Liquid explosives detection in transparent containerslaser-detect.com/wp-content/uploads/2016/09/Liquids.pdf · Liquid explosives detection in transparent containers M. Gaft a,

Liquid explosives detection in transparent containers

M. Gafta, L. Naglia,b

a Laser Distance Spectrometry, Petah Tikwa, Israel

b University Tel Aviv, Israel

Abstract. The principal possibility to recognize liquid explosives and their components in various

glass and plastic containers with different transparency in visible spectral range was demonstrated.

Acetone was used as a target, as alone and mixed with traditional liquids. The advantage of gated

Raman spectroscopy over the CW was proved. It was found that using 532 nm, 6 ns laser pulses any

real target with characteristic Raman spectrum with intensity similar to those for acetone may be

detected in 100 % of glass and in 80 % of plastic containers. The mixing with different liquid makes

detection more difficult and acetone was detected in 55 % of studied cases. The main reasons for

detection difficulties are intrinsic Raman and luminescence of plastic containers and liquids relevant

to airport passengers. In case of strong luminescence the advantages of red light excitation over

green light was demonstrated.

1. Introduction

The constant threat of terror attack on civilian targets emphasizes the need to develop a

sophisticated means of detecting explosive materials in different aggregations phases and amounts.

Lately terror organizations enhance their efforts of delivering attacks using bombs camouflaged to

fine alcohol hoping a security services won’t open a several hundred's dollar worth bottle. Thus there

is a demand for simple and reliable device for chemical analysis of liquids inside sealed bottles.

Detection of liquid explosives inside of different kind of transparent or semitransparent containers

requires the development of new analytical techniques. A suitable analytical instrument must be able

to detect and identify in almost real-time the explosive liquids materials with a reasonable level of

confidence. It is well recognized that laser-based spectroscopy is the proper technique, which may be

potentially capable for this purpose in airport scenario. Laser Induced Luminescence and Laser

excited Raman scattering may be proposed as suitable for this purpose. In most cases the liquids

Please verify that (1) all pages are present, (2) all figures are acceptable, (3) all fonts and special characters are correct, and (4) all text and figures fit within themargin lines shown on this review document. Return to your MySPIE ToDo list and approve or disapprove this submission.

7664 - 58 V. 5 (p.1 of 19) / Color: No / Format: A4 / Date: 2010-03-08 05:26:47 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

luminescence spectra are broad and not specific enough, thus luminescence may not be used alone

but must be combined with Raman spectroscopy. Several Raman based devices were proposed

recently for liquid detection inside containers, but all of them are based on continuous wave (CW)

Raman technique [1,2]. It is well known that luminescence of bottles and liquids may interfere with

CW Raman measurements. Gated Raman spectroscopy technique using a visible laser for excitation

seems to be very promising for our goal showing several advantages over CW Raman [3].

The aim of this paper was to check the ability of our laboratory gated Raman system to recognize

acetone and its mixtures with relevant liquids, such as wines, oils and different soft drinks, in more

than 60 transparent and semitransparent glass and plastic vessels. The intensity of acetone Raman

spectrum was compared with some liquid explosives components, such as nitromethane, hydrogen

peroxide and different acids to check our ability to develop a real detector. Several excitation

wavelengths in visible diapason have been compared.

2. Experimental technique

The detailed description of the experimental system was presented elsewhere [3]. The sample

holder was modified to contain different kinds of bottles. Two excitation sources were used: (i)

second harmonic of Continuum Micro-Light Nd-YAG (532 nm, 6 ns, 40 mJ) and (ii) tunable

Opotek’s Opolette 355 Optical Parametric Oscillator (OPO) in spectral range from 0.410 μm up to

2.5 μm with pulse duration about 6 ns and laser pulse energy about 1 mJ. The experiment was

performed with two different CCD exposure times: gate widths – 10 ns for gated Raman

measurements and 9 ms for a quasi-continuous Raman. In the gated Raman experiment a gate width

is of the order of magnitude of laser pulse, so the much longer luminescence processes interfere less

than in quasi-continuous experiment, where the luminescence has much greater influence. It’s

important to emphasize that both measurements were performed in exactly the same physical

conditions. Both empty and filled bottles were measured. All Nd-YAG measurements were

performed averaging over a 100 laser pulses. In the case of OPO excitation averaging was over 1000

laser pulses due to weakness of laser pulse. Raman signals intensity is measured from the base of the




line on broadband background. Signal to Noise (S/N) was evaluated as a ratio of the normalized

intensity to the electronic noise of the detection system. The transmittance spectra have been

measured using the container pieces from different bottles by Jasco spectrophotometer (model V-

530) in 400-1000 nm spectral range. The numerical transmittance data are presented for 535 and 590

nm points, which represent the limits of the spectral range where Raman spectrum has been

measured.

