Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Near field mixing of a vertical buoyant jet in a shallow crossflow : implications on adsorption and flocculation Gomm, Leslie 1999

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

Item Metadata

Download

Media
831-ubc_1999-388905.pdf [ 10.93MB ]
Metadata
JSON: 831-1.0228836.json
JSON-LD: 831-1.0228836-ld.json
RDF/XML (Pretty): 831-1.0228836-rdf.xml
RDF/JSON: 831-1.0228836-rdf.json
Turtle: 831-1.0228836-turtle.txt
N-Triples: 831-1.0228836-rdf-ntriples.txt
Original Record: 831-1.0228836-source.json
Full Text
831-1.0228836-fulltext.txt
Citation
831-1.0228836.ris

Full Text

NEAR FIELD MIXING OF A VERTICAL BUOYANT JET IN A SHALLOW CROSSFLOW: IMPLICATIONS ON ADSORPTION AND FLOCCULATION by Leslie Gomm B.A.Sc , Queen's University, Kingston, Ontario, 1985 M.Eng., University of British Columbia, Vancouver, British Columbia, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies Department of Civil Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A March, 1999 © Leslie Gomm, 1999 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 (~\i\>M FT nc;V n f p . r ^ r The University of British Columbia Vancouver, Canada Date / \ ? ^ 1 9 / ° ^ DE-6 (2/88) Abstract The behaviour and movement of pulpmill pollutants discharged into the Northern Fraser River is of significant concern due to their potential impact on this valuable aquatic ecosystem. The shallow receiving water can influence the mixing and subsequent dilution of these discharges. The association of contaminants with suspended sediment, either by direct adsorption or flocculation of contaminated solids discharged with the effluent (biosolids), also affects pollutant fate. This study examined the effects of a shallow crossflow in the near field mixing of a vertical buoyant jet, specifically dilution and trajectory. Physical mixing experiments were carried out in a shallow ambient current over a range of conditions similar to those seen in the Fraser River, specifically peak and low flow conditions. The dilution and trajectory results were then compared to those predicted by CORMLX1. The mechanism of association of contaminants with suspended sediment under these near field conditions was also investigated. A jet classification scheme was developed based on the behaviour of the jets in the shallow crossflow. Jets were classified to be Bottom, Intermediate or Surface Jets. Bottom Jets were influenced primarily by interaction of the jet with the bottom boundary layer, resulting in significantly higher levels of dilution and possible bottom attachment. The mixing of Intermediate Jets was more complicated due to interaction with both the top and bottom boundaries: the free surface inhibited mixing while interactions with the boundary layer enhanced mixing. Surface Jets were drastically affected by the free surface, with a reduction in dilution due to impingement on the free surface. The C0RMLX1 model was found to be unsuitable for predicting the dilution in this application since it does not consider the effects of either the free surface or the bottom boundary layer on jet mixing. Adsorption was found to play a limited role in the near field region. Of greater importance, is the potential for flocculation of biosolids with suspended sediment. The most important parameter in predicting where the conditions for this increased flocculation will occur was the ratio of the number of biosolid particles to the number of suspended sediment particles. in Table of Contents Abstract ii Table of Contents iv List of Figures vii List of Tables x List of Principal Symbols xi Acknowledgements xv 1 Introduction 1 1.1 Scope of the study 3 1.1.1 Pulpmills and their Pollutants 3 1.1.2 The Receiving Environment 5 1.2 Statement of the problem 7 1.3 Objectives of the Project 8 2 Literature Review 15 2.1 Discharge Mixing theory 15 2.1.1 Simple Jets, Plumes and Buoyant Jets 17 2.1.2 Vertical Jets in Uniform Crossflows 20 2.1.2.1 Chu and Goldberg (1974) 23 2.1.2.2 Wright (1977) 24 2.1.3 Vertical Discharges in Shallow Crossflows 30 2.1.3.1 Bifurcation 36 2.1.4 CORMLX Mixing Model 38 2.1.5 Summary 40 2.2 Association of Contaminants with River Sediment 41 2.2.1 Sorption to River Sediments 41 2.2.1.1 Octanol/Water Partition Coefficient (Kow) 45 2.2.1.2 Effect of Ionization on Sorption 47 2.2.1.3 Other Factors Affecting Sorption 50 2.2.1.4 Summary 52 2.2.2 Biosolids and Their Association with River Sediment 53 2.2.2.1 Characteristics of Biosolids 53 2.2.3 Fate of Biosolids in the Receiving Environment 58 2.2.3.1 Flocculation 59 2.2.3.2 Stability of Biosolids 61 2.2.3.3 Stability of River Sediment 62 2.2.3.4 Destabilization/Aggregation of Colloids 63 2.2.3.5 Effect of Temperature 67 2.2.4 Flocculation of River Sediment in the Presence of Pulpmill Effluent 69 2.2.5 Summary 73 3 Experimental Procedures 83 3.1 Near Field Mixing Experiments 83 3.1.1 Description of Experiments 83 3.1.2 Experimental Facilities 84 3.1.3 Temperature Measurement 85 3.1.4 Flow Visualization 85 3.1.5 Flow Characterization 86 3.1.6 Experimental Procedure 87 3.1.7 Characterization of the Ambient Flow 87 3.1.7.1 Vertical Velocity Profiles and Boundary Layer Determination 88 3.1.7.2 Horizontal Velocity Profile and Determination of Sidewall Effects 89 3.1.8 Mixing Experiments 90 3.1.9 Validation of video/temperature maximum 91 3.2 Experimental Determination of Adsorption for 3,4,5 TCG and D H A 92 3.2.1 Description of Experiments 92 3.2.2 Characterization of Fraser River Sediment 93 3.2.3 Determination of Kp from K o W 94 3.2.4 Determination of Kp - Batch Adsorption Studies 94 3.2.5 Materials 95 3.2.6 Analytical Techniques 96 3.2.6.1 Guaiacol Analysis - GC Method 96 3.2.6.2 Guaiacol Analysis - Autoanalyzer Method 96 3.2.6.3 Resin Acid Analysis 97 4 Results and Discussion 105 4.1 Near Field Mixing Experiments 105 4.1.1 Results, Analysis and Discussion 107 4.1.1.1 Jet Trajectory 109 4.1.1.2 Dilution 110 4.1.2 Comparison to Previous Results I l l 4.1.2.1 Trajectory I l l 4.1.2.2 Dilution 114 4.1.3 Bifurcation 120 4.1.4 Buoyancy Effects 121 4.1.5 Comparison of Jet Types 123 4.1.6 Application of Jet Types to Fraser River Conditions 123 4.1.7 Comparison with CORMLX1 Output 125 4.1.8 Summary of Near Field Mixing Experiments and CORMLX Modeling .... 128 4.2 Batch Adsorption Experiments 131 4.2.1 Nature of Fraser River Sediment 131 4.2.2 Determination of Kow for Dehydroabietic Acid 132 4.2.3 Determination of Kp and Theoretical Adsorption using K Q W 133 4.2.4 Results of Batch Adsorption Studies 136 4.2.4.1 Dehydroabietic Acid 136 V 4.2.4.2 Chlorinated Guaiacols 137 4.2.5 Summary of Kp and Adsorption Determination 138 5 Flocculation of Biosolids/Sediment in the Near Field Region 177 5.1 Factors Affecting Flocculation in Near Field Region 177 5.1.1 Temperature 180 5.1.2 Turbulence Intensity 180 5.1.3 Concentration 181 5.2 Calculation of Flocculation Potential in the Near Field Mixing Zone 183 5.3 Application to Pulpmill Discharges in Fraser River 188 6 Conclusions and recommendations 199 6.1 Near Field Mixing Experiments 199 6.2 Interactions with Sediment in the Near Field Region 202 6.3 Recommendations for Further Research 204 7 References 206 APPENDIX A HPLC Determination of Kow 217 APPENDIX B Temperature Data acquisition Programs 220 APPENDIX C Thermistor Calibration Data and Curves 225 APPENDIX D Chemical Analytical Procedures 257 APPENDIX E Details of Cross-section Video Analysis 260 APPENDIX F Experimental Data - Mixing Experiments 261 APPENDIX G Details of Calculation of Buoyancy Stabilizing Ratios 287 APPENDIX H Details of Analysis of Contaminant Levels in CANFOR Effluent.. 288 vi List of Figures Figure 1-1 Mixing Zone Immediately Downstream of Caribou Pulp M i l l located in Quesnel, British Columbia 10 Figure 1-2 Near and Far Field Regions in a Mixing Zone 11 Figure 1-3 Mean Monthly Fraser River Hydrograph measured at Shelley 12 Figure 1-4 Increased 4,5-Dichlorocatechol Levels in Suspended Sediment during Spring Freshette 13 Figure 1-5 Simultaneous Processes Acting on Contaminants in the Near Field Mixing Region 14 Figure 2-1 Simple Jet 74 Figure 2-2 Schematic of a Vertical Jet in a Crossflow 75 Figure 2-3 Depth Classifications for Jets in a Crossflow 76 Figure 2-4 Flow Regimes for Vertical Jets in a Shallow Crossflow 77 Figure 2-5 Bifurcation of a Vertical Jet in a Crossflow 78 Figure 2-6 CORMLX1 Flow Classifications for Vertical Buoyant Jets in Crossflows 79 Figure 2-7 Freudlich and Langmuir Isotherms 80 Figure 2-8 Negative Colloid with Electrostatic Field 81 Figure 2-9 Interparticle Forces Acting on a Particle 82 Figure 3-1 Layout of Experimental Facilities 98 Figure 3-2 Vertical Non-dimensional Ambient Velocity Profiles 99 Figure 3-3 Non-dimensional Vertical and Longitudinal Velocity Fluctuations 100 Figure 3-4 Vertical Reynolds Stress Distribution 101 Figure 3-5 Horizontal Non-dimensional Ambient Velocity Profiles 102 Figure 3-6 Range of Experimental Conditions Qn/d and R) Compared to Hodgson (1991) 103 Figure 3-7 Comparison of Centreline Temperature and Video Profiles at 5 and 27 cm Downstream 104 Figure 4-1 Colour Images of a) Bottom Jet Js=0.08, b) Intermediate Jet Js=0.37 and c) Surface Jet Js=0.58 143 Figure 4-2 Jet Classification Diagram 144 Figure 4-3a) Jet Centreline Trajectories (zc/hi) at (i) hi/d = 5.42, (ii) hi/d = 6.92 145 Figure 4-3b) Jet Centreline Trajectories (z c/hO at (i) h f /d = 7.16, (ii) h,/d = 8.5 146 Figure 4-4 Jet Trajectory at Various Jet Strength Ratios (Js) 147 Figure 4-5a) Minimum Centreline Dilution (S) at (i) hi/d = 5.42, (ii) h^d = 6.92 148 Figure 4-5b) Minimum Centreline Dilution (S) at (i) hi/d = 7.16, (ii) h)/d = 8.5 149 Figure 4-6 Minimum Centreline Dilution (S) Prior to Surface Effects, hi/d = 8.5 150 Figure 4-7 Comparison of Jet Trajectory and Dilution for Surface Effects 151 Figure 4-8 Jet Centreline Dilution (S) as a Function of Jet Strength (Js) 152 Figure 4-9 Jet Centreline Trajectory (z c/zm) at Various Jet Strengths (Js) compared to Trajectories predicted by Wright (1977) and Hodgson (1991) using Equation 4.2 153 Figure 4-10 Actual Jet Trajectory (zc/hl) compared to that Predicted (zp/hl) using Equation 4.5 154 vii Figure 4-11 Contour Plots of Jet Thickness for Bottom Jets a) Experiment 5 Js=0.08 and b) Experiment 12 Js=0.19 155 Figure 4-12 Contour Plots of Jet Thickness for Intermediate Jets a) Experiment 6 Js=0.38 and b) Experiment 14 Js=0.39 156 Figure 4-13 A Comparison of Jet Dilution Data (S/R) to Wright (1977) predicted using Equation 4.7 and C d = 0.76 (concentration trajectory) and 0.97 (photographic trajectory) 157 Figure 4-14 Pre-Surface Effects Dilution Coefficient (C d) as a function of J s for Experiments 1 through 15 158 Figure 4-15 M D F F Dilution Coefficient (Cmdff) as a function of J s 159 Figure 4-16 Dilution (S/R) Jet Classification Diagram 160 Figure 4-17 Comparison of Minimum Jet Centreline Dilution data (S) to Dilution (Sp) predicted by Equation 4.9 161 Figure 4-18 Comparison of Jet Dilution data (S) to that predicted by Equation 4.11 (Hodgson, 1991) 162 Figure 4-19 Lateral Dilution Contour Plots for Experiments 16-19 163 Figure 4-20 Cross-section Width Images at x = 32cm downstream 164 Figure 4-21 Jet Trajectory (zc/hi) (a) and Dilution (S) (b) at varying Froude numbers and similar Jet Strength ratios 166 Figure 4-22 Jet Trajectory (zc/hi) (a) and Dilution (S) (b) at varying Jet Strength ratios and similar Froude numbers 167 Figure 4-23 Buoyancy Effects on Jet Trajectory (zc/hi) (Exp. 20, 21 and 22) 168 Figure 4-24 A comparison of Jet Dilution (1/S) for Bottom, Intermediate and Surface Jets 169 Figure 4-25 Comparison of Dilution Data to CORMLX1 output and Equation 4.9 for a) Bottom, b) Intermediate and c) Surface Jets 170 Figure 4-26 Relative Error (AS/S) in dilution predicted by CORMLX1 for various jet strength ratios at 5, 27, 32 and 42cm downstream 171 Figure 4-27 Comparison of dilution data to CORMLX 1 output and Equation 4.9 for Experiment 1 (Js=0.16) and Experiment 5 (Js=0.08) 172 Figure 4-28 Results of Batch Adsorption Studies for Dehydroabietic Acid 173 Figure 4-29 Results of Batch Adsorption Studies for 3,4, 5 - Trichloroguaiacol (uncorrected for i nterference) 174 Figure 4-30 Interference Contribution for 3, 4, 5 - Trichloroguaicol Adsorption 175 Figure 4-31 Corrected Adsorption for 3,4, 5 - Trichloroguaiacol 176 Figure 5-1 Depth Averaged Turbulence Intensity at J s = 0.08, 0.35 and 0.58 191 Figure 5-2 Flocculation Potential Product at a) Peak, b) Mid and c)Low Flow Conditions 192 Figure 5-3 Flocculation Potential Product at a) Mid and b) Flow Conditions with [sed]=500 mg/1 193 Figure 5-4 Flocculation Potential Product at Low Flow Conditions with d b i 0 = l l ( im 194 Figure 5-5 Comparison of Number Concentration Contribution (NCC) for N b i 0 / N s e d = l and N b i o /N s e d =10 195 Figure 5-6 Coagulation Regions in a Plume (Holman, 1986) 196 viii Figure 5-7 Comparison of Number Concentration Contribution (NCC) with Field Data (Ad 5 0) (October, 1996) 197 Figure 5-8 Number of Biosolid:Sediment Floes formed at Low, Mid and Low Flow Conditions 198 Figure A - l Standard Curve for HPLC Determination of KoW 219 Figure E - l Setup for Cross Section Video Measurements for Bifurcation 260 IX List of Tables Table 2-1 Relationships for Simple Jets (Fischer et al, 1979) 18 Table 2-2 Relationships for Simple Plumes (Fischer et al, 1979) 19 Table 2-3 Wrights Coefficients 30 Table 2-4 Constants a and b for Equation 2.53 (Schwarzenbach et al, 1993) 47 Table 2-5 Biosolids Size Fractionation (Zanella et al, 1978) 56 Table 2-6 TSS Concentration in Fraser River (Krishnappan, 1994) 70 Table 2-7 Median Settling Velocities from Evans (1996) 72 Table 3-1 Chemical Characteristics of 3,4,5 T C G and D H A 92 Table 4-1 Jet Classification Criteria 106 Table 4-2 Summary of Experimental Conditions and Length Scales 108 Table 4-3 Wright's (1977) M D F F Coefficients 115 Table 4-4 Dilution Coefficients 116 Table 4-5 Comparison of Dilution (1/S) for Three Jet Types 123 Table 4-6 Particle Size Fractionation 131 Table 4-7 Log(KoW) using HPLC 132 Table 4-8 Calculated log(KoW) using Aqueous Solubility 133 Table 4-9 Calculated Partition Coefficients (K p) 134 Table 4-10 Calculated Adsorption... 136 Table 4-11 Results of D H A Adsorption Experiment 137 Table 5-1 Typical River and Discharge Conditions 179 Table A-1 Wavelengths for HPLC Determination of KoW 218 Table A-2 Log(KoW) for Standards used in HPLC 219 Table F - l Overview of Physical Mixing Experimental Conditions 262 Table G - l Variables used for Determination of Buoyancy Stabilizing Ratios 287 Table H - l Results of Analysis of Total and Dissolved Portions of CANFOR Effluent 288 List of Principal Symbols A . Jet and Mixing A port area Ay proportionality constant for M D N F used by Wright (1977) A2 proportionality constant for M D F F used by Wright (1977) A3 proportionality constant for BDNF used by Wright (1977) A4 proportionality constant for BDNF used by Wright (1977) B initial buoyancy flux C concentration or excess temperature C0 initial concentration or excess temperature cd presurfacing dilution coeffcient Cm maximum centreline concentration Cmdff M D F F dilution coefficient Q concentration of pollutant in solid phase CsDF surface dominated flow trajectory coefficient (Hodgson, 1991) Cw concentration of pollutant in aqueous phase Ci coefficient for Wright's (1977) centreline M D N F trajectory equation c2 coefficient for Wright's (1977) centreline M D F F trajectory equation C3 coefficient for Wright's (1977) centreline BDNF trajectory equation c4 coefficient for Wright's (1977) centreline BDFF trajectory equation c5 coefficient for Wright's (1977) centreline M D N F dilution equation c6 coefficient for Wright's (1977) centreline M D F F dilution equation c7 coefficient for Wright's (1977) centreline BDNF dilution equation c8 coefficient for Wright's (1977) centreline BDFF dilution equation d jet diameter e roughness height densimetric Froude number f friction factor from Moody diagram 8 gravitational acceleration go' initial effective gravitation acceleration H total depth of ambient water hi depth of water above jet exit Js jet strength ratio h jet characteristic length scale IM jet/plume characteristic length scale mi exponent for SDF trajectory from Hodgson (1991) xi m momentum pulse M initial momentum flux Q initial volume flux R velocity ratio (U/Ua) Re Reynolds number Ro jet Richardson number S minimum centreline dilution sP dilution predicted using Equation 4.9 ambient fluid temperature Tc maximum centreline temperature Tm discharge fluid temperature Ua ambient crossflow velocity initial jet velocity um maximum jet centreline velocity w velocity fluctuations in the longitudinal component of velocity u shear velocity w flume width w velocity fluctuations in the vertical component of velocity X horizontal distance downstream Xi downstream point of impingement XT downstream transition point from momentum to buoyancy dominated flow Y initial mass flux z vertical distance Zb jet/crossflow buoyancy length scale Zc vertical jet centreline trajectory measured from the jet exit Zm jet/crossflow momentum length scale zP predicted jet centreline trajectory ae entrainment rate coefficient 8 boundary layer thickness V kinematic viscosity Pa ambient fluid density Ap density difference between ambient and discharge fluid B. Adsorption and Flocculation c concentration Cs concentration of pollutant in solid phase Cw concentration of pollutant in aqueous phase cm3 volume of solution in adsorption calculations D dielectric constant of the liquid Ds distribution ratio for sorption of ionic compound dbio mediam particle diameter for effluent biosolids dc thickness of layer surrounding shear surface where charge is effective dsed median particle diameter for suspended sediment di diameter of particle 1 d2 diameter of particle 2 dn diameter of particle 1 and 2 when dj=d2 d&50 median particle diameter foe fraction of organic carbon in solid G rms velocity gradient 8 gravitational acceleration 8s amount of solid in adsorption calculations I ionic strength of a solution K equilibrium adsorption constant for Freudlich Isotherm Ka ionization constant Kf flocculation rate constant organic carbon normalized distribution coefficient Kos rate constant for orthokinetic flocculation due to shear Kot rate constant for orthokinetic flocculation due to turbulent motion Kow octanol-water partition coefficient KP equilibrium partition coefficient Kpk rate constant for perikinetic flocculation k Boltzmann's constant h characteristic length scale jet/plume characteristic length scale nij exponent for SDF trajectory from Hodgson (1991) m momentum pulse M initial momentum flux N rate of contact for flocculation Nbio number of biosolid particles in solution Nsed number of suspended sediment particles in solution N rate of formation of lasting contacts NC number concentration ratio NCC number concentration contribution to Floe Potential Product n number concentration for particle 1 and 2 when ni=n2 nf constant for slope of Freudlich Isotherm ni number concentration for particle 1 n2 number concentration for particle 2 pKa ionization constant (-\og(Ka) Q initial volume flux q charge of particle per unit area R velocity ratio (U/Ua) S minimum centreline dilution Saq aqueous solubility S specific gravity of a particle T absolute temperature Tc temperature contribution to Floe Potential Product T 1 mm minimum temperature in degrees Kelvin TI turbulence intensity TT turbulence intensity contribution to Floe Potential Product VR repulsive force between two double layers Vso median particle settling velocity W0 weight of compound originally in solution prior to adsorption Ws weight of compound associated with solid after adsorption Ww weight of compound remaining in solution after adsorption a collision efficiency factor <P fraction of total suspension volume occupied by particles P density of fluid surrounding particle Pbio biosolid particle density Ps density of spherical particle Psed suspended sediment particle density absolute viscosity V kinematic viscosity c zeta potential Acknowledgements I would like to thank my supervisors, Dr. Greg Lawrence and Prof. Jim Atwater, for their guidance, encouragement and financial support. I am thankful to all those who help with various aspects of this project from the onset. Susan Harper and Paula Parkinson for their invaluable assistance in the Civil Environmental Engineering Laboratory and Kurt Nielson from the Civi l Engineering Hydraulics laboratory for building parts for the experimental apparatus. Special thanks goes to my family for their continued support, patience, and provision of free daycare. XV 1 Introduction The Fraser River supports one of the most valuable salmon runs in Canada and the world. The majority of British Columbia's population and industrialization is also within the Fraser River Basin. Of significant concern is the introduction of pulpmill contaminants into the northern reaches of the river. This region provides both important spawning and overwintering habitat for salmon as well as habitat for other aquatic organism, birds and mammals. The discharge of these contaminants has possible detrimental effects in both the Northern Fraser River and downstream. Studies have shown some pulpmill constituents present in the system, including sediment and biota, as far down as the estuary, 800 km downstream (Carey, 1988). Not only are the pollutants transported in the water column, they are also transported in association with suspended sediment as a result of interactions between the river sediment and compounds associated with both the solid and liquid constituents in the effluent. The extent to which these processes take place, determines the amount of pollutant available for potential detrimental environmental effects; where they will occur and which biota will be affected. The prediction of the dilution and location of a discharge in a river will increase the ability to predict environmental effects in a system. This in turn can be used when setting water quality guidelines. Presently, regulations are based on the assumption of instantaneous mixing of the discharge and neglect specific characteristics of the receiving water. In actuality, there is a zone between the point of discharge and the location where the effluent is completely mixed and the final objective concentration is achieved. An example of the 1 region prior to complete mixing is shown in Figure 1.1 This is known as the "mixing zone", and may extend for many kilometers downstream of the discharge. Within this mixing zone, pollutant concentrations may exceed water quality guidelines and pose a potential hazard to aquatic biota. Maximizing dilution in the region near the mixing zone reduces the extent of this zone and minimizes environmental impacts. This is commonly done by the use of a diffuser. A diffuser discharge has a substantial amount of momentum associated with it, typically much higher than that of the receiving water. The discharge mixes initially due to the momentum and turbulence of the discharge itself and later by the turbulence of the receiving water, the former being controlled by the actual design of the discharge. For a discharge into a flowing environment, the behaviour of the discharge near the jet exit is controlled by the initial conditions of the discharge. This is called the Near Field region (Figure 1.2). Initial characteristics such as momentum, buoyancy and geometry dominate the jet dilution and trajectory. Within the near field region, complete vertical mixing may be achieved, typically occurring within a distance equal to 50 to 100 times the river depth. The buoyancy of the discharge (positive or negative) can increase the distance due to the formation of a density stratification. A significant amount of mixing can be achieved in the near field, with an initial dilution on the order of 100 or more (Yotsukura et al., 1976). Farther downstream, the initial conditions no longer dominate the flow behaviour. This Far Field region may or may not occur after vertical mixing has been achieved. At this point, ambient conditions such as turbulence and residual buoyancy effects begin to dominate. The far field region extends until the discharge is completely mixed across the entire depth and 2 f width of the river. Downstream of the mixing zone, further dilution occurs due to increased stream loading from runoff and incoming tributaries, although additional contaminant inputs are also likely from other industrial and municipal discharges. 1.1 Scope of the study 1.1.1 Pulpmills and their Pollutants There are five pulpmills discharging at four locations in the Northern Fraser River, three in Prince George and two in Quesnel. Four of these mills are bleached kraft mills while Quesnel River Pulp is a TMP (Thermal-Mechanical) and CTMP (Chemical-Thermal-Mechanical) mill. They all provide secondary effluent treatment for BOD (Biochemical Oxygen Demand), solids and toxicity removal as required by the Waste Management Branch permit. The actual amount of suspended sediment input from the mills is insignificant in comparison to the levels naturally present in the river. The major concern regarding effluent solids is that they consist of biosolids (biological solids), which may sorb a spectrum of contaminants during the treatment process. Pulpmills discharge the largest portion of industrial effluents in the Fraser River, impacting the entire reach of the river (Scheier et al., 1991). The impact of pulpmill effluent on an aquatic environment can be due to either chemical constituents in the discharge or non-chemical components such as BOD and suspended solids (McLeay, 1987). In the past, before secondary treatment methods were used, most environmental concerns were with suspended solids, BOD, and the acute toxicity of the effluent to aquatic biota. More recently, other properties such as sublethal toxicity, bioaccumulation and fish tainting have been receiving 3 considerable attention (Wilson et al., 1992; Carey et al., 1993). Although many of the contaminants are present in only trace amounts, it is their persistence in the aquatic environment and potential for bioaccumulation that is a concern. The composition of pulpmill effluent is complex and includes both high and low molecular weight compounds (Suntio et al., 1988; Lehtinen, 1992). There are several organic and inorganic compounds derived from wood, process chemicals and from interactions between wood derivative and chemicals present in pulpmill effluent. Many of these compounds have not yet been identified and therefore their toxicity is not known. The most toxic compounds found in pulpmill effluents are chemicals found in wood or derivatives of these chemicals which are formed during the pulping and bleaching process (Branion, 1991). Resin and fatty acids occur naturally in wood and are present in both bleached and unbleached mill effluent. Chlorinated organics are generated from the lignin breakdown products during chlorine bleaching. Generally, biological treatment of pulpmill effluent successfully removes acute toxicity, although this can be quite variable depending on mill operation. This does not remove the possible long-term environmental effects due to the presence of sublethal concentrations and bioaccumulation effects in the food web (McLeay, 1987). Of key concern is the low water solubility of many pulpmill contaminants including chlorinated organics and resin acids. Due to these hydrophobic characteristics, there is a tendency for these compounds to sorb out of solution and become associated with sediment and organic material. This could be sorption onto biological solids in the treatment system prior to discharge or direct sorption to the sediment once discharged into the river. 4 Due to the complexity of pulpmill effluent, this study focused on two particular constituents; 3,4,5 trichlorinated guaiacol and dehydroabietic acid (DHA). These compounds are persistent and are found in both water and sediment samples taken downstream of the pulpmill discharges in the Fraser River (Voss and Yunker, 1983; Sekela et al., 1995; Prahacs, 1994). Chloroguaiacols are found in bleached mill effluents, while DHA is present in both mechanical and bleached mill effluents. In a study on the Peace River system, both these compounds were commonly found in sediment and water samples (Monenco, 1991). Elevated concentrations of guaiacols have also been found in overwintering chinook salmon downstream of Prince George (Schreier et al., 1991). The elevated temperature of pulpmill discharges also plays an important role, especially during the colder, low flow winter period. Pulpmill discharges tend to be between 25 to 35 degrees Celsius (Dwernychuk, 1989), which can result in a possible temperature difference of 35 degrees Celsius during the critical low flow periods. At these elevated temperatures, the effluent can behave buoyantly which has the potential to influence the mechanisms of mixing. As well, temperature changes can drastically affect the various processes acting on the discharges including reactions rates and flocculation. 1.1.2 The Receiving Environment The Fraser River is among the largest rivers of Canada, draining one quarter of the province of B.C. (Dorcey et al., 1991). It has the second greatest mean annual flow (3792 m3/s) and the fifth largest drainage basin (234,000 km2). The flow of the Fraser River is driven predominantly by the melting snowpack in the 5 headwaters. This results in large streamflow variations throughout an annual cycle. Peak flows occur during the spring freshet and low flows occur in winter as shown in the hydrograph in Figure 1.3. At the Environment Canada Water Survey Station at Shelley, 20 km upstream of Prince George, the average winter low flow is 190 m3/s and the peak flow is 2170 m3/s. As a result of these seasonal variations in flow, the depth, width and velocity at discharge locations vary seasonally. The river temperature also varies seasonally, ranging from 0 degrees Celsius during the low flow winter months to temperatures of 16 to 20 degrees Celsius in August (Hall et al., 1991). The Fraser River is also known for its extremely high sediment load (Ages, 1985). Northcote and Larkin (1989) called the Fraser a "silt-laden river with large seasonal variations in response to varying discharge rates". Approximately eighty six percent of the suspended sediment load is transported during the period from May to August (Zyrmiak et al., 1986). Between Hansard and Hope, the suspended sediment concentration varied from 1.0 mg/1 in December and January to 1650 mg/1 during May and June (Dwernychuck, 1989). Approximately 60% of this load is of the clay/silt size. As well the organic fraction of the sediment also varies seasonally with an increased content during the low flow period (Prahacs, 1994; Sekelaetal. 1995). This suspended sediment provides a secondary mode of transport of pollutants in a silt-laden river system like the Fraser. The greatest potential is with the clay/silt size fraction due to its large available specific surface area. Due to the small size of this sediment, it is carried for some distances downstream, only settling out during low flow periods or in areas of minimal 6 flow. This settled contaminated sediment is later flushed out in periods of high flow. Such was the case seen in a field monitoring program carried out by Sekela et al. (1994), which measured an increase in contaminant concentration in suspended sediment during the onset of spring freshet (Figure 1.4). 