Double-Pulse Laser Fragmentation/Laser-Induced Fluorescence Method for Remote Detection of Traces of Trinitrotoluene

Abstract

This paper presents the results of an experimental study of the dynamic characteristics of the process of laser fragmentation/laser-induced fluorescence (LF/LIF) of trinitrotoluene traces on a paper surface under synchronized double-pulse laser irradiation. An Nd:YAG-laser (266 nm) was used for the fragmentation of TNT molecules, while fluorescence excitation of their NO fragments was performed using a KrF laser with a generation line of 247.867 nm in the region of the location of the bandhead of the P₁₂ branch of the γ(0, 2) absorption band of the NO molecule. It was shown that the dissociation process of TNT traces has an inertial character and continues after the cessation of the fragmenting laser pulse. It was found that with the delay values between the fragmenting and probing laser pulses in the region of 200 ns, the efficiency of the LF/LIF method can be increased by 12 times. This paper presents the results of an experimental evaluation of the efficiency of two-pulse LF/LIF compared to single-pulse laser exposure, where the fragmentation of TNT molecules and excitation of their NO fragments were simultaneously performed by KrF laser pulses. The possibility of multiple increases in the efficiency of two-pulse LF/LIF with an increase in the energy density of the fragmenting laser radiation was shown. The obtained results are important in terms of increasing the sensitivity and/or range of the LF/LIF method for remote detection of traces of nitrocompounds.

1. Introduction

A radical way to ensure the safety of people when exposed to explosive devices (EDs) is the principle of remoteness at a safe distance [1]. The same principle can be extended to the process of detecting explosive devices: remote, contactless, covert detection of an explosive device allows timeliness and a safe distance to take measures to prevent the act of using it and, thereby, minimize possible tragic consequences.

The technical means of detection used in practice make it possible to reliably detect explosive devices by both direct and indirect signs [2]. However, even being contactless but not remote, these means are not able to ensure the safety of personnel and other people at inspection points in scenarios of accidental or intentional actuation of the device.

In this work, as one of the promising methods for the truly remote detection of explosives, capable of detecting at distances of several tens of meters or more, the method of laser fragmentation (LF) of traces of explosive nitrocompounds followed by laser-induced fluorescence (LIF) of their NO fragments (nitrogen oxide molecules) is considered. The LF/LIF method was first proposed in [3] for the in situ detection of nonfluorescing molecular species. Subsequently, the method found many applications, including in studies on the possibility of detecting nitrocompound vapors [4,5,6,7,8,9,10,11,12] and their traces (microparticles) on surfaces [13,14,15].

The authors demonstrated the possibility of multiple increases in the efficiency of LF/LIF of nitrobenzene and nitrotoluene vapors with synchronized double-pulse exposure [16,17]. The purpose of this work is to experimentally determine the combination of double-pulse laser irradiation parameters that provide increased efficiency of the LF/LIF method for detecting traces of TNT on the surface. To accomplish this, we will study the dependence of the intensity of the LIF of NO fragments of TNT traces on the delay between the fragmenting and probing laser pulses. At the second stage, we will determine the most effective combination of energy densities of fragmenting and probing pulses at an optimal delay between pulses.

2. Experimental Technique

The experimental technique was discussed in previous works by the authors [16,17] and is based on the use of two independent sources of pulsed laser radiation, which make it possible to divide the LF/LIF process into its component parts—the formation of fragments and the excitation of the laser-induced fluorescence of the fragments themselves. This approach makes it possible to study both the conditions and kinetics of the formation of the fragments, as well as the conditions of excitation and the kinetics of the evolution of the fragments themselves. To implement the experimental technique, it is necessary to ensure the possibility of influencing the sample with two laser pulses, combined in space on the surface of the sample and synchronized in time. In this case, it is necessary to ensure the ability to smoothly change the delay time between pulses as well as regulate the radiation intensity of each laser pulse. It is also necessary to ensure precise registration of weak signals of the LIF of NO fragments while suppressing the unshifted scattering line by 14–15 orders of magnitude. Any pulsed UV laser, whose generation line falls within the absorption band of TNT vapor, is suitable as a source of fragmenting radiation. There are no high requirements for the width of the spectrum. The fourth harmonic of the Nd:YAG laser (266 nm) is ideal. The requirements for the spectral characteristics of the probing laser are much more stringent. It should be possible to smoothly tune in the region of 248 nm at a line width of 5–10 pm.

