N. A. Hunt1 and W. A. Pretorius2

Company/organisation and addresses

1Mondi Kraft Felixton Mill, Private Bax X05, Felixton, 3875
Water Utilisation Division, Department of Chemical Engineering, University of Pretoria, Pretoria, 0002



acidogenic, pre-acidification, anaerobic treatment, pulp and paper wastewater, solids removal


The effect of an acidogenic solids removal reactor on the properties, and biological treatability, of secondary wastewater from a bagasse based integrated pulp and paper mill was tested using acidogenic continuously stirred tank reactor (CSTR) followed by anaerobic/aerobic series treatment. COD and TSS reductions of 34% and 80% respectively, in the acidogenic stage, resulted in effluent that responded well to the anaerobic/aerobic series treatment.  The system produced final effluent in which COD and TSS had been reduced by 93%.  Acidogenic reactor performance was pH dependent with improving performance as pH decreased towards 5.   Addition of molasses was found to be beneficial in supplying carbohydrates for conversion to volatile fatty acids (VFAs). The final effluent was acceptable for reuse in some applications in the mill.


With ever-increasing focus on environmental issues and the development of more stringent laws, such as the new Water Act of 1998, the prospect of minimising water consumption and recovering wastewater for reuse has become more appealing to industry. The mill in question is investigating the possibility of eliminating wastewater discharges, which currently go to sea. To achieving this goal it is necessary to treat a wastewater stream, referred to as Mill wastewater, for reuse in the mill.

Results obtained from a number of preliminary investigations, using chemical and biological treatment methods, and published literature pointed to the use of an acidogenic solids removal stage, followed by anaerobic/aerobic series treatment, as a possible treatment option.

In this work an acidogenic continually stirred tank reactor (CSTR), a methanogenic Upflow Anaerobic Sludge Bed reactor (UASB) and an aerobic activated sludge Sequencing Batch Reactor (SBR) were used to determine the potential of such a system. The final results showed 93% reduction in both COD and TSS resulting in water that could be reused for some applications in the mill.


Mill wastewater referred to in this work is primarily a combination of backwater and reclamation water where:

  • Backwater is excess water drained from pulp during papermaking and has an average pH of about 8.5 due to residual alkalinity from the pulp cooking process.  This stream contains soluble lignin derivatives and fine suspended solids released from the pulp during cooking and pulping.
  • Reclamation water is used to recover bagasse from storage. This stream has a pH of approximately 4, and a population of acidogenic bacteria, due to the acidogenic conditions under which bagasse is preserved during storage.  This stream also contains suspended organic and inorganic material.

The combined stream as used in this work has the average quality parameters shown in Table 1.

Table 1: Properties of mill wastewater stream


Total COD



















Preliminary investigations ruled out the possibility of using conventional biological treatment, chemical precipitation and acid induced lignin precipitation as possible wastewater treatment methods. (Hunt & Pretorius, 1999).


Use of pre-acidification
Pre-acidification treatment makes use of the acidogenic stage of the anaerobic digestion process, by optimising conditions for growth of acidogenic organisms while restricting growth of methanogenic organisms. Since acidogens tend to be far hardier and metabolise faster than methanogens, this is simply achieved by applying a high nutrient loading rate to the reactor.

In single stage anaerobic processes the acidogens are forced to grow at an equivalent rate to the methanogens by restricting the food supply. Under these conditions the methanogens destroy VFAs at the same rate at which the acidogens produce them and the system can be maintained at stable pH levels suitable to the methanogens. Under high load conditions the acidogens thrive causing a drop in pH as the VFA production rate exceeds both the capacity of the methanogens to destroy them, and the buffering capacity of the water. Liu & Ghosh, (1997) showed that, once the pH is below 6.5 the methanogens are inhibited and selected out of the reactor.

The optimum pH for acidogens was reported by Alexiou et al. (1994) to be 6 at a temperature of 37oC, while Speece (1997:94) reported one case in which acidogens were active at pH 3.6 in a starch mill wastewater treatment plant.  However, methanogens tend to be restricted to a pH range of 6.5 to 8 outside of which methanogenesis stops.

Of the many uses reported for pre-acidification the most relevant was the work of Zeeman et al. (1997) who reported successful solids reduction in dairy wastewaters and sewage using acidogenic treatment.  This indicated the use of a pre-acidification stage for solids reduction before conventional biological treatment.

