Coke-Oven Emissions and Cancer

Coke is coal that has been baked. Bituminous coal is heated to between 1,000 °C and 1,400 °C with no oxygen present. Tars and light oils distill out. What is left is a hard, almost pure carbon fuel, and its main job is to feed iron-making blast furnaces. One metric ton of coal yields roughly 545 to 635 kg of coke.

"Coke-oven emissions" has a precise technical meaning. It is the benzene-soluble fraction of the total particulate matter given off during coke production. The by-product process is built to capture the volatile material. Some of it escapes anyway. It leaks through structural defects around oven doors and charging lids, through engineering controls used improperly, and through work practices that let gas out.

The leaking is not spread evenly across a shift. About 60% of total coke-oven emissions occur during charging, when coal drops into a hot oven. About 30% occur during pushing, when finished coke is rammed out. The last 10% come during quenching.

More than 60 organic compounds have been identified in air sampled at coke plants. More than 40 are polycyclic aromatic hydrocarbons, or PAHs. These are ring-shaped molecules that form when carbon-rich material burns badly.

The rest of the mixture is a chemistry lesson in itself. There is formaldehyde, acrolein, other aliphatic aldehydes, ammonia, carbon monoxide, nitrogen oxides, and phenol. There are three metals: cadmium, arsenic, and mercury. Coke-oven gas carries hydrogen, methane, ethane, and hydrogen sulfide, among other gases. The gas tar carries pyridine, tar acids, naphthalene, creosote oil, and coal-tar pitch. Benzene, xylene, and toluene can be pulled out of the light-oil fraction.

Several of those are carcinogens on their own. Our pages on PAHs and cancer, arsenic, and cadmium cover them one at a time. The National Toxicology Program adds one more point. The mixture contains agents that make other chemical carcinogens act more strongly, especially on the airways.

Before 1950, case reports already linked coke production to cancer of the skin, the urinary bladder, and the respiratory tract. Cohort studies followed in the United States, the United Kingdom, Japan, and Sweden. All reported raised lung cancer risk. Smoking was accounted for in some of them and did not explain the result.

The largest was published in 1969 and followed 59,000 steel workers. Lung cancer risk rose as duration of exposure rose, and as intensity of exposure rose. That dose-response pattern is part of why the evidence is considered strong rather than suggestive.

Kidney cancer is the second signal. Several studies of coking-plant workers reported raised risk. The National Cancer Institute is more careful here than with the lung finding. It calls kidney cancer a possible link rather than a settled one. Excess cancer of the prostate, large intestine, and pancreas turned up in no more than one study each, which is too thin to build on.

Animal work points the same way. Coke-oven emission samples put on the skin of mice weekly for up to 52 weeks caused skin cancer. Coal-tar aerosols made from coke-oven samples caused benign and malignant lung tumors in rats and mice when inhaled, and skin tumors in female mice.

The National Toxicology Program lists coke-oven emissions as known to be human carcinogens. The basis is sufficient evidence in humans. That listing dates to the Second Annual Report on Carcinogens in 1981. Coke-oven emissions have no CAS registry number, because they are a mixture rather than a single chemical.

The International Agency for Research on Cancer reviewed coke production in Volume 100F of its monographs. Volume 100 exists to re-review every agent already classed as carcinogenic to humans, Group 1. So the Group 1 label is not new either. For what those groups do and do not mean, see how the IARC monographs work and hazard versus risk.

OSHA's coke oven emissions standard is 29 CFR 1910.1029. Its exposure limit is 150 micrograms per cubic meter of air, averaged over any 8-hour period, measured as the benzene-soluble fraction. NIOSH recommends a slightly different figure: a time-weighted-average of 0.2 mg/m³, which is 200 micrograms per cubic meter. NIOSH also lists coke-oven emissions as a potential occupational carcinogen.

For scale, breathing-zone averages compiled by IARC ran from 0.39 mg/m³ for a pusher-machine operator up to 3.22 mg/m³ for a lidman, 3.14 for a tar chaser, and 3.05 for a larry-car operator. Those are many times the current limit, which is what the standard was written to end.

The rule also draws lines on a map. Regulated areas must include the coke oven battery with its topside, pushside, coke side, and battery ends, plus the wharf, the screening station, and any beehive oven. Only authorized people go in. Sampling must be full-shift, meaning at least 7 continuous hours. There must be at least one sample per shift per battery for each named job. The named jobs include lidman, tar chaser, larry car operator, luterman, and machine operator on the coke side. Also benchman on either side, heater, quenching car operator, pusher machine operator, screening station operator, wharfman, and oven patcher. Samples measure what reaches the worker, ignoring any respirator.

Housekeeping rules follow from that line on the map. Change rooms must be clean, with street clothes stored apart from protective gear. Every shift ends with a shower. Lunchrooms need temperature-controlled, positive-pressure, filtered air. Hands and face washed before eating. No food, drink, smoking products, or cosmetics inside the regulated area. Drinking water is the one exception.

Anyone working in a regulated area at least 30 days a year is covered by a medical surveillance program. It is free to the worker. A licensed physician runs it or supervises it. The first exam has seven parts.

• Work and medical history. This includes smoking history and any breathlessness, cough, sputum, or wheezing.
• A standard posterior-anterior chest X-ray.
• Lung function tests. These measure forced vital capacity and forced expiratory volume in one second.
• Weight.
• A skin examination.
• Urinalysis for sugar, albumin, and blood in the urine.
• A urinary cytology exam. It looks for abnormal cells shed from the bladder lining.

Most of those repeat at least annually. Workers aged 45 or older, or with 5 or more years in the regulated area, also get the urinary cytology every year. The physician then writes an opinion stating whether any condition puts that worker at increased risk, and what limits on exposure or protective equipment are advised.

The bladder and kidney tests are not decoration. They exist because the earliest coke-oven reports described urinary tract cancer. Urinary cytology can also find bladder cancer before it causes symptoms. Workers with a heavy smoking history have another question to raise: lung cancer screening. Quitting smoking lowers lung cancer risk on top of anything the exposure controls achieve.

In 1970 the United States had 64 coking plants running more than 13,000 ovens, with roughly 10,000 coke-oven workers. By 1998 that was 23 plants and about 3,800 ovens. Pollution controls cut emissions further over the same decades.

Exposure has not disappeared, and it reaches beyond the coke plant itself. NCI notes occupational exposure among workers in the aluminum, steel, graphite, electrical, and construction industries. The National Toxicology Program adds that people living near coking and coal-tar plants have a high potential for exposure. Both breathing it in and taking it up through skin count.

Coke-oven emissions are a listed hazardous air pollutant under the Clean Air Act. EPA's Urban Air Toxics Strategy named them among 33 pollutants posing the greatest threat to public health in urban areas. Under the Superfund law, the reportable quantity is 1 pound.

• NCI — Coke Oven Emissions
• NTP Report on Carcinogens, 15th edition — Coke-Oven Emissions profile
• OSHA 29 CFR 1910.1029 — Coke oven emissions
• IARC Monographs Volume 100F — Chemical Agents and Related Occupations