DERIVATISATION OF CHEMICAL WARFARE AGENT

DEGRADANT WITHOUT REMOVAL OF WATER

By

Xiang Le CHUA

A thesis submitted in fulfilment of the requirements for the degree of

Master of Forensic Science (Professional Practice)

in

The School of Veterinary and Life Sciences

Murdoch University

Supervisors:

Dr Kate ROWEN (Murdoch)

Dr John COUMBAROS (Murdoch)

Semester 2, 2018

ii

DECLARATION

I declare that this manuscript does not contain any material submitted previously for the

award of any other degree or diploma at any university or other tertiary institution.

Furthermore, to the best of my knowledge, it does not contain any material previously

published or written by another individual, except where due references has been made in

the text. Finally, I declare that all reported experimentations performed in this research were

carried out by myself, except that any contribution by others, with whom I have worked is

explicitly acknowledged.

Xiang Le CHUA

iii

ACKNOWLDEGEMENTS

This work would not have been possible without the invaluable contributions from Dr Kate

Rowen, Dr John Coumbaros, Mr Bob Du and Mr Lucas Dival. I would love to express my most

sincere gratitude to my supervisors, Dr Kate Rowen and Dr John Coumbaros, for their

constant support, encouragement and patience throughout the semester.

I would also like to thank Mr Bob Du from the Post-Harvest research group in Murdoch

University for his insights and advice on my research. I am exceptionally thankful towards Dr

David Berryman and Reza from the SABC, for providing me with solutions when my project

hit the wall. I am also thankful for Lucas, for providing me with suggestions and support at

the start of my research into chemical warfare degradant. Special thanks to my co-academic

chair, Mr Brendan Chapman for helping with the funding for my research. Thanks Brendan,

for being an inspiring figure/educator for the Masters course. I would like to thank Shahaziq

Shahari, Gareth Liew, and especially Saesario Putra, for being the motivating seniors of my

research semester.

My postgraduate program has been a very enjoyable and memorable one, thanks to

everyone that I’ve mentioned above, including my beloved Academic Chair, Associate

Professor James Speers, and lovely classmates/buddies: Sonita Regupathi, Kathleen Nguyen

and Anindita Warddhana. Looking forward to graduation with you guys.

Thanks, Mummy and Daddy, for stretching your last bit of finances to support my dream.

TABLE OF CONTENTS

TITLE PAGE ............................................................................................................................. i

DECLARATION ....................................................................................................................... ii

ACKNOWLEDGEMENTS ........................................................................................................ iii

PART ONE

Literature Review ............................................................................................................. 1-65

PART TWO

Manuscript ....................................................................................................................... 1-25

Blank Page – not numbered

Part One

Literature Review

DERIVATISATION OF CHEMICAL WARFARE AGENT

DEGRADANT WITHOUT REMOVAL OF WATER

1

TABLE OF CONTENTS

LIST OF FIGURES ................................................................................................................... 2

LIST OF TABLES ..................................................................................................................... 3

LIST OF ABBREVIATIONS ...................................................................................................... 4

ABSTRACT ............................................................................................................................. 8

1. INTRODUCTION .............................................................................................................. 9

2. OVERVIEW OF CHEMICAL WARFARE AGENTS............................................................. 11

3. ORGANOPHOSPHORUS NERVE AGENTS AND THEIR DEGRADANTS........................... 15

3.1 Degradation Pathways ............................................................................................ 15

4. SAMPLE PREPARATION ............................................................................................... 20

4.1 Soil Samples ............................................................................................................ 21

4.2 Aqueous Liquid Samples ......................................................................................... 22

4.3 Solid and Other Samples......................................................................................... 23

5. DERIVATISATION.......................................................................................................... 24

5.1 Derivatisation of Organophosphorus Nerve Agents and their Degradants ........... 24

5.2 Types of Derivatisation Reactions .......................................................................... 25

5.2.1 Common Derivatising Techniques in CWA Studies ....................................... 27

5.3 Disadvantages of Derivatisation ............................................................................. 30

5.4 Current Trends in Derivatisation ............................................................................ 31

5.4.1 Derivatisation Without Removal of Water ................................................... 32

6. DETECTION & QUANTITATION OF CHEMICAL WARFARE AGENT DEGRADANTS ...... 36

6.1 Gas Chromatography-Mass Spectrometry ............................................................ 37

6.2 Liquid Chromatography-Mass Spectrometry ........................................................ 40

7. CONCLUSION ................................................................................................................ 50

REFERENCES ........................................................................................................................ 51

2

LIST OF FIGURES

Figure 1. Hydrolytic pathways of various CWAs (sarin, soman, cyclosarin, VX and Russian VX)

into their various alkyl methylphosphonic acids, and subsequently into their final

degradation product- methylphosphonic acid ................................................................... 16

Figure 2. Hydrolytic pathway of tabun (GA) into its degradant products under acidic or basic

pH conditions ...................................................................................................................... 18

Figure 3. Chemical equation for general derivatisation reactions where ‘A’ denotes nitrogen,

oxygen or sulphur, while ‘D’ denotes the functional group of derivatisation reagent used.25

Figure 4. Reaction scheme of the methyl, trimethylsilyl, tert-butyldimethylsilyl, and

pentafluorobenzyl derivatives of MPA ............................................................................... 29

Figure 5. Three different modes for derivatisation methods associated with SPME; Direct

derivatisation in sample matrix, derivatisation in SPME fiber coating, or derivatisation in GC

injector port ........................................................................................................................ 33

3

LIST OF TABLES

Table 1. Names, chemical structures and classes of CWA of interest ................................ 12

Table 2. Three different types of derivatisation reactions with their corresponding

derivatising reagents ........................................................................................................... 26

Table 3. General LC-MS parameters used for analysis of organophosphorus compounds.40

Table 4. LC-MS parameters for the analysis of CWA related compounds ......................... 42

Table 5. Summary of HPLC-MS methodologies for the determination of MPA from various

sample matrices .................................................................................................................. 45

4

LIST OF ABBREVIATIONS

CWA

CWC

OPCW

MPA

GC

GC-MS

MTBSTFA

LOD

PPM

GB

GD

GF

VX

VR

HD

HN-3

L

BZ

AC

CK

AChE

Ach

Chemical Warfare Agent

Chemical Weapons Convention

Organisation for the Prohibition of Chemical Weapons

Methylphosphonic Acid

Gas Chromatography

Gas Chromatography with Mass Spectrometry

N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide

Limits of Detection

Parts Per Million

Sarin

Soman

Cyclosarin

O-ethyl, S-2-diisopropylaminoethyl methylphosphonothiolate

S-(N,N-diethylaminoethyl) Isobutyl methylphosphothiolate

Sulphur Mustard; Bis-(2-chloroethyl)sulfide

Nitrogen Mustard; Tris-(2-chloroethyl)amine

Lewisite; 2-Chlorovinyldichloroarsine

3-quinuclidinyl benzilate

Prussic Acid

Cyanogen Chloride

Acetylcholinesterase

Acetylcholine

5

IMPA

PMPA

GA

DESH

EMPA

DES2

iBuMPA

ROP

SPE

SPME

OMDEX

INDEX

NICI-MS

DMF

PFBBr

PFBHA

TBH

BSA

BSTA

HMDS

TMSI

TMS-DEA

TMCS

Isopropyl Methylphosphonic Acid

Pinacolyl Methylphosphonic Acid

Tabun

Diisopropyletylmercaptoamine

Ethyl Methylphosphonic Acid

Bis(diisopropylaminoethyl)disulphide

Isobutyl Methylphosphonic Acid

Recommended Operating Procedures

Solid-Phase Extraction

Solid-Phase Microextraction

On-matrix Derivatisation Extraction

Microemulsion mediated in-situ Derivatisation and Extraction

Negative Ion Chemical Ionisation with Mass Spectrometry

Dimethylformamide

Pentafluorobenzyl Bromide

Pentafluorobenzyl-hydroxylamine Hydrochloride

Tetrabutylammonium Hydroxide

Bis(trimethylsilyl)-acetamide

N,N-bis(trimethylsilyl)trifluoroacetamide

Hexamethyldisilzane

Trimethlsilylimidazole

Trimethylsilildiethylamine

Trimethylchlorosilane

6

MBTFA

PFBCI

PFPOH

SIM

LC-MS

TBDMS

TMS

TMSCI

TBDMSCN

Sn-Mont

HF-LME

ECD

GC-MS/MS

LC-MS/MS

NMR

FPD

FID

PID

NID

EI

CI

TIC

OCAD

N-methyl-bis(trifluoroacetamide)

Pentafluorobenzoyl Chloride

Pentafluoropropanol

Selected Ion Monitoring

Liquid Chromatography with Mass Spectrometry

t-butyldimethylsilyl

Trimethylsilyl

Trimethylsilyl Chloride

t-butyldimethylsilyl Cyanide

Tin Ion-exchanged Montmorillonite

Hollow Fiber Liquid Phase Microextraction

Electron-capture Detection

Gas Chromatography with Tandem Mass Spectrometry

Liquid Chromatography with Tandem Mass Spectrometry

Nuclear Magnetic Resonance

Flame Photometric Detection

Flame Ionisation Detection

Photoionisation Detection

Nitrogen-phosphorus Detection

Electron Impact Ionisation

Chemical Ionisation

Total Ion Current

OPCW Central Analytical Database

7

APPhI

APCPhI

1D/2D

MIM

ESI

TS-MS

API

APCI

Atmospheric Pressure Photoionisation

Atmospheric Pressure Photochemical Ionisation

Selectable One- and Two- Dimensional

Multiple Ion Monitoring

Electronspray Ionisation

Thermospray Mass Spectrometry

Atmospheric Pressure Ionisation

Atmospheric Pressure Chemical Ionisation

8

ABSTRACT

Degradation products are distinct chemical signature for the detection of chemical warfare

agents and forensic attribution in cases where the parent compound no longer exists. Being

the common degradant to nerve agents, methylphosphonic acid serves as an explicit

biomarker for the use of nerve agents. However, this non-volatile compound requires

derivatisation prior to its detection, which incurs time-limiting factor and error. Therefore,

development for a novel rapid and simple approach for identifying these degradants is of

high importance to shed light on the use of chemical warfare agents and remediation of

impacts. With the success of a recent pilot study for derivatisation of methylphosphonic acid

without elimination of water, a subsequent quantitative study can be proceeded to assess

on its practicality and efficiency.

9

1. INTRODUCTION

The presence of the hydrolysis products (degradants) of chemical warfare agents (CWA) is a

potential evidence for the possession and/or involvement with CWA.1-3 This is in relation to

the Chemical Weapons Convention (CWC), which was signed by 130 countries for the

prohibition of the development, production, stockpiling and use of CWA, and their

destruction.4 CWC entered the force officially on 29th April 1997, forming the Organisation

for the Prohibition of Chemical Weapons (OCPW).5 To uphold and enforce the requirements

of CWC, OCPW carries out verification activities on a regular basis, to conduct and challenge

former CWA facilities of the member states for any presence of CWA precursors and

degradants.4-7 Despite this, the use of CWA remains a problem, with many different forms of

CWA being developed and utilised heavily in many conflicts, terrorist attacks and wars

around the world even till this day.8

Organophosphorus nerve agents such as sarin, soman, cyclosarin, VX and VR hydrolyses in

aqueous environment to form various forms of alkylphosphonic acids, and finally into

methylphosphonic acid (MPA).9-11 Besides being the final degradant product to

organophosphorus nerve agents, MPA is in its most stable form and chemically resistant to

any further degradation.3,9,10,12,13 These degradant compounds do not exist in nature even

though MPA was suggested to be a degradant product of some fire-retardants, however, no

publications has been able to prove the latter.14 Gas chromatography (GC) is the most

widely used and approved technique for the unambiguous detection of MPA.8,15,16 However,

MPA possesses high polarity with poor volatility, therefore derivatisation must be

performed on this compound prior to gas chromatographic analysis.9,17 Derivatisation of

target analytes requires the elimination of water content; concentration of aqueous extract

10

to dryness.9,18 This step is reported by many analysts and authors to be time-consuming, as

well as its potential for errors.3,9,18,19

There are limited literature resources that report and suggest for an alternative method for

the removal of water. Earlier this year (2018), a time-efficient protocol for the qualitative

detection of MPA without the removal of water by gas chromatography combined with

mass spectrometry (GC-MS) was finally developed and proven to be successful by Dival.20 In

this study, N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide (MTBSTFA) was used for

the derivatisation of MPA, with an addition of organic layer of hexane.20 The presented

approach was successful in producing GC-amenable derivatives of MPA, with the detection

limit (LOD) to be 1000 ppm.20 Despite the success of his approach, no quantitative

assessment was performed by Dival to evaluate the efficiency of his methodology. The

presented study by Dival can be progressed by providing a quantitative study to further

assess the derivatisation efficiency and practicality of the pilot study. This literature review

aims to review and discuss the current applications and protocols in CWA analysis, with the

goal of developing a reliable and reproducible approach to quantitate the efficiency of two-

phase derivatisation of MPA without the removal of water.

11

2. OVERVIEW OF CHEMICAL WARFARE AGENTS

The history of chemical weapons has its roots in the first World War where the German

forces utilised poisonous chlorine gases on a large scale against the French army in 1915.21

The devastation of CWA in World War I resulted in about 90,000 deaths and 1.3 million non-

fatalities.22 Since then, CWAs have been utilised in many armed conflicts and terrorism.

CWAs are undoubtedly one of the most feared invention owing to their fatal ability to cause

death, temporary incapacitation or any permanent harm and injury to humans or animals.23

There are many different types and classes of CWAs, with the most lethal group suggested

to be nerve agents, with the rest being the blistering agents (also known as the vesicants),

blood agents and the incapacitating agents (Table 1).8,16 All CWAs are distinct in their

characteristics, chemical and physiological properties, volatility and chemical structures.24

12

Table 1. Names, chemical structures and classes of CWA of interest.10,16,25

Chemical Name [CAS number]

Common Name (Abbreviation) Chemical Formula

Class Schedule no.

Structure

Isopropyl methylphosphonofluoridate [107-44-8]

Sarin (GB) C4H10FO2P

Nerve agent 1.A.1

Pinacolyl methylphosphonofluoridate [96-64-0]

Soman (GD) C7H14FO2P

Nerve agent 1.A.1

Cyclohexyl methylphosphonofluoridate [74192-15-7]

Cyclosarin (GF) C7H16FO2P

Nerve agent 1.A.1

O-ethyl, S-2-diisopropylaminoethyl methylphosphonothiolate [50782-69-9]

VX C7H18NO2PS

Nerve agent 1.A.3

S-(N,N-diethylaminoethyl) isobutyl methylphosphothiolate [159939-87-4]

Russian VX (VR) C11H26NO2PS

Nerve agent 1.A.3

Bis-(2-chloroethyl)sulfide [505-60-2]

Sulphur Mustard (HD) C4H8Cl2S

Vesicant 1.A.4

Tris-(2-chloroethyl)amine [555-77-1]

Nitrogen Mustard (HN-3) C5H11Cl2N

Vesicant 1.A.6

13

2-Chlorovinyldichloroarsine [541-25-3]

Lewisite (L) C2H2AsCl3

Vesicant 1.A.5

3-quinuclidinyl Benzilate [6581-06-2]

BZ C21H23NO3

Psychotomimetic 2.A.3

Hydrogen Cyanide [74-90-8]

Prussic Acid (AC) HCN

Blood agent 3.A.3

Cyanogen Chloride [506-77-4]

(CK) CClN

Blood agent 3.A.2

Organophosphorus nerve agents exert their extreme toxicity by anticholinesterase action,

targeting the nervous system of the victims, often leading to seizures, respiratory arrests

and deaths.13,26 These nerve agents can be further divided into two categories; G-series and

V-series. Both series do not exist naturally and are also commonly known as

organophosphorus nerve agents due to their phosphorous-core structure.22,27 Nerve agents

of G-series comprises of tabun, sarin, soman, and cyclosarin. Among all nerve agents, the V-

series are found to be more dangerous than G-series due to their lower volatility and

persistence in the environment, which could be explained by the presence of a strong P – S

14

bond.22,28 Furthermore, V-series has an estimated human LD50 of 10 mg on skin of a 70 kg

human.17,27

Blistering agents comprises of sulphur mustards, nitrogen mustards and lewisites.8,22 These

blistering agents are known to inflict chemical burns onto skins; tissue blistering.22 These

cytotoxic alkylating chemical weapons were developed to induce immediate casualties,

reducing soldiers’ fighting efficiency and morale.22 The third category of CWA belongs to the

incapacitating agent 3-quinuclidinyl benzilate which targets the central nervous systems to

induce hallucinations, disorientation, coma or even medullary paralysis.8,22 Incapacitating

agents are also known as the psychotomimetic agents. The blood agents exert their toxicity

on a cellular level, interfering with the activity of oxygen transport mechanism within the

victims’ body, ultimately leading to death by respiratory failure. Toxic compounds that come

under this category are either arsenic- or cyanide based; hydrogen cyanide, cyanogen

chloride and arsine.22,29 Besides being used in World War I and II, hydrogen cyanide was

suggested to be used during the Iran – Iraq war, and the attack in Tokyo in 1995.30 Other

classes of CWAs such as the choking agents, toxins and bioregulators, and many others exist

but will not be discussed in this literature review.