To implement the methodology for studying the optimal conditions of the LF/LIF process, an experimental setup was created.

3. Experimental Setup

A block diagram of the installation for the experimental studies is presented in Figure 1. An Nd:YAG laser Q-Smart 850 (LUMIBIRD) was used to fragment the trinitrotoluene molecules. Excitation of the fluorescence of the NO fragments of TNT was carried out using a KrF laser LF-200 (IHCE SB RAS) [18] with a generation line in the region of the bandhead of the P₁₂ branch of the absorption band A²Σ⁺ (v′ = 0) − X²Π (v″ = 2) of the NO molecule. The laser parameters are provided in

Table 1. Laser parameters.

Parameter	Value
	Nd:YAG Laser	KrF Laser
Radiation wavelength, nm	266	247.867
Maximum pulse energy, mJ	100	100
Line width, pm	30	5
Pulse repetition rate, Hz	10	10
Pulse duration (τ0.5), ns	5	30
Beam divergence, mrad	<0.5	1
Output beam size, mm	Ø9	18 × 9

.

The output radiation beams of the fragmenting (LS1) and probing (LS2) lasers are combined using rotating guide mirrors M1 and M2, and a dichroic mirror DM, and directed to the object Ob. The lens systems L1–L2 and L3–L4 are used to change the radiation divergence in order to provide the required radiation energy density of each laser on the surface of the object. Laser synchronization is carried out using a PG pulse generator controlled by a PC. The optical fluorescence response generated on the surface of an object under the influence of laser radiation is collected by the optical system of a double diffraction spectrometer (holographic concave grating, focal length 380 mm, f/3.2). The spectrometer provides suppression of the unshifted scattering line by 12 orders of magnitude and performs spectral separation of the γ(0, 0) fluorescence band of the TNT NO fragments in the wavelength range 223.0–228.1 nm with an efficiency of no less than 25%. To record the spectra, a multichannel photodetector based on a time-gated intensified CCD camera iStar DH-712 (Andor Technology Ltd., Belfast, UK) was used.

Spatial alignment of laser beams was carried out using a dichroic mirror DM, which was an LP02-266RU interference filter (Semrock Inc., New York, US) installed at an angle of 30° to the direction of incidence of the radiation. At an oblique incidence of radiation, the edge of the spectral characteristic of the filter shifts to the short-wavelength region of the spectrum, providing high transmission for Nd:YAG laser radiation (266 nm) and effective reflection for excimer laser radiation (247.867 nm). The alignment of the laser beams was ensured by independent adjustments of mirrors M1 and DM.

To prepare the samples, a solution of TNT in acetone with a concentration of 10 g/L was used. The solution was prepared immediately before the start of the measurements. The solution was applied to a designated area of a paper substrate with a density of 220 g/m² using a pump marker. After drying, the substrate with traces of TNT was placed in the studied area of space at a distance of 5 m from the experimental setup. Since continuous pulsed laser exposure on the same surface area can lead to gradual degradation of TNT traces, during the experiments, the sample was smoothly moved at a constant speed in the transverse direction relative to the laser beam axis using a special trace renewal system. All the measurements, the results of which are presented below, were carried out under laboratory conditions at normal atmospheric pressure and air temperature of 25 °C. A pyroelectric laser energy detector Gentec-EO QE50LP-H-MB-D0 (Gentec Electro-Optics, Inc., Quebec, Canada) was used to control the energy of the laser pulses.

