A colorimetric sensor array for detection of triacetone triperoxide vapor

Article abstract of DOI:10.1021/ja107419t

Triacetone triperoxide (TATP), one of the most dangerous primary explosives, has emerged as an explosive of choice for terrorists in recent years. Owing to the lack of UV absorbance, fluorescence, or facile ionization, TATP is extremely difficult to detect directly. Techniques that are able to detect generally require expensive instrumentation, need extensive sample preparation, or cannot detect TATP in the gas phase. Here we report a simple and highly sensitive colorimetric sensor for the detection of TATP vapor with semiquantitative analysis from 50 ppb to 10 ppm. By using a solid acid catalyst to pretreat a gas stream, we have discovered that a colorimetric sensor array of redox sensitive dyes can detect even very low levels of TATP vapor from its acid decomposition products (e.g., H₂O₂) with limits of detection (LOD) below 2 ppb (i.e., <0.02% of its saturation vapor pressure). Common potential interferences (e.g., humidity, personal hygiene products, perfume, laundry supplies, volatile organic compounds, etc.) do not generate an array response, and the array can also differentiate TATP from other chemical oxidants (e.g., hydrogen peroxide, bleach, tert-butylhydroperoxide, peracetic acid).

Full text of DOI:10.1021/ja107419t

Published on Web 10/15/2010
A Colorimetric Sensor Array for Detection of Triacetone Triperoxide Vapor
Hengwei Lin and Kenneth S. Suslick*
Department of Chemistry, UniVersity of Illinois at Urbana-Champaign, 600 South Mathews AVenue,
Urbana, Illinois 61801
Received August 25, 2010; E-mail: ksuslick@illinois.edu
detection (LOD) below 2 ppb (i.e., <0.02% of its saturation vapor
pressure) (Figure 1 and Supporting Information (SI) Figure S1).
Abstract: Triacetone triperoxide (TATP), one of the most danger-
ous primary explosives, has emerged as an explosive of choice
for terrorists in recent years. Owing to the lack of UV absorbance,
fluorescence, or facile ionization, TATP is extremely difficult to
detect directly. Techniques that are able to detect generally
require expensive instrumentation, need extensive sample prepa-
ration, or cannot detect TATP in the gas phase. Here we report
a simple and highly sensitive colorimetric sensor for the detection
of TATP vapor with semiquantitative analysis from 50 ppb to 10
ppm. By using a solid acid catalyst to pretreat a gas stream, we
have discovered that a colorimetric sensor array of redox sensitive
dyes can detect even very low levels of TATP vapor from its acid
decomposition products (e.g., H2O2) with limits of detection (LOD)
below 2 ppb (i.e., <0.02% of its saturation vapor pressure).
Common potential interferences (e.g., humidity, personal hygiene
products, perfume, laundry supplies, volatile organic compounds,
etc.) do not generate an array response, and the array can also
differentiate TATP from other chemical oxidants (e.g., hydrogen
peroxide, bleach, tert-butylhydroperoxide, peracetic acid).
Triacetone triperoxide (TATP), a high-powered primary explo-
sive first synthesized in 1895,1 has become well-known during the
Figure 1. (a) Acid catalyzed decomposition of TATP. (b) Color difference
past decade as an explosive of choice for terrorists.2 Due to its
extreme sensitivity, TATP does not have any practical engineering
or military applications. It has been used, however, in at least three
major terrorist acts in the past 10 years,2b in part because TATP is
easy to prepare from readily available chemicals (i.e., acid catalyzed
reaction of acetone with hydrogen peroxide)3 but difficult to detect.4
While several techniques for TATP detection have been developed
in recent years5 and there are several commercial products
available,6 these methods generally require expensive and generally
nonportable instrumentation,7 need extensive sample preparation,8
are qualitative or have poor limits of detection,5 or are limited to
liquid or solid samples and cannot detect TATP vapor.9 Hence,
the development of an inexpensive, portable, and easy-to-use device
for the field detection of TATP remains a high priority. Particularly,
on-site detection of TATP vapor would have significant advantages
for rapid screening.
