Halogens are a family of chemical elements that includes fluorine (F), chlorine (Cl), bromine (Br), and iodine (I). On a global scale, natural sources, the sea in particular, give rise to most of the chlorine, bromine and iodine compounds in the atmosphere. The main natural sources of atmospheric fluorine compounds are volcanoes, fumaroles, forest fires and marine aerosols. The atmospheric fluorine compounds include inorganic gases and particles, and about a dozen organic compounds. This overview considers fluorine first because it is the most widespread and important phytotoxic halogen. Environmental problems caused by chlorine, bromine and iodine are very rare so all available information is summarised here and they are not covered in the database.

Fluorine is one of the most reactive of elements so it is not found in nature in the free, elemental form. The element is used in some processes (e.g. in rocket fuel) but it is very rare. If it is released into the atmosphere, it reacts quickly to form inorganic compounds (NAS, 1971). The term "fluorine" is usually used when referring to the elemental form and the miscellany of inorganic and organic compounds of fluorine are usually referred to collectively as "fluorides". 

Inorganic fluorides

Fluoride emissions in the UK are almost entirely due to human activities: production of fluorides for a multitude of industrial uses; aluminium reduction; steel and phosphate fertiliser production; coal burning; and glass, ceramics and brick manufacture (NAS, 1971). Coal was an important source of hydrogen fluoride before smoke control because sufficient amounts was released from domestic fires to injure plants in towns and cities throughout Britain. Emissions from most other sources have decreased significantly in the last decade but phosphate fertiliser is still an important source of soil contamination of arable and pasture land. Wood fires and entrained soil are minor sources.

Most of the fluoride emitted from anthropogenic sources occurs as inorganic gases and particulates. The most important inorganic gas is hydrogen fluoride (HF) though a few processes release silicon tetrafluoride (SiF4). Aluminium reduction is an important source of HF but it also leads to the release of small quantities of the gas perfluoromethane (CF4), and the magnesium industry releases sulphur hexafluoride (SF6). Both of these are potent greenhouse gases so emissions are being rapidly reduced. The very toxic gas sulphuryl fluoride (Vikane, SO2F2) is used in some countries as a fumigant for warehouses, stores and ships. It is a possible replacement for methyl bromide so its use may increase in the UK. However, by far the main compound of environmental concern is HF.

HF is the most phytotoxic of all air pollutants, being from 10 to 1000 times more toxic than ozone. SO2 or NO2 (Weinstein et al., 1998). The main reason for this is that it is carried in the transpiration stream, leading to accumulation of very high concentrations in the apex and margins of leaves. In sensitive species this may lead to distortion of the leaf shape, chlorosis (yellowing), red colouration and/or death of tissues. (see APIS Plant Effects Gallery). However, the high accumulation in the apex and margins means that concentrations are low (often at background level) in the rest of the leaf. This means that much of the leaf may be physiologically unaffected even when the apex is dead. 

There is an enormous range in the sensitivity of species and varieties, covering at least an order of magnitude. Field surveys by Davison (unpublished) suggest that in two areas of Britain where there were about >200 species of vascular plants about 8-12 showed visible injury and were in the most sensitive to moderately sensitive classes (Weinstein et al., 1998; Weinstein & Davison, 2003). Several monocotyledons are among the most sensitive taxa, notably Gladiolus, Allium, Crocus, Tulipa, Lilium and Polygonatum. Weather affects response, dry conditions for example, decreasing the effects because of lower stomatal conductance. Pine species vary considerably in sensitivity but some are almost as sensitive as the most sensitive monocots. In general, pine needles are only sensitive during the period of extension so the time of exposure is important. Some publications suggest that HF may reduce seed production via effects on pollination/fertilisation but the concentrations used were relatively high so the significance for the natural environment is not known. Lichens have been injured near aluminium smelters in Wales and Scotland (Perkins & Millar, 1987; Gilbert, 1985) but it is difficult to use the data to provide a critical level because atmospheric concentrations were not measured. Most of the effects occurred during start-up, a period when there were rapidly fluctuating and unusually high concentrations so it is not easy to compare their sensitivity with vascular plants.