15

3. ORGANOPHOSPHORUS NERVE AGENTS AND THEIR DEGRADANTS

Organophosphorus nerve agents are known to be the most lethal and toxic compounds

among all CWA.31 The first reported use of tabun and sarin was during the first Gulf War

among the Iraq and Iran forces, and the Kurdish rebels.22 Impure Sarin was used by

terrorists during a Japan subway attack in 1995, taking away 12 lives with 5000

casualties.16,22 Sarin was used again a year ago in 2017, killing at least 80 civilians in the

northern Syria.32,33 The latest use of a nerve agent happened earlier this year on 4th March

2018, injuring a former Russian double agent and his daughter, and a policeman in UK.34

Nerve agents remain a high priority concern due to their trend of use on civilian targets.35

Organophosphorus nerve agents are anticholinesterase compounds. Upon inhalation or

dermal absorption, these compounds irreversibly inhibit acetylcholinesterase (AChE) via the

formation of covalent P – O bond at the serine hydroxyl group in the enzyme active site. This

leads to the build-up of neurotransmitter acetylcholine (Ach) at nerve synapses, resulting in

the disruption of nervous and respiratory systems.8,26 These chain of events is extremely

fatal to any living organisms and kills within several minutes.8 Nerve agents of G-series

exhibit a common mechanism of toxicity following a hydrolysis metabolic pathway, whereas

the V-series have a lower affinity for AChE and are less prone to hydrolysis.22

3.1 Degradation Pathways

CWA degradants often serve as biological and environment markers of the use or

production of CWA.2 CWA are known to decompose (hydrolyse) relatively well in humid air

16

or in the presence of water, which may lead to the absence of the parent compound.12,35

Therefore, verification of CWA for the CWC will be dependent on the qualitative detection

for the presence of their corresponding degradants.2 Organophosphorus nerve agents are

moderately volatile substances that could be analysed by any gas chromatographic

methods.8,9 Upon degradation in aqueous environment, these agents form their

corresponding alkyl phosphonic acids degradants, and finally into MPA; the common

degradant to organophosphorus nerve agents (Figure 1).8,9,36

Figure 1: Hydrolytic pathways of various CWAs (sarin, soman, cyclosarin, VX and Russian VX) into their various alkyl methylphosphonic acids, and subsequently into their final degradation product- methylphosphonic acid. Adapted and modified from Y. Seto et. al.13

17

Sarin hydrolyses (hydrolysis of P – F bond) in aqueous environment to produce non-toxic

degradants isopropyl methylphosphonic acid (IMPA) and hydrofluoric acid.37,38 Rate of

hydrolysis is largely dependent on several factors such as temperature, pH and water

quality, and the half-lives of IMPA estimated to range between eight to 13 days.39 Hydrolysis

of sarin to IMPA could be expedited under basic pH conditions, possesses half-life of 30

seconds at pH 11 and 25°C.40 Similarly with the hydrolysis of P – F bond, soman hydrolyses

in aqueous environment under acidic and basic pH conditions to produce pinacolyl

methylphosphonic acid (PMPA) and fluoride.41,42 However, hydrolysis of soman occurs at a

slower rate as compared to sarin, taking as much as 60 hours at pH 6 and 25°C for the

formation of PMPA.41,42 Both degradant products of sarin and soman; IMPA and PMPA

respectively, degrade at a much slower rate into MPA.38 Tabun (GA) undergoes a different

degradation pathway as compared to sarin and soman (Figure 2). Under acidic pH conditions

below pH 5, tabun hydrolyses to produce O-ethylphosphorylcyanidate and dimethylamine.

Under basic pH conditions, tabun hydrolyses to produce O-ethyl N,N-

dimethylamidophosphoric acid, dimethylphosphoramidate, phosphoric acid and cyanide.

18

Figure 2: Hydrolytic pathway of tabun (GA) into its degradant products under acidic or basic pH conditions.

VX hydrolyses at a relatively slow rate under aqueous environment, and speeds up

proportionally to the increase in pH conditions.43 Unlike the rest of the nerve agents,

hydrolysis of VX can occur at either the P – O or P – S bond.28,43 Under neutral or alkaline

conditions between pH 7 to 10, VX hydrolyses to produce S-(2-diisopropylaminoethyl),

methylphosphonothioic acid and ethanol. Under pH <6 or pH >10, VX will hydrolyse into

diisopropyletylmercaptoamine (DESH) and ethyl methylphosphonic acid (EMPA) instead.10

Being an unstable hydrolysis product, DESH will either oxidise into

bis(diisopropylaminoethyl)disulfide (DES)2 or react with diisopropylethyleneimmonium ion

[(CH2)2N+(C3H7)2] to form bis(diisopropylaminoethyl) sulphide.10 As shown in Figure 1, EMPA

also hydrolyses into MPA at a slow rate. VR is less prone to hydrolysis as compared to VX,

with half-life of 12.4 days in water.44 With the cleavage of P – S bond, VR hydrolyses to

19

produce diethylaminoethanethiol and isobutyl methylphosphonic acid (iBuMPA) which goes

on to hydrolyse into MPA.22,45 However, there is no current knowledge on the conditions for

the cleavage of P – O bond to produce S-(2-diethylaminoethyl)methylphosphonothioate

from VR.22

20

4. SAMPLE PREPARATION

Sample preparation is critical for extraction of analytes from any target sample matrices and

for the subsequent analytical testing. Reliable and efficient extractions are highly

recommended for the success of any qualitative and quantitative analysis.4,15 Over the

years, comprehensive sample preparation procedures have been developed and approved

for the off-site laboratories of CWC.5,46 The Recommended Operating Procedures (ROPs)

developed for such purpose comprises comprehensive sample preparation methodologies

for solid, aqueous liquid, soil and many other types of samples.5,46

The ease of subsequent analytical testing is highly dependent on the amount, as well as the

quality of sample preparation needed to obtain a suitable extract.5 Furthermore,

distribution of analytes in their broad repertoire of matrices may be of trace levels,

therefore it is essential to bring them to limits of detection and rid of any interfering

substances.4 It is crucial to adopt simple extraction methodologies that could allow large

scale extraction of target analytes within one single extraction.47 Sample preparation

requires a comprehensive understanding of the various chemical and physical properties of

target analytes in their sample matrices and the subsequent choice of instrumental

analysis.47 CWA or their degradant products do not exist naturally and may be found in

matrices such as soil, aqueous liquid or even solid samples.5,46 Therefore, it is important to

extract them for subsequent analytical testing. Several new techniques such as liquid

microextraction, solid-phase extraction (SPE), solid-phase microextraction (SPME), on-

matrix derivatisation extraction (OMDEX) and microemulsion mediated in-situ derivatisation

and extraction (INDEX) for the extraction of CWA related compounds from complex sample

matrices have been presented in many literature resources.

21

4.1 Soil Samples

Soil samples are highly regarded due to the abundance of soil in many post chemical

warfare attacks.5 Recovery of nerve and blistering agents in soil samples are commonly

extracted with an organic solvent. Dichloromethane is reported in many studies to be the

most common solvent for the extraction of non-polar and semi-polar organic analytes from

soil, aqueous liquid and solid samples.4 However, water-soluble and non-volatile

compounds are found to be left behind after organic extraction. 5 These substances are

capable of hydrolysis at a readily fast rate in the environment, therefore, extraction with

water is incorporated to aid in the recovery.5 Centrifugation, filtration, and drying of

samples are common steps performed upon addition of the desired organic solvent.5,46

Repetition of these steps are recommended, if necessary, to ensure better recovery of

analytes.46 Degradants of organophosphorus nerve agents in soil samples of µg/g

concentration are recommended to be extracted with water using sonication, followed by

centrifugation.48,49 It is important to note that metal ions such as Na+, K+, Mg2+ and Ca2+ are

present in environmental samples, especially in soils. Divalent organophosphorus acids can

react with these ions to form insoluble salts. Contamination of chromatographic

instruments can occur, especially with divalent compounds such as MPA.5 Also, the

presence of Mg2+ and Ca2+ in soils are reported to inhibit subsequent derivatisation process

prior to chromatographic separation.1,4,5 For such cases, strong cation-exchange resin

cartridges are recommended.4,50 D’Agostino et al. reported success with the use of

chloroform for the organic extraction of CWA analytes from soil sample involved in the Iran-

Iraq conflict.51 They were able to identify mustard-related CWA compounds in their study.

22

4.2 Aqueous Liquid Samples

Similar to the sample preparation for soil samples, recovery of CWA related analytes in

aqueous liquid samples involves the use of an organic solvent such as dichloromethane or

ethyl acetate.5 Samples are required to be neutralised in terms of their pH, with either

ammonium hydroxide or hydrochloric acid.5 Following which, each neutralised sample is

subjected to liquid-liquid extraction with two portions of water-immiscible organic solvent

(ratio of 1:2 solvent to sample).4,5 This allows the separation of any non- or semi- polar

analytes in the organic extract, with the polar analytes separated into the aqueous extract.

Extraction of analytes from aqueous liquid samples often requires evaporation to dryness

which should be avoided due to the tendency of residues getting absorbed to glass

surfaces.4,5 Same procedure can be repeated separately, with an additional derivatising step

for the detection of alkaline CWA related compounds.5 Other alternatives involve the use of

ExtrelutR extraction cartridges, C18 SPE cartridges, conditioned SCX cation-exchange

cartridges.4,5,15,22 Liquid-liquid extraction was mentioned to be the preferred conventional

method, especially for analysis of complex organic extracts.4,15 Liquid-liquid extractions

allows the elimination of hydrocarbons and selective extraction of semi-polar analytes.4

Extraction with methanol and water are recommended for the extraction of polar organic

analytes. Liu et al. suggested the use of membrane filtration for studying water samples

CWA degradants of µg/cm3.1 Extraction of polar CWA related compounds such as

alkylphosphonic acids and MPA are commonly performed with water, methanol, acidified

or alkaline methanol.4 A solvent-extraction method for targeted organophosphorus CWA

compounds from painted wallboard was presented by Wahl et al.52 With dichloromethane

and acetone (ratio of 1:1), surrogate organophosphorus compounds for sarin were

23

successfully extracted; extraction efficiency of 60%.52 SPME is an excellent technique for

extraction of CWA compounds, especially for large scale extractions.18,53 This technique is

greatly recommended for aqueous and soil samples. One of the advantages offered by

SPME is its ability to incorporate derivatisation with sample extraction, and eliminates the

need for the removal of water (to be explained in later section).4,15 A SPE method using

polymeric mixed-mode strong anion-exchange (Oasis MAX) cartridges was developed by

Kanaujia et al.54 The presence of carbonyl and tertiary amine groups in polymeric backbone

of Oasis MAX cartridges are capable of eliciting secondary interactions with acidic analytes,

which aided extraction of analytes.54 This method was developed initially for extraction of

acidic CWA degradants in water but was proven to be successful in detecting

alkylphosphonic acids in soil sample too.54

4.3 Solid and other Samples

Solid samples such as active charcoal, concrete, paint, rubbers and wipes are covered under

the ROPs for CWA related compounds.5,46 Depending on the analytical purposes, extraction

methods for active charcoal requires solvents such as acetone, carbon disulphide,

dichloromethane or chloroform.5 For wipe samples, acetone, ethyl acetate,

dichloromethane and chloroform can be used.5,35 Concrete, paint and rubber samples are

required to be broken down into small fragments for extraction, with either acetone or

dichloromethane as the extraction solvent.5 All solid samples are required to be completely

submerged into the solvent. Similar to all solid samples, the organic phase will then be

decanted, filtered, centrifuged before analysis.

24

5. DERIVATISATION

Derivatisation is a procedural technique used to modify the functional group of polar

analytes, producing derivatives amenable for chromatographic separations. 8,9,55

Compounds with active hydrogen functional groups (-COOH, -NH, -OH and -SH) have the

tendency of forming intermolecular hydrogen bonds that affect their volatility; interaction

with column packing and thermal stability.55

The primary objectives of derivatisation are aimed at 55:

• Increasing resolution, with the reduced tailing of polar compounds containing active

hydrogen functional groups (-COOH, -NH, -OH and -SH).

• Allowing analysis of non-volatile compounds

• Reducing of compounds’ volatility, and/or stability prior to gas chromatography

• Enhancing analytical efficiency and detectability

5.1 Derivatisation of Chemical Warfare Agents and their Degradants

Many CWAs exhibit moderate reactive electrophilic properties while some CWA related

compounds such as phosgene, and toxic chemicals like perfluoroisobutylene, are listed in

the Annex on Chemicals of the CWC to be reactive electrophilic gases under normal ambient

temperatures.5,9,56 Both phosgene and perfluoroisobutylene are labile electrophiles capable

of reacting with nucleophiles rapidly. Reactive electrophilic properties of these compounds

result in undesirable reaction with column coatings or any free nucleophilic sites in

analytical instruments; inhibiting chromatographic analysis.9 Under ambient or sub-ambient

temperatures, perfluoroisobutylene will react with nucleophiles. It was noted that

25

underivatized phosgene can be analysed by gas chromatography, but with poor response

factor due to interference.9

Many CWA degradants are very polar compounds with insufficient volatility and thermal

instability, are chromatographic properties that contributes to peak tailing and poor

detection limits.9,18,55,57-59 As described by Black and Muir, derivatisation can be used to

improve selectivity and sensitivity of detection.9 Lewisite can be derivatised with thiol

reagents to allow sulphur selective flame photometric detection.17 Perfluorinated

derivatives produced from the conversion of phosphonic acids and thiodiglycol allows

detection by negative ion chemical ionisation-mass spectrometry (NICI-MS).17

5.2 Types of Derivatisation Reactions

Suitability, efficiency and detectability are three key factors to a successful derivatisation

prior to chromatographic analysis.55 During the derivatisation process, the derivatising

reagent replaces the active hydrogen with the functional group of the derivatisation reagent

used, rendering highly polar compounds sufficiently volatile for gas chromatographic

analysis.9,20,55 (Figure 3) The three common types of derivatisation reactions are alkylation,

acylation and silylation.60 (Table 2)

Figure 3: Chemical equation for general derivatisation reactions where ‘A’ denotes nitrogen,

oxygen or sulphur, while ‘D’ denotes the functional group of derivatisation reagent used.

