CHAPTER 1: Development, Historical Use and Properties of Chemical Warfare Agents

The year 2014 was the centenary of the commencement of the 1914–18 war [World War I (WWI)], a conflict that resulted in more than 20 million deaths and unprecedented numbers of casualties. A notorious development of that conflict was the widespread use of chemical weapons, manufactured on an industrial scale. Since 1914 more than 700 000 tonnes of chemical agents have been produced by various nations but, fortunately, since 1918 the use of chemical weapons in warfare has been the exception rather than the rule. Nevertheless, their use in conflicts in Iraq in the 1980s and more recently in Syria in 2013, particularly against unprotected civilian populations, has served as a reminder that the dangers still exist, even though a near comprehensive treaty, the Chemical Weapons Convention (CWC), entered into force in April 1997.¹  This chapter provides the reader with a brief history of the development and use of chemical weapons, a summary of the physicochemical properties that determine the primary hazard posed by chemical agents and how they might be used, and finally an overview of the main classes of chemical warfare (CW) agents known to have been weaponised, or which are known or suspected to have progressed to advanced development at some stage.

Dating from ancient times there had been sporadic exploitation of toxic chemicals for use in warfare; examples include poisoned arrows, the burning of sulfur to produce asphyxiating fumes, and the use of crude irritants to drive defenders into the open. The lachrymator ethyl bromoacetate was used in France for law enforcement operations from 1912–1914. More detailed historical accounts are available.^2–4  These relatively minor and small scale uses of chemicals had been sufficient to warrant the drafting of prohibitive articles as part of two conventions on the conduct of land warfare. The 1899 Hague Convention forbade the use of poison or poisoned weapons, and included a declaration that parties would ‘abstain from using projectiles the sole object of which is the diffusion of asphyxiating or deleterious gases’; these were reaffirmed in the 1907 Hague Convention.⁵ 

The most significant development in the decades before WWI was the rapid expansion of the chemical industries of the main protagonists. Commodity and other chemicals were being produced in large industrial plants on multi-tonne scales, and this laid the foundation for the subsequent production of CW agents on a scale suited to WWI battlefields.

The advent of the tactical and strategic large scale use of chemical weapons occurred almost one year into WWI. Major land battles, mostly in Europe and Russia, involved thousands of soldiers entrenched in high concentrations on vast, muddy and relatively flat areas of land. These attritional battles often continued for weeks or even months, many resulting in stalemate. The high concentration of relatively static entrenched combatants, initially with little protection, was seen as an ideal target for chemical weapons, offering a possible means of breaking the stalemate.

The main CW agents used in WWI were various irritants, chlorine, phosgene, and in the later stages sulfur mustard.^6–9  In addition to their direct effects, the psychological effects of fear of exposure proved to be a significant factor. It is estimated that approximately 125 000 tons of agent were used, resulting in around 1.25 million casualties, of which 90 000–100 000 were mortalities. Chemical agents accounted for approximately 4% of combat deaths in WWI.⁶  The effectiveness of chemical attacks was gradually reduced as protective countermeasures were introduced.

The first minor and ineffective use of chemical weapons in WWI occurred in August 1914 when French forces fired 26 mm grenades containing the lachrymator ethyl bromoacetate at German troops. A much larger scale dissemination of an irritant occurred in January 1915, when German forces fired 18 000 artillery shells containing xylyl bromide at Russian positions west of Warsaw during the battle of Bolinov. This too was largely ineffective because the agent froze in the cold conditions of the eastern front. Use of various lachrymatory and respiratory irritants continued, with more than twenty being used in WWI.^10,11 

The seminal incident generally regarded as the advent of large scale tactical CW occurred at the battle of Ypres, Belgium, on 22 April 1915. German troops released 168 tons of the industrial gas and lung injurant chlorine from 5730 cylinders, late in the afternoon and in a slight breeze towards entrenched allied soldiers. Chlorine is much denser than air, and lingered close to the ground and trenches; casualty numbers are uncertain but probably totalled several thousand.⁸  Ironically this first use was more effective than German forces had anticipated, and they failed to capitalise on the resultant breach in the allied defences. Further releases of chlorine occurred on both the western and eastern fronts in 1915. It was used much less successfully by British forces at Loos in October 2015, when a late change of wind direction moved the toxic cloud back towards entrenched British soldiers.⁸  As well as relying on wind direction for dissemination when released from cylinders, chlorine was readily detected by its familiar odour and the visible yellow–green cloud. Basic protective countermeasures were developed and chlorine on its own was gradually abandoned in favour of phosgene, although use continued in admixture with phosgene and chloropicrin.¹² 

The year 1915 saw the first use, initially by France, of the highly volatile industrial chemical phosgene.¹²  Phosgene is three to four times more potent than chlorine as a lung injurant and was to become the most effective of the lethal agents used in WWI, causing approximately 85% of the deaths from exposure to chemical agents. Phosgene was more insidious than chlorine, being colourless, its odour (fresh mown hay) less obvious, and onset of serious overt effects (pulmonary oedema) delayed for several hours. It was initially released from cylinders alone or mixed with chlorine, but was later liquefied and used in projectiles. Approximately 36 000 tons was manufactured for use in WWI. Germany preferred to use diphosgene (trichloromethyl chloroformate), which has a similar toxicity to phosgene but is a liquid [boiling point (bp) 128 °C]. It was therefore more persistent, and more easily handled and loaded into projectiles.⁸  Several less effective lung injurants were used in WWI.¹² 

The most effective chemical agent of WWI proved to be the liquid vesicant sulfur mustard, which caused more casualties than all of the other agents combined, even though it was introduced late in the war.^6,8  Its first use was by Germany against French forces at Ypres on 12 July 1917, producing an estimated 15 000 casualties. Development and use by British, French and US forces soon followed. Not only did sulfur mustard cause serious blistering by contact with the skin, its vapour caused serious damage to the eyes and lining of the respiratory tract. One of the enduring images of WWI is a chain of walking mustard casualties with bandages over their damaged eyes. Although only around 2% of mustard casualties were mortalities (mostly from secondary lung infections), the large number of injured soldiers caused huge logistic problems for medical treatment.

