Crystal analyser spectrometer

Detailed considerations of resolution, operation and calibration are similar to those for crystal analyser spectrometers ( 3.4.2.3) see also Appendix 3. 

The great advantage of crystal analyser spectrometers is that there are no moving parts in the spectrometer. Even physically moving the instrument is not problematic because TOSCA uses a kinematic mount system and the whole spectrometer can be removed and replaced within a day but the calibration remains unaltered. An advantage of this feature is that the conversion to S(Q,co) can be initiated automatically when the current spectrum has finished accumulating, with fiill confidence in the... 

In this Appendix we derive the analytical expression for the energy resolution of a low-bandpass spectrometer like TOSCA ( 3.1) (also known as crystal analyser spectrometers) and describe two key features of the design ( 3.2), time focussing ( 3.2.1) and the Marx principle ( 3.2.2) that improve the resolution at high and low energy transfer respectively. 

Fig. A3.1 Schematic of tfie time focusing effect on a crystal analyser spectrometer.

Note that the Marx principle is more restrictive than time focussing in the latter all that is necessary is that the sample, analyser and detector planes are parallel, in the former the planes must not only be parallel but the sample and detector planes must also be co-planar. Time and energy focussing are so successfiil that all current and planned crystal analyser spectrometers employ it. The only exception is the forward scattering bank on TOSCA where limitations on space meant that the detectors were placed slightly downstream from the sample plane. In practice, the displacement is small enough that the resolution is essentially the same in both the forward and backscattering detector banks. 

In this second class of spectrometers, X-ray radiation emitted by the sample, after it has been filtered by a sheet collimator (Soller slits), impacts on a crystal analyser... 

Filter instruments offer good intensity at modest energy resolutions and crystal analyser instruments offer good resolution with modest intensities. While a ciystal analyser instrument is conceivable at a continuous source (such an instrument would be a specialised form of the triple axis spectrometer ( 3.4.1) [17]) none have yet been constructed to exploit the advantages of low final energies. 

Of the low final energy crystal analyser instruments, there were several early spectrometers that incorporated some of the elements found in current machines. The first instrument to include all the elements was installed at KENS (Tsukuba, Japan) in die early 1980 s [18]. Similar instruments were commissioned over the next decade NERA-PR at IBR-2 (Dubna, Russia) [24], CHEX at IPNS (Argonne, USA) [25] and TFXA at ISIS (Chilton, UK) [26]. The operating principle of all these... 

Direct geometry instruments use choppers or crystal monochromators to fix the incident energy and they are found on both continuous and pulsed sources. To compensate for the low incident flux resulting from the monochromation process, direct geometry instruments have a large detector area. This makes the instruments expensive, they are generally twice the price of a crystal analyser instrument. At present, they are used infrequently for the study of hydrogenous materials, so we will limit our discussion to a chopper spectrometer at a pulsed source and a crystal monochromator at a continuous source. 

Time focussing was first introduced in the LAM-D spectrometer at KENS [1] and has been adopted in all subsequent crystal analyser instruments. The key feature is that the planes of the sample, analyser and detector are parallel as shown in Fig. A3.1. For simplicity, a point sample is assumed. From Fig A3.1 for the ray shown by the solid line it can be seen that the length of the final flight path from sample to detector, df, is given by ... 

Fig. II.l. Triple-axis spectrometer i collimator, 2 crystal monochromator, 5 sample holder, 4 sample crystal, 5 crystal analyser, 6 detector. Drawn from Iyengar (1965)...

Figure 3 An X-ray fluorescence spectrometer. 1, X-ray tube 1a, electron source 1b, target 2, substance investigated (secondary anode) 3, crystal analyser 4, registration device hv, primary radiation hv2, secondary radiation and hv, registered radiation.

Thus rapid motions require relaxed resolution, while slower motions require high resolution. Note that these are relative terms typically an energy resolution of better than 1% AE/E and 100 peV is required. The timescale to be probed can be separated into three regimes, each of which uses a different technique for r 10" s, AE is 10-100 peV and direct-geometry time-of-flight is used, for t 10" s, AE is 0.3- 20 peV and a backscattering crystal analyser is used, for t 10" s, AE is 0.005-1 peV and neutron spin echo is used. Examples of each type of spectrometer will be considered. 

An alternative type of spectrometer is the energy dispersive spectrometer which dispenses with a crystal dispersion element. Instead, a type of detector is used which receives the undispersed X-ray fluorescence and outputs a series of pulses of different voltages that correspond to the different wavelengths (energies) that it has received. These energies are then separated with a multichannel analyser. 

Simultaneous spectrometers consist of various combinations of analyser crystals and detectors, arranged around the sample at fixed angle settings. Use of a multichannel X-ray spectrometer with simultaneous determination of up to 24 elements can considerably increase the analysis speed (a few seconds to a few minutes). 

The irradiating X-ray beam cannot be focussed upon and scanned across the specimen surface as is possible with an electron beam. Practical methods of small-spot XPS imaging rely on restriction of the source size or the analysed area. By using a focussing crystal monochromator for the X-rays, beam sizes of less than 10 pm may be achieved. This must in turn correspond with the acceptance area and alignment on the sample of the electron spectrometer, which involves the use of an electron lens of low aberration. The practically achievable spatial resolution is rarely better than 100 pm. A spatial resolution value of 200 pm might be regarded as typical, and it must also be remembered that areas of up to several millimetres in diameter can readily be analysed. 

This point analysis is employed to analyse a selected region of chemically homogeneous composition, such as a phase. The electron beam is stopped and positioned carefully on the point selected on the SEM screen, and the composition of the sampling volume is determined by a crystal spectrometer. 

Figure 13.8—Reflecting crystals used in goniometers of dispersive spectrometers and Bragg s relationship. The higher the wavelength to be analysed, the higher the inter-reticular distance must be in the crystal.

---