How to Choose a UV Detector for Chromatography Applications

1. Dynamic Range

1.1. Maximum Absorbance

In any UV detector a beam of light passes through a certain width of liquid to be analyzed. The detector measures absorbance A (expressed in Absorption Units, AU) of the sample, that is determined as follows:

A = - log(intensity of light emerging from the sample cell / Intensity of light directed onto the sample cell)

where log(x) is the base-ten logarithm of x.

for A = 1.00: 90% of the photons are absorbed, 10% reach the detector;
for A = 2.00: 99% of the photons are absorbed, 1% reach the detector;
for A = 5.00: 99.999% of the photons are absorbed, 0.001% reach the detector.

Contemporary UV detectors employ very low intensity of the irradiating UV light. Higher intensities are detrimental for the analyzed samples, and they would have made the detectors dramatically larger, expensive and requiring an extended maintenance. Reliable measurements of less then 0.001% of this low-intensity light is, at the moment, not viable for commercial instruments. Therefore, NO available detector can directly measure absorbance exceeding 5 AU.

1.2. Minimum Absorbance

To be reliably observed, the absorption signal should be at least 10 times higher than the detection noise. The noise is created by minute flow rate and temperature fluctuations of the flowing elutant, and by changes in the detector's electromagnetic surrounding.

In HPLS system the detector's flow cell is maintained at the constant temperature, and the flow rate is strictly regulated, which allows the manufacturers to claim a noise level of 0.00001 AU. Be aware that this value is measured for for a short time span, and with an empty flow cell. Moreover, baseline drift is usually 10 times the noise value or more. Therefore, the minimum height of reliably observed peaks is usually 0.0001 AU.

During Flash separations, the flow rates and temperature are not precisely controlled, hence the minimum height of reliably observed peaks is usually 0.001 AU.

2. Choice of Flow Cells

2.1. Optical Pathlength

The most important factor defining if a particular UV detector may be used for a given separation is the optical pathlength of the detector's flow cell – the width of the analyzed liquid layer – variable b in the Beer-Lambert law:

A = E * b * C, where:

With the exception of our detectors – new generation (Patent pending) multiple-path-length instruments, all other available detectors have flow cells with a fixed b.

For analytical applications (HPLC), flow cells with b = 5 or 10 mm are usually provided. Most application involve 1ug - 1mg analyte applied onto at least 1000 times the analyte weight of the stationary phase. The effluent exiting the column and entering the detector contains very low concentration of the analyte and can be adequately visualized in common single-pathlength flow cells.

For preparative applications (Flash Chromatography, Protein Purification), flow cells with b = 1 or 2 mm are usually provided, which are not always adequate. Preparative chromatographic separations have an optimal ratio of the sample weight to the stationary phase weight. Not obviously, no matter whether you purify 5 mg or 50 g of a target compound, the concentration of eluted product(s) should be identical (for example, 10–50g/l for silica-filled flash columns). Visualization of such high concentrations requires b = 0.01 mm optical cells. However, such a small optical pathlength will "miss" weakly absorbing admixtures.

Optical path
b, mm
Absorbance A, AU
Methyl benzoate,
0.15M
Benzyl alcohol,
0.015M
10 2200 0.18
5 1100 0.09
2 440 0.036
1 220 0.018
0.1 22 0.0018
0.01 2.2 0.00018

The illustrating table presents absorbances A of methyl benzoate, E = 14400 l/(mol cm) at λ= 242nm, and benzyl alcohol, E=12 l/(mol cm) at λ= 250 nm, calculated for various optical pathlengths b at typical (for Flash Chromatography) purification concentrations C = 0.15 M.

Our RD4 and RD2 detectors are the only instruments on the market that simultaneously measure absorption at pathlengths range from b = 1 to 0.01 mm. Our detectors reliably visualizes both strongly and weakly absorbing analytes at essentially any concentration.

2.2. Cell Volume and Back Pressure

Flow cells with smaller total volume will result in better chromatogram resolution. Note, that distributors often report only the illuminated cell volume, which is a just a fraction (sometimes under 3% !) of the total cell volume. Inquire about the total cell volume before making a purchase.

