Disclaimer: This infomation is provided "as is" and is only provided as ADVICE.
The information in the node is compiled from military and civilian sources, they have NOT been tested by myself and I can provide no guarentee that they will work.
So if you follow this and hurt yourself, you were warned..

Unfortunately my research turned up very few "in the field" ways of decontaminating water that ordinary people can use in the event of a (terrorist) chemical attack. However, I decided I'd like to node what I have learnt about the subject while researching it.

A lot of this node is merely a compilation of other sources, some of it is paraphrasing my sources and some is applying information from one to another (as you will see).

Chemical Agents, Definition:

Nerve Agents
Vesicants (Blister Agents)
Cyanogen Agents (Blood Agents)
Choking Agents

Firstly it has to be determined if the water is contaminated, to confirm this scientifically a testing kit would be required, unless the contamination is observable, (e.g. chlorine clouds). These kits appear to be tricky to find, but it would be wise to assume everything is contaminated if chemical weapons have been used in the area.
On the web I have only discovered kits by oritest-group and the American Military's "Chemical Agent Water Testing Kit M272"

Hopefully the following is obvious to most people, but it may not be: (1)

Simple boiling is not a reliable method of decontamination. The following methods are available...
The best method for removing all of the above chemical agents involves a complete water filtration system, i.e ROWPU (Reverse Osmosis Water Purification Unit). Home versions of these can be purchased from several manufacturers found on the internet. To get an idea of how it works, check out the Reverse Osmosis node.

The level of protection these offer from chemical agents is shown here (2).
The ROWPU also removes large amounts of chemical agents.
RO removal characteristics for various chemical warfare agents are as follows:
  • GB - 99.1%, leaving the chemical cylinder to remove .7%
  • VX - 99.9%, leaving the chemical cylinder to remove .1%
  • BZ - 99.9%, leaving the chemical cylinder to remove .1%
  • GD - 99.7%, leaving the chemical cylinder to remove .3%
The agents quoted are all nerve agents, but other sources indicate that a ROWPU would bring any chemical agent contaminated water to a drinkable level.

The above source also assumes that a chemical cylinder (probably an active charcoal filter) is attached. IMHO this is probably not necessary as the remaining dose will be small enough ready (this can be tested for of course), but there is no such thing as "too much precaution" with chemical agents..

Now that I have introduced figures to the node, I may as well indicate what levels can be ingested safely: (1)
Table 12-III: Max allowable concentration of Agents in Drinking Water

-------------------------------------------------
| Agent		| Max concentration (mg/litre)* |
-------------------------------------------------
| Mustard	| 0.2				|
| Nerve Agents	| 0.02*				|
| Arsenic	| 0.3				|
| Cyanogens	| 6.0				|
-------------------------------------------------
| * 5 litres/day, no more than 7 days		|
| ** many testing kits are not sufficiently	|
|    sensitive to detect nerve agent at this	|
|    concentration				|
-------------------------------------------------
The problem here though is that fairly advanced equipment will be needed here to determing the mg/l of these chemicals, any method using just the testing kit would require knowledge of the kit's sensitivity and repeated diluting-testing.
By which I mean, test a sample, dilute it, re-test it, until the reading becomes negative, thus getting an estimate of the agent's concentration in the water. However, to dilute it pure water would be required, which you would drink with...

From this I made a quick calculation at the maximum dose in ppb, (testing kits tend to measure this way).
Is this correct ?
Firstly, 1 litre of water = 1kg..

Maximum safe dose (mass ratio):
 0.02mg : 1kg
    1 g : 50x(106) g

Molar masses:
 Nerve agent = x
 H2O = 18

Max safe dose (ppb):

     1  * (109)
 __________________
 (18/x) * 50x(106)
      
So for Sarin (mass 141) this works out to 156ppb.

Although, Sarin is soluble in water, so I am not sure whether it is still dangerous or not

Small, portable hand-pump type Reverse Osmosis equipment does exist, although I have not been able to find any information about its effectiveness at removing chemical agents (it does not boil/distill).

The next safest method would involve active charcoal filtration and/or flocculation equipment which (again) would be large and non-portable. Also there doesnt appear to be much hardware like this available for civilians, and few details on its effectiveness are available.

The next method works for all chemical agents that are filtered by a respirator, which, (for a good repsirator) should be most of them. (1)
In a small scale emergency, water may be decontaminated by running it through a spare unused respirator canister, provided that the flow rate is such that the water emerges drop by drop, any water coming through at first faster than this should be discarded. no more than 5 litres should be filtered with one canister. -- The canister cannot be used on a respirator after being being used for this purpose.
Clearly the respirator canister should be spare, as the respirator may be needed later. Interestingly enough, the canister may be able to filter biological agents also (check the specification of the respirator for this).

The final method is one I have decided may work, but I am not a good enough chemist to how to proceed...

In this situation the water source in question has been exposed to a nerve agent, (e.g. surface of a lake after a nerve agent attack). Nerve agents have the following (useful, for us) properties: (3)
Three types of chemical mechanisms have been used for decontamination... oxidation; and acid/base hydrolysis. Mustard (HD) and the persistent nerve agent VX contain sulfur molecules that are readily subject to oxidation reactions. VX and other nerve agents (Tabun (GA), Sarin (GB), Soman (GD), and GF) contain phosphorus groups that can be hydrolyzed. Therefore, most chemical decontaminants are designed to oxidize mustard and VX and to hydrolyze nerve agents (VX and the G Series).
Now, here we need to find a chemical that oxidizes or hydrolyzes and can be easily seperated from the water after the reaction.

However, a chemical like this appears to be quite tricky to find, (for me, I am not a chemist). Also, once the chemical is used any resultant compounds need to be extracted from the water (will distillation work ?).
Any chemists who can step in and help me with this ?

Hopefully this writeup has no huge errors, if it has I'll soon know, and please feel free to /msg me any corrections/suggestions/flames. Suggestions/comments from chemists are very welcome.

Sources:

 (1) NATO HANDBOOK ON THE MEDICAL ASPECTS OF NBC DEFENSIVE OPERATIONS AMedP-6(B) 
     PART III - CHEMICAL - CHAPTER 12 - DISPOSITION OF CONTAMINATED FOOD AND WATER 
	http://www.fas.org/nuke/guide/usa/doctrine/dod/fm8-9/3ch12.htm

 (2) Integrated Publishing
     Utilitiesman - Chapter 9 - Nuclear Agent Removal
	http://www.tpub.com/utilities/63.htm

 (3) Medical Aspects of Chemical and Biological Warfare
     DECONTAMINATION - Page 354
	http://usamricd.apgea.army.mil/textbook/Ch-15.pdf

Log in or register to write something here or to contact authors.