Both sides experimented with other agents, mostly volatile noxious industrial chemicals but also some solid agents.^8,13,14  Examples of eye irritants were bromoacetone, bromobenzyl cyanide, chloropicrin and ethyl iodoacetate. Chloropicrin, which was first used by Russian forces, caused a significant number of deaths when used at high concentration. More toxic chemicals included hydrogen cyanide (HCN), cyanogen chloride and hydrogen sulfide.¹⁵  HCN was largely ineffective as a lethal agent. It is slightly less dense than air and rapidly dispersed, making it difficult to sustain effective dosage levels with the munitions available. The French later used the denser cyanogen chloride. Several respiratory irritant arsenicals, some also with vesicant action, were used in the later stages of WWI, e.g. ethyl-, methyl- and phenyldichloroarsine, diphenylchloroarsine (DA) and diphenylcyanoarsine (DC).¹¹  DA and DC irritated eyes and mucous membranes of the nose and throat, causing sneezing and coughing, and induced vomiting. As solids disseminated as crude particulate aerosols rather than vapour, they were conceived by Germany as possible mask breakers, and used in combination with more toxic lung injurants. They were countered by inclusion of a mechanical filter in the mask. Two additional arsenicals, the newly developed vesicant lewisite, and the respiratory irritant/vomiting agent adamsite (DM), would have become available had the war lasted beyond 1918.

Following the experiences of WWI, a strong desire to prohibit the use of poison gas led to the 1925 Geneva Protocol, which prohibited ‘the use of asphyxiating, poisonous and other gases, and of all analogous liquids, materials and devices’ (and bacteriological methods of warfare).¹⁶  It did not prohibit production of CW agents and several signatories reserved the right to retaliate in kind following first use against them. Notwithstanding this protocol, further development of existing and new chemical agents, plus more effective munitions and delivery systems, continued.¹⁷  Examples of agents included the further development of the arsenicals lewisite and DM, the vesicant nitrogen mustards,¹⁸  and the ‘nettle gas’ or urticant phosgene oxime.¹⁹  The latter causes severe pain and lesions of the skin and was conceived as a possible means of forcing removal of respirators; fortunately it has poor stability. The lachrymator 1-chloroacetophenone (CN) was developed as the standard riot control and harassing agent by the USA and adopted in other countries.²⁰  DM was also used as a riot control agent (RCA) during this period in the USA and other countries, and became a significant component of military arsenals. By the onset of World War II (WWII) a broader variety of munitions for CW agents was available as well as the means to deliver them from artillery and aircraft. Research continued into more effective defensive countermeasures (detection, protective clothing, respirators, medical countermeasures and decontaminants).

CW agents were reportedly used in several localised, mainly colonial, conflicts in the interwar years.²¹  Of particular note was the first major use of airborne delivery when Italy sprayed sulfur mustard from aircraft during its invasion of Abyssinia in 1935/36. A year later Japan began using chemical weapons on the Chinese mainland.

The most important development during this period, which occurred during the build up to WWII, was the chance discovery and military development of organophosphorus nerve agents in Germany. In 1936, the chemist Gerhardt Schrader, working in the laboratories of the industrial conglomerate IG Farben, synthesised an experimental organophosphorus insecticide that proved to be an order of magnitude more toxic to animals than any agent used in WWI. The discovery was disclosed to the military, and the chemical was developed in secrecy into the CW agent named tabun.²²  This development was to dominate offensive and defensive chemical weapon research and development for the next 40–50 years.²³ 

In spite of fears to the contrary, chemical weapons were not used in Europe in WWII. One theory was that Hitler had an aversion to CW from his experiences in WWI. Perhaps a more convincing argument was that first use would simply result in retaliation in kind. The major protagonists had built large stockpiles of chemical weapons in the build up to and during WWII, chiefly sulfur mustard, small quantities of nitrogen mustards, lewisite, various other arsenicals, phosgene, HCN and cyanogen chloride.⁸  The UK had plans to use the persistent agent sulfur mustard on the beaches in the event of a German land invasion. HCN in the form of Zyklon B, a commercial pesticide formulation of HCN adsorbed onto a solid support, was used to kill up to a million people in the gas chambers by Nazi Germany. Japan continued to use chemical agents on mainland China, chiefly sulfur mustard, and possibly lewisite, DC and phosgene.

Germany produced approximately 12 000 tons of tabun for weaponisation before and during WWII. Systematic molecular modification of tabun led to the more potent and volatile nerve agent sarin, which was later to become one of the major components of modern chemical arsenals.^8,23  A small quantity of sarin (estimated variously from 500 kg to 10 000 tons) was produced on a pilot plant scale in Germany. A less volatile analogue of sarin, given the name soman, was developed by Germany towards the end of the war (see Section 1.4.4).

Britain and the USA had also researched organophosphorus compounds as potential CW agents during this period but the leading candidate, O,O-diisopropyl fluorophosphate (DFP), was significantly less potent than tabun.²⁴ 

Following the end of WWII, details of the German programme were obtained by allied intelligence. The tabun plant at Dyhernfurth was dismantled and rebuilt in Russia, and small stocks of tabun (given the US military designator GA, G = German) were acquired by the USA and UK. Interest in tabun then gradually waned as sarin (GB) became the favoured nerve agent. The immediate post war years saw the intensive exploration of analogues of sarin and soman (GD), by the UK, USA, Russia, France and other nations.²⁵  Sarin was later weaponised and stockpiled by the USA and the Soviet Union; soman was weaponised and stockpiled by the Soviet Union neat and in a thickened form as a more persistent nerve agent. Other nations, e.g. France, built pilot plants and acquired small stockpiles of sarin.

History was repeated in 1952 when a low volatility organophosphorus pesticide, later named amiton, was discovered in the plant protection laboratories of Imperial Chemical Industries (ICI) in the UK.²⁶  The notable feature of amiton was its unusually high percutaneous toxicity in rodents and rabbits. The discovery was disclosed to the UK Porton Down laboratory, where molecular modification led to the development of the low volatility V (venomous) agents in a collaboration with the USA and Canada. Similar agents were developed independently in the Soviet Union. The USA weaponised the analogue designated VX, based mainly on a combination of its toxicity and storage stability. The Soviet Union weaponised an isomeric analogue codenamed R-33, usually now referred to as RVX or VR. These agents were developed and stockpiled with the possibility of a major conflict between western and Soviet forces, e.g. on the northern plains of Europe, sarin as a rapid acting non-persistent casualty producing agent, and V agents and thickened soman as persistent casualty producing and disruptive agents, for terrain denial, and for attacking rear logistical sites.²⁷  Large stockpiles of sulfur mustard were retained. The relative importance of these agents is reflected in the later declarations to the CWC (see Table 1.1).

