Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Aerobic treatment of a CTMP wastewater using a rotating biological contactor Mathys, Renata Gabriela 1991

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1991_A7 M38.pdf [ 10.51MB ]
JSON: 831-1.0058872.json
JSON-LD: 831-1.0058872-ld.json
RDF/XML (Pretty): 831-1.0058872-rdf.xml
RDF/JSON: 831-1.0058872-rdf.json
Turtle: 831-1.0058872-turtle.txt
N-Triples: 831-1.0058872-rdf-ntriples.txt
Original Record: 831-1.0058872-source.json
Full Text

Full Text

AEROBIC T R E A T M E N T OF A CTMP WASTEWATER USING A ROTATING BIOLOGICAL CONTACTOR by Renata Gabriela Mathys B.A.Sc.(Bio-Resource Engineering) University of British Columbia A THESIS SUBMITTED IN PARTIAL FULFILLMENT O F T H E REQUIREMENTS FOR T H E D E G R E E O F MASTER O F APPLIED SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES CHEMICAL ENGINEERING We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH COLUMBIA August 1991 © Renata G. Mathys, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library 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 The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Effluents from high-yield TMP and CTMP mills contain biochemical oxygen demand and toxicity levels that are much higher than the effluents from low-yield kraft mills and if not properly treated before disposal can impose environmental threats to water bodies. Therefore, this study was undertaken to demonstrate the ability of a rotating biological contactor to remove organic matter from such a CTMP wastewater when operated at different flowrates and temperatures. An apparatus was constructed to allow for temperature control of the wastewater and the flexibility to change flowrates and disc rotational speed. Parameters studied included BOD 5 , COD, suspended solids, dissolved oxygen, sulfur compounds, acetic acid and nutrient uptake. Because the lab device exposes a thin film of liquid to the air over an extensive surface, evaporation from the RBC as high as 87% at temperatures of ~ 30°C were experienced during the experiment. Results of this study showed that for hydraulic residence times (HRT) of 1 to 2 days the BOD 5 removal ranged from 77 to 86% with a corresponding COD removal of 63 to 72%. Increases in flowrate (decreases in HRT) resulted in a slight decrease in percent BOD (COD) removed and an increase in loading rate and BOD (COD) removal rate. An increase in temperature from 20 to 30°C increased BOD removal from 70.1 to 85% and COD removal from 45.8 to 71.4% at HRT=1.5 days. Dissolved oxygen, pH and MLVSS had no distinct effects on the removal of B O D 5 and COD. Losses due to evaporation of major volatile compounds tested were minimal at the pH of operation. Start-up time of the RBC was of the order of 1 day during the experiment. An attempt to do a run without microorganism activity by adding 50 mg/1 HgCl 2 failed to eliminate all the organisms present in the wastewater and RBC. Tracer studies revealed that the RBC was close to being perfectly mixed. ii T A B L E O F CONTENTS ABSTRACT ii LIST O F TABLES viii LEST O F FIGURES ix A C K N O W L E D G E M E N T S xi 1 INTRODUCTION 1 2 L I T E R A T U R E REVIEW 5 2.1 Characteristics of CTMP effluent 5 2.1.1 Suspended Solids 7 2.1.2 Oxygen Demand 8 2.1.3 Lignin 10 2.1.4 Colour 11 2.1.5 Toxicity 11 2.1.6 Temperature and pH 13 2.1.7 Nutrients 14 2.2 Rotating Biological Contactor 16 iii 2.2.1 Major System Characteristics Influencing R B C Performance 17 Disc Rotational Speed 17 Dissolved Oxygen 18 Organic Loading Rate 20 Biofilm Studies and Microbiology 21 Wastewater Temperature 24 Suspended Growth 25 Staging 25 Disc Diameter and Scale-up 26 2.2.2 Other Design Parameters 27 2.2.3 Industrial Wastewater Treatment Using an R B C 29 2.3 Treatment of CTMP Effluent 30 2.3.1 Treatment of Pulping Effluent Using an R B C 31 2.3.2 Treatment of CTMP Effluent 36 3 METHODOLOGY 42 3.1 Experimental Design 42 3.1.1 Experimental Layout and Equipment Details 44 3.2 Operating Procedures 46 3.2.1 Wastewater Source 46 3.2.2 Seeding 47 3.2.3 Operation of the R B C 47 3.2.4 Sampling Procedures 49 3.3. Testing Procedures 50 iv 3.3.1 Biological Oxygen Demand 50 3.3.2 Chemical Oxygen Demand 50 3.3.3 Dissolved Oxygen Demand 51 3.3.4 pH 51 3.3.5 Oxidation Reduction Potential 51 3.3.6 Relative Humidity 51 3.3.7 Conductivity Measurements 52 3.3.8 Total Kjeldahl Nitrogen 52 3.3.9 Total Phosphorus 52 3.3.10 Solids 52 3.3.11 Volatile Fatty Acids 53 3.3.12 Sulfur 54 3.3.13 Quality Control 55 3.3.14 Microbiology 55 3.3.15 Statistics and Graphical Presentation 55 4 RESULTS AND DISCUSSION 56 4.1 Results 56 4.2 Discussion 56 4.2.1 Characterization of Pollutants 57 Characteristics of QRP CTMP Wastewater 57 BOD Curve 60 4.2.3 Evaporation 62 4.2.4 Organic Removal 66 B O D 5 Removal 66 v COD Removal 73 B O D 5 Versus COD Removal 76 Organic Loading 79 4.2.5 Effects of Temperature 85 4.2.6 Effects of Dissolved Oxygen Concentration 86 4.2.7 Effect of Oxidation Reduction Potential 89 4.2.8 Effect of pH 90 4.2.9 Analysis of Volatile Compounds 91 Acetic Acid Analysis 94 Sulfur Compounds 96 4.2.10 Solids Analysis 98 Effects of TS, VS, TSS and VSS 98 Biomass Growth in Reactor Over Time 100 Analysis of MLVSS 102 4.2.11 Nutrient Addition 104 4.3 Other Analysis 106 4.3.1 'Sterile'Run 106 4.3.2 Microbiology 106 4.3.3 Start-up Without Biomass Inoculum 108 4.3.4 Hydraulic Characterization of the Reactor 109 5 C O N C L U S I O N S A N D R E C O M M E N D A T I O N S 112 5.1 Conclusions 112 5.2 Recommendations 114 B I B L I O G R A P H Y 116 vi APPENDICES 125 Appendix I. Raw Data from Run I 125 Appendix II. Raw Data from Run II 134 Appendix HI. Raw Data from Run III . 146 Appendix IV. Raw Data from Run IV 157 Appendix V. 'Sterile'Run 165 Appendix VI. Tracer Test 172 vii LIST O F TABLES 2.1 LCso values for TMP and CTMP wastewaters 12 4.1 Comparison of CTMP wastewater characteristics from the Quesnel River Pulp Co 58 4.2 Comparison of CTMP wastewater characteristics of individual batches used for runs I, JJ, HI and IV 59 4.3 Comparison of organic removal rate at different wastewater temperatures 86 4.4 pH in RBC and pH of raw wastewater during runs I, II, III and IV 90 4.5 Equilibrium concentrations of measured chemical compounds in the CTMP wastewater 92 4.6 Equilibrium considerations of chemical compounds (not measured) in the CTMP wastewater 93 4.7 Comparison of acetic acid concentration to BOD and COD concentrations and effect of acetic acid on hydraulic retention time 94 4.8 S 0 4 = , S 0 3 = and S = concentrations during run JJ and run JJI 96 4.9 Comparison of influent and effluent solids during runs I, H, III and IV 98 4.10 Ratio of nutrient addition of Carbon:Nitrogen:Phosphorus measured as BOD 5:TKN:TP 104 4.11 Results of mass balances for nitrogen and phosporus for RBC treatment during runs I, II, and HJ 105 viii LIST O F FIGURES 2.1 TMP/CTMP flowsheet showing water recycle loops and effluent discharges from process 6 3.1 Schematic drawing of the RBC lab-scale model 43 3.2 Schematic lay-out and flowsheet of the laboratory experimental set-up of the biological RBC treatment system 45 4.1 BOD curve 61 4.2 Percent volume reduction due to evaporation in runs I to TV 63 4.3 Effects of relative humidity on evaporation at low AT during run IV 64 4.4 Percent BOD removed during runs I, II, III and TV 67 4.5 BOD removal rates during runs I, II, III and IV 68 4.6 Effects of hydraulic retention time on %BOD and %COD removal 70 4.7 Effects of hydraulic retention time on BOD and COD removal rates 70 4.8 Percent COD removal during runs I, n, HI and TV 74 4.9 C O D removal rates during runs I, II, HI and IV 75 4.10 Comparison of percent BOD and percent COD removed during runs I, II, HI and IV 77 4.11 Comparison of BOD and COD removal rates during runs I, H, III and IV 78 ix 4.12 BOD and COD concentrations in the RBC tank as lines Influent BOD's and COD's as superimposed numbers 81 4.13 Percent BOD and COD concentrations in tank relative to BOD and COD influent concentrations 82 4.14 Percent BOD removal vs. BOD loading rate 83 4.15 Dissolved oxygen concentrations in RBC bio-vessel during runs I, II, JH and IV 87 4.16 Effect of hydraulic retention time on acetic acid, B O D a c e t i c a c i d and C O D a c e t i c a c i d 95 4.17 Effect of hydraulic retention time on sulfate, sulfite and sulfide 97 4.18 Biomass growth in RBC over time during run II 101 4.19 MLVSS concentration in the bio-reactor during runs I, H, HI and IV 103 4.20 Comparison of %organic removals between bacterial and 'sterile' runs for runs I, n, and HI 107 x A C K N O WLEDGEMENTS I would like to thank my supervisor, Dr. R.M.R. Branion for his support and guidance; and particularly, I am grateful to him for his role in discussing and revising this work, I also wish to thank Dr. K.L. Pinder and Dr. V. Lo for there assistance and support in preparation of this thesis. Thanks are also due to Mr. H. Lam from the Chemical Engineering stores for his always friendly help and to Mr. J. Baranowski and the rest of the personnel in the Chemical Engineering workshop for constructing the equipment. I thank Susan Liptak and Paula Parkinson from the Civil Engineering Environmental Laboratory for their invaluable assistance in part of the laboratory analysis. I would like to express my gratitude to my family and in particular, to my husband Karim, without whom I could not have done this work, for his constant moral and intellectual support. The financial support of the Science Council of British Columbia is greatfiilly acknowledged. My thanks also go to Dr. R J . Kerekes for the use of Pulp and Paper Centre faculties. xi CHAPTER 1 I N T R O D U C T I O N Mechanical pulping processes produce higher yields than chemical pulping processes; however mechanical pulping produces lower quality fibres than those from chemical pulping. Recent advances in thermomechanical pulp (TMP) and chemi-thermomechanical pulp (CTMP) processing technology have improved pulp strength properties such that they are suitable for the production of paper and other products. Since the 1980's there has been a significant increase in the utilization of refiner mechanical pulps (TMP and CTMP). The effluent flow from such high-yield TMP and CTMP mills contains biochemical oxygen demand (BOD 5) and toxicity levels that are much higher than in the effluent from the low yield kraft mills. These increased loads in the TMP and CTMP effluents are mainly due to the absence of chemical recovery systems in TMP/CTMP mills, as well as modifications of the TMP/CTMP process which minimize water usage by internal recirculation of process water. The high oxygen demand and the biological degradability of TMP/CTMP wastewaters suggest biological treatment. Many biological systems have been investigated for this purpose. In general, wastewater treatment practised in this component of the pulp and paper industry includes one or more of the following: 1) Primary treatment for suspended solids removal; 2) Aerobic treatment such as aerated lagoons or activated sludge systems; 3) Anaerobic treatment consisting mostly of high rate processes. 1 Chapter 1. Introduction 2 Aerobic treatment has traditionally been the treatment of choice at most Canadian pulp mills where secondary treatment is practised. Anaerobic treatment has generated extensive interest on the part of governmental agencies, private enterprise and academia because of its ability to treat these high concentration effluents generated from the TMP/CTMP pulping processes. The major disadvantage of aerobic treatment processes are substantial costs associated with power requirements for mixing and aeration, whereas the major limitation for anaerobic treatment processes have been long start up times and process instability leading to system failures. These disadvantages led to the consideration of a process not among those mentioned above, the rotating biological contactor (RBC) which, taking these factors into account, would appear to be an effluent treatment system worth studying. While operating under aerobic conditions, the RBC has low power requirements for aeration with energy needed only for turning the discs and moving wastewater through the system. The purpose of this research was to demonstrate the ability of an RBC to remove organic matter from a CTMP wastewater when operated at different flowrates and different temperatures. Mechanical pulp is produced either by grinding logs into separate fibres with rotating stones (stone groundwood), which is the oldest form of mechanical pulping, or by forcing chips between grooved rotating steel plates in refiners (refiner groundwood). The advantage of refiner mechanical pulp over the stone groundwood process is that chips as well as sawmill wastes can be used as a raw material instead of logs. TMP was introduced in the 1970's as a modification of the refiner groundwood pulping process. In TMP pulping, the wood chips are presteamed in a pressurized vessel before fiberization under pressure in a disc refiner. The thermal pretreatment results in a higher strength pulp by softening the lignin to reduce damage of the long fibre component during mechanical fiberization. CTMP refers to a process which utilizes a combination of chemical, thermal and mechanical pulping stages. The impregnation of wood chips with sodium sulfite under alkaline conditions prior to preheating and refining results in pulp, which after washing, exhibits a low extractive content. Consequently, chemical pretreatment improves the quality of the final pulp by Chapter 1. Introduction 3 increasing the flexibility of the long fibre fraction (McCubbin, 1983) resulting in an increase in overall strength of about 10% over that produced by a TMP process (Jackson and Akerlund, 1984). The improved quality and increase in overall strength makes CTMP more suitable for products typically made from chemical pulps such as fluff, tissue, printing and writing grades (Jackson and Akerlund, 1984). The bleaching of CTMP is technically a brightening process which uses hydrogen peroxide or hydrosulfite as a brightening agent. Often chelating agents are also added to deactivate heavy metal ions which originate in the wood or from system corrosion and which accelerate the decomposition of the brightening agents. Finally, a further removal of extractives can be achieved by the introduction of surface active agents into the CTMP process, this leads to a additional improvement in water absorption properties and brightness of the pulp. Therefore, the quality of CTMP over TMP pulp is improved at the expense of lower yield and higher effluent load. Typical pulp yield values for the CTMP process vary from 80-90% as compared to about 91-95% yield for TMP and 43-48% yield for kraft pulp (Cornacchio and Hall, 1988). In this project, the CTMP wastewater came from the Quesnel River Pulp (QRP) Mill located on the Fraser River at Quesnel, B.C. The mill normally operates 24 hours per day, 7 days per week subject to market demand for pulp. Both TMP and CTMP are produced. The overall toxicity of the raw mill effluent is quite high. Servizi and Gordon (1986) report a mean 96 hours L C ^ toxicity of 1.27% for TMP and 0.83% for CTMP effluents. Work done in the U B C Pulp and Paper Centre on this CTMP wastewater indicated an L C ^ of 0.55%. Of the wood extractives, resin acids are a major toxicity problem. Up to 70% of the toxicity of whole effluent from pulp and paper mills has been attributed to resin acids (Environment Ontario, 1988). Other contributors to the total toxicity of the wastewater are suspended solids, fatty acids, phenols, tannin-lignins, sulphites, and sulphonated lignins. The chelating and surface agents may also contribute to the CTMP toxicity (Cornacchio and Hall, 1988). The RBC process was conceived and commercialized in 1960 in West Germany (Brenner et al., Chapter 1. Introduction 4 1984). The device consists of a series of discs which are mounted on a shaft and rotated while partially submerged in the wastewater to be treated. A microbial film develops on the surface of the discs shortly after seeding with an appropriate source of microorganisms, and provides the microbial population necessary to oxidize the biodegradable materials in the waste. The rotation of the discs leads to continuous growth of the microorganisms, as well as a continuous sloughing of biomass when the innermost layer loses its ability to adhere. The discs further serve as a mixing device for effective treatment. During immersion of the discs the BOD is removed while after emersion, the aeration of the culture is accomplished. The rotating disc is, therefore, utilized as support for the bacteria, an aeration mechanism, and a means of contacting the organism with the wastewater to be treated. Operation in this manner gives this wastewater treatment device the name the "Rotating Biological Contactor" or "RBC". CHAPTER 2 LITERATURE REVIEW This chapter provides a review of previous work that has been done in areas relevant to the present study. It begins with a review of the characteristics of wastewater from thermomechanical type pulping (TMP) processes, including a brief summary of the environmental inpact of such wastewaters if they were to be discharged without treatment. Z l CHARACTERISTICS OF CTMP EFFLUENT There is quite a variety of thermomechanical pulping processes, many of these are described in Leask (1987). Of the various pulping options this thesis is mostly concerned with chemithermomechanical pulping (CTMP) and bleached CTMP (BCTMP). In TMP processes wood chips are reduced to pulp by mechanical forces which act on the chips as they pass through a narrow gap between rotating, barred discs. The chips are in water suspension. The refining process generates a lot of heat which raises the temperature inside the refiners hence the "thermo" in TMP, CTMP etc. Some of the wood constituents are solubilized in the process. In the CTMP process sodium sulfite is added to soften the chips and make freeing of the fibres from one another easier. This results in more wood derived chemicals being solubilized along with the residual sulfite. Bleaching is often accomplished with hydrogen peroxide, aided by the addition of chelating agents, which introduces more material into solution. Figure 2.1 is a flowsheet of a possible CTMP mill configuration put together from information provided in Miotti (1990) and Wearing et al. (1985). It illustrates the prinicipal sources of wastewater 5 Chapter 2. Literature Review 6 I CHIP WISHING PRE88 SCREENS AND CLEANERS 7> HEAT RECOVERY L A T E N C Y R E M O V A L CLOUDY FILTRATE J THICK-ENER MEDIUM 0EN8ITY STORAGE 18T 8TAGE REFINING PRESSATE TANK 2N0 STAGE REFINING CLEAR FILTRATE P R E 8 S F I B R E FLOW — o VOTER R E C Y C L E «=> E F F L U E N T WASTEWATER PAPERMACHINE FRE8H WTER INTER STAGE WISHING BLEACH PLANT i THICK-ENER Figure 2.1 TMP/CTMP flowsheet showing water r e c y c l e loops and e f f l u e n t discharge from process. Chapter 2. Literature Review 7 from various subprocesses and also shows how recycling mixes the various streams. In the past water use in TMP/CTMP mills has been in the range 16 ~ 150 m3/oven dry tonne of pulp (Novatec Consultants, 1987). More recently water use in such mills is approaching 10 m3/tonne/ADT. Mills are currently under construction in which no liquid effluent is planned (Castagne, 1990 and HPDE Inc). These latter have little or no involvement with biological treatment. The composition of the organics in CTMP effluents has not been well documented. Pichon et al. (1986) reported that the data from a CTMP pilot plant indicated that the organic material in the effluent was composed of - carbohydrates (10-15%) - organic acids (35-40%) - lignins (30-40%). The release of carbohydrates and simple wood sugars, extractive composition, as well as the low and high molecular weight lignin compounds varies between wood species. Environmental regulations control the amounts of suspended solids, biochemical oxygen demand, and toxicity that can be discharged to receiving waters. pH of discharged water is also subject to regulations. In some circumstances wastewater colour and temperature may be subject to discharge limits (Bonsor et al., 1989; Environment Ontario, 1989; Sinclair, 1990). 2.1.1 SUSPENDED SOLIDS Suspended solids can be classified as settleable and nonsettleable, the two together being total suspended solids (TSS). Discharge of TSS can cause blanketing of the bottom of the receiving water leading to interference with the growth of aquatic plants and benthic fauna. They can also subject fish exposed to them to excessive levels of stress and can interfere with light transmission through the receiving water. Suspended solids in TMP/CTMP type mill wastewaters consist mostly of fibres or fibre debris that Chapter 2. Literature Review 8 is generated during the pulping process. Some sand and dirt particles are present in the chip washer effluent stream. TSS levels in TMP/CTMP wastewater have been reported to be 4.4 to 72 kg/t (Novatec Consultants et al., 1987), 10 to 50 kg/t (Virkola and Honkanen, 1985), 600 to 1000 mg/1 (Welander, 1989), 4.8 kg/adt at a concentration of 220 to 300 mg/1 (Beak Consultants, 1986), 2.7 to 17 kg/ adt at concentrations ranging from 200 -2000 mg/1 (Cornacchio and Hall, 1988). TSS loadings and concentrations in various process streams in an integrated CTMP/paper mill have been measured (Lo et al., 1991). TSS concentrations in streams ranged from 200 to 2600 mg/1. The highest concentrations were observed in a stream called CTMP washing but it is not clear if this refers to chip washing, interstage washing, or to some other pulp washing operation. 2.1.2 O X Y G E N D E M A N D If oxidizable matter is discharged into a natural body of water, microorganisms naturally present therein will oxidize it in order to supply themselves with the energy they need for survival and reproduction. In so doing they require oxygen, the only source of which is oxygen dissolved in the water. Oxygen is not very soluble in water. Thus if the microorganisms are growing rapidly they can strip the water of its dissolved oxygen (D.O.) content to the detriment of other aerobic, aquatic species, such as fish. Oxygen demand is measured by a test involving determination of the levels of D.O. in bottles containing wastewater and a microbial culture in a nutrient solution compared to levels observed in control bottles containing only the nutrient solution. This test result is called BOD. This test is time consuming and imprecise. A more precise test is the test for chemical oxygen demand (COD). This test can be done in a shorter time, however the chemical oxidation it utilizes can oxidize some compounds that microorganisms either cannot or do not oxidize in the time period available, 5 days in North America, 7 days in Europe. Chapter 2. Literature Review 9 The biologically oxidizable matter present in TMP/CTMP wastewaters is derived from the wood that is used as raw material in the pulping and bleaching processes. CTMP pulp yields, based on the wood charged to the process, are in the 80 to 90% range, thus 10 to 20% of the wood will appear in the wastewater, although a small fraction of this may be discharged to the air as volatile matter. Recycling of water used in the mill during TMP/CTMP processing decreases the effluent volume, however it increases the organic effluent concentration. Typical BOD values in North America and Europe range from 1000 to 4500 mg/1 and corresponding COD values range from 2500 to 13000 mg/1 (Cornacchio and Hall, 1988). Softwoods in general generate more BOD and COD than hardwoods. Other oxygen demanding chemicals found in CTMP wastewater include methanol, resin acids, fatty acids, triglycerides, steryl esters, and assorted sulfur compounds (Welander, 1989). 10 - 30% of the total BOD discharge of QRP's TMP/CTMP effluent is associated to the resin acid content, which is due to the fact that most of the mills raw material consists of pine, a softwood species, as well as the fact that a lot of the resins are associated with fines present in the effluent. In addition, Malinen et al. (1985) reported that the addition of alkaline hydrogen peroxide as a bleaching chemical increased the BOD and COD loads substantially. Acetic acid was found to contribute up to l/6th of the total BOD load in the QRP effluent (see table 4.1). Chemical pretreatment and alkaline conditions increase the BOD loading by breaking down lignin and by releasing acetic acid (Virkola and Honkanen, 1985). Anderson et al. (1986) reported that at the Oestrand mill in Sweden the carbohydrates and acetic acid accounted for about half of the soluble COD. However, acetic acid is readily metabolized by both anaerobic and aerobic bacteria and does not pose any problems in effluent treatment efficiency. In TMP most of the pollutants, including BOD and COD, are released from the wood in the primary refiner stage (Wong et al., 1979 and Wong et al., 1980). Little pollutant release occurs in the chip washing process. However since the wash water is recycled from elsewhere in the process the chip washer effluent contains a significant oxygen demand (Wong et al., 1979). Chapter 2. Literature Review 10 2.13 LIGNIN Lignins provide mechanical strength to wood by binding cellulose fibres together. The chemistry of hgnin is not very well defined since lignin composition is quite variable. Lignin accounts for 15 - 30% of the dry weight of plant material. The chemical pretreatment of the wood chips in CTMP manufactures enhances the solubility of low and high molecular weight lignins under alkaline conditions (Almemark et al., 1986). Urbantas and MacEwen (1985) found that CTMP effluents contain a smaller fraction of low molecular weight, biodegradable components than TMP effluents. Hydrogen peroxide bleaching may cause the production of more low molecular weight lignins and volatile organic acids (Almemark et al., 1986). The larger fraction of high molecular weight, non-biodegradable material in CTMP effluents is the cause of the lower BOD values observed per gram of dissolved material for CTMP than for TMP. The COD test will oxidize these high molecular weight organics, which results in a higher COD measurement than BOD measurement. The BOD/COD ratios measured for several Canadian TMP and CTMP wastewaters were between 0.3 to 0.45 (Hall and Cornacchio, 1988). Thus, it is reasonable to expect that biological treatment alone will not be capable of removing more than 30 - 45% of the organic material (COD) in CTMP effluents. Beak Consultants Ltd. (1986) reported that analytical data from the samples used in the bench tests gave an average ratio of 0.30:1 and 0.36:1 BODs/COD for TMP and CTMP respectively for QRP CTMP, effluent. However, it was recognized that the data was not derived from fresh composite samples and information from the QRP mill representing a few analysis suggested a higher ratio of 0.42:1 for both processes. They also found that alternating the production of TMP and CTMP has been a problem as the organic load in the untreated wastewater more than doubles during CTMP production. Chapter 2. Literature Review 11 2.1.4 C O L O U R Discharge of highly coloured wastewater to pristine receiving waters is aesthetically undesirable and may interfere with light transmission in the water and hence with the photosynthetic activities of aquatic plants. TMP/CTMP wastewater is brownish-yellowish in colour. Raw TMP wastewater has a reported colour of around 550 colour units (Wong, 1977) but during aerobic biological treatment its colour rises to around 1000 units. 2.1.5 TOXICITY Toxic matter, as far as present day Canadian environmental legislation is concerned, is matter capable of killing fish. KilUng fish via oxygen deprivation is excluded from the definition of toxicity. Toxicity is usually measured via a bioassay in which a test species is exposed to various concentrations of wastewater which has been diluted with clean water. The concentration (% dilution) which kills 50% of the test fish in a 96 hour exposure period is known as the 96 hour L C ^ The lower the L C ^ the more toxic the wastewater. Environment Canada uses an L C ^ = 65 test, meaning if 80% of the test fish survive 96 hours of exposure to 65% wastewater, that wastewater is deemed to be nontoxic. Exposure of fish to concentrations less than the L C ^ can result in a variety of undesirable, sublethal effects (McLeay and Associates Ltd., 1987). Table 2.1 provides data on the L C 5 0 values of TMP and CTMP. wastewaters. A wide variety of chemical compounds that can be found in TMP/CTMP wastewaters are known to be toxic to fish. The most toxic chemical to be found in TMP/CTMP wastewaters are believed to be the resin acids. They are toxic to both fish and anaerobic bacteria. Resin acid concentrations ranged from 2 to 21 mg/1 in four TMP effluents, and from 25 to 65 mg/1 in two CTMP mill effluents surveyed in Canada (Novatec Consultants Inc. et al., 1987). The Canadian data was obtained from softwoods which contain much higher concentrations of resin acids than hardwood species. However, other investigators found resin acid concentrations as high as 559 mg/1 in some Canadian Chapter 2. Literature Review 12 Table 2.1 LC„ values for TMP and CTMP wastewaters. Wastewater information L C 5 0 Reference TMP 0.7 - 2.4 S e r v i z i and Gordon (1986) CTMP 0.4 - 1.8 S e r v i z i and Gordon (1986) TMP 1.3 - 35.3 Cornacchio and H a l l (1988) CTMP 0.8 - 1.8 Cornacchio and H a l l (1988) TMP 17 - 79 Wong et a l . (1978) softwood mill effluents (Cornacchio and Hall, 1988). It is generally believed that resin acids can be degraded aerobically if sufficient retention time and oxygen (~ 3 ppm BOD/ppm resin acid) are provided. Resin acids are only slightly biodegraded under anaerobic conditions (Cornacchio and Hall, 1988; MacLean et al., 1990; and Servizi and Gordon,1986). Less toxic are the fatty acids. Lignans, flavonoids, tropolones, juvabiones, and thujic acid may also be toxic to fish (Wong et al., 1978). There are probably many other such chemicals for which toxicity data is not available (McLeay and Associates Ltd., 1987). Total phenol concentration in receiving waters should not exceed 1 microgram/1 to prevent tainting of fish flesh (CCREM 1987). Dwernychuk (1990) reported total phenol values of 43 microgram/1 in treated QRP effluent in his report on the environmental effects monitoring program examining physical/chemical/biological elements of systems related to pulp mill effluents of the upper Fraser River. Bosia et al. (1986) looked at CTMP pulping, papermaking and consequently, at effluent treatment in an activated sludge plant. They measured a phenol concentration of 890 mg/1 in the raw wastewater using the Folin method. This method measures various phenolic compounds as well as polyphenols. They concluded that 30% of the phenol content is not removed by the biological treatment employed by them and that this fraction which was made up for the most part of Chapter 2. Literature Review 13 polyphenols of considerable chemical stability, such as lignin, would also prove extremely difficult to be eliminated by other biological treatment processes. The disadvantages related to sulfite, which is one of the key-chemicals in CTMP production, have been studied and described by many researchers (Habets, 1986; Habets and de Vegt, 1990; Beak Consultants Ltd, 1986; Cornacchio and Hall, 1986; Anderson et al., 1986). The main part of the sulfur present in the CTMP effluent is as sulfate, because in the bleaching process, most of the residual sulfate gets oxidized to sulfate with the help of residual hydrogen peroxide, and a minor part is present as lignosulfate. Under anaerobic conditions sulfate and sulfite usually get biologically reduced to hydrogen sulfide, which when volatile will be stripped out with the biogas. If sulfide levels higher than 200 mg/1 remain in the treatment process inhibition of the anaerobic bacteria may result. Sulfide can also lead to odour and toxicity problems. The presence of sulfite does not pose significant problems for an aerobic treatment system. Bosia et al. (1986) found that the inorganic sulfite fraction of the spent liquor was almost completely oxidized by an activated sludge treatment The fate of the organically bound sulfur (lignosulfonates) is still unclear. Most authors assume that it remains unaffected. Almemark et al. (1986) suggested that sulfonated lignins may also contribute to the CTMP wastewater toxicity. Chelating agents such as DTP A during peroxide bleaching of CTMP pulp may be of some concern during anaerobic treatment. Welander et al. (1988) have indicated that excess D T P A may be toxic to anaerobic bacteria. 