Elsevier

Journal of Luminescence

Volume 129, Issue 9, September 2009, Pages 979-983
Journal of Luminescence

Narrow gated Raman and luminescence of explosives

https://doi.org/10.1016/j.jlumin.2009.04.008Get rights and content

Abstract

Narrow gated Raman spectroscopy is used to detect Raman signals of explosives, which are usually screened by their intrinsic or background luminescence. It was found that the Raman/luminescence ratio is improved by 2–10 times with gate width of 500 ps compared to the 10 ns gate. It enables in certain cases to combine the luminescence suppression by gating with higher identification ability of Raman signals achievable with green excitation.

Introduction

In the manufacture and transport of improvised explosive devices (IED), there is the possibility of trace amounts of the explosive being deposited on the corresponding outer surfaces, such as vehicle. This could be due to human handling of the IED with subsequent transfer to the surface, a contaminated area used for IED manufacture, or the IED shedding trace amounts as it is being loaded into the vehicle. There are contact trace techniques for sampling and detecting this material, but the operator is put in a potentially very hazardous situation. The use of standoff non-contact techniques would greatly reduce the risk associated with trace sampling, hopefully reduce inspection time, and improve detection. The request is for any technology that meets the standoff requirement and detects explosives amounts in the range of micro-grams (or less) to tens of micro-grams per cm2.

Raman spectroscopy, in which we can get specific shifts or signature for each molecule, is increasingly important technology for homeland defense applications. Lewis et al. [1], [2] have previously shown that bulk quantities of a wide range of explosive materials can be analyzed by Raman spectroscopy using either 785, 830 or 1064 nm excitation, while 830 nm was found to be the preferred excitation wavelength. Carter et al. [3] remotely detected certain explosives using ns gated Raman spectroscopy with 532 nm excitation. The main problem in using Raman application is its low signals relative to luminescence of a substrate or the sample itself that in many cases may screen desired signal. Using continuous wave (CW) lasers, Raman scattering is collected together with luminescence. In fact, the interaction time for Raman scattering is virtually instantaneous (less than 1 ps), whereas luminescence emission is statistically relatively slow, with a minimum time of hundreds of picoseconds elapsing between electronic excitation and radiative decay. Thus, if we illuminate a sample with a short laser pulse, all of the Raman photons will be generated during the pulse, whereas most of the luminescence photons will be emitted at much longer times after the pulse. If the detection system is gated so as to detect only those photons scattered or emitted during the laser pulse, we will collect all of the Raman photons but reject the majority of the luminescence. Ideally, such a system should achieve the highest possible rejection ratio while having high throughput and a time resolution or gating time, short enough to match the duration of the laser pulse and correspondingly the Raman flux.

Existing gated ICCD cameras and photomultipliers typically operate on nanosecond timescales, with the fastest devices reaching hundreds or tens of picoseconds. In many cases it enables to produce excellent Raman signals, which were completely obscured in non-gated spectra.

In previous work [4] we demonstrated that the detection of military grade RDX, C4 and especially Semtex, using gated Raman spectroscopy with 532 nm excitation wavelength and gate width of 10 ns, is impossible. The reason is the strong intrinsic and background luminescence that screens Raman signals of these explosives. It was also shown that UV gated Raman (λex=248 nm, gate width of 10 ns) is substantially better [5], but the spectra became less informative compared to visible green excitation. In this work we used gated Raman with the gate of 0.5 ns in order to suppress luminescence by narrower gating comparing to the previous experiments.

Section snippets

Experimental

All the measurements were done using the setup described earlier [4], which includes Leopard high-energy Nd-YAG laser (532, 355 and 266 nm, pulse width 50 ps, energy 15, 8 and 4 mJ, correspondingly), ICCD camera (Princeton Instruments) with gate width down to 0.5 ns and edge filters for 532, 355 and 266 nm. We investigated the fast gated Raman on C4, PETN, RDX and Semtex on different backgrounds.

Theoretical approach

The luminescence intensity as a function of time with decay times very close to excitation pulse duration was calculated using the rate equations in two levels system under pulsed excitation F(t)=p0exp(−t/τex), with τex=50 psdN2dt=Bp0exp(-t/τex)N1-N2Awhere N1, N2 are the populations in the lower level and upper level (respectively), B is the Einstein coefficient for absorption, A=1/τlum is the spontaneous emission probability and p0 is the excitation density. The border conditions are: N=N1+N2

Results and discussion

Fig. 2 demonstrates Raman spectra of C4 and Semtex with 532 nm excitation wavelength and gate widths of 0.5 and 5 ns. C4 spectra are very pure from background, demonstrating strong Raman spectra of RDX, which is the main component of C4. We did not found significant intrinsic luminescence of C4 and correspondingly dependence of C4 Raman spectra on gate width (Fig. 2a). This result strongly contradicts to excitation by 532 nm with 6–8 ns pulses, where very strong intrinsic C4 luminescence has been

Conclusions

The main problem for Raman application is its low signals relatively to Rayleigh scattering and luminescence of a substrate or the sample itself that in many cases may blur desired signal. Military grade RDX, C4 and especially Semtex have very strong luminescence that do not allow detect these explosives, even using gated deep UV excited Raman scattering with nanoseconds pulse durations. We found that ultra fast gated Raman spectroscopy may resolve this problem. Using excitation source with 50 

References (5)

  • I.R. Lewis et al.

    Spectrochim. Acta

    (1995)
  • M.L. Lewis et al.

    Vib. Spectrosc.

    (2005)
There are more references available in the full text version of this article.

Cited by (33)

  • Standoff Raman spectrometry for the non-invasive detection of explosives precursors in highly fluorescing packaging

    2013, Talanta
    Citation Excerpt :

    Standoff direct Raman spectroscopy has been demonstrated for the detection of surface deposits from different distances. Several literature that describe standoff direct Raman spectroscopy and its various applications have been published recently [2–18]. In this technique, a laser excitation beam is focused onto the exposed particles of the suspected substance.

  • Current trends in explosive detection techniques

    2012, Talanta
    Citation Excerpt :

    When the detection system has a narrow gate, all Raman photons will be collected with the minimum amount of luminescence. This allows a greater signal to noise ratio and enables detection of explosive substances with strong luminescent properties, such as Semtex [47]. The detection of explosive particles with backward coherent anti-Stokes Raman scattering (B-CARS) was studied by Portnov et al.

  • Preparation of zinc oxide-montmorillonite hybrids

    2011, Materials Letters
    Citation Excerpt :

    In addition, the Raman band at around 576 cm−1 assigned to the E1L mode was caused by the impurities and defects in ZnO such as oxygen vacancies and zinc interstitials [20]. On the other hand, the Raman spectra ZnO–montmorillonite and ZnO–HDTMA–montmorillonite (not shown) did not show any Raman lines in the wavenumber of 200–1200 cm−1; we assume that the products exhibited the extremely strong luminescence, which covered the weak Raman signals [22]. In FT-IR spectrum of ZnO–HDTMA–montmorillonite, the characteristic bands due to asymmetric and symmetric –CH2 stretching vibrations modes of amine [23] were slightly shifted to higher frequency regions at 2923 and 2851 cm−1, respectively, when compared to those observed for HDTMA (at around 2918 and 2850 cm−1), indicating the formation of HDTMA in montmorillonite.

View all citing articles on Scopus
View full text