Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Enhanced biological phosphorus removal using a sequencing batch RBC Simm, Robert 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A7 S55.pdf [ 5.32MB ]
Metadata
JSON: 831-1.0062691.json
JSON-LD: 831-1.0062691-ld.json
RDF/XML (Pretty): 831-1.0062691-rdf.xml
RDF/JSON: 831-1.0062691-rdf.json
Turtle: 831-1.0062691-turtle.txt
N-Triples: 831-1.0062691-rdf-ntriples.txt
Original Record: 831-1.0062691-source.json
Full Text
831-1.0062691-fulltext.txt
Citation
831-1.0062691.ris

Full Text

E N H A N C E D B I O L O G I C A L P H O S P H O R U S R E M O V A L U S I N G A S E Q U E N C I N G B A T C H R B C By Robert Simm B.A.Sc (Civil Engineering) University of British Columbia A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES CIVIL ENGINEERING We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1988 © Robert Simm, 1988 In presenting this thesis i n p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the L i b r a r y shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C i v i l Engineering The University of B r i t i s h Columbia 1956 M a i n M a l l Vancouver, Canada Date: Abstract The objective of the research program was to demonstrate the technical feasibility of removing phosphorus, by the enhanced biological phosphorus removal mechanism, from domestic wastewater using a laboratory scale Sequencing Batch Rotating Biological Contactor (SBRBC). The rotating discs of the RBC were subjected to alternating anaerobic/aerobic conditions by varying the water level in the reaction vessel. At the start of the treatment cycle, the RBC reactor would be filled submerging the rotating discs and ensuring anaerobic conditions in the RBC biofilm. Acetate would be added to the reaction vessel at this time. Following the batch anaerobic react period part of the reactor contents were decanted to either the sewage feed tank or a separate holding vessel to later become part of the influent for the next treatment cycle. With the rotating: discs of the-RBC partially submerged oxygen was available to the bacteria, in the RBC biofilm. Three operating schedules were tried with the above process. Each operating sched-ule differed in the way the decanted wastewater from the anaerobic phase was handled. Batch tests were conducted weekly to determine the nature of the biological reac-tions taking place in each of the batch anaerobic and aerobic phases. The SBRBC process showed promise for enhanced biological phosphorus removal from domestic wastewater. Carbon removal and nitrification of the wastewater were secondary benefits to this process. The success of the process was found to be dependent on the attainment of proper anaerobic conditions at the start of each treatment cycle. ii Table of Contents Abstract ii List of Tables vi List of Figures vii Acknowledgements ix 1 Introduction 1 2 Literature Review 3 2.1 Excess Biological Phosphorus Removal 3 2.2 Sequencing Batch Reactors 11 2.3 Rotating Biological Contactors (RBC's) 13 2.3.1 R B C Biofilm Studies 14 2.3.2 Nutrient Removal using R B C processes 15 3 Methodology 24 3.1 The Experimental Unit 24 3.1.1 System Rationale 24 3.1.2 Reactor and Rotating Discs 28 3.1.3 Disc Drive Unit 29 3.1.4 Raw Sewage Reservoir and Anaerobic Holding Vessel 32 3.1.5 Pumps 32 . > iii 3.1.6 Timer/Microprocessor and Solenoid Valves 33 3.1.7 Miscellaneous 33 3.2 System Operation 33 3.2.1 Disc Rotational Speed 34 3.2.2 Length of Reaction Phases 35 3.2.3 Acetate Addition 35 3.2.4 Flow Management 35 3.3 Seeding 36 3.4 Sewage Characteristics 37 3.5 Sampling Procedure 37 3.6 Analytical Techniques 38 3.6.1 Nonfilterable Residue (Suspended Solids) 38 3.6.2 Nitrogen (N) 38 3.6.3 Oxidation-Reduction Potential (ORP) 39 3.6.4 Phosphorus 39 3.6.5 Biofilm Phosphorus Content 39 3.6.6 Volatile Fatty Acids (VFA's) 40 3.6.7 Chemical Oxygen Demand (COD) 40 3.6.8 Disc Solids 40 4 Results and Discussion 42 4.1 Results 42 4.1.1 Phase One (September 1,1987- February 1, 1988) 42 4.1.2 Phase Two ( February 6,1988 - February 16, 1988) 58 4.1.3 Phase Three (February 17-March 17, 1988) 71 4.2 Data Analysis 94 iv 4.3 Discussion 97 5 Conclusions and Recommendations 105 5.1 Conclusions 105 5.2 Recommendations 106 Bibliography 108 A Statistical Analysis 116 v List of Tables 3.1 Operating Schedule 27 3.2 Typical Characteristics of UBC bio-P plant Influent Sewage . . 37 4.1 Average Characteristics and Ranges for Nov. ll/87-Feb.l/88 53 4.2 Biofilm percent phosphorus contents for Phase One 54 4.3 Reactor Feed Composition Phase Two (Feb.6/88-Feb.ll/88) 60 4.4 Effluent Composition Phase Two (Feb.6/88-Feb.ll/88) 61 4.5 Reactor Feed Composition Phase Two (Feb.l2/88-Feb.l6/88) 65 4.6 Effluent Composition Phase Two (Feb.l2/88-Feb.l6/88) 66 4.7 Phase Three Raw Sewage Data (Feb.l9/88-Mar.l7/88) 85 4.8 Phase Three Anaerobic Holding: Vessel Data (Feb.l9/88-Mar.l7/88) . . 86 4.9 Phase Three Effluent Data (Feb.20/88-Mar.l7/88) . . . . . . . . . . . . . . 87 4.10 Biofilm percent phosphorus contents for experimental period 93 vi List of Figures 3.1 Process Flow Sheet 26 3.2 Reactor Assembly 30 3.3 Floating Covers 31 4.1 Orthophosphate and NOx versus Time for Sept. 8/87 Batch test . . . . 44 4.2 soluble COD versus time for Sept. 8/87 Batch test 45 4.3 Orthophosphate and NOx versus time for Oct. 5/87 batch test 46 4.4 Soluble COD versus time for Oct. 5/87 batch test 47 4.5 Bulk liquid ORP versus time for Oct. 5/87 batch test 48 4.6 Phase One Percent P removal versus time (Nov. 11/87- Feb. 1/88 . . . . 50 4.7 Phase One Percent COD removal versus time (Nov. 11/87 - Feb. 1/88 . 51 4.8 Phase One Percent Nitrification versus time (Nov. 11/87-Feb. 1/88 . . . 52 4.9 Orthophosphate and Acetate versus time for Jan. 6/88 Batch Test . . . 55 4.10 NOx versus time for Jan. 6/88 Batch Test . 56 4.11 Bulk liquid ORP versus time for Jan. 6/88 Batch Test 57 4.12 Phase Two TP concentration in feed tank vs time from Feb. 6/88-Feb. 11/88 . . . . . 62 4.13 Phase Two Effluent orthophosphate versus time from Feb. 6/88-Feb. 11/88 63 4.14 Phase Two Feed tank soluble COD versus time from Feb. 12/88-Feb. 16/88 67 4.15 Phase Two Feed TP and ortho-P versus time from Feb. 12/88-Feb. 16/88 68 vii 4.16 Phase Two Effluent Orthophosphate versus time from Feb. 12/88-Feb. 16/88 69 4.17 Phase Two Percent P removal versus time from Feb. 12/88-Feb. 16/88 . 70 4.18 Orthophosphate and Acetate versus time for Feb. 24/88 Batch Test . . 72 4.19 NOx concentration versus Time for Feb. 24/88 Batch Test 73 4.20 Bulk liquid ORP versus time for Feb. 24/88 Batch Test 74 4.21 Orthophosphate and Acetate versus Time for March 2/88 Batch Test . 75 4.22 NOx concentration versus time for March 2/88 Batch Test 76 4.23 Bulk liquid ORP versus time for March 2/88 Batch Test 77 4.24 Acetate and orthophosphate versus time for March 9/88 Batch test . . . 79 4.25 NOx versus Time for March 9/88 Batch Test . 80 4.26 Bulk liquid ORP versus Time for March 9/88 Batch Test . 81 4.27 Acetate and orthophosphate versus Time for March 16/88 Batch Test . 82 4.28 NOx versus- Time for March 16/88 Batch Test 83 4.29 Bulk liquid ORP versus Time for March 16/88 Batch Test 84 4.30 Phase Three Anaerobic Holding Vessel and Raw Ortho-P versus time from Feb. 28/88 to Mar. 17/88 . 88 4.31 Phase Three Influent and Effluent Ortho-P versus time from Feb. 19/88 to Mar. 17/88 89 4.32 Phase Three Influent and Raw soluble COD versus time from Feb. 19/88 to Mar. 17/88 . 90 4.33 Phase Three Percent P removal versus time from Feb. 19/88 to Mar. 17/88 91 4.34 Anaerobic P release vs Aerobic P uptake data for Nov. 11/88 to Mar. 17/88 95 viii Acknowledgements I wish to express my gratitude to my thesis advisor Dr. W.K. Oldham, Civil Engineer-ing Department Head, for the assistance and encouragement he provided me. I would also like to thank the following for their much appreciated support. - Craig Peddie, Research Engineer, for his encouragement and suggestions during all phases of the research project. - Susan Liptak, U.B.C. Environmental Engmeering Laboratory Manager, and Paula Parkinson, Laboratory Technician, for their assistance and guidance with all laboratory analysis. - Guy Kirsch, Civil Engineering Workshop Technician, for his suggestions and as-sistance in manufacturing the laboratory reactor. - Janis Chang, U.B.C.Statistics Department Graduate Student, for her support, encouragement, and assistance with the statistical analysis. I would also like to thank Romy So, Fred Koch, Nelson Lee, R. Manoharan, and Yves Comeau for all the helpful suggestions they provided. ix Chapter 1 Introduction Population growth, intensified agriculture, and industrial development have increased the nutrient loading on many waterbodies, resulting in an acceleration of the eutrophi-cation process. Control of run-off, better agricultural practises, and removal of nutrients from point source discharges have been practised to rectify the problem. Many experts agree that the removal of the nutrient phosphorus from wastewaters is the key to the treatment of point source discharges. The other major nutrients, car-bon and nitrogen, are often readily available in the environment. Carbon is naturally abundant in organic compounds and the bicarbonate ion. The required forms of nitro-gen, nitrate and ammonia, can be brought into a system by nitrogen fixing organisms like blue-green algae. The only natural source of phosphorus is from the earth's crust. The majority of domestic wastewater plants removing phosphorus do so using chem-ical precipitation with alum, lime or ferric salts. In the last twenty years a technology referred to as 'excess biological phosphorus removal' has emerged as a viable alternative to chemical precipitation. Biological phosphorus removal is induced when a culture of 'acclimatized' micro-organisms is subjected to alternating anaerobic (no oxygen) and aerobic (aerated) con-ditions. Under non-aerated conditions, the bacteria release phosphorus into the bulk liquid, and store available short chain fatty acids. Under aerated conditions, the bac-teria will use the carbon stored in the anaerobic phase to trigger storage of phosphorus in excess of metabolic requirements. To date the above technology has been practised 1 Chapter 1. Introduction 2 using modifications of the activated sludge process. Due to the recent advances in the understanding of the biochemical mechanisms responsible for biological phosphorus removal, the application of processes other than activated sludge now appears feasible. It was the purpose of this research to demonstrate the ability of a rotating biological contactor to remove phosphorus by the biological phosphorus removal mechanism, when operated in a sequencing batch mode. The literature review discusses those articles which supplied the necessary informa-tion for the design of the experimental unit. The methodology section describes the methods and analytical techniques followed. All findings are presented in the results and discussion sections, and conclusions and recommendations for further research are proposed. Chapter 2 Literature Review This research deals with the combined application of enhanced biological phosphorus removal, Sequencing Batch Reactor (SBR), and Rotating Biological Contactor (RBC) technologies. A brief overview of some of the available literature covering these three areas is presented in sections 2.1, 2.2, and 2.3. 2.1 Excess Biological Phosphorus Removal Comprehensive literature reviews of excess biological phosphorus removal have been presented by Comeau (1984) and others. Only a brief review of biological phosphorus removal research is presented here. Special attention is given to recent research which discusses the biochemical model for biological phosphorus removal. Excess biological phosphorus removal was first reported by Srinath et al. (1959) in India. No explanation for why the phenomenon occurred was reported. The nature of the above phenomenon was investigated by Levin and Shapiro (1965) using mixed liquor from the District of Columbia sewage treatment plant. The authors concluded that the prospects of achieving reduction of dissolved inorganic orthophos-phate from sewage, using a modified activated sludge process, were promising. In the late sixties and early seventies, a number of activated sludge plants in the United States were removing phosphorus in excess of metabolic requirements. A num-ber of researchers, such as Vacker (1967), Wells (1969), and Milbury (1971) studied the above plants. The exact reasons for the observation of excess phosphorus removal were 3 Chapter 2. Literature Review 4 not given. The mechanism appeared to be biological in nature. Barnard (1974) reported on an activated sludge process referred to as the 'Barden-pho' process. The process was tested for 18 months in a pilot plant at the Daspoort Sewage Treatment Works in Pretoria, South. Africa. The plant consisted of four com-pletely mixed activated sludge basins in series followed by a clarifier from which sludge was returned to the first basin. The first and third basins were only stirred to keep the solids in suspension, while the second and fourth basins were aerated. Complete nitrification was achieved in the second basin, and mixed liquor from that basin was returned to the first basin. The author reported nitrogen removals of 90-95%, and phosphorus removals of 97%. Low concentrations of nitrate in the effluent coincided with good phosphorus removal. Barnard postulated that it was essential for the mixed liquor to pass through an anaerobic period during some stage of the process followed by a well aerated stage to ensure good phosphate removal. In (1975) Fuhs and Chen isolated a number of bacterial species from activated sludge plants. The two researchers attributed excess phosphorus storage to micro-organisms of the Acinetobacter genus. These organisms were found to accumulate poly-/d-hydroxybutyrate (a carbonaceous reserve material) and poly-phosphate ( a phosphate storage material). They hypothesized that the poIy-/?-hydroxybutyrate might serve as a storage of energy for excess phosphorus uptake. It was further discovered by the two researchers that the Acinetobacter organisms, isolated from enhanced biological phosphorus removal plants, were unable to use glucose as a substrate. They required short chain fatty acids such as acetate, and succinate, to proliferate. The authors concluded that the purpose of the anaerobic-aerobic sequence was to allow the establishment of a facultatively anaerobic microflora to produce the short chain fatty acids required by the Acinetobacter as a carbon source. Chapter 2. Literature Review 5 Barnard (1976) reviewed biological phosphorus removal i n the activated sludge pro-cess. U sing the results from his own experiments, as well as the published results of other researchers, he concluded that the mixed liquor or return sludge i n an activated sludge plant, to successfully remove phosphorus, must go through a period of anaer-obiosis during which phosphates are released, followed by an aerobic period i n which phosphates are taken up. He further concluded that the presence of nitrates i n the anaerobic zone would raise the oxidation-reduction potential above the minimnm value required f o r anaerobic conditions. Barnard's hypothesis was used to explain the find-ings of Vacker et a l . , M i l b u r y et a l . , and Garber. He then suggested a modification of his Bardenpho process: which included an anaerobic zone at the head end receiving both the sludge recycle and the influent stream. The new process was referred to as the M o d i f i e d Bardenpho Process or Phoredox process. Simpkins and M c L a r e n (1978) d i d p i l o t scale experiments w i t h the Phoredox pro-cess. The authors concluded that the physical-chemical precipitation of phosphorus was small compared to the removal by biological uptake, confirming the findings of other researchers. Simpkins and M c L a r e n found the anaerobic release of phosphate to be i n direct proportion w i t h the C O D input. R a p i d 'absorption' of soluble C O D was also observed i n the anaerobic zone. T h e authors also found that the denitrification rates i n the secondary anoxic zone were much lower than those i n the primary anoxic basin. It was suggested that the secondary anoxic and re-aeration zones be eliminated from the Phoredox process and the primary anoxic zone be enlarged. In (1979) Nicholls and Osborn examined sludges from the Olifantsvlei Works and Alexandra sewage treatment plants i n Johannesburg, South A f r i c a . B o t h plants were removing phosphorus i n excess of metabolic requirements. Poly-phosphate (poly-P) and poly-/?