26

Table 2. Three different types of derivatisation reactions with their corresponding derivatising reagents.55

Alkylationa Silylationb Acylationc

Reagent Reagent Reagent

Dimethylformamide (DMF) Diazomethane Pentafluorobenzyl bromide (PFBBr) Pentafluorobenzyl-hydroxylamine hydrochloride (PFBHA) Benzylbromide Tetrabutylammonium hydroxide (TBH) Boron trifluoride (BF3) in methanol or butanol

Bis(trimethylsilyl)-acetamide (BSA) N,N-bis(trimethylsilyl)trifluoroacetamide (BSTA) N-methyl-trimethylsilyltrifluoroacetamide (MSTFA) Hexamethyldisilzane (HMDS) Trimethlsilylimidazole (TMSI) Trimethylsilildiethylamine (TMS-DEA) Trimethylchlorosilane (TMCS) N-methyl-N-t-butyldimethylsilyltrifluoroacetamide (MTBSTFA)

Fluoracylimidazoles Fluorinated Anhydrides N-methyl-bis(trifluoroacetamide) (MBTFA) Pentafluorobenzoyl Chloride (PFBCI) Pentafluoropropanol (PFPOH)

aIn an alkylation reaction, the active hydrogen is replaced with an aliphatic or aliphatic-aromatic group. bIn a silylation reaction, the active hydrogen is replaced with a silyl group. cIn an acylation reaction, the active hydrogen is replaced with an acyl group.

The choice of derivatisation reagent is dependent on many factors. In a comprehensive

review written by Orata, the criteria for choosing a suitable derivatisation reagent are as

follows55:

• The reagent should produce more than 95% complete derivatives.

• It should not lead to any rearrangements or structural modification of compounds

during the formation of the derivative.

27

• It should not contribute to loss of the compound of interest during derivatisation.

• It should not produce any unstable derivatives and/or derivatives that will interact

with the chromatographic columns.

5.2.1 Common Derivatising Techniques in CWA Studies

The most commonly used derivatising reagents for OPCW verification analysis are

diazomethane, BSTFA and MTBSTFA.5,9,11,61 Diazomethane has several advantages over

many derivatisation reagents. Despite its inability to methylate thiodiglycol, derivatisation

with diazomethane is considered by many authors to be the most efficient approach due to

its ability to produce methyl ester rapidly.5,55 Also, the yield of methyl derivatives is reported

to be relatively high, with minimal side reactions.9,58 Some of the characteristics of methyl

derivatives are attributed to their short lifetime, insensitivity to trace amount of moisture

and weak signal of molecular ions upon electron ionisation.62 Fragments formed from

electron ionisation are found to complicate trace analysis in selected ion monitoring (SIM)

mode.62 In a study conducted by Enqvist and Rautio, diazomethane was used to derivatise

alkylphosphonic acids with methanol for the formation of methyl esters. Derivatisation was

achieved in 15 minutes, with yields >99%.58 Diazomethane is also known for its selective

reactivity with acidic analytes, and high volatility to facilitate easy removal of excess

reagent.17,56 Identification as a methyl ester is especially convenient for laboratories with no

access to a liquid chromatography combined with mass spectrometry (LC-MS).5,55

Despite the advantages offered by diazomethane, many laboratories prefer silylation with

BSTFA and MTBSTFA. According to the ROPs, TMS derivatives were typically used for on-site

analysis associated with inspections while t-butyldimethylsilyl (TBDMS) derivatives for off-

site analysis. 5,46 Silylation involves the substitution of active hydrogen with a trimethylsilyl

28

(TMS) group, resulting in reduced polarity of the target compound and hydrogen

bonding.55,63 Derivatives produced by these BSTFA and MTBSTFA are reported to possess

outstanding thermal stability.9,55 Furthermore, the presence of trifluoroacetyl group in

BSTFA and MTBSTFA allows quick and easy separation with low detector noise.55 The

recommended optimal conditions for derivatisation with MTBSTFA is at 60°C for an hour.55

Silylation can be enhanced with an addition of a catalyst. For example, BSTFA is may be used

with either 1 or 10% TMCS catalyst for better results.5,55 BSTFA can also be used with 1%

trimethylsilyl chloride (TMSCI), with recommended conditions at 60°C for 30 minutes.5

However, Creasy et al. was able to achieve 80-100% derivatisation efficiencies with the use

of BSTFA with 1% TMSCI in hexane at 60°C for 15 minutes.9,64 Many literature resources

supported the addition of derivatisation reagent with TBDMS group over TMS.13,18,55,65,66

One disadvantage of TMS derivatives is their tendency to hydrolyse easily by traces of

water.5,67 TBDMS derivatives are reported to provide better hydrolytic stability as compared

to derivatives produced by TMS reagents, with longer stability.9,55 TBDMS derivatives are

thermally stable esters less sensitive to moisture traces as compared to TMS esters.11 It was

even suggested by Orata that TBDMS derivatives can have 10,000 times the stability of TMS

derivatives.55 Furthermore, TBDMS derivatives are also reported to produce better mass

spectral properties (intensity of high mass ions; [M-57]+ ion) as compared to TMS

derivatives.9 However, it was suggested that the EI spectral of methyl alkylphosphonates

produce more distinctive organophosphorus-containing fragments as compared to silylated

alkylphosphonates.5 One notable advantage of TMS derivatives over TBDMS derivatives is

the derivatisation of thiodiglycol sulphoxide; a degradant of mustard agent.

29

Figure 4: Reaction scheme of the methyl, trimethylsilyl, tert-butyldimethylsilyl, and pentafluorobenzyl derivatives of MPA. Adapted and modified from Koryagina et al.11

Derivatisation can enhance sensitivity of detection in subsequent analytical testing. For

example, pentafluorobenzyl derivatives have been reported to be suitable for cases that

requires low level of detection of CWA related analytes.5,9,68 Derivatising reagents with

fluorine (perfluorobenzylation, trifluoroacetylation and heptafluorobutylation) are known to

enhance sensitivity.5 Derivatised phosphonic acids with pentafluorobenzyl bromide (PFBBr)

offers trace level detection as low as 1 ng/mL. Furthermore, these derivatives contribute to

sharp symmetrical peaks and longer retention times in GC, as compared to methyl

derivatives. Pentafluorobenzyl are well-suited for NICI-MS, and have reported lowest LOD.68

However, derivatisation with PRBBr is a relatively slow and complicated process.9,68

30

t-Butyldimethylsilyl cyanide (TBDMSCN) was suggested by Black and Buir to be a possible

alternative derivatising reagent for TBDMS derivatives.9 However, TBDMSCN was only used

in two studies, for the cyanosilylation of ketones and aldehyde. The use of TBDMSCN was

suggested by Fetterly and Verkade to be useful because of the easy removal of TMS group.69

In another study, Wang et al. used TBDMSCN with tin ion-exchanged montmorillonite (Sn-

Mont) as the catalyst for the cyanosilylation of ketones and achieved >98% yield of

cyanohydrin TBDMS ethers.70 There are no literature resources describing the use of

TBDMSCN in CWA studies.

5.3 Disadvantages of Derivatisation

Even though the enhancement of analyte’s reactivity, stability, volatility and detectability

allow ideal conditions for analytical separation, there are several disadvantages attributed

to derivatisation. Being the most common and recommended derivatising agents for OPCW

verifications, diazomethane, BSTFA and MTBSTFA are found to pose several problems.

Diazomethane are known for its toxicity, as well as its capacity for detonation.5,9,55 Methyl

esters produced by diazomethane are reported to produce poor chromatographic

properties, and peak shapes.17,55 One notable disadvantage is attributed to the ambiguity of

methyl ester identification.55 Methyl esters are undesirable for trace analysis due to the

highest mass ions falling below m/z of 200 (e.g. MPA has a m/z of 95).9,55 With regards to

the use of BSTFA or MTBSTFA, silylation of MPA has been reported to a huge issue for soil

analysis, even to the extent of failing OPCW tests for MPA due to its extreme sensitivity

towards calcium and magnesium ions.9 In studies conducted by Kataoka et al., they

performed extraction on derivatised alkylphosphonic acids and methylphosphonic acid from

soil, seawater and beverage samples and achieved low extraction and derivatisation

31

recoveries. They concluded that the yield for TBDMS derivatives, especially MPA, was

significantly affected by the presence of calcium and magnesium ions in their samples.71-73

Derivatisation has been reported in many literature resources to be a significant source of

error.9,15 Derivatisation of target compound requires the elimination of water;

concentration of aqueous target analytes to dryness.9,15,18 This process is described by many

authors to be highly time-consuming, inconvenient and a leading source of error.9,15,18

Furthermore, remaining traces of water are also reported to react with both the derivatising

reagent and derivative, with occurrence of losses on evaporation.9 Foreign substances

extracted from sample matrices are capable of inhibiting or interfering derivatisation,

requiring suitable sample extraction and clean-up procedures.9

5.4 Current Trends in Derivatisation

In a review written by Wells, it was outlined that the most prevalent method of

derivatisation is through trimethylsilylation.74 In his review, he went further to explain about

the other methods and advances in derivatisation. The use of fluorinated acyl derivatives,

on-column methylation, hollow fiber liquid phase microextraction (HF-LPME), and SPME

were reported to be increasingly popular choices of derivatisation.9,74 There are a number of

ways to introduce an acyl group to produce fluorinated derivatives, as illustrated in Table 2

above. Fluorinated acyl derivatives allow sensitive detection by electron-capture detection

(ECD) and NICI-MS.74 One common problem associated with derivatisation was the need to

use large quantities of derivatising reagents to drive reactions to complete.9,74 Excess

reagents lead to inherent background and column of the chromatographic instruments.9

Fluoracylimidazoles, fluorinated anhydrides and MBTFA derivatising reagents are found to

be able to overcome such problems.75 The use of solid support for derivatisation to hold

32

analytes, reagents and catalyst is one unconventional application many chemical warfare

researchers are looking into.76 With supported reagents, the problem of excessive reagent

can be overcome easily.76 In SPME, analytes are supported by the solid phase, and can be

coupled with in-situ derivatisation. Examples of such techniques will be explored in the later

section. Strategies of supported analyte derivatisation such as the use of ion-exchange resin

prior to the pentafluorobenzylation of acidic analytes in-situ, and 9-fluorenylmethyl

chloroformate for the adsorption of derivatised amino acids and peptides derivatives onto

silica are concepts many biomedical researches are looking into.77,78

5.4.1 Derivatisation Without Removal of Water

One common goal for the development of new derivatisation procedures is aimed at

overcoming the requirement for the removal of water, which is often characterised as time

and labour consuming, and source of error. 9,15 Solid-phase microextraction (SPME) has the

capacity to overcome such difficulties of derivatisation.9,10,18 This sample preparation

technique was originally used in gas chromatographic purposes for the determination of

volatile and semi-volatile substances in a sample matrix.10,79 This technique is based on the

use of a coated fiber for partitioning of target analytes between the sample of interest and

itself. 79 This coat could be a liquid polymer or a solid sorbent, or combination of both, for

the extraction of target analytes. Being a flexible analytical procedure, derivatisation of

analytes can be rearranged into three different modes (Figure 5).

33

Figure 5: Three different modes for derivatisation methods associated with SPME; Direct derivatisation in sample matrix, derivatisation in SPME fiber coating, or derivatisation in GC injector port. With derivatisation in SPME, analytes are converted from its unstable form into a thermally

stable, yet easily-extractable compound. Stashenko and Martinez suggested that

derivatisation with SPME allows better stability and reduced polarity of target analyte.79 In-

situ derivatisation with SPME was first performed by Pan et al. in 1995 to enhance the

sensitivity of carboxylic acids prior to analysis by GC.80 Following which, Sng and Ng applied

the concept of in-situ derivatisation with SPME to the analysis of CWA related compounds

and degradants in water.18 In their study, SPME fibres were exposed to vials containing the

derivatising reagent followed by contact with aqueous solutions of CWA analytes.18 After

stirring of these solutions, fibre was exposed to the same derivatising reagent before

analysis with GC-MS.18 Being a non-destructive extraction method, water samples are able

to be analysed repeatedly by SPME for volatiles, semi volatiles and CWA related

compounds. This potentially allows flexibility in CWA analysis, coupled with the advantage

of keeping sample depletion to the minimum. This technique was demonstrated in one of

34

the proficiency tests organised by the OPCW, and was proposed to be better than the ROPs

for on-site analysis.18

In another study, HF-LPME was found to be successful for the analysis of CWA degradant

without the removal of water.2,12 HF-LPME is a technique based on the use of a hollow fibre

containing an organic solvent for the separation of analytes of interest from aqueous

sample.2,12 With this method coupled with in-situ derivatisation, Lee et al. performed

extraction of CWAs degradants from aqueous samples using a solvent mixture (BSTFA and

octane in the ratio of 5:1) and MTBSFTA.12 Being a moisture-sensitive derivatising reagent,

MTBSFTA was protected within the hollow fibre, thus allowing successful extraction of

analytes prior to analysis by GC-MS. This methodology was reported to be simple,

convenient and has better LODs as compared to in-situ derivatisation with SPME.12

There has been a lack of publications for the derivatisation of CWA degradants without

removal of water besides the alternative method of using SPME. Some studies described the

use of chloroformate as an alternative for derivatisation in aqueous media without the need

to remove water.81-84 Several derivatisation strategies with alkyl chloroformate have been

presented and reviewed by Husek and Simek.85 Despite the advantages, it was suggested

that a post clean-up process was necessary for the removal of any excess and by-products

which is detrimental to GC instruments.55 Furthermore, there is no publications associated

with the derivatisation of CWA related compounds.

Earlier this year in 2018, a derivatisation procedure was developed and proven to be

successful without the necessity to evaporate aqueous solutions to dryness.20 In Dival’s

study, MPA was derivatised with MTBSFTA, with addition of organic layer of hexane.

Derivatisation of MPA was carried out at 60˚C for 30 minutes prior to analysis by GC-MS.

35

Qualitative detection of MPA derivatives was reported to be successful, however, with LOD

of 1000 ppm. For practical considerations, it is essential to ensure safe and smooth

derivatisation process with a positive yield of stable product. Besides that, detection of MPA

at trace level may pose a problem. Concentration of CWA degradants for OPCW analysis was

suggested to be between 1 to 10 ppm.5,31,46,86

36

6. DETECTION & QUANTITATION OF CHEMICAL WARFARE AGENT DEGRADANTS

Ever since the implementation of CWC in 1997, OPCW and many other research groups have

been developing, improving and executing methods for qualitative detection and

quantitative analysis of CWAs, their precursors and degradants.5,8 In accordance to the

OPCW inspectorate, at least two or more different spectrometric techniques and reference

standards are required for the unambiguous identification of CWA and their related

compounds.5 The most common chromatographic separation methods for such purposes

are based on GC-MS, or gas chromatography with tandem mass spectrometry (GC-MS/MS),

LC-MS, or liquid chromatography with tandem mass spectrometry (LC-MS/MS), and nuclear

magnetic resonance (NMR) spectroscopy.5,9,16,87

The most prevalent chromatographic separation method for the identification of CWA and

their related compounds is found to be GC.8,88 CWA compounds are identified by retention

indices (RI) or relative responses of GC detectors.15 Besides being described to be an

efficient and simple instrument, GC offers selective and sensitive detection of target

analytes. GC can be combined with different detectors such as flame photometric detection

(FPD), flame ionisation detection (FID), photoionisation detection (PID), or nitrogen-

phosphorus detection (NID) for the detection of compounds with phosphorus core, such as

the organophosphorus nerve agents.8 On the other hand, liquid chromatography is known

for its lack of detectors matching the sensitivity and selectivity of GC detector systems.16,89

Liquid chromatography has the capacity of separating polar and non-volatile

37

alkylphosphonic acids.8,89 Unlike GC, LC offers high versatility and direct analysis of

samples.16 Analysts do not have to take into account of target analytes’ thermolability,

polarity, or volatility. However, direct analysis with LC is only limited to compounds with

chromophores and/or fluorophores.8,16 CWA degradants such as methylphosphonic acids

and alkylphosphonic acids cannot be detected by UV nor fluorescence detectors.8 Many

