In a previous edition of the IRPS Bulletin Malcolm Cooper wrote an article of current trends in airport security ("Fighting crime with x-rays", IRPS Bulletin, Volume 12, No.1). This piece follows up on one of the techniques that Malcolm highlighted, that of low angle x-ray diffraction. This, unlike most security-based techniques has enjoyed significant reporting in open scientific literature in recent years. The "news blackout" regarding some of the other techniques usually has more to do with commercial interest than any security implications and often many of the techniques employed in security imaging systems have been taken from previous widely published developments in the medical field, for example dual energy imaging and computerised tomography. Diffraction techniques are rather different however in that they are currently being developed for both medical and industrial techniques simultaneously in both academic and industrial quarters.

One of the main criticisms that the neutron community often have of x-ray inspection is that x-rays are not material specific. Well this may be true in terms of the total attenuation but is clearly untrue if we consider the interaction mechanisms responsible. The photoelectric interaction, for example, leads directly to x-ray fluorescence, an unarguably elemental specific phenomenon (although the very low fluorescent energies of organic compounds make the technique impracticable for the detection of concealed contraband). Another photon interaction is Rayleigh scattering, defined as the elastic and coherent scattering of x-rays from bound atomic electrons and is the process that leads directly to diffraction. The phenomenon of x-ray diffraction was first investigated at the beginning of the century by the likes of the German physicist von Laue and the two English physicists W.H.Bragg and his son W.L.Bragg; the latter who went on to derive the famous diffraction law which bears their name. It was not however until the mid 1980’s that this interaction was first proposed as a non-invasive probe for materials characterisation, most notably in the medical field looking for increased contrast between biological media. Over the last ten years the potential applications of this form of analysis have increased dramatically, with the detection of concealed contraband being one of several currently being exploited.

One of the greatest threats to aviation security comes from sheet explosive concealed in passenger luggage. Plastic explosives are organic and therefore low-Z and when formed into thin sheet (of the order 5 mm) are difficult to detect by conventional transmission methods. In recent years a number of groups have investigated the use of x-ray diffraction as a material specific technique for the detection of explosives. Many explosives are crystalline in structure and therefore produce characteristic diffraction effects when irradiated by photons. This may be measured either angular dispersively whereby mono-chromatic photon scattering is measured as a function of angle or by energy dispersive means where the scattering of photons from a poly-chromatic source i.e. conventional x-ray tube, is measured at a fixed angle. Both of these methods produce data which, whilst varying with angle and energy, is invariant in momentum transfer and therefore directly comparable. Figure 1 shows the x-ray diffraction from calcium carbonate as a function of both energy and scattering angle. Whilst it can be seen that varying either the energy or the scattering angle causes an appropriate shift in the other parameter the fundamental diffraction data is unchanged. Either of the approaches produce a unique diffraction profile or "signature" of the irradiated material and figure 2 shows a typical energy dispersive diffraction profile of a notorious explosive often the stock weapon in the terrorist armoury, namely Semtex.

Whilst the profile does not have the resolution of conventional diffraction measurements it is unique to Semtex and easily identifiable. The problem comes when measuring the materials in a realistic situation, that is in a suitcase where overlying materials complicate the diffraction profiles and severely attenuate the data. In this situation reliable data analysis is essential. To this end various forms of pattern recognition analysis are being investigated, for example neural networks and multivariate statistical analysis have shown very promising results. Another approach is to measure both angular and energy dispersive data using an array of energy dispersive detectors and to locate regions in diffraction space associated with known explosives. This simplifies the analysis by using relative scattering intensities as the threat indicator. Whichever form of data collection and analysis is employed it is essential that the final system is fully automated to produce a quick "threat / no threat" decision without the need for human involvement. It is also clearly important that as with any security system the number of false positive decisions is minimised without compromising public safety. Whilst these requirements are technically

challenging they are being successfully addressed and several prototype systems are currently being deployed or in the development process. It is anticipated that in the next few years x-ray diffraction inspection will be another weapon in the increasing armoury of the law enforcement officer.

Having attended several security technology conferences over the past few years there seems to have been a definite shift in attitude regarding the range of potential technologies. Some years ago people would attend these meetings to promote their own field and to shoot down their competitors. The x-ray fraternity would say to the neutron people "Well the nitrogen in explosives looks the same as the nitrogen in broccoli, how can you possibly differentiate?", the neutron people would say to the nuclear quadropole resonance community

"What if the drugs are wrapped in tin foil, what happens to your signal then?" and of course the diffraction lobby would be asked "What if the drugs are diluted or the explosives are liquid, what happens to your long range order then?" To which of course we had no reply. More recently however people have realised there is no "Holy Grail" as far as contraband detection goes, no single technology that will accommodate all potential situations. The community, and the bodies funding the community, have realised that the solution involves all of the current technologies working in tandem and that the challenge for the future is to deploy them in the most effective manner, tailoring the systems to their specific requirements.

Previous

Main

Next