-hydroxybutyrate (PHB) were found to be present Chapter 2. Literature Review 6 i n abundance i n the above sludges. Nicholls and Osborn suggested that P H B and poly-P may play a mutually interdependent role i n assisting aerobic bacteria to survive a period of anaerobic stress. It was hypothesized that under anaerobic conditions, substrate i n the wastewater could be metabolized to acetyl-CoA and electrons and protons by aerobic bacteria. However, due to the absence of a terminal electron acceptor (oxygen) the electrons and protons could not be used to generate A T P v i a the electron transport chain. Acetyl Co A , however, would act as an electron and proton sink by its reduction to P H B , and as an energy source for A T P synthesis v i a deactivation to acetic a c i d . Poly-P was proposed to serve as a source of P for the formation of A T P above. The A T P generated would then be u t i l i z e d for cell maintenance v i a hydrolysis to A D P + phosphate ( P i ) . Under aerobic conditions, the hydrogen ions stored temporarily as P H B could pass back through acetyl-CoA as an intermediary and enter the normal Krebs cycle once more. P H B was thought to serve not only as an. energy source for cell function but also for p o l y - P generation and storage. U s i n g t h e nitrincation-denitrincation model developed at the University of Cape Town, Rabinowitz a n d Marais (1980) denned a parameter known as the anaerobic p o t e n t i a l . T h e anaerobic potential was denned as the difference between the denitrifi-cation capacity of the anaerobic reactor and the mass of nitrate entering the reactor. T h e two researchers suggested a modification of the Phoredox process to eliminate n i -trate f r o m the anaerobic zone. As suggested by Simpkins and M c L a r e n , the secondary anoxic and re-aeration zones were eliminated. The sludge recycle and mixed liquor f r o m the aerobic zone would be discharged to the anoxic basin. A n additional recycle would be added to the anoxic basin to transfer mixed liquor to the anaerobic zone. B y adjusting the recycle ratio from the aerobic to the anoxic zone, the nitrate concen-tr a t i o n i n the anoxic, and therefore the anaerobic zone, could be reduced. The above Chapter 2. Literature Review 7 process became known as the U C T process. Marias et a l . (1983) traced the development of biological excess phosphorus removal from its discovery i n 1959 up u n t i l 1983. The biochemical models of Funs and Chen (1975) and Nicholls and Osborn (1979) were reviewed. The authors hypothesized that poly-P accumulation serves as an energy reservoir, to sustain the organisms during the anaerobic stressed state, but more importantly to gain an advantage over non-P accumulating organisms by partitioning of readily biodegradible C O D ( i n the lower fatty acid form) i n the anaerobic state for its exclusive use subsequently i n the aerobic state. The authors hypothesized that during anaerobic conditions the internal A T P / A D P ratio of poly-P organisms would be low. It was also suggested that the above ratio may act as a. feedback mechanism activating ployphosphatase to catalyze the breakdown of poly-P to P i and energy for A T P production. Energy from A T P would then be used to b r i n g substrate into the cell and convert i t to a l i p i d form available for storage as aceto-acetate and/or poly-/?-hydroxybutyrate. The fate o f b oth short chain fatty acids, e. g. acetate, and a relatively 'high energy' substrate, which can b e used by facultative and aerobic organisms to yield energy through anaerobic glycolysis, was also considered. It was hypothesized that acetate entering the cell d u r i n g the anaerobic period is complexed by the enzyme Co A to form acetyl C o A , which requires the input of two A T P . P i and energy required for A T P production c o u l d be derived from the breakdown of the poly-P chain. Complexation of the acetate: to acetyl C o A would reduce the acetate concentration w i t h i n the organism, thus allowing an osmotic pressure to be created for further entry of substrate into the c e l l . Due to the l i m i t e d supply of C o A i n the c e l l , storage as acetoacetate or P H B was considered likely. The high concentration of P i i n the c e l l , resulting from poly-P breakdown, would lead to P i diffusion out of the c e l l . Chapter 2. Literature Review 8 Two possibilities f o r glucose u t i l i z a t i o n , i n the anaerobic zone, were proposed. The authors suggested that if the poly-P organisms are unable to use substrates such as glucose directly, facultative organisms w i l l break them down v i a the Embden-Meyerhof pathway to acetate which the poly-P organisms can store as discussed previously. Poly-P organisms able to use glucose would break the substrate down v i a the Embden-Meyerhof pathway to form pyruvate. A pathway u t i l i z i n g electrons as N A D to trans-f orm pyruvate to P H B was proposed. It was suggested that poly-P organisms entering an aerobic or anoxic region would ut i l i z e t h e i r stored substrate immedately, giving rise to an increase i n the A T P / A D P r a t i o . A n increase i n the above r a t i o may act as a feedback mechanism, activating poly-P kinase to catalyze poly-P production i n order to refill the poly-P p o o l . Comeau et a l . (1986) proposed a biochemical model for enhanced biological phos-phorus removal based o n experimental observations and principals of bacterial energet-ics and membrane transport. B a t c h tests were conducted using aerobic mixed liq u o r obtained from the University of B r i t i s h C olumbia biological phosphorus removal pilot p lant. Acetate a d d i t i o n to the mixed liquor, under anaerobic conditions, resulted i n P release i n proportion to the amount of acetate added. Carbon storage as P H B was also observed. Subsequent n i t r a t e addition resulted i n P uptake b y the biomass u n t i l a l l the oxidized nitrogen had disappeared from solution. T he addition of 2,4 dinitrophenoi ( D N P ) , s o d i u m hydroxide (high pH), H^S gas, and CO* gas was also found to cause P release. The release of P due to CO* addition was noted to be consistent w i t h the observa-tions of other researchers (Fuhs and Chen 1975, and Deinirna 1985). Because C02 w i l l not likely be stored i n carbon reserves, the previous proposal of Marais et a l . (1983) that polyphosphate degradation supplies energy for carbon storage was not considered entirely correct by the authors. Chapter 2. Literature Review 9 To explain the above descrepancy, Comeau et a l . proposed that polyphosphate reserves could be used to supply energy to maintain the proton motive force of the bio-P bacteria, i n addition to its role i n carbon storage. Using the above as a basis, a biochemical model was proposed using acetate as a substrate. Comeau et a l . hypothesized that the addition of acetate to a bio-P biomass, under anaerobic conditions, would result i n the formation of the ionic form of the above (for p H values greater than 6.5, more than 99% of the acetate added w i l l be i n the ionic form). Due to the requirement for electroneutrality when crossing the cell membrane, the acetate w i l l combine w i t h hydrogen ions outside o f the cell before being transported inside. T h e result w i l l be a reduction of the p H gradient of the proton motive force of one hydrogen i o n for each acetate ion transported into the c e l l . If the p H gradient is not restored, acetate uptake, and the corresponding abili t y to increase carbon storage as P H B , would cease. T o expel hydrogen ions and re-establish the p H gradient, it was argued that poly-phosphate may expel protons across the cytoplasmic membrane using a translocating enzyme. The resulting phosphate accumulation inside the c e l l , due to poly-phosphate breakdown, would give rise to a phosphate release to the outside of the c e l l , along the concentration gradient. The presence of a p H sensitive phosphate carrier protein i n the c e l l w a l l was also hypothesized. For unfavorable p H gradients (anaerobic conditions), inorganic P c o u l d not be used for the synthesis processes and would therefore be released at high concentrations. When the p H gradient is favorable (aerobic conditions), the carrier w i l l be inoperable. It was concluded that the extent of P H B accumulation would therefore be lim i t e d by the availability of poly-phosphate for the re-establishing of the p H gradient. This fact was confirmed by batch tests that indicated P release would not increase beyond a certain value regardless of how much extra acetate was added. In the aerobic zone, the bacteria would contain P H B reserves and reduced amounts Chapter 2. Literature Review 10 of poly-P. In the presence of oxygen, the electron transport chain would be opera-tional and the internal A T P / A D P ratio would increase. The bio-P bacteria would degrade their internal P H B reserves f o r energy, and possibly synthesis, and rebuild their poly-phosphate reserves using phosphorus i n solution i n the process. Nitrate re-ducing bacteria would also take up P i n solution according to this model, due to the operation of the electron transport chain. To explain the release of P when H^S and C 02 gas were added to the mixed liquor, the authors suggested t h a t the diffusion of H2S or C 0 2 and their subsequent intracellular dissociation would decrease both the outer and inner p H to such an extent that t h e p H gradient would be reduced, resulting i n P release. Wentzel et a l . (1986) stated that the biochemical model of Comeau et a l . is incom-plete because i t does not provide quantitative information on the pathways and control mechanisms governing biochemical reactions under different sets of imposed conditions of substrate and oxygen tension. The authors further stated that the mechanism pro-posed fo r maintaining the pmf (proton motive force) i n the anaerobic zone gives rise to charge and proton imbalances across the cytoplasmic membrane. A modified biochem-ical model was provided i n an attempt to overcome the above objections. Fundamental to the model's development was the effect of t h e anaerobic and aerobic phases on the intracellular N A D H / N A D and A T P / A D P ratios , a n d the influence of these ratios on the biochemical regulation of carbon and phosphorus metabolic pathways. Acetate was the substrate considered when developing the proposed biochemical model. In the anaerobic zone, the cells were proposed to have a high N A D H / N A D ratio and a low A T P / A D P r a t i o , due to the lack of oxidative phosphorylation. The external concentration of acetate i n t h e anaerobic zone would allow passive diffusion of this substrate into the c e l l . Once i n the c e l l , acetate is converted to acetyl-CoA by A T P hydrolysis. The A T P / A D P ratio is reduced to such an extent that A T P formation Chapter 2. Literature Review 11 v i a poly-P degradation is stimulated. T he intracellular concentration of P, and of the cations stabilizing the negative charges of the poly-P chain, increases, resulting i n their eventual release into the bulk solution. P H B synthesis results when N A D H is oxidized to N A D , w i t h electrons and protons reducing acetoacetyl-CoA to /3-hydroxybutyl-CoA. The decrease i n the N A D H / N A D ratio stimulates the T C A cycle, which generates more N A D H . The formation of P H B lowers the intracellular concentration of acetate, resulting i n more acetate diffusion into the c e l l . T h e authors further proposed that phosphate release occurs v i a a hydroxyl mediated antiport protein carrier w i t h P i moving out of the cell as H^PO^. C a t i o n release occurs v i a a proton mediated antiport protein carrier, and acetic acid w i l l be taken up by passive diffusion. B o t h the proton motive force, and charge neutrality, were shown to be maintained using th i s model. Th e authors suggested that oxidative phosphorylation w i l l take place i n the aerobic zone resulting i n a reduction of the N A D H / N A D ratio and an increase i n the A T P / A D P r a t i o . T h e decrease i n the N A D H / N A D ratio would stimulate further P H B degradation to acetate, providing carbon a n d energy for cell function. The high A T P / A D P ratio would stimulate poly-P synthesis, enabling the organism to establish any required pmf and u t i l i z e A T P for molecule translocation. A similar discussion of anoxic conditions was presented. 2.2 Sequencing Batch. Reactors A great deal of literature has been published on sequencing batch reactor (SBR) tech-nology. Only a brief review of recent research on the application of S B R technology to biological phosphorus removal is presented here. Alleman and Irvine (1980) traced the development of the S B R to the classic study Chapter 2. Literature Review 12 of A r d e r n and Lockett i n the early 1900's. The authors stated that interest i n the fill-and-draw mode of wastewater treatment declined i n the 1920's. However, recently, interest i n the S B R process had been rekindled. T he S B R process was described as consisting of five distinct phases mcluding F i l l , React, Settle, D r a i n , and Idle. D u r i n g the f i l l p eriod, raw sewage is brought into the reactor. Once the reactor is filled, a set amount of time is allowed for the mixed liquor suspended solids to degrade the wastewater organics. The mixers and aerators are then shut off, allowing the mixed liquor to settle, and the treated supernatant is decanted during the d r a i n phase; D u r i n g the i d l e phase, sludge wasting is practised as required. Alleman and Irvine concluded that the opportunity to follow the metabolism of the substrate species during the react period was one of the advantages of the S B R . M a n n i n g a n d I r v i n e (1985) used bench scale SBR's, treating synthetic wastewater, to remove phosphorus biologically. The basic cycle used included a two hour fill p eriod, four hour react period, and a two hour period for draw, settle and id l e . Six operating strategies were tested. T he feed addition pattern, feed addition time, and mixer status d u r i n g fill times differed for each strategy. The strategies which eliminated D O and NOx-N during the fill p e r iod, resulted i n biological phosphorus removal. The authors concluded that biological phosphorus removal could be achieved i n a n S B R w i t h a relatively low COD/TKN" r a t i o . K etchum et a l . (1987) reported results of phosphorus removal experiments using a f u l l scale S B R at Culver, Indiana. The S B R used is a fill and draw activated sludge system i n which each basin i n the system is filled and then aerated i n a batch treatment mode. F i v e discrete periods, consisting of F i l l , React, Settle, Draw, and Idle, were identified for each cycle. B o t h chemical and enhanced biological phosphorus removal were tried at this facility. The authors concluded that the f u l l scale use of an S B R for enhanced biological phosphorus removal was quite effective. Chapter 2. Literature Review 13 Okada et a l . (1987) used a laboratory scale S B R to clarify the behavior of micro-organisms responsible for phosphorus removal and to remove nitrogen, phosphorus, and B O D . Operations where anoxic/anaerobic reactions were introduced into the f i l l period gave the best performance. Appeldoom and Deinema (1987) used a laboratory scale fill-and-draw system to study biological phosphorus removal. A six hour operating schedule, consisting of two hours and forty-five minutes of aerobic time, one hour and fifteen minutes of anaerobic time, and two hours settling time, was used. When the system was r u n w i t h acetic acid and ten percent glucose, the sludge was found to contain more than 6 0 % Acinetobacter. When fifty percent glucose was used i n the feed 2 0 % fewer Acinetobacter were observed t h a n i n the first r u n . T he authors concluded t h a t fill-and-draw systems could be used to explain the discrepancies i n the behavior of pure cultures of Acinetobacter and those of sludge f r o m P-removal plants. 2.3 Rotating- Biological Contactors (RBC's) T h e development of the R B C process began i n Germany and the U n i t e d States i n the 1920's using wooden discs. B y 1959, J . C o n r a d Stangelin was manufacturing two an d three meter diameter polystyrene discs h i West Germany, and b y the late sixties a number of small West German communities were using RBC's for their wastewater treatment. T h e capital cost of these original units was quite high. However, the maintenance a n d energy costs proved to be low. The above is s t i l l a selling point used by R B C manufacturers. I n 1972, the American company A u t o t r o l developed a more compact disc for the R B C process and RBC's soon became cost competitive w i t h other treatment processes. A cost-effectiveness analysis was done by Lundberg and Pierce (1980) to compare air-drive and mechnical drive R B C processes w i t h air and Chapter 2. Literature Review 14 pure oxygen activated sludge processes. For the range of design flows studied (3-50 M G D ) RBC's were found to require 40-60% of the power needed for the pure oxygen activated sludge facilities and from 23-60% of the power needed for air activated sludge processes. The authors concluded that R B C processes were a viable alternative to activated sludge i n the 3-50 M G D design range. The R B C literature provided i n this section has been subdivided into the two main areas considered pertinent to this research: R B C biofilm studies; and nutrient removal using t h e R B C process. 