From the 1970s, the USA and Soviet Union moved away from producing unitary agents (i.e. the final product) for weaponisation because of the hazards of production, storage and transport, plus the high cost of destruction when required. Both started to develop binary weapons, wherein two reactive precursors are placed in a munition separated in rupturable compartments. On firing, the barrier is broken and the rotation caused by rifling of the munition results in rapid mixing of the precursors to produce the agent during the time of flight.²⁸  The USA developed binary munitions for sarin and VX; the Soviet Union developed binaries for sarin and RVX.^29,30  Binaries may also be advantageous if the agent has poor storage stability.

Major advances were made in defensive countermeasures during this period, such as manual and automated detectors (for warning and monitoring contamination), protective clothing and respirators, decontaminants and medical countermeasures, the latter particularly against nerve agents.³¹  It was therefore not surprising that continuing research into potential new agents, both offensive research and defensive threat agent assessment, began to focus on possible means of defeating these countermeasures. Examples were canister penetrating volatile organofluorine agents (e.g. perfluoroisobutene, trifluoronitrosomethane),^29,30  fast acting vomiting agents to force removal of the respirator,²⁷  and intermediate volatility nerve agents that could penetrate semi-permeable suits.^29,30  Most of these experimental agents had significant shortcomings.

CN was gradually replaced by the more effective and less toxic irritant 2-chlorobenzilidene malononitrile (CS) for law enforcement, riot control and military use; pepper spray was widely introduced for self defence or as an aid to arrest. The use of irritants for civilian riot control increased substantially during these years in an age of widespread protest.

From the 1950s there were programmes in several countries aimed at exploiting the rapidly increasing pace of pharmaceutical and veterinary drug discovery, particularly centrally acting drugs, for military or law enforcement purposes.^20,32  In 1963 the USA weaponised the anticholinergic compound 3-quinuclidinyl benzilate (BZ) as a military incapacitating agent.

Chemical weapons have been used in at least four conflicts in the Middle East, and it remains the most volatile region with regard to CW. Sulfur mustard and phosgene were reported to have been used by Egypt in its intervention in the North Yemen civil war in the mid-1960s.²¹  Alleged use of sarin in the Yemen was not confirmed. Iraq used sulfur mustard, tabun and sarin in the conflict with Iran from 1980–1988, causing an estimated 150 000 casualties.^33–35  Iraq also used sulfur mustard and sarin against its indigenous Kurdish population, the most infamous occurrence being in the city of Halabja in March 1988, where between 3000 and 5000 deaths were reported, mostly unprotected civilians.³⁶  Multiple agents appear to have been used in Halabja including sulfur mustard and nerve agents—probably sarin but no analytical evidence confirming the lethal agent used has been published. Four years later, retrospective investigations of an attack on the Kurdish village of Birjinni provided the first analytical evidence corroborating the use of sarin and further use of sulfur mustard.^37,38  In the aftermath of the Gulf War with Iraq, there were incidents of crude use of chlorine by insurgents. Most recently (2013) the use of sarin was confirmed in the internal conflict in Syria, although the UN investigative mission did not identify the perpetrators.³⁹ 

Although there have been incidents of small scale terrorist or criminal use of irritants, poisons and powders contaminated with the toxin ricin, the perceived threat of moderate scale dissemination of chemical agents by terrorists has not materialised other than in Japan in the 1990s.⁴⁰  In Matsumoto City, June 1994, members of the Aum Shinrikyo religious cult disseminated vaporised sarin from a van towards an apartment block, targeting three judges overseeing a land dispute with the cult. The judges survived but seven others were killed with around 270 casualties. The second use of sarin by Aum Shinrikyo was to have a major impact on home security programmes throughout the world.⁴⁰  On 20 March 1995, an estimated 20 kg of crude sarin was released by puncturing plastic bags containing the agent on trains on five subway lines converging towards government offices and the central police headquarters. The attack resulted in 12 deaths with approximately 1100 serious casualties. Up to 4000 mildly exposed personnel or ‘worried well’ presented themselves to hospitals, and the attack left a psychological imprint on many thousands of people in Japan and elsewhere. The main reasons for the low number of deaths were the crudeness of the agent and rather slow method of dissemination. The cult also used VX in an assassination.⁴¹  In some aspects Aum Shinrikyo was an exception in that it was a large and wealthy organisation operating for many years with a degree of impunity within a developed nation, a situation that is much less likely today.

With the increasing terrorist threat of hostage taking and aircraft hijacking, the search for potent knockdown incapacitating chemicals for counter-terrorist operations continued in several countries. In 2002, Russian Special Forces ended a siege of a Moscow theatre by disseminating an aerosol containing two analogues of the opioid analgesic/anaesthetic fentanyl into the theatre.^42,43  At least 130 out of the 800 hostages plus 40 terrorists died in the operation, most from exposure to the agent.

As described above, massive stockpiles of chemical weapons (approximately 70 000 tonnes declared by 2001), mainly vesicants and nerve agents, were accumulated in the cold war years, mostly by the USA and the Soviet Union. After years of negotiation, a near comprehensive chemical weapons treaty, the CWC, was opened for signature in April 1993 and entered into force in October 1997.^1,44  Unlike previous treaties, which attempted unsuccessfully to prevent first use, the CWC prohibits the development, manufacture, stockpiling and use of toxic chemicals in warfare. Furthermore, it requires all stockpiles to be declared and destroyed. The deadline for the latter was initially 2007, but this was extended first to 2012 and now to 2020 because of the technical and economic complexities of destruction. A total of 190 of the UN recognised nations have ratified the CWC, only six exceptions remain. As of December 2014, Angola, Egypt, North Korea and South Sudan had not signed the Convention, and Israel and Myanmar had signed but not ratified. Following recent conflicts, CW agent possessor states Iraq, Libya and Syria have signed and ratified the Convention. Table 1.1 summarises the total declarations of agents to the Organisation for the Prohibition of Chemical Weapons (OPCW) by 2001, underlining the importance of sulfur mustard and nerve agents.⁴⁵ 

CW agents have been classified in several ways.^46,47  They can be grouped simply by their predominant gross effect at realistic concentrations:

• Lethal, tissue damaging (casualty producing), irritant (harassing), incapacitating.