CWA degradants are undetectable electrochemically or by conductivity.16 However, these

compounds exhibit sufficient ionisation and fragmentation in mass spectrometry where

solutes can be accurately detected by their molecular mass.16 As such, mass spectrometric

detectors plays a huge role to compensate for limitations as described above. Mass

spectrometry has been widely used for CWA studies in accordance to the CWC. Mass

spectrometers can achieve excellent results from analyses based on detection, identification

of specific CWA compounds.90

6.1 Gas Chromatography-Mass Spectrometry

Gas chromatography in combination with mass spectrometric detectors remains to be one

of the most widely used technique in CWA studies due to its excellent sensitivity and

selectivity, as well as the presence of large libraries of mass spectra.88 GC-MS allows

sensitivity in analysis with the SIM, as well as recording in SIM and SCAN modes within the

same run. Ionisation techniques such as electron impact ionisation (EI) and chemical

ionisation (CI) are very useful for studying CWA related compounds in environmental

samples.16 In a study carried out by Katagi et al., GC-MS with CI mode was suggested for the

detection of EMPA in serum, achieving LOD of 50 ng/mL in total ion current (TIC) mode and

3 ng/mL in the SIM mode.91 In the method, extraction of EMPA was performed with

acetonitrile, with MTBSTFA derivatising reagent + 1% TBDMSCI catalyst.91 The common

38

method utilised for the analysis of CWA degradants is GC-MS with EI.61 EI mass spectra for

these compounds is easily available and can be found in the OPCW Central Analytical

Database (OCAD).7,16,25

In a study with GC-MS in EI mode, Kataoka and Seto were able to quantitatively determine

TBDMS derivatives of MPA in plasma and urine matrices, but only achieved yield as low as

8%.66 Singh et al. presented a study on the use of magnetic sorbent for the extraction of

CWAs from organic liquid samples prior to analysis by GC-MS in EI mode, with quantitation

study in the SIM mode.6

A comparative study of two ionisation methods; GC-MS with atmospheric pressure

photoionisation (APPhI) and GC-MS with photochemical ionisation (APCPhI) were presented

by Yashin et al. for the qualitative detection and confirmation of molecular weights of TMS

and TBDMS derivatives of MPA.61 Confirmation and identification of derivatives of MPA was

successful with both methods; TMS- and TBDMS derivatives of MPA were detected at

picogram level. However, GC-MS with APCPhI achieved tenfold lower LOD as compared to

GC-MS with APPhI.61 With selectable one- and two- dimensional (1D/2D) GC-MS, Seto et al.

were able to allow target determination of TBDMS derivatives of alkyl methylphosphonic

acid (AMPA) and MPA selectively from urine samples and acid-fortified solutions.13 1D/2D

GC-MS allows alternating between 1D and 2D GC modes within the instrument for selective

separation of analytes.13 Analytes that are unable to co-elute with matrix components can

be separated with 1D GC mode, while the parts of chromatogram inhibited by the co-elution

of matrix components can be moved and separated by 2D GC mode.13 Baseline separation of

TBDMS derivatives of MPA were separated within three minutes, with LOD of 20 ng/mL.13

39

GC-MS/MS provides excellent accuracy and stability of results over an extensive range of

measurements. Selective screening of TMS derivatives of many AMPA from complicated

sample matrices was found to be highly effective using GC-MS/MS.92 Rohrbaugh and Sarver

were able to achieve excellent selectivity and sensitivity from the base peak of m/z 153

parent ion to the m/z 75 product ion under EI conditions.92 It is also possible to reach ultra-

trace LOD of 30 – 70 pg in SIM with m/z of 99 - 79 for sarin and soman.50 SIM or multiple

ion monitoring (MIM) can be used for cases requiring low detection limits. It was described

that SIM has the capacity of detecting <1 pg per injections under optimal conditions.5 A

factor of 100 -1000 times enhanced sensitivity can be achieved. This setting is dependent on

the compound, as well as the background of sample matrices and GC column. However, GC-

MS with MIM is required to be performed on at least three of a mass spectral (>10% of the

base peak) intensity in order to facilitate a proper unambiguous identification of analytes.5

In a study for the trace level of AMPA in environmental and biological samples, GC-MS/MS

was successful in the detection of AMPA (derivatised with PFBBr) in the negative CI mode.93

Sensitivity as low as femtogram was achieved in a study of CWA degradants derivatised by

PFBBr.93 High resolution GC-MS/MS allows determination of CWA analytes in complex

sample matrices with excellent selectivity at low LOD.5 Under EI and CI conditions, GC-

MS/MS was successful with identification of TMS derivatives of alkyl methylphosphonic

acids in complicated sample matrices. In cases where identification of CWA analytes in

complicated sample matrices, the additional dimension of tandem MS to GC-MS could be

employed.

The major advantage of using GC-MS is for the unambiguous identification of CWA

degradants. However, derivatisation is often necessary due to the high polarity and low

volatility of CWA degradants, prior to analysis by GC-MS. Moreover, derivatisation is crucial

40

in recording of EI spectrum for qualitative detection, and adequate sensitivity for SIM. As

explained in earlier section, derivatisation incur a time-limiting factor with the tendency to

produce artefacts that may lead to low or even non-detection of target analytes. Thus, LC-

MS may be better suited for handling these compounds.

6.2 Liquid Chromatography-Mass Spectrometry

LC-MS is utilised for the analyses of CWAs and related compounds since the 1980s owing to

its broad range of interfaces and ionisation methods.90 This analytical technique has been

used primarily for the direct analysis of aqueous, biological and soil sample matrices. For LC-

MS analysis of organophosphorus compounds, Riches presented a set of general parameters

and settings as seen in Table 3 below.

Table 3. General LC-MS parameters used for analysis of organophosphorus compounds.94

LC-MS parameters Setting

Stationary phase C18 column: 150 by 2.0 mm (3 or 5 μm particle size)

Mobile phase composition

APCI A: 20 mM ammonium formate (or ammonium acetate) in water B: 20 mM ammonium formate(or ammonium acetate) in methanol

ESI A: 0.1% formic acid in water B: 0.1% formic acid in acetonitrile

Mobile phase flow rate

200 μL/min

Mobile-phase gradient

5% B (0–5 min) to 90% B (15 min). Hold at 90% B (5 min)

41

MS source conditions

APCI Vaporiser temperature = 400 °C Corona current = 4–6 μA ESI ESI voltage = 3–5 kV

Collision induced dissociation (CID)

5–25 V

Mass range m/z 40–400

Just like GC-MS, the basic parameters revolves around the stationary and mobile phases,

flow rate, elution type, along with the different types of MS sources and detectors.89,94 The

use of spectrophotometric or UV detectors (e.g. HPLC-DAD) are either highly limited or

impossible for detection of CWA degradants due to the lack of chromophore groups in

them; insensitivity and low-informative absorption spectra.95-98

A number of LC-MS applications for the analysis of nerve agents and their degradants are

listed below in Table 4. As explained earlier, mass spectrometers interfaced to LC allow high

sensitivity and direct analysis of many CWA degradants. Fragmentation of MPA alkyl esters

begins with the breaking of the C – O bond, separating the alkyl radical, together with the

formation of [M – H]- ions at m/z 95 and [M + H]+ ions at m/z 97 in the negative- or positive-

ion modes of LC-MS, respectively.88 A common fragment ion among soman, sarin and

cyclosarin is found to be m/z 99 due to the loss of an alkyl moiety from a R – O group

(alkoxy) with proton transfer. There is a lack current literature resources for the

chromatographic separation and identification of VR’s degradant products and

bis(diisopropylaminoethyl) sulphide.45 Most of the studies involving the degradant products

of nerve agents employs the use of electrospray ionisation (ESI) sources in positive ion mode

42

(Table 4). The choice for positive- and negative- ion modes is mainly for the selectivity or

sensitivity aspect of instrument.99

Table 4. LC-MS parameters for the analysis of CWA related compounds.

Compounds of Interest

Sample Matrix

Ionisation Stationary & Mobile Phases

References

• Alkyl methylphosphonic acids (AMPA)

• Alkyl ethylphosphonic acids

• Alkyl alkylphosphonic acids

• Dialkyl alkylphosphonates

Water ESI in positive ion mode

Mixed C8/C18 (250 X 2.1mm) column

MeOH/water (0.1%formic acid) gradient at 3200 μL/min.

97

• GB

• GD

• GF

Water ESI in positive ion mode

C18 (150 X 0.32 mm) column Acetonitrile/water (0.1% trifluoroacetic acid) gradient at 5 μL/min

100

Derivatized (p-bromophenacyl) alkyl phosphonic acids

Water Fast atom bombardment in positive ion mode

C18 (150 X 1.5 mm) column Acetonitrile/water (0.005 M ammonium acetate) isocratic at 100 μL/min

101

43

• Alkylphosphonic acids

• AMPA

• Alkyl ethylphosphonic acids

Water Ion Spray in negative ion mode

PGC (150 X 2.1 mm) column ACN/water (trifluoroacetic acid) gradient/isocratic at 200 μL/min.

98

• GB

• GD

• GF

• Degradants of GB, GD and GF

Water ESI in positive ion mode

C18 (150 X 0.32 mm) column Acetonitrile/water (0.1% trifluoroacetic acid) gradient at 16 μL/min

49

• GB

• IMPA

• MPA

Water ESI in positive ion mode

C18 (150 X 0.32 mm) column Acetonitrile/water (0.1% trifluoroacetic acid) gradient at 16 μL/min

102

AMPA Water Soil

ESI in negative ion mode

C18 (150 X 2.1 mm) column Methanol/water (0.01M ammonium formate) gradient at 200 μL/min

1

• GB

• GD

• IMPA

• PMPA

• MPA

Soil ESI in positive ion mode

C18 (150 X 0.32 mm) column Acetronitrile/water (0.1% trifluoroacetic acid) gradient at 16 μL/min.

49

44

IMPA Serum ESI in both positive and negative ion modes

PRP-X100 (200 X 0.32 mm) column. Acetonitrile/water (0.5% formic acid) isocratic at 20 μL/min.

103

LC-MS is especially effective for studying alkyl alkylphosphonic acids, alkylphosphonic acids

and degradants of VX agents using thermospray ionisation. Thermospray mass spectrometry

(TS-MS) was found to be the most successful technique associated with LC-MS, followed by

atmospheric pressure ionisation (API).16,95,104 API-MS are capable of working on polar and

non-polar compounds with various flow rates and mobile phase compositions.16

Furthermore, API-MS can be combined with collision-induced dissociation (CID), to allow

acquisition of CWA degradants’ structural information which was suggested to be a major

advantage over other instrumental analysis.16 It can also be categorised into ESI or

pneumatic-assisted ion spray and atmospheric pressure chemical ionisation (APCI-MS);

latter being most prominent technique used in CWA studies.8,16

A series of LC-MS methodologies for the determination of CWA degradants were compiled

and reviewed by Aleksenko.88 In his review, the optimal conditions of separation and

identification of such compounds in environmental and biological samples, with various

ionisation techniques were discussed.88 The LOD of LC-MS are found to be much better than

their maximum permissible concentrations. An excerpt from his review for the

determination of MPA using LC-MS methodologies:

45

Beverages

Water, so

il

Stand

ard

Solu

tion

s

Soil

Samp

le

type

Table 5

.

Sum

mary o

f HP

LC-M

S meth

od

olo

gies for th

e determ

inatio

n o

f MP

A fro

m vario

us sam

ple m

atrices. 88

Water:

me

mb

rane

filtration

.

Soil:

ultraso

nic

water

extraction

- Extraction

with

meth

ano

l,

extraction

w

ith w

ater,

evapo

ration

Extraction

Meth

od

PR

P -X

- 10

0, 5

μm

(10

0 × 4

.6) P

RP

-X- 1

00

, 5 μ

m (1

50

× 2

.1)

Zorb

ax C1

8SB

, 5 μ

m(1

50

× 2.1

), tco

l = 25

°C

Sph

erisorb

S50

DS2

(15

0 x 2

)

Sph

erisorb

S50

DS2

(15

0 x 2

)

Station

ary Ph

ase, Particle size

(colu

mn

, mm

)

Grad

ient elu

tion

1: fro

m 0

.5%

fo

rmic acid

in w

ater (pH

~ 2.3

)

con

tainin

g 5 vo

l % o

f meth

ano

l to

0.3

M am

mo

niu

m fo

rmate (p

H ~

2.3

) con

tainin

g 22

vol %

of

meth

ano

l (10

00

). Grad

ient elu

tion

2: 1

0 m

M am

mo

niu

m carb

on

ate (p

H ~ 8

.5)—

50

mM

amm

on

ium

carbo

nate (p

H ~ 8

.5)

Grad

ient elu

tion

: from

10

mM

am

mo

niu

m fo

rmate in

water to

10

mM

amm

on

ium

form

ate in

meth

ano

l (20

0)

Grad

ient elu

tion

from

1 m

M

tribu

tylamin

e solu

tion

in w

ater to

1 m

M in

meth

ano

l (30

0)

Grad

ient elu

tion

from

1 m

M

tribu

tylamin

e solu

tion

in w

ater to

1 m

M in

meth

ano

l (30

0)

Eluen

t (flow

rate, µL/m

in)

ES, ICP

ES, TOF

ES,

MS/M

S

ES,

MS/M

S

Detecto

r

105

1 97

106

Referen

ce

46

Acidic phosphonic acids such as MPA, exist in deprotonated form under typical LC

conditions, thus suitable for both negative- and positive ionisation.5 Black and Read

presented three studies on the use of positive and negative APCI and ESI screening

procedures for the CWA degradants.59,96,97 In the first study, LC-APCI-MS was used for the

screening various CWA degradants with a C8/C18 reversed-phase column with gradient

elution (0.05% trifluoroacetic acid in water and acetonitrile). LODs in the positive SIM mode

were within detectable range of between 10 to 400 ng/mL, however, MPA was found to

have the worst sensitivity in this study.96 With the replacement of 0.1% formic acid for

trifluoroacetic acid as the acid modifier, Black and Read were able to achieve better

sensitivity of the detection of phosphonic acids with LC-ESI-MS; LOD of <50 ng/mL.97 An

explanation for the increase of sensitivity was described by Zhou and Hamburger that

trifluoroacetic acid suppresses, and formic acid enhances formation of [M + H]+ ions with

ESI.107 In the following study, C18 column (150 x 2.1 mm) with ammonium formate-water-

methanol gradient elution was used, which enabled the use of positive- and negative-mode

of ACPI and ESI.59 Optimal sensitivity and selectivity were successful with neutral and basic

analytes, and acidic analytes with positive- and negative ionisation mode, respectively.59

Combination of these two screening modes with mass spectrometry allowed rapid detection

and identification (LOD of 10 – 100 ng/mL) of many CWA degradants in aqueous samples.

Chromatograms for phosphonic acids are found to be cleaner with negative ion as

compared to positive ion.108 It was also suggested that the capacity for the formation of

variable adduct ion (e.g., [M + H]+ or [M + Na]+) in APCI makes LC-APCI-MS much more

favourable technique as compared to LC-ESI-MS.59 However, the acquisition of mass spectra

obtained from either API technique would remain similar in terms of ion content. Adduct

ions produced by either technique for the determination of molecular mass and

47

characteristic product ions will nevertheless indicate for analytes’ structure.49 LC-ESI-MS is

found to be better suited for lower flow rates associated with packed capillary LC

separation.109 One notable disadvantage of LC-ESI-MS is its inability to ionise mustard

agents.110 Even though LC-MS with APCI, ESI, or TS (soft ionisation techniques) are

suggested to be well-suited for the rapid screening and preliminary identification of

alkylphosphonic acids in aqueous samples or extracts, GC-MS has better identification and

sensitivity for such compounds. Also, these soft ionisation techniques only produce simple

molecular ions without fragmentation of molecules. As such, LC-MS is more commonly

employed as a complementary technique or for rapid initial screening method, but never a

replacement for GC-MS.8,88

Gradient separations with C18 column allows flexibility for the analysis of a variety of target

analytes ranging from small, polar CWA degradants (e.g. MPA and thiodiglycol), to polar

CWA compounds of larger size.111 However, C18 columns are found to provide poor

retention for polar CWA degradants.111 Mercier et al. recommended the use of porous

graphitic carbon as the stationary phase to enhance the chromatographic separation of

phosphonic acids by LC-MS.98,112 Sorbent with such surface induces both electronic and

hydrophobic interactions; better hydrophobicity and stability as compared to a C18

sorbent.98 Polymeric packings allows better separation of phosphonic acids than on silica

packings due to the combination of apolar and ionic interactions between solutes and the

stationary phase.98 Separation of MPA from the various phosphonic acids achieved within

15 mins in mobile phase of 0.1% trifluoroacetic acid and acetonitrile in isocratic mode.98

Besides ensuring optimal pH and its non-reactivity with components within a solution, it is

essential to take into account for possible presence of adducts produced in high levels by

48

the components of buffer compounds which may have reacted degradants of sarin, soman

and cyclosarin.113 Gäb et al. discussed about the amino function of buffer compound acting

as an intramolecular proton acceptor to turn its hydroxyl groups nucleophilic, which is

reactive with the phosphorus core of many CWA degradants.113 In their study, sarin, soman

and cyclosarin reacted with TRIS and TES buffers to form stable adducts, which were

detected by LC-ESI-MS/MS system.113

An additional dimension to LC-MS (LC-MS/MS) allows trace detection of CWA analytes.

