The present invention relates to an X-Ray analysis device, in particular an X-Ray analysis device adapted to detect the presence of a crystalline material of interest in a sample, the crystalline material having a lattice structure comprising at least a first lattice spacing and a second lattice spacing.

Various techniques may be used to identify crystalline materials, including those materials classed as polycrystalline that is, comprising many crystallites, such as a sample of rock. One particular technique useful for crystalline materials is X-Ray diffraction, particularly X-Ray powder diffraction. The behaviour of X-Rays striking a crystalline structure is governed by Bragg's Law: A = 2dsinQ where A is the wavelength of the incident X-Ray radiation, d is a lattice spacing of the crystal material diffracting the X-Rays, and 2Θ is the angle through which the X-ray radiation is diffracted. The lattice spacing d is the spacing between the planes in a crystal lattice, and can be calculated for each of the 14 Bravais lattice types based upon a combination of the lengths of each side of a unit cell making up the lattice (a, b and c). This set of d spacings is unique for each crystal structure, thus making X- Ray diffraction a useful tool in the identification of crystalline materials of interest. One commonly used technique is to use a fixed X-Ray wavelength, for example a Ka X-Ray line, and, using a detector of appropriate sensitivity, scan through a range of Θ, such as in angle-dispersive X-Ray diffraction (ADXRD). The commonly-used Bragg-Brentano diffractometer works on this principle. An alternative implementation of the Bragg equation is to fix the scattering angle and scan the X-ray wavelength (equivalently, the X-ray energy). This method can also be implemented without scanning the X-ray wavelength: a broadband X-ray source, such as an X-ray tube, can be used together with an energy-resolving detector. In either case, this technique is known as energy-dispersive XRD (EDXRD). There are several advantages of using EDXRD over ADXRD, including the relatively simple implementation (fixed geometry with no moving parts), simultaneous detection of the entire diffraction pattern and shorter exposure times. However, there are also disadvantages of using the technique, since the X-Ray source has a non-uniform spectral distribution, and diffraction and fluorescence peaks may overlap, making analysis relatively complex.

However, one further issue that may occur is the effect that any sample morphology and surface roughness may have on the resultant EDXRD spectra. One solution to minimise this issue is to use a back-reflection (2Θ close to 180°) geometry, as described in Hansford, J. Appl. Cryst. (2011), 44, 514 - 525. In this geometry sample morphology, surface roughness and distance between sample and source have little effect on the spectra obtained, aside from variations in signal intensity. X- Ray fluorescence peaks observed in the recorded spectrum are likely to overlap the region where diffraction peaks are observed. Work has been done in suppressing X- Ray fluorescence peaks, such as in J. Appl. Cryst. 47, 1708 - 1715, and in the ability of a back reflection geometry to be used on samples with minimal sample preparation, such as in Hansford et al., Nucl. Instr. Meth. Phys. Rev. A, 728 (2013) 102 - 106. However, despite EDXRD being a valuable tool in materials analysis, there are still improvements to be made. In particular, X-ray powder diffraction methods are normally used to identify and possibly quantify all of the phases in a sample. However, there are some circumstances in which the user is interested in the presence (and quantity) or absence of just one key mineral or phase in the sample and has no particular interest in identifying or quantifying the remaining components. In this situation, merely detecting the presence of a material of interest may be sufficient.

The present invention aims to address this issue by providing an X-Ray analysis device adapted to detect the presence of a crystalline material of interest in a sample, the crystalline material having a lattice structure comprising at least a first lattice spacing and a second lattice spacing comprising: an X-Ray source arranged to emit a first characteristic X-Ray line and a second characteristic X-Ray line both incident on the sample; an X-Ray detector adapted to detect X-rays diffracted by the sample; wherein each of the first and second characteristic X-Ray lines is diffracted at a distinct diffraction angle determined by each of the first and second lattice spacings to produce four diffraction angles, and wherein two of the diffraction angles correspond to one another.

The correspondence of two of the diffraction angles allows a single detector, placed at the relevant diffraction angle, to detect the two characteristic X-ray energies simultaneously, constituting a signature of the crystalline material of interest.

