High Assay

11 posts / 0 new
Last post
Lysander's picture
High Assay

Hello there guys-
Hope all is well with you
Had a question - It's not a quantitative Analysis question- But it's related to that..
Wondering what could have happening here- And it has become a mysterious problem.
Thought I could have your feedback
I'm running a drug that has to have Salicylic Acid content of 2.8% - The range is between 2.50 - 3.08%-
My mobile phase is 90% Methanol and 10% (0.1% TFA)- This is what SOP has recommended
My column is Supelco LC18 ( 30cm x 4mm x 5 um)
Retention time is 3.5 minutes- Flow Rate is about 1.0 ml/min and Detection is at 280 nm@ 25 degrees
The  sequences that I have run are showing high end results and low end result. In one sequence run, it shows numbers ranging about 2.5 % ( which is low end) and on the other two sequence runs, they show a range of 3.15to 3.3%- ( the high end) Nothing in between, where I am looking for.( about 2.8%)
I'm wondering where could I have made a mistake?
The calibration curves look fine- R square values are ok-R square value =0.9999.
Two of the  sequence runs are using 4 point calibration and one uses 5 point.
I calculated propagation error results and I'm within +/- 2%. , So, I think I'm not totally off.
Anyway- I thought you could help me out with the problem

 I'm thinking somewhere in the software I may have entered a wrong entry-I'm using Chromeleon 6.4
We have sent the samples out and we got 2.8% results, where it exactly should be.
I'm making my stock standard Weight/Weight,, and then do series of dilutions of working standards.
I also have to mention that I don't have the best peak shape- The peak is tailing on the right-
Any word of wisdom will be appreciated
Thanks guys
I'll check in later

Dr. Analytical
Dr. Analytical's picture

When you analyze each sequence, are you using new standards for each sequence, or did you use the same standards?  If you make new standards, were they prepared for the original stock solution, or did you also prepare a new stock solution.
Please give us information on the standard concentrations that you inject.  Your expected range is from 2.5 - 3.1 %.  Using four or five standards is not needed over such a short range of concentration.  You should be able to use one standard at 2.8% and make three or four injections of the standard.
Other issues that can cause problems are the purity of your standard and the moisture content of both the standard and sample. 
Finally, if your peak shape is not good, integration errors could easily cause a 10% change in concentration.  Since your peak is tailing you must decide how to draw the integration baseline - where does the peak start and where does it stop?  Check that this is done the same way for samples and standards.  Check the different sequences to see if the integration was different for each sequence.
Write back with your results and we will help you solve this problem.

Lysander's picture

 Dr. A
Yes- I'm using the same Standard by Sigma (99%)- For making a stock standard  solution and I have always made a new stock standard and working standards. The  Standard by Sigma labels 99% purity. However, I checked it's melting point and it melted at 145C .The pure Salicylic Acid standard should melt at 159. So, I thought maybe moisture is playing a role here and I dried the standard . The results, didn't change.
Dr. A
You have asked about my standard concentration- I made a 2.8% solution- From there, I'm taking a gram out into 25 ml volumetric and QS it with methanol. From this solution, I'm making my daily standards A,B,C and D.
For example - If I took a 1.00 gram into 25 ml- This will be the calculation of the Stock Standard.
1/25 x 0.028 x 1x10E6 and the Stock concentration will be 1120 ug /ml
From here  we will yield series of dilutions of 1/10, 2/10, 3/10 and 4/10-
- I think this is an integration error-But , I will try your advice and we will get back to you.
Thank you sir
- Lysander

efoconnor's picture
25 degrees is a tough one to

25 degrees is a tough one to keep.  You might want to bring it to 30 or 35 C.   Increasing the temp should also reduce the tailing.
10%  what? Water?
Are you using an IS
Are you using area or ht (areas or ht ratios)?
What is your acceptance for peak shape-asymmetry?  Tailing-if it were variable could contribute to the low and high.
Simple things:  how reproducible are your injections?
do you run system suitability?
How wide is your retention time window? and how variable are your retention times?

