Hi,
In the presentation Billy gave last friday he said that the calculation of the permutation rate was simply the slope of linear best fit of a plot of mass loss against time. However in the example spreadsheet you sent through you have also forced the linear regression line to pass through the origin. Is this correct? it certainly seems to make sense considering the data involved but I did just wanted to check as you had not mentioned this in .
Thanks,
Hi Matthew,
Good spot. I had set the intercept to zero on the basis that you know the mass loss must be zero at time zero. You can do this if you have already allowed the tube to equilibrate for a reasonable period of time. If you just put the tube in directly from storage it is better to allow a non-zero intercept as there will be a non-linearity at the start as the analyte initially diffuses through. Make sense? Cheers – Billy
Comments
Why does my permeation source calibration not go through zero!
In a ideal situation the intercept should go through zero and therefore you should included it.
However in practice at the initial phase of calibration the tube is still reaching equilibrium where one of two things is occurring. (please see attachment)
The first is that the analyte is still breaking through the membrane which results in a lower mass loss for that time period. Or secondly that there is some analyte on the outside of the tubing which is being released more quickly, or that the PTFE is saturated with a high concentration (normally after long storage). Each of these will mean that you intercept will not go through zero.
So what we do is usually ignore the first couple of calibration points and plot the regression once stabilization is reached, You can then adjust you calibration to start from here to then go through the zero intercept, if you wish
Permeation source Types of calibration plots.pptx
Please sign in to leave a comment.