Alternatively, they may be classified more specifically according to their main physiological effects:

• Vesicants (blister agents), lung injurants (choking agents), blood agents, nerve agents, irritants (skin, eye and respiratory), incapacitants.

A third way of classifying agents is according to their physicochemical properties:

• Non-persistent (moderate to high volatility), persistent (low volatility).

Physicochemical properties are a major determinant of the hazard (as opposed to the toxicity) of an agent, and hence the most likely mode of use.^46–48  Although the terms mustard gas, nerve gas, etc. are in common usage, most CW agents are liquids. Efficient dissemination is key to the effectiveness of a chemical weapon. Irrespective of the agent’s properties, it needs to be dispersed evenly over the target area.²⁸ 

Non-persistent agents (e.g. sarin, phosgene and HCN) have moderate to high vapour pressure and readily vaporise; the least persistent are gases at ambient temperatures. The major portal of entry is the lung, although the eyes may also be important. Non-persistent agents have been disseminated using modified conventional munitions as a mixture of vapour, aerosol and small droplets, according to their vapour pressure and the energy used for dissemination. They soon disperse on the battlefield and need to be disseminated in proximity to the ground. Early evening to early morning is generally the most effective time to disseminate non-persistent agents, when air temperature stratification is neutral or inverted (coldest nearest the ground), and turbulence is minimal. Non-persistent agents would rarely (e.g. under very cold conditions) need to be decontaminated in the field; some are rapidly degraded in the environment. The general military philosophy has been that a non-persistent agent would be the most effective for an on- or near-target attack to cause rapid casualties, particularly in a surprise attack before defensive countermeasures are in place, and where the attacker may seek to occupy the area attacked. Sarin is the most important non-persistent agent; onset of effects by inhalation occurs in minutes. Such an agent is also attractive to terrorists because it is easier to disseminate compared with persistent agents. The most volatile non-persistent agents, e.g. phosgene, chlorine and HCN, are generally regarded as obsolete because it would be difficult to achieve effective concentrations over a modern dispersed battlefield. They could be used effectively against unprotected civilians.

Persistent agents have low vapour pressures and most produce insufficient vapour concentration to cause large numbers of casualties through inhalation. Their major portal of entry is through the skin. With the most important persistent agents, i.e. sulfur mustard and V agents, onset of effects is slow (up to several hours) by this route of exposure. Unless aerosolised, a persistent agent is most likely to be used in the form of droplets for contaminating the ground and equipment, rear logistical sites and supply routes. Against well protected and trained personnel they are predominantly disruptive, forcing defenders into individual protective equipment, with the attendant physical, physiological and psychological impositions, along with impaired communication. The most effective persistent agents have some resistance to environmental degradation. They may persist on the battlefield for days or even weeks, depending on weather conditions (temperature, wind and precipitation).²⁸  Wet and windy conditions can reduce persistence from days to hours. Decontamination is targeted mainly at persistent agents. Persistent agents can be disseminated from moderately high altitude, particularly when thickened by the addition of a few percent of a polymer to prevent dispersion before the droplets reach ground level. Thickening also improves the ballistics of munitions and makes decontamination more difficult. The Soviet Union declared thickened sulfur mustard, lewisite, soman and V agent during a confidence building visit to the Shikhany proving ground during negotiations towards the CWC.²⁹ 

There is of course no clear demarcation between persistence and non-persistence, with a continuum of volatilities across a broad spectrum, and with a strong dependence on the temperature of the environment. Chemicals with intermediate volatility can be very effective, the prime example being sulfur mustard. Depending on weather conditions, sulfur mustard can produce sufficient vapour to cause casualties through inhalation, eye exposure and contact with moist sensitive areas of the skin, but can present a persistent contact hazard on the ground when present as droplets (see Section 1.4.3.1). Unless thickened, soman is rather too volatile to be an effective intermediate volatility agent (IVA), but the less volatile cyclosarin (GF) and 2-methyl GF fall within the intermediate volatility range. Boiling points and volatilities of the main agents at 25 °C are shown in Table 1.2.⁴⁹  Volatility is defined as the maximum concentration of vapour in equilibrium with liquid agent in a confined space and at a defined temperature; it is derived from vapour pressure data. In an open space such as a battlefield only a small percentage of this value is likely to be achieved. Volatility is highly dependent on temperature, with a 10 °C increase above 20 °C approximately doubling the volatility.

Other physical properties are also important. Surface tension determines the extent of spreading (e.g. oil versus water). Most liquid CW agents resemble organic liquids or oils, and have a significantly lower surface tension than water. They tend to spread on surfaces, getting into parts that are difficult to decontaminate with water-based decontaminants. Viscosity determines thickness, how readily an agent sticks to surfaces, and drop size on dissemination.

Solid agents are a special case in that they are generally dispersed as aerosols, fine particles with particle sizes in the respiratory range (1–10 µm diameter).⁴⁸  Particulate aerosols are most efficiently generated thermally or pneumatically, targeting the lung as the primary portal of entry. One of the most efficient and rapid means of disseminating solid agents is using multiple pyrotechnic sub-munitions. The agent is mixed with a pyrotechnic formulation, it is rapidly (within seconds) vaporised at high temperature, and immediately condensed to a particulate aerosol in the cold air. In the 1960s the USA designed cluster bombs intended to be filled with multiple pyrotechnic sub-munitions containing the irritant CS or the incapacitant BZ.⁵⁰  Such cluster munitions were designed to disseminate effective concentrations over an area up to around 1 hectare within masking time. Persistence of aerosolised solid agents is generally considered to be low, although the residual hazard from impacted or deposited aggregated solid agent has not been clearly defined.

