Learning from what went wrong--two case studies.

Rande H Grotefendt; J Wayne Chastain

doi:10.1016/j.jhazmat.2004.05.005

Home/Archive/PubMed

PubMed · 15518972

Learning from what went wrong--two case studies.

Rande H Grotefendt·J Wayne Chastain

View original source ↗Export metadata ↓

Abstract

Safer design of process equipment can protect against unexpected events. Two case studies involving the design of a process vessel and the subsequent events will be reviewed. One case study will show how the original design minimized equipment damage from an operational error, and how additional safeguards will prevent recurrence. The second case study will show that over time small process changes can lead to an unexpected chemical reaction that results in a vessel rupture. We will also cover the additional safeguards added to prevent recurrence.

Explore related subjects

Chemical Industry Equipment Design Equipment Failure Analysis Hydrogen Maleic Anhydrides Nitrogen Polymers Safety Management Evaluation Study Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Rande H Grotefendt, J Wayne Chastain. 2004-11-11. Learning from what went wrong--two case studies.. https://doi.org/10.1016/j.jhazmat.2004.05.005

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2006

Mercury speciation in the hair of pre-school children living near a chlor-alkali plant.

Exposure to mercury species was assessed in the hair of 130 Spanish children (age 4) from the general population in two areas. Both areas are exposed to different sources of mercury: a point source in Ribera d'Ebre (northeastern Spain) and a diffuse source on the island of Menorca (northwestern Mediterranean). The median MeHg values in the hair of children from Ribera d'Ebre (RE) were nearly twice (0.631 microg/g vs. 0.370 microg/g) those of children from Menorca (MC) (p < 0.05). Total Hg showed a similar trend (REmedian: 0.720 microg/g vs. MCmedian: 0.476 microg/g). Nevertheless, inorganic mercury levels were similar in the two groups of children (REmedian: 0.186 microg/g vs. MCmedian: 0.210 microg/g). Two subgroups of the Ribera d'Ebre group were defined: children living in Flix (a village near a chlor-alkali plant) (RE1) and children living on the outskirts of Flix with no clear, direct influence of the plant (RE2). The mercury concentrations in RE1 were also significantly higher than those in Menorca, but no significant differences were found between Menorca and the RE2 subgroup. We evaluated the fish consumption of RE1, RE2 and MC and found that the Menorcan children consumed significantly less fish (p < 0.05) than the other two subgroups. Children who consumed fish more than three times a week had higher MeHg concentrations (beta (SE) = 0.991 (0.279) than those who ate it less than once a week. Nevertheless, the differences in MeHg levels between children from Ribera d'Ebre and Menorca remained statistically significant after adjustment for fish intake and other variables (beta (SE) = 0.779 (0.203) for children from RE1). In conclusion, local sources other than seafood contribute significantly to MeHg content in hair in the two Ribera d'Ebre subgroups.

Chemical Industry↗

PubMed2006

Chemical process based reconstruction of exposures for an epidemiological study. I. Theoretical and methodological issues.

In the occupational hygiene component of occupational epidemiological studies the goal is to assign group average exposure levels that can be used to compute individual cumulative exposures. This task requires the availability of sufficient amounts of proper individual exposure level data. Typically, the required data are either sparse, completely lacking or happenstance data collected for purposes not suitable for the aims of the study. In the epidemiological study of mortality patterns among industrial workers exposed to chloroprene and other substances, we developed and used a process analysis and modeling based exposure reconstruction to augment, extrapolate, or interpolate the available exposure data. The models developed utilize equations based on the engineering principles and chemistry associated with the processes as determined from the process documentation and task performance habits as determined from interviews of knowledgeable personnel. The resulting equations are tractable and provide a general basis for calculating exposure levels for vapors. The validation of the results with available exposure measurements suggests that comprehensive process analysis and modeling may be used to reconstruct exposures or to evaluate exposure potential with scientifically defensible methods. Furthermore, even in the absence of validating data, the methodology developed has potentially very useful applications in predicting exposure levels to newly synthesized substances. Properly interpreted, the limitations of modeling can be minimized to obtain scientifically reasonable results.

Chemical Industry↗

PubMed2006

Chemical process-based reconstruction of exposures for an epidemiological study. Part II. Estimated exposures to chloroprene and vinyl chloride.

In a four-facility occupational epidemiology study of chloroprene monomer and polymer production workers, the chloroprene (CD) and vinyl chloride monomer (VCM) exposures were modeled for plant specific job title classes. In two facilities an acetylene-based process was used and in the other two plants only a butadiene-based process was used in the monomer synthesis. In the Acetylene process VCM was an undesirable by-product to be removed. In the newer butadiene-based process, VCM was not involved and the exposures to CD were considerably lower than they were in the earlier years. One of the limiting factors was the operator rotation within a number of job titles. This rotation and inability to differentiate between job titles subsumed in job classifications recorded in the work histories required an exposure classification scheme based on an order of magnitude separation of exposure classes. In the four facilities with considerable variation in the mix of the production methods, the CD exposures were remarkably similar in both calculated and measured values. The reductions in exposures were much more dependent upon the improvement of the production methods, rather than deliberate exposure control for occupational hygiene considerations. This is reasonable since the exposures were generally lower than the coeval exposure limits and/or guidelines. The estimated exposures were less than 100 ppm in the pre-1960 era and less than 10 ppm in the 1960-1980 era, less than 1 ppm 1980-1990 era and less than 0.5 ppm thereafter. The exposures were categorized in four classes for VCM and six classes for CD. The characteristic class exposure values were used to cumulate individual exposures over time with a quantification of the potential range for exposures that are reasonably certain to ascribe correct ranking to job classes.

Chemical Industry↗