Preferably, the two diffraction angles are coincident. Preferably, the first characteristic X-Ray line has a first emission energy and the second characteristic X-Ray emission line has a second emission energy, and wherein the ratio of the first lattice spacing to the second lattice spacing corresponds to the inverse ratio of the first emission energy to the second emission energy. In some circumstances, the X-Ray source may comprise a single element capable of emitting at least two characteristic X-Ray lines. Alternatively, the X-Ray source may be a combination of at least two elements each capable of emitting at least one characteristic X-Ray line. In this latter situation, the X-Ray source is preferably an alloy of at least two elements each capable of emitting at least one characteristic X- Ray line.

Preferably, the device further comprises a collimator arranged to collimate the first characteristic X-Ray line and the second characteristic X-ray line on exiting the at least one X-Ray source.

The device may also further comprise an image detector. Preferably the image detector is a charge-coupled device. The detector may be an energy-resolving detector. The device may also further comprise a filter, the filter being arranged to filter the X- ray spectrum diffracted by the sample before entering the detector. Alternatively, the device may further comprise a filter, the filter being arranged to filter the first and second characteristic X-ray lines emitted by the X-ray source. Preferably the at least one X-Ray source is arranged in a back-reflection geometry with respect to the sample.

The crystalline material of interest has at least a third lattice spacing, and the X-ray source is arranged to emit at least a third characteristic X-Ray line, and wherein the third characteristic X-Ray line is diffracted at a further diffraction angle by the third lattice spacing, wherein the diffraction angle corresponds to those produced by the first and second characteristic X-Ray lines. The crystalline material of interest may be a single crystalline phase or a combination of at least two crystalline phases.

The present invention also provides for the above device to be adapted to be housed within a handheld unit.

In another aspect the present invention provides a method of X-Ray analysis of a sample to determine the presence of a crystalline material of interest, the crystalline material having a characteristic ratio of lattice spacings, comprising: selecting an X- Ray source having a first characteristic X-Ray line and a second characteristic X-Ray line; diffracting the first characteristic X-Ray line and the second characteristic X-Ray off the sample; detecting the X-Rays diffracted by the sample; wherein each of the first and second characteristic X-Ray lines is diffracted at a distinct diffraction angle determined by each of the first and second lattice spacings to produce four diffraction angles, and wherein two of the diffraction angles correspond to one another.

Preferably, the X-Ray source comprises a single element capable of emitting at least two characteristic X-Ray lines. Alternatively, the X-Ray source is a combination of at least two elements capable of emitting at least one characteristic X-Ray line. In this situation, the X-Ray source is preferably an alloy of at least two elements capable of emitting at least one characteristic X-Ray line.

Preferably, the first characteristic X-Ray line has a first emission energy and the second characteristic X-Ray emission line has a second emission energy, and wherein the ratio of the first lattice spacing to the second lattice spacing corresponds to the inverse ratio of the first emission energy to the second emission energy.

The present invention will now be described by way of example only, and with reference to the accompanying drawings, in which:

Figure 1 is a schematic perspective view of a device in accordance with a first embodiment of the present invention adapted to detect the presence of a crystalline material of interest in a sample;

Figure 2 is a schematic cross-section view of a device in accordance with a second embodiment of the present invention adapted to detect the presence of a crystalline material of interest in a sample;

Figure 3 is a chart illustrating the emission energies of the L-series of characteristic X-Ray lines of an X-Ray source having a Pd (Palladium) anode;

Figure 4 is a chart illustrating the emission energies of the L-series of characteristic X-Ray lines of an X-Ray source having a Pd (Palladium) anode, a simulated EDXRD spectrum for quartz at 2Θ = 154.6°, and the simulated absorption spectrum of rhodium foil;

Figure 5 is a chart illustrating the experimental confirmation of the correspondence between the X-Ray diffraction angles in quartz with a Pd X-Ray source;