However, there was no criteria established by the OPCW for trace detection by LC-

MS/MS.5,114 The lowest LOD was suggested to be obtained in negative ion modes.25 With

LC-MS/MS in negative ion detection mode, Baygildiev et al. achieved satisfactory

reproducibility and selectivity for the determination of MPA in human blood plasma; LOD of

3 ng/mL.115 In the study, separation was performed on a C18 column, with mobile phase

(0.1% formic acid in water and acetonitrile) gradient elution mode with flow rate of

0.4 mL/min.115 Samples of the averaged plasma were spiked with MPA ranging between 3 to

500 ng/mL for the development of calibration curve, LOD and LOQ.115 Rodin et al developed

an approach for the determination of a number of VX’s degradants in aqueous

environmental samples with HPLC-MS/MS.45 The aim of their study was to optimize the

conditions of the mass spectrometric detection and the gradient elution of the mobile

phase.45 Mass selective ion trap instrument equipped with ESI source in the selected

reaction monitoring mode using positive ions was employed. Separation was carried out in

gradient elution, with mobile phase consisting of 0.5% formic acid in water and acetonitrile;

produced the best chromatographic peaks.45 To optimize spectrometric detection, optimal

ion pairs of ion transitions were selected. Rodin et al. were able to achieve desirable

accuracy, reproducibility and selectivity with their proposed methodology.45 Analysis of

49

p bromophenacyl derivatives of alkyl alkylphosphonic acids in human serum and

environmental water with LC-MS/MS was presented by Katagi et al.101 Continuous flow frit

fast atom bombardment on hybrid high resolution magnetic sector quadrupole was used,

and achieved LOD range of 1-20 ppb.101 Even though the proposed technique reported good

reproducibility and sensitivity, it requires the removal water before derivatization.

In 2017, Baygildiev et al. developed a time-efficient procedure for the quantitative

determination of MPA in aqueous samples (natural water spiked with known concentration

of MPA), with LC-MS/MS equipped with APCI and ESI sources.3 The R2 value and the

calibration curve achieved in their study was 0.99 and y = 570x, respectively. This developed

technique achieved LOD of MPA in natural waters to 10 ng/mL, which is remarkable since

the LOD for detection of MPA with LC-MS was suggested to only be 50 ng/mL.3 Peak shape

and sensitivity are highly dependent on both buffer concentration and mobile phase

composition.25 In order to obtain the most suitable mobile phase composition, Baygildiev et

al. experimented on the use of different solutions for the first and second components, as

well as exploring the relationship between peak intensity and the pH of mobile phase.3

Mixture of 0.5% formic acid and 5 mM ammonium acetate was used as the first component

of the mobile phase, while acetonitrile was used for the second component. Fragmentor

voltage and collision energy were studied to allow the best sensitivity; optimisation of MS

detection parameters. Baygildiev et al. went further to enhance the mass spectrometry by

increasing the nebulizer gas flow rate and temperature. These steps and parameters

discussed and taken by Baygildiev et al. and other analysts are crucial and should be

considered for development and enhancement of LC-MS procedures for the detection

and/or quantification of any target analytes.

50

CONCLUSION

In the current society, GC-MS represents the most conventional and primary separation

technique for the unambiguous identification of chemical warfare degradants. However,

derivatisation of chemical warfare degradants requires the evaporation of aqueous extract

to dryness, which is a complicated process and major source of error. Presented approach

by Dival for the derivatisation of MPA without the elimination of water was successful in

producing GC-amenable derivatives for GC-MS, with LOD to be 1000 ppm. EI mass spectra

data from analytical techniques have been acquired over the years for the verification

purposes by OPCW where identification of CWA compounds must be confirmed by at least

two different spectrometric methods.5 It is of high importance to consider and evaluate all

methods and instrumental parameters involving the LC-MS analysis of MPA in order to

develop a robust method to quantitate the efficiency of the two-phase derivatisation. The

types of derivatising reagent, stationary and mobile phases, detection settings, ionisation

techniques and conditions play crucial roles to generate an accurate calibration curve for

the determination of MPA via LC-MS. As such, the development of a reliable, reproducible

and time-efficient analytical protocol would achieve a quantitative measurement of MPA

derivatisation, as well as serving as a baseline LC-MS, if not, a secondary analytical method

for the verification purposes by OPCW.

51

REFERENCES

1. Liu Q, Hu X, Xie J. Determination of nerve agent degradation products in

environmental samples by liquid chromatography–time-of-flight mass spectrometry with

electrospray ionization. Analytica Chimica Acta. 2004;512(1):93-101.

2. Desoubries C, Chapuis-Hugon F, Bossée A, Pichon V. Three-phase hollow fiber liquid-

phase microextraction of organophosphorous nerve agent degradation products from

complex samples. Journal of Chromatography B. 2012;900:48-58.

3. Baygildiev TM, Rodin IA, Stavrianidi AN, Braun AV, Akhmerova DI, Shpigun OA, et al.

Time-efficient LC/MS/MS determination of low concentrations of methylphosphonic acid.

Inorganic Materials. 2017;53(14):1382-5.

4. Pragney D, Vijaya Saradhi UVR. Sample-preparation techniques for the analysis of

chemical-warfare agents and related degradation products. Trends in Analytical Chemistry.

2012;37:73-82.

5. Mesilaakso M. Chemical Weapons Convention chemicals analysis: sample collection,

preparation, and analytical methods. 1st ed. Hoboken, NJ; Chichester, West Sussex, England:

Wiley; 2005.

6. Singh V, Purohit AK, Chinthakindi S, Goud RD, Tak V, Pardasani D, et al. Analysis of

chemical warfare agents in organic liquid samples with magnetic dispersive solid phase

extraction and gas chromatography mass spectrometry for verification of the chemical

weapons convention. Journal of Chromatography A. 2016;1448:32-41.

7. Witkiewicz Z, Neffe S, Sliwka E, Quagliano J. Analysis of the Precursors, Simulants and

Degradation Products of Chemical Warfare Agents. Critical Reviews in Analytical Chemistry.

2018;48(5):337-71.

52

8. Hooijschuur EWJ, Kientz CE, Brinkman UAT. Analytical separation techniques for the

determination of chemical warfare agents. Journal of Chromatography A. 2002;982(2):177-

200.

9. Black RM, Muir B. Derivatisation reactions in the chromatographic analysis of

chemical warfare agents and their degradation products. Journal of Chromatography A.

2003;1000(1):253-81.

10. Popiel S, Sankowska M. Determination of chemical warfare agents and related

compounds in environmental samples by solid-phase microextraction with gas

chromatography. Journal of Chromatography. 2011;1218(47):8457.

11. Koryagina NL, Savel’eva EI, Khlebnikova NS, Ukolov AI, Ukolova ES, Karakashev GV, et

al. Chromatography–mass spectrometry determination of alkyl methylphosphonic acids in

urine. Journal of Analytical Chemistry. 2016;71(14):1309-18.

12. Lee HK, Lee HSN, Sng MT, Basheer C. Determination of degradation products of

chemical warfare agents in water using hollow fibre-protected liquid-phase microextraction

with in-situ derivatisation followed by gas chromatography–mass spectrometry. Journal of

Chromatography A. 2007;1148(1):8-15.

13. Seto Y, Tachikawa M, Kanamori-Kataoka M, Sasamoto K, Ochiai N. Target analysis of

tert-butyldimethylsilyl derivatives of nerve agent hydrolysis products by selectable one-

dimensional or two-dimensional gas chromatography–mass spectrometry. Journal of

Chromatography A. 2017;1501:99-106.

14. Mazumder A, Gupta HK, Garg P, Jain R, Dubey DK. On-line high-performance liquid

chromatography–ultraviolet–nuclear magnetic resonance method of the markers of nerve

agents for verification of the Chemical Weapons Convention. Journal of Chromatography A.

2009;1216(27):5228-34.

53

15. Savel’eva EI, Gustyleva LK, Orlova OI, Khlebnikova NS, Koryagina NL, Radilov AS.

Modern methods for identification and quantitative determination of organophosphorus

chemical warfare agents. Russian Journal of Applied Chemistry. 2014;87(8):1003-12.

16. Kientz CE. Chromatography and mass spectrometry of chemical warfare agents,

toxins and related compounds: state of the art and future prospects. Journal of

Chromatography A. 1998;814(1):1-23.

17. Black RM, Harrison JM. The Chemistry of Organophosphorus Chemical Warfare

Agents. The Chemistry of Organophosphorus Compounds 1996.

18. Sng MT, Ng WF. In-situ derivatisation of degradation products of chemical warfare

agents in water by solid-phase microextraction and gas chromatographic–mass

spectrometric analysis. Journal of Chromatography A. 1999;832(1):173-82.

19. Birkemeyer C, Kolasa A, Kopka J. Comprehensive chemical derivatization for gas

chromatography-mass spectrometry-based multi-targeted profiling of the major

phytohormones. Journal of Chromatography A. 2003;993(1-2):89.

20. Dival L. The development of an in-field rapid derivatisation technique for the analysis

of chemical warfare agent degradants [Masters dissertation]: Murdoch University; 2018.

21. Coleman K. A History of Chemical Warfare. London: Palgrave Macmillan UK; 2005.

22. Gupta RC. Handbook of toxicology of chemical warfare agents. 1st ed. Amsterdam;

Boston: Elsevier/Academic Press; 2009.

23. Valdez CA, Leif RN, Hok S, Hart BR. Analysis of chemical warfare agents by gas

chromatography-mass spectrometry: methods for their direct detection and derivatization

approaches for the analysis of their degradation products. Reviews in Analytical Chemistry.

2018;37(1).

54

24. Ganesan K, Raza SK, Vijayaraghavan R. Chemical warfare agents. Journal of Pharmacy

& Bioallied Sciences. 2010;2(3):166-78.

25. Cooper N, Timperley C. Best synthetic methods: organophosphorus (V) chemistry.

London, England: Academic Press; 2015.

26. Mercey G, Verdelet T, Renou J, Kliachyna M, Baati R, Nachon F, et al. Reactivators of

acetylcholinesterase inhibited by organophosphorus nerve agents. Accounts of Chemical

Research. 2012;45(5):756-66.

27. Knaack JS, Zhou Y, Abney CW, Prezioso SM, Magnuson M, Evans R, et al. High-

throughput immunomagnetic scavenging technique for quantitative analysis of live VX nerve

agent in water, hamburger, and soil matrixes. Analytical Chemistry. 2012;84(22):10052-7.

28. Saint-André G, Kliachyna M, Kodepelly S, Louise-Leriche L, Gillon E, Renard P-Y, et al.

Design, synthesis and evaluation of new α-nucleophiles for the hydrolysis

of organophosphorus nerve agents: application to the reactivation of phosphorylated

acetylcholinesterase. Tetrahedron. 2011;67(34):6352-61.

29. Banoub J, Caprioli R. Molecular Technologies for Detection of Chemical and

Biological Agents. Dordrecht: Springer Netherlands; 2017.

30. Sidell FR. Chemical agent terrorism. Annals of Emergency Medicine. 1996;28(2):223-

4.

31. Kroening K. Analysis of Chemical Warfare Degradation Products. 1st ed. Chichester,

West Sussex, United Kingdom: Wiley; 2011.

32. Traces of sarin found after attack in northern Syria, says chemical weapons

watchdog. Breaking News U6 Newspaper Article. 2017.

55

33. Chemical weapons watchdog: Traces of sarin found following March 30 attack in

northern Syria that injured 50. AP English Language News (includes AP 50 State Report)

Newspaper Article. 2017.

34. Vale JA, Marrs TC, Maynard RL. Novichok: a murderous nerve agent attack in the UK.

Clinical toxicology (Philadelphia, Pa). 2018:1-5.

35. Willison SA. Investigation of the persistence of nerve agent degradation analytes on

surfaces through wipe sampling and detection with ultrahigh performance liquid

chromatography-tandem mass spectrometry. Analytical Chemistry. 2015;87(2):1034-41.

36. Baygildiev T, Zatirakha A, Rodin I, Braun A, Stavrianidi A, Koryagina N, et al. Rapid IC–

MS/MS determination of methylphosphonic acid in urine of rats exposed to

organophosphorus nerve agents. Journal of Chromatography B. 2017;1058:32-9.

37. Kingery AF, Allen HE. Ion chromatographic separation of closely related nerve agent

degradation products using an organic modifier to provide selectivity. Analytical Chemistry.

1994;66(1):155-9.

38. Kanamori-Kataoka M, Seto Y. Laboratory Identification of the Nerve Gas Hydrolysis

Products Alkyl Methylphosphonic Acids and Methylphosphonic Acid, by Gas

Chromatography-mass Spectrometry after tert.-butyldimethylsilylation. Journal of Health

Science. 2008;54(5):513-23.

39. Munro NB, Talmage SS, Griffin GD, Waters LC, Watson AP, King JF, et al. The sources,

fate, and toxicity of chemical warfare agent degradation products. Environmental Health

Perspectives. 1999;107(12):933-74.

40. Gustafson RL, Martell AE. A Kinetic Study of the Copper(II) Chelate-catalyzed

Hydrolysis of Isopropyl Methylphosphonofluoridate (Sarin). Journal of the American

Chemical Society. 1962;84(12):2309-16.

56

41. Yang YC, Baker JA, Ward JR. Decontamination of chemical warfare agents. Chemical

Reviews. 1992;92(8):1729-43.

42. Ward JR, Yang Y-C, Wilson RB, Burrows WD, Ackerman LL. Base-catalyzed hydrolysis

of 1,2,2-trimethylpropyl methylphosphonofluoridate—An examination of the saturation

effect. Bioorganic Chemistry. 1988;16(1):12-6.

43. Groenewold GS. Degradation kinetics of VX. Main Group Chemistry. 2010;9(3-4):221-

44.

44. Jansson D, Lindström SW, Norlin R, Hok S, Valdez CA, Williams AM, et al. Part 2:

Forensic attribution profiling of Russian VX in food using liquid chromatography-mass

spectrometry. Talanta. 2018;186:597-606.

45. Rodin IA, Braun AV, Baygildiev TM, Anan’eva IA, Shpigun OA, Rybalchenko IV.

Determination of the hydrolysis products of nerve agents in natural waters by liquid

chromatography–mass spectrometry. Journal of Analytical Chemistry. 2015;70(14):1671-7.

46. Rautio M, Finland U. Recommended operating procedures for sampling and analysis

in the verification of chemical disarmament. 1994 ed. Helsinki: Ministry for Foreign Affairs of

Finland; 1994.

47. McMaster MC. HPLC, a practical user's guide. 2nd ed. Hoboken, N.J: Wiley-

Interscience; 2007.

48. Hooijschuur EWJ, Kientz CE, Brinkman UAT. Application of microcolumn liquid

chromatography and capillary electrophoresis with flame photometric detection for the

screening of degradation products of chemical warfare agents in water and soil. Journal of

Chromatography A. 2001;928(2):187-99.

57

49. D’Agostino PA, Hancock JR, Provost LR. Determination of sarin, soman and their

hydrolysis products in soil by packed capillary liquid chromatography–electrospray mass

spectrometry. Journal of Chromatography A. 2001;912(2):291-9.

