sensors

Communication
Detection of Triacetone Triperoxide (TATP) and Hexamethylene
Triperoxide Diamine (HMTD) from the Gas Phase with
Differential Ion Mobility Spectrometry (DMS)
Mirosław Maziejuk 1 , Monika Szyposzyńska 1, * , Aleksandra Spławska 1 , Monika Wiśnik-Sawka 1 and
Michał Ceremuga 2

1 Military Institute of Chemistry and Radiometry, al. gen. A. Chruściela “Montera”, 105,
00-910 Warsaw, Poland; [email protected] (M.M.); [email protected] (A.S.);
[email protected] (M.W.-S.)
2 Military Institute of Armament Technology, ul. Prym. St. Wyszyńskiego, 7, 05-220 Zielonka, Poland;
[email protected]
* Correspondence: [email protected]

Abstract: One of the significant problems in the modern world is the detection of improvised
explosives made of materials synthesized at home. Such compounds include triacetone triperoxide
(TATP) and hexamethylene triperoxide diamine (HMTD). An attempt was made to construct an
instrument allowing for the simultaneous detection of both compounds despite the large difference of



vapor pressure: very high for TATP and very low for HMTD. The developed system uses differential


ion mobility spectrometry (DMS) in combination with a specially designed gas sample injection
Citation: Maziejuk, M.; system. The created system of detectors allowed for the detection of a high concentration of TATP and
Szyposzyńska, M.; Spławska, A.; a very low concentration of HMTD. TATP detection was possible despite the presence of impurities—
Wiśnik-Sawka, M.; Ceremuga, M. acetone remaining from the technological process and formed as a coproduct of diacetone diperoxide
Detection of Triacetone Triperoxide (DADP) synthesis. Ammonia added to the carrier gas improved the possibility of detecting the
(TATP) and Hexamethylene
abovementioned explosives, reducing the intensity of the acetone signal. The obtained results were
Triperoxide Diamine (HMTD) from
then compared with the detection capabilities of drift tube ion mobility spectrometer (DT-IMS), which
the Gas Phase with Differential Ion
has not made possible such detection as DMS.
Mobility Spectrometry (DMS). Sensors
2021, 21, 4545. https://doi.org/
Keywords: differential ion mobility spectrometry (DMS); ion mobility spectrometry (IMS); explosives;
10.3390/s21134545
triacetone triperoxide (TATP); hexamethylene triperoxide diamine (HMTD)
Academic Editor: Santiago Marco

Received: 31 May 2021

Accepted: 29 June 2021 1. Introduction
Published: 2 July 2021 The development of the detection of explosives is important for military and security
purposes due to the use of explosives in terrorist attacks. This is especially important
Publisher’s Note: MDPI stays neutral for materials that can be manufactured at home. Such materials include triacetone triper-
with regard to jurisdictional claims in oxide (TATP), which was used in the attack in July 2005 in London, and hexamethylene
published maps and institutional affil- triperoxide diamine (HMTD), which was used in the terrorist attack in December 1999
iations.
in the USA. Explosives detection is based on two strategies: the detection of substance
vapors and particle analysis after swipe sampling. Vapor detection is faster as no sampling
or preparation is required. However, vapor analysis is more challenging due to the low
amount of substance evaporated from the surface. The vapor pressure (VP) is the pressure
Copyright: © 2021 by the authors. of a gas above the surface of a substance. This value for explosives varies from a single hPa
Licensee MDPI, Basel, Switzerland. to 10−15 hPa.
This article is an open access article Vapor pressure measurement is difficult due to slight signal variations. The determi-
distributed under the terms and nation of VP is performed at an elevated temperature, which may contribute to material
conditions of the Creative Commons
decomposition or molecular desorption from the surface, which may result in the appear-
Attribution (CC BY) license (https://
ance of additional analytical signals. Vapor pressure can be determined by a dynamic
creativecommons.org/licenses/by/
method consisting of mass change measurements of a substance at a constant tempera-
4.0/).

Sensors 2021, 21, 4545. https://doi.org/10.3390/s21134545 https://www.mdpi.com/journal/sensors

 [1,2] and static methods, such as manometric VP determination [3] or surface phase stud-
ies using gas chromatography [4,5].
Many studies have investigated the vapor pressure of TATP [2,6–9], diacetone diper-
oxide (DADP) [1,2] and HMTD [4,10,11]. Ewing and colleagues [12] compiled the values
Sensors 2021, 21, 4545 of the vapor pressure of explosive substances available in the literature and standardized 2 of 12
them by determining their average values. Figure 1 presents the VPs for various explo-
sives.
DADP
ture is a peroxide
[1,2] and with the
static methods, highest
such VP, which isVP
as manometric 24.7 Pa; for TATP,[3]
determination this
or value is 6.39
surface phase
Pa.studies
The vapor pressure of these substances
using gas chromatography [4,5]. is compared to that of liquid nitro derivatives
such asManynitroglycerin (NG).
studies have HMTD hasthe
investigated a vapor
vapor pressure
pressure of150
TATPtimes lowerdiacetone
[2,6–9], than thatdiper-
of
TATP, which is 3.9 × 10 −2 Pa. These values are significantly higher than those of commonly
oxide (DADP) [1,2] and HMTD [4,10,11]. Ewing and colleagues [12] compiled the values
used strong
of the grinding
vapor pressurematerials such as
of explosive octogen (HMX)
substances or hexogen
available (RDX). The
in the literature and vapor pres-
standardized
sures
them ofby
thedetermining
described acetone peroxides
their average are Figure
values. higher1than that of
presents theTNT,
VPs for × 10 Pa.
9.27various−4
explosives.

Figure
Figure 1. Vapor
1. Vapor pressure
pressure values
values forfor various
various groups
groups of of explosive
explosive compounds
compounds (Reprinted
(Reprinted from
from [12]).
[12]).

