At present, raw aqueous textile effluent produced by textile mills in Hammarsdale, KwaZulu- Natal is reticulated voluntarily to the Hammarsdale Wastewater Works owned by the eThekwini Municipality and operated by Ethekwini Water and Sanitation. Thereafter the treated effluent is discharged into the Sterkspruit River which flows into the Shongweni impoundment. The cost to these textile mills of treating this effluent is calculated using a trade tariff formula administered by Ethekwini Water and Sanitation. In principle this arrangement is governed as follows; • in the case of the textile mills, by Sewage Disposal Bylaws set by Ethekwini Water and Sanitation, and • in the case of Ethekwini Water and Sanitation by a licence issued by the Department of Water Affairs and Forestry in terms of the National Water Act 36 of 1998. This Act prescribes a General Effluent Standard which specifies the quantity, quality and temperature of treated effluent which may enter a defined water resource such as the Sterkspruit River. In practice however, the Hammarsdale Wastewater Works is not licensed by the Department of Water Affairs and Forestry and operates temporarily according to an Exemption Permit issued to Umgeni Water who owned and operated the Hammarsdale Wastewater Works until 2003. It is thus incumbent upon Ethekwini Water and Sanitation to obtain a licence from the Department of Water Affairs and Forestry in order to comply with the requirements of the National Water Act and be allowed to operate the Hammarsdale Wastewater Works on a permanent basis. However, because of design limitations, the Hammarsdale Wastewater Works cannot remove the visible colour continuously and reliably from incoming raw textile effluent. The consequence of this is that the Sterkspruit River is often contaminated by coloured discharges from the Hammarsdale Wastewater Works. In terms of the National Water Act this situation is illegal and must be remedied. The approach adopted by Ethekwini Water and Sanitation has been to amend the Sewage Disposal Bylaws to oblige the textile mills (by the use of permits), by certain dates, to remove all, or most of the colour (to specified levels according to the test method used) from their effluent before it will be admitted to the Hammarsdale Wastewater Works for disposal. The costs of compliance will have to be borne by the individual textile mills. This business case study explores the impact of this obligation on the business of Textile Mill A and examines solutions to the problem. After a review of the efforts of that company to conform with the concept of Cleaner Production, it was decided to perform an end-of-pipe effluent treatment trial using a skid mounted pilot-plant utilising an adsorption and flocculation mechanism followed by cold soda ash softening, 'polishing' through a column of granular activated carbon and the removal of calcium and magnesium through a cation exchange softening column. The results obtained were; • that the permit requirements of Ethekwini Water and Sanitation could be met, • that a financial 'break-even' point could be achieved at the start of the project in 2005 should 43% of the treated effluent be recovered for reuse, • that a realistic rate of water recovery would be 50% resulting in a positive contribution in present day terms (2004) of R65 000 in 2005 increasing to Rl 377 000 in 2014, • that this rate of recovery could be increased should a demineraliser be introduced into the treatment train, and • that the project could be financed by a vendor on a Build, Own, Operate and Transfer basis with transfer of ownership to Textile Mill A occurring after 5 years. The results of that trial showed that end-of-pipe effluent treatment is a viable option, technically and commercially, for Textile Mill A considering the current inclement trading conditions being experienced by the South African textile industry. It is also a means of assisting Ethekwini Water and Sanitation to comply with the requirements of the National Water Act. / Thesis (M.B.A.)-University of KwaZulu-Natal, Pietermaritzburg, 2004.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / O descarte de efluentes tÃxteis em Ãguas superficiais representa um sÃrio problema ambiental e de saÃde pÃblica devido, principalmente, à presenÃa de corantes na sua composiÃÃo, muitos dos quais sÃo potencialmente tÃxicos e carcinogÃnicos. A remoÃÃo de cor desses compostos ainda à uma das maiores dificuldades enfrentadas pelas estaÃÃes de tratamento de efluentes das indÃstrias desse segmento. Dentre os mÃtodos de descoloraÃÃo, o tratamento anaerÃbio tem merecido bastante destaque por ser economicamente atraente. Nesta pesquisa, foram realizados dois experimentos acerca da remoÃÃo de cor de corantes de efluentes tÃxteis. O primeiro objetivou, inicialmente, avaliar e comparar a remoÃÃo de cor de efluente tÃxtil sintÃtico, contendo o corante azo Congo Red (CR), em sistemas anaerÃbios de um estÃgio, composto por um Ãnico reator UASB (R1), e de dois estÃgios (R2), composto por um reator UASB acidogÃnico (R2,A) seguido de um outro metanogÃnico (R2,M). Os reatores foram submetidos a diferentes condiÃÃes operacionais, variando-se as concentraÃÃes iniciais do corante e do substrato doador de elÃtrons (etanol) assim como o tempo de detenÃÃo hidrÃulica (TDH) dos reatores. Constatou-se que, com o aumento gradual da concentraÃÃo de CR de 0,3 para 1,2 mM, a eficiÃncia mÃdia total de remoÃÃo de cor do R1 diminuiu de 97,8 para 95,1%, enquanto a do R2 nÃo apresentou variaÃÃo relevante, sendo o R2,A responsÃvel por grande parte da descoloraÃÃo total alcanÃada (98,5%). Em relaÃÃo à concentraÃÃo inicial de etanol, observou-se queda de menos de 2% na eficiÃncia mÃdia do R1 contra quase 6% na do R2 ao se reduzir a concentraÃÃo do substrato de 1,0 para apenas 0,2 g DQO/L. E, ao se reduzir o TDH total dos sistemas de 24 para 12 horas, as eficiÃncias mÃdias de R1 e R2 passaram de, aproximadamente, 98% para 96,6 e 97,7%, respectivamente. Posteriormente, esses mesmos sistemas ainda foram alimentados com efluente tÃxtil real, e, apesar de ambos terem apresentado eficiÃncias de remoÃÃo de cor menores do que as obtidas com o CR, o R1 atingiu um valor mÃdio 7% maior do que o do R2. O segundo experimento buscou avaliar e comparar a remoÃÃo de cor de efluente tÃxtil real em sistemas anaerÃbios de um estÃgio (TDH = 12 h) suplementados ou nÃo com doador de elÃtrons (etanol) e mediador redox (AQDS). NÃo se observou nenhuma diferenÃa entre os valores mÃdios de eficiÃncia de descoloraÃÃo obtidos pelo reator suplementado com AQDS (R3) e pelo reator livre desse composto (R4). Entretanto, na ausÃncia de etanol, embora ambos os reatores tenham apresentado menores valores de eficiÃncia de remoÃÃo de cor, o R3 obteve uma eficiÃncia mÃdia 5% maior do que a do R4. Concluiu-se que os sistemas anaerÃbios empregados foram capazes de remover a cor de corantes de efluentes sintÃtico e real sob diferentes condiÃÃes operacionais. O impacto do mediador redox nÃo foi evidente no tratamento do efluente real a um TDH de 12 horas, e a ausÃncia de doador de elÃtrons adicional reduziu a eficiÃncia de remoÃÃo de cor dos reatores / The release of textile effluents into superficial water bodies represents a serious environmental problem and a public health concern because lots of dyes from wastewater and their breakdown products are potentially toxic and carcinogenic. Colour removal of dyes is still a challenge for textile industry wastewater treatment plants. Amongst the decolourisation methods, the anaerobic treatment has called attention for being economically attractive. This work reports two textile wastewater colour removal experiments. The first one aimed, primarily, to assess and compare colour removal of synthetic textile effluent, which contained the azo dye Congo Red (CR), in an one-stage anaerobic system, consisted of only a UASB reactor (R1), and in a two-stage anaerobic system (R2), consisted of an acidogenic (R2,A) followed by a methanogenic reactor (R2,M). The reactors were run under different operational conditions by varying the initial dye and electron donor (ethanol) concentrations as well as the hydraulic retention time (HRT) of the reactors. It was observed that, with gradual increase of CR concentration from 0.3 to 1.2 mM, R1 total average colour removal efficiency decreased from 97.8 to 95.1%, while R2 did not present remarkable variation, and R2,A was responsible for the major part of total decolourisation reached (98.5%). Concerning the initial ethanol concentration, a less than 2% drop was observed at R1 average efficiency against an almost 6% one at R2 by reducing the substrate concentration from 1.0 to only 0.2 g COD/L. And, by reducing the systems total HRT from 24 for 12 hours, the R1 and R2 average efficiencies changed from, approximately, 98% to 96.6 and 97.7%, respectively. Afterwards, those same systems were fed with real textile wastewater, and, although both have presented lower colour removal efficiencies than the ones achieved with CR, R1 reached a 7% average value higher than R2. The second experiment aimed to assess and compare the real textile wastewater colour removal in one-stage anaerobic systems (HRT = 12 h) supplemented or not with electron donor (ethanol) and redox mediator (AQDS). No difference was observed between decolourisation efficiency average values achieved by the reactor supplemented with AQDS (R3) and by the reactor free of this compound (R4). However, in the absence of ethanol, both reactors have presented lower colour removal efficiency values, the R3 reached a 5% average efficiency higher than R4 one. It can be concluded that the used anaerobic systems achieved good decolourisation efficiencies with both synthetic and real textile wastewaters under the different operational conditions studied. The redox mediator impact was not evident in the treatment of the real effluent at HRT of 12 hours, and the absence of an external electron donor reduced the reactors colour removal efficiency