50. D'Agostino PA, Provost LR. Analysis of irritants by capillary column gas

chromatography-tandem mass spectrometry. Journal of Chromatography A.

1995;695(1):65-73.

51. D'Agostino PA, Provost LR. Capillary column isobutane chemical ionization mass

spectrometry of mustard and related compounds. Biomedical & Environmental Mass

Spectrometry. 1988;15(10):553-64.

52. Wahl JH, Colburn HA. Extraction of chemical impurities for forensic investigations: A

case study for indoor releases of a sarin surrogate. Building and Environment.

2010;45(5):1339-45.

53. Bogdan Z, Agnieszka Z, Joanna Swiat, Jacek N. Solid Phase Microextraction Combined

with Gas Chromatography - A Powerful Tool for the Determination of Chemical Warfare

Agents and Related Compounds. Current Organic Chemistry. 2007;11(3):241-53.

54. Kanaujia PK, Pardasani D, Gupta AK, Kumar R, Srivastava RK, Dubey DK. Extraction of

acidic degradation products of organophosphorus chemical warfare agents: Comparison

between silica and mixed-mode strong anion-exchange cartridges. Journal of

Chromatography A. 2007;1161(1):98-104.

55. Orata F. Derivatization Reactions and Reagents for Gas Chromatography Analysis:

INTECH Open Access Publisher; 2012.

56. Noort D, Benschop H, Black R. Biomonitoring of Exposure to Chemical Warfare

Agents: A Review. Toxicology and Applied Pharmacology. 2002;184(2):116-26.

58

57. Kroening KK, Richardson DD, Kubachka KM, Caruso JA. Chemical warfare nerve

agents - Analyzing their degradation products. Spectroscopy 2008;23(9):34.

58. Enqvist J, Rautio M. Identification of degradation products of potential

organophosphorus warfare agents. Helsinki: Ministry for Foreign Affairs of Finland. 1980.

59. Read RW, Black RM. Rapid screening procedures for the hydrolysis products of

chemical warfare agents using positive and negative ion liquid chromatography-mass

spectrometry with atmospheric pressure chemical ionisation. Journal of Chromatography A.

1999;862(2):169-77.

60. Lin DL, Wang SM, Wu CH, Chen BG, Liu RH. Chemical Derivatization for Forensic Drug

Analysis by GC- and LC-MS. Forensic Science Review. 2016;28(1):17-35.

61. Yashin YS, Revelsky IA, Tikhonova IN. Analysis of silyl derivatives of

methylphosphonic acid esters by gas chromatography-mass spectrometry with atmospheric

pressure photoionization. Journal of Analytical Chemistry. 2014;69(14):1356-60.

62. Driskell W, Shih M, Needham L, Barr D. Quantitation of Organophosphorus Nerve

Agent Metabolites in Human Urine Using Isotope Dilution Gas Chromatography-Tandem

Mass Spectrometry. Journal of Analytical Toxicology. 2002;26(1):6-10.

63. Kashutina MV, Ioffe SL, Tartakovskii VA. Silylation of Organic Compounds. Russian

Chemical Reviews. 1975;44(9):733-47.

64. Creasy WR, Rodríguez AA, Stuff JR, Warren RW. Atomic emission detection for the

quantitation of trimethylsilyl derivatives of chemical-warfare-agent related compounds in

environmental samples. Journal of Chromatography A. 1995;709(2):333-44.

65. Purdon JG, Pagotto JG, Miller RK. Preparation, stability and quantitative analysis by

gas chromatography and gas chromatography—electron impact mass spectrometry of tert.-

59

butyldimethylsilyl derivatives of some alkylphosphonic and alkyl methylphoshonic acids.

Journal of Chromatography A. 1989;475(2):261-72.

66. Kataoka M, Seto Y. Discriminative determination of alkyl methylphosphonates and

methylphosphonate in blood plasma and urine by gas chromatography–mass spectrometry

after tert.-butyldimethylsilylation. Journal of Chromatography B. 2003;795(1):123-32.

67. Drozd J. Chemical derivatization in gas chromatography. Journal of Chromatography

A. 1975;113(3):303-56.

68. Riches J, Morton I, Read RW, Black RM. The trace analysis of alkyl alkylphosphonic

acids in urine using gas chromatography–ion trap negative ion tandem mass spectrometry.

Journal of Chromatography B. 2005;816(1):251-8.

69. Fetterly BM, Verkade JG. P(RNCH2CH2)N: efficient catalysts for the cyanosilylation of

aldehydes and ketones. Tetrahedron Letters. 2005;46(46):8061-6.

70. Wang J, Masui Y, Hattori T, Onaka M. Rapid additions of bulky tert-butyldimethylsilyl

cyanide to hindered ketones promoted by heterogeneous tin ion-exchanged

montmorillonite catalyst. Tetrahedron Letters. 2012;53(15):1978-81.

71. Kataoka M, Tsunoda N, Ohta H, Tsuge K, Takesako H, Seto Y. Effect of cation-

exchange pretreatment of aqueous soil extracts on the gas chromatographic–mass

spectrometric determination of nerve agent hydrolysis products after tert.-

butyldimethylsilylation. Journal of Chromatography A. 1998;824(2):211-21.

72. Kataoka M, Tsuge K, Seto Y. Efficiency of pretreatment of aqueous samples using a

macroporous strong anion-exchange resin on the determination of nerve gas hydrolysis

products by gas chromatography–mass spectrometry after tert.-butyldimethylsilylation.

Journal of Chromatography A. 2000;891(2):295-304.

60

73. Kataoka M, Tsuge K, Takesako H, Hamazaki T, Seto Y. Effect of pedological

characteristics on aqueous soil extraction recovery and tert-butyldimethylsilylation yield for

gas chromatography-mass spectrometry of nerve gas hydrolysis products from soils.

Environmental Science and Technology. 2001;35(9):1823-9.

74. Wells RJ. Recent advances in non-silylation derivatization techniques for gas

chromatography. Journal of Chromatography A. 1999;843(1):1-18.

75. Kientz CE, Verweij A. Trimethylsilylation and trifluoroacetylation of a number of

trichothecenes followed by gas chromatographic analysis on fused-silica capillary columns.

Journal of Chromatography. 1986;355(1C):229-40.

76. Rosenfeld JM. Solid-phase analytical derivatization: enhancement of sensitivity and

selectivity of analysis. Journal of Chromatography A. 1999;843(1):19-27.

77. Lee H-B, Peart TE. Supercritical carbon dioxide extraction of resin and fatty acids

from sediments at pulp mill sites. Journal of Chromatography A. 1992;594(1):309-15.

78. Hillmann R, Bachmann K. On‐line supercritical fluid derivatization and extraction –

capillary gas chromatography of polar compounds. Journal of High Resolution

Chromatography. 1994;17(5):350-2.

79. Stashenko EE, Martınez JR. Derivatization and solid-phase microextraction. Trends in

Analytical Chemistry. 2004;23(8):553-61.

80. Pan L, Adams M, Pawliszyn J. Determination of fatty acids using solid phase

microextraction. Analytical Chemistry. 1995;67(23):4396-403.

81. Rawlinson C, Kamphuis LG, Gummer JPA, Singh KB, Trengove RD. A rapid method for

profiling of volatile and semi-volatile phytohormones using methyl chloroformate

derivatisation and GC–MS. Metabolomics. 2015;11(6):1922-33.

61

82. Hušek P. Chloroformates in gas chromatography as general purpose derivatizing

agents. Journal of Chromatography B: Biomedical Sciences and Applications.

1998;717(1):57-91.

83. Tao X, Liu Y, Wang Y, Qiu Y, Lin J, Zhao A, et al. GC-MS with ethyl chloroformate

derivatization for comprehensive analysis of metabolites in serum and its application to

human uremia. Analytical and Bioanalytical Chemistry. 2008;391(8):2881-9.

84. Vincenti M, Biazzi S, Ghiglione N, Valsania MC, Richardson SD. Comparison of Highly-

Fluorinated Chloroformates as Direct Aqueous Sample Derivatizing Agents for Hydrophilic

Analytes and Drinking-Water Disinfection By-Products. Journal of the American Society for

Mass Spectrometry. 2005;16(6):803-13.

85. Husek P, Simek P. Alkyl Chloroformates in Sample Derivatization Strategies for GC

Analysis. Review on a Decade Use of the Reagents as Esterifying Agents. 2006.

86. Pal Anagoni S, Kauser A, Maity G, Upadhyayula VVR. Quantitative determination of

acidic hydrolysis products of Chemical Weapons Convention related chemicals from

aqueous and soil samples using ion‐pair solid‐phase extraction and in-situ butylation.

Journal of Separation Science. 2018;41(3):689-96.

87. Kostiainen O. Chapter 11D Analysis of chemicals related to the chemical weapons

convention. Handbook of Analytical Separations. 2000;2:405-35.

88. Aleksenko SS. Liquid chromatography with mass-spectrometric detection for the

determination of chemical warfare agents and their degradation products. Journal of

Analytical Chemistry. 2012;67(2):82-97.

89. Parris NA. Instrumental liquid chromatography: a practical manual on high-

performance liquid chromatographic methods. 2nd ed. Amsterdam, Netherlands: Elsevier;

1984.

62

90. Papoušková B, Bednář P, Fryšová I, Stýskala J, Hlaváč J, Barták P, et al. Mass

spectrometric study of selected precursors and degradation products of chemical warfare

agents. Journal of Mass Spectrometry. 2007;42(12):1550-61.

91. Katagi M, Nishikawa M, Tatsuno M, Tsuchihashi H. Determination of the main

hydrolysis product of O-ethylS-2-diisopropylaminoethyl methylphosphonothiolate, ethyl

methylphosphonic acid, in human serum. Journal of Chromatography B: Biomedical Sciences

and Applications. 1997;689(2):327-33.

92. Rohrbaugh DK, Sarver EW. Detection of alkyl methylphosphonic acids in complex

matrices by gas chromatography–tandem mass spectrometry. Journal of Chromatography A.

1998;809(1):141-50.

93. Fredriksson SÅ, Hammarström LG, Henriksson L. Trace determination of alkyl

methylphosphonic acids in environmental and biological samples using gas

chromatography/negative-ion chemical ionization mass spectrometry and tandem mass

spectrometry. Journal of Mass Spectrometry. 1995;30(8):1133-43.

94. Riches J. Chapter 7 - Analysis of Organophosphorus Chemicals. In: Timperley CM,

editor. Best Synthetic Methods. Oxford: Academic Press; 2015. p. 721-52.

95. Wils ERJ, Hulst AG. Determination of organophosphorus acids by thermospray liquid

chromatography-mass spectrometry. Journal of Chromatography A. 1988;454:261-72.

96. Black RM, Read RW. Application of liquid chromatography-atmospheric pressure

chemical ionisation mass spectrometry, and tandem mass spectrometry, to the analysis and

identification of degradation products of chemical warfare agents. Journal of

Chromatography A. 1997;759(1-2):79-92.

97. Black RM, Read RW. Analysis of degradation products of organophosphorus chemical

warfare agents and related compounds by liquid chromatography–mass spectrometry using

63

electrospray and atmospheric pressure chemical ionisation. Journal of Chromatography A.

1998;794(1-2):233-44.

98. Mercier JP, Morin P, Dreux M, Tambuté A. Liquid chromatography analysis of

phosphonic acids on porous graphitic carbon stationary phase with evaporative light-

scattering and mass spectrometry detection. Journal of Chromatography A.

1999;849(1):197-207.

99. Barker JPD, Ando DJ, Davis R. Mass spectrometry. 2nd ed. New York: John Wiley &

Sons; 1999.

100. D’Agostino PA, Hancock JR, Provost LR. Packed capillary liquid chromatography–

electrospray mass spectrometry analysis of organophosphorus chemical warfare agents.

Journal of Chromatography A. 1999;840(2):289-94.

101. Katagi M, Tatsuno M, Nishikawa M, Tsuchihashi H. On-line solid-phase extraction

liquid chromatography-continuous flow frit fast atom bombardment mass spectrometric

and tandem mass spectrometric determination of hydrolysis products of nerve agents alkyl

methylphosphonic acids by p-bromophenacyl derivatization. Journal of Chromatography A.

1999;833(2):169-79.

102. D'Agostino PA, Chenier CL, Hancock JR. Packed capillary liquid chromatography-

electrospray mass spectrometry of snow contaminated with sarin. Journal of

Chromatography A. 2002;950(1-2):149-56.

103. Noort D, Hulst AG, Platenburg DHJM, Polhuijs M, Benschop HP. Quantitative analysis

of O-isopropyl methylphosphonic acid in serum samples of Japanese citizens allegedly

exposed to sarin: estimation of internal dosage. Archives of Toxicology. 1998;72(10):671-5.

104. Blakley CR, Vestal ML. Thermospray interface for liquid chromatography/mass

spectrometry. Analytical Chemistry. 1983;55(4):750-4.

64

105. Ciner FL, McCord CE, Plunkett RW, Martin MF, Croley TR. Isotope dilution LC/MS/MS

for the detection of nerve agent exposure in urine. Journal of Chromatography B.

2007;846(1):42-50.

106. Tak V, Kanaujia PK, Pardasani D, Kumar R, Srivastava RK, Gupta AK, et al. Application

of Doehlert design in optimizing the determination of degraded products of nerve agents by

ion-pair liquid chromatography electrospray ionization tandem mass spectrometry. Journal

of Chromatography A. 2007;1161(1):198-206.

107. Zhou S, Hamburger M. Application of liquid chromatography-atmospheric pressure

ionization mass spectrometry in natural product analysis. Evaluation and optimization of

electrospray and heated nebulizer interfaces. Journal of Chromatography A.

1996;755(2):189-204.

108. Borrett VT, Mathews RJ, Colton R, Traeger JC. Verification of the United Nations

Chemical Weapons Convention: The application of electrospray mass spectrometry. Rapid

Communications in Mass Spectrometry. 1996;10(1):114-8.

109. Bell AJ, Murrell J, Timperley CM, Watts P. Fragmentation and reactions of two

isomeric O-alkyl S-(2-dialkylamino)ethyl methylphosphonothiolates studied by electrospray

ionization/ion trap mass spectrometry. Journal of the American Society for Mass

Spectrometry. 2001;12(8):902-10.

110. Black RM, Read RW. Methods for the analysis of thiodiglycol sulphoxide, a

metabolite of sulphur mustard, in urine using gas chromatography-mass spectrometry.

Journal of Chromatography. 1991;558(2):393-404.

111. Black RM, Clarke RJ, Read RW. Analysis of 1,1'-sulphonylbis[2-

(methylsulphinyl)ethane] and 1-methylsulphinyl-2-[2-(methylthio)ethylsulphonyl]ethane,

65

metabolites of sulphur mustard, in urine using gas chromatography-mass spectrometry.

Journal of Chromatography A. 1991;558(2):405-14.

112. Ross P, Knox JH. Carbon-based packing materials for liquid chromatography:

applications. Advances in chromatography. 1997;37:121.

113. Gäb J, John H, Melzer M, Blum M-M. Stable adducts of nerve agents sarin, soman

and cyclosarin with TRIS, TES and related buffer compounds—Characterization by LC-ESI-

MS/MS and NMR and implications for analytical chemistry. Journal of Chromatography B.

2010;878(17):1382-90.

114. Rodriguez M, Orescan DB. Confirmation and quantitation of selected sulfonylurea,

imidazolinone, and sulfonamide herbicides in surface water using electrospray LC/MS.

Analytical Chemistry. 1998;70(13):2710-7.

115. Baygildiev TM, Rodin IA, Stavrianidi AN, Braun AV, Shpigun OA, Rybalchenko IV.

Determination of methylphosphonic acid in human blood plasma by high-performance

liquid chromatography–tandem mass spectrometry. Journal of Analytical Chemistry.

2017;72(13):1307-11.