3. Experimental results

3. 1. Raman spectra of acetone in glass bottles

Mainly acetone was selected as a target compound for this study and sometimes ethanol and

acetone/ethanol mixtures were used. Table 1 represents the types of studied glass containers and

Figures 1, 2 their typical transmittance spectra and Raman spectra of acetone inside those containers.

400 500 600 700 800 900 1000

84

85

86

87

88

750 1000 1250 1500 17500

200

400

400 500 600 700 800 900 10000

10

20

30

750 1000 1250 1500 17500

10

20

400 500 600 700 800 900 1000

30

40

50

60

750 1000 1250 1500 17500

20

40

60

782.32074

1063.08381216.07171427.3138 1712.3131

779.63592

1063.08381213.4983 1424.7964

792.45913

1075.9127 1446.5278 1711.3816

Tran

smitt

ance

, %

535 nm-87.5 % 590 nm - 88 %

Vodka Absolute, transparent

Inte

nsity

(a.u

.)

Vodka Absolute

Inte

nsity

(a.u

.)

Tran

smitt

ance

, %

535 nm - 19 %

590 nm -30 %

Liquor Frangeliko, brown Liquor Frangeliko

Inte

nsity

(a.u

.)

Whisky Glenlivet, green

Tran

smitt

ance

, %

nm

535 nm - 57%

590 nm - 42 %

Whisky Glenlivet

Raman Shift, cm-1

Figure 1. Transmittance spectra of transparent, brown and green bottles and Raman spectra of

acetone inside those containers.




Table 1. The studied glass bottles and the main Raman line of acetone intensity

N Name Color Transmittance, %

Normalized Intensity

au

S/N

1 Vodka Absolute Transparent 87.5-88 400

100

2 Vodka Smirnoff Transparent 580

150

3 Shablis Red Wine Greenish 260

75

4 Sardasol, Red Wine Green 180

50

5 Frangeliko Liquor Brown 19-30 10

10

6 Bacardi Breezer Lemon Transparent 1600

500

7 Bacardi Breezer Melon Transparent 1760 500 8 Beer Grolsch Green 420 100 9 Malt Nesher Brown 10-28 30 15

10 Gerber Juice Pear Transparent 125 100 11 Liquor Coffee Transparent 150 50 12 Kikkoman Transparent 30 75 13 Olive Oil Transparent 20 50 12 Perfume NafNaf Red 17.2-17.5 30 10 13 Arak Transparent 100 25 14 Perfume Love Line Red 1.5-11.5 55 15 15 Cream Revlon Absolutes Dull 9.5-10 110 50 16 Malt Glenlivet Green 57-42 40 15 17 Perfume Toxic Red 5-16 20 10 18 Perfume Dee Deep Red 5-15.5 40 15

In all glass containers acetone was definitely detected by its characteristic Raman lines. Both

the strongest line peaking at 782 cm-1 and the weaker lines peaking at 1063, 1216, 1427 and 1712

cm-1 may be seen, which makes the identification ability very high and the false rate actually zero.

The intensity of the main Raman peak is changing from approximately 1800 to 25 arbitrary units

depending on transmittance of the corresponding glass in the spectral range of 535-590 nm, which is

relevant for Raman lines. The lowest intensities were detected for brown and red glasses. No one of

the studied glass materials revealed strong luminescence in 535-590 nm spectral range using gated

Raman technique, while it is the main obstacle for traditional CW Raman. Only one red glass

exhibits detectable luminescence. Studied glass bottles mostly do not exhibit strong intrinsic Raman




lines. The only exception with detectable Raman line near 1094 cm-1 is red perfume bottle (Fig. 2),

but it is very weak and will not prevent Raman identification of the future target compounds.

Because those containers do not reveal their own luminescence or Raman signals, even one Raman

line of suspicious material in any part of the Raman spectrum (400-3000 cm-1) will be enough.