2.1.6 T E M P E R A T U R E A N D pH Discharges of hot water to receiving streams are unlikely to cause problems in cold waters outside the immediate discharge zone. Discharges to warm waters however, if they result in a significant rise in temperature of the receiving stream may overstress the fish therein (Welch, 1980). Wastewater temperature is an important factor in various kinds of wastewater treatment technology. Typical Chapter 2. Literature Review 14 discharge temperatures for CTMP mills are of the order of 50 to 80°C (Welander, 1989). Malinen et al. (1985) concluded in their study on characterization and treatment of chemithermomechanical pulping effluent that increased temperature and increased sulfite charge in the chemical pretreatment both resulted in a higher BOD and COD load, where temperature was the more sensitive parameter. Discharges of wastewater to receiving waters or to biological treatment systems should be close to neutrality, ie pH = 7. CTMP wastewater pH's have been reported in Canada and in Europe to be in the range of 6.0 to 7.5 (Servizi and Gordon, 1986; Wilson et al., 1985; Lo et al.,1991; Bosia et al., 1986; Scarlata, 1981). 2.1.7 NUTRIENTS Nitrogen (N) and phosphorus (P) are major nutrients required for bacterial growth during biological treatment. If biological systems are to function properly, nutrients must be available in adequate amounts. Based on an average cell composition of cell tissue of C 5 H 7 N 0 2 about 12.4 percent by weight of nitrogen will be required and the phosphorus requirement is usually assumed to be about one-fifth of this value. This translates into a BOD:N:P ratio of 100:5:1 for aerobic fermentation. For anaerobic metabolism experimental data for carbohydrate waste indicate that 1.0 to 1.5 mg of N are required for every 100 mg of COD removed to provide for biomass synthesis. A survey of several Canadian TMP and CTMP mills (Hall and Cornacchio, 1988) indicated that effluent biodegradable COD/N ratios varied between 100:0.67 and 100:4. Corresponding biodegradable COD/P ratios were from 100:0.093 to 100:0.5. In 1986 when the effluent treatment plant at QRP consisted of a primary clarifier and an aerated lagoon the treated effluents failed to meet federal regulations and provincial quality criteria for detoxification. Servizi and Gordon (1986) argued, after a pilot plant treatability study, that biological treatment was sensitive to supplemental nitrogen addition rates as well as to retention time. They concluded that increasing the treatment time of CTMP effluents from five to seven days or increasing Chapter 2. Literature Review 15 the supplemental BOD:N ratio from 100:1 to 100:2.22 of TMP or CTMP effluents resulted in lethal amounts of ammonia and/or nitrite. In the former case, endogenous metabolism by a high concentration of mixed liquor suspended solids (MLSS) was the apparent cause of elevated amounts of ammonia and nitrite. In the latter case, lethal amounts of nitrite were apparently derived from excess ammonia fed to the pilot plant. The authors further suspect that nitrite and ammonia may have been responsible for some occasional unexplained acute lethalities in biologically treated mill effluents with which they are acquainted. This problem with treated effluents has also been observed by Schnell et al. (1990), nutrient solubilization resulting from the decay of biological solids was apparently responsible. The addition of a ratio of BOD:P of 100:1 did not pose any problems on the treatment quality. The aerated stabilization basin at QRP however, was unable to provide the effluent quality required by B.C. Level A guidelines. In 1986 Beak consultants Ltd. (1986) initiated a bench scale study using an anaerobic treatment system for the QRP effluents. A low nutrient level of BOD:N:P of 100:0.9:0.5 and 100:1.6:1 for CTMP respectively for TMP were added to the reactors. Yet, problems with the nutrient concentrations were noted. The ammonia nitrogen added to the raw TMP and CTMP effluents fed to the reactor was theoretically just sufficient based on the COD removal obtained. In contrast, only minor removals of N H 4 + nitrogen or filtered total P were observed and significant residual concentrations of both N and P were present in the treated effluent When McLean et al. (1990) investigated a full-scale anaerobic/aerobic effluent treatment at QRP. Nutrients based on a COD:N:P ratio of 100:0.83:0.17 were added into the first stage of the treatment plant with no additional nutrients added to the treatment process. They concluded that the effluent had met government criteria. A number of researchers have investigated CTMP effluent from the QRP mill. A comparison of CTMP wastewater characteristics from the QRP company between Beak Consultants Ltd. (1986), Cornacchio and Hall (1986), Habets and de Vegt (1990), Kennedy et al. (1990) and Servizi and Chapter 2. Literature Review Gordon (1986) is listed in table 4.1. 16 Z2 R O T A T I N G BIOLOGICAL CONTACTORS The development of the RBC process began in Germany in the 1920's using wooden discs. By 1959, J. Conrad Stengelin was manufacturing two and three meter diameter polystyrene discs in West Germany, and by the late sixties a number of small West German communities were using RBC's for their wastewater treatment. The capital cost of these original units was quite high. However, the maintenance and energy costs proved to be low. The above is still a selling point used by RBC manufacturers. The Stengelin Company licensed the RBC process in the United States in the mid 1960's. In 1972, the American company Autotrol developed a more compact disc for the RBC process which increased the available surface area of the disc by 70 to 150%; and RBC's soon became cost competitive with other treatment processes. A cost-effective analysis was done by Lundberg and Pierce (1980) to compare air-driven and mechanically driven RBC processes with air and pure oxygen activated sludge processes. For the range of design flows studied (3-50 106gal/day) RBC's were found to require 40-60% of the power needed for pure oxygen activated sludge facilities and from 23-60% of the power needed for air activated sludge processes. The authors concluded that RBC processes were a viable alternative to activated sludge in the 3-50 lf/gal/day design range. Perry's Chemical Engineering Handbook (1984) reports that rotating biological contactors have been very popular in treating industrial wastes because of their relatively small size, their low energy requirements, and consequently, their low cost. It further states that a number of problems have occurred which should be recognized prior to using RBC's. Strong industrial wastes tend to create excessive microbial growth which is not easily sheared off and which create high oxygen demand rates, and possibly the production of hydrogen sulfide and other obnoxious odours. Heavy microbial growth has damaged some of discs and has caused some shaft failures. It is recommended that RBC units work best under uniform organic loads, requiring surge tanks for many industrial wastes. A recent Chapter 2. Literature Review 17 survey of rotating biological contactors indicated that over 80 percent of the RBC's designed before 1980 have experienced operation problems (Harrison, 1988). Many of the past problems have been overcome with new designs where: (a) improved construction techniques (e.g. minimizing shaft or media failures) are used and (b) organic and hydraulic loadings have had a proper accounting. 2.Z1 MAJOR SYSTEM CHARACTERISTICS INFLUENCING RBC PERFORMANCE One of the factors which initially enhanced RBC popularity with design engineers was the apparent simplicity of the RBC system and what were perceived to be easily definable design factors. Early investigators thought to establish the design of RBC systems on substrate removal rates per unit surface area at fixed optimal rotation speeds and detention times. Subsequent investigations, however, have revealed that RBC performance is considerably more complex and is based on numerous factors, some of which are analogous to other biological treatment systems, while others are unique to the RBC systems themselves. For better design and operation of the RBC process, it is important to understand the interaction of these parameters. The major factors influencing performance are discussed in the following subsections. 23.1.1 DISC ROTATIONAL SPEED Disc rotational speed is expressed either as tip speed or peripheral velocity; it is related to the disc diameter and rotational velocity by the equation: 2 where V = tip speed (m/s) D = disc diameter (m) Chapter 2. Literature Review 18 o) = rotational speed (1/s) Most experimental investigators concerned with the effect of tip speed on RBC performance have reported that efficiency increased with increased speed, up to a limiting or maximum removal rate (Manual of Practice OM-10, WPCF, 1988; Bender, 1985). The exact or dominant mechanism by which rotational speed improves performance is not readily known, but can be assumed to be dependent on the specific rate-limiting condition, such as mixed liquor dissolved oxygen (D.O.) concentration and shearing of the biofilm. Practically, rotational velocity is a trade-off between increased process performance efficiency and power requirements. A tip speed ranging between 0.21 m/s to 0.33 m/s is widely used in bench scale, pilot and full scale RBC design (Andreadakis, 1987; Brenner et al., 1984; Wilson and Murphy, 1980; Steels, 1974; Congram, 1976; Antonie et al., 1974; Chen et al. 1974; Lee and Mueller, 1975). 2 2 1 2 DISSOLVED O X Y G E N Previous literature has identified dissolved oxygen as a major performance parameter (Brenner et al., 1984; Bunch, 1984; Friedman, 1980; Hitdlebaugh and Miller, 1980; Huang, 1980; Lagnese, 1980; Steiner, 1980; Stover and Kincannon, 1982; Chesner and Molof, 1977; Torpey, 1971; Welch, 1968). Increases in oxygen concentration in the reactor liquid (Bender, 1985; Hartung, 1982; Hitdlebaugh and Miller, 1980; Langese, 1980; Chesner and Molof, 1977; Welch, 1968) up to recommended D.O. level of 1.5 - 2.0 mg/1 produced beneficial results. Further increases didn't (Bender, 1985; Chesner and Molof, 1977; Welch, 1968). Augmenting the 0 2 of the gas surrounding the discs tended to raise liquid D.O. levels and improve performance up to a limiting D.O. level (Bender, 1985; Huang, 1980; Torpey et al., 1971). Dissolved oxygen deficiency can result if the wastewater has a high influent organic loading. This condition brings about an increase in the organic concentration in the reactor, which increases the Chapter 2. Literature Review 19 oxygen demand required for substrate utilization resulting in low D.O. D.O. depletion within the biofiim and bulk liquid can result in the growth of nuisance organisms which competitively colonize the RBC media surface, leading to reduced removal rates, due to their lower metabolic rate (Manual of Practice, OM-10, WPCF, 1988; Bender, 1985; Brenner et al., 1984; Alleman et al., 1980). Further more, when sulfide is present, either in the influent wastewater or by its production deep within the biofiim or in the bulk liquid by anaerobic sulfate reducing bacteria Desulfovibrio. sulfide oxidizing organisms such as Beggiotoa can and frequently do grow on the biofiim surface. Sulfide serves as an energy source for Beggiotoa metabolism (Alleman et al., 1982). Beggiotoa compete with heterotrophic organism for oxygen and space on the RBC media surface. Their predominance can result in an increase in the concentration of biomass on an RBC unit while at the same time causing a substantial reduction in organic removal per unit area. This slow growing population has a smooth, milky white appearance when it is predominant on RBC surfaces (Bender, 1985; Brenner et al., 1984). While low bulk liquid D.O.'s are indicative of an overloaded system, increasing D.O. levels in the bulk liquid will not necessarily overcome process performance difficulties that may develop. Two of the plants studied by Chesner and Iannone (1985) had nuisance organisms (Beggiotoa) present in the first stages despite bulk liquid D.O. levels of 1.5 to 2.0 mg/1. A similar observation was also made at an RBC facility in Edgewater, New Jersey (Gutierrez et al., 1980). Hartung (1982) discussed the effects of supplemental aeration by comparing continuous application of air to intermittent application with regards to bio-disc loading and the protection of the bio-disc equipment of a municipal RBC plant at North Huntingdon, Pennsylvania. Positive displacement blowers were used to diffuse air beneath the bio-discs units capable of producing 185 cfm at 2 psi. A diffuser header was fabricated out of 2 in diameter PVC pipe with two rows of 30 diffusing holes 1/2 in. diameter, spaced 81/4 in. apart. The diffuser header was then anchored to a 3 in. steel channel and mounted directly under the center line of the bio-disc with the diffusing direction slightly angled Chapter 2. Literature Review 20 against rotation of the bio-disc. He concluded that intermittent dispersion of large volumes of air (370 cfm at 2 psi diffused to each individual disc for 30 min.) was more beneficial than a continuous mode. The advantages were that the blasting stripped off matured biomass to allow a new culture of the young and more active feeders to grow. He concluded further that this prevented anaerobic growth from occurring, maintained the total weight of the equipment within safe limits to prevent mechanical failures and prolonged the life of the equipment, at an inexpensive operational cost. Air driven RBC systems may be an alternative solution to limiting D.O. problems for wastewaters with high organic concentrations. The drive assembly has 10 or 15 cm deep plastic cups attached to holders welded around the perimeter of the media and an air header placed beneath the media. Forced release of air (approx. 4 psi) into the cups creates a buoyant force that causes the shaft to turn (Manual of Practice OM-10, WPCF, 1988). Brenner et al. (1984) report that a major advantage of the air drive RBC system is claimed to be related to the increased turbulence which strips off excess biofilm as air bubbles rise through the media. Z2.L3 ORGANIC LOADING R A T E In the early days, and even today, many engineer's involved with RBC design have been concerned with either hydraulic flow rate or the concentration of the substrate applied to the system. Today organic loading rate (kg BOD/day/surface area) is considered the most important design parameter for an RBC system (Springer, 1986; Brenner et al, 1984; Stover and Kincannon, 1982; Bunch, 1980; Steiner, 1980; Weng and Molof, 1980). Organic loading rate design values ranging from 0.004 to 0.04 kg BOD/m2/day are recommended by manufacturers and researchers. Typical, design, organic loading rates are: 0.0293 kg/m2/day (based on max. total BOD) or 0.004883 to 0.00732 kg/m2/day (based on max. soluble BOD) for refinery and municipal wastewaters (Stover and Kincannon (1982); 0.004 to 0.04 kg/m2/day for municipal wastewater (Wilson and Murphy (1980); 0.0313 kg/m2/day (total BOD) or 0.0127 to 0.0186 kg/m2/day (soluble BOD) for industrial and municipal wastewater (Autotrol Chapter 2. Literature Review 21 Design Manual, 1978); 0.007194 kg/m2/day for kraft wastewater (Lee and Mueller, 1975). Stover and Kincannon (1982) studied the effects of organic loading rates from 1974 to 1979. They noted that organic loading rate is the product of concentration and hydraulic flowrate per unit of disc area. Thus various combinations of organic concentration and hydraulic loading can give the same loading rate. They concluded that the amount of organics removed by the system was the same at the same organic loading rate regardless of whether the loadings are accomplished by a low flow rate at a high organic concentration or vice versa. Weng and Molof (1980) tested a six-stage laboratory RBC reactor using a synthetic substrate made from glucose, proteins and fatty acids. They concluded that an increase in influent concentration at a constant flow rate resulted in a decrease of both, percent chemical oxygen demand (COD) and D.O. content of the mixed liquor, while the rate of COD removal (mg/hour) increased. Increasing flow rate at a constant influent concentration had the same effects, but resulted also in a decrease of C O D removal (mg/1) and, consequently an increase in effluent C O D (mg/1). It is not recommended to organically overload or underload the system. The problem caused by overloading usually results in limiting the oxygen transfer capability of the system and its consequences are described in section Organic underloading is manifested by sparse growth on later stages and possible poor secondary clarifier operation because of excessive detention time. BIOFILM STUDIES AND MICROBIOLOGY Biologically, each RBC consists of a complex, interrelated population of predominantly heterotrophic, attached microorganisms. In general, the majority of this attached microbial population will be comprised of aerobic, facultative and anaerobic bacteria (Kinner et al., 1985; Alleman et al., 1980; McKinney and Ross, 1977). Metcalf & Eddy (1979) report that the biological community includes also fungi, algae and protozoa. Fungi contribute to waste stabilization; algae do not take directly part in wastewater degradation but add oxygen to the system, and protozoa do not stabilize the wastewater Chapter 2. Literature Review 22 but control the bacterial population. In addition, as indicated by Kornegay (1975), a significant population of suspended microorganisms may also be present, if the system is operated at a long hydraulic retention time or high organic loading rate. With respect to substrate removal, the concept of an "active" microbial depth has been adopted by several investigators, testing synthetic wastewater or sewage, and has been estimated to be between 70 and 200 micrometers in thickness (Nilsson and Dostalek, 1984; Characklis and Trulear, 1980; Famularo et al, 1978; Kornegay, 1975; Hoehn and Ray, 1973; Kornegay and Andrews, 1970; Sanders, 1966). This hypothesis divides the total microbial film thickness into two layers. The outermost layer is termed the active layer, while the "inactive layer", if present, is closer to the support media. Sanders (1966) evaluated active depth in terms of "critical" depth at which diffusion of oxygen within the slime layer became limiting. Tomlinson and Snaddon (1966) have also suggested that the active layer consists of an aerobic microorganisms zone. However, the presence of an anaerobic zone does not appear to offer sufficient proof to conclude that the active thickness and aerobic thickness are necessarly one and the same. Kornegay (1975), Atkinson and Davies (1974) and Grieves (1972) contend that the active depth should not only be defined with respect to oxygen diffusion but also applies to the diffusion of a metabolic substrate, and that in dilute media such as sewage the concentration of the substrate may be the rate limiting factor even when low oxygen concentration exist. In all investigations, maximum removal rates where defined within these ranges of depths; two investigations (Hoehn and Ray, 1973 and Sanders, 1966) reported decreased removal per unit depth of growth when the optimum depth was exceeded. This decreasing rate was attributed to anaerobic end products diffusing outward and interfering with aerobic activity. Conceptually, substrate removal from the bulk liquid phase requires diffusion of metabolic reactants into the attached biofilm, metabolism by the organism and diffusion of the metabolic by-product back through the biofilm and into either the bulk liquid or the atmosphere. For relatively thick biofilms significant concentration Chapter 2. Literature Review 23 gradients, resulting from mass transport resistances, can exist between the bulk liquid and the active microbial layer (Famularo 1978). A shaggy biomass surface makes the definition of the active depth somewhat imprecise. To date, no universally acceptable technique exists for measurement of active depth in any fixed film system. Studies done by Namkung and Rittmann (1988), Characklis and Trulear (1980) and La Motta and Hickey (1980) have shown that biofiim processes not only utilize organic compounds for growth, but also produce a variety of organic materials. The bacteria attach to the media by excreting polysaccharides of different kinds. Many investigators have related steady state, organic removal to biofiim thickness and age. Hoehn and Ray (1973) observed that nutrient utilization rates continued to increase until the biofiim reached approximately the level of limiting thickness. They further concluded that a quasi-steady-state nutrient utilization rate would be established at a maximum film thickness of 300 to 400 micrometers. These conclusions support those of Kornegay and Andrews (1970) and Tomlinson and Snaddon (1966). Weng and Molof (1980) investigated organic removal in terms of biological film age and thickness. They noted that at film ages between 25 to 50 hours, corresponding to thicknesses between 135 to 265 micrometers the removal rates stabilized at some constant value. They further emphasized that these values were based on composite data, and therefore only useful in the determination of average results. Torpey et al. (1971) found that, after cleaning, only 18 hours were required to restore biological growth on the disk surfaces of the first stage of an RBC thereby reestablishing normal treamtent efficiency. However, they did not report the procedures used for this evaluation. Clogging or bridging of the open spaces between the discs can shield portions of the biofiim from exposure to oxygen and substrate. This would result in a reduction in effective RBC surface area and reduced performance. Huang (1980) and Antonie (1974) suggest that high hydraulic loading is beneficial in keeping a thin healthy biomass on the discs. Clogging is a particular concern in the Chapter 2. Literature Review 24 internal non-exposed portion of the RBC media which does not come in direct contact with the shearing forces of the liquid. The internal media present in newer types of RBC corrugated media configurations (Hormel, 1979 and Autotrol, 1978) are exposed to shearing forces generated by wastewater flowing in and out of the media. 23.1.5 WASTEWATER TEMPERATURE The literature indicates that most full scale RBC systems operate between 3 to 25°C, while most lab scale RBC's usually are run at room temperature (approx. 20°C). Design manuals suggest that wastewater temperatures should generally fall within a range of 7 to 38°C to insure proper process performance (Autotrol Corporation, 1978; Hormel Design Manual, 1979; Manual of Practice, OM-10, WPCF, 1988). Some investigators have conducted studies on RBC's at elevated temperatures. Congram (1976) reported on a biodisk treating oil refinery waste. The biodisk unit treated 1,200 gpm at a wastewater temperature reaching 28°C during the summer. He did not report on any factors affecting performance effects related to summer temperatures. Studies reported by Antonie (1976) indicated that organic removal efficiency was unaffected by wastewater temperatures above 13°C. Both, Congram and Antonie, concluded that process performance deteriorated below 12 to 13°C. In some highly loaded RBC systems, lower wastewater temperatures do not always result in decreased carbonaceous removal rates and, in fact, may enhance removal. When this phenomenon occurs, it may be attributed to increasing D.O. saturation values with decreasing temperature, which promote increased oxygen transfer, and possible reduction in the number of sulfide oxidizing organisms. For example, Scheible and Novak (1980) obtained a 15% increase in BOD removal when reducing the temperature from 26 to 11°C, using 12 ft. diameter RBC units. Others (Brenner et al, 1984) have reported increased removals in wintertime conditions for heavily loaded systems, such as at Alexandria, Virginia. Chapter 2. Literature Review 25 Z2.1.6 SUSPENDED G R O W T H Suspended solids in the trough of an RBC will occur naturally as a result of biological sloughing. The concentrations of such suspended solids are dependent on the organic loading rate and the RBC stage. Bender (1985) found that suspended solids measurements in RBC tanks are subject to wide variations because of the periodic nature of sloughing and difficulties in obtaining representative samples. Suspended solids in the reactor provide additional biomass available for substrate utilization. This premise is the basis of the Surfact Process which combines an activated sludge and an RBC system (Guardino et al., 1977). An excess of suspended growth, however, can reduce reactor dissolved oxygen concentration and can contribute to anaerobic conditions in the biofilm. Reported values (Antonie, 1975; Famularo et al., 1978) of the mass of fixed biomass solids relative to the normal suspended solids mass indicate that the quantity of active suspended solids under normal conditions is negligible relative to the fixed biomass. A more recent study (Kincannon and Groves, 1980) showed that suspended solids can affect kinetic performance, especially at low flowrates. This can result because the active biological suspended solids are wasted out of the system at a lower rate, therefore, playing a greater role in substrate removal. 221.7 STAGING Staging refers to the physical separation of tanks in series, or sections of RBC media, within the same tank, by the use of baffles. One effect of a staged configuration is that each successive stage experiences an organic loading level lower than the previous. In each, a different environment may be established with different predominant strains of microorganisms developing in response to that environment. In chemical reaction kinetic theory, the use and effectiveness of staging is related to the order of reaction kinetics in the R B C tank or reactor and the nature of mixing in the reactor (e.g. Chapter 2. Literature Review 26 completely mixed or plug flow). If the RBC stages of a system are completely mixed, then a series of stages is more efficient than a single stage of equivalent capacity for first or higher order reaction rates. Studies conducted by Antonie (1975) with 5.74 ft, diameter discs demonstrated that a four-stage RBC unit produced higher percentage BOD and suspended solids removals than obtained with a two-stage unit having the same overall surface area when treating the same municipal wastewater over a hydraulic loading range of 1 to 5 gpd/sq. ft. For the same hydraulic loading four-stage and six-stage operations, however, produced virtually identical performance. Bunch (1980) explains that under heavy organic loadings, the liquor from the first stage can be distinctly anaerobic and thus, any change in parameters can stress the system. He proposes that increasing the number of stages will increase the total treatment potential of the system, reducing the stress, because the excess organic material left untreated by the first stage can then be treated by successive stages. Gillespie et al. (1974) recommends that a three-stage unit is sufficient for pulp and paper effluents, as usually nitrification takes place in the fourth stage. Brenner et al. (1984) recommends ensuring that the organic loading to any individual stage is not excessive when selecting the number of stages for an RBC system design, implying that the first few stages may have to be larger than the others. DISC D I A M E T E R A N D SCALE-UP A common RBC design practice in the past has been to scale up small pilot plant results to full-scale applications by setting the media tip or peripheral speed at the same value for both size units. Studies have shown that smaller systems exhibit greater removal capacities per unit surface area. This capacity has been attributed to the greater rotational velocity of smaller diameter systems. Reh et al. (1977) estimated that the oxygen transfer capacity of their 18 in. diameter unit was about 1.6 times greater than that of a 10.5 ft. diameter unit they evaluated. Stover and Kincannon (1982) and Severin et al. Chapter 2. Literature Review 27 (1980) propose scale-up design models based on oxygen transfer. Other investigations concerned with process scale-up reported that the results of the scale-up evaluation indicated that the B O D 5 removal efficiency decreased as the RBC disc diameter increased under similar operating conditions (Wilson and Murphy, 1980; Wu et al., 1982). Today, no proven scale-up technique from smaller diameter discs to larger diameter discs has been found universally acceptable. However, Brenner et al. (1984) suggest that if small-diameter units must be operated to collect design data, it is important that each stage be loaded below the oxygen transfer capability of a full-scale unit to minimize scale-up considerations. In general, RBC design manuals recommend keeping the diameter of the discs small for a given surface area, by increasing the number of discs. Z Z 2 O T H E R DESIGN PARAMETERS From available start-up characteristics it is apparent that a measurable biofiim will result one day to two weeks after start-up begins depending on the substrate utilized. Fry et al. (1982) report that growth of a biofiim began 24 hours after start-up for a municipal wastewater. Bracewell et al. (1980) observed that establishment of an observable biomass required two weeks in an RBC treatability study for phenol-formaldehyde resin wastewater. Torpey et al. (1971) found that only 18 hours after cleaning were required to restore a biological growth on the disc surfaces of the first stage of an RBC, with the reestablishment of normal efficiency of treatment. They did not report the procedure used for this evaluation. Based on the available literature, steady state operation appears to be reached after two to three weeks of operation (Fry et al., 1982; Ahlberg and Kwong, 1974; Lue-Hing et al., 1974; Trinh (1981); Srinivasaraghavan et al., 1980). This information was based on steady-state C O D removal data obtained from the RBC treatment of sewage. Statements concerning the excellent ability of RBC units to successfully handle shock organic and Chapter 2. Literature Review 28 hydraulic loadings are frequently encountered. Wu et al. (1980) for example, noted that a major advantage of an RBC is its ability to resist organic and hydraulic shock loads. These statements are generally based on one or two characteristics of RBC plants. One important characteristic is the ability to retain the attached biomass when exposed to large hydraulic shocks. An analysis of phenol-formaldehyde resin wastewater treatment by an RBC process found effluent COD values to be a function of the influent COD concentration. More importantly the pilot plant RBC unit functioned effectively "under varying climatic and loading conditions and exhibited excellent stability in withstanding periodic shock loadings" (Bracewell et al., 1980). Trinh (1981) reported that the biological slime of an RBC system weathered shock loads without sloughing and produced consistent effluent quality. However, diurnal flow variations caused a slight deterioration of effluent quality. These comments are based on a study which compared an extended aeration activated sludge process with a full-scale RBC system. A pilot plant study done by Fry et al. (1982), treating municipal sewage, tested an RBC with controlled shock loadings. They concluded that hydraulic shock loads depressed D.O. concentrations due to decreased hydraulic detention times and/or increased biological activity. However, the response to hydraulic shock loadings was not very reproducible. For the organic shock loads a D.O. depression did not occur and they hypothesized that this could be most likely attributed to a change in the organic constituents of the wastewater. To reduce excessively heavy microbial growth on the initial discs of an RBC, influent load distribution or step-feeding can decrease the flow reaching the initial stages of the RBC and distribute the mass organic loadings more evenly to downstream shafts or stages. When compared to straight flow-through systems, step-feeding increased the total detention time of wastewater fed into the initial stages but did not affect the total detention time of wastewater fed into the later stages. Step-feeding may also be used to distribute the inflow more evenly into a multistage reactor providing for more complete mixing. However, the evaluation of the potential effectiveness of step-feeding requires fundamental knowledge of step-wise kinetics and available models have not specifically addressed Chapter 2. Literature Review the effect of step-feeding (Bender, 1985). 