DADP
Since TATPis aand
peroxide
HMTD with
do the
nothighest
containVP, which
nitro is 24.7 Pa;
or aromatic for TATP,groups,
functional this value is 6.39 Pa.
analytical
The vapor pressure of these substances is compared to that of liquid
devices for conventional explosives are unsuitable for detecting them. In recent years,nitro derivatives such
as nitroglycerin (NG). HMTD has a vapor pressure 150 times lower than that of TATP,
many techniques of−2TATP and HMTD analysis have been developed, including infrared
which is 3.9 × 10 Pa. These values are significantly higher than those of commonly used
(IR) [13,14], liquid chromatography (LC) [15–17] and IMS detectors, which play a signifi-
strong grinding materials such as octogen (HMX) or hexogen (RDX). The vapor pressures
cant role in detecting these materials [18–21].
of the described acetone peroxides are higher than that of TNT, 9.27 × 10−4 Pa.
Since TATP and HMTD do not contain nitro or aromatic functional groups, analytical
1.1. IMS Technique
devices for conventional explosives are unsuitable for detecting them. In recent years,
Ion techniques
many mobility spectrometry
of TATP and (IMS)
HMTD usesanalysis
differences
haveinbeen
ion mobility
developed, in carrier gas infrared
including under
the(IR)
influence of an electric field [22–26] (Figure 2). A classical IMS (drift tube
[13,14], liquid chromatography (LC) [15–17] and IMS detectors, which play a significant ion mobility
spectrometer, DT-IMS)
role in detecting thesespectrometer has a weak linear electric field with values from 150
materials [18–21].
to 400 V/cm. Within these fields, ion mobility is constant.
1.1.Ion
IMSseparation
Techniquein a homogeneous electric field takes place based on differences in
their mobility. The results
Ion mobility obtained
spectrometry from
(IMS) usesthedifferences
detector are presented
in ion mobility ininthe form gas
carrier of drift
under
time spectra. An IMS detector with a membrane was used in the laboratory research.
the influence of an electric field [22–26] (Figure 2). A classical IMS (drift tube ion mobility
spectrometer, DT-IMS) spectrometer has a weak linear electric field with values from 150 to
400 V/cm. Within these fields, ion mobility is constant.
Ion separation in a homogeneous electric field takes place based on differences in their
mobility. The results obtained from the detector are presented in the form of drift time
spectra. An IMS detector with a membrane was used in the laboratory research.
Sensors 2021, 21, x FOR PEER REVIEW 3 of 12
Sensors 2021, 21, 4545 3 of 12

Figure 2. Operating principle for ion mobility spectrometry (IMS).

Figure 2. Operating principle for ion mobility spectrometry (IMS).
1.2. DMS Technique
1.2. DMS
A variantTechnique
of IMS is differential ion mobility spectrometry (DMS) [27–31]. In DMS,
A variant
ions are of IMS because
distinguished is differential ion mobility
of the difference betweenspectrometry
mobilities(DMS)
at high[27–31].
and lowInelectric
DMS,
ions
fields [32,33]. For strong electric fields (exceeding 12 kV/cm), ion mobility depends onelec-
are distinguished because of the difference between mobilities at high and low the
tric fields
applied [32,33].
field, andForthe strong
dependence electricis fields (exceeding
nonlinear. 12 kV/cm),between
The relationship ion mobility dependsand
ion mobility on
the applied
intensityfield,
of theand the dependence
electric field is expressedis nonlinear.
by EquationThe relationship
(1) [34]. between ion mobility
and the intensity of the electric field is expressed by Equation (1) [34].
K ( E/N ) = K0 ·[1 + α( E/N )] (1)
K ( E / N ) = K 0 ⋅ [1 + α ( E / N )] (1)
2 − 1 − 1
where K0 is the reduced ion mobility [cm V s ], E/N is the electric field in Townsend
where
units (1K0Td is the reduced
= 10 −17 V·cm ion2 ),mobility
which is [cm 2V−1s−1], E/N is the electric field in Townsend units
known as the normalized molecular density, and
(1 Td = 10
α(E/N) is theV·cm
−17 2 ), whichfunction
normalized is known as the normalized
describing the electricmolecular density,
field mobility and α(E/N) is
dependence.
the normalized function describing the electric field mobility
DMS spectrometers consist of an ionization area and a space in which ions are sepa- dependence.
ratedDMS (Figure spectrometers
3) [35,36]. An consist of an
isotopic ionization
source emitting areaβ and a space
particles in which
is placed ionsionization
in the are sepa-
rated (Figure 3) [35,36]. An isotopic source emitting β particles
region. Carrier gas transports generated ions through the separation region, which consists is placed in the ionization
region. Carrierplates
of two parallel gas transports
on which generated
metal electrodes ions through
are placed, thethe
separation
distance region,
betweenwhich con-
electrodes
sists of two parallel plates on which metal electrodes are
(gap) should be almost 0.5 mm. The electric field produced by these electrodes is perpen- placed, the distance between
electrodes
dicular to the (gap)
gasshould be almostThe
flow direction. 0.5 voltage
mm. The electricthe
between field produced
electrodes by these
consists of electrodes
a variable
is
component called the separation voltage (SV) and compensation voltage (CV). consists of
perpendicular to the gas flow direction. The voltage between the electrodes
a variable component
The waveform of called
SV is shownthe separation
in Figurevoltage (SV) and
3c: it should be acompensation
rectangular wave voltage
with(CV).
30%
The practically
of duty; waveform of SVvery
it is is shown
similar in toFigure 3c: itline.
a dotted should be a rectangular
Voltage wave withfrom
should be regulated 30%
of
400duty;
V to practically
2000 V, 1 MHz it is very
or abovesimilar to a dotted
should line. Voltage
be applied; should be
if the frequency is regulated
too low we from 400
cannot
V to 2000
seen V, 1 MHz
the light ions. or above should be applied; if the frequency is too low we cannot seen
the light
The ions.
compensation voltages are shown in Figure 3b as a very low wave. These voltages
The
are generated compensation
as a special voltages
electricare field shown
whichinworksFigure as3banas a very
ion filter.low wave. These volt-
ages In arethegenerated
establishedas aconditions
special electric
only one fieldtype
whichof ionworks
comesas an ion filter.
through the control electrode
to theInion thecurrent
established conditions
electrode. In the only
time onet1 thetype of ion comes
mobility of ionsthrough
M1 is K1the andcontrol
in the electrode
time t2 it
to the but
is K2, ion current electrode. In
the compensation the time
voltage t1 thethese
reduces mobility of ions and
differences M1 is K1 and
allows ioninM1thetotime
reacht2
it
theis collecting
K2, but theelectrode.
compensation The othervoltageions reduces
going these
to thedifferences and allows
control electrode cannot ionbeM1observed
to reach
via the
the collecting
collecting electrode.
electrode. TheSlowly
other changing
ions goingthe to compensation voltage cannot
the control electrode allowedbeusobserved
to make
a graph (function ion current depending on compensation
via the collecting electrode. Slowly changing the compensation voltage allowed us to voltage) of all ions generated
in the aionization
make region.ion
graph (function This allowed
current us to order
depending from higher tovoltage)
on compensation lower the values
of all of α
ions gen-
(differences
erated in theofionization
mobility).region. This allowed us to order from higher to lower the values
of α (differences of mobility).
Sensors 2021, 21, x FOR PEER REVIEW 4 of 12
Sensors 2021, 21, 4545 4 of 12

