Page 1 of 1

Software (hence sensitivity) problem

Posted: Wed Jul 19, 2006 10:09 pm
by letsdoit
Hi everybody,
I am using the HP1050 HPLC system with chemstation on it.
The problem I have is that at low concentrations say at 10 ug/ml I see a peak on the computer screen which is a high peak but at the end of the run the Chromatogram shows the peak as very small peak.
This is I guess because the solvent elution front appears as a very high peak and hence the Y axis is adjusted according to tht front.
Consequently I can see the 1 ug/ml or 2 ug/ml peaks on the computer screen when they are running but at the end of the run, the chromatogram appears flat at that time.
So how do i make sure that these peaks (1ug/ml or 2ug/ml) are integrated and appear in the chromatogram...
Please help me :( :?: .............

Posted: Wed Jul 19, 2006 11:02 pm
by Mark Tracy
First of all, fear not. Your data is OK, and the integration process does not care about the magnification on the screen. If it is not integrating some small peaks, you adjust the integration parameters. To fix the view, you just need to turn off autoscaling, and use fixed scales.

Second, ChemStation, like all good chromatography data systems is a complex tool and will require you to spend time with the Operator's Manual if you are to use it effectively. You should also cultivate an experienced mentor in your lab, and ask her to explain the parts that the manual doesn't make clear.

Posted: Thu Jul 20, 2006 6:55 am
by leadazide
An other solution is to change the veiwing area in the report. This is done in the Data Analysis area of chemstation. In the menu Graphics there is a submenu called Signal Options. Here it is possible to change the veiwing area.

It is also possible to designate a peak as a solvent peak. This should also make chemstation "ingore" the peak in relation to autoscaling.

A last solution is to inhibit integration in the time of the solvent peak.. I think this will make chemstaion ingore the peak but I am not completly certain.

tried but does not work

Posted: Fri Jul 21, 2006 2:39 am
by letsdoit
Hi there,
Thankx for your replies ......but..........
I could not figure out how that works
I tried something wherein the screen actually integrates only a specified portion of the chromatogram....
Consequently after the run the signal appears large enough.
I did this by going to data analysis- calibration- signal options- and then start signal and end signal time...
However after I have finished all runs and at the end I go to view data analysis the chromatogram appears the same old one .... normalused to the solvent front and hence I have to always go there and zoom the peak area and then manually integrate every peak.............
Could someone please suggest me something better

Posted: Fri Jul 21, 2006 5:08 am
by Bruce Hamilton
I haven't used it for a while, but my suggestion would be to look in Data Analysis, Graphics, Signal Options, there are two places you have to change to stop the software full scaling printed output.

Firstly, there is the Ranges setting, and there is also the Multi-chromatogram option, which is something like "each in full Scale", change that to "all the same scale". My recollection is that the multi-chromatogram option affects the output, even for single runs.

Hope this helps, but it could be totally irrelevant.

Bruce Hamilton

Thanx Bruce

Posted: Fri Jul 21, 2006 6:48 pm
by letsdoit
Hey Bruce,
Thanx a lot........ it worked
i guess both changes were necessary for that.
I really thank all of you for your suggestions....

I just need to take care of my baseline now which beomes weird sometimes as I am running a gradient and it appears as uphill ramp (sometimes wavy and sometimes not )

Posted: Fri Jul 21, 2006 7:24 pm
by Bruce Hamilton
There seems to be a lot of friendly HPLCs around these days, lots of waving is being reported. The following is very general, there are lots of exceptional causes as well.

The uphill ramp is probably due to the increasing component of your gradient having higher absorbance at the set wavelength. As long as it's consistent and fairly low, don't worry.

If the uphill slope is a concern, and in all runs, look at each component of you mobile phase to work out the cause. If you are using HPLC grade materials with low absorbance, you probably can't change the mobile phase absorbance much.

If you are using standard lab reagents, look to purchase some HPLC Low UV grades, sometimes it worth trying different suppliers.
Ensure you have careful and consistent addition of highly-absosbing compounds ( such as some modifiers/buffer ). Just ensure that your mobile phase absorbance isn't stealing all the detector energy, which can cause a wide variety of strange baseline effects. It always pays to know the UV properties of your mobile phase.

The waves, if fairly broad and appear at irregular intervals, are likely to originate from earlier samples, and you may want to run a sample with multiple blanks after it, or try a more aggressive final gradient to try and find the source. My current record is a compound that eluted 3 injections and 165 minutes after injection.

If they appear in all samples and stds, then look to column, sample preparation, glassware, and other possible sources. If they are narrow, then they may be related to the current injection, but trying combinations of samples and multiple blanks should help identify source.

If all the above is OK, start to look for causes such as dirty column, vials, etc. The trick is to be systematic, and understand your toy. Each method offers new opportunities for your HPLC to surprise you.

Please keep having fun,

Bruce Hamilton