LABORATORY PROJECT: IMPACT OF VARIABLE LOADING ON A FIXED FILM REACTOR VERSUS A SUSPENDED GROWTH REACTOR DUE: MAY 6, 2008 PROFESSOR: DR. BELINDA MCSWAIN-STURM TA: ERIN BELLASSAI
1.0 INTRODUCTION 1.1 BACKGROUND Suspended growth activated sludge wastewater treatment plants are commonly used to treat municipal and industrial wastewater. Since the introduction of the activated sludge process, the development of wastewater treatment has been generally held back by tradition (Odegaard, 2000). As regulators continue to strengthen effluent water quality standards, design criteria for suspended growth activated sludge wastewater treatment plants must be updated accordingly. Numerous options, including more contemporary technologies, are available to wastewater treatment entities to improve effluent water quality. Recently, biofilm treatment processes have become increasingly popular compared to traditional activated sludge processes. One of the main reasons for the increased interest in biofilm, according to Odegaard et al. (1994), is that the treatment plant can be made much more compact, requiring less space. Many types of biofilm processes exist today, such as trickling filters, rotating biological contactors fixed media submerged biofilters, etc (Odegaard, 2006). Fixed film media is currently being incorporated into designs for constructing new and upgrading existing suspended growth activated sludge wastewater treatment plants to enhance treatment. The main targets of applying the fixed film media process are increased hydraulic capacity and increased COD/nutrients removal from the effluent (Azimi et al., 2007). 1.2 PURPOSE Industrial and municipal wastewater treatment plants experience variations in loading throughout any given time frame. Analysis of the impact of variable loading of a bench scale fixed film activated sludge reactor versus a bench scale conventional suspended growth reactor is the purpose of this laboratory. Given the increased surface area for microbial growth, we postulate that the fixed film activated sludge reactors will achieve more consistent COD removal results under variable load than the conventional suspended growth reactors. CE773 1 LAB PROJECT
2.0 METHODS Reagents included a premixed wastewater solution and activated sludge biomass provided by the teacher s assistant. Approximately 0.7 liters of activated sludge biomass and 1.3 liters of premixed wastewater solution were added to four 2-liter chemostats (R6, R7, R8, and R9). Two of the 2-liter chemostats (R8 & R9) were filled with 0.8 liters of Siemens AGAR type fixed film media; a fill fraction of 40%, which equates to approximately 0.665 m2 of protected surface area for fixed film growth. Chemostat R8 & R9 model a Integrated Fixed Film Activated Sludge (IFAS) and R6 & R7 model typical activated sludge suspended growth in a completely mixed reactor. Aeration and mixing was provided by compressed air at the lab bench and submersible aerators. Figure No. 1 and Figure No. 2 are photographs of the setup for the chemostats. Figure 1 R6 & R7 CE773 2 LAB PROJECT
Figure 2 R8 & R9 The chemostats began operating on April 15, 2008 at a hydraulic retention time (HRT) of 12 hours. Chemostats R6 & R7 were fed from a 20 liter carboy and chemostats R8 & R9 were fed from a second 20 liter carboy. Each carboy contained premixed wastewater solution and was fed to associated chemostats via peristaltic pumps. The pumping rate was established at 2.8 ml/min to maintain a 12-hour HRT. During the first 5-day period the carboys were refilled daily with premixed wastewater solution with a COD concentration of 500 mg/l. The purpose for allowing the reactors to operate for five days prior to taking readings was to allow the biofilm on the media and the suspended growth to reach steady state. Further, any growth on the wall of Chemostats R6 & R7 was scraped daily. Following the initial five-day period for establishment of steady state suspended growth and biofilm on the media, the feed concentration of the premixed wastewater was increased by 500 mg/l every two days when the carboys were refilled until finally the influent COD was established at 2500 mg/l. Due to odor and foaming, the experiment was cut short by one day, but data was still collected for the 2500 mg/l dose. Figure No. 3 is a photograph of the foaming chemostat. CE773 3 LAB PROJECT
