SRT and F/M ratio

Higher F/M ratio directly causes lower SRT by increasing biosolids production in a given system. Therefore, the observations of SRT effect on OTE can be directly translated to F/M ratio effects on OTE.

SRT has been known as an important factor affecting a (Rosso, 2005, 2007a, 2007b; Leu, 2010). SRT can change a-factor quite dramatically. When SRT increases from 2 days to 20 days, a-factor increases three folds from 0.2 to 0.6 according to Fig. 1. In biological nutrient removal (BNR) process, SRT is naturally high due to the extra tank capacities used for denitrification. As a result a-factor tends to be higher in BNR than CAS (Rosso, 2005; Leu, 2010).

Graphics123

Fig. 1—Effect of SRT on a-factor in full-scale conventional activated sludge (CAS) with fine bubble diffusers. Each dot shape indicates data from one full scale plant. Original graph was plotted against mean cell residence time (MCRT) in aeration basins (excluding the residence time in clarifier), but MCRT equals to SRT in MBR (Rosso, 2005).

F/M ratio can be also correlated with a-factor. The higher the F/M ratio is, the lower the a-factor is. The cause of low a-factor at high F/M ratio or low SRT is not clear, but it can be at least partially attributed to the surfactant-like substrates contained in feed wastewater that can accumulate in the gas-liquid interface of bubble surface and interfere oxygen dissolution. Since the residual surfactant like substrate concentrations are lower at lower organic loading (or higher SRT), a– factor can increase (Stenstrom, 1990). Other explanation is that the high biopolymer level at a high F/M ratio causes lower a-factors. It has been observed that polysaccharides and protein concentrations in mixed liquor tend to increase at high F/M ratio (Drews, 2010). Due to the surfactant-like nature of biopolymers, the high level biopolymers can cause low a-factor.

One may notice there are no significant differences in a between MBR and CAS as shown in figures in MLSS & viscosity effect  and above figure (Fig. 1), although MLSS are very different in the two processes. The causes are not clear, but three theories are available at this point.

  •  Firstly, the data in the figures in MLSS & viscosity effect are mostly obtained with new diffusers while the data in Fig. 1 in this page are obtained with fouled diffusers in full-scale CAS plants. The fouling factor,ff, is often included in a but reported as a instead of af. If new diffusers were used in Fig. 1, the a should be 12.5% higher assuming the average fouling factor of 0.8.
  • Secondly, it can be partially attributed to the substantially different SRT in the two processes. The high SRT of MBR partially compensates the loss in a to some extent.
  • Thirdly, diffuser fouling is accelerated at low SRT as shown in Fig. 2, where specific image010 per water depth decreases much faster at <5 day SRT. Specific image010  decreased ~10% in the first 60 months when SRT is 5-15 days while it decreased ~25% when SRT is less than 5 days. Therefore CAS is affected more by diffuser fouling than MBR.

© Seong Hoon Yoon