2.3.1 R B C B i o f i l m S t u d i e s A i l e m a n et a l . (1982) used scanning electron microscopy to evaluate the stratified com-position of an R B C biofil m . B i o f i l m samples were taken from the A l e x a n d r i a , V i r g i n i a R B C facility. T h e b i o f i l m was found to contain a stratified layering of morphologi-c a l l y diverse organisms. T h e top layer contained filamentous bacteria and the lower layer contained comma shaped rods. The authors hypothesized that the aerobic top layer contained Beggiatoa, which are able to use sulphide as an energy source. The lower anaerobic layer was thought to contain Desulphovibria, able to reduce sulphate to sulphide. Hoag et a l . (1983) investigated the types of organisms associated w i t h R B C biofilms. B o t h laboratory a n d f u l l scale units were used i n their study. B i o f i l m samples were microscopically examined and microfauna present were identified using taxonomic keys. A succession of microfauna f r o m stage to stage was observed i n both f u l l and laboratory scale RBC's. Filamentous bacteria were found to dominate the biofilms i n the first stage of the RBC's tested. Attached ciliates were the most frequently dominant microfauna! group i n the second stage laboratory and first stage full-scale units. Rotifers and Sarcodinians were the dominant group i n the t h i r d and fourth stage of lab units and Chapter 2. Literature Review 15 the t h i r d through s i x t h stage of f u l l scale units. The authors related population levels of four microfaunal groups to the concentration of C O D and ammonia. K i n n e r and Maratea (1985) examined biofilm bacteria from an R B C to obtain de-tailed information on the microflora present. Organic loading rate was varied and the effect on bacterial morphology was investigated. A great number and variety of cells was observed i n the most heavily loaded R B C compartments. The d i s t r i b u t i o n between gram-positive and gram-negative bacteria appeared to be equal. The large gram-positive cells contained fewer inclusions than similarly sized gram-negative cells. T h e smaller bacteria were observed to store large amounts of P H B - l i k e m a t e r i a l . Polyphosphate and possibly sulfur inclusions were also observed. K hmer a n d Maratea. found that the lower organic loading rates resulted i n a pre-dominantly gram-negative population, and less diversity of cell types. The amounts of P H B and polyphosphate were reduced and no sulphur inclusions were observed. B a c t e r i a resembling nitrinera were found i n the final compartment of the R B C operating at the. lowest loading rate 2gTOC/m2 — hr. The above bacteria contained extensively convoluted cytomembranes, were often enclosed i n common capsules, and often contained polyphosphate-like inclusions. P H B was not stored i n any of the above cells. 2.3.2 Nutrient Removal using' RBC processes R B C processes have been used extensively i n the past for carbon oxidation and n i t r i -fication. The use of RBC's for denitrification has been li m i t e d to date. To the authors knowledge no R B C plant has been purposefully operated for enhanced biological phos-phorus removal as of yet. However, a number of plants have incorporated chemical phosphorus removal into their process trains. The reason for the above is that many Chapter 2. Literature Review 16 researchers hold the same view as Strom and Chung (1985) that i t would be difficult to employ RBC's f o r the luxury uptake of phosphorus. Presented below is a literature review of research on nutrient removal using RBC's. Although by no means complete it is felt that some of the most pertinent information has been presented. Pretorius (1971) used a laboratory scale R B C to treat anaerobically pretreated domestic sewage. T h e author used an inoculum from an activated sludge plant to seed the process, and observed a good biological f i l m on the discs after twenty-one days of operation. The rotating, discs nearest the inlet were covered w i t h a thick filamentous growth of sphaerotilus and beggiatoa. The remaining discs consisted of a slimy indefinable film, made up of very fine filamentous cells together w i t h a zoogleal mass. Good C O D removal and ni t r i f i c a t i o n were reported. Pretorius reported very poor phosphorus removal using this system. In p a r t i c u l a r t h e T P concentration decreased slowly f r o m 13.2 to 10.6 mg/l across the R B C u n i t . The orthophosphate concentration remained v i r t u a l l y unchanged. Antonie et a l . (1974) reported on the results of a study done at the Pewankee, Wis-consin wastewater treatment plant. The above plant was the first full-scale American f a c i l i t y using the R B C process. Primary clarification, two banks of four 3.1 m diameter R B C u n i t s , secondary clarification, and aerobic digestion comprised the process t r a i n of the 0. 5 mgd plant. The authors investigated rotational disk velocity, hydraulic loading, and effect of cl i m a t i c conditions on treatment efficiency. A high degree of B O D and SS removal and ni t r i f i c a t i o n was reported. Stable operation under fluctu-ating hydraulic and organic loading and wastewater temperature was observed. Low power, maintenance, and sludge handling costs were also cited as advantages of the R B C process. Davies and Pretorius (1975) studied the denitrification performance of an anaer-obic R B C , w i t h respect to carbon requirements, p H and temperature. Their study Chapter 2. Literature Review 17 was divided into four stages including adaptation of the R B C unit to denitrification, determination of the lowest carbon to nitrogen ratio necessary for complete denitrifi-cation and the determination of the effects of pBT and temperature on denitrification rate. The carbon to nitrogen (C:N) ratio was varied by changing the methanol feed rate. Once the optimum C:N ratio was found the effects of p H and temperature vari-ations were stu d i e d . T h e optimum C:N ratio for denitrification was found to be 2.6:1. The optimum p H for denitrification was between 7 and 8.5. Values on either side of the reported range gave less t h a n complete denitrification. The optimum temperature range for denitrification was found to be between 10° and 30° C. A l l the above values were reported as being s i m i l a r to those observed for activated sludge denitrification u n i t s . The authors concluded that the mechanical simplicity and ease of operation of the R B C make i t a. system worth considering f o r denitrification. Hao et a l . (1975) reported on the results of a pilot study at the Columbus, Indiana sewage treatment plant. A n R B C u n i t was used for carbonaceous B O D , SS , a n d am-m o n i a nitrogen removals. Phosphate removal was accomplished by chemical addition. B o t h l i q u i d a lum and ferric chloride were t r i e d . Chemical addition to and after the fourth stage was investigated. The addition of l i q u i d alum directly to the fourth stage appeared to affect effluent B O D and SS adversely. The authors hypothesized that the chemical coating o n t he media decreased the biological treatment efficiency, and that chemical a d d i t i o n resulted i n poor solids settleability. A d d i t i o n of phosphorus precipi-t a t i n g chemicals after the fourth stage resulted i n improved B O D removals. When no chemicals were added the authors reported phosphorus removals between 20 and 40 percent. The authors concluded that the R B C system provides good removals of B O D , SS, and ammonia nitrogen for the Columbus wastewater under an hydraulic loading of 1.5gpd/ft2. M u r p h y et a l . (1977) compared the denitrification capabilities of a suspended growth Chapter 2. Literature Review 18 system, a submerged R B C u n i t , and four upflow submerged packed columns. The den-itri f i c a t i o n rate for the R B C was found to be independent of nitrate and n i t r i t e concen-trations. Efficient a nd predictable removals at al l temperatures normally encountered were reported. The effluent from the R B C was found to be quite low i n suspended solids, indicating the possibility of eliminating further clarification. Odegaard and Rusten (1980) studied nitrogen removal using RBC's. The exper-imental unit consisted of a submerged R B C for denitrification followed b y a semi-submerged R B C for n i t r i f i c a t i o n . N i t r i f i e d effluent was recycled from the n i t r i f y i n g unit to the deni t r i f y i n g u n i t . Raw wastewater was used as the carbon source for den-i t r i f i c a t i o n . Three: different types of raw wastewater were used i n the experiments including synthetic sewage, municipal sewage, and land f i l l leachate. The authors found that i t was unnecessary to cover the free water surface of the submerged R B C , as den-it r i f i c a t i o n took place at Oi concentrations i n the water of 1 mg Oj/Z. The proposed process was demonstrated to give good nitrogen removals for a l l three wastewaters. Organic matter was found to be consumed at a rate of 3g COD/g NOx-N removed. To achieve complete denitrification the ratio of soluble C O D to NOx-N should be greater t han seven. T h e authors concluded t h a t the main advantages of the process were that no external carbon source was required, and that the process would be economical from both an investment and operations view p o i n t . Noss et a l . (1980) studied wastewater recarbonation using an R B C . P i l o t studies were conducted using domestic sewage. Raw degritted wastewater was pumped to a r a p i d m i x tank where lime was added for phosphorus removal purposes. The wastewa-ter then flowed through a flocculation b a s i n , a primary settling tank, an R B C u n i t , and finally a secondary clar i f i e r . The concentration of phosphorus i n the wastewater was reduced from values as high as 10.5 mg/l to values of 2 mg/l or less. The R B C unit successfully recarbonated the high p H wastewater w i t h no deleterious effects on the Chapter 2. Literature Review 19 biofilm organisms being reported. Advantages to the above process included reduced R B C surface area requirements for B O D removal due to the B O D reduction achieved during chemical precipitation, and the production of an excellent environment for n i -tri f i c a t i o n i n the later R B C stages due to biological oxidation of organics i n the i n i t i a l R B C stages. Ouyang (1980) attempted to characterize the sludge from two conventional R B C pilo t facilities. One f a c i l i t y had two stages, while the other had four. He found that bi o f i l m thickness was dependent on the organic loading and rotational speed for both units. He attempted to compare the solids concentration i n the R B C w i t h the conven-tional activated sludge plant using a value he referred to as the equivalent suspended solids concentration (ESS). T h e ESS was defined as the dry weight of solids on the disks i n g/m2 d i v i d e d by the l i q u i d volume to disk surface areas as l/m2. Ouyang found the ESS to vary w i t h the organic loading. He also found an average ESS of 9,000 mg/l for the 4 stage process when the B O D loading ranged from 10-641 g BOD/m2 day. T h e 2 stage process h a d an average ESS of 11,000 mg/l for the same loading. It was argued that the high values of E S S explained the higher efficiencies of RBC's compared to the activated sludge process. U s i n g the ESS a n d surface loading rates, the author estimated the F / M ra t i o of the R B C process to be 0.1 - 0. 4 g B OD/g ESS day, which is similar to that of the activated sludge process. Analysis of the biofilm was done a n d Ouyang reported an average volatile frac-t i o n of 7 4 % w i t h a water content of 9 5 % . The reported chemical composition was CHH%NQAOI. R B C sloughed solids were reported to settle quickly. The treatability of the R B C solids using aerobic or anaerobic treatment was reported to be better than that of activated sludge. Hynek and Lemura (1980) looked at phosphorus and nitrogen removal using the R B C . In particular the authors looked at nutrient removal using Au t o t r o l s ' Bio-Surf Chapter 2. Literature Review 20 Process. B o t h conventional and submerged units were used. Conventional R B C units were used for ni t r i f i c a t i o n and B O D removal. Submerged R B C units were used for denitrification. Methanol was the carbon source used for denitrification. Phosphorus removal was achieved by alum or ferric chloride addition. The authors reported a minimum phosphorus removal of 8 0 % using the A u t o t r o l process. Knoetz et a l . (1980) studied the inhibitory effects of various heavy metal ions and organic inhibitors. Laboratory scale activated sludge and R B C units were used i n their study. The authors found that n i t r i f y i n g and denitrifying rotating disc units, a n d nitrifymg-denitrifying activated sludge units had completely different responses to inhibitory substances. T h e denitrifying R B C showed greater resistance to inhibition by C r6* and Ni+t than d i d the m t r i f y i n g R B C and activated sludge units. However, n i t r i f y i n g discs and activated sludge units tolerated Zn+3 a n d Cd+2 salts much better t h an the d e n i t r i f y i n g disc u n i t s . T he activated sludge units were not affected as greatly by mercury salts as were b o t h rotating disc u n i t s . O f a l l the organics studied only dithane and dichlorophenol had significantly differ-ent effects on the units studied. T h e denitrifying discs had a much higher tolerance for dithane, while the n i t r i f y i n g discs had a much higher tolerance f o r dichlorophenol. S m i t h and. K h e t t r y (1980) reported o n the results of a pilot study done by the Ontario ministry of the Environment using RBC's for complete nitrogen removal. The pi l o t plant consisted of two RBC's i n series. The first unit n i t r i f i e d the incoming wastewater. A second R B C was used for wastewater denitrification. The second R B C was operated i n both the submerged and semi-submerged modes to determine the effect of disc submergence on denitrification efficiency. Close to f u l l ammonia oxidation was attained i n t h e n i t r i f i c a t i o n R B C module. The D O concentration i n the n i t r i f y i n g R B C was found to increase w i t h increasing disc r p m at the expense of nitrate removal i n the denitrification module; When the denitrification module was operated i n the Chapter 2. Literature Review 21 f u l l submergence mode a nitrate removal efficiency of 9 1 % was attained. The semi-submerged mode of operation resulted i n much lower nitrate removals. Singhal (1980) reported on phosphorus and nitrogen removal at the C a d i l l a c , M i c h i -gan wastewater treatment plant. RBC's were used for ammonia nitrogen removal and addition of ferric chloride to the aeration system was used for phosphorus removal. The author found high concentrations of heavy metals i n the ferric chloride being used for phosphorus removal. The heavy metals were found to accumulate i n the biofilm of the n i t r i f y i n g R B C and i t was hypothesized that they may inh i b i t n i t r i f i c a t i o n to some degree. The Japanese company K u b o t a (1980) made a Japanese patent application for an R B C system to provide n i t r i f i c a t i o n and denitrification i n the same reactor. N i t r i f i -c ation w o u l d be carried out i n the R B C unit when the discs were pa r t i a l l y immersed i n wastewater. Denitrification would b e carried out when the discs were completely submerged i n the wastewater. The advantages claimed for this system included the fact that n i t r i f i c a t i o n and denitrification could be carried out i n t h e same R B C un i t , installation costs are saved, the addition of a carbon source for the denitrification pro-cess is not required, and the amount of alkali matter used i n the nit r i f i c a t i o n process is decreased due to the increase i n alkalinity i n the denitrification procedure. Masuda et a l . (1983) used a completely closed R B C unit to investigate simultaneous ni t r i f i c a t i o n and denitrification (SND) i n a n R B C bio f i l m . Vent holes i n the reactor were closed and the experimenters monitored the concentrations of oxygen, nitrogen, a n d total gas w i t h time. The removal rate of nitrogen due to simultaneous nitrification and denitrification was found to be dependent upon ammonia loading, organic loading, mean cell residence time, and the pressure of oxygen i n the gas phase. A long mean cell residence time accompanied by low concentrations of ammonia and organic matter re-sulted i n increased removal of nitrogen by simultaneous ni t r i f i c a t i o n and denitrification. Chapter 2. Literature Review 22 High ammonia and organic loadings and short mean cell residence time resulted i n l i t t l e nitrogen removal by SND. Experiments were conducted i n which ammonia and