adventures in analysis of peroxide explosivesenergetics.chm.uri.edu/system/files/isade peroxide...

9/27/2017

1

Adventures in Analysis of Peroxide Explosives

HO OOH OOHHOO

OHOO O OOH

H3C

C

CH3

O O

O O

C

O

C

O

H3C

H3C

CH3

CH3

NCH2H2C H2C

OOO

OOO

CH2CH2H2C

N

TATP HMTDMethyl Ethyl Ketone Peroxides

Jimmie C. OxleyUniversity of Rhode [email protected]; 401-874-2103

PhD Students: Kevin Colizza Lindsay McLennan Alex Yeudakimau

Professors: Jimmie Oxley & Jim Smith

This work is supported by the U.S. Department of Homeland Security but views & conclusions are those of the authors alone.

Studies with Peroxide ExplosivesHair as forensic evidence

Safe Scent Canine Training Aids

Methods: Gentle Destruction of Peroxides

450 g TATP digested in 25 min

9/27/2017

2

TATP Vapor by direct syringe of headspace

Sample Mean (ng TATP/mL vapor) SD (+/- ng/mL vap)

1158 354 341202 388 541208 384 1.61212 379 64

Mean 376Standard Error 8

TATP volatility is a problem!

1. Draw 2.5 mL vapor into a gas-tight syringe2. Bubble gas through 0.5 mL of 50/50 ACN/water in

a glass GC vial with PTFE septa3. Perform this 3 times for each vessel4. Directly inject solutions onto LC-MS system (N=2)

~100 mg TATP in each container

1158 1202

1208 1212

Vapor concentration independent of container size. Vapor pressure predicted for 25°C is 1.8 Pa by rising temp TGA & 6.95 Pa by headspace GC

Oxley, Smith, Brady,, PEP, 2012, 37, 215-222

Pvap = 4.1 ± 0.1 Pa

Oxley, Smith, Moran, Shinde PEP, 2005, 30, 127‐130

0

2

4

6

8

10

12

0 10 20 30 40 50 60

Con

cen

trat

ion

(μg/

mL

)

Time (minutes)

Closed container

Opened container

100 μM TATP in 10mM potassium phosphate buffer incubated 37°C for 60 min in 1.5 mL Eppendorf snap-cap tube

HN

N

N

N

hexamine N(CH3)3 HN(CH3)2

dimethylformamide

methylformamide

ethylenimine

1-methyl-1H-1,2,4-triazole

NH

O

NH

O

N,N'-methylenebis(formamide)

TATP headspace

DimethylformamideEthylenimine

Methylformamide

trimethylamine

HMTD Headspace

To make safe scent aids, headspace signature must be determined

The headspace of TATP contains only TATP. The scent aid must be TATP

The headspace of HMTD contains mainly decomposition products. These non‐explosives make up the odor.

9/27/2017

3

TATP vs DADP FormationHO

C

OOH

H3C CH3

C

H3C CH3

O

+ H2OH2O2

HOO

C

OOH

H3C CH3

C

H3C CH3

O

+ +

1:1:1 acetone:HP:acid (8.6 mmol) rt

50% HP (g) water (mL)

96.5% sulfuric acid

(g) % TATP % DADP

0.59 0.25 0.88 100

0.59 0.50 0.88 7.6 92

0.59 0.75 0.88 81 19

0.59 1.0 0.88 100

+ H2O2H2O

High [acid] low [water]

+H2O

C

O

H3CCH3

O

C

CH3

OOH

H3C

C

H3C CH3

O

H O O H

H+

HO

C

OOH

H3C CH3

HO

C

O

H3CCH3

O

C

CH3

OH

H3C

HO

C

O

H3CCH3

O

C

CH3

OOH

H3C

HP

H3C

C

CH3

O O

O O

C

H3C CH3

HP

HOO

C

OOH

H3C CH3

HOC

OOH

H3C CH3

HOO

C

O

H3CCH3

O

C

CH3

OOH

H3C

HOO

C

O

H3CCH3

O

C

CH3

O

H3C

O

C

CH3

OH

H3C

H O O H

H+

OO

C

O

OC

O

O

C

CH3H3C

CH3

CH3

H3C

H3C

aceto

ne

+ H2O

acetone

+ H2O

+ H2O

HP

I

II

III

IV

V

DADP

TATP

HOC

OOH

H3C CH3

No acid required

Acid andhigh [water]

Acid


Hig

h a

cid

lo

w w

ate

r




Low [water]

HMTD has little vapor pressure; its odor is decomposition productsMay 2013 HMTD (250mg in glass vial) exploded as chemist picked it up. As explosives start to decompose, they become more sensitive because porosity increases,

HMTD already has lower thermal stability than any military explosive (DSCexo 200-300oC) & even most HME. Friction & ESD sensitivities are high.

