The most important detector specification is sensitivity as it defines the minimum concentration of solute that can be detected. It is best defined a function of the detector response and the noise level The detector response (Rc) can be defined as the voltage output for unit change in solute concentration or as the voltage output that would result from unit change in the physical property that the detector measures, e.g. refractive index or carbon content. Detector noise is the term given to any perturbation on the detector output that is not related to an eluted solute. It is a fundamental property of the detecting system and determines the ultimate sensitivity or minimum detectable concentration that can be achieved. Detector noise has been arbitrarily divided into three types, 'shor-term noise', 'long-term noise' and 'drift' all three of which are depicted in figure 26.

Figure 26 Different Types of Noise

Short-term noise consists of baseline perturbations that have a frequency that is significantly higher than the eluted peak. Short-term detector noise can be easily removed by appropriate noise filters without significantly affecting the profiles of the peaks. Its source is usually electronic, originating from either the detector sensor system or the amplifier. Long-term noise consists of baseline perturbations that have a frequency that is similar to that of the eluted peak. This noise is the most significant as it is indiscernible from very small peaks in the chromatogram. Long-term noise cannot be removed by electronic filtering without affecting the profiles of the eluted peaks. In figure 26, it is clear that the peak profile can easily be discerned above the high frequency noise but is lost in the long-term noise. Long-term noise usually arises from temperature, pressure or flow rate changes in the sensing cell. Drift is a baseline perturbation that has a frequency that is large compared to that of an eluted peak. Drift is almost always due to either changes in ambient temperature, changes in mobile flow rate, or column bleed in GC; in LC drift can be due to pressure changes, flow rate changes or variations in solvent composition. A combination of all three sources of noise is shown by the trace at the bottom of figure 26.

The detector noise is defined as the maximum amplitude of the combined short– and long-term noise measured over a period of 15 minutes. The detector must be connected to a column and mobile phase passed through it during measurement. The detector noise (ND) is obtained by constructing parallel lines embracing the maximum excursions of the recorder trace over the defined time period as shown in figure 27.

Figure 27 Measurement of Detector Noise