B American Society for Mass Spectrometry, 2018 J. Am. Soc. Mass Spectrom. (2019) 30:501Y508DOI: 10.1007/s13361-018-2108-6

RESEARCH ARTICLE

Ammonia-Assisted Proton Transfer Reaction MassSpectrometry for Detecting Triacetone Triperoxide (TATP)Explosive

Qiangling Zhang,1,2 Xue Zou,1 Qu Liang,1,2 Hongmei Wang,3 Chaoqun Huang,1

Chengyin Shen,1,4 Yannan Chu1

1Anhui Province Key Laboratory of Medical Physics and Technology, Center of Medical Physics and Technology, Hefei Institutesof Physical Science, Chinese Academy of Sciences, Hefei, 230031, China2University of Science and Technology of China, Hefei, 230026, China3Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei, 230031, China4Hefei Cancer Hospital, Chinese Academy of Sciences, Hefei, 230031, China

Abstract. Proton transfer reaction mass spec-trometry (PTR-MS) usually detects different typesof compounds by changing the discharge gas toproduce different reagent ions in the ion source.In the present work, a novel method of changingreagent ions, ammonia-assisted PTR-MS, wasdeveloped. Through an injection port bypass, am-monia was injected into a homemade PTR-MSdevice. A conventional PTR-MS apparatus withreagent ions H3O

+(H2O)n (n = 0, 1, 2) can beconverted to an ammonia-assisted PTR-MS with reagent ions NH4

+.The new method was introduced to detecttriacetone triperoxide (TATP) explosive material. Results showed that the sensitivity is enhanced more than 37times compared with TATP detection using conventional PTR-MS and the limit of detection (LOD) could reach1.3 ppb. TATP in real complex matrixes can also be detected successfully using this method. Compared toconventional PTR-MS, ammonia-assisted PTR-MS has better sensitivity and better LOD for TATP detection, andthe technique provides common users with a convenient and quick method to change reagent ions. The users ofPTR-MS can easily obtain other reagent ions by injecting different assisted gases into an injection port to meetdifferent detection needs.Keywords: Conventional PTR-MS, Ammonia-assisted PTR-MS, Assisted gases, TATP

Received: 20 September 2018/Revised: 4 November 2018/Accepted: 4 November 2018/Published Online: 3 December 2018

Introduction

Triacetone triperoxide (TATP) can be easily synthesized, soit is often used in terrorist activities such as the July 2005

London transport bombings, the November 2015 Paris bomb-

ing, and the March 2016 Brussels airport and subway stationbombings. Therefore, it is necessary to develop a real-time andonline technology for detecting TATP.

In recent years, a series of methods for detecting TATP havebeen developed, such as gas chromatography mass spectrom-etry (GC-MS) [1], liquid chromatography mass spectrometry(LC-MS) [2], and liquid chromatography method with post-column ultraviolet irradiation and fluorescence detection [3].Unfortunately, these methods need sample pretreatment andchromatographic separation, which usually takes dozens ofminutes, so they are only suitable for laboratory analysis ofTATP. Ion mobility spectrometry (IMS) [4–6] can be used for

Electronic supplementary material The online version of this article (https://doi.org/10.1007/s13361-018-2108-6) contains supplementary material, whichis available to authorized users.

Correspondence to: Chengyin Shen; e-mail: chyshen@aiofm.ac.cn, YannanChu; e-mail: ychu@aiofm.ac.cn

on-site detection, but it may misreport because of limitedresolution. The electrochemical sensor method [7] needs sonicprocessing of tens of minutes for the electrodes in order toremove contaminants from electrode surfaces. The N,N-dimethyl-p-phenylenediamine (DMPD) sensor method [8]needs sample acid treatment. None of these methods can beused for reliable online detection of TATP.

