Wednesday, 16 March 2011 21:05

Aluminium Smelting and Refining

Rate this item
(32 votes)

Process Overview

Bauxite is extracted by open-pit mining. The richer ores are used as mined. The lower grade ores may be beneficiated by crushing and washing to remove clay and silica waste. The production of the metal comprises two basic steps:

  1. Refining. Production of alumina from bauxite by the Bayer process in which bauxite is digested at high temperature and pressure in a strong solution of caustic soda. The resulting hydrate is crystallized and calcined to the oxide in a kiln or fluid bed calciner.
  2. Reduction. Reduction of alumina to virgin aluminium metal employing the Hall-Heroult electrolytic process using carbon electrodes and cryolite flux.


Experimental development suggests that in the future aluminium may be reduced to the metal by direct reduction from the ore.

There are presently two major types of Hall-Heroult electrolytic cells in use. The so-called “pre-bake” process utilizes electrodes manufactured as noted below. In such smelters exposure to polycyclic hydrocarbons normally occurs in the electrode manufacturing facilities, especially during mixing mills and forming presses. Smelters utilizing the Soderberg-type cell do not require facilities for the manufacture of baked carbon anodes. Rather, the mixture of coke and pitch binder is put into hoppers whose lower ends are immersed in the molten cryolite-alumina bath mixture. As the mixture of pitch and coke is heated by the molten metal-cryolite bath within the cell, this mixture bakes into a hard graphitic mass in situ. Metal rods are inserted into the anodic mass as conductors for a direct current electric flow. These rods must be replaced periodically; in extracting these, considerable amounts of coal tar pitch volatiles are evolved into the cell room environment. To this exposure is added those pitch volatiles generated as the baking of the pitch-coke mass proceeds.

Within the last decade the industry has tended to either not replace or to modify existent Soderberg type reduction facilities as a consequence of the demonstrated carcinogenic hazard they present. In addition, with the increasing automation of reduction cell operations—particularly the changing of anodes, tasks are more commonly performed from enclosed mechanical cranes. Consequently worker exposures and the risk of developing those disorders associated with aluminium smelting are gradually decreasing in modern facilities. By contrast, in those economies wherein adequate capital investment is not readily available, the persistence of older, manually operated reduction processes will continue to present the risks of those occupational disorders (see below) previously associated with aluminium reduction plants. Indeed, this tendency will tend to become more aggravated in such older, unimproved operations, especially as they age.

Carbon electrode manufacture

The electrodes required by pre-bake electrolytic reduction to pure metal are normally made by a facility associated with this type of aluminium smelting plant. The anodes and cathodes are most frequently made from a mixture of ground petroleum-derived coke and pitch. Coke first is ground in ball mills, then conveyed and mixed mechanically with the pitch and finally cast into blocks in a moulding presses. These anode or cathode blocks are next heated in a gas-fired furnace for several days until they form hard graphitic masses with essentially all volatiles having been driven off. Finally they are attached to anode rods or saw-grooved to receive the cathode bars.

It should be noted that the pitch used to form such electrodes represents a distillate which is derived from coal or petroleum tar. In the conversion of this tar to pitch by heating, the final pitch product has boiled off essentially all of its low-boiling point inorganics, e.g., SO2, as well as aliphatic compounds and one- and two ring aromatic compounds. Thus, such pitch should not present the same hazards in its use as coal or petroleum tars since these classes of compounds ought not to be present. There are some indications that the carcinogenic potential of such pitch products may not be as great as the more complex mixture of tars and other volatiles associated with the incomplete combustion of coal.

Hazards and Their Prevention

The hazards and preventive measures for aluminium smelting and refining processes are basically the same as those found in smelting and refining in general; however, the individual processes present certain specific hazards.


Although sporadic references to “bauxite lung” occur in the literature, there is little convincing evidence that such an entity exists. However, the possibility of the presence of crystalline silica in bauxite ores should be considered.

Bayer process

The extensive use of caustic soda in the Bayer process presents frequent risks of chemical burns of the skin and eyes. Descaling of tanks by pneumatic hammers is responsible for severe noise exposure. The potential hazards associated with the inhalation of excessive doses of aluminium oxide produced in this process are discussed below.

