Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of DTPA and hydrogen peroxide on activated sludge performance Larisch, Belinda Cornelia V. 1998

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

Item Metadata


831-ubc_1999-389219.pdf [ 8.76MB ]
JSON: 831-1.0058648.json
JSON-LD: 831-1.0058648-ld.json
RDF/XML (Pretty): 831-1.0058648-rdf.xml
RDF/JSON: 831-1.0058648-rdf.json
Turtle: 831-1.0058648-turtle.txt
N-Triples: 831-1.0058648-rdf-ntriples.txt
Original Record: 831-1.0058648-source.json
Full Text

Full Text

EFFECT OF DTPA AND HYDROGEN PEROXIDE ON ACTIVATED SLUDGE PERFORMANCE by Belinda Cornelia v. Larisch B. Eng., McGill University, 1989 M. Eng., McGill University, 1990 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Chemical Engineering We accept this thesis as conforming to the required standard-THE UNIVERSITY OF BRITISH COLUMBIA November 22, 1998 © Belinda Cornelia v. Larisch, iMb 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 of QJLtmsL <L<zJ <£v The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT This work explores the behaviour of a standard biological treatment system when exposed to two of the most common residuals from novel bleaching processes: hydrogen peroxide and the chelant DTPA. The potential problems for an activated sludge biological treatment unit, from the introduction of a chelating agent and hydrogen peroxide, were anticipated to stem from the sequestering of vital trace elements by the chelant, and the oxidation of biomass by the hydrogen peroxide. Typical residual quantities from TCF bleaching processes are 0.875 g DTPA / L and 0.2 g FL,02 / L, although the peroxide residual concentration is variable, up to approximately 1 g/L. The effects of running a typical biological treatment system on effluent from novel bleaching processes were also determined, in addition to the effects of switching influent sources between novel and conventional bleaching effluent. Activated sludge secondary treatment systems could successfully treat elementally chlorine free (ECF) and totally chlorine free (TCF) bleached kraft mill effluents by achieving > 90% BOD removal, > 40 % COD removal, and 100% acute toxicity removal. Influent feed changes between untreated chlorine free bleaching effluent and conventional effluent (from a 60% C102 bleaching sequence) resulted in immediate changes in treatment efficiency. Switching from TCF to conventional effluent decreased BOD removal, whereas switching from ECF to conventional effluent increased BOD removal. The addition of DTPA and hydrogen peroxide was found to have significant effects on activated sludge treatment. Continuous treatment of peroxide-containing wastewater reduced floe density at peroxide concentrations greater than 200 - 500 mg/L, although treatment efficiency was maintained. Continuous treatment of DTPA containing wastewater resulted in BOD removal efficiencies of 60% at DTPA concentrations greater than 600 mg/L. The addition of both DTPA and peroxide at typical TCF bleached kraft mill effluent (BKME) residual concentrations caused biological treatment to cease entirely. Individually, the effects of hydrogen peroxide and DTPA were: the reduction of biomass metabolic activity at concentrations of 100 mg/L H 2 Q, and 500 mg/L DTPA, the induction of catalase activity upon H 2 0 2 addition, and the release of cellular material due to cell wall damage at DTPA concentrations greater than 50 mg/L. Table Of Contents ABSTRACT ii Table Of Contents iv LIST OF TABLES vii LIST OF FIGURES : viii 1. INTRODUCTION 1 2. LITERATURE REVIEW 3 2.1 Chlorine-Free Bleaching 3 2.2 Effluent Treatment in the Pulp and Paper Industry 7 2.3 Activated Sludge 12 2.4 Activated Sludge Kinetics 18 2.5 Metals 21 2.6 Chelation 23 2.7 Hydrogen Peroxide 31 3. MATERIALS AND METHODS 34 3.1 Laboratory Scale Activated Sludge Reactors 34 3.1.1 Reactor Feed 34 3.1.2 Reactor Operation 34 3.1.3 Influent changeover between TCF and conventional untreated mill effluent 36 3.1.4 Constant Peroxide Study 36 3.1.5 Ramping Peroxide Study 36 3.1.6 Constant DTPA Study 37 3.1.7 Ramping DTPA Study 37 3.1.8 DTPA and Metals Study 37 3.1.9 DTPA and Peroxide Study 38 3.2 Peroxide Effects Assays 39 3.2.1 Effect on effluent characteristics 39 3.2.2 Peroxide reduction 39 3.2.3 Shock tests 39 3.2.4 Catalase-equivalent activity 40 3.3 General Analyses 41 3.3.1 BOD 5 41 3.3.2 HBOD (headspace BOD) 41 3.3.3 COD 41 3.3.4 Sludge volume index (SVI) .* 42 3.3.5 Oxygen uptake rate (OUR) 42 3.3.6 Effluent toxicity , 42 3.3.7 Activated sludge growth 42 3.3.8 Solids Retention Time (SRT) 43 3.3.9 Yield 43 3.3.10 Statistics 43 3.3.11 Curve Fits 43 iv 3.3.12 Microscopy . . . 43 3.4 DTPA Assay 44 3.5 Sludge Kinetics Measurements 44 3.6 Cellular Solute Release 46 RESULTS AND DISCUSSION 48 4. STEADY STATE REACTOR OPERATION 48 4.1 Reactor Operation Summary • • 48 4.2 Activated Sludge Operation - TCF Effluent 55 4.3 Activated Sludge Operation - ECF Effluent 58 4.4 Activated Sludge Operation - TCF / Conventional Influent Switch 61 4.5 Activated Sludge Operation - ECF / Conventional Influent Switch 63 4.6 Summary of the Effects of Reactor Influent Type on Activated Sludge Treatment 65 5. BATCH TESTS WITH DTPA AND METALS 67 5.1 Effect of DTPA on Combined Kraft Bleach Effluent Characteristics 67 5.2 Effect of DTPA on Endogenous Respiration of Activated Sludge 72 5.3 Effect of DTPA on Foaming during Batch Tests 77 5.4 Effect of DTPA on Growth of Activated Sludge . 78 5.5 Effect of the Chelant Dose on Oxygen Uptake Rate 82 5.6 Effect of Metals and the Degree of Chelation on Oxygen Uptake Rate . . . . 87 5.7 Effect of Metals on Oxygen Uptake Rate 99 5.8 Effect of DTPA on Yield 102 5.9 Cellular Solute Release 105 6. DTPA REACTOR STUDIES 112 6.1 Continuous Reactor Study 1: Constant DTPA Concentration 112 6.2 Continuous Reactor Study 2: Ramping DTPA Concentration 118 6.3 Continuous Reactor Study 3: DTPA and Calcium 124 7. BATCH TESTS WITH HYDROGEN PEROXIDE 131 7.1 Effect of Peroxide Additions on Combined Mill Effluent Characteristics . 131 7.2 Decomposition of Peroxide in Combined Mill Effluent 134 7.3 Effect of H2O2 Shock Loads on Unacclimated Sludge 135 7.4 Effect of H 2 Q J on Activated Sludge Kinetics 139 8. PEROXIDE REACTOR STUDIES 140 8.1 Continuous Reactor Study 4: Ramping Peroxide Concentration 140 8.2 Continuous Reactor Study 5: Constant Peroxide Concentration 148 9. COMBINED STUDIES WITH DTPA AND PEROXIDE 152 9.1 Respirometric Data with Combined DTPA/Peroxide Additions 152 9.2 Effect of DTPA on Catalase-Equivalent Activity Induction 156 9.3 Effect of DTPA Addition to a Continuous Reactor Receiving 500 mg H2O2 /L Influent 158 9.4 Effect of DTPA on the Cellular Solute Release of Biomass Acclimatised to 500 mg H2O2. / L Influent 160 9.5 Effect of DTPA on Growth of Biomass from 500 mg/LH202 Unit . . . . 162 v 10. CONCLUSIONS 163 11. ENGINEERING SIGNIFICANCE 168 12. RECOMMENDATIONS FOR FUTURE WORK 169 NOMENCLATURE 170 REFERENCES 172 APPENDICES 183 Appendix A: Effluent DTPA Concentration 183 Appendix B: Activated Sludge Kinetic Measurement . 185 Appendix C: Cellular Solute Release Test 187 Appendix D: Endogenous OUR Response to DTPA under Conditions of Varying Metals 190 vi LIST OF TABLES 2.1: DTPA and EDTA stability constants with selected metals 26 4.1: Reactor influent characteristics 54 4.2: Effect of reactor influent type on average COD:BOD ratios and BOD removal rates 66 5.1: Effect of DTPA on BOD measurements 70 5.2: Microtox® toxicity of DTPA 71 5.3: Effect of calcium and DTPA on Rmax in differing substrates 91 5.4: Effect of chelants and metals on cellular solute release I l l 8.1: Percent Microtox® toxicity removal 150 vii LIST OF FIGURES C H A P T E R 2 2.1 Gram positive and gram negative cell wall structures 15 2.2: Monod kinetics 20 2.3: The origin of the word "chelant" 24 2.4: Diethylenetriaminepentaacetic Acid 25 2.5: Ethylenediaminetetraacetic Acid 25 C H A P T E R 3 3.1: Lab-scale activated sludge schematic 35 3.2: Respirometer 45 C H A P T E R A 4.1 Typical appearance of healthy activated sludge floes 50 4.2 Typical appearance of unhealthy activated sludge floes 50 4.3 Activated sludge microorganisms: stalked ciliates 51 4.4 Activated sludge microorganisms: sizeable colonies of stalked ciliates 51 4.5 Activated sludge microorganisms: swimming ciliates 52 4.6 Activated sludge microorganisms: rotifers 52 4.7 Activated sludge microorganisms: nematodes 53 4.8 Activated sludge microorganisms: large unidentified organism in active pursuit of a rotifer 53 4.9 Reactor solids concentrations in 2 parallel reactors treating TCF mill effluent . . . . 57 4.10 Reactor solids concentration in a reactor treating ECF mill effluent (Control reactor, study #1) • 57 4.11 Reactor solids concentrations in a reactor treating ECF mill effluent (Control reactor, study #2) 60 4.12 Reactor solids concentration in a reactor treating ECF mill effluent (Control reactor, study #3) 60 4.13 COD reduction during feed switch between TCF and conventionally-bleached reactor influent 62 4.14 BOD reduction during feed switch between ECF and conventionally-bleached reactor influent 62 4.15 Activated sludge floes, ECF-bleached reactor influent . . . 64 4.16 Activated sludge floes, after switch from ECF to conventionally-bleached reactor influent 64 C H A P T E R 5 5.1 Effect of DTPA on COD measurement 69 5.2 Effect of DTPA on BOD measurement 69 5.3 Effect of DTPA on endogenous OUR 73 5.4 Effect of 0.25 g/L DTPA on respirometric measurements 73 5.5 OUR traces upon injection of substrate 76 5.6 Effect of DTPA on growth of activated sludge 80 5.7 Effect of DTPA on growth as measured by solids concentration 80 5.8 Effect of DTPA and excess calcium on growth rate 81 5.9 Effect of DTPA on oxygen uptake rate upon substrate injection 83 viii 5.10 Effect of DTPA on Monod parameters 83 5.11 Effect of DTPA and EDTA on OUR upon substrate injection of untreated TCF and ECF effluent 86 5.12 Effect of time on maximum oxygen uptake rate of DTPA - affected biomass . . . . 86 5.13 Effect of excess calcium on biomass kinetics 88 5.14 Effect of iron and potassium on biomass kinetics 88 5.15 Effect of time on oxygen uptake rate in the presence of DTPA and calcium . . . . . 90 5.16 Difference between the addition of DTPA and EDTA on the protective effect of calcium 90 5.17 Effect of DTPA and calcium on respirometric measurements 94 5.18 Effect of DTPA and magnesium on oxygen uptake rate 96 5.19 Effect of DTPA and iron on oxygen uptake rate 96 5.20 Effect of DTPA and calcium on oxygen uptake rate 97 5.21 Effect of DTPA and manganese on oxygen uptake rate 97 5.22 Effect of DTPA, zinc and copper on oxygen uptake rate 98 5.23 Effect of metals on maximum oxygen uptake rate 101 5.24 Effect of metals on endogenous respiration 101 5.25 Effect of DTPA on OUR trace 104 5.26 Effect of DTPA on total oxygen consumption 104 5.27 Effect of chelant on cellular solute release 106 5.28 Effect of low chelant concentrations on cellular solute release 106 5.29 Schematic representation of the proposed method of interaction between activated sludge floes and DTPA 108 CHAPTER 6 6.1 Effect on BOD reduction of the addition of a DTPA concentration of 0.875 g/L to a continuous reactor 113 6.2 Reactor solids after addition of a DTPA concentration of 0.875 g/L to a continuous reactor 113 6.3 Effluent DTPA concentration during addition of 0.875 g/L to a continuous reactor 115 6.4 BOD reduction in a reactor treating influent with increasing DTPA concentration . 119 6.5 Reactor biomass yield with the addition of increasing DTPA concentrations . . . . 119 6.6 Effect of increasing influent DTPA concentration on reactor kinetics 121 6.7 Mixed liquor prior to DTPA addition 122 6.8 Mixed liquor after DTPA addition . 122 6.9 Mixed liquor after DTPA addition 123 6.10 BOD reduction upon addition of DTPA and calcium to reactor 124 6.11 Mixed liquor prior to addition of DTPA and calcium 126 6.12 Mixed liquor prior to addition of DTPA and calcium 126 6.13 Mixed liquor 1 day after addition of DTPA and calcium 127 6.14 Mixed liquor 3 days after addition of DTPA and calcium 127 6.15 Mixed liquor 6 days after addition of DTPA and calcium 128 6.16 Effect of the addition of DTPA and calcium to a continuous reactor on cellular solute release 130 6.17 Effect of the metal addition on cellular solute release 130 CHAPTER 7 7.1 Effect of H2O2 additions on combined mill effluent BOD and toxicity 133 7.2 Effect of peroxide concentration and temperature on reaction rate of peroxide ix reduction in effluent 133 7.3 Effect of varying peroxide shock load levels on OUR of activated sludge 136 7.4 Recovery of sludge viability as a function of time after exposure to 960 mg/L peroxide shock load 136 7.5 Effect of hydrogen peroxide on oxygen uptake rate 137 7.6 Effect of single and cumulative hydrogen peroxide shocks on oxygen uptake rate 137 7.7 Effect of hydrogen peroxide shocks on endogenous and maximum oxygen uptake rates 138 7.8 Effect of hydrogen peroxide on activated sludge kinetics 139 CH A P T E R 8 8.1 Reactor performance after addition of H2O2 141 8.2 Mixed liquor floe structure before addition of H2O2 into reactor 142 8.3 Mixed liquor floe structure after addition of H2O2 into reactor 142 8.4 Rate of peroxide decomposition by unacclimated and acclimated sludge 144 8.5 Effect of peroxide shock dose on viability of sludge from reactors acclimated to different peroxide levels 144 8.6 Induced peroxide decomposition ability expressed as catalase-equivalent activity . 147 8.7 Effect of 500 mg H2O2/L influent on BOD removal 149 8.8 Effect of 500 mg H2O2/L influent on activated sludge kinetics 149 8.9 Effect of activated sludge acclimation to 500 mg/L H2O2 on catalase activity . . . 151 CH A P T E R Q 9.1 Effect of DTPA, followed by hydrogen peroxide, on activated sludge kinetics . . 154 9.2 Effect of hydrogen peroxide on activated sludge kinetics 154 9.3 Effect of DTPA on activated sludge kinetics after a 100 mg/L hydrogen peroxide shock 155 9.4 Effect of DTPA after a hydrogen peroxide shock on activated sludge kinetics . . . 155 9.5 Effect of DTPA on catalase-activity induction 157 9.6 Effect of DTPA on COD removal of sludge acclimated to 500 mg/L peroxide . . . 159 9.7 Effect of peroxide acclimation on cellular solute release 161 9.8 Effect of peroxide acclimation and DTPA on sludge growth 162 A P P E N D I C E S A . l Calibration curves, DTPA in effluent 184 A. 2 Calibration curve, DTPA in effluent 184 B. 1 Sample DO trace during substrate injection 186 B. 2 Sample OUR trace during substrate injection 186 C. 1 UV spectrograph: effect of boiling biomass for 10 minutes 188 C.2 UV spectrograph: effect of buffer alone 188 C.3 UV spectrograph: effect of DTPA without biomass 189 C. 4 UV spectrograph: effect of 1 g/L DTPA on biomass 189 D. l Endogenous OUR with magnesium 192 D.2 Endogenous OUR with iron 192 D.3 Endogenous OUR with calcium 193 D.4 Endogenous OUR with manganese 193 D.5 Endogenous OUR with zinc and copper 194 x Acknowledgements I would like to thank: • Dr. Sheldon Duff, my supervisor • Dr. Richard Branion, Dr. Jack Saddler, and Dr. Ken Pinder, my committee • Al Strang, at Howe Sound Pulp and Paper • the Pulp and Paper Centre and staff, NSERC, • and the BC Science Council. Special mention to • the PPC staff: Tim and Peter for their excellent work and support, Rita for superior librarian services, and Lisa for keeping everything organised. • fellow inmates of lab 311: Tazim, Elod, Raj, and Yee-Tak the humour in the lab. • the multitudes of bacteria and protozoa that gave their lives for this project. Most of all, special thanks to Steve. 1. INTRODUCTION The environmental impact of the pulp and paper industry has been reduced in recent years through the use of advanced pulp production and waste treatment technologies, due largely to increased market pressure for pulp and paper products which are considered to be environmentally friendly (Beaton, 1994; Hileman, 1993). The result has been a move away from the use of chlorine, the bleaching compound with the worst environmental reputation, and towards both elementally-chlorine free (ECF) and totally chlorine free (TCF) bleaching (Albert, 1994; Slinn, 1992). These relatively novel technologies have been implemented commercially over the last decade; however, the research necessary to fully understand the impact of effluent from these new bleaching technologies has lagged behind. Accompanying the introduction of novel bleaching technologies has been the construction of treatment systems which reduce the impact which a pulp mill has on its immediate environment. Although also driven by market demand, the use of treatment systems gained most of its popularity through government regulations and enforcement. Treatment systems have been designed based upon successful experience with the treatment of municipal wastewater and conventionally bleached pulp and paper effluent. However, novel bleaching processes introduce new effluent components, the treatment of which has not been tested in standard or traditional treatment systems. The concern that these new compounds may disrupt or interfere with biological treatment prompted this study. Generally, the perceived dangers for an activated sludge unit from the introduction of a chelating agent and hydrogen peroxide were anticipated to stem from the sequestering of vital trace elements by the chelant, and the oxidation of biomass by the hydrogen peroxide. Current mill practice at the time of this study was to remove hydrogen peroxide from the influent to the treatment system by souring with S0 2. It is the overall goal of this work to explore the behaviour of a standard treatment system when exposed to the most common 1 residuals from the novel bleaching processes. Additional'experiments were run to determine the general effects of switching the influent to the treatment system between chlorine-free and conventional bleached mill effluent, as this is representative of common mill practices. A series of bench scale activated sludge reactor runs were performed over a total of 921 days of operation. Nine distinct reactor runs were performed in order to gauge the response of activated sludge to hydrogen peroxide and diethylenetriaminepentaacetic acid (DTPA), which are perceived to be two of the most common and potentially problematic residual compounds from the totally chlorine free bleaching process. Although experiments were conducted in a somewhat random order, governed largely by the available mill effluent supply, results are presented and discussed based upon a division between the two compounds in the order in which they are introduced in the bleach plant, and between the batch tests and reactor runs used to obtain the data for each compound. Where applicable, batch test data is discussed in advance of the corresponding continuous reactor run data, in order to lend support to the reactor data. Steady state reactor operation on TCF and ECF influent is presented in Section 4, as are the effects of switching operation to conventionally bleached influent from either TCF or ECF operation. This is followed by the series of experiments on DTPA: batch tests in Section 5 and reactor runs in Section 6, and the series of experiments on hydrogen peroxide: batch tests in Section 7 and reactor studies in Section 8. Finally, combined DTPA and hydrogen peroxide studies are presented in Section 9. 2 2. LITERATURE REVIEW 2. I C H L O R I N E - F R E E B L E A C H I N G In response to the demand for white paper products, pulp is bleached. This is accomplished largely through the removal and decolourisation of lignin. Lignin, a major component of wood, is comprised of phenyl-propane subunits bound randomly into a three dimensional macromolecule with no repeating subunits or regular stereochemistry (Jurasek, 1995). Unlike cellulose, another major wood component, lignin colours with age, darkens and loses strength, thus necessitating its prior removal for the production of strong paper products with good brightness stability (Reeve, 1992). The random nature of the lignin molecule precludes specific chemical processes which attack known bond configurations. Instead, in processes such as kraft bleaching, harsh chemical conditions are required to break as many lignin bonds as possible, allowing for the dissolution and removal of the smaller lignin breakdown products. Delignification can be achieved through the use of chlorine (C), chlorine dioxide (D), oxygen (O), or ozone (Z)1. Further fragmentation is then achieved through oxidation with chlorine dioxide (D), peroxide (P) and sodium hypochlorite (H). The lower molecular weight fragments can be removed in subsequent extraction stages (Canovas, 1992). Significant research has been done for the optimization of each stage of delignification and bleaching, particularly in recent years with the introduction of novel bleaching processes, to ensure the use of minimal quantities of the relatively expensive newer bleaching compounds. Initial attempts to reduce the use of chlorine replaced some or all of the chlorine with chlorine dioxide, C102. The formation of adsorbable organic halides (AOX) and total organic chlorine (TOC1) compounds was found to be linearly proportional to the consumption of elemental chlorine (Axegard, 1989). C\O0 substitution up to 50% increases the formation of i See Nomenclature, p. 170 for a complete listing of bleaching sequence abbreviations. 3 chlorinated phenolic compounds, while substitutions of 50 - 100% result in decreased chlorinated phenol formation (Pryke, 1989). In addition, increased C102 substitution consistently decreases bleach plant effluent colour (Pryke, 1989). Another significant advance in the reduction of chlorine use in kraft bleaching has been the installation of oxygen delignification stages prior to the bleaching sequence. After kraft pulping, oxygen delignification removes 50% of the remaining pulp lignin, resulting in a 50% reduction in subsequent bleach demand (Allison, 1991). The search for non-chlorine bleaching sequences has increased interest in diverse technologies such as the use of hydrogen peroxide (P), ozone (Z), peracids (A), or dimethyldioxirane ("activated oxygen") (T) for bleaching, and enzymes (X) as pre-bleaching compounds (Hunt, 1995; Paice, 1995; Colodette, 1994; Troughton, 1994; Lavielle, 1993; Forrest, 1992; Sinner, 1992). In general, the bulk of research on chlorine free bleaching alternatives has focused on the use of hydrogen peroxide, already in widespread use in the pulp and paper industry, and ozone (Germgard, 1993), in comparatively limited industrial use. One of the driving forces behind the move towards chlorine free bleaching, in addition to producing environmentally acceptable products whose production meets with effluent quality regulations, is to aid in the development of the effluent free mill (Albert, 1994). Totally effluent free (TEF) pulp manufacture can be considered to be the ideal situation from the perspective of a general public which is concerned about the environmental impact of industrial processes. The elimination of chlorine in many mills around the world has brought them many steps closer to effluent free operation, by allowing for increased internal recycle of mill water without the corrosive effects of chlorine. Scandinavian countries, in particular Sweden and Finland, have led the way in the research and implementation of chlorine free technologies (Meadows, 1995). Kraft bleaching sequences have evolved from the traditional CEH (chlorine gas bleaching, caustic extraction, sodium hypochlorite bleaching) or CEHEH sequences to those using less harsh chemicals: CEDED (where D represents C102 bleaching) and DEDED (elementally 4 chlorine free), and finally to QZP (chelation, ozone, peroxide) -type totally chlorine free sequences. As mentioned previously, ozone does not have widespread industrial use, and a basic TCF bleaching sequence would involve oxygen delignification, chelation, and peroxide bleach (OQP or OQPP) (Gellerstedt, 1997; Dalhman, 1995; Dunlop-Jones, 1994). An additional advantage to the move to TCF bleaching is the relative energy demand of the processes. In general, energy demand has been found to follow the trend: peroxide bleaching < ozone < C109 (Folke, 1996). Of the novel bleaching processes currently in use around the world, those based upon bleaching with oxygen - derived chemicals are the most common, and, of these, the most industrially widespread is hydrogen peroxide (H202). Hydrogen peroxide is a well known brightening agent which has been used for over 50 years in the pulp and paper industry (Croon, 1993; Strunk, 1990). Initially, PL,02 was used for bleaching of mechanical pulp and as a reinforcement in the caustic extraction stage in the kraft process; more recently it has been used as a distinct bleaching stage in the kraft process. Effluent treatment considerations with respect to this bleaching stage stem from incomplete hydrogen peroxide consumption (Robitaille, 1988). Residual peroxide is responsible for a significant portion of the toxicity of untreated TCF effluent (Nelson, 1995; Saunamaki, 1995). It is hypothesized that residual quantities of such a strong oxidizing agent could affect not only the characteristics of the effluent, but also a downstream biological secondary treatment system. The quantity of residual peroxide within the effluent stream is variable and depends as much upon peroxide bleaching efficiency as it does on pulping and bleaching conditions and wood furnish. Effluent from a CTMP process using peroxide can contain a hydrogen peroxide concentration of 0.1-1 g/L (Andersson, 1987). In an effort to reduce or eliminate the amount of peroxide contacting the activated sludge unit, mills are known to sour the effluent prior to treatment, using sulphur dioxide (Casey, 1980). This achieves the purpose of reducing any remaining oxidants after 5 oxidative bleaching. Additional benefits of the use of SO, include the reduction of the acute toxicity of kraft effluent by 87% (Donnini, 1985). For the purpose of hydrogen peroxide reduction, a holding period for a mixture of bleach plant effluent and waste activated sludge (Cocci, 1985), and the use of an enzymatic pretreatment (Molin, 1987) have also been recommended. A peroxide bleaching stage is preceded by a chelation stage; it is required for economic reasons, to prevent the premature decomposition of peroxide by metal ions prior to bleaching. A concentration as little as 1 ppm of manganese (added in the oxygen delignification stage immediately prior to bleaching) can result in higher peroxide costs (Colodette, 1987). A non-specific chelant is used for this task, as there is a variety of metals present in the pulp. Ethylenediaminetetraacetic acid (EDTA) and diethylenetriaminepentaacetic acid (DTPA), which bind metals in a 1:1 molar ratio, are used most frequently (Lapierre, 1995). Control of the process conditions, particularly the pH, can ensure that harmful metals are removed while beneficial metals (magnesium in particular) remain within the pulp. The most common metals of concern are iron, manganese, and copper, which originate in the wood, process water, and the process equipment. The major source of the metals is the wood itself (Jayawant, 1994; Bambrick, 1985). TCF bleaching results in "rather large" EDTA / DTPA discharges (Folke, 1996), when compared with the chelant dosages commonly used in peroxide stages associated with TMP production. CTMP effluent contains 0.02 - 0.05 g/L DTPA (Andersson, 1987); TCF effluent has been reported to contain DTPA concentrations of 0.185 g/L combined mill effluent (Verta, 1996) and 0.875 g/L bleach plant effluent (Strang, 1996). 6 2 . 2 E F F L U E N T T R E A T M E N T IN T H E P U L P A N D P A P E R I N D U S T R Y The move toward the use of environmentally friendly bleaching agents has been accompanied by the installation of secondary treatment systems. These systems perform biological treatment of combined mill effluent, and are normally preceded by a primary separation stage to remove solids. The purpose of secondary effluent treatment is to remove biodegradable material from the wastewater streams which are sewered. Traditionally, these streams were added directly into the receiving waters, causing a local increase in the oxygen demand and depositing toxins such as dioxins into the environment. It was found that the increased oxygen demand adversely affected the receiving water by removing oxygen from the water. This had the effect of depriving higher organisms such as fish of oxygen, and encouraging the growth of simpler organisms such as algae. The proliferation of algae further disrupts the ecosystem in the receiving water by restricting the passage of light deeper into the water (Bailey, 1986). By reducing the allowable measured oxygen demand of the waste streams, these potential problems in the receiving waters can be greatly reduced. Primary effluent treatment traditionally involves only a physical screening or sedimentation of the effluent to remove solids such as pulp fibres. Oxygen demand reduction is achieved through the use of secondary treatment systems. Secondary treatment alternatives include stabilization ponds, aerated lagoons, anaerobic reactors, aerobic reactors, and various combinations of these. The simplest system, stabilisation ponds, require the largest area, a 20 - 60 day hydraulic retention time (HRT), and are successful at removing 50 - 90% of the BOD at a loading of 50 -60 lb (22 - 27 kg) BOD/acre/day (Springer, 1993). The next degree of complexity is the aerated stabilization basin (ASB), which is able to treat 2000 lb (900 kg) BOD/acre/day, at a removal rate of 80 - 90%. ASB's usually operate at an HRT of 5 - 10 days, and require mechanical aeration to maintain a dissolved oxygen concentration of 0.5 ppm. Nitrogen is normally added at a BOD:N ratio of (50 - 100): 1. The advantage of an ASB 7 is that, provided a temperature of 20 - 43 °C is maintained, excess sludge is not generated. Activated sludge units represent the next degree of effluent treatment complexity. These units operate at an HRT of 3 - 8 hours, with a solids retention time (SRT) of 5-10 days achieved through the use of a clarifier for solids separation. BOD removal rates are comparable to those achieved with an ASB, 80 - 90%. BOD loading, however, can be in excess of 100 lb BOD/1000 ft3 ( 160 kg BOD/100 m3) aeration volume. Solids levels are usually 2000 - 5000 mg/L. Vigorous aeration is required, and nutrient supplements are traditionally applied at a BOD:N:P ratio of 100:5:1. In addition to the reduced land area required for an activated sludge unit, the main advantage is the insensitivity of the unit to winter conditions. However, untreated effluent with a temperature greater than 37 °C requires cooling, and problems associated with the use of an activated sludge unit are the substantial production of solids and the risk of operational upsets due to sludge bulking. A variation of the activated sludge system involves the use of oxygen rather than air for aeration. In an oxygen system, higher solids levels are attained (5000 - 7000 mg/L), in addition to potentially higher BOD removal rates (87 - 91%). The oxygen activated sludge units can be significantly smaller in size than their air-based counterparts, allowing for an HRT of only 1.5-3 hours. Dissolved oxygen must be maintained at a concentration greater than 5 ppm. An alternative to the previously discussed aerobic systems is the use of anaerobic treatment. The advantages of anaerobic systems include: 1) the potential for energy production, in the form of methane, 2) lower solids production, at a rate of 20 - 35% that of aerobic treatment of similar substrates, 3) less nitrogen and phosphorus requirements, and 4) no aeration, requiring less power. Disadvantages include: 1) lower substrate removal rates per unit biomass (10 - 25% that of a similar aerobic system), 2) production of malodourous and corrosive compounds (hydrogen sulphide, mercaptans, organic acids and aldehydes), 3) the use of a large land area due to a minimum HRT of 7 - 10 days and 4) the need, often, for further treatment (i.e. aerobic polishing) to meet discharge criteria. BOD removal rates are 75 - 90%. (Springer, 1993). 