Purity Peak

[Bryanfs9010]

I have a Question

Purity Peak test is necessary for Degradation products?

Thanks!

[vmu]

I guess you mean a spectral purity test of a chromatographic peak. This test is not a good way to demonstrate the specificity of a method. It cannot confidently confirm that there is no impurity hidden under the main component peak when the concentration of the impurity is small (ca. 0.1-1%) and the spectra of the compounds are similar (this is not uncommon for related substances). Apparent spectral purity of a peak is not a sufficient condition to say that the peak is chemically pure and the method is specific.

[Fernando]

I guess you mean a spectral purity test of a chromatographic peak. This test is not a good way to demonstrate the specificity of a method. It cannot confidently confirm that there is no impurity hidden under the main component peak when the concentration of the impurity is small (ca. 0.1-1%) and the spectra of the compounds are similar (this is not uncommon for related substances). Apparent spectral purity of a peak is not a sufficient condition to say that the peak is chemically pure and the method is specific.

Hi vmu

And then? What?

Best regards

Fernando

[vmu]

And then? What?

Then one has to try to vary the chromatographic conditions and to compare the results:
- test several "similar" columns;
- use a more efficient column (higher length or lower particle size);
- change the mobile phase or use a gradient instead of an isocratic separation (provide enough retention for the components of interest; an isocratic method with k = 1-2 for the main peak is not the best choice);
- vary the column temperature (in a wider range than the one usually used for a robustness check);
- reduce the sample load (injection volume or sample concentration);
- try to use orthogonal separation conditions or even 2D-LC to confirm that the original simple method is selective enough.

All these tests do not give one a 100% guarantee that the method is specific but they greatly enhance the confidence in the method.

Also one can try to use MSD to confirm the results obtained with a UV detector.

One example about the sample load: related substances in methionine according to USP41.
Column: 250 mm x 4.6 mm, 5 um, C18.
Mobile phase: water/ACN/H3PO4, 1 mL/min, 30 °C, isocratic for 6 min, followed by a long linear gradient.
Detection: UV 205 nm.
Injection: 50 uL of a 30 mg/mL sample solution.
SST: Rs NLT 5.0 between methionine (RRT 1.0) and methionine sulfoxide (RRT 0.5) in a solution containing 0.1 mg/mL of each component.
Disregard limit: 0.05%.

The large injected amount of the sample results in a huge overloaded peak of methionine. The sensitivity of the method is so high that one can safely reduce the injection volume (allowed by pharmacopeias) 10-fold without compromising the LOQ. This volume reduction leads to a narrower and less tailing main peak and can uncover a previously hidden peak of an unspecified impurity eluted after the main peak.