Comparing the gated and not gated spectra it was found that for glass the short-gated spectra have

major advantage over quasi-continuous ones (Fig. 3). There is a much weaker luminescence that can

mask a Raman signal in the short-gated spectra. It appears that while gated Raman demonstrates no

differences between various bottles presenting a strong and clear quartz lines, the quasi-continuous

spectra present clear variations connected with intrinsic luminescence. Acetone and ethanol have

been measured inside glass bottles by gated and not gated modes. In such scenario, we have to move

the focus inside the bottle. It makes the signal coming from the walls weaker while the main signal

comes from the liquids. Thus, the glass material which bottles are made of is less important as long

as it is transparent for the visible light. Nevertheless, comparison between the gated Raman spectra

and actually not-gated demonstrates that gating enables to remove totally the interference from bottle

material luminescence and Raman signatures are much clearer, allowing to better detect and

recognize the substance inside the bottle (Fig. 4).

400 500 600 700 800 900 10000

10

20

30

750 1000 1250 1500 17500

20

40

60

80

400 500 600 700 800 900 1000

16

17

18

19

750 1000 1250 1500 17500

20

40

60

400 500 600 700 800 900 10000

10

20

30

750 1000 1250 1500 17500

20

40

785.23091

1065.6168

1436.18031705.7927

789.7976

1065.5555

1441.5486 1713.797

785.2321

1065.61571230.27831431.2089

1708.1324

Tran

smitt

ance

, %

Perfume Naf-Naf, red

535 nm - 1.5 %

590 nm - 11.5 %

Inte

nsity

(a.u

.) Perfume Naf Naf

Tran

smitt

ance

, %

535 nm - 17.2 % 590 nm - 17.5 %

Perfume Love Line, red

Inte

nsity

(a.u

.)

Bottle

Tran

smitt

ance

, %

nm

535 nm - 5 %

590 nm - 16 %

Perfume Toxic, red

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Perfume Toxic

Figure 2. Transmittance spectra of red bottles and Raman spectra of acetone inside those containers.




500 1000 1500 2000 25000

50

100

150

500 1000 1500 2000 25000

200

400

600

500 1000 1500 2000 25000

500

1000

500 1000 1500 2000 25000

1000

2000

3000

4000

500 1000 1500 2000 25000

100

200

300

400

500 1000 1500 2000 25000

100

200

300

500 1000 1500 2000 25000

200

400

600

500 1000 1500 2000 25000

100

200

Inte

nsity

(a.u

.) 1

Beer Tuborg, brow n

Tuborg beer, green

Inte

nsity

(a.u

.)Carlseberg beer, bright green

Tuborg beer,bright green

Inte

nsity

(a.u

.)

Edelw eiss beer, brow n

6

Utenos beer, brow n

Inte

nsity

(a.u

.)

Raman Shift, cm -1

Santa Ana Cabarne Souvignon,dark green)

Raman Shift, cm -1

Angove's Shiraz, dark green

Figure 3. Raman spectra of different glass bottles with gating of 10 ns (low line) and without gating

(upper line).

500 1000 1500 2000 25000

1000

2000

3000

4000

5000

6000

500 1000 1500 2000 25000

500

1000

1500

500 1000 1500 2000 25000

100

200

300

400

500

500 1000 1500 2000 25000

500

1000

1500

2000

95.38

752.056

1037.791396.791677.22

754.773

1032.541396.791677.22

852.0031058.75

1424.31849.318

1019.41

1242.39

1421.81

Inte

nsity

(a.u

.)

Acetone in green beer bottle

Acetone in brown red wine bottle

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Ethanol in green beer bottle

Raman Shift, cm-1

Ethanol in dark green red wine bottle

Figure 4. Raman spectra of acetone and ethanol in different glass bottles with gating of 10 ns (low

line) and without gating (upper line).




3. 2. Raman spectra of acetone in different plastic bottles

Actually all plastic containers exhibit very strong luminescence under excitation by 532 nm

both in not gated and gated modes. Nevertheless the gated technique enables to substantially reduce

the luminescence interference (Fig. 5). Table 2 represents the types of studied plastic containers and

Figures 6, 7 their typical transmittance spectra and Raman spectra of acetone inside those containers.