8 Although the use of both aerated lagoons and activated sludge plants is well established in the pulp and paper industry, activated sludge technology represents a growing fraction of biological treatment installations (Springer, 1993) and is in fact preferred to the aerated lagoon in many countries (Boman, 1991). Treatment by the activated sludge process reduces biochemical oxygen demand (BOD), chemical oxygen demand (COD), and acute toxicity in bleached kraft mill effluent. BOD and toxicity are almost completely removed, and up to 70% of COD and AOX are removed (Simpura, 1993; Rempel, 1992). COD removal from TCF bleached kraft pulp effluent has been found to be higher (55-65%) than that from either conventionally or ECF bleached kraft effluent (35-35%) (Saunamaki, 1995). Some work has been performed on the general characteristics, treatability and toxicity of effluent emanating from novel bleaching processes. TCF bleaching of both hardwood and softwood produced effluent with much higher COD and total organic carbon (TOC) values, and lower Microtox® toxicity, than effluent from ECF bleaching of the same woods (Cates, 1995). TCF bleaching in that study was accomplished with the sequences: ZQEpZP for softwood and XZQP for hardwood. The higher TOC and COD values were attributed to the release of hydrolysed hemicellulose and associated lignin by the xylanase treatment, and to carbohydrate loss due to the non-specificity of ozone as a bleaching chemical. In a separate study on the high molecular weight effluent fractions of TCF and ECF bleaching it was found that hydrogen peroxide-based TCF bleaching effluent contained more carbohydrates than comparable samples from ECF bleaching effluent (Dahlman, 1995). In other comparisons of conventionally bleached, ECF, and TCF effluent, the higher acute toxicity found in the untreated TCF effluent was attributed to an unknown concentration of residual peroxide (Stauber, 1996; Saunamaki, 1995). This toxicity was successfully removed by activated sludge (Saunamaki, 1995). TCF bleaching was also shown to result in higher total nitrogen content due to the use of EDTA / DTPA (Verta, 1996; Saunamaki, 1995). The addition of ozone to a TCF bleaching sequence using peroxide was not found to further increase the treatability of the effluent (Saunamaki, 1995). The toxicity of effluents from different bleaching sequences varies depending on the organism used in the toxicity test. Untreated bleaching effluents (TCF, ECF and conventional) were all found to be inhibitory to algal growth. Treated effluent, on the other hand, stimulated algal growth (Ahtiainen, 1996). Untreated TCF and ECF effluents have also been found to be slightly more toxic than conventionally bleached effluent, as measured by the luminescence of the marine bacteria used in the Microtox test, V. fisheri (Ahtiainen, 1996; Stauber, 1996; Lancaster, 1992). It was concluded by O'Connor et al. (1994) that a battery of toxicity tests should be conducted, including acute, chronic and mixed function oxidase induction toxicity tests. Effluents from a variety of TCF, ECF, and conventional bleaching sequences were found to respond differently to each test (O'Connor, 1994; O'Connor 1993). At least one set of toxicity tests, comprised of assays using a variety of organisms to test mill effluent, found that TCF effluent was less toxic than ECF effluent (Lovblad, 1994). Effluent colour is often considered an important characteristic, as many mills around the world use rivers as their water source, and therefore also discharge their liquid effluent into the rivers, downstream of the plant. Depending on the size of the river, a change in water colour due to mill effluent is often one of the more publicly visible impacts of the plant. The move to chlorine-free bleaching has resulted in an improvement in effluent colour. TCF bleaching was found to produce an effluent which is very light in colour, in contrast with the effluent obtained from a conventional bleaching sequence, which is almost black. (Verta, 1996; Saunamaki, 1995; Anon, 1993). Previous work has suggested that treatment of phenolic wastewaters with H 2 0 , may lower the COD and toxicity load to aerobic biological treatment through partial oxidation of phenolic compounds. Biological treatment of the same effluent was improved following treatment with H 9 0 2 . Decreased final effluent toxicity was also observed (Bowers, 1989). This suggests a beneficial result from the presence of hydrogen peroxide in effluent prior to 10 treatment. Hydrogen peroxide has been found to reduce the COD of phenolic waste by 28 -69%, the higher levels of reduction being achieved with greater H 2 0 2 concentrations (Kibbel, 1972); reductions of up to 80% of the COD of toxic phenolic organics have more recently been reported (Eckenfelder, 1991). Anaerobic biodegradability was also improved after H 2 0 2 treatment. The biodegradable fraction of o-cresol was increased with increasing doses of hydrogen peroxide. However, the same increase in biodegradability was not observed with 2,4-dinitrophenol (Wang, 1990). In addition to improved treatability of effluents from a peroxide bleaching sequence, hydrogen peroxide provides additional benefits to biological treatment systems. It was found that 200 mg/L peroxide added to an activated sludge reactor prevented bulking, and the growth of filamentous bacteria, whereas partial deflocculation occurred at a peroxide concentration of 400 mg/L (Cole, 1973). Large doses of H 9 Q, have been used in the secondary clarifier recycle to remedy cases of filamentous infestation of an oxygen activated sludge unit (Zitrides, 1980). Due to the known bactericidal properties of peroxide (Kibbel, 1972), it is anticipated that the addition of peroxide to an activated sludge reactor may be detrimental to the performance of the reactor and could lead to the failure of the secondary treatment system. Limited research has been performed to assess the effect of DTPA on biological treatment systems. DTPA has been found to have a negative effect on upflow anaerobic sludge blanket reactors treating pulping effluent (Kennedy, 1991). It was postulated that these compounds sequestered metals essential to bacteria in the treatment system. While previous work has shown that peroxide and chelants may pose problems, knowledge is incomplete about the impact of such compounds on aerobic activated sludge reactors treating pulp mill effluent. This work seeks to more clearly define the effects which peroxide and DTPA have on an activated sludge reactor treating TCF and ECF bleached pulp mill effluent. 11 2 . 3 A C T I V A T E D S L U D G E Activated sludge is comprised of a consortium of microorganisms, which form a partial ecosystem capable of treating wastewater to an extent considered acceptable for release into the environment. The microorganisms which perform the bulk of the treatment, or decomposition of organic substrates, are bacteria. There is a variety of bacterial species present in the mixture, including many unidentified species, but important bacterial families known to be present are Pseudomonas, Alcaligenes, and Zoogloea (Hantula, 1991; Seiler, 1982). The work of Hantula et al. (1991) has led them to suggest that a few thousand bacterial species exist in the dispersed phase of municipal activated sludge, but that the most common belong to a relatively low number of genera. The bacterial cells adhere to each other through a number of distinct mechanisms, which include electrostatic interaction between bacterial surfaces, extracellular polymers and polyvalent metal ions (Eriksson, 1991). The primary step in flocculation is the production of extracellular polymers which form a slime layer on the exterior of the cells, allowing the cells to form floes. Metal ions, also considered vital to flocculation, serve as ionic bridges between negatively charged sites on the polymer. Hoc formation is advantageous, as it results in a mixture of dense solids which are easily separated from the liquid (the treated effluent) by sedimentation. The presence of the extracellular polymer layer also helps the bacteria trap and use soluble nutrients, and provides a barrier to toxins. Growing in this way, with the cells surrounded by polymer, was found to promote access of nutrients to cells deeper within the floes. Nutrient access is promoted due to the large amount of fibrous material which comprises the biofilm and prevents the occlusion of deeper cells by overlying cells (Costerton, 1985). The extracellular polymer, also referred to as the slime layer or glycocalyx, is composed largely of polysaccharides (Fletcher, 1985). The bacterial glycocalyx is not visible by conventional light or electron microscopy unless specific stains are used (Costerton, 1981). 12 However, upon staining it has been found to occupy a significant amount of space surrounding the individual cells, on the order of 50 - 100% of the length of the cell. The glycocalyx is comprised of polysaccharide-conlaining structures which are of bacterial origin (Costerton, 1981). The glycocalyx, in addition to aiding in the attachment of neighbouring cells to each other, can also be used to attach a cell to an inert surface. Once this has taken place, the cell produces very large amounts of exopolysaccharide, enabling the further concentration of nutrients in the microenvironment near the cell (Costerton, 1981). It is also possible, in liquid media, for capsular polysaccharide material to be synthesized by extracellular enzymes at distances remote from the cell (Stacey, 1960). Some bacteria, Zoogloea for example, are known to form profuse quantities of slime (Seiler, 1982). The glycocalyx is affected by nutrient conditions, and the surfactant nature of the media surrounding the cells (Govan, 1975). Govan observed that the presence of surfactant in the environment to which the bacterial cells were exposed induced a non-mucoid strain of Pseudomonas aeruginosa to change to a mucoid strain. This effect was determined not to be attributed to a selection of mucoid over non-mucoid variants. Experiments on Aerobacter aerogenes have found that most polysaccharide production occurs after growth has ceased, and that production is 10 - 20 times greater if growth stops while excess substrate (sugar) is still available. This occurred, in these experiments, due to exhaustion of nitrogen or phosphorus, and it was found that the excess sugar was used to form excess polysaccharide (Duguid, 1953). The glycocalyx is an active component of the bacterial cells, as a certain amount of cell wall polymer is regularly lost through natural cell processes. It has been found that, although variable, approximately 20% of cell wall polymers are lost per generation time, and that it is the outer 20 - 40% of the wall which is affected by turnover activity (de Boer, 1981). The capsular layer, therefore, is in a constant state of change. In addition to the presence of capsular polysaccharide material, filamentous bacteria also play an important role in floe formation. These bacteria are considered beneficial in small 13 quantities, as they form the floe backbone to which other bacterial cells may adhere through metal ion assisted bonding and exopolysaccharide production. However, a predominance of filamentous bacteria results in a condition referred to as "bulking", in which the floes become too voluminous to settle properly during clarification, and biomass is lost with the treated effluent. One of the conditions favouring filamentous growth is low dissolved oxygen concentration (Lo, 1994a). Viscous bulking, with microorganisms dispersed in an extracellular mass, with a jelly-like consistency, has also been found on one occasion at a mill (Lo, 1994a) under conditions of high F/M and high BOD loading. With respect to interactions with their environments, bacterial cell walls are the first to come into contact with any changes in the surrounding media. Cell walls are divided into two general types, gram positive (+) and gram negative (-), based upon their reaction in the gram staining test. On average, it has been found that the bacteria in wastewater treatment systems from pulp mills are comprised of equal numbers of gram positive and gram negative bacteria (Whiteman, 1998; Lewandowski, 1990). A basic diagram of the two cell walls is shown in Figure 2.1. Gram positive cell walls are less complex in nature, being comprised of many layers of cross-linked muramic (peptidoglycan) complexes attached to the cytoplasmic membrane. Gram negative cells have both an outer membrane and a cytoplasmic membrane, connected through occasional attachment points, with the space in between occupied by the periplasm. The outer membrane exists as a lipid bilayer, with the two lipid types, lipopolysaccharide and phospholipid, preferentially occupying the outer and inner membrane faces, respectively. Of the two, the lipopolysaccharide, which points into the media, contains the greatest density of electronegative sites (Hughes, 1989; Umbreit, 1976). 14 Cross-linked peptidoglycan and polysaccharides Gram positive cell wall Gram negative cell wall Figure 2.1: Simplified Gram-positive and Gram-negative cell wall diagrams \ 15 The existence of the cell wall is not always sufficient to shield the bacteria from potentially harmful compounds. The cells also have access to various internal enzymatic processes through which damaging compounds can be degraded and neutralized. Catalase, a ferric hemoprotein, is a common enzyme which catalyses the decomposition of H 2 0 9 . It is present in nearly all aerobic organisms, due to the fact that hydrogen peroxide, a compound which is toxic to cells, is generated as a by-product of the autooxidation of some organic compounds. (Hanzlik, 1976; Eichhorn, 1973). Catalase-like activity is not limited to the catalase enzyme itself, but can be exhibited by similar proteins such as peroxidase and in higher organisms haemoglobin and myoglobin, and by free iron. Catalase activity in bacterial cells is considered inducible (Halliwell, 1979). Microorganisms other than bacteria are also present in activated sludge, and include protozoans, rotifers, and nematodes. These organisms feed on free-floating bacterial cells, thus aiding in effluent clarification (Eikelboom, 1981; Jones, 1976). Protozoans commonly found in activated sludge include stalked and swimming'ciliates, flagellates, and amoebae. The ciliated organisms are the most abundant. Microscopic observation of the protozoan community is very useful in determining, in a qualitative manner, the overall health of the system. The presence of a large number of stalked ciliates indicates a healthy system, whereas a predominance of rotifers is indicative of a dying bacterial population, and a possible precursor to poor effluent treatment. The rotifers feed primarily on the dead cells and protozoans in the mixture, and their proliferation is thus an indicator of an unbalanced, largely dying, population (Eikelboom, 1981). Activated sludge is considered to be a robust form of treatment able to treat wastewaters with changing characteristics, due largely to the variety of microorganisms present. However, switching between TCF and chlorine gas bleaching was found to have an adverse effect on activated sludge floes (Saunamaki, 1995). Switching from TCF to conventionally bleached effluent resulted in an initial increase in effluent turbidity which subsequently subsided, and an 16 increase in the protozoans of the rotifer family, while stalked ciliates disappeared almost completely. A sensitivity to changing reactor feed quality is of particular importance in the pulp and paper industry, where wood furnish and bleaching conditions can be expected to vary on a daily basis. 17 2. A ACTIVATED SLUDGE KINETICS In measuring the activity of biomass, it is useful to attribute quantifiable parameters to the behaviour of the actual biological system. One of the simpler measurements which may be made of a complex biological system is the overall metabolic activity, as determined through rates of oxygen utilization. Microbial kinetics (rates of substrate metabolism) for a complex mixed culture such as activated sludge have been previously determined to follow the Monod equation, with the quantifiable parameters, u and K~s representing mean values of the corresponding parameters of the individual species (Cech, 1984; Chudoba, 1985). The Monod equation is given as: — — = = s dt X P (K,+.S) 1 J where: pi is the growth rate, ]imax is the maximum microbial growth rate, K is the half saturation constant, S is the substrate concentration, and X is the biomass concentration. This relationship is graphically represented in figure 2.2. The growth rate is converted to the substrate uptake rate through the yield coefficient, Y (biomass produced per substrate consumed). The observed yield coefficient is used to account for discrepancies between the theoretical and actual yield. dt °bs dt L_J Combination of equations [1] and [2] gives: dS 1 5 dt X 7 ,„ K+S [3] obs 18 Defining q and qmax as substrate removal rates, equation [3] becomes: S The measurements actually made in respirometric tests, however, are not of the growth rate or the substrate removal rate, but of the oxygen uptake rate. The Monod equation can be converted into a more useful form by relating the substrate removal rate, q, to the oxygen removal rate, R. Since the true yield defines the partitioning of the energy derived from the substrate into the respiration and growth fractions, the following relationship can be written: R=~(\-Y) = q(\-Y) [5] Assuming a constant yield, q is therefore proportional to R, and equations [4] and [5] can be combined to give the Monod equation for the rate of oxygen uptake: R = R ™ K ^ S  [6] Yield can also be calculated based upon available oxygen uptake data. The quantity of substrate consumed is directly proportional to the oxygen consumed, as it is assumed that the total substrate is consumed through: 1) oxidation for energy and 2) the production of biomass. If only the oxygen utilised is measured (as in respirometric tests), the yield may be calculated through the difference in the amount of substrate added (measured in terms of oxygen demand) and the oxygen used in the metabolism of that substrate. Therefore: OC r = i — m where: S is the substrate concentration in oxygen units (BOD or COD), mg/L OC is the oxygen consumed, mg/L. 19 s Substrate concentration, S Figure 2.2: Monod kinetics 2 . 5 M E T A L S Metals play an important role in the lives of bacterial cells. Calcium and magnesium are important components of the cell wall, while metals such as potassium and magnesium are associated, within the cell, with proteins and DNA (Hanzlik, 1976). Metals functioning as components of the cell wall serve to stabilize the organic charge density in the membrane, particularly the phosphoryl groups of the lipopolysaccharides. They affect molecular motion and produce an optimal lipid packing order. Magnesium and calcium are used preferentially over other metals by the organic building blocks of the cell wall as an organic cement (Beveridge, 1989). If calcium and magnesium are removed from the cell wall, the motion and packing order of the lipids is disturbed, causing the lipids to require a more highly curved surface than that which is provided by the cell. Thermodynamically, this results in the formation of small, curved vesicles which no longer fit within the membrane (Beveridge, 1989). Potassium and magnesium are required by many enzymes for activation or activity enhancement. Inside the cell, trace metals are also required by enzymes for their catalytic functions, for instance, the formation of bridges between enzymes and substrates. Metals are very often involved in oxidation, reduction, and carbon-carbon coupling (the three most important reactions in organic chemistry). Metals have a variety of effects upon microbial systems. Calcium, magnesium and iron were found to be required for the growth of cells isolated from sewage activated sludge (Kakii, 1986b), as media deficient in all three supported little or no bacterial growth. In addition to affecting cell growth, metals have also been found to play a role in cell metabolic rates. The respiration rate of Klebsiella aerogenes has been shown to increase with decreasing bulk media potassium concentrations due to the increase in ion gradient. In these experiments, the response to the changing potassium concentration was immediate (Hueting, 1979). The oxidation state of metals also has an effect upon activated sludge. Activated sludge reduced 21 Fe + 3 to Fe + 2 at a rate of 0.9 - 3.7 mg Fe/g VSS h but upon aeration, Fe + 2 was immediately oxidized back to Fe + 3. The conversion of Fe + 3 to Fe + 2 resulted in reduced floe strength (Nielsen, 1996). Lastly, some metals, such as copper and zinc, have been found to be toxic to activated sludge (Beyenal, 1997). However, in separate tests the inhibitory effect of these metals on oxygen uptake rate was found to decrease with increasing MLVSS concentration. Under some conditions, calcium was determined to be significantly stimulatory to activated sludge (Hartz, 1985). 22 2 . 6 C H E L A T I O N Chelating agents are introduced in the pulp bleaching sequence prior to the peroxide bleaching stage, to ensure that the oxidative strength of the hydrogen peroxide is directed towards pulp bleaching rather than metal oxidation (Lapierre, 1995; Basciano, 1990). In alkaline solution (the condition in the peroxide bleaching tower) hydrogen peroxide is susceptible to decomposition, especially in the presence of contaminants such as metals (Wallace, 1962). Chelation is the process by which metal ions are sequestered by another compound, referred to as the chelant or ligand. If the ligand is attached to the metal ion by more than one donor atom, thus forming a ring, the metal complex becomes a metal chelate. Binding in this way, to multiple sites on the ligand, increases the stability of the complex. Functional groups commonly associated with chelants are: acidic groups, such as: - C O O H , - S 0 3 H , - O H (enolic and phenolic), - NH 2R, - N H , and coordinating groups such as: = 0, - N = , - O - R -These functional groups must be situated to permit ring formation, with a metal ion as the closing member. Some metal ions, such as Cu 2 + , N i 2 + , and Co 2 + prefer nitrogen donor atoms, while others, such as M g 2 + and Ca 2 + prefer oxygen (Mellor, 1964). In general, a chelant which contains n donor atoms capable of coordination to a single metal ion will form (n-l) chelate rings. The greater the number of rings, the greater the stability of the complex; therefore, a larger number of donor atoms implies that the chelant will be able to form more stable complexes. The number of bonds, up to n, which a chelant is capable of forming with a metal ion is determined by the coordination number, historically referred to as the auxiliary valency, of the metal. This number, defined as the number of donor or ligand atoms bound 23 directly to a central metal atom, is not necessarily fixed for a given metal, although for metals in lower oxidation states (ie. +1, +2, +3) the coordination number is often 4 or 6, but can be any number from 2 to 10 (Bell, 1977; Mellor, 1964). The word chelate is derived from the Greek word X^X'i) (chele), meaning a lobster's claw (Dwyer, 1964), as illustrated in figure 2.3. A wide variety of chemical compounds can act as chelants under the proper conditions. Figure 2.3: The origin of the word "chelant" DTPA and EDTA are both referred to as multidentate ligands, since they bond to a metal through more than one of their functional groups (or "teeth"; hence the term "dentate"). DTPA (Figure 2.4) is an octadentate ligand, whereas EDTA (Figure 2.5) is a sexadentate ligand (Dwyer, 1964), each bonding with both oxygen and nitrogen atoms to a metal ion. Both chelants are commonly used in the chelation stage during pulp bleaching, due to their high stabilities with metals commonly found in pulp (Lapierre, 1995). Chelants are also commonly used as an ingredient in media for the maintenance of bacteria (Umbreit, 1976) and protozoans (Kirsop and Doyle, 1991). The presence of the chelant in media aids in holding metals in solution. Additional common usages for chelants are as therapeutic agents in instances of metal poisoning (Jones, 1983). 24 H C O C ^ H 2 ^ ^CH2 -COOH H—N- CH2-CH2 -N-CH2-CH2 - N - H / I \ HOOC-CH2 C H 2 CH2-COOH COOH Figure 2.4: Diethylenetriaminepentaacetic Acid (Frost, 1956) HOOC-CH2 CH2-COOH H—N-CH2-CH2 — N - H / \ HOOC-CH2 CH2-COOH Figure 2.5: Ethylenediaminetetraacetic Acid The affinity of a specific metal for a specific chelant is indicated by stability, or equilibrium, constants. The relationship between a metal (M) and a chelant or ligand (L) can be written as: M + L ^ M L [8] where ML is the metal chelate. The equilibrium constant is therefore: K - [ M L ] rei e* [MJ[L] L J Higher equilibrium constants are a result of the equilibrium in equation [8] shifting to the right, and are indicative of greater stability of the metal chelate. As generally similar compounds, EDTA and DTPA have similar affinities for metal ions; however, it may be noted in table 2.1 that DTPA, with a greater number of binding sites, produces more stable complexes than does EDTA for the same metal ions (Lapierre, 1995). 25 Table 2.1: DTPA and EDTA stability constants with selected metals Metal Ion Log K * DTPA Log K * EDTA C a + 2 10.75 10.61 C u + 2 21.4 18.7 Fe + 2 16.4 14.27 Fe+3 28.0 25.0 K+ n/a 0.8 (20°C) Mg+2 9.34 8.83 Mn+2 15.51 13.81 Z n + 2 18.3 16.44 * at 25 °C, 0.1 ionic strength (Martell, 1974). Large chelants such as EDTA and DTPA were originally thought to be non-toxic to bacterial cells due to their inability to diffuse through the cell membrane. However, they have been found to affect cells in other ways. Chelation by EDTA has been shown by numerous researchers to increase cell wall permeability (Weiser, 1968; Wilkinson, 1967; Lieve, 1965; Gray, 1956a,b). EDTA extracts portions of the lipopolysaccharide, phospholipid, and protein cell wall components by removing the divalent metal ions which bind these cell wall components together. These compounds are extracted as distinct, small, outer membrane vesicles which, in the case of E.coli, have been shown to comprise 33, 24, and 10% of the lipopolysaccharide, phospholipid, and protein (respectively) from the wall. Once portions of the outer membrane of E. coll have been extracted with EDTA, the membrane becomes deficient in calcium (Ferris, 1986). The effect of EDTA upon cells was found by Goldschmidt (1966) to be of a permanent nature; once exposed, the effect of the EDTA remained despite repeated washing. EDTA was not found to bind to the outer membrane of E. coli, as 99% of the chelant remained soluble after treated cells were washed (Ferris, 1986). EDTA treatment was found to cause cells to become more hydrophobic, due to an interfacial free energy increase caused by loss of a portion of the lipopolysaccharides which were capable of 26 hydrogen bonding to the aqueous phase (Tanford, 1980). EDTA addition to Pseudornonas aeruginosa caused the formation of osmotically fragile rods which were restored by the addition of multivalent cations to stable forms, capable of normal respiration. The site of attack of the chelant was inferred to be the lipopolysaccharide component of the cell wall (Asbell, 1966a,b). Increased cell wall permeability in E. coli caused by EDTA was also prevented by the addition of magnesium sulphate or calcium chloride (Weiser, 1968). The concentration of metals, in particular magnesium, during growth has been found by Brown (1975; 1969) to control cell sensitivity to EDTA for Pseudornonas aeruginosa. Divalent magnesium ion concentrations greater than the minimum required for growth resulted in increased sensitivity. Other divalent cations have similar effects, but with varying effectiveness for inducing sensitivity to EDTA: M g + 2 > M n + 2 > C a + 2 > Ba + 2 > Sr + 2 Z n + 2 was found to have no effect, and Fe + 2 was found to have a protective effect. In addition, cultures sensitive to EDTA were found to have high wall concentrations of M g + 2 and Ca + 2 , whereas those insensitive to EDTA were found to have low concentrations of M g + 2 and C a + 2 in the cell wall. Brown (1975) hypothesized that there are 2 possible sites of action for EDTA: 1) the surface of the cell envelope and 2) the cytoplasmic membrane. Electron microscopy has shown that cell envelopes which originally had a close-knit appearance had gaps, attributed to the loss of cell envelope material, after treatment with EDTA (Stinnett, 1973). Different microbial populations exhibit differing sensitivities to EDTA; Pseudornonas aeruginosa has been found to be unusually sensitive to metal binding compounds (Wilkinson, 1975). However, the presence of slime, the extracellular capsular layer surrounding bacterial cells under some conditions, was found to significantly reduce the sensitivity of Pseudornonas aeruginosa to EDTA. In addition to growth conditions discussed above, immediate environmental conditions were also found to affect sensitivity. Cells stored for several hours were less sensitive to lysis 27 by EDTA and lysozyme than were freshly harvested cells. Sensitivity of the stored cells was restored by aeration (Repaske, 1958). Although, as the above review indicates, abundant work has been conducted on the direct effects of EDTA on cell walls, comparatively limited research has been performed on the effects which chelants have upon cellular functions such as metabolism. Chelated metals have been found to affect respiration rates in mammalian cells, and cell isolates. Cobalt with chelant caused hyperglycaemia in rats (Koch, 1955), and iron chelates were found to stimulate respiration in rat brain cells (but Cu, Mn, Ni, and Co chelates did not) (Shulman, 1964). Tris chelates were shown to uncouple phosphorylation from oxidation in rat heart mitochondria (Yang, 1958). Other researchers attempted to determine the mechanism of respiratory stimulation by certain metal chelates of EDTA (i.e. with iron) by investigating the effect on isolated components of the respiratory chain of liver mitochondria (Lenta, 1960). They found that the ferric chelate was able to stimulate oxidation of DPNH (reduced diphosphopyridinenucleotide) and that the feme chelate could compete with a cytochrome component for electrons. They suggested that the most likely site of action of the ferric chelate was at the level of the flavoproteins. Similar redox potentials existed when comparing the flavoprotein systems and the ferric chelate system. EDTA has also been shown, at a concentration of 1 mM, to completely inhibit growth of bacteria isolated from activated sludge (Kakii, 1986b). Strong, non-specific chelants such as DTPA and EDTA present additional concerns with respect to their persistence. If they are able to pass through wastewater treatment systems, they may be able to remobilize heavy metals from river sediments and/or affect metal balances of the receiving waters (Bolton, 1993; Gardiner, 1976). Degradation of these nitrogen-containing compounds in the environment external to the treatment system could result in increased eutrophication (Folke, 1996). Highly variable results have been obtained by numerous researchers pertaining to the biodegradability of the chelants EDTA and DTPA. EDTA 28 degradation of up to 89%, by microorganisms from an aerated lagoon, has been reported (Belly, 1975). EDTA has also been shown to be degraded by an enriched bacterial mixed culture, which was able to use EDTA at a concentration of 0.2 g/L as its sole source of carbon and nitrogen for growth (Nortemann, 1992). In a separate study which was unable to isolate pure bacterial cultures capable of using EDTA or DTPA as a sole source of carbon or nitrogen, EDTA and DTPA were found to be tolerated at concentrations of up to 1000 and 500 mg/L respectively when in the presence of 100 mg/L magnesium (Neilson, 1977). Various bacterial strains were tested, including members of the Pseudornonas, Acinetobacter, Alcaligenes, Bacillus and Klebsiella genera, many of which were able to provide at least 75% of the growth yield of control cultures in the presence of 100 mg/L magnesium. However, numerous bacterial strains could give similar performance only with significantly decreased DTPA concentrations (ie. 200, 100, 50 or 0 mg/L), indicating a greater sensitivity of these bacteria to the chelant. Additionally, as the magnesium concentration in the growth media was decreased to 8 mg/L, gram positive bacteria and Pseudornonas aeruginosa became increasingly sensitive to the chelant. In a separate study, up to 4.7 g/L EDTA was degraded only by growing cells from industrial sewage sludge, and only if divalent metal ions were present in stoichiometric excess. Cell growth was precluded at any metal concentrations less than this (Henneken, 1995). In the same study, EDTA was found to remain inert if complexed with Fe 3 +, with which it has a high thermodynamic stability constant (Table 2.1). The study also concluded that when EDTA degradation occurred, it was completely metabolised to C0 2 , H 2 0, N H 4 + and biomass. Tests which were performed using activated sludge for chelant degradation yielded significantly less promising results. Researchers in two such studies concluded that effectively no EDTA breakdown was achieved through operation of an activated sludge plant (Kari, 1996; Saunamaki, 1995; Alder, 1990). Recently, an average EDTA degradation of 50% in activated sludge was reported. This performance was attributed to pH regulation, and was obtained 29 from a feed whose EDTA concentration was only 6 mg/L (Virtapohja, 1998), significantly less than the concentration used in other studies, and the concentrations expected as residual quantities in the pulp and paper industry. In summary, it would appear as though inconsistent success has been achieved in the breakdown of EDTA using biological means. In spite of the minimal research performed using DTPA, it is likely that DTPA, considered more recalcitrant than EDTA (Neilson, 1977), would remain intact through biological treatment. Overall physical effects of chelants on activated sludge have also been observed. EDTA has been shown by numerous researchers to increase deflocculation of activated sludge biomass (Silverstein, 1994; Eriksson, 1991; Kakii, 1986a). Silverstein (1994) used a low concentration (0.5%) to break up activated sludge floes for enumeration purposes, based on the belief that EDTA tied up the multivalent ions, particularly Ca + 2 , important to cell aggregation. A similar calcium-specific chelant, ethylene glycolbis(B-aminoethyl ether)-N,N-tetraacetic acid (EGTA) was used to cause a biofilm to detach from its support within 5 minutes. The result was attributed to the removal of essential calcium from the biofilm (Turakhia, 1983). The addition of calcium chloride to cells prior to EDTA treatment was found to afford complete protection to Azotobacter cells (Goldschmidt, 1966). The addition of iron caused recovery of activity loss thought to be due to DTPA in a combined anaerobic/aerobic wastewater treatment plant (Driessen, 1994). 