All workers involved in the Bayer process should be well informed of the hazards associated with handling caustic soda. In all sites at risk, eyewash fountains and basins with running water and deluge showers should be provided, with notices explaining their use. PPE (e.g., goggles, gloves, aprons and boots) should be supplied. Showers and double locker accommodations (one locker for work clothing, the other for personal clothing) should be provided and all employees encouraged to wash thoroughly at the end of the shift. All workers handling molten metal should be supplied with visors, respirators, gauntlets, aprons, armlets and spats to protect them against burns, dust and fumes. Workers employed on the Gadeau low-temperature process should be supplied with special gloves and suits to protect them from hydrochloric acid fumes given off when the cells start up; wool has proved to have a good resistance to these fumes. Respirators with charcoal cartridges or alumina-impregnated masks give adequate protection against pitch and fluorine fumes; efficient dust masks are necessary for protection against carbon dust. Workers with more severe dust and fume exposure, particularly in Soderberg operations, should be provided with air-supplied respiratory protective equipment. As mechanized potroom work is remotely performed from enclosed cabins, these protective measures will become less necessary.

Electrolytic reduction

Electrolytic reduction exposes workers to the potential for skin burns and accidents due to molten metal splashes, heat stress disorders, noise, electrical hazards, cryolite and hydrofluoric acid fumes. Electrolytic reduction cells may emit large quantities of dusts of fluoride and alumina.

In carbon-electrode manufacturing shops, exhaust ventilation equipment with bag filters should be installed; enclosure of pitch and carbon grinding equipment further effectively minimizes exposures to heated pitches and carbon dusts. Regular checks on atmospheric dust concentrations should be made with a suitable sampling device. Periodic x-ray examinations should be carried out on workers exposed to dust, and these should be followed up by clinical examinations when necessary.

In order to reduce the risk of handling pitch, transport of this material should be mechanized as far as possible (e.g., heated road tankers can be used to transport liquid pitch to the works where it is pumped automatically into heated pitch tanks). Regular skin examinations to detect erythema, epitheliomata or dermatitis are also prudent, and extra protection can be provided by alginate-base barrier creams.

Workers doing hot work should be instructed prior to the onset of hot weather to increase fluid intake and heavily salt their food. They and their supervisors should also be trained to recognise incipient heat-induced disorders in themselves and their co-workers. All those working here should be trained to take the proper measure necessary to prevent the occurrence or progression of the heat disorders.

Workers exposed to high noise levels should be supplied with hearing protection equipment such as earplugs which allow the passage of low-frequency noise (to allow perception of orders) but reduce the transmission of intense, high-frequency noise. Moreover, workers should undergo regular audiometric examination to detect hearing loss. Finally, personnel should also be trained to give cardiopulmonary resuscitation to victims of electric shock accidents.

The potential for molten metal splashes and severe burns are widespread at many sites in reduction plants and associated operations. In addition to protective clothing (e.g., gauntlets, aprons, spats and face visors) the wearing of synthetic apparel should be prohibited, since the heat of molten metal causes such heated fibers to melt and adhere to the skin, further intensifying skin burns.

Individuals using cardiac pacemakers should be excluded from reduction operations because of the risk of magnetic field induced dysrhythmias.

Other Health Effects

The hazards to workers, the general population and the environment resulting from the emission of fluoride-containing gases, smokes and dusts due to the use of cryolite flux have been widely reported (see table 1). In children living in the vicinity of poorly controlled aluminium smelters, variable degrees of mottling of permanent teeth have been reported if exposure occurred during the developmental phase of permanent teeth growth. Among smelter workers prior to 1950, or where inadequate control of fluoride effluents continued, variable degrees of bony fluorosis have been seen. The first stage of this condition consists of a simple increase in bone density, particularly marked in the vertebral bodies and pelvis. As fluoride is further absorbed into bone, calcification of the ligaments of the pelvis is next seen. Finally, in the event of extreme and protracted exposure to fluoride, calcification of the paraspinal and other ligamentous structures as well as joints are noted. While this last stage has been seen in its severe form in cryolite processing plants, such advanced stages have rarely if ever been seen in aluminium smelter workers. Apparently the less severe x-ray changes in bony and ligamentous structures are not associated with alterations of the architectural or metabolic function of bone. By proper work practices and adequate ventilatory control, workers in such reduction operations can be readily prevented from developing any of the foregoing x-ray changes, despite 25 to 40 years of such work. Finally, mechanization of potroom operations should minimize if not totally eliminate any fluoride associated hazards.