29 2 3 3 INDUSTRIAL WASTEWATER T R E A T M E N T USING A N R B C Friedman (1980) investigated and described some of the failure modes and suggested preventative evaluation procedures for the consideration of designers working with industrial wastewaters. He found that RBC's have not proven to be quite the panacea promised, especially for industrial wastewaters. Even though they have been applied to or pilot tested with a large variety of industrial wastewaters with good success, failures with some industrial wastewaters have occurred and will probably continue to occur in the future. He found that the RBC process has many applications for the treatment of wastewaters. Escalating power costs make the process even more promising for the long term, cost effective treatment of industrial wastewaters. However, RBC's have distinct fundamental limitations that can lead to process failures for some wastewaters. Because of unique wastewater characteristics, gross traditional parameters such as B O D 5 and suspended solids alone are inadequate for assessing RBC treatment for many industrial wastewaters. Friedman gives an example of bench scale RBC studies with tannery beamhouse wastewaters which illustrates a potential problem when a wastewater is not carefully characterized. After several months of continuous operation, an inorganic precipitate probably consisting of calcium sulfate-calcium carbonate complex was found throughout the biomass attached to the discs. Even though this one quarter to three-eigths inch thick layer did not appear to interfere with soluble COD removal for these experiments, it would likely put extra stress on the discs and possibly lead to bridging of the biomass between the discs, as shearing of biomass would be harder, because of the greater adhesive forces of the precipitate. Therefore, industrial wastewaters should be carefully characterized for constituents that may form precipitates in the RBC units as a result of oxidation and/or pH changes. High loading conditions have created problems in the past by decreasing the D.O. content of the RBC and/or by developing excess biomass growth on the discs causing mechanical stresses, additional Chapter 2. Literature Review 30 power requirements and possible bridging conditions. Improper pilot plant operating conditions and poor data interpretation have also led to problems encountered in the use of RBC systems for industrial wastewaters. Friedman concludes that only complete chemical characterization studies and careful pilot plant evaluation programs can prevent subsequent process failures and allow owners to obtain the full benefits potentially available for the RBC process. Bracewell et al. (1980) used an RBC to treat a high strength phenol-formaldehyde resin wastewater. The wastewater was diluted 4-fold with domestic, primary treated sewage resulting in an average influent COD of 1550 mg/1 before it was fed to the RBC. The experiment was conducted in the winter in a pilot plant, consisting of an RBC unit followed by a settling pond from which a recycling stream was drawn, both leased as a unit from Autotrol. The temperature of the wastewater ranged from 13 to 14°C. A 60% average COD removal was obtained throughout the operation. From this they concluded that the pilot system as a whole and the rotating biological contactor in particular performed well but less efficiently then expected when compared to literature values for activated sludge and trickling filters. Since the pilot program was carried out during winter conditions they hypothesized that the resulting low temperatures in the small pilot system undoubtedly reduced bacterial activity and thus prevented optimum removal rates from being attained. 23 T R E A T M E N T OF CTMP E F F L U E N T Published literature indicates that TMP and CTMP effluents can be treated aerobically to reduce toxicity to acceptable levels. Combined anaerobic and aerobic treatment can reduce toxicity from highly concentrated TMP/CTMP effluent even more. Chemical treatment has found little application. Neither anaerobic nor chemical treatment alone has been shown in previous studies to reduce toxicity to acceptable levels. Cornaccio and Hall (1988) observed that, because of the high BOD concentrations CTMP effluents Chapter 2. Literature Review 31 would be ideally suited to anaerobic treatment in which sludge production and energy consumption are minimized. The much higher aerobic sludge yields are potential drawbacks to aerobic treatment. They further concluded that the long term impacts of very high resin and fatty acid levels are not known. Evidence currently available suggests that these toxic compounds can be degraded aerobically. The presence of sulfite, hydrogen peroxide and chelating agents has little impact on aerobic treatment. Some of these toxic compounds can also be removed in an anaerobic process, however, if these compounds are present at high concentrations, inhibition of or toxicity to anaerobic bacteria could result. 23.1 T R E A T M E N T O F PULPING E F F L U E N T USING A N RBC In the 1970's some investigators evaluated the rotating biological contactor as a potential method for secondary biological treatment of pulp and paper mill effluents. Influent B O D 5 concentrations in these studies ranged from 170 to 1200 mg/1. Some of the investigators fed the pulping wastewater at elevated temperatures to the RBC, but nobody performed investigations with temperature controlled tanks. The wastewater temperature fluctuated seasonally when tested with pilot plants at mills, lab scale models were usually run at room temperature. The performance of an RBC pilot plant unit to treat groundwood pulp effluent was evaluated by Bennett et al. (1973). The RBC pilot plant was rented from the Autotrol Corporation. Their overall objective was to provide an effluent comparable to other biological treatment systems. Specific test objectives included evaluation of the RBC process during mill upset, investigation of sludge settling characteristics, establishment of the degree of toxicity removal, and generation of data with regards to colour and phosphate content of the RBC treated effluent. The B O D 5 influent concentration ranged from 246 to 295 mg/1 and the TSS concentration ranged from 134 to 208 mg/1. Nutrients were added at a conventional BOD:N:P ratio of 100:5:1. It was concluded that the RBC pilot plant unit provided reduction levels of B O D 5 ranging from 67 to 89% and TSS ranging from 49 to 81%. The Chapter 2. Literature Review 32 biomass generated by this RBC treatment exhibited good settling characteristics and the unit recovered quickly during simulated mill upsets. The toxic properties were sufficiently reduced to comply with the Canadian federal effluent limitations. However, no reductions in effluent colour were observed. Influent phosphate was 60 to 70% removed and no major operational difficulties were encountered with the pilot plant Bennett et al. further reported that the overall dissolved oxygen concentration in the tank increased from 0.78 to 3.13 mg/1 from stage 1 to stage 4. The mixed liquor suspended solids (MLSS) or concentration of biomass in the tank was reported to range from 370 to 770 mg/1, and the attached biomass MLSS concentration ranged 24,000 to 5,100 mg/1 from the first stage to the last stage. The authors recommended installation of a full-scale RBC process based on their findings. Osterman (1974) evaluated the treatability of bleached kraft pulp mill effluent in Castelgar, B.C. using Autotrol's Bio-Surf pilot plant process. The influent wastewater was aerated in a holding tank to lower the temperature from 38 to 32 °C and was then fed to the RBC. At an influent BOD of 200 mg/1 Osterman reported an efficiency of over 90% with a retention time less than 2 hours. He also obtained data on sludge characteristics, toxicity, reduction and recovery time from mill upsets and concluded that the Bio-Surf system was an economically viable alternative to more conventional systems. The variability in the rate of solids formation and the low percent solids of the clarifier sludge were considered as disadvantages of the system. Chen et al. (1974) compared an oxygen activated sludge, an RBC, a trickling filter and an aerated stabilization basin for the biological treatability of bleached kraft and refiner groundwood pulp effluents. The influent BOD ranged from 123 to 230 mg/1 and the influent temperature ranged from 30 to 35 °C. Nutrients for microbial growth were fed continuously to maintain a BOD:N:P ratio of 100:6:1. For the purpose of their investigations four pilot plants were installed at the Wauna mill, Oregon. The H R T s in the oxygen activated sludge and the RBC were set at 2 hours and at 1.5 hours, and longer for the two other systems. It was concluded that both the RBC and the oxygen activated Chapter 2. Literature Review 33 sludge system provided 90% BOD removal at moderate loading conditions. The oxygen activated sludge achieved consistently near 90% BOD removal for both retention times. The RBC required an HRT > 2 hours, while at an H R T < 1.5 hours the percent BOD removal dropped noticeably. In order to obtain the same degree of treatment, Chen et al. found that more than 10 days of H R T would be necessary for an aerated stabilization basin, and the degree of BOD removal in the trickling filter was less than expected. They further found that no reduction of TSS or colour was achieved by the biological treatment systems under study and the suspended solids did not settle without further treatment. A pilot plant study using two RBC's was conducted at an unbleached kraft mill by McAliley (1974). The small RBC had 18 in. diameter discs and the large RBC pilot plant had of 12 ft. diameter discs. The purpose of the study was to compare the RBC process with the activated sludge system operated at the mill. Waste from a lignin plant along with unbleached kraft effluent were tested. It was concluded that the small pilot plant was capable of achieving 85% BOD removal from an influent containing 210 mg/1 BOD at an H R T of one hour. However, the large pilot plant did not perform as well as the small pilot plant with respect to BOD removal per unit area. The author did not know whether this low performance was related to the irregular operation of the large pilot plant or to a scale-up problem. The correlation between BOD removal rate and BOD concentration was not significantly affected by hydraulic loading, rotational speed, or type of waste (kraft or kraft plus lignin) over the range of values they studied. The retention times of these RBC pilot plants were about 25% shorter than the retention times required for the activated sludge process. It was further calculated that the horsepower requirements for the RBC plants were about 40% of the horsepower requirements for the activated sludge process. Sludge filterability was about the same for both processes. Finally, it was concluded that the RBC pilot plants required very little operator skill or attention. Gillespie et al. (1974) did a pilot scale evaluation of RBC treatment of pulp and paper mill wastes. Chapter 2. Literature Review 34 Different effluents from boardmill, kraft and sulfite mills were tested by installing pilot plant equipment from the two U.S. RBC manufacturers at various mill sites. They found for influent BOD concentrations up to 500 mg/1 the RBC system was able to achieve high levels of BOD removal, producing effluent B O D levels in compliance with emerging U.S. Federal regulations. To achieve this, however, the systems would have to be designed with considerably lower hydraulic loading rates than those quoted by the manufacturers for domestic wastewater. For effluents of high strength (>500 mg/1), they suggested special design considerations, possibly including an enlarged first stage. They further reported that these systems gave excellent resistance to drastic changes in influent conditions. The sludge generation was much lower than that considered normal for activated sludge systems and it possessed good settling properties. Even at relatively modest BOD removal rates, the systems showed good ability to remove effluent toxicity. However, effluent samples were taken after the second clarifier, which was an integral part of the pilot plant, and not directly after the RBC. Finally, they concluded that no operational difficulties were encountered with any of the pilot plant equipment used in the study and that operation was quite simple in all cases. Lee and Mueller (1975) tested an RBC lab model treating bleached kraft mill effluent. The lab scale model consisted of three stages and testing was conducted at room temperature. The influent BOD concentration ranged from 165 to 185 mg/1, the wastewater was treated at H R T s of 4, 8 and 16 hours. The dissolved oxygen of the unit increased from 3.3 mg/1 in the first stage to 7.5 mg/1 in the third stage. Microbial growth developed within the first three days of batch treatment. Lee and Mueller concluded that the RBC was capable of removing toxicity from bleach kraft mill effluents, as well as yielding high BOD removal efficiency (>90%) provided that nutrient levels, oxygen supply, pH control and mixing were proper. Lee and Costantino (1977) did a labscale, RBC study at the Powell River mill treating groundwood washer effluents and TMP effluents. The influent BOD concentration ranged from 440 to 972 mg/1. The laboratory scale RBC model consisted of a 7 liter total trough which was divided into 3 stages, Chapter 2. Literature Review 35 each stage consisted of 33 discs 8 inch in diameter. Nutrients were added according to a BOD 5:N:P ratio of 100:5:1. Biomass development on the surface of the discs was achieved after a period of approximately 2 weeks. Tests were conducted at retention times of 2 and 4 hours and the influent temperature was set at 20°C. The groundwood washer effluent was used to test the RBC for temperature shocks by feeding hot effluent at 50 and 60°C to the reactor, as well as pH shocks to pH 9 and 10. Results of the temperature shocks showed a noticeable increase in biomass growth at the discs in the first stage when the biomass had cooled down to 32 to 38°C. At these temperatures a slime activity increase was observed. Results of pH shocks showed that the system had recovered after 16 hours, when the BOD removal efficiency was stabilized again. Their investigations with TMP effluents indicated the RBC unit was highly efficient at removing BOD. At an H R T of 4 hours 94% BOD removal and 100% detoxification of the effluent were achieved. However, data from colour tests on the TMP wastewater indicated an overall colour increase of 81%. Lee and Costantino hypothesized the colour increase might be due to the high organic content of the TMP effluent making it more prone to oxidation. Overall, the authors recommended the RBC system for the pulp and paper industry as a touch-up unit on selected streams for mills facing problems in meeting government guidelines pertaining to these pollutants. A pilot plant study of biological treatment of magnesium base sulphite mill effluent was made by Baker et al. (1978). Investigations were conducted by comparing an aerated stabilization basin to an RBC. The two systems were independent but shared a common feed of the sulphite mill effluent The influent BOD ranged from 1000 to 2000 mg/1. The 140,000 U.S. gallon, aerated lagoon was operated at 7 and 20 days retention time and achieved a satisfactory BOD removal ranging from 48 to 71% during winter months and from 80 to 91% during the summer months. The RBC was a 6 feet diameter pilot unit supplied by Autotrol, it was operated at different loading rates and two different rotational speeds. The BOD removal ranged from 62 to 80% at 3 R P M and from 71 to 85% at 4.5 RPM. Because of the anticipated high influent BOD concentration the flowrates were kept Chapter 2. Literature Review 36 considerably lower than recommended. Flowrates of 1, 2 and 3 GPM were tested. This created problems in trying to maintain a constant flowrate to the pilot unit, in addition some problems were encountered in keeping a constant flow from the sulphite mill because of electrical power failures in the region. This created problems in that the overall performance of the R B C unit was erratic. Because of its long residence time the lagoon was not as much affected by these feed problems as was the RBC with its much shorter hydraulic residence time. The reported data in this study did not include the period for which there were obvious known system problems. It was found that some BOD reduction was achieved under all operating conditions and an increase in rotational speed improved the BOD reduction. However, the authors were not able to tell if this performance represented an acclimatized biomass. As the pilot plant came complete with an integral clarifier the reported data was for the clarifier effluent and not only for the RBC, therefore it was suspected that the data for B O D removal was actually lower than reported. Baker et al. further reported that dewatering of the sludge by vacuum filtration was very difficult They concluded that the RBC performance was very sensitive to fluctuations in feed properties and suggested that this may be a property of all high rate systems operating on a variable strength waste such as a sulphite mill effluent. 233. TREATMENT OF CTMP EFFLUENT Turk (1988) has reviewed aerobic and anaerobic treatment technologies presently used for treating CTMP wastewater in Canada, these included aerated lagoons with and without post settling, activated sludge systems and high rate anaerobic treatment processes with post aerobic lagoons. He presented a table comparing the advantages and disadvantages of these treatment processes with respect to fourteen parameters, as well as a table of cost comparisons between the different processes. The aerated lagoon ranked best in that it had the best rating for most parameters under investigation, however, its disadvantages include the large land requirement and the total annual cost which is Chapter 2. Literature Review 37 lowest only at lower wastewater flow rates. The activated sludge system ranked worst in both, parameter comparison and cost comparison. The high rate anaerobic treatment with post aerated lagoon rated between the aerated lagoon and the activated sludge process for the parameter comparison, however, it rated best in the cost comparison for higher effluent flow rates. This was attributed to the higher methane production which offset both capital and operating costs. McAllen (1988) reviewed aerobic waste treatment for a bleached CTMP operation. He looked at the most common treatment systems in the pulp and paper industry in Canada which included aerated stabilization basins (lagoons), mixed air activated sludge and pure oxygen activated sludge systems. He concluded that all three systems were capable of high organic removal rates. Lagoons require more land than the other two systems, have very long retention times and the BOD removal efficiency is strongly dependent on wastewater temperatures usually limited to below 35°C. The major disadvatage of the air and oxygen activated sludge systems is the high production of biological solids. The advantages are high BOD removal efficiencies of 90 to 95% at optimum retention times. McAllen found that oxygen activated sludge systems were able to detoxify high strength wastewaters at short retention times. He concluded that an aerated lagoon has a high capital cost but low operational and energy costs. Both air and oxygen activated sludge systems exceed the total cost of an aerated lagoon by 2.5 times. Since the QRP mill started production of TMP in 1981 and after modifications to include CTMP production, the effluent treatment consisted of a primary clarifier followed by an aerated stabilization basin. Servizi and Gordon (1986) studied the aerobic lagoon treatment system after numerous complications were encountered with the lagoon as its effluent consistently failed to meet with federal regulations. They concluded that such treatment fo CTMP and TMP effluents was sensitive to retention time and supplemental nitrogen addition. Beak Consultants Ltd. (1986) proposed in September 1986 a treatment process consisting of a pretreatment for fiber removal and detoxification followed by an anaerobic treatment for COD/BOD 5 Chapter 2. Literature Review 38 removal and an aerated polishing treatment for odour and toxicity removal at this mill. An anaerobic/aerobic pilot study was then conducted starting in 1987. Based on the positive results of this study a full-scale upflow anaerobic sludge blanket (UASB) was built to treat the effluent prior discharging it to the existing aerated lagoon. The wastewater is first passed through a heat-exchanger to cool it to 35°C before it is passed through the acidogenic and methanogenic phases of the UASB. The hydraulic retention time in the anaerobic system is 9 to 16 hours overall and in the aerobic polishing state a 2 day hydraulic retention time is needed to polish the anaerobic effluent. Odourous gases produced in the UASB were collected and vented through a caustic scrubber for treatment and subsequent discharge to the atmosphere. McLean et al. (1990) concluded that the system had met the government Level A criteria consistently after initial primary clarifier problems were solved. These problems had occured because large fibers and wood chips accumulating in the anaerobic system were causing reduced sludge activity. Some investigators have looked at aerobic treatment of the CTMP effluents from mills in Canada and other countries. Scarlata (1981) in Italy treated a bleached semichemical pulp effluent using a trickling filter and an activated sludge system. The influent BOD 5 concentration was 1600 mg/1 and the COD concentration averaged 5000 mg/1. With a hydraulic retention time of 3.7 days a 90% BOD reduction was achieved when both systems were employed in series, however, the COD reduction was only 57 percent which exceeded the limit of 500 mg/1 imposed by Italian law. The COD/BOD 5 ratio for the influent was 3, while for the effluent it was as high as 43. Sulphite levels dropped from 30 mg/1 to 4 mg/1 during treatment but the effluent level again exceeded the Italian limit of 2 mg/1. In a further attempt to reduce C O D levels Scarlata set up activated sludge pilot plants containing powdered activated carbon. No difference in treatment efficiency was observed over a longer time period. Shere and Day (1982) tested the treatability of a TMP effluent in a deep shaft, activated sludge bench scale model. The study was undertaken at PAPRICAN in Point Claire, Quebec. The influent Chapter 2. Literature Review 39 B O D and C O D were 1000 mg/1 and 2650 mg/1 respectively. The system was run at hydraulic retention times of 1.6 to 7 hours while the D.O. level was maintained at 3 mg/1. The performance of the system was evaluated in terms of food/microorganism (F/M) ratios. The authors observed that the system achieved "under intense agitation and high oxygen availability a complete removal of high initial toxicity". They reported that BOD and COD removals of 95% and 80% were achieved at F/M values up to 1.7 per day. However, at F/M ratios of 3 per day higher reductions in COD, but not BOD, removal were measured. They further found that a low sludge yield was produced. Colour did not change during treatment and they encountered problems with excess foaming because of a high level of agitation. A comparison between chemical precipitation with alum and activated sludge treatment was performed in Sweden by Malinen et al. (1985). Wastewater influent concentrations and the retention times of the activated sludge system were not given. The efficiency of these two processes were evaluated by looking at COD, BOD and colour removal. With alum precipitation the BOD was reduced by 30%, COD by up to 50% and colour removal was as high as 70 to 80%. Biological effluent treatment reduced BOD by 60 to 80%, COD by 40 to 60% and colour by about 30 to 40%. They concluded that chemical precipitation with alum was an inefficient treatment for BOD and C O D removal, but for colour removal fairly good. Bosia et al. (1986) reported on the performance of an activated sludge lab scale model in Italy. A CTMP effluent with a COD of 8,800 mg/1 and a BOD 5 of 3,200 mg/1 was tested. At a sludge retention time of more than 5 days a 95% BOD removal and a 70% COD removal were obtained. The sulphite level was reduced from 1,400 mg/1 to 6 mg/1. They concluded that a full-scale activated sludge system for this high retention time would be very costly and that further studies should be concentrated on a treatment process comprising sulphite precipitation and anaerobic treatment. Combined anaerobic/aerobic treatment of high strength CTMP effluent has been proposed by many researchers. Wilson et al. (1985) tested the feasibility of an aerated stabilization basin, an activated Chapter 2. Literature Review 40 sludge system including a secondary clarifier and a polishing basin, an anaerobic basin followed by an aerobic polishing basin and an anaerobic filter followed by an aerobic polishing basin. The wastewater tested was a combined NSSC and CTMP effluent from the Bathurst mill in New Brunswick. The BOD concentration of the influent was 1290 mg/1 and the corresponding COD concentration was 2740 mg/1. Because of future mill water flow recycling these concentrations were estimated to double. It was concluded that the aerated lagoon with an overall 7 to 9 days retention time was able to produce a non-lethal effluent, the activated sludge basin with a 2.5 days retention time however, was unable to produce a non-lethal effluent. Nevertheless both aerobic systems were able to remove 80% BOD under normal summer conditions. Both high rate and low rate anaerobic systems did not achieve the targeted 80% BOD removal during the pilot trials. However, following a three to five day aerated basin both combined anaerobic/aerobic systems produced an effluent exceeding 95% BOD removal. The data showed that toxicity removal appeared to be dependent on hydraulic residence time with a total of six to seven days required to produce non-lethal aerobic effluent. Anaerobic treatment alone had only a small effect in reducing toxicity. As a result of this study it was concluded that a combined anaerobic/aerobic treatment system was feasible and worth further investigation. Wilson et al. (1988) then investigated, in 1986, three different anaerobic pilot processes consisting of a fiuidized bed, a UASB and a contact basin, all followed by an aerated polishing basin. They concluded that the UASB pilot reactor performed best, with the result that a UASB followed by an aerated polishing basin was finally selected for the full scale installation at the Bathurst mill. In a CTMP mill in Domsjoe in Sweden an anaerobic treatment plant followed by an aeration basin was chosen. Hultman and Hillstroem (1986) concluded that this was the cheapest solution for the "unbleached effluents" and provided a very efficient method to remove BOD. Rintala and Vuoriranta (1988) treated a TMP effluent with a combined anaerobic/aerobic lab-scale treatment system in Tampere, Finland. The anaerobic system alone removed 60 to 70% COD, while the combined anaerobic/aerobic system was capable of removing 80 to 85% COD. The aerobic post Chapter 2. Literature Review 41 treatment using an activated sludge system produced effluent with stable quality, but it also included a reddish colour change. The change of the colour occurred also when anaerobically treated effluent alone was aerated for 30 minutes. PAPRICAN investigated techniques for the improvement of effluent management at the Spruce Fall Power and Paper Company Ltd. integrated newsmill in Kapuskasing, Ont. A cost effective system had to be chosen for the mill. A high rate process was favoured because of restricted land availability. An activated sludge process was rejected, because although the degree of treatment and the operational reliability are usually high, the costs associated with aeration and the increased production of excess biological sludge are substantial. A pilot anaerobic process followed by an activated sludge system was investigated. Influent BOD ranged from 1100 to 2100 mg/1 and influent C O D ranged from 1900 to 4100 mg/1. Schnell et al. (1990) concluded that the anaerobic treatment alone produced insufficient removals of BOD and toxicity. Combined anaerobic/aerobic treatment gave a consistently high reduction in BOD resin and fatty acids. However, this required a total minimum H R T of 27 to 30 hours with the whole mill effluent. The effluent detoxification was inconsistent, due to lethal levels of non-ionized ammonia from nitrogen nutrient supplementation of the raw wastewater. The anaerobic high rate system had a considerable degree of operational risk, due to serious losses of excess sludge at high applied loading rates. Finally, Schnell et al. measured an increase in colour of 50-68% for the anaerobic system and 68-130% after aerobic treatment. CHAPTER 3 METIIODOLOGY This chapter details the experimental methods used in this study. Section 3.1 specifies the equipment details, section 3.2 describes the operating procedures and section 3.3. outlines the testing procedures. 3.1 E X P E R I M E N T A L DESIGN Initial operating experience including some quantitative analysis was obtained using a bench-scale RBC reactor obtained from the Civil Engineering Department at U.B.C. That 20 litre unit consisted of thirty four, 19.5 cm diameter polyethylene discs and was separated by baffles into two stages. A more elaborate RBC unit was designed to allow for temperature control and rotational speed control. A schematic drawing of this rotating biological contactor is shown in Figure 3.1. The main design parameter used was the organic loading rate. After reviewing the literature (see section the system was designed for an organic loading rate of 0.008 kg BOD/m2/day. However, at this time the B O D 5 parameters had yet to be estimated. Special design consideration was given to the disc configuration. Because of the potential occurrence of excess biomass growth and possible bridging as described in section and glass discs were considered with a 1 inch spacing between them. The choice of glass as a material for the bio-discs arose from concerns related to the deformation of materials under elevated temperatures, i.e. it was aimed in this study to keep a liquid temperature in the trough as high as 30 to 35°C and maybe increase it up to 70°C for future thermophilic studies. At these temperatures most materials used for RBC lab-scale discs, such as 42 Chapter 3. Methodology 44 acrylic and polyethylene warp, which could result in easier bridging between some of the discs. The glass discs were sandblasted to give a rough surface enabling better biomass attachment. 1 inch spacing between the discs allowed for better observation of biomass growth and estimation of the biomass thickness during the runs. In order to allow for easy access into the trough for purposes of maintenance or taking grab samples a hydraulic jack attached with two levers to the shaft enabled easy lifting of the discs. 3.1.1 E X P E R I M E N T A L L A Y O U T A N D EQUIPMENT DETAILS A schematic layout of the process is shown in Figure 3.2, where the elements of this experimental set-up are outlined. Initially, a 14 litre acrylic reaction vessel rectangular in plan with a cylindrical bottom and a heating jacket spaced 1 inch (2.54 cm) apart was used. However, during run 1 leaks in the bio-reactor caused problems. These were caused by glueing problems and elevated temperatures in the waterbath. The majority of leaks were between the waterjacket and the bio-reactor. A stainless steel bio-reactor was built after run I. This 12 litre vessel measured 13 by 24 in. (32.5 by 60 cm). Twenty three, 10 inch (25 cm) diameter, glass discs were attached to a 3/4 inch (1.9 cm) diameter, stainless steel shaft. The total surface area of the discs was 1.89 m 2 and the total surface area including the wetted bio-reactor walls was 1.93 m2. The shaft was 23 in. (57.5 cm) long. Individual discs were fitted to the shaft 1 inch (2.54 cm) apart, and the group of 23 discs was spaced 1.5 in. (3.8 cm) from each end of the shaft. The shaft was connected, using a sliding stainless steel rod to allow for the lifting and lowering of the discs, to a 1/2 hp Dayton D C gear motor model 6Z417 used for rotation of the discs. The motor was connected to a Penta KB Power variable speed controller (CHE no. 3896). The discs were rotated at a constant tip speed of 0.31 m/s throughout the whole experiment. A parallel to the disc shaft influent/effluent configuration was used, however, the option for a NUTRIENT SUPPLY EFFLUENT SAMPLER FEED TANK EFFLUENT TANK I I Fig . 