2 . 7 H Y D R O G E N P E R O X I D E Hydrogen peroxide can serve a number of purposes in a pulp mill, as a distinct bleaching chemical, or an added chemical in an extraction stage. Hydrogen peroxide bleaches pulp through decomposition into oxygen and water via a vast number of reactions which produce intermediate species, including hydroxyl and perhydroxyl anions, hydroxyl and perhydroxyl radicals, and many others. The intermediate believed to be responsible for much of the bleaching is the perhydroxyl anion (OOH-) via a nucleophilic reaction with the lignin-based chromophoric groups (Stromberg, 1994; Anderson, 1992). The following simplified reactions involving the decomposition of hydrogen peroxide play a role (Stromberg, 1994): H202*OOH-+H+ [10] H202*OH++OH~ [11] H202*OOH» + H<> [12] H202*OH» + OH* [13] H202+OH-*OOH- + H20 [14] H202 + OOH-*H20+OH~ + 02 [15] H202 + H+*OH+ + H20 [16] H202«H20+\02 [17] The active bleaching agents, perhydroxyl anions, are formed preferentially at high pH (reaction [14]). Other intermediate species formed during peroxide decomposition include free radicals such as OH* and OOH», which are highly oxidizing, non-selective, and degrade cellulose 31 (Jayawant, 1994). The generation of these destructive radicals can be controlled by the regulation of pH, temperature, and transition metal concentration (Anderson, 1992). In addition to bleaching, peroxide is also used to delignify pulp. At high pH, the phenolic hydroxyl components of the lignin molecules are ionized, and the lignin anion reacts, forming radicals and unstable intermediates which result in the opening of the aromatic rings (Anderson, 1992). Under conditions of high pH, therefore, peroxide is a valuable chemical in the bleach plant. Peroxide is very stable if maintained in pure form under acidic conditions; its decomposition is catalysed by both base and metals. Transition metals, especially manganese, iron, and copper, are able to catalyse peroxide degradation (Anderson, 1992; Basciano, 1990). Peroxide decomposition in the presence of metals is thought to proceed through the following free radical mechanism (Bambrick, 1985): Fe+2 + HOOH -Fe+3 + OFT + HO • [18] 2 0H~-0+H20 I [19] \02\ In addition, therefore, to the consumption of hydrogen peroxide in a non-bleaching, non-delignifying reaction in the presence of metals, the resulting hydroxyl radical created will have a detrimental effect on the pulp through a degradation of pulp strength. Not all metals catalyse peroxide degradation. Some tin, barium, and aluminum compounds stabilize peroxide, but are detrimental to bleaching efficiency. Sodium silicate, Na^SiOy2, is also known to have a stabilizing effect, while increasing bleaching efficiency (Kutney, 1985). Of the metals present in the pulp mixture, the iron and copper which are 32 strongly attached to the pulp do not contribute to the catalysis of hydrogen peroxide decomposition (Colodette, 1989). With respect to effluent treatment considerations, hydrogen peroxide is generally deemed to be incompatible with biological treatment systems due to its strongly oxidative nature and the knowledge that a 3% H 2 0 2 solution is commonly used as a disinfectant (which is by definition antimicrobial in nature) for cleansing wounds (Kibbel, 1972). Normal residual peroxide concentrations in effluent are 150 - 600 mg/L, to a maximum of 1200 mg/L (Andersen, 1995; Schnell, 1993). The impact of such levels of H ? 0 2 on activated sludge has not been previously studied. 33 3 . MATERIALS AND METHODS 3. I L A B O R A T O R Y S C A L E A C T I V A T E D S L U D G E R E A C T O R S 3.1.1 Reactor Feed. Untreated mill effluent was obtained from Howe Sound Pulp and Paper (Port Mellon, BC) and stored at 4°C until use. Howe Sound Pulp and Paper Ltd. operates an integrated kraft and thermomechanical pulping (TMP) operation and produces 1000 tonnes per day bleached kraft pulp and 585 tonnes per day of quality newsprint. The bleached kraft mill is a single line operation and is capable of producing high chlorine dioxide substitution, elemental chlorine free or totally chlorine free bleached kraft pulp in response to market demands. The TCF process uses a peroxide bleaching stage, preceded by a chelation step. Combined mill effluent is comprised of approximately 75% (by volume) kraft mill effluent and 25% TMP effluent, and is treated in a primary clarifier. TCF combined mill effluent was obtained for the first peroxide study and TCF steady state study, and ECF kraft effluent was used in the DTPA studies to ensure an initial DTPA concentration of 0 g/L, and in the second peroxide study due to the unavailability of TCF effluent. ECF bleaching sequences use conventional bleaching with 100% C102 substitution. 3.1.2 Reactor Operation. Two continuous lab-scale 3.5-litre working volume activated sludge reactors were operated (figure 3.1). One reactor was used to provide sludge samples for batch tests and as a control reactor, and the other for the continuous reactor studies. Each reactor consisted of a 4-litre cylindrical jacketed Plexiglas aeration tank and a 0.5 litre conical glass clarifier. Feed, waste sludge, and sludge recirculation rates were regulated by Chrontrol (San Diego, CA) timers controlling Masterflex peristaltic pumps (Cole-Parmer, Chicago, IL). Reactor temperature was controlled at 34°C by circulating water from a constant temperature bath 34 (VWR Scientific) through the annular Plexiglas jacket surrounding the aeration tank. Initial reactor startup was accomplished by seeding the reactors with waste activated sludge from the Howe Sound Pulp and Paper mill's oxygen-activated sludge plant. Smoothest start-up was achieved by dilution of waste activated sludge to the desired final concentration using tap water prior to feeding the reactors with untreated mill effluent. All activated sludge used in this study was obtained from the lab-scale activated sludge units. W ASTE S L U D G E AIR | TREATED E F F L U E N T Figure 3.1: Lab-scale activated sludge schematic The operating conditions of the activated sludge units were as follows: hydraulic retention time (HRT) = 12 hours (except for the ramped peroxide study: HRT = 30 hours), solids retention time (SRT) = 5 days, nutrient addition (BOD:N:P) = 100:5:1. Feed pH was adjusted to 6.5 - 7.5 and reactor dissolved oxygen content (DO) was maintained at approximately 5 mg/L through aeration. Nutrients were added in the form of concentrated NH 4 OH (Fisher Scientific, Nepean, ON) and H 3 P0 4 (BDH, Toronto, ON). Biochemical oxygen demand (BOD), chemical oxygen demand (COD), and toxicity analyses were performed throughout the studies to monitor effluent treatment and reactor performance, and reactor solids concentrations were measured regularly. Steady state operation was defined as the passage of at least 3 SRT's with stable performance and mixed liquor solids concentrations. 35 INFLUENT W A T E R J A C K E T Sludge age was controlled through the sludge waste rate. Waste sludge was removed directly from the aeration tank by a peristaltic pump on a timer, set to remove sludge once every hour. Readjustment of this rate was done when needed to achieve an SRT of 5 days. Additional operational details for individual reactor runs follow in sections 3.1.3 - 3.1.9. Between reactor runs, both reactors were shut down and cleaned; each new set of continuous tests began with fresh waste activated sludge from the operational treatment system at Howe Sound Pulp and Paper. 3.1.3 Influent changeover between TCF and conventional untreated mill effluent The influent to one activated sludge unit, treating TCF effluent at an HRT of 12 hours, was switched to conventionally bleached (60% CIO, substitution) for a period of 17 days and subsequently switched back to TCF effluent. BOD removal efficiency for this period was monitored using the HBOD (headspace BOD) test (Logan, 1993). 3.1.4 Constant Peroxide Study After achieving steady state operation while treating ECF effluent, 500 mg H 2 Q, (Fisher Scientific, Nepean, ON) /L influent was added to the reactor. After 22 days, reactor influent was switched to conventionally bleached effluent (60% CIO,, substitution) due to the unavailability of ECF effluent. 3.7.5 Ramping Peroxide Study. After achieving steady state operation while treating TCF effluent, the reactor influent was supplemented with increasing concentrations of H 9 0 9 , beginning at a level of 5 mg H 2 0 2 / L influent and increasing to 1000 mg H,0 2 /L influent. Hydrogen peroxide (50% w/vv, Fisher Scientific, Nepean, ON) was added directly to the reactor by continuous operation of a Cole Parmer 74900 syringe pump (Cole Parmer, Chicago, IL). A minimum of 2 - 3 sludge 36 ages passed between step changes in peroxide addition. At each level of H 0 O 2 dosing, samples of mixed liquor were removed from the reactor and assayed to determine the response to H 2 0 2 shock loading and catalase-equivalent activity. An HRT shock was administered to the reactor at the completion of the above study, while the reactor was receiving 1000 mg H 2 Q,/L influent. The feed rate to the reactor was doubled, decreasing the HRT from 30 h to 15 h. BOD removal efficiency during the HRT shock was monitored using the HBOD (headspace BOD) test. 3.1.6 Constant DTPA Study After steady state operation was achieved using ECF effluent, 0.875 g DTPA (Sigma, St. Louis, MO)/L feed was added to the reactor influent. This chelant concentration was chosen to match the average bleach plant DTPA concentration at Howe Sound Pulp and Paper mill during TCF operation. DTPA was added to the feed through elevation of the pH of a feed sample until dissolution, and subsequent mixing of the smaller feed sample with the remainder of the influent. 3.1.7 Ramping DTPA Study After steady state operation was achieved by a reactor treating ECF effluent, increasing concentrations of DTPA were added in increments of 0.2 g/L, beginning with 0.2 g/L. A maximum concentration of 0.8 g/L was added. 3.1.8 DTPA and Metals Study DTPA at a concentration of 0.875 g/L and excess calcium chloride (to bind DTPA) was added to the reactor influent. This experiment was performed twice. On the first occasion, an unknown blend of effluent was being treated, most probably ECF, and on the second 37 occasion, conventionally bleached effluent was being treated. 3.1.9 DTPA and Peroxide Study DTPA was added, at a rate of 0.875 g DTPA/L influent to the reactor previously acclimatised to receiving a constant addition of 500 mg H ? 0 2 /L influent, while treating conventionally bleached effluent. A hydrogen peroxide concentration of 500 mg/L was chosen based upon the results of the initial ramping peroxide study (section 3.1.5), in which it was determined that this was a concentration to which the biomass could acclimatise. 38 3 . 2 P E R O X I D E E F F E C T S A S S A Y S 3.2.1 Effect on effluent characteristics 1 M and 10 M stock solutions of hydrogen peroxide were made up before each set of assays. Varying concentrations (0-1000 mg/L) of H 9 0 2 were added to untreated effluent and, after the peroxide concentration had decreased to non-detectable levels, BOD and toxicity were measured. 3.2.2 Peroxide reduction Analysis of hydrogen peroxide decomposition in both effluent and activated sludge was carried out using Merkoquant hydrogen peroxide test strips (Merck, NJ). Effluent or mixed liquor samples of 10 mL were placed in a stirred beaker and temperature was controlled using a circulating water bath. Known initial concentrations of hydrogen peroxide were obtained by micropipet addition of H 2 0 2 and the concentration of H ? 0 2 as indicated by the hydrogen peroxide test strips was monitored over time. Experimental data for peroxide reduction were fit to a first order kinetic model, and have therefore been analysed and presented using a calculated pseudo-first order rate constant, k, as a means of comparison. 3.2.3 Shock tests The ability of activated sludge to tolerate sudden increases in hydrogen peroxide concentration was determined by comparing the viability (as measured by oxygen uptake rate (OUR)) of mixed liquor samples from the continuous activated sludge reactor. OUR was measured both before and 15 minutes after rapid introduction of a range of peroxide concentrations (10 - 960 mg H 2 0 2 /L mixed liquor). After exposure to H 9 0 2 , the sludge sample was centrifuged (1000 x g, 5 minutes) and resuspended in mill effluent (to ensure 39 removal of residual H 2Q,) before each oxygen uptake rate measurement. Cumulative shock tests were conducted by adding peroxide to a previously shocked biomass sample. In tests where endogenous respiration rate measurement was specified, the peroxide shock procedure did not involve resuspension in mill effluent. Complete peroxide decomposition for these tests was verified using Merkoquant hydrogen peroxide test strips (Merck, NJ). 3.2.4 Catalase-equivalent activity Catalase-equivalent activity was, measured using the method of Welander (1989). Briefly, 2 mL of mixed liquor sample was added to a stirred vessel containing 18 mL of 0.1 M phosphate buffer at pH 7. Twenty mL of 10 M H 2 0 2 were added to give an initial H 2 0 2 concentration of 10 mM. To determine the remaining concentration of H,,Q,, samples (0.5 mL) were removed over a 1 hour time period and added to 4.5 mL of reagent containing 1 mM 4-aminophenazone, 6 mM 4,5-dihydroxynapthalene-2,7-disulphonic acid and 5000 units/L peroxidase (Sigma P-8375, Sigma, St. Louis, MO). After 15 minutes, to allow for colour stabilization, blue colour was measured @ 572 nm using a Hach spectrophotometer. Log [H202] in mM was plotted versus time, and catalase activity was calculated as: -23.02585 x slope (/min). 40 3 . 3 GENERAL ANALYSES 3.3.1 BOD5 BOD 5 was measured in standard 300 mL BOD bottles using standard methods (Greenberg, 1992). Polyseed (Polybac Corporation, Bethlehem, PA) was used as biological seed for the BOD tests during the ramping peroxide study, and laboratory reactor seed was used during the remainder of the reactor scale studies. 3.3.2 HBOD (headspace BOD) HBOD was measured based upon the method of Logan and Wagenseller (1993), using 60 ml BOD bottles. Measurements were made in 24 h. Briefly, HBOD is calculated using the formula: VxMPiW3 cf HBOD=lv^rjw (l-^^-^ where: V = volume of gas V= total volume M= molecular weight of oxygen p = partial pressure of oxygen (atm) R = universal gas constant (0.0821 atm/mol °K) T= absolute temperature (K) ^ f s a t = concentration of oxygen in liquid phase (initial, final, saturation) Untreated and treated effluent HBOD tests used gas volumes of 30 ml and 7 mL, respectively. Polyseed (2 mL) (Polybac Corporation, Bethlehem, PA) was used as a seed for HBOD studies. 3.3.3 COD COD was measured using the standard colorimetric method (Greenberg, 1992). 41 3.3.4 Sludge volume index (SVI) Sludge volume index was measured based on standard methods (Greenberg, 1992), but using 100 mL instead of 1 L sample volumes to avoid disturbance of the reactors. SVI is calculated as follows: ^ settled sludge volume (mL/L)xl000 suspended solids (mg/L) 3.3.5 Oxygen uptake rate, (OUR) Oxygen uptake rate was measured in standard 300 mL BOD flasks using the following procedure. II necessary, mixed liquor samples from the main aeration tank were aerated for 2 -3 minutes, until the dissolved oxygen (DO) was greater than 5 mg/L. The DO decrease was monitored using the Model 59 dissolved oxygen meter (YSI Inc., Yellow Springs, Ohio) over a period of 15 minutes, or until the dissolved oxygen reached a value of 1.0 mg/L. Oxygen uptake rate was calculated as the slope of DO vs time (mg/L min). Specific OUR was calculated by dividing the OUR by the concentration of mixed liquor volatile suspended solids, MLVSS, in the sample. 3.3.6 Effluent toxicity Effluent toxicity was measured using the Microtox® 500 toxicity analyser and the basic test protocol supplied by Microbics Corporation (Carlsbad, CA). The microorganism used in this test is the luminescent marine bacteria, Vibrio fisheri. 3.3.7 Activated sludge growth Growth was measured by OUR (Gaudy, 1988). Mixed liquor samples of 2 mL were added to 70 mL of untreated effluent, and incubated in a shaker (Brunswick, NJ) at 150 rpm and 34 °C. OUR was monitored over time by periodically transferring the samples to 60 mL 42 BOD flasks. 3.3.8 Solids Retention Time (SRT) Solids retention time in the continuous reactors was calculated as: MLVSS • Reactor Volume SRT = (MLVSS • Waste Rate) + (Effluent VSS • (Feed Rate - Waste Rate) 3.3.9 Yield Yield in the continuous reactors was calculated as follows: MLVSS. HRT Y = B O D r e m o v c d . S R T Yield in respirometric tests was calculated through manipulation of the Monod relationship (Section 2.4, Equation 7): Y = 1 - ° C S 3.3.10 Statistics Unless otherwise specified, confidence intervals for all data analysis were calculated as 95% confidence intervals (Himmelblau, 1980). 3.3.11 Curve Fits Non-linear regression of respirometric data was performed using the software program "Deltagraph" (SPSS). 3.3.12 Microscopy Microscopic analysis was performed using a light microscope to which a Nikon camera was attached. Magnifications of 50,100, and 400X were used without staining. Activated sludge samples, taken from the aeration tank were examined on glass microscope slides. 43 3 . A D T P A A S S A Y DTPA was assayed by titration with CaCl2, based upon DOW Chemical Method #90947A (1981). Samples were prepared by filtration through 1.5 micron glass fibre filters, and pH was adjusted to 12 using NaOH. CaCl2 (0.2M) was added in 100 microlitre increments by micropipet to 25 mL samples, allowed to mix for 5 s, and the absorbance was measured at 450 nm. Marked increases in turbidity upon saturation of the DTPA with calcium were evident by the change in slope of the data when the absorbance (A 4 5 0 ) was plotted versus the CaCl2 volume added. Calibration with known DTPA concentrations in lab-treated effluent allowed for measurement of DTPA in treated effluent samples. Each assay required separate calibration, as the changing characteristics of the effluent influenced the test. Calibration curves can be found in Appendix A. This assay was found to give acceptable results when used for the basic and rapid determination of the presence or absence of chelant, and as such was acceptable for use in this study. However, if more accurate results are desired for the determination of actual chelant degradation, a chromatographic method is recommended. 3 . 5 S L U D G E K I N E T I C S M E A S U R E M E N T S Activated sludge kinetics were determined through respirometric methods (Cech, 1984) and modeled based upon the Monod relationship. Briefly, reactor mixed liquor samples were placed in a modified sealed and jacketed BOD flask (Figure 3.2). Injection ports allowed for rapid introduction of untreated effluent, and kinetics were determined through the measurement of the change in oxygen uptake rate after an injection of a known quantity of untreated effluent. Metal addition (Ca+ 2, Fe + 3, Zn + 2 , Cu + 2 , K + , Mg + 2 , Mn + 2) was accomplished through the addition of CaCl2 (Fisher Scientific, Fair Lawn, NJ), FeCl3, ZnS0 4 7H 2 0 (Anachemia, Toronto, ON), CuS0 4 5H 2 0 (Fisher Scientific, Fair Lawn, NJ), KC1 (Fisher Scientific, Fair Lawn, NJ), MgS0 4 7H 2 0 (Matheson, Coleman & Bell, Norwood, Ohio), MgCl 2 6H 2 0 44 (BDH, Toronto, ON) and MnS04- FL,0 (Matheson, Coleman & Bell, Norwood, Ohio). The respirometric measurements were made using either of: a 210 mL or a 230 mL jacketed glass vessel (Canadian Scientific, Richmond, BC), each with 4 ports and a neck conforming to the standard BOD bottle size neck. .The 4 ports were used as follows: 1) aeration; 2) overflow, using a funnel to extend the height of the glass port; 3) substrate injection through a rubber septum; and 4) available for extra measurements, such as the insertion of a pH probe, otherwise the fourth port remained sealed with a rubber septum (Figure 3.2). The DO probe was inserted vertically through the open "BOD bottle neck" of the respirometer. A water bath was used to ensure constant temperature during the experiments, as biomass activity is extremely sensitive to temperature. It was found in preliminary work that the oxygen uptake rate of the biomass could vary by 20% as a result of a 4 °C temperature fluctuation. OVERFLOW WATER OUTLET AIR STONE WATER INLET WATER BATH STIR BAR J Figure 3.2: Respirometer 45 Kinetic measurements were based upon the OUR response of the mixed liquor to known quantities of substrate. Unless otherwise noted, the substrate injected was undiluted, untreated combined mill effluent taken from the reservoir feeding the reactors from which the mixed liquor was taken. DTPA was added during aeration prior to injection of substrate. Samples were mixed for two minutes after DTPA addition. Chelant was added as a stock chelant solution. Stock chelant solutions, at a concentration of 33.3 g/L (either DTPA or EDTA) were made as follows: 3.33 g chelant was added to 50 mL distilled water. Sufficient IM NaOH was added for dissolution of the chelant, and the remaining volume was made up with R,0. The pH of the DTPA stock solutions was unregulated for storage, but adjusted using HC1 prior to use in experiments. Mixed liquor pH adjustment, using 0.1M HC1, and chelant addition were accomplished simultaneously upon introduction to the respirometer. Sample test results are described in Appendix B. 3 . 6 C E L L U L A R S O L U T E R E L E A S E Measurements of cellular solutes were performed based upon methods used by Wilkinson (1967) and Lieve (1965). Biomass to be used for the test was taken from the aeration tank of the activated sludge unit, centrifuged for 10 min at 1000 x g, and resuspended in 0.05M borate buffer. It was re-centrifuged for 10 min at 1000 x g, and resuspended in 0.05M borate buffer to give a final biomass concentration of between 4000 - 7000 mg/L. One mL concentrated biomass was added to 10 mL borate buffer, allowed to incubate at 20 °C for 1 hour, and centrifuged for 5 min at 2300 x g. Where applicable, DTPA or metals were added to the borate buffer prior to the addition of concentrated biomass. Samples were withdrawn carefully by micropipet from the supernatant, and absorbance at 260 nm was measured. Maximum solute release values, for comparison with sample values, were obtained by maintaining the biomass-buffer mixture at 100 °C for 10 min, 46 followed by centrifligation and supernatant removal. Verification that DTPA does not absorb at 260 nm was made using a scanning UV spectrophotometer. Other researchers have described the material released-from the cells as a mixture of phospholipids, proteins, and carbohydrates (comprised of significant portions of lipopolysaccharide), from the bacterial cell walls. No effort was made in this study to identify which components were present in the released material, and it is assumed, especially given the i mixed bacterial culture present in activated sludge, that a "general mixture" of the common polysaccharides involved in cell wall structures, including capsular material, was present. Verification that 260 nm was an appropriate wavelength to use was made by scanning spectroscopy (included in Appendix C). 47 RESULTS AND DISCUSSION 4. STEADY STATE REACTOR OPERATION A. I R E A C T O R O P E R A T I O N SUMMARY Prior to a discussion of the effects of DTPA and peroxide on activated sludge, a summary of normal reactor operation on both elementally chlorine free (ECF) and totally chlorine free (TCF) influents will be given. General reactor influent properties throughout the course of all continuous experiments are presented in table 4.1. Although every effort was made to supply a consistent influent to the continuous reactors, this was rarely possible, as regular changes to the wood furnish mixture at the mill occurred. In addition, storage of each batch of influent over an average of 6 weeks appeared to introduce further fluctuations in influent strength (BOD, COD) and toxicity over time. Included in this section for completeness is a collection of photographs taken during regular microscopic analysis of the biomass (figures 4.1 - 4.8). These figures illustrate the typical appearance of activated sludge floes when healthy (figure 4.1), and unhealthy (figure 4.2), and some of the non-bacterial life forms also present within the mixture (figures 4.3 -4.8). Healthy floes (figure 4.1) were identified as those with well defined bacterial agglomerates of moderate to heavy density, and surroundings which were relatively free of debris (comprised largely of unagglomerated cells). The onset of the appearance of unhealthy floes was often preceded by a transition floe form (figure 4.2). An initial sign of unhealthy floe formation was the gradual appearance of "spidery" and large floes. The spidery appearance was attributed to the preferential growth of Zoogloea bacteria types, associated with sludge bulking. The passage of time, if no corrective action was taken, would result in the formation of larger, more disperse floes, and an increase in the amount of free-floating bacterial cells in the surrounding media. Healthy floe formation was always accompanied by the 48 presence of higher life forms, such as protozoans, nematodes, and rotifers. Among the protozoans, the stalked ciliates (figures 4.3 and 4.4) are the most immediate and visible indicators of healthy biomass. These organisms anchor themselves to the floes, and feed on free-floating and dying bacterial cells. In a particularly healthy system, it is possible to support sizeable colonies of stalked ciliates (figure 4.4). Swimming ciliates of many types were also often observed (figure 4.5) moving freely through the media between the floes. Another common higher life form in activated sludge are the rotifers (figure 4.6). These organisms feed predominantly on dead biomass, and although a certain number of rotifers is considered healthy, a proliferation is indicative of an aging bacterial population which is unable to rejuvenate itself. These organisms are extremely mobile, and move through the mixed liquor in search of food. Also present in activated sludge, but to a much lesser degree, are nematodes (figure 4.7). These organisms also feed on the waste of the other organisms within the sludge. Over the course of all reactor runs, most common protozoans associated with activated sludge were observed. Occasionally, uncommon organisms were seen, which therefore often went unidentified. One such organism (figure 4.8) was observed in active pursuit of a rotifer. Approximately half of the organism is visible in this photograph, as it was very large -approximately 1 - 2 orders of magnitude larger than the rotifer, and visible to the naked eye. Evidence of the presence of a large predator-type organism is included in this collection of photographs to illustrate the seemingly complete nature of the ecosystem in activated sludge. It is to this comprehensive character of the microbial consortium that the robust nature of activated sludge, observed over the course of these studies, is attributed. 49 Figure 4.1: Typical appearance of healthy activated sludge floes Figure 4.2: Typical appearance of unhealthy activated sludge floes Figure 4.6: Activated sludge microorganisms: rotifers 52 Table 4.1: Reactor influent characteristics Reactor Run (#, title) Operation Day Influent Typeva) -wood furnish BOD Range (mg/L) COD Range (mg/L) Microtox® Range (IC50) 1: Constant DTPA •0-41 ECF - hem 353 - 404 - 12.9 42-92 ECF - hem/ced 410 - 450 - 7.7 - 15.8 93 - 133 ECF - hem 413 - 17.2 134 - 153 ECF - hem/ced 231 -328 - -2: Ramping DTPA 0-43 ECF - hem/ced 205 - 253 1085 22.3 44-90 ECF - hem/fir 236 - 400 1690 - 1710 35.9 91 - 143 ECF - hem/ced 314-514 2177 - 2226 -3: DTPA and calcium 0-53 ECF - hem/ced 280 - 441 1098 -54 - 59(b) 60% C102 -SPF/cedar 379 1834 -54-82 ECF - hem/ced 441 - 509 1920 - 2627 -4:Perox Ramp 0-260 TCF 87 - 400 346 - 1787 2-70 5: Constant Peroxide Concentration .0-41 ECF - hem/ced 280 - 475 1594 - 1949 13.9 - 29.8 42-98 60% C102 - hem 100 - 139 611-764 16.1 99 - 107 ECF - hem/ced 408 1663 -108 - 173 ECF - hem 455 - 659 2028-2231 10.1 6: TCF Steady State 0- 126 TCF 199 - 325 753 - 769 18.2 - 47.2 Summary by Reactor Influent Type K e y (a) : T C F = totally chlorine free combined mill effluent ••• E C F = elementally chlorine free bleach plant effluent hem = hemlock ced = cedar SPF = spruce, pine, & fir (b) : fed to the control reactor only data not available Influent Type BOD Range (mg/L) COD Range (mg/L) Microtox® Range (IC50) TCF 87 - 400 346 - 1787 2-70 ECF - hemlock 353 - 659 2028 - 2231 10.1 - 17.2 ECF - hem/fir 236 - 400 1690 - 1710 35.9 ECF - hem/cedar 231-514 1594 - 2627 7.7 - 29.8 60% C102 - hem 100 - 139 611-764 16.1 60% CIO? -SPF/cedar 379 1834 -54 • • 4 . 2 A C T I V A T E D S L U D G E O P E R A T I O N - T C F E F F L U E N T Continuous operation of two parallel activated sludge units treating totally chlorine free effluent yielded solids data as shown in figure 4.9. The data obtained from both reactors illustrates that although reactor fluctuations occur, they are, generally, reproducible fluctuations, likely due to unobserved variations in operating conditions. Suggestions for the causes of these variations are the changes in effluent properties throughout storage time and fluctuations in reactor ambient conditions, in spite of the presence of water jackets for temperature regulation. For example, reactor temperature fluctuations of 3°C were observed over 24 hour periods during hot summer months. It was also determined from the operation of these reactors that parallel activated sludge units can support distinctly different and varying levels of biomass under steady state operation. In these experiments (figure 4.9), the average mixed liquor volatile suspended solids was 1050 ± 120 mg/L in one reactor and 1416 + 162 mg/L in the other. Throughout all subsequent reactor runs, it was observed that in spite of all diligence to maintain as identical a pair of reactors as possible, reactor solids levels were consistently, and often substantially, different. This is assumed to be a result of the complex nature of the microbial community present in activated sludge. The myriad of unobserved interactions between the members of the microbial community is beyond the control of the activated sludge plant operator, and leads to the establishment of steady state at one of a number of possible operating states. It was observed throughout the course of this study that treatment efficiency was unaffected by the steady state at which the reactor had settled. Effluent solids levels (figure 4.9) during TCF operation remained at 47.5 ±11.1 mg/L and 49.4 ± 10.8 mg/L through most of the reactor mn, in the two parallel reactors. Occasional higher effluent solids measurements were a symptom of the fluctuations of indeterminate source occurring in reactor operation. Treatment efficiency, as monitored by BOD removal, remained above 90% in both 55 reactors, indicating the acceptable performance of activated sludge units operating with untreated TCF effluent as reactor influent; Average BOD removal was 97.7 ± 0.9%. This can be compared with standard treatment efficiencies of 80 - 90% for similar units operating with conventional influent (Springer, 1993), and agrees with previous findings of higher BOD removal efficiencies for TCF effluent (Saunamaki, 1995). TCF peroxide bleached effluents appear to be more amenable to biological treatment than conventionally bleached effluents. 56 2500 2000 "S, 1500-1 T~ 300 co co > 1000 500H - | — i — i — i — | — i — i — i — | — i — i — i — p 40 60 80 100 Time (days) Figure 4.9: Reactor solids concentrations in 2 parallel reactors treating TCF mill effluent (study #6): Reactor 1 MLVSS • , Reactor 2 MLVSS • . Reactor 1 effluent VSS • , Reactor 2 effluent VSS O 4000-a • i i i i i i i •250 h-200 -150 E, co co > •100 § E U J •50 1—1—j—I—I—I—|—I—I—1—j—1—1—I—|—I—I—I—j-0 20 40 60 80 100 120 140 160 Time (days) Figure 4.10: Reactor solids concentration in a reactor treating ECF mill effluent (Control reactor, study #1): MLVSS • , Effluent VSS • , Switch in reactor effluent type-(table 4.1). 57 4 . 