Sodium Hypochlorite Shortage Affects Water Treatment

I ran into an interesting article over at BoombergLaw.com (thanks to Fred Gossen on LinkedIn for pointing at the article) about the problems that a number of water treatment and wastewater treatment plants are having with getting ahold of adequate supplies of commercial grade sodium hypochlorite for their water disinfection processes. Many have, according to the article, petitioned the EPA for legal assistance in acquiring the needed chemicals.

The article pins at least part of the blame for this chemical shortage on a fire at a chemical plant last year that shutdown a major supplier of hypochlorite and a pandemic driven increase in the use of family swimming pools. Both of these have undoubtedly had an affect the availability of sodium hypochlorite, but that is not the root cause of this problem. That can be traced back to the ‘inherently safer technology’ push away from using chlorine gas as a water treatment disinfectant back in the middle 2000’s (see one of my 2008 blog posts on the topic here). A fear of a large chlorine gas release in a major city because of an industrial accident or terrorist attack caused a number of environmental activists to successfully push many treatment facilities to switch to ‘safer alternatives’ and the easiest was from chlorine gas to commercial grade liquid bleach also known as sodium hypochlorite.

The big problem with bleach is that it degrades in quality (read – decreasing concentration while releasing chlorine gas) fairly quickly, limiting how far it can practically be shipped by rail. That, combined with the NIMBY opposition to the construction of new bleach production facilities, set up a tight supply situation for commercial grade bleach that left the market susceptible to production upsets. There are a number of alternative disinfection processes available, but they are almost all more technologically challenging with lengthy lead times and high switchover costs.

One potential long-term solution for larger facilities is the on-site generation of chlorine gas. The technology is well understood and requires water, salt (common NaCl), and electrical power as the major process feeds. It is still inherently safer than railcar loads of chlorine gas because it is produced as needed with relatively small amounts of process storage required. The biggest problem is getting rid of the byproduct sodium hydroxide (caustic soda). There are, however, lots of places that use that chemical as an important feedstock or process aid.

Common Chemical Accident Causes Building Evacuation

Yesterday a 49-story office building in downtown Chicago was evacuated when a common chemical accident occurred on the roof of the building resulting in the release of chlorine gas. Six people were injured severely enough to be transported to local hospitals.

The Incident

Very little information is available in the news reports on the incident (here, here, here and here). The common thread is that “chlorine and acid were accidentally mixed on the roof of the building”. Based upon that this is likely what happened.

A maintenance crew was cleaning/disinfecting the water side of the cooling tower for the building HVAC system. These systems have been implicated in a number of Legionnaire outbreaks, so the cleaning/disinfection of these roof top systems is a fairly normal maintenance task. The ‘acid’ was likely muriatic acid (dilute hydrochloric acid); it is commonly used for pH adjustment, and cleaning metal or concrete. The ‘chlorine’ was almost certainly a solution of sodium hypochlorite (bleach); it is commonly used as a disinfectant and cleaning solution.

In disinfecting a small body of water the muriatic acid is added to lower the pH of the water. Then the chlorine is added to kill off bacteria. With adequate mixing or an appreciable time between adding the two chemicals to the water there is no problem. If the two chemicals are added in close physical or temporal proximity the they remain concentrated enough to allow a very quick exothermic action to occur. A byproduct of that reaction is the release of chlorine gas.

Unless someone is really stupid in the amount of bleach added to the water, there will not be enough chlorine gas released to kill anyone unless they are in a small, confined space above the surface of the water. Relatively small, non-fatal, amounts of chlorine gas will cause severe irritation to the eyes, nose and respiratory tract. Prompt medical evaluation is routinely recommended for anyone experiencing eye pain or difficulty breathing after relatively minor chlorine gas exposures.

Commentary

In hind sight, there was almost certainly no need to evacuate the building. The amount of chlorine gas released would not have been medically significant beyond the immediate are of the release on the roof. I suspect that enough gas got into one of the HVAC air intakes to allow some people to detect the odor of chlorine (detectable by the average person at very low levels). Complaints of a strange chemical odor reported in the building coupled with the report of a chemical incident on the roof would be sufficient, however, to make any emergency response incident commander order a precautionary evacuation.

One of the reasons for this is that the same reaction between hypochlorite and muriatic acid make for a pretty interesting improvised chemical munition in closed quarters like a building. Without the diluting effect of a small body of water, the fast and strong exotherm results in low order explosion (no flame but and expanding gas cloud) that releases chlorine gas. Both chemicals are easy to buy and the only difficulty in constructing these bombs is how to keep the two chemicals apart until you want the reaction to take place. Again, unless the ‘bomb’ is really large, there is little real danger outside of the immediate area of ‘detonation’, but the loud bang and chlorine odor will do a nice job of starting a panic in a crowded building.

I expect that the investigation of this incident will ultimately place the blame for this incident on ‘human error’ and inadequate training of the maintenance personnel. People really can handle these two relatively innocuous industrial chemicals safely with just a modicum of training and supervision. But, the reason that this is such a common accident is that the process looks so simple and the chemicals look very common, so no one really takes the safety issues seriously until it is too late.