Figure 6 is a chart illustrating the X-Ray diffraction spectrum obtained at 2Θ = 154° for a mudstone rock specimen known to contain quartz, with the prominent X-Ray fluorescence peaks labelled;

Figure 7 is a chart illustrating the X-Ray diffraction spectrum obtained at 2Θ = 154° for the same mudstone rock specimen mounted in three different positions: a reference position, tilted by 30° and shifted by 2 mm away from the source and detector;

Figure 8 is a chart showing the simulation of the EDXRD spectrum of a sample of steel containing (by volume) 5% austenite, 60% martensite, 20% ferrite and 15 % cementite; and

Figure 9 is a chart showing the enhanced detection of austenite in a steel sample using a source which emits In and Ti characteristic X-rays simultaneously.

The present invention takes the approach of maximising the X-Ray diffraction signal obtained at fixed 2Θ by exploiting a coincidence of diffraction angles created by different X-Ray emission energies. An X-ray analysis device can be adapted to detect the presence of a crystalline material of interest. The crystalline material will have a lattice structure, which comprises at least a first lattice spacing and a second lattice spacing. For example, imagining the unit cell of a crystalline lattice being assigned a Cartesian co-ordinate system, the distance between adjacent lattice planes in the x-direction gives rise to a first lattice spacing, and the distance between adjacent lattice planes in the y-direction gives rise to a second lattice spacing. An X- Ray source can be arranged to emit a first characteristic X-Ray line and a second characteristic X-Ray line, both of which are incident on the sample. An X-Ray detector adapted to detect X-Rays diffracted by the sample is used to detect any spectrum obtained. Each of the first and second characteristic X-Ray lines is diffracted at a distinct diffraction angle determined by each of the first and second lattice spacings to produce four diffraction angles, and two of these diffraction angles correspond to each other. This correspondence, or in highly accurate cases, coincidence, can be used to enhance the signal detected when X-rays are diffracted from a material of interest. The simultaneous observation of the two X-ray energies constitutes a signature of the material of interest and may be used to detect the presence of a crystalline material of interest in a sample. In order to determine the relevant diffraction angles in an energy-dispersive context it is useful to re-write Bragg's equation above in terms of energy:

Ed sine = ½hc = 6.19931 keV A where E is X-ray energy, h is Planck's constant and c is the speed of light; 6.19931 keV A is equal to ½ 7c when E is expressed in units of keV and d is expressed in units of A. Considering two characteristic energies emitted by the X-ray source and two lattice spacings:

Eidi sine, = 6.19931 keV A

E ₂ d ₂ sin¾ = 6.19931 keV A

£iCf ₂ sine ₃ = 6.19931 keV A

E ₂ d-, sin0 ₄ = 6.19931 keV A where Ei and E ₂ are the first and second characteristic energies emitted by the X- Ray source, di and d ₂ are the first and second lattice spacings. Thus, there are four diffraction angles though in each case the inequality Ed≥ 6.19931 keV A must be satisfied otherwise the corresponding X-ray energy/lattice spacing combination will not diffract because the X-ray wavelength is too long for the lattice spacing. If Qi ~ Q ₂ then the first two equations above may be rewritten:

E ₁ d ₁ sin(0 - <5) = E ₂ d ₂ sin(0 + <5) = 6.19931 keV A where 2Θ = θι+θ ₂ and δ is small. When correspondence between diffraction angles occurs, an enhanced signal is detected due to the superposition of the diffracted X- Ray beams. The size of δ represents the degree to which the two diffraction angles are not exactly coincident. The above equation can be rearranged as follows:

E ₂ /E ₁ = {d-,ld ₂ ) sin(0 - <5)/sin(0 + <5) When this equation is solved, two crystal planes diffract through substantially the same diffraction angle at two different incident X-Ray energies. Consequently when an X-Ray source emits a first characteristic X-Ray line having a first emission energy and a second characteristic X-Ray line having a second emission energy, the ratio of the first lattice spacing to the second lattice spacing is substantially equal to the inverse ratio of the first emission energy to the second emission energy. Therefore characteristic X-Ray lines may be chosen to enhance the signal detected in an X- Ray spectrum when looking for a crystalline material of interest.