Proton transfer reaction mass spectrometry (PTR-MS) isan online detection technology with fast response time anda good limit of detection (LOD) [9–11]. Detailed descrip-tions of the PTR-MS technology can be found elsewhere[12–15]. In recent years, PTR-MS has been widely appliedin the field of public safety, such as the detection ofexplosives [16–19] and illicit and controlled prescriptiondrugs [20]. Our research team had pioneered the study ofdetecting TATP using PTR-MS. The results showed thatwhen the discharge gas in the ion source was changed fromwater vapor to ammonia under a low reduced-field condi-tion, the sensitivity for detecting TATP could be improvedeffectively [21]. However, for a common user of PTR-MS,it is inconvenient to switch the discharge gas betweenwater vapor and ammonia. Furthermore, the voltage con-ditions of ion source need to be changed accordingly.Nazarov and Satoshi Inomata respectively introducedmethods of using dopants in IMS and PTR-MS systemsto get some new reagent ions [22, 23], inspiring the ideathat a dopant with ammonia in PTR-MS may be a goodchoice for TATP detection based on our previous study[21]. This work developed ammonia-assisted PTR-MS.Through an injection port bypass, ammonia was injectedinto a homemade PTR-MS device instead of changing thedischarge gas. The drift tube voltage and ammonia flowwere optimized respectively. The characteristic mass spec-tra for TATP detection were gained under optimized con-ditions. Further detection research was carried out on sam-ples of TATP mixed with white sugar or salt.

ExperimentalExperimental Device

The TATP detection was performed on a homemade PTR-MS device (Figure 1). Briefly, the PTR-MS instrumentconsists of an ion source, drift tube, and ion detectionsystem. The reagent ions H3O

+(H2O)n (n = 0, 1, 2) weregenerated in the ion source by glow discharge on the watervapor and then were injected into the drift tube. The flowof water vapor was maintained at 120 mL h−1 during theexperiment. Driven by air through the injection port, thesample gas TATP vapor, including the assisted gas ammo-nia, was introduced into the drift tube. At this time, theammonia underwent a proton transfer reaction withH3O

+(H2O)n (n = 0, 1, 2) (Eq. (1)) [24] because the protonaffinity (PA) of ammonia (PA = 850.6 kJ/Mol) is higherthan that of H2O (PA = 691.0 kJ/mol) [25] and its cluster

ions [26]. In addition, the TATP underwent a reaction withNH4

+ (Eq. (2)).

NH3 þ H3Oþ H2Oð Þn→NH4

þ þ H2Oð Þnþ1 n ¼ 0; 1; 2ð Þ ð1Þ

TATPþ NH4þ→ TATPþ NH4½ �þ ð2Þ

After these series of reactions in the drift tube, the productions first passed through a differentially pumping intermediatechamber, in which they were focused by ion lens before enter-ing the ion detection system. The ion detection system, whichis mainly made up of a quadrupole mass spectrometer, detectedthe m/z and number of ions and finally obtained the massspectrum. During this experiment, the data sampling rate was1 Hz and averaged 120 times for one mass spectrum. Thetemperature of the drift tube was maintained at 60(± 1) °Cand the pressure of the drift tube was 0.7 Torr.

Experimental Process

In the drift tube, the product ions generated from the reaction ofreagent ions H3O

+(H2O)n (n = 0, 1, 2) or NH4+ with the sam-

ples will be affected by the ionic collision energy. When thecollision energy is too high, many fragment ion peaks will beformed. When the collision energy is too low, many cluster ionpeaks will be formed. Both of these conditions make theanalysis of mass spectra complicated. The ionic collision ener-gy in the drift tube is closely related to the voltage of the drifttube. To test the proper drift tube voltage condition for TATPdetection, the voltage was optimized. The assisted gas ammo-nia was used to change the reagent ion, so the flow of ammoniawas also optimized. The ammonia came from the saturatedsteam of ammonia water, and an injection pump controlled itsflow.

Under the optimized conditions, the characteristic massspectra of TATP detection were obtained under the conditionof conventional PTR-MS and ammonia-assisted PTR-MS. Thevolume specific concentration C of TATP was calculated byEq. (3). The equation was derived from the ideal gas equation.

C ¼ PS

P

FS

FTð3Þ

In this equation, Ps (5.25 × 10−2Torr) is the saturated vaporpressure of TATP at room temperature [27], and P (760 Torr) isthe environmental pressure of the laboratory. Fs is the saturatedvapor flow of TATP, which was controlled by an injectionpump, and FT (244.2 mL h−1) is the total flow through theinjection port, which was determined by the instrument.