1.2 Statement ofthe problem There is a concern about overwintering salmon being exposed to elevated levels of contaminants in the winter low flow period when available dilution and mixing depth is minimal (Birtwell et al., 1988). This exposure may be increased by a potential attraction of the salmon to the warm, nutrient rich plume. Determining the fate of pulpmill contaminants in the Fraser River is essential to understanding this and other potential impacts on the receiving environment. Of primary importance is the understanding of what is happening to the effluent within the near-field region where a significant amount of dilution can be achieved. Within this near field region, constituents in the effluent are present at elevated concentrations and are undergoing complex mixing processes that influence the subsequent dilution and location of the discharge in the river. In order to quantify the distribution of pollutants in the mixing zone, it is necessary to understand the various factors affecting the mixing (dilution) of a buoyant discharge in a shallow receiving water. The most basic case is that of a vertical buoyant jet. While there has been a substantial amount of work done investigating this type of discharge in unconfined environments, a limited amount of work has been done on buoyant jets in confined crossflows, similar to the discharge conditions in the Fraser River. The key variables affecting the mixing of these discharges are: 7 • the proximity of both the bottom and free surface • the seasonal variation in the flow rate, especially during the critical low flow period • the elevated temperature (buoyancy) of the effluent. Once the distribution and subsequent dilution of this type of discharge is quantified, it is possible to look at the important simultaneous physical, chemical and biological processes that further affect the fate of pulpmill contaminants in the Fraser River (Figure 1.5). The most important of these processes is the potential sorption of contaminants with ambient river sediment and the flocculation of biological solids with the river sediment. 1.3 Objectives ofthe Project The present study examines the near-field mixing characteristics of a vertical buoyant jet discharging into a shallow crossflow. It then investigates the potential for association of specific hydrophobic compounds with the ambient sediment within this region. Specifically, the objectives were: • to study the key factors affecting the near-field mixing of vertical buoyant discharges under various discharge scenarios, including seasonal variations of ambient velocity and depth. • to compare these results with those empirical relationships used to describe the trajectory and dilution of the discharges and that predicted by CORMLX (Cornell Mixing Zone Expert System). • to investigate the potential for association of hydrophobic compounds with the ambient river sediment either through direct sorption or flocculation of contaminated solids within the near-field region. • to apply the above to the specific discharge conditions seen in the Northern Fraser FLiver in order to get a good understanding of the factors affecting the distribution of hydrophobic pulpmill contaminants in the Fraser River Basin. The research consisted primarily of physical and chemical experiments. A physical mixing model was setup in an experimental flume and run under various discharge scenarios similar to those seen in the Northern Fraser River. Batch adsorption studies were also conducted using both DHA and various chloroguaiacols and uncontaminated Fraser River Sediment. The results of these experiments were then applied to specific discharge conditions in the Fraser River to provide for a better understanding of the fate of pulpmill contaminants. Specifically discussed are factors that affect the overall dilution at various times throughout the year and the mechanisms by which pulpmill contaminants become associated with sediment. Due to the complexity of pulpmill effluent, this study targeted two specific compounds in pulpmill effluent; the resin acid, dehydroabietic acid and 3,4,5-trichloroguaiacol. As well, all discharge conditions and pulpmill loadings are based on the discharge from the Northwood Pulp and Timber Ltd. pulp mill, located farthest upstream on the river system. This is a kraft bleach pulp mill with secondary treatment system consisting of an Aerobic Stabilization Basin (ASB) and a pre-discharge basin. 9 Figure 1.1 Mixing Zone Immediately Downstream of Caribou Pulp and Paper Mill Co. discharge located in Quesnel, British Columbia (photo from Environment Canada). Effluent discharge rate is approximately 1.05 m3/s. 10 a) Plan View Near Field Region ; ^ - - " - — • x 1 1 _ _ _ _ _ — — '"' Surfacing of Discharge T U a ^ / Initial Jet Buovanc>. Momentum, and Geometry Dominate Ambient Condition* and Residual B u o \ a n c \ Dominate z / Near Field Region Far Field Region" ~-4 b) Profile View Figure 1.2 Near and Far Field Regions in a Mixing Zone Figure 1.3 Mean Monthly Fraser River Hydrograph measured at Shelley (Environment Canada) Suspended Sediment (mg/1) ^M4,5 PCC Concentration (ng/l) - * - F l o w r a t e (m3/s) 250 1600 1400 1200 1000 E 800 2 600 v •-» 2 •t o u. 400 200 30-Mar-93 7-Apr-93 15-Apr-93 22-Apr-93 Figure 1.4 Increased Contaminant Levels (4,5-Dichlorocatechol) in Suspended Sediment during Spring Freshet (Sekela et al., 1994) 13 Interaction between sediment and biosolids Entrainment of Ambient water and sediment X x Advection Downstream Deposition t Resuspension Pulpmill Effluent X Biosolid • Suspended Sediment Figure 1.5 Simultaneous Processes Acting on Contaminants in the Near Field Mixing Zone 14 2 Literature Review Upon discharge into a river, there are simultaneous physical, chemical and biological processes which act on the discharge. The extent to which these processes take place determine the amount of pollutant available for environmental effects, where they will occur and which biota will be effected. The primary processes acting on the contaminants in pulpmill discharges in the Northern Fraser River are: • mixing with the receiving water, • adsorption to biosolids or riverine sediment, • potential flocculation of contaminated solids with suspended sediment. The following literature review will deal with work that has been done in these areas as it pertains to the discharge of hydrophobic contaminants in a vertical buoyant jet in a crossflow. The physical mixing processes are covered in Section 2.1. The various mechanisms of association of contaminants and suspended sediment are then discussed in Section 2.2, specifically adsorption in Section 2.2.1 followed by a discussion of biosolids in Section 2.2.2 and their fate in the receiving water in Section 2.2.2. 2.1 Discharge Mixing theory In order to understand the dynamics of buoyant jets in shallow cross flows, it is important to take a few steps back and look at the theory and length scales used to describe simple jets, plumes and buoyant jets in quiescent and infinite environments. This is covered in section 2.1.1. This theory is then applied to buoyant jets in infinite and shallow cross flows in 15 section 2.1.2. Section 2.1.3 then looks at previous relevant research in the area of vertical buoyant jets in shallow cross flows. Section 2.1.4 describes the computer mixing model, CORMLX. The behavior and subsequent dilution of a turbulent jet is a function of three sets of parameters: jet parameters, environmental parameters and geometrical parameters (Fischer et al., 1979). Jet parameters include initial jet discharge conditions such as mass, momentum and buoyancy flux. Environmental parameters include ambient velocity, turbulence and receiving water density. Geometrical factors are orifice shape, configuration and proximity of the discharge to boundaries. Near the jet exit, the flow is controlled by the initial volume (Q), momentum ( A / ) and buoyancy fluxes (B) which, for a round jet are defined as: Q = Uj —a v4 , = 17.A 2.1 M=QUj=Uj2A 2.2 where Uj is the initial mean jet velocity, d is the port diameter, A is the port area, gn is the initial effective gravitational acceleration, Ap is the density difference between the discharge fluid and the ambient and pa is the ambient fluid density. Depending on the importance of the initial momentum and/or buoyancy of the discharge, the discharges can be classified as pure jets, pure plumes and buoyant jets. A pure jet has no buoyancy associated with it and the flow is driven solely by its initial momentum. A pure 16 plume has no initial momentum and the flow is driven by either a positive or negative buoyancy resulting from a density difference (Ap) between the discharge fluid and the ambient. Positive buoyancy acts vertically upwards while negative buoyancy acts downward. Buoyant jets have both initial momentum and buoyancy and are typical of many environmental discharges. In general, the flow is initially dominated by the momentum and then the buoyancy begins to dominate the mixing process when the momentum diminishes relative to the ambient current. With all buoyant discharges, buoyancy will eventually dominate the flow behaviour. 2.1.1 Simple Jets, Plumes and Buoyant Jets Simple Jets The simplest case is the discharge of a round jet into a body of quiescent ambient fluid of the same density. Simple jets such as the one in Figure 2.1 have been studied extensively. Fischer et al. (1979) provides an extensive review of the work done in this area. Within this type of jet discharge there are two distinct regions: Zone of Flow Establishment (ZFE), and the Zone of Established Flow (ZEF). In the ZFE, the shear layer between the jet fluid and the ambient fluid diminishes the velocity core (potential core) of the jet from its initial pipe velocity distribution to a Gaussian distribution. The ZFE is approximately 6 jet diameters long (Fischer et al., 1979). In the ZEF, the jet is fully established and begins to expand and entrain ambient fluid. This subsequently decreases the mean jet velocities and concentrations which both take on a self-similar profile or Gaussian distribution and can be defined in terms of a maximum centreline 17 value and a width. By dimensional analysis a characteristic length scale, lQ, is often used where: The length scale lQ provides a measure of the distance in which the initial geometry z influences the flow. As a result, when — > 10, the initial volumetric flux becomes less important in defining the flow and the momentum flux begins to dominate. Therefore, downstream flow variables such as maximum centreline velocity (Um) and maximum centreline concentration (Cm) are functions of lQ and the distance from the jet exit (z). For example: QM 1 2.5 where the function ( / ) can be determined by dimensional analysis combined with empirical data. A summary of these relationships is shown in Table 2.1. Table 2-1 Relationships for Simple Jets (Fischer et al., 1979) Parameter Relationship Maximum time-averaged velocity U m Umj[ = a.0±0.l)l/{ 2.6 Maximum time-averaged concentration C m ^ = ( 5 . 6 ± 0 . 1 ) / e / 2.7 Simple Plume In an ideal plume, there is no initial momentum or volume flux just buoyancy. As a result the downstream flow variable such as velocity (Um) and concentration (Cm) are a function of the initial buoyancy flux (5), the mass flux ( Y ) and the distance from the pipe exit ( z ) . Again by dimensional analysis in combination with experimental data, the relationships shown in Table 2.2 have been developed. Table 2-2 Relationships for Simple Plumes (Fischer et al., 1979) Parameter Relationship Maximum time-averaged velocity U m Um = (4.7 ± 0.2)5 V K 2.8 Maximum time-averaged concentration C m cm = (9.1 ± o.5) y / r V K 2- 9 Buoyant Jet A buoyant jet has both initial momentum and buoyancy compared to the receiving water. The buoyancy will act to increase the momentum flux (List, 1982). At a given distance from the jet exit, the momentum generated by the buoyancy will be larger than that due to the initial momentum of the jet. As a result, a buoyant jet typically goes from exhibiting "jet-like" to "plume-like" behaviour. For a round (axisymmetric) buoyant jet, this transition point can characterized by a characteristic length scale, lM , where 2 - 1 0 For a buoyant jet in quiescent ambient, there are two length scales that can be used to define the behaviour of the flow, lQ and lM . The first length scale describes the behaviour of jets while the second can be used as a measure of the buoyancy effects. The ratio of these two length scales is known as the jet Richardson number (R0) which is the ratio of buoyant and inertial forces. k QB^ K= = y- 2.11 19 In buoyant jets, initially momentum is the driving force and as momentum diminishes buoyancy forces come into play. After flow establishment (z > 10/ e) the initial volume flux z can be neglected and the flow is now a function of lM and z or -—. As discussed previously, *-M all buoyant jets, given sufficient time, will eventually become plumes. The dimensionless quantity y- can be used to determine approximately if and when the buoyant jet will exhibit M z jet-like or plume-like behaviour. If — » 1 the flow is buoyancy dominated and will behave z as a plume and if - — « 1 , the flow is momentum dominated and will be jet-like. If lM and lQ are of the same order of magnitude, Rn ~ 1, the flow will be immediately plume-like following flow establishment. 2.1.2 Vertical Jets in Uniform Crossflows The investigation of jets in a crossflow has become increasingly popular due to their relevance in many engineering applications including river and atmospheric discharges, cooling of turbine blades, jets in combustion units and V/STOL aircraft. As a result, there has been a considerable amount of work done in this area for discharges into infinite environments. The following is a summary of some of this work that is relevant to this project. Based on observations by Rajaratnam (1976), Keffer and Baines (1963), and Pratte and Baines (1967), three distinct regions exist in a jet deflected by a cross flowing stream (see Figure 2.2). The first region is called the potential core or zone of flow establishment (ZFE). 20 Based on Keffer and Baines (1963), when the velocity ratio (R = — ) is greater than 4, the a potential core terminates immediately above the jet exit. If R < 4 , this point is pushed downstream by the ambient current and the jet actually begins to deflect at the end of the ZFE (Keffer and Baines, 1963). The length of the potential core varies with the velocity ratio and is typically less than that of a jet being discharged into a stagnant ambient, approximately one-half that of a free jet. Within this region, only the outer perimeter of the jet is mixed with the ambient fluid. The flow spreads, slows down and curves in a downstream direction. The second region is known as the zone of maximum deflection and is characterized by the kidney-shaped cross sectional area consisting of two counter rotating vortices (Keffer and Baines, 1967). In this region, the vortices continue to grow in size but their angular velocity or circulation remains constant. A significant amount of entrainment of ambient fluid occurs within this region as well as the decay of the jet's velocity. The final region is called the vortex zone in which the vortices continue to grow but their angular velocity begins to decrease. The flow at this point is essentially parallel to the ambient current and carried along by the ambient flow. Although there is no defined end of this region, vortices have been seen to remain until approximately 1000 jet diameters downstream in a wind tunnel (Keffer and Baines, 1967). Each individual discharge scenario consists of several different regions of flow exhibiting specific characteristics,the transition between each region is defined by length scales (Wood et al.,1993). 21 For a momentum dominated discharge, the two key parameters influencing the flow are the jet momentum flux M and the cross flow velocity Ua. Combining these results in a characteristic length scale zm where: 2.12 The length scale zm provides an indication of when a jet would bend due the ambient velocity. For z«Zm t n e effect of the cross flow is negligible and for z » z m , the ambient current dominates the flow. Similarly, for a buoyancy dominated discharge into a cross flowing ambient, the two key factors are the initial buoyancy flux (B) and the ambient velocity (Ua). Combining these two variables forms the following characteristic length scale, zb, which provides an approximation of when a plume would bend under the influence of the free stream velocity. B Z b = ~ n J . 2 - 1 3 a When z«zb> the effect of the ambient flow is negligible and when z » z & , the cross flow dominates the behaviour of the buoyant jet. The use of the length scales to describe the mechanisms controlling the behaviour of the flow is summarized in the following diagram. 22 effects of jet exit are negligible z « IM momentum dominated z » IM buoyancy dominated Z « Z m Z » Z m Z « Zb Z » Zb jet momentum ambient current jet buoyancy ambient current dominates dominates dominates dominates Hodgson (1991) provides an extensive review of work pertinent to jets in crossflows. The following is a summary of key experimental investigations into the trajectory and dilution of vertical buoyant jets issuing into a cross flow that builds a basis for the work carried out in the present study. 2.1.2.1 Chu and Goldberg (1974) Chu and Goldberg (1974) investigated buoyant forced plumes in a uniform cross flow using a negatively buoyant jet injected vertically downward. The work was carried out in a 0.30 m x 0.45 m x 9.00 m flume over a range of velocity ratios between 4.0 and 19.4. The jet trajectory was determined using photographic techniques. For non-buoyant discharges the jet trajectory followed a one-third power law given by: ( V 3 — =1.44 — 2.14 For buoyant discharges, the trajectory was given by: 23 = 1.441- x + X 2.15 The downstream transition point (xT) between the momentum dominated flow and the buoyancy dominated flow was found to be given by: 2U U j a S{Aplpa)' 2.16 These relationships were developed based on the assumption that the entrainment coefficient (cce) is 0.5. The entrainment coefficient is a constant of proportionality for the equation for the growth of a jet discharge. dr Uj — = c r — dx eU/ 2.1.2.2 Wright (1977) Wright (1977) carried out a comprehensive investigation of buoyant jets in a cross flow. He obtained solutions for the mean jet variables based on dimensional analysis combined with simple conservation of momentum flux arguments and assuming self-similarity. In order to do this, the assumption was made that within a specific region, the flow behaviour is controlled by only one jet parameter and one ambient parameter. The following is an overview of the analysis Wright carried out to develop approximate solutions for the mean jet trajectory and dilution. The development of these asymptotic solutions is covered in detail as they provide the basis for the analysis of the physical mixing experiments carried out in the present study. 24 Momentum Dominated Near-Field For a pure jet in a cross flow, zm is the only characteristic length scale which describes the flow behavior. Close to the jet exit, when the effects of the cross flow are negligible, the flow can be described by the velocity relationship for a simple jet given in Table 2.1. Substituting for lQ, this becomes: Z where A ; is a constant, Um the mean centreline velocity and z is the height above the jet exit. Due to the presence of the cross flow, the simple jet is advected downstream, not bent, by the cross flow. The slope of the jet trajectory is given by: 2.18 dz=UJ, dx U„ where Ua is the ambient velocity. Combining equations 2.17 and 2.18: &-EJ!L = A M ^ 2.19 dx ' Un 1 U„z • Using z„, = ——, equation 2.14 becomes — =A— 2.20 dx 1 z 25 By integration and setting the virtual origin as x = 0 and z = 0, this becomes equation 2.21 for z«zm X ( - V l 2.21 v Z"< J This region is known as the momentum dominated near field (MDNF). Wright (1977) found that Ci was a weak function of ZJIQ. Momentum Dominated Far Field When z»Zm the cross flow begins to dominate the flow behaviour, bending the jet in a direction parallel to the cross flow. In this region, the flow is similar to a cylindrical momentum puff from a line source (Scorer, 1978). A momentum puff is a non-buoyant element with significant momentum (Lee and Rodi, 1993). The flow is a function of the momentum pulse m=M/Ua (Dugan, 1994) and the vertical distance above the jet exit. By dimensional analysis; U, f~i = constant = A2 or 2.22 ml z 2.23 Combining the above with the slope for the jet trajectory (equation 2.18) and integrating results in the following equation which describes the flow in the momentum dominated far field region (MDFF). 26 2.24 Wright found that C2 was a function of ZJIQ. Buoyancy Dominated Near Field Similarly, the same analysis can be applied to the two extreme cases for a pure plume in a cross flow. The first being the case where z<zb and the plume buoyancy dominates the flow and the plume behaves like a pure plume simply advected downstream by the cross flow. Again, the slope of the trajectory is given by equation 2.18 and the flow can be described by the equation for a pure plume discharging into a quiescent ambient given in Table 2.2. which describes the behaviour of a plume in a cross flow when z«zb. This is also known as the buoyancy dominated near field (BDNF). 2.25 Combining 2.18 and 2.25 and integrating results in 2.26 27 Buoyancy Dominated Far Field In the region where z»zb, the plume is bent and the flow is similar to a buoyant thermal or puff of strength b = B/Ua (Dugan, 1994). The velocity of this bent plume is a function of the cross flow velocity, b and z. By dimensional analysis: r i f 2.27 v z J Combining this with equation 2.18 for the slope of the trajectory and integrating results in -=C — ^ 4 z, •b X u , 2.28 b J which describes the behaviour of the flow in the buoyancy dominated far field (BDFF). A similar analysis by Wright (1977) develops asymptotic relationships for dilution. The minimum dilution, S, can be defined as: £ = - - _ 2.29 where CQ = initial pollutant concentration Cm = maximum concentration at any downstream distance (x). The following relationships were developed for the four asymptotic regions of a buoyant jet in a crossflow. 28 SQ Unzl = C. f z^ momentum dominated near field 2.30 SQ 2 a ~m = C< \z>" J momentum dominated far field 2.31 SQ Uaz2b = c7 buoyancy dominated near field 2.32 SQ UaZl = cs V2 buoyancy dominated far field 2.33 Based on these four asymptotic flow situations, the behaviour of a buoyant jet discharged into a flowing ambient can be determined based on the discharge parameters M, B, and Q by comparing zm and zb. If Zm«zb-> initially the flow is jet like. When z is on the order of lM, the flow switches to plume like behaviour. At z approximately equal to Zb the plume begins to bend due to the influence of the cross flow velocity. In summary, this trajectory follows a M D N F - B D N F -BDFF or jet - plume - bent plume sequence. If zm»zb, the flow begins as a jet. At approximately zm the jet bends under the influence of the ambient velocity and then ultimately behaves like a plume following a M D N F - M D F F -BDFF or jet - bent jet - bent plume sequence. In order to verify these relationships and determine the constants, Wright (1977) carried out experiments using a negatively buoyant jet discharged vertically downward and towed 29 through a stationary fluid. The flume was 0.6m x 0.6m x 9m and the velocity ratios varied from 0.8 to 116. Jet trajectories were determined using photographic techniques and the dilution was measured fluoroscien dye concentration. Table 2.3 contains a summary of the relationships and constants determined by Wright (1977). Table 2-3 Wrights Coefficients Flow Regime Photographic Data Trajectory Concentration Trajectory Dilution M D N F Cj« 1.8 Ci « 2.3 C5 « 0.35 M D F F C 2 « 1.6 C 2 « 2 . 1 « 1.2Lf? 1 / 6 a ) C6 « 0.38 b ) B D N F C i « 1.35 C j « 1 . 8 C 7 = 4.6 BDFF C4~0.Z5(zJzb)1'6 C4~lA(zn/zb)m C8 « 3.3 a) Relationship determined by Hodgson (1991) based on Wright's data b) Corrected by Wright (1978); Lee (1993) 2.1.3 Vertical Discharges in Shallow Crossflows A l l of the previous studies discussed have dealt with discharges into essentially infinite ambient flows in which the effect of the free surface did not come into play. According to MacLatchy (1993) there are four classifications of depth when dealing with jets and plumes (see Figure 2.3). These are deep water (hj/d » 20), shallow water (6 < hj/d < 20), very shallow water (2 < h/d < 6), and extremely shallow water (hi/d < 2) where hi is the depth above the jet exit and d is the jet diameter. Recently research has begun to focus on shallow water receiving environments similar to that seen in river discharges and jet turbines. The following is a review of some of the relevant research in this area, specifically looking at how the proximity of a shallow free surface can affect the trajectory and dilution of a discharge. Stoy and Ben-haim (1973) investigated the deflection and impingement of turbulent jets in confined crossflows with application to gas turbine blade cooling schemes. Their work 30 consisted of both experimental and analytical work. They developed the following relationship for the downstream point of impingement (*,-) where the jet trajectory intersected the opposing wall (h[/d = 3.05). As the velocity ratio increases, the point of impingement becomes closer to being vertically above the jet exit. Labridis (1989) found that when a vertical jet is discharged into an ambient current, the relative values of the discharge and current velocities dictate the surfacing location, plume spreading and the overall dilution. He defined three possible flow regimes that can occur (see Figure 2.4) in this type of discharge scenario. Regime 1 occurs when the ambient current dominates the discharge. This results in the jet being bent and surfacing some distance downstream of the jet. Dilution is high due to the increased length of entrainment and dilution is defined as the excess temperature at the discharge divided by the maximum excess temperature at any point. In Regime 2 flow, the jet surfaces directly above the discharge. There is an upstream extent associated with this type of flow. The dilution is lower than that achieved in Regime 1. In Regime 3 flows, there is a large surface disturbance caused by a large discharge velocity. There is extensive recirculation of the discharge fluid, resulting in relatively unstable conditions and minimal dilution. As discussed previously, for momentum dominated discharges, an important length scale is the momentum length scale, zm • If z « zm, the jet is not influenced by the free stream and essentially rises vertically into the crossflow. When z » zm , the ambient current begins to 2.34 31 dominate the flow behaviour bending the discharge in the downstream direction, resulting in an almost horizontal flow. With a discharge into a finite shallow crossflow, the depth of the ambient will also play an important role, limiting the vertical height of rise of the jet. This can be defined nondimensionally by hjd. Hodgson (1991) introduced two dimensionless relative jet strength parameters for discharges into shallow crossflows: R/^ ^ and ^ / n j^-These combine the relative strength of the discharge (velocity ratio) and the depth of flow. Hodgson (1991) carried out a photographic analysis of a circular non-buoyant jet issuing vertically into a relatively shallow flow. Experiments were done in a 0.6m x 1.22m x 17.6 m channel with hi/d ranging from 15.8 to 23.5 and velocity ratios of 2.32 to 13.05. Based on jet trajectory measurements, relationships similar to those of Wright (1977) were developed for the momentum dominated near field and far field respectively: 7 r =c, Rd 1 J L Y 2 2.35 \Rd) ' 7 f v-Rd *.Y3 2.36 The constants were found to be larger than those determined photographically by Wright (1977) and both increased with increasing R. Due to the effects of the free surface, two new regions were defined, the Surface Dominated Field (SDF) and the Terminal Level region. In the SDF, the shallow crossflow inhibited the growth of the jet resulting in trajectories lower than that given equation 2.36 for the MDFF. Within the Terminal Level Region, the jet was essentially parallel to the freestream flow and 32 remained at a stable terminal level for some distance downstream. Farther downstream, the flow was further dispersed by the ambient turbulence. In the SDF region the growth of the jet was inhibited by the proximity of the free surface. The jet trajectory was found to follow a power law relationship of the form: Rd = C SDF KRd; 2.37 The constant, CSDF, was a function of the velocity ratio and had a maximum value of 1.44 for R > 6.3. The exponent, mi, was found to be a function of the relative jet strength. Within the Rd SDF region, two distinct jet trajectories were found: transition or deep water jets ( — <0.34) hx and shallow water jets. The outer edge of the transition jets did not reach the surface and as a result they were not strongly influenced by the crossflow. The exponent for the transition zone jets is given by: mx = 0.34 (x_\.2Rd^ 2.38 As the jet strength decreased, the slope approached that of jets in unconfined cross flows, approximately 1/3. For the shallow water jets, the influence of the free surface was more prominent, resulting in a rapidly decreasing exponent given by: mx = 0.03 rRd^ 2.39 33 The momentum dominated near field did not extend farther than approximately 3 jet diameters downstream and the momentum dominated far field extended to about x -• 0.5 Rd for R > 4.5. When R < 4.5, the jet transition was from momentum dominated near field to surface dominated flow immediately. Hodgson (1991) also carried out a separate laboratory investigation on the dilution of non-buoyant vertical jets in cross flows. The same channel as described before was used with flow depths (hj/d) ranging from 15.7 to 41.3 and velocity ratios ranging from 1.46 to 10.55. Rhodamine WT dye and a fluorometer was used to measure the concentration. From these measurements, Hodgson developed the following relationship for jet centreline dilution: This equation was valid for — values between 1.1 and 990. Bimodal concentration profiles d Rd were seen for discharge scenarios with — >0.36. Although these discharges displayed \ bifurcation with peak nodal concentrations 1.3 to 1.5 greater than centreline concentrations, their centreline dilution was represented by equation 2.40. An equation for jet centreline trajectory was also developed based on maximum centreline concentration. 