Opposite to the glass containers, acetone was definitely detected by its characteristic Raman lines

only in 26 plastic bottles from 35 while in two bottles only the strongest acetone Raman line may be

identified because of influence of intrinsic plastic Raman. Thus acetone may be detected in 28 from

35 plastic containers (80 %). The intensity of the main Raman peak is changing from ~ 300 to 0

arbitrary units. The lower maximal intensity compared to glass containers is connected to

substantially lower (~ 100 times) excitation energy which may be used for Raman excitation. The

reason is plastic burning under excitation density more than 1 MW/cm2. It appears that the main

reason when the identification is not possible is not the low transparency of plastic containers, but

their intrinsic Raman and luminescence. In our experiments it appears that focusing laser light inside

the plastic bottle partly prevent the bottle burning but not improve Raman signal quality. The

identification of acetone was not possible in seven plastic bottles. In two of them the intrinsic Raman

is so strong that it covers the target compound Raman. In other five bottles the strong luminescence

even using gated Raman technique prevents detection.

500 1000 1500 2000 25000

500

1000

1500

2000

2500

3000

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Figure 5. Typical Raman spectra of plastic bottle with gating of 10 ns (low line) and without gating

(upper line).




400 500 600 700 800 900 1000

80

82

84

86

750 1000 1250 1500 17500

20

40

60

80

100

400 500 600 700 800 900 10000

1

2

3

750 1000 1250 1500 17500

50

100

150

400 500 600 700 800 900 10000

10

20

750 1000 1250 1500 17500

10

20

30

782.32074

1063.0838

1429.8305 1717.1906

785.004891429.8305

1712.3131

768.14893

1049.0539 1202.7509 1421.5916

Tran

smitt

ance

, % 535 nm - 84 %590 nm - 84.5 %

Fanta Ornage Mango, transparent

Fanta Orange mango

Tran

smitt

ance

, %

Tran

smitt

ance

, %

535 nm - 0.3 % 590 nm - 0.3 %

Johnson Baby Oil, white semi-transparent

Johnson Baby Oil

nm

535 nm - 9.8 %

590 nm - 3.3 %

Bleach Economica, blue, semi-transparent

Raman Shift, cm-1

Acetone 100%

Bleach Economica

Figure 6. Transmittance spectra of transparent, white and blue semi-transparent plastic bottles and Raman spectra of acetone inside those containers.

400 500 600 700 800 900 1000

40

50

60

70

750 1000 1250 1500 17500

50

100

400 500 600 700 800 900 100060

70

80

90

750 1000 1250 1500 17500

20

40

60

400 500 600 700 800 900 10000

20

40

60

80

750 1000 1250 1500 17500

20

40

60

80

790.610791288.60321431.2089

1612.22471724.7764

Tran

smitt

ance

, %

590 nm - 53.1 %535 nm - 50 %

Retarding Lavander, white semi-transparent

Retarding Lavander

Tran

smitt

ance

, %

535 nm - 65 %

590 nm - 85 %

Life, pink transparent

Inte

nsity

(a.u

.)

Life

Tran

smis

sion

, %

nm

535 nm - 1.2 %

590 nm - 42.0 %

Washing Softener, red semi-transparent

Inte

nsity

(a.u

.)

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Washing Softener

Figure 7. Transmittance spectra of white, pink and blue semi-transparent plastic bottles without Raman spectra of acetone inside those containers.




Table 2. The studied plastic bottles and the main Raman line of acetone intensity

N Name Color Transmittance, %

Normalized Intensity, au

S/N

1 Fanta Orange Mango Transparent 84.0-84.5 70 35 2 Primor Fruit Juice Transparent 140 75 3 Tapuzina Diet Transparent 190 100 4 Prigat Apple Honey Transparent 25 20 5 Baby Oil Transparent 135 70 6 Sod Max Transparent 100 60 7 Johnson Baby Oil White 0.3 – 0.3 10 3 8 Glass Cleaner Transparent 170 50 9 Nestea Mango Apple Transparent 80 10

10 Tapuzina Fresh Apple

Transparent 90 12

11 Prigat Citrus Honey Transparent 73.7 – 74.1 30 15 12 Pepper Mouth Wash Transparent 30 30 13 Cool Mint Mouth

Wash Transparent 25 25

14 Coca Cola Zero Transparent 30 20 15 Fairy Original Transparent 60 50 16 Pashtan Oil Transparent 15 25 17 Canola Oil Transparent 20 20 18 Bleach Economica Bluish 9.8-3.3 7 5 19 Retarding Lavender White 50-53.1 0 0 20 Souse Soya Transparent 88-88.2 0 0 21 Life Pink 65-85 0 0 22 Life Blue 82-73 100 100 23 Life Purple 120 80 24 Life Transparent 150 125 25 Body Wash Transparent 120 100 26 Bath Gel Transparent 86.5-87 27 20 27 Body Scrub Transparent 32 25 28 Thai Sweat Chili