In recent years, our group has developed a colorimetric sensor
array methodology10 that has been applied successfully for the
identification of a wide range both of toxic industrial gases and
vapors11 and of organic analytes in aqueous liquids.12 Herein, we
describe a colorimetric sensor array for the sensitive and selective
detection of the vapor phase of TATP. Owing to the lack of UV
absorbance, fluorescence, or facile ionization, TATP is extremely
difficult to detect directly: indeed, even redox sensitive dyes are
not very responsive to TATP vapor. By using a solid acid catalyst
to pretreat a gas stream, we have discovered that a colorimetric
sensor array can detect even very low levels of TATP vapor from
its acid decomposition products13 (e.g., H2O2) with limits of
maps of TATP vapor at concentrations specified after 5 min (top row) and
10 min (bottom row) of exposure at 50% relative humidity and 298 K. For
display purposes, the color range of these difference maps is expanded from
4 to 8 bits per color (RGB range of 4-19 expanded to 0-255). The identity
of the nanoporous pigments and a full database of the color differences are
given in Supporting Information Tables S1 and S2.
We have extensively tested a colorimetric sensor array against
a wide range of TATP vapor concentrations, using a solid acid
catalyst to decompose the TATP in a gas flow to H2O2. After
screening several possible acid catalysts, we chose Amberlyst-15,
the acid form of a sulfonated highly cross-linked polystyrene ion-
exchange polymer;14 see SI for experimental details and cautions.
Using an ordinary flatbed scanner, digital images of an array were
acquired before and after exposure to TATP vapor at various
concentrations. The red, green, and blue values of each spot in the
array were measured before and after TATP vapor exposure, and
color difference maps were generated. These difference maps
provide a pattern that effectively identifies different concentrations
of TATP vapor, as shown in Figure 1b. This colorimetric array is
capable of semiquantitative detection of TATP vapor well below
50 ppb. For comparison, the array response to TATP vapor without
the use of an acid catalyst is nearly a 100-fold less (SI Figure S3),
and therefore prior redox sensitive dyes have not previously proved
useful for the detection of TATP vapor.9
TATP vapor concentrations were confirmed by in-line analysis
in real time using an FT-IR multigas analyzer (SI Figure S4). The
amount of catalyst required was optimized (Figure 2); Amberlyst-
15 proved to be extremely efficient in the decomposition of TATP
9
10.1021/ja107419t  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 15519–15521 15519

C O M M U N I C A T I O N S
vapor, and the observed amount of acetone corresponded to the
stoichiometry expected from TATP decomposition (3:1). These
results indicate that Amberlyst-15 is an excellent solid-acid catalyst
for TATP vapor decomposition because negligible amounts of
TATP vapor or its decomposition products were adsorbed.
after 10 min of exposure, i.e., <0.02% of its saturated vapor
concentration at room temperature (∼69 ppm at 25 °C).16
Figure 4. Total Euclidean distance of the array plotted versus concentration
of TATP vapor after 5 min of exposure. Inset shows concentration vs total
Euclidean distance plot for the lower concentrations with a linear fit. The
average of three trials is shown with error bars. The Euclidean distance is
simply the total length of the 48-dimensional color-difference vector, i.e.
the total array response.
Figure 2. Decomposition of TATP vapor by a solid-acid Amberlyst-15
catalyst. At t ) 0, a gas stream containing 10 ppm of TATP vapor is
initiated; no acetone is observed. At t ) 15 min (i), a small Teflon tube
containing 20 mg of Amberlyst-15 was inserted quickly into the gas stream,
and TATP and acetone were monitored: TATP concentration immediately
drops and acetone concentration increases with the expected stoichiometry.
At t ) 40 min, the catalyst tube was removed, and TATP returned rapidly
to 10 ppm and acetone to 0 ppm. This process was repeated with (ii) 10
mg and (iii) 5 mg of catalyst: as the amount of catalyst was decreased, so
did the extent of TATP decomposition.