Figure 3. Principle of differential mobility spectrometry. (a) Schematic diagram of DMS spectrometers; (b) CV waveform;
Figure 3. Principle of differential mobility spectrometry.
(c) SV waveform.

The analyte
The analyteintroduced
introducedinto
intothe
theIMS/DMS
IMS/DMS detector is ionized
detector by charge
is ionized transfer
by charge (pro-
transfer
(proton) between the reaction ions and the neutral molecule of the analyte. The changethe
ton) between the reaction ions and the neutral molecule of the analyte. The change in in
enthalpy
the of this
enthalpy reaction
of this taking
reaction takingplace
placein in
the gas
the gasphase
phaseisiscalled
calledproton
protonaffinity
affinity (PA). For
(PA). For
the ionization
the ionization of
of each
each substance,
substance, the
the values
values are
are different for TATP
different for TATP and
and HMTD.
HMTD. They They are
are
860 kJ/mol [37] and 940 kJ/mol [11], respectively.
860 kJ/mol [37] and 940 kJ/mol [11], respectively.

2. Materials and Methods

The DMS chamber used to build the measuring system was made of ceramic sub-
strates [32]. The
Thegasgaswas
wasionized
ionizedwith
withananelectrode
electrodemade made ofof nickel
nickel isotope
isotope (63 Ni).
(63Ni). TheThe
de-
detector drift
tector drift areaconsisted
area consistedofoftwo
twoparallel
parallelelectrode plates (5 ××25
electrode plates 25 mm)
mm) placed 0.635 mm mm
apart. A high-amplitude 3 MHz asymmetric voltage waveform waveform was was generated
generated by by the HSV
generator.
generator. Voltage
Voltagewas wasapplied
appliedtotothe
theelectrodes
electrodes across
acrossthethe
flow direction
flow of the
direction of carrier gas.
the carrier
The applied voltage range of separation was from 7 Td to 118 Td (peak
gas. The applied voltage range of separation was from 7 Td to 118 Td (peak to peak). A to peak). A constant
weak compensation
constant voltage ranging
weak compensation voltage from −5.4 from
ranging Td to −5.4
+2.1 Td
Td to
was applied
+2.1 Td was to applied
the electrodes.
to the
At the end of the drift section, two collecting electrodes (5 × 5 mm)
electrodes. At the end of the drift section, two collecting electrodes (5 × 5 mm) were were installed, which in-
in the DMS
stalled, which detector allowed
in the DMS the simultaneous
detector obtaining of aobtaining
allowed the simultaneous signal from of apositive and
signal from
negative ions,negative
positive and as shown in Figure
ions, as shown 3. Measurements were made using
in Figure 3. Measurements air as using
were made carrierair
gas,as
dried with 13 X molecular sieves (Merck, France). Internal gas flow
carrier gas, dried with 13 X molecular sieves (Merck, France). Internal gas flow throughthrough the detector
was 3.2 L/min.
the detector wasThe3.2 detector
L/min. The temperature was 45 ◦ C. was 45 °C.
detector temperature
The
The gas
gas system
systemisispresented
presentedininFigure
Figure4.4.The
Themeasurement
measurement system
system was
was constructed
constructed at
WIChiR (Warszawa) for the purpose of detection and identification
at WIChiR (Warszawa) for the purpose of detection and identification of TATP and of TATP and HMTD
(Figure
HMTD 5). The system
(Figure 5). The was constructed
system with the with
was constructed use ofthethree
userotation
of threepumps
rotation (Thomas)
pumps
(P1, P2, P3) with flow rates respectively of 1, 1, 3.2 L/min and two
(Thomas) (P1, P2, P3) with flow rates respectively of 1, 1, 3.2 L/min and two three-waythree-way valves (Z1,
Z2). The advantage of the constructed system is the possibility to work
valves (Z1, Z2). The advantage of the constructed system is the possibility to work in the in the membrane
mode
membranewhenmode
detecting
whensubstances with high volatility
detecting substances with highand in the and
volatility membrane-free
in the membrane- mode
when detecting substances with low volatility, such as
free mode when detecting substances with low volatility, such as HMTD.HMTD.
 Sensors 2021, 21, x FOR PEER REVIEW 5 of 12
Sensors 2021, 21, 4545 5 of 12

Figure 4. Diagram of the measurement system.

