Perspective: Preventing Catastrophic Chemical Attacks
Preventing Catastrophic Chemical Attacks
A terrorist attack on a single 90-ton chlorine tank car could generate a cloud of toxic gas that travels 20 miles. If the attack took place in a city, it could kill 100,000 people within hours. Now multiply that nightmare by another 100,000. That’s the approximate number of tank cars filled with toxic gases shipped every year in the United States.
We are vulnerable to catastrophic acts of chemical terrorism such as this plausible scenario. There are 360 sites sprinkled across the United States where, if terrorists attacked, more than 50,000 people at each could be harmed or killed. Many of them are in heavily populated areas of New Jersey, New York, Pennsylvania, Texas, Louisiana, and California. Yet the federal government has made no progress whatever in addressing this threat in the five years since the 9/11 attacks.
Private industry has taken a few steps to make us more secure from chemical terrorism. The American Chemistry Council (ACC), which represents many of the nation’s chemical manufacturing plants, has required its member facilities to undertake a set of security initiatives. They have invested several billion dollars in security upgrades since 2001. Although these private-sector investments may help with both safety (prevention of accidents) and security (prevention of attacks), two key shortcomings keep the ACC’s efforts from providing the country with the level of security commensurate with the threat. First, the facilities that have invested in improved security number only 1,100, a small fraction of the 15,000 facilities that store or produce large amounts of hazardous chemicals. Even worse, they are not, by and large, the facilities that would injure or kill the largest numbers of people if they were attacked. Second, the stark reality is that a resourceful terrorist group can compromise any chemical facility or shipment that it puts its mind to.
The government finally appears to have become aware of the frightening state of our chemical plant security (although, as explained later, not of chemical transportation security). President Bush supports bills in the House and Senate aimed at hardening security at chemical plants based on the risks they pose. The legislation calls for the Department of Homeland Security (DHS) to place chemical plants into different tiers according to the results of offsite consequence analyses and set tier-dependent security standards. In turn, chemical plants would be required to assess their own vulnerabilities and choose how to improve security, if need be, to satisfy these standards. Facilities in the highest-risk tier would begin first.
Unfortunately, the bills do not go far enough. We need an approach that is both risk-based (so that investments in protection are made at the most dangerous facilities) and focused on the use of alternative, much safer chemicals (so that true prevention can be achieved.) The bills focus only on the degree of risk posed by a plant, not on safer processes.
The three main culprits
Chemical facilities vary widely in their attractiveness to terrorists. Obviously, measures should be aimed at plants and rail shipments that could generate mass casualties if attacked. As a result of the Clean Air Act, chemical facilities have already analyzed their potential worst-case offsite consequences from a chemical accident. That has made it possible to identify the most dangerous needles in the haystack of chemical facilities and shipments. We know that the most dangerous chemicals are not the flammable ones, such as liquid propane gas. The larger danger comes from chemicals that, on release, form heavier-than-air clouds that can travel 10 to 20 miles. We also know that there are three main culprits: chlorine (used in the production of building materials); anhydrous ammonia (used for agricultural fertilizer); and, worst of all, hydrofluoric acid (used in transportation fuels.) There may be several plants using less common chemicals that are equally dangerous. But their off-site consequence analysis data are exempt from the Freedom of Information Act and are not on the Internet. That makes it difficult for the general public to identify these facilities, but it makes it difficult for the terrorists, too.
How can we prevent or mitigate an attack on sites with these particular chemicals? One approach is to harden security at the facility. Although installing fences and security guards and performing background checks on employees are marginally useful, they will not prevent a suicide attack by a group of terrorists using several large trucks or a small airplane. Another approach is to place water curtains around large storage tanks. These are essentially elaborate sprinkler systems that attempt to force the gas to the ground before it can form a toxic cloud. These systems can be effective against the leaks or slow releases that are typical of many industrial accidents, but they will not help forestall a massive release resulting from a truck or airplane crashing into a storage tank, which is the most likely scenario for a terrorist attack. Moreover, these water curtains, which are activated by gas detection, probably would not work after such an attack.
With neither security hardening nor water curtains offering a robust response to a terrorist attack, we must resort to measures that prevent a toxic cloud from being released in a heavily populated area. Products and processes need to be redesigned so that they cannot be the catalyst for a deadly plume.
Particularly problematic are the nation’s 148 oil refineries. Of these, 50 use hydrofluoric acid in their alkylation process, which provides high octane while maintaining low sulfur and nitrogen content. Although only 4% of the nation’s hydrofluoric acid is used by these 50, they top the list of most dangerous chemical facilities because the scale of their operations is immense. Some refineries store hundreds of thousands of pounds of hydrofluoric acid, which could seriously harm or kill hundreds of thousands of people. Collectively, these 50 refineries have more than 10 million pounds of hydrofluoric acid on their premises.