3.2 Schematic lay-out and flowsheet of the laboratory experimental set-up of the b i o l o g i c a l RBC treatment system Chapter 3. Methodology 46 perpendicular inlet/outlet configuration was available. The feed was pumped into the reactor using a peristaltic, Masterflex, variable speed pump Model 7523-10 (CHE no. 3668E). A number 16 Masterflex pump head was used for this pump, requiring Masterflex tubing of 4 mm inner diameter. The effluent left the reactor through an effluent hole which was set 3.5 in. (9 cm) from the inner bottom of the vessel to ensure 40% submergence of the discs in the liquid. For controlling the temperature in the RBC a Lauda constant temperature bath (CHE no. 1517) was used. 6 mm inner diameter Tygon tubing was utilized to connect the waterbath to the jacket of the RBC. Thermometers where placed at each end of the reactor, i.e. of the inlet and outlet. The temperature of the liquid in the reactor was read manually every day. 33 OPERATING PROCEDURES The entire experimental period lasted approximately 9 months. This period was sub-divided into a qualitative analysis period using the bench scale reactor, borrowed from the Environmental Engineering group of U.B.C.'s Civil Engineering Department, which lasted about 2 months, followed by runs I, JJ, HI and IV, using the RBC equipment described above. Each run lasted between 21 and 23 days followed by 3 weeks of analytical evaluation for each run. Further, a 'sterile' run was attempted lasting 2 weeks, a hydraulic characterization experiment was conducted during 5 days and a 1 day start-up experiment running wastewater without the addition of biomass innoculum through the bio-reactor was performed. 33.1 WASTEWATER SOURCE The wastewater used during this study was obtained from the QRP mill in Quesnel, B.C. Wastewater, in 1 m 3 containers, was shipped from the mill to the Pulp and Paper Centre at U.B.C. It was subsequently stored in 5 gallon plastic buckets in the laboratory cold room at 2°C. Chapter 3. Methodology 47 322 SEEDING Seeding of the borrowed RBC was undertaken by using seed sludge previously grown on sewage and CTMP wastewater. Five litres of a 50:50 mixture of raw sewage, obtained from a pilot scale activated sludge plant at the University of British Columbia and well adjusted CTMP sludge, obtained from an aerobic sequencing batch lab reactor treating CTMP effluent at the Pulp and Paper Centre at U.B.C., were added to the reactor. The MLSS concentration of the added sludge was approximately 15,000 mg/1. The reactor was then filled with a mixture of CTMP wastewater and deionized water, and operated in batch mode at a disc speed of 3 RPM. Additional air was supplied to the sludge with an aquarium air stone. Each day, for the next five days, the disc rotation and air supply were stopped and the 'mixed liquor' in the reaction vessel was allowed to settle. Approximately 2 litres of the supernatant was decanted and that removed volume was replaced with fresh CTMP wastewater. Supplemental N and P in the form of urea (CO(NR^)2) a n a " phosphoric acid (H 3P0 4) at a conventional ratio of 100 BOD 5:5N:lP were added to the wastewater to ensure nutrient levels were adequate. After five days the batch mode was replaced by continuous feeding of CTMP effluent to the reactor at a flowrate resulting in a 5 day HRT and aeration was discontinued. Seeding in the larger glass disc RBC was undertaken simply by adding 5 litres of acclimatized mixed liquor from the borrowed RBC diluted with 9 liters of deionized water to the acrylic trough unit and 7 litres of deionized water to the smaller stainless steel trough of a 50:50 mixture of CTMP wastewater diluted with deionized water. Immediately after, fresh wastewater was continuously fed to the reactor at the flowrate under investigation for the particular run. 323 OPERATION O F T H E RBC For each run only one parameter was changed to investigate its effect. For runs I, II and III the flowrate was changed resulting in a retention time of 2 days for run I, 1 day for run II, and 1.5 days Chapter 3. Methodology 48 for run III. The temperature in the trough was maintained at 30°C for these three runs. Run TV was conducted by changing the temperature in the trough to 20°C at a retention time of 1.5 days. The 'sterile' run was performed at 1.5 days retention time and a temperature of 30°C, while the tracer test was undertaken at a 1.5 day retention time and a temperature of 20°C in the trough. Runs I, JJ , HI and TV were done using the following procedure: Every day the time, the influent volume, the effluent volume, the temperature, the D.O. and pH were measured and recorded. On top of these measurements the oxidation reduction potential was measured daily during run LTI and the relative humidity was measured daily during run IV. Dissolved oxygen, pH and oxidation reduction potential were measured along the flow path at three different places, the beginning, the midpoint and the end within the trough. In addition, pH and ORP where measured on influent and effluent samples. The feed to and the effluent from the RBC were stored in calibrated plastic tanks (each with a capacity of 20 litres). The feed tank was refilled every one to two days with QRP, CTMP wastewater, previously stored at 2°C in the laboratory cold room. Nutrients were added directly into the feed tank by pipetting. To ensure that N and P would not be limiting the conventional concentration ratio BOD 5:N:P = 100:5:1 was used throughout the experiment. The feed plastic tank had to be cleaned every day to minimize any bacterial activity in the feed tank itself prior to RBC treatment. After each run the inlet, outlet and sampling tubing were cleaned or if necessary replaced. Biomass which tended to bridge between the discs was removed, as well as excess sludge accumulating in the tank. For the attempted 'sterile' runs the same procedure as for runs I to III were used. However, dissolved oxygen, oxidation reduction potential and relative humidity were not measured. 50 mg/1 of mercuric chloride (HgCl2), which according to Gledhill (1975) would sterilize the solution, were mixed in a large bucket with 60 litres of CTMP wastewater. Each run was carried out lasting two theoretical hydraulic retention times. Chapter 3. Methodology 49 Prior to the start-up run experiment, using CTMP wastewater without the addition of microorganism innoculum, the discs and the reactor were cleaned. The trough was then filled with the raw CTMP effluent at the same setting as for run m. Measurements were taken and recorded, these included B O D 5 and MLVSS, dissolved oxygen, pH and temperature readings, taken every half hour. The tracer test was conducted in order to obtain information on the residence time distribution and flow regime in the RBC reactor. This was accomplished by employing a step change of sodium chloride tracer concentration to the reactor and measuring the tracer concentration in the effluent. The experiment was performed following run rv and none of the parameters were changed. A large 20 IMP gallon container was used to mix enough raw CTMP wastewater to last for the whole experiment The background conductivity of wastewater was measured and an amount of sodium chloride about 6 times the background conductivity was added to the feed bucket, measured and recorded. At time zero conductivity measurements of the effluent at the sampling port were taken and recorded every half hour. Later, during the experiment, measurements were recorded every hour, every two hours, and finally towards the end every 4 to 6 hours. The conductivity in the feed bucket was remeasured every day during the five days of the run. 3.2.4 SAMPLING PROCEDURES Sampling was carried out at the influent and effluent sampling ports, except for the MLSS which were sampled in the reactor vessel. These ports were located 6 to 8 cm away from the bio-reactor (see Figure 3.2). For the first two weeks effluent samples for BOD 5 and COD analysis were taken daily, and feed samples were taken for every new bucket used from the cold storage. It was assumed that after two weeks steady state was reached (Fry et al., 1982; Tringh, 1981; Srinivasarghavan et al., 1980; Ahlberg and Kwong, 1974; Lue-Hing et al., 1974). During this time until the end of the experiment samples for B O D 5 and C O D analysis were withdrawn every two days. In addition to the regular B O D 5 Chapter 3. Methodology 50 and COD samples TKN, TP, and solids tests of the inlet, outlet and in the trough (for MLSS) were carried out once a week during the last 2 weeks of each run. For runs JJ and HI additional samples of the influent and effluent were withdrawn for analysis of acetic acid and sulfur compounds. Acetic acid samples were taken every 3 days and sulfur samples were taken twice during each run at the same time the solids samples were withdrawn. Two BOD 5 , COD, acetic acid and sulfur samples were taken during the 'sterile' run, at the beginning and end of the second HRT. The procedure for taking MLSS samples involved lifting the discs out of the vessel using a hydraulic jack. The liquor was then mixed by stirring and a grab sample was withdrawn. 33. TESTING PROCEDURES All filtrations of the samples during this study were done using Whatman No. 4 filter, except during solids analysis in which Whatman 934-AH glass fibre filters were used. Samples for TKN, total phosphorus and acetic acid were preserved with concentrated sulfuric acid (pH < 2.0) in accordance with the Standard Methods (AP.H.A, 16th ed., 1985). If samples were not filtered prior to testing, the sample content was allowed to settle and the supernatant was withdrawn for analysis. 33.1 BIOLOGICAL O X Y G E N D E M A N D (BOD 5) The determination of 5-day BOD was done immediately after sampling in accordance with the Standard Methods (AP.H.A, 16th ed., 1985). 33.2 C H E M I C A L O X Y G E N D E M A N D (COD) The samples were preserved by freezing and analyzed in accordance with the Standard Methods (AP.H.A, 16th ed., 1985) using the open reflux method. However, for run 1 the Hach DR/2000 spectrophotometer was used and the reagents were prepared as outlined in the Standard Methods Chapter 3. Methodology (A.P.H.A., 16th ed., 1985) for the closed reflux, colorimetric method. 51 333 DISSOLVED O X Y G E N (D.O.) Dissolved oxygen concentrations in the bio-reactor and for BOD 5 analysis were measured manually by using a Model 54 A R C dissolved oxygen meter (Yellow Spring Instrument Co.). The membranes and electrolytes in the probes were replaced once a week on average. Prior to taking measurements the probe used in the bio-reactor was calibrated with air, while the BOD probe was calibrated with distilled water. The exact procedures are outlined in the manufacturer's manual and the Standard Methods (AP.H.A, 16th ed., 1985). 33.4 pH The pH was measured with a Cole Palmer Microcomputer pH-meter Model 05669-20 with glass and reference electrode combined in a single probe. The meter was standardized against pH 10 and pH 7 buffer solution as outlined in the Standard Methods (AP.H.A, 16th ed., 1985). 33.5 OXIDATION REDUCTION POTENTIAL (ORP) ORP measurements were made using a Broadley James Corporation's Ag/AgCl probe. The probe was connected to a Orion Research digital ion analyzer Model 601A The probe was calibrated in accordance with the manufacturer's manual. 33.6 R E L A T I V E HUMIDITY (RH) The relative humidity in the lab was measured because of the high evaporation of the wastewater in the bio-reactor experienced during the experiment. A humidity probe HP1 was used for this purpose, connected to a Fluke 77 multimeter. The R H probe was calibrated weekly as outlined in the manufacturer's instruction manual. Chapter 3. Methodology 52 3 3 . 7 C O N D U C T I W r Y M E A S U R E M E N T S Conductivity measurements were performed using a Jenway 4020 Conductivity meter. The programmable conductivity meter was equipped for print interval modes. The printout parameter consisted of date/time, sample number, conductivity and temperature readings. The equipment was operated and calibrated according to the manufacturer's operating manual (4020/REV B). 3 3 . 8 T O T A L K J E L D A H L NITROGEN (TKN) Samples were digested in a Technicon block digester 40 and analyzed according to Technicon Block digester instruction manual (1974). 3 3 . 9 T O T A L PHOSPHORUS The samples were first subjected to acid digestion using a Technicon block digester 40, prior to analysis by Technicon Auto-analyzer II, according to the Technicon Industrial Method No.327-73W (1974). 3 3 . 1 0 SOLIDS All TS, VS, TSS, VSS were examined to see the contribution of solids in the wastewater (TS,VS) initially and after RBC treatment, as well as the contribution of biomass (VSS) before and after RBC treatment. (a) T O T A L SOLIDS (TS) A known sample was put into a crucible and oven dried for a minimum of 1 hour at 104°C, as outlined in the Standard Methods (A.P.H.A, 16th ed., 1985). Chapter 3. Methodology 53 (b) V O L A T I L E SOLIDS (VS) This was determined by igniting the solid obtained in 3.3.10 (a) at 550°C, as outlined in the Standard Methods (A.P.H.A., 15th ed., 1985). (c) T O T A L SUSPENDED SOLIDS (TSS) This was done by vacuum filtering a known volume of sample through a pre-washed and oven dried Whatmann 934-AH glass fibre filter and oven drying for a minimum of 1 hour at 104°C, as outlined in the Standard Methods (A.P.H.A., 16th ed., 1985). (d) V O L A T I L E SUSPENDED SOLIDS (VSS) This was determined by igniting the nonfilterable solids obtained in 3.3.10 (c) at 550°C, as outlined in the Standard Methods (A.P.H.A., 16th ed., 1985). (e) MIXED LIQUOR V O L A T I L E SUSPENDED SOLIDS (MLVSS) This was determined as described in 3.3.10 (c) and (d). (f) BIOFILM DRYING Biofiim was scraped off the disc at the end of run 1 and run 3. The total volume and weight were recorded. A known sample was then taken and was proceeded as described in 3.3.10 (c) and (d). 33.11 V O L A T I L E F A T T Y ACIDS (VFA) The volatile fatty acids measurements were done using a computer-controlled Hewlett-Packard 5880A gas chromatograph, equipped with a flame ionization detector (FID) and using helium as the carrier gas. The glass column (0.91 m long with a 6 mm external diameter and a 2 mm internal diameter) was packed with 0.3% Carbowax/0.1% H 3 P 0 4 on Supelco Carpopak C (supplied by Supelco Inc.). The Chapter 3. Methodology 54 column was conditioned according to the procedure outlined in the Supelco Bulletin 751E (1982). Experimental conditions for the chromatograph were as follows: Injection port temperature = 150°C Detector port temperature = 200°C Isothermal oven temperature = 120°C Flow rate of carrier gas (helium) = 20 ml/min. Volatile fatty acids analysis included acetic, propionic and butyric acids with quantification done by the external standard methods , using reagent grade standards. However, concentrations propionic and butyric acids were < 0.01% compared to acetic acid concentrations and often not detected by the analyzer. 33.12 SULFUR (a) SULFATE (S04) Sulfate analysis were done by using the Hach DR/2000 spectrophotometer as well as the Hach reagents specific for the sulfate test. The analysis was performed in accordance with the Hach Procedure Manual (1988). (b) SULFITE (SOs) Sulfite tests were conducted using a back-titration of the method given in the Standard Methods (AP.H.A, 16th ed., 1985). This allowed for a sharper end-point. (c) SULFIDE (S2-) Sulfide analysis were done by using the Hach DR/2000 spectrophotometer as well as the Hach reagents specific for the sulfide test. The analysis was performed in accordance with the Hach Procedure Manual (1988). Chapter 3. Methodology 55 33.13 QUALITY C O N T R O L Quality control was performed on all the chemical analysis done. Standard solutions with known concentrations were tested in accordance with the instrument procedure manuals and the Standard Methods (AP.H.A, 16th ed., 1985). In the case of B O D 5 and COD analysis two different U.S. EPA samples (no. 3405 WP 1088, Batch 823 and Batch 860) with a know B O D 5 and COD concentration were tested. 33.14 MICROBIOLOGY A solid medium of glucose agar in a petri dish was supplied from the Microbiology department at U.B.C. for determination of mixed bacterial cultures. A sterile loop was used to streak the plate with the wastewater sample under investigation. Three replicate plates were used for each sample. The plates were then incubated for 48 hours at 35°C. 33.15 STATISTICS A N D GRAPHICAL PRESENTATION The data was processed using the software Lotus, release 2.2. The graphical solutions and linear regressions were done using Grapher (version 2) from the Golden Software, Inc., Golden, Colorado. The three dimensional graphs were done using Surfer (version 4.15) from the Golden Software, Inc., Golden Colorado. The constants for the BOD curve were determined using the least square curve fitting program from the N A G Fortran Library (function E04FDF) and the curve fitting for the flow characterization was done using the numerical analysis program Polymath from Cache Corporation, Austin, Texas. CHAPTER 4 RESULTS AND DISCUSSION The results of the bench scale experiments conducted in this study are detailed and discussed in this chapter. Experimental results of the four individual runs are presented in section 4.1. Data analysis and discussion of these four runs are presented in section 4.2. Finally, section 4.3 outlines an attempted 'sterile' run and subsequent tests to reveal some problems related to this experiment, as well as a tracer test done to determine the residence time distribution of the RBC reactor. 4.1 RESULTS The daily logs from runs I, U, HI, and IV and the corresponding raw data are listed in Appendices I, IJ, HJ, and IV respectively. 42. DISCUSSION Data from the experiments are shown as time plots in what follows. Most of these graphs plot various information for all four runs on one page. They are plotted in the order in which the runs were performed, ie from left to right run I (HRT=2 days) and run II (HRT=1 day), followed below by run HI (HRT=1.5 days, Temp.=32°C) and run TV (HRT=1.5 days, Temp.=20°C). Any bargraphs and tables which follow show values of parameters averaged after 10 days of ninning the system for each run when steady state response was assumed. With run IV data was measured only after 10 days of operation. 56 Chapter 4. Results and Discussion 57 4.Z1 CHARACTERIZATION OF POLLUTANTS 43.1.1 CHARACTERISTICS OF QRP CTMP WASTEWATER D A T A ANALYSIS: Table 4.1 presents a summary of QRP, CTMP wastewater characteristics as selected from the literature which are then compared to the measured values of the raw wastewater (Appendices I-IV, raw data) used in this study. Four different batches of the QRP CTMP wastewater have been used, one per run, throughout this work. The values for the characteristics have been averaged over all four runs. The BODj/COD ratio was 0.42 and compared well with the other studies. This indicates that any biological system alone will not be capable of removing all of the organic matter from this CTMP effluent. The pH was lower than reported in the studies from Servizi and Gordon (1986) and Beak Consultants Ltd. (1986). Measured data of the raw wastewater differed for each batch. Table 4.2 shows wastewater characteristics data for the four individual batches used during this study. DISCUSSION: The values of this study (Table 4.1) correlate well with literature values, considering that the mill supplying the wastewater has gone through process changes during the past years. The normal operating pH range for the CTMP stock is 6.3 - 6.8 (Thakore et al., 1987), the present observations are somewhat lower. Measured values for individual batches (Table 4.2) indicate that BOD values were highest during run II and IV and pH values were lowest during run I and VI. C O D data from run I do not compare very well to the three other runs. A colourimetric method (see section 3.3.2) was used for COD measurements during run I. Problems with this method were encountered because of interferences caused by particles and precipitates. Another problem with this method was the underestimation of measurements at high COD concentrations which are said to be typical for the colourimetric method. COD values from run I do not compare very well to literature values and to C O D analysis done by other researchers on the same raw QRP, CTMP effluent. Table 4.1 Comparison of CTMP Wastewater Characteristics from the Quesnel River Pulp Co. SOURCE Beak Consultants Ltd., 1986 Habets and de Vegt, 1990 Kennedy et al . , 1990 Cornacchio and Hall, 1988 Servizi and Gordon, 1986 This study COD (mg/1) Range Average 8180 7200 5000 - 9000 | 3400 - 8500 6750 - 8600 7960 COD(soluble) (mg/1) Range Average 3000 - 6000 ! 5800 - 7450 6660 BOD5 Range Average 2920 3100 1549 - 4500 3302 2500 - 4550 3310 BOD5/COD Range Average 0.36 0.43 0.30 - 0.45 0.42 Acetic Acid (mg/1) Range Average 400 - 800 500 - 1000 383 - 492 436 Sulfate (mg/1) Range Average 300 254 - 300 430 - 470 445 Sulfite (mg/1) Range Average 0 100 - 300 56.0 - 69.5 61.5 TKN (mg/1) Range Average minimal 28 16.2 - 16.4 16.3 Phosphorus total(mg/1) Range Average 5.0 2.2 2.7 - 6.1 4.0 TSS (mg/1) Range Average 2200 400 1000 - 1600 220 221 900 - 3600 1480 VSS (mg/1) Range Average 700 - 1400 300 - 1000 700 PH Range Average 7 - 8 7.15 5.37 - 6.55 6.11 Table 4.2 Comparison of CTMP Wastewater C h a r a c t e r i s t i c s o f I n d i v i d u a l Batches Used f o r Runs I , I I , I I I , and I V . SOURCE Run I Run I I Run I I I Run I V COD (mg/1) Range Average 3840 - 4920 4600 7300 - 8500 7970 6750 - 8600 7960 7250 - 8500 7960 BOD (so luble ) (mg/1) Range Average 2350 - 4450 3400 2350 - 2900 2520 BOD5 Range Average 2600 - 3450 3010 5400 - 2600 3690 2500 - 3900 3025 2900 - 4300 3530 BOD5 /COD Range Average 0.65 0.46 0.38 0.44 A c e t i c A c i d (mg/1) Range Average 383 - 463 422 421 - 492 449 S u l f a t e (mg/1) Range Average 430 - 440 435 440 - 470 455 S u l f i t e (mg/1) Range Average 61.5 - 69.5 65.5 56 - 59 57.5 TKN (mg/1) Range Average 16.3 16.4 16.2 Phosphorus to t a l (mg/1 ) Range Average 2.69 3.0 6.12 TSS (mg/1) Range Average 3600 4850 - 5000 4925 900 - 1100 1000 900 - 1000 950 VSS (mg/1) Range Average 250 800 - 900 850 900 - 1000 950 900 PH Range Average 5.8 6.42 6.15 - 6.55 6.37 5.37 - 5.70 5.56 Chapter 4. Results and Discussion 60 B O D C U R V E OBJECTIVE: To find out if the concentration of nitrogen in the wastewater is present at a level were nitrifying bacteria compete with the carbonaceous organism for available oxygen. PROCEDURE: BOD tests were done as described in section 3.3.1. Enough BOD bottles were prepared and incubated to read the D.O. daily for 30 days. The test was done from June 20,1990 to July 10, 1990 and repeated on July 3, 1990 until August 2, 1990. RESULTS: See Figure 4.1 DISCUSSION: The biochemical oxygen demand is equal to the amount of dissolved oxygen which is consumed by a substrate incubated for a specific length of time at 20°C. The BOD determined after a 5-day incubation period is the most common. Five days is not long enough for interference by slow growing bacteria which is exerted only after 6 to 10 days. The presence of nitrogenous oxygen demand can be determined from such a BOD curve as it appears on the graph like a second stage curve superimposed on the original curve. No nitrogenous interference was found in the QRP wastewater. Figure 4.1 shows that the BOD curve exhibits first order behaviour. y(t) = K(l - e"*) where y = B O D 5 t = time K = steady-state gain = 5600 mg/1 T = time constant of process = 5.6 days Figure 4.1 indicates that the B O D 5 is between 60 and 65 percent of the ultimate BOD and oxidation is more or less complete in 18 to 20 days. Metcalf and Eddy (1979) state that within a 20-day period, the oxidation of sewage is about 95 to 99 percent complete, and in the 5-day period used for the Chapter 4. Results and Discussion 6000 -i Time (days) Figure 4.1 BOD curve. Chapter 4. Results and Discussion 62 B O D 5 test, oxidation is from 60 to 70 percent complete. Therefore, the rate of oxidation of the CTMP wastewater is comparable with that of sewage. 4.23 EVAPORATION D A T A ANALYSIS: Since most of the experiments were conducted at approximately 30°C in a device which exposes a thin film of liquid to the air over an extensive surface, a lot of evaporation occurred. The amount of this evaporation had to be quantified. Figure 4.2 shows the effects of such evaporation during the four runs. Runs 1,11, and HI were all conducted at temperatures ranging from 30 to 34°C, but at different HRT's. The evaporation data from run I (HRT=2 days) were erratic. This was caused by leakage of water from the waterjacket into the bio-reactor. During the time when there were no leaks, evaporation would go as high as 90%, however it averaged 74%. During run II (HRT=1 day) evaporation slowly increased from 62% to 90% at the end of run II and averaged 78%. Run III (HRT=1.5 days) showed a constant evaporation throughout the run of 87% average. Following these three runs, run IV (HRT=1.5 days) was conducted by lowering AT (temperature difference between wastewater in the trough and room air temperature). It resulted in a much lower average evaporation of 29%. The results of Figure 4.3 for run TV show the effects of relative humidity on evaporation while AT was almost zero. As expected, when the relative humidity in the room went up less water was evaporated from the RBC. DISCUSSION: Evaporation of water in the RBC is a function of available surface area, volume to surface area ratio, mixing, temperature difference (AT), and relative humidity. A lot of water evaporates from the surfaces of the discs which carry a thin film of liquid on the surface of the layer of biomass attached to the discs. For this small, lab-bench RBC the trough volume to disc surface area ratios were 0.00074 m for the 14 litre vessel used for run I and 0.00063 m for the 12 litre vessel used during run II to IV. In these lab scale RBC's the discs are usually flat, whereas in full scale units Chapter 4. Results and Discussion 63 HRT - 1.5 days Temperature - 31*C I I 1 1 1 1 1 1 1 1 10 Time TTTTTgTTTTT (days) • I i n I I I I i •• 20 100 40 HRT - 1.5 days Temperature - 20*C 1 1 1 1 1 1 1 1 1 1 1 1 I I 1 1 1 1 10 Time r T T T T T T T T (days) • 11111111111 80 25 Figure 4.2 Percent volume reduction due to evaporation i n runs I to IV. Chapter 4. Results and Discussion 64 Figure 4 .3 E f f e c t s of r e l a t i v e humidity on evaporation at low A T during run IV. Chapter 4. Results and Discussion 65 extended surfaces are used, which increase the area by 70 to 150%, giving a lower volume to surface ratio (Autotrol, 1978). Disc rotation tends to cause turbulence in the trough wastewater, which facilitates mass transfer of water from the trough liquid surface to the air. Run I was performed during July/August 1990. The temperature in the room at that time ranged between 24 and 28°C resulting in a AT ranging from 4 to 8°C; the relative humidity ranged from 30 to 45%. Run II (HRT=1 day) was started up during mid-November 1990 with the new stainless steel vessel which had a 15% smaller volume to surface area ratio. The temperature difference was approximately 11 to 16°C, considering a constant room temperature of 16 to 20°C. Evaporation started to increase during this run which lasted until the beginning of December 1990. The relative humidity decreased gradually from 35% to 14%. By the middle of November the outside temperature started to drop and the relative humidity began to decrease. The temperature inside the room was at a constant level of 20°C during the day, however, temperatures dropped to 12 to 15°C during the night, when heating was lowered in the building. During run III (December 7,1990 - December 28,1990) a constant low relative humidity ranging from 14 to 18% and a AT ranging between 9 and 11°C resulted in a constant high evaporation of 87%. By decreasing the temperature in the bio-vessel during run IV (May 1, 1991 - May 21, 1991) the water loss was decreased by 58% of run HI. To summarize these results, the following general observations were made based on the evaporation data: 1. With the dependent parameters, surface area, volume to surface area ratio, and mixing kept constant, AT and % R H were major parameters affecting the water loss in the RBC during this study. 2. The fact that run I was performed during the summer, while run JJ and LTJ were done at the end of fall into the winter, and finally run IV was done in the spring, indicated that seasonal changes can have a major impact on evaporation when running an RBC system inside a temperature controlled building. Chapter 4. Results and Discussion 66 3. Since many of the parameters measured in this study are concentrations, corrections had to be made because the evaporation of solvent water tends to increase the concentration of things like BOD and COD. 424 ORGANIC REMOVAL Flowrate of wastewater to the RBC is one factor in determining the rate of organic addition and defines the retention time of the liquid in the R B C system. The effect of flow rate was studied at 7 litres/day (run I), 12 litres/day (run II) and 9 litres/day (run HI). Run HI and run IV were conducted at the same flowrate but at different wastewater temperatures of 31°C and 20°C respectively. Since hydraulic retention time (HRT) equals RBC volume divided by influent flowrate these flowrates are equivalent to HRT's of 2 days, 1 day and 1.5 days. Graphs for percent organic removal, organic removal rate and percent of organic removal vs. loading rate (Figures 4.4 to 4.11 and Figure 4.14) were corrected for water loss. 424.1 BOD 5 REMOVAL D A T A ANALYSIS: Figure 4.4 shows plots of percent BOD removal over the time periods during which each of the four experiments were run. Figure 4.5 shows BOD removal rates (mg/hour) during the four runs. Since the active disc surface area was constant in all of these runs at 1.75 m 2 the rates shown can be put on a per unit disc area basis by dividing by 1.75. For run I (HRT=2 day) BOD removals ranged from 62 to 98% and averaged 86%, while the BOD removal rates ranged from 429 to 1915 mg/hr and averaged 817 mg/hr. For run II (HRT=1 day) BOD removals ranged from 70 to 91% and averaged 77%, while BOD removal rates ranged from 902 to 1756 mg/hr and averaged 1432 mg/hr. For run HI (HRT=1.5 days) BOD removals ranged from 77 to 86% and averaged 83%, while BOD removal rates ranged from 787 to 1373 mg/hr and averaged 1001 mg/hr. For run IV BOD removals ranged from 64 to 77% and averaged 70%, while BOD Chapter 4. Results and Discussion 67 120-100 A "2 60-03 > O B o 60-Q 40-o m 20-HRT _ 2 days Temperature - 32 *C ° ° ° o n o a 11111111111111111111IIi 10 16 Time (days) 20 26 120 100-1 *-< BO-03 > O B 60 (i Q 40 o PQ 20-1 HRT - 1 days Temperature - 33 *C O • D O o 1 111111111111111111II111 6 10 Time (days) 11111111II11111 16 20 26 120-i HRT . 1.5 days Temperature - 31 *C 100-I • a * ° -> o B 60-o u § «: m 20H a • 111111111111111 6 10 Time (days) i II1111111111111111111 15 . 20 26 120 100 ti B u BOH 60-Q 40 o PQ 20-1 HRT . 1.5 days Temperature - 20 *C M I 111 n 1111111 n 111111111111111 n 6 10 16 20 Time (days) Figure 4.4 Percent BOD removed during runs I, I I , I I I and IV. Chapter 4. Results and Discussion 68 3000-4 HRT . 1.5 days Temperature - 31 C a i>2000H 1 o a « Q o ra looo H 111II11111 J1111111111111111111111111111111111111111 10 . 15 20 25 Time (days) 1 3 0 0 0 - 1 so a .2 2000H « o a tf Q O m 1000 HRT . 1.5 days Temperature - 20 "C 111111111111111111111111111111111111111111 5 10 15 20 25 Time (days) Figure 4 . 