& density decreases. Pores introduce heterogeneity which increases sensitivity.

dry HMTD

HMTD+2uLH2O

0

10

20

30

40

50

60

70

80

90

100

'1 week

Humidity 0

29%

74.50%

100%

Week 1 Week 2 Week 4

29

75

100

May 2013Jan 2015

171oC

142oC

9/27/2017

4

Mass Identity amount

1 73 L DRY & HUMID

2 75 L

MATCHED TO AUTHENTIC SAMPLE; MAINLY SEEN IN HUMID CONDITIONS

3 103 M


4 88 S

5 84, 102 L

MATCHED TO AUTHENTIC SAMPLE; DRY CONDITIONS

6 116 L DRY CONDITIONS

7 178 SBOTH IN DRY &

HUMID CONDITIONS

8 140 M

MATCHED TO AUTHENTIC

SAMPLE; MAINLY SEEN IN HUMID OR

ACIDIC CONDITIONS

9 208 L

10 143 S

11 171 L


12 157 SMAINLY SEEN IN DRY CONDITIONS

[M+H]+ Identity amount

1 74.06004 L C3H8ON

2 103.0501 L

C3H7O2N2 MATCHED TO AUTHENTIC

SAMPLE

3 106.0499 S C3H8O3N

4 117.0659 L C4H9O2N2

5 117.1022 L C5H13ON2

6 120.0768 S C3H10O2N3

7 133.0608 S C4H9O3N2

8 141.1131 L C6H13N4

9 144.0768 M C5H10O2N3

10 145.0608 M C5H9O3N2

11 155.1289 M C7H15N4

12 157.1083 L C6H13ON4

NH

N

O O

13 158.0923 L C6H12O2N3

14 160.0717 L C5H10O3N3

15 172.0712 L

C6H10O3N3 MATCHED TO AUTHENTIC

SAMPLE

16 172.1078 S C7H14O2N3

17 174.0873 M C6H12O3N3

18 174.1235 S C7H16O2N3

19 185.1032 S C7H13O2N4

20 201.0982 L C7H13O3N4

21 205.0931 S C6H13O4N4

22 207.0611 TMDDD M C6H11O6N2

23 209.0768 HMTD M C6H13O6N2

GC-MSLC-MS R1-A.1LC‐MS GC‐MS

HMTD Decomposition Products

C6N4H12 + 6 H2O2 + 6 OCH2 2 C6N2H12O6 + 6 H2O

C6N4H12 + 3 H2O2 1 C6N2H12O6 + 2 NH3

With added formaldhyde reaction is fasterYield 130% based on 1 HMTD : 1 hexamine;

70% based on 2 HMTD : 1 hexamine

9/27/2017

5

13C is incorporated in HMTD with added 13COH2 under formation conditions.

N O

O

O

O

O

O

N13COH2

209 210 211 212 213 214m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relative Abundance

211.0837

210.0804

212.0871

209.0771

213.0904

214.1553

209.0771C6H13O6N2

+

210.0804C5

13CH13O6 N2+

211.0837C4

13C2H13O6N2+

212.0871C3

13C3H13 O6N2+

213.0904C2

13C4H13O6N2+

HMTD

141.0 141.5 142.0 142.5 143.0 143.5m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relativ

e A

bundance

141.1136C 6 H13 N4

142.1169C 5 13C H13 N4

143.1202C 4 13C 2 H13 N4

141.2977140.9294

Hexamine

142.1169 C5

13CH13 N4+

Should be ~6.5 % relative to 12CIs about 20%

143.1202 C4

13C2H13 N4+

Should be ~0.2% relative to 12CIs about 1.6%

NN

N

N

141.1136 C6H13N4

+ Some 13C also appeared in remaining hexamine

N O

O

O

O

O

O

N

{ }

141 142 143 144 145 146 147m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bunda

