This issue sponsored by

http://www.aiche.org/CCPS/Publications/Beacon/index.aspx
www.aiche.org/ccps www.iomosaic.com
Messages for Manufacturing Personnel
Conduct of Operations June 2015

In January 1997 there was an explosion and fire in a In April 1998 an explosion followed the release of
hydrocracking unit in a refinery in California. A pipe in flammable chemicals from a 2000 US gallon batch reactor
the unit ruptured releasing a flammable mixture of into a building in a specialty chemical plant in New Jersey.
hydrocarbons and hydrogen, which ignited resulting in a Operators were unable to control the temperature of the
fire and explosion. There was one fatality and 46 people batch, and the runaway reaction partially vented through the
were injured. One of the causes was excess temperature in reactor manway into the production building. 9 people were
one of the hydrocracking reactors. The specified maximum injured, 2 seriously, and chemicals were released into the
temperature in the reactors was 800ºF (425ºC), and the surrounding community. It is believed that the initial
system was supposed to be shut down if the temperature temperature of the batch was higher than normal, making it
exceeded this value. The reactor and the pipe which more difficult for operators to control the batch temperature
ruptured were believed to have reached a temperature with the available cooling.
greater than 1400ºF (760ºC).
In 8 of the previous 32 batches produced, operators had
Previous temperature excursions in excess of the difficulty in controlling the batch temperature. The
specified 800ºF (425ºC) maximum had occurred, but the temperature and the rate of temperature rise for individual
system had not been shut down. This led operators to steps of the process were beyond the limits specified by the
believe that these excursions were acceptable. Also, some procedure. In some cases, the temperature exceeded the
of these temperature excursions were not investigated, and maximum range of the reactor temperature recorder (150ºC
recommendations from those that were investigated were or 300ºF). In those batches, operators were able to regain
not all implemented. control of the batch temperature without a runaway
reaction. These temperature excursions were not
investigated, and no action was taken in response to them.

What can you do?

Although these two incidents occurred in completely different types of manufacturing plant, they have one important
thing in common. In both incidents, the process had exceeded specified safe operating limits during operations before the
incident. The abnormal conditions became accepted – this is called “normalization of deviation.” These warning signs
were either not investigated, or actions recommended by the investigation were not implemented. “Conduct of
Operations” can be summarized in two simple concepts: (1) Say what you intend to do (procedures), and (2) Always do
what you say. This means, for example, that if your operating procedures say to shut down if a critical safety parameter
exceeds a specified value, you must always take this action!
• Know what the critical safety process parameters are for your plant, know the consequences of exceeding them, and
know what to do if they are exceeded.
• Always take the required actions if critical safety parameters are violated.
• If critical safety parameters are exceeded, report it to management so an appropriate investigation can be done.

What are your plant’s critical safety control limits?

©AIChE2015. All rights reserved. Reproduction for non-commercial, educational purposes is encouraged. However, reproduction for any
commercial purpose without express written consent of AIChE is strictly prohibited. Contact us at ccps_beacon@aiche.org or 646-495-1371.

The Beacon is usually available in Arabic, Afrikaans, Chinese, Czech, Danish, Dutch, English, French, German, Greek, Gujarati, Hebrew, Hindi, Italian, Japanese, Korean,
Malay, Marathi, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Telugu, Thai, Turkish, and Vietnamese.
 Questions:

1. (5 min) What two errors led to the 1997 hydrocracking unit explosion? Describe how
the errors could have been corrected to prevent or mitigate the incident.

2. (5-10 min) For the 1998 batch reactor explosion, what went wrong and how could it
have been prevented? Consider whether operators had a false sense of safety with regard
to large temperature variations.

3. ***(20-30 min) Whenever reactor temperatures are exceeded, significant temperature

variation is observed, or reactor temperature cannot be properly controlled, what should
be done to the system from a design or maintenance standpoint? What kinds of things
would you check? Hint: You may find helpful the safety recommendations outlined in
the Measures section of this article.

4. (5 min) What did you learn?

What lessons have you learned from this article and how can you apply them to your
chemical engineering career?