3 A C T I V A T E D S L U D G E O P E R A T I O N - E C F E F F L U E N T Throughout three of the continuous reactor studies performed using ECF effluent, control reactors were maintained (for each run) which provided information on ECF activated sludge operation. Operation of one activated sludge unit (study #1 control) treating ECF effluent yielded solids data as shown in figure 4.10. Reactor solids levels climbed inexplicably and then dropped after switching to a new effluent supply, although the BOD of the new effluent did not differ from the older effluent. It is suggested that the change in reactor influent supply, which included a change in wood furnish, provided the microorganisms with an altered environment which may have resulted in the generation of greater amounts of biomass. Reactor solids levels returned to previous values after switching to a third batch of effluent whose wood furnish was the same as that of the first batch. BOD removal remained acceptable throughout the reactor run, at greater than 90%. Average BOD removal was 94.3% ± 1.67% which was statistically lower than that obtained with TCF effluent, indicating that a smaller portion of effluent BOD from an ECF bleaching sequence than from a TCF bleaching sequence is degradable in an activated sludge unit. Figure 4.11, plotted with data from a separate run, (study #2 control) clearly illustrates the general response of reactor solids levels to the increasing level of influent BOD. Solids levels, after decreasing initially from start-up values, climbed steadily in response to a general increase in feed BOD. The greater concentration of available food in the reactor influent was used by the biomass to support a larger concentration of microorganisms. As available energy increased, the biomass solids concentration increased towards higher steady state values. Figure 4.12, plotted with data from a third run, (study #3 control) is included to show that it is possible to run an activated sludge unit under stable conditions when treating ECF effluent. Average solids concentration was 1900 + 92 mg/L. Average reactor biomass yield, 58 which is most accurately calculated on a reactor under stable operation, was 0.5 ± 0.05 mg MLVSS/mg BOD consumed for this run. This lies within the range of normal yields expected in activated sludge units (Orhon, 1994). 59 Figure 4.11: 600 h-500 400 300 E. Q O m 200 100 I i i i I '' ' ' I ' 0 20 40 60 80 100 120 140 160 Time (days) Reactor solids concentrations in a reactor treating ECF mill effluent (Control reactor, study #2): Influent BOD • , MLVSS , Switch in reactor influent type (table 4.1). I 1 1 1 1 l 1 1 1 ' l 1 1 1 1 I 10 20 30 40 50 Time (days) 70 Figure 4.12: Reactor solids concentration in a reactor treating ECF mill effluent (Control reactor, study #3). 60 4 . 4 A C T I V A T E D S L U D G E O P E R A T I O N - T C F / C O N V E N T I O N A L I N F L U E N T S W I T C H A marked decrease in COD reduction was observed upon switching from TCF bleached mill effluent to conventionally bleached (60% C109 substitution) mill effluent (figure 4.13). This effect was reversed immediately upon switching the reactor influent back to TCF mill effluent. This indicates that a greater portion of TCF effluent is biodegradable. This can be confirmed through the calculation of the COD:BOD ratio for each untreated effluent. The COD:BOD ratio for TCF effluent averaged 3.2 + 0.4, whereas the ratio for conventionally bleached effluent averaged 5.1 ± 1.4. Eliminating the use of chlorine therefore results in effluent which has a larger proportion of biodegradable material. This is in agreement with the higher BOD reduction values obtained during the treatment of TCF effluent, as discussed in Section 4.1. Acceptable effluent treatment was not adversely affected by switching the influent source from TCF to conventional and back to TCF as measured by toxicity and BOD reductions, which remained at 100 % and over 95 % respectively. 61 60-50 H £ 4 0 -c o t i -g30-<D DC Q § 2 0 -10H Conventional TCF TCF "1 " f" T I"1 I ! ' 1 'J 'I "I !' | 'I 'I" 1 I |' ' 1" T I 1 ] ]—T-* 0 20' 40 60 80 100 120 Time (days) Figure 4.13: COD reduction during feed switch between TCF and conventionally-bleached reactor influent. 100-90-80-- 70-.1 6 0 " o 73 50-40-30-20-10-0 ECF Conventional - o t o -ECF ' ' 1 I 1 1 1 I ' 1 1 1 I 1 1 1 I 1 1 ' I • • 1 I 1 ' 0 20 40 60 80 100 120 140 Time (days) Figure 4.14: BOD reduction during feed switch between ECF and conventionally-bleached reactor influent. 62 4 . 5 A C T I V A T E D S L U D G E O P E R A T I O N - E C F / C O N V E N T I O N A L I N F L U E N T S W I T C H Switching from ECF to conventionally bleached (60% CIO,,) reactor influent appeared to have a beneficial effect on activated sludge operation, resulting in a slight improvement in BOD removal efficiency (figure 4.14). This was confirmed by a subsequent slight decrease in BOD removal efficiency upon reintroduction of ECF mill effluent after steady state operation had been achieved with conventionally bleached effluent. BOD removal prior to conventional mill effluent introduction was 89.2 ± 3%. While treating conventional effluent, BOD removal averaged 96.5 + 0.8%. This implies that a greater portion of the BOD in conventional effluent than the BOD in ECF effluent is treatable by activated sludge. Activated sludge quality, as observed under a microscope, also improved after the introduction of conventionally bleached effluent. Hoes had previously exhibited a severe filament problem, and significant amounts of debris (non-aggregating cells) were visible in the surrounding media (figure 4.15). The most marked difference upon switching reactor influent was a reduction in the quantity of filaments and debris (figure 4.16). A proliferation of stalked ciliates, an indicator of good reactor health was also observed after the introduction of conventionally bleached effluent (previously shown in figure 4.4). 63 Figure 4.15: Activated sludge floes, ECF bleached reactor influent 4 . 6 S U M M A R Y O F T H E E F F E C T S O F R E A C T O R I N F L U E N T T Y P E O N A C T I V A T E D S L U D G E T R E A T M E N T Average COD:BOD ratios and BOD removal over the course of these studies is presented for ease of comparison in table 4.2. The COD:BOD ratio for the ECF mill effluent used in all ECF studies averaged 4.2 + 0.4, and can therefore be seen to lie between the values for TCF and conventional mill effluent, at 3.2 ± 0.4 and 5.1 ± 1.4 respectively. It would appear, based upon these averages, that with the decrease in the use of chlorine compounds in pulp bleaching, the biodegradable fraction of the wastewater increases. However, it can also be observed that the treatment efficiencies, as measured by BOD removal, do not follow the same relationship. Although the BOD fraction of totally chlorine free mill effluent was the most successfully treated by activated sludge (section 4.2), conventionally bleached mill effluent was found to be the next most successfully treated, followed by elementally chlorine free mill effluent (table 4.2). This is likely a simple indication of the different distributions of partially degraded lignin and cellulose compounds which emanate from each of the bleaching sequences. Previous work has indicated that the selectivity for lignin over cellulose during delignification varies with the delignification chemicals used as: D > C > O (Reeve, 1992). Equated with the terminology used here to define mill effluent types, this translates to: ECF> Conventional > TCF, which corresponds to the order of increasing biological treatability observed in these studies. This suggests that greater selectivity for lignin degradation over cellulose may result in the release of a biodegradable fraction of wastewater which is slightly more difficult to treat in an activated sludge secondary treatment system. 65 Table 4.2: Effect of reactor influent type on average COD:BOD ratios and BOD removal rates Reactor Influent Type COD:BOD % BOD Reduction TCF 3.2+0.4 97.7+0.9 ECF 4.2+0.4 94.3+1.7 Conventional 5.1±1.4 96.5+0.8 66 5 . BATCH TESTS WITH DTPA AND METALS 5. I E F F E C T O F . D T P A ON C O M B I N E D K R A F T B L E A C H E F F L U E N T C H A R A C T E R I S T I C S ' DTPA affects bleach plant effluent by contributing significantly to the chemical oxygen demand. COD is raised by approximately 790 mg/L per g/L DTPA (figure 5.1). This compares with the theoretical oxygen demand value for DTPA ( C , 4 H 7 6 N 3 O 1 0 ) of 1080 mg/L, with the difference between the two numbers attributed to incomplete oxidation during the COD reaction. These numbers are considered to be in good agreement, as a difference of approximately 20% between actual and theoretical oxygen demand values was anticipated in COD tests. With a typical DTPA concentration of 875 mg/L, an increase of 690 mg/L in the COD of the bleach plant effluent would result in a substantial increase in COD input to the activated sludge reactor downstream from the bleach plant. As previously seen in table 4.1, the bleach plant COD entering treatment is highly variable, but usually in the range of 1000 - 2000 mg/L. COD contributions from DTPA would therefore also be highly variable, at approximately 25 - 40 % of the total COD input from the bleach plant. Assuming DTPA to be as stable during biological treatment as other researchers (Saunamaki, 1995; Means, 1980) have found it and similar chelants to be, the higher COD input implies a decreased efficiency of COD removal during treatment. DTPA entering an activated sludge treatment system will pass through into the effluent and into the receiving waters. Although the chelant is chemically oxidizable, attempts at measuring its potential for biological oxidation using the BOD test appeared to indicate that DTPA is recalcitrant to microbiological treatment (table 5.1), indicating its potential for release into the environment. Effluent BOD was measured both with and without DTPA in order to gauge the contribution of the chelant to the BOD. For ease of comparison, the mill effluent BOD with DTPA was divided by the mill effluent BOD without DTPA and given as a percentage. The contribution of DTPA to the BOD measurements is therefore represented numerically as a deviation from 67 100%, in the third column of table 5.1. At concentrations typically encountered in TCF bleach effluent (up to approximately 1 g/L), DTPA does not contribute significantly to the BOD of elementally chlorine free bleach plant effluent. This is illustrated in figure 5.2, in which normal BOD test eiTor determined over the course of these studies is indicated by the parallel dotted lines. The majority of the variation in BOD readings between samples with and without DTPA falls within the normal error of the test. At concentrations up to those found in bleach plant effluent and over the course of a 5-day BOD test, DTPA is not inhibitory to the seed bacteria used in the test. At higher concentrations, however ( > 15 g/L), DTPA can completely inhibit the BOD test (table 5.1). BOD test inhibition is likely accomplished through the prevention or interruption of the respiratory processes of the test's seed bacteria. Although the higher concentrations are unrealistic with respect to the environment in an activated sludge unit, these results are nevertheless indicative of the potentially toxic nature of the chelant to biomass originating from an activated sludge unit. 68 1000-900-^ 800H 7 0 0 -5" 600-3 CT) & 500-] Q 8 4 0 0 ^ 300^ 200-j 100^ 0 i i i I i i i I i i i I i i i I i i 0.2 0.4 0.6 0.8 DTPA (g/L) T-|—I I I 1.2 1 Figure 5.1: Effect of DTPA on COD measurement. CT) E, I -Q 3 £ 2 2 c e a> 3= 60 40 20 0 -20H -40 - 6 0 -- 8 0 -100 111111111111111111111111111111111111111111 M 11 M 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 DTPA (g/L) Figure 5.2: Effect of DTPA on BOD measurement: DTPA in effluent • , normal BOD test accuracy range Table 5.1: Effect of DTPA on BOD measurements DTPA BOD BOD with DTPA BOD effluent , Finn x 1 0 0 % D K J L J effluent A DTPA 0 290.2 290.2 ioo 0.2 252 253.8 100.7 0.2 250.2 245.4 98.1 0.4 332.4 378.6 113.9 0.4 356.4 319.2 89.6 0.4 357.6 341.4 95.5 0.4 399.6 399.6 100 0.4 352.8 370.8 105.1 0.6 372.6 346.2 92.9 0.6 405 462 114.1 0.8 505 495 98 1 462 381 82.5 1 . 555 548 98.7 15 290.2 0 0 30 290.2 0 0 52.5 290.2 0 0 The sensitivity to DTPA of a species used in standard toxicity tests was also measured. At typical bleach plant effluent concentrations DTPA was also found to exhibit no toxicity to the marine bacteria Vibrio fisheri (table 5.2), both by an insignificant increase in effluent toxicity, and a "non-toxic" toxicity measurement in water. The results of this test imply negligible acute effects of the release of DTPA in to the environment, but give little information on the potential toxic effects of the chelant on activated sludge bacteria. With respect to the operation of an activated sludge unit receiving influent DTPA, the toxicity load to the biomass would be unchanged. 70 Table 5.2: Microtox® toxicity of DTPA Sample IC50 95% confidence interval Effluent 18.6 17.6 - 19.6 Effluent & 0.875 g/L DTPA 22 18.2 - 26.6 Water & 0.875 g/L DTPA 100 n/a 5 . 2 E F F E C T O F D T P A O N E N D O G E N O U S R E S P I R A T I O N O F A C T I V A T E D S L U D G E Addition of DTPA to activated sludge affected the biomass in numerous measurable ways, including a marked and immediate, but inconsistent, effect on the endogenous respiration rate of the biomass. DTPA addition at concentrations over 0.25 mg/L appeared to result in an increase in endogenous respiration by as much as 65%. In tests where repeated additions of DTPA were made, as in the respirometry tests, the endogenous respiration rate remained elevated. If, however, the biomass was simply left aerating after being exposed to a single dose of DTPA, the endogenous respiration rate gradually returned to its original value. After the addition of 0.25 g/L DTPA, approximately one hour was required for the' endogenous respiration rate to return to its original value (figure 5.3). Other researchers (Lenta, 1960; Koch, 1955) who have also observed increased respiration rates in the presence of a chelant did not comment on the persistence of the phenomenon. Changing endogenous respiration rates were also found to play a significant role in the interpretation of the data obtained by respirometry, as illustrated in figure 5.4. Repealed substrate injections were administered, and all respirometric responses to substrate injection were recorded: 1) endogenous OUR, and 2) maximum OUR. The difference between the endogenous and maximum OUR, R, was calculated from these values. The endogenous OUR exhibited a response similar to that observed in figure 5.3, although with repeated substrate injections, the time required to return to pre-test endogenous respiration rates was increased. The maximum OUR obtained upon substrate injection remained constant over a period of 200 minutes, and appeared to be unaffected by the change in endogenous OUR. However, the calculation of R, the increase in oxygen uptake rate attributable to the addition of substrate, was directly dependant upon which endogenous OUR was used in its calculation (Appendix B). As a result of the changing actual endogenous OUR, R decreased initially and subsequently increased. If, however, R was calculated based upon the original endogenous OUR, an 72 0.35 DC 3 o 0.1-co co 0.05-0 20 40 60 80 100 120 Time (min) Figure 5.3: Effect of 0.25 g/L DTPA on endogenous OUR. Injected substrate was ECF mill effluent. Figure 5.4: Effect of 0.25 g/L DTPA on respirometric measurements: Endogenous OUR • , Change in OUR upon substrate injection, R • , Adjusted R • , and Maximum OUR • . Injected substrate was ECF mill effluent. "adjusted was obtained, and also plotted in figure 5.4. With respect to further data analysis, the question of whether R or the adjusted R is the correct data must be answered. However, this cannot be answered without first determining whether or not the true endogenous respiration rate has changed under conditions of elevated DTPA concentration. For each individual injection during a respirometric test, the measurement of the change in OUR (R) in response to the injection of known amounts of substrate is made. An elevated endogenous OUR would result in the calculation of a decreased value of R (figure 5.5). For example, the actual value of R calculated for a sample containing 0.25 g DTPA/L was 0.65 mg/L min, whereas R calculated based on the original endogenous OUR was 0.95 mg/L min, a difference of approximately 30%. If the true endogenous respiration rate of the biomass had remained unchanged, this implies that the basal metabolism was being masked by an additional respiratory function, which may or may not require subtraction from the observed response. Ideally, if the endogenous respiration rate had in fact changed over the course of the test, then this would be taken into account during the analysis of the data. Careful analysis of all respirometric data obtained over the course of this study revealed no incontrovertible evidence that the true endogenous respiration rate had been altered by the presence of DTPA. Instead, the data, on average, indicated that more consistent results were obtained if the effects of DTPA on endogenous OUR were ignored for the purposes of data analysis. Therefore, calculations in this study are based upon the measured endogenous respiration rates for each test, and did not involve corrections based on changes in endogenous respiration rates. Care was taken, however, to avoid drawing conclusions which were wholly dependant upon endogenous respiration rate values. It should also be pointed out that an increased endogenous respiration rate in response to DTPA addition was not observed throughout the course of this project. The presence of elevated endogenous respiration varied from day to day and, in fact, from year to year. The effect was never observed while using TCF mill effluent, likely due to the appreciable 74 concentrations of DTPA already present in the effluent. The inconsistent nature of this response makes it difficult to determine its cause, meaning and the appropriate manipulation of the data. Detailed examination has shown that, when the endogenous respiration increased, it did so for all concentrations of DTPA, from 0.25 g/L to 2 g/L and above, and that the new endogenous respiration rate remained approximately constant at all chelant concentrations. The question raised by an increase in endogenous respiration rate pertains to the origin and nature of the substrate being metabolised. Endogenous respiration is a complex and little understood aspect of cellular metabolism. Normally, the energy expended during endogenous metabolism is thought to be used in the maintenance of a specific internal cellular environment, and therefore involves, for example, the regulation of cell wall ion gradients. The energy for normal endogenous respiration is derived from storage compounds within the cells, some of which are themselves polysaccharide components, similar to some cell wall components (Umbreit, 1976). It is proposed, therefore, that endogenous respiration may increase for one (or both) of two main reasons:' 1) the cell responds to a need for increased maintenance and 2) the storage compounds which were previously metabolised under endogenous respiration become more readily available for consumption by the cell. It is also suggested that a contributing factor to the increase in endogenous respiration may be the physical availability of oxidizable substrate to the individual cells. Chelants have been shown to deflocculate activated sludge (Silverstein, 1994; Eriksson, 1991; Kakii, 1986a), resulting in a mixture of bacterial cells which would experience less mass transfer resistance of the substrate through the floe. 75 1.9-1.7-1.5-1.3H c 'E «d E o 0.9H 0.74 0.5 w i t h 0.25 g / L D T P A C o n t r o l Base O U R | I I | I I M ) l l l l | l l l l | l l l l | I l I I | n i : | i l l l | l l l i 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Time (min) Figure 5.5: OUR traces upon injection of substrate: with DTPA , and without DTPA-76 5.3 E F F E C T O F D T P A O N F O A M I N G D U R I N G B A T C H T E S T S The addition of DTPA to aerated mixed liquor samples can cause an immediate physical response, dependant on the concentration of DTPA present in the mixture. As the DTPA concentration was increased during batch respirometry tests, foaming was observed, to a degree markedly more severe than previously observed in the same sample with lesser concentrations of DTPA present. This would seem to be indicative of an increase in the surfactant concentration of the mixture. There appeared to be a threshold DTPA concentration value, below which excess foaming was not induced, and above which foaming increased dramatically. The exact level of this threshold varied from experiment to experiment, although foaming was usually first observed as the DTPA concentration was increased to 0.75 or 1 g/L. Foaming often became so severe that normal aeration, necessary between tests to elevate the dissolved oxygen concentration of the mixture, was not possible due to the carryover of biomass. Aeration rates were decreased, and biomass lost in the foam was recovered through mechanical disruption of the foam. Although surfactants are present in pulp mill wastewaters, particularly from the kraft process, in the lignin and wood resin fractions (Allen, 1993), these compounds are unlikely to undergo a reaction over the course of a batch test which would cause a sudden increase in the surface active nature of the mixture. It is hypothesized instead that components released from the bacterial cells through the action of the DTPA on the cell walls (section 5.9) are a direct cause of the foaming due to the surfactant nature of some of these components. This is discussed in greater detail in section 5.9. Additional surface active material may also be released from the extracellular components of the cell wall which may be released upon chelant-induced deflocculation (Silverstein, 1994; Eriksson, 1991; Kakii, 1986a). 77 5 . 4 E F F E C T OF D T P A ON GROWTH OF ACTIVATED S L U D G E Anticipated long-term effects of a chelant upon activated sludge can be determined by examining the effect on sludge growth rates. The addition of DTPA to untreated ECF effluent had a strong effect on the growth of activated sludge biomass in the effluent (figure 5.6) in shake flask experiments. The lag time, before the onset of microbial growth, increased noticeably with increasing DTPA concentrations, and both growth rate and overall growth yield decreased. This is comparable with the results of a study performed using EDTA, which was found, at a concentration of 1 mM (0.37 g/L), to completely inhibit the growth of bacteria isolated from activated sludge (Kakii, 1986b). The increase in lag time (figure 5.6) indicates an interruption of normal metabolic processes, while the decrease in overall growth and growth rate may indicate a decreased ability of the biomass to utilize the substrate present. Although residual BOD was not measured in this test, continuous reactor studies (section 6) verified that substrate removal was significantly reduced in the presence of DTPA. These impacts on substrate metabolism and microbial reproduction could have a profound effect on the performance and stability of continuous activated sludge reactors. Although growth was measured by oxygen uptake rate, due to the inaccuracy of optical cell density measurements in the presence of floes, actual overall cell growth was confirmed by solids determination (figure 5.7). Growth rates at varying DTPA concentrations are plotted in figure 5.8. A relatively low concentration of DTPA, 0.2 g/L, resulted in a significant decrease in overall growth rate. In addition, it is evident that within the range of DTPA concentrations tested, a minimum overall growth rate between 20 and 30% of the original was reached as the chelant concentration increased. The presence of calcium added to the growth media prior to DTPA and biomass inoculation, in excess (a 4.8:1 molar ratio) of the amount required to bind to the chelant, mitigated the effect of the chelant on the growth rate. However, the negative effect was not 78 eliminated, indicating that the presence of the chelant was still able to affect the biomass when bound strongly to a metal. This may provide evidence for the preference of DTPA to chelate cell-wall calcium over free calcium, of, may indicate that additional metals required for cell growth were sequestered by the DTPA. The evidence for the latter explanation would only be manifest during the longer-term growth tests, and not during short-term respirometric tests, where calcium was found to completely mitigate the effects of the chelant (section 5.6). Microbial growth, which is dependant upon the overall health of the cells, was more sensitive to the presence of DTPA. This occurred in spite of the excess calcium, as additional metals, normally used by the cells as trace nutrients, were made unavailable. / 79 0 500 1000 1500 2000 Time (min) Figure 5.6: Effect of DPTA on growth of activated sludge: Control • , 0.5 g/L DTPA • , 0.875 g/L DTPA • , 1.25 g/L DTPA A. 8-1 | 1 7H o -S -x 6 -_c -_ • 00 -« ^ c o k_ c CO c 3 -o O Ui o _ o co : 1-U 0 0.2 0.4 0.6 0.8 1 1.2 1.4 DTPA (g/L) Figure 5.7: Effect of DTPA on growth as measured by solids concentration: Biomass • , Biomass with metals • . 80 1-* 0 0.2 0.4 0.6 0.8 1 1.2 1.4 DTPA (g/L) Figure 5.8: Effect of DTPA and excess calcium on growth rate: Growth rate • , Growth rate with calcium (calcium:DTPA molar ratio = 4.8:1) • . 5 . 5 E F F E C T O F T H E C H E L A N T D O S E ON O X Y G E N U P T A K E RATE The determination of the immediate effect which a chelant exerts on a microbial system was accomplished by the examination of chelant effects on biomass activity. In contrast to the long-term effects which may be gauged by the measurement of growth rates and overall growth, the immediate consequences of chelation are best measured by the influence of the chelant on the metabolic activity of the biomass, as measured by respiration. The addition of DTPA to biomass treating TCF effluent clearly illustrates the detrimental effect which the chelant has on the metabolic activity of the biomass (figure 5.9). Increasing concentrations of DTPA resulted in increasingly severe reductions in the respiration rate, and therefore in the substrate uptake rate, of biomass samples. Each data set in figure 5.9 has been fitted to the Monod curve as described in Section 2.4: R R S ™*(KS + S) The data in figure 5.9 were summarized by plotting the kinetic constants Rmax and Ks (figure 5.10). This representation more clearly illustrates the chelant effects by only plotting the effect at the maximum substrate uptake rate, as a function of chelant concentration. This simplified method was used for the presentation of the majority of the data in this and subsequent sections. Wherever appropriate, for the comparison of data from separate experiments performed on different biomass, data were normalized and presented as a ratio (i.e. either WRmnr o r ^ m / , y ^ m / , r r>) ^ is illustrated in the figures 5.11 and 5.12. The use of ECF effluent in the same tests shifted the Rmax curve to the right, as shown in figure 5.11. The maximum oxygen uptake rate of the biomass was not negatively affected by concentrations of DTPA less than approximately 0.5 g/L in ECF (DTPA-free) bleach plant effluent. The use of TCF effluent, in contrast, which originally contained DTPA, resulted in a decrease in oxygen uptake rate at chelant concentrations less than 0.5 g/L. This indicates the presence of a 82 0.6-0.5H Control 0.25 g / L D T P A 2 3 4 5 6 Substrate Concentration (mg BOD/L) Figure 5.9: Effect of DTPA on oxygen uptake rate upon substrate injection: DTPA Concentration (g/L): 0 • , 0.25 • , 0.5 • , 0.875 1.25 • . Lines are non-linear curve fits to the Monod equation. - T — i — i | i—i—p-|—r-i—i | i i i |—i—i—i—p 0.2 0.4 0.6 0.8 1 1.2 1.4 DTPA concentration (g/L) Figure 5.10: Effect of DTPA on Monod parameters: R m a x •,!<••. 83 threshold value of chelant concentration, below which the kinetics of the biomass was unaffected. These experiments determined this value to lie in the range of 0.5 to 0.75 g/L. It is evident that the DTPA concentration in TCF effluent meets or exceeds the threshold value. The response of Rmax to the addition of EDTA was determined to be similar in nature to that of DTPA (figure 5.11). The general shape of the curves is the same, with a steepening of the EDTA curve in TCF effluent. At higher chelant concentrations, therefore, the biomass appears to be more sensitive to further addition of EDTA than DTPA. However, the overall effect of either chelant at concentrations greater than 1 g/L appears to be that of reducing the maximum metabolic rate of the biomass to 20 - 30 % of its original value. Explanations for the effect that the chelants have upon the biomass fall into two main categories: 1) the activity of all or most of the microorganisms is reduced, on average, by 70 - 80 %, and 2) the chelants strongly affect 70 - 80 % of the biomass, however the remaining 20 - 30% of the biomass is resistant to the damaging effects. There is insufficient evidence to determine which explanation is correct. In either case, a contributing factor is likely to be the deflocculation observed (Silverstein, 1994; Eriksson, 1991; Kakii, 1986a) in the presence of the chelant, thereby increasing the exposure of the cells to the chelant. Within the range of chelant concentrations which may be expected during TCF bleaching operation (up to approximately 1 g/L in bleach plant effluent), it appears as though a maximum negative effect is reached at the higher concentrations. The implication of a non-zero minimum metabolic activity is that the chelants are not killing all the microorganisms. At worst, a fraction of the microorganisms may be killed, but not all. Equally likely is that the metabolic activity of the microorganisms is being severely hindered by the presence of the chelant, but that this does not result in the death of the microorganisms. Rather, the perceived loss of activity may be due to a redirection of cellular functions towards the reparation of damage being done to the cells by the chelants. Additional evidence for a redirection of cellular functions, manifested as an increased endogenous respiration rate, was discussed in section 5.2. 84 The length of time for which the biomass was in contact with the chelant did not appreciably alter the threshold chelant concentration, below which there was a minimal negative effect on biomass activity (figure 5.12). Biomass in contact with 0.5 g/L DTPA lost only 20% of its activity over a three hour period, in comparison with a 70% loss of activity observed using biomass in contact with 0.875 and 1.25 g/L DTPA. The full effect of the DTPA at higher concentrations is evident after 25 - 30 minutes exposure to the chelant during this test. This was the longest such lag time observed; it was consistently found that the normal lag time was no more than a few minutes. These results confirm that, after the time required for an initial response to the chelant, the negative effects of the chelant on oxygen uptake rate are unlikely to worsen. 85 1 I ' 1 1 I 1 0.2 0.4 0.6 0.8 1 Chelant (g/L) 1.2 1.4 Figure 5.11: Effect of DTPA and EDTA on OUR upon substrate injection of untreated TCF and ECF effluent: DTPA, TCF • , EDTA, TCF • , EDTA, ECF O , DTPA, ECF • , DTPA, ECF replicate (different effluent batch) O. 1-0.9-0.8-0.7-°- 0.6-£ C£ 0.5-X 18 r£E 0.4-0.3-0.2-0.1-0-• • i • • • A * • i i i I i i i I i i i I i i i I i i i I i i i I i i | ' ' , | | , , , | 0 20 40 60 80 100 120 140 160 180 Time (min) Figure 5.12: Effect of time on maximum oxygen uptake rate of IDT PA - affected biomass: DTPA Concentration (g/L): 0.5 • , 0.875 • , 1.25 A . 86 5 . 6 E F F E C T O F M E T A L S A N D T H E D E G R E E O F C H E L A T I O N O N O X Y G E N U P T A K E R A T E Given the affinity of chelants for sequestering metal ions, it is reasonable to assume that this property played a significant role in the negative effects that DTPA was observed to have on activated sludge biomass. A number of different metals, with differing stability constants with DTPA, were added in conjunction with the chelant to determine the extent to which the chelant's ability to sequester metals was responsible for the negative effects. The OUR was not decreased by DTPA when the active chelation sites were occupied by a cation with a high affinity for DTPA, such as calcium (figure 5.13). Almost complete binding of DTPA molecules to calcium ions is likely given the relatively high stability constant of calcium with DTPA (10.75, table 2.1) and the stoichiometric excess of calcium (4.6 moles Ca+2/mole DTPA). This clearly indicates that it is the compound's action as a chelant which is responsible for previously observed detrimental effects, since altering its ability to chelate additional metals strongly alters its effect upon the respiratory response of the biomass. The presence of the calcium in conjunction with the DTPA also appeared to enhance the ability of the biomass to metabolise substrate at substrate concentrations greater than the approximate original K value. This resulted in an increase in the maximum oxygen uptake rate, Rmax, by 17 %. The half saturation constant, which remained the same in this experiment for the control biomass and the biomass with 1 g/L DTPA concentration, increased by 40% with the addition of both calcium and DTPA. The use of metals other than calcium resulted in a differing oxygen uptake response by the biomass (figure 5.14). Iron and potassium, when added to the biomass prior to DTPA introduction, illustrate the difference between the use of metals with widely differing stability constants with DTPA. Iron, with a stability constant with DTPA greater than that of calcium (28.0, table 2.1), is also able to mitigate the negative effects of the chelant. Potassium, with a very low stability constant value (approximately 1, table 2.1), does not afford the biomass 87 1.4H 0 2 4 6 8 10 12 Substrate (mg BOD/L) Figure 5.13: Effect of excess calcium on biomass kinetics: Control A , 1 g/L DTPA • , Calcium & 1 g/L DTPA (molar ratio Ca:DTPA = 4.6: i) • ; Half saturation constants: Control and Calcium & 1 g/L DTPA samples - - \ 1 g/L DTPA I 1 1 1 I 1 1 1 l 4 6 8 Substrate (mg BOD/L) Figure 5.14: Effect of iron and potassium on biomass kinetics: FeCl3 concentration (g/L): 0 I 1.3 • , KC1 concentration (g/L): 0 A , 1.3 A. DTPA (1 g/L) was present in . samples • and A. 88 protection against the chelant. These results further support the proposition that the chelating ability of the chelant is the characteristic which controls whether or not the activity of the biomass is negatively affected by the presence of the compound. The detrimental effects of DTPA on oxygen uptake rate were apparent within 2 minutes and appeared to grow somewhat more severe over periods of time which were much longer than would be experienced during a typical test of oxygen uptake rate (figure 5.15). The control biomass lost a maximum of approximately 20% of its activity over a period of two hours, whereas biomass in the presence of 1 g/L DTPA lost 70% of its activity over the same period. A maximum loss of 80% of activity was apparent after four hours. Calcium, at a concentration of 1.3 g/L, added to the mixture after the biomass appeared to have reached a maximum negative effect with 1 g/L DTPA was able to partially mitigate the negative effect, in spite of the extended period of time for which the biomass had been in contact with the chelant. The presence of calcium alone resulted in a response which was, in general, the same as that of the control biomass. The addition of both calcium and DTPA resulted in an enhancement of the oxygen uptake rate, similar to that observed in figure 5.13, which remained at or above the initial values for the duration of the test. Of most significance was that the oxygen uptake rate remained approximately 15 - 25% higher than both the control and the sample with calcium alone, confirming that the activity enhancement with DTPA and calcium exists, and is of a permanent, rather than a temporary, nature. 