The critical levels are based on prevention of visible injury by HF. They are the result of many experiments, and years of observations so they are considered to be reliable (NAS, 1971; Weinstein et al., 1989). There is insufficient published research on the effects of HF on growth but evidence indicates that visible injury starts to occur at concentrations lower than those that affect growth so it is reasonable to base the critical levels on visible injury. In addition, there is some evidence that HF may cause stimulation of growth at low concentrations (probably by changing resource allocation), which complicates the issue.

Particulate fluorides differ in size, chemistry and reactivity. For example, wind may entrain soil minerals that contain from <100 to >1000 mg F kg-1, while aluminium reduction releases the minerals chiolite, cryolite and fluoridated alumina. The size range emitted from a smelter is typically from sub-micron to around 5 µm. The larger the particle size, the greater the rate of deposition so particulates are usually only a problem in the near vicinity of industrial sources. In most cases, the reactivity of particulate fluorides is low so they are not usually considered to be a hazard to plants. 

Both HF and particulates contribute to the total fluoride content of plants and therefore to the diet of herbivores. Therefore both a critical level and a critical load are provided for fluorides. Until about 40 years ago, the condition called fluorosis was a major constraint on livestock production over a large area of England, including Rotherham to Chesterfield, large parts of the Bedfordshire brickworks area, the Potteries area of Stoke on Trent and Teeside. In general, the first symptoms of fluorosis appear when the diet has an annual mean greater than about 30-40 mg F kg-1 dry wt. Long-term application of phosphate fertilisers can be an additional source of fluorides in the diet of cattle and sheep because of the amount of soil that they ingest so it is important to take this into account when applying the critical load. Fluorosis has been recorded in several small mammals in Britain (Cooke et al., 1996) but it is not known if soil is an important source in their diet. Although there are authoritative reports of significant effects of fluorides on silk worms, and some inorganic fluorides are used as insecticides, there is not sufficient evidence to provide a critical load for any invertebrate group. However, invertebrates appear to be less sensitive than livestock so it is concluded that it is safe to use a single critical load until there is contradictory evidence.


About 1 million organofluorides have been synthesised. They include: CFCs and other propellants, foam blowers and refrigerants; pesticides; herbicides; surfactants; solid polymers such as Teflon; and anaesthetics. The effects of the CFCs are well known. The CFC replacements (HCFCs and HFCs) do not have such ozone-depleting or warming potential, partly because they breakdown in the troposphere. The chemical nature and toxicology of the breakdown products have been thoroughly investigated (http://www.afeas.org/issue_areas.html) and the main product is trifluoroacetic acid (TFA). TFA is resistant to chemical and biological breakdown (i.e. it accumulates in the environment), and there is a natural, but unidentified source in the environment. The toxicological evidence indicates that TFA does not currently present an environmental hazard but concentrations in precipitation are rising faster than predicted so further research is needed. 

Research on the potential sources of TFA in the environment have revealed that significant amounts originate from the metabolism of anaesthetics and the thermal decomposition of polymers such as the linings of frying pans. The organo-fluorine surfactants that are used as fabric protectors have low volatility but some have appeared in the blood of the human population and in the tissues of animals living in remote regions of the world (Kannan et al., 2001).

Although fluoro-organic pesticides and herbicides are subjected to stringent toxicology testing, little is known about the long-term effects, if any, on soil biota. In general, compounds with trifluoromethyl groups are not defluorinated and aromatic rings are not broken. The long-term fate of these and many other organofluorine compounds is not known (Key et al., 1997).

Chlorine is the major naturally-ocurring halogen in the atmosphere, mostly being present as chloride. The gas chlorine is potentially very hazardous but it is very rare for it to be released in sufficient quantities to pose a risk outside industrial premises. Occasional releases are due to accidental spills or leakage from sources such as chlorine storage tanks, chlorination chemicals used to treat swimming pools, and chlor-alkali plants. Dissipation of leaked or spilled gas depends on the weather at the time of the incident but a report from the USA states that "the affected areas are usually small, often extending less than 0.5 km downwind of the source" (Temple et al., 1998). There have been some examples where vegetation was damaged, one when a valve on a storage tank ruptured and the other due to emissions from an open air swimming pool. The former killed all leaves on willow trees (Salix phyllicifolia) for a distance of about 800m downwind (but the plants survived) and the latter caused brown necrotic spots on plants growing around the pool.