organic loadings, as well as the p a r t i a l pressure of oxygen were varied. The rate l i m i t i n g step for S N D was goverened by the par t i a l pressure of oxygen. When the p a r t i a l pressure of oxygen was low, nitrif i c a t i o n limited SND i n the biofilm. As the p a r t i a l pressure of oxygen increased, the diffusion of organic matter to the denitrifying bacteria limited S N D . T h e value of par t i a l pressure governing the rate l i m i t i n g step was dependent on the organic loading. Experiments were carried out to determine the optimum C/N ratio f o r S N D . Masuda et a l . developed a biofilm model to explain their results. It was proposed that the b i o f i l m was comprised of an outer heterotrophic layer, an autotrophic layer, and an inner anaerobic layer. The gradients of organic carbon, dissolved oxygen, and ammonia would determine the dominant bacterial species. The rate l i m i t i n g step i n S N D would also be determined by the above gradients. The authors concluded that i t was theoretically possible to accomplish organic oxidation, n i t r i f i c a t i o n , and denitrifi-c ation i n the same R B C reactor. M u c h of the research discussed i n the above literature review lends credence to the concept of using an R B C , operated i n a sequencing batch mode, to remove phosphorus f r o m wastewater by the enhanced biological phosphorus removal mechanism. W i t h the recent advances made i n the understanding of the biochemical mechanisms governing enhanced biological phosphorus removal, the application of this technology to treatment processes, other than activated sludge, appears possible. Research done by Irvine and others has shown that SBR's can be used for biological phosphorus removal at both lab and f u l l scale. T h e patent application brought forth by K u b o t a (1980) suggests RBC's can be operated successfully i n a sequencing batch mode w i t h alternating aerobic and anaerobic conditions. The discovery of both poly-P and P H B inclusions i n the biofilm Chapter 2. Literature Review 23 bacteria of RBC's by K hmer and Maratea (1985) suggests that the proper 'bio-P' bacteria can be encouraged to grow i n a fixed-film culture. The primary objective of this research was to prove that a lab scale sequencing batch R B C could i n fact be used for enhanced biological phosphorus removal. It was hoped that knowledge gained w i t h the lab scale reactor could be used to provide guidelines for the design of a pilot-scale facility. C h a p t e r 3 M e t h o d o l o g y This section describes the rationale for the proposed process, the individual process components, system operation, seeding, sewage characteristics, sampling procedures, and analytical techniques used. 3.1 The Experimental Unit 3.1.1 System Rationale D u e to recent advances i n the understanding of the mechanisms governing biological phosphorus removal, the application of this technology to fixed film wastewater treat-ment systems now appears possible. I n January 1987 an investigation into the use of an R B C system for biological phosphorus removal was started. The R B C was chosen due to its low operating and maintenance costs. For enhanced biological phosphorus removal to take place 'acclimatized micro-organisms' must be subjected to alternating anaerobic and aerobic conditions. The presence of short chain fatty acids and absence of nitrates i n the anaerobic phase is also essential. To achieve'the necessary anaerobic/aerobic sequence, the rotating discs of a lab-oratory scale R B C were designed to be subjected to varying degrees of submergence. Ini t i a l l y , the reactor was filled to ensure the rotating discs were completely submerged. 24 Chapter 3. Methodology 25 D u r i n g the fill period, acetate was added to the influent line of the reactor. Follow-ing the fill period the reactor was operated i n the batch mode. Enough time was allowed for the necessary anaerobic processes to take place. Theoretically, the bacteria attached to the R B C discs would take up the added acetate for carbon storage and release phosphorus into the bulk solution from their poly-P. reserves. A f t e r the anaerobic react period part of the reactor contents were then drawn off into a holding vessel to later become part of the influent for the next cycle. The pa r t i a l l y submerged discs were then rotated during the batch aerobic react period. Oxygen would be available from the atmospheric a i r v i a the bulk l i q u i d . W i t h the availability of oxygen, the bi o f i l m bacteria would be able to use the carbon stored d u r i n g the anaerobic p e r i o d to store phosphorus. A n y nit r i f i c a t i o n of the wastewater was considered to be a secondary benefit of the proposed process. FoUowing the aerobic period treated effluent and sloughed solids were decanted to a l i q u i d / s o l i d separation u n i t . Removal o f accumulated phosphorus from the system coincided w i t h the removal of sloughed solids i n the effluent draw. The end of the effluent draw signaled the beginning of the next fill p eriod. R a i s i n g a n d lowering the rotating discs was considered as an alternative to the above mode of operation. T h i s option was not pursued i n this research due to the difficulty of constructing such a system f o r laboratory scale operation. Process control was achieved using an electronic timer/microprocessor. Predeter-mined times for pumping, decanting, and reaction could be programmed into the in-d i v i d u a l c i r c u i t s . Solenoid valves were used to control the water level i n the reactor. The disc rotational speed was kept constant at al l times. A process flow sheet of the experimental unit is shown i n Figure 3.1. Table 3.1 shows the operating schedule w i t h the desired reactions. circuit 1 circuit 2 cetate^  reservoir Microprocessor Influent Sewage pump Sequencing Batch RBC circuit 3 circuit 4 F i g u r e 3 . 1 P r o c e s s P l o w S h e e t e Anaerobic Draw Valve Effluent Valve P A Wastewater Power Acetate Chapter 3. Methodology 27 Table 3.1: O p e r a t i n g S c h e d u l e Process Step Microprocessor status Desired Reaction 1. pump circuits 1 and 2 on circuits 3 and 4 off Sewage pumped from the raw sewage reservoir and anaerobic holding vessel into the R B C . Rot a t i n g discs submerged to create anaerobic conditions. Acetate added during pump phase. 2. Anaerobic react al l circuits off Submerged rotating discs. B i o f i l m bacteria store added acetate as P H B and release phosphorus. 3. Anaerobic decant circuit 4 o n a l l other circuits off P a r t i a l l y treated wastewater drawn into anaerobic holding vessel. G r a v i t y flow when anaerobic valve on. Draw stops at draw line level. Discs now 5 0 % submerged. 4. Aerobic react a l l circuits off B i o f i l m bacteria now exposed to oxygen. Stored P H B used to store phosphorus as poly-P. N i t r i f i c a t i o n and majority of C O D reduction take place. 5. Effluent decant circuit 3 on al l other circuits off Treated effluent and P rich sloughed solids decanted for s o l i d / l i q u i d separation. Chapter 3. Methodology 28 P r i o r to the start-up of this system a number of concerns of peers w i t h respect to the potential success of the process were given special consideration. Paramount among their concerns was the a b i l i t y of such a system to remove phosphorus i n the sloughed solids at a rate equal to the rate of phosphorus accumulation. Other concerns brought f o r t h included the difficulty of achieving anaerobic conditions i n the laboratory u n i t . The above concerns were kept i n mind i n the design of the system prototype. 3.1.2 R e a c t o r a n d R o t a t i n g D i s c s The laboratory-scale reactor consisted of three main pieces including the reaction vessel, the shaft and attached discs, and a plexiglass cover with its attached bearing supports. T h e 20 l i t r e reaction vessel was rectangular i n plan w i t h a c y l i n d r i c a l bottom. Plexiglass construction was used throughout. The reaction vessel was formed i n two parts due to the difficulty of forming long plexiglass pieces. Inlet, anaerobic draw, and effluent pipes were 6 mm i n diameter. T h i r t y four, 19.5 cm diameter polyethylene discs were attached to a 6 mm diameter stainless steel shaft. The shaft was 46 cm long. The discs were fitted to the shaft i n two groups of 17 discs. E a c h group of discs was spaced 3.5 cm from the end of the shaft center. Individual discs were spaced 1 cm apart on centers. A 25 mm diameter stainless steel sprocket was welded to the shaft at a distance of 1 cm offset from center. A 6 m m inch pitch was used for the drive sprocket. The cover and attached bearing supports were made of 1 c m thick plexiglass. Each bearing support was 19.5 by 22 cm. A tenon friction bearing was set into each support 16 c m f r o m the top of the plexiglass cover. The cover was constructed w i t h two 5 centimeter diameter air vents, a 1.5 by 7 cm slot for the drive chain, a 14 by 18 cm plexiglass pad on which to mount the drive motor, and a 2 cm diameter hole at the influent end of the R B C reaction vessel to accommodate an O R P probe. The entire Chapter 3. Methodology 29 reactor assembly is shown i n Figure 3.2. The drive motor was bolted to the plexiglass pad. Due to the structural flexibility of the plexiglass bearing supports the disc/shaft assembly could be easily fitted into the Teflon bearings. The drive chain was then attached to the stainless steel sprockets. The entire assembly including the discs and the shaft would fit into the reaction vessel, and be supported by the side walls of the reaction vessel, and the center compartment w a i l . A slot was cut into the center wall to accommodate the shaft, drive sprocket, and chain. T h e assembly is shown i n Figure 3.2. The above construction allowed for easy shaft and media maintenance. Special metal clips were constructed to support two floating covers for the reaction vessel. The styrofoam covers were provided to minimize the surface exchange of oxygen between the atmosphere and the bulk l i q u i d during the anaerobic phase. The metal clips allowed the styrofoam covers to rest 6 cm above the rota t i n g discs during the aerobic react period. Figure 3.3 shows the details of the metal clips and floating covers. P r i o r to the anaerobic react period the reaction vessel would be fi l l e d to a level above t h a t of the metal clips allowing t h e covers to float. Disc Drive Unit The mechanical drive unit consisted of a D C gearmotor, a variable speed controller, two stainless steel sprockets and 6 m m p i t c h stainless steel chain. A 1/8 hp Dayton gearmotor model 4Z135B was used to rotate the shaft and at-tached discs. T h e motor was connected to a Dayton S C R Controller model 60648. T h i s configuration allowed for motor rotational speeds from 0 to 100 R P M . A 2.5 cm diameter stainless steel sprocket was welded to the gear motors haft. A stainless steel chain was connected between the 2.5 cm sprocket and the 5 cm sprocket on the R B C shaft. The 2:1 sprocket ratio allowed for disc rotational speeds between 0 and 50 rpm. Chapter 3. Methodology FIG.3.2. R E A C T O R A S S E M B L Y . Chapter J . Methodology 31 FIG.3;3. F L O A T I N G C O V E R S . Chapter 3. Methodology 32 Stainless steel sprockets and drive chain were used due to their resistance to corro-sive agents present i n domestic sewage. The abilit y to vary the disc rotational speed over such a wide range gave more process flexibility. 3.1.4 Raw Sewage Reservoir and Anaerobic Holding Vessel A t the start of each treatment cycle raw wastewater and anaerobically treated wastew-ater would be used to fill the reaction vessel. The raw wastewater was stored i n a 10 gallon plastic garbage can . The anaerobically treated wastewater was excess wastewa-te r f r o m the anaerobic react period. A five gallon plastic garbage can was used to store this l i q u i d . B o t h reservoirs were covered and mixed to keep solids i n suspension. The raw sewage reservoir was mixed w i t h a model 4Z135B 1 /8 hp Dayton gearmotor. T h e m i x i n g shaft was a 13 m m diameter stainless steel shaft. A propeller shaped mixer was attached to the end of the m i x i ng shaft. A Dayton model 2Z807 12 R P M gearmotor was used to keep t h e contents of the anaerobic holding vessel i n suspension. 3.1.5 Pumpa Two pumps were used i n the R B C process. One pump was used to pump the raw and anaerobically treated wastewater into the reaction vessel at the start of each treatment cycle. A chemical feed pump was used to inject acetate solution into the influent line. B o t h pumps were operated f r o m separate microprocessor circui t s . T h e influent sewage pump was a Masterflex model 7520-00 variable speed u n i t . Two number 15 Masterflex pump heads were used for this pump. One pump head was used to pump from each of the raw sewage and anaerobic holding vessels. A Cole-Palmer model C-1714SP chemical feed pump was used for acetate addition. A 2 litre graduated cyclinder served as the reservoir for the concentrated acetate solution. Chapter 3. Methodology 33 3.1.6 T i m e r / M i c r o p r o c e s s o r a n d S o l e n o i d V a l v e s Process control was achieved using an electronic timer/microprocessor and two solenoid valves. T h e microprocessor used was a Chrontrol Model CD03 4 circuit microprocessor. T h e two solenoid valves and a l l pumps were operated by the microprocessor. A l l pumps used one circuit each. Each solenoid valve was connected to a seperate c i r c u i t . C o n t r o l programs were written for the three circuits used to control pumping, react, a nd decanting times. T w o Asco 13 T r i m model 8030A17 Solenoid valves were used to control disc submer-gence i n the reaction vessel. 3.1.7 Miscellaneous S i x m i l l i m e t e r diameter Tygon tubing was used for a l l flow connections. 3.2 S y s t e m O p e r a t i o n T h e three m a i n process operations of fill, react, and decant were time controlled by the microprocessor. T i m e requirements for filling and decanting were calibrated prior to system start-up. The amount of time allowed for reaction is discussed i n section 3.2.2 • A t the beginning of the treatment cycle the circuits controlling pumping were acti-v ated. A l l other microprocessor circuits were off at this t i m e . Equal volumes of raw and anaerobically treated wastewater were pumped fr o m the respective reservoirs into the reaction vessel. Acetate solution was injected into the influent line at a predetermined rate and concentration to give the desired concentration i n the reaction vessel. During the anaerobic react period all,microprocessor circuits were turned off. Chapter 3. Methodology 34 At the end of the anaerobic react period, the circuit controlling the solenoid valve for the draw line at the reactor midpoint was activated. A portion of the reactor contents were decanted into the anaerobic holding vessel to form part of the influent for the next cycle. W i t h the discs partially submerged, the necessary aerobic reactions could take place. At the end of the aerobic time period, the circuit controlling the solenoid valve for the effluent draw line, at the bottom of the reaction vessel, was opened. Treated effluent and phosphorus rich sloughed solids were then decanted to an effluent bucket. At the end of the effluent decant period, the pump circuit was activated once more, to signal the start of the next treatment cycle. 3.2.1 Disc Rotational Speed The U.S. Environmental Protection Agency design manual for rotating biological con-tactors suggests that the peripheral velocity of the rotating discs of an R B C unit be 0.3 m/s for good oxygen transfer and biofilm management. For the 19.5 cm diameter discs employed in these experiments this corresponds to a disc rotational speed of 30 R P M . Prior to the installation of floating covers, a disc rotational speed of 10 R P M was employed. The lower speed was chosen in an attempt to strike a balance between the requirements of the anaerobic and aerobic phases. It was hoped the lower rotational speed would result in lower surface turbulence, and therfore reduced oxygen transfer, in the anaerobic phase. However, the lower rotational speed resulted in an undesirable lag in the time required to go from anaerobic to aerobic conditions following the anaerobic decant period. Biofilm bridging between discs also resulted due to the inadequate shear given at the lower rotational speed. Following the installation of the floating covers, the disc rotational speed was increased to the suggested value and the above was corrected. Chapter 3. Methodology 35 3.2.2 L e n g t h of R e a c t i o n Phases The length of anaerobic and aerobic time required for the necessary reactions to take place was chosen on the basis of the past experiences of other researchers with activated sludge. A n anaerobic react period of one and a half hours and an aerobic react period of three hours were used. Batch tests proved these times to be quite conservative. In retrospect, these times could have been shortened considerably. 