Hollow wrote:
What happens if you
- change flow rate? -> change in frequency of ripple?
Assessment of frequency is a bit difficult because it increases during the chromatogram, parallel to the increase of %B: At 0.55 mL/min flow using channels A and B, I have measured the following values (each time point is a "minimum" in the chromatogram):
Time %B
4.13 16.2
5.12 17.7
6.03 19.1
6.87 20.3
7.65 21.5
8.37 22.6
9.07 23.6
9.68 24.5
I changed the flow rate to 0.4 mL/min (decrease to 75 % of the initial value), the result was roughly a decrease of the frequency by factor 1.4 (reciprocal of 73 %), so correlation to flow rate and gradient composition is for sure.
- change (increase / decrease) the concentration of TFA in one of the mobile phases -> change in amplitude?
No change in amplitude when H2O + 0.1 % TFA stays constand and TFA in ACN is increased from 0.085 to 0.1 % TFA. Only the baseline drift is much greater (as expected due to the different extinction coefficient of TFA in water and ACN).
- put both solvent lines in same reservoir
No ripple at all
- test a different lot/brand of TFA and/or ACN
pending, all bottles in house are from the same batch
- remove solvent reservoir frits
These were already replaced => No change
[edit] Tried without frits => no change[/edit]
- have you done a gradient accuracy/mixing test?
Yes, accuracy is sensationally good with calculated deviation < 0.05 % and ripple < 0.005 % (Thermo PQ procedure), reproducibility is 0.006 % RSD between three runs.
My thoughts to see if the ripple is correlated to one solvent channel or solvent.
I also tried C vs. D instead of A vs. B => Amplitude and frequency remain almost identical.
Next experiment B vs. A instead of A vs. B => Amplitude and frequency remain almost identical.
...or mixing quality (remove and/or add one mixer)
pending
And what if you additionally degas the mobile phases offline maybe with He?
Unfortunately we don't have helium.
This seems to be a difficult one