89 rr tr 111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 M 1111111 1 1 1 1 1 1 1 1 i 0 50 100 150 200 250 300 350 400 Time (min) Figure 5.15: Effect of time on oxygen uptake rate in the presence of DTPA and calcium: Control • , 11.7 mM CaCl2 • , 2.5 mM DTPA A , 2.5 mM DTPA & 11.7 mM CaCl2 • . i r Control 1 g/L DTPA & 1 g/L EDTA & Ca (5.2:1 molar Ca (4.9:1 molar ratio) ratio) Figure 5.16: Difference between the addition of DTPA and EDTA on the protective effect of calcium. 90 In order to further examine the activity enhancement observed when DTPA was added to activated sludge biomass in conjunction with calcium, further tests were conducted using other substrates. Data previously discussed in this section were obtained using untreated mill effluent as the substrate. The enhancement of the rate of substrate metabolism was also observed when the test was performed with single substrates such as methanol or formic acid (table 5.3). This implies that the enhanced oxygen uptake rate is a symptom of a general, non substrate-specific effect which the chelant and calcium have upon bacterial cells. It will be seen in section 6.3 that the continuous addition of DTPA and calcium together to activated sludge resulted in increased permeability of cell membranes. It is therefore proposed that the activity enhancement observed only with the combination of DTPA and calcium was a result of the reduced resistance to the transfer of substrate, oxygen, or both across cell membranes. The negative effect on biomass activity caused by DTPA was completely mitigated by the calcium, and the only remaining effect on the cell was the increase in permeability. Table 5.3: Effect of calcium and DTPA on R in differing substrates Substrate Rmax p r i o r t o a d d i t i o n o f c h e l a n t a n d m e t a l R a f t e r 1.42 g/L CaCl2 a n d 1 g/L DTPA Increase in Rmax m i l l e f f l u e n t 1.121 1.313 17% m e t h a n o l 0.220 0.251 14% f o r m i c a c i d 0.274 0.296 8% Although DTPA and calcium were able, in combination, to increase the maximum metabolic activity of the biomass in mill effluent, the same result was not observed with EDTA, as shown in figure 5.16. This clearly illustrates that EDTA and DTPA, often thought to be equally effective with respect to their performance in pulp bleaching sequences, have distinctly different effects on activated sludge biomass. Also observed during this experiment was the difference in foaming exhibited by the different samples. The control and DTPA/Ca samples 91 did not foam excessively. The third sample, with EDTA/Ca, began to foam immediately upon introduction of the chelant and metal, indicating a significant change in the system and possible damage to the biomass. It is clear that the sample with DTPA and calcium, which exhibited a 10% increase in activity, was not damaged to an extent which would cause the release of cell wall components hypothesized to be responsible for foaming (sections 5.3 and 5.9). However, the sample with EDTA & calcium was likely to have suffered this damage, in addition to a decrease of approximately 48% in biomass activity. This further suggests a link between the decrease in the ability of the biomass to metabolize substrate and the release of surface active material into the media. Examination in greater detail of the effect of a fixed concentration of calcium at varying DTPA concentrations yielded the data in figure 5.17. This set of experiments was performed using a single substrate, methanol, to eliminate variation which may have been the result of the multiple substrate nature of untreated mill effluent. The three components which comprise kinetic measurements are included in figure 5.17: the endogenous OUR, the maximum OUR after substrate injection, and the difference between the two, Rmax- The effect of DTPA on the endogenous OUR in calcium-containing media was similar to that of the control biomass (not containing calcium). Both exhibited the increase in endogenous OUR discussed in section 5.2, but the control biomass exhibited a slightly greater increase. In both cases the endogenous OUR remained elevated for the duration of each test which, in the calcium experiment, involved DTPA concentrations as high as 6 g/L. The maximum change in oxygen uptake rate (the lower set of curves) clearly illustrates the protective effect of the calcium against the DTPA. As elevated DTPA concentrations, 0.75 g/L and greater, were administered to the biomass, the Rmax in the presence of calcium remained approximately constant, while the Rmax for the control sample dropped significantly, to 31.4% of the calcium Rmax- The Rmax of the biomass in the presence of calcium eventually decreased, at a DTPA concentration of 3.5 g/L. This corresponds to a Ca:DTPA stoichiometric ratio of 1.5:1. Below this ratio, it would appear that 92 the protective effect of the calcium was weakened. The actual maximum oxygen uptake rate attained for each methanol injection (the upper set of curves) also exhibited the difference between the control and metal-containing biomass samples. At lower DTPA concentrations, the maximum activity attained by the biomass was higher in the control sample. This corresponds to the relative effect of the DTPA on the endogenous OUR. However, the maximum OUR of the calcium containing samples was maintained up to DTPA concentrations of 3.5 g/L. ( 93 0 \ i i — i — i — i — i — i — i — i — i — i — i — i — I — i — I — r — 1 — i — | — i — i — i — i — | — i — i — i — r 0 1 2 3 4 5 6 DTPA (g/L) I — i — i — i — | — i — i — i — | — i — i — i — | — i — i — i — | — i — i — I — | — i — i — i — | — i — i — i — | — i — i 0 2 4 6 8 10 12 14 DTPA (mM) Figure 5.17: Effect of DTPA and calcium on respirometric measurements: Control: Filled symbols (• • • ), 13.1 mM CaCl2: Open symbols (• O A). These experiments were repeated with the use of other metals (Fe+3, Zn + 2 , Cu + 2 , Mg + 2 , Mn+2),with varying affinities for DTPA (figures 5.18 - 5.22). Of the metals tested, calcium, iron and magnesium were able to mitigate the effect of the chelant. Calcium (figure 5.20) was the only metal whose introduction resulted in an actual improvement in sludge kinetics ^maJ^max o > ^ a s m e a s u r e d by the ability of the sludge to metabolize oxygen. The greater the metal concentration, the greater the protection afforded to the biomass. This trend also held true for magnesium (figure 5.18) and iron (figure 5.19). Manganese (figure 5.21) and zinc (figure 5.22) proved to offer some protection at higher concentrations of DTPA, but at lower DTPA concentrations were more toxic than the chelant itself, with increasing metal concentrations resulting in greater levels of toxicity. The metals which offered protection against DTPA, calcium, iron, and magnesium, have relatively high stability constants (table 2.1) with the chelant, at 10.75,28.0, and 9.34 respectively. Although the other metals tested, copper, manganese, and zinc, also had high stability constants with DTPA, the concentrations required to bind significant amounts of chelant were found to be toxic to activated sludge biomass. It is apparent that both the ability to bind strongly to DTPA and the toxicity of the metal determines whether a particular metal ion is effective in protecting biomass from DTPA. Given this sensitivity to a combination of conditions (stability constant and toxicity), it is perhaps of no surprise that of the metals tested, only one, calcium, was able to enhance the activity of the biomass when added in conjunction with DTPA. Additional data obtained from this set of experiments, presented in Appendix D, provides additional evidence of the increase in endogenous respiration rates with DTPA addition. This was considered of interest because, taken as a whole, this data confirms the hypothesis that it is the presence of the chelant, independent of whether it is bound to a protective or a toxic metal, which causes an increase in the maintenance energy expended by the cells. 95 0 "I i i i i i i i i i i i i i i i i i i i i i i i i i i i i i [ i i i i i i i i i i i i i i I 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 DTPA (g/L) Figure 5.18: Effect of DTPA and magnesium on oxygen uptake rate: M g S 0 4 concentration (mM): 0 A , 1.9 2 • , 3.9 • , 6 • , 9.9 O, MgCl , concentration (mM): 7 • . 4 DTPA (g/L) Figure 5.19: Effect of DTPA and iron on oxygen uptake rate: F e d , concentration (mM): 0 0.06 • , 0.5 1.8 • , 5.3 A . 1.4-0.2-j 0—111 11 11 i 111 111 111 111 111 i | i 11 i | i i i i | i i i i 0 0.5 1 1.5 2 2.5 3 3.5 4 DTPA (g/L) Figure 5.20: Effect of DTPA and calcium on oxygen uptake rate: CaCI, concentration (mM): 0 • , 0.5 • , 3.3 A, 12.9 • , 13.1 • . 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 DTPA (g/L) Figure 5.21: Effect of DTPA and manganese on oxygen uptake rate: M n S 0 4 concentration (mM): 0 • , 2.8 A, 24.4 • . 97 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 DTPA (g/L) Figure 5.22: Effect of D T P A , zinc and copper on oxygen uptake rate: Z n S 0 4 concentration (mM): 0 • , 0.2 1.5 A, C u S 0 4 concentration (mM): 0.02 • . 5.7 • . 98 5 . 7 E F F E C T O F M E T A L S O N O X Y G E N U P T A K E R A T E The effect of the metals alone on activated sludge was also examined. Different metals, at different concentrations, had varying effects on the maximum oxygen uptake rate of activated sludge (figure 5.23). Calcium and iron had a positive effect (Rmax I Rmax 0 > 1) up to a metal concentration of approximately 0.4 g/L. The effect of adding ten times the normal "excess" concentration of calcium was also examined, in order to determine whether the mitigation afforded by the calcium would be limited by metal toxicity at much higher concentrations. The addition of 14 g/L CaCl 2 resulted in a oxygen uptake rate approximately 60% of the control value, indicating that the protective qualities of calcium are overridden by metal toxicity at elevated calcium concentrations. However, lower concentrations of calcium were able to enhance biomass activity, possibly through augmented ion transport near the cell wall. Previous work has found that the cell walls of E. coli are able to bind a wide range of metal ions (Beveridge, 1981), indicating that a close relationship between added metal ions and the bacterial cell is maintained. Of the remaining metals tested, copper, zinc and manganese exhibited definite toxicity as their presence alone resulted in a decrease in the maximum activity of the biomass. It is possible, however, that the presence of low (< 0.3 g/L) concentrations of these metals would not disrupt activated sludge treatment, provided the treatment system was operating at a loading rate well below the maximum capacity of its biomass. The lab-scale units in these studies were operated under these conditions, although this particular facet of activated sludge operation was not explored in detail. Magnesium, sodium, molybdenum, and potassium did not exhibit toxicity at the concentrations studied, and should therefore not have an impact upon treatment efficiencies. The effect of the metals on the endogenous OUR was generally less severe (figure 5.24) than the effect on maximum oxygen uptake rate, indicating that although some metals at certain 99 concentrations were able to interfere with active exogenous oxygen uptake by the cells, they were less effective at disrupting the cells' resting metabolism. Of the metals which elicited a toxic response as measured by the maximum oxygen uptake rate (figure 5.23), only copper appeared to elicit a similar effect on the endogenous respiration rate (figure 5.24). Manganese, which can be seen in figure 5.23 to have been moderately toxic to the biomass with respect to its ability to metabolize substrate, appeared in figure 5.24 to enhance the endogenous respiration rate. The remaining metals exhibited either little effect, or a positive effect, on endogenous respiration rates. In general, the effect of the metals alone did not account for the significant increases in endogenous respiration observed when metals and DTPA were present together, as discussed in section 5.6. 100 0 1 2 3 4 14 Metal Concentration (g/L) Figure 5.23: Effect of metals on maximum oxygen uptake rate: CaCl2 • , FeCij #, CuS04 • , ZnS04 A, CuS04 • , MgS04 • , MgCl2 O, MnS04 A, KC1 O. 1.4-1.2 1 if 0.8 O I 0.6 O 0.4 0.2-J 0 1 I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | M I I | I I I I | I I I I 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Metal Concentration (g/L) Figure 5.24: Effect of metals on endogenous respiration: CaCl2 • , FeClj • , CuS04 • , ZnS04 A, CuS04 • , MgS04 • , MgCl2 O, MnS04 A, KC1 O. 101 5 . 8 E F F E C T O F D T P A ON YIELD The effect of adding DTPA to activated sludge biomass was found to involve more than a change in the rate of oxygen uptake rate (section 5.5). DTPA also had an effect on the amount of oxygen consumed in the metabolism of the substrate (figure 5.25). When DTPA was added, less oxygen was consumed (as measured by the area under the curve) than when no DTPA was added. This would seem to indicate that the relative proportion of substrate used in respiratory processes and growth was altered. The oxygen uptake rate data presented in figure 5.25 was obtained after injection of equal amounts of substrate under differing conditions, and should, had yield remained unchanged, have resulted in three OUR traces with identical areas under the curves. Biomass yield appeared, however, to increase with the presence of DTPA, as seen by the smaller area under the OUR curve obtained with DTPA, but was not changed in the presence of both DTPA and an excess of calcium. Performing this experiment with varying amounts of substrate showed this response to be reproducible (figure 5.26). From this figure, the differences in biomass yield under the three conditions can be determined from the slopes of the graphs of 0 2 consumed versus substrate consumed, calculated as: [1 - the slope] (section 2.4, equation 7). This calculation determined that the yield was 15% and 22% higher with DTPA than in the control and DTPA/Ca samples, respectively. Clearly, biomass yield does not actually increase with the addition of DTPA, as revealed through the measurement of biomass growth in section 5.4. DTPA is therefore affecting the yield measurement by some method other than through the generation of biomass. The maximum OUR (figure 5.25) was much lower in a biomass sample with DTPA than without. This can be partially explained by the biomass metabolising only a portion of the given substrate, or by the substrate being metabolised by only a portion of the biomass. The second possibility can be eliminated, as this would result in a shorter (vertical) but longer (horizontal) curve with an unchanged total oxygen consumption (the area under the curve). It is 102 therefore evident that only a portion of the added substrate was being metabolised. This hypothesis was supported by data obtained from continuous reactor runs #1 and #2 (sections 6.1 and 6.2), during which effluent BOD was high, and 40% - 50% of the influent BOD was found to be untreated. It is proposed that, although there may be many other contributing factors, there is sufficient evidence to state that the apparent increase in yield is likely a result of incomplete substrate metabolism. 103 25 0 1 1 1 1 1 1 I I 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 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Time (min) Figure 5.25: Effect of DTPA on OUR: Control , 1 g/L DTPA , 1 g/L DTPA & excess calcium (molar ratio, CaCL:DTPA = 4.8.1) 0 0.2 0.4 0.6 0.8 1 1.2 Substrate (mg BOD) Figure 5.26: Effect of DTPA on total oxygen consumption: Control • , 1 g/L DTPA • , 1 g/L DTPA & excess calcium (molar ratio, CaCl2:DTPA = 4.8:1) A. 104 5.9 CELLULAR SOLUTE RELEASE The addition of DTPA to activated sludge results in the release of material into the surrounding fluid. Other researchers have determined the released material to be comprised of portions of the lipopolysaccharides (Gray, 1965b; Lieve, 1965), phospholipids and lipoproteins (Voss, 1967) from the gram negative bacterial cell wall. These compounds will be referred to as cellular solutes, in keeping with the nomenclature chosen by these researchers. The present study confirmed that the chelants DTPA and EDTA were both responsible for increased cellular solute release, to approximately the same degree (figure 5.27). Chelant addition resulted in an approximate doubling of the amount of material released under test conditions. The amount of released material was determined to be independent of the chelant concentration, at concentrations relevant to pulp mill discharge values. The concentration of chelant at which the biomass began to exhibit this response was found to be significantly lower than the concentrations used for the majority of other batch tests, and in the reactor studies (figure 5.28). Biomass from an activated sludge unit treating ECF effluent was found to be sensitive to a DTPA concentration as low as 50 mg/L. A maximum solute release value, for a given set of biomass, was found, indicating that higher DTPA concentrations could produce no further solute release. A comparison of the DTPA concentration required to release cellular material (50 mg/L) with the DTPA concentration required to affect cell respiration during respirometric tests (500 mg/L, section 5.5), illustrates the tolerance of the cells to lower concentrations with respect to cell metabolism, but not to cell leakage. These results suggest that the leakage of material is not the sole factor governing the microbial response to DTPA. It is proposed that the initial "attack" of the chelant occurs at the outermost boundary of the bacterial glycocalyx, a polysaccharide-containing structure which surrounds both gram positive and gram negative bacterial cells. Extracellular polysaccharide alone may be released at lower DTPA 105 45.0 0.0 I i i i i i i i i i i i i i i i i i i i i i—i—i i i—i i I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Chelant Concentration (g/L) Figure 5.27: Effect of chelant on intracellular solute release: DTPA • , EDTA • . Smaller numbers indicate less leakage. 40 0 \ i i i i | i i i i | i i i i |—i—i i i—|—i—i—i i j 0 0.05 0.1 0.15 0.2 0.25 DTPA (g/L) Figure 5.28: Effect of low chelant concentrations on intracellular solute release. Smaller numbers indicate less leakage. concentrations, followed by some cell wall material at intermediate DTPA concentrations, and disruptive portions of cell wall material at higher DTPA concentrations. The fact that the extracellular polysaccharide layer is no longer present in sufficient quantities to protect the cell may contribute to the leakage of important compounds. Cultures of Pseudomonas aeruginosa were found by other researchers to be significantly more resistant to EDTA if they possessed an extracellular layer of slime (capsular material) (Brown, 1971). The release of similar fractions of cellular solute at chelant concentrations equivalent to those used in previous batch tests indicates that the initial effect which the chelant has upon the activated sludge is the same, regardless of chelant concentration. A proposed diagram of the mechanism and result of chelant attack on the cell walls of the bacteria in activated sludge is shown in figure 5.29. The presence of DTPA in the media surrounding the bacterial cells causes the divalent metal ions (calcium and magnesium) in the cell walls to be either partially or completely removed through the sequestering action of the chelant. This action is sufficiently severe to harm the structural integrity of the cell wall. Without the binding effects of the metal ions, those cell wall components which were involved in metal binding are able, to some extent, to diffuse into the surrounding fluid. As was seen in figure 5.27, above a threshold value the concentration of DTPA does not significantly affect the extent of cellular solute release. It may be surmised instead that it is the nature and composition of the bacterial cell wall which might play a more important role in determining the extent of cellular solute release. This is in agreement with research performed on Pseudomonas aeruginosa, in which the bacteria were found to be resistant to EDTA if the cell wall concentration of magnesium was minimised during growth (Brown, 1975). It is assumed that activated sludge bacteria have access to ample metal ions for uptake into the cell wall material, and that this leaves the cells vulnerable to the "metal-robbing" effects of a chelating agent. The effect of the chelant on bacteria, once the metals have been sequestered from the cell walls, is complex. It is proposed that the damage to the cell wall results in a more permeable membrane, resulting in the 107 Figure 5.29: Schematic representation of the proposed method of interaction between the bacterial cells in activated sludge Hoes and DTPA: (a) intact cell, (b) initial attack by DTPA, disrupting the capsular layer of extracellular polysaccharide (EP) and complexing bivalent cell wall metal ions, leaving gaps in the cell wall, (c) release of lipopolysaccharide (LPS) from the cell wall. 108 observed increase in endogenous respiration rate (Section 5.2). This is likely to be accompanied by a response from the bacterial cell to increase energy expenditure towards cell • wall reparation and maintenance of the gradients across the cell wall, which would also be manifest through an increased endogenous respiration rate. In addition to the effects on respiration, cells capable of producing extracellular polymers would likely also respond by increasing polymer production for the purpose of protection against a harsh environment, although evidence for this would be more likely to be found in continuous reactor runs than in batch experiments. Discussion of the excess production of extracellular polymers will be accomplished in Section 6.1, but is mentioned here for completeness. Lastly, the cellular solutes released by the chelant from the cell walls of gram negative bacteria contain a large fraction of lipopolysaccharides (figure 5.29). Gram positive bacteria contain analogous compounds, such as lipoteichoic acids, that share many of the biological properties of lipopolysaccharides, and which are probably released under similar circumstances. Some peptidoglycan, present in large quantities in the gram positive cell wall is also likely released. However, as much of the previous research comparable to this portion of the present study was performed on gram negative bacteria, the description of the attack by DTPA on the cell wall is described using a generic gram negative cell. Lipopolysaccharides and lipoteichoic acids are amphiphilic in nature, having both hydrophobic and hydrophillic regions (Wicken, 1985), and are therefore surface active. As a result, once a sufficient amount of cellular amphiphilic material has been released into the surrounding media, the surfactant nature of the mixture causes foaming upon aeration, as discussed in Section 5.3. Since it has been determined (figure 5.27) that above 0.25 g/L DTPA, greater chelant concentrations did not result in greater release of cellular material, it is suggested that the sudden onset of foaming was attributable to a change in the nature of the released material, likely from material associated with the glycocalyx, to lipopolysaccharides or their equivalents associated with the cell wall. The glycocalyx was probably the source of the majority of the released material. The relatively smaller proportion 109 of material contributed by the amphiphilic material from the cell wall was responsible for the foaming and was indicative of the extent of the damage to the cells. It was not, however, of a sufficient concentration to contribute significantly to the overall material released by the cell and measured in the cellular release test. The presence of metals provided protection to the biomass from the chelant-induced polysaccharide release (table 5.4). Numbers in the table are given as percent of the maximum cellular solutes released (by boiling the cells), and are directly comparable to the highlighted value in the table representing the amount of material released by control cells. Administering the metals alone to the sludge samples revealed that the presence of either calcium or magnesium reduced the amount of cellular solutes "naturally" released during the test. Potassium had no effect, and iron appeared to somewhat enhance solute release. It is believed to be significant that those ions already present in the cell walls, and known to play important roles with respect to cell wall structural integrity, were able to decrease cell wall permeability when added to activated sludge samples. It is surmised that the added metals served to further bind the structural elements of the cell wall together through increased ionic bridging, thus the strengthening the wall. The addition of either DTPA or EDTA resulted in a release of approximately half of the cellular solutes released by boiling, representing an increase of 252% over the amount of cellular solutes released from the control samples. Of the metals tested in conjunction with the chelants, at the concentrations at which they were tested, only calcium afforded complete protection against this increase. Magnesium also protected the biomass against solute release, especially against DTPA, and could possibly afford greater protection with a larger metal concentration. Potassium did not appear to protect the biomass, and the addition of iron increased cell wall permeability. The observed effects were similar for each of the two chelants tested. 110 Table 5.4: Effect of chelants and metals on cellular solute release % of Cellular Solutes Releasable by Boiling no metal Ca 5 mM Fe 0.02 mM Mg 1.8 mM K 6.2 mM no chelant 19.4 9.9 24.0 13.2 19.3 DTPA 2.5 mM 48.6 11.3 53.4 41.5 44.7 EDTA 0.675 mM 46.7 14.2 54.1 43.6 46.4 1.35 mM 48.2 2.7 mM 49.1 (smaller numbers indicate less leakage) 6. DTPA REACTOR STUDIES Three reactor studies involving DTPA were conducted: 1) with constant DTPA concentration, 2) with step-wise increases in DTPA concentration, and 3) with a metal added in conjunction with a constant DTPA concentration. All studies were conducted at an HRT of 12 hours, and an SRT of 5 days. 6 . I C O N T I N U O U S R E A C T O R S T U D Y I : C O N S T A N T D T P A C O N C E N T R A T I O N .... Reactor performance was significantly affected by chelant addition at the rate of 0.875 g DTPA/L feed (figure 6.1). After the introduction of this concentration of DTPA in the reactor influent, BOD removal efficiency dropped to below 60%. This indicates a decreased utilization of available substrate, which is consistent with results obtained during growth tests (section 5.4), and provides evidence that the decrease in overall growth may have been a result of an inability to metabolise substrate. Further evidence that DTPA causes incomplete substrate metabolism by biomass was discussed in section 5.8, where yield was observed to increase in the presence of DTPA. As discussed previously, this may have been at least partially attributable to the metabolism of a reduced fraction of the wastewater, causing an artificially high yield to be calculated due to the indirect method of yield calculation - that of a difference between available substrate and the substrate assumed to be taken up during oxidation. Approximately two weeks were required for the reactor to gradually reach a plateau of 60% BOD removal. During this time, the biomass concentration in the reactor also decreased (Figure 6.2). This is indicative of a lower growth rate, as predicted by batch tests (section 5.4). The BOD loading to the reactor, per volume, remained constant throughout this period of declining solids levels. 112 100-90-80^ 70-I 60-o :J . 50-8 40-CD 30-20-10^ 0 dP X • _ • • _ • u • • • 0.875 g/L DTPA •o 2 e> in reactor feed I 1 1 1 I 1 1 ' I ' ' ' I ' 1 ' I 20 40 60 80 100 120 140 160 Time (days) Figure 6.1: Effect on BOD reduction of the addition of a DFf A concentration of 0.875 g/L to a continuous reactor. Figure 6.2: Reactor solids and effluent VSS • after addition of a DPTA concentration of 0.875 g/L to a continuous reactor. 113 The SRT for this run averaged 6.7 ± 0.8 days. Initially, the reactor was operating sufficiently below the maximum capacity of the biomass to continue to efficiently remove BOD, in spite of the reduced activity of the microorganisms. As the chelant began to affect the growth rate of the biomass, the solids levels began to decrease, and effluent treatment began to deteriorate. The passage of one SRT saw treatment efficiency drop significantly below acceptable levels (ie. < 90% BOD removal); the passage of 2 SRT's marked the beginning of a treatment level of < 60% BOD removal (figure 6.1). This plateau is due to the steady solids level in the reactor over this time, and is a direct measurement of the reduced activity that the microorganisms were capable of maintaining in that environment. Effluent solids levels (figure 6.2) were, on average, higher during DTPA addition to the reactor than prior to DTPA addition. This was likely a result of the deflocculation effect of DTPA, similar to that of EDTA found by other researchers (Silverstein, 1994). Settling tests performed on sludge unacclimated to DTPA (from the control reactor) failed to support this explanation. However, the settling tests were carried out over 30 minutes under quiescent conditions. It is believed that the effect of the chelant upon deflocculation would be significantly enhanced in the continuous reactor by both the length of exposure of the biomass to the DTPA, and the agitation experienced due to stirring and vigorous aeration. Upon removal of the DTPA from the reactor influent, treatment efficiency improved immediately (figure 6.1), and recovered to near original levels within 1 SRT, indicating that no permanent damage to the microorganisms was done, and that in spite of the decreased solids level, treatment could be maintained during normal operation. This confirmed that the decreased BOD removal obtained while DTPA was being administered to the reactor was a result of the presence of the chelant, and not a result of the lower solids levels. It also confirmed the cause and effect relationship between DTPA and solids levels: the chelant caused a disruption of metabolic processes resulting in lower BOD removal efficiencies and an inability of the biomass to rej uvenate i tsel f. 114 1200 1000-H • • • DTPA concentration I ' I ' I YA' " | ' " | II ! | . M | I I I | I . I | I 4 6 8 10 20 30 40 50 60 70 80 90 100 Time (days since DTPA addition) Figure 6.3: Effluent DPTA concentration during addition of 0.875 g/L to a continuous reactor. 