As shown in SI Figure S5, one may also use hierarchical cluster
analysis (HCA) to analyze the color change patterns of the array.
Dendrograms are based on the clustering of the array response data
in the 48 dimensional ∆RGB color space (i.e., 16 changes in red,
green, and blue values). In quadruplicate trials at different
concentrations of TATP vapor, there were no misclassifications out
of 44 trials, even at the lowest concentrations (50 ppb).
In real world applications, interferents, including changes in
humidity, can be highly problematic. With the indicators used in
this sensor array, however, only strong oxidants will elicit any
response, and therefore the array should be immune to interference
from nearly all common chemicals, vapors, or odorants. We have
examined a wide range of relevant potential interferents (18 in total)
and also the effects of humidity on our sensor array and found no
significant response (SI Figure S6). Common potential interferents
might include personal care products typically found in travelers’
baggage: e.g., mouthwash, body wash, toothpaste, shampoo,
perfume, lotion, liquor, vinegar, laundry supplies, and various
volatile organic compounds (e.g., acetone, ethyl acetate, benzene,
toluene, and petroleum ether, which are possible volatile compo-
nents in paints, finger nail polish, etc.). Arrays were exposed at
room temperature to these interferents at 5% and 10% of their
saturated vapor at 50% RH for 10 min: importantly, none of these
interferents invoked any significant response from the sensor array.
In addition, no change in the array was observed after 10 min of
exposure to a relative humidity (RH) range from 10 to 90% RH
compared to 50% RH. The specificity of this colorimetric sensor
array for TATP was rationally planned: the array components were
chosen to be responsive only to strong oxidants, and thus, our array
is indeed insensitive to the presence of nearly all odorants, VOCs,
and common potential interferents, or to changes in humidity.
In addition, the shelf life stability and reproducibility are of the
utmost importance if these arrays are to be used in practical
applications. Exposure of the array to air has no effect for a period
of several days. Additional investigations showed that the response
of the array to TATP vapor was unaffected by the presence or
absence of O2 in the gas flow. For long-term storage, the arrays
are best kept under N2, and we observe very little change over 2
months in their response to 1 ppm or 10 ppm TATP vapor (SI
Figure 3. Two-dimensional principal component analysis (PCA) plot.
TATP vapor at different concentrations, other peroxides (encircled in gray
ellipsoids), and controls for changes in humidity (50% RH and 50% vs
10% RH and 50% vs 90% RH). TBHP ) tert-butylhydroperoxide.
To measure the array response at different concentrations of
TATP, principal component analysis (PCA)15 was performed and
a two-dimensional plot was obtained (Figure 3) with excellent
separation even at low concentrations. In addition, the array’s
response to changes in relative humidity was negligible, clustering
inseparably with the controls.
We next examined the relationship between the total Euclidean
distances of the color changes (i.e., square root of the sums of the
squares of the ∆RGB values) of the array as a function of TATP
vapor concentration. As shown in Figure 4, after a 5 min exposure
time, the total Euclidean distance tracks monotonically with
increasing TATP vapor concentration from 50 ppb to 10 ppm. When
the data were examined, a linear response was obtained for lower
concentrations (Figure 4 inset). An extrapolated limit of detection
based on 3 × S/N of the three largest color changes was 2 ppb
9
15520 J. AM. CHEM. SOC. VOL. 132, NO. 44, 2010

C O M M U N I C A T I O N S
(2) (a) Naughton, P. Times Online, July 15, 2005. http://www.timesonline.co.uk/
tol/news/uk/article544334.ece. (b) TATP has been used during various
recent terrorist acts, including in (i) Richard Reid’s shoe bomb on American
Airlines Flight 63 (Cooper, R. T. Los Angeles Times, December 29, 2001,
p A12. http://articles.latimes.com/2001/dec/29/news/mn-18822); (ii) the
July 7th 2005 London bombings. (Townsend, M. The real story of 7/7.