In the membrane mode, the Z1 valve is switched to gas flow through the membrane,
which is forced by the P1 pump at this time. The Z2 valve at the P2 pump outlet allows
gas to flow through the carbon filter, and then the sample is taken from the membrane
exchanger (sample introduction system), and goes to the DMS.
In the membrane-free mode, the gas to be analyzed is charged directly to the DMS.
Setting the valve Z1 to take air allows direct sampling from the test area.
The P2 pump allows you to regulate the flow with the required amount of gas for
analysis (external circuit). In this mode, the valve Z2 at this time is switched over to allow
gas to be purged after the analysis.
The P3 pump is used to regulate the carrier gas flow in the DMS chamber at 3.2 L/min
(internal circuit). The applied molecular sieves clean the gas in the internal circuit of the
DMS chamber.
Due to the high vapor pressure of TATP, most measurements for this substance are
made in the membrane mode, while for HMTD all measurements are made in the mem-
brane-free mode. In the membrane mode, the appearance of the HMTD peaks was not rec-
orded as the intensity of peaks was below the limit of detection (LOD). TATP, HMTD, and
DADP were obtained synthesized at the Military Institute of Armament Technology
Figure 4. Diagram of the measurement system.
(Zielonka).
Figure 4. Diagram of the measurement system.

In the membrane mode, the Z1 valve is switched to gas flow through the membrane,
which is forced by the P1 pump at this time. The Z2 valve at the P2 pump outlet allows
gas to flow through the carbon filter, and then the sample is taken from the membrane
exchanger (sample introduction system), and goes to the DMS.
In the membrane-free mode, the gas to be analyzed is charged directly to the DMS.
Setting the valve Z1 to take air allows direct sampling from the test area.
The P2 pump allows you to regulate the flow with the required amount of gas for
analysis (external circuit). In this mode, the valve Z2 at this time is switched over to allow
gas to be purged after the analysis.
The P3 pump is used to regulate the carrier gas flow in the DMS chamber at 3.2 L/min
(internal circuit). The applied molecular sieves clean the gas in the internal circuit of the
DMS chamber.
Due
Figure5.5.The
to the high vapor
Themeasurement
measurementsystem.
pressure of TATP, most measurements for this substance are
system.
Figure
made in the membrane mode, while for HMTD all measurements are made in the mem-
brane-free mode. Inmode,
In the membrane the membrane mode,
the Z1 valve the appearance
is switched of thethrough
to gas flow HMTD thepeaks was not rec-
membrane,
which is forced by the P1 pump at this time. The Z2 valve at the P2 pump outlet allowsand
orded as the intensity of peaks was below the limit of detection (LOD). TATP, HMTD,
gas DADP
to flowwere
throughobtained synthesized
the carbon at then
filter, and the Military
the sampleInstitute
is takenof from
Armament Technology
the membrane
(Zielonka).
exchanger (sample introduction system), and goes to the DMS.
In the membrane-free mode, the gas to be analyzed is charged directly to the DMS.
Setting the valve Z1 to take air allows direct sampling from the test area.
The P2 pump allows you to regulate the flow with the required amount of gas for
analysis (external circuit). In this mode, the valve Z2 at this time is switched over to allow
gas to be purged after the analysis.
The P3 pump is used to regulate the carrier gas flow in the DMS chamber at 3.2 L/min
(internal circuit). The applied molecular sieves clean the gas in the internal circuit of the
DMS chamber.
Due to the high vapor pressure of TATP, most measurements for this substance
are made in the membrane mode, while for HMTD all measurements are made in the
membrane-free mode. In the membrane mode, the appearance of the HMTD peaks was
not recorded as the intensity of peaks was below the limit of detection (LOD). TATP,
HMTD, and DADP were obtained synthesized at the Military Institute of Armament
Technology (Zielonka).
Figure 5. The measurement system.
 Sensors2021,
Sensors 2021,21,
21,4545
x FOR PEER REVIEW 66ofof12
12

3.3.Results
Results
The
Thegas
gassystem
systempresented
presentedin inFigure
Figure44 was
was used
used to analyze pure TATP,
TATP,DADP,
DADP,HMTD
HMTD
and
andaamixture
mixtureofofTATP
TATP and
andDADP.
DADP.Scattering spectra
Scattering spectrawere recorded
were for for
recorded a separation voltage
a separation volt-
of
age118
ofTd.
118 Td.
For
ForTATP
TATP(99%(99%purity),
purity),monomer
monomerand anddimer
dimerionsionswere
wereobtained
obtainedininpositive
positivemode
mode
(Figure 6). The monomer ion peak was issued at − 0.6 Td and the dimer
(Figure 6). The monomer ion peak was issued at −0.6 Td and the dimer ion peakion peak at +0.1 Td.
at +0.1
No peaks for negative ions were observed.
Td. No peaks for negative ions were observed.