nce

141.1135C 6 H13 N4

145.1016C 6 H13 15N4

142.1169C 3 H16 O3 N3

146.1049C 2 H15 O3 N 15N3144.0656

C 6 H10 O3 N143.0815

C 6 H11 O2 N2 145.3487141.3505 144.8547140.8770

H2O2 + citric acid{ }

202 204 206 208 210 212 214 216 218 220m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rela

tive A

bunda

nce

210.0739C 6 H13 O6 N 15N

209.0769C 6 H13 O6 N2

211.0710C 6 H13 O6 15N2

205.1548C 9 H21 O3 N2

208.0583C 6 H11 O6 N 15N

212.0745C 2 H15 O9 N 15N

202.1438C 10 H20 O3 N

217.0972C 12 H13 O2 N2

219.1704C 10 H23 O3 N2

206.1582C 4 H22 O5 N4

15N15N

15N

15N

+

15N15N

15N

15N

Mechanism where hexamine breaks down is supported

Formation Mechanism Evidence

15N label scrambled during HMTD formation, but no scrambling in starting materials observed.

9/27/2017

6

NH

CH H

HO OHH+

H2NC

H H

OOHNH

CH H

H+

O OH2N NH2

O OH2N NH2

O OHO OH

NHHN

O O

O O

O OHO OH

N O

O

O

O

O

O

N

2

DHMP

4H2O O

H H4 +

NH

CH H

2 + NH32

O

H H

2 HO OH+ OO

HOOH

2X

Mechanism for HMTD Formation involves C & N scrambling

BHMP

Bis(hydroxymethyl)peroxide

232 234 236 238 240 242 244 246 248 250m/z

0

20

40

60

80

1000

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

1000

20

40

60

80

100239.1120

243.1684

237.1436235.1676 241.2143234.1325 244.1886240.1423 249.1829

238.1280 242.2461

232.1529 247.6226235.4929 237.7898

235.2038

240.1421243.1683

234.1316236.1108 249.1828244.1522239.1623232.1524 246.1678

237.1828 248.1481233.1154

240.1421

235.1672

234.2043 241.2142244.2250

231.1143 236.1109 249.1828246.1678239.1621 243.1935233.1152

240.1423

235.1674 241.1455 244.2252231.1144 249.1829246.1680237.1830 239.1984233.1155

240.1422

241.1455235.1674 249.1829239.1962 244.1887231.1144 233.1882 237.1831 246.1679

NL: 2.69E5APCI+_TATP_50ACN_50Aq_1dec2015_151026171724#4-21 RT: 0.05-0.33 AV: 18 T: FTMS + p APCI corona Full ms [230.00-250.00]

NL: 8.27E4apci+_tatp_20acn_30meoh_50aq_1dec2015_151026171724#2-21 RT: 0.02-0.33 AV: 20 T: FTMS + p APCI corona Full ms [230.00-250.00]

NL: 2.33E5apci+_tatp_5acn_45meoh_50aq_1dec2015_151026171724#3-23 RT: 0.03-0.35 AV: 21 T: FTMS + p APCI corona Full ms [230.00-250.00]

NL: 1.42E6apci+_tatp_50meoh_50aq_1dec2015_151026171724#3-21 RT: 0.04-0.32 AV: 19 T: FTMS + p APCI corona Full ms [230.00-250.00]

NL: 1.42E7apci+_tatp_true0acn_50meoh_50aq_1dec2015_2_151026171724#3-21 RT: 0.04-0.33 AV: 19 T: FTMS + p APCI corona Full ms [230.00-250.00]

52% ACN

22% ACN

7% ACN

2% ACN

0% ACN

2.33x105

1.42x106

1.42x107

8.27x104

2.69x105

Detecting Peroxide at low concentration Signal response of TATP [M+NH4]

+ m/z 240.1442 with & without acetonitrile (ACN)

Response 240.1442

9/27/2017

7

%ACN→ 50 40 30 20 10 5 2 0

Sample prep: aqueous is a constant 50% 10 mM NH4OAc while organic ratio of ACN and methanol (MeOH) is varied

*FIA- Flow injection analysis of 20 µL injections (N=3) of 50 µg/mL into mobile phase of 100% 10 mM ammonium acetate (NH4OAc) at 300 μL/min flowrate.

