Worked on 2 separate HPLC methods, first one uses 100% of 5mM sodium phosphate pH6.8 and aqueous compatible HPLC column and column oven temperature of 40C. Get resolution between imipenem peaks.
Mobile phase used for the second method is 95% aqueous and 5 % methanol, both contain 0.05% ammonia solution (35%). Column oven temperature 22C. Only one peak seen and pek purity analysis showed only one pek unfortunately peak purity can't distinguish isomers.
Column used for the first method lasted for 200 injections and the column used for the second only lasted for 500 injections.
Peak symmetry about 1.0 and it looses it's symmetry after those injections?
Any idea what's going on with the columns?
Precolumn filters are used.
Quick answer: It could be that the columns used are simply fouled with sample material because your isocratic method does not have enough organic in the mobile phase to remove the late eluters (or are not using a strong enough wash solvent after each run). "Pre-columns" do not help in this regard at all. BTW: 500 injections is not terrible. Some methods are good for thousands of injections while others less than a few hundred (depends on the quality of the method and sample injected).
Please provide us with details about the exact column(s) you are using. Specifically, their make. model and dimensions (Length x ID , particle size) and flow rate used. Please also provide the peak retention times seen using both methods and the detection settings (wavelength/bandwidth).
I have run imipenem samples (and their related substances) before and have been a consultant to many labs running this same sample where highly aqueous isocratic HPLC methods provided to me for review (similar to yours) showed little to no retention of the antibiotic on the columns used (and all of them were "Validated" HPLC methods too. Proof that "Validated" does not mean the methods are scientifically good). Please provide more details and we will try and help.
Thanks, Bill.
For the 1st method, column used is Hypersil Gold aQ, (50 X 2.1)mm, 1.9um. Flow rate 0.5ml/min, retention time ~0.7minute. Capacity factor 1.3
For both the methods, detection wavelength 300nm, BW 4nm and reference W.L 400nm and ref BW 100nm.
For 2nd method, column is Ace Super C18 ( 50 X 2.1) 1.9um. Flow rate 0.5ml/min. Retention time 0.7 minute.
Thank you for the info. Two comments.
(1) The approx void volume of a 2.1 mm x 50 mm column is ~ 120 ul. At a flow rate of 0.500 ml/min your Tzero time would be ~ 0.24 min.
- Note: These types of low volumes will require that the HPLC systems have fully optimized flow paths to minimize dead volume. This includes ultra-low A/I and A/S flow paths and very low detector cell volumes. Since this is an isocratic method, the critical volumes which must be minimized will be found in the injector and detector, plus the interconnecting lines in between them. Any larger volumes in these areas may easily result in loss of resolution between systems so be sure to optimize each system before use.
A Rt of 0.7 minutes gives you a K prime (Cap Factor) of ~ 1.92. *You should of course measure it to obtain an exact value, but this estimate should be close. So that establishes that you do have retention on the column. Based on that info, I think that a few simple steps should help you solve this problem. Dissolving the sample in mobile phase, filtering the samples before injection and most importantly, incorporating proper wash steps after each analysis (use a stronger column wash solution, such as 80% MeOH with the 20% buffer to flush the column down after each run) may improve things for you. You may also want to run this analysis at a standard temperature, say 30C.
(2) Please turn the detector's Reference Wavelength feature 'OFF'. It is invalidating your method and destroying the original raw data obtained by the detector. It may also be adding to the signal noise as you have programmed it to delete all data obtained between 350 and 450 nm. If anything ever shows up in this region, you will never know, because the system is programmed to subtract it from your raw data at 300 nm. This results in the data that is collected at 300 nm being modified before you see it. It is now invalid (and corrupt because you have now lost the original data). *You can read more about the use of Reference Wavelengths in the linked article, "REFERENCE WAVELENGTHS (as used in HPLC UV/VIS):"
http://hplctips.blogspot.com/2015/06/k-prime-also-known-as-capacity-factor.html
http://hplctips.blogspot.com/2011/03/reference-wavelengths-as-used-in-hplc.html
- Badges
- Science method