Dr. Analytical
Dr. Analytical's picture

Since you are preparing all new standards each time, look at this information first.  Compare the raw peak areas or response factors (area/concentration) for the standards to see if there are any differences between the different preparations.  Do the same for the samples.  It may be a simple preparation difference in the stock solution, although the differences are larger than I would expect.
Also, you should not have significant tailing for this compound if your pH is low.  Your column may have a problem.  Also check fittings and connections.

Lysander's picture
Good afternoon

Good afternoon
Hope everyone is doing well-
Thank you for the replies-
To Efoconnor
I will raise the temperature to see what is happening and I'll get back to you on that-
The 10% is (0.1% TFA) and 90% MeOH and yes it is Isocratic.
The Chromeleon is set to calculate Heights- For the rest of the information, and SST  I don't have an answer till I get to my lab tomorrow-Retention times are not varying that much- They all are coming out within 3.5 and 3.51 minutes- It is  consistent retention timing.
Dr. A
I'm calculating Height/Concentration ratios and they do vary from sequence to sequence
0.73 ( In a good sequence from past) and I have had the following ones- - 0.36 ( High End Assay )
 and 0.42( Low End Assay)
However, the Area/Concentration ratios are practically same in whatever I have calculated so far. They all come about 0.09 ~ 0.1
Yes Dr. A
I also belive that this is an SST plus Column problem-My SST is  failing. I have a low Column  Theoretical plate count and like I said for Assymetry, Noise and Capacity factor  I will run some tests tomorrow and will get back to you guys
Please stand by
 Thank you again

Dr. Analytical
Dr. Analytical's picture

As a general rule, for tailing peaks, area is more reproducible than height.  You may have a bad column.  Try reversing it and injecting again to see if the peak shape improves.

Lysander's picture

Hey guys
Hope all is well with you-
 I did some work with my LC analysis-
First- I changed the newer column with my old column-The results were even higher- This time they came about (ranging from 3.9- to 4.10)- Chromatography SST was failing in Plate counts- Same problem that I had with my newer column.
This is what I discovered-
If I changed the mobile phase composition, I can get results, that are in range, which I was looking for – But the peaks have a rider peak coming with them or a shoulder peak when the peak is coming down near to the base line-
The software integration part takes the major peak and that’s how I got good numbers- but I’m not happy with Chromatography.
My previous Mobile Phase Isocratic Condition was this
90% MeOH- 10 % (0.1% TFA) 1 ml/min- SOP recommended
This time I tried this condition-
80% MeOH- 20% (0.1% TFA) with 1 ml/min
Significant change in numbers and two peaks instead of one- Is this a co elution problem?
Is that why I was getting a High Assay numbers?
I filtered my 0.1% TFA solution and checked the linings for making sure the column and the fittings are ok and clean-
I usually clean my column with 100% MeOH for about ½ hour before next use.
I also have to mention that, My standards do not have the rider peak or the shoulder peak coming along with analysis- My samples, on the other hand, do have this problem.
Now then
If I went to 70:30 instead of 80: 20 mobile phase- The peaks get distinct and there is no rider peak anymore – But I’m still getting a 3.3% instead of 2.8% assay-
By the way - I changed the Height integration with Area-

Dr. Analytical
Dr. Analytical's picture
If you get a separate peak

If you get a separate peak that moves away from the main peak when you switch to a weaker mobile phase, then you do have a separate compound.  (In general, the longer things stay on the column, the better the separation.)  That would explain the high values for the earlier tests.
If you are still getting high values now, then your problem is more difficult.  Can you post a chromatogram for us to look at? 
Do you have a different C18 column to try? 
I would also try a mobile phase with 0.1% phosphoric acid rather than TFA.

Lysander's picture
Yes Dr.A

Yes Dr.A
Thank you for the reply- I will try to post a chromatogram. Meanwhile , I will try a different column as well- Let's see what I'm going to come up with. Give me a day or two-
Thanks Dr.A

Santosh@India's picture


Check the following points
1)Check the stability of your sample  in the matrix over time =>unstable analyte tends to show varying results over the time.
2)Check the calibration of your HPLC system includes-injection repeatability,linearity and carry over factor.
3)Where ever possible use Autointegration facility for standards as well sample, maual integration incorporates many error.
4)Use syringe flush after each sample/standard injection.
5)If possible modify the method without using TFA.