Chromatography and Mass Spectrometry Application Note
L.Pereira,1 M.Woodruff,1 P.Deland,2 D.Elmashni,2 M.Euerby3
1Thermo Electron Corporation, Runcorn,UK
2Thermo Electron Corporation, San Jose CA,USA
3Astra Zeneca R&D, Charnwood,UK
Comparison of the performance of various diode array detectors in the analysis of trace related substances of a pharmaceutical drug substance are discussed. We consider the variables of stationary phase in the optimization of analysis sensitivity, both by altering selectivity and improving peak shape.
Our results show that by utilizing the total internal reflectance potential of the flow cell in the Finnigan Surveyor PDA that significant increases in the signal/noise ratio can be achieved, therefore leading to improved limit of detection (LOD).
We have also shown how the correct choice of stationary phase both in terms of its quality and its physical characteristics can affect the overall sensitivity of the method.
HPLC is widely used in the analysis of trace components, i.e. <0.1%, in complex mixtures. The success of this technique in terms of determining accurate concentrations of impurities relies on gaining good separation and excellent peak shapes for all of the analytes involved. This will be dependent upon system parameters such as sample preparation, column resolution, sample injection, and sample detection. In order to achieve the best measurement accuracy, trace components of interest need to be completely resolved from adjacent peaks.
Quantitation of trace components is regularly performed by measurement of peak height, therefore the chromatographic separation needs to produce sharp, efficient peaks.
The accuracy and sensitivity of the analysis is determined by the chromatographic parameters that affect separation, namely stationary phase and mobile phase. Particle size will also have a significant effect on the peak height of the related substances present in the crude drug substance. Changing particle size will increase resolution by increasing N column plate number. N can be estimated by the calculation in equation 1:
As shown in Figure 1 for 3 μm and the 5 μm particle sizes we would have N 175,000 and 105,000 respectively, the smaller particle size giving rise to increased peak height of N. Which equates to a potential gain of 30% in peak height.
The other major parameter, which determines sensitivity, is detector type. The signal-to-noise ratio of a UV detector can be maximized by increasing the flow cell path length, according to Beers Law (Equation 1):
The limit of detection (LOD), usually defined as a peak with a signal-to-noise (S/N) ratio of at least 3:1, is the smallest concentration that can be detected. If it is possible to increase the signal with no significant increase in the noise level, substantial decreases in the LOD will be seen which will allow better accuracy and precision at the limit of quantitation (LOQ). Noise is made up of two components: short-term and long-term noise. The former is more critical when the flow cell path length is increased, as it is associated with low light intensity and detector electronics.
ChromQuest software settings for PDA (UV6000LP) with 50 mm path flow cell: Analogue sampling 10 Hz; rise time 0.5 s; wavelength 214 nm, bandwidth 11 nm, scan rate 10 Hz.
ChromQuest software settings for a diode array detector with 10 mm path flow cell: Analogue sampling 10 Hz; response time 0.5 s; wavelength 214 nm, bandwidth 11 nm, slit 4 nm.
Noise (ASTM) was taken from blank run (1 min centered around peak), calculated by ChromQuest software. Signal was taken as the peak height.
It has been demonstrated how the flow cell path length for a PDA detector can be increased to five times that of traditional flow cells by utilizing total internal reflectance. It has been proved that this leads to a significant increase in the S/N ratio afforded by UV detection allowing much lower LODs and therefore LOQs when trying to accurately quantify trace components in a pharmaceutical drug.
It has also been demonstrated how the chromatography can play a major role in sensitivity as the efficiency and height of the peaks will be affected by the particle size in which the stationary phase is chosen.
In this methodology the variables in the chromatography have first been optimized to ensure good efficiency and peak shape and then a new flow cell with LightPipe technology has been used to remove one of the constraints in the Beer Lambert law, allowing much greater S/N values to be obtained.
Reference: (1) Practical HPLC Method Development, Snyder, Kirkland and Glajch.