3 and drift
Figure6.6.Dispersion
Figure Dispersionplots
plots for
for positive
positive ions
ions of
of triacetone
triacetone triperoxide (TATP) (500
triperoxide (TATP) µg/m3))and
(500 µg/m drift
spectra
spectrafor
for118
118Td.
Td.
Sensors 2021, 21, x FOR PEER REVIEW 7 of 12
For
ForTATP
TATP(99%
(99%purity),
purity),monomer
monomerand
anddimer
dimerions
ionswere
wereobtained
obtainedininpositive
positivemode
mode
(Figure
(Figure7).
7).
Traces of dimer ions were observed for DADP (low intensity compared to monomer
ions). When increasing the DADP concentration, the RIP and NH3 peaks disappeared.
This means that the peak intensity of the DADP monomer increased but the peak of the
DADP dimer did not increase.
In further studies, a mixture of TATP and DADP (approximately 78% w/w TATP)
was analyzed. This research doped NH3 at two concentrations (50 ppb and 200 ppb) in the
internal circuit. The results for 118 Td are shown in Figures 8 and 9.
In the case of a TATP + DADP mixture for positive ions, there were generally four
peaks: one from the TATP + DADP monomer, the other from the acetone monomer, and
two peaks from the dimer ion, from acetone and TATP. Due to the rather low resolution
of DMS, the peaks derived from the TATP and DADP monomers were not separated and
appeared as one peak. In the presence of a low concentration of ammonium, the dominant
peaks were those of acetone (Figure 9).
For a system with a 200 ppb ammonium concentration, two reaction peaks were visi-
ble: one for ammonium and the other for H+(H2O)n, both of which showed small ampli-
Figure 7. Dispersion plots for positive ions of diacetone
diacetone diperoxide
diperoxide (DADP)
(DADP) (600 µg/m3)3 and
(600 µg/m ) anddrift
drift
tudes. The ammonium
spectra for 118 Td. peak was only twice as high as the noise.
From the spectra presented in Figure 9, we can see that acetone significantly disturbed
the accuracy
Traces ofofdimer
the measurement. In particular,
ions were observed for DADPthe acetone dimer peak
(low intensity interfered
compared with the
to monomer
analysis.
ions). Theincreasing
When acetone peaks were higher
the DADP than thethe
concentration, TATP
RIP peaks.
and NH Doping
3 peaks the carrier
disappeared.gas with
This
means that the peak intensity of the DADP monomer increased but the peak of
200 ppb NH3 caused a decrease in the intensity of the acetone peak (Figure 10). Therefore, the DADP
dimer didconcluded
it can be not increase.
that the minimum value of ammonia concentration in the carrier gas,
whichInisfurther studies,
necessary a mixture
for the of TATP
separation and
of the DADP
peaks from (approximately 78% w/w TATP)
TATP, was approximately 200 was
ppb.
analyzed. This research doped NH3 at two concentrations (50 ppb and 200 ppb) in the
internal circuit. The results for 118 Td are shown in Figures 8 and 9.
Sensors 2021, 21, 4545 7 of 12
Figure 7. Dispersion plots for positive ions of diacetone diperoxide (DADP) (600 µg/m3) and drift
Figure 7. Dispersion plots for positive ions of diacetone diperoxide (DADP) (600 µg/m3) and drift
spectra for 118 Td.
spectra for 118 Td.

Figure8.8. Differential
Figure Differentialion
ion mobility
mobility spectrometry
spectrometry (DMS)
(DMS) spectrum
spectrum for
for 118
118 Td
Td for
for TATP
TATPpositive
positiveions,
ions,
Figure 8. Differential ion mobility spectrometry (DMS) spectrum for 118 Td for TATP positive ions,
withNH
with NH33 atat aa concentration
concentrationof
of50
50ppb
ppbin
inthe
theinternal
internalcircuit.
circuit.
with NH3 at a concentration of 50 ppb in the internal circuit.

Figure 9. DMS spectrum for TATP positive ions, with NH3 at a concentration of 200 ppb in the
internal circuit.

In the case of a TATP + DADP mixture for positive ions, there were generally four
peaks: one from the TATP + DADP monomer, the other from the acetone monomer, and
two peaks from the dimer ion, from acetone and TATP. Due to the rather low resolution
of DMS, the peaks derived from the TATP and DADP monomers were not separated and
appeared as one peak. In the presence of a low concentration of ammonium, the dominant
peaks were those of acetone (Figure 9).
For a system with a 200 ppb ammonium concentration, two reaction peaks were
visible: one for ammonium and the other for H+ (H2 O)n , both of which showed small
amplitudes. The ammonium peak was only twice as high as the noise.
From the spectra presented in Figure 9, we can see that acetone significantly disturbed
the accuracy of the measurement. In particular, the acetone dimer peak interfered with the
analysis. The acetone peaks were higher than the TATP peaks. Doping the carrier gas with
200 ppb NH3 caused a decrease in the intensity of the acetone peak (Figure 10). Therefore,
 Sensors2021,
Sensors 2021,21,
21,4545
x FOR PEER REVIEW 88of
of12
12

itFigure
can be9. concluded that for
DMS spectrum theTATP
minimum value
positive ions,ofwith
ammonia
NH3 atconcentration
a concentrationinofthe
200carrier
ppb ingas,
the
which is necessary
internal circuit. for the separation of the peaks from TATP, was approximately 200 ppb.

Figure 10.
Figure 10. DMS
DMS spectra
spectra for
for HMTD
HMTD positive
positive ions.
ions.

Figure 10 shows
Figure 10 showsthethespectrum
spectrumfor
for HMTD
HMTD positive
positive ions.
ions. For For this substance,
this substance, both
both mon-
monomer and dimer ions were formed. Table 1 presents the compensating voltage
omer and dimer ions were formed. Table 1 presents the compensating voltage for all pos- for all
positive ions.
itive ions.

Table1.1.Compensating
Table Compensatingvoltage
voltagevalues
valuesfor
forpositive
positiveions.
ions.

Substance
Substance Positive
Positive IonIon CV(Td)
CV (Td)
RIPRIP −1.3
−1.3
TATP
TATP99% 99%purity
purity monomer
monomer −−0.6
0.6
Dimer
Dimer 0.1
0.1
RIPRIP − 1.3
−1.3
DADP 99% purity monomer −0.6
DADP 99% purity monomer −0.6
dimer 0.1
dimer 0.1
RIP −1.3
RIP −1.3
acetone monomer −0.9
TATP + DADP acetone monomer −0.9
TATP + DADP acetone dimer −0.2
(50 ppb NH3 )
acetone
TATP dimer
+ DADP −−0.2
0.7
(50 ppb NH3) TATP
TATPdimer
+ DADP 0.2
−0.7
TATP
RIP dimer 0.2
−1.3
acetone monomer
RIP − 1.0
−1.3
TATP + DADP
acetone dimer −0.2
(200 ppb NH
TATP + DADP 3 ) acetone monomer −1.0
TATP + DADP monomer −0.6
(200 ppb NH3) acetone
TATP dimer
dimer −0.2
0.1
TATP + DADP monomer −0.6
RIP −1.3
HMTD TATP
HMTD monomerdimer 0.1
−1.0
HMTD RIPdimer −−1.3
0.5
HMTD HMTD monomer −1.0
To determine the possibility of detecting HMTD dimer
the mixture of TATP and HMTD,−0.5
tests were
carried out with the DT-IMS detector. The tests were carried out under the same conditions
as with a DMS detector, with concentrations of TATP totaling 250 µg/m3 and 5 ppm NH3
in the internal cycle.
Sensors 2021, 21, x FOR PEER REVIEW 9 of 12