3.2.3 Acetate Addition Acetate was added to the reactor to give a concentration of 25 m g / l as C O D during the anaerobic react phase. The results of Comeau (1984) and others have shown that a concentration of 15 mg / l would be more than adequate for lab scale activated sludge units. The higher concentration was chosen to overcome any mass transfer limitations that might be present. Again, the choice proved to be very conservative. 3.2.4 Flow Management Due to the unusual shape of the reaction vessel, more wastewater entered the anaerobic holding vessel during the anaerobic decant period than was reintroduced to the R B C from this vessel during the reactor fill period. The resulting flow accumulation in the anaerobic holding vessel of 11. 5 liters per day was transferred to the raw wastewater reservoir at the head end of the plant. This situation was unavoidable due to the difficulty of predicting the amount of fluid displaced by the biomass at any period of time. To alleviate the problem, a more sophisticated process control system would have been required. One of the methods which was attempted to alleviate this problem included pumping the anaerobic holding vessel dry during each fill period. This practise increased the proportion of anaerobically treated wastewater to raw wastewater in the Chapter 3. Methodology 36 influent, and also introduced oxygen into the anaerobic phase. 3.3 Seeding Seeding of the R B C reactor was undertaken to reduce reactor start-up time, and in-crease the likelihood of fixing the desired micro-organisms to the rotating discs. Fletcher (1979) discussed the attachment of bacteria to surfaces in aquatic envi-ronments. She found that an increase in either culture concentration or exposure time resulted in an increase in the number of bacteria attached to a surface. Using Fletcher's results as a guide, a seeding program was devised. Thickened sludge was obtained from the pilot-scale enhanced biological phosphorus removal plant at the University of Brit ish Columbia. The M L S S concentration of the thickened sludge was approximately 5,000 mg/ l . A 50:50 mixture of raw sewage and thickened sludge was added to the reaction vessel. A t this time the reactor was operated in a batch mode, at a disc rotational speed of 10 R P M . A n acetate slurry was added to the reaction vessel four times daily to supplement carbon levels. The concentrations of C, N , and P were monitored daily to ensure nutrient levels were adequate. Each day, the disc rotation was stopped and the 'mixed liquor' in the reaction vessel was allowed to settle. The supernatant was decanted and the removed volume was replaced with fresh sewage. This practise ensured adequate concentrations of N , P, and trace elements daily. Supplementing carbon levels with acetate kept measured bacterial growth rates at accelerated levels. Nilsson and Dostalek (1984) showed that the ability of a pure culture of Pseudomonas putida to form a biofilm increased with increasing growth rate. Chapter 3. Methodology 37 Table 3.2: T y p i c a l Charac ter i s t ics of U B C b i o - P plant Influent Sewage Characteristic Concentration Range (mg/l) C O D 100-300 SS N A T P 3.5-4.5 T K N 15-25 3.4 Sewage Charac te r i s t ics Raw sewage was obtained from the U . B . C. biological phosphorus removal pilot plant inflow. The plant treats domestic sewage from a two thousand person residential com-plex. The typical composition of the raw sewage during the experimental period is given in Table 3.2. 3.5 S a m p l i n g P rocedu re Sampling was carried out twice weekly. One sample run consisted of raw, influent (raw wastewater plus wastewater from the anaerobic holding vessel), anaerobic draw (wastewater decanted to the anaerobic holding vessel immediately after the anaerobic react period), and effluent samples. Total and soluble C O D , NOx, T K N , TP , and orthophosphate analysis were run on all of the above samples. Raw and effluent samples were also analyzed for suspended solids concentration. A biofilm sample was taken for a % P determination once weekly. The above sampling procedure gave a quick scan of the process treatment performance. Batch testing was conducted during the second weekly sampling run. In addition to the regular raw wastewater, influent and effluent samples volatile fatty acid, solu-ble C O D , orthophosphate, and NOx samples were drawn from the reaction vessel at predetermined times, during the anaerobic and aerobic react phases. These samples Chapter 3. Methodology 38 were taken using a 60 ml syringe. Batch testing results gave the concentration of the bulk liquid soluble species with time. A strong phosphorus release during the anaer-obic period, followed by a subsequent phosphorus uptake, during the aerobic period would be considered as confirmation of enhanced biological phosphorus removal. O R P measurements were also taken during all batch tests. 3.6 Analytical Techniques 3.6.1 Nonfilterable Residue (Suspended Solids) Nonfilterable residue (SS) was determined by filtering a known volume of sample through a standard glass fiber filter (Whatman 934AH), and drying the sample to a constant weight at 104° C. The difference in the weight of the dried filter before and after sample filtration divided by the filtered volume was taken as the concentration of nonfilterable residue in mg/ l . 3.6.2 Nitrogen (N) (a) Nitrate plus Nitrite-Nitrogen (NOx) NOx was analyzed using the Technicon Method 100-70W (1973) with the Technicon Autoanalyzer II. In the above procedure, nitrate is reduced to nitrite by a copper-cadmium reductor column. Nitrite ion reacts with sulfanilamide under acidic conditions to form a diazo compound. A reddish purple azo dye is formed after the coupling of diazo compound with N-l-naphthylethylenediamine dihydrochloride. A detection limit of 0.04 mg JV03-N/1 is achieved. (b) Total Kjeldahl Nitrogen ( T K N ) T K N samples were digested, prior to sample analysis, using the block digestor Chapter 3. Methodology 39 method w i t h sulfuric acid. Sample analysis was carried out on the autoanalyzer ac-cording to Technicon autoanalyzer II , Method No. 146/71A (1972). 3.6.3 O x i d a t i o n - R e d u c t i o n P o t e n t i a l ( O R P ) O R P was measured during a l l batch tests. A Broadley James Corp. A g / A g C l probe was used for al l measurements. 3.6.4 P h o s p h o r u s Ortho and to t a l phosphorus were both measured. The percentage of phosphorus i n the b i o f i l m was also measured. (a) O r t h o p h o s p h a t e (P04) Samples were filtered using a 0.45 micron membrane filter. T he automatic ascorbic acid reduction method (Technicon Autoanalyzer IT, Method No. 94-70W, 1973) was used to analyze filtered samples for ortho-phosphate. The detection l i m i t is 0.2 mgP/I. (b) T o t a l P h o s p h o r u s Total Phosphorus samples were digested p r i o r to sample analysis, using the block di-gestor method w i t h sulfuric acid. Sample analysis was carried out on the Autoanalyzer according t o Technicon Autoanalyzer DT, M e t h o d No. 327-73W (1974). 3.6.5 B i o f i l m P h o s p h o r u s C o n t e n t B i o f i l m samples were scraped from the rotating discs once weekly. Scrapings were dried to a constant weight at 104° C, and then ground into a powder. A weighed aliquot of the above powder was then analyzed for total phosphorus using the method described above. Results were expressed as (mass P/mass SS) on a percentage basis. Chapter 3. Methodology- AO 3.6.6 Volatile Fatty Acids (VFA's) Concentrations of three V F A ' s were determined: acetic, propionic, and butyric acids. Samples were preserved by filtration and freezing. Sample analysis was done using the Hewlett-Packard H P 5880 gas chromatograph with a flame ionization detector (FID). The analysis procedure followed is outlined in Supelco Bulletin 751E (1982). Sample injection was carried out using microsyringes (Hamilton Model 75N, #87900, 5 micro liters). The gas chromatograph column was 0.91 m long with a 4 mm internal diameter. Sulpeco 60/80 Carbopack C/0.3% Carbowax 20M/0.1% H3P04 was used to pack the column. Prior to sample injection, the pH was adjusted to below 3 by adding phosphoric acid giving a 1% acid solution. The experimental conditions for the chromatagraph were: injection port temperature=150° C, detector temperature=110° C, and carrier gas fiowrate=20ml/min. The standards were prepared using reagent grade acetic (99.9%), propionic (96%), and butyric (98%) acids. 3.6.7 Chemical Oxygen Demand (COD) C O D was determined according to method 220 of Standard Methods for the Examina-tion of Water and Wastewater 13th edition (1971). 3.6.8 Disc Solids Biofilm was washed from the discs and onto 74 x 48 x 4 cm drying dishes, at the end of the experiments. The drying dishes were placed in a fume hood for 48 hours to allow the majority of the water to evaporate. The biofilm cake was then scraped into a two litre pyrex beaker and dried to a constant weight at 104° C, to determine the total biofilm weight. The pyrex beaker was then fired at 550° C in a muffle oven to destroy all volatile compounds. The above allowed the determination of both total and organic Chapter 3. Methodology 41 solids present in the biofilm. Chapter 4 Results and Discussion In this chapter experimental results are presented first. D a t a analysis is presented i n the second section of the chapter. F i n a l l y , the results are discussed i n the context of the data analysis provided. T h e experimental results are presented i n three separate sections each corresponding to a different time period. The first time period was from September 1,1987 to February 1, 1988. F o r the first t i m e period the anaerobic holding vessel was included i n the process. T h e second t i m e period r a n f r o m February 6, 1987 to February 16, 1987. D u r i n g the second time period the anaerobic holding vessel was taken off line and the anaerobic draw was returned directly to the raw sewage vessel. D u r i n g the t h i r d time period the anaerobic holding vessel was put back on line. The pumping rate from this vessel was adjusted to double the value i t had been during the September 1,1987 to February 1,1988 time period. The increase i n pumping rate ensured the anaerobic holding vessel was pumped dry during each reactor fill period. 4.1 Results 4.1.1 Phase One (September 1,1987- February 1, 1988) Seeding of the reactor was started on J u l y 30th, 1987. A good biofilm was evident on the discs by m i d August. Sampling was started September 1st, 1987. Initially, phosphorus removal i n excess 42 Chapter 4. Results and Discussion 43 of growth requirements was not observed. The results of a batch test conducted on September 8/87 are shown i n Figures 4.1 and 4.2. No anaerobic phosphorus release was observed at this time. It was suspected that true anaerobic conditions were not being achieved. A shaft failure on September 19/87 supplied the opportunity to retrofit the system w i t h floating covers for the R B C reaction vessel, the raw sewage reservoir, and the anaerobic holding vessel. It was hoped that the above changes would drastically reduce any surface exchange of oxygen between the atmosphere and the l i q u i d . The system was put back on l i n e o n September 22/87. Effluent soluble phosphorus levels of less t h a n 1 mg/l were observed on September 29 and 30. A batch test was r u n on October 5. T h e results for the above test are shown i n Figures 4.3,4.4, and 4.5. The sudden decrease i n t h e reactor PO*, a n d increases i n NO* concentration and O R P at ninety minutes coincides w i t h the draw off o f a portion of the reactor contents to expose the discs for aerobic conditions. B u l k l i q u i d O R P reached a level of -120 m V (vs. A g/AgCl) d u r i n g the anaerobic phase of the treatment cycle. T h e O R P value i n the immediate v i c i n i t y of the b i o f i l m was probably much lower. A phosphorus release of 1.6 mg/l of anaerobic volume was observed i n the anaerobic phase; Uptake i n the aerobic phase was 5.6 mg/l of aerobic volume. Overall percent phosphorus removal was 9 3 % . Twenty-five milligrams per l i t r e o f acetate as C O D was added to the anaerobic volume. A second shaft failure on October 7 resulted i n the system being taken off line f o r mechanical design modifications. B y October 15, 1987 the system was again back on line and seeding of the reactor was i n i t i a t e d . B y November 1 a healthy biomass covered the rotating discs. Effluent sampling was started on November 3, and phosphorus removal i n excess of seventy per-cent was observed by November 10,1987. A batch test r un on November 14 confirmed the removal mechanism to be 'enhanced biological phosphorus removal'. D u r i n g the Chapter 4. Results and Discussion 44 Orthophosphate and NOx concent ra t ions ( m g / l ) ^ Chapter 4. Results and Discussion 46 Chapter 4. Results and Discussion 47 F i g u r e 4 . 5 B u l k l i q u i d O R P v e r s u s T i m e , _ f o r O c t . 5 / 8 7 B a t c h T e s t  Elapsed Cycle Time (min.) Chapter 4. Results and Discussion 49 above run a phosphorus release of 1.4 mg/ l of anaerobic volume was observed during the anaerobic phase. Eighty-three percent phosphorus removal was achieved at this time. Very little nitrification was evident. By November 18, 1987 P removals were consistently above 90%. Percent C O D and phosphorus removals for the period from November 11/87 to February 1/88 are shown in Figures 4.6 and 4.7. Percent nitrification for the same time period is shown in Figure 4.8. Percent phosphorus removals were calculated using raw total phosphorus (TP) and effluent orthophosphate (P04) values. C O D removal was calculated on the basis,of raw total C O D and effluent soluble C O D . Percent nitrification was calculated as the ratio of effluent NOx to raw T K N expressed as a percent. It is likely that some simultaneous nitrification- denitrification took place in the biofilm during the aerobic phase of each treatment cycle, although this was difficult to quantify. The calculation method chosen for percent nitrification was used only to get a relative measure of this parameter between treatment cycles. November 11, 1987 is referred to as day zero of the above time period. The decline in percent phosphorus removal on day 19 (October 30/87) of the ex-perimental run coincided with a malfunction of the acetate feed pump. No acetate had been added to the anaerobic phase of the treatment cycle for three cycles prior to sam-pling. The average characteristics of the raw sewage, the reactor combined influent, the anaerobic draw, and the effluent for the time period November 11/87-February 1/88 are shown in Table 4.1. Figures 4.9 to 4.11 give the results of a typical batch test for the above time period. Table 4.2 shows the observed values for percent phosphorus content of the biofilm. F i g . 4 . 7 P h a s e O n e P e r c e n t C O D R e m o v a l v s T i m e Nov. l/87-Feb.l/88 > O K Q O O -P d 100 Time (days) since Nov. 11/87 F i g . 4 . 8 P h a s e O n e P e r c e n t N i t r i f i c a t i o n v s T i m Nov. 11/87-Feb. 1/88 Time (days) since Nov. 11/87 Chapter 4. Results and Discussion Raw Sewage Concentration (mg/l) Value CC sol. )D tot. I PO, > TP 1 NOx NT TKN TSS Avg. Min. Max. 74 35 220 154 67 504 NA NA NA 4.6 3.0 6.T 0 0 0 20.5 15.2 30 50 10 86 Coi nfluenl icentra t Sewage tion (mg/l) Avg. Min. Max. 62 35 101 121 53 194 5.4 3.2 9.0 6.6 4.5 10.1 0 0 0.2 17.6 12.2 22.5 NA NA NA J Coi Lnaero icentra >ic Draw tion (mg/l) Avg. Min. Max. 45 26 69 NA NA NA 8.0 4.3 13.1 9.3 5.7 20.0 0 0 0 14.9 5.0 30.0 NA ,NA NA Effluent Concentration (mg/l) Avg. Min. Max. 29 9 54 NA NA NA 0.2 0.0 1.4 1.4 0.3 5.2 7.7 0.0 14.5 7.9 2.2 18.5 61 14 186 Table 4.1 : Average Characteristics and Ranges for Nov. 11/87-Feb. 1/88 Chapter 4. Results and Discussion 54 Date Biofilm %P content (total solids basis) Biofilm %P content (volatile solids basis) Dec. 17/87 1.84 3.29 Jan. 6/88 2.04 3.64 Jan. 13/88 2.39 4.27 Jan. 20/88 2.30 4.11 Jan. 25/88 2.27 4.05 Feb. 1/88 2.00 3.57 Table 4.2: Biofilm percent phosphorus contents for Phase One Chapter 4. Results a n d Discussion 55 Orthophosphate and Acetate concent ra t ions (mg / l ) N U * . Ol 05 ^ (ft. Chapter 4. Results and Discussion 56 NOx concentra t ion ( m g / l ) F i g . 4 . 1 1 B u l k l i q u i d O R P v e r s u s T i m e Typical batch test for phase one (Jan. 6 / 8 8 ) Elapsed Cycle time (min.) Chapter 4. Results and Discussion 58 Due to the large draw of wastewater from the reaction vessel following the anaerobic phase, more fluid was returned to the anaerobic holding vessel than was pumped from this vessel during the influent period. Up until February 1, 1988 approximately 11.5 litres of P rich sewage accumulated in the anaerobic holding vessel daily. To alleviate the problem, excess anaerobically treated sewage was transfered to the raw sewage vessel as required. The above practise lead to large fluctuations in the raw and influent sewage characteristics reported in Table 4.1. Influent P levels as high as 10.1 mg/l were observed during the above time period with no apparent effect on effluent quality. Even higher influent P levels may have been observed if both the raw and anaerobic holding vessels had not been emptied and cleaned weekly. 