The refractive index (RI) detector is unique among common liquid chromatography (LC) detectors because it is truly universal
in its detection capabilities. LC detectors based on the absorbance of ultraviolet (UV) light are the most popular detectors
because they are simple, reliable, sensitive, and respond to a wide range of sample compounds, but only if the analytes have
sufficient UV absorbance to detect. Fluorescence detectors are much more selective and can be more sensitive, but compounds
must fluoresce to be detected. Mass spectrometry (MS) detectors are increasing in popularity and can provide extremely sensitive
and selective detection, but only if the sample can be ionized. RI detectors respond to a universal, bulk property of the
analyte — its refractive index. Usually referred to as differential refractive index detectors, these detectors detect peaks based on the difference in refractive index between the analyte and the background mobile phase.
This is a benefit that makes the detector universal, but also a problem in that the detector also is sensitive to any other
factor that affects refractive index. The major factors are temperature, pressure, and mobile-phase composition. This month's
installment describes how RI detectors work and discusses some good practices to follow to get the most out of this powerful
How It Works
Let's first consider how the RI detector works. There are specific design differences between detectors from different manufacturers,
but most have the elements of the generic detector shown in Figure 1 in common. All RI detectors depend on the fundamental
property of light's refraction, or change of angle, as it passes through different materials. In the case of the RI detector,
light passes through the clear walls of the flow cell and through the fluid in the cell. With each transition, refraction
takes place and the direction of the light changes slightly. Rather than detect the absolute refractive index (which some
detectors can), most detectors measure the differential refraction between a sample flow cell and a static reference cell
filled with mobile phase. This, in effect, subtracts the mobile-phase background signal from the sample signal. Because light
of longer wavelengths refracts more than shorter wavelengths, a tungsten lamp or light-emitting diode (LED) is used as the
light source in most RI detectors. In a quick survey I did of commercial RI detectors, various manufacturers used light sources
producing wavelengths of 660–880 nm. After the light has passed through the sample and reference cells, it must be detected.
Most commonly this is done with a pair of photodiodes. As the refractive index changes, the position of the light beam on
the photodiodes shifts so that more or less light shines on each diode. This shift of position can then be detected by comparing
the relative intensity of the signal produced by the two photodiodes. In Figure 1, you can see that most of the light strikes
the upper diode. With a change in refractive index, the position of the light beam might move down, causing less light to
strike the upper diode and more on the lower one.
Figure 1: Schematic of a generic refractive index detector, showing the key components.
The basic components of the RI detector shown in Figure 1 are supplemented in real detectors by hardware to stabilize the
detector and simplify operation. The reference cell needs to be filled with mobile phase of the same composition as that filling
the sample cell (without the analyte, of course). To facilitate this, a switching valve commonly is included to direct mobile
phase through the reference cell to refresh or replace the resident liquid. Because it can take several hours for the detector
to stabilize, the switching valve may be capable of routing the waste line back into the mobile-phase reservoir to allow the
mobile phase to be recycled during warm-up so as to reduce the waste of mobile phase.
A change in environmental temperature can be a major problem with RI detectors, because the refractive index of a fluid is
dependent on its temperature. For this reason, RI detectors are contained in an insulated compartment. Most commercial detectors
can control the temperature above room temperature, typically 30–35 °C up to 50–60 °C, although some models can cool the detector
as well. Also, the incoming mobile phase must be at the same temperature as the thermostated portion of the detector, so heat
exchangers are included to stabilize the temperature of the mobile phase. Although flow-cell volumes are relatively small,
typically 8–10 μL, the heat exchanger volume may be 5–10 times this, or even more. This added volume means that RI detectors
usually generate broader peaks than their UV counterparts with smaller total detector volumes.
The inherent design and operating principles of RI detectors leave them susceptible to several problem areas. Specifically,
anything that causes changes in the temperature, pressure, or mobile-phase composition will create corresponding changes in
the refractive index of the mobile phase as it passes through the sample cell. If this is not compensated by the static mobile
phase in the reference cell, baseline disturbances will occur. We'll look at each of these problem areas next.