The present invention is concerned with new and improved substrates and methods for Raman spectroscopy, and especially substrates capable of enabling detection of materials that are generally difficult to detect/identify through use of surface enhanced Raman spectroscopy (SERS). The substrates and methods are especially suitable for enabling detection/identification of explosives, such as 1, 3, 5-trinitroperhydro-l, 3, 5-triazine (RDX) and pentaerythritol tetranitrate (PETN), with Raman spectroscopy.

A known disadvantage of Raman spectroscopy is that the Raman effect is generally inherently very weak, and thus detection and identification of materials using Raman spectroscopy can suffer from low sensitivity, and generation of weak Raman signals. One way to overcome this may be to use highly optimised methodologies or instrumentation. Often, surface enhanced Raman spectroscopy (SERS) is used to overcome the problem of the inherently weak Raman signals, to provide much higher sensitivity. However, some materials, such as the explosives RDX and PETN, are generally difficult to detect/identify through use of SERS, and an alternative approach to detecting these types of materials with high sensitivity is required.

The ability to identify non-visible quantities of explosives in the field using a high confidence and reproducible system has long been desirable. The ideal system would be fast, easy to use, require very little or no sample preparation and be hand-portable without adding to the already high burden of first responders and military personnel. Raman spectroscopy is one method that has been used for detecting trace explosives, such as TNT (2,4,6-trinitrotoluene), RDX and PETN, three commonly used explosives, however although TNT can generally be fairly easily detected using SERS, materials such as RDX and PETN are not strongly SERS active and consequently are generally difficult to detect through use of SERS. A number of solutions have been proposed to enhance detection of materials using Raman spectroscopy, including to enhance SERS detection, especially though modification of the substrates used for Raman spectroscopy. One solution has been to generate hydrophobic Raman substrate surfaces, in order to concentrate samples of materials on the substrates prior to analysis. For example, a Teflon coated steel substrate (sold as the m-Rim™ slide or SpectRim™ slide) is reportedly able to detect RDX and PETN at a concentration of 1 mg/ml in the solvent acetonitrile. However, these are high concentrations of material, and further it has been shown that the Teflon coated steel substrate is restricted in the solvents that can be used for applying samples to the substrate. Some solvents, such as acetone, do not form a droplet on the substance surface but instead wet the surface, and consequently do not produce the desired increased concentration of material on the surface, and thus not the required improvement in sensitivity. It remains difficult, for example, to identify/detect certain explosives, such as RDX and PETN, with sufficient sensitivity, such as at least 10 pg/ml, which is desirable.

A solution to the problem of detecting materials generally difficult to detect/identify through use of surface enhanced Raman spectroscopy (SERS), such as, but not limited to, the explosives RDX and PETN, with improved/sufficient sensitivity, is still required.

In a first aspect, the present invention provides for a Raman substrate having a surface comprising a fluorocarbon polymer, wherein the fluorocarbon polymer comprises, or is assembled from, perfluoroalkyl acrylate monomer units. The fluorocarbon polymer generally provides for a Raman substrate having a super hydrophobic surface.

The surface comprising the fluorocarbon polymer may be generated through plasma polymerisation deposition of a perfluoroalkyl acrylate monomer, by itself, or in combination with a linking chemical moiety, or through deposition with a different second monomer or fluorocarbon (or perfluoroalkyl acrylate) monomer, and in particular may be through pulsed plasma deposition. The Applicant has surprisingly found that a Raman substrate having a surface coated with a fluorocarbon polymer comprising, or assembled from, perfluoroalkyl acrylate monomer units, especially generated by plasma polymerization (or pulsed plasma polymerisation) is capable of improving the sensitivity of detection of certain materials by Raman spectroscopy. The Applicant has especially found that plasma polymerisation (or pulsed plasma polymerisation) of monomers containing polymerisable unsaturated acrylate groups and perfluoroalkyl chains is capable of improving the sensitivity of detection of certain materials by Raman spectroscopy. An improvement of sensitivity up to about 100 to 1000 fold over Raman substrates in the art, such as the Teflon coated steel substrate, is achievable. In particular, such Raman substrates can be used for detecting the explosives RDX and PETN with improved sensitivity.