The method described could be used to detect a class of minerals, such as micas or other subsets of clay minerals, rather than an individual crystalline phase, by selecting lattice spacings which are characteristic of that class. The method could also be used to detect two distinct crystalline materials of interest simultaneously by selecting at least one lattice spacing from each material. Therefore the crystalline material of interest may be a single crystalline phase or a combination of at least two crystalline phases.

In the following examples and embodiments characteristic X-Ray lines are chosen such that the diffraction angles for various lattice spacings correspond to one another. For small values of δ in the above equation, the diffraction angles of the lattice spacings in question will tend towards becoming coincident. The ideal case is where δ is zero. In the following examples the value of δ has been minimised to produce good overlap between peaks in the resultant spectra and the diffraction angles. Figure 1 is a schematic perspective view of a device in accordance with a first embodiment of the present invention adapted to detect the presence of a crystalline material of interest in a sample. A sample stage 1 is provided with a sample holder 2 in which a sample 3 of a material of interest is placed. A device 4 in accordance with a first embodiment of the present invention is positioned a distance L away from the sample stage 1 . A back-reflection geometry is utilised. The device 4 comprises an X-ray source 5 arranged to emit a first characteristic X-Ray line 6 and a second characteristic X-Ray line 7, both of which are incident on the sample 3 as indicated by arrow I. An X-ray detector 8 adapted to detect the X-Rays emitted by the sample 3 is positioned out of the line of the incident X-Rays, and receives diffracted X-Rays as indicated by lines D. Preferably the X-ray detector 8 is an energy dispersive X- Ray detector, such as a silicon drift detector, but may be an imaging device such as a charge coupled device (CCD). A collimator 9 is positioned in front of the X-Ray source 5, and arranged to collimate the first characteristic X-Ray line and the second characteristic X-Ray line on exiting the X-Ray source 5, and a filter 10 is positioned between the sample 3 and the X-Ray detector 8. The filter 10 is optional, and may be placed in a position 10a in the incident X-Ray beam I or in a position 10b in the diffracted X-Ray beam D adjacent the X-Ray detector 8, or in between as desired. The filter 10 may therefore filter the characteristic X-Ray lines emitted by the X-Ray source 5 or diffracted by the sample 3 before entering the X-Ray detector 8. The characteristic lines are part of a spectrum emitted by the X-Ray source 5.

Figure 2 is a schematic cross-section view of a device in accordance with a second embodiment of the present invention adapted to detect the presence of a crystalline material of interest in a sample. A housing 1 1 is provided to house an X-Ray source 12 arranged to emit a first characteristic X-Ray line and a second characteristic X- Ray line, and an X-Ray detector 13 adapted to detect X-Rays diffracted by a sample (not shown). The housing 11 comprises a main body 14 housing the X-ray source 12 and the X-Ray detector 13, and a handle portion 15 adapted to be gripped easily by a user. A screen 16 may be provided in the housing 1 1 to display information to a user, with a touchscreen being most preferable, such that a user may input commands to a processor controlling the X-Ray source 12 and the X-Ray detector 13. It may be desirable to include a collimator 17 positioned in front of the X-Ray source 12, and a filter 18, where the filter may be in the form of a fixed window 19 or an interchangeable filter arrangement.

The use of a device in accordance with the first embodiment of the present invention will now be described. It is particularly advantageous to be able to determine the presence of quartz in various mining and related applications. For example, quartz can present a hazard to miners when in the form of respirable silica, since this can cause silicosis. Quartz may also be present as an impurity in iron ore, where its hardness inhibits processing and thus increases the required energy, the detection of quartz in this situation could lead to an optimised processing method.

