Sustainable treatment of rubber latex processing wastewater : the UASB-system combined with aerobic post-treatment

The main objective of this PhD-thesis is to assess the applicability of UASB-process for treating RLP wastewater and the feasibility of some adequate post-treatment processes for the effluent of the anaerobic treatment process.The studies were carried out in The Netherlands during November 1990-May 1991 and May 1993-October 1993, and in South Vietnam during 1991-1995.The survey was carried out in six RLP factories to determine the quality and quantity of RLP wastewater. In the RLP factory, wastewater originates particularly from two main processing steps, viz. the coagulation step (serum) and the milling, cutting step. The serum wastewater is high in COD-content, viz. up to 5,015-9,962 mg/L, and its pH-values is low with values in the range of 4.95 to 5.24. The quantity of wastewater produced in crushing and milling contributes up to 80-90% of the total amount of wastewater. The COD-content of this wastewater is lower, viz. 1,800-2,400 mg/L, and the pH is higher with values in the range 6.12-6.43. The total quantity of RLP waste amounts to 25-35 m3/ton of dry rubber. The RLP wastewater contains mainly readily biodegradable organic matter (95%), such as volatile fatty acids (acetic/formic acid), and sugars, proteins and lipids, and in addition mineral salts. Considering the presence of these compounds the RLP-wastewater in principle is quite suitable for biological treatment. The raw wastewater also contains uncoagulated rubber particles at concentration up to 350 mg/L. These particles may affect an anaerobic biological process quite adversely.Based on information available concerning the start-up procedure of the up-flow anaerobic sludge bed (UASB) process, four types of potentially attractive seed sludge available in South Vietnam were selected, viz. Fresh pig manure (FPM), digested pig manure sludge (DPMS), city polluted canal sludge (CS) and septic tank sludge (STS), for assessing the specific methanogenic activity (SMA).The SMA of DPMS, CS, STS and FPM amount to 0.13-0.26 g COD/g VSS.d, 0.12-0.15 g COD/g VSS.d, 0.11-0.12 g COD/g VSS.d and negligible values, respectively. The DPMS are strongly recommended to use as seed sludge for full-scale UASB-reactor start-up.Using a SMA test, which is simpler and therefore more adequate for use in developing countries like Vietnam, additional experiments were conducted with the aim to assess the effect of pH, oxygen and mixing conditions. In addition also the effect of the sludge storage time was investigated. The results obtained from these experiments are useful to establish the appropriate procedure for SMA test suitable for Vietnamese conditions.In order to determine the design parameters for pilot and full-scale RLP wastewater treatment plants some lab-scale continuous flow experiments were carried out. From the results obtained in these experiments it appeared that the UASB-process can be applied for RLP wastewater at an organic loading rate up to 28.5 kg COD/m3.d., providing a COD removal efficiency in the range of 79.8-87.9%. The corresponding hydraulic loading rate are up to 7.3-9.1 m3/m3.d at a HRT as low as 3.3-2.6 h. and an upflow liquid velocity up to 0.4 m/h (9.6 m/d). Granular sludge formed in the UASB-reactor after 10-21 days operation. A quite adversely effect of low pH values of the RLP wastewater (i.e. as low as pH = 4.8) was found on the UASB-reactor performance. The period of time passing before the detrimental effect of low pH on the performance of the UASB-system will manifest depends, apart from the pH of the influent and the amount of non acidified organic matter present in the feed solution, particularly on the buffer capacity of sludge bed (i.e. on the accumulated amount of alkalinity in the sludge) and the loading rate imposed to the system. The sludge can recover quite well and sometimes even in a relatively short period of time from an acidic pH-upset, i.e. the UASB-system recovered completely within 2-3 days when using feed pH values in the range of 6.0-6.2.The low influent pH, i.e. 4.5-4.8, and high effluent pH-value that can be reached, viz. in the range of 8.5 to 9.4, may represent true problems in the operation of a UASB-reactor for RLP wastewater, when acid formic is used in the plant for coagulation of latex. These problems can be overcome by using effluent recirculation.Water hyacinth (WH) can accommodate COD concentrations in RLP wastewater up to 2,480 mg/L, and some cases up to 2,900 mg/L and the admissible concentration for algae (ALG) anyhow is lower than 2,280 mg/L, depending significantly on the adaptation procedure applied. Organic surface loads (OL) up to 100 and 120 kg COD/ha.d can be applied for design of WH and ALG ponds respectively. At influent COD concentrations below 300 mg/L effluent COD concentrations will be achieved below 100 mg/L. In order to have oxygen in the effluent, in ALG pond low OL should be used.Laboratory experimental data were used to design a pilot and full scale wastewater treatment plant in Lai Khe (Binh Duong province) and Long Thanh (Dong Nai province) Rubber Latex Processing Factory. The full-scale wastewater treatment plant in Long Thanh has been operated for 4 months, starting in August 1995.