Fluorocarbon polymer surfaces, or coatings, generated by plasma polymerization (or pulsed plasma polymerisation) of monomers (precursor molecules) containing at least one polymerisable unsaturated acrylate group and at least one perfluoroalkyl chain are especially applicable to the first aspect of the invention due to their exceptional properties. The exceptional properties include speed of polymerization, generally mild processing conditions and exceptional hydrophobic, as well as oleophobic, properties. The exceptional properties of these monomers, and subsequent polymers, are believed to be attributed to the peculiar chemical architecture of the perfluoroalkyl polymer chains. Examples of such monomers include lH,lH,2FI,2H-perfluorodecyl acrylate (PFDA, also known as PFAC-8), and lH,lH,2H,2H-perfluorooctylacrylate (PFOA, also known as PFAC-6), which, by themselves, are capable of generating poly(lH,lH,2H,2H-perfluorodecyl acrylate) or poly(lH,lH,2H,2H-perfluorooctylacrylate), respectively. Surfaces with regularly aligned and closely packed CF groups, as would be generated with such monomers, have been reported to exhibit a surface energy of 6.7 mN/m, which is well below the 18-20 mN/m value for polytetrafluoroethylene (PTFE), otherwise known as Teflon. The fluorocarbon polymer surfaces could alternatively be generated from PFAC-10 (1H, 1H, 2H, 2H-perfluorododecyl acrylate), or indeed the corresponding methacrylates of PFAC-6, PFAC-8 or PFAC-10. The fluorocarbon polymer of the first aspect may be generated via plasma polymerisation of a monomer having the structure: CH ₂ =CR ¹ C(0)0(CH ₂ )x(CF ₂ ) _v CF ₃ wherein Ri is hydrogen, alkyl, such as methyl, or haloalkyl, but preferably hydrogen, x is an integer of at least 2, but most likely 2, and y is an integer of from 1 to 9, for instance between 4 to 9, and preferably 7, with polymerisation via the acrylate functional group. The fluorocarbon polymer could thus comprise the unit, or repeating unit, -[CH ₂ CRi(C(0)0(CFl2)x(CF2)yCF3)]-, providing repeating side chains of -C(0)0(CH ₂ ) _x (CF2)yCF ₃ , which unit, or repeating unit, could be separated by spacer molecules, or alternatively the fluorocarbon polymer could be a co-polymer, where at least one unit comprises the formula - [CFl2CRi(C(0)0(CH2) _x (CF2) _y CF3)]-, or both units could comprise that formula, but differing in at least one of Ri, x and y. The Raman substrate of the first aspect provides a means for undertaking Raman spectroscopy of materials that are not strongly SERS active, such as the explosives RDX and PETN.

The fluorocarbon polymer may be a polymer comprising, or assembled/generated from, the monomer 1H, 1H, 2H, 2FI-perfluorodecyl acrylate (PFAC-8) or the monomer 1H, 1H, 2H, 2H- perfluorooctyl acrylate (PFAC-6), either by itself or in combination with a linking chemical moiety, for example divinyl adipate, or a different second monomer or fluorocarbon monomer, which may be generated at/on the surface of any suitable support material using plasma polymerisation, to generate the Raman substrate.

The Applicant has especially found that fluorocarbon polymers generated from PFAC-8 or PFAC-6 by plasma polymerisation, are capable of concentrating, supporting and enabling crystallisation of particular agents at its surface, using numerous solvents. The extent of concentration and quality of crystallisation is proportional to the possible improved sensitivity achieved, and the quality of the Raman spectra produced.