To test the applicability of ammonia-assisted PTR-MS de-tecting TATP in real complex matrixes, the method was ap-plied to detect TATP mixed with sugar or salt under theoptimized conditions.

502 Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP

TATP Synthesis and Reagents

The small amount of TATP used in the experiment was syn-thesized according to a method described in the previousliterature [28]. The purity of the TATP sample was analyzedusing gas chromatography mass spectrometry (GC-MS)(Figure S1). The results showed that there was only a littleresidual acetone in the TATP sample. This is because, forsafety reasons, the synthetic TATP sample was not dried. Theammonia solution (25~28% NH3) was purchased fromSinopharm Chemical Reagent Co., Ltd. (Shanghai, China,http://www.sinoreagent.com). The sugar and salt used in theexperiment were purchased from a nearby grocery store.

Results and DiscussionOptimization of the Drift Tube Voltageand Ammonia Flow

When the injection port bypass ammonia flow was closed inconventional PTR-MS, 7.58 ppm TATP steam wasinjected into the injection port. The ion [TATP+H]+ at m/z223 was chosen as the characteristic ion. The drift tube voltagewas regulated from 50 to 240 V with 10 V increments. Theintensity variation of [TATP+H]+ with the drift tube voltage isshown in Figure 2a. With the increase of the drift tube voltage,the [TATP+H]+ ionic signal intensity first increased and thendecreased, reaching its maximum at 140 V. The normal drifttube voltage was 358 V for this homemade PTR-MS apparatus.This phenomenon indicates that the drift tube voltage neededfor TATP detection is lower than that of usual detection. Inother words, the [TATP+H]+ ions fragment easily.

When the injection port bypass ammonia flowwas 1 mL h−1

in ammonia-assisted PTR-MS, 7.58 ppm TATP steam wasinjected into the injection port. The ion [TATP+NH4]

+ at m/z240 was chosen as the characteristic ion. The drift tube voltagewas regulated from 50 to 330 V with 10 V increments. Theintensity variation of [TATP+NH4]

+ with the drift tube voltageis shown in Figure 2b. With the increase of drift tube voltage,the [TATP+NH4]

+ ionic signal intensity first increased and

then decreased, reaching its maximum at 230 V. Comparedwith Figure 2a, the optimized drift tube voltage is higher, whichshows that [TATP+NH4]

+ is more stable than [TATP+H]+.Also, the [TATP+NH4]

+ ionic signal intensity was higher thanthat of [TATP+H]+ for the same concentration of TATP steam.This phenomenon indicates that ammonia-assisted PTR-MS ismore sensitive to TATP than conventional PTR-MS.

When the drift tube voltage was set to 230 V in ammonia-assisted PTR-MS, 7.58 ppm TATP steam was injected into theinjection port. The ammonia flow optimization was carried outby changing the ammonia flow. The result is shown inFigure 3. One can see that the intensity of the characteristicions [TATP+NH4]

+ gradually leveled off after the initial ramp-up, and the optimum ammonia flow is 9 mL h−1. The reason forthis trend should be as follows. With the increase of ammoniaflow, more reagent ions H3O

+(H2O)n (n = 0, 1, 2) should beconverted to NH4

+, so the ion [TATP+NH4]+ intensity will

increase. When the ammonia flow is large enough,

Figure 1. Schematic diagram of PTR-MS apparatus. GD glow discharge, SD source drift region, IC intermediate chamber, EMelectron multiplier

Figure 2. Dependence of the characteristic ionic intensity onthe drift tube voltage using (a) conventional PTR-MS for detect-ing TATP and (b) ammonia-assisted PTR-MS for detectingTATP

Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP 503

H3O+(H2O)n (n = 0, 1, 2) should be completely converted to

NH4+. At this time, the increase of ammonia flow will not

increase the intensity of NH4+, so the intensity of

[TATP+NH4]+ will tend to level off.