Blank Page – not numbered

Part Two

Manuscript

DERIVATISATION OF CHEMICAL WARFARE AGENT

DEGRADANT WITHOUT REMOVAL OF WATER

1

ABSTRACT

With the success of a recent pilot study to derivatise the chemical warfare agent degradant,

methylphosphonic acid without removal of water, a quantitative assessment was undertaken

to assess its efficiency. An accurate and rapid approach using liquid chromatography-mass

spectrometry was developed for quantitating the efficiency of the two-phase derivatisation

of chemical warfare degradant without removal of water. Chromatographic conditions, a

gradient elution program, and mass spectrometric detection parameters were optimised. The

developed method revealed the efficiency for the two-phase derivatisation of

methylphosphonic acid into methylphosphonic acid-derivative bis[(dimethyl)(tert-butyl)silyl]

esters to be 14.6% at 1000 mg/L. Satisfactory reproducibility, reliability and linearity were

demonstrated with the quantitation of chemical warfare degradant in comparison with

quality controls of known concentration. The limits of detection and quantification of

methylphosphonic acid in water were 1.6 mg/L and 5.3 mg/L, respectively.

Keywords: Chemical warfare agents, chemical warfare degradant, derivatisation,

methylphosphonic acid, liquid chromatography-mass spectrometry, silylation

2

INTRODUCTION

The Organisation for the Prohibition of Chemical Weapons (OPCW) develops and executes

methods for the enforcement of Chemical Warfare Convention (CWC), to conduct and

challenge former chemical warfare agent (CWA) facilities of the member states for the

presence of CWA precursors, degradants or any related compound.1-3 Despite this, the use

of CWA remains a problem, with many different forms of CWA being developed and utilised

heavily in many conflicts, terrorist attacks and wars around the world even till this day.4

The most lethal group of chemical warfare agents (CWA) is known to be the

organophosphorus nerve agents, which consist of sarin (GB), soman (GD), cyclosarin (GF),

VX and VR.4 In the presence of aqueous environment, these nerve agents hydrolyse rapidly

to form their corresponding alkyl methylphosphonic acids (AMPA), and finally into

methylphosphonic acid (MPA); the common degradant product among the nerve agents

(Figure 1).5-7

3

Figure 1. Hydrolytic pathways of organophosphorus nerve agents (sarin, soman, cyclosarin, VX and Russian VX) into their corresponding alkyl methylphosphonic acids, and subsequently into their final degradant product- methylphosphonic acid. Adapted and modified from Y. Seto et. al.8

Besides being the final degradant product to organophosphorus nerve agents, MPA is

known to be in its most stable form, and is chemically-resistant to any further

degradation.5,6,8-10 Furthermore, MPA does not exist naturally in the environment, therefore

the detection of this compound serves as a potential evidence for the possession and/or

involvement of CWA.5,9 Also, in the likely event of the absence of the parent compound,

verification of CWA for the CWC would be highly dependent on the unambiguous

identification of CWA degradants.1,11

4

Gas chromatography coupled to mass spectrometry (GC-MS) is one of the most common

and approved technique for the study of CWA degradant products.1,5,12 Owing to its high

polarity and low volatility properties, MPA requires an additional step of derivatisation prior

to GC analyses.12 Derivatisation is a procedural technique for the modification of the

functional groups of target compound to produce GC-amenable derivatives for GC

analyses.4,5 The basic reaction scheme of derivatisation reaction is shown in Figure 2.

Figure 2. Chemical equation for general derivatisation reactions where ‘A’ denotes nitrogen, oxygen or sulphur, while ‘D’ denotes the functional group of derivatisation reagent used.

Derivatisation requires the elimination of water content; concentration of aqueous extract

to dryness, which is reported in many literature resources to be both time-consuming and a

major source of error.5,9,13,14 There are no literature resources that report nor suggest for an

alternative method to derivatise compounds without the removal of water until 2018,

where Dival has successfully developed a method to produce GC-amenable derivatives of

MPA without the removal of water.15 Dival eliminated the need to evaporate the aqueous

extract to dryness by using hexane, an organic solvent, together with the derivatising

reagent N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide (MTBSTFA).15 Despite being

successful in producing GC-amenable derivatives at detection limit (LOD) of 1000 mg/L, the

presented approach necessitates a quantitative assessment for its efficiency and

practicality.15

5

Being a highly sensitive and versatile instrumental technique, liquid chromatography-mass

spectrometry (LC-MS) can be employed to quantitate the efficiency of the two-phase

derivatisation of MPA developed by Dival. LC-MS is known for its low detection limits, offers

detection limits as low as 50 ng/mL and a short analysis time for the detection of MPA.9 In

this study, derivatising reagent MTBSTFA with 1% tert-butyldimethylsilyl chloride

(TBDMSCl), was used in replacement of pure MTBSTFA to produce hydrolytically-stable

derivatives for the two-phase derivatisation of MPA (Figure 3).

Figure 3. Reaction scheme for the silylation of MPA.

Chromatographic conditions, a gradient elution program, and mass spectrometric detection

parameters were optimised for the development of an accurate and rapid quantitative LC-

MS protocol for MPA.

6

EXPERIMENTAL

Reagents and solvents

Solid MPA (98%) from Sigma-Aldrich (St. Louis, MO, USA), hexane and acetonitrile of

analytical grade reagents (AR) from Mallinckrodt (St. Louis, MO, USA) were used.

Derivatising reagent MTBSTFA with 1% TBDMSCl catalyst were purchased from Sigma-

Aldrich (St. Louis, MO, USA) for performing two-phase derivatisation.

Standard solutions and samples

MPA stock solution of 1000 mg/L was prepared by dissolving 255 mg (±0.1 mg) of solid MPA

in 250 mL of deionised water. From this stock solution, a series of ten-fold dilutions

(1000 mg/L, 100 mg/L, 10 mg/L and 0.1 mg/L) were prepared in glass vials. A separate set of

serial dilutions were prepared at concentration range of 1500 mg/L, 1000 mg/L, 500 mg/L,

250 mg/L and 100 mg/L for the development of a calibration curve. The calibration curve

established was used to quantitatively assess the amount of aqueous MPA derivatised by

MTBSTFA with 1% TBDMSCl catalyst. Blank solutions were prepared with deionised

water (AR). A set of three quality control samples were prepared with serially diluted MPA

of 1000 mg/L from the calibration standard solutions to serve as a set of known

concentration (1000 mg/L) for quantitative comparison.

Two-Phase Derivatisation

Hexane (2.00 mL) and MTBSTFA with 1% TBDMSCl catalyst (200 μL) were mixed thoroughly

into each aqueous MPA solution (1.00 mL) using an oil bath on a Yellow Line MST Digital

magnetic stirrer and heater, at 60 ˚C for 30 minutes, as recommended by Dival.15 Following

7

this, the MPA samples were left aside at room temperature for 15 minutes to allow

separation of the organic and aqueous layers of MPA derivatives. The organic layer of each

glass vial was removed carefully with a pasteur pipette into each GC vial for the

unambiguous detection of MPA derivatives by GC-MS. The remaining aqueous solution in

each glass vial was removed carefully and reconstituted with DI water to a known volume of

1.00 mL prior to quantitative analysis by LC-MS with the quality control samples.

Apparatus and analytical conditions

Analyses with GC-MS were performed on a Shimadzu GCMS-QP2010S (Shimadzu

Australasia, Rydalmere, NSW, Australia) interfaced to a single quadrupole mass

spectrometer with electron ionisation source. Separation was carried out using a BPX-5 (5%

phenyl polysilphenylene-siloxane) capillary column (30 m x 0.25 mm x 0.25 μm). Split

injections were performed at a ratio of 10:1. The temperature program for the analysis of

MPA derivatives was: 80 ˚C, held for 1 minute, ramp at 20 ˚C/min to 280 ˚C held for 6

minutes. The carrier gas was ultra-high purity helium (BOC, Sydney, NSW, Australia) at

30 cm/sec. Injection temperature was maintained at 250 ˚C. GC analyses were performed in

SIM (selected ion monitoring) and TIC (total ion chromatogram) modes on m/z 267 and over

the range of m/z 45 – 310 at scan rate of 555 scans/sec, respectively.

Quantitative analyses with LC-MS were performed using AcquityTM UHPLC system interfaced

to a Micromass MS Technologies triple quadrupole time-of-flight mass spectrometer. LC

separation was performed on a 50 mm x 2.1 mm Acquity UPLC BEH® C18 (1.7 µm) column.

The mobile phases were 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile

(B). The optimised gradient elution was 90% A (0 – 4 min), 1% A (4 – 7 min), 90% A (7 –

8

10 min), delivered at flow rate of 0.3 mL/min. Runtime was 10 minutes, with seal wash of 5

minutes. Mass spectrometric detection was performed using electrospray ionisation in the

positive ion detection with the following parameters: capillary voltage of 3.5 kV, sample

cone voltage of 3 V, extraction cone voltage of 3 V, desolvation temperature 350 °C, source

temperature 100 °C, and desolvation gas at flow rate of 500 L/hr. Data visualisation and

analysis were performed with MassLynxTM software.

Construction of Calibration Curve

Calibration standards of concentration range: 1500 mg/L, 1000 mg/L, 500 mg/L, 250 mg/L

and 100 mg/L were used for the development of the calibration curve. Each analytical run

consisted of blank(s) and a series of serially diluted MPA derivatives. Data from three runs

were evaluated to assess accuracy and precision (RSD%), linearity, limits of quantitation

(LOQ) and LOD for each MPA analytes. Calibration curves were constructed by plotting the

peak area versus concentration (mg/L) using Microsoft Excel.

9

RESULTS & DISCUSSION

Qualitative detection of MPA-derivatives

To improve on the precedent study conducted by Dival, derivatisation reagent MTBSTFA

with 1% TBDMSCI catalyst was used as the derivatisation reagent for the two-phase

derivatisation of MPA instead of pure MTBSTFA.15 Even though no current literature

described the optimal amount of derivatising reagent for the derivatisation process, a

mixture of MPA and derivatising reagent in a ratio of 1:2 is required for the stoichiometry of

the reaction due to the presence of two hydroxyl functional groups in the former

compound. Furthermore, MPA is known to be a residual contaminant in chromatographic

system in the case of incomplete derivatisation.7 Therefore, an excess volume of 200 μL of

MTBSTFA with 1% TBDMSCl catalyst was used to maximise the yield of the target reaction

and ensure the absence of any MPA residual contaminant. The MPA derivative was only

detected in the organic later following the two-phase derivatisation of aqueous MPA of

1000 mg/L, consistent with results achieved by Dival despite the addition of catalyst. The

organic layer from two-phase derivatisation was analysed by GC-MS. The mass spectrum of

the peak at retention time 8.133 mins in the GC-MS chromatogram gave the main peak with

m/z 267 for [M – Bu]+ (TBDMS), bearing 96% similarity to the NIST database of the desired

MPA-derivative bis[(dimethyl)(tert-butyl)silyl] ester on the Shimadzu database. A

reconstructed ion chromatogram and mass spectrum showing the detection of MPA-

derivative bis[(dimethyl)(tert-butyl)silyl] ester are shown in Figure 4.

10

Figure 4. GC-MS chromatogram (a) and mass spectrum (b) for the detection MPA-derivative bis[(dimethyl)(tert-butyl)silyl] ester at 8 min retention time and m/z 267, respectively.

Even though the derivative as only detected for 1000 mg/L MPA solutions post-

derivatisation, minute peaks with m/z 267 were found to be present in the chromatograms

for the organic layer from the derivatisation of 100 mg/L, 10 mg/L, 1 mg/L and 0.1 mg/L

solutions which were not observed in Dival’s results.15 However, the intensity of signals

were insufficient to qualitatively confirm the presence of MPA-derivative bis[(dimethyl)(tert-

butyl)silyl] ester. As such, the replacement of derivatisation reagent with MTBSTFA with 1%

TBDMSCI catalyst may have played a role in the yield of these peaks. With the lack of

11

success at detecting MPA-derivative below 1000 mg/L, decision was made to adjust the

analytical range to 100 – 1500 mg/L for the generation of calibration curve.

Optimisation of LC-MS parameters

Previous studies involving the chromatographic separation of the degradant products of

organophosphorus nerve agents employ the use of electrospray ionisation source in positive

ion mode.16-21 Comparison between the use of electrospray ionisation source in positive and

negative ion modes was performed prior to the decision of employing electrospray

ionisation in positive ion mode. Contrasted with electrospray ionisation in negative ion

mode, the mass spectrum produced by positive ion mode was less complex with fewer

adducts or other undetermined ions (Figure 5), as opposed to the results obtained in the

study conducted by Liu et al.11 The mass spectra generated by electrospray ionisation in

negative ion mode were described by Liu et al. to be simpler than those of positive ion

mode.11 It was also explained that ionisation under negative ion mode would be easier since

MPA are acidic analytes.11 Apart from the fragment ion derived from the deprotonated form

of MPA, [M – H]- with m/z of 95, the mass spectra generated by negative ion mode showed

a prominent peak at m/z 79 corresponding to a molecule formed from MPA upon losing its

methyl group in collision cell.9

12

Figure 5. Mass spectra of MPA (1000 mg/L) analysed by electrospray ionisation in positive (top) and negative ion mode (bottom), with fragment ions derived from protonated form of MPA, [M + H]+, with m/z of 97, and deprotonated form of MPA, [M – H]-, with m/z of 95, respectively.

To achieve the best shape of chromatographic peaks and shortest retention time for the

elution of MPA, elution in gradient mode was selected and optimised. The best results were

achieved with the gradient elution parameters as illustrated in the experimental section,

with MPA eluted within the first minute of analysis as shown in Figure 6.

Figure 6. LC-MS chromatograms of three MPA calibration standard solutions of 1000 mg/L.

13

Many similar studies employed the use of formic acid as the acid modifier in mobile phases

for the chromatographic separation of CWA related compounds. Black and Read were able

to achieve better sensitivity for the detection of phosphonic acids with the use of formic

acid instead of trifluoroacetic acid.16 Ion intensity obtained from electrospray ionisation

modes is highly dependent on the concentration of the buffering salt in mobile phases.11 It

was explained by Zhou and Hamburger that formic acid enhances, while trifluoroacetic acid

suppresses the formation of [M + H]+ ions with electrospray ionisation.22 Formic acid was

used here as the acid modifier for mobile phases to suit the acquisition of positive ion

spectra, as recommended by Black and Read.16

The mass spectrum of MPA shown in Figure 7 features a prominent peak at m/z of 97,

corresponding to protonated MPA ([M + H]+).

Figure 7. ESI+ mass spectrum of MPA, [M + H]+, with m/z of 97. Initial Quantitative Assessment of MPA Calibration Standards

The first attempt at generating a calibration curve was relatively unsuccessful due to the

inconsistent peak area for calibration standards of same concentration levels despite all

standard solutions being originated from the same stock solution. The precision results are

shown in Table 1, where %RSD ranged between 35 – 94.1%, suggesting poor reproducibility.

Furthermore, the peak areas are found to be increasing across all MPA calibration standard

14

solutions of the same concentration over time. For example, the first MPA calibration

standard solution of 100 mg/L revealed a peak area of 1.29 arb. units, while the second and

third calibration standards of the same concentration were 19.3 and 38.1 arb. units,

respectively (Table 1). This anomaly was consistent across all three sets of MPA calibration

standards where the peak areas increased from each set of calibration standards to another.

Table 1. Initial quantitative results of MPA calibration standards.

Concentration (mg/L)

100 250 500 1000 1500

Peak areas from each triplicate run

1.29 39.8 104 176 241

19.3 77.2 161 316 459

38.1 111 225 404 505

%RSD* 94.1% 46.8% 36.4% 38.5% 35.1%

* %RSD = Standard Deviation (MPA calibration standard of interest from each run)

Average (MPA calibration standard of interest from each run)

To address the possibility of instrumental error, seven repeated injections of a MPA

standard solution of 1000 mg/L was performed. As illustrated in Table 2, the peak areas

across all seven injections from the same solution are consistent with each other, thus

invalidating the possibility of instrumental error.

Table 2. Quantitative results for seven repeated injections on MPA standard solution of 1000 mg/L.

