Nick and Wes are analytical chemists. Both were assigned to test the stability of assay on some benzoyl peroxide. Stability testing is the continual monitoring of a pharmaceutical product that remains on the market to ensure the product being sold maintains the proper strength and no unacceptable impurity levels are reached. Other tests may be performed on stability if there is a chance the product could change throughout the aging of a product.
The benzoyl peroxide sold by Nick and Wes’ company is sold in 2.2%, 4.4% and 9.3% concentrations. Each analyst grabbed their respective studies. Nick’s study was to be tested on the 53th week following formulation. Wes’ study was to be tested on the 25th week. Each analyst believes their sample has a benzoyl peroxide concentration of 4.4%. The specification for the stability study for assay is 85%-115% %LC.
Both analysts followed the following preparations:
Standard:
- Accurately weigh 55 mg of benzoyl peroxide standard and quantitatively transfer into a 100 mL volumetric flask.
- Dissolve and dilute to volume with water
Sample:
2.2% - 2.5 g
4.4% - 1.25 g
9.3% - 591 mg
- Accurately weigh a sample according to the information provided above and transfer into a 100 mL volumetric flask.
- Dissolve and dilute to volume with methanol.
Nick’s sample weight is 1.25042 g
Wes’ sample weight is 1.23092 g
Since they are running the same type of sample, Nick and Wes decide to analyze their samples together on the same HPLC. The flow rate is 1.0 mL per minute analyzed at 215 nm with a run time of 13 minutes (780 seconds) per injection. The standard weight is 53.21 mg with a purity factor of 0.986.
Once the samples and standard were analyzed, the following data was obtained.
The average peak area of the chromatograms of the system suitability standard injections was 49814.014 area units.
Nick’s sample had a peak area of 49573.967 au
Wes’ sample had a peak area of 48049.003 au
Nick and Wes signed off their data and submitted it for review. During review, the data auditor found that Nick’s sample was a 9.3% instead of a 4.4%. We definitely have a problem.