2.40 Rx 0.26 2.41 34 This equation could only be applied in the region of the flow that was not affected by the free f surface x ^ — < 10 ]. Although the effect of the shallow cross flow was noted with respect to the jet trajectory, Hodgson made no attempt to correlate the surfacing of the jet trajectory and impacts on subsequent dilution. Effects of Shallow Crossflows on other Discharge Configurations Johnston et al. (1993) (Johnston et al., 1994) investigated a horizontal buoyant jet discharging into shallow moving water. They found that the major factors influencing this type of discharge are the proximity to the bed and free surface and the velocity ratio. Small velocity ratios (0.10 - 0.20) resulted in significant increases in the jet dilution and width when compared to that of discharges into quiescent ambient. For larger velocity ratios, jet dilution and width was increased close to the discharge but were inhibited farther downstream. As the jet reaches the free surface it is essentially a buoyant surface jet with a reduced level of entrainment. The transverse width increases while the vertical width of the jet decreases. The shallow flows were found to restrict the penetration of the discharge into the receiving water and resulted in reduced dilution rates. They found that the mixing process in a confined flow was much more complex due to instabilities that form near the free surface, influences of the Coanda effect and changes in the jet cross-sectional area as it approaches the boundary. The Coanda effect is the interaction of the plume and the bottom or free surface due to low pressure effects (Jirka et al., 1996) Rajaratnam and Langat (1995) investigated the mixing of a circular turbulent wall jet at right angles to a crossflow at velocity ratios between 2 and 12 and cross flow depths (h]/d) of 20, 10 and 5. This type of discharge is similar to that used by Caribou Pulp and Paper in the 35 Fraser River at Quesnel. They developed an empirical relationship to predict mean dilution Dilutions of 50 to 100 were obtained within the mixing region which included the momentum dominated near field, momentum dominated far field and the ambient region. They found that for the shallower flows (h/d=5 and 10) the effects of the free surface could be seen and the dilution began to deviate from equation 2.42 at approximately — =100 and d for h/d = 5, the dilution actually leveled off at this point. Further investigation of their results indicates that for low R, the exponent is greater than 0.63 at farther distances downstream while for larger velocity ratios, the exponent is less. 2.1.3.1 Bifurcation The power law relationships discussed previously describe the mean behaviour (trajectory and dilution) of a non-buoyant or buoyant jet discharging into a crossflow. They do not deal with the physical nature of the flow such as the formation of vortices (Figure 2.5). These are formed as a result of the pressure drag on the jet flow which distorts the jet and induces vortices in the flow (List, 1982). In certain situations, the vorticity generated is strong enough that it splits the plume into two concentration nodes or maxima. This phenomena is called bifurcation and is further enhanced by buoyancy (Abdelwahed and Chu 1978). As well, as a vortex pair approaches a free surface, bifurcation is enhanced (List, 1982). From in terms of a transformed downstream distance 2.42 36 the side, this bifurcation may not be evident, but it can be seen clearly from underneath a plume (Scorer, 1978). This splitting is associated with line vortices which are the trailing arm of the horseshoe vortex generated by the strong shearing between the crossflow, the jet flow and the initially existing pipe flow vorticity. Bifurcation is due to internal circulation induced by buoyancy (Scorer, 1978). Turner (1960) investigated two dimensional vortex pairs in bent-over plumes. The buoyancy acted to increase the momentum of the pair. Koh (1973) noticed this bifurcation or thermal wake when the ambient current is relatively strong compared to the jet discharge. Abdelwahed and Chu (1978) investigated bifurcation and found that without buoyancy, the character of each of the bifurcated nodes is identical to a non-bifurcated jet. The cross-section and trajectory are essentially the same as in a non-bifurcating jet that is undeflected by the free surface, following the 1/3 power law. Buoyant situations were found to be much more complex. Essentially a bifucating jet was found to consist of 3 regions: non-bifurcating region, impingment region around bifurcation point which occurred when the top edge touched the free surface, and downstream region where the jet splits into two separate elements. The second region is very complicated with a radially spreading layer forming around the bifurcation point. This layer would eventually bifurcate at some distance downstream. Under buoyant conditions, the process is much more complicated and the experiments were insufficient to describe the phenomena under these conditions. 37 The environmental implications of this bifurcation is that the resulting plume is much wider and the measured center line concentration may not be the maximum at that location. There is also a potential that one of the arms may drift toward the side of the channel, causing high concentrations in potentially sensitive areas. 2.1.4 C O R M I X Mixing Model With increasing pressure from regulatory agencies for the provision of plume delineation, there is an increasing use of discharge mixing models such as CORMLX. Under contract to the US EPA the Cornell Mixing Zone Expert System (CORMLX) was developed by Cornell University for the analysis and design of aqueous discharges into aquatic environments (Jirka et al., 1996). CORMLX1 is used for positively and negatively buoyant submerged single port discharges-. CORMLX2 is used for submerged multiport (>3) discharges and CORMIX3 for surface buoyant discharges. In CORMLX 1 and CORMLX2 the jet must be highly submerged with a small effective diameter compared to the depth of the river. A l l models can deal with stagnant or flowing ambients with or without density stratifications. There is a wide range of models available for predicting mixing zones, ranging from simple analytical formulae to complex numerical solutions of differential. CORMIX is based on an expert system which attempts to mimic the way an experienced person would solve a specific mixing problem. The core of the model is a flow classification system (Doneker and Jirka, 1991). Based on the dimensionless length scales discussed in Section 2.1, the model classifies the discharge configuration into generic flow classifications. A flow classification chart is shown in Figure 2.6 for vertical and horizontal flow classes. 38 Once the flow has been classified, the model assembles and executes a sequence of submodels to simulate the hydrodynamic behaviour of the discharge and calculates the plume trajectory, dilution, width and position using a control volume concept. The version 3.1 of CORMLX was used in the present study. An extensive study looking at the suitability of a variety of models was carried out by McCorquodale et al., 1992. Of the models evaluated, CORMIX was the only one that has been developed to deal with positively and negatively buoyant discharges in flowing ambient currents. They investigated the specific use of CORMIX 1 and CORMIX2 in connecting channels of the Great Lakes. This study found that the use of C O R M I X 1 and CORMIX2 was restricted to discharges that have a ratio of depth of flow to pipe diameter (hi/d) greater than 3 and the vertical distance of the jet exit above the river bed less than 1/3 of the flow depth. Essentially, the discharge must have a large submergence. When compared to actual field data from a mixing study carried out in the Lesser Slave River in Alberta (McCorquodale et al., 1992), CORMIX1 developed the same dilution trend but underestimated the measured dilution resulting in the overestimation of the contaminant concentration. The same study also compared CORMIX 1 output to field data from the Ingleside STP in the St. Lawrence River for three measured effluent cases. In this case it was found to give reasonable estimates of dilution. Marks (1996) also compared C O R M I X 1 output to laboratory experiments using a buoyant jet issuing horizontally at an angle of 45° to a shallow ambient current. She found that in most cases, C O R M I X 1 underestimated the dilution. The difference between predicted and observed dilution reduced as the velocity ratio increased. When the velocity ratio is less 39 than one, C O R M I X 1 overestimated the initial dilution due to the assumption of bottom or wake attachment. She also found that CORMIX 1 did not accurately predict the trajectory and width of the discharge in the near field region, although in general the predicted trends were consistent with laboratory observations. 2.1.5 Summary The majority of the previous research has focused on unconfined receiving environments. Those that have dealt with shallower ambient flows, neglected to investigate thoroughly the effect of the proximity of the free surface on the dilution and behaviour of the discharge. This cannot be neglected when dealing with discharges into the Fraser River (hjd < 15), especially when looking at the critical low flow case where the available dilution is dramatically reduced. To improve the available knowledge in this area, the mixing experiments performed in this study investigated the effects of the proximity of the free surface on the behaviour of a discharge in a shallow crossflow, specifically on dilution and subsequent downstream concentrations. The results were compared to the commonly used empirical relationships of Wright (1977) as well as Hodgson (1991). The results of these experiments were then compared to that of the CORMLX. This provided insight into the suitability of this program for the prediction of dilution of shallow water discharges similar to those in the Fraser River and to provide additional data that can be used for further refinement of the model for shallow crossflow conditions. 40 2.2 Association of Contaminants with River Sediment Once in the aquatic environment, there are several complex physical, chemical and biological processes occurring simultaneously with mixing, that determine the fate of pollutants. This includes degradation (chemical, biological or photolytic), volatilization, partitioning to sediment, flocculation, and bioaccumulation. The predominant pathways are determined by environmental conditions, such as pH and temperature, chemical characteristics, such as solubility and ionization, and whether the contaminants are associated with biological solids or remain dissolved in solution. Each of these processes dictate the concentration and distribution of a pollutant in the water, air and sediment compartments and subsequent availability for biological uptake. For pulp mill discharges In the Fraser River, the key processes affecting the distribution of contaminants in the receiving water are the potential sorption of contaminants to the ambient sediment and the potential interaction between the biological solids and natural river sediment. Both of these processes result in the association and subsequent transport of pulp mill contaminants with the sediment. The following sections will review the theory behind these processes and their applicability to discharges in the Fraser River, specifically the near field region. 2.2.1 Sorption to River Sediments When investigating the fate of pollutants in the aquatic environment, sorption from solution onto sediments is of significant importance. Sorption is the overall process by which compounds (sorbate) in solution become associated with solid material (sorbent) such as 41 sediment or biota. It consists of adsorption and absorption. Adsorption is the taking up of molecules on the internal and external surfaces of solids. Absorption is the taking up of molecules into the bulk solid phase. Adsorption occurs on surfaces due to attractive forces of the atoms and molecules making up the surfaces. The reactions occurring at this solid-liquid interface determine the rate and extent of adsorption. There are three steps that take place during adsorption: • film diffusion to the solid surface, • pore diffusion to an adsorption site, • adsorption to the surface. There are two types of adsorption, physical and chemical, depending on how the compound becomes attached to the particle. Physical adsorption is when attachment is by physical forces and chemical is by the formation of chemical bonds. Based on compound concentrations in the aqueous (Cw) and solid (Cs) compartments at constant temperatures, there are several relationships that can be applied called adsorption isotherms. Two most common isotherms are the Langmuir and the Freundlich isotherms (Benefield et al., 1982). The Langmuir isotherm assumes that the adsorbed layer is one molecule thick (physical) while the Freundlich isotherm assumes multiple layers (chemical). Langmuir Isotherm C 2.43 ' 1 + C... 42 Cs = concentration in solid phase C = concentration of material in solution after adsorption a ,b = constants determined empirically Freundlich Isotherm Cs=K{Cw)X>f 2.44 where K and rif are constants for each solute and temperature. The shape of the isotherm depends on the mechanism dominating the adsorption (Schwarzenbach et al., 1993). Examples of Freudlich isotherms are shown in Figure 2.7 (a) through (c) and (d) is an example of a Langmuir Isotherm. In Figure 2.7 (a), a linear isotherm results in a constant 7\" value indicating that the adsorption is non-site specific (Podoll et al., 1984). This type of isotherm is typical of low molecular weight organic solutes with low aqueous solubilities at low concentrations similar to those found in the receiving environment. Isotherms with decreasing slope (l/tif< 1) and increasing slope (lMf> 1) are shown in Figure 2.7 (b) and (c) respectively. The decreasing slope is known as unfavourable adsorption, while an increasing slope is called favourable adsorption. These nonlinear isotherms are indicative of site specific adsorption which is common for polar and ionizable solutes. A Langmuir Isotherm is shown in Figure 2.7(d) where Cs is the amount sorbed in one layer. The adsorption of organochlorines and other organic contaminants is primarily by chemical processes and therefore is best represented by the Freundlich isotherm (Anderson et al., 43 1985). Generally, for low concentrations similar to those seen in the environment, IMf is unity and the previous expression simplifies to: C, = KCW 2 .45 Error is introduced as IMf diverges from unity. As discussed previously, partitioning of contaminants to solids consists of both adsorption and absorption. Factors affecting the overall sorption are chemical composition, organic content of the solid, available surface area, pH, temperature, clay content and ion exchange capacity. Sorption is also affected by water solubility, increasing as solubility decreases. This is of particular importance in the case of pulp mill effluents. Many of the toxic constituents are generally hydrophobic, chlorinated and have low water solubility. This solubility decreases as chlorine substitution increases. Studies have indicated that the fraction of organic carbon (f ) in the solid is a controlling factor in the adsorption of neutral organic contaminants (Carey, 1985). The Partition Coefficient Kp, is given by the ratio of a equilibrium concentrations in the sorbed phase and in the solution (Schwarzenbach et al., 1993): K = C/C 2 . 4 6 where Cs is the concentration of pollutant associated with a weight of solid and C is the concentration of pollutant associated with the same weight of water. The partition coefficient varies with the type of material and varies from system to system. Due to the high degree of variability in sediment and suspended solid composition in natural 44 systems, it was found necessary to develop a simple procedure for predicting adsorption parameters. It has been seen that it is possible to normalize partition coefficients to organic content (Karickhoff, 1981): K =K xf 2.47 p oc J oc where Kgc is the organic carbon normalized distribution coefficient for a pollutant and foc is the fraction of organic carbon associated with the sediment. KQC can be estimated from water solubility and octanol/water partition coefficients ( K o W ) using the following relationship (Schwarzenbach et al., 1985). \ogKoc = a\ogKow+b 2.48 where a and b are constants. Combining equations 2.47 and 2.48, log Kp = l og / o c + log Koc = log foc + a log Kow +b 2.49 It should be noted that this method doesn't apply to ionic compounds and poorer correlations were found for more polar compounds. As well it is only applicable for cases where foc > 0.001 and the organic content dominates sorption. In summary K can be determined experimentally from sorption isotherms or calculated based on Kow and/ o c . 2.2.1.1 Octanol/Water Partition Coefficient (Kow) The octanol/water partition coefficient is a distribution coefficient of a compound between an aqueous phase and a hydrophobic organic phase, octanol. That is the equilibrium ratio of the amount of a compound in the octanol phase to the amount of a compound in the aqueous 45 phase. Octanol has become a commonly used reference organic phase for characterizing the partitioning of organic compounds due to its suitability as a surrogate for various environmental and physiological organic matter (Schwarzenbach et al., 1993). The larger the value of Kow, the more hydrophobic the compound and therefore the higher the tendency for that compound to partition out of the aqueous phase into sediment or biota organic phase. The octanol/water partition coefficient can be determined directly using the "shake flask" method and indirectly using High Performance Liquid Chromatography (HPLC) (discussed in Appendix A) or calculated from aqueous solubilities. Water solubility is a key determinant of Kow (Chiou et al., 1985). Many empirical relationships have been developed relating these two properties and can be used if the relationship is derived from similar kinds of compounds with hydrophobicity in the same range. Chiou et al., 1977 determined the following correlation based on a variety of compounds including aliphatic and aromatic hydrocarbons, aromatic acids, organochlorines and polychlorinated byphenyls. log Kow= 5.00 -0.670 log Sail 2.44 The aqueous solubility, Saq, is in umole/1. This equation estimated K 0 w within one order of magnitude. Factors affecting this are temperature, analytical methods used for K 0 w determination as well as variations is solubility values for the individual compounds. Caution should be taken when dealing with polar compounds or compounds with polar functional groups that can undergo hydrogen bonding such as -OH" and - N H 2 46 (Schwaxzenbach et al., 1993) and these simple correlations may not be suitable for organic acids. Other similar correlations have been developed relating K o w to solubility, S a q , in moles/1. Chiou et al., 1985 log Kow = -0.960log Saq + 0.432 Andren et al., 1987 log Kow = 0.960 - 0.806 log Saq Schwarzenbach et al., 1993 log Kow - -a log Saq + b 2.51 2.52 2.53 In equation 2.53, a and b are constants that depend on the type of compound and are given in Table 2.4. Table 2-4 Constants a and b for Equation 2.53 (Schwarzenbach et al., 1993) Set of Compounds a b Substituted benzenes only nonpolar sub. including polar sub. 0.86 (± 0.03) 0.72 (+0.05) 0.75 (±0.09) 1.18 (±0.16) Miscellaneous pesticides 0.84 (±0.12) 0.12 (±0.49) 2.2.1.2 Effect of Ionization on Sorption When looking at the adsorption of weak organic acids such as guaiacols, other mechanisms come into play. An important parameter in determining the fate of weak organic acids, such as chlorophenols, chloroguaiacols and resin acids, in the aquatic environment is its ionization (or dissociation) constant, Ka or pKa=-log(Ka). In the aquatic environment, these acids are present in two forms; anionic and neutral species, both exhibiting different chemical properties. The anionic species tends to be more water soluble. The relative amount of each species is determined by the following equilibrium relationship: 47 HA^H+ + A-_ k k _ ] 2.54 [HA] This equation can also be written as: W pH-pKa =log [HA] 2.55 From these equations, it is apparent that the pH of the receiving water determines the extent of dissociation of the organic acid. As chlorine substitution increases, Ka increases. For 3,4, 5-trichloroguaiacol and tetrachloroguaiacol, pKa values are 7.56 and 6.26 respectively (Xie et al., 1986). Therefore at a pH of 7, the higher chlorinated guaiacols will be primarily in the ionic form while the lower compounds are present in both forms. For example at a pH of 7, 3,4,5-trichloroguaiacol is 21% dissociated while tetrachloroguaiacol is 85% dissociated. The degree of dissociation also affects the extent to which individual processes act on the compound. For example, for pentachlorophenols, with increased dissociation, photolysis increases while adsorption and volatilization tend to decrease. At pfTs normally seen in the Fraser River (7.9) (Shaw et al., 1995), guaiacols are present in both neutral and ionic forms, both exhibiting different sorption characteristics. Partitioning of the neutral species is well correlated to the organic content of the sediment. The anionic species may adsorb to the sediment phase by different mechanisms, depending on the receiving water characteristics. This preferential partitioning of the undissociated portion was observed in the Fraser River by Prahacs (1994) with an increased rate of sediment partitioning as pKa increased. 48 Extensive work has been done on the mechanisms of adsorption of both neutral and anionic forms of chlorophenols. Schellenberg et al. (1984) found that for sorbents with foc > 0.001, the sorption of both neutral (HA) and anionic (A") species can be described using a linear isotherm: [HA]+ A~ •1 \[HA]+ A" 1 where Ds is a distribution ratio and is pH dependent. For chlorinated phenols with 3 or less chlorine substitutions the phenolate sorption can be neglected and the distribution ratio can be calculated using: The degree of sorption of the ionic species was found to be dependent on the ionic strength the undissociated form and the contribution of the phenolate ion can be neglected when pH - pKa < 1. For higher chlorinated phenols, sorption of the phenate species must be considered since at ambient pH values they are predominantly in the anionic form. Schellenberg et al., 1984 also derived the following relationship for estimating Kp values for chlorinated phenols: 1 2.57 ofthe aqueous phase. In waters of low ionic strength, I, (I < 10" 3M) the sorption is primarily Kp=foc.l.05(Kj ,0.82 2.44 49 Schwarzenbach (1985) found that chlorinated phenate ions are sorbed significantly in the presence of calcium and magnesium ions. In this case, processes such as ion-pair formation and subsequent adsorption to the organic fraction may occur. A study by Westall (1985) showed that the distribution of tetra- and pentachlorophenol in water and octanol depended on both pH and ionic strength. At high pH and ionic strength, the phenate ions in association with potassium ions were the primary species in the non-aqueous phase. This was also found in experiments carried out by Jafvert et al. (1990). Other mechanisms such as surface complexation between ionic species and sediment surfaces are being suggested but they are not well understood. 2.2.1.3 Other Factors Affecting Sorption Organic Macro-molecular Material The presence of humic/fulvic acid has also been found to affect the adsorption of hydrophobic organic compounds. It is thought that the hydrophobic compound partitions out of solution to bind with the macro-molecule at "hydrophobic centers" (Yu-Ping et al., 1989). Murphy et al. (1990) found that increasing the amount of humic substances increased the adsorption of hydrophobic organic compounds out of solution. This reduces the pollutants availability for biological uptake or further modification through chemical reactions. These humic substances cause colour in natural waters and it may be possible that the same effects may be seen with the colour causing agents in bleached pulp mill effluent; lignosulphonates. Voice et al. (1983) found that the perceived sorption increased as the total concentration of solids decreased. It was thought that this may be due to an increased amount of dissolved organic macro-molecular or micro-particulate material sorbing some of the compounds in 5 0 question, the amount of which increased as the concentration of solids increases. During the phase separation, this smaller material was not removed from the liquid phase and therefore was subsequently included in the bulk liquid analysis. This resulted in an apparent reduced sorption. Even after high-speed centrifugation, some of the particles remained in solution. This has important implications for the sorption of pulpmill contaminants due to the presence of a large amount of biological solids in the effluent, especially that from an Aerobic Stabilization Basin. Temperature Similar to most chemical and biological processes, the adsorption of ionic and polar substances (weak organic acids) is temperature dependent. The extent of this dependency depends on the specific sediment/solution system. Many studies have shown that a decrease in temperature results in decreased sediment adsorption of these compounds (Roy et al., 1991). This dependency is most likely due to the actual thermodynamics of the adsorption process; changes in enthalpy and entropy. Time Adsorption is a time dependent process and equilibrium between the sediment and solution concentrations of a compound does not happen instantaneously. Depending on the compound and sediment characteristics, equilibrium time can vary from 30 minutes to 2 weeks (Roy et al., 1991). Typically, equilibrium time is assumed to be 24 hours. The time for a system to reach equilibrium is very important when dealing with a dynamic system such as the near field mixing zone. 51 2.2.1.4 Summary The extent of adsorption of pulpmill contaminants to the river suspended sediment in the near field region can play an important role in determining the fate and effects of the contaminants in the receiving environment. The relative importance of this mechanism and the factors affecting it was investigated and is discussed in Section 3.2 and 4.2. 52 2.2.2 Biosolids and Their Association with River Sediment As discussed previously, contaminants in the effluent are also associated with biological solids or "biosolids" that are discharged with the effluent. Once in the receiving environment, these solids may then interact with the ambient sediment. This is an important factor affecting the distribution of pulp mill contaminants in the receiving environment. The following sections will introduce the characteristics of biosolids, their potential as a source of contamination and their fate in the receiving environment. 2.2.2.1 Characteristics of Biosolids Aerobic Stabilization Basins (ASBs) can remove significant amounts of chlorinated and unchlorinated organic compounds from bleached kraft mill effluent (Tomar et al., 1991). In many cases, these basins operate facultatively, with an anaerobic region where biological solids accumulate (Hall, 1993). It is in this anaerobic benthal layer that much of the removal of these organic compounds takes place by anaerobic degradation and dehalogenation. Adsorption from the water column on to the biomass provides the transport mechanism for this degradation (Amy et al., 1988). The biomass is capable of adsorbing both high and low molecular weight compounds. The concentration of these biosolids increases across the lagoon. The amount of low molecular weight compounds also increase across the lagoon while high molecular weight compounds decrease. Typically, there is no secondary clarification and the ASB effluents have higher levels of suspended solids than effluents treated using activated sludge. The suspended solids not removed in treatment are discharged to the environment with the effluent. These 53 contaminated biosolids are subsequently incorporated into the food chain by grazing and filter feeding invertebrates (Costa et al., 1979). Typical suspended solid concentrations released into the aquatic environment range from 30 to 200 mg/1 (NCASI, 1978). Once discharged, 95% of these biological solids remained in suspension in the receiving environment, even when the stream velocities were conducive to settling (Costa et al., 1980). Amy et al. (1988) carried out an extensive study on the biosorption of organochlorine compounds (TOX) in a bleached Kraft mill lagoon in order to determine the importance of sorption versus degradation for the removal of these compounds from the water column. Experiments were carried out at various pH levels and temperatures using both live and dead biosolids. Significant sorption (average 7^=459 cm3/g) was seen with both the live and dead biosolids, reaching equilibrium within 4 hours. The sorption kinetics was characterized by an unfavourable isotherm. The most effective sorption occurred under low pH conditions and temperature was found to have minimal effect on adsorption. The pH of the system affected both the characteristics of the biomass and the compounds. At a pH of 7, bacteria were found to have a negative charge density, which decreased with pH. The same is true for the compounds investigated. Lower pH values resulted in decreased charge repulsion between the sorbent and the sorbate. In addition, there are more non-ionized compounds at lower pHs which sorb better. As well, the unionized portion of the compound partitioned to the cell lipids rather than adsorbed to the surface. Overall, temperature had little effect, with sorption of the >1000 weight fraction decreasing and the <1000 fraction increasing as temperature increased. 