WaterSubject

The UK water industry recognises that the Water Framework Directive (WFD) (2000/60/EC) may have a major impact on wastewater treatment throughout Europe. A detailed understanding of the load and sources of priority pollutants entering wastewater treatment works (WwTW) is necessary in order to identify possible reduction measures to meet new legislation, should they be needed. There are few reported studies on the levels of metals entering the WwTW of urban catchments and even fewer for organic determinands. The lack of data on organic compounds may be attributed to the difficulty in their detection in a complex matrix such as crude sewage; however, improvements in limits of detection for organic chemicals in recent years, along with more efficient sample ‘cleanup’ methods, has now made the quantification more accurate. This lack of data prompted the UKWIR to commission a study to identify the presence and range of concentrations of substances on the priority list, consisting of the sampling of crude sewage entering approximately 30 works across England. Sites were selected to provide a good geographical coverage of the country, including a variety of different types of catchment (e.g. combinations of rural, urban, separated, combined, industrial, mixed and domestic, large and small). It provides information on the key characteristics of the chosen sampling sites. Composite samples were collected and analysed for a range of determinands, including all of those present on the WFD priority hazardous substances list, the Dangerous Substances Directive (DSD) (76/464/EEC) List I and within (or proposed to be included in) the European Union Sludge Directive (86/278/EEC).

Sampling

Suitable contacts for each selected works were found and arrangements were made to use any on-site automated samplers capable of taking a composite sample of crude sewage at the inlet. For works with a permanent on-site influent sampler, bottles were couriered to the works and a qualified professional at the works sampled the influent at a predetermined time and date. For works without permanent samplers, or where it was inconvenient to use the on-site one, automated samplers were installed by WRc-NSF (Marlow, Buckinghamshire, UK) in order for sampling to be undertaken.

To perform the entire suite of analysis, a 10 L sample was required. For practical reasons of sampling, handling and transportation, the sampling was carried out in two different ways depending on whether an on-site sampler or a WRc-NSF sampler was used. Most on-site samplers have the capacity to hold only one 5 L bottle at a time; consequently, two 5 L bottles were couriered to the site, and 200 mL samples were collected hourly for 24 h in one bottle, after which a fresh bottle was placed into the sampler for another 24 h. Thus, two lots of 5 L of sample were obtained over a 48 h period. Samples were then couriered back to WRc-NSF, combined, mixed and analysed. For sites without permanent samplers, it was necessary to install a WRc-NSF sampler to the works. WRc-NSF samplers have the capacity to hold a 10 L bottle, and so at these sites a 10 L composite sample was taken over a 24 h period (400 mL hourly).

Flow data were requested for each of the works for the period over which samples were taken, along with dry weather flow details (in some cases, calculated from effluent flow data rather than influent values). Where possible, data for at least hourly intervals over the 24 or 48 h sampling period were obtained. Composite samples provided the opportunity to ensure that representative samples were taken, rather than spot samples, which may have been biased towards ephemeral events. The sampling timetable was designed to provide the best opportunity to obtain samples taken under a mixture of dry and wet weather conditions, so that the potential of inputs of priority substances from runoff (e.g. polynuclear aromatic hydrocarbons (PAHs), pesticides, heavy metals, etc.) could be assessed.

Analysis

Each sample was analysed for 48 determinands (and groups of compounds), which included the WFD list of priority substances, those on the dangerous substances List I (not already on the WFD list) and those listed in the Sludge Directive. Analysis was carried out at WRc-NSF; an outline of the analytical methods is given below.

Metals

Metals (with the exception of mercury) were analysed using aqua regia digestion followed by inductively coupled plasma-mass spectrometry determination (Perkin Elmer Elan 6000 Atomic Spectrometer; Perkin Elmer, Germany). Mercury was determined using cold-vapour atomic absorption spectrometry. All forms of mercury were converted into the inorganic form and reduced to elemental mercury by the addition of acidified tin (II) chloride, before being swept by argon to a gas cell and measured spectrometrically.

Semi-volatiles, phenols and polybrominated diphenyl ethers

The samples were transferred to a separating funnel, and a mixture of labelled internal standards was added. After adjustment of the pH to 2, samples were extracted using dichloromethane, and concentrated. The concentrated extracts were analysed using gas chromatography-mass spectrometry (GCMS) operated in the selected ion-monitoring mode, using positive ion electron impact ionisation. For phenols only, extracts were derivatised prior to GCMS analysis.

Volatile organic chemicals (VOCs)

For VOCs, samples were analysed using purge and trap GCMS. Using an autosampler, an aliquot of each sample was transferred to the purge and trap vessel, purged with nitrogen and condensed onto a cold trap. After flash injection, the purged samples were analysed using GCMS operated in the selected ion-monitoring mode, using positive ion electron impact ionisation.

Polynuclear aromatic hydrocarbons

The samples were transferred onto a separating funnel, and a mixture of labelled internal standards was added. Samples were then extracted using dichloromethane and concentrated, after which they were analysed using GCMS operated in the selected ion monitoring mode using positive ion electron impact ionisation.

Urons

For urons, samples were filtered to remove particulates, and the liquid phase was extracted using an octadecyl (C18)-bonded silica solid-phase extraction (SPE) cartridge (Varian, Palo Alto, California, USA), which was eluted with methanol. The extract was concentrated using a nitrogen blow-down apparatus, and analysed using liquid chromatography mass spectrometry (LCMS) operated in the selected ion monitoring mode using positive ion electrospray ionisation (ESI).

Linear alkylbenzene sulphonates (LAS)

LAS was determined using the standard methylene blue active substances (MBAS) colourimetric method for detecting anionic substances in effluents and waters. Although analysis of anionic substances using MBAS is not substance specific, it does provide a good indication of the overall concentration of anionic surfactants present in the sample. Determination using LCMS would have offered a more accurate measurement of specific isomers; however, the cost of analysis would have been prohibitive and may not have included all of the relevant isomers.