FIA* depicting TATP [M+NH4]+ response in ESI with

decreasing concentration of ACN

Average Area

%ACN TATP response50 192614640 217597730 276651620 343766410 61099705 101718182 152025700 24112311

Normalized %

%ACN TATP response50 840 930 1120 1410 255 422 630 100

5 10 15 20 25 30 35Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bund

ance

TATP

DADP

HMTD

Hexamine

MEKP DHP1*

MEKP DHP2

MEKP C3

MEKP DHP3

MEKP DHP4

MEK

MEKP C1

MEK/AP C2

MEKP C2

AP DHP2

MEK/AP DHP2MenadioneCyclopentanoneCyclohexanone

1, 2, 4-Triazole

TBAH Michler's Ketone

Diphenyl isophthalate

[M+NH4]+ Only Not affected by ACN

[M+H]+ Only

[M+NH4]+ and [M+H]+

*Also produced sodium adduct equally affected

Compounds tested‐‐ions formed & ACN effect on ionization

9/27/2017

8

Decreasing Signal Intensity with Increasing ACN

FIA in APCI

FIA in ESI

HMTDcyclohexanone1,2,4‐triazolemenadioneMEK

Hexamine

HMTD1,2,4‐triazolemenadioneacetonecyclohexanoneMEKdiphenyl isophtha

cyclopentanone

H+ affini ty

water 691

MeOH 754

ACN 779

cyanamide 806

pivalonitrile 811

acetone 812

cyclopentanone 824

MEK 827

cyclohexanone 841

ammonia 854

1,2,4‐triazole 886

This is curious since ACN has a lower proton affinity than most of the analytes.

Polar interaction between nitrile and analyte causes formation of neutral aggregatePolarization

R groups block site of ionizationWhen analyte in cis configuration

Analysis with electron donating (-NH2), electron withdrawing (-Br) and steric (trimethyl) nitriles support this mechanism.

Peroxides in cis formation have large dipole.

TATP and HMTD are forced into cis configuration

Large, linear peroxides cannot form cis isomer without self-steric interaction

DHP3 and DHP4 were not affected by ACN

Addition of heat (HESI) showed significant effect for DHP3 not DHP4!

Peroxide Detection

O

O

HO

O O

O O

OH

O

OO

OO

HO

O

O

OH

O

MEKP DHP3

MEKP DHP4

Proposed Mechanism for Ion Suppression by ACN

Rapid Communications in Mass

Spectrometry 2016, 27(1), 1796‐1804.

9/27/2017

9

Full Scan of HMTD in MeOH/H2O—why two 207 peaks?

207.0 207.5 208.0 208.5 209.0 209.5

m/z

0

20

40

60

80

100

0

20

40

60

80

100R

elat

ive

Abu

ndan

ce207.0979 209.0771

208.1011207.0621

209.0772

207.0980

NL: 7.12E5

HMTD_100%MeOH_Infuse_APCI_3june2014#8-41 RT: 0.06-0.29 AV: 34 T: FTMS + p APCI corona Full ms [150.00-500.00]

NL: 6.39E5

hmtd_10%meoh_90%water_infuse_apci_3june2014#6-35 RT: 0.05-0.25 AV: 30 T: FTMS + p APCI corona Full ms [150.00-500.00]

207.0617

100% MeOH

10% MeOH

[HMTD+H++MeOH-H2O2]+

ΔPPM = 1.70

[HMTD-H2+H]+

ΔPPM = 2.60

C6H13O6N2

[HMTD+H]+

ΔPPM = 1.40

C7H15O5N2

C6H11O6N2

HMTD infused in CD3OD/D2O or CH318OH/H2O

[HMTD+D++CD3OD+-D2O2]+

ΔPPM = 0.67

[HMTD+H]+

ΔPPM = 0.91(non-deuterated solvent impurity)

[HMTD+D]+

ΔPPM = 1.24

All peroxide hydrogens → lost from the solvent!

