Page 70 - Plant design and economics for chemical engineers

52 PLAN-F DESIGN AND ECONOMICS FOR CHEMICAL ENGINEERS

inventory of all materials present in the various stages of the process. Even
when materials are present in only trace amounts, there is more than enough
present to produce a potentially hazardous situation in a localized work area.
Generally, feedstocks and products of a process are well known. Intermediates,
by-products, and waste materials may be less conspicuous and may not even
have been identified. Other materials, such as catalysts, additives, cleaning
agents, and maintenance materials need to be identified to complete the
inventory.
An estimate of the toxicity or intrinsic hazard is needed for each material
identified in the inventory. Such information for many chemicals in the form of
a Material Safety Data Sheet (MSDS) are required by the OSHA Hazard
Communication Standard. (Other countries have similar requirements.) Stan-
dard hazard-data sources may need to be consulted for those chemical com-
pounds for which no MSDSs are presently available. Adequate hazard data may
be lacking for various mixtures that are unique to the plant. For such mixtures,
it may be necessary to analyze the contents and then estimate the overall hazard
based on the individual components.
To perform a risk assessment and then prioritize the exposure measure-
ment effort requires an approximate initial exposure potential assessment. For
each chemical present and for each source of exposure for that chemical, an
estimate of exposure can be made. These exposure estimates combined with a
toxicity estimate from the hazard data can then be combined to yield a risk
estimate which can be used as a basis for prioritization of the measurement and
monitoring effort.
It is generally not necessary to make an exposure estimate for every
chemical/exposure source combination since many will be of such low signifi-
cance that they can be neglected. For those chemical/exposure source combina-
tions that could be near the top of the priority list, the exposure estimate is
probably not needed beyond an order of magnitude. Methods for making this
type of estimate have been developed by the EPA for the purpose of evaluating
Premanufacturing Notifications (PMNs).
Contaminant concentrations in a typical plant environment are highly
variable. The background level of exposure in a chemical plant is generally the
result of a large number of small fugitive emissions, each varying with time.
These variability aspects in the contaminant concentrations and the exposure of
workers require that a sufficient number of samples be taken to permit charac-
terization of the statistical distribution and permit estimation of exposure over
the appropriate averaging time. In mathematical terms, the averaging time
should be no longer than the biological half-time of a substance acting in the
body. Although the range of biological half-times is continuous, for simplicity
only a few discrete averaging times are commonly used. For fast-acting sub-
stances 15 and 30 min are used, while 8 h is most often used for substances with
biological half-times longer than 8 h. The latter is generally labeled as the 8-h
time weighted average (TWA).
The most commonly used methods for the analysis of airborne contami:
nants are listed in Table 2. Any method used for a particular contaminant must