The ObserVer, May 7, 2006. http://www.guardian.co.uk/uk/2006/may/07/
theobserver.uknews); and (iii) the attempted bombing in Northwest Airlines
Flight 253 on Christmas Day, December 25, 2009 (“Indictment in U.S. V.
Abdulmutallab”. January 6, 2010. http://www.cbsnews.com/htdocs/pdf/
Abdulmutallab_Indictment.pdf (retrieved January 10, 2010).
(3) Laine, D. F.; Roske, C. W.; Cheng, I. F. Anal. Chim. Acta 2008, 608, 56–
60.
Figure S7); practically no variation was observed in the color
difference maps, and a tight clustering of data collected regardless
of array age was obtained. In addition, three separate printing
batches of arrays gave nearly identical results for tests at 1 ppm
and 10 ppm TATP vapor (SI Figure S8), which demonstrates
excellent reproducibility for printing of the arrays and their response
to TATP.
Finally, it should be noted that our array can also distinguish
TATP from H2O2 or other volatile oxidants (e.g., hypochlorite
bleach, peracetic acid, and tert-butyl hydroperoxide). As clearly
shown in Figures 3 and S9, there is no confusion in the array
response among TATP, H2O2, Clorox bleach, peracetic acid, or tert-
butyl hydroperoxide in sextuplicate trials at 10 ppm. Furthermore,
we note that Amberlyst-15 does not affect the array response to
H2O2 or the other oxidants but dramatically changes the array
response to TATP vapor (due to its acid catalyzed decomposition),
which provides additional discrimination between TATP vapor and
other oxidants (SI Figure S9).
While the laboratory studies reported here made use of inex-
pensive flatbed scanners for imaging, we have recently constructed
a fully functional prototype hand-held device, as shown in SI Figure
S10, which has a 3-fold improved S/N compared to the flatbed
scanners. Combined with a low dead volume cartridge (SI Figure
S11), a hand-held device could present a rapid, inexpensive, and
highly sensitive method for portable monitoring of TATP vapor,
e.g., for screening of luggage.
In conclusion, we have created a simple, disposable colorimetric
sensor array that is capable of sensitive and semiquantitative
detection of the vapor phase of the primary explosive TATP with
limits of detection below 0.02% of its saturation vapor pressure.
The array is highly selective for TATP, is unaffected by changes
in humidity or by the presence of many common potential
interferents, and can differentiate TATP from other chemical
oxidants.
(4) Dubnikova, F.; Kosloff, R.; Zeiri, Y.; Karpas, Z. J. J. Phys. Chem. A 2002,
106, 4951–4956.
(5) (a) Schulte-Ladbeck, R.; Vogel, M.; Karst, U. Anal. Bioanal. Chem. 2006,
386, 559–565. (b) Burks, R. M.; Hage, D. S. Anal. Bioanal. Chem. 2009,
395, 301–313.
(6) (a) ACRO Security Technologies: ACRO-P.E.T. . (b) Mistral Security Inc.
www.mistralgroup.com/SEC_explosives.asp. (c) Bauer, C.; Sharma, A.;
Willer, U.; Burgmeier, J.; Braunschweig, B.; Schade, W.; Blaser, S.;
Hvozdara, L.; Mu¨ller, A.; Holl, G. Appl. Phys. B 2008, 92, 327–333. (d)
Santillan, J.; Brown, C.; Jalenak, W. Proc. SPIE 2007, 6540, 65400P. (e)
Staples, E. In Electronic Noses & Sensors for the Detection of ExplosiVes;
Gardner, J., Yinon, J., Eds.; Kluwer: Dordrecht, 2004.