A fluorocarbon polymer generated from PFAC-6 or PFAC-8 not only prevents spreading of the materials to be analysed over the substrate surface, but thereby also enables or aids crystallisation, thus focussing and concentrating the materials in addition to the crystallisation. Substrates coated with a polymer comprising, or assembled from, PFAC-6 or PFAC-8, or coated with polymerised PFAC-6 or PFAC-8 are capable of producing a Raman signal through the Raman effect as a result of the concentration and crystallisation of material at the surface.

Crystallisation can be further optimised by careful selection of the solvents from which to crystallise the materials.

The support material of the Raman substrate may be selected from numerous possibilities. The Applicant has in particular found that the substrate does not need to be or include a metal, such as gold, a feature essential for SERS, but could be any material. For example, the support material could be plastic, fabric, glass, metal, or combinations thereof. The support material may be nickel. The surface of the substrate may be textured, as texturing of the surface has been shown to lower the surface energy at the surface.

The solvent for crystallisation of explosives could be any suitable organic solvent, though the solvent should be one that enables good crystallisation on the Raman substrate surface, such as acetone, methanol, or ethanol. Acetone is particularly advantageous for detecting explosives. In a second aspect, the present invention provides a method for detecting the presence of a material in a sample using Raman spectroscopy comprising i. crystallising the sample on a Raman substrate according to the first aspect of the invention with an appropriate solvent; ii. undertaking Raman spectroscopy on the crystallised sample to generate a Raman spectrum, and iii. interrogating the Raman spectrum to determine whether the material is present in the sample.

In one embodiment the material is an explosive, for example TNT, RDX or PETN. For TNT the appropriate solvent may be any organic solvent, though in a preferred embodiment would be ethanol, methanol, acetone or isopropanol. Suitable solvents for RDX and PETN also include ethanol, acetone, methanol and isopropanol.

In a third aspect, the present invention provides a method of manufacturing a Raman substrate comprising undertaking plasma polymerisation deposition of a perfluoroalkyl acrylate monomer in the presence of a support material suitable for Raman spectroscopy.

The plasma polymerisation deposition may comprise deposition of the perfluoroalkyl acrylate monomer by itself, or in combination with a linking chemical moiety, for example divinyl adipate, or in combination with a different second monomer or fluorocarbon monomer, which may be a different perfluoroalkyl acrylate monomer. The support material may be plastic, fabric, glass, metal, or combinations thereof. In one embodiment the support material is nickel. The support material may be smooth or roughened, or textured.

The plasma polymerisation deposition may be pulsed plasma deposition.

The perfluoroalkyl acrylate monomer may be any of the monomers discussed for the first aspect of the present invention.

The present invention shall now be described with reference to the following non-limiting examples and drawings in which

Figure 1 is four Raman spectra, each of a 10 mI_ droplet of RDX at the following concentrations in the solvent acetone; 1 mg/ml (A), 100 pg/ml (B), 10 pg/ml (C) and 1 pg/ml (D) on nickel (support material) coated with a fluoropolymer of PFAC-8 (coated by pulsed plasma polymerisation). The lower line shows a 5 s integration time, the middle line a 10 s integration time and the higher line a

20 s integration time. Spectra were collected using a 785 nm laser excitation; Figure 2 is four Raman spectra, each of a 10 pL droplet of PETN at the following concentrations in the solvent acetone; 1 mg/ml (A), 100 pg/ml (B), 10 pg/ml (C) and 1 pg/ml (D) on nickel (support material) coated with a fluoropolymer of PFAC-8 (coated by pulsed plasma polymerisation). The lower line shows a 5 s integration time, the middle line a 10 s integration time and the higher line a 20 s integration time. Spectra were collected using a 785 nm laser excitation;