54 Increasing the mean cell residence time increased biomass contaminant loading of TOX. Cell viability had little effect on the removal of the compounds from the liquid phase. This suggests the predominant mechanism is sorption to the biomass rather than metabolic uptake and biodegradation. Liver (1990) carried out batch adsorption studies of Resin Acids using both anaerobic and aerobic biomass and found that the partitioning rates were significantly higher than the rate of removal by degradation. The percent of D H A in the aerobic biomass was 62.1 ± 20% of the original loading (pH 7.5 - 8.1). Freudlich isotherms, which describe surface adsorption, were fitted to this data and resulted in the following relationship: Cs=Kc/»f 2.59 where is K = 0.07 and l/nf = 1.41. Liver (1990) found the solid phase loading to be linearly related to the concentration of resin acids in the supernatant. This linear isotherm may be indicative of actual partitioning of the compound to the organic phase in the bacteria rather than adsorption to the surface. From this data a partition coefficient, K , of 310 cm3/g was calculated (Hall and Liver, 1996). Organic Content Many studies have been carried out on the characteristics of both wastewater and pulp mill biosolids. The organic content of wastewater biosolids ranges from 65% to 85% (Dobbs et al., 1989). The suspended solids in the pulp mill effluent are primarily bacterial cells and cell fragments with a specific gravity of approximately 1 (Costa et a l , 1980). Zanella et al. (1978) found the organic content of biosolids generated in a bleach Kraft mill ASB range 55 from 20% to 40%. The organic content was found to be higher in the winter. Biological material made up 45 to 64% of the total suspended solids, again with higher levels present in the winter. Of this, 3 to 45% were viable (3 to 11% in the winter). The non-biological material consist of low levels of ash and inorganic elements. Tomar et al. (1991) found the organic content of pulp mill biosolids to range from 18 to 58% while a study by the Philadelphia Academy of Natural Sciences determined the volatile fraction (organic content) of biosolids to range from 64 to 90%, with 4 to 20% viable. Particle Size Particle size of the biosolids range from 1 to 16 microns with very few at the large end (Costa et al., 1980). They found that 62 to 78% were 1.1 to 1.6 microns and 80 to 91% were 1.1 to 2.0 micron. The latter range includes bacterial cells, cell debris, inorganic coating materials and grit. The large material consisted of bacterial floes, fibers and protozoans. Similar size fractionations were seen by Zanella et al. (1978). Scanning Electron Microscopic analysis of pulp mill ASB biosolids indicated very few wood fibers and flocculant solids (NCASI 303). Their study also included size gradation and settling studies of ASB biosolids, which is summarized in Table 2.5. Sedimentation had little effect on the <8.0 micron size fraction, increasing it by only 5% over the 10 day period, indicating that there was minimal removal of this fraction due to settling. Table 2-5 Biosolids Size Fractionation (Zanella et al., 1978) ASB Effluent 5 Days Settling 10 Days Settling 8.0 microns 5 1 0 1.0 microns 84 82 82 0.8 microns 7 12 3 0.4 microns 4 5 15 5 6 Mechanism of Adsorption The mechanism of sorption to the biological solids includes both partitioning into the cell and surface adsorption. The high molecular weight compounds are too large to be biologically active and adsorb to the surface of the biosolids (Amy et al., 1988). The low molecular weight fraction was found to partition to the cell lipids. This partitioning increases as a compounds solubility in water decreases. The octanol-water partition coefficient can be used as an indicator for potential partitioning to cells lipids. Wang et al. (1993) investigated the mechanism of sorption of organic contaminants to waste water biological solids. This included both adsorption on the surface of the biomass and partitioning into the interior of the cell. The latter process dominates for compounds with octanol-water partition coefficients (Kow) greater than 1000. The effect of competing species was also included for sorption in multi-component systems such as a waste water treatment lagoon. There was a clear effect of the presence of competing species for compounds with Kow <500. A small effect was seen when Kow ranged from 500 to 1000 and for Kow values above 1000, competition can be ignored. This is due to the fact that when partitioning dominates, the competing species has no effect. Previous investigations by the authors on the sorption of high molecular weight organic compounds on soil and sediment show that it was also unaffected by competing compounds. When surface adsorption dominates, one adsorption site can only accommodate one adsorbate molecule and competing compounds can decrease the sorption of a given compound. For example, pentachlorophenol, with a Kow of approximately 104, is only sorbed to the biomass via partitioning and was not affected by the presence of a competing species. 57 For the compounds being used in this research project, partitioning to the cell lipids is the dominant mechanism of sorption to the biological solids in the treatment basins. The slime layer of the bacteria may also sorb some of the compounds in question from the lagoon. The slime layer is the outer most layer of bacteria and is composed of polysaccharide materials (McKinney, 1962). The formation of this layer is a normal result of the metabolism of all bacteria. The presence of large quantities of slime indicates an old system of non-motile cells. These extracellular structures are important in adsorption and aggregation of bacteria into floes as they contribute to the surface hydrophobicity of bacteria (Christensen et al., 1990). These extracellular polymers can also help bind particles together (Droppo and Ongley, 1990), alter sediment geochemistry and increase surface stability (Montaque et al., 1993). 2.2.3 Fate of Biosolids in the Receiving Environment The major mechanism influencing the fate of biological solids in the river is flocculation (Krishnappan et al., 1994a; Krishnappan et ah, 1994b; Krishnappan, 1994; Krishnappan, 1996; Evans, 1996). This could be flocculation of biosolids alone or flocculation with the ambient sediment. It is the process of flocculation that brings the individual particles together, allows them to aggregate to a size large enough to settle out to be resuspended at a later time. Flocculation and subsequent deposition of these floes may contribute to the contamination that is found associated with the sediment downstream of the pulp mills. Prior to discussing the potential for flocculation of biosolids in the Fraser River, the following sections will introduce the basic concepts of colloidal stability, destabilization and flocculation. 58 2.2.3.1 Flocculation Both the biosolids and the ambient suspended sediment are colloidal in nature. A colloid is defined as a particle that is large enough to reflect light but too small to settle in a reasonable length of time in a quiescent environment. They typically range from 1 nm to 1 pm in diameter. These particles are kept in suspension and "stable" due to repulsive forces that inhibit aggregation into larger particles with larger settling velocities. For most colloids, this repulsiveness is due to electrostatic forces (Benefield et al., 1982) resulting from equally charged particles. Colloidal solids in water can be classified as either hydrophilic or hydrophobic depending on their affinity for water (Reynolds, 1982). Organic colloids such as microbes are hydrophilic due to the presence of water soluble groups on the surface of the colloid. These include amino, carboxyl, sulfonic and hydroxyl groups. The presence of these groups results in a water layer or film that surrounds the colloid. Inorganic colloids such as clay tend to be hydrophobic. Colloidal particles have electrostatic forces that help maintain the stability of the system. The surface of the colloids tend to acquire a charge due to the ionization of surface groups, adsorption of ions from the surrounding solution, and for inorganics, due to an ionic deficit within the mineral lattice (Reynolds, 1982). Generally, most naturally occurring hydrophillic colloids have a negative charge if the pH is at or above neutral and hydrophobic colloids, such as clay, also have a negative charge. It is the repulsive forces resulting from like charges that keeps colloids in suspension or "stable". 59 Electric double layer. A negative colloidal particle with its electrostatic field is shown in Figure 2.8. Ions of the opposite charge are attracted to the surface of the colloid from the surrounding liquid. This compact layer of oppositely charge ions or counterions is termed the fixed layer (Reynolds, 1982). Adjacent to this layer is the diffused layer. These two layers represent a region surrounding the colloid where an electrostatic potential exists and is called the electric double layer. The concentration of counterions decreases from a maximum at the particle surface to that of the bulk solution at the outer edge of the double layer. This boundary is referred to at the shear plane or shear surface which encloses the water envelope that moves with the particle (Reynolds, 1982). The electrostatic potential at the shear surface is called the zeta potential which is given by: r = 4n4d< 2.60 D where £ = zeta potential q = charge per unit area dc = thickness of the layer surrounding the shear surface where the charge is effective D = dielectric constant of the liquid. The stability of a colloid system depends on the relative strength of the forces of attraction and the forces of repulsion. Forces of attraction are due to Van der Waals' forces which are 60 only effective close to the particle. The magnitude of the repulsive forces can be measured by the zeta potential. The greater the zeta potential, the greater the repulsive forces and therefore, the more stable the colloidal system. Another source of repulsion is steric hindrance due to adsorbed layers of polymers on the surface of colloidal particles (Russel et al., 1989). These macromolecules act as a barrier that prevents the particles from getting close. This method of stabilization can provide more stability than electrostatic repulsion for a longer time and at higher concentrations (Russel et al., 1989). The range or "thickness" of this barrier is determined by the effective length of the polymer tail's extending into solution (Strenge, 1993). This thickness, similar to a double layer is a function of temperature with more floe particles at lower temperatures (Russel et al., 1989). 2.2.3.2 Stability of Biosolids The system pH can significantly affect the stability of colloids. Bacteria have a Zero Point Charge ranging from 2 to 4 (Mongomery, 1985). Zero Point Charge is defined as the pH corresponding to a zero surface charge. Above ZPC, bacteria have a negative charge and below a positive charge. At both the pH of the ASB and river, bacteria and therefore biosolids will mostly have a negative surface charge. Zeta potential determinations can also be used to determine the stability of biosolids. The results of zeta potential determinations have shown a variation in the value of biosolids zeta potential. One ASB effluent with sludge recycle had a mean negative value which would indicate that the stability of the biosolids is due to mutual repulsion (NCASI, 1978). A second effluent's zeta potential was near the isoelectric point, with no trend in increased 61 solids separation. This implies that the mechanism of stability is not due to mutual repulsion but by the adsorption of hydrated colloids, steric hindrance or both. A study by C H 2 M Hil l found zeta potentials for biosolids ranging from -55 to 40 mV (Costa et al., 1979). No correlation could be found between zeta potential and again, increased solid separation was not ensured at the isoelectric point. Therefore charge repulsion may not be solely responsible for the stability of the biosolids in the ASB. 2.2.3.3 Stability of River Sediment The clay-silt fraction of the river sediment load exists in essentially a colloidal state, with a large specific surface area. Also the presence of a negative surface charge results in the attraction of both ions and water molecules (Brady, 1974). Clay colloid particles typically carry a negative charge to which positively charged ions (cations) are attracted. This results in an electric double layer. The inner layer being the negatively charged colloid and the outer layer being the surrounding cations and associated water molecules. Clays are typically completely dispersed in a water solution with each particle repelling each other. This is due to the negative charge of each particle and hydration which is enhanced by the surrounding hydrated cations. As pH increases, the negative charge increases and therefore the stability of the clay colloids. The presence of hydrated monovalent cations, such as N a + , do not reduce the colloids zeta potential, and therefore do not affect the stability of the system. Exchange with higher valency cations can reduce the zeta potential. The zeta potential of the clay particles, and therefore the stability of the colloids, can be reduced by the following: 62 - reducing the pH, - exchanging of monovalent ions by di- or trivalent ions, - adding simple salts which increase the concentration of cations around each colloid, subsequently reducing the zeta potential. 2.2.3.4 Destabilization/Aggregation of Colloids The removal of colloidal particles from water is a two step process: coagulation and flocculation (Harris, 1966). Coagulation is the destabilization of the colloidal particles and initial aggregation, typically by reducing the repulsive forces to a point where the attractive Van der Waals' forces can dominate. Flocculation is the subsequent aggregation of the destabilized particles into large floes that may then settle out due to gravity. Figure 2.9 shows interparticle forces acting on a colloid. At a given distance away from the particle, the net force acting between two particles is attractive, after this point it is repulsive. It is within this distance that particles can coalesce and form larger particles. Two processes affecting successful coagulation of particles (Pearson et al., 1984): • particle transport to bring particles close together, • interparticle forces which stabilize the particles. The probability of success is a function of the product of the mean velocity gradient, quantity of floes in the system, floe size and a rate constant (Harris et al., 1966). The rate constant is generally associated with the destabilization of the particles and is a function of physio-chemical properties. Essentially, the success of forming larger particles through flocculation 63 depends on the ability of the particles to come together and collide, the concentration of particles, and the strength of the particles repulsiveness. There are three mechanisms of particles transport: Brownian motion, shear bulk fluid motion, and differential settling. An excellent review of these mechanisms can be found in Letterman (1981) and is summarized in the following section. The rate at which two particles of different sizes collide by any mechanism is given by where Kf is the rate constant determined by transport mechanism and n(dj) and n(d2) are the number concentrations for particles 1 and 2. A) • Brownian or Perikinetic Flocculation Brownian motion is small scale random motion which is a function of thermal energy. The rate constant for perikinetic flocculation (Kpk) is given by: where k is Boltzmann's constant, T is the absolute temperature and p is the absolute viscosity. With perikinetic flocculation the rate of collision increases with temperature. The minimum collision rate occurs when the two particles are the same size (dj=d2)-B) Shear (Velocity Gradient) or Orthokinetic Flocculation N( dx ,d2) = Kf(d{,d2)- n(dx) • n(d2) 2.61 2kT {d,+d2)2 3 (i dxd2 2.62 64 In orthokinetic flocculation the relative motion of particles is caused by the bulk fluid motion. This can be laminar shear or turbulent shear. For laminar shear the collision rate depends on the velocity gradient and is given by (<j, +d2)3 du 2.63 For orthokinietic flocculation due to turbulent motion, the collision rate is a function of the kinetic energy of the small-scale turbulence motion and is given by = (d]+d?yQ 2 6 4 6.18 where G = (^yf2> £ = rate of energy dissipation and v = kinematic viscosity. G is defined as the rms velocity gradient (Cleasby, 1984) and is proportional to turbulence intensity (McConnachie, 1991). As temperature increases, the viscosity decreases and this result in an increase in G and therefore an increase in the rate constant. Assuming the initial suspension has a uniform particle size, dl=d2=d\2 and n(dx)=n{dn), the rate of contact becomes N(dn) = *dniGn1. 2.65 rnzdy3 Using <p = — - — = the fraction of total suspension volume occupied by the particles, equation 2.65 becomes 65 N(d12) = —tpGn. 2.66 K This means that the rate of orthokinetic flocculation for an initially monodispersed suspension is first order with respect to n and is proportional to (f). C Differential Settling In this case, collisions occur where a rapidly settling particle overtakes and collides with a slower settling particle. The rate constant (Ks) is given by Ks(drd2)=K8^2J\dl + d2)\dl-d2) 2.67 where s is the specific gravity of the particle. When the particles are of the same diameter (J[=<i2), the rate constant is zero. These rate constants describe the rate at which collisions occur but not the rate at which lasting or successful collisions occur. A successful collision is one that leads to aggregation or agglomeration. Hydrodynamic forces and interparticle repulsion reduce the formation of lasting contacts. The collision rate is corrected to account for this reduction using a collision efficiency factor (a) defined as: rate of formation of lasting contacts N(d^, d7) 2 68 collision rate N(dud2) The collision efficiency factor is a function the stability of the individual particles, which in most cases is due electric double layer characteristics. If the repulsion due to the double 66 layer is great, the collision efficiency factor approaches zero. When the double layer is compressed or charge neutralization occurs subsequently destabilizing the colloid, the efficiency factor approaches unity. In natural river systems, the velocity gradients and turbulence dominates and the effects of Brownian motion can be neglected. Brownian motion is the dominant process for smaller colloids with shear or differential settling becoming more important as size increases (Lick et al., 1993). The maximum size density and strength of the newly formed floes is a function of the collision mechanism (Partheniades, 1993). Floes formed by Brownian motion and differential settling tend to be large with low density and shear strength. Those formed by orthokinetic flocculation have an overall higher density and strength. A key factor affecting the overall rate and ultimate extent of flocculation and distribution of floe sizes is disaggregation. Both aggregation and disaggregation occur simultaneously (Lick et al., 1993). The break up and/or erosion of floes is a function of the intensity of the turbulence, G (Parker et al., 1972) or shear stress. As shear increases the floe diameter first increases than decreases due to disaggregation (Winterwerp, 1998). In orthokinetic flocculation, at a high G, there is a rapid distribution of primary particles and rapid formation of small floes. At lower shears, larger floes form. In natural systems, disaggregation is significant in the bottom boundary layer (Leuttich et al., 1993). Floe diameter also increases with particle concentration (Winterwerp, 1998). 2.2.3.5 Effect of Temperature The overall affect of lower temperatures on flocculation success has generally thought to be negative, primarily due to the decrease in settling velocity as temperature drops due to 67 viscous effects as defined by Stake's Law. This is generally true for quiescent environments where floe contact is due to Brownian motion or differential settling. Colour removal experiments carried out by Velz (1934) found that colour removal was significantly affected by temperature. When minimum alum dosages are used, there was a marked improvement in colour removal, speed of floe formation, and quantity and quality of floes. At temperatures near the freezing point, floe formation was almost instantaneous. This temperature dependence was thought to be due to electrochemical versus chemical processes governing the flocculation. Experiments carried out by Lau (1994) in a annular flume using kaolinite in distilled water, salt water and natural river water and sediment found that decreasing temperature resulted in a larger proportion of material initially suspended settling out. Observed settling velocities were also higher at lower temperatures. This is contrary to experiments carried out in settling tubes where settling velocities followed Stake's Law, decreasing with temperature. The main reason for this discrepancy is the presence of flow and turbulence in the annular flume. As discussed previously this turbulence plays a significant role on flocculation in natural systems for particle interaction. Another significant effect of temperature is the reduction in repulsive forces between two double layers. This force (V^), given by Equation 2.69, increases linearly with temperature. r6AnkT^ V X J 7 2 exp(-2^) 2 6 9 The attractive Van der Waal forces do not vary with temperature. As a result, at higher temperatures, the repulsive forces are greater while the attractive forces remain unchanged. 68 The floes formed at the higher temperatures are weaker and easily broken up by the turbulence resulting in an overall smaller floe size. Lau (1994) found that this decrease in floe size at higher temperatures had more impact on the settling velocity than the decrease in viscosity. Therefore in turbulent flows, due to the effects of electrostatic forces, larger more stable floes are formed at lower temperatures. 2.2.4 Flocculation of River Sediment in the Presence of Pulpmill Effluent Recent research has indicated that presence of pulp mill effluents affects the physical transport characteristics of ambient river sediment. An extensive review and analysis of this work was carried out by Evans (1996) and will be discussed in the following section. Annular flume experiments by Krishnappan et al. (1994a) consisted of two experiments with Fraser River sediments, one with the pulp mill effluent from the Northwood mill added at a 3% dilution. Both depostional and erosional tests were performed. These experiments indicated that the presence of the pulp mill effluent increased the settleability of the suspended solids in the river water. There was also some increase in the median particle diameter due to the presence of the effluent. Athabasca River Study As part of the Northern River Basins Study, a research program was carried out by Krishnappan et al. (1994b) to characterize the size distribution and transport processes of suspended particles in the Athabasca River below the Weldwood Pulp Mill outfall at Hinton, Alberta. The results of this program indicate that the presence of pulpmill effluent affected the transport processes of the ambient sediment, increasing the deposition rate of the effluent-sediment mixture. Two different surveys were done in the winter and fall of 1993. 69 Both showed that the effluent promoted flocculation of the sediment and increased deposition rates. In low flow condition, it was observed that 74% of the incoming sediment was deposited within 20 km of the outfall. There were no measurements taken in the near field, with the closest sampling location being 1 km downstream. As well, all the sampling, including background, were each carried out on a different day. Fraser River Study A similar field program was carried out by Krishnappan (1994) during October 1994 upstream and downstream of the Northwood discharge in the Fraser River as part of Fraser River Action Plan (FRAP). Measurements of river cross-section, flow velocities, in-situ aggregate size distribution and suspended sediment concentrations were made at various transect: 2 km upstream and downstream at 30, 100, 300 and 1000 m. Each transect was carried out on consecutive days, which has important ramifications on the interpretation of the data collected. The average suspended sediment concentration at each transect is given in Table 2.6 shows a dramatic drop in the suspended sediment concentration (TSS) downstream of the diffuser. Transect Location TSS (mg/1) 2000 m u/s 255.6 30md/s 129.6 100 m d/s 67.5 300 m d/s 49.1 1000 m d/s 33.1 Analysis of this data by Evans (1996) indicates that these dramatic changes in concentration are primarily due to changes in flow, shear and sediment supply rather than solely the 70 presence of pulp mill effluent. Analysis of the median diameter data by Evans (1996) indicates that there was a slight increase in the median diameter due to the presence of the effluent plume (Adso=0-5 urn). The maximum increase in median diameter occurred at the 100 m transect (Adso^l.O |im). At the 30 m and 100 m transects, there was a slight increase in the proportion of larger aggregates in the size distribution which may be due, in part, to sediment concentration changes and turbulent velocity gradients from transect to transect. Evans (1996) also carried out several experiments investigating the effect of pulp mill effluent on Fraser River suspended sediment. Quiescent settling tests using turbidity as a measurement showed that there was an instantaneous turbidity reduction due to the addition of effluent. There was no enhanced turbidity removal subsequent to this initial reduction. The maximum instantaneous reduction occurs at an effluent to river water ratio of 1:1. This reduction in turbidity was thought to be due to a shift in the aggregate size distribution of the sample. Settling tests with shear were carried out with illite and river sediment in the presence of filtered and unfiltered effluent (Evans, 1996). Filtering the effluent removed any solids associated with the effluent. Measurements of total suspended solids (TSS) were made at various times and locations over the height of the column. Air was used to maintain shear in the settling column similar to that in the river environment. The settleability of illite was enhanced by the addition of the effluent, with the most significant affect in the presence of unfiltered effluent. Therefore, the flocculation of the sediment was further enhanced by the biosolids. 71 Evans (1996) found that the settleability of the river sediment, collected July 1992, was not enhanced by the pulp mill effluent, and the settling velocity actually decreased due to the addition of the effluent. Particle size distribution analysis of this sediment using an E L Z O N E found that the river sediment collected using the trap described in Section 3.2 was significantly larger than those collected in the suspension. This was due to the configuration of the sediment trap which resulted in a bias towards larger particles with larger settling velocities. Sediment particles must be smaller than 10 pm for the flocculation of biosolids and sediment to be noticeable. Settling tests were also carried out with the river sediment at 4°C and 23°C and the median settling velocities (V 5 0) calculated from these are given in Table 2.7. Table 2-7 Median Settling Velocities from Evans (1996) V 5 0 @ 230 C V 5 0 @ 4°C Sediment + Effluent 1.22 1.24 Sediment + Distilled Water 3.55 1.80 Effluent + Distilled Water 0.057 0.049 The decrease in settling velocity of the sediment due to viscous effects is consistent with Stoke's Law: V » = ^ ( P , - P ) . 2.70 The effluent sample did not follow Stoke's Law, which was expected since biological solids cannot be considered as spherical particles. The settling velocity for the effluent alone only decreased by a factor of 1.16. Of particular interest is the fact that the settling velocity of the effluent/sediment mixture stayed roughly the same. Given that the viscous effects should 72 decrease the settling velocity somewhat, this points to the possibility that larger floes were being formed at the lower temperature, similar to results seen by Lau (1994). This may also be the reason why the settling velocity of the effluent sample only decreased slightly with temperature, as larger floes may have formed. This temperature effect is important especially when considering potential flocculation of biosolids alone in the river as the transition from the warm lagoon to the cooler river may result in increased flocculation. 2.2.5 Summary It is apparent that the presence of pulpmill effluent enhances the settleability of suspended sediment. This is especially true for unfiltered effluent containing biological solids, or biosolids. The conditions by which this occurs is not exactly known, but the mechanism causing it is thought to be increased flocculation. This results in large median diameters and therefore higher settling rates. The key factors that influence the formation of these floes within the near field mixing zone are discussed in Chapter 5. Specifically, the conditions that exist in the near field that promote successful flocculation and the subsequent association of contaminated solids with the suspended sediment. 73 Figure 2-1 Simple Jet Vortex Zone U : Jet Exit Figure 2-2 Schematic of a Vertical Jet in a Crossflow (adapted from Rajaratnam, 1976) 75 Deep h 1 > 20d Shallow 6d < h, < 20d T 1 Very Shallow 2d < hj < 6d H ± Extremely Shallow hj <2d Nozzle Flush with Bed hj =H Figure 2-3 Depth Classifications for Jets in a Crossflow (adapted from MacLatchy, 1993) 76 Surfacing position Flow Regime 1 U Flow Regime 2 Row Regime 3 Figure 2-4 Flow Regimes for Vertical Jets in Shallow Crossflow (adapted from Labridis, 1989) 77 Figure 2-5 Bifurcation of a Vertical Jet in a Crossflow (adapted from Moussa et al., 1977) 78 Figure 2-6 C0RMIX1 Flow Classifications for Vertical Buoyant Jets in Crossflows. Reproduced with permission of ASCE from Hydrodynamic Classification of Submerged Single-Port Discharges, Jirka, G.H. and Doneker, RX., ASCE Journal of Hydraulic Engineering, 117 (9), September 1991. 79 Figure 2-7 Freudlich Isotherms a) Linear b) Unfavourable c) Favourable and d) Langmuir 80 Shear Layer Bulk Solution Fixed Layer Diffused Layer Figure 2-8 Negative Colloid with Electrostatic Field 81 Figure 2-9 Interparticle Forces acting on a Colloid 3 Experimental Procedures Both physical and chemical experiments were carried out as part of this research project, each of which is described in the two following sections. 3.1 Near Field Mixing Experiments 3.1.1 Description of Experiments Experiments were carried out in a laboratory flume to observe the mixing of vertical buoyant jets in a shallow cross flow as shown in Figure 3.1. The objective of the experiments was to obtain data to develop techniques for predicting jet behaviour under various discharge conditions indicative of those seen in a river discharge situation. This includes jet trajectory, dilution and bifurcation. The experiments consisted of a continuous centerline discharge of a positively buoyant fluid into a shallow cross flow. The density difference was achieved by discharging hot fresh water into cooler fresh water. Fluoroscein dye was added to the hot water discharge to enable flow visualization. A l l flows were measured with flow meters. The distribution of each buoyant discharge was obtained by measuring centreline and lateral temperature profiles at regularly spaced intervals downstream of the jet. Acoustic Doppler Velocimetery (ADV) supplied by Sontek Inc. of California was used to characterize the ambient crossflow. 83 3.1.2 Experimental Facilities The jet mixing experiments were conducted in a 20m recirculating flume located in the U.B.C. Civi l Engineering Hydraulics Laboratory. The height of the flume is 1.0 m and the width is 0.5 m. The velocity and depth of the ambient water were controlled by an inlet valve and an adjustable gate at the end of the flume. The flume bottom is made of finished wood and the vertical sidewalls are Plexiglas. A series of grids, decreasing in size along the flow direction, were used to ensure uniform flow and turbulence and to eliminate secondary flow effects in the experimental region. The test starts 2.8 meters downstream from the screens. Figure 3.1 shows the layout of the experimental facilities. The temperature of the ambient water varied from 9 to 19 degrees Celsius. The ambient velocity was measured using an Ott Mechanical Current Meter type C2 (No. 47916) with propeller No. 3 (No. 20484). A vertical circular jet was set up in the flume bottom at the center of the flume 3.8 meters downstream from the flume inlet. The height and diameter of the jet are 12.44 and 12.65 mm respectively. The jet water came from a constant head tank located on the top of the flume. This tank was continuously fed with hot water approximately 50 degrees Celsius. The outlet flow was controlled by a valve at the bottom of the tank and the exit flow passed through a King Instrument Company hot water rotometer, model number F45500 (4400k52), for flow ranges 0.5 to 5 U.S. gpm (1.9 to 19 1pm). Fluorescein dye was added to the jet water to facilitate flow visualization. The dye was sufficiently diluted with the jet water prior to discharge in order to ensure negligible density effects. 