The evaporative light scattering detector (ELSD) is a universal, mass-based
detector for HPLC, ultraperformance liquid chromatography (UPLC), gel permeation chromatography (GPC), and supercritical fluid chromatography (SFC) that has
received a considerable degree of
acceptance over the last two decades.
It can provide quantitative information
about essentially all compounds
in the sample and does not require
that the compound(s) of interest contain
a chromophoric, fluorophoric,
electroactive, or other type of functional
group to provide a response.
The sensitivity of the evaporative
light scattering detector (ELSD) is
several orders of magnitude greater
than other “universal” detectors such
as the refractive index detector, and
allows for the use of gradient elution.1
The ease of operation of the detector
and the sensitivity that the ELSD can
provide has made it the detector of
choice in a broad range of fields such
as drug discovery and high-throughput
screening. The ELSD is commonly
used for the detection of such diverse
compounds as phospholipids, carbohydrates,
small peptides, nutraceuticals,
and synthetic polymers.
The sensitivity of the ELSD, like that
of all HPLC detectors, is a critical
issue for many chromatographers.
While the sensitivity of the detector is
dependent on many of the same factors
as other detectors (e.g., the compound
should be eluted as a sharp
peak), there are a number of specific
issues that relate to the sensitivity of
the ELSD. As is discussed in detail
here, perhaps the most critical condition
for which to optimize the sensitivity
is to employ a low temperature
for the evaporation of the mobile
phase, especially for compounds that
have a significant vapor pressure
and/or are thermally labile.2,3 In addition,
a number of design features
are described that can optimize
the sensitivity of the detector,
such as the selection of the
appropriate nebulizer and the
use of a focusing gas in the optical
All chromatograms were collected
on a variety of SEDEX LT-ELSD
light scattering detectors)
(SEDERE, SA, Alfortville,
Cedex, France). The column,
solvents (HPLC grade), and conditions
for each chromatogram
are indicated in Figures 1–5.
Steps in the detection of an analyte via ELSD
There are three discrete
stages in ELSD detection
Figure 1 - The three steps involved in low-temperature
evaporative light scattering detection.
- Nebulization. The eluent from the HPLC system is forced through a narrow orifice with a stream of gas flowing through a Venturi tube that sheaths the mobile phase to form small droplets that can be easily evaporated. A flow of N2 at a pressure of 2–4 bar is typically used.
- Vaporization. The mobile phase is removed from the nebulized droplets that contain the compound(s) of interest. The evaporation is performed by passing the droplets through a heated tube.
- Detection. The stream of solute particles that exits the vaporization chamber enters the optical cell where the amount of light scattering, which is related to the mass of the compounds of interest, is measured. The sensitivity of the detector is ultimately determined by the number and shape of particles that are detected by the light scattering chamber in a given period.
While each of these steps is described as
an individual process, the reader should
recognize that the detector is an integrated
system, and if efforts are taken to
optimize one of the steps, these factors
may reduce the effectiveness of another
step. As is typical with multistep detection
processes, optimization of the overall
process will therefore require that a
series of compromises be made.
Role of temperature in optimizing sensitivity via ELSD
Role of temperature in
optimizing sensitivity via ELSD sample contains compounds
that are volatile
and/or thermo labile, a
low temperature is desirable
for vaporization of
the mobile phase since it
will minimize the loss of
analyte and hence optimize
The appropriate temperature
for the detection of a
given compound can be
empirically determined by
collecting the chromatogram
at various temperatures.
Figure 2 presents
of caffeine, which is thermosensitive (it
sublimes), using 30 °C and 50 °C to
evaporate the mobile phase. At 30 °C,
the peak height is approx. 10 times
greater than when 50 °C is employed.
Similarly, Figure 3 demonstrates that
the sensitivity of the detector is
improved by greater than a factor of 10
for urea (another thermosensitive compound)
when the temperature is lowered
from 39 °C to 25 °C.
Figure 2 - Detection of caffeine at 30 and 50 °C evaporation
temperatures. ODS column: 30 × 4.6 mm, particle size: 5 μm,
mobile phase: 70:30 water/acetonitrile, flow rate: 1 mL/min.
Low-temperature detection with gradient separation via ELSD
Figure 3 - Detection of urea at various evaporation temperatures.
Column: Asahipak (Showa Denko, K.K., Tokyo,
Japan) 5 μm, NH2; mobile phase: CH3CN/H2O (85:15);
flow rate: 1 mL/min.
A major benefit of ELSD vis-à-vis other “universal” detection methods (e.g., the use of a refractive index detector) is that gradient mobi le phases can be used. An example of evaporative light scattering detection with a gradient is shown in Figure 4, where glucose and sucrose can be readily determined at the 5-ppm level.* (Note: The background noise of the detector, which is due to issues such as unevaporated solvent particles and residue after evaporation of the mobile phase, sorbent that has been removed from the column packing, and particulate matter from the mobile phase, is a major determinant in the sensitivity of the evaporative light scattering detectors. An example of the level of noise with the gradient used for the separation depicted in Figure 4 is shown in Figure 5.)