Injections

1st 2nd 3rd 4th 5th 6th 7th

Peak areas 520 525 539 537 535 539 515

Despite the poor reproducibility and anomaly in the initial results, the increase in peak area

between each serially-diluted solution was found to be relatively consistent. For example,

there was an increase of 1.45 times (close to the expected mathematical increase of

15

1.5 times) peak area between MPA calibration standard of 1000 mg/L to 1500 mg/L. As

such, the differences in the peak area for MPA calibration standards of each concentration

level can be used as correction factors to enhance the linear regression of the calibration

curve (Table 3).

Table 3. Correction factors for MPA standards based on first analytical run.

Concentration (mg/L)

100 250 500 1000 1500

Correction Factor* 0.06 0.24 0.51 1 1.45

*Correction Factor = Peak Area of MPA calibration standard of interest

Peak Area of MPA calibration standard 1000 mg/L

Evaluation of Calibration Curve

Method precision was assessed using data from subsequent analytical runs with triplicate

analysis of the MPA calibration standards. The precision results are shown in Table 4,

with %RSD below 8.23% across the six concentration levels.

Table 4. Reproducibility of the optimised procedure.

Concentration (mg/L)

0 100 250 500 1000 1500

%RSD* from analytical runs

0%

6.53%

8.23%

4.05%

3.51%

6.32%

* %RSD = Standard Deviation (MPA calibration standard of interest from each run)

Average (MPA calibration standard of interest from each run)

Despite the acceptable reproducibility results, the increase in peak area between each MPA

calibration standard was found to be out of the expected range. For example, there should

be 10 times differences in terms of the peak area between MPA calibration standard of

100 mg/L and 1000 mg/L, which was not observed in this analytical run (Table 5). With the

16

inclusion of correction factors from the previous run, the peak area for each calibration

standard can be rectified to fit the series of serial dilution (Table 6).

Table 5. Quantitative results of MPA calibration standards.

Concentration (mg/L)

100 250 500 1000 1500

Peak areas from each triplicate run

130 275 510 761 1073

148 320 540 772 968

143 317 499 813 960

Average peak area 140 304 515 782 1000

Table 6. Final quantitative results of MPA calibration standards.

Concentration (mg/L)

MPA calibration standards

Original Peak Area* Adjusted Peak Area**

0 0 0

100 140 47.7

250 304 191

500 515 398

1000 782 782

1500 1000 1135

*Derived from the average peak areas of each triplicate MPA calibration standards. **Adjusted Peak Area = (Original Peak area of MPA calibration standard 1000 mg/L) × (Correction Factor from Table 3), where correction factor for each MPA calibration standards of 100 mg/L, 250 mg/L, 500 mg/L, 750 mg/L, 1000 mg/L, and 1500 mg/L are 0.06, 0.24, 0.51, 1, and 1.45, respectively.

The calibration dependences were plotted over the concentration range of 100 –

1500 mg/L, with the y intercept set to 0 (Figure 8). The line of best-fit was evaluated, with

calibration curve equation and correlation coefficient to measure the linearity. The

performance characteristics of presented approach were calculated and shown in Table 7.

17

The detection limits (LOD) was calculated to be 1.6 mg/L, with the limit of quantitation

(LOQ) to be 5.3 mg/L.

Table 7. Performance characteristics of developed approach for the quantitative determination of MPA.

Analytical Range (mg/L) Calibration Curve Equation R2 LOD (mg/L) LOQ (mg/L)

100 – 1500 y = 0.766 x 0.9985 1.6 5.3

Figure 8. Calibration curve of MPA calibration standards in the concentration range of 100 –

1500 mg/L.

Efficiency of Two-Phase Derivatisation

The developed LC-MS method for determining the concentration of MPA remaining in the

aqueous layer after derivatisation was used to analyse the aqueous layer of MPA

solutions (1000 mg/L) post-derivatisation simultaneously with a set of three quality control

MPA samples of 1000 mg/L. Due to limited resources (budget) and time constraint, only

MPA samples of 1000 mg/L were assessed.

y = 0.766xR² = 0.9985

0

200

400

600

800

1000

1200

1400

0 500 1000 1500

Pe

ak A

rea

(arb

. un

its)

Concentration (mg/L)

MPA Calibration Standards

Linear Regression

18

Only 0.9 mL of aqueous MPA was found to remain in each glass vial after performing two-

phase derivatisation. Each aqueous MPA solution was reconstituted to a known volume of

1.00 mL prior to quantitative analysis by LC-MS. In order to ensure accuracy in the

quantitation of 1000 mg/L MPA post-derivatisation, three reference controls of known

concentration were introduced and analysed with target samples in the same analytical run.

The reference control or described as quality control used here were made by taking 0.9 mL

of each MPA calibration standard solutions of 1000 mg/L and reconstituted to a known

volume of 1.00 mL. The peak area and calculated concentration of both quality control and

MPA samples are shown in Table 8.

Table 8. Quantitative results of MPA in aqueous extract post-derivatisation.

Samples of 1000 mg/L

Calculations of un-derivatised MPA in 3 significant figures

Peak Area Before** (mg/L) After** (mg/L) True Concentration (mg/L)

Quality Control 462* 603 670 1000

MPA (Post-derivatisation)

427* 557 620 854 (± 3.51%)

* Derived from the average peak areas of each triplicate sample. ** ‘Before’ and ‘After’ denotes the concentration of sample of interest before and after calculation with 0.1 × dilution factor, respectively. Both quality control and MPA concentration were calculated with the calibration curve equation of y = 0.766 x.

Upon derivatisation, MPA-derivative bis[(dimethyl)(tert-butyl)silyl] esters partitions at the

organic layer of the solution, which was calculated to be 146 mg/L (± 3.51%), while the

concentration of un-derivatised MPA remaining in the aqueous layer was approximately

854 mg/L (± 3.51%). As such, the efficiency of derivatisation without removal of water

presented by Dival is 14.6% at 1000 mg/L, in comparison with the quality controls and linear

regression equation.

19

Apart from the quantitation results for the efficiency of the two-phase derivatisation, it is

evident that the peak area for the quality controls was significantly lower than the MPA

calibration standards of the same concentration (Table 5 and 8). The original peak area of

the three MPA calibration standards of 1000 mg/L amounted to an average of 782 arb.

units, whereas the peak area of three MPA quality controls of the same concentration

amounted to an average of 514 arb. units (after taking 0.1 × dilution into account from

original value of 462 arb. units). A line graph illustrating the drop in peak areas of the

compounds is shown in Figure 9.

Figure 9. Peak area of each MPA calibration standards (1000 mg/L) versus the time passed after each

analytical run. denotes three MPA calibration standard solutions (1000 mg/L) being run in

duplicate, with at least 210 minutes apart from each run. denotes three quality control samples

(1000 mg/L), with at least 60 minutes apart from the previous analytical run. To obtain accurate data for the generation of calibration curve, each calibration standard

solution was analysed twice for their peak areas. The first analytical run on the MPA

calibration standard solutions of 1000 mg/L indicated peak area of 761 – 813 arb. units.

With analysis of each sample in LC-MS instrument to be 15 minutes, 210 minutes (15 min ×

20

14 samples; 15 samples per analytical run) would have passed before being analysed again.

The second analytical run on the same MPA calibration standards indicated a drop in peak

areas (637 – 698 arb. units), as shown in Figure 9. The third analytical run was performed

approximately 60 minutes later to accommodate for the derivatisation process. Our results

suggested possible degradation of MPA with loss of peak area over time. According to

current literature resources, MPA is known to be a highly stable compound in aqueous

environment, and possesses a relatively high solubility rate in water; > 20,000 mg/L. In an

article written by Mill and Gould, MPA was described as a highly stable hydrolysis (or

degradant) product of nerve agents, with half-life for the reaction with OH-radicals in water

of 18 years.23 Furthermore, MPA is resistant to photolysis, and do not absorb any UV light

present in the natural nor laboratory environment.23 One possible explanation for the loss

of peak areas in samples over time could be due to the irreversible adsorption of MPA on

the surfaces of glassware. Fenimore et al. explained that polar compounds are capable of

the irreversible adsorption of themselves on the surfaces of laboratory glassware.24 The

work done by Subramaniam et al. corroborates this.25 Subramaniam et al. reported a loss of

yield of their target analytes post-derivatisation, and explained that the loss of yield was

caused by the irreversible adsorption of alkylphosphonic acids on the surfaces of

glassware.25 Results of their study showed that the use of silylated glassware improved the

yield of MPA by 20%.25 Also, the limitations of presented method can be overcome by the

introduction of an internal standard. With the inclusion of a known compound with known

concentration, an accurate concentration of the internal standard will produce a separate

chromatographic peak.26 In this manner, any loss of sample will be accompanied by the loss

of an equivalent amount of internal standard.

21

CONCLUSION

The degradant product of organophosphorus nerve agents was shown to be successfully

derivatised into MPA-derivatives at 1000 mg/L using the developed method by Dival.15

Degradation products are distinct chemical signature for the detection of chemical warfare

agents and forensic attribution in cases where the parent compound no longer exist. In

situations where time is a determining factor, this presented analytical procedure would be

much more time-efficient as compared to the conventional method of derivatisation due to

the eliminated requirement for the evaporation of aqueous extracts to dryness. For

practical considerations, it is essential to ensure a safe and smooth derivatisation process

with a positive yield of stable product. As such, a rapid and simple approach for the

quantitative determination of MPA was developed and presented here with the use of LC-

MS and three quality control samples. In this method, effective elution of MPA with good

(peak shape and resolution) chromatographic peaks was achieved in less than a minute. The

developed method offered good linearity, with LOD and LOQ calculated to be 1.6 mg/L and

5.3 mg/L, respectively. As a quantitative assessment for the time-efficient protocol

developed by Dival to derivatise MPA without the removal of water, the efficiency of the

presented derivatisation approach was revealed to be approximately 14.6 %. However, with

the reduced peak area of MPA observed over time, this method provides avenues for future

research to address this anomaly.

22

REFERENCES

1. Mesilaakso M. Chemical Weapons Convention chemicals analysis: sample collection,

preparation, and analytical methods. 1st ed. Hoboken, NJ; Chichester, West Sussex, England:

Wiley; 2005.

2. Baygildiev TM, Rodin IA, Stavrianidi AN, Braun AV, Shpigun OA, Rybalchenko IV.

Determination of methylphosphonic acid in human blood plasma by high-performance

liquid chromatography–tandem mass spectrometry. Journal of Analytical Chemistry.

2017;72(13):1307-11.

3. Grigoriu N, Epure G, Ginghina R, Mosteanu D. An overview of the OPCW’s

programme for biomedical samples analysis. 2015;21(3):815.

4. Hooijschuur EWJ, Kientz CE, Brinkman UAT. Analytical separation techniques for the

determination of chemical warfare agents. Journal of Chromatography A. 2002;982(2):177-

200.

5. Black RM, Muir B. Derivatisation reactions in the chromatographic analysis of

chemical warfare agents and their degradation products. Journal of Chromatography A.

2003;1000(1):253-81.

6. Popiel S, Sankowska M. Determination of chemical warfare agents and related

compounds in environmental samples by solid-phase microextraction with gas

chromatography. Journal of Chromatography. 2011;1218(47):8457.

7. Koryagina NL, Savel’eva EI, Khlebnikova NS, Ukolov AI, Ukolova ES, Karakashev GV, et

al. Chromatography–mass spectrometry determination of alkyl methylphosphonic acids in

urine. Journal of Analytical Chemistry. 2016;71(14):1309-18.

23

8. Seto Y, Tachikawa M, Kanamori-Kataoka M, Sasamoto K, Ochiai N. Target analysis of

tert-butyldimethylsilyl derivatives of nerve agent hydrolysis products by selectable one-

dimensional or two-dimensional gas chromatography–mass spectrometry. Journal of

Chromatography A. 2017;1501:99-106.

9. Baygildiev TM, Rodin IA, Stavrianidi AN, Braun AV, Akhmerova DI, Shpigun OA, et al.

Time-efficient LC/MS/MS determination of low concentrations of methylphosphonic acid.

Inorganic Materials. 2017;53(14):1382-5.

10. Lee HK, Lee HSN, Sng MT, Basheer C. Determination of degradation products of

chemical warfare agents in water using hollow fibre-protected liquid-phase microextraction

with in-situ derivatisation followed by gas chromatography–mass spectrometry. Journal of

Chromatography A. 2007;1148(1):8-15.

11. Liu Q, Hu X, Xie J. Determination of nerve agent degradation products in

environmental samples by liquid chromatography–time-of-flight mass spectrometry with

electrospray ionization. Analytica Chimica Acta. 2004;512(1):93-101.

12. Desoubries C, Chapuis-Hugon F, Bossée A, Pichon V. Three-phase hollow fiber liquid-

phase microextraction of organophosphorous nerve agent degradation products from

complex samples. Journal of Chromatography B. 2012;900:48-58.

13. Sng MT, Ng WF. In-situ derivatisation of degradation products of chemical warfare

agents in water by solid-phase microextraction and gas chromatographic–mass

spectrometric analysis. Journal of Chromatography A. 1999;832(1):173-82.

24

14. Birkemeyer C, Kolasa A, Kopka J. Comprehensive chemical derivatization for gas

chromatography-mass spectrometry-based multi-targeted profiling of the major

phytohormones. Journal of Chromatography A. 2003;993(1-2):89.

15. Dival L. The development of an in-field rapid derivatisation technique for the analysis

of chemical warfare agent degradants [Masters dissertation]: Murdoch University; 2018.

16. Black RM, Read RW. Analysis of degradation products of organophosphorus chemical

warfare agents and related compounds by liquid chromatography–mass spectrometry using

electrospray and atmospheric pressure chemical ionisation. Journal of Chromatography A.

1998;794(1-2):233-44.

17. D’Agostino PA, Hancock JR, Provost LR. Packed capillary liquid chromatography–

electrospray mass spectrometry analysis of organophosphorus chemical warfare agents.

Journal of Chromatography A. 1999;840(2):289-94.

18. Katagi M, Tatsuno M, Nishikawa M, Tsuchihashi H. On-line solid-phase extraction

liquid chromatography-continuous flow frit fast atom bombardment mass spectrometric

and tandem mass spectrometric determination of hydrolysis products of nerve agents alkyl

methylphosphonic acids by p-bromophenacyl derivatization. Journal of Chromatography A.

1999;833(2):169-79.

19. D'Agostino PA, Provost LR. Capillary column isobutane chemical ionization mass

spectrometry of mustard and related compounds. Biomedical & Environmental Mass

Spectrometry. 1988;15(10):553-64.

25

20. D'Agostino PA, Chenier CL, Hancock JR. Packed capillary liquid chromatography-

electrospray mass spectrometry of snow contaminated with sarin. Journal of

Chromatography A. 2002;950(1-2):149-56.

21. Noort D, Hulst AG, Platenburg DHJM, Polhuijs M, Benschop HP. Quantitative analysis

of O-isopropyl methylphosphonic acid in serum samples of Japanese citizens allegedly

exposed to sarin: estimation of internal dosage. Archives of Toxicology. 1998;72(10):671-5.

22. Zhou S, Hamburger M. Application of liquid chromatography-atmospheric pressure

ionization mass spectrometry in natural product analysis. Evaluation and optimization of

electrospray and heated nebulizer interfaces. Journal of Chromatography A.

1996;755(2):189-204.

23. Mill T, Gould CW. Free-radical oxidation of organic phosphonic acid salts in water

using hydrogen peroxide, oxygen, and ultraviolet light. Environmental Science &

Technology. 1979;13(2):205-8.

24. Fenimore DC, Davis CM, Whitford JH, Harrington CA. Vapor phase silylation of

laboratory glassware. Analytical Chemistry. 1976;48(14):2289-90.

25. Subramaniam R, Åstot C, Juhlin L, Nilsson C, Östin A, Umeå U, et al. Direct

derivatization and rapid GC-MS screening of nerve agent markers in aqueous samples.

Analytical Chemistry. 2010;82(17):7452-9.

26. Parris NA. Instrumental liquid chromatography: a practical manual on high-

performance liquid chromatographic methods. Rev. i.e. 2nd ed. Amsterdam, Neth: Elsevier;

1984.