84 3.1.3 Temperature Measurement Discharge, ambient and plume temperatures were measured using T-type thermistors set up in an adjustable array. Ambient and discharge temperatures were measured continuously during each experiment. A data acquisition system was used to collect all voltage readings from the thermistors and calculate calibrated temperatures, variances and standard deviations. The system consisted of a Metrabyte EXP-16 expansion multiplexer, a Metrabyte DAS-8 analog/digital converter and an I B M compatible computer. The temperature data was collected at a rate of 10 Hz and 1034 data points where collected from each thermistor at each location. The data acquisition programs were written in Quick Basic and based on vendor supplied programs. See Appendix B for Data Acquisition programs. The thermistors were calibrated using a constant temperature water bath and a certified mercury calibration thermometer. Linear regression analysis on the cold junction compensated voltages and actual temperature readings resulted in linear calibration equations for each thermistor. These equations were incorporated into the data acquisition program. See Appendix C for calibration data and curves. 3.1.4 Flow Visualization The vertical centreline cross section of the plume was also measured photographically using both 35 mm still photographs and a Hi8 video. A vertical cross section of the test region was illuminated by a narrow band of light located above the discharge. Prior to filming each run, a scale was placed in the same plane and filmed. This scale was then used to determine the centimeter to pixel conversion for each run. The position and focal length of the camera was not changed once the scale has been photographed and therefore no corrections for 85 perspective and diffraction were required. The video was digitized using the Image Software Program. Due to the turbulent nature of the flow, an instantaneous capture of a single video frame results in a distorted intensity profile. In order to remove these time dependent fluctuations, 30 sequential frames were captured and digitally averaged. This averaged intensity data was used to determine the jet centreline trajectory, shape, and width. To ensure that the intensity peak corresponded to the temperature peak, simultaneous video and temperature measurements were made for verification during two experimental runs. 3.1.5 Flow Characterization A SoniTek 3-D Acoustic Doppler Velocimeter (ADV) was used to measure ambient velocity profiles and turbulence levels. Both horizontal and vertical velocity profiles were taken to determine sidewall and bottom boundary layer effects. A sideward facing probe was used for the horizontal profiles while a downward facing probe was used for the vertical profiles. The sample volume of each probe is approximately 1 cm^ and is located 10 cm from the transducer for the side facing probe and 5 cm from the transducer for the downward probe. The probe operates at 25 hertz. Output from the probe was collected and converted into velocities using a vendor supplied real-time data acquisition program written in C software on a I B M compatible computer. Acoustic doppler velicometry is based on the Doppler Principle and is derived from the scattering of sound signals by small particles. A short acoustic pulse of known frequency is transmitted from the transmit transducer. The echo is then received at the three elements on the arms. The shift in frequency between the transmit pulse and the received echo at each element is proportional to velocity in that direction. For a high quality echo, there must be 86 sufficient particles in the flow to scatter the sound signal. In these experiments the ambient flow supply had particles and as a result, seeding was not required. 3.1.6 Experimental Procedure Prior to carrying out each the experiment, uniform flow was established in the flume. This was achieved by adjusting the inlet valve and the exit gate and allowing the flow to maintain a constant depth. Once this was achieved, the constant head tank was then opened to the desired discharge flow rate and allowed to reach steady state. Experiments were carried out over a range of jet/ambient velocity ratios (R) at a given hj/d ratio and ambient velocity. The velocity ratio was varied by changing the buoyant jet discharge rate. Only Regime 1 flows (Labridis, 1989) were analyzed in these experiments. That is, the ambient current dominates the discharge and the jet is bent and surfaces some distance downstream with minimal surface disturbance. For each velocity ratio, centreline and/or lateral temperature profiles were measured at regularly spaced intervals and photographs/video taken. Ambient and discharge temperatures were measured continuously throughout each experiment. The discharge temperature was varied in one set of experiments, carried out to determine the effects of buoyancy on the jet trajectory and subsequent dilution. The ambient cross flow velocity was measured before and after each experiment. The hot water discharge rate was measured continuously throughout each run. 3.1.7 Characterization of the Ambient Flow Prior to carrying out any mixing experiments it was necessary to determine both the bottom and sidewall boundary layers. This must be done to determine whether or not the discharge 87 is subject to a uniform ambient flow. The interaction of the effluent with a boundary layer can result in a very complex flow situation, drastically affecting the behaviour of the discharge. 3.1.7.1 Vertical Velocity Profiles and Boundary Layer Determination Vertical velocity profiles were taken in the experimental region using two A D V probes discussed in Section 3.1. Due to the configuration of the probes and the shallow depth of the cross flow used in the experiments, velocity measurements could not be taken between the dimensionless depths (z/H) of 0.4 to 0.8. The non-dimensional cross flow velocity profiles as a function of height above the flume bottom for three velocities and depths indicative of those used in the mixing experiments are shown in Figure 3.2. The cross flow was self-similar and the boundary layer, defined as V'a/u a = 0.99, occupied 20% of the flow depth. The mean velocity (Ua) was calculated based on the weighted average of the velocity measurements in each profile. Figure 3.3(a) and (b) are non-dimensional profiles of the longitudinal and vertical velocity fluctuations: TT and jj~ , respectively. According to Tritton (1988), large velocity a ^ a fluctuations are typically observed to have a maximum close to the wall and drop to a minimum at the edge of the boundary layer (z/S= 1). This occurs at z/H = 0.2, corresponding / .2 to a boundary layer thickness of 20%. An increase in JJ~ is apparent near the free surface. 88 This was most likely due to surface disturbances distorting the reflected sound signal and interfering with the A D V measurement. u w A plot of Reynolds Stress distribution ( T T - 2 ) across the flow depth in Figure 3.4 also a verifies the 20% boundary layer. The Reynolds Stress is defined as - u w . At the edge of the boundary layer, the value of the Reynolds stress should go to zero, as the velocity fluctuations decrease and the vertical velocity gradient levels off. Such is the case in Figure 3.4. The significance of this boundary layer and its potential impact on the downstream mixing is negligible in most of the discharge scenarios investigated. This is due to the fact that the jet exit is 1.244 cm above the flume bottom. As a result, the majority of the jet flow was well above the boundary layer in essentially uniform flow. The initial region of the jet passes through the upper part of the boundary layer, but is relatively unaffected due to the significant shear present in the jet itself compared to that in the boundary layer. 3.1.7.2 Horizontal Velocity Profile and Determination of Sidewall Effects Velocity measurements were taken across the width of the flume in the experimental region using the transverse A D V probe. Again, due to the probe's configuration and the shallow flow, velocity measurements were restricted to one flow depth (z/H=0.5). Figure 3.5 shows the non-dimensional horizontal velocity profile.. 89 The extent of the boundary layer due to the sidewalls is 8% on either side of the flume. The effect of this on the subsequent mixing experiments is negligible as the maximum width of the plume during the experiments occupied approximately 60% of the centre flume flow. 3.1.8 Mixing Experiments A total of 27 experiments were carried out of which 15 were used for centreline dilution and jet trajectory analysis. Lateral temperature profiles were taken in Experiments 16 to 19 and Experiments 20 to 22 were carried out to investigate the effects of buoyancy on the near field mixing. Experiments 23 to 27 were carried out to investigate bifurcation or "splitting" of the plume using cross-sectional video images. The range of experimental conditions (hjd and R) covered is shown in Figure 3.6 along with a comparison of the experimental conditions used in this study with those of Hodgson (1991). The following is a summary of the experimental parameters. h/d 5.42 - 10.4 R 0.41-3.83 zm 0.46 - 4.29 cm zb 1.03 X 10"2 - 6.13 X10" 2 cm Refer to Table 4.2 for more details on the individual experimental conditions. In all cases zm»zb and therefore all jets were momentum dominated jets. 90 3.1.9 Validation of video/temperature maximum For all the experiments, temperature was used as a measure of concentration and maximum centreline video image intensity was used for jet trajectory. To ensure that the locations of the centreline temperature maxima and intensity maxima corresponded, simultaneous measurements were taken. Non-dimensional intensity and temperature profiles at 5 and 27 cm downstream are shown in Figures 3.7 (a) and (b). It is evident that both the temperature and intensity maxima occur at the same location and therefore video image intensity can be accurately used for the determination of the jet centreline trajectory. 91 3.2 Experimental Determination of Adsorption for 3,4,5 TCG and DHA 3.2.1 Description of Experiments As discussed previously Kp provides information on the potential for sorption of specific contaminants onto sediment. K can either be calculated based on K and f using the p OW J OC 0 equations discussed in Section 2.1 or determined experimentally from aqueous and solid phase equilibrium concentrations in batch sorption studies. Both methods were used in an attempt to quantify the potential for sorption onto Fraser River sediment of two pulp mill constituents: 3,4,5 trichloroguaiacol (TCG) and dehydroabietic acid (DHA). These two compounds are both weak organic acids and are partially dissociated at the pH of the river. Basic chemical characteristics of each are given in Table 3.1. Table 3-1 Chemical Characteristics o l f 3,4,5 TCG and DHA Formula Molecular Weight log(Kuw) pKa Dehydroabietic Acid (DHA) C 2 0 H 2 9 O 2 301.43 unknown 7.25 (Liver, 1990) 3,4,5 Trichloroguaiacol (TCG) CyH^O^Clj 227.49 4.11 7.56 (Xie etal., 1986) A preliminary method for determining Kp experimentally was developed based on the method for Batch Adsorption Studies developed by the EPA (Roy et al., 1991). Experiments were carried out in the Environmental Engineering Laboratory at U .B.C. using Fraser River suspended sediment. The sediment was analyzed for size and organic content. In Section 4.2 the results from these Batch Adsorption tests were compared to those calculated based on the theory previously discussed in Section 2.2. As there was no documented octanol-water partition coefficient for DHA, it was necessary to determine it 92 experimentally using High Performance Liquid Chromatography (HPLC) which is explained in detail in Appendix A. 3.2.2 Characterization of Fraser River Sediment Prior to determining K it was necessary to characterize the nature of the ambient sediment. The nature of the sediment in the Fraser River plays an important role in the potential sediment contamination and subsequent transport. A size fractionation of Fraser River sediment from the Prince George region carried out by Prahacs (1994) showed a composition of 4.3 % coarse sand (>250 pm), 78.5 % fine sand (63 - 250 pm) and 17.2 % clay/silt (<63 pm). Partitioning to sediment is related to particle size, with the greatest amount of partitioning being associated with the fine size fraction. Prahacs (1994) also carried out qualitative X-ray diffraction analysis on the silt fraction. The clay minerals were primarily chlorite (60%) and illite (40%). The organic content, measured by loss on ignition, was 3.1% (July 1991) for a sample taken in spring freshette and 7.8% (February 1992) for a low flow sample. Increased sediment inputs from the river bed and banks during spring freshet account for the reduced organic fraction during this period. Suspended sediment samples were collected upstream of the pulp mill discharges at Shelley, B.C. on the Fraser River in July 1992 for use in the Batch Sorption Experiments. The sediment sampler used by Prahacs (1994) was used for this purpose. It was developed by Bi l l Duncan of the B. C. Ministry of Environment in Prince George. It consists of 4 Plexiglas tubes, 30 cm long and 4 cm wide, open at one end. These tubes are fixed in a plastic holder. The trap is placed in the river at approximately mid-depth with the tubes 93 perpendicular to the flow, open to the surface. The suspended sediment was continuously collected over a 5 day period from July 1 to 6, 1992. Sediment was wet sieved to less than 180 micron to remove any large organic debris (leaves, twigs) that might have interfered with the adsorption experiments. This resulted in a more uniform mixture of sediment. Sediment was then air dried to ensure minimal change in characteristics due to oven drying (Roy et al., 1991). Organic content of the sediment was determined based on loss during ignition at 550°C in a muffle furnace (Standard methods, 1995). 3.2.3 Determination of Kp from Kow Using Equation 2.44, K can be calculated using Kow and the organic content of Fraser River sediment (foc) and assuming that both D H A and 3,4,5 TCG behave similarly to weak organic acids such as chlorinated phenols. 3,4,5 T C G has a log(Kow) value of 4.11 (Xie et al., 1986). As there was no documented Kow value for DHA, it had to be determined experimentally using High Performance Liquid Chromatography (HPLC). This analysis is discussed in detail in the Appendix A and the results are summarized Section 4.2. 3.2.4 Determination of Kp - Batch Adsorption Studies Adsorption studies were carried out in 50 ml vials with Teflon lined screw caps. To protect against photodegradation, the vials were coated with black enamel. As 40 ml samples were used, minimal air space was remaining in the vials, inhibiting volatilization. 94 Chemical solutions were prepared using distilled water at a pH of 6 which was spiked with the specific compound to be used: 3,4,5-trichloroguaiacol or dehydroabietic acid. Standard solutions of these compounds were made up in methanol. Twenty four hour adsorption studies where carried out at a constant solution concentration of 1 ppm and varying sediment concentrations. Previously weighed sediment samples were placed in the vials and the 1 ppm solution of D H A or 3,4,5 -TCG was then added. Vials were sealed and quickly shaken by hand. The vials were then placed in a rotating apparatus and rotated at 30 rpm for 24 hours. Chemical blanks were run to check for procedural losses and sediment blanks for any desorption. After the 24 hour period, the vials were centrifuged at 2000 rpm for 20 minutes. 25 ml of the supernatant was removed for analysis. The partition coefficient (Kp) and percent adsorption was calculated based on the change in concentration in the supernatant. 3.2.5 Materials The compounds used for the adsorption tests were dehydroabietic acid (DHA) and 3,4,5 trichloroguaiacol (3,4,5 TCG). The D H A was obtained from Helix Biotech Corporation, Richmond B.C.. The guaiacols were obtained from Aldrich Chemical Company, Milwaukee, WI. The chemicals were chosen for this research due to their presence in pulp mill effluent and association with downstream sediments. DHA is present in both bleached and unbleached effluent. 3,4,5 TCG is indicative of the lower chlorinated organics being discharged under the new mill operation methods. As well, at the typical river pH (=7.9), it is the guaiacol with the highest tendency for adsorption (Xie et al., 1986). 95 3.2.6 Analytical Techniques The following is a summary of the analytical techniques used for dehydroabietic acid and 3,4,5 trichloroguaiacol. Details of each analytical procedure can be found in Appendix D. 3.2.6.1 Guaiacol Analysis - GC Method Aqueous samples were acetylated prior to extraction with hexane using 5M K 2 C O 3 and acetic anhydride according to the method of Voss et al. (1981) and 1 ml of the extract was then analyzed on a Hewlett Packard 5890 Series II Gas Chromatograph using an electron capture detector. The surrogate used was 2,6 dibromophenol and 2,4,6 tribromophenol was used as an internal standard. 3.2.6.2 Guaiacol Analysis - Autoanalyzer Method In an attempt to simplify the analysis of the supernatant, a method for analyzing chlorinated phenolics (Prokopy, 1993) on the Lachat Autoanalyzer was modified for use with chlorinated guaiacols. This method is suitable for halogen (chlorine) substituted phenols. The reagents used in the autoanalyzer are phosphoric acid, copper sulfate, sodium hydroxide, 4-aminoantipyrene colour reagent and a buffered potassium ferricyanide solution. Based on the method for phenol analysis, the compounds react with the 4-Aminoantipyrene in the presence of potassium ferricyanide to form a coloured dye (Standard Methods, 1995). The absorbance of the dye is then measured. 9 6 3.2.6.3 Resin Acid Analysis The analysis of dehydroabietic acid involved extraction with methyl-t-butyl ether according to Paprican Method 120.1 (in house). The extract is concentrated and then diluted to 1 ml using di-ethyl ether. The samples are then derivatized using diazomethane. During this derivatization, the dehydroabietic acid is converted to its methyl ester. Full derivatization is indicated by a pale yellow colour. The derivatized samples are then concentrated in a gentle stream of nitrogen and diluted to 1 ml with isooctane and analyzed by GC using a flame ionization detector. O-methylpodocarpic acid was used as a surrogate and tricosanoic and heneicosanoic acid as the internal standards. 97 a) Plan View Hot Water Overflow 1 Constant Head Tank From Thermistors to Data Acquisition System Hot Water to Discharge Port b) Profile View Hot Water from Tank ;ure 3-1 Layout of Experimental Facilities t K • 0.5 Ua/Oa 1.5 o H 1 x H 2 + H 3 Figure 3-2 Vertical Non-dimensional Ambient Velocity Profiles for HI (H=ll cm, Ua = 0.40 m/s), H2 (H=10.7 cm, TTa= 0.35 m/s) and H3 (H=ll cm, TTa= 0.30 m/s) 99 a) Vertical rms velocity profile 1.2 1 0.8 0.6 0.4 0.2 0 K to . v H X X ^ 0.01 0.02 0.03 sqrt(w,2)/Ua 0.04 o H 1 X H 2 + H 3 0.05 0.06 b) Flow direction rms velocity profile 1.2 1 0.8 0.4 0.2 0 ft**** IjX* * ++ + + *x x * + 0.05 0.1 sqrt(u,2)/Ua 0.15 o H 1 x H 2 + H 3 0.2 Figure 3-3 Non-dimensional a) Vertical and b) Longitudinal Velocity Fluctuations for H I (H=ll cm, U~= 0.40 m/s), H 2 (H=10.7 cm, TTa= 0.35 m/s) and H 3 (H=ll cm, TTa = 0.30 m/s) 100 1 X : r < X + ** -0.005 -0.0025 0 0.0025 0.005 u'wVU2 oH1 xH2 +H3 Figure 3-4 Vertical Reynolds Stress Distribution for HI (H=ll cm, Ua = 0.40 m/s), H2 (H=10.7 cm, L>7= 0.35 m/s) and H3 (H=ll cm, TTa= 0.30 m/s) 101 1.2 0.8 § 0.6 0.4 0.2 0.2 0.4 0.6 0.8 1 Non-dimensionalized Flume width (y/w) Figure 3-5 Horizontal Non-dimensional Ambient Velocity Profiles at z/H = 0.; 4.5 • • Temp./Video 0 0 Width 4 * * Buoyancy 3.5 9 9 <T7 ^0 75 ^ 74 20,21,22 T1! 9 93 72 2 4 6 8 Dimensionless Depth l^/d 10 * Present Study o Hodgson Trajectory (1991 0 Hodgson Dilution (1991) o 0 o o © o o 0 oo o 0 , -W* sly oo o 0 w * i f * * * 0 0 o 0 10 20 30 Dimensionless Depth h^d 40 Figure 3-6 a) Range of Experimental Conditions (hx/d and R) and b) Range of Experimental Conditions (hi/d and R) Compared to Hodgson (1991) a) x = 5 cm 0.5 1 Normalized Temp./lntensity • Temperature + Video 1.5 0.5 1 Normalized Temp./lntensity • Temperature + Video 1.5 Figure 3-7 Comparison of Centreline Temperature and Video Profiles a) 5cm and b) 27cm Downstream 104 4 Results and Discussion The experimental results and discussion for both the physical and chemical investigations are provided in this chapter. Section 4.1 covers the near field mixing experiments and CORMLX modeling and Section 4.2 covers the batch adsorption study. 4.1 Near Field Mixing Experiments Similar to the results of Hodgson (1991), the influence of the shallow crossflow resulted in the jet exhibiting a Surface Dominated Flow Region within the region traditionally defined as the momentum dominated far field region. Based on experimental trajectory and dilution results from this study, the SDF region was found to consist of three distinct types of flow: • Bottom Jets • Intermediate Jets • Surface Jets The colour images of each flow type shown in Figure 4.1 help illustrate these three classifications of jet behaviour. The most characteristic parameter defining the type of flow is the Jet Strength parameter or Js (——). In Figure 4.2, a Jet Classification diagram is hjd presented in terms of the nondimensional flow depth (hi/d) and the velocity ratio (R). Table 4.1 outlines the criteria for this jet classification which is then discussed in detail in the following section. 105 Table 4-1 Jet Classification Criteria Jet Classification Jet Strength Ratio Dilution Trajectory Surface 7,>0.44 Cmdff<0J5 ^ > 0.8 at - = 10 hy d Intermediate 0.30 <JS< 0.44 0.75 <Cmdff< 0.90 0.55 < ^ < 0.8 at --= 10 ht d Bottom 7, < 0.30 CW>0-90 0.55 at - = 10 hx d Bottom jets are weak discharges that were not strongly affected by the shallow ambient flow. They are defined as jets with Js approximately less than 0.30. This roughly corresponds to the deep water jets previously defined by Hodgson (1991) (Js < 0.34). Although the vertical growth of the jet was inhibited by the presence of the shallow free surface, the dilution was enhanced due to interactions of the flow with the shear flow in the turbulent boundary layer. The dilution of the Bottom Jets followed the MDFF power law relationship initially defined by Wright (1977) with coefficients greater than 0.90. That is the dilution is greater than that assuming no shallow water effects. Intermediate and Surface Jets in this project correspond to those defined as shallow water jets by Hodgson (1991). Their behaviour is influenced by the proximity of the free surface, the extent to which is determined by Js. Intermediate jets are stronger discharges whose trajectory and dilution are both inhibited by the free surface. They are defined as jets with 0.30 <JS< 0.44. The trajectory is affected by the free surface at a point downstream where it's growth begins to gradually be reduced from that predicted using the M D F F 1/3 power law due to the presence of the free surface. The dilution continues to follow the M D F F power law until the jet actually approaches the surface and dilution is reduced due to the lack of vertical entrainment at the surface. The 106 dilution of jets of this type follows the M D F F power law with coefficients ranging from 0.75 to 0.90. The proximity of some of these intermediate jets to the bottom boundary layer further complicates this flow, actually increasing entrainment and dilution if the jet edge reaches this boundary layer. Surface jets are very strong jets with Js greater than approximately 0.44. The trajectory of these jets were relatively unaffected by the shallow crossflow, and followed the M D F F 1/3 power law until they impinge the free surface. After surfacing, these jets were abruptly bent in the crossflow direction, essentially travelling parallel to the ambient current. The dilution was unaffected until surfacing, after which a reduction in the overall level of dilution could be seen. The surfacing of the jet had the most pronounced affect on the dilution of these surface jets, with M D F F dilution coefficients of less than 0.70. The stronger of these jets also exhibited bifurcation after surfacing. 4.1.1 Results, Analysis and Discussion The length scales for all of the experiments have been calculated and are given in Table 4.2. The discharges are all momentum dominated, since zm » Zb, with the exception of Experiment 5. Due to the shallowness of the flow, the discharges should not be significantly affected by the buoyancy since IM> hi. The height of all ambient flows were beyond the region of the potential core or flow establishment defined as being 1 to 4 jet diameters or 6.3 cm above the flume bottom. The majority of the data collected for both trajectory and dilution falls within the momentum dominated far-field (MDFF) defined as z > z„, and x < lM. This is the region of interest investigated in this project. For some of the weaker discharges, 107 data was collected in the downstream region where buoyancy began to influence the mixing, but due to lack of data in this region, no conclusions were drawn. Table 4-2 Summary of Experimental Conditions and Length Scales Exp. Jet Type Uj H hi/d R Js g' lq Im zm zb (m/s) (m/s) (cm) (m/s2) (m) (m) (m) (im) 1 Bottom 0.34 0.38 10 6.92 1.12 0.16 0.109 0.011 0.12 0.013 1.34E-04 2 Intermediate 0.34 1 10 6.92 2.97 0.43 0.119 0.011 0.31 0.033 3.89E-04 3 Surface 0.27 0.82 8.7 5.89 3.05 0.52 0.114 0.011 0.26 0.034 6.13E-04 4 Bottom 0.36 0.82 11.5 8.11 2.24 0.28 0.112 0.011 0.26 0.025 2.38E-04 5 Bottom 0.24 0.1 8.1 5.42 0.41 0.08 0.119 0.011 0.03 0.005 1.03E-04 6 Intermediate 0.24 0.5 8.1 5.42 2.04 0.38 0.118 0.011 0.15 0.023 5.09E-04 7 Surface 0.24 0.77 8.1 5.42 3.15 0.58 0.116 0.011 0.24 0.035 7.75E-04 8 Intermediate 0.31 0.8 10.3 7.16 2.54 0.36 0.113 0.011 0.25 0.029 3.68E-04 9 Surface 0.31 1.03 10.3 7.16 3.30 0.46 0.117 0.011 0.32 0.037 4.95E-04 10 Surface 0.31 1.19 10.3 7.16 3.81 0.53 0.117 0.011 0.37 0.043 5.74E-04 11 Bottom 0.31 0.6 10.3 7.16 1.91 0.27 0.115 0.011 0.19 0.021 2.81E-04 12 Bottom 0.36 0.6 12 8.5 1.64 0.19 0.114 0.011 0.19 0.018 1.78E-04 13 Bottom 0.36 0.9 12 8.5 2.46 0.29 0.117 0.011 0.28 0.028 2.75E-04 14 Intermediate 0.36 1.19 12 8.5 3.28 0.39 0.117 0.011 0.37 0.037 3.65E-04 15 Surface 0.36 1.39 12 8.5 3.83 0.45 0.117 0.011 0.43 0.043 4.26E-04 16 Bottom 0.24 0.4 8.5 5.74 1.63 0.28 0.104 0.011 0.13 0.018 3.59E-04 17 Surface 0.24 0.7 8.5 5.74 2.85 0.5 0.105 0.011 0.23 0.032 6.32E-04 18 Intermediate 0.29 0.7 10 6.92 2.40 0.35 0.11 0.011 0.22 0.027 3.95E-04 19 Surface 0.29 0.99 10 6.92 3.43 0.5 0.113 0.011 0.31 0.038 5.80E-04 20 Intermediate 0.29 0.72 10 6.92 2.47 0.36 0.111 0.011 0.23 0.028 4.10E-04 21 Intermediate 0.29 0.72 10 6.92 2.47 0.36 0.035 0.011 0.41 0.028 1.28E-04 22 Intermediate 0.29 0.72 10 6.92 2.47 0.36 0.005 0.011 1.06 0.028 1.89E-05 23 Surface 0.24 0.77 9.4 6.45 3.15 0.49 0.119 0.011 0.24 0.035 7.89E-04 24 Surface 0.24 0.95 9.4 6.45 3.87 0.6 0.119 0.011 0.29 0.043 9.70E-04 25 Bottom 0.24 0.32 9.4 6.45 1.30 0.2 0.119 0.011 0.10 0.015 3.26E-04 26 Bottom 0.24 0.4 9.4 6.45 1.63 0.25 0.119 0.011 0.12 0.018 4.08E-04 27 Intermediate 0.24 0.55 9.4 6.45 2.24 0.35 0.119 0.011 0.17 0.025 5.61E-04 Note: 1. Experiments 16 through 19 used for lateral temperature (bifurcation) and trajectory. 2. Experiments 20, 21, and 22 used for buoyancy effects. 3. Experiments 23 through 27 used for bifurcation. To help illustrate the effect of the shallow flow on the behaviour of a vertical buoyant jet, the results of both jet trajectory and dilution at various depths and velocity ratios are described 108 below. They are then compared to the previous results of Wright (1977) and Hodgson (1991). 4.1.1.1 Jet Trajectory The jet centreline trajectories were analyzed to determine the effect of the shallow crossflow on the vertical growth of the jet. As shown in Figure 2.3, after the jet exit, the discharge flow penetrates vertically into the ambient current. At some point downstream, the flow is bent by the ambient flow, eventually traveling parallel to the free stream current. The extent of each of these regions was found to be a function of both the ambient depth and the velocity ratio. Using the method outlined in Section 3.1.4, the jet centreline trajectory was determined from digitized video images. For a given flow depth, the height of vertical rise is dependent of the velocity ratio (Figure 4.3). The larger the velocity ratio, the stronger the jet and the higher the vertical jet trajectory. The vertical height of rise before bending for the strongest jets is determined and limited by the actual depth of the crossflow. The effect of the flow depth on the stability of the trajectory also varied depending on the velocity ratio. For weaker discharges, the trajectories were influenced by the free surface, the bottom boundary layer or both, depending on the vertical location of the jet in the free stream. For example, for Experiment 12 in Figure 4.3 (b), the trajectory can be seen to be oscillating up and down in the crossflow. This is due to the interaction of the bottom edge of the flow with the shear flow in the boundary layer and the inherent complex nature of the flow. Wake attachment due to the discharge structure may also have drawn the jet flow towards the flume bottom. Similarly, for slightly stronger flows that approach the surface asymptotically, such as Experiment 8 and 14, the upper edge of the jet flow interacts with the surface boundary layer. This results in the flow being drawn up towards the surface and becoming attached, similar to the Coanda Effect. For all experiments, within the momentum dominated far field region, the flow depth and velocity ratio dominated the behaviour of the flow. Overall, the jet trajectory was found to be dependent on the relative jet strength, Js (Figure 4.4). 4.1.1.2 Dilution Temperatures were measured according to the method laid out in Section 3.1.3. The dilution ratio (S) defined as: Cm where C0 = initial discharge excess temperature = Te- Ta Cm = maximum centreline excess temperature = Tm - Ta. This gives the minimum centreline dilution. A comparison of dilution as a function of downstream distance for a given flow depth, shown in Figure 4.5, also indicates a dependency on the velocity ratio, R. The larger the velocity ratio, the lower the overall growth of the dilution curve, with the rate actually leveling off at a given distance downstream. Prior to this point, dilution increased with R due to the higher levels of turbulence and therefore entrainment in this region (Figure 4.6). Overall, the 110 stronger jets at a given flow depth, the more the dilution is restricted by the shallow cross flow. A comparison of dilution, as 1/S, and trajectory for four experiments is shown in Figure 4.7. From this it is evident that the leveling off of the growth of the dilution curve corresponds roughly to the downstream location where the jet impinges the free surface. The dilution at this point is inhibited by both the limited vertical entrainment from above, and the recirculation and reentrainment of 'contaminated' jet fluid in the underside of the bend. Altogether, the dilution, as depicted in Figure 4.8, is a function of both the velocity ratio and the flow depth, or collectively a function of Js, increasing as Js decreases. 