English　　Technical-paper

Modelling the Effect of Shear History on
Activated Sludge Flocculation

C. A. Biggs and M. J. Hounslow, University of Sheffield
P.A. Lant, University of Queensland

The aim of this paper is to investigate the effect of shear history on activated sludge
flocculation dynamics and to model the observed relationships using population balances.Activated sludge flocs are exposed to dramatic changes in the shear rate within the treatment process, as they pass through localised high and low mixing intensities within the aeration basin and are cycled through the different unit operations of the treatment process. We will show that shear history is a key factor in determining floc size, and that the floc size varies irreversibly with changes in shear rate. A population balance model of the flocculation process is also introduced and evaluated.

Online analysis of the floc size during activated sludge flocculation was provided using a recently developed experimental technique. To determine the effect of shear history on activated sludge flocculation step changes in the shear rate were conduct-ed. For each experiment, activated sludge was flocculated at G = 19.4s -1 until the fast dynamics of aggregation and breakage were complete. For the first experiment (E1), after 85 minutes, the shear rate was increased to G = 113s -1 . This value was maintained for 35 minutes, after which the shear rate was returned to the original value of G=19.4s-1 . For the second experiment (E2), two consecu-tive step changes were performed cycling between values of G=19.4s-1 to 113s -1 . The results of the two experiments can be seen as a change in floc size with time in Figure 1.

Changing the shear rate during the flocculation experiment confirmed that aggregation and breakage are key mechanisms of activated sludge flocculation. Once the shear rate was increased the floc size decreased since rate of breakage is greater than the rate of aggregation. Conversely, decreasing the shear rate resulted in an increase in the floc size as the rate of aggregation exceeded the rate of breakage.

A step change in shear rate also gives insight into the re-growth behaviour of activated sludge flocs. From Figure 1, it can be seen that final floc size after the shear rate was returned to the original value is less than the final floc size before the shear rate was increased. This has been described as irreversible behaviour and is characteristic of systems flocculated with polymers. At the higher shear rate, particle-flocculent bonds are broken resulting in fragmentation of the flocs and a reduction in floc size. Breakage of the bonds during fragmentation reduces the efficiency of subsequent aggregation since the bonds are not able to reform to the same extent (Spicer et al., 1998). Therefore, a reduction in the final floc size once the original shear has been re-applied results, as demonstrated in Figure 1.

Population balances have been used to successfully model the flocculation of inorganic particulate systems. Since the experimental observations of activated sludge flocculation are comparable to inorganic particulate systems it was proposed that population balances would be suitable to describe activated sludge flocculation. Preliminary simulations of the change in mass mean with time for experiment E1 as predicted by the population balance model developed by Hounslow et al. (1988) and Spicer and Pratsinis (1996) can be seen in Figure 2.

From Figure 2, it can be seen that the model follows the trend in the floc size during the step change in shear rate. The model provides a good approximation of the initial flocculation period and the flocculation period after the shear rate is returned to its original value. However, a discrepancy occurs between the model and experimental results can be seen when predicting the re-growth behaviour after the change in shear rate. The experimental results demonstrate irreversible re-growth behaviour while the floc size predicted by the model is the same before and after the increase in shear rate. This suggests that the kinetics currently used in the population balance model to describe activated sludge flocculation may not be entirely suitable and hence this will be the focus of future work.

Hounslow M. J. et al. (1988). AIChE Journal 34(11): 1821-1832.
Spicer P. T et al. (1998). Powder Technology 97(1): 26-34.
Spicer P. T. and Pratsinis S. E. (1996). AIChE Journal 42(6): 1616-1620.

close

www.wxjxgy.cn