The following example illustrates how the correspondence of diffraction angles can be used to detect a crystalline material of interest, such as quartz. Figure 3 is a chart illustrating the emission energies of the L-series of characteristic X-Ray lines of an X-Ray source having a Pd (Palladium) anode. The lines are essentially monochromatic although they sit on a continuum with an approximate comparable intensity when integrated across the entire energy range of the X-Ray source. To ensure a good signal intensity it is desirable to use the most intense X-Ray lines if possible, which in this situation are the La _! and lines. The X-Ray source is therefore a single element capable of emitting at least two characteristic X-Ray lines. The ratio of the energies of these two characteristic X-Ray lines is 2.990 keV : 2.838 keV, or 1.0536. The ratio of the d-spacings for quartz for the (1 11) and (200) reflections is 2.2366 A : 2.1277 A, or 1.0512. These ratios lead to a value of Θ = 77.3° and δ = 0.6°. Using the value of Θ, the diffraction angle 2Θ of interest is 154.6°.

Figure 4 is a chart illustrating the emission energies of the L-series of characteristic X-Ray lines of an X-Ray source having a Pd (Palladium) anode, and a simulated EDXRD spectrum for quartz at 2Θ = 154.6° and the simulated absorption spectrum of rhodium foil. Figure 4 was simulated using the ray tracing program, PoDFluX (see Hansford, Rev. Sci. Instrum., 80 (2009), 073903), and shows an energy-dispersed X- Ray diffraction (XRD) spectrum for quartz at 2Θ with the L-series emission lines of Pd overlaid thereon. The EDXRD spectrum of quartz is incident on the detector. Only two of the diffraction peaks overlap the characteristic energies of two of the Pd X- Ray lines. This indicates that they can be used for the specific enhancement in the quartz spectrum. This particular example of where there is a correspondence between the angles of diffraction involves the strongest two Pd characteristic X-Ray lines. In addition, Figure 4 also illustrates the transmission of a 1 μηι thickness rhodium (Rh) foil on a 7 μηι thickness polyester support. Rhodium has an absorption edge at 3.013 keV, just above the Pd L-βι characteristic X-Ray line at 2.990 keV. Incorporating the rhodium foil as a filter into the X-Ray source - sample beam or the sample - X-Ray detector beam, as illustrated in Figure 1 above, will suppress any spectral features immediately above the quartz - Pd correspondence lines and thus significantly enhance the signal-to-background noise ratio and consequently sensitivity of the measurement.

Figure 5 is a chart illustrating the experimental confirmation of the correspondence between the X-Ray diffraction angles in quartz with a Pd X-Ray source. The experimental set up forming the analysis device was as shown in Figure 1 , with a distance between the X-Ray source and the sample of 400 mm, a distance between the collimator (a 2 mm diameter aperture in a sheet of aluminium (Al)) and the sample of 200 mm and a distance between the sample and the detector of 70 mm. A filter was also used, to suppress unwanted spectral features, comprising a 1 μηι thick rhodium coating on a 7μηι thick polyester support. The sample used was a pressed-powder pellet of quartz. The X-Ray source was based on a copper (Cu)- anode with a thin plate of Pd mounted on the surface. The electron incidence angle was 75° relative to the anode surface, and the X-Ray take off angle 15°. The detector used was a charge-coupled device (CCD), a CCD-22 available from e2v technologies (UK) Ltd, 106 Waterhouse Lane, Chelmsford, Essex, M 1 2QU, operated in frame-transfer mode. The CCD has an imaging area of 600 ^χ 600 pixels with a 40μηι pixel width, and when configured in a back reflection geometry covers a 2Θ range of 148 - 168°. Two spectra were taken from different areas of the detector in the approximate ranges 2Θ = 152 - 157° and 20 = 158 - 163°, illustrating the enhanced signal detected when the correspondence technique is used. The quartz (1 1 1) and (200) diffraction peaks are enhanced by a factor of approximately 10. Fig. 5 also shows the effect of introducing a rhodium filter between the X-ray source and sample, and illustrates the suppression of intensity in the spectrum at energies above the (200) - Pd-L peak at 2.990 keV. The use of a high 2Θ diffraction angle also reduces the sensitivity to sample morphology considerably. This allows unprepared or minimally prepared samples to be analysed. In order to determine the effect of a minimally prepared sample on the quality of the results, a mudstone rock known to contain a significant proportion of quartz was placed in a sample mount and positioned in the analysis device. Figure 6 is a chart illustrating the X-Ray diffraction spectrum obtained in the ranges 20 = 152 - 157° and 20 = 158 - 163°, with the prominent X-Ray fluorescence peaks labelled. The X-Ray diffraction peaks resulting from the correspondence between the (1 1 1 ) and (200) quartz diffraction peaks and the Pd-Lcn and Pd-L^ are highlighted, again showing the success of the correspondence technique.

