Hi everyone

When you have a multicomponent sample how do you determine which wavelength is suitable for all actives. If you vaildate the assay method is this wavelength valid, although some of the actives may not have max absorbance at the chosen wavelength.
Many thanks

Try using a diode array detector... or Multiple wavelength detector set at the max for each of your components. I think you can monitor upto 5 wavelengths with a MW detector.

I have never seen ANYTHING stating that the wavelength of maximum absorbance be used. Ideally, one would use such wavelengths through time-programming or DAD/MWD, but since you're validating the method and documenting precision, accuracy, etc., should be non-issue.

You do not have to be at the lamda max for each component. The USP has validated multiactive methods that are run at a single wavelength.

Collecting multiple channels of data and processing them individually for each component would be my last choice, and using a time program to switch wavelengths would also be low on my list.

I would just pick a wavelength where each component has sufficient absorbance, even if this was low uv (210-215)and not on a plateau.

Modern detectors have sufficient wavelength stability and wavelength reproducibility that this generally is not too much of a concern.

If the impurities or a representative number of impurities are available make up solutions of each at the same concentrations. Scan each solution on a UV spectrophotometer. The point at which the spectra overlap is called the isosbestic point. By definition, this is the most appropriate wavelength.

<By definition, this is the most appropriate wavelength.

Why?

If you have more than two components you probably do not have an isobestic point. If you are quantitating against external standards the isobestic point offers no advantage over any other wavelength.

I thought the isobestic point is more commonly used in UV spectroscopy where you have two forms of an analyte and you don't care about the ratio, but want to measure the total concentration. With HPLC you have probably separated the two forms.

Noname,
Learn your matrix. Take the peak representing the compound with the lowest concentration, or the peak representing the most important compound. Dilute it until a run clearly shows the baseline noise. Measure the noise and calculate the signal to noise ratio. Run the rest of your compounds, e.g., just dilute a typical sample. Calculate the s/n for all the analytes. If the detection and quantitation limits meet your specifications, you are home free. Otherwise you will have to compromise your expectations.

For multicomponent chromatography of non-homologous series, selection of the best wavelength for a compound comprises with the absorbance of the mobile phase and absorbances of the compounds.

The point of isobestic points is that for two compounds, you can usually find a wavelength that either compound will absorb the same amount of light for a given concentration. That condition allows quick visual evaluation of chromatograms to compare concentration. Compromising for a group of compounds would give you chromatograms with peaks approximately proprotional to concentration amoung the compounds. While desirable for visual evaluation, the optimization may not give you optimization for assay accuracy and especially precision.
Jim

Another important point is to evaluate unknown impurities or substances (you can't have a standard solution for this impurity and also impossible to find isobestic point).

You can asume that the response factor will be similar than the main peak. In my opinion, then it's better to be near the max absorbance for the main peak.