Fortunately, there is an obvious fix. Quite simply, we need to discontinue the use of unmodified hydrofluoric acid in the alkylation process. We are on the way to doing that. The other 98 refineries use two safer alternatives. First, the alkylation process can be converted from using hydrofluoric acid to using sulfuric acid, which does not form a dense cloud on release. Indeed, 86% of the new alkylation units introduced in the 1990s used sulfuric acid, which also leads to a reduction in the fractionation capacity required. The conversion cost is $20 million to $30 million per refinery, but a good-sized oil refinery refines the equivalent of approximately one billion gallons of gasoline annually. A second approach is to modify the hydrofluoric acid with an agent that causes about three-quarters of the acid in the cloud to fall to the ground.
Dealing with chlorine and anhydrous ammonia is much more difficult. These two chemicals are used in approximately half of the 15,000 facilities with large amounts of dangerous chemicals. Although some water and sewage treatment plants have successfully substituted hypochlorite bleach or ultraviolet light for chlorine, these treatment plants represent only 6% of the nation’s chlorine use. Similarly, some paper and pulp manufacturers have switched to less dangerous alternatives such as chlorine bleach, but the paper industry uses only 5% of total industrial chlorine.
The place to focus efforts is polyvinyl chlorine (PVC) plastics manufacturing, which consumes approximately 40% of the nation’s chlorine; another 40% is used in the production of a wide variety of organic and inorganic chemicals. 75% of PVC is used in buildings, primarily in piping, siding, and roofing membranes. Because of environmental hazards associated with PVC (dioxin is generated during production and PVC releases deadly gases during a fire), there are several alternatives. For example, piping can be made of cast iron, concrete-vitrified clay, and high-density polyethylene. Siding alternatives include fiber-cement board, stucco, brick, or polypropylene.
Similarly, although the anhydrous ammonia in some pollution-control processes has been replaced by urea or aqueous ammonia, 72% of ammonia use is in agricultural fertilizers. Several thousand facilities make fertilizers. Unlike oil refineries, these facilities tend to be small and near farmland rather than near large population centers. Ammonia-free alternatives that use urea or liquid nitrogen are in the fertilizer marketplace.
Chlorine and anhydrous ammonia are not stored in the vast quantities that hydrofluoric acid is at oil refineries, and the biggest threat from these chemicals may be during their transport in tank cars. To prevent these well-marked tank cars from injuring or killing tens of thousands of people, hazardous shipments must be routed away from densely populated areas.
But when Washington, DC, tried to do just that in January 2005 by passing a bill preventing hazardous shipments in its downtown, it was sued by the rail company CSX Transportation, which got help, incredibly enough, from DHS. Although a federal judge ruled in favor of the city in April 2005, a three-judge appellate panel reversed this decision a month later.
Why does CSX oppose the city’s desire to make its citizens safer? Rerouting in this oligopolistic industry, where many customers have one option, would sometimes require handoffs of shipments between companies. If that practice became widespread, it would eat into rail industry profits. Sen. Joseph Biden (D-DE) has introduced legislation that would force the rerouting of hazardous rail shipments away from likely terrorist targets. It lies dormant.
Congress must move beyond the approach embedded in the pending legislation, because it would provide only the illusion of security. To achieve true security, legislation should require the 50 oil refineries to convert to either sulfuric acid or modified hydrofluoric acid. It should also eliminate hazardous shipments of dense cloud–forming toxic gases in urban areas. Rather than shift the entire burden for protecting these gases to the transportation sector, we need to dramatically reduce the demand for dense cloud–forming toxic gases such as chlorine and anhydrous ammonia by requiring the use of safer technologies in cases where such alternatives already have a proven track record, such as ammonia-free fertilizers and PVC-free building products.
This nonvoluntary approach will level the playing field, allowing the chemical industry to pass these costs on to consumers. The Chemical Security and Safety Act bill introduced in March 2006 by Sen. Frank Lautenberg (D-NJ), which would require certain plants to use safer technologies if practical alternatives exist, coupled with Biden’s legislation banning hazardous rail shipments in densely populated areas, would go a long way toward reliably reducing the possibility of a catastrophic chemical attack.
Unlike catastrophic biological or nuclear terror attacks, catastrophic chemical attacks can be avoided by removing the targets. The White House and Congress must choose which should prevail: special interests or the safety of our citizens.
Lawrence M. Wein ([email protected]) is the Paul E. Holden Professor of Management Science at Stanford University’s Graduate School of Business.