A chemical property that has been one of the major factors in determining the extent of proliferation of CW agents is the ease of production.^51–55  To a crude rule of thumb, around 1 tonne of agent is considered the minimum required for a small but militarily significant attack, although disruption can be caused by smaller quantities, or even by a credible threat of use. For terrorist purposes, considerably lower quantities could be effective, particularly if panic and publicity are the desired effects.

In WWI, chlorine, phosgene and some other industrial chemicals were available in multi-tonne quantities. Sulfur mustard was easily synthesised from industrially available precursors by a one or two stage process.⁵⁵  Nerve agents present a somewhat greater challenge but should not be too difficult for a nation with a moderately developed chemical industry. Nerve agents require between three and eight stages depending on the agent and precursors available, the ease of synthesis being tabun > sarin > cyclosarin > soman > RVX/VX.^51–53  It is instructive to follow the order of development of chemical agents by Iraq in the 1980s. Sulfur mustard was the first agent to be produced and weaponised, followed in order by tabun, sarin, sarin-GF mixture, and finally VX, which proved problematic. This is in the order of complexity of production. Cyclosarin was selected rather than soman because of the greater availability of a key precursor, cyclohexanol, rather than pinacolyl alcohol required for the production of soman.

Stability on storage is an important aspect of chemical production.⁵⁴  US and Soviet Union cold war stockpiles were produced for an uncertain future conflict. Long term storage stability (e.g. 10–20 years) was therefore imperative. In contrast, Iraq in its conflict with Iran tended to produce and use. Stability is partly dependant on the purity of the agent and on the addition of stabilisers, particularly for nerve agents. Decomposition of stored agents tends to be self-accelerating. Reaction of nerve agents and sulfur mustard with traces of moisture releases hydrofluoric or hydrochloric acid, which accelerate further degradation. Stabilisers such as diisopropylcarbodiimide or tributylamine are added to scavenge any moisture or acid formed.

Chlorine, Cl₂, is a corrosive industrial gas (bp −34 °C) with a pungent and characteristic odour, and yellow–green colour.¹²  It is produced industrially by the electrolysis of brine and used extensively in the production of industrial chemicals and consumer products, and for disinfection. Chlorine is a strong oxidizing agent and is destructive towards lung and eye tissues, but a relatively high concentration is required to cause death. Although obsolete as a military agent after WWI, there have been isolated incidents of crude opportunistic chlorine dissemination by insurgents in Iraq and Syria, and allegations of chlorine use in Bosnia in 1993.

Phosgene, COCl₂ (military designator CG), is a colourless gas (bp 8 °C) except at low ambient temperatures, with an odour of freshly mown hay.¹²  It is produced on a multi-million tonne scale from carbon monoxide and chlorine, and is used for the manufacture of a broad range of chemical products. As a lethal CW agent it is approximately 3–4 times more potent than chlorine and, as described in Section 1.2.2.4, was the most effective lethal agent used in WWI. Its effects are much more insidious than chlorine, initially causing mild irritation of the throat followed by a latent period of up to 24 hours to induce pulmonary oedema, by which time its effects are life threatening. It is assumed that phosgene reacts with various nucleophilic sites on macromolecules in the lung but the precise mechanism of action is unknown. Effective concentrations of phosgene would be difficult to sustain on a modern dispersed battlefield. As a military agent it is generally regarded as obsolete although it could still be used effectively against unprotected civilians.

Diphosgene or trichloromethyl chloroformate, ClCO₂CCl₃, (DP) is a volatile liquid (bp 128 °C).¹²  Industrially it is used in the same way as phosgene. It has a similar toxicity to phosgene but is more persistent, and offers the advantage of being more easily handled. The name diphosgene derives from its disproportionation into two molecules of phosgene on heating or catalysis.

Perfluoroisobutene, (CF₃)₂C=CF₂ (bp 7 °C), included in Schedule 2 of the CWC, is also a lung injurant that causes pulmonary oedema. It is a by-product of Teflon production. Like phosgene it is a reactive electrophile. It is not known to have been weaponised but was studied as a potential hydrophobic canister penetrant.⁵⁶ 

HCN (AC) is a colourless liquid or gas (bp 25.7 °C).¹⁵  Its odour of bitter almonds is not perceptible by some people, although a bitter taste in the mouth may be evident. It is produced industrially by a catalytic reaction of methane with ammonia. Its many uses include chemical and polymer production, in electroplating, and as a fumigant pesticide and rodenticide. Liquid HCN is prone to violent polymerisation and requires stabilisation when stored in bulk. Unlike chlorine and phosgene, HCN vapour is marginally less dense than air and its sparse use in WWI was largely unsuccessful because it was difficult to obtain effective concentrations with the munitions available. In stark contrast to phosgene, its effects at lethal concentrations are rapid in onset (seconds to minutes). It prevents cells, including blood cells, from utilising oxygen by inhibiting the enzyme cytochrome C oxidase.⁵⁷ 

Cyanogen chloride, ClCN (CK), is also a volatile liquid or gas (bp 13.1 °C); it was used more successfully than HCN in WWI mainly due to its higher density.¹⁵  It is less potent than HCN as a lethal agent, but is a respiratory irritant at sub-lethal exposure concentrations and is thus more easily perceived. Cyanogen chloride is widely used in the chemical industry. HCN and cyanogen chloride are regarded as obsolete CW agents.

Sulfur mustard, bis(2-chloroethyl)sulfide (Scheme 1.1), military designator in distilled form HD, is a medium to low volatility liquid (bp 213 °C) with a consistency resembling a light oil, and an odour (due to impurities) reminiscent of mustard or garlic. It was first synthesised, and its vesicant properties noted, in the 19th century, but was not fully characterised until its development as a CW agent by German scientists during WWI. After devastating use in WWI, sulfur mustard has remained one of the major threat agents, as illustrated by the quantities shown in Table 1.1 declared to the OPCW (and these figures exclude late signatories to the CWC, Iraq, Syria and Libya).