4.1.2 Phase Two (February 6, 1988 - February 16, 1988) In an attempt to streamline process operation, and eliminate the flow accumulation problem, the anaerobic holding vessel was taken off line on February 6, 1988. There-after, the anaerobic draw line returned directly to the raw sewage tank. The rationale behind the above change was that, if returning the accumulated flow from the anaero-bic holding vessel to the raw sewage vessel did not affect effluent quality, neither would direct return of the excess anaerobic draw volume. In fact, the above modification proved to be detrimental to effluent quality. Tables 4.3 and 4.4 show the raw sewage and effluent characteristics for the time period February 6 to February 11, 1988. Daily sampling of the feed tank (raw sewage vessel) was carried out before and after each sewage addition. Samples for effluent phosphate and soluble C O D were taken during the treatment cycles prior to raw sewage addition. Treated effluent was collected, between rawsewage additions to the feed tank, to obtain composite samples for T P , T K N , and TSS. Figure 4.12 shows the progressive increase in the total phosphorus concentration of the feed tank between February 6/88 and Chapter 4. Results and Discussion 59 February 11/88. The sharp drops in T P concentration correspond to the addition of fresh sewage to the feed tank. Figure 4.13 shows the steady increase in effluent phosphate concentrations for the same time period. Chapter 4. Results and Discussion 60 Feed Tank Sewage Composition Concentration (mg/l) Date Time elapsed Time when COD (tot. ) POZ* TF TKN TSS since feed tank sample taken cleaned (days) before(b) or after(a) fresh sewage add Feb. 6/88 0 a 130 2.6 3.5 16.3 25 Feb. 7/88 I b 70 4.8 5.4 16.8 25 Feb. 7/88 1 a 116 3.4 5 15.5 60 Feb. 8/88 2 b 64 5.6 6.6 12.9 38 Feb. 8/88 2 a 110 3.4 4.7 20 37 Feb. 9/88 3 b 176 5.1 7.4 18.1 119 Feb. 9/88 3 a 181 3.8 5.5 16.8 86 Feb. 10/88 4 b 115 4.9 6.8 18.7 67 Feb. 10/88 .4 a 123 4.0 5.6 18.7 65 Feb. 11/88 5 b 75 NA 7.0 12.2 38 Feb.11/88 5 a 119 NA 4.6 17.1 64 Table 4.3: Reactor Feed Composition Phase Two (Feb.6/88-Feb.ll/88) Chapter 4. Results and Discussion 61 Effluent Characteristics Concentration (mg/l) Date Time elapsed COD (sol.) PO? TP TKN TSS since feed tank cleaned (days) Feb. 6/88 0 18 0.3 0.7 8.7 2.5 18 Feb. 7/88 1 55 0.9 1.7 9.5 2.7 23 Feb. 8/88 2 26 1.9 2.2 8.2 3.6 44 Feb. 9/88 3 18 2.2 2.4 8.0 2.7 25 Feb. 10/88 4 22 1.2 2.6 8.0 4.0 57 Feb. 11/88 5 22 2.3 2.7 8.7 4.0 42 Table 4.4: Effluent Composition Phase Two (Feb.6/88-Feb.ll/88) to l£> Fig. 4.13 Phase Two Eff. Orthophosphate vs Time f r o m Feb, 6/88 to Feb. 11/88 Time (hours)- since Feb. 6/88 Chapter 4. Results and Discussion 64 On February 12, 1988 the feed tank was cleaned, and refilled w i t h fresh sewage. The raw sewage and effluent characteristics for the time period February 12 - February 16/88 are shown i n Tables 4.5 and 4.6. Figure 4.14 shows the concentration of soluble C O D i n the feed tank w i t h time. The sharp increases i n soluble C O D correspond to the addition of fresh sewage to the feed tank. The concentrations of T P and PO< i n the feed tank versus time are shown i n F i g u r e 4.15. A g a i n , the sharp decreases i n T P and F O 4 concentrations coincide w i t h the daily addition of fresh sewage to the feed tank. Effluent PO* concentrations for the Febi.12 -Feb.16/88 time period are shown i n Figure 4.16. F i g u r e 4.17 shows the corresponding percent phosphorus removals. Chapter 4. Results and Discussion 65 Feed Tank Sewage Composition Concentration (mg/l) Date Time elapsed Time when COD (sol.) PC*? T P T K N TSS since feed tank sample taken cleaned (days) before(b) or after(a) fresh sewage add Feb. 12/88 0 a 61 2.1 2.9" 21 65 Feb. 13/88 1 b 49 3.2 4.3 18.5 81 Feb. 13/88 1 a 61 3.3 3.9 19 46 Feb. 14/88 2 b 49 3.5 4.4 13.7 30 Feb. 14/88 2 a 61 2.6 4.3 17.6 50 Feb. 15/88 3 b 45 5.2 5.7 12.2 24 Feb. 15/88 3 a 65 3.7 4.8 18 61 Feb. 16/88 4 b 38 5.3 6.1 13.7 50 Table 4.5: Reactor Feed Composition Phase Two (Feb.l2/88-Feb.l6/88) Chapter 4. Results and Discussion 66 Effluent Characteristics Concentration (mg/l' Date Time elapsed COD (sol.) PO? TP NO, TKN TSS since feed tank cleaned (days) Feb. 12/88 0 20 0.1 NA 13.3 NA NA Feb. 13/88 1 24 0.3 1.0 11.1 4 31 Feb. 14/88 2 20 0.9 2.3 8.9 4 65 Feb. 15/88 3 20 1.3 2.6 8.9 4 39 Feb. 16/88 4 22 2.2 2.9 8.3 4 56 Table 4.6: Effluent Composition Phase Two (Feb.l2/88-Feb.l6/88) Chapter 4. Results and Discussion 67 Feed Tank soluble COD c o n c e n t r a t i o n ( m g / l ) 4 . 1 5 P h a s e T w o F e e d T P a n d O r t h o - P v s T i m e f r o m F e b . 1 2 / 8 8 t o F e b . 1 6 / 8 8  Time (days) s ince Feb. 12/88 Chapter 4. Results and Discussion Eff luent Orthophosphate c o n c e n t r a t i o n ( m g / l ) 69 ° F i g . 4 . 1 7 P h a s e T w o P e r c e n t P r e m o v a l v s T i m e f r o m F e b . 1 2 / 8 8 t o F e b . 1 6 / 6 8 J2 Time (days) since Feb. 12/68 o Chapter 4. Results and Discussion 71 Unfortunately, no batch test data is available for the time period from Feb. 6 to Feb. 16, 1988. A biofilm scraping was taken on Feb. 10 and a %P value of 2.3% (total solids basis) was recorded. 4.1.3 Phase Three (February 17 - March 17, 1988) On February 17, 1988 the anaerobic holding vessel was again put back on line. However, the pumping rate from this tank was now double of what it had been in phase one. The increase in pumping rate ensured that the entire contents of the anaerobic holding vessel were emptied during the start of each treatment cycle, thus eliminating any flow accumulation in this tank. The anaerobic holding vessel now supplied 66% of the combined influent to the reaction vessel, as opposed to the previous 50%. Percent phosphorus removals increased following the above change. Phosphorus removals of 75 to 95% were realized from Feb. 19 to Feb. 25, 1988. A batch test was run on Feb. 24/88 and the results are shown in Figures 4.18 to 4.20. Mean effluent P04 for the above run was 0.7 mg/l as P. The initial increase in bulk liquid ORP, observed at the start of the anaerobic phase of the Feb. 24 treatment cycle, was believed to be caused by influent air entrainment. On Feb. 25, 1988 both the raw sewage and anaerobic holding vessels were cleaned. Percent phosphorus removals from Feb. 26-March 1, 1988 ranged from 17 to 50%. On March 1, 1988 a new batch of sewage was collected. Again, the raw sewage and anaerobic holding vessels were cleaned, and refilled. The results of a batch test conducted on March 2 are presented in Figures 4.21 to 4.23. Eighty to ninety percent phosphorus removal was achieved at this time and con-tinued up until March 8. A batch test run on March 9 showed that percent phosphorus removal had decreased to 70% . However, the characteristic anaerobic P release and aerobic P uptake associated with biological phophorus removal was still observed. Chapter 4. Results and Discussion NOx c o n c e n t r a t i o n ( m g / L ) 73 Chapter 4. Results and Discussion 74 B u l k l i q u i d ORP (mV vs A g / A g C l ) o o o o o o o o o o o o o o o F i g . 4 . 2 1 O r t h o p h o s p h a t e a n d A c e t a t e v s T i m e ^ f f o r M a r c h 2 / 8 8 B a t c h T e s t \ 21 -4 1 S* Elapsed cycle time (minutes) o Chapter 4. Results and Discussion NOx concen t r a t i on ( m g / L ) 76 Chapter 4. Results and Discussion 77 Chapter 4. Results and Discussion 78 Figures 4.24 to 4.26 show the batch test results for March 9, 1988. Percent phosphorus removals between 45 and 62% were observed up until March 14, 1988. A t that time, both the raw sewage and anaerobic holding vessels were cleaned. A n effluent sample was taken one cycle after both tanks had been cleaned.Percent phophorus removal had increased to 75%. Mean effluent PO4 was 0.7 mg/ l as P. A percent phosphorus removal of 35% was observed by March 15. A batch test was run on March 16. A n anaerobic phosphorus release of only 0.6 mg/ l of anaerobic volume was observed. The presence of nitrates in the influent was also observed. Figures 4.27 to 4.29 show the results from the March 16 batch test. Sampling was ceased on March 17, 1988. Tables 4.7, 4.8, and 4.9 show the raw sewage, anaerobic holding vessel, and effluent characteristics for the time period February 1.9, 1988 to March 17, 1988. Figures 4.30 to 4.33 show the anaerobic holding vessel and raw PO4 concentrations versus time, the influent and effluent PO4 concentrations versus time, the influent and raw soluble C O D concentrations versus time, and the percent phosphorus removal versus time, for the time period February 27 to March 17, 1988. Chapter 4. Results and Discussion 79 Aceta te and or thophosphate concen t r a t i ons ( m g / l ) i—» P CQ P. O ft 6J CO 0^3 F i g . 4 . 2 5 N O x v e r s u s T i m e f o r M a r c h 9 / 8 8 B a t c h T e s t Elapsed Cycle Time (minutes) F i g . 4 . 2 7 A c e t a t e a n d O r t h o p h o s p h a t e v s T i m f o r M a r c h 1 6 / 8 8 B a t c h T e s t ti m rt o •i-4 +-> a +-> rt o rt o a <u +-> a xi fr m O A fr o +-> u o •d rt <u +-> a +-> a aerobic phase orthophosphate 200 240 280 Elapsed Cycle Time (minutes) Chapter 4. Results and Discussion NOx c o n c e n t r a t i o n ( m g / L ) 83 O h~ CO 03 Ol -sZ CD co Chapter 4. Results and Discussion 84 B u l k l i q u i d ORP (mV vs A g / A g C l ) I er 4. Results and Discussion Raw Sewage Characteristics Date Concentrations (mg/l) C O D (sol.) C O D (tot.) P04 T P NOx T K N TSS Feb.19/88 83 143 2 3.1 0 19.2 30 Feb.20/88 103 135 2 2.9 0 18.4 N A Feb.21/88 82 163 2.2 3 0 18.8 N A Feb.22/88 91 115 2.2 2.8 0 18 32 Feb.23/88 95 135 2.1 2.9 0 18.4 39 Feb.24/88 71 104 2.1 2.9 0 18.4 33 Feb.25/88 79 136 2.1 3.1 0 20 59 Feb.26/88 83 118 2.2 3.1 0 19.2 44 Feb.27/88 79 106 2.1 2.7 0 14.5 23 Feb.28/88 84 128 2.1 2.9 0 14.5 46 Feb.29/88 94 126 2.2 2.9 0 14.1 35 Mar.1/88 111 181 2.4 3.6 0 16.9 99 Mar.2/88 65 132 2.3 3.6 0 16.9 67 Mar.3/88 6T 130 2.7 3.7 0 22.2 62 Mar.4/88 86 136 2.8 3.5 0 20.4 54 Mar.5/88 76 132 2.7 3.5 0 20.7 48 Mar.6/88 120 209 2.7 3.5 0 20 28 Mar.7/88 56 115 2.7 3.5 0 20.4 63 Mar.8/88 56 91 2.7 3.2 0 18.5 28 Mar.9/88 53 132 2.7 3.4 0 18.9 55 Mar.10/88 55 100 2.8 3.3 0 20 35 Mar.11/88 47 123 1.8 3 0 20 74 Mar.12/88 51 107 2 2.8 0 19.6 53 Mar.13/88 53 94 1.8 2.9 0 19.6 46 Mar.14/88 53 115 1.8 2.8 0 19.2 49 Mar.15/88 51 135 1.8 2.8 0 19.6 66 Mar.16/88 59 121 2 2.8 0 18.1 56 Mar.17/88 66 127 2 2.7 0 18.5 51 Table 4.7: Phase Three Raw Sewage D a t a (Feb.l9/88-Mar.l7/88) Chapter 4. Results and Discussion Anaerobic Holding Vessel Characteristics Date Concentrations (mg/l) C O D (sol.) C O D (tot.) P O 4 T P NOX Feb.19/88 83 143 2 3.1 0 Feb.20/88 N A N A N A N A 0 Feb.21/88 32 N A 8 8.9 0 Feb.22/88 N A N A N A N A 0 Feb.23/88 N A N A N A N A 0 Feb.24/88 N A NA. N A N A 0 Feb.25/88 79 136 2.1 3.1 0 Feb.26/88 N A N A N A N A 0 Feb.27/88 35 N A 7.1 N A 0 Feb.28/88 49 N A 6.8 N A 0 Feb.29/88 47 N A 6.7 N A 0 Mar.l/88 53 N A 2.3 N A 0 Mar.2/88 36 N A 3.9 N A 0 Mar.3/88 37 N A 7.3 N A 0 Mar.4/88 39 N A 9 N A 0 Mar.5/88 70 N A 9.8 N A 0 Mar.6/88 43 N A 9.7 N A 0 Mar.7/88 31 N A 10.3 N A 0 Mar.8/88 37 N A 9.8 N A 0 Mar.9/88 31 N A 8.9 N A 0 Mar.10/88 31 N A 7.8 N A 0 Mar.11/88 29- N A 8 N A 0 Mar.12/88 31 N A 6.8 N A 0 Mar.13/88 41 N A 6.7 N A 0 Mar.14/88 53 N A 6 N A 0 Mar.15/88 53 N A 5.3 N A 0 Mar.16/88 66 N A 6 N A 0 Mar.17/88 41 N A 5.8 N A 0 Table 4.8: Phase Three Anaerobic Holding Vessel D a t a (Feb.l9/88-Mar.l7/88) Chapter 4. Results and Discussion 87 Effluent Characteristics Date Concentrations (mg/l) COD (sol.) PO< TP NOs TKN TSS Feb.20/88 26 0.15 1 7.5 2.9 NA Feb.21/88 20 0.35 1 7.5 2.5 NA Feb.22/88 26 0.2 1.2 8.3 2.5 NA Feb.23/88 18 0.3 1 8.3 3.3 36 Feb.24/88 39 0.7 1.2 7.2 2.9 29 Feb.25/88 35 0.6 1.9 8.3 4 49 Feb.26/88 22 1.8 2.9 8.5 3.3 42 Feb.27/88 35 2.1 3.4 8.5 4.8 33 Feb.28/88 29 2.4 3.4 9.3 3.7 40 Feb.29/88 31 2.2 3.4 9 4.1 55 Mar.I/88 30 1.7 2.9 7.3 3.6 49 Mar.2/88 18 0.7 2.3 8.2 4.1 72 Mar.3/88 31 0.5 0.9 8.5 2.5 36 Mar.4/88 31 0.3 1.3 8.2 2.9 40 Mar.5/88 29 0.3 0.9 8.3 2.9 36 Mar.6/88 25 0.3 3.2 8.6 3.6 58 Mar.7/88 27 0.3 1.5 7.9 3.6 48 Mar.8/88 25 1 2.7 8.2 4.4 79 Mar.9/88 NA 1 NA 8.6 NA NA Mar.10/88 31 1.8 2.8 8.4 2.5 34 Mar.11/88 20 2 3.3 9 5.4 56 Mar.12/88 25 1.8 3.3 8.4 2.5 41 Mar.13/88 25 1.1 2.2 8.2 2.9 49 Mar.14/88 20 0.7 1.6 8 2.5 37 Mar.15/88 25 1.8 3.2 7.8 4 61 Mar.16/88 29 2.2 4.3 8.5 9.6 136 Mar.17/88 33 1.7 2.2 8.3 2.2 34 Table 4.9: Phase Three Effluent Data (Feb.20/88-Mar.l7/88) CD 00 u -a 9 3 8 8-8 o o <u +-> ecj A fr m O fl fr o +-> O -P fl ej l-H «H <H W fl ed +-> fl t—I F i g . 4 . 3 1 P h a s e T h r e e I n f . a n d E f f , O r t h o - P v s T i m e f r o m F e b . 1 9 / 6 8 t o M a r . 1 7 / 8 8 Influent Ortho-P Time (days) since Feb. 19/88 Chapter 4. Results and Discussion 90 O) I n 3 O .a T3 9 "3 8 ft* -to 8-8 F i g . 4 . 3 3 P h a s e T h r e e P e r c e n t P r e m o v a l v s T i m e f r o m F e b . 1 9 / 8 8 t o M a r c h 1 7 / 8 8 •a > o 3 4) U tn ? u o fl fr O fl PH 4-> <v V 100 90 H 80 H 70 H 60 H 50 H 40 30 H 20 H 10 Time (days) siilce Feb. 19/88 Chapter 4. Results and Discussion 92 Following the experimental runs, the biofilm was scraped from the discs and weighed. The dried biofilm weight was 66 grams. Thirty-seven grams (56%) of the total mass of the biomass was volatile. Table 4.10 shows %P contents of the biofilm for the entire experimental period. Chapter 4. Results and Discussion 93 D a t e B i o f i l m %P content Bi o f i l m % P content (total solids basis) (volatile solids basis) Dec. 17/87 1.84 3.29 J a n . 6/88 2.04 3.64 J a n . 13/88 2.39 4.27 J a n . 20/88 2.30 4.11 J a n . 25/88 2.27 4.05 Feb. 1/88 2.00 3.57 Feb. 10/88 2.3 4.11 Feb. 23/88 2.3 4.11 M a r . 1/88 2.27 4.05 M a r . 8/88 2.4 4.29 Table 4.10: B i o f i l m percent phosphorus contents for experimental period Chapter 4. Results and Discussion 94 4.2 Data Analysis The computer package SAS (statistical analysis system) version 5 was used to get a correlation matrix of all the parameters measured in these experiments. Parameters having high correlation coefficients were then used to explain the results obtained. From the correlation matrix, a relationship was found to exist between anaerobic P release and aerobic P uptake. Multiple regression analysis showed a linear relationship between influent TP, anaerobic ORP, and anaerobic P release. Figure 4.34 shows the scatter plot for anaerobic phosphorus release versus aerobic phosphorus uptake. The best fit line for the data is also provided. Both anaerobic P release and aerobic P uptake are calculated as the difference in initial and final P04 concentrations during the respective phases of the treatment cycle. R2 for the best fit line was found to be 0.765. In other words, 76.5% of the aerobic P uptake ob-served during these experiments could be explained by anaerobic P release. The higher the anerobic P release, the higher the subsequent P uptake under aerobic conditions. Anaerobic P release has been reported by Barnard (1976) and others to be one of the prerequisites for good phosphorus removal. Previous research into the mechanism of biological phophorus removal has determined anaerobic P release to be dependent on the availability of simple carbons, and the absence of nitrates and oxygen in the anaerobic phase. The correlation matrix generated using SAS showed weak correlations between influent NOx and anaerobic P release, and influent acetate COD and anaerobic P release. It was felt that the poor correlations above were due in part to the small ranges of NOx and influent acetate concentrations observed in these experiments. It is also felt that, in the range of acetate concentrations used in these experiments, acetate was never limiting to the biological phophorus removal mechanism. Therefore, small variations in influent acetate concentration would not effect anaerobic P release. Chapter 4. Results and Discussion 96 Influent T P gave the best correlation of all the chemical data provided. R2 for P release versus influent T P (total phosphorus) was 0.41. The above suggests that P was in fact limiting to the biological phosphorus mechanism for the amount of acetate added to the anaerobic phase. Influent NOx, P 0 4 , C O D , T P , T K N and acetate concentrations were correlated with anaerobic P release. The R2 value for the above multiple linear regression was found to be 0.556. Again, the most significant parameter was influent T P . Batch test O R P values were then correlated with anaerobic P release. The O R P values after 15, 30, 60 and 90 minutes of anaerobic time were used. The correlation matrix showed an R2 value of 0.45 between the O R P after 30 minutes and anaerobic P release. The O R P readings at 60 and 90 minutes gave similar correlation cofficients. The correlation between P release and O R P after 15 minutes was not as good. A multiple linear regression was then run for P release versus influent T P and anaerobic