To test the above speculation, the mass spectra for thelaboratory air were measured at 230 V drift tube voltage; theresults are shown in Figure 4. Figure 4a is the mass spectrumfor the conventional PTR-MS. The peaks of the reagent ionsappear at m/z 19, 37, and 55, and they can be attributed toH3O

+(H2O)n (n = 0, 1, 2). Figure 4b is the mass spectrum forammonia-assisted PTR-MS with an ammonia flow of 9 mL h−1. The peaks of the reagent ions appear at m/z 18, 35, and 36and can be attributed respectively to NH4

+, NH4+NH3, and

NH4+H2O. One can see that almost all the H3O

+ and its clusterions can be transformed to NH4

+ and its cluster ions. And thisphenomenon confirms the above speculation.

Contrast of Sensitivity and LOD of TATP DetectionUsing Conventional PTR-MSand Ammonia-Assisted PTR-MS

With the drift tube voltage set to 140 V in conventionalPTR-MS, the concentration of TATP steam was prepared as7.58, 17.5, 26.2, 40.8, and 53.6 ppm. The dependence of theintensity of [TATP+H]+ at m/z 223 and the concentration ofTATP steam are shown in Figure 5a. When the drift tubevoltage was set to 230 V and the ammonia flow was 9 mL h−1 in ammonia-assisted PTR-MS, the concentration ofTATP steam was the same as above. The dependence ofthe intensity of [TATP+NH4]

+ at m/z 240 and the concen-tration of TATP steam is shown in Figure 5a. The ratio ofcharacteristic ion [TATP+NH4]

+ intensity to [TATP+H]+

intensity is shown by numbers on arrows in Figure 5a.The background signal intensity at m/z 223 and 240 wasless than 5 cps (counts per second), which is far less thanthe ion intensity of TATP seen in Figure 5a. Therefore, theratio of characteristic ion intensity is approximately equal tothe ratio of sensitivity. From Figure 5a, one can see that thesensitivity of TATP detection is enhanced more than 37times by ammonia-assisted PTR-MS.

To investigate the LOD of TATP detection, 7.58 ppmTATP steam was detected using conventional PTR-MS andammonia-assisted PTR-MS. The monitor graphs of character-istic ions are shown in Figure 5b and c. After a signal-to-noiseratio of 2:1 is considered, the LOD can be calculated respec-tively using the results in Figure 5b, c [29]. The LOD of TATPdetection was 33.6 ppb for conventional PTR-MS and was1.3 ppb for ammonia-assisted PTR-MS. The LOD of TATPfor ammonia-assisted PTR-MS is better even than that ofammonia discharge PTR-MS due to the higher reagent ionsin this PTR-MS instrument [21].

Analysis of Typical Mass Spectra for DetectingTATP

When the drift tube voltage was set to 140 V in conventionalPTR-MS, the typical mass spectrum was measured for the airinside the laboratory; the result is shown in Figure 6a. Toreduce the damage of the electronic multiplier, the water clusterions at m/z 55 and 73 were skipped in this scan. In Figure 6a,some obvious peaks at m/z 36, 37, 57, 75, and 91 can be seen.These peaks can be attributed respectively to NH4

+H2O,H3O

+H2O, and O18 isotopic peaks of H3O+(H2O)2, and to

O18 isotopic peaks of H3O+(H2O)3 and H3O

+(H2O)4. Thereagent ions are all in the form of water clusters because ofthe low drift tube voltage.

When the drift tube voltage was set to 140 V in conventionalPTR-MS and the concentration of TATP steam was 40.8 ppm,the mass spectrum of TATP was measured; Figure 6b showsthe results. In contrast to Figure 6a, the new rising ionic peaksappeared atm/z 43, 59, 74, 75, 77, 79, 91, 92, 93, 97, 109, 117,and 223. The ion at m/z 223 is the characteristic ion[TATP+H]+. Methane positive ion chemical ionization massspectrometry (PICI-MS) [30] and PTR-MS [21] used for

Figure 3. Dependence of the [TATP+NH4]+ intensity on the

ammonia flow F

Figure 4. Mass spectra of laboratory air measured by (a) con-ventional PTR-MS and (b) ammonia-assisted PTR-MS