Figure No. 3 Foaming Chemostat After each increase in chemical oxygen demand concentration the chemostats were analyzed for influent and effluent chemical oxygen demand (COD), mixed liquor suspended solids (MLSS), total biomass, dissolved oxygen (D.O.), oxygen uptake rate (OUR) and ph. D.O., OUR, and ph were measured with probes and recorded by each laboratory group. The OUR data was not comprehendible and was excluded from the report; a faulty probe is suspected. Samples to be analyzed for COD and MLSS were collected by each group and delivered to the classroom Teacher s Assistant (TA). COD and MLSS were measured by Standard Method No. 5220 D Closed Reflux, Cholorimetric Method and Standard Method No. 2540 D Total Suspended Solids Dried at 103-105 oc respectively; these Standard Methods were referenced from the Standard Methods for the Examination of Water and Wastewater and were performed by the TA. The total biomass for the fixed film media was calculated by adding the suspended MLSS solids to the solids on 5 media removed from each reactor. The media were weighed before and after heat drying to capture the weight of the CE773 4 LAB PROJECT
fixed biomass. The five media removed from each reactor were replaced with five new media. 3.0 RESULTS Chart No. 1 shows %COD consumed relative to influent COD concentration. As originally postulated, the IFAS reactors (R8 & R9) achieved more consistent results than the conventional suspended growth reactors (R6 & R7). Chart No. 1 - % COD Consumed vs. Influent COD 100 90 80 Legend 6 7 8 9 70 60 50 40 30 20 10 0 0 500 1000 1500 2000 2500 Inffluent COD mg/l Chart No. 2 shows dissolved oxygen (DO) relative to influent COD concentration. The DO did not vary with the exception of the 1,500 mg/l COD concentration. The data points associated with the 1,500 mg/l COD concentration are not likely correct and could potentially be associated with the DO meter being out of CE773 5 LAB PROJECT
calibration. The DO concentration in R6 & R7 does not correlate well to the other data for R6 & R7 and the data for R8 & R9 is well above saturation for the 1,500 mg/l COD data point. Further, the %COD removal performance of R6 & R7 would have significantly deteriorated and then rebounded, but Chart No. 1 shows that they did not. Otherwise, the data was consistently at or above oxygen saturation. Chart No. 2 DO vs. Influent COD g m O D 20 15 10 Legend 6 7 8 9 5 0 0 500 1000 1500 2000 2500 Inffluent COD mg/l Chart No. 3 shows MLSS relative to influent COD concentration. The MLSS measured the suspended solids concentration of the interstitial fluid in R8 & R9 and the suspended biomass in R6 & R7. Visually, the interstitial fluid was clearer, or less murky, in R8 & R9. R6 & R7 exhibited fluid typical of a conventional suspended growth reactor. Refer to photos in Methods section. CE773 6 LAB PROJECT
Chart No. 3 MLSS vs. Influent COD g S S L M 3.00 2.25 1.50 Legend 6 8 9 7 0.75 0.00 0 500 1000 1500 2000 2500 Inffluent COD mg/l Chart No. 4 shows Total Biomass relative to influent COD concentration. The total biomass measured the suspended solids concentration of the interstitial fluid plus the mass of the solids on the fixed media in R8 & R9 and the suspended biomass in R6 & R7. Visually, the interstitial fluid was clearer, or less murky, in R8 & R9. R6 & R7 exhibited fluid typical of a conventional suspended growth reactor. Refer to photos in Methods section. CE773 7 LAB PROJECT
Chart No. 4 Total Biomass vs. Influent COD 20 15 Legend 6 8 9 7 10 5 0 0 500 1000 1500 2000 2500 Inffluent COD mg/l 4.0 DISCUSSION As shown in Chart No. 1, under lower COD concentration R6 & R7 achieved similar results to R8 & R9 and as the influent COD concentration increased R8 & R9 performed significantly better. The better performance of R8 & R9 is likely attributable to the higher biomass concentration sustainable in the IFAS due to increased surface area for microbial growth; additional information regarding the quantity and type of microbes on the media is necessary to verify. Visible growth did occur on the media, but to determine the quantity and type of microbes an additional study where broad DNA analysis of microbes present on the fixed film should be performed. IFAS type reactors typically require more oxygen than conventional suspended growth reactors. Recent IFAS models and research suggest that dissolved oxygen CE773 8 LAB PROJECT
