The mid-infrared spectral region is particularly important for optical spectroscopy, since it contains many absorption lines which are characteristic for certain molecules. In particular, that applies to the “fingerprint region” with wavelengths between 7 μm and 11 μm. The absorption lines in that region are related to quantized molecular vibrations, and they can be used to distinguish many molecules and to measure their identity. Typical applications are in environmental monitoring and medical sciences.
According to ISO 20473:2007, the mid infrared spans the whole wavelength region from 3 μm to 50 μm. Some spectrometers can operate in much of that wide region.
Challenges in the Mid-infrared Region
Spectrometers for operation in the mid-infrared spectral region can in principle be made similar to other spectrometers, but substantial challenges arise for various reasons:
- A main problem is the limited availability of well performing photodetectors in that spectral region. Particularly multi-element photodetectors such as photodiode arrays are hard to make for the mid infrared.
- For laser spectroscopy, there is a limited choice of narrow-linewidth tunable mid-infrared laser sources.
- Another challenge can be the thermal radiation emitted by all absorbing objects at room temperature; note that the thermal energy kB T is larger than the photon energy in the mid-infrared region, permitting substantial emission. This is actually a substantial part of the problem with mid-infrared detectors; many of them have to be used at cryogenic temperatures (e.g. with liquid nitrogen at 77 K) in order to sufficiently suppress thermal noise.
For those reasons, mid-infrared spectrometers are not as widely available as those for the near-infrared and visible spectral regions, and often exhibit lower performance e.g. in terms of sensitivity (or signal-to-noise ratio) or spectral resolution.
Fourier Transform Spectroscopy
A common solution is to use Fourier transform spectroscopy. (For IR applications, that is often called FTIR, meaning Fourier transform infrared.) Here, a key advantage is that a single-element photodetector can be used.
As a broadband light source for absorption spectroscopy in the mid infrared, one often uses a thermal source – for example, a Nernst glower based on an electrically heated rod made of zirconium/yttrium ceramics.
A limitation of Fourier transform spectroscopy is that such a spectrometer operates in a scanning mode, requiring substantial time for recording a single spectrum, and a higher precision of the used optical delay line.
Although the availability of mid-infrared lasers for applications in spectroscopy has been very limited for many years, some developments have allowed for substantial progress. In particular, quantum cascade lasers are now available, which emit significant optical power with a small linewidth and are tunable over some wavelength range. Although a quantum cascade laser can cover only a quite limited spectral range, such lasers can be made in a very wide range of emission wavelengths, and due to their compactness one combine multiple lasers in one spectrometer.
Even frequency comb sources are now available in the mid infrared, and are particularly useful in spectroscopy. By employing supercontinuum generation in an optical fiber, one can make such sources covering a large wavelength range in the infrared.
Tunable mid-IR laser sources can also be realized with optical parametric oscillators. These can cover wide wavelength regions with a high output power and narrow linewidth, but such OPO systems tend to be relatively complex and expensive.
A high spectral resolution often results from the narrow linewidth of the laser light source used. In other cases, it results from high wavelength discrimination in the detection.
The problem of mid-infrared photodetectors can be circumvented by upconverting mid-infrared light into shorter wavelength regions (typically the near infrared around 0.8 μm to 1 μm), where well performing photodetectors are available. The upconversion can be achieved with sum frequency generation (SFG) in a nonlinear crystal, in which the mid-infrared light interacts with laser light, e.g. from a Nd:YAG laser emitting at 1064 nm. Typically, the mid-infrared light to be detected is quite faint, while laser light with a substantial optical power is applied such that a significant part of the mid-infrared light is converted into the near infrared.
Note that sum frequency generation can be understood to be based on a nonlinear process where one photon from each input source is converted into a single photon at the output wavelength. In the ideal case, with full conversion of all mid-infrared input photons and a near-infrared detector with a high quantum efficiency, essentially shot-noise-limited detection is possible.
The nonlinear conversion is constrained by the requirement of phase matching. Ideally, the used nonlinear crystal material should not only be highly transparent in the complete relevant spectral region, but also offer a suitable phase matching scheme (e.g. with sufficiently broad phase-matching bandwidth) and a high nonlinear coefficient (in order to reduce the amount of required laser power). It has been found that silver gallium sulfide (AgGaS2, AGS) is quite suitable, having a wide transparency range from 0.85 μm to 11 μm, a reasonably high nonlinear coefficient (around 16 pm/V) and suitable phase-matching properties for using a YAG laser at 1064 nm, and promising results have been demonstrated . The sum frequency wavelengths are then in the region around 0.9 μm, which is very suitable for using silicon-based photodetector technology. For example, linear photodiode arrays and two-dimensional focal plane arrays (image sensors) are available with good performance at low cost.
Using non-collinear phase matching in conjunction with a laser beam having a substantial beam radius, one can utilize the resulting angular dispersion: different mid-infrared wavelengths result in different cones of the output light, and by detecting that light with a focal plane array, one later associate different detector pixels with different wavelengths. The same nonlinear crystal may be used at different angular orientations to cover different parts of the spectral region of interest.
For continuous detection, one has to use a continuous-wave laser, but it is also possible to use a pulsed laser, obtaining high sensitivity at the times of the laser pulses and no sensitivity at other times. Such time gating may actually be useful for some applications.
Questions and Comments from Users
Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the author’s answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.
Please do not enter personal data here; we would otherwise delete it soon. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him e.g. via e-mail.
By submitting the information, you give your consent to the potential publication of your inputs on our website according to our rules. (If you later retract your consent, we will delete those inputs.) As your inputs are first reviewed by the author, they may be published with some delay.
|||P. Jacquinot, “New developments in interference spectroscopy”, Rep. Prog. Phys. 23 (1), 267 (1960), doi:10.1088/0034-4885/23/1/305|
|||F. Adler et al., “Mid-infrared Fourier transform spectroscopy with a broadband frequency comb”, Opt. Express 18 (21), 21861 (2010), doi:10.1364/OE.18.021861|
|||J. Mandon et al., “Fourier transform spectroscopy with a frequency comb”, Nature Photon. 3, 99 (2009), doi:10.1038/nphoton.2008.293|
|||P. Tidemand-Lichtenberg et al., “Mid-infrared upconversion spectroscopy”, J. Opt. Soc. Am. B 33 (11), D28 (2016), doi:10.1364/JOSAB.33.000D28|