504 Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP

Figure 5. (a) Dependence of characteristic ions intensity on TATP concentration using conventional PTR-MS and ammonia-assisted PTR-MS; monitor graphs of characteristic ions using (b) conventional PTR-MS and (c) ammonia-assisted PTR-MS

Figure 6. Conventional PTR-MS for detecting (a) air and (b) TATP; ammonia-assisted PTR-MS for detecting (c) air and (d) TATP

Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP 505

detecting TATP all detected the ions at m/z 43, 59, 74, 75, and91; these are fragment ions from [TATP+H]+ and have beenassigned as CH3CO

+, C3H7O+, C3H6O2

+, C3H7O2+, and

C3H7O3+ respectively [30]. The ions at m/z 75 and 91 also

have a minor contribution from water clusters. To investigatethe influence of residual acetone, acetone was added to theinjection port at this voltage. The rising ion peaks were found atm/z 59, 77, and 117, which should be protonated acetone, aclustering ion of m/z 59 with a water molecule, and dimer ofacetone (Figure S2a). This means that the ions at m/z 59 havepartial contributions from residual acetone and the ions at m/z77 and 117 are also the result of residual acetone. The ions atm/z 79 and 97 may be the clustering ions of m/z 43 with twowater molecules C2H3O

+(H2O)2 and ofm/z 43 with three watermolecules C2H3O

+(H2O)3. As for the ions at m/z 92, 93, and109, they should be the clustering ions of m/z 74, 75, and 91with a water molecular and are assigned as C3H6O2

+ H2O,C3H7O2

+ H2O, and C3H7O3+ H2O.

When the drift tube voltage was set to 230 V and ammoniaflow was 9 mL h−1 in ammonia-assisted PTR-MS, the typicalmass spectrum measured for the air inside the laboratory is asshown in Figure 6c. The reagent ion is mainly NH4

+ at m/z 18(see Figure 4b), which was skipped in this scan because of itstoo strong intensity. Furthermore, some other obvious peaks atm/z 30, 35, and 36 can be seen; they may be NO+, NH4

+NH3,and NH4

+H2O.When the drift tube voltage was set to 230 V and

ammonia flow was 9 mL h−1 in ammonia-assisted PTR-MS, 40.8 ppm TATP steam was injected into the injectionport. The measured mass spectrum is shown in Figure 6d. Incontrast to Figure 6c, the rising ionic peaks appear at m/z43, 58, 73, 74, 75, 76, 89, 91, and 240. The ion at m/z 240is the characteristic ion [TATP+NH4]

+. The experimentalresults of TATP detection by PTR-MS [21] showed thatwhen ammonia was used as the discharge gas in the ion

source, the reagent ions changed from H3O+ to NH4

+ andNH4

+NH3, and the signal peak of characteristic ion[TATP+NH4]

+ could be obtained. Meanwhile, there weresome fragment ion peaks in the mass spectrum, such as ionsat m/z 43, 58, 74, 75, and 91. As for ions at m/z 58, Ewinget al. suggested that this peak was due to protonated 2-propanimine C3NH8

+ generated by the interaction between[TATP+NH4]

+ and NH3 [5]. In addition to the ions reportedin the literature [21], new ion peaks appeared at m/z 73, 76,and 89 in this paper. To investigate the influence of residualacetone, acetone was added to the injection port under thesame conditions of drift tube voltage and ammonia flow,and arising ion peak was found at m/z 76, which should beadduct ions [C3H6O+NH4]

+ (Figure S2b). Therefore, the ionpeak at m/z 76 is the result of residual acetone. As for ionpeaks at m/z 73 and 89, Marr et al. [31] used IMS to detectTATP; they found that when the temperature of the migra-tion tube was higher than 130 °C, the [TATP+NH4]

+ peakdisappeared and new ion peaks appeared at m/z 73 and 89.Therefore, the ions at m/z 73 and 89 may be thermaldecomposition fragment ions of [TATP+NH4]

+.