Industrial Chemical Incompatibility – Bleach

Hardly a week goes by when we don’t see a news story about a hazmat incident caused by the mixing of incompatible chemicals. Yesterday there was a story from a water treatment plant in Weslaco, TX. In this case there was no one hurt nor any long term damage to the facility. I haven’t written about one of these incidents lately, so this will be a good chance to review the problems associated with bulk chemical storage.

The Incident

As is usual the news reports (here, here and here) provide incomplete and conflicting data about what happened. Here is what I think we know.

Apparently just before 2 pm yesterday a tank truck showed up at the Weslaco water treatment facility for a routine chemical delivery. It is not clear (from news reports) what chemical was in that truck; one report claims is was ‘sodium chloride’. A sodium chloride solution is not a chemical normally used by water treatment facilities. Maybe they meant ferric chloride.

The truck contents unloaded into a sodium hypochlorite (industrial strength bleach) storage tank. It was never mentioned who unloaded the truck, but this is very often the truck driver. In this case there is at least one report that the truck driver did not know what he was carrying (I hope this was not true).

It is also not clear if a chemical reaction actually took place; one news report claims that one did occur, producing ‘hydrogen chloride’. Again, this is not typically a reaction product of sodium hypochlorite, we generally see chlorine gas as a reaction product.

Local residents (within 820 or 1,000 feet depending on the news story) downwind of the plant were told to shelter in place. Apparently a nearby (within 250 ft according to one report) residence and daycare center were evacuated. At about 5:00 pm (CDT) the all clear was apparently given.

Public Communications

It is easy to blame the news media for the conflicting information be provided to the public about this incident. It is apparent, however, that the local emergency response community was the source of most of the information provided and that community did not speak with one voice. If this had been a more serious incident this could have caused significant problems as the community reacted to the conflicting messages.

Sodium Hypochlorite (NaClO)

I have talked about this chemical on a number of different occasions. In water treatment and waste water treatment it is used as a source of chlorine for water disinfection. It is not quite as effective as chlorine gas, but it is not nearly as dangerous if released into the environment.

It is however a very reactive chemical. It reacts explosively with ammonia and it reacts violently with acids and reducers. It also reacts with heavy metals. In every case chlorine gas is given off as a byproduct of these reactions and the reactions are generally exothermic; frequently producing enough heat to produce steam. This will increase internal tank pressures significantly.

A number of chemicals routinely used in water and waste water treating facilities react violently with hypochlorite, even in very dilute solutions. They include: ammonia, ferric chloride, muriatic acid and sulfuric acid.

Sodium hypochlorite degrades over time and it degrades faster at higher temperatures. The byproducts produced during that degradation process are chlorine gas, water and sodium chloride (salt). For this reason there are typically frequent deliveries of sodium hypochlorite, typically by tank wagon or rail car for most treatment facilities.

Storage Tank Management

When dangerous chemicals (and due to reactivity at least sodium hypochlorite is a dangerous chemical) are stored in bulk storage tanks special care must be taken to ensure that the design and management of those tanks take into account both the stand alone hazards of the chemicals, and the reactions of those chemicals with those that could ‘reasonably’ be accidentally introduced into the storage tank. The EPA has a simple chemical compatibility chart for water treatment plants; it is, however, short on information about sodium hypochlorite.

Sodium hypochlorite it typically stored in a vented storage tank that allows vapors to escape the tank during filling operations and when decomposition vapors are produced. A common industry handbook for bleach recommends sizing the venting device to take into account the decomposition outgassing rather than the fill rate outgassing, meaning that the outgassing rate for decomposition may be significantly higher than for filling operations. Typically, decomposition rates are determined by maximum ambient temperature, not the much higher rates of decomposition associated with filling the tank with incompatible materials. That means that there may be a significant hazard of catastrophic failure of the storage tank in an incompatible filling situation.

Proactive strategies to prevent misfiling the tank are thus very important. The first step is to identify all of the undesirable chemical reactions that can take place at a facility based on the bulk chemicals that are received. The facility management and employees must be fully aware of the potential hazards associated with the improper mixing of chemicals in storage tanks.

While most medium to large chemical manufacturing facilities have a dedicated bulk unloader to handle transferring liquids from tank trucks and rail cars to storage tanks, most water treatment facilities and smaller chemical companies will not have that luxury. This means that it is very common for truck drivers to unload their own trucks. Unloading procedures must take that into account.

The facility needs to make it as difficult as possible to unload the wrong chemical into the wrong tank. The easiest technique to use is to have all storage tanks in widely different areas. This is frequently not possible due to physical layout constraints. Additionally, the requirement to have storage tanks in diked areas makes it impractical to have completely separate storage tank areas.