115 DTPA was not removed by activated sludge treatment (figure 6.3). Although there is a significant amount of scatter associated with the data due to the nature of the test used to determine DTPA concentration in the effluent, the average effluent DTPA concentration over the course of the run was 875 mg/L ± 106 mg/L, which is equal to the influent concentration and therefore indicates no DTPA biodegradation. Microscopic observations of the biomass were made during this run. Prior to the introduction of DTPA to the reactor, the consortium of microorganisms observed included bacterial cells bound into dense floes, and the presence of protozoans. Within a few days of DTPA addition, no higher life forms were observed, and floes were significantly decreased in size, to approximately 5 -10 % of their pre-DTPA sizes. Free-floating material, likely bacterial cells, was abundant. This material corresponds to the general increase in effluent solids which was observed during the administration of DTPA to the reactor (figure 6.2), as discussed previously in this section. Another visual observation which deserves comment is the presence of an unidentified sticky, viscous substance adhering to the inside walls and lid of the aeration tank in the DTPA reactor, above the liquid level. Microscopic analysis revealed rotifers, stalked ciliates, and much smaller motile microorganisms within the substance. Unfortunately, long-term malfunction of the photographic equipment precluded the procurement of a visual record of this phenomenon. This substance was observed approximately one month af ter the addition of DTPA into the reactor. Although it is hypothesised that this substance is an extracellular polymer, produced perhaps to encourage reflocculation of the dispersed bacterial cells, it is unclear as to 1) why the bacterial cells within the bulk phase of the reactor were still dispersed, and the floes small, and 2) how a relatively low solids level within the reactor (<1000 mg/L) could have produced sufficient quantity of polymer to coat the available reactor surface. Nevertheless, excess extracellular polymer has been found to be produced in large quantities once bacterial adhesion has taken place (Costerton, 1981) and in the presence of surfactants (Govan, 1975), which have 116 been proposed to be released through the action of the chelant. (See section 5.9.) It is proposed that those bacterial cells already deposited on the walls of the reactor through foaming and vigourous aeration, would anchor themselves in the most favourable location, away from the DTPA in the bulk fluid. Creation of a thick layer of extracellular polymer would enable a sheltered environment to be created for the consumption of substrate and micronutrients, while providing a protective barrier against the stress of the chelant present in the bulk fluid. The majority of the bacteria within the reactor would be either unable to access the protective benefits of the excess polymer or simply under too great a stress from the incoming chelant to benefit from these protective effects. Additional bacteria and protozoans would be added to the slime phase through further foaming and aeration, establishing a second micro-ecosystem. In addition, it is possible that some of the polymeric substance, initially created by the microorganisms in the main aeration tank, through either natural or enhanced shedding of cell wall polymers (de Boer, 1981), was simply deposited on the reactor walls through foaming. The appearance of an unidentified slime is not without precedent. The reactor conditions at the time of the observation of the unidentified substance were, as indicated in figures 6.1 and 6.2, those of high food to microorganism ratio (F/M of approximately 1.0) and high continuous BOD loading due to incomplete substrate utilisation (effluent BOD of approximately 100 -150 mg/L). These conditions are comparable to those associated with the observation of viscous bulking in a media of "jelly-like consistency" in a separate study (Lo, 1994a). Other research has found that excess polysaccharide, on the order of 10 - 20 times the normal amount, is generated if growth stops while excess substrate is still available (Duguid, 1953). The conditions in the reactor during study #1 would seem to correspond to the conditions imposed by Duguid: negligible growth in the presence of DTPA, as discussed in section 5.4, and high reactor BOD (figure 6.1). It is likely that the excess sugar was used to form excess polysaccharide (Duguid, 1953). 117 6 . 2 C O N T I N U O U S R E A C T O R S T U D Y 2 : R A M P I N G D T P A C O N C E N T R A T I O N Reactor performance, as measured by BOD removal, during this study was affected only at DTPA concentrations in excess of 0.6 g/L (figure 6.4), implying that a threshold DTPA concentration existed. Below the threshold, treatment was able to continue at acceptable levels; above the threshold, treatment decreased below acceptable levels. This threshold value appeared to be at approximately 0.6 g DTPA/L influent, as BOD removal efficiency under these conditions remained above 90% for one week upon augmentation of the DTPA concentration to 0.6 g/L, but subsequently decreased over a period of several weeks to approximately 80%. Further increases in influent DTPA concentration to 0.8 g/L resulted in a drop in BOD removal efficiency, to 50%, before a partial recovery to 60% BOD removal. It is of interest to note that the threshold DTPA concentration of approximately 0.6 g/L in this continuous reactor study agrees with the threshold value obtained from batch tests (section 5.5) of between 0.5 and 0.75 g/L. This concurrence between batch and continuous experimental data would appear to reinforce both the immediacy and the persistence of the effects which DTPA exerts on the biomass in activated sludge. Biomass yield also changed with elevated DTPA concentrations. As can be seen in figure 6.5, biomass yield decreased significantly below the level in the control reactor at a DTPA concentration of 0.6 g/L. Until that point, both reactors had been averaging almost identical yields, which increased steadily due to a change in influent supply to a value of approximately 0.47. The decrease in apparent biomass yield in the DTPA reactor indicates that the biomass was no longer converting as much BOD into biomass, and occurred simultaneously with the onset of the reduction in BOD removal efficiency (figure 6.4). The yield dropped to approximately 0.27, where it appeared to reach a plateau. It was at this time that the plateau in BOD removal efficiency was established at 60% (figure 6.4). It is proposed that the apparent yield was decreased as a consequence of the following two responses by the biomass to 118 0.8-50.6-O) <0.5-0. 0.3-0.2-mm \ • ~] 1 1 1 T 1 1 • | i i I | 1 1 '. I 1 1 1 I .1 1 1 I 20 40 60 80 100 120 140 160 Time (days) r 100 r90 r80 r70 Co" r60 tion r50 due - a> rr r40 Q : o -30 CD r20 r10 -0 Figure 6.4: BOD reduction in a reactor treating influent with increasing DTPA concentration: B O D reduction • , Reactor influent DTPA concentration . 1 1 I 1 1 1 I 1 1 1 I 20 40 60 80 100 120 140 160 Time (days) Figure 6.5: Reactor biomass yield with the addition of increasing DTPA concentrations: DTPA reactor - - - , Control reactor ; Reactor influent DTPA concentration . DTPA. First, the decreased BOD removal implies that there was less energy available for all cell functions, including reproduction. In addition, the effect of a shift in energy distribution within the cell towards increase maintenance would decrease the apparent yield. Batch tests described in section 5 have been presented which lend considerable support to the concept of an increase in maintenance energy in response to the presence of DTPA. These results further support this proposition by illustrating the same effect under continuous reactor operation. Kinetic analysis of the oxygen uptake rates of the biomass was found to follow a distinct pattern throughout the duration of reactor study #2. Upon each increase in influent DTPA concentration, biomass oxygen uptake kinetics decreased before increasing to activities higher than those at the previous reactor DTPA concentration (figure 6.6). In spite of scatter in the data, the repetition of this trend for each increase in DTPA concentration provides confirmation of this observation. It would appear as though repeated increases in influent DTPA concentration are initially detrimental to biomass activity, but that as the biomass becomes acclimated to the elevated DTPA concentration, biomass activity, on a specific basis, is able to recover. Although overall activity has been shown to be insufficient at higher chelant concentrations to maintain treatment efficiency (figure 6.4), the individual bacteria have, in response to environmental stress, increased their maximum oxygen uptake rate. This likely has a mitigating effect. At the time the reactor was receiving 0.8 g/L DTPA the addition of 1.3 g/L (11.7 mM) CaCl2 to a mixed liquor grab sample did not affect the kinetics of the biomass. In spite of previous indications from batch test results (section 5.6) that the addition of a protective metal such as calcium was able to mitigate the effects of DTPA, this mitigation was not observed when the affected biomass was acclimated to elevated influent DTPA concentrations. DTPA can be expected to have exerted selective pressure for those microorganisms which could grow in its presence. These bacteria would have been growing for a sufficient period of time in the presence of DTPA, and therefore also in an environment in which metal concentration was 120 0 20 40 60 80 100 120 Time (days) Figure 6.6: E f fec t o f increasing inf luent D T P A concentrat ion on reactor k inet ics: Speci f ic m a x i m u m oxygen uptake rate, R 1 T m Reactor in f luent D T P A concentrat ion . limited. The lack of mitigation by calcium indicated that the microorganisms had a modified cell wall, in agreement with the results published by Brown (1975), and as discussed in section 5.9. Visual observations of the reactor revealed that mixed liquor in the reactor receiving DTPA was consistently lighter in colour than that in the control reactor during this study. Microscopic analysis of the sludge showed that the protozoans previously present (figure 6.7), were no longer present by the time that 0.4 g/L DTPA was being administered to the reactor. Floe size was large and the floes were dense until the administration of 0.8 g/L DTPA, at which point the floes became less dense, with many free-floating cells in the surrounding fluid (figures 6.8 and 6.9). 121 Figure 6.8: Mixed liquor after DTPA addition 122 123 6 . 3 C O N T I N U O U S R E A C T O R S T U D Y 3 : D T P A A N D C A L C I U M The addition of an excess of calcium in conjunction with 0.875 g/L DTPA did not negatively affect reactor treatment efficiency. However, it should be noted that both reactors suffered a decrease in BOD removal efficiency during the course of the run. This has been attributed to the change in reactor influent to an influent with higher BOD (figure 6.10). The increase in influent BOD may have been accompanied by an increase in the recalcitrant fraction of the wastewater, causing the universal decrease in BOD reduction. Since the reactor receiving both the metal and the chelant did not exhibit inferior treatment performance when compared to the control reactor, it can be concluded that an excess of calcium administered with the chelant protected the biomass from the negative effects of the chelant (discussed, for the same concentration of DTPA, in Section 6.1). This result agrees with the numerous batch tests (Section 5.6) which explored the combined effects of metal and chelant, and determined that calcium mitigates the negative effects of DTPA, provided that it is present in stoichiometric excess. 100-95 H c o TJ rr 90-85-Q § 80-75-• • • • " 1 • • • • • • * addition of D T P A and calcium 0 10 20 30 40 50 60 70 80 90 Time (days) Figure 6.10: BOD reduction upon addition of DTPA and calcium to reactor: DTPA-Ca reactor • , Control reactor • . 124 Microscopic analysis showed that despite the protective effect of the calcium, the biomass was nevertheless affected over the course of this run. Upon addition of the chelant and metal, motility patterns of the smaller swimming ciliates changed. Normal swimming motion was not observed and they instead exhibited a tendency to vibrate, or shake. This was observed the day after administration of the DTPA and calcium. Vibration of organisms which normally exhibit smooth movement is indicative of a significant disruption of microbial life with respect to the protozoans present in the reactor. Numerous photographs were taken of the biomass before (figures 6.11 and 6.12) and after (figures 6.13 - 6.15) introduction of the metal and chelant. Prior to chelant and metal addition, floes were in general very healthy in appearance, in spite of a slight tendency towards a fingered structure (figure 6.11), exhibiting moderately dense floes and a healthy diverse mixture of protozoans (figure 6.12). Within one day of chelant and metal addition, the smaller swimming ciliates were all observed to be vibrating (and are, as result, out of focus), and floe structure appeared to have suffered some degree of disintegration (figure 6.13). After three days (figure 6.14) and six days (figure 6.15) of operation with DTPA and calcium in the reactor influent, floe structure became increasingly disperse, and vibrating protozoans continued to be observed. 125 Figure 6.11: Mixed liquor prior to addition of DTPA and calcium Figure 6.14: Mixed liquor 3 days after addition of DTPA and calcium 127 Figure 6.15: Mixed liquor 6 days after addition of DTPA and calcium 128 Visual observations that the biomass was affected in spite of the mitigating effect which calcium had upon the treatment efficiency of the reactor were confirmed through analysis of the cellular solutes releasable with DTPA (figure 6.16). The biomass from the DTPA-Ca reactor was more susceptible to cell wall leakage than was the biomass from the control reactor. This would tend to indicate a weakened cell wall in at least a portion of the biomass, which was likely due to the presence of the chelant in the mixture, indicating that DTPA was not rendered completely "inert" by metal ions which were strongly bound to it. This confirms suspicions first described in section 5.6 that the chelant-calcium complex is able to interact with the biomass, in spite of the stability of the complex. At least one manifestation of this interaction is an increase in cell wall permeability, in spite of the generally protective effect of the calcium. Increasing metal concentrations were found to protect against cell wall leakage (figure 6.17). This confirmed previous results, given in table 5.4. Although DTPA-Ca reactor biomass was significantly more permeable than control biomass, this difference was eliminated at high metal ion concentration. It is also interesting to note that when comparing metal ion mixtures to calcium ions alone, calcium provides more protection against cell wall permeability than does a mixture of Ca, Fe, Mg, and K. 129 100.0 Figure 6.16: i i i I i i i I i i i | i i i | i i i | i i i 0 0.2 0.4 0.6 0.8 1 1.2 D T P A (g/L) Effect of the addition of DTPA and calcium to a continuous reactor on intracellular solute release: DTPA-Ca reactor • , Control reactor • . Smaller numbers indicate less leakage. 60.0 I I I I | I I M | I I I I | I I I I | I I I I | 0 0.01 0.02 0.03 0.04 0.05 Metal Concentration (Ionic Strength) 0.06 Figure 6.17: Effect of the metal addition on intracellular solute release: DTPA-Ca reactor, with Ca, Fe, Mg, K • , Control reactor, with Ca, Fe, Mg, K • , DTPA-Ca reactor with Ca A , Control reactor with Ca • . Smaller numbers indicate less leakage. 130 7. BATCH TESTS WITH HYDROGEN PEROXIDE As residual hydrogen peroxide from the bleaching stages comes into contact with pulp and paper mill effluent constituents, and subsequently with activated sludge, the oxidative properties of peroxide can be expected to affect the characteristics of both media. 7 . I E F F E C T OF PEROXIDE ADDITIONS ON COMBINED M I L L E F F L U E N T CHARACTERISTICS Hydrogen peroxide residuals in the combined mill effluent can be 200 mg/L or greater during TCF bleached pulp production (Andersen, 1995). Since H 2 0 2 is a strong oxidant it is expected that it will have an effect on effluent characteristics. Untreated mill effluent BOD was reduced approximately 25% by H 2 0 2 additions at all but the lowest (10 mg/L) concentration applied (figure 7.1). Since a lower BOD indicates a reduced amount of substrate available for degradation by the biomass in downstream secondary treatment operations, one possible effect would be a decrease in the solids concentration in the reactor at a given SRT. Another possible effect is a change in composition of the microbial community, as competition for a smaller concentration of substrate would favour the faster growing microorganisms over the slower growing, or more substrate-specific, microorganisms. This selection would favour organisms whose kinetic constant, K s , is smaller (Koch, 1979), as these organisms would therefore be able to more rapidly utilize smaller substrate concentrations. Hydrogen peroxide additions, followed by complete reduction of the added peroxide, did not appreciably affect untreated effluent toxicity even after addition of up to 640 mg/L (figure 7.1). Wood extractives, the components of effluent which are responsible for the bulk of effluent toxicity (Murray, 1993), are present in smaller concentrations than many of the compounds which contribute significantly to BOD (low molecular weight alcohols, organic acids, wood sugars etc.). This probably accounts for the fact that BOD is much more affected by H 2 0 2 addition than is effluent toxicity. It is interesting to note that the lack of reduction in toxic loading to the secondary treatment system, in conjunction with reduced BOD, indicates 131 that the risk of both negative impact on the treatment system and poor treatment efficiency may be high in the event of process variations, as fewer microorganisms may be available to treat and detoxify the effluent. This may be particularly true for toxic compounds which are more effectively degraded in the presence of an easily degraded co-metabolite. 132 o CD 80-60-40-20-0-•10 1 M I 1 1 1 1 I 1 " 1 I 1 1 1 1 I 1 " 1 I " 1 1 I 0 100 200 300 400 500 600 700 Hydrogen Peroxide Concentration (mg L) -9 -8 7 (A/A -6 -5 ;iC5o - 4 city i - 3 Toxi - 2 -1 -0 Figure 7.1: Effect of f f ,0 2 additions on combined mill effluent BOD • and toxicity w«ra,, So value of k Figure 7.2: Effect of peroxide concentration and temperature on reaction rate of peroxide reduction in effluent. Lighter shading represents higher first order rate coefficients. 133 7 . 2 D E C O M P O S I T I O N O F P E R O X I D E IN C O M B I N E D M I L L E F F L U E N T Since temperature can vary widely in pulp mill effluent streams, it was of interest to examine the effect of temperature on the kinetics of peroxide reduction in the effluent. The rate of peroxide reduction appeared to be higher at 35°C than it was at either extreme of the 4°C to 70°C range tested (figure 7.2), a finding which may indicate that the reduction of peroxide in effluent is biologically mediated. This is supported by the observation that H 2 0 2 in autoclaved effluent did not degrade appreciably within a time period of 1.5 hours. A concentration of 20 mg H 2 0 2 /L in a sample of mixed liquor from the lab-scale reactor resulted in an 8 fold increase in the rate of peroxide decomposition over the rate of decomposition in effluent alone. Repetition of this experiment using autoclaved sludge yielded negligible peroxide degradation within 12 hours, providing further support to the contention that peroxide decomposition is likely catalysed by viable microorganisms. While the rate of peroxide decomposition at 4°C is of no significance in industry, the fact that H 2 0 2 reduction did occur at 4°C precluded H 2 0 2 addition to lab scale reactors by direct dosing of the reactor feed reservoirs. Of more relevance in the pulp and paper industry is the fact that hydrogen peroxide appears to be stable in 70°C effluent. Between the bleach plant and the secondary treatment system, effluent temperature is likely to be at 50 to 70 °C until it reaches the cooling tower immediately prior to treatment. The higher the temperature, therefore, the more likely it is that little or no decomposition of residual peroxide would occur prior to effluent introduction into the biological treatment system. The rate of peroxide reduction in effluent appears to be inversely related to the initial concentration of hydrogen peroxide (figure 7.2), likely due to the inhibitory effect of the strong oxidant on the microbes which catalyse the decomposition. Because of this inhibitory effect, the peroxide in streams with higher residual peroxide content is less likely to be degraded, increasing the probability that residual peroxide from the bleach plant will reach the secondary treatment system. 134 7 . 3 E F F E C T O F H 2 0 2 S H O C K L O A D S ON UNACCLIMATED S L U D G E Residual quantities of hydrogen peroxide can be expected to fluctuate with changing conditions in the peroxide bleaching stage. Any sudden increases in peroxide concentration would result in a shock load to biomass in downstream treatment systems. It was therefore of interest to study the effect of shock loads of peroxide on activated sludge. The OUR of unacclimated activated sludge was inhibited by exposure to shock doses of hydrogen peroxide (figure 7.3). Residual peroxide concentrations typical of untreated mill effluent (200 mg/L) resulted in the reduction of biomass oxygen uptake rate to 60 % of its original value. The oxidative properties of hydrogen peroxide were responsible for the decrease in biomass activity. While the viability of the sludge was profoundly affected, the effect was reversible, with full recovery of metabolic activity within approximately 10 hours of exposure to the initial shock dose of 960 mg H 2 0 2 I L mixed liquor (figure 7.4). It is proposed that once the peroxide was fully decomposed through the oxidation of a portion of the biomass, the bacteria were able to recover through normal growth processes. Sludge shocked with 960 mg H 2 0 2 / L mixed liquor in the presence of added catalase (5.33 units/mL mixed liquor) exhibited a less severe response to the shock, as measured by OUR (figure 7.5); activity was 140% higher than without the catalase. Catalase decomposed a portion of the peroxide, thereby reducing the concentration of peroxide to which the microorganisms were exposed. It is proposed that the presence of catalase not only catalyses peroxide decomposition, but in so doing, prevents the biomass from coming into contact with the full strength of the peroxide. Shocks of hydrogen peroxide appear to have approximately the same effect on endogenous OUR whether administered as single doses or cumulative shocks. No additional negative effect was observed upon introduction of successive hydrogen peroxide shocks, when compared with single dose shocks equal in magnitude to the final cumulative amount of hydrogen peroxide (figure 7.6). 135 Figure 7.3: Effect of varying peroxide shock load levels on OUR of activated sludge. I I I I | I I I I | I I I I | I I I I | I ! I I | I I I 0 5 10 15 20 25 30 Time from shock (h) Figure 7.4: Recovery of sludge viability as a function of time alter exposure to 960 mg/L peroxide shock load. 136 8 0 | I I I | I I I | I I I | I M | I I I | I I I | I I I | I I I I 0 2 4 6 8 10 12 14 16 Time (min) Figure 7.5: Effect of hydrogen peroxide on oxygen uptake rate: Control • , After catalase-mitigated shock • , After unmitigated shock • . 1-P 0.9-0 . 8 -0 . 7 -o DC 0 . 6 ^ D Q 0 . 5 -DC O 0 . 4 ^ 0 . 3 -0 . 2 ^ 0 . 1 -0 -0 5 0 0 • 1000 1500 2 0 0 0 2 5 0 0 3 0 0 0 Hydrogen Peroxide Concentration (mg/L) Figure 7.6: Effect of hydrogen peroxide shocks on oxygen uptake rate: Cumulative shocks • , Single shocks • . 137 Hydrogen peroxide had a more significant effect on the maximum OUR (as measured after resuspension in untreated effluent) than on the endogenous OUR (figure 7.7). Endogenous OUR remained within 80% of its original value after treatments (shocks) with up to 1000 mg H 2 0 2 / L mixed liquor. Approximately twice this dosage (2000 mg H 2 0 2 / L mixed liquor) is required under endogenous respiration conditions to obtain the response obtained with 1000 mg H 2 0 2 / L mixed liquor under maximum respiration conditions. 500 1000 1500 2000 2500 3000 Hydrogen Peroxide Concentration (mg/L) Figure 7.7: Effect of hydrogen peroxide shocks on oxygen uptake rate: Endogenous OUR I Maximum OUR • . 138 7 . 4 E F F E C T O F H 2 0 2 O N A C T I V A T E D S L U D G E K I N E T I C S The administration of peroxide shocks of 100 and 500 mg/L had a negative effect on the metabolic activity of the biomass (figure 7.8). For both peroxide concentrations, it appeared as though a similar response was exhibited, resulting in a decrease in the maximum oxygen uptake rate to approximately 38% of the control value. As observed in similar tests with the addition of DTPA (section 5.5), this may indicate a reduced ability of the biomass to treat incoming effluent or to handle increased BOD loading to a continuous unit. However, as seen above (section 7.3), the effects of peroxide shocks did not appear to be as permanent in nature as those observed with DTPA. The rapid recovery times from peroxide shocks would seem to indicate that any reduced effluent treatment efficiency may be temporary, and that, provided sufficient time passes between a peroxide shock load and an increase in influent BOD, the biomass should be capable of maintaining acceptable treatment. 0 f i i A | i 0 0.5 I i i i i | i i i. i | i i i i | i i i i 1 1.5 2 2.5 Substrate (mg BOD/L) 3 3.5, 4 Figure 7.8: Effect of hydrogen peroxide on activated sludge kinetics: Control • , 100 mg H 2O 2 /L«,500mgH 2O 2 /LA. 139 8. PEROXIDE REACTOR STUDIES 8 . I C O N T I N U O U S R E A C T O R S T U D Y 4 : RAMPING P E R O X I D E C O N C E N T R A T I O N As an activated sludge reactor becomes acclimatised to receiving pulp mill effluent containing hydrogen peroxide, it is hypothesized that the biomass will adapt to the presence of the peroxide. To test this hypothesis, once steady state operation had been achieved, gradually increasing quantities of hydrogen peroxide were continually added to a lab-scale reactor treating TCF influent. The hydraulic retention time was 30 hours for this portion of the study. Acclimatization to H 2 0 2 was measured by three methods: changes in the sludge's ability to decompose peroxide, changes in catalase-equivalent activity, and changes in the response of activated sludge to shock loads of hydrogen peroxide. Throughout this continuous reactor study, acceptable treatment efficiency, as defined by: > 95% BOD removal, > 40 - 50% COD removal, and 100% acute toxicity (Microtox®) removal, was maintained after the introduction of hydrogen peroxide up to an influent concentration of 1000 mg/L (figure 8.1). The concentration of MLVSS in the reactor appeared to exhibit a transient response to the introduction of H 2 0 2 . After the addition of low levels (5 to 50 mg H 2 0 2 /L influent) the reactor MLVSS concentration decreased (figure 8.1). Some of the highest levels of effluent VSS concentration, which were variable throughout the study, were observed at a peroxide influent concentration of 50 mg/L. MLVSS concentrations returned to previous levels at H 2 0 2 concentrations greater than 100 mg/L as the reactor became acclimated to higher levels of peroxide. As stated previously, BOD and toxicity removal were not adversely impacted by this apparent response; however, this may have been due to the long (30 h) HRT used in this study. Microscopic analysis of the biomass revealed that as influent peroxide concentrations increased to 1000 mg/L, floes, originally densely packed (figure 8.2) became less strongly bonded together (figure 8.3). Structurally, the floes appeared to have suffered significant damage, resulting in the loss of some biomass (increased effluent VSS 140 I 70-Z3 X J 0> rr | 3 0 i Q- 20-E 10 1200-1000-|> 800-w 600-co > ^ 400-200-5 1000-CT) E, 800-cT 600-J 400-| 200-s o-cP°oo <b 0^>(g) o .CP o _o o o o O ^ Q D o o o a D $ » 0 o n o_oW <b° o o ° o o G D O 0 ° o • L U C I D DrjQDQTjLlIID D T J L T J L I I IIIIIIIIIJ I auD H^T) ll|lll|lliyiljl.'in-f 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 20 40 60 80 100 120 140 160 180 Time (days) Figure 8.1: Reactor performance after addition of Hfl^ Lower graph: Influent peroxide concentration (mg/L) • , Middle graph: Mixed liquor volatile suspended solids (mg/L) O, Upper graph: Removal of COD • , BOD • , and toxicity A. 141 values) and a subsequent decrease in reactor MLVSS. However, protozoans such as stalked ciliates were present in the mixed liquor for the duration of the study (figures 8.2 and 8.3), which is indicative of an otherwise healthy microbial community. The observation of a complete and healthy biomass is consistent with the maintenance of acceptable treatment levels (figure 8.1). As the activated sludge reactor became acclimatised to H 2 0 2 , the rate of decomposition of hydrogen peroxide by mixed liquor samples increased (figure 8.4). Additionally, acclimated sludge was capable of decomposing increasing concentrations of peroxide without an increase in the amount of time required, indicating that acclimated sludge was less affected by an increase in the concentration of peroxide administered to the reactor. The faster the mixed liquor was able to degrade the incoming peroxide, the less likely it was to be adversely affected by it, as previously discussed in section 7.3. Peroxide in the aeration tank was undetectable throughout the study, indicating rapid peroxide decomposition and a microbial population able to handle a steady influx of hydrogen peroxide. 143 I I I I I I I I I I I I I I I I I I I I I 40 60 80 100 120 140 160 180 200 Peroxide Decomposed (mg/L) Figure 8.4: Rate of peroxide decomposition by unacclimated and acclimated sludge: Reactor influent H 2 0 2 concentration (mg/L): 0 • , 1000 • . 120-1 100-. 9 0 -8 0 -7 0 -6 0 -50 -4 0 -3 0 -20 -10-0 -% OUR Remaining after 320 mg/L Peroxide Shock L O O O O O O O O T - OJ IO o o o o T — CM IO O Reactor Acclimation Level (mg H„0 9 /L ) -rm-rp 100 200 300 400 500 600.700 800 Peroxide Shock Dose (mg/L) Figure 8.5: Effect of peroxide shock dose on viability of sludge from reactors acclimated to different peroxide levels: Reactor influent H 2 0 , concentration (mg/L): 0 • , 500 • . 144 Biomass activity, as measured by oxygen uptake rate, of the acclimated sludge was less strongly affected by the peroxide administered to the reactor than was unacclimated sludge (figure 8.5). Results given in the main chart in figure 8.5 represent the response to varying shock levels of peroxide and are presented as percentages of base (pre-shock) OUR. The inset graph illustrates the response of sludge at varying levels of acclimatisation to the same peroxide shock. Comparison of shock response of unacclimated sludge and sludge acclimated to 500 mg H 2 0 2 / L influent shows that activated sludge is less negatively affected if it is acclimatised to the presence of peroxide in its feed. The trend was similar for all peroxide shock concentrations (5 - 960 mg/L). An attempt was made to use the degree of acclimation discussed above to quantify the ability of the biomass to decompose peroxide. Comparing the improved response to shocks with the response obtained with the addition of 5.33 units catalase/mL (figure 7.5), the equivalent level of mitigation was achieved at a reactor influent concentration of 10 mg H 2 0 2 /L. The microorganisms at that hydrogen peroxide dosage level, therefore, can be said to be producing'5.33 units of catalase (or a combination of catalase equivalents) per millilitre of mixed liquor. Knowing the biomass concentration during this period (figure 8.1), a catalase expression of approximately 10 units/mg MLVSS can be calculated. The peroxide decomposition ability of activated sludge has been previously expressed as catalase-equivalent activity (Welander, 1989), and as the rates of peroxide decomposition were of more interest than the precise method by which the decomposition occurred, catalase-equivalent activity was monitored throughout this reactor study. As the experiment progressed towards higher rates of peroxide dosing, catalase-equivalent activity appeared to have been induced (figure 8.6). Under the reactor conditions in this study a maximum level of catalase induction in activated sludge, represented by an activity level of approximately 1.3-1.5 /min g biomass, appeared to exist, based on an extrapolation of the results. The maximum catalase-equivalent activity found in these tests is likely an indication of the maximum capacity of the 145 biomass to produce compounds (enzymes) to catalyse such decomposition. It is to be expected that limits to cellular products exist, as the cells have a finite supply of energy available to them, to be shared by all cell functions. One possible explanation for the initial decrease in reactor solids levels (in addition to cell lysis through oxidation by peroxide) is that energy which would have been used for biomass reproduction, was routed instead to the more immediate task of catalase (or catalase-equivalent) production. Once catalase-equivalent activity was at or near its maximum (at a reactor peroxide dose of 200 mg/L), the reactor solids concentration increased and stabilized at a higher level (figure 8.1). Acclimation of mixed cultures to increasing concentrations of hydrogen peroxide could be a result of either the induction of peroxide-degrading enzymes such as catalase in existing members of the microbial consortium, the selection of microorganisms with enhanced abilities to degrade hydrogen peroxide, or both. These results are not able to distinguish between these two mechanisms. However, the decrease in reactor MLVSS concentration upon exposure to H 2 0 2 supports the idea that peroxide is exerting a selective pressure on the microbial community, allowing those organisms which can tolerate peroxide to flourish, and perhaps resulting in the death and washout of those which lack the capacity to accommodate the addition of peroxide. Such acclimation may result in a reduction in species diversity, a condition commonly associated with a decreased tolerance to perturbations. 