Figure 3 is four Raman spectra, each of a 10 pL droplet of TNT at the following concentrations in the solvent acetone; 1 mg/ml (A), 100 pg/ml (B), 10 pg/ml (C) and 1 pg/ml (D) on nickel (support material) coated with a fluoropolymer of PFAC-8 (coated by pulsed plasma polymerisation). The lower line shows a 5 s integration time, the middle line a 10 s integration time and the higher line a 20 s integration time. Spectra were collected using a 785 nm laser excitation;

Figure 4 is two overlaid Raman spectra, a Raman spectrum for bulk PETN from a library, and a Raman spectrum for PETN (1 mg/ml; 5 pi) in methanol crystallised on a Raman substrate having a smooth nickel support material coated with plasma polymerised PFAC-8;

Figure 5 is two overlaid Raman spectra, a Raman spectrum for bulk PETN from a library, and a Raman spectrum for PETN (1 mg/ml; 5 pi) in methanol crystallised on a Raman substrate having a textured nickel support material coated with plasma polymerised PFAC-8;

Figure 6 is two overlaid Raman spectra, a Raman spectrum for bulk PETN from a library, and a Raman spectrum for PETN (1 mg/ml; 5 pi) in methanol crystallised on a p-RIM substrate;

Figure 7 is two overlaid Raman spectra, a Raman spectrum for bulk RDX from a library, and a Raman spectrum for RDX (1 mg/ml; 5 pi) in ethanol crystallised on a Raman substrate having a smooth nickel support material coated with plasma polymerised PFAC-8;

Figure 8 is two overlaid Raman spectra, a Raman spectrum for bulk RDX from a library, and a Raman spectrum for RDX (1 mg/ml; 5 pi) in ethanol crystallised on a Raman substrate having a textured nickel support material coated with plasma polymerised PFAC-8; Figure 9 is two overlaid Raman spectra, a Raman spectrum for bulk RDX from a library, and a Raman spectrum for RDX (1 mg/ml; 5 mI) in ethanol crystallised on a m-RIM substrate;

Figure 10 is two overlaid Raman spectra, a Raman spectrum for bulk TNT from a library, and a Raman spectrum for TNT (1 mg/ml; 5 mI) in methanol crystallised on a Raman substrate having a textured nickel support material coated with plasma polymerised PFAC-8;

Figure 11 is two overlaid Raman spectra, a Raman spectrum for bulk TNT from a library, and a Raman spectrum for TNT (1 mg/ml; 5 mI) in methanol crystallised on a m-RIM substrate;

Figure 12 is a Raman spectrum for TNT (lmg/ml; 5 m I) in acetone crystallised on a m-RIM substrate;

Figure 13 is a Raman spectrum for TNT (lmg/ml; 5 mI) in acetone crystallised on a Raman substrate having a smooth nickel support material coated with plasma polymerised PFAC-8;

Figure 14 is a Raman spectrum for bulk TNT from a library;

Figure 15 is a Raman spectrum for RDX (lmg/ml; 5 mI) in ethanol crystallised on a m-RIM substrate;

Figure 16 is a Raman spectrum for RDX (lmg/ml; 5 mI) in ethanol crystallised on a Raman substrate having a smooth nickel support material coated with plasma polymerised PFAC-8; Figure 17 is a Raman spectrum for bulk RDX from a library;

Figure 18 is a Raman spectrum for PETN (lmg/ml; 5 mI) in methanol crystallised on a m-RIM substrate;

Figure 19 is a Raman spectrum for PETN (lmg/ml; 5 mI) in methanol crystallised on a Raman substrate having a smooth nickel support material coated with plasma polymerised PFAC-8; Figure 20 is a Raman spectrum for bulk PETN from a library; Figure 21 is Raman spectra for RDX (100 mg/ml) in acetone, ethanol, and methanol crystallised on a Raman substrate coated with plasma polymerised PFAC-6 [in the presence of a fixed volume percentage of divinyl adipate (DVA)]; and

Figure 22 is a Raman spectrum for RDX (lO pg/ml) in methanol crystallised on a Raman substrate coated with plasma polymerised PFAC-6 [in the presence of a fixed volume percentage of divinyl adipate (DVA)].

