Kinetics of Removal of Particulate Chemical Oxygen Demand in the Activated-Sludge Process Jose A. Jimenez, Enrique J. La Motta, Denny S. Parker ABSTRACT: The existing theories incorporated to state-of-the-art, activated-sludge-consensus models indicate that the removal of particulate substrate from the liquid in the activated-sludge process is a two-step process: instantaneous enmeshment of particles and hydrolysis followed by oxidation. However, experimental observations indicate that the removal of particles is not instantaneous and needs a more accurate description. This removal process can actually be described as a three-step process: floc- culation, hydrolysis, and oxidation. The principal objective of this research was to observe and model the kinetics of the removal of suspended particles and colloidal particles. A first-order, particulate-removal expression, based on flocculation, accurately described the removal rates for supernatant suspended solids and colloidal chemical oxygen demand (COD). The rate of reaction for removal of colloidal COD was slow and comparable to that for soluble organic matter. Water Environ. Res., 77, 000 (2005). KEYWORDS: activated sludge, particles removal, bioflocculation, aera- tion basin. Background The primary purpose of wastewater treatment is to remove a complex mixture of particulate and soluble substances, both inorganic and organic constituents, that range in size from less than 0.001 microns to well over 100 microns. A major fraction of the organic material in municipal wastewater is in the particulate form—colloidal and suspended solids. These constituents contrib- ute 60 to 70% to the total organic load in domestic wastewater plants (Hunter and Heukelekian, 1965; Levine et al., 1985 and 1991; Rickert and Hunter, 1967). Wastewater characterization is a critical step in activated-sludge- process design and modeling. Total chemical oxygen demand (COD) can be defined as the sum of particulate COD (PCOD) and soluble COD present (SCOD) in the wastewater. For the purpose of this investigation, the PCOD is made up of organic suspended solids (SSCOD) and organic colloids (CCOD) present in the wastewater (PCOD 5 SSCOD 1 CCOD). On the other hand, the dissolved COD is defined as the truly soluble organic material present in the wastewater. Modeling of activated-sludge processes has become a common part of the design and operation of wastewater treatment plants (WWTPs). The International Association on Water Quality (IAWQ) task group on mathematical modeling for design and operation of biological wastewater treatment processes has introduced the activated-sludge models (ASM) No. 1, 2, 2D, and 3. The ASM1 was developed to simulate removal of nitrogen and organic pollutants in activated-sludge processes (Henze et al., 1987). The ASM1 was modified and expanded into ASM2 and ASM2D to cover enhanced biological phosphorus removal (Henze et al., 1995 and 1999). Recently, ASM3 was proposed, advocating biochemical storage as the primary mechanism of substrate use for organic carbon and nitrogen removal (Gujer et al., 1999). This rapid progress in activated-sludge modeling is carried out by introducing new concepts to existing activated-sludge-kinetic models. Nevertheless, some phenomena and their corresponding theoretical descriptions still need to be incorporated to the latest models. Although flocculation seems to be important to constantly produce good effluents, its role in the overall particulate-substrate- removal process has not been fully described. Most researchers agree that flocculation is the process responsible for producing settleable particles that can be separated easily by gravity in the final sedi- mentation tank. However, researchers have overlooked the effect of the kinetics of flocculation on the overall particulate-substrate- removal process and have concentrated their attention on the kinetics of hydrolysis and oxidation when modeling activated-sludge sys- tems. For example, the commonly used International Water Associa- tion (IWA)/IAWQ activated-sludge models take into account a two-step reaction for the removal of slowly biodegradable substrate: rapid enmeshment and hydrolysis of particulate substrate followed by oxidation of soluble biodegradable substrate. These models assume that slowly biodegradable substrate (primarily particulate substrate and colloidal substrate) is considered to be removed from suspension instantaneously by entrapment in the biofloc (Henze et al., 2000; Insel et al., 2002) and is then either degraded or removed with the excess sludge from the system. It may be that the developers of the IWA/IAWQ activated-sludge models did not focus on bioflocculation because of the very high hydraulic retention times (HRTs) required with the design of nitrification and biological nutrient-removal systems, which they emphasized in their modeling. In an attempt to model the contact stabilization process with the general aerobic activated-sludge model, Alexander et al. (1980) identified that a fraction of the particulate COD, which is not adsorbed onto the active organisms, does not become enmeshed in the sludge flocs to escape with the effluent. They found that, at 20 minutes of contact time, only approximately 50% of the PCOD was enmeshed into the floc. Therefore, the enmeshment between the particles and sludge flocs were not instantaneous. The implications of this work have not been generally recognized. For instance, Henze et al. (2002) stated that the particles of the wastewater are quickly absorbed to the activated-sludge flocs, so that a model description of this adsorption will not be necessary, even in a high-rate biochemical oxygen demand process preceding a second-state nitrification step. Also, Sarioglu et al. (2003) did not consider the limits of May/June 2005