Table 1. Process materials inputs and pollution outputs for aluminium smelting and refining


Material input

Air emissions

Process wastes

Other wastes

Bauxite refining

Bauxite, sodium hydroxide

Particulates, caustic/water


Residue containing silicon, iron, titanium, calcium oxides and caustic

Alumina clarification and precipitation

Alumina slurry, starch, water


Wastewater containing starch, sand and caustic


Alumina calcination

Aluminium hydrate

Particulates and water vapour


Primary electrolytic
aluminium smelting

Alumina, carbon anodes, electrolytic cells, cryolite

Fluoride—both gaseous and particulates, carbon dioxide, sulphur dioxide, carbon monoxide, C2F6 ,CF4 and perfluorinated carbons (PFC)


Spent potliners


Since the early 1980s an asthma-like condition has been definitively demonstrated among workers in aluminium reduction potrooms. This aberration, referred to as occupational asthma associated with aluminium smelting (OAAAS), is characterized by variable airflow resistance, bronchial hyperresponsiveness, or both, and is not precipitated by stimuli outside the workplace. Its clinical symptoms consist of wheezing, chest tightness and breathlessness and non-productive cough which are usually delayed some several hours following work exposures. The latent period between commencement of work exposure and the onset of OAAAS is highly variable, ranging from 1 week to 10 years, depending upon the intensity and character of the exposure. The condition usually is ameliorated with removal from the workplace following vacations and so on, but will become more frequent and severe with continued work exposures.

While the occurrence of this condition has been correlated with potroom concentrations of fluoride, it is not clear that the aetiology of the disorder arises specifically from exposure to this chemical agent. Given the complex mixture of dusts and fumes (e.g., particulate and gaseous fluorides, sulphur dioxide, plus low concentrations of the oxides of vanadium, nickel and chromium) it is more likely that such fluorides measurements represent a surrogate for this complex mixture of fumes, gases and particulates found in potrooms.

It presently appears that this condition is one of an increasingly important group of occupational diseases: occupational asthma. The causal process which results in this disorder is determined with difficulty in an individual case. Signs and symptoms of OAAAS may result from: pre-existing allergy-based asthma, non-specific bronchial hyperresponsiveness, the reactive airway dysfunction syndrome (RADS), or true occupational asthma. Diagnosis of this condition is presently problematic, requiring a compatible history, the presence of variable airflow limitation, or in its absence, production of pharmacologically induced bronchial hyperresponsivity. But if the latter is not demonstrable, this diagnosis is unlikely. (However, this phenomenon can eventually disappear after the disorder subsides with removal from work exposures.)

Since this disorder tends to become progressively more severe with continued exposure, affected individuals most usually need be removed from continued work exposures. While individuals with pre-existent atopic asthma should initially be restricted from aluminium reduction cell rooms, the absence of atopy cannot predict whether this condition will occur subsequent to work exposures.

There are presently reports suggesting that aluminium may be associated with neurotoxicity among workers engaged in smelting and welding this metal. It has been clearly shown that aluminium is absorbed via the lungs and excreted in the urine at levels greater than normal, particularly in reduction cell room workers. However, much of the literature regarding neurological effects in such workers derives from the presumption that aluminium absorption results in human neurotoxicity. Accordingly, until such associations are more reproducibly demonstrable, the connection between aluminium and occupational neurotoxicity must be considered speculative at this time.

Because of the occasional need to expend in excess of 300 kcal/h in the course of changing anodes or performing other strenuous work in the presence of molten cryolite and aluminium, heat disorders may be seen during periods of hot weather. Such episodes are most likely to occur when the weather initially changes from the moderate to hot, humid conditions of summer. In addition, work practices which result in accelerated anode changing or employment over two successive work shifts during hot weather will also predispose workers to such heat disorders. Workers inadequately heat acclimatized or physically conditioned, whose salt intake is inadequate or who have intercurrent or recent illness are particularly prone to development of heat exhaustion and/or heat cramps while performing such arduous tasks. Heat stroke has occurred but rarely among aluminium smelter workers except among those with known predisposing health alterations (e.g., alcoholism, ageing).

Exposure to the polycyclic aromatics associated with breathing of pitch fume and particulates have been demonstrated to place Soderberg-type reduction cell personnel in particular at an excessive risk of developing urinary bladder cancer; the excess cancer risk is less well-established. Workers in carbon electrode plants where mixtures of heated coke and tar are heated are assumed to also be at such risk. However, after electrodes have been baked for several days at about 1,200 °C, polycyclic aromatic compounds are practically totally combusted or volatilized and are no longer associated with such anodes or cathodes. Hence the reduction cells utilizing prebaked electrodes have not been as clearly shown to present an undue risk of development of these malignant disorders. Other neoplasia (e.g., non-granulocytic leukaemia and brain cancers) have been suggested to occur in aluminium reduction operations; at present such evidence is fragmentary and inconsistent.