Figure 4 - Detection of glucose and sucrose. Column: carbohydrate, 5 μm, 150 × 4.6 mm;
mobile phase: H2O/CH3CN gradient, T = 0 min, 80% B, T = 10 min, 50% B, T = 11 min, 80%
B, T = 14 min, 80% B; flow rate: 1.0 mL/min.
Importance of instrument design on sensitivity
Although the analyst can rapidly
determine the optimum temperature
to maximize the sensitivity, a number
of other instrument-related features
can be employed to enhance the sensitivity
of the detector. These include:
1. Use of a nebulizer to match the appropriate
flow rate. The role of the nebulizer
is to form droplets from the
mobile phase. Since the HPLC
flow rate can range from the low
μL/min range to several mL/min,
the design of the nebulizer is a critical
factor. As an example, a nebulizer
that is designed to optimize
the nebulization of mobile phase at
a flow rate of 0.1 mL/min will not
do a very good job if the flow rate is
2 mL/min. If the nebulizer is
designed to handle low flow rates,
relatively large particles will be
formed when a considerably higher
flow rate is used. This will, of
course, require a higher temperature
to vaporize the mobile phase.
Similarly, if a nebulizer that is
designed for a 2-mL/min flow rate
is used with a 0.2-mL/min flow
rate, the particles that are formed
will be very small.
It is suggested that the instrument
manufacturer provide a set of nebulizers
to cover the flow rate range typically
used for HPLC. In this regard,
nebulizers for capillary chromatography
(~2 μL/min to 75 μL/min), low
HPLC flow (~20 μL to 1 mL/min),
HPLC (~200 μL/min to 2.5 mL/min),
and high HPLC flow (~1 mL/min to
5 mL/min) should be available so
that the analyst can then select the
most appropriate nebulizer for the
2. Use of a drift tube to remove large particles
after nebulization. The nebulizer
typically creates a stream with a broad
range of particle sizes. Since it is desirable
that only the smaller particles
enter the evaporation region (so that
a lower evaporation temperature can
be employed and higher S/N can be
reached, i.e., sensitivity), it is desirable
that the large particles be eliminated.
A simple way to eliminate the large
particles is to allow the nebulized particles
to traverse a drift tube (Figure 5)
in which the larger size particles will
fall to the bottom of the drift tube and
can then be drained from the system.
3. Design of the oven to minimize the pathlength
and ensure rapid heating. The oven should be designed to maximize
the heat transfer to the coil through
which the nebulized particles traverse
during the evaporation process.
All materials should be specifically
selected and configured to the minimum
temperature required to evaporate
the mobile phase.
4. Focusing of the particles. The particle
flux that exits the evaporation
tube in the ELSD is directed to the
optical cell in which the light scattering
of the solid particles is monitored. The signal from the detector
can be increased by using a secondary
flow of inert gas to sheathe
the particles and focus the particles
in the center of the cell to
optimize detection. An additional
benefit of this design is to minimize
the deposition of the particles
on the walls of the detector cell,
reducing the need for periodic
cleaning of the cell.
Figure 5 - Special drift tube designed to eliminate large eluent particles.
The sensitivity provided by an ELSD
is extremely dependent on the design
of each of the components of the
detector. The nebulizer should be
selected to match the flow rate of the
mobile phase, the drift tube should be
designed so that large particles do not
enter the oven, and the particles from
the oven should be focused into the
These features allow the analyst to
employ a lower temperature in the
oven used to evaporate the mobile
phase, so that even thermolabile,
volatile, and semivolatile compounds
can be readily detected at exceedingly
low levels. Typically, low ppm levels of
volatile compounds can be readily
detected, even when low temperature
evaporation is employed.
- Dreux M, Lafosse M, Morin-Allory L. LC·GC Int March 1996.
- Herbreteau B, Morin-Allory L, Lafosse M, Dreux M. J High Res Chromatogr 1990; 13:343.
- Lafosse M, Elfakir C, Morin-Allory L, Dreux M. J High Res Chromatogr 1992; 15:312.
- Pennanec R. LC·GC Sept 2004.
*The sensitivity of an evaporative light scattering detector is typically defined as the minimum
detectable quantity of an analyte that can be detected with a given S/N. Many manufacturers
of ELSD systems present the minimum sensitivity by collecting a chromatogram
with a considerably larger S/N (e.g., S/N = 30 or higher) using a high concentration and
then extrapolating to determine the concentration that would provide a S/N of 3. This
approach is not ideal, since this assumes that the signal at the maximum is due entirely to
the compound of interest. If sensitivity specifications are employed, they are obtained from a
chromatogram with the selected S/N criteria, rather than via extrapolation.
Dr. Pennanec is Applications Engineer,
SEDERE, SA, Alfortville, Cedex, France,
and Dr. Froehlich is President, Peak Media,
10 Danforth Way, Franklin, MA 02038,
U.S.A.; tel.: 508-528-6145; e-mail: firstname.lastname@example.org.