146 1.8 1.6H 0 200 400 600 800 1000 Reactor Peroxide Dose (mg/L feed) Figure 8.6: Induced peroxide decomposition ability expressed as catalase-equivalent activity. 147 8 . 2 C O N T I N U O U S R E A C T O R S T U D Y 5 : C O N S T A N T P E R O X I D E C O N C E N T R A T I O N This reactor study was performed at an HRT of 12 hours, allowing a more direct comparison of the effects with those observed in the DTPA studies (also performed at an HRT of 12 hours). Treatment efficiency was not affected by the continuous addition of 500 mg hydrogen peroxide/L influent (figure 8.7). BOD removal decreased slightly after 5 days of receiving hydrogen peroxide; however, a similar decrease was observed in the control reactor, which indicates that the decrease was not attributable to the presence of H 2 0 2 . COD removal efficiencies exhibited a similar response. The improvement in BOD removal efficiency in both reactors after day 42 was attributed to a switch in reactor influent, as discussed in Section 4.5. These results confirm the finding in section 8.1 that hydrogen peroxide addition, at concentrations likely to be experienced as residual concentrations in peroxide bleached pulp mill effluent, does not interfere with the removal of the oxygen demand of the effluent. Removal of acute toxicity appeared to worsen with the addition of hydrogen peroxide (Table 8.1). Under normal activated sludge operation, 100% toxicity removal is regularly attained (Simpura, 1993; Rempel, 1992). Toxicity removal during this study was consistently lower in the peroxide reactor during the time that peroxide was administered. The failure of the activated sludge to completely remove acute toxicity may be due to the 12 h HRT used in this study, in contrast with the 30 h HRT used in study #4. However, the control reactor also failed occasionally to successfully remove toxicity. This is indicative of the presence of an unusually recalcitrant effluent compound which precluded effective treatment. Slight decreases in BOD and COD removal rates over the same period would tend to support a hypothesis of a reactor influent which was exceptionally difficult to treat. Therefore, although the peroxide reactor exhibited unacceptable levels of toxicity removal, under normal operating conditions it is likely that acceptable treatment would be maintained. 148 100-90 -i 80-70-. | 60-3 T3 •g 50^ e 30-3 20 10-3 0 l a Peroxide Addition (500 mg/L) i i i i i i i i I i i i i | i i i i | i i i i | i i i i | i i i i | i i 0 10 20 30 40 50 60 70 Time (days) Figure 8.7: BOD removal: Peroxide reactor, receiving 500 mg H 20 2/L influent • , Control reactor • . 0.8-0.7-» 0.6-> _ i 2 0.5-O) I 0.4^ CM o E 0 . 3 H J= 0.2-0.1-0-Peroxide Addition (500 mg/L) 10 i i i i i i i i i i i i i i i i i i i i i 20 30 40 50 60 Time (days) Figure 8.8: Effect of 500 mg H,0,/L influent on activated sludge kinetics: R 149 Table 8.1: Percent Microtox® toxicity removal Toxicity Removal (%) Peroxide Reactor Control Reactor Prior to H 2 0 2 addition 97 100 After H 2 0 2 addition 55 100 29 46 After H2O2 addition and influent switch to conventional 49 84 100 na After H 2 0 2 removal and switch to ECF 100 100 Microscopic examination of the biomass has shown that the addition of 500 mg/L H 2 0 2 resulted in an increase in both debris and filaments within a few days of initial peroxide introduction. An excessive number of rotifers, which is normally indicative of an unhealthy biomass, was also observed in the peroxide reactor, but not in the control reactor. These effects appeared to be transient in nature, as the passage of two weeks (corresponding to approximately 3 SRT's) resulted in a substantial improvement in the amounts of debris, filaments, and protozoans. Examination of the kinetics of the biomass before and after the introduction of H 2 0 2 (figure 8.8) reveals a reduction in sludge kinetics after peroxide introduction. The same reduction was not observed in the control reactor. These results are a clear indication that the capacity of the biomass to handle changes in influent strength had been substantially reduced. Catalase activity was also monitored during this reactor study, and compared directly with the catalase activity in the control reactor (figure 8.9). The ability of the biomass in the reactor receiving peroxide to rapidly decompose the peroxide was approximately one order of magnitude greater than in the control reactor. However, this ability was lost upon removal of peroxide from the reactor influent, indicating the need for a constant peroxide influx for the maintenance of peroxide-decomposition capability within the biomass. It also implies that this 150 ability involves the expenditure of energy, as the removal of the stress to which the microorganisms were responding resulted in the immediate cessation of catalase activity. 1.2 0 20 40 60 80 100 120 Time (days) Figure 8.9: Effect of activated sludge acclimation to 500 mg/L H,02 on catalase activity: Peroxide reactor • , Control reactor • . 9. COMBINED STUDIES WITH DTPA AND PEROXIDE 9. I R E S P I R O M E T R I C DATA WITH C O M B I N E D D T P A / P E R O X I D E ADDITIONS The addition of both chelant and peroxide to biomass was accomplished by a sequential addition of the chemicals, in two ways: (1) administering a peroxide shock to biomass previously exposed to DTPA, and (2) administering DTPA to biomass previously exposed to a peroxide shock. In the first test, the addition of peroxide after DTPA resulted in a further decrease of the activity of the biomass, already significantly reduced by the DTPA (figure 9.1). An initial decrease of approximately 50% of the maximum oxygen uptake rate was observed upon the introduction of 1 g/L DTPA. The maximum oxygen uptake rate after peroxide addition was 33% of the uptake rate after DTPA, and 16% of the original maximum uptake rate. The negative effect of adding 100 mg/L peroxide was enhanced by the previous addition of 1 g/L DTPA. This effect can be compared with that of the addition of peroxide alone (figure 9.2), which resulted in a decrease of only 53% in maximum oxygen uptake rate. Clearly, peroxide has a more detrimental effect on biomass previously exposed to DTPA. In the second test, the addition of DTPA after a peroxide shock (figure 9.3) resulted in a response similar to that observed with DTPA alone (section 5.5). The maximum oxygen uptake rate, at DTPA concentrations greater than 1 g/L, was reduced by approximately 50%, the same decrease noted above (figure 9.1) for DTPA alone. Clearly, the prior shock of peroxide had no effect on the level of disturbance caused by the DTPA. This test was performed using sequential doses of DTPA, for comparison with the results in chapter 5. A similar test was conducted to isolate the effects of a single dose of DTPA after a peroxide shock of 500 mg/L. The addition of 1 g/L DTPA, after measuring the effect of 500 mg/L peroxide on activated sludge kinetics for 2 hours, resulted in a further decrease in the kinetics of the biomass (figure 9.4), although this decrease did not bring the activity of the biomass significantly lower than 152 after the original dosage of 500 mg/L peroxide. The kinetics were however, again approximately 50% lower than immediately prior to the addition of DTPA, confirming the independent effect of DTPA. During the initial peroxide shock, therefore, the peroxide randomly oxidized both the microorganisms resistant to DTPA and those that were susceptible to the effects of DTPA. Comparison of the overall effect of administering peroxide prior to or following the addition of DTPA revealed that the biomass was more severely affected when DTPA was added first. Interpretation of figures 9.1 through 9.3 benefits from the realization that DTPA addition (at a concentration of 1 g/L in each test) resulted in a 50% decrease of available biomass activity whether or not peroxide was previously added. Peroxide addition, however, (at a concentration of 100 mg/L in each test) resulted in a decrease in biomass activity which was dependant on whether or not DTPA had been added. It is proposed that peroxide, through random oxidation of biomass, killed or severely damaged the bacterial cells until it was completely decomposed, and therefore had a more potent effect if the cells had been previously exposed to the damaging effects of DTPA. It is probable, based on the findings in this study, that the increased cell wall permeability caused by DTPA (section 5.9) allowed for a facilitated attack of peroxide on the cell due to less hindered peroxide diffusion through the wall. This effect would also be augmented by greater access of the peroxide to the bacteria due to deflocculation when in contact with DTPA. 153 0 ' I I I I I I I I I I I I I I I I I I I I I I I I I II I I I I I I I I I I l.l I I I : I | I I I I | 0 1 2 3 4 5 6 7 8 9 10 Substrate (mg BOD/L) Figure 9.2: Effect of hydrogen peroxide on activated sludge kinetics: Control • , After 100 mg H O / L t 2 2 154 0.45 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 DTPA (g/L) Figure 9.3: Effect of D T P A on activated sludge kinetics after a 100 mg/L hydrogen peroxide shock. 1HP 0.9-^  0.6-0.7-o X to 0.6-. E • CC 0.5-X -_ 1 rr 0A-. 0.3-i 0.2-0.1-0-^ A d d e d 500 m g / L peroxide Added 1 g / L DTPA I I i i i I i i i I i i i I i i i I i i i I i—r-T-|—r—r 0 20 40 60 80 100 120 140 160 y Time (min) Figure 9.4: Effect of D T P A after a hydrogen peroxide shock on activated sludge kinetics. 155 9 . 2 E F F E C T O F D T P A ON C A T A L A S E - E Q U I V A L E N T ACTIVITY INDUCTION Previous tests for catalase-equivalent activity (ch. 8) were performed on reactor biomass operating under steady state conditions. When considering the possible combined effects of DTPA and peroxide, it was anticipated that a response such as the induction of enzyme production might be strongly affected by the additional environmental stress of a chelant. Batch tests showed that induction of catalase-equivalent activity was achieved over a period of approximately 6 hours after the introduction of 500 mg/L H2C»2 to reactor feed (figure 9.5). The presence of 0.25 g/L DTPA reduced the overall specific catalase-equivalent activity, and the presence of 1.25 g/L DTPA precluded the induction of catalase-equivalent activity, probably due to the overall reduction in activity of the biomass under these conditions. Peroxide addition in the presence of 0.25 g/L DTPA initially induced catalase-equivalent activity at the same rate as catalase-equivalent activity was induced in the control biomass. However, after a period of approximately 5 hours, catalase-equivalent activity induction began to taper off, resulting in an overall catalase-equivalent activity approximately 25% lower than in the control biomass. Since this concentration of DTPA was found to be non-inhibitory over the course of respirometric tests (section 5.5) and a continuous reactor study (section 6.2), it is likely that 0.25 g/L DTPA sufficiently sensitized the biomass to the incoming peroxide to reduce the ability of the biomass to respond the peroxide. It has already been shown in section 5.9 that a concentration of 0.25 g DTPA/L had a significant effect on the release of cellular material, likely through damage to the cell wall. In spite of the damage inflicted by this chelant concentration the cells were still able to respond to an environmental stress (that of hydrogen peroxide addition) in the same way in which they would normally have responded. These results, added to those previously elaborated upon, illustrate that lower chelant concentrations are disruptive, but not incapacitating, to activated sludge biomass, allowing the microorganisms to continue respiration (section 5.2), oxygen uptake (section 5.5), and enzyme induction. 156 Time (h) Figure 9.5: Effect of D T P A on catalase-activity induction: D T P A concentration (g/L): 0 • , 0.25 • , 1.25 A. 157 9 . 3 E F F E C T O F D T P A ADDITION T O A C O N T I N U O U S R E A C T O R R E C E I V I N G 5 0 0 M G H 2 0 2 / 1 _ I N F L U E N T The combination of the effects of adding DTPA and hydrogen peroxide simultaneously to an activated sludge reactor was examined for the purpose of determining whether, under the conditions of a continuous reactor, the effects of the two separate compounds were additive. The hydraulic retention time in this study was 12 hours, and the solids retention time was 5 days. The addition of 0.875 g DTPA/L influent to a reactor acclimatised to receiving 500 mg H 2 0 2 / L influent had a profound effect upon effluent treatment. COD removal, previously approximately 50% during hydrogen peroxide addition, dropped to between 10 and 20% in the first week after the addition of DTPA to the reactor feed (figure 9.6). After 1 week, a removal rate of 0% was observed. The ability of the biomass to treat biodegradable substances within the effluent was severely affected, to a greater degree than observed after the addition of the same concentration of DTPA alone (section 6.1). Individually, H 2 0 2 and DTPA had much less severe effects on activated sludge. Hydrogen peroxide alone, at a concentration of 500 mg/L, did not have a negative effect on treatment efficiency (section 8.2), and DTPA alone, at a concentration of 0.875 g/L, reduced BOD removal efficiency to 60%. A COD removal of 0% (and therefore also a BOD removal of 0%) implies that the effects of the two compounds were synergistic. It would appear that those organisms, either as a fraction of the whole biomass or as a distinct group of species, which would ordinarily be capable of at least partial biological treatment in the presence of elevated DTPA concentrations had been rendered incapable of such treatment by the combined effects of hydrogen peroxide and DTPA. It is suggested that the more severe damage was rendered by the chelant, through disruption of the cell wall structure, increased cell wall permeability, and decreased ability to oxidize substrate. To these damaged cells, an oxidative compound was added, and the biomass was unable to: 1) shield itself from the peroxide due to increased cell wall permeability, 2) produce a sufficient quantity of 158 peroxide-degrading enzyme due to reduced catalase activity, and 3) reproduce itself due to reduced growth rates. The result, after 1 week, or approximately 1 SRT, was the complete failure of treatment. 11111111111111111111111M | II 11 j 111 75 76 77 78 79 80 81 Time (days) Figure 9.6: Effect of DTPA on COD removal of sludge acclimated to 500 mg/L peroxide. 9 . 4 E F F E C T O F D T P A ON T H E C E L L U L A R S O L U T E R E L E A S E O F B I O M A S S A C C L I M A T I S E D T O 5 0 0 M G H 2 O A / L I N F L U E N T Biomass obtained from the reactor acclimatised to 500 mg H 2 0 2 / L influent (continuous reactor study #5) responded differently to DTPA than did unacclimatised biomass (figure 9.7). Biomass from the peroxide reactor released more cellular solute material than did the control biomass, indicating that it was more sensitive to cell wall disruption by DTPA. The proportion of leakage between the two biomass samples appeared to be approximately constant regardless of chelant or metal concentration, giving the impression of a generally weaker biomass, with little dependence of the biomass strength on the short-term conditions imposed during the test. The difference in biomass strength between the two samples may have been a result of the direct action of hydrogen peroxide on the cells, or a property of the consortium of microorganisms acclimated to the addition of peroxide into the reactor. The selective power of the hydrogen peroxide, hypothesized to select for a distinct set of microorganisms (section 8.1), may have selected for those microorganisms which were more susceptible to the influence of DTPA (figure 9.7). An additional possibility for the increased cellular solute release is the increased ability of DTPA to access bacterial cells in floes affected by hydrogen peroxide. As seen in section 8.1 (figure 8.3), floes are much less dense in a reactor receiving peroxide, and this may simply allow the DTPA to affect a greater number of cells during a test of cellular solute release. 160 161 9 . 5 E F F E C T O F D T P A ON G R O W T H O F B I O M A S S F R O M 5 0 0 M G / L H 2 0 2 UNIT As previously discussed, the growth rate of activated sludge biomass is decreased upon addition of DTPA to the growth medium. However, it has been found that if the inoculum is acclimatised to receiving 500 mg/L H 2 0 2 in its influent, the effect is less severe (figure 9.8). Of the tests conducted on biomass affected by the combined effects of DTPA and peroxide, this is the only test which indicated a benefit from the addition of both chemicals to an activated sludge system. This result is indicative of the different microbial communities likely to be present in the two different inocula, caused by the selective effect of the hydrogen peroxide. It would appear as though the microorganisms present in the H 2 0 2 acclimated inoculum were less susceptible to lower concentrations (0 - 0.6 g/L) of DTPA, but that at concentrations greater than 0.6 g/L, the effect is similar. Each inocula also appeared to benefit by approximately the same degree to the mitigating effect of excess calcium. 100H 90 -_ 8 0 -_ <o : I 7 0 ^ .c -? 6 0 - \ "H. * 2. 5 50-E S : \ \ with excess calcium I 40- { O : o 3 0 --8 20-. ' 10-E u I I I | i i I | I I I | I I I | i l I | I i I | i i I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 D T P A (g/L) Figure 9.8: Effect of peroxide acclimation and DPrA on sludge growth: Control reactor • , Peroxide reactor • , Control reactor with excess calcium A, Peroxide reactor with excess calcium • , excess calcium molar ratio, DTPA:Ca = 4.8. 162 10. CONCLUSIONS This work summarizes the results of an investigation into the consequences of the implementation of a totally chlorine free bleaching sequence in a bleached kraft mill. The effects were determined to result from 1) a change in bulk effluent characteristics, and 2) the introduction of compounds unique to the TCF bleaching process into the effluent. Both changes were found to alter the performance of a secondary biological effluent treatment system. The overall effects of switching from the treatment of conventionally-bleached mill effluent to TCF mill effluent were positive in nature. The replacement of chlorine by peroxide resulted in a greater biodegradable fraction in mill wastewater, and a more easily treatable effluent with higher BOD removal efficiencies. The specific effects of the common TCF residual compounds, the chelant DTPA and the bleaching chemical hydrogen peroxide, were found to be more detrimental than the overall change in effluent characteristics. Examination of the effects of DTPA and peroxide on activated sludge has revealed that both compounds have a significant influence on the operation and health of the aerobic biological treatment system. The effects of the compounds individually and in combination, were studied under batch and continuous reactor test conditions. DTPA was found to interact with the cell walls of the microorganisms in activated sludge. Contact between the cell wall and the chelant, at concentrations as low as 50 mg/L, resulted in the release of cellular material, implying an increase in cell wall permeability. However, continuous tests revealed that effluent treatment efficiency did not decrease until a threshold DTPA concentration (approximately 0.5 - 0.6 g/L) was reached. At concentrations high enough to cause treatment efficiency to decrease, the released material initiated foaming during batch tests. The material was therefore surface active in nature, supporting the hypothesis that a 163 portion of the released material was amphiphilic material from the cell wall and glycocalyx. Greater chelant concentrations were not found to result in a measurably greater release of cellular material, and it was concluded that the sudden onset of foaming was attributable to the release of smaller quantities of surface active lipopolysaccharides, or lipoteichoic acids, from a more structurally important area of the cell wall. Concurrent with cellular solute release under batch test conditions, however, a decrease in the maximum oxygen uptake rate of the biomass was observed in response to added substrate. The resulting decreased substrate removal left less energy available for cellular functions such as cell growth. The maximum oxygen uptake rate was reduced to approximately 20-30% of its original value, likely a result of the redirection of energy to the reparation of the damage done to the cell walls by the chelant. Examination of the endogenous respiration rates alone provided additional evidence of the redirection of available energy. Endogenous respiration rates were observed to increase in the presence of the chelant, likely also in response to the more permeable cell wall, and the requirement for increased maintenance energy. It was hypothesized that the substrates metabolised for maintenance were possibly supplied by the increasingly available storage compounds which were originally more firmly bound to the cell wall. The presence of DTPA in growth media was found to hinder the growth of activated sludge microorganisms, at DTPA concentrations similar to those used in batch and continuous tests. The lag time, before the onset of microbial growth, increased noticeably with increasing DTPA concentrations, and both growth rate and overall growth yield decreased. It was DTPA's action as a chelant which was determined to be responsible for the detrimental effects, since altering its ability to chelate metals strongly altered its effect upon biomass respiration. Each metal tested had a different influence on the effect of DTPA on activated sludge. The negative effect on cell respiration was mitigated by the addition of appropriate amounts of calcium, magnesium, or iron, but not by the addition of manganese, copper, or zinc, which were each found to exhibit toxicity towards activated sludge. Of 164 particular interest was the addition of calcium in conjunction with the DTPA, which was found to enhance the ability of the biomass to metabolise substrate. The result was in an increase in the maximum oxygen uptake rate, R m a x , by approximately 20% and the half saturation constant, K s , by 40%. It was proposed that this effect, uniquely observed upon the addition of both calcium and DTPA, was a result of the combination of two separate effects: 1) the increase in cell wall permeability caused by DTPA and calcium together, and 2) the protection, by calcium, against the negative effects of DTPA on respiration. The addition of calcium did not eliminate the measurable negative effects of the chelant, however, as the decrease in growth rate was only partially mitigated by the presence of calcium added at a stoichiometric ratio of 4.8:1 in the growth media. Addition of DTPA and calcium to a continuous reactor resulted in changed motility patterns of the protozoa, and a biomass which was more susceptible to cell wall leakage. A weakened cell wall in at least a portion of the biomass further indicated that DTPA was not rendered completely "inert" by metal ions which are strongly bound to it, confirming that the chelant-calcium complex was able to interact with the biomass. Additionally, calcium mitigation was not observed for biomass previously exposed to elevated influent DTPA concentrations. The general susceptibility of activated sludge bacteria to the effects of DTPA was likely a result of the access to ample metal ions during growth in pulp mill effluent. Maximum metal ion uptake into the cell wall material would leave cells vulnerable to the "metal-robbing" effects of a chelating agent. The overall effects of the chelant on bleached kraf t mill effluent were also determined. DTPA was found to contribute approximately 25 - 40 % of the total COD input in DTPA-containing bleach plant effluent, thus significantly affecting effluent strength. DTPA was found to be recalcitrant to microbiological treatment, and at concentrations higher than those found in pulp mill effluent, inhibitory to the BOD test's seed bacteria, likely through the prevention or interruption of the respiratory processes. This was concluded to be indicative of the potentially 165 toxic nature of the chelant to biomass. DTPA at concentrations typically found in TCF BKME was found to exhibit no toxicity to Vibrio fisheri. Hydrogen peroxide, the second TCF bleaching residual to be studied, did not affect effluent treatment up to a feed concentration of 1000 mg/L. However, biomass quality deteriorated notably at peroxide concentrations greater than 200 - 500 mg/L. Biomass acclimation was achieved through an increased ability of the cells to rapidly decompose hydrogen peroxide, as measured by an increase in catalase-equivalent activity by approximately one order of magnitude. Reactor biomass concentration.decreased during periods of reactor operation with hydrogen peroxide addition, proposed to be a result of direct oxidation. Another contributing factor was likely that the energy which would have been used for biomass reproduction, was routed instead to the more immediate task of catalase (or catalase-equivalent) production. Once catalase-equivalent activity was at or near its maximum (at a reactor peroxide dose of 200 mg/L), the reactor solids concentration increased and stabilized at a higher level. The effects of F^Oj on effluent characteristics were determined to likely have minimal impact on activated sludge treatment. Hydrogen peroxide reduced the BOD of untreated mill effluent by 25%, but did not affect the toxicity. Effluent treatment and detoxification may therefore be more difficult with less substrate available due to the presence of residual hydrogen peroxide from the bleach plant. Results from one of the continuous reactor studies indicated that a reactor acclimatised to an influent peroxide concentration of 500 mg/L may be susceptible to toxicity breakthrough. Residual peroxide was found to remain within the effluent at higher effluent temperatures, indicating its persistence under typical mill operating conditions. In addition, the greater the hydrogen peroxide concentration in the effluent, the slower the degradation process, thus increasing the probability that significant residual peroxide concentrations reach the" treatment system. Shock loads of peroxide adversely affected the activity of activated sludge, although activated sludge acclimatised to the presence of H 2 0 2 was less adversely affected due to a more 166 rapid decomposition of the peroxide by the enhanced catalase activity of the biomass. The effects of DTPA and hydrogen peroxide were also examined in combination. A 1 g/L DTPA concentration was found to have the same effect on biomass activity whether or not a shock of 100 mg/L peroxide was administered first. DTPA was also found to hinder the catalase induction which would normally take place in response to the addition of hydrogen peroxide. Additionally, biomass acclimatised to peroxide was more susceptible to cell wall leakage. These effects may have severe repercussions for the operation of a reactor. In continuous reactor trials, the addition of DTPA and hydrogen peroxide together resulted in complete treatment failure after 1 week, or just over 1 SRT. It can be concluded from these experiments that the presence of hydrogen peroxide and DTPA in pulp and paper mill effluent being treated by an activated sludge unit have the potential to significantly impact the operation of such a unit. Conditions under which reactor performance can be predicted to deteriorate or fail, based upon this study, would involve sudden increases in peroxide concentration, the incomplete complexation of chelant with metals during the bleaching process, or a combination of these events. 167 11. ENGINEERING SIGNIFICANCE Activated sludge treatment systems are sensitive to an influx of hydrogen peroxide and the chelant DTPA, individually or together. As a result, treatment systems at risk of receiving significant concentrations of either or both of these compounds are vulnerable to the effects described in this thesis. Disruption of activated sludge treatment efficiency is a likely result of the introduction of uncomplexed DTPA at concentrations greater than 500 mg/L. However, immediate removal of the chelant will result in the recovery of treatment efficiency. These problems will not be encountered if the DTPA is sufficiently complexed with metals from the chelation stage, or if a DTPA concentration of < 0.5 g/L exists in the effluent. The introduction of unchanging concentrations of hydrogen peroxide, up to concentrations of 1000 mg/L, do not reduce treatment efficiency, provided that uncomplexed DTPA is not also present. Shock loads of hydrogen peroxide will have temporary detrimental effects, from which the activated sludge will recover upon removal of the shock. Given the choice of operating an activated sludge unit with constant H 2 0 2 input, or no regular input with the risk of occasional shock loads of H 2 0 2 , it is recommended that the former option be chosen, as the microorganisms will adapt to the presence of peroxide and will have enhanced capabilities of handling unforeseen shock load conditions should they occur. The introduction of both peroxide and uncomplexed DTPA should be guarded against through complexation of the chelant, in order to avoid complete treatment failure. Switching effluent sources between TCF- and conventional- or ECF-kraft bleaching sequences has no serious implications for biological treatment, although the treatment of TCF effluent will likely result in higher BOD removal rates. In summary, the treatment of TCF effluent, containing novel bleaching compounds, is feasible provided that proper precautions are taken. 168 12. RECOMMENDATIONS FOR FUTURE WORK The following areas of study are suggested as extensions of the work presented here: 1. Determination of the mechanism of DTPA attack, with particular focus on the resulting endogenous OUR and the specific substrates removed. 2. Further study of the combined effects of DTPA and peroxide, with attention to the effects on cellular solute release and the identification of the released solutes. An examination of which bacterial species are affected would be of interest in the determination of whether different species are affected by DTPA and peroxide. 3. Analysis of the microbial response to FL;Q,, to determine the exact nature of the catalase-equivalent activity. Determination of the effect of catalase production on maintenance energy would also be of interest. 4. Exploration of the relationship between wastewater biodegradability and the bleaching process used, to determine whether greater selectivity for lignin degradation over cellulose results in the release of a more recalcitrant biodegradable wastewater fraction. 