The FTS group has two FT spectrometers, similar in design but slightly different in a few respects. Both are scanning Michelson interferometers, with catseye retroreflectors in place of plane mirrors. The moving catseye has a travel of 10 cm either side of zero path difference, giving a resolution limit of 0.025 cm -1 - i.e., a resolving power of 2 million at 200 nm. The double-sided interferogram is important for optimising phase correction in emission spectra. The focal ratio is f/20 (25 mm diameter beam, 500 mm focal length). The angle of incidence on the beamsplitter is 5 degrees, and the whole interferometer is enclosed in a vacuum tank only 1.5 m x 30 cm x 30 cm. Particular attention was paid to the key components of the beamsplitter and the guidance system and drive for the moving catseye. In both instruments the guidance system is a vee-block with large (14 mm) balls. In the older instrument the beamsplitter consists of a silica disc with accurate plane-parallel faces, one of which acts as beamsplitter and the other as beam recombiner. Slightly wedged silica pieces are optically contacted to the non-reflecting sides of the disc to decohere back-surface reflections and reduce shear effects. The drive is hydraulic, with a wobble pin connecting the oil-driven piston to the moving catseye. In the newer instrument the beamsplitter is made in the same way, but the material is magnesium fluoride instead of silica so that the spectrometer can be used (in principle) down to 120 nm. The drive in the newer instrument is a linear motor. A stabilised He-Ne laser beam follows the same path as the signal beam through the interferometer in order to monitor the path difference.
Owing to the catseye retroreflectors, the complementary interferogram returning in the direction of the source is offset from the input beam and is steered to a second detector. The two beams can be added, to halve the effective integration time, or they can be used independently - e.g., with two different detectors. In absorption spectroscopy we have placed the absorption cell in front of one detector and used the other to monitor the continuum background simultaneously, thus establishing a baseline independent of drifts in the intensity of the background source.
The design and performance of the UV FT spectrometer were described in a paper soon after its implementation in the mid-1980s. Progress into the VUV region with the magnesium fluoride beamsplitter and various improvements to the other optical components are summarised in a 1996 paper.
The greater part of our work with these two instruments has been done on emission spectra, using demountable high current water-cooled hollow cathode lamps, with either argon or neon as the carrier gas. The cathodes usually have an internal diameter of 8 mm, and the lamps can be run at currents up to 1 amp from stabilised high voltage power supplies. Intensity calibration can be carried out by recording the spectra of our standard lamps: tungsten halogen, calibrated at NPL for the visible region down to 350 nm, deuterium, also calibrated at NPL, for the UV from 400 nm to 200 nm, and deuterium with magnesium fluoride window, calibrated at PTB, for the UV down to 120 nm.
Absorption spectra require much longer integration times to achieve acceptable signal-to-noise ratios, and the demands on the background source are for both high photon flux and high stability. The high pressure "quiet" xenon arc has proved reasonably satisfactory from the visible to about 240 nm. For shorter wavelengths we used a home-made high current hydrogen capillary discharge lamp until recently, but we now have a commercial high current deuterium lamp of similar intensity but greater stability. Signal-to-noise ratios can also be improved by restricting the spectral bandwidth to the region of direct interest. This can sometimes be done with appropriate filters, but we have also built and used extensively a double monochromator of variable bandwidth and net zero dispersion.
Imperial College Prototype UV Fourier Transform Spectrometer
The Imperial College prototype UV Fourier transform (FT) spectrometer is the original high resolution UV FT spectrometer. The system was designed and built by the Spectroscopy Laboratory and has been used as the prototype for comercial UV FT spectrometers, see the Chelsea FT spectrometer below. The instrument parameters are;
- High resolution, R = 2 000 000 at 200 nm
- Wavelength coverage, 190 to 800 nm
- Portable & compact design, 1.5 x 0.3 x 0.3 m
Chelsea Instruments VUV FT spectrometer
Georgina: zero dispersion monochromator
Hollow Cathode Lamps
The hollow cathode lamp (HCL) is an ideal light source for neutral and singly ionised atomic species. The Imperial College HCL's have accessible, interchangeable cathodes allowing the measurement of metals and many other solid elements.
Penning Discharge Lamp
The Penning Discharge Lamp (PDL) is used for singly and doubly ionised atomic species. As with the HCL the PDL has interchangeable cathodes allowing metals and other solid elements to be observed.
Xe Arc Lamp
The Xe arc lamp has a high intensity continuum from 200 to 500nm and is used for molecular absorption measurements, including our current research on low temperature SO2 photoabsorption cross sections.
High Pressure Deuterium Lamp
The High Pressure Deuterium (HP D2) lamp provides a line free continuous UV spectrum from 180 to 370nm. As with the Xe arc lamp the HP D2 lamp has applications in UV to VUV molecular absorption measurements.
Radiance Standard Lamps
We have both Tungsten (visible to IR) and Deuterium (VUV to visible) intensity calibration lamps. The lamps have been calibrated at the National Physical Libratory, UK and Physikalisch-Technische Bundesanstalt (PTB), Germany. The calibration has a maximum accuracy of 3% at 2s.d. providing relative spectral radiance from 140 to 800nm.
Last updated: 27th March 2012