The symptoms of chlorine toxicity are described as; brown necrotic lesions along the edges of the leaves; brown spots scattered along the leaf; upper surface bleaching; epinasty (distorted growth); chlorosis (yellowing) and leaf drop. Laboratory fumigation suggests that symptoms start to develop after exposures above 0.1 ppm for four hours. Concentrations in the 0.4-2.5 ppm range cause severe symptoms (Temple et al., 1998). 

Hydrogen chloride

The atmosphere contains chloride in the form of sea salt aerosols but in most terrestrial environments, chlorine mostly occurs as the acidic gas hydrogen chloride. It arises mostly from emissions from coal-fired power stations and incinerators. Burning of PVC leads to the formation of HCl and there have been occasional leaks from industrial processes. The direct effects of HCl on plants are local but there is little information available about dose-response relations. Temple et al.  (1998) cite work in which exposure to concentrations >3 ppm for several hours caused visible injury. Symptoms of injury are described as various forms of brown necrosis.

The regional patterns of HCl concentration in the atmosphere and of deposition onto soil and freshwater are to some extent similar to those of sulphur dioxide, the main acidifying emission from power stations. However, because HCl is much more water-soluble than SO2, it is deposited in rain within a few tens of kilometres from major sources. A further source of HCl in the atmosphere is as the reaction product of sea salt (NaCl) and nitric acid (HNO3), which derives from NOx emissions. This reaction contributes to a wider regional increase in HCl concentrations. The concentrations of hydrogen chloride in and deposition from the atmosphere are much less that those of SO2.

The highest concentrations and deposition of non-marine chlorine in the United Kingdom occur in eastern England, downwind of coal-burning power stations. The halogen content of coal can vary by a factor of ten or more, depending mainly on the proximity of the mine to the sea during coal formation. Combustion conditions and pollution control equipment may appreciably reduce the emissions of halogens to the atmosphere from coal combustion. Those processes designed to control emissions of SO2 are very effective in reducing emissions of acidic halogen gases (Sloss 1992).

HCl contributes, with the products of NOx, SO2 and NH3 emissions, to acidifying deposition in the UK. However, at a regional level this is only a minor input.

Bromine and iodine emissions are not usually of environmental significance and there is virtually no scientific literature on either element.

For additional information on HCl, see:



Cooke, J.A.; Boulton, I.C.; Johnson, M.S. 1996 Fluoride in Small Mammals. In: Environmental Contaminants in Wildlife: Interpreting Tissue Concentrations Society of Environmental Toxicology and Chemistry Special Publication. Lewis, Boca Raton 22 473-482
Kannan, K.; Koistinen, J.; Beckmen, K.; Evans, T.; Gorzelany, J.F.; Hansen, K.J.; Jones, P.D.; Helle, E.; Nyman, M.; Giesy, J.P. 2001 Accumulation of perfluorooctane sulfonate in marine mammals Environmental Science and Technology 35 1593-1598
Key, B.D.; Howell, R.D.; Criddle, C.S. 1997 Fluorinated organics in the biosphere Environmental Science and Technology 31 2445-2454
NAS, 1971 Fluorides pp 295
Perkins, D.F.; Millar, R.O. 1987 Effects of airborne fluoride emissions in Wales, Part 2 Environmental Pollution 48 185-196.
Weinstein, L.H.; Davison, A.W.; Arndt, U. 1998 Fluoride. In: Recognition of Air Pollutant Injury to Vegetation: A Pictorial Atlas. 2nd Edition pp 4-1 to 4-27
Sloss, L.L. 1992 Halogen emissions from coal combustion IEA Coal Research
Temple, P.J.; Sun, J.E.J.; Krause, G.H.M. 1998 Peroxyacetyl nitrates (PAN) and other minor pollutants Recognition of Air Pollutant Injury to Vegetation: A Pictorial Atlas. 2nd Edition. Air & Waste Management Association, Pittsburgh, PA. pp 6-1 to 6-2
Weinstein, L. H.; Davison, A. W. 2003 Native plant species suitable as bioindicators and biomonitors for airborne fluoride Environmental Pollution 125 3-11
Weinstein, L.H.; Laurence, J.A. 1989 Indigenous and cultivated plants as bioindicators Biologic Markers of Air-Pollution Stress and Damage in Forests 194-204