169 NOMENCLATURE Abbreviation Explanation Units BOD biochemical oxygen demand mg/L COD chemical oxygen demand mg/L DO dissolved oxygen mg/L DTPA diethylenetriaminepentaacetic acid ECF elementally chlorine free EDTA . ethylenediaminetetraacetic acid HBOD headspace biochemical oxygen demand mg/L HRT hydraulic retention time h IC50 toxicity measurement: concentration mg/L at which 50% of luminescence is extinguished K s half saturation constant mg/L LX low halogen content (conventionally bleached) y. growth rate MLVSS mixed liquor volatile solids mg/L }imax maximum growth rate HQ initial or control growth rate OUR oxygen uptake rate mg/L»min OURo initial or control oxygen uptake rate mg/L#min OUR e nd endogenous OUR mg/L»min O U R m a x maximum OUR upon substrate injection mg/L»min q substrate uptake rate mg substrate/L min q m a x maximum substrate uptake rate mg substrate/L min R oxygen uptake rate mg Oo/L min R m a x maximum oxygen uptake rate due to mg O2/L min substrate utilisation R m a x o initial or control maximum OUR due to mg O2/L min substrate utilisation S substrate concentration mg/L SDOUR specific delta OUR mg/L»min»mg biomass SRT solids retention time d TCF totally chlorine free TMP thermomechanical pulping X biomass concentration mg/L Xn initial biomass concentration mg/L Y yield Y0bs observed yield 170 Bleaching sequence abbreviations: Abbreviation Explanation A peracid bleach \ C chlorine bleach D chlorine dioxide bleach E extraction Eo oxygen enhanced extraction Eop oxygen and peroxide enhanced extraction Ep peroxide enhanced extraction O oxygen delignification P peroxide bleach Q chelation T dimethyldioxirane ("activated oxygen") X xylanase (enzyme) bleach Z ozone bleach REFERENCES Ahtiainen, J., Nakari, T. and Silvonen, J. 1996. Toxicity of TCF and ECF Pulp Bleaching Effluents Assessed by Biological Toxicity Tests. Nordic Pulp and Paper Research Journal, 4:365-378. Albert, R.J. 1994. Worldwide Survey: State-of-the-Art TCF Bleaching. Proceedings: 1994 International Non-Chlorine Bleaching Conference, Paper no. 11-1. Alder, A.C., Siegrist, H., Gujer, W. and Giger, W. 1990. Behaviour of NTA and EDTA in Biological Wastewater Treatment. Water Research, 24 (6):733-742. Allen, S.L., Lawrence, H.A., and Flaherty, T.H. 1993. Defoaming in the Pulp and Paper Industry in: Surfactant Science Series, vol 45, ed: P.R. Garrett, Marcel Dekker, Inc., New York.: 151-176. Allison, R.W. 1991. Potential of Ozone in Kraft Pulp Bleaching. Appita, 44 (6):405-409. Andersen, R. and Hallan, P. 1995. State-of-the-Art Effluent Treatment for a Scandinavian Pulp and Paper Mill. Proceedings: 1995 Tappi Environmental Conference, :645-651. Anderson, R. 1992. Hydrogen Peroxide Use in Chemical Pulp Bleaching. Tappi Notes: 1992 Bleach Plant Operations Short Course, : 123-133. Andersson, P-E., Gunnarsson, L., Olsson, G., Welander, T., and Wikstrom, A. 1987. Anaerobic Treatment of CTMP Effluent. Pulp and Paper Canada, 88 (7):T233-236. Anon. 1993. Chlorine Free: TCF or ECF? Papermaker, March:38-42. Asbell, M.A. and Eagon, R.G. 1966a. The Role of Multivalent Cations in the Organization Structure, and Assembly of Cell Wall of Pseudornonas aeruginosa. Journal of Bacteriology, 92 (2):380-387. Asbell, M.A. and Eagon, R.G. 1966b. The Role of Multivalent Cations in the Organization and Structure of Bacterial Cell Walls. Biochemical and Biophysical Research Communications, 22 (6):664-671. Axegard, P. 1989. Improvement of Bleach Plant Effluent by Cutting Back on Cl 2 . Pulp and Paper Canada, 90 (5):78-82. Bailey, J.E. and Ollis, D.F. 1986. Biochemical Engineering Fundamentals. McGraw-Hill Book Company, New York. Bambrick, D.R. 1985. The Effect of DTPA on Reducing Peroxide Decomposition. Tappi Journal, 68 (6):96-100. Basciano, C.R., and Hiemburger, S.A. 1990. Importance of Chemical Pretreatment on the Hydrogen Peroxide Brightening of Mechanical Pulps. 76th Annual Meeting Technical Section CPPA Jan 30-Feb 2,1990, Montreal, PQ., :A7-A18 172 Beaton, A. 1994. Developing Markets Push Industry to Consider Using TCF Processes. Pulp and Paper, February:77-78. Bell, C F . 1977. Principles and Applications of Metal Chelation, Clarendon Press, Oxford. Belly, R.T, Lauff, J.J. and Goodhue, CT. 1975. Degradation of Ethylenediaminetetraacetic Acid by Microbial Populations from an Aerated Lagoon. Applied Microbiology, 29 (6):787-794. Beveridge, T.J. 1989. Interactions of Metal Ions with Components of Bacterial Cellwalls and their Biomineralization, in: Metal Microbe Interactions, Poole, R.K and Gadd, G.M., (eds.) :65-83. Beveridge, T.J. and Koval, S.F. 1981. Binding of Metals to Cell Envelopes of Escherichia coli K-12. Applied and Environmental Microbiology, 42 (2):325-335 Beyenal, N. Y., Ozbelge (Baser), T. A. and Onder Ozbelge, H. 1997. Combined Effects of Cu+2 and Zn+2 on Activated Sludge Process. Water Research, 31 (4):699-704. Bolton, H. Jr., Li, S.W., Workman, D.J. and Girvin, D.C 1993. Biodegradation of Synthetic Chelates in Subsurface Sediments from the Southeast Coastal Plain. Journal of Environmental Quality, 22:125-132. Boman, B., Ek, M., Heyman, W. and Frostell, B. 1991. Membrane Filtration Combined with Biological Treatment for Purification of Bleach Plant Effluents. Wat. Sci.Tech. 24 (3/4), 219-228. Bowers, A.R., Gaddipati, P., Eckenfelder, W.W. Jr. and Monsen, R.M. 1989. Treatment of Toxic or Refractory Wastewaters with Hydrogen Peroxide. Water Science and Technology, 21:477-486. Brown, M.R.W. and Foster, J.H.S. 1971. Effect of Slime on the Sensitivity of Pseudomonas aeruginosa to EDTA and Polymyxin. J. Phartn. Pharmac. Science Communications, 23:Suppl. 236S. Brown, M.R.W. and Melling, J. 1969. Loss of Sensitivity to EDTA by Pseudomonas aeruginosa Grown under Conditions of Mg-Linidation. Journal of General Microbiology, 54:439-444. Brown, M.R.W., ed. 1975. Resistance of Pseudomonas aeruginosa. John Wiley and Sons Ltd., Toronto. Canovas, R. V. 1992. The Alkaline Extraction (E) & Oxygen-Alkali Extraction. Tappi Notes: 1992 Bleach Plant Operations Short Course, : 117-121. Cardinal, L.J., and Stenstrom, M.K. 1991. Enhanced Biodegradation of Polyaromatic Hydrocarbons in the Activated Sludge Process. Research Journal of the Water Pollution Control Federation, 63 (7):950-957. Casey, J.P.(ed.) 1980. Pulp and Paper Chemistry and Chemical Technology, third ed., John Wiley & Sons, Toronto. 173 Cates, D.H., Eggert, C , Yang, J.L. and Eriksson, K.-E.L. 1995. Comparison of Effluents from TCF and ECF Bleaching of Kraft Pulps. Tappi Journal, 78 (12):93-98. Cech, J.S., Chudoba, J., and Grau, P. 1984. Determination of Kinetic Constants of Activated Sludge Microorganisms. Water Science and Technology, 17 :259-272. Chudoba, J., Cech, J.S., Farkac, J., and Grau, P. 1985. Experimental Verification of a Kinetic Selection Theory. Water Research, 19 (2): 191-196. Cocci, A.A., Landine, R.C, Brown, G.J., Tennier, A.M. 1985. Pilot-Scale Anaerobic Treatment of Peroxide Bleachery Waste, Paper Machine Effluent and Waste Activated Sludge. Proceedings of the 40th Industrial Waste Conference, Purdue University, West Lafayette Indiana, May 14-15,1985. Bullerworths, Boston, MA., :335-341. Cole, C.A., Stamberg, J.B., and Bishop, D.F. 1973. Hydrogen Peroxide Cures Filamentous Growth in Activated Sludge. Journal oftlie Water Pollution Control Federation, 45 (5):829-835. Colodette, J.L. 1987. Factors Affecting Hydrogen Peroxide Stability in the Brightening of Mechanical and Chemimechanical Pulps. PhD Dissertation, UMl Dissertation Services, Ann Arbor, MI. Colodette, J.L., and Dence, CW. 1989. Factors Affecting Hydrogen Peroxide Stability in the Brightening of Mechanicla and Chemimechanical Pulps, Part IV: The Effect of Transition Metals in Norway Spruce TMP on Hydrogen Peroxide Stability. Journal of Pulp and Paper Science, 15(3): J79-J83. Colodette, J.L., Ghosh, A.K., Dhasmana, B., Singh, U.P., Gomide, J.L. and Singh, R.P. 1994. Bleaching Processes for Market Grade TCF Pulps. Proceedings: 1994 International Non-Chlorine Bleaching Conference, Paper no. 7-1 Costerton, J.W. and Irvin, R.T. 1981. The Bacterial Glycocalyx in Nature and Disease. Ann. Rev. Microbiol, 35:299-324. Costerton, J.W. Marrie, T.J. and Cheng, K.-J. 1985. Phenomenon of Bacterial Adhesion, in: Bacterial Adhesion, Savage, D.C. and Fletcher, M., eds., Plenum Press, New York. Croon, I. 1993. Remarkable Advance in Kraft Pulping Technology. European Papermaker, 1 (3): 18-21. Dahlman, O., Morck, R., Ljungquist, P., Relmann, A., Johansson, C , Boren, H. and Grim vail, A. 1993. Chlorinated Structural Elements in High Molecular Weight Organic Matter from Unpolluted Waters and Bleached-Kraft Mill Effluents. Environmental Science and Technology, 27:1616-1620. Dahlman, O.B., Reimann, A.K., Stromberg, L.M. and Morck, R.E. 1995. High-Molecular-Weight Effluent Materials from Modern ECF and TCF Bleaching. Tappi Journal, 78 (12):99-109. 174 de Boer, W.R., Kruyssen, F.J., and Wouters, J.T.M. 1981. Cell Wall Turnover in Batch and Chemostat Cultures of Bacillus subtilis. Journal of Bacteriology, 145 (1):50-60. Dean, A.C.R., Ellwood, D.C.. Melling J and Robinson, A. 1976. The Action of Antibacterial Agents on Bacteria Grown in Continuous Culture, in: Continuous Culture 6: Applications and New Fields, Ellis Horwood Ltd., Chichester. Donnini, G.R, Mosher, S.C, and Scroggins, R.P. 1985. Mill Scale Application of Sulpur Dioxide to Reduce Bleaching Effluent Toxi city. Pulp and Paper Canada, 86 (12): 190-193. Driessen, W.J.B.M. and Wasenius, C.-O. 1994. Combined Anaerobic/Aerobic Treatment of Peroxide Bleached TMP Mill Effluent. Water Science and Technology, 29 (5-6):381-389. Duguid, J.P. and Wilkinson, J.F. 1953. The Influence of Cultural Conditions on Polysaccharide Production by Aerobacter aerogenes. Journal of General Microbiology, 9:174-189. Dunlop-Jones, N., and Gronberg, V. 1994. Recent Developments in the Application of Xylanase Enzymes in Elemental Chlorine-Free (ECF) and Total Chlorine-Free (TCF) Bleaching. Proceedings: 80th Annual Meeting, Technical Section, CPPA, :A 191-196. Dwyer, F.Pand Mellor, D.P., eds. 1964. Chelating Agents and Metal Chelates. Academic Press, New York. Eckenfelder, W.W. 1991. Toxicity Reduction in Industrial Wastewaters. Journal of Water Science and Technology, 24 (7): 187 Eichhorn, G.L. 1973. Inorganic Biochemistry, vols 1,2. Elsevier Scientific Publishing Company, New York. Eikelboom, I.D.H. and van Buijsen, H.J.J. 1981. Microscopic Sludge Investigation Manual: Report A 94a, TNO Netherlands Organization for Applied Scientific Research, Delft, the Netherlands. Eriksson, L. and Aim, B. 1991. Study of Flocculation Mechanisms by Observing Effects of a Complexing Agent on Activated Sludge Properties. Water Science and Technology, 24 (7):21-28. Ferris, F.G. and Beveridge, T.J. 1986. Physicochemical Roles of Soluble Metal Cations in the Outer Membrane of Escfierichia coli K-12. Canadian Journal of Microbiology, 32:594-601 Fletcher, M. 1985. Effect of Solid Surfaces on the Activity of Attached Bacteria, in: Bacterial Adhesion: Mechanisms and Physiological Significance, Savage, D.C and Fletcher, M., eds. :339-362. Folke, J., Renberg, L. and McCubbin, N. 1996. Environmental Aspects of ECF vs TCF Pulp Bleaching. Environmental Fate and Effects, St. Lucie Press 681-691. 175 Forrest, R. 1992. Focus Sharpens on Bio-bleaching as Way to Reduce Chlorine Use. Pulp and Paper Journal, March/April:39. Frost, A.E. 1956. Polyaminopolycarboxylic Acids derived from Polyethyleneamines. Nature, 178:322. Gardiner, J. 1976. Complexation of Trace Metals by Ethylenediaminetetraacetic Acid (EDTA) in Natural Water. Water Research, 10:507-514. Gaudy, A.F., Rozich, A.F., Garniewski, S., Moran, N.R., Ekambaram, A. 1988. Methodology for Utilizing Respirometric Data to Assess Biodegradation Kinetics. Proc. 42nd Purdue Ind. Waste Conf.,:573-584. Gellerstedt, G., and Heuts, L. 1997. Changes in the Lignin Structure During a Totally Chlorine Free Bleaching Sequence. Journal of Pulp and Paper Science, 23 (7) :335-339. Germgard, U. and Norden, S. 1993. Superbatch Cooking and Ozone Bleaching. Paper Southern Africa, April:23-26. Goldschmidt, M.C. and Wyss, O. 1966. Chelation Effects on Azotobacter Cells and Cysts. Journal of Bacteriology, 91 (1): 120-124. Govan, J.R. W. 1975. Mucoid Strains of Pseudornonas Aeruginosa: The Influence of Culture Medium on the Stability of Mucus Production. Journal of Medical Microbiology, 8 :513-522. Gray, G.W. and Wilkinson, S.G. 1965a. The Effect of Ethylenediaminetetra-acetic Acid on the Cell Walls of Some Gram-Negative Bacteria. Journal of General Microbiology, 39:385-399. Gray, G.W. and Wilkinson, S.G. 1965b. The Action of Ethylenediaminetetra-actic Acid on Pseudornonas aeruginosa. Journal of Applied Bacteriology, 28 (1): 153-164. Greenberg, A.E., Clesceri, L.S. and Eaton, A.D. eds. 1992. Standard Methods for the Examination of Water and Wastewater. 18th edition. American Public Health Association. Washington, D.C. Halliwell, B. 1979. Oxygen-Free-Radicals in Living Systems: Dangerous but Useful?, in: Strategies of Microbial Life in Extreme Environments, M. Shilo (ed).: 195-221. Hantula, J., Kurki, A., Vuoriranta, P., and Bamford, D.H. 1991. Rapid Classification of Bacterial Strains by SDS-polyacrylimide Gel Electrophoresis: Population Dynamics of the Dominant Dispersed Phase Bacteria of Activated Sludge. Applied Microbiology and Biotechnology, 34 :551-555. Hanzlik, R.P. 1976. Inorganic Aspects of Biological and Organic Chemistry. Hartz, K.E.,. Zane, A.T and Bhagat, S.K. 1985. The Effect of Selected Metals and Water Hardness on the Oxygen Uptake of Activated Sludge. Journal of the Water Pollution Control Federation, 57 (9):942-947. 176 Henneken, L, Nortemann, B. and Hempel, D.C. 1995. Influence of Physiological Conditions on EDTA Degradation. Applied Microbiology and Biotechnology, 44:190-197. Hileman, B. 1993. Concerns Broaden over Chlorine and Chlorinated Hydrocarbons. Chemical and Engineering News, April 19:11-20. Himmelblau, D.M. 1980. Process Analysis by Statistical Methods. University of Texas. Hueting, S., de Lange, T., and Tempest, D.W. 1979. Energy Requirement for Maintenance of the Transmembrane Potassium Gradient in Klebsiella aerogenes NCTC 418: A Continuous Culture Study. Archives of Microbiology, 123 :183-188. Hughes, M.N. and Poole, R.K. 1989. Metal Mimicry and Metal Limitation in Studies of Metal-Microbe Interactions, in: Metal Microbe Interactions, Poole, R.K and Gadd, G.M., (eds.) Hunt, K, and Lee, C.-L. 1995. Dimethyldioxirane ("Activated Oxygen"), A Selective Bleaching Agent for Chemical Pulps. Part II. Dimethyldioxirane (T) Used as the Interstage Treatment in an OTO Sequence. Journal of Pulp and Paper Science, 21 (8):263-267. Jayawant, M.D. and De Graw, E.J. 1994. Practical Implications of Metals Management in Totally Chlorine-Free (TCF) Pulp Production. 1994 International Non-Chlorine Bleaching Conference, Paper no. 7-3, Jones, G.L. 1976. Microbiology and Activated Sludge. Process Biochemistry, 1 :3-5,24. Jones, M.M. 1983. Therapeutic Chelating Agents, in: Metal Ions in Biological Systems, vol. 16, Siegel, H., ed., Marcel Dekker, Inc., New York: 47-83. Jurasek, L. 1995. Toward a Three-Dimensional Model of Lignin Structure. Journal of Pulp and Paper Science, 21 (8):274-279. Kakii, K., Sugahara, E., Shirakashi, T., and Kuriyama, M. 1986a. Isolation and Characterization of a Ca++-Dependent Floc-Forming Bacterium. Journal of Fermentation Technology, 64(l):57-62. Kakii, K., Yamaguchi, H., Iguchi, Y., Teshima, M., Shirakashi, T., and Kuriyama, M. 1986b. Isolation and Growth Characteristics of Nitrilotriacetate-Degrading Bacteria. Journal of Fermentation Technology, 64 (2): 103-108. Kari, F.G. and Giger, W. 1996. Speciation and Fate of Ethylenediaminetetraacetate (EDTA) in Municipal Wastewater Treatment. Water Research, 30 (1): 122-134. Kennedy, K.J., Andras, E., Elliott, C M . , and Methven, B. 1991. Effect of a Chelating Agent (DTPA) on Anaerobic Wastewater Treatment in an Upflovv Sludge Blanket Filter. Catuidian Journal of Civil Engineering, 18:53-57. 177 Kibbel, W.H., Raleigh, C.W., and Shepherd, J.A. 1972. Hydrogen Peroxide for Industrial Pollution Control. Proc. 27th Ind. Waste Conf., Purdue Univ., West Lafayette, Ind., :824-839. Kirsop, B.E., Doyle, A., eds., 1991. Maintenance of Microorganisms and Cultured Cells: A Manual of Laboratory Methods. Academic Press, Toronto. Koch, A.L. 1979. Microbial Growth in Low Concentrations of Nutrients, in: Strategies of Microbial Life in Extreme Environments, M. Shilo (ed). :261-279. Koch, J.H. 1955. Cobalt Chloride and the Alpha-Cells of the Pancreas. Nature, 175 (4463):856-857. Kutney, G.W. 1985. Hydrogen Peroxide: Stabilization of Bleaching Liquors. Pulp & Paper Canada, 86 (12): 182-189. Lancaster, L.M., Yin, C. and Phillips, R.B. 1992. The Effects of Alternative Pulping and Bleaching Processes on Product Performance - Economic and Environmental Concerns. Proceedings: International Symposium on Pollution Prevention inthe Manufacture of Pulp and Paper, August 18-20, 1992, Washington, D.C., : 194-205. . Lapierre, L, Bouchard, J., Berry, R.M., and van Lierop, B. 1995. Chelation Prior to Hydrogen Peroxide Bleaching of Kraft Pulps: An Overview. Journal of Pulp and Paper Science, 21 (8) :268-273. Lavielle, P. 1993. Xylanase Pre-bleaching. Papermaker, Sept.:29-31. Lenta, M.P. and Riehl, M.A. 1960. The Influence of Metallic Chelates on the Diphosphopyridine Nucleotide Oxidase and Diphosphopyridine Nucleotide-Cytochrome c Reductase Systems. Journal of Biological Chemistry, 235 (3):859-864. Lewandowski, G.A. 1990. Batch Biodegradation oflndustrial Organic Compounds Using Mixed Liquor from Different POTWs. Research Journal WPCF, 62 (6):803-809. Lieve, L. 1965. Release of Lipopolysaccharide by EDTA Treatment of E. Coli. Biochemical and Biophysical Research Communications, 21 (4):290-296. Lo, S.N., Lavallee, H.C., Rowbottom, R.S., Meunier, M.A., andZaloum, R. 1994a. Activated Sludge Treatment of TMP Mill Effluents Part 1: Effluent Characterization and Treatability Study. Tappi Journal, 77 (11): 167-178. Lo, S.N., Lavallee, H.C., Rowbottom, R.S., Meunier, M.A., and Zaloum, R. 1994b. Activated Sludge Treatment of TMP Mill Effluents Part 2: Biokinetic Parameters and Effluent Treatment Strategies. Tappi Journal, 77 (11): 179-184. Logan, B.E. and Wagenseller, G.A. 1993. The HBOD Test: A New Method for Determining Biochemical Oxygen Demand. Water Environment Research, 65 (7):862-868 Lbvblad, R. and Malmstrom, J. 1994. Biological Effects of Kraft Pulp Mill Effluents - A Comparison Between ECF and TCF Pulp Production. Proceedings: 1994 International Non-Chlorine Bleaching Conference, Paper no. 11-2 178 Martell, A.E. 1974. Critical Stability Constants. Plenum Press, London. Meadows, D.G. 1995. Monsteras going 100% TCF as it Increases Production and Minimizes Emissions. Tappi Journal, 78 (12):49-52. Means, J.L., Kucak, T. and Crerar, D.A. 1980. Relative Degradation Rates of NT A, EDTA and DTPA and Environmental Implications. Environmental Pollution (Series B), 1:45-60. Mellor, D.P. 1964. in: Chelating Agents and Metal Chelates, Dwyer, F.P and Mellor, D.P, eds. Academic Press, New York. Molin, N.L., Welander, T.G., Hansson, B.G., Andersson, P.E., Olsson, B.A.G. 1987. Method of Treating Peroxide-Containing Wastewater. Murray, W.D. and Richardson, M. 1993. Development of Biological and Process Technologies for the Reduction and Degradation of Pulp Mill Wastes that Pose a Threat to Human Health. Crit. Rev. Env. Sci. Tech., 23(2): 157-194. Neilson, A., and Hoglund, C. 1977. EDTA and DTPA: Their Effect on the Growth of Bacteria and Attempts to Isolate Bacteria Capable of Using Them as Carbon or Nitrogen Sources. IVL Report B349. Nelson, P.J., Stauber, J.L., Gunthorpe, L., Deavin, J.G., Munday, B.L., Krassoi, R. and Simon, J. 1995. Study Shows ECF, TCF Effluents Have Long-term Toxic Impact on Sea Life. Pulp and Paper, August: 103-109. Nevalainen, J., Rantala, P.R., Junna, J., Lammi, R. 1991. Activated Sludge Treatment of Kraft Mill Effluents from Conventional and Oxygen Bleaching. Water Science and Technology, 24 (3/4):427-430. Nielsen, P.H. 1996. The Significance.of Microbial Fe(III) Reduction in the Activated Sludge Process. Water Science and Technology, 34 (5-6): 129-136. NOrtemann, B. 1992. Total Degradation of EDTA by Mixed Cultures and a Bacterial Isolate. Applied and Environmental Microbiology, 58 (2): 671 -676. O'Connor, B.I., Kovacs, T.G., Voss, R.H., Martel, P.H. and Van Lierop, B. 1994. A Laboratory Assessment of the Environmental Quality of Alternative Pulp Bleaching Effluents. Pulp and Paper Canada, 95 (3):47-56. O'Connor, B.I., Kovacs, T.G., Voss, R.H., Martel, P.H. and Van Lierop, B. 1993. A Laboratory Assessment of the Environmental Quality of Alternative Pulp Bleaching Effluents. Proceedings: EUCEPA International Environmental Symposim, Paris 27-29 April 1993, 1 Sessions A-C-D:273-297. Orhon, D. and Artan, N. Modelling of Activated Sludge Systems. Technomic Publishing Co., Inc., Lancaster, Pennsylvania, 1994. Paice, M.G., Bourbonnais, R., Reid, I.D., Archibald, F.S., and Jurasek, L. 1995. Oxidative Bleaching Enzymes: A Review. Journal of Pulp and Paper Science, 21 (8) :280-284. 179 Pryke, D.C. 1989. Mill Trials of Substantial Substitution of Chlorine Dioxide for Chlorine: Part II. Pulp and Paper Canada, 90 (6):93-97. Reeve, D.W. 1992. The Principles of Bleaching. Tappi Notes: 1992 Bleach Plant Operations Short Course : 1-12. Rempel, W., Turk, O., and Sikes, J.E.G. 1992. Side-by-Side Activated Sludge Pilot Plant Investigations Focusing on Organochlorines. J. Pulp Pap. Sci. 18(3), J77-J85. Repaske, R. 1958. Lysis of Gram-Negative Organisms and the Role of Versene. Biochimicaet biophysica acta, 30:225-232. Robitaille, M.A. 1988. Hydrogen Peroxide: A Versatile Bleaching Agent. Pulp and Paper Canada, 89 (12):T411-414 Saunamaki, R. 1995. Treatability of Wastewaters from Totally Chlorine-free Bleaching. Tappi Journal, 78 (8): 185-192. Schnell, A., Skog, S, and Sabourin, M.J. 1993. Chemical Characterization and Biotreatability of Alkaline-Peroxide Mechanical Pulping Effluents. Proceedings: 1993 Tappi Environmental Conference -.187-199. Seiler, H and Blaim, H. 1982. Population Shifts in Activated Sludge from Sewage Treatment Plants of the Chemical Industry: A Numerical Cluster Analysis. European Journal of Applied Microbiology and Biotechnology, 14 :97-104. Shulman, A and Dwyer, F.P. in: Chelating Agents and Metal Chelates, Dwyer, F.P and Mellor, D.P, eds. Academic Press, New York, 1964. Silverstein, J., Hess, T.F., Mutaari, N.A. and Brown, R. 1994. Enumeration of Toxic Compound Degrading Bacteria in a Multi-Species Activated Sludge Biomass. Water Science and Technology, 29 (7):309-316. Simpura, E., and Pakarinen, K. 1993. Super Low Loaded Activated Sludge Process Incorporating an Aerobic Selector for the Treatment of Pulp and Paper Waste Water. Tappi Environmental Conference. 865-877. Sinner, M. and Preselmayr, W. 1992. Chlorine is Out, Bring in the Enzymes. PPI, Sept.:87-89. Slinn, R.J. 1992. Markets and End Uses for Bleached Pulps. Tappi Notes: 1992 Bleach Plant Operations Short Course : 13-26. Springer, A.M. Industrial Environmental Control. Pulp and Paper Industry. Second Edition. Tappi Press, 1993. Stacey, M. and Barker, S.A. 1960. Polysaccharides of Micro-Organisms. Clarendon Press, Oxford. 180 Stauber, J., Gunthorpe, L., Woodworth, J., Munday, B., Krassoi, R., and Simon, J. 1996. Comparative Toxicity of Effluents from ECF and TCF Bleaching of Eucalypt Kraft Pulps. Appita, 49 (3): 184-188. Stinnett, J.D., Gilleland, H.E. and Eagon, R.G. 1973. Proteins Released from Cell Envelopes of Pseudornonas aeruginosa on Exposure to Ethylenediaminetetraacetate: Comparison with Dimethylformamide-Extractable Proteins. Journal of Bacteriology, 114 (l):399-407. Strang, A. 1996. Personal Communication. Stromberg, B, and Szopinski, R. 1994. Pressurized Hydrogen Peroxide Bleaching for Improved TCF Bleaching. International Pulp Bleaching Conference Papers, 199-209. Strunk, W.G. 1990. Kraft Bleach Plants Increase Use of Hydrogen Peroxide as Benefits Mount. Pulp & Paper, October: 112-116. Tanford, C , 1980. The Hydrophobic Effect: Formation of Micelles and Biological Membranes. John Wiley & Sons, Toronto. Troughton, N.A., Desprez, F., and Devenyns, J. 1994. Peracids: The Pathway to High Brightness TCF Kraft Pulps. Proceedings: 1994 International Non-Chlorine Bleaching Conference. Paper no. 10-1. Turakhia, M.H., Cooksey, K.E. and Characklis, W.G. 1983. Influence of a Calcium-Specific Chelant on Biofilm Removal. Applied and Environmental Microbiology, 46 (5): 1236-1238. Umbreit, W.W. 1976. Essentials of Bacterial Physiology. Burgess Publishing Company, Minneapolis. Verta, M., Ahtiainen, J., Nakari, T, Langi, A. and Talka, E. 1996. The Effect of Waste Constituents on the Toxicity of TCF and ECF Pulp Bleaching Effluents. Environmental Fate and Effects, St. Lucie Press :41-50. Virtapohja, J., and A\6n, R. 1998. Accelerated Degradation of EDTA in an Activated Sludge Plant, in: Proceedings, CPPA Technical Section, 84th Annual Meeting, :B375-B377. Voss, J.G. 1967. Effects of Organic Cations on the Gram-negative Cell Wall and Their Bactericila Activity with Ethylenediaminetetra-acetate and Surface Active Agents. Journal of General Microbiology, 48:391-400. Wallace, J.G. 1962. Hydrogen Peroxide in Organic Chemistry. E.I. du Pont de Nemours & Co. Wang, Y.-T. and Latchaw, J.L. 1990. Anaerobic Biodegradability and Toxicity of Hydrogen Peroxide Oxidation Products of Phenols. Research Journal of the Water Pollution Control Federation, 62 (3):234-238. Welander, T., 1989. Anaerobic Treatment of CTMP Effluent. Doctoral Dissertation. Department of Applied Microbiology, Lund University. Lund, Sweden. 181 Weiser, R., Asscher, A.W. and Wimpenny, J. 1968. In Vitro Reversal of Antibiotc Resistance by Ethylenediamine Tetra-acetic Acid. Nature, 219:1365-1366. Whiteman, R. 1998. Improving Treatment Performance with "Natual" Augmentation. Proceedings: 1998 Tappi Environmental Conference :785-788. Wicken, A.J. 1985. Bacterial Cell Walls and Surfaces, in: Bacterial Adhesion, Savage, D.C. and Fletcher, M., eds. Plenum Press, New York: 45-70. Wilkinson, S.G. 1975. Sensitivity to Ethylenediaminetetraacetic Acid in: Resistance of Pseudomonas aeruginosa, Brown, M.R.W. ed., John Wiley and Sons, Toronto: 145-177. Wilkinson, S.G. 1967. The Sensitivity of Pseudomonads to Ethylenediaminetetra-acetic Acid. Journal of General Microbiology, 47:67-76. Wolin, M.J. 1966. Lysis of Vibrio succinogenes by Ethylenediaminetetraacetic Acid or Lysozyme. Journal of Bacteriology, 91 (5): 1781-1786. Yang, W.C., Yanasugondha, D., and Webb, J.L. 1958. The Inhibition of Mitochondrial Respiration by 1,10-Phenanthroline and 2,2'-Bipyridine and the Possible Relationship to Oxydati ve Phosphorylation. Journal of Biological. Chemistry, 232:659-668. Zitrides, T.G. 1980. Mutant Bacteria Control Filamentous Growth in Mill Wastewater Treatment. Pulp and Paper, 54 (2): 172-174. 182 APPENDICES A P P E N D I X A : E F F L U E N T D T P A C O N C E N T R A T I O N Calibration data for the test to measure DTPA concentration in treated effluent is shown in figure A. 1. Increasing the total volume of calcium chloride in the test sample, in stepwise additions of a 0.2M calcium chloride solution, resulted in a sudden increase in sample turbidity. The greater the concentration of DTPA in the sample, the greater the volume of calcium chloride required to initiate the onset of turbidity. It may also be noted in figure A. 1 that increasing DTPA concentrations resulted in initial "negative adsorbance" values. This was due to a partial clarification of the effluent through interaction between DTPA and effluent components. Correlation of the volume of calcium chloride solution required to cause the sudden increase in turbidity to the DTPA concentration of the sample resulted in an approximately linear relationship (figure A.2). 183 0.8 [ I I T 1 J I' I'TT'j T TT T j ' H ' l T"]"T TTT'|"TTTT'|'T 7 J I yrr r v | r r i i j 'i 0 100 200 300 400 500 600 700 800 900 1000 DTPA (mg/L) Figure A.2: Calibration: DTPA = 2500 * Vol - 741.75 where Vol = the volume at which the absorbance begins lo increase. 184 A P P E N D I X B : A C T I V A T E D S L U D G E KINETIC M E A S U R E M E N T Sample experimental data clearly illustrates the response of activated sludge during a test of the biomass kinetics (figures B. 1 and B.2). Prior to an injection of substrate (usually untreated mill effluent, although simple substrates such as methanol and formic acid were also used), the biomass used oxygen at a constant rate (figure B. 1), the endogenous metabolic rate. Upon substrate injection, the biomass responded by increasing the rate of oxygen consumption. The OUR remained elevated until the substrate had been consumed and subsequently returned to the endogenous respiration rate. Plotting the OUR over the course of the test allowed for a more intuitive representation of the injection response (figure B.2). In addition to the endogenous OUR, represented in figure B.2 as an average "baseline" and the maximum OUR, the shaded area under the curve illustrates the total amount of oxygen used in the metabolism of the added substrate. All calculations of biomass kinetics were based upon these three parameters. The rate of oxygen utilisation, R, was calculated as follows: R-OUR^-OUR^ [18] 185 0 I I I I I I I I I I I I I I I I I I I I I I I I I I M I I I I I I I I I I I I I I I I I I M I I I 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time (min) Figure B. 1: Sample D O trace during substrate injection. Figure B.2: Sample O U R trace during substrate injection: Average endogenous O U R prior to i n j e c t i o n , O U R . A P P E N D I X C : C E L L U L A R S O L U T E R E L E A S E T E S T Verification of cellular solute release test validity was accomplished through the use of a scanning spectrophotometer. The wavelength used in normal test measurements was 260 nm; the absorbance at both this wavelength and the wavelength of maximum absorbance is marked on each figure (figures C. 1 to C.4). When biomass was present alone and boiled to release cellular material, the released material absorbed most strongly at 208 nm, with a visible second peak at approximately 260 nm (figure C. 1). Performing the test'without boiling the biomass resulted in a much sharper maximum peak at 196 nm, and a visible although significantly smaller shoulder (~ 25% that of the boiled sample) at 260 nm (figure C.2). Clearly, under the conditions of the test, the biomass released a certain quantity of material into the supernatant, which absorbed at approximately 260 nm. This provided verification that 260 nm was the proper wavelength for the purposes of measuring material released from biomass. The response of the test to DTPA alone (figure C.3) showed that DTPA did not interfere with the measurement of released cellular material, as DTPA alone does not absorb at 260 nm. For comparative purposes, the response of the test to the addition of DTPA in the presence of biomass resulted in the release of 45% of the amount of material released from the same biomass by boiling (figure Cl ) , and 186% of the amount of material released from the biomass alone during the test (figure C.2). 187 Boiled biomass Figure C. 1: Effect of boiling biomass for 10 minutes: Marked Wavelengths: Test absorbance: 260 nm, absorbance = 0.82219 Maximum absorbance: 208 nm, absorbance = 2.2807 Buffer alone Figure C.2: Effect of buffer alone: Marked Wavelengths: Test absorbance: 260 nm, absorbance = 0.19933 Maximum absorbance: 196 nm, absorbance = 0.4619 188 DTPA (1 g/L) Control (no biomass) . , , , f ^ ^ ^ ^ ^ y ^ f ^ x ^ m% ssa *ao Figure C.3: Effect of DTPA without biomass: Marked Wavelengths: Test absorbance: 260 nm, absorbance = 0.04665 Maximum absorbance: 208 nm, absorbance = 2.2181 1 g/L DTPA (with biomass) i«c i , 3 ? 3 * l r 5 •3<*a9«.i "i£aS S B 8 m> sat Figure C.4: Effect of 1 g/L DTPA on biomass: Marked Wavelengths: Test absorbance: 260 nm, absorbance = 0.36940 Maximum absorbance: 208 nm, absorbance = 2.3182 189 A P P E N D I X D : E N D O G E N O U S O U R R E S P O N S E T O D T P A UNDER CONDITIONS O F VARYING M E T A L S The addition of DTPA to activated sludge biomass often resulted in an increase in the endogenous OUR, as discussed in section 5.6. Figures D. 1 to D.5 have been included to provide additional evidence of the almost universally observed increase in endogenous respiration rate. Taken as a whole, this set of data confirms the hypothesis that it is the presence of the chelant, independent of whether it is bound to a protective or a toxic metal, which causes an increase in the maintenance energy expended by the cells. Magnesium, found to be one of the metals which was protective against DTPA, appeared to decrease the endogenous oxygen uptake rate of the biomass below that of the control biomass (without magnesium). However, magnesium did not completely prevent the elevation of endogenous OUR upon the addition of DTPA, as the ratio OUR/OUR0 remained consistently greater than 1 (figure D. 1). In general, OUR enhancement appeared to be a function of magnesium concentration, as can be observed in figure D. 1. On average, the samples with greater magnesium concentrations exhibited OUR/OUR0 ratios closer to 1, whereas the OUR/OURg ratios of the samples with less magnesium were higher. The presence of iron, found to be somewhat protective of biomass activity against DTPA, also did not significantly affect the rise in endogenous OUR exhibited upon addition of DTPA (figure D.2). Calcium, the metal found in this study to be most protective of activated sludge biomass against DTPA, significantly reduced, and at higher concentrations eliminated, the rise in endogenous OUR exhibited upon addition of DTPA (figure D.3). This lack of endogenous OUR rise may partially explain the greater metabolic activity observed with calcium and DTPA (section 5.6). The remainder of the explanation lies in the protective action of the calcium against the damaging effects of the chelant. 190 Manganese, zinc, and at low concentrations, copper, exhibited endogenous OUR responses similar to magnesium and iron (figures D.4 and D.5). The ratio of endogenous OUR/OUR 0 remained, in general, over 1 upon DTPA addition, and may have been somewhat reduced at higher metal concentrations. 191 o 1-1 rr ( o 0 .8 -0 .6 -0 .4 -0 .2 -0 -0 0.5 1 1.5 2 2.5' 3 3.5 4 4.5 DTPA (g/L) Figure D . 1: Endogenous O U R with magnesium, M g S O ^ concentration (mM): O A , 1.9 • , 2 A, 3.9 • , 6 • , 9.9 O , M g C L concentration (mM): 7 • . 0.6-0 . 4 J 0 . 2 -0 1—1—1—1—I—1—1—1—1—I—1—1—1—1—I—r- i—1—1—1—1—1—1—1— 0 0.5 1 1.5 2 2.5 DTPA (g/L) Figure D.2: Endogenous O U R with iron, FeCl^ concentration (mM): 0 • , 0.06 • , 0.5 • , 1.8 • , 5.3 A . 192 0.6-0.4-0.2-0 — i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i M | i i i i 0 0.5 1 .1.5 2 2.5 3 3.5 4 DTPA (g/L) Figure D.3: Endogenous OUR with calcium, CaC^ concentration (mM): 0 • , 0.5 • , 3.3 12.9 • , 13.1 • . Figure D.4: Endogenous OUR with manganese, MnSO^ concentration (mM): 0 • , 2.8 • , 24.4 • . 2 0.6-0.4-E1 0 2 -0 I i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 DTPA (g/L) Figure D.5: Endogenous OUR with zinc and copper, ZnSO^ concentration (mM): 0 • , 0.2 • , 1.5 • , CuSO. concentration (mM): 0.02 • , 5.7 • . 194 


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