The results shown in Fig. 6 demonstrate the feasibility of the phase-specific XRD method for unprepared samples exhibiting significant surface morphology. To further illustrate this feature of the technique, data were acquired for the same sample at a reference position, rotated through 30° and moved 2 mm away from the source and detector. The same on- and off-coincidence regions were selected for the three datasets and the extracted spectra are shown in Fig. 7. The spectra have been offset on the vertical axis for greater clarity. There is very good correspondence between both the on- and off-coincidence spectra, demonstrating the tolerance of the method to changes in the sample position and orientation. It follows that the technique is tolerant to sample morphology on the same scale. These experimental changes of position and orientation represent movements of the sample of a size which conventional XRD methods would be unable to cope with. Another example of an embodiment of the present invention is the detection of austenite in carbon steels. A complicating factor for austenite is that the lattice spacings depend on the carbon content of this phase. According to R. Abbaschian, L. Abbaschian and R. E. Reed-Hill in Physical Metallurgy Principles (Wadsworth Publishing Co Inc, 4th edition, 2008), the size of the austenite cubic unit cell is given by the equation: a = 3.555 + 0.044x A where a is the unit cell dimension and x is the weight percentage content of carbon. The relationship between the unit cell dimension and the lattice spacing, d, of the set of lattice planes with Miller indices (hkl) is given by: d = a I V(h ² +k ² +l ² )

The ratio of the lattice spacings for any two given sets of lattice planes is not affected by variations in the unit cell size. For example, the lattice planes (200) and (220) have lattice spacings of a/2 and a/V8 respectively, and therefore a fixed ratio of V8/2 = V2 = 1.4142. Austenite, or y-iron (Fe) has a small crystallographic unit cell and high symmetry, and therefore a sparse diffraction pattern. This makes it difficult to find a single element X-Ray source for which two angles of diffraction can be found to correspond. However, in an extension of the above method it is possible to use characteristic X-Ray lines from two separate elements to generate a correspondence between two X-Ray diffraction angles. The ratio of the (200) and (220) lattice spacings, 1.4142, in this example, leads to using the indium (In) characteristic X- Ray line at 3.487keV and the titanium (Ti) Κβ line at 4.932 keV, giving a ratio of 1.4144. Whilst the variation of the austenite unit cell dimension does not affect the correspondence of diffraction angles using these two characteristic X-ray lines, it may affect the value of the diffraction angle. For example, if the carbon content of the austenite in a steel sample is 0.82 wt%. The d(200) and d(220) lattice spacings are then 1.7955 and 1.2696 A respectively, and solving the equation for the coincidence criterion gives 2Θ = 163.86° and δ = 0.05°. This correspondence is very accurate and the high diffraction angle favours an insensitivity to sample morphology and to precise positioning of the sample. To investigate the range of variation of 2Θ, the minimum and maximum carbon content of austenite in carbon steel may be taken as 0.05 and 2.0 wt%:

0.05 wt% carbon: d(200) = 1.7786 A 2Θ = 176.40°

d(220) = 1.2577 A δ = 0.24°

2.0 wt% carbon: d(200) = 1.8215 A 20 = 154.83

d(220) = 1.2880 A δ = 0.03° A desirable possibility is to use an imaging detector that covers the full angular range, and that could therefore selectively detect and quantify austenite in steel over the full range of carbon content. The selected lattice spacings are unlikely to cause any overlap or confusion with other possible phases that may be present in a steel sample, such as ferrite (cr-Fe), martensite (ferrite supersaturated with carbon) and cementite (Fe ₃ C).