Sauce Purple 40 35

29 Pachuki Lavander Purple 10 5 30 Shock White 20 15 31 Vaseline White 0.55-0.65 0 0 32 Washing Softener Purple 10-12 0 0 33 Body Lotion Pink 42.5-71 0 0 34 Washing Softener Red 1.2-42 0 0 35 Empty Keter Transparent 70 75




3. 3. Raman spectra of acetone in glass bottle mixed with different liquids

Table 3 represents the types of studied mixtures and Figure 8 their typical Raman spectra inside

transparent glass bottle. The identification probability becomes lower compared to identification in

empty bottles. Acetone was definitely detected in mixtures with 18 liquids from 34 while in mixtures

with 14 liquids only the strongest acetone Raman line may be identified and the detection is

potentially possible, but very difficult. In mixtures with two liquids the detection is impossible. The

reasons for the difficult detection are the influence of intrinsic liquids Raman and luminescence.

750 1000 1250 1500 17500

50

100

150

750 1000 1250 1500 17500

50

100

150

200

750 1000 1250 1500 17500

200

400

750 1000 1250 1500 17500

100

200

300

787.68836

875.86429

1076.14

1228.92881424.7964

1707.4328 790.37116

1704.9917

793.053281148.9429

1514.9978

790.37116

1614.1955

Vodka Absolute+Acetone, 50+50 %

Acetone

Acetone

Raman Shift, cm-1Raman Shift, cm-1

Bacardi Breezer Melon+Acetone, 50+50 %

Inte

nsity

(a.u

.)

Inte

nsity

(a.u

.)

Grape fruite+acetone, 50+50%

Acetone

Acetone

Prigat Apple Honey+Acetone, 50+50 %

Figure 8. Typical Raman signal of liquid-acetone mixtures measured in transparent glass bottle.

All studied liquids are characterized by very strong luminescence with relatively weak ethanol

lines for alcohol drinks. This strong luminescence present both in gated and in quasi-continuous

modes, but it is several times lower for the first technique (Fig. 9). Thus common liquids that can be

transported in passenger’s luggage present a strong luminescence when excited by 532 nm




wavelengths. This luminescence can mask the signal coming from explosive or other potentially

dangerous materials. One known way to avoid luminescence excitation is to use red or NIR

excitations. We checked several liquids with intensive luminescence using different excitation

wavelengths and arrived to conclusion that such approach may be really helpful. For example, Figure

10 represents cognac luminescence spectra under different visible excitations. It may be seen that

cognac exhibits slightly different luminescence with different excitations. The band’s maximum is at

500, 540 and 580 nm for 410, 480 and 532 nm excitations, respectively. The excitation spectra

present a clear tendency of growing luminescence intensity towards the blue region. This tendency

takes place both for measurements at the center of luminescence band and for the off-center ones.

Starting from excitation by 590 nm the cognac luminescence becomes very weak. Similar behavior

was found for other alcohol liquids, such as whisky. In red wine, however, the luminescence

behavior is different and the band’s center is constant peaking at 650 nm (Fig. 11). But once again,

starting from excitation by 590 nm the red wine luminescence becomes very weak.

500 1000 1500 2000 25000

1000

2000

3000

500 1000 1500 2000 2500

0

1000

2000

3000

500 1000 1500 2000 25000

1000

2000

3000

4000

5000

500 1000 1500 2000 25000

500

1000

1500

2000

2500

3000

849.3181421.81

849.3181045.661419.31

Inte

nsity

(a.u

.) Whisky

Cognac

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Martini

Raman Shift, cm-1

Red wine

Figure 9. Raman spectra of several non-explosive liquids in transparent glass bottle with gating of

10 ns (low line) and without gating (upper line).