4.1.2 Comparison to Previous Results 4.1.2.1 Trajectory Both Hodgson (1991) and Wright (1977) proposed the following relationship for trajectory of a vertical jet in the Momentum Dominated Far Field: ^2 Z,„ fx* 7 \ m J 4.2 where C2 is a function of the velocity ratio given by: C2=1.21 Rm (postulated by Hodgson (1991) based on Wright, (1977)) 4 3 C2=1.09 R022 (Hodgson, 1991) 4 4 111 For the purposes of comparison to the results in this project, Hodgson's and Wright's coefficients (C2) were calculated using the mean experimental velocity; resulting in coefficient values of 1.34 and 1.41, respectively. Figure 4.9 shows the trajectory as a function of x/zm for over the range of experimental jet strength ratios in addition to the trends predicted by equation 4.2. Prior to the influence of the free surface, all the jet trajectories in the momentum dominated far field (MDFF) region, z> zm and x < IM, followed the 1/3 power law. In contrast to the unconfined conditions, the vertical penetration of the jets in the confined receiving water was lower. The maximum level of vertical rise of the jet centreline is dictated by the free surface. The vertical growth of these jets seemed to be slightly higher than that predicted by Hodgson (1991). This is most likely due to the different methods used for the determination of jet trajectory: photographs versus digitized video images. The trajectory determined from photographs is the location of the centre of the jet axis whereas that determined using the video image is the location of maximum intensity. The latter has been shown to correlate to that of maximum temperature or concentration. The location of the jet centreline trajectory from photographs is lower than that of maximum concentration as shown in the comparison of the two M D F F trajectory coefficients determined by Wright (1977): 1.6 from photographs and 2.1 from concentration. Due to the presence of the finite cross flow, all the jet trajectories 'deviated' from this 1/3 slope. The extent of this deviation is a function of Js and the proximity of the flow to either the top or bottom boundary. This can be demonstrated more effectively by comparing the actual jet centreline trajectory (z</hi) to that predicted (z,Ai) using equation 4.2 where C2 was 112 determined using experimental trajectory data. Assuming that prior to surfacing, the shallow crossflow has minimal effects on the trajectory of the strongest jet, a coefficient (C2) value of 1.48 was determined based on pre-surfacing trajectory data from Experiment 10. Rearranging equation 4.2 in terms of Zp/hi gives: z„ z x \ Z m J 4.5 A comparison of predicted (zp/hi) and actual (zjhi) jet trajectories is shown in Figure 4.10 for a range of jet strength ratios. The shallow crossflow influenced the jet trajectories, the extent of which is a function of Js. Jets with lower jet strength ratios had a lower overall vertical height of rise. The stronger jets (Js>0.44) all followed the appropriate power law relationship, relatively unaffected by the crossflow, until the jet actually impinged on the free surface (z/hi=0.90). At this point the flow followed a trajectory essentially parallel to the free surface. Jets exhibiting this type of behaviour are defined as Surfacing Jets. The growth of all other jet trajectories was inhibited by the presence of the shallow crossflow, even prior to surfacing. The weakest jets, defined as Bottom Jets (7S<0.30), bent away from the trajectory predicted by the 1/3 power law sooner, resulting in a flow travelling essentially parallel to the ambient current within the bottom 60 percent of the flow. These jets never reached the free surface within the M D F F region. The proximity of the turbulent boundary layer had more influence on the behaviour of this jet type than the free surface. This can be seen in the contour plots of jet thickness in Figure 4.11. The outer edge of the jet is defined as the point were 50 percent of maximum centreline concentration is achieved (C/Cm=0.50). The lower edge of the weaker jets reached the turbulent boundary layer region 113 (z/H=0.20) which may have drawn jet fluid towards the bottom. This interaction of the jet fluid with the shear flow in the boundary layer increased the dilution measured for these weaker jets. Wake attachment, caused by the presence of the discharge structure, also resulted in some of the weaker jets being drawn toward the bottom immediately following the jet exit. This is most apparent in the contour plot for Experiments 5 in Figure 4.11 (a). The trajectory of the Intermediate Jets (0.30 < 7* < 0.44) were also found to be influenced by the limited ambient depth, deviating away from the predicted path. The vertical height of rise is higher than that of the Bottom Jets and the upper edge of the jet (C/Cm=0.5) approaches the free surface asymptotically within the momentum dominated far field region (Figure 4.12). In some cases, as the edge of the jet reached the free surface, the jet trajectory became rapidly attached to the surface as shown in Figure 4.12 (a). This is due to the slight reduction in the ambient flow velocity at the free surface. This causes a pressure drop (drag) and causes the flow to become attached to the surface by the Coanda Effect. The behaviour of some of the Intermediate Jets was further complicated by the proximity of the bottom jet edge to the bottom. Although the growth of the trajectories of these flows was inhibited by the shallow flow, interaction with the flow in the boundary layer also influenced the behaviour of the jet flow. This will be discussed in more detail in the next section. 4.1.2.2 Dilution According to Wright (1977), the minimum centreline dilution in the momentum dominated far field (MDFF) region can be represented by the following relationship: 114 M 7 V '» J SQ U„zl 4.6 Substituting the jet trajectory relationship given in equation 4.2 and introducing the velocity ratio (/?), gives the following relationship: cjcm=i = c( * V 3 R R \z»< J 4.7 where Cd = C^C? and is given in Table 4.3. Table 4-3 Wright's (1977) M D F F Coefficients c2 c6 cd 1.41 (concentration) 0.38 0.76 1.6 (photographically) 0.38 0.97 The nondimensional centreline dilution as a function of x/zm is plotted in Figure 4.13 for various jet strength ratios. The growth rate of the dilution data is consistent with the results of Wright (1977), following the power law given in Equation 4.7. The effect of the finite flow depth on the overall dilution coefficients varied with Js, increasing with decreasing jet strength ratios. Prior to any effects of the free surface, the 2/3 slope fit the data reasonably well. The coefficient (Cd) for each experiment was calculated using the method of least squares regression assuming a slope of 2/3. Only data points that are within the M D F F region, prior to any influence of the free surface are used to calculate this coefficient. Refer to Table 4.4 for a listing of the calculated coefficients. The coefficients are within the region of those determined by previous research. A plot of the coefficient versus relative jet strength (Js) in 115 Figure 4.14 shows that, with the exception of the two extremely high values, no trend could be found. The coefficient Cd was essentially constant over the range Js with a median value of 0.90±0.03. The dilution in the two extreme cases, Experiment 1 and 9, was significantly enhanced by the turbulent boundary layer. Immediately following the jet exit, these jets were influenced by the shallow crossflow and the proximity of the bottom boundary, and as such were not included in the calculation of a pre-surfacing dilution coefficient. Table 4-4 Dilution Coefficients Exp. No. Js Q Cmdff 1 0.16 1.27 1.25 2 0.43 1.00 0.85 3 0.52 0.95 0.66 4 0.28 0.95 0.93 5 0.08 1.66 1.66 6 0.38 1.03 0.91 7 0.58 0.90 0.67 8 0.35 0.89 0.84 9 0.46 0.89 0.75 10 0.53 0.88 0.66 11 0.26 0.92 0.96 12 0.19 0.96 1.07 13 0.29 0.87 0.94 14 0.38 0.87 0.80 15 0.45 0.89 0.70 After surfacing, 2/3 slope also adequately fit the data. Using the method discuss previously, overall coefficients (Cmdff) where calculated using all the data points in the M D F F region (Table 4.4). The coefficient varies with the jet strength ratio, increasing with decreasing Js values. A plot of Cmdff against the jet strength ratio is shown in Figure 4.15. The variation in the coefficient with jet strength can be approximated by the following equation with a high degree of correlation (R2=0.95): Cmdff=0.5(Jsras 4.8 This trend is contrary to the variation in the dilution coefficient for unconfined flows. The coefficient calculated based on Wright's trajectory coefficient (Q) increases with velocity ratio for unconfined ambient currents. In his experiments, the depth of flow was held constant and as such the velocity ratio is proportional the jet strength ratio. Therefore, for unconfined flows, the dilution coefficient can be said to increase, not decrease with jet strength ratio. The difference in this behaviour is due to the proximity of both the confining boundaries in the shallow crossflow. The mixing and subsequent dilution in the stronger jets is inhibited by interactions free surface while the presence of the turbulent boundary layer increased the dilution of the weaker jets. Furthermore, Wright's experiments were carried out using a negatively buoyant jet towed through a quiescent ambient fluid. As a result, the ambient flow structure in his experiments is dramatically different, with significantly reduced levels of ambient turbulence and minimal boundary layer effects. Marks (1996) saw the same decreasing trend in C m ^ w i t h velocity ratio, and subsequently Js, for a single buoyant jet discharging at an angle of 45 degrees into a shallow crossflow of constant depth. A jet classification based on M D F F dilution coefficients is shown in Figure 4.16. For flows with Js less than approximately 0.30, the proximity of the flow to the bottom and the shear flow of the boundary layer results in levels of dilution significantly higher than that predicted assuming a coefficient value of 0.90. That is, the dilution is higher than that expected if the influence of the finite cross flow depth was negligible. The effect of the shallow cross flow on the Intermediate jets (0.30 < Js < 0.44) is a slight reduction in the rate of dilution due to limited vertical entrainment as the outer edge of the jet 117 approached the free surface. These jets had M D F F dilution coefficients between approximately 0.75 and 0.90. In the shallower flows, although the dilution of these jets was inhibited at the upper edge, the dilution at the bottom edge may actually be increased due to interaction with the shear flow in the boundary layer. This is most evident in Experiment 6, with a M D F F coefficient of 0.91 The interaction of the surface jet (Js > 0.44) with the finite depth results in an obvious drop at surfacing and subsequent reduction in the overall rate of dilution ( C , ) U # F < 0.75). The growth of the dilution curve of stronger jets appears to level off at the surface, restricted by the shallow flow conditions. After surfacing, once the flow has been reestablished, the dilution begins, once again, to increase at a rate less than that prior to surfacing, with a reduction in the overall dilution after surfacing. For some of these jets, the flow after surfacing is further complicated by the onset of bifurcation or splitting of the plume. This will be discussed in more detail in Section 4.1.2. Combining equations 4.7 and 4.8, the dilution in the M D F F (\<x/ <10) region of a / Zm vertical buoyant jet in a shallow crossflow can be estimated using: f .. \Vi 4.9 x c / „ vs -=o.5(js r 5 R 7 \ "> J Substituting Js =yn j d and zm = -Jn/ARd into equation 4.9, the dilution relationship can be given in terms of the downstream distance (x/d), velocity ratio (R) and flow depth (hj/d): 118 The overall M D F F dilution of a vertical buoyant jet in a shallow crossflow at a given depth will decrease as the velocity ratio increases. The experimental data for dilution (5) and those predicted by equation 4.9 (Sp) are compared in Figure 4.17. Generally the dilution data fit the predicted trend, especially Intermediate and Bottom Jets, with the following exceptions: a) Farther downstream, the dilution of the weakest Bottom Jets was significantly higher than predicted. In these cases, this increased level of dilution occurred outside of the M D F F region and therefore outside the applicable region for equation 4.9. At this point, the mixing may be influence by any residual buoyancy. b) The dilution of Experiments 2 and 6 was higher than predicted due to the influence of the boundary layer. In some of the shallower flows, although the dilution is limited at surfacing, the proximity of the flow of the boundary layer enhanced the dilution of these jets. c) Generally, the dilution of the Surface Jets was slightly underestimated prior to surfacing and slightly overestimated after surfacing. A noticeable exception to this is Experiment 7 with increased levels of dilution after. This is predominantly due to the onset of bifurcation or splitting of the plume. This results in an apparent higher level of centreline dilution. This will be discussed in more detail in Section 4.1.2. The dilution data for various jet strength ratios were also compared to the following relationship defined by Hodgson and shown in Figure 4.18: 119 S = 1.09 V d J This relationship adequately fit the dilution data prior to any effects due to the proximity of the free surface or the flume bottom. Once again, the interaction of the jet flow with either the top and bottom boundaries resulted in either higher or lower dilution than those predicted using Equation 4.11, depending on the relative jet strength (Js). 4.1.3 Bifurcation For some of the stronger jets, the flow after surfacing may be complicated by the onset of bifurcation. This bifurcation results in the splitting of the plume into two concentration nodes, with the resulting centreline dilution being much lower than that in the two nodes. This may account for the perceived increased dilution growth after surfacing seen in Figure 4.17 for Experiment 7. For Experiments 16 through 19, temperature profiles were taken laterally across the jet flow at a various distances downstream of the jet exit. Contour plots of C/Cm are shown in Figures 4.19 (a) through (d). Bifurcation of the jet flow was found to occur in Experiment 17 and 19. These jets correspond to Surface Jets, both with a jet strength ratio of 0.50. Temperature differences, indicative of concentration differences, were found to be 10 to 15 percent higher in the concentration nodes. An additional qualitative video analysis was done on five different discharge scenarios. Cross sectional video images were collected 32 cm downstream of the jet exit (see Appendix E for details of this video analysis). Colour images of these cross sections are shown in 120 Figure 4.20. Within the test section (x <32 cm), bifurcation was found to occur in discharge scenarios with jet strength ratios greater than 0.47. This corresponds to Surfacing Jets. For these jets, there is sufficient vorticity at surfacing to cause the splitting of the plume or bifurcation into two distinct nodes. 4.1.4 Buoyancy Effects For all experiments, within the MDFF region, the buoyancy, represented by the Froude number, was found to have minimal impacts on the jet trajectory and dilution. A comparison of both jet trajectory and dilution for experiments with varying Froude numbers and similar jet strength ratios is shown in Figure 4.21. A similar comparison shown in Figure 4.22 is made with experiments with varying jet strength ratios and similar Froude number. Within the range of experimental Froude numbers examined, the flow depth and velocity ratio, or collectively the jet strength ratio dominated the behaviour of the flow. Three further experiments were carried out to investigate the effects of large variations in buoyancy on the near field mixing (Experiments 20, 21 and 22). A l l experimental variables were held constant (Js = 0.36) with the exception of the temperature of the discharge which was varied between 17 and 50 degrees Celcius (g' = 0.005 to 0.111). The vertical jet trajectories are plotted in Figure 4.23 as a function of the dimensionless downstream distance. Immediately downstream of the jet exit, buoyancy effects on the jet trajectory were negligible. The positive buoyancy did marginally begin to influence the flow farther downstream, resulting in a slightly higher jet trajectory for the higher temperature discharge prior to reaching the free surface. Regardless of this, all three jets reached the free surface at essentially the same distance downstream. 121 There may be effects in the far field region when the jet momentum is diminished and any residual buoyancy may enhance transverse mixing and limit vertical mixing. Taking the ratio of the stabilizing influence of the buoyancy per unit depth or width to the mixing power of the ambient current gives two dimensionless groups, *.-, and W , (Fischer et al., / H(u y / w(u y 1979). These can be used to estimate when the influence of the effluent buoyancy on vertical and transverse mixing must be considered. According to Fischer et al. (1979), if W . » , is / H(u )' less than five, buoyancy effects on transverse mixing are negligible. Bruno et al. (1990) found that the minimum value of this parameter ranged from 4.3 to 23.6 depending on the individual discharge scenario and channel roughness. When W , *.-,is less than one / w(u y (Fischer et al., 1979), buoyancy can be said to have negligible effects on the vertical mixing. These two parameters were calculated for a Bottom Jet (Experiment 5) and a Surface Jet (Experiment 15): Exp. 5 B/ 3 = 6.6 y . .,3=1-1 / H(u ) /w(u ) Exp. 15 B/ =24.8 y' . . , = 5.9 Details of this calculation can be found in Appendix G. These results indicate that the buoyancy of the effluent may begin to influence the mixing farther downstream, when the initial discharge momentum no longer dominates the mixing. This influence would be more pronounced for Surface Jets than for the weaker Bottom Jets. 122 4.1.5 Comparison of Jet Types When dealing with the discharge of contaminants that may have a detrimental effect on the receiving environment, the downstream dilution is a key factor in the design of the discharge system. Therefore, a comparison of the dilution achieved by the three jet types provides insight into which scenario will provide optimal dilution under various ambient conditions: specifically low flow and peak flow. Dilution for Bottom, Intermediate and Surface Jets in terms of fraction of effluent concentration (1/5) is shown in Table 4.5 and Figure 4.24. Initially, the dilution is greater for the stronger Surface Jet until the effect of the free surface begins to dominate the behavior and subsequent mixing of the discharge. At this point, the dilution for both the Intermediate and Surface Jets becomes inhibited by the proximity of the free surface, while the dilution of the Bottom Jet is enhance by interactions with the boundary layer. At approximately 22 centimetres downstream, this jet type has a pollutant concentration approximately 58% lower than the Surface. Table 4-5 Comparison of Dilution (1/S) for Three Jet Types Jet Type Js x=2 x=ll x=22 x=32 x=42 Bottom 0.08 0.56 0.13 0.07 0.05 0.04 Intermediate 0.35 0.63 0.18 0.12 0.10 0.09 Surface 0.58 0.52 0.26 0.17 0.11* 0.08* (* onset of bifurcation) 4.1.6 Application of Jet Types to Fraser River Conditions The above jet characterization can be applied to discharge conditions in the Fraser River. The following are typical river and discharge flow data for both the peak and low flow periods (Marks, 1996). The flow rate is from the mean monthly values from the 123 Environment Canada Water Survey Station at Shelley. The river velocity was calculated based on the assumption of a rectangular channel 300 m wide within the vicinity of the discharge. The depth was calculated using Manning's formula using a slope of 0.005 (Northwest Hydraulics, 1994) and a Manning's n of 0.03 (Vine, 1996; Henderson, 1966). The effluent discharge information is for Northwood and was provided by Environment Canada (Marks, 1996). The discharge velocity is calculated by dividing the flow rate by the number of ports and the port area. The Northwood diffuser is a multiport system with three risers located 15m apart. Each riser consists of two ports at 90 degree angles to each other, discharging horizontally at an angle of 45 to the river flow. In total there are six ports; with a diameter of 0.5 m. Low Flow River Discharge <2= 183 m 3/s Q= 1.8m3/s Ua = 0.6 m/s Uj= 1.5 m/s h{ = Im R = 2.5 Peak Flow River Discharge 2 = 2200m 3/s 2 = 1.4 m 3/s Ua= 1.9 m/s Uj= 1.2 m/s hj = 3.9 m R = 0.6 J,= 1.2 Js = 0.08 The low flow discharge is a Surfacing Jet which surfaces immediately downstream of the point of discharge, resulting in an overall decreased centreline dilution due to the free surface. In addition bifurcation may occur, splitting the plume into two concentration peaks. This will result in elevated concentrations of contaminants for a farther distance downstream, 124 potentially affecting aquatic habitat including overwintering chinook salmon and potential bank attachment of the split plume. The peak flow discharge is a Bottom Jet which will not be influenced by the effects of the free surface and will achieve optimal dilution due to potential interactions with the bottom boundary layer. This interaction with the river bottom may also have negative impacts on the receiving environment, resuspending contaminated sediments and exposing benthic habitat to high levels of contaminants. 4.1.7 Comparison with CORMIX1 Output The suitability of the CORMIX 1 model for discharges into shallow crossflows was then investigated. CORMIX 1 version 3.1 was run for the discharge and ambient conditions of Experiments 1 through 15 and the model output compared to actual dilution and trajectory data. Based on the model's flow classification, all except Exp. 1 and 5 were determined to be Class V2, a vertical momentum dominated buoyant jet. For discharges of this type, the following flow zones are predicted to exist. 1) Weakly deflected jet in a crossflow dominated by the initial momentum of the jet (indnf). 2) Strongly deflected jet in a crossflow bent over by the ambient current. 3) Layer boundary approach consisting of a bent-over jet/plume with uniform concentration distribution. 4) Passive ambient mixing dominated by the ambient shear flow. Typical comparisons of experimental dilution data and those predicted by both C O R M I X and equation 4.9 are presented in Figure 4.25 for the three jet types: bottom, intermediate and surfacing. For Surfacing Jets (Js > 0.44), CORMIX 1 slightly underestimates the minimum dilution in the region prior to surfacing. At surfacing, CORMLX1 assumes that the concentration profile changes from a Gaussian to a uniform top hat distribution. This assumption results in a rapid jump in the predicted dilution as the discharge surfaces and subsequent overestimation of the dilution downstream. The same result is seen with Intermediate Jets (0.30 < Js < 0.44). In reality, the surfacing of the jet caused a reduction in the dilution due to limited vertical mixing and re-entrainment of jet fluid. C O R M I X was able to predict the location where the free surface would begin to influence the mixing of the discharge. The results are quite different for Bottom Jets (J, < 0.30). CORMIX 1 underestimates the minimum dilution within the experimental region. Similar results were seen by Marks (1996). This difference is most likely due to the fact that C O R M I X does not take into account the presence of a turbulent boundary layer, which enhanced the dilution of the weaker Bottom Jets. In all cases, equation 4.11 more adequately predicted the dilution. The relative error in the dilution predicted by CORMLX is plotted as a function of jet strength ratio in Figure 4.26. Relative error is defined as the error in the predicted result divided by the actual dilution (Sacnml - SC0RMlx /Sacmal). Prior to substantial effects of the shallow crossflow, CORMIX underestimates the minimum dilution. As the jet strength ratio increases, the error between predicted and actual dilution reduces. C O R M I X was able to adequately predict the dilution of the stronger jets (Js > 0.44) prior to surfacing, but farther 126 downstream, where the proximity of either of the boundaries influence the jet behaviour, CORMLX did not adequately predict minimum dilution. CORMIX underestimates the dilution for the weaker Bottom Jets (Js < 0.30) with the error increasing as the jet strength ratio decreases, underestimating the dilution by as much as 80 percent. For both Surfacing and Intermediate Jets (Js > 0.30), CORMIX overestimates the dilution by 20 to 40 percent. Experiment 1 was classified a V2(A2) which is a vertical momentum dominated buoyant jet with bottom wake attachment without lift off. The following flow regions are predicted to occur in this discharge scenario. 1) Wake recirculation in which the discharge is strongly deflected by the ambient flow and subsequent bottom attachment. 2) Passive ambient mixing of the deflected discharge. Experiment 5 was classified at V I (A 1) which is defined as a vertical momentum dominated buoyant jet with lift off. This discharge scenario is predicted to follow the following flow zones. 1) Wake recirculation with subsequent bottom attachment. 2) Lift off form the bottom due to the positive buoyancy of the discharge. 3) Strongly deflected plume in a crossflow rising slowly to the surface. 4) Layer boundary approach as plume reaches the surface. 5) Passive ambient mixing. 127 Comparisons of experimental dilution data and those predicted by C O R M I X and equation 4.9 are presented in Figure 4.27 for Experiment 1 and 5. The initial dilution in Experiment 5 is overestimated by CORMLX 1 due to a prediction of rapid dilution during wake recirculation. Following this is a region of negligible dilution during plume lift off from the bottom. In reality, rapid turbulent mixing was evident due to the interaction of the discharge with the shear flow in the boundary layer. The predicted dilution of Experiment 1 is very close to the actual levels achieved. A similar comparison was made with trajectory data. Generally the predicted trajectory closely follows the actual trajectory, lower in the near field and higher after surfacing. For Bottom Jets, as the velocity ratio (R) increases, the difference between predicted and actual decreases. For jets where wake attachment is predicted, the jet centreline is predicted to remain on the bottom. In summary, C O R M I X 1 is not a suitable tool for modeling of near field mixing in shallow crossflows when the proximity of the free surface and the bottom can drastically impact the flow. The overall dilution is underestimated in weaker Bottom Jets and overestimated in the stronger Intermediate and Surfacing Jets. In the context of pulpmill discharges in the Fraser River, this means an overestimation of the dilution by up to 25 percent in the critical winter low flow period (Surfacing Jet). This could have major implications if decisions on discharge levels are being based on CORMLX output. 4.1.8 Summary of Near Field Mixing Experiments and CORMIX Modeling The presence of a shallow ambient current was found to have significant effects on both the jet trajectory and dilution of a vertical buoyant jet. The extent of the impacts was found to be 128 primarily a function of the jet strength ratio (Js). Based on the jet strength ratio, three jet classifications were identified for this type of discharge in the momentum dominated far field region (MDFF). a) Bottom Jets are defined as weak strength jets (Js < 0.30) whose trajectory and dilution are influenced by the proximity of the flow to the shear flow in the turbulent boundary layer. The dilution of Bottom Jets was significantly increased by the enhanced mixing due to interactions with the boundary layer flow. The trajectory of these jets was inhibited by the shallow flow and some of the weaker flows exhibited attachment to the flume bottom. These jet never surfaced within the experimental region. b) Intermediate jets are defined as jets with Js between 0.30 and 0.44. The behaviour of these jets is more complicated due to the influence of both the top and bottom confining boundaries. Although the growth of the trajectory and the overall jet dilution was inhibited as the jet approached the free surface, dilution was sometimes enhanced due to interactions with the boundary layer. c) Surface Jets are defined as jets that impinged on the free surface immediately downstream of the jet exit (Js > 0.44). Prior to surfacing, the dilution and trajectory were relatively unaffected by the shallow flow. After surfacing, the jet dilution was inhibited due to the lack of vertical entrainment, and the overall M D F F dilution was reduced. These jets also tended to bifurcate after surfacing. Within the momentum dominated far field (MDFF) the dilution was found to follow a power law similar to that previously defined by Wright (1977). 129 S ( x Y3 - = 0.5(Jsr05 — R In general, due to the presence of the free surface, the dilution was found to be a function of the jet strength ratio, Js. The weaker the discharge, the better the dilution. This is contrary to the behaviour of a jet in an unconfined current, with stronger jets having higher rates of dilution. The CORMLX 1 model was shown to be unsuitable for the predicting of dilution of vertical buoyant jets in shallow crossflows. This was most pronounced for Bottom and Surfacing Jets. These discrepancies were primarily due to the lack of consideration of the turbulent boundary layer enhancing mixing of the weaker Bottom Jets and the prediction of increased dilution at surfacing. Incorporating the results of this study, including the dependency of dilution on jet strength (equation 4.9), will provide for better predictions of the dilution and therefore potential impacts of discharges in shallow ambient currents such as rivers. In general, due to the presence of the free surface, the dilution was found to be a function of the jet strength ratio, Js. The weaker the discharge, the better the dilution. This is contrary to the behaviour of a jet in an unconfined current, with stronger jets having higher rates of dilution. 130 4.2 Batch Adsorption Experiments The extent of potential adsorption to Fraser River sediment for dehydroabietic acid and 3,4,5 trichloroguaiacol was determined both theoretically based on Kow and experimentally using Batch Adsorption Studies. Prior to conducting these studies, the organic content: of the sediment used was determined along with the value of Kow for DHA. The theoretical and experimental results are then discussed with respect to the conditions in the near-field mixing zone. 4.2.1 Nature of Fraser River Sediment The sediment used for the adsorption experiments was wet sieved to less than 180 microns. This provided for a fairly homogeneous sediment, removing irregularities such as organic debris. The organic content of the <180 pm sediment was determined to be 1.8% based on loss on ignition at 550°C. This value was used in subsequent calculations for the fraction of organic content. The total fraction of organic material of the entire sediment sample was found to be 2.8%, which is consistent with that found by Prahacs (1994). A particle size fractionation analysis was also carried out on another sample of the sediment and the results are summarized in Table 4.6. Table 4-6 Particle Size Fractionation Size Fraction % Weight % Organic >180pm 6.0 21.0 63 - 180 pm 38.4 1.75 < 63 pm 55.6 1.50 131 4.2.2 Determination of Kow for Dehydroabietic Acid The results for the HPLC determination of the octanol-water partition coefficient (Kow) of dehydroabietic acid are discussed in detail in Appendix A and reported in Table 4.7. Table 4.7 Log {Kow) using H P L C Compound log (Kow) logtfT™) (Xie et al., 1984) Dehydroabietic Acid 4.55 4, 5, 6 - Trichloroguaiacol 3.67 3.92 Tetrachloroguaiacol 4.75 4.76 The values determined for the two guaiacols compare relatively well to those in the literature (Xie et al., 1984). This provides confidence in the value determined for DHA. Although there were no previously documented values of Kow for DHA, similar examinations were carried out by Rohr (1994) and Breen (1995) at U.B.C. Their results give quite different log(Kow) values of 6.2 and 6.3 respectively. Contrary to these findings, Rohr (1994) found that the amount of D H A taken up by hexane in semi-permeable membranes did not reflect these higher Kow values, indicating that the lower value of 4.55 may be more representative of DHA's adsorption capability. To provide for further insight into the accuracy of this octanol-water partition coefficient, it was also calculated based on aqueous solubility using the equations discussed in Section 2.2. These calculations were also done for two reference compounds, pentachlorophenol and DDT. The results are given in Table 4.8. 