Importantly, the smaller the total cell volume the greater back-pressure it will present to the effluent flow. In addition, backpressure always rapidly rises with increasing flow rate or solvent viscosity.

For analytical applications (HPLC), a detector's backpressure can be neglected due to generally low flow rates and the presence of the high-pressure pump. Therefore, these applications benefit from flow cells with the smallest available total volume.

For preparative applications (Flash Chromatography, Protein Purification), detector's backpressure should not exceed 0.2 - 2 psi at working flow rates (20 - 60 ml/min). Therefore, the appropriate flow cells usually would have total volume of at least 100 ul.

2.3. Cell Materials

Flow cell materials and all tubing/connectors must be chemically/biochemically compatible with the sample used. PTFE/Teflon, PFA, FEP, polyethylene and polypropylene are usually appropriate for almost all samples. Glass, quartz, and sometimes PEEK are not compatible with RNA/DNA (too much binding) and must be surface-treated (siliconized). Stainless steel will corrode at a low pH if chloride or bromide is present in the aqueous solution. Rubber tubing always leaks plasticizer if an organic solvent is used.

3. The Choice of Light Source and Monitored Wavelength(s).

Due to the very nature of the interactions between UV light and matter, shorter wavelengths are, on average, absorbed by matter more strongly than longer wavelengths. Most known compounds absorb UV rays with wavelengths shorter than 160 nm, and are transparent to 300 nm - 900 nm light. Therefore, in order to have the highest possible chance of detecting all components of the reaction mixture, it is advisable to use the shortest possible UV wavelength.

However, chromatographic UV detection can only be done at wavelengths at which the absorbance of the dissolved sample is much stronger than the absorbance of the solvent itself. Each transparent solvent has a characteristic UV wavelength, when the solvent's absorbance begins to gradually increase on decreasing the wavelength of irradiating UV light. This wavelength is presented in blue in the table below. On reaching the wavelength, presented in red, the rise of absorbance accelerates, and very soon the solvent becomes completely opaque. Generally, chromatography detection in a certain solvent is recommended only at wavelengths longer than given in red. Precise quantitative UV measurements should be done at the wavelengths larger than given in blue.

Solvent λ(nm)* λ(nm)* Solvent λ(nm)* λ(nm)* Solvent λ(nm)* λ(nm) *
water 190 185 n-butanol 245 215 tetrahydrofuran 280 220
acetonitrile 200 190 n-propanol 250 210 ethyl acetate 280 260
cyclopentane 220 195 2-butanol 250 200 p-dioxane 290 220
pentane, hexane, heptane 230 200 dichloromethane 245 230 benzene 295 280
cyclohexane 235 200 chloroform 260 240 toluene 315 285
methanol, ethanol, 2-propanol240 205 n-butyl acetate 275 255 acetone 340 330

Obviously, to observe absorbtion of radiation of a certain wavelength you need this particular wavelength to be produced by your light source. Common types of light sources used in UV/VIS spectrophotometers are listed:

TypeSourceWavelength(s)Life Time
Metal vapor discharge lamps;
produce discrete spectrum
Mercury, Hg, low pressure184, 2532000-7000 h
Mercury, Hg, medium pressure365, 405, 4362000-7000 h
Mercury, with Phosphor coating, Hg/P280100-6000 h
Cadmium, Cd225100-6000 h
Zinc, Zn214100-6000 h
Gas discharge lamps;
produce continuous spectrum
Deuterium, D2180-3702000-5000 h
Xenon, Xe250-7003000-5000 h
Incandescent lamps;
produce continuous spectrum
Tungsten/halogen, W370-9001000-4000 h
Solid state sources;
Each produces about 20 nm wide band
Deep UV LEDs240-400 range10-15 years

4. Fixed vs Wariable Wavelength Detectors

Fixed wavelength detector (FWD). The light from the source travels thought a narrow bandpass optical filter, and the exiting nearly monochromatic light irradiates your sample. The wavelength that you monitor is determined by your choice of filter. Usually, you manually change the filter, and only one wavelength can be selected and monitored at a time.

