FREE shipping to on qualifying orders when you spend or more. All prices ex. VAT.

UV-Vis and IR Spectroscopy: How to Use a Spectrometer

The Ossila Broadband White Light Source and Spectroscopy Transmission Holder set up for visible spectroscopy

Optical spectroscopy concerns light in the UV, visible and IR regions of the electromagnetic spectrum with is either emitted by, or interacts with, a sample.

In optical spectroscopy, light is split into its constituent wavelength components by a grating or prism inside an optical spectrometer in order to study how it is affected by interaction with a certain material. The spectrometer outputs the intensity of the light as a function of wavelength and from this, important information on the electronic structure of the material can be obtained.

A typical set up for spectroscopy consists of a spectrometer, the sample (usually held in either a transmission holder or cuvette holder), and a light source. These may be connected together with optical fibers ("coupled") and/or bolted to an optical table or breadboard plate.

UV, visible and IR Spectroscopy can be performed using a range of different light sources which can typically be categorised as being either monochromatic or broadband. The type of light source used in any spectroscopy set up will depend on the type of measurement being taken. In general, monochromatic sources are used to excite materials (for example, in fluorescence measurements) whereas broadband sources are used for absorption, transmission and reflection measurements.

Complete Optical Spectroscopy Bundle

  • Spectrometer
  • Broadband Light Source
  • And More

£1600.00With Free Shipping

To get started with the Ossila Optical Spectrometer, simply plug it in to a computer via the supplied USB-C cable and launch the spectroscopy software. The spectrometer can be controlled via serial commands (please see the Optical Spectrometer user manual for details) but the spectroscopy software is the easiest way to interface with the device, take measurements, and save and analyse data.

We recommend waiting for around 30 minutes for the internal temperature of the spectrometer to stabilise before taking any measurements.

Transmission and Absorption

Ambient light

We recommend performing measurements in the dark where possible. This reduces the effects of ambient light, which can interfere with spectroscopic measurements. The measurement apparatus (i.e. spectrometer, light source, and sample holder) should also be firmly secured (e.g. on an optical breadboard) before starting measurements as any changes in position during the measurement will affect the results.

Transmission measurements make it possible to quantify how opaque a sample is to a particular wavelength of light. The transmission, T, of a sample is defined as the ratio of the intensity of light incident on the sample, I0, to the intensity of light emerging from the other side of the sample, I:

The transmission of a sample is wavelength-dependant; optical spectrometers measure the transmission of a sample for many wavelengths simultaneously.

To perform a transmission measurement, both the light that is incident on the sample and the light that passes through it need to be measured. This can be done by directing a beam of light through a sample and into the optical spectrometer.

Spectroscopy set up for transmission measurement

In practice, a transmission spectrum requires three separate spectra to be recorded. These are:

  1. A background spectrum. The detector in the spectrometer will record a signal even when there is no light falling on it. This is due to the thermal generation of electron-hole pairs, which produce a “dark” signal. This background is not related to the sample and so needs to be subtracted from all subsequent spectra in order to account for these effects. A background spectrum is taken by turning off the illumination source and recording a spectrum.
An example background spectrum, taken with the excitation source/lamp turned off or blocked. The spectrometer will record a non-zero signal even when there is no light falling on its detector.
  1. A reference spectrum, I0. This is the spectrum that will measure how much light is incident on the sample. This spectrum can be taken by turning on our illumination source and recording a spectrum without the sample present. It is often necessary to use a reference sample in place of our actual sample. For example, if you are interested in the transmission of a thin film that is deposited on a glass substrate, you may wish to exclude the effect of the glass in the measurement. In this case, place a blank glass substrate in the light path for the reference measurement. Or, if you are measuring the transmission of molecules in solution inside a cuvette, you may want to exclude the effect of both the solvent and the cuvette. In this case, use a cuvette containing just the solvent as the reference sample. The background should be subtracted from the measured spectrum.
  2. A measurement spectrum, I. Place the sample in the light path and take the measurement spectrum. The background should again be subtracted from this spectrum.
A reference signal (I0) must be taken with the lamp turned on/unblocked but without the sample in place (or by using a reference sample). The sample is then placed into the setup and the lamp spectrum transmitted through it (I) recorded.

If the 'Transmission' measurement type is selected in software, the background subtraction and calculation of the transmission is performed automatically, and the final transmission spectrum displayed. A transmission spectrum should have values between 0 and 1; 0 indicating that all light has been absorbed by the sample, and 1 indicating that no light was absorbed in the sample. Any values less than 0 or greater than 1 suggest a problem with the background spectrum and/or reference spectrum.