In the vicinity of the electrolytic cells, the use of pneumatic crust breakers in the potrooms produce noise levels of the order of 100 dBA. The electrolytic reduction cells are run in series from a low-voltage high-amperage current supply and, consequently, cases of electric shock are not usually severe. However, in the power house at the point where the high-voltage supply joins the series-connection network of the potroom, severe electrical shock accidents may occur particularly as the electrical supply is an alternating, high voltage current.

Because health concerns have been raised regarding exposures associated with electromagnetic power fields, the exposure of workers in this industry has been brought into question. It must be recognized that the power supplied to electrolytic reduction cells is direct current; accordingly, the electromagnetic fields generated in the potrooms are mainly of the static or standing field type. Such fields, in contrast to low frequency electromagnetic fields, are even less readily shown to exert consistent or reproducible biological effects, either experimentally or clinically. In addition, the flux levels of the magnetic fields measured in present day cell rooms are commonly found to be within presently proposed, tentative threshold limit values for static magnetic fields, sub-radio frequency and static electric fields. Exposure to ultra-low frequency electromagnetic fields also occur in reduction plants, especially at the far-ends of these rooms adjacent to rectifier rooms. However, the flux levels found in the nearby potrooms are minimal, well below present standards. Finally, coherent or reproducible epidemiological evidence of adverse health effects due to electromagnetic fields in aluminium reduction plants have not been convincingly demonstrated.

Electrode manufacture

Workers in contact with pitch fumes may develop erythema; exposure to sunlight induces photosensitization with increased irritation. Cases of localized skin tumours have occurred among carbon electrode workers where inadequate personal hygiene was practised; after excision and change of job no further spread or recurrence is usually noted. During electrode manufacture, considerable quantities of carbon and pitch dust can be generated. Where such dust exposures have been severe and inadequately controlled, there have been occasional reports that carbon electrode makers may develop simple pneumoconiosis with focal emphysema, complicated by the development of massive fibrotic lesions. Both the simple and complicated pneumoconioses are indistinguishable from the corresponding condition of coalworkers’ pneumoconiosis. The grinding of coke in ball mills produces noise levels of up to 100 dBA.

Editor’s note: The aluminium production industry has been classified as a Group 1 known cause of human cancers by the International Agency for Research on Cancer (IARC). A variety of exposures have been associated with other diseases (e.g., “potroom asthma”) which are described elsewhere in this Encyclopaedia.



Read 14972 times Last modified on Wednesday, 10 August 2011 23:13

" DISCLAIMER: The ILO does not take responsibility for content presented on this web portal that is presented in any language other than English, which is the language used for the initial production and peer-review of original content. Certain statistics have not been updated since the production of the 4th edition of the Encyclopaedia (1998)."


Metal Processing and Metal Working Industry References

Buonicore, AJ and WT Davis (eds.). 1992. Air Pollution Engineering Manual. New York: Van Nostrand Reinhold/Air and Waste Management Association.

Environmental Protection Agency (EPA). 1995. Profile of the Nonferrous Metals Industry. EPA/310-R-95-010. Washington, DC: EPA.

International Association for Research on Cancer (IARC). 1984. Monographs on the Evaluation of Carcinogenic Risks to Humans. Vol. 34. Lyon: IARC.

Johnson A, CY Moira, L MacLean, E Atkins, A Dybunico, F Cheng, and D Enarson. 1985. Respiratory abnormalities amongst workers in iron and steel industry. Brit J Ind Med 42:94–100.

Kronenberg RS, JC Levin, RF Dodson, JGN Garcia, and DE Griffith. 1991. Asbestos-related disease in employees of a steel mill and a glass bottle manufacturing plant. Ann NY Acad Sci 643:397–403.

Landrigan, PJ, MG Cherniack, FA Lewis, LR Catlett, and RW Hornung. 1986. Silicosis in a grey iron foundry. The persistence of an ancient disease. Scand J Work Environ Health 12:32–39.

National Institute for Occupational Safety and Health (NIOSH). 1996. Criteria for a Recommended Standard: Occupational Exposures to Metalworking Fluids. Cincinatti, OH: NIOSH.

Palheta, D and A Taylor. 1995. Mercury in environmental and biological samples from a gold mining area in the Amazon Region of Brazil. Science of the Total Environment 168:63-69.

Thomas, PR and D Clarke. 1992 Vibration white finger and Dupuytren’s contracture: Are they related? Occup Med 42(3):155–158.