A combination of at least two elements, each of which is capable of emitting at least one characteristic X-Ray line, such as an alloy of indium (In) and titanium (Ti), say a 50/50 alloy, may be used to create the simultaneous emission of the two characteristic X-Ray lines of interest. Figure 8 is a chart showing the simulation of the EDXRD spectrum of a sample of steel containing (by volume) 5% austenite, 60% martensite, 20% ferrite and 15-% cementite. The strong peaks at approximately 3.0keV and 4.3keV are due to martensite diffraction, with the other two peaks being as a result of the correspondence technique. Identifying and quantifying the amount of retained austenite is a critical parameter in the manufacture of various products, where its presence may be seen as beneficial or harmful. For example, the amount of retained austenite indicates the success or failure of any heat treatment used to tailor mechanical properties of tools and dies, whereas in the preparation of bearings and gears the presence of austenite produces a mechanism by which the lifetime of such components may be extended by suppression of crack propagation.

Experimental results demonstrating the enhanced detection of austenite in steel are shown in Fig. 9. The sample is a high-Mn Twining-lnduced Plasticity (TWIP) steel with an estimated 40% austenite content with the balance made up with martensite and/or ferrite. In order to generate In and Ti characteristic X-rays simultaneously, a Ti plate was coated with a thin layer of In using electro-deposition methods and mounted on the anode of the X-ray tube source. Experiments were performed using the same laboratory set-up as for the detection of quartz using Pd. In this case, a 10 μηι thick Ti foil was mounted between the sample and the detector instead of the rhodium foil mounted between the source and sample. Ti has an absorption edge at 4965 eV and the foil serves to suppress diffraction and fluorescence peaks at higher energies and also at lower energies, below approximately 2.5 keV for example. The spectra shown in Fig. 9 are for on- and off-coincidence regions. The main fluorescence and scattering peaks have been labelled, including the two coincidence diffraction peaks which are the strongest peaks in the on-coincidence spectrum. The (200) peak is weaker than the (220) peak in the off-coincidence spectrum, illustrating how the two enhanced peaks can be equalised in intensity by optimising the thickness of the In layer on the InTi source. The Ti-Κα peak is due to fluorescence from the Ti foil in front of the CCD, there is no Ti in the sample. There will also be a small contribution to the Ti-Κβ peak in both spectra arising in the same way.

In the above examples, correspondence between the diffraction angles of two lattice planes at two particular characteristic X-Ray energies occurs. In each example utilising an embodiment of the present invention, the correspondence leads to an enhanced signal in the energy-dispersed X-Ray diffraction pattern obtained. In addition to the examples above, where quartz and austenite are detected, many other applications are possible, such as the identification of shale in coal deposits by the detection of quartz within the shale, the detection of polymorphs and hydration states of crystalline materials in pharmaceutical manufacturing, quality control of alloys during manufacture where crystalline impurities and poor nucleation and growth of desired crystalline phases may be detected and the detection of deposited crystalline phases in the preparation of coatings. In the above examples, a CCD camera was used as a detector to capture the X-Ray spectra. However, it may be preferable to use an energy-dispersive detector, such as a silicon drift detector, or a Si(Li) detector in place of or in addition to the CCD. This is particularly advantageous in a handheld device. In the above examples a filter comprising a thin foil of a material that suppresses X-Ray fluorescence peaks may be placed between the X-Ray source and the sample or between the sample and the X-Ray detector. In some circumstances positioning in front of the detector may be more suitable to achieve X-Ray fluorescence peak suppression.

In a further alternative embodiment at least a third diffraction angle correspondence may be used in the detection of a crystalline material of interest. In this situation the X-Ray source is arranged to emit at least a third characteristic X-Ray line, and the crystalline material of interest has at least a third lattice spacing. The third characteristic X-Ray line is diffracted at a distinct diffraction angle determined by each lattice spacing. The resulting X-Ray spectrum will have three peaks representing the three lattice spacings and the correspondence between the diffraction angles.

These and other embodiments of the invention will be apparent from the appended claims.