5 BOD removal rates during runs I, I I , I I I and IV. Chapter 4. Results and Discussion 69 removal rates ranged from 806 to 1073 mg/hr and averages 901 mg/hr. Figure 4.6 shows average B O D 5 percent removal versus H R T and Figure 4.7 shows average B O D 5 removal rates versus HRT for runs I, II and HI. DISCUSSION: BOD analysis were started from day 2 and it appears from Figure 4.4 (runs I to III) that the system was at steady state during the experiment Figure 4.4 and Figure 4.6 indicate that the average percent BOD removed slightly decreased with an increase in flowrate (consequently a decrease in HRT) and decreased with a lower wastewater temperature at a constant flowrate (runs HI and TV). Figure 4.5 shows that for an increase in flowrate (decrease in HRT), BOD removal rate increased, while the percent BOD removed decreased. Rate of BOD (COD) removal can be affected by several factors BOD (COD) concentration, organic loading rate, disc speed, temperature, MLVSS concentration, D.O. concentration, nutrient concentration, etc. For runs I, H, III and IV the disc speed was constant so this can be ehminated. Runs I, II and HI were at constant temperature, more or less, so this can be eliminated as a factor for these three experiments. The organic loading rate increased with increasing flowrate (see section it was lowest in run I at 0.0128 kg BOD/m2/day average and increased in run HI to 0.0166 kg BOD/m2/day average and was highest during run H at 0.0257 kg BOD/m2/day average, so perhaps organic loading rate explains the difference in BOD removal rate which was lowest in run I and increased from run III to run H. Furthermore, Figure 4.12 shows the actual levels of BOD (COD) concentrations in the trough of the RBC. BOD levels were higher in runs H and HI than in run I. However there is not much difference in BOD concentration between runs H and HI, yet run HI has a lower removal rate than run H. Similar remarks apply to C O D removal rate although this was not constant throughout runs n and m. Chapter 4. Results and Discussion 70 90 r-1 1.5 2 Hydraulic retention time (days) %B00 removed R^SI SCCOD removed Figure 4 . 6 E f f e c t of hydraulic retention time on %BOD and %COD removal. 2.6 h 1 1.5 2 Hydroulic retention time (doys) BOD removol rote COO removol rote Figure 4 . 7 E f f e c t of hydraulic retention time on BOD and COD removal rates. Chapter 4. Results and Discussion 71 Run I had a higher D.O. concentration (see section 4.2.6) than run JJ or JJI, yet had a lower BOD 5 removal rate thus D.O. does not appear to play any part in explaining the differences in BOD removal rate for these three runs. The results presented below in section indicate that MLVSS levels cannot shed any light on these BOD (COD) removal rate differences. MLVSS measures only those organisms in suspension in the RBC trough. The amount of active biomass on the discs should be more or less constant if one assumes that there is a thin layer of active organisms at the surface of the biofiim. Nutrients were added at a constant BOD5:nitrogen:phosphorus ratio. Data presented in section 4.2.11 suggest that nutrient concentrations were not rate limiting. The effects of percent BOD removal (%BOD r e m) and BOD removal rate (BOD^ can also be explained from a mathematical point of view. % B O D r e m is calculated by BODiato-BODou =%BOD ( i ) BODinto and BOD„ is calculated (BODlnto-BODout) flowrateinCo=BODZI (2) equating (1) and (2) gives *BODzemfl owra t e i n t o B O ° i » t o =BODZI (3) Equation (3) indicates that both %BOD r e m and flowrate into the RBC are directly related to BOD,,. The increase in flowrate during runs I, JJ and JJI were proportionally larger than the decrease in %BOD r e m . Therefore BOD,,, increased with increasing flowrate, while % B O D r e m decreased slightly. Chapter 4. Results and Discussion 72 In addition, increasing the flowrate will increase the organic loading into the RBC ie the food supply to the bacteria will increase, which in turn may explain the increase in organic removal rate and support the above argument. The same trends were observed for COD. Comparing runs HI and IV (see Figure 4.4. and 4.5) shows that, for a decrease in temperature, percent BOD removed and the BOD removal rate both decreased. The decrease in BOD removal rate is due to the lower rate of reaction at lower temperature. Increasing the hydraulic retention time (see Figure 4.6) will increase percent BOD and COD removals, these results indicate that removal might increase still further above HRT=2 days. However, without modifying the RBC to control evaporation loss, conducting tests at a greater H R T might result in a complete evaporation of the water at a temperatures of 32 ±2°C. Lo et al, (1991) used an activated sludge system to treat a CTMP wastewater with a B O D 5 influent concentration of 3000 mg/1 and a corresponding COD concentration of 6200 mg/1. The best BOD and C O D removal was achieved between 2 to 2.5 days H R T with 98% BOD reduction and 80% COD reduction at 20°C. It has been noted, that because of evaporation the hydraulic residence time in the reactor increased as did the organic concentration. Stover and Kincannon state that the amount of organics removed by the system is the same at the same loading rate regardless of weather. The loadings are accomplished by a low flow rate at a high organic concentration or a high flow rate at a low organic concentration. Other investigators have treated CTMP wastewaters of similar strength by different kinds of biological treatment systems and achieved the following BOD and COD removals: (1) Baker et al. (1978) compared an aerated lagoon to an RBC using wastewater from a magnesium base sulphite mill with an influent BOD ranging from 1000 to 2000 mg/1. The aerated lagoon achieved BOD removal rates of 48 to 71% during the winter months and 80 to 91% during the summer months, while the RBC's removal ranged from 62 to 85%, they did not report the actual hydraulic residence time used during their investigations. (2) Scarlata (1981) used a trickling filter and an activated sludge system Chapter 4. Results and Discussion 73 which removed 90% of the BOD from an influent B O D 5 concentration of 1600 mg/1 and removed 57% of the C O D from an influent COD concentration of 5000 mg/1 with a total hydraulic retention time of 3.7 days. (3) Wilson et al. (1985) compared an aerated lagoon, an activated sludge system and a combined anaerobic/aerobic treatment system treating CTMP effluent with a BOD concentration of 1290 mg/1 and a C O D concentration of 2740 mg/1. The aerated lagoon with a 7 to 9 days HRT produced a non-lethal effluent, the activated sludge system with a 2.5 day H R T did not produce a non-lethal effluent but both systems achieved 80% BOD reductions during the summer months. The combined anaerobic/aerobic treatment system (high rate or low rate anaerobic system followed by an aerated lagoon) produced an effluent exceeding 95% BOD removal after a 6 to 7 days total hydraulic retention time. (4) Schnell et al. (1990) treated wastewater from an integrated newsprint mill with an influent BOD ranging from 1100 to 2100 mg/1 and an influent C O D ranging from 1900 to 4100 mg/1. A combined anaerobic/aerobic treatment system gave a consistently high reduction in BOD, requiring a total H R T of 27 to 30 hours. C O D R E M O V A L D A T A ANALYSIS: Figure 4.6 and Figure 4.7 show the percent COD removed versus H R T and the C O D removal rate versus H R T for runs I, II and HI. Figure 4.8 shows the graphed results of the percent C O D removed during runs I to IV, while Figure 4.9 shows the C O D removal rate during runs I to IV. For run I (HRT=2 days) the average percent COD removed was 72%, while the average COD removal rate was 1063 mg/hr. The percent COD removed during run II (HRT=1 day) ranged from 54 to 82% and averaged 63%, while the COD removal rate ranged from 2116 to 3091 mg/hr and Chapter 4. Results and Discussion HRT . 2 days Temperature « 32 "C 0 j 11111111111111111111111111111111111111111111111111 0 6 10 15 20 26 Time (days) 120 l o o H *—' • a • o fi 0) 80H 60 H Q 40-O 20 H HRT _ 1 days Temperature - 33 *C l l 11111111 i n 11 11| 11111111111111111111 i I I 111 I I 11 6 10 16 20 25 Time (days) 120 100 1 80 60 Q 40-O 20 HRT . 1.5 days Temperature » 31 *C i mTUT111 n 11 r m I M T T i n II 11 n i i n T r i m i n i n i 6 10 . 15 20 25 Time (days) 120 100' •-J 60' O B 60 Q 40 O 20 HRT - 1.5 days Temperature - 20 *C 1111111111M1111111111111111 6 10 16 Time (days) " M i l SO 25 F i g u r e 4 .8 P e r c e n t COD remova l d u r i n g runs I , I I , i n and I V . Chapter 4. Results and Discussion F i g u r e 4 .9 COD r e m o v a l r a t e d u r i n g runs I , I I , m and I V . Chapter 4. Results and Discussion 76 averaged 2511 mg/hr. The percent COD removed during run HI (HRT =1.5 days) ranged from 59 to 78% and averaged 71%, while the COD removal rate ranged from 1982 to 2829 mg/hr and averaged 2289 mg/hr. The percent COD removed during run TV ranged from 41 to 51% and averaged 46%, while the C O D removal rate ranged from 1173 to 1450 mg/1 and averaged 1337 mg/hr. DISCUSSION: The same trends as for the BOD analysis were observed for the C O D results. It is interesting to note that the difference between the COD removal rates of run HI and run TV is much larger than the difference between the corresponding BOD removal rates (Figure 4.5 and 4.9). For a 13% decrease of percent BOD removed and a 10% decrease in BOD removal rate from run UJ to run IV the corresponding percent COD removed indicates a 24% decrease while the COD removal rate indicates a 42% decrease. Comparing Figure 4.6 and 4.7 show that the ratio of COD removal rate to BOD removal rate is much larger than the corresponding ratio of percent COD removed to percent BOD removed. When comparing Figure 4.7 to 4.12 COD and BOD ratios indicate a similar trend, which may indicate that organic concentrations (BOD and COD) in the trough are dependent variables of organic removal rates. 43.43 B O D 5 VERSUS C O D R E M O V A L D A T A ANALYSIS: Comparisons of percent BOD and COD removed and of BOD and COD removal rate for all four runs are presented in Figure 4.10 and Figure 4.11. The average %COD/%BOD removed from start-up to the end of a run was 0.842 during run I, 0.817 during run H, 0.842 during run HI, and 0 654 during run IV (see Figure 4.10). Figure 4.11 shows the COD and BOD removal rates for the four runs. The average ratio of COD/BOD removal rate from start-up to the end of a run was 1.30 during run 1,1.75 during run II, 2.29 during run HI, and 1.48 during run IV. Samples for B O D 5 and COD analysis were usually withdrawn from settled wastewater. However, Chapter 4. Results and Discussion HRT « 2.0 days Temperature - 32 *C • BOD remova l * COD remova l 0- TTTTT 1111111I1111111 e io 16 Time (day) 20 I 25 120-1 K loo f> 80 O e o 60-40-o 20-HRT _ 1.0 days Temperature - 33 *C « o a a° a' • a a « • BOD remova l # COD removal 0- 1111111111111111IIHn111111111n111II11II111111 n 0 6 10 15 Time (day) 20 25 HRT . 1.5 days Temperature - 31 *C « * * • BOD remova l * COD remova l I I 111 I I 11111111111111 n 11111111111111111111111111 6 10 15 20 25 Time (day) 120 K too ! n u o • a o B0 SO 40 20 HRT - 1.5 days Temperature - 20 *C • BOD remova l * COD removal 0- 111111111111111111II111 n i n 1111111111II111111111| li 6 10 16 . 20 25 Time (day) Figure 4.10 Comparison of percent BOD and percent COD removed during runs I, I I , I I I and IV. Chapter 4. Results and Discussion 78 Figure 4.11 Comparison of BOD and COD removal rates during runs I, I I , I I I and IV. Chapter 4. Results and Discussion 79 during runs LT and III BOD and COD analyses (see Appendix I to IV, raw wastewater data) were also done on filtered influent and effluent samples. The percent BOD and COD removal using filtered samples showed little difference from the results of the settled BOD and COD analysis. The reason for this was that the solids content was high in both the raw wastewater and the RBC effluent (see Table 4.9, and Appendix I, II, HI and IV - filtered BOD and COD data). This resulted in a proportional decrease of the filtered influent and effluent organic concentration (BOD and COD), eliminating the effect when taking the difference to calculate percent organic removal. DISCUSSION: The results plotted in Figure 4.10 show that at a temperature of 30°C (run 1,11 and HI) a higher percentage C O D can be removed than at lower temperature of 20°C (run TV). The ratio of %COD/%BOD removed was lower at the lower temperature. No significant difference could be seen for runs I and HI. Figure 4.11 also suggests that the removal rate of COD compared to the removal rate of BOD is much higher at elevated temperatures (comparing run I,H and HI to run TV). However, when comparing COD/BOD removal rates for runs I, H and HI no explainable difference could be observed. For the raw wastewater the average BODs/COD ratio was 0.42. The average effluent BOD5/COD ratios were calculated, after correcting the effluent values for water loss, to be 0.33 for run I, 0.29 for run H, 0.22 for run HI, and 0.24 for run TV. Meaning that more B O D 5 in relation to C O D was removed during all four runs. The same trend is observed here as in Figure 4.4: Decreasing flowrate increases the BOD removal. However, again the reason for the high ratio during run I may be because of underestimation in the COD analysis. 43.4.4 O R G A N I C L O A D I N G Because of evaporation in the bio-reactor during the four runs the organic concentration in the vessel tended to increase, while the hydraulic residence time tended to decrease. It was felt that the Chapter 4. Results and Discussion 80 following graphs could describe the effect of the actual concentration distribution in the tank over time caused by evaporation and biological degradation. D A T A ANALYSIS: The tracer test results (see section 4.3.4) indicate that it can be assumed that the RBC behaves very like a CSTR, from that it follows that the organic concentration leaving the reactor is equal to the organic concentration in the reactor. Figure 4.12 shows both B O D 5 and COD concentration in the reactor throughout each run. The numbers on the plots indicate the influent concentrations. These can be compared to the actual concentrations in the tank which are graphed as the lines on the plots. For run I the maximum BOD concentration in the tank was 2700 mg/1 for a corresponding 3100 mg/1 influent BOD, while the maximum COD concentration was 3720 mg/1 for a corresponding 4660 mg/1 influent COD; for run II the maximum BOD concentration in the tank was 4800 mg/1 for a 2800 mg/1 influent BOD, while the maximum COD concentration was 22800 mg/1 for an 8350 mg/1 influent COD; for run III the maximum BOD concentration in the tank was 6300 mg/1 for a 3450 mg/1 influent BOD, while the maximum COD concentration was 22450 mg/1 for an 8350 mg/1 influent COD; and for run IV the maximum BOD concentration in the tank was 2000 mg/1 for a 4300 mg/1 influent BOD, while the maximum COD concentration was 6200 mg/1 for an 8000 mg/1 influent COD. Figure 4.13 complements Figure 4.12 by showing the percent organic concentrations (ie organic concentration in the tank divided by the influent concentration times 100) in the tank relative to the influent concentrations plotted versus time. For 100% organic influent concentration the maximum percent concentration in the tank for run I was 68% BOD and 80% COD, for run II it was 172% BOD and 273% COD, for run HI it was 183% BOD and 269% COD, and for run IV it is 49% BOD and 81% COD. The plot in Figure 4.14 shows the distribution of percent BOD removed over the range of organic loading rates tested. The organic concentration of each CTMP batch shipped to us varied (see table 4.2). DISCUSSION: The BOD and COD concentration in the tank (see Figure 4.12) did not exceed the Chapter 4. Results and Discussion 81 Figure 4.12 BOD and COD concentrations i n the RBC tank as l i n e s . Influent BOD's and COD's as superimposed numbers. Chapter 4. Results and Discussion 82 Figure 4.13 Percent BOD and COD concentrations i n tank r e l a t i v e to BOD and COD i n f l u e n t concentrations. Chapter 4. Results and Discussion 83 ; v—' 60-•a >>• o 0 o 60-« Q • O 03 40-HRT » 2.0 days Temperature - 32 *C a a a a 111111111111111111111111111111 I I 11111111111111n 0.005 0.010 0.015 0.020 0.025 0.030 Org. Loading Rate (kg BOD m day"1) 120 100 K BO •a % a 8 40-20-1 HRT _ 1.0 days Temperature - 33 *C • a tv a a a a a a a 0.015 11111111111111111111 0.020 0.025 1111111111 0.030 11111111111111 0.035 0.040 Org. Loading Rate (kg BOD m~* day -1) 120 100-I 60H R § 40-1 20H HRT _ 1.5 days Temperature - 31 *C v II1111n1111 in11111111n1111n•• iII1111111111111mini 0.012 0.014 0.016 0.018 0.020 _0.022 0.024 Org. Loading Rate (kg BOD m day-1) 120 100 H BOH 20' HRT . 1.5 days Temperature - 20 *C 0' | 11111111111111111 ., 0.014 0X16 0.018 • 11111111111111 „ 0.020 6.022 Org. Loading Rate (kg BOD m day"1) Figure 4.14 Percent BOD removal versus BOD loading. Chapter 4. Results and Discussion 84 influent concentration during run I. However, leakages into the reactor would dilute the wastewater and may have distorted the picture of the actual concentration distribution in the tank during this run. A new reactor was re-started for run II and was continuously run until the end of run HI. Evaporation during these two runs was highest throughout the experiment with an average water loss of 78% during run II and 87% during run III. BOD concentration in the tank stayed almost constant throughout run II and HI. It is interesting to note the gradual increase of COD concentration relative to BOD concentration during run H, while during run HE the COD concentration indicates a decreasing trend. The COD concentration increased during run II, as it was not biodegraded, but only concentrated in the tank, while the BOD concentration, which was biodegraded, stayed at an almost constant concentration level throughout run II and HI despite the evaporation. Thus, the decrease in COD concentration in the plot of run HI suggests that more of the high molecular weight compounds, which contribute to COD, were able to be attacked by the bacteria after about 35 days of acclimating them in a continuously running RBC reactor. Run IV never exceeded the influent concentration and shows a steady organic concentration in the tank throughout the run. Figure 4.13 is another way of presenting the data graphed in Figure 4.12. During runs H and HI the maximum percent COD concentration in the tank was almost doublethe corresponding maximum percent BOD concentration in the tank. Percent BOD concentration and COD concentration show a simultaneous increase during run n and a simultaneous decrease during run HI. Originally, the RBC was designed for an organic loading rate of 0.008 kg BOD/m2/day as described in section 3.1. The actual organic loading rate varied from 0.0079 to 0.0366 kg BOD/m2/day during the four runs. Figure 4.14 indicates that percent BOD removal was not much affected, over the range of organic loading rates tested for each HRT, by organic loading rate. Most of the data points of the organic loading rate curve for run I were distributed around the lower end of the scale, however one data Chapter 4. Results and Discussion 85 point had a higher value indicating an increasing slope towards that one point at the higher end of the x-axis. This may be missleading and indicating that the reactor would be able to withstand a higher organic loading rate for the same degree of treatment The organic loading rate increased with decreasing HRT. However, it almost doubled from run I (HRT=2 days) to run JJ (HRT=1 day). This was caused because the flowrate was increased and the BOD i n was higher during run JJ (see table 4.2) The organic loading rate was higher during run IV (HRT=1.5 days) than during run JJI (HRT =1.5 days) because the BOD concentration in run IV was higher than the one in run III (see table 4.2). The organic concentration was calculated using the influent B O D 5 concentration and the influent flowrate. However, considering the decrease in hydraulic residence time and the increase in concentration within the bio-vessel caused by the evaporation, the organic loading rate was also calculated using the effluent B O D 5 concentration and the effluent flowrate. These calculations showed little change from the original results which were used to plot Figure 4.14. Stover and Kincannon (1982) showed that the removal rate and efficiency were indeed dependent on the amount of total organics applied to the RBC rather than its concentration or flowrate. A2J5 EFFECTS O F T E M P E R A T U R E D A T A ANALYSIS: The effect of treatment temperature was investigated because of the large evaporation experienced during runs I to in. Treatment efficiency was investigated with all conditions fixed as during run IJJ and temperature was changed to 20°C. Table 4.3 shows a comparison of averaged percent B O D 5 and COD removal during these two runs. Chapter 4. Results and Discussion 86 Table 4.3 Comparison of organic removal rate at d i f f e r e n t wastewater (WW) temperatures WW Temp. = 31°C WW Temp. = 20°C Parameters HRT =1.5 days HRT = 1.5 days BOD5 removal (%) 83.0 70.1 COD removal (%) 71.4 45.8 %COD/%BOD5 0.861 0.654 removal DISCUSSION: Temperature of the wastewater determines the rate of reaction. Treatment efficiency of aerobic biological systems will increase with increasing temperature up to a limiting temperature (usually 40°C in the mesophyllic range). The data from table 4.3 indicate that the organic removal increases with increasing temperature and that more COD relative to BOD can be removed at higher temperature. This is not in agreement with the findings of Lo et al. (1991), who tested the effect of treatment temperature for a CTMP wastewater (BOD=3000 mg/1 and COD=6200 mg/1) using an activated sludge system. Both BOD and COD removals were reduced when the temperature was raised from 20 to 30°C. 426 E F F E C T O F DISSOLVED O X Y G E N CONCENTRATION D A T A ANALYSIS: Three-dimensional (3-D) plots were used to illustrate surface trends revealed by the D.O. concentration as a function of time and reactor length (from inlet to outlet). To fit the 3-D curve a minimal curve method was used. Figure 4.15 shows the data for runs I, II, III and IV. DISCUSSION: Run I revealed an average D.O. concentration of 0.5 mg/1 throughout the reactor rising to peak values between day 12 to 15 and rising again between 18 and 19 and finally between day 22 and 23. Run II and run III (Figure 4.15) show an average D.O. concentration of 0.26 mg/1 and 0.28 mg/I respectively, throughout the RBC reactor and during the entire length of these runs. Run IV showed an initially high average D.O. concentration of 6.5 mg/1 at start-up. The D.O. was 81 Time (daysj * « ii la Time (days) *> ii its Time (days) T ronc©1^* i n t rougn ^ived oX£»~ and A 15 0 i s S ° i XX, XXX a Chapter 4. Results and Discussion 88 measured every 1/2 hour and after 4 hours the D.O. concentration had decreased to an average of 2 mg/1 when it was observed that biomass started to adhere to the discs. At low D.O. concentrations (below 0.5 mg/l) one would expect that BOD removal rate might be limited by the availability of D.O. Yet, comparison of the D.O. data (Figure 4.15) for run I with the BOD removal rate data of Figure 4.5 shows no evidence of such limitation. Since BOD removal rates were as high or higher in runs JJ and HI and D.O. levels were lower it appears that D.O., as measured in the trough, is not an indicator of the availability of D.O. to the organisms in the biofiim. Additional support for this argument is found by examining the MLVSS data. In run II (section MLTSS in the trough increased throughout the run (see Figure 4.18) yet, BOD removal rate (see Figure 4.5) decreased. The conclusion could be made that either the microorganisms in suspension in the trough were inactive or were limited in their ability to remove BOD by lack of D.O during run JJ. This inactivity my have come about by a change of population (from aerobic to anaerobic) in the trough and a period of readjustment would be required. This argument is supported by Figure 4.12 which shows an increase in COD accumulation during run JJ and a decrease during run III, which was continued after run H. The higher levels of D.O. observed in run IV are due to the greater solubility of oxygen in water at lower temperatures and the lower B O D concentration (see Figure 4.12) in the trough due to a decrease in water loss. In general, dissolved oxygen levels did not appear to have an effect on BOD and COD removal. Lo et al. (1991) tested the effect of D.O. on treatment efficiency. They used an activated sludge reactor at 3 days HRT, 20°C and a CTMP wastewater with similar organic composition to that used in this study. They found that D.O. had an insignificant effect on B O D 5 removal and COD removal increased slightly with a large increase in D.O. concentration. However, their tests were conducted at D.O. concentrations ranging from 2.5 to 8.2 mg/1. Tests were also performed to observe the effects of variation in disc rotational speed (RPM) and the addition of air by means of a submerged diffuser on D.O. concentration. At the end of run HI, Chapter 4. Results and Discussion 89 at which time the D.O. level was very low, neither an increase in the RPM of the discs nor the provision of additional air by means of aerators was sufficient to change the D.O. concentration in the RBC reactor. However, when the same tests were repeated after run IV, in which the D.O. concentration was not limiting, it was found that the increase in D.O. was more sensitive to an increase in RPM than to aeration. D.O. measurements were performed at the inlet end of the RBC vessel which measured a D.O. concentration of 0.7 mg/1 at a tip speed of 0.31 m/s. Increasing the R P M 1.5 times increased the D.O. to 1.4 mg/1. Doubling the R P M increased the D.O. to 2.2 mg/1, and when tripled the D.O. concentration was measured at 3.15 mg/1. The tip speed was then again set at 0.31 m/s and an airs tone was added to the reactor at the inlet end. Airflow was not measured, but the D.O. did not respond as well as during the RPM test. Increasing the airflow from small dispersed air bubbles to the point where air bubbles created intense turbulence resulted in a D.O. concentration which varied from 1.5 to 2 mg/1 maximum. 427 E F F E C T O F OXIDATION REDUCTION POTENTIAL D A T A ANALYSIS: Based on the low D.O. concentration in the reactor during run II, the oxidation reduction potential (O.R.P.) in the RBC was measured daily during run HI. The O.R.P. in the trough was measured to be -325 ± 57 mV and the O.R.P. of the influent was measured at -230 ± 24 mV. These values were averaged over run in. DISCUSSION: The oxidation reduction potential measures the oxidizing or reducing properties of a solution. The chief oxidant is oxygen, and its presence raises the potential. Reducing agents lower it. Aerobes oxidize organic materials and aerobic processes are dominant when oxygen levels and ORP's are high (positive). Anaerobic processes are prominent when the oxygen level is low and the ORP is also low (negative). An electron activity (pE-pH) diagram can be used to illustrate major oxidation-reduction zones in aqueous systems (Sawyer and McCarty, 1978, p.210). It shows that for an average pH of 8.93 (during run Ul) completely aerobic conditions are associated with an O.R.P. Chapter 4. Results and Discussion 90 of about 750 mV and completely anaerobic conditions are associated with an O.R.P. of about -400 mV. In between these extremes lies a transition zone. Based on the above values the results from this experiment seem to indicate that conditions in the trough during run IH, and probably during run II, were anaerobic. 43.8 EFFECT OF pH D A T A ANALYSIS: pH was measured daily during all four runs at three points in the trough (see section 3.2.2). An average of the pH values was taken because no significant deviation in the data as a function of position was observed. Table 4.4 shows the averaged values for each run. Table 4.4 pH i n RBC and pH of raw wastewater during runs I, I I , III and IV Parameters pH i n RBC Range Average pH of i n f l u e n t HRT = 2 days Temp. = 32°C 8.26 - 8.89 8.68 5.9 HRT = 1 day Temp. = 33°C 7.54 - 9.11 8.68 6.42 HRT = 1.5 days Temp. = 31°C 8.35 - 9.07 8.93 6.37 HRT = 1.5 days Temp. = 20°C 8.46 - 8.81 8.59 5.56 DISCUSSION: It was observed during all four runs that the pH became basic after RBC treatment. The biochemical oxidation of organic matter by microorganisms can be described by the following equation: CTMP wastewater + O z -> C 0 2 + H 2 0 + more sludge The acids in the CTMP wastewater are oxidized to C 0 2 which is exchanged between the atmosphere Chapter 4. Results and Discussion 91 and the water. This will increase the pH and shift the equilibrium to the bicarbonate form. C O z volatility decreases with increasing pH. This process continuous until the carbonic acid stays in its buffer form, and at pH of - 9 almost all alkalinity is in the bicarbonate (HC0 3 ) form. The volatility of carbon dioxide increases with increasing temperature. From the data there is no apparent trend noticeable when relating the influent and effluent pH with temperature or hydraulic residence time. Keeping in mind that each CTMP batch had a different composition of organic compounds (see table 4.2), different amounts of acids will be oxidized by microorganisms. 43.9 ANALYSIS O F V O L A T I L E COMPOUNDS High evaporation rates especially during run LI and HI led to the suspicion that other volatile compounds besides water vapour might be escaping from the RBC. Our measurements, as do literature values, indicate that a significant amount of acetic acid and sulfur compounds are present in the CTMP wastewater. Because of the low D.O. concentration in the trough during run U and III, hydrogen sulfide may have been generated by anaerobic microorganisms, as well. Both these compounds are volatile under certain conditions, and could be removed simply by evaporation. To determine whether or not this was important, inlet and outlet samples of acetic acid and of sulfur compounds were analyzed during runs II and III and compared to the 'sterile' runs because the pH differed (see section 4.3.1). Equilibrium calculations of the concentrations of the volatile and nonvolatile forms of known compounds in the CTMP wastewater at the pH of operation have been done according to Benefield et al. (1982). D A T A ANALYSIS: Table 4.5 shows the equilibrium concentrations of the measured compounds during runs II and HJ compared to the 'sterile' run at HRT=2 days, 1.5 days and 1 day. Other volatile compounds in the CTMP wastewater, some of which were measured by other investigators and which are present in small quantities and others which are by-products from bacterial fermentation, were investigated for their volatility. They are listed in Table 4.6. Chapter 4. Results and Discussion 92 Table 4.5 Equilibrium concentrations of measured chemical compounds i n the CTMP wastewater Compounds: Chemical equation: E q u i l i b . const. 1: A c e t i c A c i d CH3COOH »* CH3COO_ + H+ pKa =4.7 S u l f i d e H2S * HS" + H+ pKa = 6.992 Components: v o l a t i l e CH3COOH non-v o l a t i l e CH3COO" v o l a t i l e H2S non-v o l a t i l e HS" Concentration f o r HRT=1 day (mg/1): - at pH 8.932 - at pH 7.433 0.0265 1.58 449.21 443.63 0.041 1.48 3.51 2.10 Concentration f o r HRT=1.5 days (mg/1): - at pH 8.942 - at pH 7.243 0.024 1.29 422.13 443.92 0.038 1.29 3.40 2.29 Concentration f o r HRT=2 days (mg/1): - at pH 8.82 - at pH 7.153 not done 0.827 not done 444.38 not done 0.96 not done 2.62 E f f e c t of pH: Increasing the pH s h i f t s the reaction e q u i l i b r i u m to the r i g h t which decrea-ses the v o l a t i l i z a -t i o n of CH3OOH. Mini-mal v o l a t i l i z a t i o n occurs at pH > 9. Increasing the pH moves the reaction equilibrium to the r i g h t which decrea-ses the v o l a t i l i z a -t i o n of H2S. Minimal v o l a t i l i z a t i o n occurs at pH > 10. E f f e c t of temperature: pKa i s temperature independent. A l a r g e r increase i n temperature moves the reaction e q u i l i -brium to the r i g h t which decreases the v o l a t i l i z a t i o n of H2S. Values f o r pKa and enthalpy are from Handbook of Chemistry and Physics, 64th e d i t i o n , 1983-84. Temperature corrections are adjusted to 30° C. 2pH values f o r runs I, II and I I I (HRT =2, 1, and 1.5 days) were averaged over the steady s t a t e period. 