ORP . R2 for the above regression was found to be 0.654. Considering the facts that the O R P values were taken with two different probes, and only sixteen data points were available for the regression, this was considered to be a reasonable fit to the data. To summarize, the correlation coefficient for anaerobic phosphorus release and aer-obic phosphorus uptake was found to be 0.765. The correlation coefficient for influent T P and anaerobic O R P versus anaerobic phosphorus release was 0.654. Considering the facts that the above parameters were not purposely varied to determine their in-terrelationships, and that the number of data points was small, the above correlations are quite good. Chapter 4. Results and Discussion 97 4.3 Discuss ion The data collected between September 1/87 and February 1/88 suggests that consistent enhanced biological phosphorus removal is possible using a Sequencing Batch Rotating Biological Contactor(SBRBC). However, data collected from Feb.6/88-March 17/88 appears to contradict the above claim. The discussion section provides 'reasonable explanations' for the data observed. The average percent phosphorus removal between November 11/87 and Feb. 1/88 was 94.8%. One hundred percent phosphorus removal was observed on occasion. The average value includes the data collected on November 30 following three cycles with no acetate addition, as well as, the time period from November 11 to November 20, 1988, when it appeared that steady state with respect to phosphorus removal had not yet been reached. Even without acetate addition for three cycles, phosphorus removal was 72%. This suggests that some volatile fatty acids were being provided by the sewage or that extra stored P H B ' s were available for more than one cycle. Steady state with respect to C O D removal appeared to be reached more quickly than for phosphorus removal (see Figure 4.7). The fluctuations in C O D removals may have been caused by the variability in raw total C O D values. It is difficult to tell whether nitrification reached steady-state during phase one of these experiments (Sept. 1/87-Feb. 1/88). Figure 4.8 appears to show an increasing trend in percent nitrification over time. However, percent nitrification for the remainder of the experiments did not increase appreciable above the levels observed during phase one. Since the reactor contents were emptied after each aerobic period, the presence of NOx in the anaerobic phase of the treatment process was not expected. It is likely that lower biofilm layers did not receive any oxygen during the aerobic portion of the treatment cycle, due to oxygen depletion in the upper biofilm layers. Wi th NOx Chapter 4. Results and Discussion 98 available, denitrification likely resulted. It would be difficult however, to determine how much denitrification actually took place in the biofilm. The one fact that appears to come out of the results up until February 1, 1988 is that enhanced biological phosphorus removal using a 'fixed film process' is definitely possible on a short-term basis. The batch test results presented in Figures 4.3 to 4.6 and Figures 4.9 to 4.11, were typical for the above time period. Each batch test showed an anaerobic phosphorus release upon acetate addition followed by an aerobic uptake in excess of metabolic requirements. Looking at the plots of P(9 4 versus time (Figures 4.3 and 4.9) it is apparent that the values of bulk liquid PO4 at 60 minutes and 90 minutes of the anaerobic phase did not vary appreciable. The bulk liquid P04 concentrations at the 180 minute and 280 minute marks for the above plots did not vary appreciable either. The above facts suggest that the anaerobic time might have been shortened by 30 minutes and the aerobic time by 100 minutes, with treatment efficiences for phosphorus similar to the ones observed. In enhanced biological phosphorus removal treatment plants, phosphorus is removed from the system with the phosphorus rich biomass. In the activated sludge process, controlled sludge wasting is practised to remove the phosphorus from the system. In a fixed-film system such as the S B R B C biofilm sloughing provides the required solids wasting. However, unlike the sludge wasting in activated sludge plants, biofilm slough-ing is uncontrolled and is dependent on sewage strength, oxygen availabilty, and disc rotational speed. It is this lack of control that worried some my colleagues prior to the start up of this project. In particular, the concern was that the required steady state biofilm %P, at the low wasting rate, would be in excess of the phosphorus stor-age capacity of the bacteria responsible for biological phosphorus removal, resulting in system failure. The accumulation of phosphorus beyond the systems capacity did not appear to Chapter 4. Results and Discussion 99 be a problem. One of the reasons for this could be the apparent P limitation that was indicated by the statistical analysis. Influent T P gave the best correlation with anaerobic P release. Higher levels of influent T P corresponded to higher levels of anaerobic P release. Influent acetate C O D did not correlate well with anaerobic P release. As discussed in the data analysis section of this thesis, the poor correlation may have been due to the fact that the amount of acetate added was in excess of that required to remove the amount of P present in the bulk liquid. In fact, lower than expected values of biofilm percent P were observed in these experiments, when percent P was measured on a total solids basis. When percent P in the biofilm was reported on a volatile solids basis, the 3.5 to 4.5 % phosphorus content expected was observed. It is believed that the R B C biofilm acted as an inorganic solids trap. The measurement of the total and volatile solids contents of the biofilm supports the above argument. Only 56% of the total biofilm mass was volatile, as opposed to a typical value of 80% for activated sludge systems operating in a phosphorus removal mode. Therefore, it appears that the biofilm percent P for these experiments is comparable to values expected for enhanced biological phosphorus removal activated sludge plants, when the comparison is made on a volatile solids basis. Data collected during phase two (Feb. 6-Feb. 16/88) and phase three (Feb. 19-March 17/88) appear to contradict the claim for consistent enhanced biological phosphorus removal using an S B R B C . Each time period gave inconsistent removals for apparently slightly different reasons. During the second phase of the experiments the anaerobic draw flow was returned directly to the raw sewage vessel. The sewage vessel was cleaned on Feb. 6 and again on Feb. 12, 1988. Percent phosphorus removal was high after the above vessel was cleaned and refilled with fresh sewage. However, percent phosphorus removal declined with time following cleaning of the raw sewage vessel. Referring to Figure 4.14, showing the Chapter 4. Results and Discussion 100 influent soluble C O D versus time for the Feb. 12 to Feb. 16 portion of phase two, a steady decrease in soluble C O D concentration is observed after each sewage addition. Between Feb. 15 and Feb. 16 the above decrease was almost 30 mg/ l . The soluble C O D data for the raw sewage vessel between Feb. 6 and Feb. 11 is not available. However, the total C O D data for this time period shows a similar trend to the influent soluble C O D data for Febl2 to Feb. 16/88. Part of the reason for the decrease in feed vessel (raw sewage vessel) soluble C O D , between raw sewage additions, was likely the dilution effect caused by returing the anaerobic draw volume directly to this vessel.Aerobic C O D reduction likely took place at the start of the anaerobic phase, due to the presence of air entrained in the influent during the fill period. Influent air entrainment is indicated by all of the batch test O R P plots provided. For each O R P trace, there was a brief period at the beginning of the anaerobic cycle in which the O R P value was either positive or increased prior to the steady drop in O R P expected.Storage of soluble C O D and some anaerobic fermentation also reduced the C O D of the anaerobic draw being returned to the feed after each anaerobic time period. Some bacterial reduction of C O D in the feed tank may have also taken place. The source of the above bacteria was believed to be the sloughed solids returned to the feed tank in the anaerobic draw. A n increase in the phosphate concentration in the feed tank between raw sewage additions was also observed during phase two. This was as expected, for phosphorus would be released during the anaerobic phase of the treatment cycle. The anaerobically treated wastewater returned to the feed tank would therefore have a higher phosphate concentration than the wastewater entering the R B C reactor at the begining of the treatment cycle. The hypothesized sequence of events responsible for the progressive decrease in percent phosphorus removal, following each feed tank cleaning during phase two, is as follows. A t time zero, the feed tank was cleaned and filled with raw sewage. Percent Chapter 4. Results and Discussion 101 phosphorus removal for the first two or three treatment cycles was quite good. A t the start of each treatment cycle, some aerobic C O D reduction would likely take place in the R B C reaction vessel due to the presence of entrained air introduced during the fill period. Soluble C O D storage by 'bio-P' bacteria and some anaerobic fermentaion would also reduce the C O D of the bulk liquid in the reaction vessel. Soluble C O D storage by 'bio-P' bacteria would result in phosphorus release. Biofilm sloughing would also take place during this phase. The anaerobic draw volume reintroduced into the feed tank following the anaerobic phase, would have a lower C O D concentration, and higher phosphate concentraion than the wastewater originating from the feed vessel during the time the R B C reaction vessel was filled. The dilution of carbon strength and concentration of phosphate in the feed tank would be the result. The introduction of sloughed solids to the feed tank, during the anaerobic draw period, could only contribute to the above. Sloughed solids introduced into the feed vessel could use up more soluble C O D for carbon storage with resulting phosphorus release. Some anaerobic fermentation in the feed tank may also have taken place. Over time, the oxygen demanding capability of the R B C influent sewage would be reduced as a result of the dilution of carbon strength. The proportion of the influent soluble C O D being used by the aerobic bacteria to remove entrained air, introduced during the pumping cycle, would increase. Therefore, less soluble C O D would be available for carbon storage by 'bio-P' bacteria at a time when the influent loading of phosphorus was increasing. When the feed vessel was cleaned and refilled with fresh sewage, the carbon strength was once again high enough to satisfy the demands of influent air entrainment depletion, and of necessary anaerobic carbon storage. However, the progressive dilution of the feed tank carbon strength would result in the decrease in the oxygen demanding capability of the influent, and therefore percent phosphorus removal. When comparing this situation to that encountered in phase one of these experiments, it is noted that only 11.5 liters of Chapter 4. Results and Discussion 102 anaerobic draw was returned to the raw sewage vessel daily in the phase one case. In phase two, more than four times the 11.5 liters was returned to the feed tank daily. A n increase in the volume of anaerobic draw returned to the feed tank would result in a subsequent increase in the number of sloughed solids in this tank, in addition to the dilution of carbon strength. The reader is also referred to the model for anaerobic P release formulated in the data analysis section. Anaerobic phosphorus release was found to be dependent on influent T P and anaerobic ORP . The presence of entrained air at the beginning of the treatment cycle would explain the poor correlation of influent soluble C O D versus anaerobic P release. The data collected during phase three of the experiments can be explained in a sim-ilar manner to the above. To avoid the problem of flow accumulation in the anaerobic holding vessel, the pumping rate from this tank was doubled. Pumping the anaerobic holding vessel 'dry' ensured a degree of aeration in the reactor prior to the anaerobic phase. Influent samples taken on March 9 and March 16, 1988 showed NOx concentra-tions of 0.2 mg/l . Figures 4.26 and 4.29 show a slow rate of O R P decrease for both the above batch tests. In fact, both tests showed very low anaerobic phosphorus release, indicating poor anaerobic conditions. It is therefore suspected that poor anaerobic con-ditions were again set up during the above times, accounting for the resultant decrease in phosphorus removal. Figures 4.30 to 4.33 lend even more credence to the above argument. In Figure 4.30, both influent and raw phosphate concentrations are plotted with time. The concentra-tion of phosphate in the raw sewage vessel appears relatively constant. The variation in influent phosphate concentration can therefore be explained by the variation in the concentration of this species in the anaerobic holding vessel. The concentration of Chapter 4. Results and Discussion 103 phosphate in the anaerobic holding vessel would be governed by the degree of anaero-bic phosphorus release taking place in the reaction vessel during the anaerobic phase of the treatment cycle. Figure 4.31 shows that the lowest effluent phosphate concentrations correspond to the highest influent concentrations. The reader is referred to the argument in the above paragraph. Since the variation in influent phosphate concentration was due to the variation in the concentration of phosphate in the anaerobic holding vessel, and therefore the degree of anaerobic P release, the poor phosphorus removals can once again be traced to poor anaerobic P release. Periods of high anaerobic P release would result in carbon storage as P H B and an increase in the concentration of phosphate in the anaerobic holding vessel. The good anaerobic P release would result in good aerobic P uptake, as discussed in the data analysis section of this thesis. FigUre 4.32 shows influent and raw soluble C O D values with time. It is noted here that the high values of influent C O D correspond to the high percent phosphorus re-movals. High influent C O D gives a high immediate oxygen demand and good anaerobic conditions. Good anaerobic conditions, and available volatile fatty acids result in good phosphorus release. As discussed above, good anaerobic phosphorus release, with it's resulting carbon storage, results in good aerobic phosphorus uptake later. To summarize, the proposed S B R B C process appears to hold promise for biological phosphorus removal from wastewater. In each test situation in which high effluent PO4 values were observed, the cause of the problem could be traced to improper anaero-bic conditions during the anaerobic phase. The problems caused in phase two seem to be due to the progressive dilution of the feed tank carbon strength by the direct return of the anaerobic draw to this tank. The introduction of sloughed solids into the feed tank may have also caused some problems. The results from phase three of these experiment show variable results. Poor phosphorus removals corresponded to low Chapter 4. Results and Discussion 104 influent soluble C O D . The extra air entrained when the anaerobic holding vessel was pumped dry in this phase may have caused many of the problems during this time period. Nitrates were found in the reactor influent, at the start of the anaerobic phase, on ocassion. Batch tests showed that the runs with poor overall phosphorus removal exhibited poor anaerobic phosphorus release. The data analysis supports the above ar-guments. Anaerobic phosporus release correlated well with aerobic phosphorus uptake. Good anaerobic phosporus release was dependent on influent T P and anaerobic ORP . Chapter 5 Conclusions and Recommendations Conclusions drawn from this research are presented in the first section of this chapter. Recommendations for future research and for process improvements are then discussed. 5.1 Conclusions The main objective of this research from the outset was to study the technical feasibility of using a Sequencing Batch Rotating Biological Contactor ( S B R B C ) for 'enhanced biological phosphorus' removal. It was felt that, if the proposed R B C process proved to be technically feasible, that some of the 'economic benefits' inherent to R B C technology might be realizable for the removal of phosphorus as well as carbon. The above research project appears to have been reasonably successful, considering the fact that this was a first attempt at the above problem. Below is a list of the major conclusions that have been derived from this research. (1) Enhanced biological phosphorus removal is definitely possible using a Sequencing Batch Rotating Biological Contactor ( S B R B C ) . (2) The attainment of proper anaerobic conditions during the anaerobic phase of the process appeared to be the key to making the process work. A i r entrained during the influent period of the treatment cycle enabled aerobic bacteria to use added acetate and influent soluble C O D . The utilization of these materials by aerobic bacteria left less for carbon storage by 'bio-P' bacteria, and resulted in lower percent phosphorus removals. 