I measure the same substance (sewage treatment plant influent) every day to see how it changes (molecular weight, etc.), so I measure SEC-UV every day.
Question 1) When measuring SEC, what do you do when the same sodium polystyrene sulfonate standard(mw16500) measures differently every day? Shouldn't it be measured in the same place as the original?
Question 2) Since the standard material measures like this, I want to set the standard material to the same one point and move the measurement material as much as possible. Is there any way? It was difficult with excel, but is it possible with sigma-plot? Can you teach me how to do it?
I don't know what the graphs are meant to show, but for Q1, yes the standard should be the same every day. However, what do we mean by the 'same'? You will not get an identical elution time (i.e. measured to the second) but the values should be pretty close. If not, you have some variation in your process that needs to be eliminated, but it could be one of many possibilities.
How different are the elution times for the standard? Equally, if the standard elutes a bit later than expected, do the samples elute a bit later too?
As for Q2, I'm not sure exactly what you are suggesting, but the approach may not be scientifically valid. Better to fix the problem than 'move' the data.
Ideally you should have a few standards as well, to calibrate the column and compute the molecular sizes of your sample from a graph of Ve/Vo vs log molecular weight.
Hello, I guess you have to calibrate your instrument since the standards should give you reproducible results. Try calibrating with at least three standards and carefully purge your column and pre-column. To use the data you’ve already got, you can compare your curves simply by finding the standard maximum peak values in your Excel data and shifting the columns up or down to bring the maximums on the same row. However, data collected without proper calibration shouldn’t be taken as reliable.
You should check whether the deviation is systematic or statistical. For this, I would suggest to simply plot the peaks you find for the standard as a function of date.
If you can find a clear trend something is happening to your SEC, e.g. the columns are ageing. If it is statistical, you need to ask yourself what the source of the uncertainty is. Possibly, the injection time and the start of the counter don't agree or the flow rate is not constant.
- Badges
- Science method