Table 3. The studied mixtures of acetone and liquids

N Name Normalized Intensity au

S/N

1 Vodka Absolute 110 100 2 Vodka Smirnoff 275 150 3 Shablis Red Wine 125 75 4 Sardasol, Red Wine 6 3 5 Frangeliko Liquor 10 10 6 Bacardi Breezer Lemon 80 50 7 Bacardi Breezer Melon 10 10 8 Beer Grolsch 30 15 9 Malt Nesher 5 5 10 Fanta Orange Mango 45 25 11 Grape Fruit Juice 50 25 12 Tapuzina Diet 65 50 13 Prigat Apple Honey 50 25 14 Gerber Juice Pear 40 15 15 Baby Oil 20 10 16 Sod Max 15 10 17 Johnson Baby Oil 25 15 18 Glass&Window Cleaner 50 25 19 Nestea Mango Pine Apple 35 20 20 Tapuzina Fresh Apple 60 25 21 Whisky Glenlivet 70 45 22 Liquor Coffee 150 50 23 Prigat Citrus Honey 10 10 24 Pepper Mint Mouth Wash 10 10 25 Cool Mint Mouth Wash 10 10 26 Coca Cola Zero 0 0 27 Fairy Original 50 25 28 Kikkoman 0 0 29 Olive Oil 15 10 30 Pashtan Oil 12 10 31 Canola Oil 13 10 32 Bleach Economica 10 10 33 Arak 65 25 34 Perfume Love Line Rouge 5 5




500 550 600 650 700 7500

50

100

150

200

250

300

350

Inte

nsity

(arb

. uni

ts)

Wavelength (nm)

530nm 540nm 550nm 560nm 570nm 580nm 590nm 600nm 610nm 620nm 630nm

Figure 10. Cognac luminescence with different excitations (sharp picks are laser excitation lines).

500 550 600 650 700 750

15

30

45

60

75

90

Inte

nsity

(arb

. uni

ts)

Wavelength (nm)

530nm 540nm 550nm 560nm 570nm 580nm 590nm 600nm 610nm 620nm 630nm

Figure 11. Red wine luminescence with different excitations. (Sharp picks are laser excitation lines).

It may be concluded that application of red and not green Raman excitation appears preferable to

overcome the luminescence interference. Thus we start to check the red Raman spectra of several




from studied objects. Figure 12 demonstrates such spectra for acetone and ethanol. In contrast with

excitation by 532 nm, the difference between short gated and long gated spectra is substantially less.

Relatively strong luminescence exists under red excitation also, but its influence is less prominent.

For example, under excitation by 532 nm almost no intrinsic ethylene Raman signal was visible on

strong luminescence background in alcohol drinks, while it may be seen even in red wine spectrum,

while in cognac, whiskey and white wine they are strong and clear (Figure 13).

600 800 1000 1200 1400 1600 18000

200

400

600

600 800 1000 1200 1400 1600 18000

100

200

300

784.92508

1065.828371277.638141430.04781712.57823

1096.066781276.01805

1455.3253

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Ethanol

In

tens

iuty

(a.u

.)

Acetone

Figure 12. Raman spectra of acetone and ethanol under excitation by 650 nm in transparent glass

bottle with gating of 10 ns (low line) and without gating (upper line).

3.4. Raman spectra of TNT and RDX dissolved in acetone and ethanol

The Raman spectra of TNT and RDX diluted in acetone and ethanol have been studied.

Comparing those four different possibilities for explosive solution demonstrates that the most




recognizable features relate to TNT solution in acetone, evidently due to good dissolving properties

of acetone and strong TNT Raman signal. Example of such spectra are presented in Figure 14.

Subtracting a pure acetone spectrum from TNT-acetone mixtures and plotting the highest TNT peak

(ν = 1326 cm-1) height against the TNT concentration, the good linear correlation was received

between TNT concentration and its Raman signal (Fig. 15). It may be concluded that it is possible to

detect TNT mixtures in acetone down to a concentration of 10 mg/cm3.

Figure 13. Raman spectra of several non-explosive liquids under excitation by 650 nm in transparent

glass bottle with gating of 10 ns (low line) and without gating (upper line).

3.4. Raman spectra of TNT and RDX dissolved in acetone and ethanol

The Raman spectra of TNT and RDX diluted in acetone and ethanol have been studied.

Comparing those four different possibilities for explosive solution demonstrates that the most

recognizable features relate to TNT solution in acetone, evidently due to good dissolving properties

800 1000 1200 1400 1600 18000

100

200

300

400

500

800 1000 1200 1400 1600 18000

500

1000

1500

2000

2500

3000

800 1000 1200 1400 1600 18000

50

100

150

200

250

300

350

400

800 1000 1200 1400 1600 18000

50

100

150

200

1082.643141276.018051453.74546

874.18827

1082.643141276.01805

1452.17567

872.45045

1077.60173

1276.01805

1452.17567

Inte

nsity

(a.u

.)