Sensors 2021, 21, 4545

To determine the possibility of detecting the mixture of TATP and HMTD, tests9were of 12
carried out with the DT-IMS detector. The tests were carried out under the same condi-
tions as with a DMS detector, with concentrations of TATP totaling 250 µg/m3 and 5 ppm
NH3 in the internal cycle.
The
The spectrum
spectrum showed
showed (Figure
(Figure 11)
11) the
the RIP
RIP peak
peak with
withNH NH33 and
and the
the most
most insensitive
insensitive
TATP
TATPmonomer
monomerpeak.peak. The
The DADP
DADP peak
peak was
was invisible
invisible (it
(it appeared
appeared on on the
the leading
leading edge
edge of
of
the
the TATP
TATPmonomer
monomer peak). One
peak). dimer
One peak
dimer waswas
peak observed
observedas a mixture of TATP
as a mixture of and
TATPDADP
and
ions.
DADP In ions.
the case of recording
In the the spectrum
case of recording for DADP,for
the spectrum peak
DADP,separation was not observed,
peak separation was not
and only the shift
observed, and only of the shift of the peak from RIP + DADP ions towardstime
peak from RIP + DADP ions towards increasing values from
increasing time
9.3 ms to
values a maximum
from 9.3 ms toof a 9.7 ms was of
maximum observed. Table
9.7 ms was 2 shows Table
observed. drift times for the
2 shows mixture
drift of
times for
TATP and DADP
the mixture for aand
of TATP concentration 250 µg/m3 . of 250 µg/m3.
DADP for aofconcentration

Figure 11.
Figure 11. Drift time
time spectrum
spectrum of
of TATP.
TATP.

Table2.2.Drift
Table Drifttime
timefor
forTATP
TATPand
andDADP
DADPions.
ions.

PositivePositive
Ion Ion Time (ms)
Time (ms)
RIP (NH3) 9.3
RIP (NH3 ) 9.3
DADP
DADP monomer monomer
ion ion 9.7 9.7
TATP monomer
TATP monomer ion ion 11.2 11.2
TATP +TATP
DADP+dimer
DADP ion
dimer ion 13.2 13.2

In the
In the measurement
measurement system
system shown
shown in
in Figure
Figure 2,
2, no
no recorded
recorded peaks
peaks from
from HMTD
HMTD were
were
observed. After removing the inner membrane, the spectrum showed a peak with
observed. After removing the inner membrane, the spectrum showed a peak with reduced reduced
mobility of
mobility of approximately
approximately1.51.5cm
cm22/Vs.
/Vs. HMTD
HMTD detection
detection using
using DT-IMS was possible
DT-IMS was possible only
only
under laboratory conditions (for dried air).
under laboratory conditions (for dried air).

4.
4. Discussion
The
The acetone residues
residues reduced
reducedthe thedetection
detectioncapability
capabilityofofthethe DMS
DMS detector.
detector. TheThePA
PA for acetone is 812 kJ/mol [38], which is lower than the PA of the
for acetone is 812 kJ/mol [38], which is lower than the PA of the added ammonia (853.6 added ammonia
(853.6
kJ/molkJ/mol [39]); therefore,
[39]); therefore, the use the use of ammonium
of ammonium as an admixture
as an admixture in thegas
in the carrier carrier gas
will help
will help to
to reduce reduce
the the appearance
appearance of acetoneofpeaks.
acetone
Whenpeaks. When improvised
detecting detecting improvised
explosivesexplo-
using
sives using differential
differential ion mobilityion mobility spectrometry,
spectrometry, the carrier the
gas carrier
shouldgas be should be doped
doped with with a
a minimum
minimum of 200 ppb NH to reduce the effect of acetone (TATP contamination).
of 200 ppb NH3 to reduce the effect of acetone (TATP contamination). During the analysis
3 During
the analysis
of TATP of TATP DADP
containing containing
as anDADP as anan
impurity, impurity,
increasean inincrease in the amplitude
the amplitude of the
of the monomer
monomer was observed, the peak of which was at the same location
was observed, the peak of which was at the same location as that of the TATP monomer, as that of the TATP
monomer, without a significant increase in the dimer amplitude. This may indicate a lack
of separation of the dimer for DADP.
In the case of TATP measurements with the DT-IMS detector, the measurement system
used a carrier gas with NH3 content at the level of several ppm. The appearance of a peak
close to the reaction ions (NH3 ) was observed in both the TATP and DADP studies.
Sensors 2021, 21, 4545 10 of 12

Differential ion mobility spectrometry was more effective than DT-IMS. For DT-IMS,
the peaks derived from TATP and DADP were located close to the peaks of the reaction
ions; in the case of DMS, the peaks were spaced apart and it was possible to separate them
more efficiently. Moreover, for the gas system with the membrane, no HMTD peaks were
observed in the case of IMS. Furthermore, it is impossible to detect HMTD in generally
used DT-IMS systems with membranes. In this case, there was no peak observed from
the analyte.
In the DMS studies, it was observed that the peaks moved towards higher compen-
satory voltages in membrane-free mode, due to a humidity effect. When adjusting the
effect of humidity on the location of peaks, the identification of HMTD and TATP was
maintained at the same level—no shift of adjusted peaks was observed.
The response time for instruments with membranes for TATP is a few seconds, while
that for HMTD is from several dozen seconds to two minutes. The response time for HMTD
results is the time required for membrane permeation, whereas for a membrane-free system,
TATP and HMTD are detected in seconds.