3pH values f o r the ' s t e r i l e ' runs were averaged. Equilibrium considerations of chemical compounds (not measured) in the CTMP wastewater. Components: Chemical equation: Equilibrium constant (pKa) 1: Comments: Phenol CtH5OH «• C,H!0- + H* v o l a t i l e : CsH5OH non-volatile: C6H50" 9.9' Increasing the pH moves the reaction equilibrium to the right which decreases the v o l a t i l i z a t i o n of C6H5OH. Minimal v o l a t i -l i z a t i o n occurs at pH >13. Alcohol R - OH The OH group does not ionize and chemistry of alcohol relays en t i r e l y on OH group (Sawyer and McCarty, 1978). Therefore alcohols are not affected by pH. V o l a t i l i z a t i o n depends on molecular weight and vapour pressure of species. Alcohols have generally a high vapour pressure. They are v o l a t i l e up to a molecular weight of - C 1 0 at 1 atm. Carbonic Acid * C02 + HjO - HCOj- + H* v o l a t i l e : C02 non-volatile: HC03" b HCOj" - CO/ + H* 6.433 10.263 "Increasing the pH moves the reaction equilibrium to the right which decreases the v o l a t i l i z a t i o n of C02. Minimal v o l a t i l i -zation occurs at pH > 9. Almost a l l a l k a l i n i t y i s i n the bi-carbonate form at a pH ranging from 7.5 - 9. A large increase i n temperature moves the reaction e q u i l i b r i -um to the right which decreases the v o l a t i l i z a t i o n of C02. bIncrease i n pH moves the reaction equilibrium to the right, increasing the carbonate concentration i n solution. A large increase i n temperature moves the reaction e q u i l i b r i - . um to the l e f t increasing the bi-carbonate concentration. Hydrocarbons many forms: -saturated -unsaturated V o l a t i l i z a t i o n depends on molecular weight and vapour pressure of species. Saturated and unsaturated hydrocarbons are generally v o l a t i l e up to a molecular weight of - C 5 resp. - C, at 1 atm. Resin acids many forms Resin acids are generally non-volatile because of the i r high molecular weight and complex structure. 'Values for pKa and enthalpy are from Handbook of Chemistry and Physics, 64th edition, 1983-84. JpKa values for 25 °C. 3pKa values adjusted to 30 °C. Chapter 4. Results and Discussion DISCUSSION: From Table 4.5 it appears that at the pH of the wastewater in the trough during the bacterial runs acetic acid and hydrogen sulfide exist almost completely in their non-volatile forms. However, the pH during the 'sterile' experiment was at a lower average value of 7.3. At that pH approximately 40% of HjS could be volatile. Table 4.6 shows that the volatilization of hydrocarbons and alcohols can only be determined if the form of the hydrocarbons or the type of alcohol is known. Phenols, which are usually present in CTMP wastewaters at smaller concentrations, are volatile up to a pH of 13. ACETIC ACID ANALYSIS D A T A ANALYSIS: Table 4.7 shows the results of the acetic acid analysis. All effluent concentrations were corrected for water loss. Because of the high acetic acid concentration in the CTMP wastewater, B O D 5 and COD analyses of known concentrations of acetic acid were performed to find out the contribution of acetic acid to the total B O D 5 and COD concentration. It was found that the BOD of acetic acid was equal to 0.9 times the acetic acid concentration and that the COD of acetic acid was equal to 1.033 times the acetic acid concentration. Figure 4.16 then shows a bar graph comparing influent and effluent concentrations of acetic acid at various HRTs. Table 4.7 Comparison of a c e t i c a c i d concentration to BOD and COD concentrations and e f f e c t of a c e t i c a c i d on hydraulic retention time. Parameters: (days) A c e t i c Acid Concentration (mg/1) i n out BOD cone, of Acetic Acid (mg/1) i n out COD cone, of Ace t i c A c i d (mg/1) i n out HRT = 1 HRT = 1.5 422.16 449.25 13.64 3.49 379.95 404.31 12.28 3.61 436.24 464.21 14.09 3.61 DISCUSSION: The BOD value for acetic acid is very close to the actual acetic acid concentration, Chapter 4. Results and Discussion Figure 4.16 E f f e c t of hydrau l i c re tent ion time on a c e t i c a c i d concentrat ion . BOD.^,. . . and COD ' a c e t i c a c i d a c e t i c a c i d Chapter 4. Results and Discussion 96 indicating that 90% of acetic acid is biodegradable. This shows that acetic acid is readily available to the organisms and may be one of the compounds in the CTMP wastewater which is consumed by them first. The theoretical oxygen demand of acetic acid was calculated to be 1.067. This value differs by 3% from the measured COD value. Figure 4.16 shows that almost all of the acetic acid was biodegraded and a larger amount was removed when the H R T was increased from 1 day to 1.5 days. S U L F U R C O M P O U N D S D A T A ANALYSIS: Table 4.8 shows the results from the sulfur compounds analysis. Analysis was done for sulfate (S0 4 =), sulfite (SO s =) and sulfide (S=). All effluent concentrations were corrected for water loss. Figure 4.17 shows the effects of HRT on sulfate, sulfite and sulfide. Table 4.8 S0 4 =, S0 3 = and S= concentrations during run II and run I I I . Parameters: (days) i n s o -(mg/1) out i n s o 3 = (mg/1) out i n S = (mg/1) out HRT = 1 435. 0 514.5 65. 5 8.95 3.55 0.97 HRT = 1.5 455. 0 498.0 57. 5 5.70 3.44 0.26 DISCUSSION: Sulfate and sulfite compounds were analyzed because they exist in large concentrations in the CTMP wastewater; sulfide was analyzed to see if anaerobic bacteria were reducing S 0 4 = and S O s = to S = to produce hydrogen sulfide. Sulfur may also exist, in organic form, in the wastewater as lignosulfate. However, lignosulfate is said to be present only as a minor part of the raw CTMP effluent (Almemark et al., 1986; Cornacchio and Hall, 1986). The results in Table 4.8 show that most of the sulfite was oxidized to sulfate during RBC treatment. It appears that small amounts of H 2S escaped, but it is not evident from the data Chapter 4. Results and Discussion 97 Figure 4.17 E f f e c t o f h y d r a u l i c retention time on s u l f a t e , s u l f i t e and s u l f i d e . Chapter 4. Results and Discussion 98 if any sulfide was produced by anaerobic fermentation in the trough during RBC treatment. Sulfate and sulfite were present at different concentrations in the raw wastewater during run II and run III and no apparent relationship can be seen from the data. From Table 4.8 and Figure 4.17 it appears that more hydrogen sulfide evaporated at a longer HRT. 4.2.10 SOLIDS ANALYSIS EFFECTS OF TS, VS, TSS AND VSS D A T A ANALYSIS: The results for TS, VS, TSS and VSS obtained during runs I, U, III and IV are presented in Table 4.9. Effluent concentrations are shown both, corrected and not corrected for water loss. Table 4.9 Comparison of i n f l u e n t and eff l u e n t s o l i d s during runs I, I I , III and IV. Solids (mg/1) HRT=2 days Temp.= 32°C HRT=1 day Temp.= 33°C HRT=1.5 days Temp.= 31°C HRT=1.5 days Temp.= 20°C TS i n out corr. 6200 6925 7350 7250 6700 26675 22350 6800 4850 3350 3000 5000 VS i n out corr. 4200 4925 5225 5250 1900 12875 9100 4000 1375 1625 1225 2950 TSS i n out corr. 3600 900 1000 950 4950 8925 4350 2250 3575 1125 575 1650 VSS i n out corr. 250 850 950 900 1400 6700 3775 2100 1000 850 500 1550 Chapter 4. Results and Discussion 99 DISCUSSION: Total solids (TS) and volatile solids (VS) analysis were done to compare to total suspended solids (TSS) and to volatile suspended solids (VSS) data. Most of the matter in TS and VS tests is in the dissolved form and consists mainly of inorganic salts, small amounts of organic matter and dissolved gases (Sawyer and McCarty, 1978). Undissolved substances are measured as TSS and VSS. Comparing the TS with the TSS test indicated that approximately 87% of the solids in the CTMP wastewater during runs n, HI and IV consisted of dissolved matter. However, during run I approximately 42% of the raw CTMP effluent solids consisted of dissolved material. The TS test can also be compared to B O D 5 and COD analysis. It was originally devised as a means of evaluating the pollutant matter present in wastewaters. TS values compare well to the B O D 5 and COD measurements obtained in this study (see Figure 4.12, section and Appendices I to IV, raw data). However, B O D 5 and COD tests are said to evaluate the strength of such materials more exactly (Sawyer and McCarty, 1978). The volatile content, VS, of the total solids test is used to measure the amount of organic matter present in the wastewater. The large difference in value, when comparing the VS data to the VSS data confirms the amount of inert organic matter, such as the lignin present in the CTMP wastewater. The corrected VS values after RBC treatment indicated that during run I (HRT=1 day) and run II (HRT=1 day) 67% of the VS was removed while during run III (HRT=1.5 days, T=31°C) 70% of VS was removed and during run IV (HRT=1.5 days, T=20°C) 44% was removed. This confirms the observations made in section 4.2.5 that a larger amount of high molecular weight compounds such as lignin can be removed biologically at a higher wastewater temperature. The best VS removal was attained during run HI at HRT=1.5 days, while the best BOD removal was achieved during run I at HRT=2 days. TSS values show the amount of undissolved matter in the wastewater. These amount to about 13% in the raw wastewater during runs II, III and IV, however, they account for about 58% during run Chapter 4. Results and Discussion 100 L Data for the corrected RBC effluent values show that the best removal of undissolved material was reached during run HI with 43% removed, while the VSS concentration during run II and IV actually increased. Because almost all of the TSS is in form of VSS during runs H, in and IV the increase in TSS is attributed to bacterial washout. Run I was not compared because its TSS and VSS composition seemed to differ from the CTMP batches tested during runs JJ, HI and IV. VSS data in table 4.9 indicate that a large amount of organic solids is present in the raw CTMP effluent prior to treatment About 95% of the TSS was in the form of VSS in the CTMP wastewater used for runs H, HI and VI. However, only about 7% of TSS was in the form of VSS during run I. Corrected effluent data shows that the VSS concentration increased because of bacterial washout. The data reflect a decrease in VSS with a decrease in effluent flowrate, which was most noted in run IV followed by runs I, H and run HI were the effluent flowrate was lowest. BIOMASS G R O W T H IN R E A C T O R O V E R TIME OBJECTIVE: To investigate the growth of biomass in the trough over time. PROCEDURE: During run n (HRT=1 day) from November 28, 1990 over a weeks period until December 5,1990 daily samples of the mixed liquor were withdrawn as described in section 3.2.3 and the mixed liquor total suspended solids (MLTSS) concentration was determined as described in section 3.3.10 (c). RESULTS: See results of the raw data Appendix H (Run H) and Figure 4.18. DISCUSSION: During run H it was observed that the sludge production in the trough increased visibly. Some of the bacterial increase was contributed to biomass sloughing off the discs, as well as to microbial growth occurring in the trough. The result of this experiment shows that the sludge concentration during the run doubled within a week from 6000 to 11,950 mg/1. These MLTSS values correspond to MLVSS concentrations ranging from - 5,275 mg/1 (calculated value based on the data that the MLVSS is 87.9% of the MLTSS) to 10,500 mg/1. This confirmed that the biomass was Chapter 4. Results and Discussion 101 12000-1 10000 bo B, C Q C Q I 8000 H 6000 E q u a t i o n o f best f i t : Y = 2500 • LN(X) + 6600 4000 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 n 0 2 4 6 8 10 d a y s 12000-1 IOOOOH C Q C Q E H 8000 H 6000-t 4000 d a y s Figure 4.18 Biomass growth i n RBC over time during run II. Chapter 4. Results and Discussion 102 actively being created, however, during the time the MLTSS analysis was done, biomass was sloughing off most the discs and because of the high evaporation (87% average) during this run hardly any washout of sludge occurred. Because of the large concentration of biomass present in the trough any D.O. transferred into the trough liquid was probably consumed instantaneously. 4.2.103 ANALYSIS O F MLVSS D A T A ANALYSIS: Figure 4.19 shows a bar graph of the average MLVSS concentrations in the RBC for runs I, U, Ul and IV. Samples were taken after running the RBC for 10 days. For run I (HRT=2 days, T=32°C) the MLVSS concentration was 2400 mg/1, for run H (HRT=1 day, T=33°C) the MLVSS was 9800 mg/1, for run HI (HRT=1.5 days, T=31°C) the MLVSS was 10,650 mg/1, and for run IV (HRT=1.5 days, T=20°C) the MLVSS was 3800 mg/1. DISCUSSION: Not much in the way of conclusion can be drawn from the MLVSS data. That run III has a higher MLVSS than run II is because run HI followed directly on run II without removing sludge. Washout of sludge in these two runs was low due to high levels of evaporation. Thus sludge which sloughed off the discs tended to accumulate in the RBC trough. This suspended biomass did not seem to contribute to BOD removal as noted above in section 4.2.5 perhaps because of low levels of D.O. However, may have contributed to COD removal in run HI (see Figure 4.12). Sludge was scraped of the discs and measured after runs I and HI. The collected sludge was drained and weighed. The weighed amount was then divided by the total disc area. 0.446 kg/m2 of disc were found for run I and for run JH 0.526 kg/m2 of disc. Further, a sample of the drained sludge was withdrawn for TSS and VSS and measured according to section 3.3.10. The TSS was 55,800 mg/1 and the VSS was 32,000 mg/1 for run I, and the TSS was 58,600 mg/1 and the VSS was 17,000 mg/1 for run HI. Chapter 4. Results and Discussion 1 2 R u n I Run II Run III Run IV Figure 4.19 MLVSS concentration i n the bio-reactor during runs I I , I I I and IV. Chapter 4. Results and Discussion 104 4.2.11 NUTRIENT ADDITION D A T A ANALYSIS: Nutrients were measured as total Kjeldahl nitrogen (TKN), which determines the total organic nitrogen present, and as total phosphorus (TP). Table 4.10 shows data of the nutrient ratios for run I, TJ and m for influent and effluent CTMP wastewater. Table 4.10 Ratio of nutrient addition of Carbon:Nitrogen:Phosphorus measured as BOD5:TKN:TP BOD:TKN:TP r a t i o HRT = 1 day HRT =1.5 days HRT = 2 days Raw CTMP wastewater 100:0.46:0.08 100:0.54:0.2 100:0.54:0.09 Influent with nutrient 100:5:1 100:5:1 100:5:1 RBC e f f l u e n t 100:8.15:0.52 100:4.68:0.29 100:16.2:1.53 RBC e f f l u e n t corrected 100:8.25:0.53 100:4.47:0.28 100:16.3:1.54 A mass balance was also done on the nitrogen and phosphorus addition to the RBC treatment system during steady state. Nutrient - nutrient = nutrient consumed + ammonia flowrate in flowrate out by bacteria loss to the environment Table 4.11 shows the results of the mass balance. Chapter 4. Results and Discussion 105 Table 4.11 Results of mass balance for nitrogen and phosphorus flowrate f o r RBC treatment during runs I, II and I I I . Nutrients HRT (days) Nutrient flowrate i n (g/day) Nutrient flowrate out (g/day) Nutrients consumed by bacteria + ammonia lo s s TKN 1 2.000 0.716 1.285 1.5 1.296 0.178 1.118 2 1.145 0.260 0.885 TP 1 0.401 0.046 0.355 1.5 0.260 0.011 0.249 2 0.228 0.025 0.203 DISCUSSION: The data from Table 4.10 indicated that the BOD 5:TKN:TP ratio measured in the raw wastewater was too low to encourage optimal microbial growth. Nitrogen and phosphorus were added to the wastewater at a conservative ratio of 100:5:1 to assure that nitrogen and phosphorus would not become limiting factors during this study. However, the RBC effluent data showed an increase of T K N during runs I and II (HRT = 2 days resp. 1 day). These results agree with previous research (Servizi and Gordon, 1986; Beak Consultants Ltd, 1986; and Schnell et al., 1990). During run HI (HRT = 1.5 days) however, the nitrogen concentration did not change much. The results of the mass balance from Table 4.11 indicate that nutrient consumption of the bacteria and possible ammonia loss increased when the BOD (COD) removal rate increased (see Figure 4.5 and 4.9). Concentrations of effluent T K N ranged from 391 to 86 mg/1 in the concentrated effluent and from 66 to 20 mg/1 when correcting the effluent T K N concentration for evaporative water loss. These nitrogen concentrations are very high considering that the lethal level for ammonia is 0.2 mg/1 (Servizi and Gordon, 1986). Chapter 4. Results and Discussion 106 43. O T H E R ANALYSIS 43.1 'STERILE' R U N OBJECTIVE: To investigate if volatile compounds could be removed simply by evaporation baseline runs though to be devoid of microbial activity were conducted at various H R T s as described in section 3.2.2. P R O C E D U R E : 50 mg/1 of mercuric chloride (HgClj) was added to the raw wastewater as described by Gledhill (1975) to try to sterilize the CTMP wastewater. RESULTS: The raw data is presented in Appendix V, A.V.1. Equilibrium concentration results are shown in Table 4.5 and results of percent organic removals (BOD 5 and COD) are shown in Figure 4.20. DISCUSSION: Each of the runs was carried out for two theoretical hydraulic retention times. After analyzing the BOD and C O D data, it was a surprise to find out that both BOD and C O D reductions were higher with HgCl 2 addition, than when it was not present, especially during run JJ (HRT=1 day) and run HI (HRT=2 days). At first it was assumed that the cause was an analytical error and both, BOD and COD analyses were repeated twice with other sample effluents which had been collected during these runs, but the results were consistent. Since the organics had been somehow degraded the only other conclusion was that HgCl 2 addition failed to kill the active microorganisms. To investigate further, the wastewater was plated on an agar nutrient media to find out if the runs were indeed sterile (see section 4.3.2). These tests revealed that the addition of mercuric chloride did not kill all the organisms present in the wastewater. The BOD^COD removal ratio during these runs was almost 1:1 compared to the lower values (around 0.83) noted in section 43.2 MICROBIOLOGY OBJECTIVE: To investigate if the sterile runs (section 4.3.1) were indeed sterile. Chapter 4. Results and Discussion 107 Figure 4.20 Comparison of %organic removals between b a c t e r i a l and • s t e r i l e 1 runs f o r runs I, II and I I I . Chapter 4. Results and Discussion 108 PROCEDURE: A grab sample of CTMP wastewater was plated as described in section 3.3.14 under the following conditions: (1) raw wastewater, (2) settled residue from wastewater, (3) raw wastewater with mercuric chloride addition, (4) raw wastewater inoculated with aerobic microorganism, (5) raw wastewater inoculated with aerobic microorganism and with mercuric chloride addition. RESULTS: Photographs of the plates were taken after 48 hours of incubation (see Appendix V, Figure A V . l ) . DISCUSSION: Comparison of picture (4) and (5) in Figure A.V.1 indicates that the addition of HgCl 2 kills most bacteria, however organisms believed to be fungi were not affected by it. These photos suggest that bacteria and the fungi could be competing with each other and when bacteria are not inhibited, as by HgCl^ they dominate over the fungi. It is well known that fungi such as the white-rot fungus have been used in the past to degrade lignin (Joyce et al., 1983; Matsumoto et al., 1985), which may explain the high C O D removal during these runs. It was also noted that the pH of the wastewater did not go up as much during the runs (see table 4.5) to which HgCl 2 was added as it did during the runs without HgClj. Picture (1) did not show any bacterial growth in raw CTMP wastewater and consequently, when adding HgCl 2 to the same sample no growth was found on the plate in picture (3). However, when plating settled wastewater (2) some bacterial growth, pink in colour, was found on the plates. These bacteria may be facultative anaerobes. 433 START-UP R U N WITHOUT BIOMASS INOCULUM OBJECTIVE: To investigate and confirm that BOD is not removed simply by evaporation as was indicated by the 'sterile run'. PROCEDURE: Raw CTMP wastewater was added to the reactor without bacterial inoculum and the reactor was run under the same conditions as during run HI for three hours. During this time measurements of D.O., pH, temperature in the RBC, BOD i n , B O D o u t and MLVSS were done every Chapter 4. Results and Discussion 109 1/2 hour. After the test a sample was plated for possible microbial growth during the run. RESULTS: Results are presented in Appendix V, A.V.2 and Figure A.V.1 (6). The pH increased from 5.91 to 6.4 and the BOD reduction increased from 3.17 to 10.08% during the three hours of the run. The D.O. concentration decreased from 7.7 to 6.2 mg/1 while the temperature changed from 28 to 31°C, all other parameters stayed constant. DISCUSSION: During the three hours of the run acetic acid may have been evaporated as its volatility is increased at an average pH of 6.25 or maybe microorganism started to grow on the discs. This decrease of acetic acid in the wastewater increased the pH as well as the BOD reduction. Plating of the wastewater in the trough did not show any microbial growth (Figure A V . l (6)). The increase of temperature was probably responsible for the decrease in D.O., which did not reach steady state as fast as the other parameters (see also section 4.2.5). 43.4 H Y D R A U L I C CHARACTERIZATION O F T H E R E A C T O R OBJECTIVE: To obtain information on the hydraulic residence time distribution (RTD) and flow regime in the RBC. PROCEDURE: A step change of sodium chloride tracer of concentration Co was added to the fluid stream entering the reactor and the time dependent tracer concentration in the exit stream (C) from the vessel was measured. The experimental procedures are described in chapter 3, section 3.2.2 and 3.3.7. RESULTS: The raw data and the results are listed in Appendix VI. The effluent sodium chloride concentration obtained and measured as C/C„ is plotted as a function of time to show residence time distribution. Figure A.VI.1 (Appendix VI) shows a plot of the F-curve (step-change input). DISCUSSION: The effluent conductivity (C) was corrected for a 25% water loss. The observed residence time distribution data for the RBC reactor was statistically analyzed to estimate the mean residence time, the variance and the dispersion number (Levenspiel, 1984). The analysis gave Chapter 4. Results and Discussion 110 estimates of the mean residence time of 35.1 hours which is close to the theoretical residence time of 36 hours. The variance or distribution about the mean (a2) was 1150.4 hours2 and the corresponding ag2 was 0.932. To characterize the nonideal flow within the RBC vessel the dispersion model and the model for a network of ideal mixers (CSTR) in series were used. The dispersion model indicates plug flow condition, if the dispersion number is zero, and CSTR conditions if the dispersion number goes to oo. The condition for using the dispersion model implies that there exists no stagnant pockets and no gross bypassing or short-circuiting of fluid in the vessel (Levenspiel, 1972). For a first approximation, assuming there was no backmixing the dispersion model for large extents of dispersion was used. Fitting this model gave a dispersion number of 0.2385 which is considered a large amount of dispersion and is close to the mixed flow regime (Levenspiel, Ch.9, Fig. 17). Because the dispersion number and the variance both are large the tracer data of the RBC was also fitted to the CSTR in-series model which is given by the equation: No. tanks=— °l This model indicates plugflow conditions for a large number of tanks and CSTR conditions for one tank. Its value was calculated as 1.073 tanks in-series which shows indeed that the reactor behaves like a CSTR. For the RBC reactor design, these analysis indicate that: (1) good mixing is provided in the vessel leaving essentially no dead space; (2) short circuiting is eliminated due to the presence of the rotating media plates; (3) almost complete mixing is achieved in the reactor shown by the large dispersion number and by the CSTR in-series model. It is pointed out that this model describes residence time distribution in a reactor with the biofiim on the media. It is postulated, though not tested, that the biofiim influences the movement of the Chapter 4. Results and Discussion 111 sodium chloride tracer through the system by causing tracer diffusion into and out of the biofilm (Riemer et al., 1980). The biofilm could therefore distort the residence time distribution in the reactor, resulting in a different degree of mixing. The same argument may be made for the water loss in the RBC which by decreasing the flowrate as water moves through the RBC could also distort the mixing regime resulting in poorer mixing. CHAPTER 5 CONCLUSIONS A N D RECOMMENDATIONS Based on the literature review and the experimental work of this study, the major results, together with the conclusions and recommendations drawn from them, are summarized as follows. 5.1 CONCLUSIONS (1) For the ranges of HRT tested the percent BOD and COD removal were 86% BOD and 72% COD reductions at HRT=2 days; 83% BOD and 71% COD reductions at HRT=1.5 days; and 77% BOD and 63% COD reductions at HRT=1 day. (2) Since most experiments were carried out at approximately 30°C high levels of evaporation occurred, because the lab scale RBC exposed a thin film of liquid to the air over an extensive surface. Water loss during runs I to JJ1 ranged between 75% and 87%. The average waterloss at 20°C (run IV) was 29%. AT and % R H were major parameters affecting the water loss of the RBC during this study. It was concluded that seasonal changes can have a major impact on evaporation when running an RBC system inside a temperature controlled building. Evaporation did not appear to have an effect on BOD and COD removal. (3) Increase in flowrate (decrease in HRT) resulted in a slight decrease in percent BOD (COD) removed, an increase in organic loading rate and in BOD (COD) removal rate. (4) For the range of temperatures tested, more B O D 5 and COD can be removed at higher temperatures. 85% BOD was removed at 31°C and 70.1% was removed at 20°C. A higher %COD/%BOD removal ratio of 0.861 at 31°C was found compared to a %COD/%BOD ratio of 0.654 at 20°C (5) For the ranges of treatment conditions studied, levels of dissolved oxygen, pH and MLVSS had 112 Chapter 5. Conclusions and Recommendations 113 no distinct effect on B O D 5 and COD removal. (6) Tests and calculations done to reveal the volatility of major volatile compounds present in the CTMP wastewater and possible BOD loss from the RBC trough to the environment caused by the high evaporation indicated that losses of acetic acid, HjS and B O D 5 were negligible at the pH's tested. (7) Start-up time was minimal; steady state seemed to prevail throughout the experiments. (8) The RBC appeared to respond well to shock loadings caused by the flowrate change in going directly from run II to run III. (9) A 'sterile' run using 50 mg/1 of HgCl 2 revealed that mercuric chloride may inhibit bacteria, however other microorganisms such as fungi are not affected by it. Fungi appear to be able to attack COD much better than bacteria. (10) B O D 5 oxidation lies between 60 to 65% of the ultimate BOD indicating that the oxidation of the CTMP wastewater used during this study is comparable to sewage. No nitrogenous oxygen demand was observed. (11) The residence time distribution curve observed in the RBC reactor using a sodium chloride tracer revealed that the flow regime can be considered to be perfectly mixed. (12) Levels of B O D 5 eliminated by the RBC treatment are comparable to those reported in the literature. At higher temperature COD removal results seemed promising compared to literature values. (13) The H R T s required to achieve high levels of BOD removal are long suggesting that this process may be impractical for full-scale pulp mill installation. (14) The BOD levels in the effluent are high even after correction for evaporation, thus more than one stage of treatment would be necessary to produce a dischargeable effluent. Chapters. Conclusions and Recommendations 114 5.2 RECOMMENDATIONS Future research is recommended in the following areas. (1) An increase in temperature indicated higher treatment efficiency at the temperatures tested during this study, especially for COD removal. Enclosing the RBC system to test for optimum treatment temperature may be the next logical step to do. Temperatures may be increased to levels were thermophilic bacteria become active. (2) Using RBC's with more than one stage, separated by baffles is said to improve treatment efficiency. It is recommended to test an RBC with stages, possibly two to three. This may improve B O D 5 and COD removal further. (3) Nitrogen levels of the RBC effluent were high during this study and it appeared that nutrient addition may be reduced. It is recommended to test if nutrients need to be added to the raw CTMP wastewater. (4) Literature has shown that high BOD and COD reduction does not imply toxicity removal and it is suggested that the RBC effluent should be tested for toxicity. (5) The 'sterile' run conducted in this study, as does some literature, indicates that fungi may better attack high molecular weight compounds, such as lignin. It would be of particular interest to use fungi in an RBC reactor, to see if COD could be reduced dramatically. (6) Literature has shown that it is possible in CTMP effluent treatment to use anaerobic treatment followed by an aerobic post-treatment to reduce toxicity. It would be interesting to see how effective the RBC would be in effluent polishing after anaerobic treatment. (7) The high levels of evaporation and the significant reductions in BOD experienced in this work with an RBC might be a way of getting rid of a lot of water in an attempt to approach a zero pulp mill effluent. If enough water could be removed perhaps the concentrated effluent could be incinerated. However, it should be kept in mind that in these experiments the temperature was kept high by adding heat. In a full-scale unit the temperature of the wastewater would decrease as a result Chapter 5. Conclusions and Recommendations of evaporation. BIBLIOGRAPHY Ahlberg N.R., Kwong T.S. Process Evaluation of a Rotating Biological Contactor for Municipal Wastewater Treatment, Research Paper No. W2041, Wastewater Treatment Section, Pollution Control Planing Branch, Ministry of Environment, Ontario Canada, November 1974. Alleman J.E., Veil J.A., Canaday J.T. Scanning Electron Microscope Evaluation of Rotating Biological Contactor Biofiim, Water Res., Vol. 16, 1982:543-550. Almemark M., Boman B., Frostell B., Solyon P. CTMP Production and Environmental Aspects, State of the Art, Institute for Vatten-Och Luftvardsforskning, T860316, 1986. Amer. Public Health Assn. (A.P.H.A.) Standard Methods for Examination of Water and Wastewater, Washington, D.C.: A.P.H.A., 16th ed., 1985. Andersson P.E., Gunnarsson L., Welander T , Wikstroem A. Detoxification of CTMP Effluent to Enable Anaerobic Treatment, Biotechnology in the Pulp and Paper Industry, 3rd Int. Conf. (Stockholm), June 16019, 1986:207-208. Andreadakis A.D. Design of Multistage Rotating Biological Contactors, Journal of Environmental Engineering, Vol. 113, No. 1, February 1987:199-205. Antonie R.L., Kluge D.L., Mielke J.H. Evaluation of a Rotating Disk Wastewater Treatment Plant, Journal WPCF, Vol. 46, No. 3, March 1974:498-511. Antonie R.L. Fixed Biological Surfaces - Wastewater Treatment: The Rotating Biological Contactor, CRC Press, Cleveland, Ohio, 1976. Atkinson B., Davies LJ. The Overall Rate and Substrate Uptake (Reaction) by Microbial Films, Transaction of the Institute of Chemical Engineers, Vol. 52, No. 3, July 1974. Autotrol Wastewater Treatment Design Manual Autotrol Corporation, Bio-Systems Division, Milwaukee, Wisconsin, 1978. Baker C D . , Nachbar R.H., Summer R.E. Pilot Plant Studies of Biological Treatment of Magnesium Base Sulphite Mill Effluent - Aerated Stabilization Basin and Rotating Biological Surface, Preprints: 1978 International Sulphite Conference, September 12-14, 1978:69-80. Beak Consultants Ltd Anaerobic Treatment of TMP/CTMP Wastewater, Report for Environment Canada, Wastewater Technology Centre, Burlington, Ontario, September 1986. Benefield L.D., Judkins J.F., Weand B.L. Process Chemistry for Water and Wastewater Treatment, Prentice Hall Inc., N.J., 1982. Bennett D.H., Reeser D., Cambell A G , Falter M.C. Pre-Treatment of CTMP Effluent by Lime to Reduce Resin Acids, Sulfite, and Toxicity, TAPPI Proc: Environmental Conference, 1988:199-207. 