Analysis of Mass Spectra of TATP Mixed with RealComplex Matrixes

TATP is like white sugar or salt in appearance. To test theapplicability of ammonia-assisted PTR-MS for detectingTATP in real complex matrixes, 45 mg solid TATP, 45 mgsolid TATP mixed with 450 mg white sugar, and 45 mg solidTATP mixed with 450 mg salt were respectively placed in theinjection port. When the drift tube voltage was set to 230 V andammonia flow was 9 mL h−1 in ammonia-assisted PTR-MS,the measured mass spectra are as shown in Figure 7. One cansee that the peak positions and relative abundances of theproduct ions are basically the same as the result in Figure 6d

Figure 7. Mass spectra of (a) only TATP, (b) mixture of TATP and white sugar, and (c) mixture of TATP and salt, all detected usingammonia-assisted PTR-MS

506 Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP

except for ions atm/z 76. The ion peak atm/z 76 is the result ofresidual acetone (Figure S2b). A possible reason is as follows.The TATP measured in Figure 6d was saturated vapor, whichwas obtained from hours of sublimation of solid TATP in asealed glass bottle. In this situation, acetone steam originatingfrom residual acetone will accumulate in the TATP saturatedvapor, which increased the concentration of acetone in thatvapor. The TATP measured in Figure 7 was solid TATP, sothere was no cumulative process, which reduced the ion inten-sity at m/z 76. Figure 7 shows that even if TATP is mixed withsugar or salt at a mass ratio of 1:10, the characteristic ion[TATP+NH4]

+ can still be detected. This result shows thatammonia-assisted PTR-MS can be used to detect TATP in realcomplex matrixes.

ConclusionA novel method, ammonia-assisted PTR-MS, was developedto detect TATP explosive through changing the reagent ions.The drift tube voltage and the ammonia water headspace steamflowwere optimized. The optimal drift tube voltage was 230 V,and the optimal ammonia water headspace steam flow was9 mL h−1 for ammonia-assisted PTR-MS. The sensitivity wasenhanced more than 37 times compared with TATP detectionusing conventional PTR-MS, and the LOD reached 1.3 ppb.Ammonia-assisted PTR-MS can also be applied for detectingsamples of TATP mixed with sugar or salt at a mass ratio of1:10. Compared to the conventional PTR-MS, ammonia-assisted PTR-MS has better sensitivity and LOD for TATPdetection. Ammonia-assisted PTR-MS provides common usersof PTR-MS with a convenient and quick method to changereagent ions. In the same vein, the users of PTR-MS couldeasily obtain other reagent ions by injecting different assistedgases into the injection port to meet different detection needs.

AcknowledgementsThis work was supported by the National Key R&D Programof China (No. 2016YFC0200200), the National Natural Sci-ence Foundation of China (Nos. 21777163, 21477132,21577145), the Anhui Provincial Program for Science andTechnology Development, China (No. 1604d0802001), andthe Key Program of 13th Five-Year Plan, CASHIPS (No.KP-2017-25).

Publisher’s Note Springer Nature remains neutral with regard to jurisdictionalclaims in published maps and institutional affiliations.

References

1. Ezoe, R., Imasaka, T., Imasaka, T.: Determination of triacetonetriperoxide using ultraviolet femtosecond multiphoton ionization time-of-flight mass spectrometry. Anal. Chim. Acta. 853, 508–513 (2015)

2. Gamble, S.C., Campos, L.C., Morgan, R.M.: Detection of trace peroxideexplosives in environmental samples using solid phase extraction andliquid chromatography mass spectrometry. Environ. Forensic. 18, 50–61(2017)

3. Schulte-Ladbeck, R., Kolla, P., Karst, U.: Trace analysis of peroxide-based explosives. Anal. Chem. 75, 731–735 (2003)

4. Fan, W., Young, M., Canino, J., Smith, J., Oxley, J., Almirall, J.R.: Fastdetection of triacetone triperoxide (TATP) from headspace using planarsolid-phase microextraction (PSPME) coupled to an IMS detector. Anal.Bioanal. Chem. 403, 401–408 (2012)