Conventional systems
Both anaerobic and aerobic systems are used to treat wastewater from the pulp and paper industry (Lee et al., (1989). However, a number of workers including Kortikaas et al. (1994), Eroglu et al. (1994) and Dalentoft & Jonsson (1994) have found that anaerobic/aerobic series treatment systems produce better final effluent at lower capital and running costs than anaerobic or aerobic processes on their own.  The anaerobic stage is credited for removing most of the COD while producing smaller quantities of sludge than equivalent aerobic systems, while the aerobic stage is capable of more complete COD removal.  These findings pointed to the use of anaerobic/aerobic series system for further treatment of the wastewater.

Aim and objectives
The aim of this research was to determine if mill wastewater could be biologically treated to a quality acceptable for reuse in the mill. 

The objectives were thus:

  • Evaluate the solids reduction potential of pre-acidification
  • Determine how well the effluent from pre-acidification responded to anaerobic-aerobic biological treatment.


A 2 l reactor placed on a heating magnetic stirrer, and equipped with a 7mm stirrer magnet, served as an acidogenic CSTR.  A thermometer suspended in the liquor was used to monitor the temperature. Mill wastewater was pumped to the reactor using a small variable-rate diaphragm pump.  The temperature was maintained within the wastewater temperature range of 35 to 40oC. A 1 l separating funnel was used as a settler. 

A CSTR was used because of its ability to select for specific organisms (in this case acidogens) and because it is not dependent on the accumulation of sludge. 

Supernatant from the settles overflowed into a 2 l methanogenic reactor half filled with pelletised anaerobic sludge obtained from a brewery wastewater UASB reactor. This reactor was equipped with a recycle pump, to maintain fluidisation of the sludge, and was submerged in a warmed (37oC) water bath.

An UASB was selected because of the availability of the active UASB sludge and because of the technological advantages of this type of reactor pointed out by Lee et al. (1993).

Aerobic treatment was done in a 10 l activated sludge SBR with a 23-hour aeration cycle. This reactor was maintained at about 25oC.   A SBR was used because of simplicity of operation.

The feed was collected daily from the wastewater stream and sieved through a 53m screen to prevent pump blockages.

Molasses was added at a feed rate of 1ml.l 1, which was experimentally determined to increase the VFA concentration by 390 mg.l-1 (as acetic acid), this being sufficient to lower the pH from 6 to about 5.2.  Molasses was used as it can be easily and cheaply sourced from a neighbouring sugar mill.

Effluent from the methanogenic reactor was collected in a beaker for addition to the aerobic reactor. The air supply to the aerobic reactor was cut to allow the sludge to settle for an hour before the supernatant was drained off and fresh feed added.

Sampling and test methods
Samples were collected from the feed tank, settler overflow, methanogenic overflow and aerobic reactor supernatant. The following test methods were used:

1) Measurement of pH was done using a standard pH electrode and meter.

2) Total suspended solids (TSS) was determined using TAPPI test method T 656 (TAPPI Test Methods 1998-1999). 

3) The four point VFA and alkalinity titration method given by Moosbrugger et al. (1992) was used.

4) COD was measured using Hach Photospectrometer Method 8000 (Hach, 1988: 445).

5) MLSS (Mixed liquor suspended solids) was measured as reactor content TSS.

Operational conditions
The average operational conditions for the three reactors used are given in Table 2.

Table 2: Average COD loading rates and MLSS and temperature figures for the three reactors used.






















Correlation between pH and acidogenic reactor performance, with respect to solids removal, is illustrated in Figure 1. From this it can be seen that acidogenic pH is critical to performance with optimum TSS reduction occurring below pH 5.

Figure 1

Figure 1: Acidogenic reactor TSS reduction as a function of  pH

Variation in pH through the system is shown in Figure 2. As the VFA concentration increased from 1385 in the feed to 1840 mg.l-1 in the acidogenic reactor, the alkalinity dropped from 211 to 134 mg.l-1 and the pH dropped from 5.8 to 5.3. The pH in the methanogenic reactor remained within the functional range of methanogens, of 6.5 to 8, the average being pH 7.5. After aerobic treatment the pH rose to 9. This was possible as aerobic organisms are not as pH inhibited as methanogens and were therefore able to digest more VFAs.

Figure 2

Figure 2: pH variation through the system showing averages and 95% confidence level range

Figure 3 shows the average COD readings for each sample point. From a feed COD of 11441 mg.l-1 the COD in the effluent from the settler was reduced to 7781 mg.l-1 due largely to settling out of organic solids.  The COD dropped to an average of 2482 mg.l-1 in the methanogenic reactor due to conversion of VFAs to biogas.  Aerobic treatment reduced the COD to 862 mg.l-1, confirming the claims in the literature that aerobic treatment was required to achieve optimum COD reduction.

