Fresh water use in paper mills varies according to the type of paper produced, but it can be as high as 100 m3 per ton of paper. In order to reduce water use in paper mills, the reclamation of whitewater is an attractive option. However, it is not always possible to completely recover the whitewater in a closed-cycle manner due to the poor quality of this stream and the possible problems caused by the accumulation of contaminants in the mill. One way to overcome this problem is to remove these contaminants prior to water recycling. The first part of this work compared the efficiency of treatment and the feasibility of whitewater reuse of an anaerobic conventional bioreactor (ACBR) and an anaerobic membrane bioreactor (AMBR), using the whitewater collected from a writing and printing paper mill. In Phase 1, the conventional system operated with a hydraulic retention time (HRT) of 27 h and a chemical oxygen demand (COD) load of 0.24 kg COD m-3 d-1; and in Phase 2, with a HRT of 9 h and a COD load of 1.36 kg COD m-3 d-1. In Phase 1, the AMBR operated with a HRT of 26 h and a COD load of 0.33 kg COD m-3 d-1; and in Phase 2, with a HRT of 9 h and a COD load of 0.85 kg COD m-3 d-1. In the ACBR configuration the removal efficiencies were 66% and 74% for COD, 97% and 83% for total suspended solids (TSS) and 99% and 49% for turbidity, in Phases 1 and 2, respectively. In the AMBR configuration the removal efficiencies were 92% and 79% for COD, 99% and 90% for TSS and 100% and 92% for turbidity, in Phases 1 and 2, respectively. Both configurations showed good removal of total hardness. A slight increase in pH and in electric conductivity was observed. These results indicated that the AMBR has advantages over the conventional anaerobic treatment for reclaiming the treated whitewater, especially due to the high removal of suspended solids and the very low organic content present in the AMBR effluent. However, high values of hardness and electric conductivity in the effluent, not removed by the biological process, might become a drawback for a fully closed whitewater circuit. The second part of this research compared the efficiency of treatment of an ACBR and an AMBR, using the whitewater collected from a recycling paper mill (Old Corrugated Cardboard). The conventional system operated, in Phase 1, with HRT of 7 h and a COD load of 14.8 kg COD m-3 d-1; in Phase 2, with a HRT of 14 h and a COD load of 8.1 kg COD m-3 d-1; and in Phase 3, with a HRT of 24 h and a COD load of 4.7 kg COD m-3 d-1. The AMBR operated, in Phase 1, with a HRT of 8 h and a COD load of 13.4 kg COD m-3 d-1; in Phase 2, with a HRT of 15 h and a COD load of 8.0 kg COD m-3 d-1; and in Phase 3, with a HRT of 23 h and a COD load of 4.8 kg COD m-3 d-1. The anaerobic treatment of the whitewater from OCC paper machine showed high concentrations of volatile organic acids, reflecting in bad removal for COD, turbidity and total hardness. For TSS, the AMBR showed a better performance than the ACBR. The third part of this work compared the efficiency of treatment and the feasibility of whitewater reuse of a mesophilic anaerobic membrane bioreactor (35ºC AMBR) and a thermophilic anaerobic membrane bioreactor (55ºC AMBR), using the whitewater collected from a writing and printing paper mill. In the 35ºC AMBR configuration, the removal efficiencies were 70% for COD, 95% for TSS and 68% for turbidity, operating with a HRT of 10.1 h and a COD load of 1.41 kg COD m-3 d-1. In the 55ºC AMBR configuration, the removal efficiencies were 55% for COD, 97% for TSS and 77% for turbidity, operating with a HRT of 9.8 h and a COD load of 1.46 kg COD m-3 d-1. Both configurations showed slight removal of total hardness and an increase in electric conductivity and real colour. Although COD removal at 55ºC was lower than at 35ºC, the 55ºC AMBR has advantages over the mesophilic anaerobic treatment mainly because it does not need the whitewater cooling. With recent developments in membrane materials, it may be possible to use polymeric membranes at an operating temperature of 55ºC. The major disadvantage associated with using polymeric submerged hollow fiber membranes is that their long-term use at elevated temperatures has not been well documented.