105 Chapter 5. Conclusions and Recommendations 106 (3) Anaerobic biofilm sloughing may cause problems if the anaerobic draw is re-turned to the sewage feed tank. (4) A n alternative to the anaerobic holding vessel appears to be necessary to mini-mize operating problems with the S B R B C . (5) Carbon removal and nitrification are secondary benefits to the S B R B C process for phosphorus removal. (6) Greater anaerobic phosphorus release resulted in greater aerobic phosphorus uptake. (7) Anaerobic phosphorus release was dependent on influent total phosphorus and anaerobic O R P in these experiments. The facts that influent T P correlated with anaer-obic P release, and influent acetate did not, suggested that P was limiting in these experiments. 5.2 Recommendations The most important question regarding S B R B C technology appears to have been an-swered. The process appears to work. The next logical step for future research is to provide alternatives to the anaerobic holding vessel. Pilot scale studies could be con-ducted to investigate the possible economic advantages of the above process. It is felt by the author that the S B R B C technology may share many of the economic advantages of the conventional R B C — namely lower operating costs, less sludge production, and ease of operation. The above have yet to be proven. The solution to the flow accumulation problem and the use of an anaerobic holding vessel could be solved using two S B R B C units. The actual raising and lowering of the R B C discs, instead of flow recycle, may be a worthwhile alternative. Both the above alternatives would use a microprocessor, solenoid valves, and possibly level switches to Chapter 5. Conclusions and Recommendations 107 achieve process control. Other modifications to the S B R B C process are possible, and might be worthwhile investigating. In the process option involving two S B R B C units, both units would be operated in a sequencing batch mode. One unit would be operated under anaerobic conditions at the same time as the other unit was being operated under aerobic conditions. After a predetermined reaction time in the aerobic R B C , a valve would be opened to remove the treated effluent from the reactor. Following the effluent decant, part of the contents of the anaerobic reactor would be transfered to the now empty aerobic reactor. The aerobic reactor would then be filled to capacity, submerging the rotating discs, with raw sewage. The past aerobic reactor would now become the anaerobic reactor. The past anaerobic reactor, with its partialy submerged discs, would now become the aer-obic reactor. The above series of events would continue with each reactor alternating between anaerobic and aerobic conditions. Acetate or some other suitable volatile fatty acid source would be added to the anaerobic phases as required. A l l valves and pumps would be controled by a microprocessor. The experience gained in these experiments could also be very useful from a process operation point of view. To reduce the likelihood of experiencing the same problems encountered in these experiments, during the anaerobic period of the treatment cycle, it is recommended that acetate be added to the anaerobic phase only after the O R P in this phase has reached a predetermined value, indicating true anaerobic conditions have been attained. The above would also prevent added acetate from being used by aerobic bacteria, present in the biofilm, in the presence of reactor influent entrained air. Bibliography [1] Alleman J.E., and Irvine R.L. "Nitrification in the sequencing batch biological reactor." Journ. Water Pollut. Control Fed. , 52:11:2747(1980) [2] Alleman J.E., Veil JA, and Canaday JT "Scanning Electron Microscopic Evalua-tion of Rotating Biological Contactor Film" Water Res. 16, 543 (1982) [3] Amer. Public Health Assn. (A.P.H.A.).Standard Methods:for the Examination of Water and Wastewater. Washington, D.C.: A.P.H.A., 13th ed., 1971. [4] Antonie R.L., Kludge D.L., and Mielke J.H. (1974) "Evaluation of.a Rotating disc wastewater treatment plant." Journ. Water Pollut. Control Fed.. 46, 3, 498-511 [5] Appledoom K.J. , and Deinema M.H. "Biological Phosphorus Removal under de-fined conditions in a fill-and-draw system." [6] Barnard J.L.,"Cut P and N without Chemicals."Water and Wastes  Engineering,11(1974) ,33-36 [7] Barnard J.L., "A Review of Biological Phosphorus Removal in the Activated Sludge Process." Water S.A.. 2 (1976), 3, 136-144 [8] Chittenden J.A. and Wells J.W. "Rotating Biological Contactors following Anaer-obic Lagoons." Journ. Water Pollut. Control Fed.. 43 (1971), 5, 746-754. [9] Comeau Y. (1984) "Biochemical Models for Biological Excess Phosphorus Removal from Wastewater." Masters thesis The University of British Columbia, Vancouver, Canada. 108 Bibliography 109 [10] Comeau Y., Hall K.J. , Hancock R.E.W., Oldham W.K., (1986) "Biochemical Mod-els for Enhanced biological phosphorus removal." Water Res.. 20, 1511-1521. [11] Davies T.R. and Pretorius W.A. (1975) "Denitrification with a bacterial disc unit ."Water Res.. 9, 459-463. [12] Eckenfelder W. (1987) "Biological phosphorus removal: State of the Art Review.",Pollution Engineering, Sept.(1987), 88-93. [13] Fletcher Madilyn (1979) "The Attachment of Bacteria to Surfaces in Aquatic Environments" Adhesion of Microorganisms to Surfaces Elwood, Melling and Rut-ter(ed.)(1979) , 87-108. [14] Friedman A.A., Tait S.J. " Energy recovery from anaerobic rotating biological contactor (AnRBC) treating high strengh carbonaceous wastewaters." from Pro-ceedings:" First National Symposium/Workshop on Rotating Biological Contactor Technology" at Champion, Pennsylvania: University of Pittsburg, 1980, page 759. [15] Fuhs G.W. and M. Chen (1975) "Microbilogical Basis of Phosphate Removal in the Activated Sludge Process for the Treatment of Wastewater."Microbial Ecology,2, 119-138. [16] Hoa O, Hendricks G F,"Rotating biological contactors remove nutrients."Water  Sewage Works. 122(10):70 (Part 1) and 122(11):48 (Part Two). [17] Hoag G.E., Widmer W.J., and Havey W.H-"Microfauna and RBC performance: Laboratory and Full-Scale systems."from Proceedings:First International Con-ference on Fixed-Film Biological Process," at Kings island, Ohio: University of Pittsburg, 1982, page 206. Bibliography 110 [18] Hynek R.J., Iemura PL-"Nitrogen and phosphorus removal w i t h rotating biological contactors." from Proceedings: " F i r s t National Symposium/Workshop on Rotat-i n g Biological Contactor Technology." at Champion, Pennsylvania: University of P i t t s b u r g , 1980, page 295. [19] Ketchum L.H., Irvine R.L., Breyfogle R.E., M a n n i n g J.F. (1987) Jo u r n . Water P o l l u t . Control Fed., 59, 1. [20] K i n n e r et a l . , "An eletron microscopic evaluation of bacteria inhabitib-i n g r o t a t i n g biological contactor biofilms during various loading conditions." E n v i r o n . Technol. Lett.T 6, 455 (1985). [21] Knoetze C , Davies T.R., a n d Weichers S.G. : "Chemical Inhibition of biological nutrient removal processes.", Water SA.T 6(4), 171 (1980). [22] K u b o t a , Patent : T h e treatment of wastewater containing nitrogenous material by nit r i f i c a t i o n and denitrification i n the same vessel using a rota t i n g disc.", patent as-signee: ( K U B I ) K u b o t a KK.,Japanese patent numbers #55035964 and #8302633. [23] L e v i n G.V. and J.Shapiro (1965) "Metabolic uptake of Phosphorus by Wastewater Organisms.'*, Journ. Water P o l l u t . Control Fed.T 37, 6, 800-821. [24] Lundberg L.A., Pierce J.L., "Comparative cost-effectiveness analysis of rotating biological contactor and activated sludge processes for carbon oxidation." from Pro-ceedings:First National Symposium/Workshop on Rotating Biological Contactor Technology, at Champion, Pennsylvania: University of P i t t s b u r g , 1980, page 1413. [25] M a n n i n g J.F. and Irvine R.L, (1985) "The Biological Removal of Phosphorus i n a Sequencing B a t c h Reactor.", Journ. Water P o l l u t . Control Fed.. 57, 87. Bibliography 111 [26] Marias G.v.R., R.E. Lowenthal, and I. Siebritz (1983) "Review: Observations Sup-porting Phosphate Removal by Biological Excess Uptake.",Water Sci. Technol.T 15,15-41. [27] Masudo S., Watanabe Y., Ishiguro M.: "Simultaneous ni t r i f i c a t i o n and denitrifi-cation i n a r o t a t i n g biological contactor.", from Proceedings: F i r s t International Conference on F i x e d - F i l m Biological Processes., at Kings Island Ohio: Unversity of P i t t s b u r g , 1982, page 802 [28] M i l b u r y W.F., D. McCauley and C H . Hawthorne (1971) "Operation of Conventional A c t i v a t e d Sludge for M a x i m u m Phosphorus Removal.", Jour. Water P o l l u t . Control Fed.T 43, 9, 1890-1901. [29] M u r p hy K.L., Sutton P.M., Wilson R.W., and Jank B.E. ^ " N i t r o -gen Control: Design Considerations for Suspended Growth Systems." Jo u r n . Water P o l l u t . Control Fed.T 49, 4, 549-547 (1977) [30] Nicholls H.A. and D.W. Osborn (1979) "Bacterial Stress: a prerequisite of Biolog-i c a l Removal of Phosphorus." TJour. Water P o l l u t . Control Fed., 51, 3, 577-569. [31] N i l s s o n I., and M . Dostalek (1984) "Estimation of the biofilm formation a b i l -i t y of Pseudomonas p u t i d a i n descrete samples from continuous culture.", A p p l i e d Microbiology and Biotechnology 20: 183-188. [32] Noss C.L, M i l l e r R.D., Smith E.D., "Recarbonation of wastewater using the RBC"*, from Proceedings: F i r s t National Symposium/Workshop on Rotating B i -ological Contactor Technology, at Champion, Pennsylvania: University of P i t t s -b urg, 1980, page 861. Bibliography 112 [33] Odegaaxd H., Rusten B.: "Nitrogen Removal i n rotating biological contactors with-out the use of an external carbon source."from Proceedings: F i r s t National Sym-posium/Workshop on Rotating Biological Contactor Technology, at Champion, Pennsylvania: University of P i t t s b u r g , 1980, page 1301. [34] Okada M., Murakami A., and Sudo R. "Ecological Selection of Phosphorus-Accumulating. Bacteria i n Sequencing B a t c h Reactor Activated Sludge Processes f o r Simultaneous Removal of Phosphorus, Nitrogen and Organic Substances.", from Proceedings of the I A W P R C Specialized Conference Held i n Rome Italy, 28-30 Sept., 1987 on Biological Phosphate Removal from Wastewater. [35] Ouyang C.F., "The characteristics of rotating biological contactor sludge." from Proceedings: F i r s t National Symposium/Workshop on Rot a t i n g Biological Con-tactor Technology, at Champion, Pennsylvania: University of P i t t s b u r g , 1980, page 189. [36] Pretorius W.A. (1971) "Some Operational Characteristics of a bacterial disc unit.", Water Res.T 5T1141-1146. [37] Rabinowitz B. and G.V.R. Marais (1980) "Chemical and Biological Phosphorus Removal i n the Activated Sludge Process.*, Research report W32, Dept. of C i v i l Eng., University of Cape Town. [38] Rabinowitz B. (1985) "The Role of Specific Substrates i n Excess Biological Phos-phorus Removal.", P h D thesis The University of B r i t i s h Columbia, Vancouver, Canada. [39] Schroeder E., Water and Wastewater Treatment., New York: M c G r a w - H i l l , 1977. Bibliography 113 [40] Singhal A.K., "Phosphorus and nitrogen removal at Cadillac, Michigan.", Jour. Water Pollut. Control Fed.. 52 (1980), 2761. [41] Smith A.G., Khettry R.K., "Nitrification/denitrification studies with rotating bio-logical contactors.", from Proceedings: First National Symposium/Workshop on Rotating Biological Contactor Technology, at Champion, Pennsylvania: University of Pittsburg, 1980, page 1319. [42] Smith E.D., Miller R.D. Wu Y.C. (eds)."Proceedings First National Sympo-sium/Workshop on Rotating Biological Contactor Technology," at Champion, Pennsylvania: University of Pittsburg, 1980, 1545 pp. [43] Smith E.D., Bandy J.T.,"A history of the RBC process.", from Proceedings: First National Symposium/Workshop on Rotating Biological Contactor Technology, at Champion, Pennsylvania: University of Pittsburg, 1980, page 11. [44] Smith E.D., Bandy J.T., "State of knowledge for rotating biological contactor technology."from Proceedings: First International Conference on Fixed-Film Bio-logical Processes, at Kings Island, Ohio: University of Pittsburg, 1982, page 1. [45] Srinath E.G., C A . S as try and S.C. Pillai, "Rapid removal of Phosphorus from sewage by activated sludge." Water and Waste Treatment, 11 (1959), 410. [46] Strom P.F. and Chung J.C., "The Rotating Biological Contactor in Wastewater Treatment.",Adv. Biotechnological Processes, 5 (1985), 193. [47] Sullivan R.A., Hynek R.J., "Upgrading existing waste treatment facilities utilizing the Biosurf process.", from Proceedings: First National Symposium/ Workshop on Rotating Biological Contactor Technology, at Champion, Pennsylvania: University fo Pittsburg, 1980, page 1015. Bibliography 114 [48] Supelco. "Separating Aqueous Free Carboxylic Acids (C2-C5) at ppm Concentra-tions.'' Bellafonte, PA: Supelco Inc., Bull. 751E, 1982. [49] Technicon Autoanalyzer EL "Nitrate and Nitrite in Water and Wastewater (Range: 0-2.0 mgN/1)." Tarrytown, N.Y.: Technicon Lnd. Systems, Ind. Method No. 100-70W, Sept. 1973. [50] Technicon Autoanalyzer DT. "Orthophosphate in Water and Wastewater (Range: 0-10 mg/l) ."Tarrytown, N.Y.: Technicon had. Systems. Lnd. Method No. 94-70W, July 1973. [51] Technicon Autoanalyzer IL"Phosphorus/BD Acids Digests-Range: from 0-0.40 mg/l to 0-25 mg/l." Tarrytown, N.Y.: Technicon Lnd. Systems. Lnd. Method No. 327-73W, Sept. 1974. [52] U.S. Environ. Protection Agency (EPA). Design Information on Rotating  Biological Contactors.T Washington, D.C: U.S. EPA, Office of Technol. Trans-fer, Publ. No. EPA-600/2-84-106, 1984. [53] Vacker D., C H . Connell and W.N. Wells, "Phosphate removal through Municipal Wastewater Treatment at San Antonio, Texas." Journ. Water Pollut. Control Fed., 39 (1967) ,5,750-771. [54] Wells W.N., "Differences in Phosphate Uptake Rates Exhibited by Activated Sludges", Journ. Water Pollut. Control Fed.. 41 (1969), 5, 765-771. [55] Wentzel M.C., Lotter L.H., Lowenthal R.E., and Marais G.v.R. "Metabolic Be-haviour of Acinetobacter Spp. in enhanced biological phosphorus removal-a bio-chemical model.".Water SA. 12:4 (1986). Bibliography 115 [56] Wu Y.C., Smith E.D., Miller R.D., Optaken E.J. (eds) :"Proceedings: First In-ternational Conference on Fixed-Film Biological Processes," at Kings Island, Ohio: University of Pittsburg, 1982, 1859pp. Appendix A Statistical Analysis The statistical analysis for this research was done using version five of the S AS computer package. S AS (Statistical Analysis System) is a software system for information storage and retrieval, data modification and programming, report writing, statistical analysis, and file handling. The statistical analysis procedures range from simple descriptive statistics to complex multivariate techniques. The SAS output for the linear regression of anaerobic P release (PREL) versus aerobic P uptake (PUPT) is provided in the following pages. The output generated for the multiple linear regression of anaerobic "P release (APREL), influent TP (ITP), and anaerobic ORP (ORP30) is also provided. 116 Appendix A. Statistical Analysis 117 S A S D E P V A R I A B L E : P U P T A N A L Y S I S O F V A R I A N C E S O U R C E M O D E L E R R O R C T O T A L D F 1 3 0 31 R O O T M S E D E P M E A N C . V . S U M O F S Q U A R E S 1 8 9 . 6 6 4 1 4 5 8 . 1 4 7 2 8 2 0 4 2 4 7 . 8 1 1 4 2 1 . 3 9 2 2 0 8 7 . 4 8 1 5 6 2 1 S . 8 0 8 S Z MEAN S Q U A R E 1 8 9 . 6 6 4 1 4 1 . 9 3 8 2 4 2 7 3 R - S Q U A R E A O J R - S Q F V A L U E 9 7 . 8 5 4 0 . 7 6 5 4 0 . 7 5 7 5 P R O B > F 0 . 0 0 0 1 P A R A M E T E R E S T I M A T E S V A R I A B L E O F I N T E R C E P P R E L P A R A M E T E R E S T I M A T E 2 . 5 5 4 0 8 2 8 3 1 . 9 8 0 8 9 6 3 5 S T A N D A R O E R R O R 0 . 5 5 5 6 0 4 4 3 0 . 2 0 0 2 5 0 3 2 T F O R H O : P A R A M E T E R S 4 . 5 9 7 9 . 8 9 2 P R O B > | T | 0 . 0 0 0 1 0 . 0 0 0 1 Appendix A. Statistical Analysis 118 Appendix A. Statistical Analysis '.AS DEP VARIABLE: APREL ANALYSIS OF VARIANCE SUM OF MEAN SOURCE OF SQUARES SQUARE F VALUE PROB>F MODEL ERROR C TOTAL 2 9 11 ROOT MSE DEP MEAN C.V. 8.29338055 4.38328811 12.67668667 0.8978786 3.518687 19.84483 4.14869028 0.48703179 R-SQUARE AOJ R-SQ 8.514 0.8542 0.5774 0.0084 PARAMETER ESTIMATES VARIABLE OF INTERCEP 0RP3O ITP PARAMETER ESTIMATE 2.80297057 •0.02953982 -0.20450558 STANOARO ERROR 1.14337460 0.007191805 0.18301380 T FOR HO: PARAMETERS 2.45T •4.107 •1.255 PROB > |TI 0.0367 0.0028 0.2412 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0062691/manifest

Comment

Related Items