Xylene (C₆H₄(CH₃)₂) occurs in three isomers (o-, m- and p-) which vary in the site of attachment on the benzene ring of the two methyl groups. They are liquids at room temperature and pressure, and are soluble in water (134 - 230 mg l^-1). They have moderate log Kow values (2.77 to 3.20) and tend to volatilise.

Production of xylenes is primarily associated with the petrochemical and coal industries, with most being produced by the catalytic reforming of naphtha, a derivative of crude oil fractionation. Most (c. 89%) xylene is produced as a mixture of isomers, along with benzene and toluene, with subsequent isolation of xylenes as required. Technical (mixed) xylene contains typical proportions of o-, m- and p- isomers of 20-24%, 42-48% and 16-20% respectively, with 10-11% ethylbenzene as an >impurity=.

Xylene is produced widely in the EC at high tonnages (total production in the range 500 to 1,000 Ktonnes, according to the EU's IUCLID database). In the UK, production of mixed xylenes is significant, but of the isomers, only p-xylene is produced in isolation in substantial quantities. Estimated annual production capacities in the UK for the early 1990s were 150 Ktonnes mixed xylenes and 200 Ktonnes p-xylene, with major production at only two sites (ChemInform 1992).

Most xylenes (>90%) are used in mixed xylene as a solvent and a constituent (BTX - benzene-toluene-xylene) of vehicle, aviation and other fuels. This latter use has increased significantly in the EC with the widespread introduction of unleaded petrols (Crookes et al 1993). Globally, the increasing use of vehicular transport in developing nations is also increasing the use of BTX and, thus, xylenes. The industrial importance of individual isomers decreases in the order p-, o-, m-, with major uses including the following (Micromedex 1996):

During their production and use, xylenes are released primarily to the atmosphere. The principal anthropogenic releases of xylene remain consistent, i.e.:

• accidental and deliberate release of crude oils and petrochemical products (including during refining of crude oils and distribution and use of products);
• production of xylenes and derivative chemicals; and industrial and domestic discharge of solvents and other products.).

Entry into water may be direct or via atmospheric deposition, runoff and leaching.

Xylenes have been regularly reported in the oceans, estuaries, precipitation, rivers, groundwaters, potable sources and drinking water, as well as aquatic sediments and biota (Hedgecott 1990, Crookes et al. 1993). High usage and release, high mobility in the atmosphere, and natural sources all contribute to widespread occurrence in waters. Post-1988 data for UK waters were summarised in Hedgecott and Lewis (1997). Most of the data are for estuarine and coastal waters, with xylenes apparently detected in relation to industrial releases of xylenes and releases during oil extraction and transport. None of the published values exceed the concentration of 30 mg l^-1 of total xylenes which has been proposed as the annual average EQS value.

Hedgecott and Lewis (1997) reviewed the fate and behaviour of xylenes.

Atmospheric xylenes are subject primarily to photo-enhanced oxidation by reaction with hydroxyl radicals, and this (and other reaction processes) is enhanced by nitric oxides and solids. Half-lives in the atmosphere have been variously estimated between 0.83 and 29 hours. Oxidation in water is considerably slower with estimated half-lives between 30 and 300 days. Hydrolysis is unlikely (Hedgecott and Lewis 1997).

Xylene molecules are relatively simple and biodegradation is widespread in environmental media, although o-xylene appears to be slightly more recalcitrant than the other isomers. Groundwater inocula have been reported to completely degrade low concentrations of xylenes in 2 to 20 days (varying with pre-exposure) under aerobic conditions (when dissolved oxygen levels are not limiting). Although anaerobic degradation has also been observed, this is somewhat slower. Typically there is a substantial lag period of around 30 (m- and p-) or >140 (o-) days before significant degradation in unacclimated aquifer material, but with pre-exposure and adaptation, degradation can be significant (>80%) after an additional 26 (m- and p-) to 100 (o-) days, although it may take longer (Hedgecott and Lewis 1997).

A moderate tendency to sorb to organic solids is suggested by the log Kow values of 2.77 to 3.20 and log Koc values of 2.1 to 2.5. Xylenes may sorb to aquatic sediments but higher proportions remain in solution; low sediment-water partition coefficients of 8.9 for o-xylene and 10.5 for p-xylene have been measured for the Tamar Estuary (Hedgecott and Lewis 1997).

Xylenes readily volatilise from water and this is probably the major single removal process in most surface waters, with a half-life of a few to tens of hours, depending on the degree of mixing (Hedgecott and Lewis 1997).

An exhaustive literature review on the toxicity of xylenes to marine organisms has not been carried out for the purposes of this profile. The information provided in this section is taken from existing review documents (Hedgecott and Lewis 1997). The most sensitive groups of organisms have been identified.

Hedgecott and Lewis (1997) reviewed data on the aquatic toxicity of xylenes. A previous review (Hedgecott 1990) found saltwater data were limited to acute studies only, with the most sensitive of those species tested being the bay shrimp Crago franciscorum with 96 hour LC50s of 1.1, 3.2 and 1.7 mg l^-1 for o-, m- and p-xylene, respectively (mean 2.0 mg l^-1). The most sensitive fish was the striped bass Morone saxatilis, with 96 hour LC50 values of 9.6, 7.9 and 1.7 mg l^-1 for o-, m- and p-xylene respectively. These invertebrate and fish results are very similar to the lowest ones for comparable freshwater species.

Hedgecott and Lewis (1997) found few data had become available since the previous review (Hedgecott 1990), and again relate only to acute exposure. The lowest effect concentration is an EC50 for the Microtox bioassay (using Vibrio fischeri) of 8.5 mg l^-1 mixed xylene (Calleja et al. 1994). This is similar to a 15 minute Microtox EC50 of 9.2 mg l^-1 determined previously for p-xylene (Hedgecott 1990), and does not indicate greater sensitivity than previously determined for this or other saltwater species.

No data could be located for sediment-dwelling organisms.

Data summarised in (Hedgecott 1990) from fresh and saltwater studies indicated that bioaccumulation of xylenes was not significant, with BCF values ranging from 1 to 15 in freshwater fish and 1 to 24 in saltwater fish and invertebrates, and uptake and depuration both occurring rapidly. Hedgecott and Lewis (1997) found few additional bioaccumulation data had become available since the 1990 review; reported BCFs from a few freshwater test with algae are in excess of 200 but are based on dry algal weight (Herman et al. 1991b). Thus, the new data do not indicate higher bioaccumulation potential than that indicated previously.

Xylenes in marine fish in Japan have previously been implicated in tainting problems (Hedgecott 1990). Jardine and Hrudey (1988) determined a taste tainting threshold of 9 mg kg^-1 for p-xylene spiked into a freshwater fish (the walleye, Stizostedium vitreum). Assuming a fish BCF of 15, this implies that a water concentration of 0.6 mg l^-1 or above might lead to detectable tainting of edible fish.

Potential effects include:

• toxicity of xylenes to invertebrates and fish at concentrations above the proposed EQS of 30 mg l^-1 (annual average) of total xylenes in the water column.

Next Section                     References