209.1022

209.0774

207.0 207.5 208.0 208.5 209.0m/z

0

20

40

60

80

100

120

Rel

ativ

e A

bund

ance

[HMTD+H]+

ΔPPM = 2.82

[HMTD+H++Me18OH-H2O2]+

ΔPPM = 1.96

All peroxide oxygens → lost from HMTD

210.1165

210.0834

209.2 209.4 209.6 209.8 210.0 210.20

20

40

60

80

100

Rel

ativ

e A

bund

ance

209.0770

CD3OD/D2O

CH318OH/H2O

C7H12D3O5N2+

C6H12DO6N2+

C7H15O418ON2

+

C6H13O6N2+

9/27/2017

10

Proposed Mechanism for Formation of (A) Protonated HMTD or (B) Alcohol‐Bonded Species

[M+H]+

m/z = 209.0768

R m/zH- 193.0819

CH3- 207.0976CH3CH2- 221.1132C2H6CH2- 235.1289

C4H9- 249.1445C6H11- 275.1601C8H17- 305.2071

Rapid Comm Mass Spectrometry 2015, 29(1), 74-80. Oxygens lost from HMTDHydrogens lost from solvent

0

10

20

30

40

50

60

70

80

90

100

0

5

10

15

20

25

30

0 2 4 6 8

% M

ethan

ol in M

obile P

hase

m/z

89.

0597

/240

.144

2 In

ten

sity

Rat

io

Time (min)

Increase of m/z 89.0597 relative to m/z 240.1442 with increasing MeOH

Post-column addition TATP in MeOH/water into LC flow of std gradient (10 mM NH4OAc/MeOH) monitoring ratio of m/z 89.0597 to m/z 240.1442 with increasing methanol.

Which mass & column for best detection limits? With ammonium modifier present → 240.1442 (TATP-NH4

+) or 89.0597 (C4H9O2+)?

TATP elution time on PFP column

TATP elution time on C18 column

~65% MeOH

~85% MeOH

Gradient

9/27/2017

11

NL:1.96E6

88 90 92 94 96 98 100 102 104m/z

05

101520253035404550556065707580859095

100

Rel

ativ

e A

bund

ance

97.0776

91.0400

105.0556

89.0608

95.098488.0530

103.0764

98.117294.0589 101.060792.0434 104.0716

or

or

92.0795

O

OH

O

D3C CD3

OO

D3C CD3

O

OH

O

Direct infusion (20 μL/min) of TATP (5 μg/mL), d18-TATP (5 μg/mL) & MEKP (10 μg/mL) in 90% MeOH/10% 10 mM NH4OAc. Only m/z 88-105 shown; incorporated ROH in red.

Infusion of TATP, d18-TATP & MEKP in CH3OH in to APCI Source

OO

OHO

OO

What is Mass 89.0597? C4H9O2+ is unlikely fragment!

OO

O

O

O

O +3H

-2CH3

d18-TATP produced m/z 95 C4H3D6O2

+ not m/z 98 C4D9O2+

and m/z 92 C4H6D3O2

+

NL: 2.63E6

88 90 92 94 96 98 100 102104

m/z05

101520253035404550556065707580859095

100

Rel

ativ

e A

bund

ance

91.0651

105.0807

89.0609

97.102793.0693

103.0765102.1289

104.0967

88.1132 100.076895.0985

96.0881

94.0662 101.097292.0684 99.081590.0925 98.1173

88.0531

96.0632

91.0401

97.0777

105.0557

D3C CD3

18OO

D3C

18O

OH

O

18OO

O

OH

O

93.0443

Direct Infusion (20 μL/min) of TATP (5 μg/mL), d18-TATP (5 μg/mL) and MEKP (10 μg/mL) in 90% Me18OH/10% 10 mM NH4OAc. Only m/z 88-105 are shown for resolution purposes.

Infusion of TATP, d18-TATP & MEKP in CH318OH into the APCI source

18OO

OO

Lower gas flow promoted the formation of 95 and 92. Higher flow favored [M+NH4]+ and 97 formation. Lower gas flow may mean more time to react with MeOH in the discharge region or more time in the heated region of ceramic tube.