Examples

The explosives PETN and RDX are particularly difficult to detect using surface enhanced Raman spectroscopy (SERS) - they are not strongly SERS active. These materials are usually applied to a Raman substrate in an organic solvent, and then dried. Research was thus undertaken to investigate minimising the spread of the organic solvent on the substrate, which could then concentrate the explosive material on the surface. A number of materials were investigated, including the commercial available m-RIM slide, which has a stainless steel support material coated with Teflon. The Applicant also investigated potential novel Raman substrates, including several support materials (e.g. plastic, glass, metal) coated with fluorocarbon polymers, and especially a fluorocarbon polymer generated through plasma polymerisation deposition with the monomer PFAC-8.

Example Manufacture of novel Raman substrate 1

The base (support material) of a new Raman substrate was prepared from a nickel sheet, which in this case was electroformed for 55 amp hours (AH) in a nickel sulfamate tank, from a silver-coated 8 x 10" glass plate.

The nickel sheet was cut, in this case using a Coherent Talisker laser in 355 nm UV mode, max power input 4W at 200 kHz with a pulse duration of 15 ps. For a plain nickel substrate, the substrate was machined at full power on continuous loops until it had fully cut through the thickness of the substrate. For the roughened substrates, the surface was hatched with the laser at 25% power at four different angles, and then cut on full power for continuous loops until it had fully cut through the substrate.

Once the base of the substrate had been prepared, it was plasma treated as described below.

Plasma treatment was carried out in an inductively coupled glass cylindrical glow discharge reactor (10 cm diameter, 4.3x10 ³ m ³ volume, base pressure typically better than 1x10 ² mbar). The reactor was connected to a two stage Edwards rotary pump via a liquid nitrogen cold trap with a thermocouple pressure gauge inline. In this example, a monomer tube containing lH,lH,2H,2H-perfluorodecyl acrylate (PFAC-8, Fluorochem, UK) was purified by freeze-thaw cycles prior to use and attached to the air inlet side of the reactor. All connections were grease free. An L-C matching unit was used to minimise the standing wave ratio (SWR) of the transmitted power between the 13.56 MHz RF generator and the electrical discharge.

A base substrate (support material; Nickel sheet) was placed in the centre of the reactor. The chamber was then evacuated to the base pressure of the apparatus, typically 1 x 10 ² mbar, and the chamber was heated to 32°C. Once base pressure had been reached, the PFAC8 vapour was introduced into the reactor. The reactor was purged with the vapour for five minutes, and then the RF generator was switched on to create a 40 W continuous wave plasma, for 30 s. The pulse generator was then turned on, in this case at a pulsing sequence of 40ps on, 20ms off. Once the plasma deposition had recovered, as indicated by an input power of 40 W and a stable pulse envelope (confirmed using an oscilloscope), the deposition was allowed to run for 20 min. At the end of the treatment the RF generator was switched off and the reactor purged for 2 minutes with monomer vapour, prior to being evacuated back to base pressure. Once base pressure was reached the vacuum chamber was isolated from the pump and the system was brought up to atmospheric pressure to allow the new Raman substrate to be removed.