132 Table 4.8 Calculated log(Kow) using Aqueous Solubility Compound Solubility Calculate log (Kow) using: (mol/1) Eq. 2.1 Eq. 2.2 Eq. 2.3 Eq. 2.4a Eq. 2.4b Average DDT 1.82 X 10" / c ) 5.49 6.90 6.39 5.78 6.14 PCP 5.26 X 10" i d ) 3.85 4.54 4.41 4.42 4.30 D H A 2.19 X 10" i e ) 4.10 4.91 4.72 4.76 4.73 4.64 a) Calculated using a and b for sub. benzenes. b) Calculated using a and b for misc. pesticides. c) Schwarzenbach et al., 1993. d) Suntio etal., 1988. e) Unkulavasapaul, 1984. The log(Kow) determined by the author using HPLC is consistent with that calculated based on solubility. Information on the relative values of Kow can also be drawn from a comparison of aqueous solubility. hog(Kow) is inversely proportional to solubility, decreasing as aqueous solubility increases. The \og(Kow) values determined by both Breen (1995) and Rohr (1995) indicate that D H A is even more hydrophobic than the well known pesticide DDT. This is not consistent with DHA's solubility, which is approximately two orders of magnitude greater than DDT's. Furthermore, the level of solubility of D H A is similar to that of PCP, and therefore should have a relatively similar log(A"mv) value. In summary, given the above discussion, the value of log(^f m v . ) for dehydroabietic acid is taken to be 4.55 and this value will be used in all subsequent calculations. 4.2.3 Determination of Kp and Theoretical Adsorption using K„w Using Equation 2.44, K can be calculated using Kow and an organic content of 1.8%. This is based on the assumption that both D H A and 3,4,5 T C G behave similarly to other weak 133 organic acids such as chlorinated phenols. 3,4,5 TCG has a log(Kow) value of 4.11 (Xie et al., 1986). Adsorption to a sediment containing organic content can be calculated using the equations previously outlined in Section 2.2. The percent ionization is given by: pH - pKa = log ([A-]) {[HA]] 4.12 At a pH of 6 for 3, 4, 5 -trichloroguaiacol and dehydroabietic acid, the percent ionization is 2.7% and 5.3% respectively. The partition coefficients, K , for the undissociated compounds are calculated using Kp=focl.05(Kj ,0.82 4.13 based on Schellenberg et al. (1985) and are shown in Table 4.9. Compound Partition Coefficient (Kp) Dehydroabietic Acid 101.7 3,4,5 Trichloroguaiacol 44.32 cm The partition coefficient, K , has the units of and is the ratio of equilibrium gs concentrations of the contaminant in the sediment and water fractions: Cs _ /gs _ concentration of pollutant in sediment 4.14 w ww / concentration of pollutant in solution / cm1 134 where ws is the weight of compound associated with the solid, gs is the amount of solid, ww is the weight of compound remaining in solution and cm3 is the volume of solution. Rearranging equation 4.14 to solve for ws yields: w s = K n — T x w = Kx[sed]xw 4.15 ' cm' v where [sed] is the concentration of sediment. It should be noted that Kp is an equilibrium constant and w is the weight of compound in solution in equilibrium with the sediment and is defined as the difference between the original amount initially added (vv„) and the amount associated with the sediment (ws), or wa - ws. Using this and rearranging equation 4.15 gives us the following equation for ws. Kp[sed] ws=~,—771—TT 4.16 \+Kp[sed\ The percent adsorption of the undissociated component is then calculated by: w. K [sed] % Adsorption„nr i = — x 100 = - — — R — - r x 100 4.17 wo 1 + Kp[sed] The overall percent adsorption is given by multiplying the above equation by the fraction of the total compound that is in the undissociated form. The results of these calculations for various sediment concentrations are given in Table 4.10 for both D H A and 3,4,5 TCG. As expected the theoretical amount of adsorption decreases with sediment concentration as available adsorption sites decrease. 135 Table 4-10 Calculated Adsorption Sediment Concentration (mg/1) % Adsorption - 3,4,5 TCG % Adsorption -D H A 10000 31.5 47.7 5000 18.8 31.9 2000 8.5 16.0 1000 4.4 8.7 500 2.3 4.6 200 0.9 1.9 133 0.6 1.3 4.2.4 Results of Batch Adsorption Studies 4.2.4.1 Dehydroabietic Acid The results of the batch adsorption studies for D H A are presented in Table 4.11 and Figure 4.28. Generally, the results are extremely variable with only the 5 g/1 sediment concentration showing any significant adsorption of DHA. Further analysis of the results indicate that procedural (surrogate) losses of approximately 80 to 90 % during analysis may largely account for this apparent adsorption. Similar to the calculated results in Section 4.3.2, there is a trend in the data indicating that the amount .of adsorption decreased as the concentration of sediment decreased. No actual estimate of the equilibrium partition coefficient, Kp, could be made due to the extreme variability of the results. The actual increase in concentration of D H A in some of the samples indicates that there were similar compounds being released from the sediment that interfered with the actual analysis. 136 Table 4-11 Results of DHA Adsorption Experiment Sediment Concentration % Adsorption mg/l 10000 -8.21 5000 19.2 2000 5.18 1000 -0.61 500 -6.18 200 -5.36 133 -0.3 4.2.4.2 Chlorinated Guaiacols Batch adsorption results for the 3, 4, 5 - trichloroguaiacol are shown in Figure 4.29. These results are based on analysis of the guaiacols using the autoanalyzer method discussed in Section 3.2. Similarly, it is evident that there are compounds similar to guaiacols being released during the 24 hour batch adsorption experiment which interfered with the analysis, resulting in a perceived increase in concentration or desorption. In order to quantify this, a 24 hour test was carried out with sediment only at various concentrations. The results are shown in Figure 4.30. An attempt was made to correct for this interference using the following correlation between amount of interference (y) in mg/l and the sediment concentration (x) in mg/l: y = 0.000111* 4.18 The corrected adsorption data can be seen in Figure 4.31. There is no apparent adsorption, but the interference was not successfully removed. In order to verify these results, the supernatants from the 1:100, 1:500 and 1:1000 batch adsorption experiments were analyzed using the gas chromatograph method discussed in Section 3.2 and Appendix D. The was no 137 change in the concentrations of these three samples from the original solution concentration (1 ppm) confirming that no significant adsorption occurred. 4.2.5 Summary of Kp and Adsorption Determination Although there seems to be some partitioning of D H A out of solution, no conclusive results could be drawn due to the extreme variability of the results. Any perceived adsorption was found to be a function of sediment concentration, with reduced levels at lower suspended sediment concentrations. In the 3,4,5-trichloroguaiacol batch adsorption studies, no significant partitioning out of solution could be measured at the sediment concentrations used. The calculated values of adsorption for both compounds exceed those determined experimentally. It should be noted that this was a preliminary experimental method to determine how much, if any, partitioning to the sediment fraction was taking place. The intent was to further refine the experimental procedure, remove many of the interferences and enhance the reliability of the results. The following is a summary of various factors that may have influenced the Batch Adsorption Studies and caused the discrepancy between the calculated and experimental results. Duration of the Batch Adsorption Experiment The calculated results are based on the assumption that equilibrium has been reached between the sediment and the compounds in solution. This may not have been the case during the actual adsorption tests which were cut off at 24 hours. 138 Assumption of Uniform Organic Content For purposes of the calculations, the fraction of organic content in the sediment was assumed to be uniform throughout and set at 1.8 %. This most likely was not the case in the Batch Adsorption Studies. An attempt was made to create a homogeneous mixture of sediment for use in these tests, but complete uniformity could not be guaranteed. As a result, variations in the organic content as well as size fractionation may have occurred between each test vial. This would explain the extremely variable results seen, especially in the D H A analysis. Release of Dissolved Organic Matter As noted by Voice et al. (1983), the presence of dissolved organic material initially associated with the sediment may sorb some of the compounds from solution. This smaller material may have remained with the liquid fraction during phase separation and have subsequently been included in the analysis of the bulk solution, resulting in a reduction in the measured sorption. This release of dissolved organic material was seen by the presence of interfering organic matter released and measured during the sediment-only test described in Section 4.2.4. This effect could not be taken into account in the theoretical calculations. Dissolution of Ionic Compounds During the Batch Adsorption Tests, ionic compounds may dissolve from the sediment into solution, increasing the ionic strength of the mixture and affecting the overall adsorption. As well, some of the compounds released could have been basic in nature and, therefore, increased the actual pH of the solution during the test. As the pH increases, the amount of the undissociated compound is reduced, potentially reducing the overall level of adsorption. 139 Taking these factors into account, the following procedural changes would improve the Batch Adsorption Studies and remove much of the variability: • pH and ionic strength measurement before and after mixing period, • determination of time to equilibrium • measurement of the organic fraction of the remaining sediment in each test after phase separation. Sekela et al., 1995, calculated log(^foc) values for 3, 4, 5-trichloroguaiacol based on suspended sediment and water concentrations measured at various locations downstream of the pulpmills on the Fraser River. These values ranged from 3.88 to 5.25, with higher concentrations associated with the sediment in the winter. The log(Koc) value for 3, 4, 5-trichloroguaiacol predicted, using equation 4.13, is 2.46, 10 to 1000 times less that those measured in the river. The river log(Koc) values are actually equal to or greater that the octanol-water partition coefficient. This implies that there must be another mechanism other, than direct adsorption or partitioning, causing this apparently high level of contaminants associated with the suspended sediment. When looking at the potential for association of contaminants with suspended sediment in the near-field region, the time dependency of the system must also be considered. The results from Section 4.1 show that within this region, dilution is rapid, reaching a level of 10 % of the discharge concentration approximately 30 x/d. In terms of the discharges in the Fraser River, this level of dilution may be achieved within 15 meters of the discharge. For the typical flow conditions seen in the Fraser River (average velocity of 1.0 m/s), this will occur 140 within approximately 15 to 20 seconds. Within this region the mixing is extremely rapid and given the fact that adsorption is not an instantaneous reaction, it should have negligible effects within this region. Even if equilibrium is reached rapidly within 30 minutes, at this river velocity, this would not occur until roughly 2.7 kilometers downstream. Also the effects of the colder river temperature may play an important role in the extent to which adsorption occurs within this region and downstream. As with many chemical and biological processes, adsorption can be affected by temperature, with rates of adsorption being reduced at lower temperatures. Taking these various factors into account, direct adsorption may not play a significant role in the association of pulpmill contaminants with river sediment within the near-field region. There has been an increasing concern about the impact of biological solids in the effluent on the receiving water. This was further highlighted by an apparent increase in flocculation and deposition due to the presence of pulpmill effluent in both the Fraser and Athabasca River systems. Given the conditions in the secondary treatment basins including retention time, elevated temperatures and high fractions of organic content in the solids, significant partitioning of hydrophobic compounds to biological solids should occur in the lagoon. Many of these solids are settled out of solution in the basin prior to treatment, but some of the smaller unsettleable solids remain in solution and are discharged with the effluent. This is especially true in systems that do not provide for any formal secondary clarification. An example of the extent of association of contaminants with small biological solids in the effluent was demonstrated in a preliminay analysis of both dissolved and total fractions of 141 whole mill effluent from Canfor (April, 1994). Guaiacols were analyzed for in samples of total and filtered (0.45 micron) effluent. Anything less than 0.45 microns was assumed to be dissolved and free of biological solids. Well mixed samples were split, half being centrifuged and filtered (glass fiber) and the other remaining as a whole mill sample. Details of this can be found in Appendix H. The concentration peaks of 3,4,5 trichloroguaiacol and 3,4,5 trichlorovanillin were the most pronounced. A comparison of the concentration of each compound in both the dissolved and total mill effluent samples indicated that approximately 40% of the vanillin and 65% of the guaiacol is associated with the dissolved fraction. Therefore significant amounts of these compounds are associated with small solids in the effluent. As a result, a portion of contaminants being measured in whole mill effluent are actually associated with small non-settleable biosolids which are being discharged with the effluent where they can become associated with ambient sediment through various mechanisms including flocculation. The extent to which these solids flocculate with ambient suspended sediment within the near field is discussed in Section 5.0. 142 a) Bottom Jet - Experiment 5 - J s = 0.08 b) Intermediate Jet - Experiment 6 - J s = 0.37 c) Surface Jet - Experiment 7 - J s = 0.58 Figure 4-1 Colour Images of a) Bottom Jet Js=0.08, b) Intermediate Jet Js=0.37 and c) Surface Jet Js=0.58 143 Figure 4-2 Jet Classification Diagram 144 1 0.9 ^ 0 . 8 u % 0.6 'ST £ 0.5 CD | 0 . 4 S o . 3 3 0.2 0.1 0 o O i) hJ6 = 5.42 _ o 0 O O O O O O o O O O o o O O o0 0 0 0 0 0 0 ° + + + + + + + + + + + + + + + + + + + + + ° + + o * * Exp.5 R=0.41 + + • Exp.6 R=2.04 o o Exp.7 R=3.2 * l t—#- _* * 10 15 Dimensionless Downstream Distance (x/d) 20 25 1 0.9 o £-0.7 -§ 0.6 a> CD > H 0.5 CD I 0.4 +-* § 0.3 O +—* •3 0.2 0.1 0 ii) h 1/d = 6.92 1 1 r + + + + ' + + + + + + + + + + * * Exp.1 R=1.12 + + Exp. 2 R=2.97 + + 0 6 8 10 12 14 16 Dimensionless Downstream Distance (x/d) 18 20 Figure 4-3 a) Jet Centreline Trajectories fa/hi) at (i) hi/d = 5.42, (ii) hi/d = 6.92 145 i) 1^/(1=7.16 1 £-0.9 N° £.0.8 10.7 0.6 £ 0.5 +-> c cu O 0.4 w v> | 0.3 o to _ _ c 0 2 E Q 0.1 0 + + o o o o + + + + o + + + 0 o + + x + * * * x X X X X X X x X x X X X X X X x X X X 8 R=2.5 + + 9 R=3.3 O O 10R=3.8 X X 11 R=1.9 10 15 20 Dimensionless Downstream Distance x/d 25 ii) h 1/d=8.5 ^ - 0 . 9 N° ^.0.8 10.7 o 0.6 c £ 0.5 +-* c cu O 0.4 w co a) c o 55 « « 5= 0.2 X X x x x x x x x x x x X *• X X o o o O O O O O O O x O O O O + + o ° o o o o ° o 0 0 o ° o o + + + + + + + + + + + + + + + + + + + + + + ° + + + * ^ * * * * * * * * * * * * * * T * * * * •E 0.3? •+ * * cu E 5 0.1 * * 12R=1.6 + + 13R=2.5 o o 14R=3.3 X X 15R=3.8 10 15 20 Dimensionless Downstream Distance x/d 25 Figure 4-3 b) Jet Centreline Trajectories ( z M at (i) hx/d = 7.16, (ii) hx/d = 8.5 146 0 0 o ir ir ft ir ft ir * ^ ir A A A A A A A A A A A A A A A A A A A A A " " A A A ir * A A * ir A . * 0 0 0 0 0 0 A A * 0 0 . o o o o ° o 0 0 ° 0 0 o . o o o o o n ° o 0 O o o 0 o 0 o-o o o o o o o o o o o o o o ir 0 O o I 0 O * * * * * * * * * * 0.2 h o * _ i L i l l * * Exp.12 J s=0.19 o O Exp.11 J s=0.27 0 0 Exp.8 J s=0.35 ir ir Exp.15 J s=0.45 A A Exp.10 J s=0.53 10 15 20 Dimensionless Downstream Distance (x/d) 25 Figure 4-4 Jet Trajectory at Various Jet Strength Ratios (Js) 147 i)h1/d=5.42 * * Exp.5 R=0.41 + + Exp.6 R=2.04 co 25 o o Exp.7 R=3.15 * * * + + + + o o o o o o o $ + o o + 10 15 20 25 Dimensionless Downstream Distance x/d 30 35 ii) h 1/d = 6.92 12 g 1 0 c o •43 3 E 8 CD c "55 £ 6 CD O •4-* I * Z3 E c 1 2 + + * + I- + * + * * Exp.1 R=1.12 + + Exp.2 R=2.97 10 15 20 Dimensionless Downstream Distance (x/d) 25 30 Figure 4-5 a) Minimum Centreline Dilution (S) at (i) l^/d = 5.42, (ii) V d = 6.92 148 i) (1^(1=7.16 + o o o x + o o * * Exp.8 R=2.5 + + Exp.9 R=3.3 o o Exp.10R=3.8 X X Exp. 11 R=1.9 10 15 20 25 Dimensionless Downstream Distance x/d 30 35 ii) h 1/d=8.5 o x + O x + O o x + o * * Exp. 12R=1.6 + + Exp. 13R=2.5 o o Exp. 14R=3.3 x X Exp. 15 R=3.8 10 15 20 25 Dimensionless Downstream Distance x/d 30 35 Figure 4-5 b) Minimum Centreline Dilution (S) at (i) hi/d = 7.16, (ii) hx/d = 8.5 149 X O + o o o + o Exp. 12R=1.6 + + Exp. 13R=2.5 o o Exp. 14R=3.3 X X Exp. 15R=3.8 0.5 1.5 2 2.5 3 3.5 Dimensionless Downstream Distance x/d 4.5 Figure 4-6 Minimum Centreline Dilution (S) Prior to Surface Effects, hj/d = 8.5 I50 Figure 4-7 Comparison of Jet Trajectory and Dilution for Surface Effects (solid line represents dilution curve without surface effects) 151 16 - * * Exp.12 J s=0.19 14 -o O Exp.11 J s=0.27 0 0 Exp.8 J s=0.35 12 - ir ir Exp.15 J s=0.45 A A Exp. 10 J s=0.53 10 2 h V * O „ © A A 4 O O 0 o 0 o A O A 10 15 20 25 Dimensionless Downstream Distance x/d 30 35 Figure 4-8 Jet Centreline Dilution (S) as a Function of Jet Strength (Js) 152 2.5 E N N ° % 21 cu cu —) cu +—* c CD o CO CO 0) c o CO c CD E 0.5 Wright (1977) C =141 9 9 1 ) 0 =1.34 o ° O" *f + + + + + + + + + + + + + 4 1 A A + + + + + + + X X Exp. 5 J =-.08 s * * Exp. 12 J =0.19 s O O Exp. 11 J =0.27 s 0 0 Exp. 8 J =0.35 s •& & Exp. 15 J =0.45 s A A Exp. 10 J =0.53 s + + Exp. 7 J =0.58 s 3 4 5 6 7 Dimensionless Downstream Distance x/z 10 m Figure 4-9 Jet Centreline Trajectory (zc/zm) at Various Jet Strengths (Js) compared to Trajectories predicted by Wright (1977) and Hodgson (1991) using Equation 4.2 153 X X Exp. 5 J s=-.08 * * Exp. 12 J g=0.19 O O Exp. 11 J s=0.27 0 0 Exp. 8 J =0.35 s T* •sSr Exp. 15 J s=0.45 A A Exp. 10 J s=0.53 + + Exp. 7 J s=0.58 X * ) * X I 1111 If-0.2 0.4 0.6 0.8 1 Predicted Jet Trajectory (z /r^) 1.2 1.4 1.6 Figure 4-10 Actual Jet Trajectory (zjhi) compared to that Predicted (Zp/hj) using Equation 4.5 154 a) Exp. 5 , J =0.08 b) Exp. 12, J s=0.19 Dimensionless Downstream Dischance (x/d) Figure 4-11 Contour Plots of Jet Thickness for Bottom Jets a) Experiment 5 Js=0.08 and b) Experiment 12 Js=0.19 155 b) Exp. 14, J s=0.39 10 15 20 25 30 Dimensionless Downstream Dischance (x/d) 35 40 Figure 4-12 Contour Plots of Jet Thickness for Intermediate Jets a) Experiment 6 Js=0.38 and b) Experiment 14 Js=0.39 156 Figure 4-13 A Comparison of Jet Dilution Data (S/R) to Wright (1977) predicted using Equation 4.7 and C d = 0.76 (concentration trajectory) and 0.97 (photographic trajectory) 157 0.2 0.4 J. 0.6 0.8 Figure 4-14 Pre-Surface Effects Dilution Coefficient (Cd) as a function of J s for Experiments 1 through 15. 3 2.5 2 0.5 -\ Js=0.44 J=0.30 • \ + , 1 1 1 i 1 1 1 1 0 0.2 0.4 0.6 0.8 J s Figure 4-15 MDFF Dilution Coefficient (Cmdff) as a function of J s Figure 4-16 Dilution (S/R) Jet Classification Diagram 160 30, 25 20 CO c g co o < no I 0 Bottom Jets O Intermediate Jets Surface Jets 0 0 5 J =0.0.08 s 1 J =0.16 s • • 12 J =0.19 s X X 11 J =0.27 s + + 4 J =0.28 s rt rt 13 J =0.29 s + + 8 J =0.35 s # * 6 J =0.37 s 0 0 14 J =0.39 s X X 2 J =0.43 s • • 15 J =0.45 s rt rt 9 J =0.46 s + + 3 J =0.52 s X X 10 J =0.53 s 0 0 7 J =0.59 s 10 15 20 Predicted Dilution ( S ^ using Equation 4.9 25 30 Figure 4-17 Comparison of Minimum Jet Centreline Dilution data (S) to Dilution (Sp) predicted by Equation 4.9 161 Jet Dilution at Various Jet Strengths (J g) 45 -X X 5 J =0.08 s 40 - * * 12 J =0.19 s O O 11 J =0.27 s 35 0 0 8 J =0.35 s 30 -•& 15 J =0.45 s X A A 10 J =0.53 s 25 - + + 7 J =0.58 s X 20 X 60 80 100 120 140 Dimensionless Downstream Distance xR/d 160 180 200 Figure 4-18 Comparison of Jet Dilution data (S) to that predicted by Equation 4.11 (Hodgson, 1991) 162 a) Exp. 16 J =0.28 b) Exp. 17 J s =0.50 0 0.2 0.4 0.6 0.8 1 Dimensionless Lateral Distance (y/w) 0 0.2 0.4 0.6 0.8 1 Dimensionless Lateral Distance (y/w) c) Exp. 18 J g=0.35 d) Exp. 19 J s =0.50 0 0.2 0.4 0.6 0.8 1 Dimensionless Lateral Distance (y/w) 0 0.2 0.4 0.6 0.8 1 Dimensionless Lateral Distance (y/w) Figure 4-19 Lateral Dilution Contour Plots for Experiments 16 -19 163 Js = 0.20 Js = 0.48 Js = 0.60 Figure 4-20 Cross-section Width Images at x = 32cm downstream (Cont'd) 165 a) 0 0 o 0 0 & 0 . 7 | o .2,0.6 h So * o ° 0 10 15 20 Downstream Distance (x/d) b) 25 30 4 -2 -0 * * J =0.45 F =36.2 s 0 0 0 J =0.46 F =26.9 s 0 10 15 20 25 30 Downstream Distance (x/d) 35 40 45 Figure 4-21 Jet Trajectory (Zc/hi) (a) and Dilution (S) (b) at varying Froude numbers and similar Jet Strength ratios 166 a) + + + o •§ 0.6t a> 'ST a> 0.4 0.2 1 0 0 0 + + + + + 0 + + + ^ + + + + -*- + + _L + 0 + + + * * * * + * 10 15 20 Downstream Distance (x/d) 25 30 15, b) 0 0 ~ 1 0 co c o 3 0 0 $ + o 0 + 0 + * * J =0.29 F =23.2 s o + + J =0.34 F =21.1 s o 0 0 J =0.58 F =20.0 s o 10 15 20 25 30 Downstream Distance (x/d) 35 40 45 Figure 4-22 Jet Trajectory (Zc/hi) (a) and Dilution (S) (b) at varying Jet Strength ratios and similar Froude numbers 167 10 15 20 Dimensionless Downstream Distance X/d Figure 4-23 Buoyancy Effects on Jet Trajectory (Zc/hi) (Exp. 20, 21 and 22) 168 ir . * 0 * ft 0 * ir 0 * * Bottom J =0.08 s 0 0 Intermediate J =0.35 s ir ir Surface J =0.58 s 0 ir 0 ir 10 15 20 25 30 Distance Downstream x (cm) 35 40 45 Figure 4-24 A comparison of Jet Dilution (1/S) for Bottom, Intermediate and Surface Jets 169 a) b) 20.00 a •a 3 Q Js=0.17 0.00 20.00 40.00 60.00 Distance Downstream (cm) -*—Actual -x --Cormix -Equation 4.9 J,=0.39 0.00 20.00 40.00 60.00 Distance Downstream (cm) -Actual - x--- Cormix Equation 4.9 c) 20.00 J,=0.53 1 0.00 20.00 40.00 60.00 Distance Downstream (cm) -Actual Cormix Equation4.9 Figure 4-25 Comparison of Dilution Data to CORMIX1 output and Equation 4.9 for a) Bottom, b) Intermediate and c) Surface Jets 170 C/3 0.8 0.6 0.4 -0.2 -0 o u u W o -0.2 > 03 "33 -0.4 -0.6 -0.8 -1 6 8 x + o + 10 20 30 40 Distance Downstream (cm) 50 o0.19 nO.29 oQ.35 A 0.39 x 0.45 + 0.53 Figure 4-26 Relative Error (AS/S) in dilution predicted by CORMIX1 for various jet strength ratios at 5,27,32 and 42cm downstream 171 a) Exp. 1 J=0.16 14.00 j 12.00 -10.00 -a o 8.00 -Dilut Dilut 6.00 --4.00 -2.00 -0.00 7 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 Distance Downstream (cm) -*— Actual — -x- - - • Cormix •Equation 4.9 b) Exp. 5 J =0.08 30.00 j 25.00 --a 20.00 --o 15.00 -p 10.00 --5.00 -0.00 ¥ 0.00 10.00 20.00 30.00 Distance Downstream (cm) 40.00 50.00 - Actual - - - -x- - - Cormix -Equation 4.9 Figure 4-27 Comparison of dilution data to CORMIX1 output and Equation 4.9 for Experiment 1 (Js=0.16) and Experiment 5 (Js=0.08) 172 DHA Adsorption 10000 5000 2000 1000 500 Sediment Concentration (mg/l) 200 133 Figure 4-28 Results of Batch Adsorption Studies for Dehydroabietic Acid 173 3,4,5 T C G Adsorption 10000 5000 2000 1000 500 200 Sediment Concentration (mg/l) 100 25 Figure 4-29 Results of Batch Adsorption Studies for 3,4,5 - Trichloroguaiacol (uncorrected for interference) 174 Interference Release from Sediment 12000 Sediment Concentration (mg/l) Figure 4-30 Interference Contribution for 3,4,5 - Trichloroguaicol Adsorption Corrected 3,4,5 T C G Adsorption 10000 5000 2000 1000 500 200 Sediment Concentration (mg/1) 100 25 Figure 4-31 Corrected Adsorption for 3,4,5 - Trichloroguaiacol 176 5 Flocculation of Biosolids/Sediment in the Near Field Region There is a potential for contamination of the receiving environment due to the discharge of biosolids with the effluent. The fate of these biosolids within the mixing zone and downstream dictate the effects this source of contamination will have on the system. Of importance, is the potential for flocculation of biosolids in the river. Increased flocculation of suspended sediment has been seen due to the presence of pulpmill effluent, especially unfiltered effluent containing biological solids (Evans, 1996; Krishnappan et al., 1994a; Krishnappan et al., 1994b; Krishnappan, 1994; Krishnappan, 1996). Evans (1996) found that this flocculation phenomena occurred rapidly and as such the conditions in the near field are most likely to dictate where, how and when this increased flocculation will occur. This chapter looks at the mixing conditions (dilution, trajectory, etc.) that exist within the near field region and identifies the major factors influencing this potential flocculation and the relevance to the pulpmill discharges in the Northern Fraser River. 5.1 Factors Affecting Flocculation in Near Field Region As discussed in Chapter 2, the rate of formation of lasting contacts, N(d},d2), is proportional to the collision rate and is given by equation 2.68. Rearranging this equation becomes N(di,d2) = a- N(d],d2)=a-Kf(dl,d2)-n(dl)-n(d2) 5-1 where a is the collision efficiency factor, Kf is the rate constant and n(di) and n(d2) are the number concentrations for sediment and biosolids respectively. The collision efficiency is a function of the stability of the individual particles which is assumed to be due to double layer repulsion, VV, defined by equation 2.69. The dominant flocculation mechanism for sediment and biosolids within the mixing zone is orthokinetic flocculation and the collision rate constant, Kf, is given by equation 2.64. Combining equation 5.1 and 2.64, the rate formation of lasting contacts between biosolids and suspended sediment can be estimated by: N(dl,d2) = cc-(d] + d l ) •G-n(dl)-n{d2) 5'2 6.18 where G is the rms velocity gradient and di (d2) is the particle diameter. Assuming that d, and d 2 are constant, the rate of formation of lasting or successful floe formation is a function of the velocity gradient (G), the number concentration of each particle, and the efficiency of collision. The latter being dependent on the particle stability, or double layer repulsive forces. N(dl,d2) = f(yR,G,n(dl),n(d2)) 5 3 An approximation of the probability of successful collisions and flocculation in the near field mixing zone can be calculated using equation 5.3 based on the following assumptions. 1. The double layer repulsive force is the primary source of particle stability for both biosolids and sediment and is given by equation 2.69, increasing linearly with temperature. 2. Turbulence intensity is proportional to the rms velocity gradient (G) and therefore can be used as a measure of shear in the near field region (Cleasby, 1994). 178 3. The number concentration of both particles and biosolids can be determined from the concentration of each compound within the mixing zone. In summary, the key factors influencing floe formation in the near field mixing zone are temperature, turbulence intensity and concentration of biosolids and sediment. The probability of increased flocculation can therefore be estimated based on the experimental results from the near field mixing experiments discussed in Chapter 4. This was done for discharge scenarios given in Table 5.1, indicative of peak, mid and low flow conditions in the Fraser River, with Jet Strength ratios (Js) of 0.08, 0.38 and 0.58 respectively. Table 5-1 Typical River and Discharge Conditions Fraser River Northwood Q U a [Sediment] River Depth Q u ( Js ( d ) (m7s) (m/s) (mg/1) (m) (nrVs) (m/s) Low Flow 190 1.5 10 1.0 1.8 0.6 0.58 Mid Flow 653 1.4 100 1.9 1.6 1.4 0.38 Peak Flow 2200 1.2 1000 3.9 1.4 1.9 0.08 Notes: a) River discharge rate is from Environment Canada's Water Survey station at Shelly. b) Ua is calculated based on a rectangular channel with an average width of 300m. c) The river depth was determined using Manning's formula (Vine, 1996). d) Uj was calculated by dividing the flow rate by the number and area of the ports where the port diameter is 0.5 m. e) Js is the jet strength of the experimental conditions that most closely approximate the peak, mid and low flow conditions. f) The suspended sediment concentration is calculated from the mean monthly suspended sediment loadings (Zyrmiak and Tassone, 1986) at Hansard. 5.1.1 Temperature Floe formation depends on particle interaction and repulsive forces. Van Der Waal's attractive forces do not vary with temperature but repulsive forces do, as discussed in Section 2.3. The weak floes formed in the lagoon at higher temperatures may break up, then as the temperature drops within the mixing zone, newer, more stable floes may form either with other biosolids or sediment. As discussed in Section 4.1, dilution and therefore temperature within the mixing zone is a function of jet strength. Higher strength jets have lower overall levels of dilution and therefore elevated temperatures within the mixing zone. This corresponds to the winter low flow conditions. 