9/27/2017

12

m/z = 88.0519

TATP formation of m/z 89.0597 (corresponds to MEKP m/z 103.0754)

m/z = 89.0597

Proposed mechanism of gas phase formation of m/z 89 for TATP or 103 for MEKP & m/z 88 for MEKP or m/z 74 for TATP

MEKP formation of m/z 88.0519 (corresponds to TATP m/z 74.0362)

+ +

++

+

RT:0.00 - 8.01

0 1 2 3 4 5 6 7 8Time (min)

0

20

40

60

80

1000

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

1002.07

4.80

4.80

4.802.09

RT:0.00 - 8.01

0 1 2 3 4 5 6 7 8

Time (min)

0

20

40

60

80

1000

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

1004.88

2.11

4.88

4.88

4.88

2.11

Krawczyk suggested converting HMTD to TMDDD would improve detectable, but…. We attempted to form in-situ by injecting HMTD (100 ng) onto 5cm PFP column with APCIsource & raising temperature 210°C 300°C. More TMDDD formed, but not enough.

We observed HMTD always had TMDDD contamination.

TMDDD TMDDDHMTD

m/z 207.0611

HMTD

m/z 207.0975

m/z 209.0768

m/z 224.0877

210°C 300°C

6.9e47.0e4

1.0e6 1.4e6

0

7.3e6 5.8e6

3.1e6 2.5e6

3.1e5

4.2e62.2e5

HN

O O O

O O O

N

TMDDD

TMDDD unaffected by temperature rise. HMTD formation of TMDDD is affected by temperatureNumbers by peaks are area counts.

9/27/2017

13

HMTD infused with MeNH2 produced different peaks in APCI & ESI.

ESI+_HMTDinACNwMeNH2_10ugmL_11nov2016_161024142316 #13T: FTMS + p ESI Full ms [60.00-600.00]

100 150 200 250 3000

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relative Abu

ndance

238.1032

220.1365

113.1071

161.1646

72.0803

171.1490

191.0661287.0807

279.1593

90.0910

310.

ESI

APCI+_HMTDinMeOHwMeNH2_10ugmL_11nov2016_161024142T: FTMS + p APCI corona Full ms [100.00-600.00]

150 200 250 3000

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relative Abu

ndan

ce

248.2218

240.1188

209.0766145.0605

163.0711114.0910

195.0874

279.1588

310.

APCI

240.1188

238.1032

209.0766

220.1365

Could an amine replace ammonium in aiding detection of HMTD?

HN

OO

O

O OO

N

H2N

oror

Could detection of HMTD be improved by in-situ formation of TMDDD & interaction with amine? Thus, HMTD with post-column addition of EtNH2

RT: 0.00 - 8.00

0 1 2 3 4 5 6 7 8Time (min)

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

4.01

4.04

4.09

4.12

4.184.53

4.01

3.99 4.03

4.07

4.10

3.94

4.183.863.561.59 2.940.86 1.90

1.87 1.881.84

1.911.961.79

1.74 2.002.031.40

1.29 2.051.07

2.13

3.20 3.172.25 4.07 4.090.97 7.18

NL: 1.43E6m/z= 209.0731-209.0810 F: FTMS + p ESI Full ms [100.00-500.00] MS HMTD_20ugmL_PostColEtNH2_ESI_15nov2016_3



RT: 0.00 - 7.50

0 1 2 3 4 5 6 7Time (min)

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

3.44

3.59 3.76 4.19 4.67 5.633.233.44

3.47

3.403.52

3.64 3.744.20 5.03 5.341.911.69 3.20 5.66

3.453.47

3.401.761.00 3.560.93

1.821.07 3.72 3.75

NL: 1.04E7m/z= 209.0721-209.0815 F: FTMS + p APCI corona Full ms [80.00-280.00] MS HMTD_20ugmL_PostColEtNH2_APCI_15nov2016_4



APCI ESI

209.0768 209.0768

254.1340254.1340

252.1190 252.1190

HN

OO

O

O O O

N

TMDDD TMDDDHMTD HMTD

1.0 e7

1.7 e7

8.6 e5

1.4 e6

1.0 e6

8.1 e6

Numbers in blue are peak area counts.

Formed TMDDD

contamination contaminationdecomposition decompositionRetention time Retention time

Note: HMTD is best observed by APCI, and TMDDD in ESI.

