why is the peak area of standard decreasing?

[moonchips]

I did see decreasing of std signal.

I did a study to see how flushing can solve the problem.
I inject one std, then flush, two std, then flush, three std, then flush, four std, then flush, five std, then flush, six std, then flush, one std. The following is peak areas obtained.

1 std injection
22.5
flush
2 std injections
23.5, 22.8, RSD%=2.1%
flush
3 std injections
24, 23.4, 23.1, RSD%=1.95%
flush
4 std injections
24.2, 23.9, 23.3, 22.9, RSD%=2.48%
flush
5 std injections
24.3, 24.1, 23.5, 23, 22.5, RSD%=3.19%
flush
6 std injections
24.3, 24, 23.5, 22.9, 22.7, 22.3, RSD%=3.35%
flush
1 std injection
24.3

The same std vial was used throughout the study.
RSD% is not that bad. The thing that bothers me is that there is a decreasing trend between flushing. That is because adsorption, right?
Are there any way to solve this problem?

Thanks!

[max_planck]

(.....)

What is activated hydrolizing HPLC vial?

Have a look on the label of your vial package. What does it say?

[max_planck]

double post. sorry

[moonchips]

it says" National Scientific, Target DPTM vials C4000-1w"

[ym3142]

did you ever tell us what is in your std?

Thanks.

[max_planck]

it says" National Scientific, Target DPTM vials C4000-1w"

These are GC vials.

[moonchips]

ym3142,
it is a kind of antioxidant BKF. here is the link to the chemical:
http://webbook.nist.gov/cgi/cbook.cgi?Name=BKF&Units=SI

max_planck,
what is the difference between GC vials and HPLC vials?
Again, Could you please explain a little bit what "activated hydrolizing" doing?

Thanks

[max_planck]

ym3142,
it is a kind of antioxidant BKF. here is the link to the chemical:
http://webbook.nist.gov/cgi/cbook.cgi?Name=BKF&Units=SI

max_planck,
what is the difference between GC vials and HPLC vials?
Again, Could you please explain a little bit what "activated hydrolizing" doing?

Thanks

It means that the vials are hydrolizing some substances, hence removing the water. This can change the chemical structure of a substance and therefore decreasing the peak area over time.
De-activated (non-hydrolizing) vials on the other hand are inert and do not react with the samples.

It think GC vials are especially hydrolized to prevent any water contamination in the volatilve GC samples.

[unmgvar]

Hi moonchips,

can you explain what it means when you say that you flush? how much time does it take?

what type of septa do you use (slit, non slit, crossed)?

you gave details of your system, do i understand correctly that the diluent for your sample is 100% iso-propanol?

[Alfred88]

Dear moonchips:
Your BKF has hydrophobic and hydrophilic moieties.
Could you check if your system can give reproducible results, by injecting Caffeine ~150ug/mL ten times (without column, using only a capillary or a back pressure regulator). Parameters: FR 0.3 mL/min. UV 270nm, Inj volume 15uL. Run time 2-3min. MP is water. Dissolve Caffeine in water with serial dilution.
If the results are good (i.e. RSD for Area and RT both <0.5%), then check if you can improve the elution by reducing the organic e.g:
Gradient:
Time water ACN
0 50 50
10 0 95
30 0 95
31 50 50
35 50 50

Good luck

[AdrianF]

From your data it seems clear that it is a column effect. It would be worth trying a more modern column eg ACE with high purity silica.

[danko]

Hi Adrian,

Would you be kind to explain the rationale for concluding that a more modern column would reduce the observed phenomenon?
If you imply that the effect is due to irreversible retention of the analyte on the column, one should expect an increasing recovery, proportional to the number of injections - up to a point naturally, because the interacting sites are going to be saturated sooner or later.
But as I understand, Moonchips experiences exactly the opposite effect.
Anyway, if I were to follow up on this hypothesis (irreversible retention) I would collect data (chromatograms) under the flush run as well as under the sample injection run, so that I could confirm the elution of the retained analyte.

Best Regards

[AdrianF]

Danko

I don't have a clear rationale apart from the fact that the problem appears to lie with the column. It seems sensible to try other columns.

Moonchips I have noiced that you seem to flush the column with 100% acetonirile in each run. What is the extra flush you use?

[moonchips]

AdrianF,
If you notice, I did not use any buffer in my analysis gradient.
Flushing with ACN or equilibrating with ACN/H2O did not change peak area of BKF.
I used 20 mM KH2PO4 (80): ACN (20) to do the flushing. It seems ACN/H2O is not the best mobile phase for BKF. But it is good for my other component.

Alfred88,
I tried your gradient and my other components are bad(some antioxidants too).
I do not have caffein handy. My other components (also antioxidants) have very good reproducibility.

unmgvar
dilute of my sample is 70% IPA.
I had a flushing program. and it takes about 30min

Time (min) % Water % ACN 20 mM KH2PO4
0 0 20 80
12 0 20 80
17 80 20 0
20 50 50 0
30 50 50 0


max_planck,
I think I can rule out sample degradation or structure change by getting normal peak area after flushing.

[Bruce Hamilton]

Were you other antioxidants that gave good precision also in 70%IPA?.

I'd be very wary of any IPA-water sample/standard solvent, especially between 5 - 75% Water, because the viscosity is actually higher than pure IPA
( water = 1, IPA = 3.1, 50% IPA = 4.2, 70% IPA = 4 cSt ).

I would recommend dissolving your sample in mobile phase, or at least a solvent of more similar rheological properties. I'd also have blanks with the same solvent as the standard/sample, rather than different "flush" solvents.

There's always a possibility than you are forming a vacuum in the vial, but slitting one cap with a scalpel after a couple of injections, and then performing a couple more would eliminate that.

There are other possible reasons for what you are seeing, but the above would be my first two experiments, and my first concern would be the sample/standard solutions viscosity.

Bruce Hamilton