Advantages. The design is often able to produce more precise and reliable readings, especially at higher absorption values (1-3 AU). FWDs are smaller, lighther, cheaper and generally more robust for laboratory conditions. FWDs are advised for preparative chromatography.

Disadvantages. Usually, you manually change the filter, and only one wavelength can be selected and monitored at a time.

Variable wavelength detector (VWD). Typically, the light with continuous spectrum produced by the source directly irradiates the sample. The exiting light is separated into different wavelengths by a prism or diffraction grating, and each separate wavelength can then potentially be monitored. In Scan Detectors, the photometer(s) are moved, usually with step motors, into position to monitor the wavelength that you have chosen to observe. In Photo Diode Array Detectors (called PDA or DAD) the light falls onto a multitude of photo diodes arranged on a single microchip, each photodiode measuring the intensity of a 1-2 nm wide portion of the spectrum. VWDs can monitor the absorption of several wavelengths simultaneously, or even record the entire spectrum.

Advantages. VWDs can monitor the absorption of several wavelengths simultaneously, or even record the entire spectrum. VWDs are advised for HPLC chromatography.

Disadvantages. VWD are more suitable for diluted samples, absorbing less then 1 AU and often become unacceptably noisy at over 2 AU. In general, they are less reliable and do not respond well to being moved arount the lab.

5. Reference Cell

It is important to understand that the presence of a reference cell is

advantageous for stationary spectrophotometers, where sample is not moving through the measuring cell;

not advantageous for chromatography detectors, where solvent/elutant is moving through the measuring cell.

During the UV measurements, reflection losses the incident light occur due to partial (specular) reflections off of the interfaces between different transparent media used within a spectrophotometer.

The fractions of light lost on reflections from the interfaces between detector’s interior/cell’s material are constant for a given detector at a given temperature and atmospheric pressure. Correction for these losses do not require a reference cell.

The fractions of light lost on reflections from the interfaces between cell material/sample media varies depends on the refraction index of sample solution. One of the approaches to partially correct for these losses is to employ a reference cell containing the pure solvent, identical to that in the sample cell.

For the reference cell to be beneficial it must be filled exactly the same solvent as the measuring cell. This is impossible to achieve in chromatography detectors, even if the elutant flow is split before the column to feed the reference cell. The only exceptions are isocratic runs with a single-component elutants. For mixed elutants (hexane/ethylacetate mixture), the solvent components are non-equally absorbed by the stationary phase during the separation, therefore the solvent leaving the column does not have the same components ratio as the solvent fed to the column. The gradient elution will make things even worse because of the added delay due to the column volume.

Moreover, a refractive index gradient is always present in a moving liquid, causing the occurrence of so-called “dynamic fluid lenses," which compromise the measurements. Fluid lenses occurring in the reference cell can have an additional negative effect on the detector's readings.

REACH Devices RD4 detector contains an advanced optical system that actually compensates the true refraction losses in flow-through liquids.

Note, that some detectors use the reference cell not as it is intended, but to cancel out fluctuations in lamp output to steady the baseline. This is a sign of a poor detector design. Better designs steady the baseline by diverting a known fraction of the source light onto a reference photometer (not cell) by a beam splitter.

6. Chromatogram Display

To the best of our knowledge, Reach Devices instruments are the only detectors with embedded graphic monitor, allowing you to see the chromatogram. On the screens of other detectors you can only observe the absorption value at the present moment. Visualization of actual chromatograms requires either a chart recorder (outdated), or a PC. The use of a PC often requires specialized proprietary software from the same company as your detector.

Many detectors have a set of absorption ranges you need to choose from before starting the separation. That way, a chart recorder would allow to fit the largest absorption on the chart. Astonishingly, in many cases, you need to pre-set the range of the absorbencies even when real-time monitoring you chromatogram on a PC.