9/27/2017

14

APCI+_HMTD_IPamine2mM_14feb2017_1 #72-103 RT: 1.81-2.66 AV: 28 NL: 1.65E5T: FTMS + p APCI corona Full ms2 [email protected] [55.00-400.00]

100 150 200 250 300 350 400m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relativ

e A

bund

ance

163.0705

268.1489

209.0758

234.1436145.0600

197.0758

72.0801

267.1365272.9864179.065288.0386 108.6934 250.1383 303.6513 383.1518

142.0694

327.9850147.1121

MS/MS fragmentation of HMTD-i-propylamine product formed in APCI

HN

O O O

O O O

N

H2N

-H2O

-H2O2Amine+1C HMTD-1C

or

Result is a product not an adduct Amine removes 1 C from HMTD

Use of deuterated solvent (D2O/CD3OD) confirmed assignments.

APCI+_HMTD_1mMcyclohexAmine_inf_23Feb2017 #19-38 RT: 0.37-0.97 AV: 20 NL: 3.49E5T: FTMS + p APCI corona Full ms2 [email protected] [80.00-400.00]

100 120 140 160 180 200 220 240 260 280 300 320m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Relative Abu

ndance

308.1813

112.1118

163.0710

274.1758

100.1117

145.0605 177.0867209.0765

187.1438 233.1494 307.169088.0389 144.1016

158.1173

112.3879278.1711

245.1494

273.6242

225.1599 309.2330

HMTD with Cyclohexylamine in APCI+

HN

N

O O

O O

N

290.1707

-H2O

197.0766

HN

O O H2O

O O O

N

9/27/2017

15

Reaction of HMTD and TMDDD with Amines

Organic Amine Expected mass Observed mass ΔPPM pKa

Methylamine 240.1190 240.1185 -2.1 10.62Ethylamine 254.1347 254.1342 -2.0 10.87

Dimethylamine 254.1347 254.1341 -2.4 10.73Triethylamine* 310.1973 310.1963 -3.2 10.78Isopropylamine 268.1503 268.1489 -5.2 10.63

Cyclohexlyamine 308.1816 308.1813 -1.0 10.63Aniline 302.1347 302.1338 -3.0 4.6

2-nitroaniline 347.1197 NR -0.28(2-aminoethyl)trimethylammonium 311.1925 NR na

* Triethylamine forms an adduct to HMTD with no observable chemical reactionNR-no reaction, na-not available

Amines infused with HMTD into APCI source

If amine reacted with HMTD, the quaternary amine would make a charged product

• TMDDD forms intense adducts with amines in ESI source• Adduct formation was confirmed since no fragments were detected for TMDDD + amine• When mass of amine was > 50 amu → detected fragment was only the ionized amine

• HMTD formed products with basic amines; these gave abundant MS fragments• Fragments suggest incorporation of the amine into the HMTD structure• Intensity of the product is related to the amine basicity• Triethylamine only forms an adduct with HMTD.

• Trimethylamine, 2-nitroaniline, (2-aminoethyl)trimethyl ammonium, & choline did not form covalent products with HMTD.

N

O O

O O O

N

N R

H H

N R

HN

O O

OO O O

N

H

N

R

HH

H

H

N

O O

O O

N

N R

H

HOOH

N

O O

O O

N

N R

H

11

12

13

19

14 16

17

HN

O O

O O

N

OOH2

15

Thoughts on Mechanism for the Fragmentation of the HMTD/amine Product in CID

10

9/27/2017

16

1. TATP is extremely volatile, even in solution, but in headspace TATP is intact, unlike HMTD where the headspace is decomposition products.

2. ACN in MP of LC‐MS suppresses charge for some peroxides & ketones.3. TATP and HMTD react with MeOH in gas phasefor TATP this results in LC‐MS peak at 89 indicating addition of 1 or 2 MeOHfor HMTD this results in 207

4. Reactions of TATP or HMTD with ROH can be used to lower detection limits but variable results will be obtained if LC‐MS analytical conditions are changed.Analytical conditions include

columnionization source –TATP & HMTD prefer APCI (linear peroxides ESI)solvent gradienttemperaturesheath & auxiliary gas flow rates

5. HMTD is generally contaminated with oxidation product TMDDD.6. HMTD reacts with amines in gas phase. As with ROH, R2NH or RNH2 attack is at C

unfortunately, unlike with ROH, addition of amines did not improve detection limits. TATP did not react with amines under the same conditions.

7. TMDDD tends to form adducts rather than products with amines

Summary of Lessons Learned

adventures in analysis of peroxide explosivesenergetics.chm.uri.edu/system/files/isade peroxide...

Documents