Chemistry Reference
In-Depth Information
7.2 Human Studies
The literature over the past two decades is replete with
expressions of concern regarding exposures of workers
in the semiconductor industry (Flora, 2000; Fowler and
Sexton, 2002; Fowler and Silbergeld, 1989; Fowler and
Woods, 1979; Nordberg and Nordberg, 1996; Tanaka,
2004; Webb,
et al
., 1984;Yamauchi,
et al
., 1989). Although
frequently based on substantial animal data, such as cited
in this chapter, which establish toxicity levels in identi-
fi ed organ systems for specifi c metals, such as gallium
arsenide, these concerns have resulted in a fairly mod-
est and uneven database regarding human health. The
problems, such as access to workers and records, chang-
ing work environments and sites, the identifi cation and
documentation of the specifi c exposure chemicals and
levels, have limited the accumulation of evidence and
understanding for this ever-growing number of work-
ers (Fowler and Silbergeld, 1989). Some editorials have
attributed at least some of the lag in assembling a body
of evidence to the industry's desire to disregard any
fi ndings suggestive of a health problem (Wadman, 2004;
Watterson and LaDou, 2003). Certainly, factors such as
the movement of production sites from developed to
developing countries (Tenenbaum, 2003), law suits
(Wadman, 2004), technological changes in manufactur-
ing (Chepesiuk, 1999; Flora, 2000), and industry-control-
led studies (Brumfi el, 2004; Watterson and LaDou, 2003)
have made the assembly of a solid base of data regard-
ing the human health effects resulting from exposure to
chemicals used in the semiconductor industry diffi cult.
Overall, the scientifi c evidence is far less than needed.
Earlier studies focused on reproductive effects (Correa
et al
., 1996; Pastides
et al
., 1988) and more recent stud-
ies on other conditions (Wang
et al
., 2002). The results
can be examined for consistency of fi ndings with those
from animal and cellular studies. In addition, gallium,
and other semiconductor materials, such as indium,
have been used at lower doses as adjuvant therapeutic
agencies. Some properties of these chemicals (e.g., pro-
duction of apoptosis) at high doses have been found to
be useful in medical treatment of quite distinct health
problems, such as treatment of cancer or wound heal-
ing. Nevertheless, as noted in the following, questions
of safety for workers exposed to these agents during the
manufacture of III-V semiconductors or light-emitting
diodes remain.
carcinogenic to humans
(Group 1). For gallium arsenide
itself, no data on human cancer were available, and
the evidence of carcinogenicity in experimental ani-
mals was considered
limited
on the basis of increased
incidence of bronchioloalveolar neoplasms observed
in female rats in one study. However, once in the body,
gallium arsenide releases a small amount of its arsenic
moiety, which behaves as inorganic arsenic, already
evaluated as
carcinogenic to humans
in Volume 84 (in
press) and Supplement 7 (1987) of the
IARC Mono-
graphs
. The gallium moiety may be responsible for
the pulmonary neoplasms observed in rats in view of
the apparent resistance of this species to the carcino-
genic potential of arsenic” (International Agency for
Research on Cancer, in press). Despite the classifi cation
by organizations such as IARC and the U.S. National
Toxicology Program of gallium arsenide along with
other semiconductor compounds as extremely haz-
ardous, there is very little in the literature on the bio-
monitoring of workers. Liao and colleagues (2004) in
Taiwan found blood indium and urine gallium to be
elevated in the 103 workers tested relative to the 67
controls. The higher level was also found to be related
to job title, gender, educational level, and use of pre-
ventive equipment.
Although no longer available, the United States
Bureau of Labor statistics earlier provided a database
from which an assessment of incidences of reported
illnesses in semiconductor workers could be made.
Data by the Standard Industrial Classifi cation (SIC)
codes were reported separately for workers in the
manufacture of semiconductors and related devices
(SIC 3674) for more than a decade. Fowler and Sex-
ton (2002) analyzed the available data and concluded
that relative to workers in other manufacturing jobs,
semiconductor workers seemed to have higher rates
of illness and that the rates were higher in the 1980s
than in the 1990s. Chee and Rampal (2003) conducted
a survey of female production workers in 18 semicon-
ductor factories in Malaysia. They found that those in
wafer polishing jobs had higher illness rates compared
with assembly workers.
In animals, pulmonary (Goering
et al
., 1988), hemat-
opoietic (Goering
et al
., 1988), renal (Goering
et al
.,
1988), immunological (Sikorski
et al
., 1991), and repro-
ductive (Tanaka, 2004) effects from gallium have been
reported. However, there is little documentation relat-
ing these conditions to workers exposed to gallium
arsenide. Nevertheless, the limited correlative human
data are consistent with those from animals.
7.2.1 Toxicity
The International Agency for Research on Cancer
(IARC) has classifi ed gallium arsenide as a Group 1
carcinogen to humans. In reaching this conclusion,
they stated: “Gallium arsenide was evaluated as
7.2.1.1 Pulmonary
As a part of a clinical survey, 249 semiconductor
workers in one plant in Taiwan were examined for