5. Ewing, R.G., Waltman, M.J., Atkinson, D.A.: Characterization oftriacetone triperoxide by ion mobility spectrometry and mass spectrom-etry following atmospheric pressure chemical ionization. Anal. Chem. 83,4838–4844 (2011)

6. Kozole, J., Levine, L.A., Tomlinson-Phillips, J., Stairs, J.R.: Gas phaseion chemistry of an ion mobility spectrometry based explosive tracedetector elucidated by tandem mass spectrometry. Talanta. 140, 10–19(2015)

7. Mamo, S.K., Gonzalez-Rodriguez, J.: Development of a molecularlyimprinted polymer-based sensor for the electrochemical determinationof triacetone triperoxide (TATP). Sensors. 14, 23269–23282 (2014)

8. Can, Z.Y., Uzer, A., Turkekul, K., Ercag, E., Apak, R.: Determination oftriacetone triperoxide with a N,N-dimethyl-p-phenylenediamine sensoron nafion using Fe3O4 magnetic nanoparticles. Anal. Chem. 87, 9589–9594 (2015)

9. Ellis, A.M., Mayhew, C.A.: Proton transfer reaction mass spectrometry:principles and applications. Wiley, New York (2014)

10. Lindinger, W., Hansel, A., Jordan, A.: Proton-transfer-reaction massspectrometry (PTR-MS): on-line monitoring of volatile organic com-pounds at pptv levels. Chem. Soc. Rev. 27, 347–354 (1998)

11. Zou, X., Kang, M., Li, A.Y., Shen, C.Y., Chu, Y.N.: Spray inlet protontransfer reaction mass spectrometry (SI-PTR-MS) for rapid and sensitiveonline monitoring of benzene in water. Anal. Chem. 88, 3144–3148(2016)

12. Yuan, B., Koss, A.R., Warneke, C., Coggon, M., Sekimoto, K., de Gouw,J.A.: Proton-transfer-reaction mass spectrometry: applications in atmo-spheric sciences. Chem. Rev. 117, 13187–13229 (2017)

13. Zou, X., Lu, Y., Xia, L., Zhang, Y.T., Li, A.Y., Wang, H.M., Huang,C.Q., Shen, C.Y., Chu, Y.N.: Detection of volatile organic compounds ina drop of urine by ultrasonic nebulization extraction proton transferreaction mass spectrometry. Anal. Chem. 90, 2210–2215 (2018)

14. Warneke, C., Veres, P., Murphy, S.M., Soltis, J., Field, R.A., Graus,M.G., Koss, A., Li, S.M., Li, R., Yuan, B., Roberts, J.M., de Gouw,J.A.: PTR-QMS versus PTR-TOF comparison in a region with oil andnatural gas extraction industry in the Uintah Basin in 2013. Atmos. Meas.Tech. 8, 411–420 (2015)

15. Pan, Y., Zhang, Q.L., Zhou, W.Z., Zou, X., Wang, H.M., Huang, C.Q.,Shen, C.Y., Chu, Y.N.: Detection of ketones by a novel technology:dipolar proton transfer reaction mass spectrometry (DP-PTR-MS). J.Am. Soc. Mass Spectrom. 28, 873–879 (2017)

16. Sulzer, P., Petersson, F., Agarwal, B., Becker, K.H., Juerschik, S., Maerk,T.D., Perry, D., Watts, P., Mayhew, C.A.: Proton transfer reaction massspectrometry and the unambiguous real-time detection of 2,4,6 trinitro-toluene. Anal. Chem. 84, 4161–4166 (2012)

17. Gonzalez-Mendez, R., Reich, D.F., Mullock, S.J., Corlett, C.A.,Mayhew, C.A.: Development and use of a thermal desorption unit andproton transfer reaction mass spectrometry for trace explosive detection:determination of the instrumental limits of detection and an investigationof memory effects. Int. J. Mass Spectrom. 385, 13–18 (2015)

18. Gonzalez-Mendez, R., Watts, P., Olivenza-Leon, D., Reich, D.F.,Mullock, S.J., Corlett, C.A., Cairns, S., Hickey, P., Brookes, M.,Mayhew, C.A.: Enhancement of compound selectivity using a radiofrequency ion funnel proton transfer reaction mass spectrometer: im-proved specificity for explosive compounds. Anal. Chem. 88, 10624–10630 (2016)