Sulfur mustard is very easily made from single stage processes from industrial chemicals (sulfur monochloride or dichloride plus ethylene; or thiodiglycol plus hydrogen chloride).⁵⁵  It has close to ideal physicochemical properties for a disruptive CW agent except for a high freezing point (melting point 14 °C when pure). It is generally persistent when dispersed as droplets posing a prolonged contact hazard but, under moderate temperatures, still produces sufficient vapour to damage eyes, lungs and sensitive areas of the skin. Larger sized drops can be much more persistent than predicted from the vapour pressure. This results from oligomers being formed at the air–liquid interface. Mustard reacts with water at the two electrophilic carbon atoms, and with oxygen at the nucleophilic sulfur atom.⁵⁸  Large ‘footballs’ of sulfur mustard, protected by a polymeric outer coating, are still being dredged up by fisherman from stockpiles dumped in the Baltic Sea after WWII.⁵⁹ 

Although sulfur mustard reacts rapidly with water when in solution, its degradation in the environment is limited by its very low affinity for water (solubility 0.092 g/100 g at 22 °C).¹⁹  The latter property also makes it one of the more difficult agents to decontaminate, and more robust for long term storage compared with nerve agents.

The freezing point of sulfur mustard has been reduced in a number of ways to prevent the agent from solidifying in weapons in cold weather. In WWI, mustard was mixed with various solvents, e.g. carbon tetrachloride and benzene. In WWII, Britain produced it from thiodiglycol and hydrogen chloride as a 6 : 4 mixture with the oligomer ‘T’ (Scheme 1.1), also known as O mustard. T has somewhat greater vesicant activity than sulfur mustard, is less volatile and more persistent. Other nations mixed mustard with lewisite, which also accelerated the onset of effects and increased the vapour hazard.

Medical treatment of mustard injuries causes major logistical problems, and military casualties may be unable to perform duties for weeks or permanently. It seems remarkable that there is still no effective treatment for mustard lesions other than symptomatic and palliative treatment. One of the reasons is that, unlike nerve agents, which undergo a selective and catalytic chemical reaction with the enzyme acetylcholinesterase (analogous to Erlich’s ‘magic bullet’ in a drug context), sulfur mustard is more akin to a ‘shotgun’, alkylating many nucleophilic groups on DNA, proteins and other macromolecules. Thus far, the key targets mediating lesions (other than DNA for carcinogenicity) have not been identified.

A number of analogues and oligomers of sulfur mustard with similar or slightly greater vesicant activity were developed as agents of lesser importance (Scheme 1.1). The homologue Q, sesquimustard, is a solid (melting point 57 °C) and like T was mixed with sulfur mustard.

The three nitrogen mustards HN-1, HN-2 and HN-3 (Scheme 1.2) are tertiary amines substituted with 2-chloroethyl groups similar to sulfur mustard. As free bases they are low volatility liquids, generally with poor stability, but form more stable water soluble solid hydrochloride salts.¹⁸  They were partially developed as CW agents during the 1930s but there has been no confirmed use. In WWII Germany produced 2000 tons of HN-3; the USA produced approximately 100 tons of HN-1 in a pilot plant.³  The most important of the N-mustards is HN-3, which is very easily made from the widely used industrial chemical triethanolamine by chlorination with thionyl chloride. It is more stable on storage than HN-1 and HN-2, and its vesicant activity as a liquid approaches that of sulfur mustard. HN-3 (bp 257 °C) is significantly less volatile than sulfur mustard and the vapour hazard is low except under very hot conditions. It might therefore be more effective than sulfur mustard as a persistent agent in hot climates. HN-2 has been used to treat some types of cancer.

Lewisite, CHCl=CHAsCl₂, 2-chlorovinyldichloroarsine, named after its discoverer W. L. Lewis, was produced by the USA and shipped to Europe in 1918, too late to be used in WWI. Between the wars it was also produced by Japan and the Soviet Union. It is relatively easily made from arsenic trichloride and acetylene, although the process is technically more difficult than the production of sulfur mustard. Lewisite is more volatile (bp 190 °C) than sulfur mustard and hence it is less persistent; it also appears to be more sensitive to environmental moisture. In contrast to sulfur mustard, its initial effect (skin pain or irritation) is almost instant, and blisters appear within a few hours.¹⁹  There has been no confirmed instance of use, although Japan is suspected of having used lewisite in China in WWII. In addition to being stockpiled as a neat agent, lewisite was mixed with sulfur mustard to speed up the onset of action and to depress the freezing point of the latter.

Several other liquid arsenicals with vesicant, lung and eye damaging effects were developed and used in WWI. Examples are methyl-, ethyl- and phenyl-dichloroarsine (MD, ED and PD) known as ‘dicks’. These agents were of low importance compared with sulfur mustard and are considered obsolete.

Tabun (GA), O-ethyl N,N-dimethyl phosphoramidocyanidate, was the first nerve agent to be weaponised following its discovery by Schrader in Germany in 1936.²²  It evolved from structure–activity studies of organophosphates related to O,O-diethyl fluorophosphate (diethyl phosphorofluoridate) (Scheme 1.3), whose toxicity had been reported some years earlier.⁶⁰  A feature of these organophosphates was a displaceable ‘leaving group’ (F, CN) on phosphorus, later shown to be displaced by a covalent reaction with a serine hydroxyl group in the active site of the enzyme acetylcholinesterase. By the end of WWII 12 000 tonnes of tabun had been produced in a plant at Dyhernfurth, which was later dismantled and reconstructed in Russia. The USA produced small stocks of tabun. The first known use of tabun was more than 40 years later by Iraq in the conflict with Iran.³⁴ 

Tabun is the easiest of the nerve agents to produce, essentially by a two to three stage process from industrially available chemicals.^51–54  It has less favourable physicochemical properties than the other weaponised nerve agents; its vapour pressure is quite low (bp 248 °C) and it is the least stable towards moisture in the environment. Added to this, its lower inhalation toxicity compared with sarin and soman, and its much lower percutaneous toxicity compared with VX, it is regarded as being obsolete as a military agent. Its ease of synthesis might make it attractive to proliferators with a limited chemical industry or to terrorists.

UK and US chemists were less successful in developing a nerve agent during WWII. The primary candidate was DFP (Scheme 1.3), studied by Saunders and colleagues at Cambridge University.^24,61  DFP had toxicity approximately one fifth to one tenth that of sarin, with volatility closer to soman. Its only advantage over sarin was ease of synthesis.

