Page 1 of 1

shoulder peak problems, overlapping peak problems

Posted: Wed Apr 01, 2009 10:51 am
by keepingthefight
Here's the situation:

We spiked hydrocortisone and prednisone (IS) in urine. We used 40:60 ACN:H2O for our mobile phase and a C18 column. Flow rate is 0.70 mL/min. After adjusting pH to 9.0 and increasing ionic strength, we extracted analytes using solvent extraction. Our first run, we noticed in the chromatogram that a peak appears very closely to the hyrdocortisone peak. After increasing the hydrocortisone concentration, we noticed that there appears to be overlapping of peaks. Any suggestions on how to separate them?

I know this should be easy...but we are rushing. We need solutions ASAP. Decreasing the flow rate will probably separate the peaks a little but it will not completely eliminate the problem. A relatively high concentration of hydrocortisone (greater than 10 ppm) will most likely end up having a shoulder peak because of the overlapping peaks.

Posted: Wed Apr 01, 2009 11:57 am
by Bryan Evans
You can:
a. decrease particle size from 5um to 3um
b. increase column length
c. replace ACN with MeOH or THF

Below is data showing this separation on Cadenza CD-C18 and Cadenza CL-C18:

http://www.imtakt.com/TecInfo/TI039E.pdf
http://www.imtakt.com/TecInfo/TI324E.pdf

Posted: Wed Apr 01, 2009 12:04 pm
by Uwe Neue
Check out a column with an embedded polar group (SymmetryShield RP18, XBridge Shield RP18). Such columns give a different selectivity for analytes with a phenol function.

Posted: Wed Apr 01, 2009 2:42 pm
by HW Mueller
How about the most obvious: fiddle with the H2O/org ratio? Or even before that, check the lit., there should be a trillion examples.

Posted: Thu Apr 02, 2009 11:27 am
by keepingthefight
You can:
a. decrease particle size from 5um to 3um
b. increase column length
c. replace ACN with MeOH or THF

Below is data showing this separation on Cadenza CD-C18 and Cadenza CL-C18:

http://www.imtakt.com/TecInfo/TI039E.pdf
http://www.imtakt.com/TecInfo/TI324E.pdf

ok ill fo with increasing the column length first...