Figure 3

Figure 3: COD variation through the system showing averages and 95% confidence level range.

Figure 4 shows the TSS variation in the system.  As would be expected the TSS dropped after settling of the flocs formed in the reactor from 2356 mg.l-1 in the feed to 464 mg.l-1 after settling.  A further drop in TSS to 304 mg.l-1 is noted after methanogenic treatment, due to entrapment and hydrolysis of solids. 

The TSS dropped to 154 mg.l-1 after aerobic treatments thanks largely to entrainment and digestion of solids by the activated sludge.

Figure 4

Figure 4: TSS variation through the system showing averages and 95% confidence level range.


The acidogenic reactor made it possible to run both methanogenic and aerobic treatment systems by reducing TSS levels sufficiently to prevent a build up of inert MLSS fractions in the reactors.  Of particular significance is that TSS was reduced to less than 10% of the COD value as is recommended for UASB reactors by Lettinga & Hulshoff Pol (1991).

Generation of a pH of 5 or less appears to be an important factor in ensuring good coagulation in the acidogenic reactor. Addition of molasses provides a source of carbohydrates that are readily converted to VFAs to help achieve this.

Since laboratory studies using acids to induce lignin precipitation showed that lignin only precipitated at pH 4 or less, it is thought that the coagulation seen in the acidogenic reactor is due to the formation of exo-polymers by the acidogens, and that either the release or coagulating ability of these exo-polymers is affected by pH.

Performance of the overall treatment system in terms of COD and TSS reduction is summarised in Table 3.

Table 3: Percent reductions in COD and TSS based on the average results relative to feed values. Confidence levels (95%) indicated in brackets.





34 (5.06)

80 (4.49)


82 (2.03)

89 (2.43)


93 (0.68)

93 (1.20)

These results show that there is potential for the mill to use such a system for treating its wastewater for reuse.  The final water quality would be acceptable for dilution and pulp washing applications, but some tertiary treatment would be required to lower the TSS levels to the recommended 50 mg.l-1 for spray applications.


1. Alexiou, I.E., Anderson, G.K. and Evison, L.M. (1994) "Design of pre-acidification reactors for the anaerobic treatment of industrial wastewaters", Wat. Sci. Tech. 29(9) pp.199-204.

2. Dalentoft, E. and Jonsson, M. (1994) "Anaerobic-Aerobic Treatment of Total Waste Effluent from a Pulp Mill Including Sludge Hydrolysis", 1994 International Environmental Conference, TAPPI Press, Atlanta, 145-152.

3. Eroglu, V., Ozturk, I., Ubay, G., Demir, I. and Korkurt, E.N. (1994) "Feasibility of Anaerobic Pre-treatment for the Effluentts from Haedboard and Laminated Board Industry", Wat. Sci. Tech. 29 (5-6), 391-397.

4. Hunt, N.A. and Pretorius, W.A. (1999) "Acidogenic treatment for coagulation of wastewater from a Bagasse and wastepaper integrated fluting mill", TAPSA Journal, November 1999.

5. Kortikaas, S., Doma, H.S., Potapenko, S.A., Field, J.A. and Lettinga, G. (1994) "Sequenced anaerobic-aerobic treatment of Hemp Black Liquors", Wat. Sci. Tech. 29 (5-6), 409-419.

6. Lee, J.W., Peterson, D.L. and Stickney, A.R. (1989) "Anaerobic Treatment of Pulp and Paper Mill Wastewaters", TAPPI Environmental Conference Proceedings, TAPPI Press ,Atlanta.

7. Lettinga, G. and Hulshoff Pol, L.W. (1991) "UASB-process design for various types of wastewaters", Wat. Sci. Tech., 24 (8), 87-107.

8. Liu,T., and Ghosh, S. (1997) "Phase separation during anaerobic fermentation of solid substrates in an innovative plug-flow reactor",  Wat. Sci. Tech. 36(6-7), 303-310.

9. Speece, R.E. (1997) Anaerobic Biotechnology, Archae Press. ISBN 0-9650226-0-9.

10. Zeeman, G., Sanders, W.T.M., Wang, K.Y. and Lettinga, G. (1997) "Anaerobic treatment of complex wastewaters and waste activated sludge Application of and upflow anaerobic solids removal (UASR) reactor for the removal and pre-hydrolysis of suspended COD". Wat. Sci. Tech. 35(10), 303-310.


[Home] [Title] [Author] [Organisation] [Keywords]