Llanwern Steelworks

Llanwern Steelworks (formerly the Spencer Works) is located in Llanwern, east of the City of Newport, South Wales. Built by Richard Thomas & Baldwin Ltd, the works was originally referred to locally as “The RTB”, before being called Spencer Works and later Llanwern. It was the first oxygen-blown integrated steelworks in Britain when it opened in 1962.The hot strip mill pioneered the first successful use of a computer for complete mill control.

Steel making at the site ceased in 2001 and the ‘heavy end’ of the works was demolished in 2004. While the site no longer manufactures steel, Tata Steel’s remaining operations at Llanwern include a hot strip mill, a cold strip mill and a hot dip galvanising line. The plant rolls 1.5 million tonnes of steel coil per annum for automotive, construction and general engineering applications.

During production of coke, a concentrated liquid effluent arises which must be treated before the water can be safely discharged or recycled. This effluent contains a range of substances of concern, including ammonia, phenols, tars, mono and polycyclic nitrogen containing compounds, cyanides, sulphides and PAH’s. While many of these substances are amenable to traditional biological treatment process, the complex composition, along with the inclusion of ammonia remains a problem.


Date Type Size Result Challenges
1988 Coke oven effluent, no pre-treatment 180,000m2 | 1,500– 2,500m3 per day Winner of the Prince of Wales Award Cyanides and thiocyanates

In order to meet consent discharge conditions, the Llanwern plant had previously employed an ammonia distillation/incineration process. The process was very energy intensive, with fuel bills approaching £1m per annum, and it also created considerable direct and indirect atmospheric pollution.

Post-distillation effluent in this process still required biological treatment prior to discharge and could not always be guaranteed to comply with consent conditions.

In 1986, Professor Lucian Gill was contacted by British Steel who were looking for a way to dispose of the wastewater effluent produced on site. Prof. Gill oversaw a pilot project which enabled a live feed effluent treatment assessment along with economic comparisons. Following initial phase laboratory trials, two small pilot beds were constructed in 1987. After a reed adaptation period full strength effluent was applied to both beds throughout 1988 and a highly significant level of treatment was achieved, with average treatment rates of 70% – 85% load reduction within the primary system, and meeting full consent after secondary phase treatment.

Construction of the full scale reed beds was completed in 1989, occupying an 18 hectare site of former rough grazing land between the coking plant and the Severn Estuary. In October 1989, the first coking plant effluent was introduced. As flow was increased, the ammonia distillation processes were reduced, generating fuel savings and drastically improving recovery rates due to optimisations and maintenance capacity. The redundant biological treatments systems were redeployed as tar separators, allowing harvested tars to be sold as a valuable by-product.


  • Robust and Effective: The composition of micro-organisms in a reed bed system is very versatile with more than 5,000 different types of bacteria compared with 200 – 300 in other biological treatment plants.
  • Consistent Discharge Quality: Reed Beds tolerate wide variations in effluent characteristics, and can withstand shock loadings, without damage or disruption to treatment, maintaining a consistent discharge quality.
  • No By-Products: Reed bed systems produce no noise or smells and typically no sludge or other by-products removing any associated additional costs of disposal.
  • Versatility: Oceans ESU have knowledge of the decomposition of several thousand compounds in reed bed systems. Reed bed systems can be used to treat effluents which are highly concentrated as well as effluents which are difficult to treat chemically.
  • Long Lasting with Low Operational & Maintenance Costs: Reed beds have little or no electrical or mechanical parts, and are therefore long lasting, wear free and naturally regenerative.
  • Lower Capital Costs than Alternative Conventional Systems: Often, the capital costs of reed bed systems are found to be cheaper than conventional treatment technologies.
  • Environmentally Safe & Beneficial: Reed bed systems are aesthetically pleasing and environmentally beneficial, attracting a range of wildlife and improving biodiversity.

Related Case Studies