Another technique is the use of access control measures. The most effective that I have seen in use is a double lock system on each unloading line. There are two locks on each line and every lock is uniquely keyed. Two unload a tank wagon or rail car the unloader first gets a key from the operations supervisor; this is done to confirm that a load is expected and to stage the delivery truck at the correct location. The second key is obtained from the quality control lab; there the quality of the material being delivered is confirmed (either by testing or COA checking) before the second key is provided. In 20 years at one medium sized chemical manufacturing facility where I know this was employed there was never a successful unload of material into the wrong tank.

For very dangerous or highly reactive chemicals additional measures should be taken. There should be signs in the unloading area clearly stating what chemicals should never be unloaded. The sign should also provide immediate emergency response information for actions to be taken if an unloading accident should occur. Finally, and most importantly, specific advance coordination needs to be made with emergency response personnel, on and off site, so that they are well familiar with the actions they need to take if an unloading accident of this type occurs.

Emergency Response

Once an unloading accident occurs there is not much remedial action that can be taken. Particularly with sodium hypochlorite reactions, the reactions are practically instantaneous. By the time emergency response personnel arrive on scene there will be a chlorine gas cloud present if one is going to be produced so emergency responders need to approach the site from upwind.

For most metal tanks the danger for catastrophic failure of the tank will have passed by the time emergency responders arrive on scene. For plastic and fiberglass tanks the hot fluids in the tank may continue to weaken the tank shell over time so those tanks should be cooled if they are not already compromised.

Chlorine gas procedures should be implemented immediately and downwind testing for chlorine gas should be started. Particular attention should be paid to low lying areas where chlorine tends to collect since it is heavier than air. Evacuation and shelter in place criteria should be set in place prior to the incident based upon the worst case chlorine release from the tank. Local hospitals should be notified to prepare for chlorine gas casualties. All personnel with even the slightest exposure should be identified for follow-up medical checks due to the long term hazards of chlorine exposure.

As always in incidents involving industrial chemicals, advance planning is the key to an effective emergency response.

2009 “Chemical Weapon Attack” in Tucson

Yesterday the FBI arrested a ‘Tucson businessman’ for a 2009 chemical weapon attack on a couple in Tucson, AZ.

The Attack

According to the FBI press release “Fries placed chemical devices in the front and back yard of a couple living on the northwest side of Tucson. When ignited, the devices produced a football field-sized cloud of chlorine gas that hovered over the neighborhood and resulted in the evacuation of numerous families in the area.”

According to a news report from a couple of days after the attack, “Investigators said Sunday's incident in Tucson included derogatory graffiti written in Spanish [NOTE: it was actually German]; dead animals; an incendiary device; and chlorine tablets covered in an unknown liquid that created a large toxic-gas cloud.”

Interestingly a copy of the police report from the initial officer on the scene mentions nothing about a ‘toxic cloud’ or ‘chemical weapons’ or even the odor of chlorine, a very distinctive and irritating odor even at low, sub-lethal concentrations.

The Device

It looks like the perpetrator used a very simple ‘chemical weapon’; a pile of ‘chlorine tablet’ (actually sodium hypochlorite) available from Wal-Mart or any pool supply store. The chemical reaction was not started by an ‘ignition source’ (someone PLEASE help the FBI with their technical descriptions) but by simply pouring an ‘unidentified’ liquid {which could have been anything from a household ammonia-cleaner (probably not; that reacts too fast and you’re in the cloud before you’re done pouring) to Coca-Cola to simple tap water with a little bit of vinegar}on the tablets.

It would have had to have been a very large pile of commercial hypochlorite tablets to produce a ‘toxic’ cloud the size of a football field. Since there are no reports of injuries or hospitalizations I would assume that at most it was an irritating cloud not a toxic cloud. It wasn’t even that irritating when the initial officer on the scene did not even notice it nor did any of the witness statements found in press accounts mention it.

WMD Overreaction

In a confined space where the concentration could reach lethal levels or even levels that could just cause serious medical consequences (damaged lungs, eyes, etc) I suppose that one might consider this ‘hypochlorite’ attack to be a weapon of mass destruction or a chemical weapon. Jason Siggers of the late Armchair Generalist blog would be proud that the FBI applied a chemical weapons charge against an inept white man; he had reported on numerous occasions that WMD type charges only appeared to be filed by the FBI against non-whites or Muslims.

Lets face it; this was not a ‘chemical weapon attack’ any more than the dead animals or feces smeared across the scene were a bio-weapon attack (though they were more likely to do harm than the ‘chlorine gas’ at these low concentration levels). Charge him with a hate crime? Absolutely. Charge him with criminal mischief? Yes, multiple charges. Charge him with criminal stupidity? Surely there must be such a charge on the books somewhere. But PLEASE, charging him with making and using a chemical weapon (two separate offenses)? That is nothing but political grandstanding on the part of the FBI and the Federal Prosecutor.

Potential Threat

The basic chemical reaction used in this case could certainly be used to manufacture an effective chemical weapon. It would take much more sophistication than a pile of pool chlorine tablets on the front lawn to be effective however. There are, in fact, a large number of relatively simple chemical reactions that can produce toxic gasses, reactions that use similarly unregulated, commercially available chemicals.