Qualifying API material

[ydna6969]

hi ,

We have some API and standard material of known potency 99.4%

We carried out a study to qualify this API as standard ,but results are > 100.00 when quantified versus std.
water content for API is 0.20%

what do you do if % purity is > 100.00 versus known std material concentration ?

[danko]

One of them (the API or the known standard) is determined erroneously. Could be the water content.
Do you determine the water content simultaneously – in the same analytical series?
Or the calculation is erroneous.

[tom jupille]

How much over 100.00 is it? and what's the confidence interval (how many replicates, and what was the standard deviation)?

[ydna6969]

thanks for your reply.

To confirm ,,, Water content was carried out in same analytical series (found to be 0.20%)

Six replicates were prepared st dev = 0.88 Mean = 102.1

CI of (101.3 - 102.

Any thoughts ,, fyi,, standard verification was on the high side (101.4%) but precision ok (0.5% RSD)


Thanks in advance for help

[Consumer Products Guy]

Not uncommon. Still within 2%.

[Peter Apps]

thanks for your reply.

To confirm ,,, Water content was carried out in same analytical series (found to be 0.20%)

Six replicates were prepared st dev = 0.88 Mean = 102.1

CI of (101.3 - 102.

Any thoughts ,, fyi,, standard verification was on the high side (101.4%) but precision ok (0.5% RSD)


Thanks in advance for help

Against what did you verify the standard ?

If measurement result on the standard is 101.4 (s.d = ?) and the result on the sample is 102.1 s.d. = 0.88 then the two results are not significantly different.

What is the uncertainty on the 99.4% purity of the standard ?

Peter

[ydna6969]

Hi ,

2 standards were prepared using certified material , one as a control and one as standard for quantification ( criteria 98.0 -102.0)

99.4% taken from Certificate of Analysis ,, so no uncertainty

Further testing has given std check 99.4 % and API (prepared as per standard) mean result = 104.3 % RSD N=6 of 0.32%

So not sure what is going on ,, so qualifiying API for standard use, purity is 104-0.20 (water) ,, how is this possible?

[danko]

If the same material used both as a standard and unknown does not match the expectations i.e. the certified value, you either have problems with the weighing procedure or signal drift in the HPLC determination.
The first mention is very easy to check out. Just use the same solution both as a standard and unknown.
The second could be investigated injecting the same solution a number of times corresponding to a typical analytical series. Then compare the obtained peak areas and see if there is a drift upwards for instance.

Best Regards

[Peter Apps]

Hi ,

2 standards were prepared using certified material , one as a control and one as standard for quantification ( criteria 98.0 -102.0) is the "certified material" the same as the "standard" - are you using these terms interchangeably or are they different specimens of the same substance. If you are verifying a "standard" with 99.4% purity against a different "certified material" then what is the purity of the "certified material". If you are verifying the standard against itself then the expected result is always 100.0% with an uncertainty depending on the precision of the analysis - is you have an imprecise analysis then it is no surprise that the result is more then 100%

99.4% taken from Certificate of Analysis ,, so no uncertainty. There is always uncertainty - no measurement result is free of bias and imprecision

Further testing has given std check 99.4 % and API (prepared as per standard) mean result = 104.3 % RSD N=6 of 0.32%

So not sure what is going on ,, so qualifiying API for standard use, purity is 104-0.20 (water) ,, how is this possible? Analytical variability or the "standard" not having the composotion stated are two possibilities - human error is another. How do you do the analysis? - dilute and shoot ?- multistep dilutions ?. Do you run replicates through the whole process, or are you just doing replicate injections from one vial of solution ?

Peter