5. Conclusions
Detection of triacetone trioxide and hexamethylene triperoxide diamine can provide
early warning of a terrorist threat. DMS is currently one of the leading new technologies
for the separation and detection of chemicals in the gas phase.
When we used the membrane-free mode, the differential ion mobility spectrometry
detected TATP and HMTD correctly.
The use of ammonia as the dopant carrier gas in system with and without membrane
allows for the detection of inaccurately purified TATP containing the by-product DADP
and residual acetone. The use of ammonia in the carrier gas as a dopant helps in TATP
analysis for a membrane or membrane-free system.
Due to the differences resulting from the vapor pressure, it is necessary to use both
a membrane and membrane-free system at the same time. In the case of the membrane-
free mode, it is necessary to apply a peak position correction due to the variable value
of wetness.
Regarding laboratory analysis, the detection times of TATP and HMTD are short and
allow for quick scanning tests. Such a system allows the detection of HMTD at a level
lower than that of the gas chromatography used in WIChiR.
This study is the first to report detection of TATP and HMTD by differential ion
mobility spectrometry.

Author Contributions: Conceptualization, M.M.; data curation, M.S. and A.S.; formal analysis, M.M.,
M.S. and A.S.; funding acquisition, M.C.; investigation, M.M.; methodology, M.M. and M.C.; project
administration, M.M. and M.C.; validation, M.M. and M.W.-S.; visualization, M.S. and M.W.-S.;
writing—original draft, M.M. and M.S.; writing—review and editing, M.M. and M.W.-S. All authors
have read and agreed to the published version of the manuscript.
Funding: The work was carried out as part of a project financed by the National Centre for Research
and Development as part of the program for Programmers and Projects Defence, Security No. 9/2018,
project No. DOB-BIO9/28/01/2018.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
Sensors 2021, 21, 4545 11 of 12

References
1. Oxley, J.C.; Smith, J.L.; Brady, J.; Naik, S. Determination of urea nitrate and guanidine nitrate vapor pressures by isothermal
thermogravimetry. Propellants Explos. Pyrotech. 2010, 35, 278–283. [CrossRef]
2. Damour, P.; Freedman, A.; Wormhoudt, J. Knudsen effusion measurement of organic peroxide vapor pressures. Propellants Explos.
Pyrotech. 2010, 35, 514–520. [CrossRef]
3. Menzies, A.W.C. The explanation of an apparent anomaly outstanding in the results of measurement of dissociation pressures. J.
Am. Chem. Soc. 1920, 42, 1951–1956. [CrossRef]
4. Oxley, J.C.; Smith, J.L.; Luo, W.; Brady, J. Determining the vapor pressures of diacetone diperoxide (DADP) and hexamethylene
triperoxide diamine (HMTD). Propellants Explos. Pyrotech. 2009, 34, 539–543. [CrossRef]
5. Oxley, J.C.; Smith, J.L.; Shinde, K.; Moran, J. Determination of the vapor density of triacetone triperoxide (TATP) using a gas
chromatography headspace technique. Propellants Explos. Pyrotech. 2005, 30, 127–130. [CrossRef]
6. Takada, Y.; Nagano, H.; Suzuki, Y.; Sugiyama, M.; Nakajima, E.; Hashimoto, Y.; Sakairi, M. High-throughput walk through
detection portal for counter terrorism: Detection of triacetone triperoxide (TATP) vapor by atmospheric-pressure chemical
ionization ion trap mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25, 2448–2452. [CrossRef] [PubMed]
7. Giordano, C.B.; Lubrano, A.L.; Field, C.R.; Collins, G.E. Dynamic headspace generation and quantitation of triacetone triperoxide
vapor. J. Chromatogr. A 2014, 1331, 38–43. [CrossRef] [PubMed]
8. Mbah, J.; Knott, D.; Steward, S. Thermogravimetric study of vapor pressure of TATP synthesized without recrystallization. Talanta
2014, 129, 586–593. [CrossRef]
9. Egorshev, V.Y.; Sinditskii, V.P.; Smirnov, S.P. A comparative study on two explosive acetone peroxides. Thermochim. Acta 2013,
574, 154–161. [CrossRef]
10. DeGreeff, L.E.; Cerreta, M.M.; Katilie, C.J. Variation in the headspace of bulk hexamethylene triperoxide diamine (HMTD) with
time; environment; and formulation. Forensic Chem. 2017, 4, 41–50. [CrossRef]
11. Aernecke, M.J.; Mendum, T.; Geurtsen, G.; Ostrinskaya, A.; Kunz, R.R. Vapor pressure of hexamethylene triperoxide diamine
(HMTD) estimated using secondary electrospray ionization mass spectrometry. J. Phys. Chem. A 2015, 119, 11514–11522.
[CrossRef]
12. Ewing, R.G.; Waltman, M.J.; Atkinson, D.A.; Grate, J.W.; Hotchkiss, P.J. The vapor pressure of explosives. Trends Anal. Chem. 2013,
42, 35–48. [CrossRef]
13. Schulte-Ladbeck, R.; Vogel, M.; Karst, U. Recent methods for the determination of peroxide-based explosives. Anal. Bioanal. Chem.
2006, 386, 559–565. [CrossRef]
14. Vodochodský, O.; Jalový, Z.; Matyáš, R.; Novotná, M. Determination of triacetone triperoxide and hexamethylene triperoxide
diamine in various matrices using infrared spectroscopy. Appl. Spectrosc. 2019, 73, 195–202. [CrossRef] [PubMed]
15. Xu, X.; Craats, A.M.; Kok, E.M.; Bruyn, P.C. Trace analysis of peroxide explosives by high performance liquid chromatography-
atmospheric pressure chemical ionization-tandem mass spectrometry (HPLC-APCI-MS/MS) for forensic applications. J. Forensic
Sci. 2004, 49, 1230–1236. [CrossRef] [PubMed]
16. Schulte-Ladbeck, R.; Kolla, P.; Karst, U. Trace analysis of peroxide-based explosives. Anal. Chem. 2003, 75, 731–735. [CrossRef]
[PubMed]
17. Dunn, L.; Obaidly, H.S.A.A.; Khalil, S.E. Development and validation of fast liquid chromatography high-resolution mass
spectrometric (LC-APCI-Q ToF-MS) methods for the analysis of hexamethylene triperoxide diamine (HMTD)and triacetone
triperoxide (TATP). Forensic Chem. 2018, 10, 5–14. [CrossRef]
18. Buttigieg, G.A.; Knight, A.K.; Denson, S.; Pommier, C.; Denton, M.B. Characterization of the explosive triacetone triperoxide and
detection by ion mobility spectrometry. Forensic Sci. Int. 2003, 135, 53–59. [CrossRef]
19. Rasanen, R.M.; Nousiainen, M.; Perakorpi, K.; Sillanpaa, M.; Polari, L.; Anttalainen, O.; Utriainen, M. Determination of gas phase
triacetone triperoxide with aspiration ion mobility spectrometry and gas chromatography–mass spectrometry. Anal. Chim. Acta
2008, 623, 59–65. [CrossRef] [PubMed]
20. Marr, A.J.; Groves, D.M. Ion mobility spectrometry of peroxide explosives TATP and HMTD. Int. J. Ion Mobil. Spectrom. 2003, 6,
59–61.
21. Oxley, J.C.; Smith, J.L.; Kirschenbaum, L.J.; Marimganti, S.; Vadlamannati, S. Detection of explosives in hair using ion mobility
spectrometry. J. Forensic Sci. 2008, 53, 690–693. [CrossRef] [PubMed]
22. Borsdorf, H.; Eiceman, G.A. Ion mobility spectrometry: Principles and applications. Appl. Spectrosc. Rev. 2006, 41, 323–375.
[CrossRef]
23. Stach, J.; Baumbach, J.I. Ion mobility spectrometry—basic elements and applications. Int. J. Ion Mobil. Spectrom. 2002, 5, 1–21.
24. Eiceman, G.A. Ion-mobility spectrometry as a fast monitor of chemical composition. Trends Anal. Chem. 2002, 21, 259–275.
[CrossRef]
25. Vautz, W.; Franzke, J.; Zampolli, S.; Elmi, I.; Liedtke, S. On the potential of ion mobility spectrometry coupled to GC pre-
separation–A tutorial. Anal. Chim. Acta 2018, 1024, 52–64. [CrossRef] [PubMed]
26. Sorribes-Soriano, A.; de la Guardia, M.; Esteve-Turrillas, F.A.; Armenta, S. Trace analysis by ion mobility spectrometry: From
conventional to smart sample preconcentration methods. A review. Anal. Chim. Acta 2018, 1026, 37–50. [CrossRef]
27. Kaur-Atwal, G.; O’Connor, G.; Aksenov, A.A.; Bocos-Bintintan, V.; Paul Thomas, C.L.; Creaser, C.S. Chemical standards for ion
mobility spectrometry: A review. Int. J. Mass Spectrom. 2009, 12, 1–14. [CrossRef]
Sensors 2021, 21, 4545 12 of 12