An example transmission spectra, calculated by dividing the sample spectrum by the reference spectrum. This can also be calculated automatically by the spectrometer.

As light propagates through a material, its intensity will fall exponentially with distance. If a light with intensity I0 and wavelength λ is incident on a surface, the intensity at a distance d below the surface will be:

Here, α(λ) is the absorption co-efficient of the material for a given wavelength. This is known as the Beer-Lambert law. The larger the absorption coefficient, the more light the material absorbs for a given thickness. A material’s absorption coefficient will be wavelength dependent. For example, it may have a large absorption coefficient for UV light but a low absorption coefficient for visible light (e.g. glass) or a large absorption coefficient for visible wavelengths but small in the IR (e.g. silicon). A low absorption coefficient leads to high transmission.

The absorbance, A, of the sample can be found by taking the base 10 logarithm of the transmission:

The factor of 0.43 comes from us taking the base 10 logarithm of the transmission rather than the natural logarithm. In some scientific fields the absorbance is defined by the natural logarithm of the transmission, but generally in the study of liquids and solids the base 10 logarithm is used. From the equation above, the absorbance of a sample can be calculated by first calculating the transmission and then simply taking the negative logarithm.

The absorbance spectra (also known as the optical density (OD)) of the sample.

The table below shows the relationships between different values of transmission and absorbance.

Transmission Absorbance
0.01 2
0.02 1.7
0.05 1.3
0.1 1
0.2 0.7
0.5 0.3
1 0

Note that there is a large caveat to calculating absorbance from transmission: it assumes that no light is reflected from the sample surface and no light is scattered and lost in the sample. These are additional loss mechanisms, in addition to absorption, that prevent light from passing through the sample. This technique for measuring absorption is therefore not recommended for turbid/rough media where high scattering losses are expected, or media with a high refractive index where significant reflection can occur at the interface with air.

Transmission and absorption measurements are particularly susceptible to noise. All three spectra involved in the calculation will have their own associated noise which sum when the calculations made. There are some approaches to reduce noise and achieve a clean transmission or absorption spectrum.

  1. Choose the right light source (see spectroscopy light sources). There should be a high intensity of light over the wavelength range being measured. If there is no light, there is nothing to be absorbed by the sample and nothing to be measured by the spectrometer and the data will be meaningless.
  2. Use the correct integration time. To maximise the signal, use an integration time that gives a clear signal for the reference spectrum. About 90% of the spectrometer saturation values is ideal. Light sources with higher intensities allow shorter integration times to be used.
  3. Use spectral averaging. By averaging several spectra, the noise in the average spectrum is reduced by the square root of the number of samples. Using a bright source, leading to short integration times means that many spectra can be recorded and averaged quickly.
  4. Be aware of limitations and prepare the sample accordingly. If a sample is very highly absorbing and almost no light can pass through, this leads to a similar situation as using a dim light source i.e. there will be no signal (or the signal will be obscured by noise) and so nothing to measure. Ideally, the transmission of the sample should remain above 0.05 for all wavelengths (i.e. the absorption is below 1.3). Practically, this can be achieved by reducing the concentration of molecules in solution if working with liquids or reducing film thickness if working with thin films. If this is not possible, the points above become particularly important as the noise will limit the maximum measurable absorption. Spectra showing very high absorption may not be accurate.

Light Sources for Spectroscopy

Monochromatic Light Sources

Light emitting diodes (LEDs)

LEDs are a popular monochromatic light source due to their narrow emission spectrum, low power consumption, high stability, long lifetime and fast switching. In the past they were made from inorganic semiconductors, such as gallium arsenide (GaAs) and gallium arsenide phosphide (GaAsP). However, organic LEDs (OLEDs) have now become popular due to the broad range of colours which they are able to produce; by adding different functional groups to organic molecules, it is possible to alter their emission wavelength, making it relatively easy to fabricate LEDs of any colour.


Lasers produce monochromatic, coherent, collimated light through the process of stimulated emission (hence LASER, or “Light Amplification by Stimulated Emission of Radiation”).

There are two types of lasers, continuous wave lasers and pulsed lasers. Continuous wave (CW) lasers produce a constant beam of photons with no fluctuation in power over time. Diode CW lasers are similar in design to LEDs, and are often used for measurements where very high powers are not necessary, such as fluorescence measurements. Compared to continuous wave lasers, diode lasers are much more affordable.

