV and are adsorbed on stat. phases, highly useful, nevertheless.

Hans

Interesting data. Thank you.

The question arises, is the difference caused by the actual amount of dissolved air in the water, or is the difference due to some trace volatile found in the water which the helium sparge or air sparge does not remove completely, but the evacuation by vacuum does remove?

Or, does a residual amount of dissolved gas allow the N2 to reenter the water 'faster' and be seen in the UV detector?

Your work is thought provoking. Thanks again.

Why put a restrictor after the detector?

In normal chromatographic circomstances, there is no restriction. Most detector cells dont allow high pressure, so the pressure, build up by the colomn, decrease when the fluid left the column.

I hope it was clear, but I can't explain it as I want.

Anon 9:37am:
The air bubbling was done to have a standardized normal H2O, namely, H2O which stands around, exposed to air, for a long time. The pressure vessels are filled with solvent/moph via vacuum. That gets rid of most of the air in H2O.
The He was pushed (bubbled) through the H20 from the bottom exit of the container (a pipe). The He, most likely, removes residual air, but now you have He dissolved in H2O. Apparently, itīs enough to cause almost as much scattering in the UV-cell as air!? (Both air and He produced lots of spikes, air more strongly).
The N2 (for the pressure)probably takes weeks to diffuse down to the outlet, there are only problems if you work with the last 5-10mL of the liquid.
Anon 11:41 am:
The restrictor was used to control flow more easily and to suppress spiking (wasnīt perfect here). Sometimes the restrictor is useful in normal chrom. when it was not possible to remove the air sufficiently, or when air gets on the column one can get rid of it much faster with a in-line restrictor. (Apparently, some cells have a strong tendency to provide "nucleii" for outgassing, especially when under atm. presure).

I always thought that helium is less soluble in water and I can't imagine any inteaction of helium with water that causes UV absorption.
However, we have had porblems with air injected into the system, causing a broad peak at ca. k= 3. It is still not clear if it was an autosampler malfunction or if the sample solvent wasn't degassed porperly.

Alex

To add an observation of ours:
We ran a gradient of aqueous to acetonitrile, through a column and two detectors in series with non-sparged, non-evacuated eluents. The detectors were a standard Waters 996 PDA(10 mm flowcell, 0.009" tubing) and a Water 474 fluorescence detector (16 ul).

We got bad results with the second detector. The problem went away with through helium sparging. It was only minimally affected by vacuum degassing.

We no longer do either of these. We get good results with the second detector now that we use a 0.01" outlet to waste. We were using 0.02" outlet tubing.

I was careful to check the pressure increase the switch in tubing would cause. That 474 flow cell is rated at 145 psi -- and they're not kidding, I cracked it with a few seconds at 250psi.

At any rate, you've discovered as we have, that dissolved air affects different flow cells differently.

In our experience, spargeing is great for removing the particular detector problems we were having -- it was just impractical for us. Yes, helium sparging just switches one set of dissolved gases (nitrogen and oxygen with helium), but there has historically been a benefit for HPLC systems and detectors.

But you can't discount the effect of backpressure caused by your restrictor. And I'm still pondering the effect of using compressed nitrogen to drive fluid.

Like everyone else here, I'm fascinated by your experiment and would like to hear more.

I was always told that N2 was worse than not sparging at all. If you are using it to push solvents into a detector flow cell, certainly some is being absorbed into the solvent under test. Different solvents would most likely absorb different amounts of N2 and could, in theory, give different levels of background as well as noise. Just my thoughts.

Intersting aside:
We once mistakenly replaced a He tank with an argon tank, and all of the Perkin Elmer sparged HPLCs failed the gradient accuracy calibration test by a wide margin, if I remember.

A.Buske: Specs taken of these baselines are best interpreted by assuming scattering and?/or? refraction phenomena, not absorption. Air injection is a broad subject in itself.

RC: Didnīt want to mention this as the saturated air H20 has not been looked at, namely, our membrane pump flow regulator system gives lower noise (no spiking), but similar upshift when He-H2O is used instead of evakuated H2O.
On N2: We use up to almost 2m tubes for longer sessions, even a 25cm tube gives no problem for 75-90% of the liquid, even with 200 bar (here we used 40 bar N2).
Also to Anon 10:16am, The Fickīs diffusion equations let you calculate how slow lateral diffusion is. We and others have used N2 pressure vessels for 10+ years.
The restrictor just lowers degassing and makes flow control easier in the pumpless, columnless setup.

Anon 1:27pm: Maybe Ar scatters more, or causes a greater shift in refr. index?

Hans