28. Kolakowski, B.M.; Mester, Z. Review of applications of high-field asymmetric waveform ion mobility spectrometry (FAIMS) and
differential mobility spectrometry (DMS). Analyst 2007, 132, 842–864. [CrossRef]
29. Jakubowska, M.; Maziejuk, M.; Ceremuga, M.; Siczek, J.; Gallewicz, W. Ceramic DMS—type detector. Int. J. Ion Mobil. Spectrom.
2012, 15, 99–108. [CrossRef]
30. Pathak, P.; Shvartsburg, A.A. Low-Field Differential Ion Mobility Spectrometry of Dipole-Aligned Macromolecules. Anal. Chem.
2020, 92, 13855–13863. [CrossRef]
31. Winter, D.L.; Wilkins, M.R.; Donald, W.A. Differential ion mobility–mass spectrometry for detailed analysis of the proteome.
Trends Biotechnol. 2019, 37, 198–213. [CrossRef] [PubMed]
32. Maziejuk, M.; Szczurek, A.; Maciejewska, M.; Pietrucha, T.; Szyposzyńska, M. Determination of benzene; toluene and xylene
concentration in humid air using differential ion mobility spectrometry and partial least squares regression. Talanta 2016, 152,
137–146. [CrossRef] [PubMed]
33. Szczurek, A.; Maziejuk, M.; Maciejewska, M.; Pietrucha, T.; Sikora, T. BTX compounds recognition in humid air using differential
ion mobility spectrometry combined with a classifier. Sens. Actuators B Chem. 2017, 240, 1237–1244. [CrossRef]
34. Krylov, E.V.; Coy, S.L.; Nazarov, E.G. Temperature effects in differential mobility spectrometry. Int. J. Mass Spectrom. 2009, 279,
119–125. [CrossRef]
35. Guevremont, R. High-field asymmetric waveform ion mobility spectrometry: A new tool for mass spectrometry. J. Chromatogr. A
2004, 1058, 3–19. [CrossRef]
36. Miller, R.A.; Nazarov, E.G.; Eiceman, G.A.; King, A.T. A MEMS radio-frequency ion mobility spectrometer for chemical vapour
detection. Sens. Actuator A Phys. 2001, 91, 307–318. [CrossRef]
37. Shen, C.; Li, J.-Q.; Xu, G.; Wang, H.-M.; Zheng, P.-C.; Li, H.; Wang, Y.-J.; Chu, Y.-N. Proton Transfer Reaction Mass Spectrometry
of Triacetone Triperoxide. Chem. J. Chin. Univ. 2009, 30, 274–278.
38. Wróblewski, T.; Ziemczonek, L.; Szerement, K.; Karwasz, G. Proton affinities of simple organic compounds. Czechoslov. J. Phys.
2006, 56, B1110–B1115. [CrossRef]
39. Canaval, E.; Hyttinen, N.; Schmidbauer, B.; Fischer, L.; Hansel, A. NH4+ Association and Proton Transfer Reactions with a Series
of Organic Molecules. Front. Chem. 2019, 7, 191. [CrossRef]