Cognak

Red wine

Inte

nsity

(a.u

.)

Raman Shift, cm-1

Wisky

Raman Shift, cm-1

White semidry wine




of acetone and strong TNT Raman signal. Example of such spectra are presented in Figure 14.

Subtracting a pure acetone spectrum from TNT-acetone mixtures and plotting the highest TNT peak

(ν = 1326 cm-1) height against the TNT concentration, the good linear correlation was received

between TNT concentration and its Raman signal (Fig. 15). It may be concluded that it is possible to

detect TNT mixtures in acetone down to a concentration of 10 mg/cm3.

Figure 14. Raman spectra of TNT, acetone and 30 % TNT solution in acetone

Figure 15. Correlation between Raman signal of TNT and its concentration in acetone

1000 1500 2000 25000

50

100

Inte

nsity

(a.u

.)

Raman shift cm-1

TNT powder Acetone 30w% TNT in acetone

10 20 30 40 50 60 70

0.3

0.4

0.5

0.6

1326

cm

-1 p

eak

heig

ht

TNT concentration (mg/cm3)




Other mixtures show worse results. The ratio between the RDX and acetone peaks is relatively

small even for high concentrations of RDX. The same picture is for TNT solution of ethanol – only

in saturated solution the TNT spectral fingerprint can be clearly seen. The worst case is RDX

solution in ethanol. The poor diluting power of ethanol combines with the fact that the highest RDX

peak overlaps with highest ethanol peak and consequently even in saturated solution no RDX

features can be clearly seen.

3.5. Raman spectra of liquid and solid explosive components

Figures 16 and 17 demonstrate Raman spectra of some potentially dangerous liquids

components measured at the same experimental conditions. It may be seen that fortunately all have

specific Raman signatures with intensity similar to those for acetone, which makes Raman

spectroscopy a potential tool for their identification inside transparent containers.

800 1000 1200 1400 1600 1800

20

40

60

80

100

120

Inte

nsity

, a.u

.

Raman shift cm-1

Acetone

30 % H2O2

Nitromethane

Figure 16. Raman spectra of acetone, 30% solution of hydrogen peroxide and nitromethane




800 1000 1200 1400 1600 1800

20

40

60

80

100

120

140

Inte

nsity

a.u

.

Raman shift cm-1

Acetone

H3PO4

H2SO4

Figure 17. Raman spectra of acetone, H2SO4 and H3PO4 acides

4. Conclusions

This study was devoted to Raman detection of pure acetone in different glass and plastic

containers, and of acetone mixed with different liquids in transparent containers. Laboratory device

used for this task includes excitation laser (second harmonic, 532 nm, 6 ns) and gated ICCD iStar

detector. Seventeen glass containers with different transparencies and colors have been tested.

Acetone was detected in 100 % of them. Thirty five plastic containers with different transparencies

and colors have been tested. Acetone was be detected in 80 % of them. Thirty-four different liquid

mixtures with acetone in transparent containers have been tested. Acetone detected in 55 % of them.

The main obstacle for acetone detection is strong intrinsic luminescence of certain plastics

and liquids under excitation by 532 nm. Gated Raman technique with red excitation with optical

parametric oscillator (OPO) pulsed laser as excitation source is evidently the better approach.

Besides better luminescence removing, plastic materials are burning under green excitation density

more than 1 MW/cm2. Lower excitation intensities lead to relatively weak Raman signals. Once

again, using less energetic red excitation may help to overcome this problem.




References

[1] "Fist Defender" www.ahuracorp.com

[2] Matousek, P., Clark,I., Draper,E., Morris,M., Goodship,A., Everall,N., Towrie,M., Finney, W.

and Parker,A.,"Subsurface probing in diffusely scattering media using spatially offset Raman

spectroscopy", Appl. Spectrosc. 59,393-400 (2005).

[3] Gaft, M. and Nagli, L., "Laser-based spectroscopy for standoff detection of explosives." Optical

Materials, 30, 1739-1746 (2008).




Documents

Liquid explosives detection in transparent containerslaser-detect.com/wp-content/uploads/2016/09/Liquids.pdf · Liquid explosives detection in transparent containers M. Gaft a,