How to Choose

You are here > SofTA Corporation> ELS Detectors> How to Choose an ELSD

How to Choose an ELS Detector

DO try a variety of instruments in your lab with your samples. SofTA will provide you with an instrument to evaluate. Ask other manufactures to do the same.

DO compare how easy each detector is to set up and use. Compare how each detector handles gradients, semi-volatile compounds, and fast chromatography.

DO NOT rely only on published sensitivity specifications. The ultimate sensitivity is limited by the physics of light scattering, and is very similar for all instruments. However, other important properties, such as baseline stability, dynamic range, signal to noise ratios at working concentrations, peak width and peak shape are profoundly influenced by instrument design. Choose the detector that demonstrates the best chromatography for your application, not the optimized method used for published sensitivity data. See more about sensitivity.

DO look inside. Are the common maintenance items, such as nebulizer, light source, and light trap, easily accessible? Is the instrument neatly assembled? Are all thermal components well insulated? Are electrical components clear of, and not underneath, liquid lines?

DO consider cost of ownership. Ask for pricing on common replacement parts. A replacement nebulizer is a good test case. Inquire about recommended routine maintenance procedures, frequency and fees.

How to Choose an ELSD based on Sensitivity

How do you determine if one ELSD is more sensitive then the other? First and foremost you MUST test each detector side by side on the same HPLC system. HPLC systems consist of many individual components to provide the separation, the mobile phase, the pumping mechanism, the sample, the injection mode, the column, all the connecting tubing and fittings and the data collection system. All of these components affect the out come of the data produced by the ELSD. There are just too many variables that affect the outcome to accurately compare two ELSD. SofTA Corporation is pleased to provide you with an ELSD for testing and comparison before you purchase. If you are considering another detector, ask the manufacture to demonstrate their detector at the same time.

When selecting a new detector for you HPLC system, one of the most common parameters to compare is sensitivity as it defines the minimum concentration that can be detected. Most if not all manufactures will provide a detection specification in the product literature. Do not rely only on this published value for your comparison, especially when the exact condition for determining this value is not given. In most cases this value is achieved using artificially optimized conditions. It is unlikely that you would be able to reproduce this level for you separation. The ultimate sensitivity of an ELSD is limited by the physics of light scattering and is very similar for all instruments. Other important detector properties should also be considered such as, ease of use, baseline stability especially during gradients, dynamic range, peak width and peak shape. Choose the detector that demonstrates the best chromatography for your application.

It is very important to define your actual detection requirements before getting engulfed in the sensitivity issue. Does you application really require picogram sensitivity or would high nanogram levels be sufficient. Sensitivity usually comes at a cost. Are you prepared to pay a premium for a product with capabilities you will never use?

Next consider your analyte. If you are looking at purchasing an ELSD, you most likely have an analyte without chromaphoric properties. ELS detectors will detect any analyte that is not volatile at the temperatures required to volatilize the mobile phase. An ELSD will be more sensitive then a UV detector for compounds without chromaphores, but may be less sensitive then UV for chromaphoric analytes that are semi-volatile.

To compare the sensitivity of two different detectors you need to look at more than the total peak height or peak area value. These values differ greatly between detector brands due to the signal processing inside the detector, the gain or attenuation settings, and the full scale voltage output settings. Detectors can not be compared based on either the noise or response alone. In order to compare the sensitivity of a detector, it is necessary to determine the minimum concentration that can unambiguously be identified. This measure of sensitivity is frequently referred to as the Signal to Noise ratio, or S:N. Getting to the lowest detectable concentration can be very time consuming. Generally, sensitivity differences between detectors can be seen at signal to noise ratios of 10:1 or greater.

Noise

Detector noise is any change of the detector output that is not related to a peak. Detector noise can been divided into three types, short term noise, long term noise and drift.

Short term noise results from baseline deflections that have frequencies significantly higher than those of an eluted peak. Short term noise is not a serious problem as it is easily removed by appropriate noise filters without significantly affecting the profiles of the peaks. Its source usually originates from either the detector sensor system or the amplifier.

Long term noise results from baseline deflections that have frequencies similar to those of an eluted peak. This type of noise is the most damaging as it can not be differentiated from very small peaks. Long term noise cannot be removed by electronic filtering without affecting the profiles of the eluted peaks. A peak profile can easily be identified above the high frequency noise but is lost in the long term noise. Long term noise ultimately limits the detector sensitivity.

Drift results from baseline deflections that have frequencies that are significantly larger than those of the eluted peak. Drift is almost always due to either changes in ambient temperature or changes in mobile composition, during a gradient, or flow rate. Drift is easily constrained by modifying operating parameters.

Detection and Quantitation Limits

The limit of detection or LOD is the lowest concentration of an analyte in a sample that can be detected, but not quantitated. It is a limit test that specifies whether an analyte is above or below a certain value. The LOD is usually defined as a signal to noise ratio of 2:1 or 3:1 and is expresses as the concentration of the analyte.

The limit of quantitation or LOQ is the lowest concentration of an analyte in a sample that can be determined with acceptable precision and accuracy using the operational conditions of the method. The LOQ is usually defined as a signal to noise ratio of 10:1.

In practice it is almost never necessary to determine the actual LOD or LOQ. Instead the respective limits are shown to be sufficiently low to be able to reliably detect or quantitated the analyte at the desired concentration.

How to measure signal to noise ratios

Sometimes signal to noise ratios are calculated from the standard deviation of the response and the slope of the calibration curve. ELS detection is non-linear, and the response significantly drops off as the concentration decreases. So S:N for an ELSD must be determine experimentally by measuring the peak height in relation to the noise. To do this, make serial dilutions of your analyte until the signal is small enough to see a measurable amount of noise.

Take special note of the differences in signal filtering of each ELSD. Many detectors automatically introduce electronic filters at high sensitivity ranges that may reduce the size of small peaks. To even the playing field, sensitivity measurements should be made without signal filtering. The SofTA ELSD should be evaluated with the filter set to 0 or off. Other detectors should be set at the lowest attenuation (or highest amplification) that does not include noise filtering devices and then corrected to an attenuation of 1.

The shape of the peak used to measure the signal can significantly affect the S:N measurement. It is important to use a peak height measurement instead of a peak area measurement when looking at minimum detection levels because tailing or fronting peaks are difficult to accurately measure at lower concentrations. Signal to noise ratios are easiest to determine by printing the chromatogram and using a ruler to measure both the noise and the signal.

The detector noise is measured by constructing two parallel lines embracing the maximum and minimum signal over the defined time period. The distance between the parallel lines measured in millivolts is taken as the measured noise (v_n), and the normalized noise level (N_D) is calculated with the formula.

where A is the attenuation factor and B is the amplification factor.

To calculate the signal to noise ratio, measure the peak height in millivolts and divide by the normalized noise value.

S:N = S

N_D