levels are especially critical to nitrification in an IFAS reactor (Takács et al., 2007), and may be of minimal importance to COD removal (Sriwiriyarat, et al., 2008). The dissolved oxygen saturation shown in Chart No. 2 confirms that the microbial activity in R8 & R9 was not limited by oxygen supply. Rather, microbial activity was limited by HRT and influent COD concentration, which was the intent of the experiment. Fixed film media has a propensity to slough older biomass and re-grow new biomass. One suspected caused of such biomass detachment is nutrient starvation (Hunt etal., 2004). The older, sloughed biomass in a typical wastewater plant would be settled in clarifiers and wasted to sludge disposal; the laboratory setup for this experiment did not allow for wastage of sloughed material off the biomass. For this reason, chart No. 3 shows significantly more MLSS in R8 & R9. Chart No. 4 shows that the total biomass in R8 & R9 increased as a response to additional load. The growth on the media likely responds to increased COD concentration, thus increasing the total biomass in the reactor further providing consistent removal of COD. The additional biomass in the IFAS reactors is the key to the more consistent COD removal. With more biomass, more COD can be removed as long as enough oxygen is present to support the microbial growth under sufficiently long HRT s. CE773 9 LAB PROJECT
5.0 CONCLUSION Prior research has shown that biofilm treatment plants may have more efficient COD removal than activated sludge plants at increasingly high COD loads (Odegaard et al., 2000). Andreottola et al. (2000) suggested that minimum values for specific surface area of the media must be met in order to ensure the IFAS is competitive with an activated sludge system. This experiment confirmed that IFAS type reactors (R8 & R9), employing sufficient surface area, achieve more consistently high COD removal than conventional suspended growth type reactors (R6 & R7). More efficient removal of COD is documented in the experiment under conditions of oxygen saturation, as was directly correlated to increasing total biomass in relationship to increasing influent substrate (COD). The next step would be to further explore the microbial mechanics of the IFAS by determining the quantity and variability of microbial populations present on the media. CE773 10 LAB PROJECT
REFERENCES Andreottolla, G., Foladori, M., Ragazzi, M., Tatàno, F. 2000. Experimental comparison between MBBR and activated sludge system for the treatment of municipal wastewater. Wat. Sci. Tech. 41(4), 375-382. APHA, AWWA, WEF. 2005. Standard methods for the examination of water and wastewater. 21 st ed. Washington, D.C. Azimi, A. A., Hooshyari, B., Mehrdadi, N., Nabi Bidhendi, G. 2007. Enhanced COD and nutrient removal efficiency in a hybrid integrated fixed film activated sludge process. Iranian Journal of Science & Technology. 31(B5), 523-533. Hunt, S. M., Werner, E. M, Huang, B., Hamilton, M. A., Stewart, P. S. 2004. Hypothesis for the role of nutrient starvation in biofilm detachment. Appl. Env. Micro. 70(12), 7418-7425. Odegaard, H. 2000. Advanced compact wastewater treatment based on coagulation and moving bed biofilm processes. Wat. Sci. Tech. 42(12), 33-48. Odegaard, H. 2006. Innovations in wastewater treatment: the moving bed biofilm process. Wat. Sci. Tech. 53(9), 17-33. Odegaard, H., Rusten, B., Westrum, T. 1994. A new moving bed biofilm reactor Applications and results. Wat. Sci. Tech. 29(10-11), 157-165. Odegaard, H., Skrovseth, A. F. 1997. An evaluation of performance and process stability of different processes for small wastewater treatment plants. Wat. Sci. Tech. 35(6), 119-127. Sriwiriyarat, T., Ungkurarate, W., Fongsatitkul, P., Chinwetkitvanich, S. 2008. Effects of dissolved oxygen on biological nitrogen removal in integrated fixed film activated sludge (IFAS) wastewater treatment process. Journal of Env. Science and Health Part A. 43, 518-527. Takács, I., Bye, C. M., Chapman, P. L., Fairlamb, P. M., Jones, R. M. 1994. A biofilm model for engineering design. Wat. Sci. Tech. 55(8-9), 329-336.