(7) (a) Schulte-Ladbeck, R.; Kolla, P.; Karst, U. Anal. Chem. 2003, 75, 731–
735. (b) Wilson, P. F.; Prince, B. J.; McEwan, M. J. Anal. Chem. 2006,
78, 575–579. (c) Schulte-Ladbeck, R.; Edelmann, A.; Quinta´s, G.; Lendl,
B.; Karst, U. Anal. Chem. 2006, 78, 8150–8155. (d) Cotte-Rodr´ıguez, I.;
Herna´ndez-Soto, H.; Chen, H.; Cooks, R. G. Anal. Chem. 2008, 80, 1512–
1519. (e) Lindley, R.; Normand, E.; Howieson, I.; McCulloch, M.; Black,
P.; Lewis, C.; Foulger, B. Proc. SPIE 2007, 6741, 67410P. (f) Lindley,
R.; Normand, E.; McCulloch, M.; Black, P.; Howieson, I.; Lewis, C.;
Foulger, B. Proc. SPIE 2008, 7119, 71190K.
(8) (a) Schulte-Ladbeck, R.; Karst, U. Anal. Chim. Acta 2003, 482, 183–188.
(b) Malashikhin, S.; Finney, N. S. J. Am. Chem. Soc. 2008, 130, 12846–
12847.
(9) (a) Germain, M. E.; Knapp, M. J. Inorg. Chem. 2008, 47, 9748–9750. (b)
Lu, D.; Cagan, A.; Munoz, R. A. A.; Tangkuaram, T.; Wang, J. Analyst
2006, 131, 1279–1281. (c) Sella, E.; Shabat, D. Chem. Commun. 2008,
5701–5703. (d) Apblett, A.; Kiran, B.; Malka, S.; Materer, N.; Piquette,
A. Ceram. Trans. 2005, 172, 29–35.
(10) (a) Rakow, N. A.; Suslick, K. S. Nature 2000, 406, 710–713. (b) Suslick,
K. S.; Bailey, D. P.; Ingison, C. K.; Janzen, M.; Kosal, M. E.; McNamara,
W. B., III; Rakow, N. A.; Sen, A.; Weaver, J. J.; Wilson, J. B.; Zhang, C.;
Nakagaki, S. Quim. NoVa 2007, 30, 677–681.
(11) (a) Janzen, M. C.; Ponder, J. B.; Bailey, D. P.; Ingison, C. K.; Suslick,
K. S. Anal. Chem. 2006, 78, 3591–3600. (b) Lim, S. H.; Feng, L.; Kemling,
J. W.; Musto, C. J.; Suslick, K. S. Nature Chem. 2009, 1, 562–567. (c)
Feng, L.; Musto, C. J.; Kemling, J. W.; Lim, S. H.; Suslick, K. S. Chem.
Commun. 2010, 46, 2037–2039. (d) Feng, L.; Musto, C. J.; Suslick, K. S.
J. Am. Chem. Soc. 2010, 132, 4046–4047.
Acknowledgment. This work was supported through the NIH
Genes, Environment and Health Initiative through Award
U01ES016011.
(12) (a) Zhang, C.; Suslick, K. S. J. Am. Chem. Soc. 2005, 127, 11548–11549.
(b) Musto, C. J.; Lim, S. H.; Suslick, K. S. Anal. Chem. 2009, 81, 6526–
6533.
(13) Armitt, D.; Zimmermann, P.; Ellis-Steinborner, S. Rapid Commun. Mass
Spectrom. 2008, 22, 950–958.
Supporting Information Available: Experimental details, Figures
S1-S11, and Tables S1-S2. This material is available free of charge
via the Internet at http://pubs.acs.org.
(14) Amberlyst-15 is manufactured by Rohm and Haas and available from
Sigma-Aldrich.
(15) Jolliffe, I. T. Principal Component Analysis; Springer-Verlag: New York,
2002.
(16) Oxley, J. C.; Smith, J. L.; Shinde, K.; Moran, J. Propellants, Explos.,
Pyrotech. 2005, 30, 127–130.
References
(1) Wolfenstein, R. Chem. Ber. 1895, 28, 2265–2269.
JA107419T
9
J. AM. CHEM. SOC. VOL. 132, NO. 44, 2010 15521