Chromatography Forum

I've tried a literature method which didn't work, using 1% ammonium acetate in methanol/acetic acid 99:1. I injected a solution of tamoxifen citrate, clomiphene citrate (as I.S.) and 4-hydroxytamoxifen and peaks co-eluted (almost as one peak).

I'm going to reduce the amm. ace. conc. and see what happens

my column is slightly longer but same I.D., but my guess is this shouldn't affect things too much... (but will I get better resolution with a longer column???)

What column is it? Also, the mobile phase is? 99% methanol? Try running a gradient from 2% methanol up to 90% methanol and see if you get a bunch of peaks. If you have a reversed phase column your components would likely coelute in the void volume with 99% methanol.

it's normal phase... hence using 99% MeOH. the compound is extremely hydrophobic

You are far off in the conditions that have chosen.

Here are two possible starting points on Symmetry C18:

http://www.waters.com/webassets/cms/lib ... try162.pdf

http://www.waters.com/webassets/cms/lib ... try163.pdf

I can also send you a publication with gradient chromatograms of tamoxifen and its impurities, or you can look it up yourself in J. Pharm. Biomed. Anal. 15 (1997) 1389-1395.

that's brilliant, thank you. I searched high and low but could only find the paper using 99% MeOH. looking into it further it seemed that normal phase was the way forward. I'll try the RP and let you know how it went. thanks again

Hi Uwe, I haven't made a start with the new conditions yet, busy with other things...

I don't suppose you'd have a manual for a Waters 470, scanning fluorescence detector. we've just unearthed one of these in our lab and want to resurrect it - couldn't find a manual on the waters site or on the web. if you can help that'd be great, thanks

this worked really well, thank you.

the 4-hydroxytamoxifen comes off earliest at 4 minutes, nice sharp peak

I tweaked it a bit from 5 minutes onwards, ramping up the ACN so the other peaks come off sharply

the clomiphene citrate isomerises... will this prevent it being accepted as I.S.??

thanks

Please take a look at the graph below.

I ran the following tamoxifen standards: 5, 10, 25, 50, 100, 200 ng/mL
Each of these contained 200ng/mL internal standard

Relative response = (area analyte/standard conc. (M))
_________________________
(area I.S./I.S. conc. (M))

Plotted the values (recovered from plasma) against standard conc. (ng/mL), and power trendline seems to fit really well

Is there a way to convert this to linear?

[IMG]http://i47.tinypic.com/20szpr8.jpg[/IMG]

Forgive me, but I'm confused. Are you intending to plot a calibration curve??

Or is there something different going on here that the NBA Finals is distracting me from....

yes I'm trying to plot calibration curve using external stds., but because they are processed I'm also using int. std.

I crunched the numbers some more and got a linear curve, thus:

(area I.S./area analyte) = F (conc. I.S./conc. analyte)

I express conc. in terms of molarity

I then plot F values against the standard conc, i.e. 5 up to 200 ng/ml...... this gives a fairly linear curve (R2=0.99)