19. Gonzalez-Mendez, R., Watts, P., Reich, D.F., Mullock, S.J., Cairns,S., Hickey, P., Brookes, M., Mayhew, C.A.: Use of rapid reducedelectric field switching to enhance compound specificity for protontransfer reaction-mass spectrometry. Anal. Chem. 90, 5664–5670(2018)

20. Agarwal, B., Petersson, F., Juerschik, S., Sulzer, P., Jordan, A., Maerk,T.D., Watts, P., Mayhew, C.A.: Use of proton transfer reaction time-of-flight mass spectrometry for the analytical detection of illicit and

Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP 507

controlled prescription drugs at room temperature via direct headspacesampling. Anal. Bioanal. Chem. 400, 2631–2639 (2011)

21. Shen, C.Y., Li, J.Q., Han, H.Y., Wang, H.M., Jiang, H.H., Chu, Y.N.:Triacetone triperoxide detection using low reduced-field proton transferreaction mass spectrometer. Int. J. Mass Spectrom. 285, 100–103 (2009)

22. Nazarov, E.G., Miller, R.A., Eiceman, G.A., Stone, J.A.: Miniaturedifferential mobility spectrometry using atmospheric pressure photoioni-zation. Anal. Chem. 78, 4553–4563 (2006)

23. Inomata, S., Tanimoto, H.: Differentiation of isomeric compounds bytwo-stage proton transfer reaction time-of-flight mass spectrometry. J.Am. Soc. Mass Spectrom. 19, 325–331 (2008)

24. Bohme, D.K., Mackay, G.I., Tanner, S.D.: An experimental study of thegas-phase kinetics of reactions with hydrated H3O+ ions (n = 1-3) at 298K. J. Am. Chem. Soc. 101, 3724–3730 (1979)

25. Szulejko, J.E., McMahon, T.B.: Progress toward an absolute gas-phaseproton affinity scale. J. Am. Chem. Soc. 115, 7839–7848 (1993)

26. Kawai, Y., Yamaguchi, S., Okada, Y., Takeuchi, K., Yamauchi, Y.,Ozawa, S., Nakai, H.: Reactions of protonated water clustersH+(H2O)(n) (n=1-6) with dimethylsulfoxide in a guided ion beam appa-ratus. Chem. Phys. Lett. 377, 69–73 (2003)

27. Oxley, J.C., Smith, J.L., Shinde, K., Moran, J.: Determination of thevapor density of triacetone triperoxide (TATP) using a gas chromatogra-phy headspace technique. Propellants Explos. Pyrotech. 30, 127–130(2005)

28. Dubnikova, F., Kosloff, R., Almog, J., Zeiri, Y., Boese, R., Itzhaky, H.,Alt, A., Keinan, E.: Decomposition of triacetone triperoxide is an entropicexplosion. J. Am. Chem. Soc. 127, 1146–1159 (2005)

29. Zhang, Q.L., Zou, X., Liang, Q., Zhang, Y.T., Yi, M.J., Wang, H.M.,Huang, C.Q., Shen, C.Y., Chu, Y.N.: Development of dipolar protontransfer reaction mass spectrometer for real-time monitoring of volatileorganic compounds in ambient air. Chin. J. Anal. Chem. 46, 471–478(2018)

30. Sigman, M.E., Clark, C.D., Fidler, R., Geiger, C.L., Clausen, C.A.:Analysis of triacetone triperoxide by gas chromatography/mass spectrom-etry and gas chromatography/tandem mass spectrometry by electron andchemical ionization. Rapid Commun. Mass Spectrom. 20, 2851–2857(2006)

31. Marr, A.J., Groves, D.M.: Ion mobility spectrometry of peroxide explo-sives TATP and HMTD. Int. J. Ion Mobil. Spectrom. 6, 62–65 (2003)

508 Q. Zhang et al.: A Novel Ammonia-Assisted PTR-MS for Detecting TATP