Background and Causes of the Disaster
The Union Carbide plant was established in 1969 and had expanded to produce carbaryl in 1979; MIC is an intermediate in carbaryl manufacture.
The chemical accident was caused by the introduction of water into MIC holding tank E610, due to slip-blind water isolation plates being excluded from an adjacent tank's maintenance procedure. The resulting reaction generated a major increase in the temperature of liquid inside the tank (to over 200°C). The MIC then gave off a large volume of toxic gas, forcing the emergency release of pressure.
A number of background causes contributed to the explosion and the disaster’s intensity.
Cost-Cutting Measures
As a long-term cause of the catastrophe, authorities had tried and failed to persuade Carbide to build the plant away from densely populated areas. Carbide explained their refusal on the expense such a move would incur[7].
Union Carbide previously produced their pesticide, Sevin (the commercial name of Carbaryl), without MIC but, after 1979, began using MIC because it was cheaper. Other manufacturers, such as Bayer, made Sevin without MIC, although this caused greater expenses[7].
In the early 1980s, the demand for pesticides had fallen: the factory was making a loss and overproducing MIC that was not being sold, leading to a series of cost-cutting measures from around 1982 onwards. These measures affected the two interrelated areas of workers and their conditions, and the equipment and safety regulations installed at the plant[7].
Workers and Their Conditions
Attempts to reduce expenses affected the factory’s employees and their conditions:
* Kurzman argues that “cuts… meant less stringent quality control and thus looser safety rules. A pipe leaked? Don’t replace it, employees said they were told… MIC workers needed more training? They could do with less. Promotions were halted, seriously affecting employee morale and driving some of the most skilled… elsewhere”[8].
* Workers were forced to use English manuals, despite the fact that only a few had a grasp of the language[9].
* By 1984, only six of the original twelve operators were still working with MIC and the number of supervisory personnel was also cut in half. No maintenance supervisor was placed on the night shift and instrument readings were taken every two hours, rather than the previous and required one-hour readings[8].
* Workers made complaints about the cuts through their union but were ignored. One employee was sacked after going on a 15-day hunger strike. 70% of the plant’s employees were fined before the disaster for refusing to deviate from the proper safety regulations under pressure from management[8].
* In the words of the International Campaign for Justice in Bhopal, “poorly trained personnel, rapid turnover, leaking valves, shoddy gauges and inadequate water spray protection were all identified as representing “a higher potential for a serious incident or more serious consequences if an incident should occur”"[10].
* In addition, some observers, such as those writing in the Trade Environmental Database (TED) Case Studies as part of the Mandala Project from American University, have pointed to “serious communication problems and management gaps between Union Carbide and its Indian operation”, characterised by “the parent companies [sic] hands-off approach to its overseas operation” and “cross-cultural barriers” [11].
Equipment and Safety Regulations
Cost-cutting initiatives affected the quality of equipment and the effectiveness of safety regulations:
* It emerged in 1999, during civil action suits in India, that, unlike Union Carbide plants in the USA, its Indian subsidiary plants were not prepared for problems. No action plans had been established to cope with incidents of this magnitude. This included not informing local authorities of the quantities or dangers of chemicals used and manufactured at Bhopal[7].
* The MIC tank’s alarms had not worked for 4 years[12].
* There was only one manual back-up system, not the four-stage system used in the USA[12].
* The flare tower and the vent gas scrubber had been out of service for 5 months before the disaster. The gas scrubber therefore did not attempt to clean escaping gases with sodium hydroxide (caustic soda), which may have brought the concentration down to a safe level[12]. Even if the scrubber had been working, according to Weir, investigations in the aftermath of the disaster discovered that the maximum pressure it could handle was only one-quarter of that which was present in the accident. Furthermore, the flare tower itself was improperly designed and could only hold one-quarter of the volume of gas that was leaked in 1984[4].
* To reduce energy costs, the refrigeration system, designed to inhibit the volatilization of MIC, had been left idle – the MIC was kept at 20 degrees Celsius, not the 4.5 degrees advised by the manual, and some of the coolant was being used elsewhere[12].
* The steam boiler, intended to clean the pipes, was out of action for unknown reasons[12].
* Slip-blind plates that would have prevented water from pipes being cleaned from leaking into the MIC tanks via faulty valves were not installed. Their installation had been omitted from the cleaning checklist.
* Water sprays designed to “knock down” gas leaks was poorly designed – set to 13 metres and below, they could not spray high enough to reduce the concentration of escaping gas[12].
* The MIC tank had been malfunctioning for roughly a week. Other tanks had been used for that week, rather than repairing the broken one, which was left to “stew”. The build-up in temperature and pressure is believed to have affected the explosion and its intensity[12].
* Carbon-steel valves were used at the factory, despite the fact that they corrode when exposed to acid[7]. On the night of the disaster, a leaking carbon-steel valve was found, allowing water to enter the MIC tanks. The pipe was not repaired because it was believed it would take too much time and be too expensive[12].
* According to Lepowski, “virtually every relevant safety instrument” was “either in short supply, malfunctioning or designed improperly”, and “internal documents show that the company knew this prior to the disaster, but did nothing about it”[12].