Kriiss Drop Shape Analyser (PSA) The DSA is capable of measuring contact angles of liquids/solvents on surfaces. The surface tension, and surface free energy, is then calculated from the contact angles measured between the liquid/solvent droplets and the surface. This approach was used to measure the surface free energy of the Raman substrate for crystallisation. The Applicant compared the m-RIM slide to a potential substrate comprising a nickel support material coated with a fluoropolymer generated from PFAC-8 through plasma polymerisation. Three contact angle measurements were taken for each liquid/solvent used, and the average of these three measurements was used for the calculation of the surface free energy. The lower the surface free energy, the higher the repellency from the surface, and thereby the greater the concentration effect prior to crystallisation. Both Raman substrates were tested with four possible solvents, water, ethylene glycol, diiodo methane, and n-hexadecane, however n-hexadecane did not produce a droplet on the m-RIM slide but spread out over the surface (wetted the surface), which meant no contact angle could be obtained for that material. The contact angles for the m-RIM slide were 96.7 degrees, 72.3 degrees, and 41.3 degrees, for water, ethylene glycol and diiodo methane respectively, with a total surface free energy (Total Interfacial Tension - IFT), based on the contact angles for all three solvents on this surface, of 35.3 mN/m. The contact angles for the PFAC-8 derived fluoropolymer substrate were 113.8 degrees, 82.0 degrees, 108.2 degrees, and 104.6 degrees, for water, n-hexadecane, ethylene glycol and diiodo methane respectively, with a total surface free energy (Total Interfacial Tension - IFT), based on the contact angles for all four solvents on this surface, of 8.3 mN/m. From these results it is apparent that there is a huge difference in the surface free energies of the commercial m-RIM substrate, and the PFAC-8 derived substrate. These results also show that there is more scope to use a wider variety of solvents with the PFAC-8 derived substrate, and also that material in a particular solvent of choice will be concentrated into a smaller area due to the better repellency on the surface of this substrate.

Experiments with potential Raman Substrate comprising a Nickel Support Material coated with a Fluoropolymer generated from PFAC-8 through Plasma Polymerisation. Flaving regard to Figures 1, 2 and 3, RDX, PETN, and TNT were applied to the substrate in acetone at concentrations of 1 mg/ml, 100 pg/ml, 10 pg/ml and 1 pg/ml, and interrogated by Raman spectroscopy with laser excitation at 785 nm. The experiments are conclusive in that on this new substrate detection of all three of the materials is possible with high sensitivity (at least 1 mg/ml), in this case when each were crystallised from acetone. This result is at least 100 fold lower than was suggested previously for detection on the m-RIM slide. In addition, it is noted that acetone, the favoured solvent for crystallisation of all three of RDX, PETN and TNT, does not generate a droplet on the m-RIM slide, and samples applied to this surface in acetone instead wetted the surface. The new substrate is thus also advantageous because of the variety of solvents that can be applied to it, and especially acetone when interrogating samples that may contain explosives.

Comparison of detection on the new Raman substrate comprising the Nickel Support Material coated with a Fluoropolvmer generated from PFAC8 through Plasma Polymerisation, with the Teflon coated u-RIM substrate

Flaving regard to Figures 4, 5 and 6, the explosive material PETN (1 mg/ml; 5 mI) in methanol was crystallised on three different Raman substrates, and Raman spectra were generated for each of them, and compared to each other and to a library spectrum of the bulk explosive. The three substrates were a smooth nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 4), a roughened/textured nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 5), and a m-RIM substrate (Figure 6). At a concentration of 1 mg/ml of PETN, it was not possible to detect/identify PETN from the m-RIM substrate, but it was possible to detect/identify PETN from both of the new Raman substrates, and especially effective from the textured Raman substrate.

Having regard to Figures 7, 8, and 9, the explosive material RDX (1 mg/ml; 5 mI) in ethanol was crystallised on three different Raman substrates, and Raman spectra were generated for each of them, and compared to each other and to a library spectrum of the bulk explosive. The three substrates were a smooth nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 7), a roughened/textured nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 8), and a m-RIM substrate (Figure 9). At a concentration of 1 mg/ml of RDX, it was not possible to detect/identify RDX from the m-RIM substrate, but it was possible to detect/identify easily RDX from both of the new Raman substrates, and especially effectively from the smooth Raman substrate. Having regard to Figures 10 and 11, the explosive material TNT (1 mg/ml; 5 mI) in methanol was crystallised on two different Raman substrates, and Raman spectra were generated for each of them, and compared to each other and to a library spectrum of the bulk explosive. The two substrates were a roughened/textured nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 10), and a m-RIM substrate (Figure 11). At a concentration of 1 mg/ml of TNT, it was not possible to detect/identify TNT from the m-RIM substrate, but it was possible to detect/identify TNT from the new Raman substrate.