4. Results and Discussion

As an example, Figure 2 shows the spectrum of the optical response under double-frequency pulsed laser irradiation of a paper substrate with traces of TNT. As can be seen from the figure, in the selected spectral range of 223.0–228.1 nm, bands are found that coincide with the calculated positions of two components of the γ(0, 0) fluorescence band of the NO molecule and are spaced from each other by the amount of doublet splitting of the ground state X²Π [19]. The abscissa axis below shows the serial number of the pixel of the CCD matrix in the direction coinciding with the direction of dispersion of the spectrograph. On the top, there is a scale in wavelengths. The optical response was recorded in the photon counting mode when the signal was accumulated over 100 laser pulses. The radiation energy density of the fragmenting and probing lasers on the surface of the object was maintained at the levels of 25.0 ± 0.8 and 23.2 ± 0.8 mJ/cm², respectively, and the delay Δt between the moments of exposure was 100 ns.

In order to determine the dynamic characteristics of the process of formation of NO fragments of TNT traces, the fluorescence intensity measurements were carried out at different delays Δt. To increase the intensity of the optical responses, the signal was integrated over the entire γ(0, 0) fluorescence band in the wavelength range 225.0–227.5 nm (Figure 2).

To assess the contribution of variations in the surface concentration of a trace in the samples to the error of subsequent measurements, the repeatability of the technology for creating a given concentration of TNT traces on the surface of an object was first checked. The level of the surface concentration fluctuations was assessed based on the results of processing the LIF signals recorded in the photon counting mode in ten successive accumulation series over 100 laser pulses (Figure 3). Assuming Poisson statistics for the distribution of photocounts, the error of the measurement results was calculated at a confidence level of 95%. It has been established that the relative theoretical error (coefficient of variation) of each measurement due to the Poisson statistics of the number of photocounts is about 7%. The empirical coefficient of variation for ten series of measurements was 4%, which indicates that variations in the surface concentration of the trace in the samples do not exceed the errors in the photocounts statistics.

During the study of the dynamics of the formation of NO fragments of TNT traces, the delay between the fragmenting and probing pulses varied from 0 s to 2 μs with a variable step. The temporal instability of the position of the lasing pulses (temporal jitter) of the fragmenting (Nd:YAG laser) and probing (KrF laser) pulses was ±1 and ±20 ns, respectively. The measurement results are presented in Figure 4.

As can be seen from Figure 4, the intensity of the LIF of NO fragments increases in the initial region and reaches a maximum in the delay region of about 200 ns. Then, a relatively slow exponential decay is observed with a characteristic time of τ = (863 ± 156) ns. Based on the shape of the obtained dependence, it can be stated that the process of fragmentation of TNT traces is inertial in nature and continues for 200 ns after the termination of the dissociating laser pulse. The maximum concentration of NO fragments is achieved in a time several times longer than the standard fragmentation pulse duration. Obviously, the single-pulse method of implementing the LF/LIF method, when the fragmentation of the TNT molecules and the excitation of their NO fragments are simultaneously performed by KrF laser pulses, does not allow for achieving maximum efficiency, which will be discussed below.

It has been established that the use of an optimal delay of the probing laser pulse relative to the fragmenting one (Δt ≈ 200 ns) leads to an increase in the LIF lidar response by 12 times compared to simultaneous double-pulse exposure (Δt = 0 ns). It was precisely this delay that was established during subsequent measurements of the intensity of the LIF of NO fragments of TNT depending on the energy density of the fragmenting and probing laser radiation.

The obtained dependences of the LIF intensity of NO fragments S(w₂₆₆, w₂₄₈) on the energy density of the fragmenting radiation w₂₆₆ at fixed energy densities of the probing radiation w₂₄₈ are presented in Figure 5. It can be seen from the figure that with the increasing w of both lasers, a monotonic increase in the LIF intensity is observed. The colored lines in the figure show quadratic approximations of the obtained data. The approximation reliability coefficient is R² > 0.98.

