The biorotor system for post-treatment of anaerobically treated domestic sewage

This thesis describes the evaluation of the applicability of biorotor system for post-treatment (polishing) of different effluent qualities of an UASB reactor treating raw domestic sewage, with emphasis on the elimination of various COD fractions, ammonia and E.Coli.The removal mechanism of E.Coli from UASB effluent using a RBC has been investigated. The results obtained revealed that an adsorption process and sedimentation comprise the most important removal mechanism of E.Coli in the biofilm. Die-off is relatively minor importance as removal mechanism in a RBC system.The performance of an anaerobic versus aerobic RBC system treating a high quality UASB reactor effluent was investigated at the same HRT and OLR. The results obtained indicated that the removal efficiency of the COD fractions and of E.Coli fractions found in the aerobic RBC significantly exceeds that of the anaerobic unit. Therefore, the results of our investigations strongly support the use of an aerobic RBC as a post-treatment step of UASB reactor effluents.When applying a single and two stage aerobic RBC at the same OLR of 14.5 g COD total .m -2.d -1and at a HRT of 2.5 h., but at different temperatures of 24 and 17 °C respectively, both systems provided the same residual effluent values for COD total (72 mg l -1), for COD suspended (16 mg l -1), for COD colloidal (5 mg l -1) and for COD soluble (51 mg l -1). Moreover, also the removal efficiency of E.Coli was almost the same, viz. amounting to 94 %. However, the ammonia removal in the single stage RBC amounted to 50 % of which 71 % was nitrified, compared to only 23 % in the two-stage system. This better performance can be attributed to the higher temperature of the wastewater during the operation of the single stage RBC system. In view of these results, we recommend to use a single stage RBC system for COD removal and for a partial nitrification and E.Coli removal at OLR of 14.5 g COD total .m -2.d -1and at HRT of 2.5 h for post-treatment of a high quality UASB reactor effluent.We investigated the use of anoxic reactor followed by a segmental two stage aerobic RBC for nitrogen removal from the nitrified effluent. The results obtained reveal that the introduction of an anoxic reactor as a 1 ststage combined with recirculation of the nitrified effluent of the 2 ndstage RBC is accompanied with a conversion of nitrate into ammonia, at least in case the content of COD biod. in the UASB effluent is low. Therefore, the introduction of a separate anoxic reactor for denitrification as final post-treatment step can not be recommended in such a situation.In one of the experiments the UASB reactor was operated at two different operational temperatures viz. of 30 and 11°C resulting in quite different COD biod. concentrations in the UASB effluent. For the post-treatment of this highly different effluent a single stage RBC was operated at a constant HRT of 1.25 h., consequently at COD biod. loading rates of 17.7 and 36.8 g m -2. d -1. The results clearly show that the residual values of COD fractions and E.Coli are significantly lower at the lower imposed COD biod. loading rate of 17.7g COD biod. m -2. d -1. We also compared the efficiency of the two-stage RBC system for this highly different UASB effluent, viz. once again at the same HRT (2.5 h) and at COD biod. loading rate of 9 and 18 g m -2. d -1. The results reveal that with the two-stage RBC system the residual values of distinguished COD fractions in the final effluent were almost the same, but the residual value of E.Coli in the final effluent amounted to 3.4 x 10 5at the higher COD biod. loading rate and to 7.6 x 10 4/100 ml at the lower one. Moreover, the calculated nitrification rate in the 2 ndstage of two stage RBC system dropped from 1.56 to 1.1 g NO 3 -N.m -2.d -1with an increase the COD biod. loading rate from 11.3 to 16 g m -2.d -1. The results clearly demonstrate that the introduction of a well performing UASB reactor not only improves the nitrification rate but also the E.Coli removal in the post-treatment system.We compared the performance of the single with that of a two-stage RBC for the treatment of poor quality UASB reactor effluent. The results obtained showed that the COD fractions and the E.Coli content in the final effluent of a two stage were lower than in the effluent of the single stage RBC. Moreover, The calculated nitrification rate in the single stage was much lower compared with the two stages RBC. Based on these results we recommend a two stage RBC system for post-treatment of poor quality UASB reactor effluent. The two-stage system was operated at different HRT's and OLR's in order to assess better design criteria for the system. The removal efficiencies for the various COD fractions decreased only slightly when decreasing the HRT from 10 to 2.5 h., and increasing the OLR from 6.45 to 24 g COD total m -2.d -1. However, the overall nitrification efficiency and E.Coli were negatively affected when increasing the loading conditions in the range investigated. The results found for E.Coli removal revealed that the major part of suspended E.Coli (>4.4 µm) was eliminated by sedimentation or by adsorption in the biofilm of the 1 ststage (99.66 %). However, E.Coli present in the colloidal fraction (< 4.4 - > 0.45 µm) was eliminated in the 2 ndstage of two stage RBC (99.78 %). Based on these results we recommend for the treatment of a poor UASB effluent quality the use of two stages RBC system for the removal of COD fractions and ammonia and for a partial removal of E.Coli at HRT of 10 h and OLR of 6.45 g COD total .m -2.d -1.The effluent of two stages still cannot be used for unrestricted irrigation purposes, at least according to the standards provided by the WHO with respect to the E.Coli content. Therefore, in order to meet these (very stringent) standards, we investigated the use of a three stage RBC-system for post-treatment of an effluent from a rather poorly performing UASB reactor. This three stage RBC was first operated at a HRT of 3.0 h. Under these conditions the E.Coli count in the final effluent was still too high. However, when applying an HRT of 10 h., the E.Coli content complied almost the WHO standards for unrestricted irrigation purposes. Therefore, when such high removal efficiency for E.Coli really would be required, the best solution is to use three independent stages at HRT of 10 h., which then obviously implies very significantly investment and operational costs.