Further molecular modification of tabun-like compounds in Schrader’s laboratory produced sarin (named after the team that discovered it, Schrader Ambros Rüdriger and Van der Linde). Sarin (GB), O-isopropyl methylphosphonofluoridate (bp 150 °C), is more volatile and more potent than tabun (Scheme 1.4). It was produced in Germany on a pilot plant scale. Schrader subsequently developed the less volatile soman (GD), O-pinacolyl methylphosphonofluoridate (bp 198 °C). Soman was later to achieve high importance because of the resistance of exposed animals to medical countermeasures (specifically the resistance of inhibited acetylcholinesterase to reactivation with oximes). Soman became a leading candidate for weaponisation in the USA, and was weaponised on a moderate scale in Russia, neat and thickened.

After disclosure of the German nerve agent programme, chemists from several nations pursued further structure–activity studies over the next 20 years. One avenue explored was analogues of sarin and soman with intermediate volatility, i.e. between that of soman and tabun. Such agents (IVAs) could penetrate semi-permeable protective clothing, and present both an inhalation and contact hazard. Leading candidates in the US programme were cyclosarin (GF), O-cyclohexyl methylphosphonofluoridate (bp 239 °C), and 2-methyl GF, O-2-methylcyclohexyl phosphonofluoridate (bp 247 °C; Scheme 1.4). Neither is known to have been weaponised during this period, although a mixture of cyclosarin and sarin was used in crude binary form by Iraq in the 1980s.

In the 1940s it was recognised that nerve agents act by binding to and inhibiting the enzyme acetylcholinesterase.⁶²  This enzyme is the body’s mechanism for inactivating the neurotransmitter acetylcholine. Inhibition of the enzyme causes an excess of acetylcholine at nerve junctions and cholinergic neurones, producing excessive stimulation of cholinergic receptors. A logical avenue to increase affinity for the enzyme was to explore pesticide or nerve agent analogues with a structural feature that mimicked the natural neurotransmitter. Tammelin and co-workers^63,64  in Sweden published a series of papers on some highly toxic compounds known as Tammelin esters (Scheme 1.5), but these were solids and had poor stability.

At the same time, chemists in the plant protection laboratories of ICI in the UK were studying systemic pesticides with such features. This research produced amiton, which possessed unusually high percutaneous toxicity.²⁶  Modification of amiton in a UK/US/Canada military collaboration led to the development of the V series of nerve agents, characterised by low volatility, high percutaneous toxicity and high systemic toxicity. The analogue O-ethyl S-(2-diisopropylaminoethyl) methylphosphonothiolate, given the military designator VX, was assessed as possessing the optimum combination of toxicity and storage stability, and was later produced and weaponised by the USA. Chemists in Russia independently discovered the same series of compounds, eventually leading to the weaponisation of a close analogue of VX, O-isobutyl S-(2-diethylaminoethyl) methylphosphonothiolate, known as R-33, RVX or VR (Scheme 1.6).

Of the three types of weaponised nerve agents, sarin became the primary non-persistent agent of modern arsenals, and VX or RVX the primary persistent agent together with sulfur mustard. Tabun was the least effective nerve agent and gradually became redundant.

Two additional series of nerve agents are worthy of mention. Research on IVAs in several countries led to the analogue known as GV, O-(2-dimethylaminoethyl) N,N-dimethyl phosphoramidofluoridate (Scheme 1.7). The name GV was coined by Czech chemists to indicate properties of both G and V agents.⁶⁵  The US military designator was GP. GV is a hybrid structure incorporating structural features of tabun, sarin and V agent. GV had true intermediate volatility properties (bp 226 °C, volatility 527 mg m⁻³ at 25 °C),⁶⁶  producing sufficient vapour to cause an inhalation hazard, and possessing percutaneous toxicity approaching that of the V agents. GV might have become an important threat agent had it not had very poor storage stability. It has been suggested that a binary version might be feasible.⁶⁵ 

In recent years, there has been much speculation that a fourth generation of nerve agents, ‘Novichoks’ (newcomer), was developed in Russia, beginning in the 1970s as part of the ‘Foliant’ programme, with the aim of finding agents that would compromise defensive countermeasures.^67,68  Information on these compounds has been sparse in the public domain,^30,68–70  mostly originating from a dissident Russian military chemist, Vil Mirzayanov.⁶⁹  No independent confirmation of the structures or the properties of such compounds has been published.

RCAs are peripheral chemosensory irritants that target the eyes, airways and/or skin.^71,72  The 1997 CWC defines them as ‘Any chemical not listed in a Schedule, which can produce rapidly in humans sensory irritation or disabling physical effects which disappear within a short time following termination of exposure’.¹  Use for riot control purposes is permitted under the CWC, but not for military harassment, and stocks of RCAs must be declared. Most of the major RCAs [CN, CS and dibenz[b,f]-1,4-oxazepine (CR); Scheme 1.8] are low volatility solids and, unless they are used in solution in a spray, they need to be aerosolised for efficient use, for example using pyrotechnic munitions or dispersed as micronised powders.

More than 20 eye irritants (lachrymators) and upper airway irritants were used in WWI.⁷³  Bromobenzyl cyanide (CA, BBC) emerged as the most effective lachrymator. DM became available shortly after WWI. Both CA and DM were used for law enforcement and stockpiled as military harassing agents. CA was soon replaced in the 1920s by the more effective and safer lachrymator CN. CN in turn was largely superseded in the 1950s by the more potent and safer CS, which has remained the RCA of choice in many countries. DM is now considered too toxic to be used as a RCA. The most potent eye irritant of the RCAs, CR, was developed in the UK in the 1960s following its chance discovery in a university laboratory. CR has an exceptionally high safety ratio but, because it is difficult to decontaminate and residues may persist for long periods, it has rarely been used in civilian environments. Small stocks were declared to the OPCW. More recently, pepper spray [oleoresin capsicum (OC)], containing capsaicin and related irritants, or a synthetic analogue of capsaicin such as nonivamide [pelargonic acid vanillylamide (PAVA)], have been widely adopted as aids to arrest and in personal defence sprays. One other irritant is worthy of mention: 1-methoxycycloheptatriene (CHT) Scheme 1.8 is a volatile liquid (bp 115 °C) with powerful irritant action, particularly on the eyes. Although easily disseminated as a vapour, CHT has not been adopted because of toxicological concerns.⁷⁴ 