116 Bibliography 117 Bennett D.J., Needham T., Summer R.E. Pilot Application of the Rotating Biological Surface Concept for Secondary Treatment of Insulating Board Mill Effluents, TAPPI, Vol. 56, No. 12 Dec. 1973:182-187. Bonsor N. , McCubbin N. , Sprangue J.B. Municipal-Industrial Strategy forAbatment (MISA), Kraft Mill Effluents in Ontario, Environment Ontario, 1989. Bosia A., Nisi D., Cappelletto P. Chemithermomechanical Pulps from Hardwood and their Possible Use in Papermaking, Proc: XXII E U C E P A Conf. (FLorence), Vol. 1, No. 1-21, October 6-10,1986:4.1-4.19. Bracewell L.W., Jenkins D., Cameron W. Treatment of Phenol-Formaldehyde Resin Wastewater using Rotating Biological Contactors, Proc.: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, February 1980. Brenner R.C., Heideman J .A , Opatken E J . , Petrasek Jr. A C . Design Information on Rotating Biological Contactors, EPA-600/2-84-106, Cincinnati, Ohio, June 1984:1.1-2.54. Bunch R.L. Rotating Biological Contactors - Are All System Go, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, Pennsylvania, E P A 600/9-80-0436A, February 4-6, 1980:1-4. Castagne M.R. Miller Western's Success in Marketing Bleached CTM Pulp Leads to a $ 250 Million Second Mill, Paper Age, May 1990:10. C C R E M Canadian Water Quality Guidelines, Water Quality Branch, Inland Waters Directorate, Environment Canada, 1974. Characklis W.G., Trulear M.G. Dynamics of Microbial Film Processes, Proc: 1st National Symposium on Rotating Biological Contactor Technology, February 1980. Chen H.T., Fredrickson E E . , Cormack J.F., Young S.R. Evaluation of Four Biological Systems on Integrated Paper Mill Effluent, TAPPI, Environmental Conf., April 17-19, 1974:143-156. Cheng N.W., Molof A H . Effect of Organic Loading on RBC Process Efficiency and Fixed-Film Thickness, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, EPA-600/9-80-046A February 406, 1980:137-165. Chesner W.H., Molof A H . Biological Rotating Disk for Scale-up Design: Dissolved Oxygen Effects, Progr. Water Technology, Vol. 9, 1977:811 Chesner W.H., Iamone J. Review of Current RBC Performance and Design Procedures, Report prepared for U S E P A Municipal Environmental Research Laboratory, Cincinnati, Ohio, Contract No. 68-02-2775, by Roy F. Weston Inc., 1985. Congram G.E. Biodisk Improves, The Oil and Gas Journal, February 23, 1976:126-132. Cornacchio L A , Hall E.R. Characteristics of CTMP Wastewaters, Preprints: Treatment of Chemithermomechanical Pulping Effluents (Vancouver, B.C.), October 27, 1988:1-14. Bibliography 118 Del Borghi M., Palazzi E. , Parisi F., Ferraiolo G. Influence of Process Variables on the Modelling and Design of a Rotating Biological Surface, Water Res., Vol. 19, No. 5, 1985:573-580. Duvernychuk L.W. The Receiving Environment of the Upper Fraser River: A Pilot Environmental Effects Monitoring Program Examining Physical/Chemical/Biological Elements of the System Related to Pulpmill Effluents, Hatfield Consultants Ltd., West Vancouver, B.C., 1990. Environment Ontario, Resin Acids, Scientific Criteria Document for development of Provincial Water Quality Objectives and Guidelines, ISBN 0-7729-4247-8, September 1988. Environment Ontario, Development Document for the Effluent Monitoring Regulation for the Pulp and Paper Sector, 1989. Famularo J., Mueller J., Mulligan T. Application of Mass Transfer to Biological Contactors, Journal of WPCF, Vol.50, No.4, April 1978. Forgie D.J.L. Principles and Applications of the Rotating Biological Contactor, Proc.: 2nd Int. Symp. (Waterloo, Ont.), June 18-20, 1980:253-269. Friedman A A RBC Design Consideration for Industrial Wastewaters, Proc.: 1st National Symposium/Workshop on Rotating Biological Contactor Technolgy, EPA-600/9-80-046B, February 4-6, 1980:909-919. Fry F.F., Smith T.G., Sherrard J.H. Start-Up and Shock Loading Characteristics of a Rotating Biological Contactor Package Plant, Proc.: 1st International Conference on Fixed-Film Biological Process, University of Pittsburg, 1982:542-569. Gillespie W.J., Duane W.M., Springer A M . A Pilot-Scale Evaluation of Rotating Biological Surface Treatment of Pulp and Paper Mill Wastes, TAPPI, Vol. 57, No. 9, September 1974:112-117. Gledhill W.E. Screening Test for Assessment of Ultimate Biodegradability: Linear Alkylbenzene Sulfonates, Applied Microbiology, December 1975:922-929. Grieves, C.G. Dynamics and Steady State Model for the Rotating Biological Disc Reactor, Ph.D. Thesis, Clemson University, South Carolina, 1972. Guardino C.F., Nelson M.D., Marazzo W J . Philadelphia Develops Surfact, 50th WPCF Conference, 1977. Gutierrez A , Bogert A L . , Scheible O.K., Mulligan T J . Upgrading Primary Tanks with Rotating Biological Contactors, EPA-6-/2-80-003, NTIS PB80-203771, March 1980. Habets L . H . A , de Vegt A L , Anaerobic Treatment of Bleached TMP and CTMP Effluents in the BIOPAQ UASB System, 3rd IAWPRC Symposium on Forest Industry Wastewaers (Tampere, Finland), 1990. Harrison R . A (ed.) O&Mof Trickling Filters, RBC's and Related Processes, Manual of Practice O M -10, WPCF, 1988. Bibliography 119 Hartung J.W. RBC Supplemental Air: Continuous or Intermittent? Proc.: 1st International Conference on Fixed-Film Biological Process, University of Pittsburg, 1982:1688-1694. Hitdlebough J. A , Miller R.D. Full-Scale Rotating Biological Contactor for Secondary Treatment and Nitrification, Proc.: 1st National Symposium on Rotating Biological Contactor Technology, February 1980. Hoehn R.C, Ray A D Effects of Thickness on Bacterial Film, Journal WPCF, Vol. 45, No. 11, November 1973:2303-2320. Hormel Design Manual for RBC, 1979. Howard T.E., Walden C C . Effluent Characteristics and Treatment of Mechanical Pulping Effluents, presented at Int. Mech. Pulping Conf. (San Francisco) July 16-20, 1975:197-201. HPD Inc. Freeze Crystallization for Pulp Mill Effluent Treatment, HPD Inc., 305 E . Shuman Boulevard, Naperville, 111. Huang J. Operational; Experiences of Oxygen-Enriched Rotating Biological Contactor, Proc: 1st National Symposium on Rotating Biological Contactor Technology, February 1980. Hultman B., Hillstroem R. Domsjoe Sulphite Mill - Environmental Protection by the Development and Use of New Technology, Proc: E U C E D A Symposium Environ. Protection in the 90's (Helsinki, Finland), May 19-22, 1986:78-85. Jackson M. , Ackerlund G. CTMP Production and End-Use Applications in Scandinavia, TAPPI Proceedings, Pulping Converence, 1984:115-122. Joyce T.W., Chang H-m., Cambell A G . Jr., Gerrard E.D., Kirk T.K.A Continuous Biological Process to Decolorize Bleach Plant Effluents, Biotechnology in the Pulp and Paper Industry, London 12th, 13th, and 14th September 1983:30-38. Kennedy K.J., Andras E. , Elliott C M . , Methven B. Effect of Intermittent Feeding on Anaerobic CTMP Wastewater Treatment, submitted to Environmental Technology Letters, 1990. Kincannon D.F., Groves S. Role of Suspended Solids in Kinetics of RBC Systems, Proc: 1st National Symposium on Rotating Biological Contactor Technology, February 1980. Kinner N.E., Maratea D., Bishop P.L. An Electron Microscopic Evaluation of Bacteria Inhabiting Rotating Biological Contactor Biofilms During Various Loading Conditions, Environmental Technology Letters, Vol. 6, 1985:455-466. Kornegay B.H., Andrews J.F. Characteristics and Kinetics of Fixed Film Biological Reactors, FWPC Grant WP-01181, E P A PB 199 834, 1970. Kornegay B.H. Modeling and Simulation of Fixed Film Biological Reactors for Carbonaceous Waste Treatment, Mathematical Modeling for Water Pollution Control Processes, ed. by Keinath T.M. and Wanielista M. , Ann Arbor Science Publisher Inc., 1975. Bibliography 120 Lagnese J.F. Jr. Use of Supplemental Air to Correct an Oxygen Limitation Condition of an Operating RBC System, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, Pennsylvania, February 4-6, 1980. La Motta E.J., Hickey R.F. Factors Affecting Attachment and Development of Biological Films on Solid Media, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology , Pennsylvania, February 4-6, 1980:803-827. Leask, R.A. (ed.) Pulp and Paper Manufacture, Volume 2, Mechanical Pulp, Joint Textbook Committee of the Paper Industry, 1987. Lee B., Costantino J. Rotating Biological Disk System, Pulp and Paper, Canada, Vol. 78, No. 4, April 1977:T86-T92. Lee E.G.H., Mueller J.C. Rotating Biological Treatment of Kraft Mill Effluents, Journal of Water and Pollution Control, May, 1975:25-29. Levenspiel O. Chemical Reaction Engineering, 2nd ed., Johen Wiley & Sons Inc., 1972. Levenspiel O. The Chemical Reactor Omnibook + Distributed by OSU Book Stores Inc., Corvallis, Oregon, July 1984. Lo S.-N., Liu H.W., Rousseau S., Lavallee H.C. Characterization of Pollutants at Source and Biological Treatment of a CTMP Effluent, APPITA Vol. 44, No. 2, March 1991:133-138. Lue-Hing G , Obayashi A.W., Zenz D.R., Washington B., Sawyer B.M. Nitrification of a High Ammonia Content Sludge Supernatant by Use of Rotating Discs, Proc: 29th Ind. Waste Conf., Purdue University Vol. 245, 1974. Malinen R., Virkola N.E., Turunen E. Effluent Characterization and Treatment of ChemitheromechanicalPulping, Proc: SPCI Int. Mech. Pulping Conf. (Stockholm, Sweden), May 6-10, 1985:267-273. Manual of Practice No.8, Wastewater Treatment Plant Design, WPCF, ASCE Manual on Engineering Practice No.36, 1977 (Second Printing 1982). Manual of Practice OM-10, O&Mof Trickling Filters, RBC's, and Related Processes, WPCF, 1988. MacLean B., de Vegt A., van Driel E . Full-Scale Anaerobic/Aerobic Treatment of TMP/BCTMP Effluent at Quesnel River Pulp, TAPPI Proc: Environmental Conference (Seattle, Wash.), April 9-11, 1990:647-661. Matsumoto Y., Yin C.F., Chang H-m., Joyce T.W., Kirk T.K. Degradation of Chlorinated Lignin and Chlorinated Organics by a White-Rot Fungus, Int. Symp. Wood & Pulping Chem. (Vancouver) Tech. Papers, August 26-30,1985:45-46. McAliley J.E. A Pilot Plant Study of a Rotating Biological Surface for Secondary Treatment of Unbleached Kraft Mill Waste, TAPPI, Vol. 57, No. 9, September 1974:106-111. Bibliography 121 McAllen V . E . Treatment of Chemi-Thermomechanical Pulping Effluents, Preprints: Treatment of Chemithermomechanical Pulping Effluents (Vancouver, B.C.), October 27, 1988:47-78. McCubbin N. The Basic Technology of the Pulp and Paper Industry and its Environmental Protection Practices, Report EPS 6-EP-83-1, Environment Canada, Ottawa, Ont., October 1983. McLeay D. and Associates Ltd. Aquatic Toxicity of Pulp and Paper Mill Effluents, Environment Canada Report, EPS 4/PF/l. April 1987. Metcalf & Eddy Inc. Wastewater Engineering Treatment/Disposal, Reuse, 2nd ed., McGraw-Hill Series in Water Resources and Environmental Engineering, 1979. Miotti R. Distribution of BOD and Toxicity in BCTMP Plants, Preprints, 1990 Spring Conference, Canadian Pulp and Paper Association, Jasper, Alberta, 1990. Namung E. , Rittmann B.E. Effects ofSMPon Bioftlm-Reactor Performance, Journal of Environmental Engineering, Vol. 114, No.l, February 1988:199-210. Nilsson I., Dostalek M . Estimation of the Biofilm Formation Ability of Pseudomonas Putida in Discrete Samples from Continuous Culture, Appl. Microbiology Biotechnol. Vol. 20, 1984:183-188. Novatec Consultants Inc., Hydroqual Consultants Inc., Sandwell Swan Wooster Inc., M.R.2 MacDonald Associates Ltd. Managements of Effluents from CTMP Mills, Part 1, Alberta Environment, 1987. Osterman W . A Pilot Plant Evaluation of BIO-SURF Biological Treatment system at Castelgar, B.C., Canadian Cellulose Ltd., April 23, 1974:14p. Perry R.H., Green D. (ed.) Perry's Chemical Engineers' Handbook, 6th ed., McGraw-Hill Book Company, New York, 1984. Pichon M. , Meyer F., Francois E. Anaerobic Treatment of CTMP Wastewater, Proc: E U C E P A Symposium (Helsinki,Finland), 1986. Reh C.W., Wilson T.E., Srinivasaraghavan R. An Approach to Design of RBC's for Treatment of Municipal Wastewater, ASCE National Environmental Engineering Conference, Nashville, Tennessee, July 1977. Riemer M. , Kristensen G.H., Harremoes P. Residence time distribution on submerged biofilters, Wat. Res., Vol. 14, 1980:949-958. Rintala J., Vuoriranta J*. Anaerobic-Aerobic Treatment of Thermomechanical Pulping Effluents, TAPPI Journal, Vol. 71, No. 9, 1988:201-207. Rogers I.H., Davies J.C., Kruzynski G.M., Mahood H.W., Servizi J .A, Gordon R.W. Fish Toxicants in Kraft Effluents, Tappi V o l 58(7), 1975:136. Sanders W.M. Oxygen Utilization by Slime Organism in Continuous Culture, International Journal of Air and Water Pollution, Vol.10, 1966. Bibliography 122 Sawyer N.C., McCarty P.L. Chemistry for Environmental Engineering, 3rd ed., McGraw-Hill Pub. Company, New York, 1978. Scarlata V. NSSC Wastewater Treatment: Pilot Plant Results, AITTPE Congr. Latinoam, Celulosa y Papel (Toremolinos, Spain) 2, Vol. 2, Sect. 6, June 1981:27-41. Schafer D.E., O'Shaughnessy J.C., Blanc F.C. Analysis of Steady State Substrate Removal Models for the RBC, Proc.: 1st National Converence on Fixed-Film Biological Process, University of Pittsburg, 1982:438-473. Scheible O.K., Novak JJ . Upgrading Primary Tanks with Rotating Biological Contactors, Proc.: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, Vol. II, EPA 600/9-80-046b, NTT.S PB81-124547, June 1980:961-996. Schnell A., Dorica J., Ho G , Ashikawa M., Munnoch G, Hall E.R. Anaeraobic and Aerobic Pilot-scale Effluent Detoxification Studies at an Integrated Newprint Mill, Pulp and Paper Canada, Vol. 91:9, 1990:T347-T352. Servizi J.A., Gordon R.W. Detoxification of TMP and CTMP Effluents Alternating in a Pilot Scale Aerated Lagoon, Pulp and Paper Canada, Vol. 87:11, 1986:T405-T409. Severin B.F., Brociner H., Dumanowski J.E., Su J.T., Gurvitch M.M. Empirical Oxygen Transfer Relation in the RBC Process, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, EPA 600/9-80-04613, Pennsylvania, February 4-6, 1980:1077-1100. Shere S.M., Daly P.G. High Rate Biological Treatment of TMP Effluent, Pulp and Paper Canada, Vol. 83:C, 1982:T91-T95. Sinclair W.F. Controlling Pollution from Canadian Pulp and Paper Manufacturers: A Federal Perspective, Environment Canada, Conservation and Protection, 1990. Springer A .M. Industrial Environmental Control: Pulp and Paper Industry, Wiley Interscience (N.Y.), 1986:251-253. Srinivasaraghavan R., Reh C.W., Lilegren S. Performance Evaluation of At Driven RBC Process for Municipal Waste Treatment, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, 1980. Steels I.H. Design Basis for the Rotating Disc Process, Effluent and Water Treatment Journal, August 1974:431-445. Steiner C.G. The Biological Approach to the Rotating Disc Process, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, Pennsylvania, February 4-6, 1980:791-801. Stover E X . , Kincannon D.F. Rotating Biological Contactor Scale-Up and Design, Proc: 1st International Converence on Fixed-Film Biological Process, University of Pittsburg, 1982:1667-1687. Bibliography 123 Thakore A., Collins G, Smiley D.W. Toxicity and its Behaviour on Pulp Fibres, Preprints: CPPA Spring Conf. (Whistler, B.C.), Session 5, Paper No. 7, May 14-16 1987:4p. Tomlinson T.G., Snaddon D.M.H. Biological Oxidation of Sewage by Films of Microorganism, International Journal of Air and Water Pollution, Vol. 10, 1966. Trinh D.T. Exploration Camp Wastewater Characterization and Treatment Plant Assessment, Report No. EPS 4-WP-81-1, Environmental Protection Service, Environment Canada, Ottawa, Canada, 1881. Turk Sam Options for Treatment of CTMP Effluents, Preprints: Treatment of Chemithermomechanical Pulping Effluents, Vancouver B.C., October 27, 1988. Turpey W.N., Heukelekian H., Kaplovski A J . , Epstein R. Rotating Discs with Biological Growth Prepare Wastewater for Disposal and Reuse, Journal WPCF, Vol.43, 1971:2181. Urbanatas R.G., MacEwen H.D. Characterization of Effluents from Chemimechanical, Chemithermomechanical and Thermomechanical Pulping of Black Spruce, Preprints: 71st Annual CPPA Meeting (Montreal, Quebec), January 29-30 1985:A213-A219. Virkola N.E., Honkanen K. Wastewater Characteristics in Anaerobic Treatment of Forest Industry Wastewaters, ed. by Rantala P. and Luonsi A , 1st IAWPRC Symposium on Forest Industry Wastewaters, Permagan Press, Great Britain, 1985:1-28. Walden C.C., Howard T.E. Toxicity of Pulp and Paper Mill Effluents - a Review, Pulp and Paper Canada, Vol. 82, No. 4, April 1981:T143-T147. Wearing J.T., Huang S., Piggott M.D., Ouchi M.D., Wong A New White Water Management Strategies for Integrated Newsprint Mills, Pulp and Paper Canada, 86(5), T139, 1985. Weast R.C. (ed.) Handbook for Chemistry and Physics, 64th ed., CRC Press, Florida, 1983-1984. Welander T., Malmqvist A , Yu P. Anaerobic Treatment of Toxic Forest Industry Wastewaters, Anaerobic Digestion, Pergamon Press, London, 1988. Welch F.M. Preliminary Results of a New Approach in the Aerobic Biological Treatment of Highly Concentrated Wastes, Proc.:23rd Ind. Waste Conf., Purdue University, 1968:428. Welch E.B. Ecological Effects of Waste Water, Cambridge University Press, Cambridge, UK, 1980. Wellisch M.T., Gnyp A W., Bewtra J.K., StPierre C.C. The BCTMP Process - Identification of its Process Emissions, TAPPI Poceedings, Environmental Conference, 1988:157-164. Wilson R.W., Murphy K L . , Stephenson J.P. Scaleup in Rotating Biological Contactor Design, Journal WPCF, Vol. 52, No.3, March 1980:611-621. Wilson R.W., Murphy K.L., Frenette E.G. Aerobic and Anaerobic Treatment of NSSC and CTMP Effluent, C P A A Environmental Conference Toronto, September 24-26, 1985:113-119. Bibliography 124 Wilson R.W., Schneider E , Frenette E .G. Anaerobic/Aerobic Treatment of CTMP and NSSC Wastewater, TAPPI Proc: Environmental Conference, 1988:315-325. Wong A Quality of Effluents from Mechanical Pulping Processes, Pulp and Paper Canada, Vol. 78, No.6, T132, 1977. Wong A , Breck D., Constantino J. TMP Effluents a Review, Tappi Vol. 61, No.8, 1978:19. Wong A , Constantino J., Breck D., Herschmiller D., Rae J.G. Toxicity and BOD of Effluents Derived from TMP of Eastern Canadian Wood Species, Pulp and Paper Canada, Vol. 80, No.l, T14, 1979. Wong A , Breck D., Constantino J., TMP Effluents are Affected by Refining Conditions, Pulp and Paper Canada, Vol. 81, No.3, T72, 1980. Wu Y.C., Smith E.D., Gratz J. Prediction of RBC Plant Performance for Municipal Treatment, Proc: 1st National Symposium/Workshop on Rotating Biological Contactor Technology, 1980:887. Wu Y.C., Smith E.D., Chen C.Y., Miller R. Evaluation of RBC Scale-Up, Proc: 1st International Conference on Fixed-Film Biological Processes, University of Pittsburg, 1982:474-486. APPENDIX I A.L1 DAILY LOG OF RUN I AL2 RAW DATA FROM RUN I 126 A.L1 DAILY LOG OF RUN I (Jury 20,1990 - August 12,1990) Seeding of the reactor was started on July 19th, 1990. A biofilm started to grow right away, especially at the liquid surface and the media interface close to the shaft. After one day the biofilm was found to have a thickness of approximately 0.1 to 0.2 mm. The growth of the biofilm was monitored and after 3 days the biofilm had reached a thickness of approximately 0.5 to 0.8 mm. Its colour appeared whitish beige. The biomass started to slough off on day 5 (July 24, 1990), and at that time the thickness was estimated to be 0.8 to 0.9 mm for the discs closest to the wastewater inlet. The biomass thickness on the media decreased progressively towards the effluent end of the reactor to approximately 0.3 to 0.5 mm on the last few discs. The colour of the biomass on the media, by then, slowly changed to dark brown. It was hard to estimate the thickness of the biomass because of its shagginess. However, during the run the biomass started to slough off regularly at a thickness of approximately 1 mm forming large floes in the trough. Biofilm regrew on the media quite rapidly. This phenomena occurred about once a week untill the end of the run. On day 10 (July 29th, 1990), a visible biofilm covering the plexiglass reactor blocked the view into the reactor. This biofilm increased however, the active surface area available for organic removal. The flowrate was maintained at 7 1/day resulting in a H R T = 2 days. The temperature was set at 32 to 33°C, and sufficient wastewater from one batch of the QRP effluent was stored in the refrigerator to meet the requirements of run I. After running the R B C for two days it was noticed that the effluent flowrate was much less than the influent flow rate and hardly any effluent was collected in the effluent bucket. Because of this high evaporation rate the influent and effluent volumetric flows were measured beginning on day 3 (July 23, 1990) of run I, to run IV to correct the outlet parameters for the water loss. Measurements were taken from day 1 (July 20,1990) and sampling was started on July 21st, 1990, 127 after one day of operation. A colorimetric method was used for COD measurements. Problems were encountered with this method, because of interferences caused by particles and precipitates. Underestimated measurements for high COD values were another problem known to exist in the colorimetric method. For the other three runs, a standard method, the open reflux method was used for COD analysis. From July 30th to July 31st, 1990 the effluent volume had exceeded the influent volume and it was noticed that the glue joint at the top of the reactor between the waterbath and the reactor had a leak. The waterbath had to be shut off for regluing. On August 2nd, 1990 the reactor was reheated again. Thereafter, leakages between the waterbath and reactor occurred on August 8th, 9th, list and 12th, 1990. The effects of these leakages on the overall performance of the system were governed by variations in temperature (ranging from 21 to 33°C) and organic loading rate (ranging from 0.0079 to 0.0267 kg BOD/m2/day). Finally, on August 12th, 1990 after 23 days of operation the reactor had to be shut down after leaks kept persisting. The whole reactor was rebuilt in stainless steel after it was discovered that the acrylic material was warped. Part of the leakage problem originated, because the heat transfer between heating bath and the bio-reactor was much larger than originally thought. The high evaporation (up to 90%) in the trough resulted in a large temperature difference between the waterbath and the wastewater exceeding the maximum temperature before failure of acrylic. The following observations were made throughout run I: (1) the colour of the effluent changed from an initial opaque milky colour of the raw wastewater to a dark reddish to purple brown colour after RBC treatment; (2) dewatering of the mixed liquor by vacuum filtration was found to be very difficult. Dewatering of the sludge taken from the discs by vacuum filtration was practically impossible. However, when the samples were subjected to 20 minutes of centrifuging, most of the water could be separated from the sludge. A.L2 R A W D A T A F R O M R U N I 128 Run I DATA FOR ROTATING BIOLOGICAL CONTACTOR 2.0 d a y s HRT DATE DAY TIME VOLUME % FLOWRATE HRT i n o u t n e t volume i n o u t v a p o u r h o u r l i t e r r e d . l i t e r / d a y d a y s 20/07/90 1 21/07/90 2 21.75 1.70 22/07/90 3 19.75 23/07/90 4 28.50 11.50 3.20 8.30 72.17 9.68 2.69 6.99 1.51 24/07/90 5 21.25 8.80 2.20 6.60 75.00 9.94 2.48 7.45 1.47 25/07/90 6 29.58 8.10 3.60 4.50 55.56 6.57 2.92 3.65 2.22 26/07/90 7 21.42 6.90 2.52 4.38 63.48 7.73 2.82 4.91 1.89 27/07/90 8 27.42 8.20 1.16 7.04 85.85 7.18 1.02 6.16 2.04 28/07/90 9 18.33 5.30 3.12 2.18 41.13 6.94 4.09 2.85 2.11 29/07/90 10 31.75 8.05 2.64 5.41 67.20 6.09 2.00 4.09 2.40 30/07/90 11 25.25 7.90 5.57 2.33 29.49 7.51 5.29 2.21 1.95 31/07/90 12 23.25 5.60 8.15 -2.55 -45.54 5.78 8.41 2.53 01/08/90 13 20.42 4.50 3.24 1.26 28.00 5.29 3.81 1.48 2.76 02/08/90 14 24.00 5.80 4.60 1.20 20.69 5.80 4.60 1.20 2.52 03/08/90 15 24.50 6.70 1.40 5.30 79.10 6.56 1.37 5.19 2.23 04/08/90 16 24.50 5.20 0.84 4.36 83.85 5.09 0.82 4.27 2.87 06/08/90 17 17.00 11.00 1.14 9.86 89.64 15.53 1.61 13.92 0.94 07/08/90 18 29.33 8.75 2.12 6.63 75.77 7.16 1.73 5.43 2.04 08/08/90 19 24.25 5.80 4.20 1.60 27.59 5.74 4.16 1.58 2.55 09/08/90 20 23.58 6.30 4.10 2.20 34.92 6.41 4.17 2.24 2.28 10/08/90 21 24.17 8.35 1.86 6.49 77.72 8.29 1.85 6.44 1.76 11/08/90 22 22.17 7.00 2.80 4.20 60.00 7.58 3.03 4.55 1.93 12/08/90 23 19.25 5.20 3.55 1.65 31.73 6.48 4.43 2.06 2.25 A v e r a g e 52.67 7.37 3.17 4.56 2.11 S t a n d , d e v . 31.54 2.25 2.25 2.90 0.45 129 Run I 2.0 d a y s HRT DATE DAY TEMPERATURE D.O. i n o u t i n m i d d e g r e e C e l s i u s mg/1 20/07/90 1 31.5 30.5 0. 20 0.70 21/07/90 2 32.0 32.0 0. 10 0.30 22/07/90 3 34.0 34.0 0. 30 0.50 23/07/90 4 33.0 33.0 0. 60 0.85 24/07/90 5 34.0 34.0 0. 40 0.80 25/07/90 6 35.0 33.0 0. 50 0.60 26/07/90 7 35.0 33.0 0. 50 0.60 27/07/90 8 35.0 33.5 0. 40 0.40 28/07/90 9 34.0 33.0 0. 90 0.95 29/07/90 10 34.5 33.5 0. 35 0.85 30/07/90 11 34.0 33.0 0. 90 1.30 31/07/90 12 35.5 33.0 0. 90 2.45 01/08/90 13 22.0 21.0 1. 75 2.65 02/08/90 14 23.0 22.5 3. 20 2.70 03/08/90 15 32.0 32.0 0. 60 1.20 04/08/90 16 33.0 32.5 0. 65 0.75 06/08/90 17 33.0 33.0 1. 35 1.25 07/08/90 18 34.5 31.0 0. 70 1.20 08/08/90 19 33.0 28.0 0. 50 3.20 09/08/90 20 34.0 32.0 0. 10 0.40 10/08/90 21 34.0 33.0 0. 25 0.45 11/08/90 22 34.0 31.0 0. 25 0.85 12/08/90 23 24.0 23.0 0. 90 1.00 A v e r a g e S t a n d , d e v . pH BOD o u t i n m i d o u t i n o u t mg/1 0 .50 8 .26 8 .32 8 .35 0 .20 8 .38 8 .39 8 .38 3100 2700 0 .60 8 .50 8 .52 8 .53 3100 1950 0 .70 8 .67 8 .69 8 .70 3100 1000 0 .90 8 .65 8 .68 8 .69 3100 1100 0 .50 8 .68 8 .70 8 .71 3400 1760 0 .50 8 .65 8 .69 8 .70 3400 2070 0 .40 8 .75 8 .76 8 .77 3400 1200 1 .15 8 .82 8 .85 8 .87 2800 1800 1 .05 8 .81 8 .82 8 .83 2800 1910 1 .60 8 .72 8 .75 8 .76 2800 1060 3 .15 8 .54 8 .60 8 .63 2800 700 4 .40 8 .52 8 .55 8 .56 2600 640 2 .10 8 .55 8 .55 8 .54 2600 400 0 .60 8 .73 8 .72 8 .71 2600 590 0 .30 8 .71 8 .73 8 .72 3010 750 0 .80 8 .80 8 .80 8 .81 3010 480 2 .00 8 .73 8 .76 8 .82 3010 710 3 .20 8 .44 8 .68 8 .77 3010 300 0 .20 8 .62 8 .66 8 .65 0 .15 8 .83 8 .85 8 .86 3430 510 0 .20 8 .86 8 .89 8 .88 3430 530 1 .00 8 .85 8 .88 8 .88 2700 440 8 .66 8 .69 8 .70 3010 0 .16 0 .14 0 .14 278 130 Run I 2.0 d a y s HRT DATE DAY BODred. BODremov BOD O r g . L o a d . r a t e accum. r a t e % mg/hr i n t a n k k g BOD/ % day*m2 20/07/90 1 21/07/90 2 22/07/90 3 23/07/90 4 91. 02 1139 32. 26 0. 0172 24/07/90 5 91. 13 1170 35. 48 0. 0176 25/07/90 6 76. 99 717 51. 76 0. 0128 26/07/90 7 77. 76 852 60. 88 0. 0150 27/07/90 8 95. 01 966 35. 29 0. 0139 28/07/90 9 62. 16 503 64. 29 0. 0111 29/07/90 10 77. 63 551 68. 21 0. 0097 30/07/90 11 73. 31 642 37. 86 0. 0120 31/07/90 12 63. 62 429 25. 00 0. 0092 01/08/90 13 82. 28 471 24. 62 0. 0079 02/08/90 14 87. 80 552 15. 38 0. 0086 03/08/90 15 95. 26 677 22. 69 0. 0098 04/08/90 16 95. 97 613 24. 92 0. 0088 06/08/90 17 98. 35 1915 15. 95 0. 0267 07/08/90 18 94. 28 847 23. 59 0. 0123 08/08/90 19 92. 78 668 9. 97 0. 0099 09/08/90 20 10/08/90 21 96. 69 1146 14. 87 0. 0163 11/08/90 22 93. 82 1016 15. 45 0. 0149 12/08/90 23 88. 87 648 16. 30 0. 0100 A v e r a g e 86. 04 817 31. 30 0. 0128 S t a n d , d e v . 10. 85 346 17. 43 0. 0044 COD CODred.CODremov i n o u t r a t e mg/1 % mg/1 4720 3240 4720 3240 80 .90 1541 4720 3660 80 .61 1576 4660 3660 65 .09 831 4660 3720 70 .85 1063 4660 2640 91 .99 1282 4920 3680 55 .97 796 4920 3230 78 .47 979 4920 2980 57 .29 882 4920 2430 28 .12 333 4520 2380 62 .09 618 4520 2640 53 .68 586 4520 2640 87 .80 1085 4460 2800 89 .86 851 4460 1590 96 .31 2779 4460 1820 90 .11 1199 4460 2240 63 .63 679 4560 2280 88 .86 1400 4560 2320 79 .65 1147 3840 2520 55 .20 573 4609 72 .45 1063 239 17 .25 522 131 Run I 2.0 d a y s HRT DATE DAY COD O r g . L o a d . accum. r a t e i n t a n k k g COD/ % day*m2 20/07/90 1 21/07/90 2 22/07/90 3 23/07/90 4 68. 64 0. 0261 24/07/90 5 77. 54 0. 0268 25/07/90 6 78. 54 0. 0175 26/07/90 7 79. 83 0. 0206 27/07/90 8 56. 65 0. 0191 28/07/90 9 74. 80 0. 0195 29/07/90 10 65. 65 0. 0171 30/07/90 11 60. 57 0. 0211 31/07/90 12 49. 39 0. 0163 01/08/90 13 52. 65 0. 0137 02/08/90 14 58. 41 0. 0150 03/08/90 15 58. 41 0. 0170 04/08/90 16 62. 78 0. 0130 06/08/90 17 35. 65 0. 0396 07/08/90 18 40. 81 0. 0182 08/08/90 19 50. 22 0. 0146 09/08/90 20 10/08/90 21 50. 00 0. 0216 11/08/90 22 50. 88 0. 0197 12/08/90 23 65. 63 0. 0142 A v e r a g e 59. 84 0. 0195 S t a n d , d e v . 12. 21 0. 0060 i n TKN T.PHOS. o u t o u t i n o u t o u t mg/1 c o r r e c t . mg/1 c o r r e c t . 16.3 85.7 34.3 2.69 8.11 3.24 132 Run I 2.0 d a y s HRT DATE DAY 20/07/90 1 21/07/90 2 22/07/90 3 23/07/90 4 24/07/90 5 25/07/90 6 26/07/90 7 27/07/90 8 28/07/90 9 29/07/90 10 30/07/90 11 31/07/90 12 01/08/90 13 02/08/90 14 03/08/90 15 04/08/90 16 06/08/90 17 07/08/90 18 08/08/90 19 09/08/90 20 10/08/90 21 11/08/90 22 12/08/90 23 TS i n o u t mg/1 VS i n o u t mg/1 i n TSS o u t mg/1 i n VSS o u t mg/1 6200 6700 4200 1900 3600 4950 250 1400 A v e r a g e S t a n d , d e v . Run I 2-0 d a y s HRT DATE DAY MLVSS mg/1 20/07/90 1 21/07/90 2 22/07/90 3 23/07/90 4 24/07/90 5 25/07/90 6 26/07/90 7 27/07/90 8 28/07/90 9 29/07/90 10 30/07/90 11 31/07/90 12 01/08/90 13 02/08/90 14 03/08/90 15 04/08/90 16 06/08/90 17 07/08/90 18 08/08/90 19 09/08/90 20 10/08/90 21 11/08/90 22 12/08/90 23 A v e r a g e S t a n d . d e v . A P P E N D I X I I AIL1 D A I L Y L O G O F R U N I I AIL2 R A W D A T A F R O M R U N U 135 A.TL1 DAILY LOG OF RUN U (November 15,1990 - December 12,1990) The stainless steel reactor was operational from mid-November 1990. This new reactor had a smaller liquid volume of 12.07 litres. After achieving a good BOD and COD removal in run I at H R T = 2 days, the flowrate for run II was set at 12 1/day resulting in an HRT = 1 day, with all other parameters remaining the same as in run I. Seeding of the reactor was started on November 15th, 1990. Within one day a biomass of 0.5 mm thickness formed on the discs. After running the system for 3 days on November 18th, 1990 the biofilm thickness was approximately 1 mm. Sloughing and regrowing of the biomass occurred the same way as in run I. Every week, the biomass thickness would increase to 3 to 4 mm before sloughing off the media, especially on the first ten discs at the influent end. It was observed that the biomass was growing much faster because of the increased organic loading rate. This fast growth rate of sludge and the very small effluent flowrate resulted in a negligible washout of biomass. The biomass concentration in the trough was increased rapidly, and consequently the dissolved oxygen concentration was close to 2ero throughout run II indicating possible anaerobic conditions in the trough. Measurements were taken starting November 16th, 1990 (day 2) and sampling was started on November 17th, 1990 (day 3). Our measurements, in tune with literature values, suggested the presence of volatile compounds in the wastewater. The possibility of these volatile compounds being removed by evaporation lead to further analysis of the RBC effluent. To determine whether or not this was of importance, influent and effluent samples were analyzed for acetic acid concentration and because of the suspected anaerobic conditions in the bio-reactor tests for hydrogen sulfide (H2S) were also performed. On November 20th, 1990 the constant temperature bath had to be replaced by a larger unit with 136 higher heating capacity to keep an average temperature of 33°C. On November 27th, 1990 power failure at the University of British Columbia for 4 hours provided the opportunity to clean out 700 ml of sludge accumulated in the bio-reactor. The following observations were made throughout run II: (1) The colour of the RBC effluent was not so dark as observed in run I and its colour was reddish brown; (2) the colour of the biomass attached to the media was brown and also seemed lighter than in run I. The overall performance of the RBC appeared less efficient during this run as compared to run I, despite the leakage problems in run I. AJL2 R A W D A T A F R O M R U N I I 137 Run I I 1 day HRT DATE DAY 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 06/12/90 22 07/12/90 23 A v e r a g e S t a n d , d e v . DATA FOR ROTATING BIOLOGICAL CONTACTOR TIME VOLUME % FLOWRATE HRT i n o u t n e t volume i n o u t v a p o u r h o u r l i t e r r e d . l i t e r / d a y d a y s 25 .08 14.50 5 .50 9 .00 62 .07 13.88 5 .26 8 .61 0. 87 24 .88 13.20 3 .90 9 .30 70 .45 12.73 3 .76 8 .97 0. 95 22 .27 11.70 3 .40 8 .30 70 .94 12.61 3 .66 8 .94 0. 96 21 .98 11.40 3 .50 7 .90 69 .30 12.45 3 .82 8 .63 0. 97 23 .47 11.75 3 .40 8 .35 71 .06 12.02 3 .48 8 .54 1. 