As can be seen from Figure 5, single-pulse laser exposure at w₂₆₆ = 0, when the probing radiation simultaneously fragments the TNT traces and excites their NO fragments, expectedly generates some LIF response. However, the use of double-pulse laser exposure at an optimal delay value Δt makes it possible to increase the LIF intensity several times. For the set of values of the energies of the fragmenting and probing pulses w₂₆₆ and w₂₄₈ implemented in the experiment,

Table 2. The value of k for different combinations of pulse energies w₂₆₆ and w₂₄₈.

w266 (mJ/cm2)	w248 (mJ/cm2)
	8.1 ± 0.5	13.0 ± 0.5	18.3 ± 0.5	23.2 ± 0.5
0	1	1	1	1
5.1 ± 0.2	2.2 ± 2.0	2.1 ± 1.6	3.3 ± 1.9	4.1 ± 2.1
10.0 ± 0.2	2.9 ± 2.7	4.9 ± 3.2	5.2 ± 2.7	5.9 ± 2.8
15.1 ± 0.2	7.5 ± 6.5	7.3 ± 4.5	7.5 ± 3.7	9.3 ± 4.1
20.0 ± 0.2	18.7 ± 14.6	17.0 ± 9.5	16.0 ± 7.3	18.1 ± 7.4
25.0 ± 0.2	19.3 ± 15.0	18.3 ± 10.1	20.3 ± 9.1	22.2 ± 9.0

shows the values of the multiplicity of the LIF intensity for various combinations. The magnitude of the multiplicity was calculated through the ratio of LIF signals using the following formula.

$$k =S\left(w_{266}, w_{248}\right)/S\left(w_{266}=0, w_{248}\right)$$

(1)

Thus, the experiments conducted confirm that in order to increase the efficiency of the LF/LIF method, it is necessary to perform synchronized double-pulse laser action on TNT molecules, separating the processes of fragmentation and excitation of the LIF NO fragments. The choice of the optimal value of the time interval between the fragmentation pulse and the excitation pulse allows for multiple increases in the efficiency of the TNT trace detection process.

Comparison of the above results with the data from similar experiments for nitrobenzene and nitrotoluene vapors [16,17] confirms the authors’ conclusions regarding the possible multiple increases in the efficiency of double-pulse LF/LIF when acting on more complex molecules, such as TNT, as well as an increase in the time of formation of their NO fragments. If we take the statement about the inverse dependence of the optimal time of NO fragment formation on the dissociation rate of the initial molecules [17] as true, the results of the present study can be assumed to indicate that the TNT photodissociation rate for 266 nm is about 5 × 10⁶ s − 1.

Of great practical importance is the study of two-pulse LF/LIF for traces of other nitro-group-containing explosives (RDX, PETN, HMX, etc.) and mixtures based on them (Composition B, Composition C, etc.). As estimates show, multiple increases in the efficiency of LF/LIF allow us to hope to achieve the high sensitivity of the method (1 ng/cm²).

5. Conclusions

The results obtained in the course of this work confirm the possibility of significantly increasing the efficiency of the LF/LIF method in detecting traces of nitrocompounds due to synchronized double-pulse laser exposure. Using TNT as an example, it was shown that the organization of the optimal delay between the fragmenting (266 nm) and probing (247.867 nm) laser pulses allows for increasing the intensity of the LIF responses of TNT NO fragments by 12 times compared to simultaneous double-pulse action at Δt = 0 ns. In addition, with an optimal delay, double-pulse LF/LIF allows for increasing the sensitivity of the method several times compared to the single-pulse LF/LIF used in early studies on the detection of nitrocompounds.

The possibility of increasing the efficiency of the LF/LIF method demonstrated in this work makes us take a new look at the prospects for using this method in solving the problems of the remote detection of traces of explosives on the surface of objects. The use of a double-pulse detection method makes it possible to reduce the energy density of the acting pulses and simplify the process of technical implementation through the use of standard commercially available lasers. The development of optimal conditions for using the double-pulse detection method for other types of nitro-group-containing explosives requires additional research.