Having regard to Figures 12, 13, and 14, the explosive material TNT (1 mg/ml; 5 mI) in acetone was crystallised on two different Raman substrates, and Raman spectra were generated from a single point for each of them, and compared to each other and to a library spectrum of the bulk explosive (Figure 14). The two substrates were a smooth nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 13), and a m-RIM substrate (Figure 12). At a concentration of 1 mg/ml of TNT, it was not possible to detect/identify TNT from the m-RIM substrate, but it was possible to detect/identify easily TNT from the new Raman substrate.

Having regard to Figures 15, 16, and 17, the explosive material RDX (1 mg/ml; 5 mI) in ethanol was crystallised on two different Raman substrates, and Raman spectra were generated from a single point for each of them, and compared to each other and to a library spectrum of the bulk explosive (Figure 17). The two substrates were a smooth nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 16), and a m-RIM substrate (Figure 15). At a concentration of 1 mg/ml of

RDX, it was not possible to detect/identify RDX from the m-RIM substrate, but it was possible to detect/identify easily RDX from the new Raman substrate. Having regard to Figures 18, 19, and 20, the explosive material PETN (1 mg/ml; 5 mI) in methanol was crystallised on two different Raman substrates, and Raman spectra were generated from a single point for each of them, and compared to each other and to a library spectrum of the bulk explosive (Figure 20). The two substrates were a smooth nickel support material that had been coated with plasma polymerised PFAC-8 (Figure 19), and a m-RIM substrate (Figure 18). At a concentration of 1 mg/ml of PETN, it was not possible to detect/identify PETN from the m-RIM substrate, but it was possible to detect/identify easily PETN from the new Raman substrate.

The Applicant also found that the material used as the support material could be any suitable material, such as glass, plastic or metal, or even fabric, without having any significant effect on the ability of the Raman substrate (coated with a fluorocarbon polymer, such as poly-PFAC-8) to generate Raman spectra for materials such as TNT, RDX and PETN.

Example Manufacture of novel Raman substrate 2

A Raman substrate was prepared in a plasma polymerisation chamber using PFAC-6 in the presence of a fixed volume percentage of divinyl adipate (DVA), used as a linking chemical moiety, and using helium as a carrier gas. The deposition process consisted of a 1 min continuous wave step with 250W power and a pulsed step with 340W power and an RF pulse duty cycle of 0.010%.

Aluminium was used as support material for the test substrate and the contact angle of the coated substrate was determined using the Krijss drop shape analyser. Water, n-hexadecane and ethylene glycol were used and gave a total surface free energy (Total Interfacial Tension - IFT), based on the contact angles for the three solvents of 11.2 mN/m. Use of PFAC-6 alone (without linking chemical moiety), which was not tested as a Raman substrate, but which can be expected to perform at least as well as that with the linker, provided an IFT of 4.8 mN/m.

Having regard to Figure 21, a ten-fold dilution of a 1 mg/ml solution of RDX was prepared in acetone, ethanol and methanol, and Raman spectra generated using a 785 nm Raman spectrometer, thus a final concentration of 100 pg/ml RDX. RDX could be identified at this concentration, irrespective of the solvent used to apply to the Raman substrate.

Having regard to Figure 22, a one hundred-fold dilution of a 1 mg/ml solution of RDX was prepared in methanol, and Raman spectra generated using a 785 nm Raman spectrometer, thus a final concentration of 10 pg/ml RDX. This appears to be close to the limit for detection of RDX using this substrate, and applied to the substrate in methanol. The limit of detection is still however at 100 fold lower than previously used substrates.