01 24 .60 12.60 3 .35 9 .25 73 .41 12.29 3 .27 9 .02 0. 98 23 .37 11.50 3 .10 8 .40 73 .04 11.81 3 .18 8 .63 1. 02 24 .27 12.00 2 .10 9 .90 82 .50 11.87 2 .08 9 .79 1. 02 22 .27 11.00 3 .75 7 .25 65 .91 11.85 4 .04 7 .81 1. 02 25 .20 12.30 3 .40 8 .90 72 .36 11.71 3 .24 8 .48 1. 03 24 .95 11.75 2 .90 8 .85 75 .32 11.30 2 .79 8 .51 1. 07 21 .50 8.90 2 .00 6 .90 77 .53 9.93 2 .23 7 .70 1. 22 20 .62 10.40 2 .40 8 .00 76 .92 12.10 2 .79 9 .31 1. 00 27 .38 13.30 3 .70 9 .60 72 .18 11.66 3 .24 8 .41 1. 04 28 .00 14.60 1 .70 12 .90 88 .36 12.51 1 .46 11 .06 0. 97 18 .07 8.60 0 .75 7 .85 91 .28 11.42 1 .00 10 .43 1. 06 25 .38 13.20 1 .10 12 .10 91 .67 12.48 1 .04 11 .44 0. 97 23 .22 10.75 0 .90 9 .85 91 .63 11.11 0 .93 10 .18 1. 09 25 .72 12.40 1 .60 10 .80 87 .10 11.57 1 .49 10 .08 1. 04 18 .32 8.30 1 .40 6 .90 83 .13 10.87 1 .83 9 .04 1. 11 23 .30 11.70 1 .70 10 .00 85 .47 12.05 1 .75 10 .30 1. 00 23 .12 11.60 1 .90 9 .70 83 .62 12.04 1 .97 10 .07 1. 00 77 .97 11.92 2 .65 9 .27 1. 02 8 .59 0.76 1 .13 0 .98 0. 07 138 Run I I 1 d a y HRT DATE DAY 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 06/12/90 22 07/12/90 23 A v e r a g e S t a n d , d e v . TEMPERATURE i n o u t d e g r e e C e l s . 18.0 17.0 32.0 32.0 31.0 31.0 30.0 30.5 31.0 31.5 32.0 31.5 32.0 31.0 33.0 31.5 33.0 31.0 32.5 31.5 33.0 31.0 33.5 32.0 31.0 32.0 31.0 31.0 35.0 35.0 35.0 36.0 33.5 34.5 32.0 33.5 32.0 33.5 32.0 33.5 34.0 34.0 33.0 33.0 32.45 32.40 1.28 1.51 D. 0. i n m i d mg/1 0 .25 0. 10 0 .20 0. 25 0 .10 0. 15 0 .20 0. 15 0 .20 0. 40 0 .10 0. 05 0 .40 0. 30 0 .40 0. 30 0 .30 0. 25 0 .10 0. 15 0 .40 0. 35 0 .30 0. 30 0 .35 0. 25 0 .10 0. 20 0 .30 0. 25 0 .15 0. 25 0 .55 0. 60 0 .05 0. 10 0 .40 0. 40 0 .30 0. 50 0 .20 0. 30 0 .26 0. 27 0 .13 0. 14 o u t i n 0.20 8.02 0.10 7.75 0.15 8.04 0.30 8.35 0.50 8.50 0.10 8.29 0.25 7.58 0.20 8.48 0.20 7.55 0.25 7.68 0.30 7.54 0.25 8.88 0.20 8.81 0.15 8.77 0.20 8.94 0.30 9.06 0.50 8.93 0.05 8.99 0.30 9.08 0.50 8.92 0.30 8.96 0.25 8.43 0.13 0.55 P H m i d o u t 8.08 8.09 8.44 8.45 8.61 8.64 8.74 8.75 8.76 8.79 8.80 8.80 8.83 8.83 8.83 8.83 8.81 8.80 8.83 8.83 8.83 8.83 8.87 8.89 8.81 8.80 8.81 8.82 8.92 8.87 9.07 9.07 8.96 8.95 8.98 8.98 9.11 9.10 8.93 8.93 8.98 8.95 8.81 8.81 0.22 0.21 i n f l . e f f l . 8.39 8.67 8.68 8.80 8.84 8.81 8.82 8.78 8.76 8.82 8.72 8.87 9.08 9.13 6.49 8.99 6.34 9.01 6.42 0.08 139 Run I I 1 d a y HRT DATE DAY i n BOD BODred.BODremov BOD O r g . L o a d . F i l t e r e d BOD o u t r a t e accum. r a t e i n o u t r e d . mg/1 % mg/hr i n t a n k k g BOD/ mg/1 % % day*m2 15/11/90 1 16/11/90 2 3400 1800 79.92 1571 52.94 0.0270 17/11/90 3 3100 2000 80.94 1331 64.52 0.0226 3000 1600 84.24 18/11/90 4 3500 3100 74.26 1366 88.57 0.0252 19/11/90 5 3500 3000 73.68 1338 85.71 0.0249 20/11/90 6 4500 4600 70.42 1586 102.22 0.0309 21/11/90 7 4500 4500 73.41 1692 100.00 0.0316 4250 4200 73.73 22/11/90 8 3900 3700 74.43 1428 94.87 0.0263 23/11/90 9 24/11/90 10 5400 4350 72.54 1935 80.56 0.0366 25/11/90 11 4550 4250 74.18 1647 93.41 0.0305 4450 4050 74.84 26/11/90 12 27/11/90 13 28/11/90 14 4300 3550 80.95 1756 82.56 0.0297 29/11/90 15 4150 3600 75.87 1529 86.75 0.0276 4050 3100 78.71 30/11/90 16 01/12/90 17 2600 2750 90.78 1123 105.77 0.0170 02/12/90 18 03/12/90 19 3000 3150 91.21 1267 105.00 0.0190 04/12/90 20 05/12/90 21 2800 4800 71.08 902 171.43 0.0174 2350 3800 72.72 06/12/90 22 07/12/90 23 2700 4200 74.52 1010 155.56 0.0186 2300 3850 72.58 A v e r a g e 3685 77.21 1432 97.99 0.0257 3400 76.14 S t a n d , d e v . 859 6.23 274 29.36 0.0056 972 4.16 140 Run I I 1 day HRT DATE DAY COD CODred.CODremov COD O r g . L o a d . F i l t e r e d COD i n o u t r a t e accum. r a t e i n o u t r e d . mg/1 mg/hr i n t a n k k g COD/ mg/1 % % . day*m2 15/11/90 1 16/11/90 2 7300 5150 7 3 . 24 3091 70.55 0.0579 17/11/90 3 7300 8500 65. 60 2541 116.44 0.0531 5800 6750 65.62 18/11/90 4 8000 9400 6 5 . 85 2768 117.50 0.0576 19/11/90 5 8000 11150 57. 21 2374 139.38 0.0569 20/11/90 6 7850 11650 5 7 . 06 2242 148.41 0.0539 21/11/90 7 7850 12400 58. 00 2332 157.96 0.0551 7200 11600 57.16 22/11/90 8 8000 11600 60. 91 2398 145.00 0.0540 23/11/90 9 24/11/90 10 8450 11500 5 3 . 60 2237 136.09 0.0572 25/11/90 11 8350 13250 56. 14 2288 158.68 0.0559 7000 10600 58.14 26/11/90 12 27/11/90 13 28/11/90 14 7850 12400 6 3 . 55 2516 157.96 0.0543 29/11/90 15 7750 11150 5 9 . 98 2258 143.87 0.0516 6500 9650 58.70 30/11/90 16 01/12/90 17 7550 15600 8 1 . 98 2946 206.62 0.0493 02/12/90 18 03/12/90 19 8350 22800 7 7 . 14 2982 273.05 0.0530 5800 15250 77.99 04/12/90 20 05/12/90 21 8500 22700 54 . 95 2116 267.06 0.0528 6650 17800 54.85 06/12/90 22 07/12/90 23 8400 19950 6 1 . 10 2575 237.50 0.0578 6850 18750 55.17 A v e r a g e 7973 6 3 . 09 2511 165.07 0.0547 6543 61.09 S t a n d , dev. 422 7. 93 294 55.00 0.0025 556 7.65 141 Run II 1 d a y HRT DATE DAY 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 06/12/90 22 07/12/90 23 A c e t i c i n o u t mg/1 463.40 123.63 413.13 87.70 442.71 115.35 382.79 100.25 408.79 33.03 A c i d o u t r e d . c o r r e c t . % 37.96 91.81 29.90 92.76 26.62 93.99 8.74 97.72 5.57 98.64 BOD A c e t . A c i d i n o u t mg/1 417.06 111.27 371.82 78.93 398.44 103.82 344.51 90.23 367.91 29.73 COD A c e t . A c i d i n o u t mg/1 mg/1 478.85 127.75 426.90 90.62 457.47 119.20 395.55 103.59 422.42 34.13 A v e r a g e S t a n d , d e v . 422.16 31.36 379.95 28.22 436.24 32.41 142 Run I I 1 day HRT DATE DAY 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 06/12/90 22 07/12/90 23 A v e r a g e S t a n d , d e v . F i l t . TKN F i l t . TKN T.PHOS. T.PHOS. i n o u t o u t o u t o u t i n o u t o u t o u t mg/1 c o r r e c t , mg/1 c o r r e c t . mg/1 c o r r e c . mg/1 16.4 391 66.0 221 37.3 3 25 4.2 12 143 Run I I 1 d a y HRT DATE DAY 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 06/12/90 22 07/12/90 23 A v e r a g e S t a n d , d e v . F i l t . T.PHOS. o u t c o r r e c . TS i n o u t mg/1 i n VS o u t mg/1 i n TSS o u t mg/1 VSS i n o u t mg/1 2000 15400 900 10350 5000 13900 800 9050 2.0 2000 12500 900 7500 4850 11850 900 4350 2000 13950 900 8925 4925 12875 850 6700 0 1450 0 1425 75 1025 50 2350 144 Run I I 1 d a y HRT DATE DAY MLVSS MLTSS mg/1 mg/1 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 6000 29/11/90 15 8800 30/11/90 16 9750 01/12/90 17 9100 10350 02/12/90 18 10600 03/12/90 19 8450 04/12/90 20 10700 05/12/90 21 10500 11950 06/12/90 22 07/12/90 23 A v e r a g e 9800 9575 S t a n d , d e v . 700 1704 i n S u l f a t e S04 c o r r e c t , o u t o u t mg/1 i n S u l f i t e S u l f i t e S03 c o r r e c t , o u t o u t mg/1 430 2750 463.86 440 3450 565.09 69.5 61.5 65.5 4.0 59.5 48.0 10.0 7.9 Run I I 1 day HRT S u l f i d e DATE DAY S i n o u t mg/1 15/11/90 1 16/11/90 2 17/11/90 3 18/11/90 4 19/11/90 5 20/11/90 6 21/11/90 7 22/11/90 8 23/11/90 9 24/11/90 10 25/11/90 11 26/11/90 12 27/11/90 13 28/11/90 14 29/11/90 15 30/11/90 16 01/12/90 17 02/12/90 18 03/12/90 19 04/12/90 20 05/12/90 21 3. 87 6. 06/12/90 22 07/12/90 23 3. 22 5.' A v e r a g e 3. 55 S t a n d . d e v . 0. 33 APPENDIX HI AIIL1 DAILY LOG OF RUN III A U L 2 RAW DATA FROM RUN TH 147 AJIL1 DAILY LOG OF R U N III (December 7, 1990 - December 28, 1990) After completing run II, 2 litres of excess biomass were removed from the reactor and the tubing for the inlet and the outlet was replaced. Then the flowrate was started for run HI. Because of the excess biomass growth caused by the high organic loading rate of run II, the flowrate for this run was set at 8 1/day resulting in an HRT = 1.5 days. All other parameters were kept the same as in run I and H. Measurements were taken starting December 8th, 1990 and sampling was started on the same day. It was hoped that because of the increase in H R T the D.O. concentration would also increase, but this was not the case. It was then decided to measure the oxidation reduction potential to verify if conditions were indeed anaerobic as was suspected. All other measurements performed were the same as in run H. The evaporation rate was higher during this run as expected, considering the decrease in the flowrate. Relative humidity measurements were taken every five days during this run to correlate the evaporation to various parameters. The following observations were made throughout run HI: (1) The colour of the effluent and of the attached biomass did not change from run II; (2) the growth and sloughing of biomass did not considerably change from run II. On December 28th, 1990 all data needed for run HI had been collected and the reactor was shut down and cleaned out. Shortly after a 'sterile run' was started. This experiment is described and discussed in section 4.3.1 to section 4.3.3. A . I I L 2 RAW D A T A F R O M R U N III 148 Run I I I DATA FOR ROTATING BIOLOGICAL CONTACTOR 1.5 d a y s HRT DATE DAY TIME VOLUME % FLOWRATE HRT i n o u t n e t v o l u m e i n o u t v a p o u r h o u r l i t e r r e d . l i t e r / d a y d a y s 07/12/90 1 08/12/90 2 24. 63 12. 60 2.00 10. 60 84.13 12 .28 1.95 10.33 0.98 09/12/90 3 18. 03 8. 10 1.00 7. 10 87.65 10 .78 1.33 9.45 1.12 10/12/90 4 21. 28 9. 30 1.00 8. 30 89.25 10 .49 1.13 9.36 1.15 11/12/90 5 24. 33 10. 40 1.10 9. 30 89.42 10 .26 1.09 9.17 1.18 12/12/90 6 26. 00 11. 10 1.00 10. 10 90.99 10 .25 0.92 9.32 1.18 13/12/90 7 29. 33 12. 20 1.50 10. 70 87.70 9 .98 1.23 8.76 1.21 14/12/90 8 17. 05 7. 10 0.90 6. 20 87.32 9 .99 1.27 8.73 1.21 15/12/90 9 31. 00 12. 40 1.50 10. 90 87.90 9 .60 1.16 8.44 1.26 16/12/90 10 17/12/90 11 23. 33 9. 30 1.10 8. 20 88.17 9 .57 1.13 8.44 1.26 18/12/90 12 26. 17 11. 40 1.45 9. 95 87.28 10 .45 1.33 9.12 1.16 19/12/90 13 21. 50 8. 20 1.20 7. 00 85.37 9 .15 1.34 7.81 1.32 20/12/90 14 21. 75 8. 20 1.20 7. 00 85.37 9 .05 1.32 7.72 1.34 21/12/90 15 23. 58 8. 10 1.00 7. 10 87.65 8 .24 1.02 7.23 1.47 22/12/90 16 22. 58 8. 80 1.10 7. 70 87.50 9 .35 1.17 8.18 1.29 23/12/90 17 27. 72 10. 30 1.40 8. 90 86.41 8 .92 1.21 7.71 1.35 24/12/90 18 23. 93 8. 70 1.20 7. 50 86.21 8 .73 1.20 7.52 1.38 25/12/90 19 26. 97 9. 70 1.30 8. 40 86.60 8 .63 1.16 7.47 1.40 26/12/90 20 24. 17 9. 00 1.20 7. 80 86.67 8 .94 1.19 7.75 1.35 27/12/90 21 20. 40 7. 60 0.90 6. 70 88.16 8 .94 1.06 7.88 1.35 28/12/90 22 23. 22 8. 70 1.10 7. 60 87.36 8 .99 1.29 8.94 1.34 A v e r a g e 87.36 9 .63 1.23 8.47 1.33 S t a n d . d e v . 1.49 0 .92 0.20 0.83 0.08 149 Run I I I 1.5 d a y s HRT DATE DAY TEMPERATURE D.O. i n o u t i n m i d d e g r e e C e l s i u s mg/1 07/12/90 1 08/12/90 2 32.0 32.0 0. 35 0.30 09/12/90 3 31.0 31.0 0. 40 0.30 10/12/90 4 31.0 32.0 0. 20 0.25 11/12/90 5 31.0 31.0 0. 30 0.30 12/12/90 6 31.0 31.0 0. 20 0.30 13/12/90 7 31.0 31.0 14/12/90 8 30.0 31.0 0. 20 0.10 15/12/90 9 30.5 31.0 0. 30 0.25 16/12/90 10 17/12/90 11 31.0 30.0 0. 30 0.25 18/12/90 12 30.0 32.0 0. 20 0.15 19/12/90 13 29.0 29.0 0. 50 0.40 20/12/90 14 29.0 29.0 0. 60 0.40 21/12/90 15 31.0 31.0 0. 40 0.50 22/12/90 16 30.0 31.5 0. 20 0.30 23/12/90 17 31.5 31.5 0. 20 0.20 24/12/90 18 31.0 32.0 0. 10 0.20 25/12/90 19 28.0 29.0 26/12/90 20 30.5 31.5 0. 30 0.40 27/12/90 21 32.5 31.0 0. 20 0.30 28/12/90 22 31.0 30.0 0. 20 0.20 A v e r a g e 30.60 : 30.88 0. 29 0.28 S t a n d , d e v . 1.05 0.99 0. 13 0.10 O.R.P. o u t i n m i d o u t i n f l . e f f l . mV 0.20 0.30 0.30 0.30 0.30 0. 20 -270 -265 -260 -210 -179 0. 20 -380 -340 -325 -275 -327 0. 20 -350 -310 -340 -200 -336 0. 25 -355 -390 -330 -201 -339 0. 50 -345 -370 -320 -214 -309 0. 60 -350 -400 -310 0. 50 -330 -390 -375 -223 -330 0. 20 -355 -400 -330 -242 -180 0. 20 -285 -270 -275 -246 -214 0. 20 -250 -240 -225 0. 30 -260 -255 -250 -265 -290 0. 30 -380 -380 -340 -226 -167 0. 20 -380 -390 -300 -231 -272 0. 29 -330 -338 -306 -230 -268 0. 12 47 59 41 24 66 150 Run I I I 1.5 d a y s HRT DATE DAY pH i n m i d o u t i n f l . 07/12/90 1 08/12/90 2 8 .98 8. 96 8 .94 09/12/90 3 9 .04 9. 03 9 .03 6. 37 10/12/90 4 9 .07 9. 03 9 .01 6. 18 11/12/90 5 9 .02 9. 03 9 .04 6. 38 12/12/90 6 9 .05 9. 00 8 .98 6. 53 13/12/90 7 14/12/90 8 8 .98 8. 97 8 .96 6. 40 15/12/90 9 9 .04 9. 06 9 .04 6. 45 16/12/90 10 17/12/90 11 8 .96 8. 97 8 .95 6. 52 18/12/90 12 8 .97 8. 72 8 .98 6. 53 19/12/90 13 8 .87 8. 55 8 .87 6. 55 20/12/90 14 8 .95 8. 65 8 .99 21/12/90 15 8 .95 8. 69 8 .92 6. 36 22/12/90 16 8 .95 8. 35 8.9 6. 26 23/12/90 17 8 .98 8. 97 8 .95 6. 30 24/12/90 18 8 .93 8. 94 8 .95 25/12/90 19 26/12/90 20 8 .97 8. 95 8 .94 6. 15 27/12/90 21 8 .99 9. 00 9 .01 6. 35 28/12/90 22 8 .51 8. 83 8 .93 6. 22 A v e r a g e 8 .96 8. 87 8 .97 6. 37 S t a n d , d e v . 0 .12 0. 19 0 .05 0. 13 BOD BODred. BODremov. e f f l . i n o u t r a t e mg/1 % mg/hr 9 .03 3400 5200 81. 12 1239 9 .03 3450 6300 80. 36 1212 8 .99 8 .95 2800 4800 84. 56 1011 8 .91 3900 4750 84. 56 1373 8 .97 8 .97 2750 3700 84. 09 922 8 .95 8 .95 2900 4500 77. 29 855 8 .97 3000 3900 83. 95 865 8 .91 9 .00 2800 2900 85. 92 894 9 .01 2500 2900 84. 53 787 9 .01 8 .97 2750 3350 82. 47 850 3025 82. 89 1001 404 2. 47 191 151 Run I I I 1.5 d a y s HRT DATE DAY BOD O r g . L o a d . F i l t e r e d BOD COD CODred. accum. r a t e i n o u t r e d . i n o u t i n t a n k k g BOD/ mg/1 % mg/1 % % day*m2 07/12/90 1 08/12/90 2 8450 21800 59.05 09/12/90 3 152.94 0.0209 8050 20350 68.79 10/12/90 4 182.61 0.0207 2900 4650 82.76 8500 22350 71.73 11/12/90 5 8600 21900 73.07 12/12/90 6 171.43 0.0164 2400 4050 84.80 8600 21900 77.06 13/12/90 7 8350 22450 66.94 14/12/90 8 121.79 0.0223 8400 20150 69.59 15/12/90 9 7500 20500 66.94 16/12/90 10 17/12/90 11 134.55 0.0150 8000 21400 68.36 18/12/90 12 7700 19550 67.71 19/12/90 13 155.17 0.0152 2500 3900 77.17 7700 17100 67.50 20/12/90 14 21/12/90 15 130.00 0.0141 7350 15700 73.63 22/12/90 16 6750 16100 70.19 23/12/90 17 103.57 0.0143 2450 2850 84.19 7700 16200 71.40 24/12/90 18 25/12/90 19 26/12/90 20 116.00 0.0128 7700 12650 78.10 27/12/90 21 28/12/90 22 121.82 0.0141 2350 3100 81.02 8000 14500 73.92 A v e r a g e 138.99 0.0166 2520 81.99 7959 70.99 S t a n d , d e v . 24.26 0.0032 196 2.74 518 3.44 152 Run I I I 1.5 d a y s HRT DATE DAY CODremov. COD O r g . L o a d . r a t e accum. r a t e mg/hr i n t a n k k g COD/ % day*m2 F i l t e r e d COD A c e t i c A c i d i n o u t r e d . i n o u t mg/1 % mg/1 07/12/90 1 08/12/90 2 2553 257 .99 0. 0593 09/12/90 3 2488 252 .80 0. 0496 10/12/90 4 2664 262 .94 0. 0509 7050 20100 69. 34 11/12/90 5 2686 254 .65 0. 0504 12/12/90 6 2829 254 .65 0. 0504 6800 18500 75. 49 491. 50 13/12/90 7 2325 268 .86 0. 0476 14/12/90 8 2434 239 .88 0. 0480 7450 18200 69. 03 15/12/90 9 2008 273 .33 0. 0411 16/12/90 10 17/12/90 11 2180 267 .50 0. 0437 6000 14400 71. 61 469. 00 18/12/90 12 2271 253 .90 0. 0460 6650 13900 73. 41 19/12/90 13 1982 222 .08 0. 0403 20/12/90 14 21/12/90 15 1859 213 .61 0. 0346 6800 15900 71. 13 4 2 1 . 05 22/12/90 16 1846 238 .52 0. 0361 23/12/90 17 2043 210 .39 0. 0392 6800 15550 68. 92 431. 38 24/12/90 18 25/12/90 19 26/12/90 20 2239 164 .29 0. 0393 6800 12250 75. 98 27/12/90 21 433. 28 28/12/90 22 2216 181 .25 0. 0411 6600 12100 73. 62 A v e r a g e 2289 238.54 0.0449 6772 72.06 449.24 S t a n d , d e v . 291 31.01 0.0063 362 2.56 26.63 153 Run III 1.5 d a y s HRT DATE DAY A c e t i c A c i d BOD A c e t . A c i d COD A c e t . A c i d o u t r e d . i n o u t i n o u t i n c o r r e c t , . % mg/1 mg/1 07/12/90 1 08/12/90 2 09/12/90 3 10/12/90 4 11/12/90 5 12/12/90 6 1.92 99.61 442.35 19.20 507.88 22.04 13/12/90 7 14/12/90 8 15/12/90 9 16/12/90 10 17/12/90 11 2.68 99.43 422.10 20.42 484.63 23.45 18/12/90 12 19/12/90 13 20/12/90 14 21/12/90 15 3.21 99.24 378.95 23.41 435.09 26.88 22/12/90 16 16.2 23/12/90 17 4.80 98.89 388.24 31.76 445.76 36.47 24/12/90 18 25/12/90 19 26/12/90 20 27/12/90 21 2.47 99.43 389.95 18.74 447.72 21.51 28/12/90 22 A v e r a g e 404.31 464.21 S t a n d , d e v . 23.96 27.51 TKN o u t o u t mg/1 c o r r e c t . 159 19.9 154 Run I I I 1.5 d a y s HRT DATE DAY 07/12/90 1 08/12/90 2 09/12/90 3 10/12/90 4 11/12/90 5 12/12/90 6 13/12/90 7 14/12/90 8 15/12/90 9 16/12/90 10 17/12/90 11 18/12/90 12 19/12/90 13 20/12/90 14 21/12/90 15 22/12/90 16 23/12/90 17 24/12/90 18 25/12/90 19 26/12/90 20 27/12/90 21 28/12/90 22 F i l t e r e d TKN o u t o u t i n mg/1 c o r r e c t . T.PHOS. o u t o u t mg/1 c o r r e c t . F i l t e r e d T.PHOS. TS o u t o u t i n o u t mg/1 c o r r e c t . mg/1 141 17.6 6.12 9.76 1.22 4.22 0.53 6900 22250 7350 22450 A v e r a g e S t a n d , d e v . 7350 22350 225 100 155 Run III 1.5 days HRT DATE DAY 07/12/90 1 08/12/90 2 09/12/90 3 10/12/90 4 11/12/90 5 12/12/90 6 13/12/90 7 14/12/90 8 15/12/90 9 16/12/90 10 17/12/90 11 18/12/90 12 19/12/90 13 20/12/90 14 21/12/90 15 22/12/90 16 23/12/90 17 24/12/90 18 25/12/90 19 26/12/90 20 27/12/90 21 28/12/90 22 Average Stand, dev. Sulfate VS TSS VSS MLVSS S04 correct, i n out i n out i n out i n out out mg/1 mg/1 ' mg/1 mg/1 mg/1 5050 7350 900 2450 900 2350 11850 440 4000 500 5400 10850 1100 6250 1000 5200 9450 470 4190 496 5225 9100 1000 4350 950 3775 10650 455 175 1750 100 1900 50 1425 1200 15 Run III 1.5 days HRT S u l f i t e Sulfide DATE DAY S03 correct. S i n out out i n out mg/1 mg/1 07/12/90 1 08/12/90 2 09/12/90 3 10/12/90 4 11/12/90 5 12/12/90 6 13/12/90 7 14/12/90 8 15/12/90 9 16/12/90 10 17/12/90 11 18/12/90 12 19/12/90 13 20/12/90 14 21/12/90 15 22/12/90 16 59.0 53.5 6.7 3.50 2.38 23/12/90 17 24/12/90 18 25/12/90 19 26/12/90 20 27/12/90 21 56.0 40.0 4.7 3.37 1.8 28/12/90 22 Average 57.5 3.44 Stand, dev. 1.5 0.07 APPENDIX IV AJV. l DAILY LOG OF RUN IV AIV.2 RAW DATA FROM RUN IV 158 A J V . l DAILY L O G O F R U N IV (May 1, 1991 - May 21, 1991) After evaluating the data gathered up to this date it was evident that one more run was needed, this time with a temperature close to room temperature. After examining the data it was found that reasonably good organic removal was achieved during run HI at HRT = 1.5 days. The flowrate for run IV was therefore set at 8 1/day (HRT = 1.5 days), temperature at 20°C constant in the bio-reactor, with all other parameters remaining the same as in run I, II, and III. The reactor was seeded and started up on May 1st, 1991. On the first day biomass started to grow on the discs but mostly at the liquid level and media interface, close to the shaft. After 2 days on May 2nd, 1991 the biomass covered all the discs and its thickness was approximately 0.3 mm. The colour of the biomass was brownish beige. The biomass grew well and by May 6th, 1991 (after 6 days) the biofiim had reached a thickness of about 1 mm. On May 11th, 1991 the biomass on the first 5 discs (on the influent end) had a thickness of approximately 2 mm which decreased progressively to about 0.8 mm thickness at the last disc (effluent end). On May 17th, 1991 it was noted that while the first 6 discs had biomass of beige white colour, the other 17 discs had biomass of more brownish beige colour. No sloughing of biomass from the discs occurred until May 24th, 1991, three weeks after start-up. After 24 days, bridging of biomass had started and progressed to cover the first half of the discs close to the shaft. The thickness of the biomass being 2-3 mm at the influent side, decreasing to about 1 mm on the discs at the effluent side. The thickness of the biomass at that point was very hard to estimate because of its shaggy surface. Measurements were taken starting May 1st, 1991 and sampling was started on May 11th, 1991. For this run the following measurements were taken daily: Row volume in and out, temperature in the reactor, room temperature, %RH, D.O., and pH. After evaluating the data obtained in the previous runs, it was observed that steady state had been reached approximately after 2 days. However, the 159 literature indicates that steady state for aerobic treatment systems should have been reached after 10 to 14 days. Based on this information sampling was started after 10 days of running the reactor. The analysis for the following parameters was done every second day for steady state data: B O D 5 and COD; and TS, VS, VSS, and MLVSS. Samples were taken beginning May 11th, 1991 until May 21st, 1991. At this lower temperature lesser evaporation occurred than during the other runs and the bacterial wash out increased with the increasing effluent flowrate. The following observations were made during this run: (1) The effluent colour was light reddish brown; (2) the biomass on the discs had a noticeably stronger texture than was observed during run I, II, and HI and this might have contributed to the fact that sloughing occurred occurred after 24 days. In general, the operation of the RBC during this run required very little attention and the run was much smoother than the other three runs. A.IV-2 R A W D A T A F R O M R U N IV 160 Run I V DATA FOR ROTATING BIOLOGICAL CONTACTOR 1.5 d a y s HRT DATE DAY TIME VOLUME i n o u t h o u r l i t e r 01/05/91 1 02/05/91 2 24. 30 10. 55 7. 50 03/05/91 3 25. 63 10. 80 7. 40 04/05/91 4 24. 37 10. 30 7. 40 05/05/91 5 22. 83 8. 30 5. 50 06/05/91 6 24. 28 9. 10 6. 70 07/05/91 7 25. 57 9. 60 6. 40 08/05/91 8 21. 32 8. 30 5. 90 09/05/91 9 24. 00 8. 80 6. 50 10/05/91 10 24. 17 9. 40 6. 90 11/05/91 11 27. 75 11.00 8. 00 12/05/91 12 19. 75 7. 30 5. 10 13/05/91 13 24. 07 8. 90 6. 50 14/05/91 14 23. 57 8. 20 6. 00 15/05/91 15 22. 42 7. 40 5. 10 16/05/91 16 23. 58 8. 70 6. 10 17/05/91 17 24. 50 9. 00 6. 20 18/05/91 18 24. 58 8. 90 6. 50 19/05/91 19 24. 25 8. 40 6. 20 20/05/91 20 25. 13 8. 70 6. 40 21/05/91 21 23. 53 9. 30 6. 90 % FLOWRATE HRT n e t v o l u m e i n o u t v a p o u r r e d . l i t e r / d a y d a y s 3 .05 28 .91 10 .42 7. 41 3 .01 1. 16 3 .40 31 .48 10 .11 6. 93 3 .18 1. 19 2 .90 28 .16 10 .14 7. 29 2 .86 1. 19 2 .80 33 .73 8 .73 5. 78 2 .94 1. 38 2 .40 26 .37 9 .00 6. 62 2 .37 1. 34 3 .20 33 .33 9 .01 6. 01 3 .00 1. 34 2 .40 28 .92 9 .34 6. 64 2 .70 1. 29 2 .30 26 .14 8 .80 6. 50 2 .30 1. 37 2 .50 26 .60 9 .33 6. 85 2 .48 1. 29 3 .00 27 .27 9 .51 6. 92 2 .59 1. 27 2 .20 30 .14 8 .87 6. 20 2 .67 1. 36 2 .40 26 .97 8 .87 6. 48 2 .39 1. 36 2 .20 26 .83 8 .35 6. 11 2 .24 1. 45 2 .30 31 .08 7 .92 5. 46 2 .46 1. 52 2 .60 29 .89 8 .85 6. 21 2 .65 1. 36 2 .80 31 .11 8 .82 6. 07 2 .74 1. 37 2 .40 26 .97 8 .69 6. 35 2 .34 1. 39 2 .20 26 .19 8 .31 6. 14 2 .18 1. 45 2 .30 26 .44 8 .31 6. 11 2 .20 1. 45 2 .40 25 .81 9 .49 7. 04 2 .45 1. 27 A v e r a g e 28.62 9.04 6.46 2.59 1.39 S t a n d , d e v . 2.42 0.64 0.49 0.29 0.09 161 Run IV 1.5 days HRT DATE DAY TEMPERATURE REL.HUM. D.O. pH i n out room RH in mid out i n mid degree Celsius % mg/1 01/05/91 1 20.0 20.0 19.5 39.0 6. 50 6.30 6.70 8.46 8.50 02/05/91 2 19.5 20.0 19.9 27.6 0. 60 1.60 2.30 8.52 8.60 03/05/91 3 20.0 20.0 20.0 26.3 1. 60 2.10 3.00 8.54 8.60 04/05/91 4 20.0 20.0 19.5 41.5 2. 90 3.10 3.70 8.60 8.65 05/05/91 5 20.0 20.0 19.2 46.7 2. 90 2.80 3.30 8.57 8.57 06/05/91 6 20.0 20.5 18.9 56.2 3. 20 2.50 2.60 8.58 8.58 07/05/91 7 20.5 20.0 19.5 52.0 1. 80 2.30 2.40 8.56 8.56 08/05/91 8 19.5 20.0 19.5 42.1 1. 50 3.20 1.90 8.62 8.63 09/05/91 9 19.5 20.0 18.6 42.1 0. 80 2.60 3.20 8.50 8.56 10/05/91 10 20.0 20.2 19.6 41.5 0. 50 1.40 1.50 8.50 8.51 11/05/91 11 20.0 20.0 19.5 50.7 0. 30 0.50 0.80 8.54 8.56 12/05/91 12 20.0 20.0 19.3 46.1 0. 80 1.60 1.80 8.54 8.54 13/05/91 13 20.0 20.0 19.1 50.7 0. 40 1.30 1.30 8.61 8.58 14/05/91 14 20.0 20.0 20.1 42.9 0. 80 2.50 1.80 8.58 8.57 15/05/91 15 20.0 20.2 19.5 39.4 0. 40 2.50 2.20 8.57 8.56 16/05/91 16 20.0 20.0 19.5 52.4 1. 50 3.50 3.10 8.63 8.66 17/05/91 17 20.0 20.2 19.0 49.1 2. 20 2.60 1.80 8.61 8.60 18/05/91 18 20.0 20.2 19.8 50.7 1. 80 4.50 4.70 8.80 8.81 19/05/91 19 20.0 20.2 19.2 61.0 0. 70 1.60 2.10 8.61 8.61 20/05/91 20 20.0 20.3 19.3 59.6 1. 80 3.00 3.10 8.60 8.61 21/05/91 21 19.9 20.2 18.8 47.8 1. 10 2.30 2.60 8.59 8.60 Average 19.95 20.10 46.32 1. 42 2.42 2.47 8.58 8.59 Stand, dev. 0.22 0.14 8.72 0. 88 0.88 0.91 0.06 0.06 162 Run IV 1.5 d a y s HRT DATE DAY PH BOD BOdred.BODremov BOD O r g . L o a d o u t i n f l . i n o u t r a t e a c cum. r a t e mg/1 % mg/hr i n t a n k k g BOD/ % day*m2 01/05/91 1 8.51 5.62 02/05/91 2 8.62 5.65 03/05/91 3 8.62 5.48 04/05/91 4 8.66 5.45 05/05/91 5 8.58 5.37 06/05/91 6 8.58 ' 5.53 07/05/91 7 8.56 5.49 08/05/91 8 8.61 5.62 09/05/91 9 8.58 5.70 10/05/91 10 8.49 5.57 11/05/91 11 8.51 5.55 3600 1700 65.66 937 47.22 0.0196 12/05/91 12 8.54 5.57 13/05/91 13 8.56 5.61 3550 1750 64.00 840 49.30 0.0180 14/05/91 14 8.51 5.61 15/05/91 15 8.56 5.56 3550 1200 76.70 899 33.80 0.0161 16/05/91 16 8.65 5.57 17/05/91 17 8.57 5.58 4300 2000 67.96 1073 4 6 . 5 1 0.0217 18/05/91 18 8.80 5.57 19/05/91 19 8.61 5.52 3250 1250 71.61 806 38.46 0.0154 20/05/91 20 8.60 5.55 21/05/91 21 8.61 5.49 2900 1000 74.42 853 34.48 0.0157 A v e r a g e 8.59 5.56 3525 1483 70.06 901 41.63 0.0177 S t a n d , d e v . 0.06 0.07 423 354 4.58 88 6.28 0.0023 163 Run I V 1.5 d a y s HRT DATE DAY 01/05/91 1 02/05/91 2 03/05/91 3 04/05/91 4 05/05/91 5 06/05/91 6 07/05/91 7 08/05/91 8 09/05/91 9 10/05/91 10 11/05/91 11 12/05/91 12 13/05/91 13 14/05/91 14 15/05/91 15 16/05/91 16 17/05/91 17 18/05/91 18 19/05/91 19 20/05/91 20 21/05/91 21 A v e r a g e S t a n d , d e v . COD CODred.CODremov COD O r g . L o a d . T S i n o u t r a t e accum. r a t e i n o u t mg/1 % mg/1 i n t a n k k g COD/ mg/1 % day*m2 7250 5900 40. 82 1173 81. 38 0. 0394 7250 6925 8000 6200 43. 40 1284 77. 50 0. 0406 8500 6100 50. 54 1418 71. 76 0. 0385 8150 6100 48. 44 1450 74. 85 0. 0411 7950 5850 45. 69 1258 73. 58 0. 0378 7900 5750 46. 00 1436 72. 78 0. 0428 7225 6650 7958 5983 45. 81 1337 75. 31 0. 0400 7238 6788 518 160 3. 16 104 3. 26 0. 0017 13 138 Run IV 1.5 days HRT DATE DAY 01/05/91 1 02/05/91 2 03/05/91 3 04/05/91 4 05/05/91 5 06/05/91 6 07/05/91 7 08/05/91 8 09/05/91 9 10/05/91 10 11/05/91 11 12/05/91 12 13/05/91 13 14/05/91 14 15/05/91 15 16/05/91 16 17/05/91 17 18/05/91 18 19/05/91 19 20/05/91 20 21/05/91 21 VS TSS VSS MLVSS i n out i n out i n out mg/1 mg/1 mg/1 mg/1 5225 4000 1000 2300 900 2200 3600 5275 4000 900 2200 900 2000 4000 Average 5250 4000 950 2250 900 2100 3800 Stand, dev. 50 0 25 50 0 100 200 A P P E N D I X V A V . l R A W D A T A F R O M 'STERILE' R U N AN2. RESULTS F R O M MICROBIAL P L A T I N G A.V.3 R A W D A T A F R O M START-UP R U N WITHOUT M I C R O B I A L I N O C U L U M 166 AN2. R A W D A T A F R O M " S T E R I L E ' R U N DATA FOR ROTATING BIOLOGICAL CONTACTOR S t e r i l e Runs RUNS DATE TIME VOLUME % FLOWRATE HRT i n out net volume i n out of gas hour l i t e r red. l i t e r / d a y days 1.0 days HRT: 02/01/91 6.25 3.0 0.2 2.80 93.33 11.52 0.77 10.75 1.05 11.52 0.77 Average 1.5 days HRT: 04/01/91 17.73 6.2 0.2 6.00- 96.77 8.39 0.27 8.12 1.44 8.39 0.27 Average 2.0 days HRT: 06/01/91 19.50 5.2 0.15 5.05 97.12 6.40 0.18 6.22 1.89 6.40 0.18 Average 167 S t e r i l e Runs RUNS DATE TEMPERATURE pH BOD i n out i n mid out i n f l . e f f l . i n out degree Cels. mg/1 1.0 days HRT: 02/01/91 30.1 30.0 7.10 7.15 7.20 6.42 7.32 3650 4550 3300 4000 Average 3475 4275 1.5 days HRT: 04/01/91 30.0 31.5 7.20 7.21 7.30 6.42 7.67 3650 5550 3300 4000 Average 3475 4775 2.0 days HRT: 06/01/91 31.0 31.5 7.35 7.40 7.55 6.41 7.97 3650 5350 3300 6550 Average 3475 5950 168 S t e r i l e Runs RUNS DATE BODred. COD CODred A c e t i c A c i d i n o u t i n o u t o u t r e d . % mg/1 % mg/1 c o r r e c t . % 1.0 d a y s HRT: A v e r a g e 02/01/91 91.69 91.90 91.80 6900 8450 8000 7650 7450 8050 91.84 93.61 92.73 426.93 463.49 445.21 608.30 577.96 593.13 40.55 38.63 39.59 90.50 91.67 91.09 1.5 d a y s HRT: A v e r a g e 04/01/91 95.10 96.10 95.60 6900 11450 8000 11950 7450 11700 94.65 95.19 94.92 426.93 463.49 445.21 784.78 773.25 779.02 25.32 94.07 24.88 94.63 25.10 94.35 2.0 d a y s HRT: A v e r a g e 06/01/91 95.77 94.42 95.10 6900 10650 8000 14800 7450 12725 95.55 94.80 95.18 426.93 463.49 445.21 998.05 892.00 945.03 28.07 93.26 25.09 94.59 26.57 93.93 S t e r i l e Runs S u l f i d e RUNS DATE BOD AceticAcid COD AceticAcid S correct i n out i n out i n out out mg/1 mg/1 mg/1 1.0 days HRT: 02/01/91 384.24 547.47 441.16 628.58 3.58 5.40 0.36 417.14 520.16 478.94 597.23 Average 400.69 533.82 460.05 612.91 1.5 days HRT: 04/01/91 384.24 706.30 441.16 810.94 3.58 6.54 0.21 417.14 695.93 478.94 799.03 Average 400.69 701.12 460.05 804.99 2.0 days HRT: 06/01/91 384.24 898.25 441.16 1031.32 3.58 6.22 0.18 417.14 802.80 478.94 921.73 Average 400.69 850.53 460.05 976.53 171 A.V3 RAW DATA FROM START-UP RUN WITHOUT MICROBIAL INOCULUM S t a r t - u p r u n u s i n g CTMP w a s t e w a t e r w i t h o u t m i c r o o r g a n i s m i n n o c u l u m D a t e o f E x p e r i m e n t : 30 A p r i l 1991 TIME DO pH TEMP. BODin BODout BODout % B 0 D r e d MLVSS c o r r e c t h o u r mg/1 d e g . C mg/1 mg/1 mg/1 mg/1 13:00 0.0 7.70 5.91 28.0 3600 4200 3486 3.17 450 13:30 0.5 7.10 6.35 30.0 3600 4100 3403 5.47 600 14:00 1.0 6.70 6.32 30.5 3600 4100 3403 5.47 500 14:30 1.5 6.40 6.46 30.5 3600 3800 3154 12.39 550 15:00 2.0 6.20 6.40 31.0 3600 3900 3237 10.08 400 15:30 2.5 16:00 3.0 6.15 6.25 31.0 3600 4000 3320 7.78 550 A v e r a g e 30.2 508 S t a n d . d e v . 1.0 67 APPENDIX VI A.VL1 R A W D A T A F R O M T R A C E R TEST A.VL2 F - C U R V E A.VL1 R A W D A T A F R O M T R A C E R TEST 173 TIME d a y s h o u r s CONDUCT. noBACKGR. INLET C/Co OUTLET EVAPO. NOISE CONDUCT TEMP. CONDUCT. CORRECT. (C) (Co) mS mS mS mS d e g r e e C H y d r a u l i c C h a r a c t e r i s t i c s o f t h e RBC r e a c t o r : T r a c e r T e s t D a t e o f E x p e r i m e n t : 25 May - 30 May 1991 A v e r a g e E v a p o r a t i o n d u r i n g E x p e r i m e n t : 24.7% OUTLET CONDUCT. noBACKGR. TIME d a y s h o u r s C/Co OUTLET EVAPO. CONDUCT. CORRECT. mS mS INLET NOISE CONDUCT TEMP. (C) mS (Co) mS d e g r e e C 0.00 0.000 4.44 3.56 0.000 25. 5 18.9 0.50 0.008 4.65 3.73 0.168 19.6 0.82 0.013 4.80 3.85 0.289 19.4 0.89 0.017 4.89 3.92 0.361 19.6 1.00 0.019 4.96 3.98 0.417 19.3 1.47 0.031 5.28 4.23 0.674 19.5 1.48 0.032 5.30 4.25 0.690 19.5 2.00 0.040 5.53 4.43 0.874 19.6 3.00 0.054 5.90 4.73 1.171 19.7 3.28 0.057 6.00 4.81 1.251 24. 8 19.7 3.50 0.070 6.35 5.09 1.532 19.4 4.00 0.074 6.46 5.18 1.620 19.6 4.50 0.083 6.70 5.37 1.812 19.7 5.00 0.093 6.96 5.58 2.021 19.7 5.50 0.102 7.22 5.79 2.229 19.8 6.00 0.111 7.45 5.97 2.414 19.8 6.50 0.121 7.72 6.19 2.630 19.8 7.00 0.131 7.99 6.41 2.847 19.8 7.50 0.141 8.28 6.64 3.079 19.7 8.00 0.150 8.53 6.84 3.280 19.7 8.50 0.159 8.75 7.02 3.456 19.6 9.00 0.168 9.02 7.23 3.673 19.6 9.50 0.176 9.22 7.39 3.833 25. 8 19.6 10.00 0.199 9.86 7.91 4.346 19.5 11.00 0.210 10.15 8.14 4.579 19.4 12.00 0.228 10.63 8.52 4.964 19.2 13.00 0.242 11.03 8.85 5.285 19.4 14.00 0.262 11.57 9.28 5.718 19.5 15.00 0.280 12.06 9.67 6.111 19.2 16.00 0.319 13.10 10.51 6.945 19.3 17.00 0.335 13.55 10.87 7.306 19.3 18.00 0.346 13.85 11.11 7.546 19.2 19.00 0.358 14.18 11.37 7.811 19.2 20.00 0.374 14.60 11.71 8.148 19.3 21.00 0.386 14.92 11.96 8.404 19.2 22.00 0.399 15.30 12.27 8.709 19.3 174 CONDUCT. noBACKGR. I N L E T TIME C/Co OUTLET EVAPO. NOISE CONDUCT TEMP. CONDUCT. CORRECT. (C) (Co) d a y s h o u r s mS mS mS mS d e g r e e C 23.00 0.410 15.59 12.50 8.941 19.1 24.00 0.420 15.85 12.71 9.150 25. 1 19.6 25.00 0.434 16.25 13.03 9.471 19.7 26.00 0.449 16.65 13.35 9.792 19.7 27.00 0.466 17.11 13.72 10.160 19.9 28.00 0.479 17.45 13.99 10.433 19.7 29.00 0.489 17.73 14.22 10.658 20.0 30.00 0.504 18.14 14.55 10.986 20.0 31.00 0.519 18.54 14.87 11.307 20.0 32.00 0.531 18.87 15.13 11.572 20.1 33.00 0.543 19.20 15.40 11.836 19.8 34.00 0.558 19.60 15.72 12.157 25. 1 19.6 35.00 0.565 19.80 15.88 12.318 19.6 36.00 0.580 20.20 16.20 12.638 19.6 37.00 0.594 20.60 16.52 12.959 19.7 38.00 0.605 20.90 16.76 13.200 19.6 39.00 0.620 21.30 17.08 13.520 19.5 40.00 0. 628 21.50 17.24 13.681 19.5 41.00 0.639 21.80 17.48 13.921 19.5 42.00 0.650 22.10 17.72 14.162 19.6 43.00 0.655 22.25 17.84 14.282 19.6 44.00 0. 663 22.45 18.00 14.443 19.6 45.00 0.674 22.76 18.25 14.691 19.6 46.00 0.679 22.90 18.36 14.804 19.8 47.00 0.686 23.10 18.52 14.964 19.7 49.00 0.701 23.50 18.85 15.285 19.6 51.00 0.708 23.70 19.01 15.445 25. 4 19.7 53.00 0.720 24.00 19.25 15.686 19.6 59.00 0.764 25.20 20.21 16.648 19.6 70.00 0.845 27.40 21.97 18.412 19.8 72.00 0.856 27.70 22.21 18.653 19.9 75.00 0.867 28.00 22.45 18.893 20.2 77.00 0.878 28.30 22.69 19.134 20.5 79.00 0.885 28.50 22.85 19.294 19.2 94.00 0.966 30.70 24.62 21.059 19.3 95.00 0.966 30.70 24.62 21.059 19.5 97.00 0.973 30.90 24.78 21.219 19.8 100.00 0.992 31.40 25.18 21.620 25. 1 19.6 101.00 0.988 31.30 25.10 21.540 19.3 109.00 0.999 31.60 25.34 21.780 19.4 119.00 1.003 31.70 25.42 21.860 19.5 121.00 0.995 31.50 25.26 21.700 19.5 123.00 0.999 31.60 25.34 21.780 19.7 125.00 1.003 31.70 25.42 21.860 19.7 A v e r a g e S t a n d , d e v . 25.26 0.33 175 Figure A.VI.1 F - C U R V E 26 o 0 1 1 1 ' ' 1 I I I I 1 _ 0.00 3.00 6.50 11.00 19.00 27.00 35.00 43.00 59.00 97.00 time (hours) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items