A typical fluorescence setup for spectroscopy, where the sample is illuminated by a monochromatic light source - in this case, a CW laser - and the emitted light is collected by a spectrometer via an optical fibre. The between the sample and the detector blocks light from the excitation source.

Pulsed lasers are very powerful as they are able to deposit very high amounts of energy in a short space of time. The pulse length used for fast spectroscopy is usually on the order of picoseconds (10-12 s) or femtoseconds (10-15 s), though attosecond (10-18 s) pulses are also possible. Pulsed lasers are often used in time-resolved measurements, such as transient absorption (pump-probe), or measurements that require very high energies - for example, as an excitation source for other lasers.

Pulsed lasers can also be used in non-linear optics to produce pulses of different wavelengths, such as in second harmonic generation (frequency doubling) or optical parametric amplification.

Broadband Light Sources

Broadband LEDs

Although individual LEDs produce light with a very narrow spectrum, multiple LEDs can be combined to produce a broader spectrum. It is also possible to coat the LEDs with phosphors - materials that absorb UV and blue light and re-emit in the visible - in order to cover the entire UV-vis spectrum. In this way, the Ossila Broadband White Light Source is able to produce light covering a spectral range of 360–900 nm.

LED light sources are typically more expensive than incandescent and gas discharge lamps, but their extended lifetimes mean they need to be replaced much less often, making them cheaper in the long run. They are also significantly more efficient as there is no energy lost through heat and their “warm up” and “cool down” times are instantaneous. Broadband LEDs can be powered over USB, are less fragile than lamp type light sources, and do not contain hazardous gases.

Broadband White Light Source

  • USB-C powered
  • 360 nm - 900 nm
  • Low Price

Available Now £120.00

Spectrometer software showing white light source spectrum
Spectrometer software showing the spectrum for the Ossila Broadband White Light Source

Tungsten halogen

Tungsten halogen lamps (also referred to simply as halogen lamps or as quartz iodine lamps) are a type of incandescent lamp that emit from the UV-visible light boundary to the infra-red region. The exact spectral range depends on the temperature of the filament, but they are generally not suitable for measurements in the UV.

Tungsten halogen lamps consist of a tungsten filament inside a glass bulb. For this, quartz glass is used as it has a high melting point and is capable of withstanding high pressures without breaking. The capsule is filled with a mix of an inert gas, such as krypton or xenon, and a halogen, such as iodine or bromine.

The tungsten filament is heated by passing an electric current through it so that the filament becomes incandescent (it emits light). Most of the energy is emitted in the infrared, making tungsten halogen lamps very inefficient for day-to-day lighting but suitable for spectroscopy measurements in the IR region.

The inert gas in the glass bulb reduces the evaporation and oxidation of the tungsten filament, while the halogen helps to redeposit the tungsten particles back onto the filament through the “halogen cycle”. This increases the lifetime of the filament compared to incandescent lamps that do not contain any halogen, and reduces blackening caused by the deposition of tungsten particles on the inside of the glass.

Typical UV-vis transmission setup, with the sample illuminated by a light source. The transmitted light is collected by a spectrometer via an optical fibre.

Gas discharge/arc lamps

Arc lamps are a type of gas discharge lamp which produce light by sending an electric discharge current through a plasma (an ionised gas). Generally, an electric field is applied between two electrodes inside a heat-resistant glass tube filled with the gas. The atoms become excited through ionisation or through collisions with electrons or other gas atoms or ions. When the atoms or ions relax back to the ground state, a photon is emitted. The wavelength of this photon is characterised by the gas used.

Deuterium arc lamps are commonly used in UV spectroscopy as they produce a continuous spectrum from around 180-370 nm (though there is non-continuous emission up to 900 nm). They are almost always combined with a tungsten halogen lamp to allow measurements in the UV, visible, and NIR.

Xenon arc lamps typically produce a continuous spectrum over a wavelength range of 190-1100 nm. This makes them more efficient than deuterium/tungsten halogen lamps as they can cover the same spectral range with only one lamp. However, they are both more expensive and less stable.

Light from discharge gas lamps is unpolarised and incoherent. Often they take a while to reach full light output power, but despite this, they are still more efficient than incandescent lamps.


Back to Top

How Useful Was This Page?

Contributing Authors

  • David Coles
  • Kristy McGhee
Return to the top