A search for military incapacitating agents began in the 1950s, and continued for at least four decades.^32,75–77  Incapacitants were later sought for law enforcement and counter-terrorism purposes, particularly after a spate of aircraft hijackings and other hostage situations in the 1960s and 1970s. This period saw a major expansion of the pharmaceutical industry, which invested heavily in research centres seeking new drugs. Large numbers of experimental drugs were studied in animals, in contrast to modern drug research, which uses animals more sparingly. Particular advances were made in drugs that affected the brain. Examples are morphine-like analgesics (opioids), various classes of intravenous anaesthetics, benzodiazepines (prescribed and manufactured in very large quantities), dopamine antagonists for the treatment of schizophrenia, and dopamine agonists for Parkinson’s disease. Some of the dopamine agonists investigated, such as derivatives of apomorphine, induced vomiting in dogs at microgram doses.^27,78 

An early candidate incapacitant in the USA was phencyclidine (sernyl, PCP; Scheme 1.9), which was marketed as an intravenous anaesthetic but soon withdrawn because of complex psychotomimetic effects. PCP has a simple structure and is exceptionally easy to synthesise and it became a major drug of abuse in the USA. It was eventually discarded as a potential military incapacitant because of its low potency and a tendency to induce unpredictable and sometimes violent behaviour. A related drug, ketamine, has more recently been studied by Czech scientists as a possible incapacitant in admixture with other depressant drugs.⁷⁹  The hallucinogen LSD was extensively researched as a disruptive military agent by the USA but was rejected because of the unpredictability of its effects and cost of synthesis.⁷⁷ 

The US military eventually selected the centrally and peripherally acting cholinergic antagonist BZ (Scheme 1.9), and manufactured approximately 50 tonnes for weaponisation in the 1960s. Several other nations are suspected to have weaponised small quantities of BZ or an analogue. BZ had a number of disadvantages, particularly its slow onset and long duration of action, incapacitation in terms of the ability to conduct military operations was difficult to judge, and it was expensive to manufacture. It was eventually declared obsolete by the USA in 1976 and destroyed in the late 1980s.⁷⁷ 

During the 1960s, potent opioids were the major focus of incapacitant research in several countries. An example is etorphine (M-99), a semi-synthetic hexacyclic opioid that is approximately 1000 times more potent than morphine and marketed for veterinary use as a knockdown agent for large game animals. Some prodeine analogues were patented as military incapacitants.⁸⁰  In the 1970s and 1980s attention moved to a structurally simpler and totally synthetic class of opioids known as the fentanyls (Scheme 1.10).⁸¹  The parent drug fentanyl, initially a product of the Belgian drug company Janssen Pharmaceutica, is one of the most widely used intravenous analgesic/anaesthetics, with potency in humans 50–100 times that of morphine. Some of its analogues are up to 10 000 times more potent than morphine and rank amongst the most potent drugs known. One such drug, carfentanil, is also used to knock down large game animals for veterinary or conservation purposes. In 2002, Russian Special Forces used an aerosolised mixture of two fentanyls, carfentanil and the clinically used short acting, fast onset analogue remifentanil (Scheme 1.10), to end the siege of a Moscow theatre by Chechen terrorists.⁴³  Approximately 130 of the 800 hostages died in this operation, a major factor being the inherent respiratory depressant activity of morphine-like compounds. Other depressants that have attracted serious attention as possible incapacitating agents include α-adrenergic agonists such as medetomidine and benzodiazepines.⁸² 

Incapacitants have a somewhat ambiguous status under the CWC. Development and use is permitted for ‘law enforcement purposes’. However, neither of the terms incapacitant and law enforcement are defined by the Convention, and this could become a grey area, for example in the context of peace keeping operations. Much concern has been expressed that this ambiguity would allow development of agents that clearly have dual use potential, military as well as law enforcement.⁸³ 

Predicting the future of CW has proved notoriously difficult. Although many hundreds of compounds have been assessed for CW potential, in defensive as well as offensive programmes, the threat with regard to agents has remained largely unchanged over the past 40 years. Sulfur mustard and G- and V-type nerve agents have remained the most important threats. These agents are relatively easy to manufacture, and they have close to the ideal physicochemical properties required of an agent. The most notable developments over this period have been the proliferation of such agents to other nations, particularly in the Middle East, use by terrorists in Japan, and the emergence and use of fentanyl analogues as incapacitants.

Advances in the life sciences are accelerating at an unprecedented rate, and concern has been widely expressed that some developments might pose a challenge to the CWC if they were seriously applied to the development of new agents.⁸⁴  Examples are accelerated drug discovery (e.g. parallel synthesis with high throughput screening), advances in neuroscience (leading to new incapacitants, e.g. derived from bioregulators),⁸⁵  and the growing convergence of chemistry and biology (e.g. synthetic biology, bio-production).⁸⁶  Some of these concerns are arguably being overstated. The Scientific Advisory Board of the OPCW, in its last five year report to the Director General on advances in science and technology,⁸⁷  noted these developments and others as warranting surveillance and periodic review, but did not see any near term threat to the Convention. It is pertinent to note that the most modern CW agents known to have been weaponised, the V agents and BZ, are products of 1950s research, and the fentanyls used as incapacitants in Moscow are products of classical drug research undertaken in the 1960s to 1980s.

Although the CWC specifies certain categories of CW agent and their precursors for declaration and verification purposes (in the Annex on Chemicals), any emerging agents would still be captured under what is unofficially referred to as the ‘general purpose criterion’. Article II states ‘For the purposes of the Convention, Chemical Weapons means... toxic chemicals and their precursors, except where intended for purposes not prohibited under this convention, as long as the type and quantities are consistent with such purposes...’.¹  It is therefore to be hoped that the near universality of the CWC, together with ever more sophisticated means of surveillance and verification, will dissuade potential proliferators from risking the opprobrium and political or economic repercussions of contravening the Convention.