5.1.2 Turbulence Intensity In order for particles to flocculate, a sufficient opportunity of collision must exist. The rate of particle collision is a function of the velocity gradient (G). Turbulence intensity is proportional to velocity gradient and will be used as a measure of the shear in the mixing zone for these calculations. Initially as the turbulence increases, so does the probability of collision between the particles and therefore the probability of floe formation increases. The resultant floe diameter is also a function of shear stress or turbulence intensity (Schroeder, 1977; Reynolds, 1982; Winterwerp, 1998). Increases in shear stress initially increase the median floe diameter as larger floes form. At higher levels of shear the strength of the floe particle is exceeded and floe break up occurs, resulting in a reduction in floe formation and diameter. Turbulence intensity is the standard deviation of the temperature fluctuations, normalized by the maximum average temperature at each downstream location. The depth averaged 180 turbulence intensities were calculated for the three flow conditions and are shown in Figure 5.1. Downstream of the jet exit, the turbulence intensity increased rapidly to its maximum level on order of 0.10 then decreases as the jet enters the Zone of Established flow. This behaviour is similar to turbulence intensity levels normally calculated from velocity data (Fischer et al., 1979). The maximum turbulence intensity levels were higher and occurred earlier for jets with higher jet strength ratios, resulting from the high levels of mixing and entrainment prior to surfacing and the complex turbulent flow that occurs when these jets impinge the free surface. Farther downstream, as these jets bifurcated, the turbulence intensity increased slightly due to the complex flow in the counter-rotating concentration nodes. The weaker discharge (Js = 0.08) has a larger region of elevated levels of turbulence intensity. This is most likely due to the interaction of the jet with the bottom boundary layer, increasing the overall level of turbulence in the jet. Regardless of the jet strength, the presence of a discharge resulted in a region of increased turbulence downstream of the jet exit. 5.1.3 Concentration The probability of collision and subsequent successful flocculation is a function of the number concentration of both sediment and biosolid particles. The resultant floe diameter also increases with solid concentration (Winterwerp, 1998). Within the near field region, both concentrations are changing as the plume travels downstream: the number of biosolids is decreasing while suspended sediment concentration is increasing. 181 Assuming that one biosolid particle floes with one sediment particle, optimal flocculation conditions exist when the number concentration of both biosolid and suspended sediment particles are equal. If more than one biosolid floculates with each sediment particle, optimal flocculation will occur earlier or later if each biosolid floes with several sediment particles. Based on the method used by Evans (1996) the number of sediment and biosolid particles can be calculated based on the river and discharge conditions given in Table 5.1 and the assuming the following sediment and biosolid characteristics: • sediment particle density (psed) = 2.6 g/ml (Evans, 1996), • biosolid particle density (pbio) =1.1 g/ml (Evans, 1996), • effluent (biosolids)suspended solids concentration ([biosolid]) = 80 mg/1 (Evans, 1996), • suspended sediment concentration ([sediment]) from Table 5.1, • median particle diameter of effluent (c4,-0) = 4.1 p:m (Evans, 1996) • median particle diameter of sediment (dsed) = 11.8 urn (Krishnappan, 1994) • dilution (5) is from near field mixing experiments. The number of suspended sediment and effluent particles were then calculated based on the following method outlined in Evans (1996): • the mass of each particle is calculated from density and median diameter assuming that the particles are spherical, 182 • the initial number of particles per litre of sample is calculated by dividing the concentration by the individual particle mass, • the ratio of biosolid particles to suspended sediment particles {NbiJNsed) is then calculated based on dilution data from the near field mixing experiments. When the ratio is greater than one, the plume consists of mostly effluent solids and primarily suspended sediment when the ratio is less that one. Assuming a 1:1 biosolid/sediment ratio, optimal flocculation will occur at the location downstream when there is the same amount of both effluent and sediment particles. 5.2 Calculation of Flocculation Potential in the Near Field Mixing Zone The probability of successful flocculation in the near field mixing zone is directly proportional to the turbulence intensity and the number concentration ratio and indirectly proportional to the temperature. The total effect of these three variables on potential flocculation can be examined by taking the product of their relative positive contributions. That is, each of the variables were normalized in such a way that they have a value of 1.0 when they contribute the most to successful flocculation. Temperature Contribution Floe stability decreases and consequently potential flocculation increases as the temperature is reduced in the mixing zone. Accordingly, temperature will have the greatest possible contribution to successful flocculation when it is a minimum value. The Temperature Contribution (Tc) can be calculated by: 183 using depth averaged centreline temperatures in degrees Kelvin. Particle Number Contribution The Number Concentration ratio was calculated using the method described in Section 5.1 using depth averaged dilution data. Assuming a one to one particle flocculation, the particle number concentration ratio has the most positive effect when it has a value of unity. Prior to this point, the number of floes formed is initially limited by the number of sediment particle and afterward by the number of biosolid particles. Based on this, the Number Contribution ratio (NQ was normalized and the Number Concentration Contibution (NCC) can be calculated as follows: NC = ^-)l, NC = —, 5.5 Nj NC NC = - ^ 2 - < 1, NCr = NC. 5.6 i y sed Turbulence Intensity Contribution The contribution of turbulence intensity (77) is the greatest when it is a maximum and accordingly its contribution (77c) was normalized by dividing by the maximum value at each downstream location. 184 These three contribution factors were combined to give the Flocculation Potential Product (FPP): FPP = NCCTCTIC 5 8 The Flocculation Potential Product was calculated for the three discharge scenarios in Table 5.1 is shown in Figure 5.2. The location of the maximum Flocculation Potential Product (FPP) corresponds to that of the maximum Number Concentration Contribution (NCC) or where the number of sediment and biosolids particles is the same (Nsed=Nt,i0). This was true for all three discharge scenarios. This location is also a function of the Jet Strength (Js), occurring closer to the jet exit for weaker jets. Optimal flocculation conditions were not achieved within the experimental region for the two stronger jets. The effect of temperature was found to be negligible compared to turbulence intensity and particle concentration for the 1:1 flocculation of sediment and biosolids. This doesn't rule out the importance of temperature reduction on the flocculation biosolids alone immediately downstream of the jet exit. Immediately downstream of the jet exit (X/IQ < 5), there is a region of high turbulence intensity which had little impact on the FPP, especially the low and mid flow cases. Although the physical turbulent conditions exist for particle collisions, the concentrations of sediment in this region are very low relative to that of the biosolids. This limits the actual number of successful sediment-biosolid collisions. 185 A simple sensitivity analysis was carried out to determine the relative importance of the turbulence intensity and particle concentration to the Flocculation Potential Product. This was done by increasing the suspended sediment concentration, essentially changing the location of equal particle concentration. New Flocculation Potential Products are shown in Figure 5.3 for a suspended sediment concentration of 500 mg/l. The onset of optimal flocculation conditions is shifted closer to the jet exit for both cases, associated with a similar shift in the number concentration ratio (NC). From this it is evident that the NCc, and therefore the particle ratio, is the dominant factor in determining the location of maximum flocculation potential within the near field mixing zone. The same effect is seen in Figure 5.4. In this case, the biosolid particle diameter was increased to 11 pm, again shifting the Number Concentration ratio. In general, modifying any variables used to determine the number of sediment and biosolid particles shifts the downstream location of optimal flocculation conditions. These variables include physical characteristics of solids, such as density, shape, diameter, and the concentration of each particle type. The latter of which is a function of the dilution achieved within the mixing zone and the initial concentrations of biosolids and suspended sediment. Another, very important factor is the assumption of 1:1 particle flocculation that was made for these calculations, which may or may not be the case. Optimal flocculation conditions will occur earlier if more than one biosolid particle floes with each sediment particle and later if the converse is true. This is illustrated in Figure 5.5, which compares the Number Concentration ratio for 1:1 particle flocculation and 10:1 biosolid/sediment particle flocculation. As well, this location will shift upstream if biosolids are coalescing with other biosolids or sediment with sediment. The Flocculation Potential Product curve can be divided into three regions similar to those initially defined by Holman (1986) and shown in Figure 5.6. Initially there is a region of rapid mixing and high turbulence intensity levels. Due to the low relative concentration of sediment particles, flocculation of suspended sediment and biosolids is limited. Farther downstream, after sufficient time for mixing, the probability of successful flocculation reaches a maximum as the number of sediment particles equals the number of biosolid particles. Downstream of the maximum flocculation potential, the effluent is further diluted and the number of biosolids available for flocculation is decreasing. Accordingly, the probability of flocculation begins to drop. This rapid mix/slow mix is similar to that used in the design of flocculation facitities in water and waste water treatment plants. The rapid mixing provides for intense mixing to ensure complete mixing of reactants and allow adequate contact between particles (Reynolds, 1982) to form microflocs. In the slow mix region, the flocculation of the microflocs into larger particles is dependent on time and gentle agitation. Too much shear at this point can break the newly formed floes apart. In summary, the dominant factor in determining the potential for flocculation in the near field mixing zone is the particle number concentration of both biosolids and suspended sediment, or the Number Concentration Ratio. The major variables that determine this ratio are the physical characteristics of the individual sediment and biosolid particles, solid concentrations prior to mixing and the dilution levels achieved downstream of the jet exit. It should be 187 noted that these calculations were based on the assumption of 1:1 flocculation. Although the actual mechanism of flocculation of biosolids and suspended sediment has yet to be determined, the role of the particle number concentration will be similar. That is, dictating the location where optimal flocculation conditions exist based on the relative number of each particle flocculating together. 5.3 Application to Pulpmill Discharges in Fraser River Applying this to discharges in the Fraser River, there will be a region downstream of the discharge where conditions will be optimal for the flocculation of sediment and biological solids. The time of year (low, mid or peak flow) or jet strength (Js) will affect the location of this optimal flocculation region. During the peak flow discharges (Bottom Jets) in the spring freshet, the location of optimal flocculation conditions occurs immediately downstream, primarily due to the high suspended sediment load. As the flow rates and suspended sediment loads decrease, the location for optimal potential flocculation moves farther downstream. The extent of the slow mix region for the low flow case (Surface Jets) is larger and extends farther downstream due to the slow down in dilution at surfacing and the large region of increased turbulence farther downstream (Section 5.1.2). During this period the temperature effects are also significant, as the river temperature is significantly lower. This may be negated by the fact that the suspended sediment concentrations are much lower, approximately 15 mg/l compared to 150 mg/l during spring freshet (Carey, 1988). The three regions described in the previous sections have also been seen in field measurements of median particle size diameter taken by Krishnappan (1996) downstream of 188 the Northwood discharge in October 1996. These samples were all taken on the same day to avoid any fluctuations in the median diameter of the background suspended sediment due to natural flow variations. The increase in median diameter downstream indicates potential increased levels of flocculation due to the presence of the pulpmill discharge. As a comparison, the dilution data for the mid flow case (7t = 0.38) was extrapolated using equation 4.9 and the Number Concentration Contribution curve is also shown in Figure 5.7. Both curves have a similar shape, reaching a maximum at a dimensionless downstream distance of approximately 225 to 275, then decreasing due to dilution in the mixing zone. A comparison of the absolute number of floes formed within the mixing zone is shown in Figure 5.8. More floes, and therefore contaminated solids, are formed in the peak flow period. In this case, the location of equal number of sediment and biosolids particles occurs earlier in the mixing zone when the concentration of biosolids has not been sufficiently diluted. The number of floes formed decreases with the falling hydrograph and suspended sediment loads. One of the major concerns of this increased flocculation is the effect of increased levels of contaminated solids on the receiving environment, specifically on overwintering salmon. The impacts of this increased flocculation and subsequent transport and deposition of these contaminated solids varies depending on the seasonal flowrates. Although, more floes are formed in peak flow conditions, most of them would not remain in the system but are flushed downstream to the estuary. The most important time of year would be during the falling hydrograph in the late summer and early fall. At these flow conditions, the levels of newly formed floes of suspended sediment and biosolids are still elevated, but the flow rate is decreasing. The drop in ambient current at this time of year allows for the deposition of these 189 contaminated solids in potentially sensitive aquatic habitat where they will remain during the winter. With the onset of spring freshet, many of these deposited solids will be resuspended and deposited farther downstream, eventually making it to the estuary. 190 Turbulence Intensity 0.12 0.00 10 15 20 25 30 Dimensionless Downstream Distance (x/lQ) •D--Js=0.08 • Js=0.37 A Js=0.58 35 40 Figure 5.1 Depth Averaged Turbulence Intensity at J s = 0.08,0.35 and 0.58 191 a) Peak Flow Conditions J =0.08 1.00 10 15 20 25 30 35 — h — Tic D -NCc - X - Tc FPP 40 1.00 -preBJ»*--.>g- -x- x- x x 0.80 0.60 0.40 0.20 0.00 b) Mid Flow Conditions Js=0.37 — X X X X -••+--+... •+. t I 1+ • e - f i -B - * ' 43 - B - -•- + --TIC - - 0 - - N C C - -x- • Tc FPP 0.00 1 1 1 1 1 1 1 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 1.00 0.80 H 0.60 0.40 0.20 H 0.00 0 c) Low Flow Conditions Js=0.58 x- x x x-n n n n n , . n +• - —+-10 20 30 Dimensionless Downstream Distance (x/lQ) — h TIc - ---a --- NCc - X - Tc FPP 40 Figure 5.2 Flocculation Potential Product at a) Peak, b) Mid and c)Low Flow Conditions 192 a) Mid Flow Conditions Js=0.37 • - Q - . . . - - -+ - - - G 0.00 5.00 10.00 15.00 20.00 25.00 30.00 Dimensionless Distance Downstream (x/lQ) - + --TIc - - o - - N C c Tc FPP 35.00 40.00 b) Low Flow Conditions Jg=0.58 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 + • /+---+-. -3K-10 15 20 25 30 Dimensionless Downstream Distance (x/lQ) + Tic ° NCc - x Tc FPP 35 40 Figure 5.3 Flocculation Potential Product at a) Mid and b) Flow Conditions with [sed]=500 mg/l 193 Low Flow Conditions Js=0.58 0.00 -I 1 1 1 1 1 1 , 1 0 5 10 15 20 25 30 35 40 Dimensionless Downstream Distance (x/lQ) + Tic NCc - - Tc FPP Figure 5.4 Flocculation Potential Product at Low Flow Conditions with d W o = 11 pm 194 Mid Flow Conditions J =0.37 o c o c <L> O o U t-<u s 0.00 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 Dimensionless Distance Downstream (x/lQ) • o - -Neff/Nsed=10 —•—Neff/Nsed=l Figure 5.5 Comparison of Number Concentration Contribution (NCC) for Nbio/Nsed=l and Nbio/Nsed=10 195 Ambient Current 1 Reduced Dilution Low Dilution and / \ and Increased ' Particle Flocculation / \ Flocculation 1 Region /Rapid Dilution \ 1 / and low \ / Flocculatioja---" 1 t Pulpmill Effluent Figure 5.6 Flocculation Regions in a Jet (adapted from Holman, 1986) / 196 Fraser River Mid Flow Flocculation Potential 1.2 0 0.00 100.00 200.00 300.00 400.00 Dimensionless Downstream Distance (x/lQ) —®— Number Concentration Ratio —•— Delta Median Diameter Figure 5.7 Comparison of Number Concentration Contribution (NCC) with Field Data (Ad50) (October, 1996) 197 1.00E+09 g l.OOE+08 o \-< S ;§ L00E+07 1.00E+06 . A A A A A A A ^ A A 10 15 20 25 30 35 Dimensionless Downstream Distance (x/lQ) 40 - o — - Peak Flow Mid Flow A Low Flow Figure 5.8 Number of BiosolidrSediment Floes formed at Low, Mid and Low Flow Conditions 198 6 Conclusions and Recommendations The following is a summary of the results of the studies discussed in Chapters 4 and 5, Near Field Mixing Experiments and the Association of Contaminants with River Sediment, and highlight areas where significant contributions have been made. This is followed by some recommendations for research that will help provide for further insight into the areas studied within this dissertation. 6.1 Near Field Mixing Experiments The near field mixing of a vertical buoyant jet in a shallow crossflow is controlled by the ambient and discharge conditions, of which the most important is the depth and the velocity ratio, defined collectively as the jet strength ratio (Js). Within the momentum dominated far field (MDFF), the behavior of a discharge could be classified into three jet types based on the jet strength parameter. The influence of the shallow free surface is greater for jets with high values of this jet strength parameter, while the proximity of the bottom influences the jets with lower jet strength ratios. Prior to any influences due to the free surface, the dilution in the M D F F can be quantified by the following power law relationship similar to that previously defined by Wright (1977) for unconfined flows: 1 rUf D ~ d R \Zm ) where the coefficient Q was found to be a constant, approximately 0.90. 199 After the effects of the shallow crossflow begin to dominate the mixing, the dilution, in the entire M D F F (1 < x / <10) region could be approximated by: In general, due to the presence of the free surface, the dilution was found to be a function of the jet strength ratio, Js. The weaker the discharge, the better the dilution. This is contrary to the behaviour of a jet in an unconfined current, with stronger jets having higher rates of dilution. The presence of a shallow ambient current was found to have significant effects on both the jet trajectory and dilution of a vertical buoyant jet. The extent of the impacts was found to be primarily a function of the jet strength ratio (Js). Based on the jet strength ratio, three jet classifications were identified for this type of discharge in the momentum dominated far field region (MDFF). 1. Bottom Jets are defined as weak strength jets (Js < 0.30) whose trajectory and dilution are influenced by the proximity of the flow to the shear flow in the turbulent boundary layer. The dilution of the Bottom Jets was significantly increased by enhanced mixing due to interactions of the jet with the boundary layer flow. The trajectory of these jets was inhibited by the shallow flow and some of the weaker flows exhibited attachment to the flume bottom. These jets never surfaced within the experimental region. m 4.9 200 2. Intermediate jets are defined as jets with Js between 0.30 and 0.44. The behaviour of these jets is more complicated due to the influence of both the top and bottom confining boundaries. Although the growth of the trajectory and the overall jet dilution was inhibited as the jet approached the free surface, dilution was sometimes enhanced due to interactions with the boundary layer. 3. Surface Jets are defined as jets that impinged on the free surface immediately downstream of the jet exit (Js > 0.44). Prior to surfacing, the dilution and trajectory were relatively unaffected by the shallow flow. After surfacing, the jet dilution was inhibited due to the lack of vertical entrainment, and the overall M D F F dilution was reduced. These jets also tended to bifurcate after surfacing. In general, due to the presence of the free surface, the dilution was found to be a function of the jet strength ratio, Js. The weaker the discharge, the better the dilution. This is contrary to the behaviour of a jet in an unconfined current, with stronger jets having higher rates of dilution. The CORMDCl model was shown to be unsuitable for the predicting of dilution of vertical buoyant jets in shallow crossflows. This was most pronounced for Bottom and Surfacing Jets. These discrepancies were primarily due to the lack of consideration of the turbulent boundary layer enhancing mixing of the weaker Bottom Jets and the prediction of increased dilution at surfacing. Incorporating the results of this study, including the dependency of dilution on jet strength (equation 4.9), will provide for better predictions of the dilution and therefore potential impacts of discharges in shallow ambient currents such as rivers. 201 The implications of these findings on the near field mixing of pulpmill discharges is the overall decrease in dilution due the presence of a shallow receiving water typical of the critical winter low flow period, and the increase in the level of dilution in the peak flow due to interactions with the bottom boundary layer. The application of the 2/3 power law derived in this dissertation provide for a better approximation of the dilution in a shallow crossflow by taking into account the effects of both confining boundaries on the flow. The use of relationships and models developed previously primarily on data from deeper crossflows, will result in errors in the estimation of downstream dilution. This could have severe ramifications if these estimates are used for discharge management decisions 6.2 Interactions with Sediment in the Near Field Region The role of direct sorption of two pulpmill constituents, dehydroabietic acid and 3,4,5 trichloroguaiacol, was found to be relatively unimportant within the rapidly changing near field mixing zone. Of greater significance is the association of contaminated biosolids with the suspended sediment during the mixing process. The presence of pulpmill effluent has been found to cause increased flocculation of sediment, especially unfiltered effluent containing biosolids (Krishnappen et al., 1994; Evans, 1996). Analysis of the mixing zone dilution, temperature and turbulence intensity data indicated that the ideal conditions for enhanced flocculation exist within the experimental near field region. The most significant contributor to increased flocculation in this region is the Number Concentration Ratio. Assuming a 1:1 biosolid to sediment flocculation, the optimal conditions were found to occur when the number of biosolids particles equals the 202 number of suspended sediment particles. This condition occurred earlier in the peak flow conditions during spring freshet, when ambient suspended sediment loads are elevated. From an impact point of view, the most important flow conditions exist during the rising hydrograph, when there are still significant sediment loadings to the system but the flow rate is decreasing. As the flow is reduced, the potential for flocculation enhancement still exists and due to the lower river velocity the larger floes formed between the sediment and contaminated biosolids may be deposited. In spring freshet, as the flow begins to increase, many of these contaminated solids are then resuspended and carried farther downstream, away from the original source. Therefore, contaminated sediment is distributed throughout the river system, potentially impacting sensitive aquatic habitat. Eventually, all of the contaminated solids will be flushed down to the estuary where they are deposited and accumulated. In summary, the work in this dissertation provided some insight into the factors affecting mixing of buoyant discharges in conditions similar to that of the Northern Fraser River. Of significance is the determination of the importance of both the free surface and the turbulent boundary layer in the dilution of jets in a shallow crossflow. This information provides a basis for the incorporation of shallow water effects into the CORMIX model. The mixing parameters and conditions were then applied to sorption and flocculation theory, to determine under what discharge conditions the association of contaminants with river sediment is probable and under what mechanism this occurs. It was determined that, the discharge of biosolids, and their subsequent flocculation with the 203 river sediment, is the probable mechanism for this sediment contamination within the near field mixing zone. 6.3 Recommendations for Further Research Further research in the areas covered by this dissertation in recommended to provide for a better understanding of the fate of hydrophobic contaminants in shallow water discharges. 1. The examination of the mixing of solids in a buoyant jet in a shallow crossflow to provide information on the interaction between the solids and the ambient current, specifically conditions that lead to flocculation. 2. Field sampling program including flow visualization and non-intrusive sampling to determine the composition and structure of the floes within the mixing zone. 3. Laboratory experiments to better quantify the dilution of the Surfacing Jets, specifically the effects of bifurcation. 4. Inclusion of these results along with those of additional analysis building on the experiments carried out as part of this dissertation on various configurations of shallow water jets into the CORMIX 1 model. 5. Physical modeling of jets of various configurations in crossflows (including diffusers), specifically looking at the role of the turbulent boundary layer including Coanda Effect and Wake Attachment. 204 Carry out batch adsorption studies in well buffered solutions incorporating the factors discussed in Section 4.2.5 including: • pH and ionic strength measurements before and after adsorption tests, • use of biosolids as well as sediment for the sorbent, • experiments done over a range of pH values covering those typically seen in the Fraser River, • experiments carried out over several equilibration times ranging from 1 to 7 days. 205 References Ages, A . (1985). Hydrodynamics and Sediment Transport of the Lower Fraser. In the Workshop Proceedings of Toxic Chemical Research Needs in the Lower Fraser River (pp 55-61). Vancouver, British Columbia. Abdelwahed, M.S.T. and Chu, V . H . (1978). Bifurcation of Buoyant Jets in Cross Flow. Technical Report No. 78-1. Department of Civil Engineering and Applied Mechanics. McGi l l University. Montreal, Quebec. Amy, G.L., Bryant, C.W., Alleman, B.C., and Barkley, W.A. (1988). Biosorption of organic halide in a kraft mill generated lagoon. Journal Water Pollution Control Federation, 60 (8), 1445-1453. Anderson, M.R. and Pankow, J.F. (1985). Determination of the Sorption Behavior of Polvchlorinated Biphenvls (PCB's) on Soil Core Samples from the Vicinity of the Federal Pioneer Ltd. Transformer Manufacturing Plant. Regina, Saskatchewan, Canada. Groundwater Research Laboratory, Department of Chemical, Biological, and Environmental Sciences, Oregaon Graduate Center. Benefield, L.D. , Judkins J.F., and Weand B.L. (1982). Process Chemistry for Water and Waste Water Treatment, Prentice.-Hall Inc. (182 pp). Birtwell, I.K., Levings C D . , MacDonald J.S., and Rogers L H . (1988). A review of fish habitat issues in the Fraser River system. Water Pollution Research Journal of Canada. 23 (1), 1-30. Brady, N.C. (1974). The Nature and Properties of Soils 8 t h Edition. Macmillan Publishing Company, New York. Branion, R.M.R. (1991). Environmental Protection and Pollution Control by the Pulp and Paper Industry, unpublished (31 pp). Breen, D. (1995). Personal communication on HPLC determination of Ko W . Bruno, M.S., Muntisov, M . , and Fischer, H.B. (1990). Effect of buoyancy on transverse mixing in streams. A S C E Journal of Hydraulic Engineering, 116(12). 1484-1494. Carey, J. (1985). Chemical Pathways. In Workshop Proceedings of Toxic Chemical Research Needs in the Lower Fraser (pp 62-70). Vancouver, British Columbia. Carey, J.H. (1988). Pathways of chlorophenols in the Fraser River Estuary, British Columbia. In Proceedings of the Canada - British Columbia Workshop on Water Quality Guidelines and Objectives: Focus on the Fraser (pp 29-34). Environment Canada, Inland Waters, Pacific and Yukon Region, Vancouver B.C. 206 Carey, J.H., Hodson, P.V., Munkittrick, K.R., and Servos, M.R. (1993). Recent Canadian Studies on the Physiological Effects of Pulp M i l l Effluent on Fish. Environment Canada and Department of Fisheries and Oceans (20 pp). Chiou, C.C.T., Freed, V . H . , Schmedding, D.W., and Kohnert, R.L. (1977). Partition coefficient and bioaccumulation of selected organic chemicals. Environmental Science and Technology, 11 (5), 475-478. Chiou, C C T . (1985). Partition coefficients of organic compounds in lipid-water systems and correlations with fish bioconcentration factors. Environmental Science and Technology, 19 (1), 57-62. Christensen, B.E., and Characklis, W.G. (1990). Physical and chemical properties of biofilms (pp 93-130). In Characklis, W.D., and Marshall, K . C (Ed.) Biofilms. John Wiley and Sons Inc. Chu, V . H . , and Goldberg, M.B . (1974). Buoyant forced-plumes in cross flow. A S C E Journal of Hydraulic Hydraulics Division. 100 (9). 1203-1213. Cleasby, J.L. (1984). Is velocity gradient a valid turbulent flocculation parameter? Journal of Environmental Engineering, 110 (5), 875-897. Costa, H.S., McKeown, J.J., and Blosser, R.O. (1979). Studies on characterization, fate, and impact of residual solids of biological treatment origin. TAPPI, 62 (10), 41-46. Costa, H.S., McKeown, J.J., and Borton, D.L. (1980). Fate of suspended biological solids discharge to receiving waters. Journal Water Pollution Control Federation, 52 (10), 2432-2441. De Kock, A . C , and Lord, D.A. (1987). A simple procedure for determining octanol-water partition coefficients using reverse liquid phase high performance liquid chromatography (RPHPLC). Chemosphere. 16 (3). 133-142. Dobbs, R.A., Wang, L., and Govind, R. (1989). Sorption of toxic compounds on wastewater solids: correlation with fundamental properties. Environmental Science and Technology 23 (9), 1092-1097. Doneker, R.L. and Jirka, G.H. (1991). Expert systems for mixing-zone analysis and design of pollutant discharges. A S C E Journal of Water Resources Planning and Management. 117 (6), 679-697. 207 Dorcey, A.H.J. (1991). Water in the sustainable development of the Fraser River Basin. In Dorcey A.H.J, and Griggs J.R. (Ed.) Water in Sustainable Development: Exploring Our Common Future in the Fraser River Basin (pp 3-18). Westwater Research Centre, University of British Columbia, Vancouver, British Columbia. Droppo, I.G., and Ongley, E.D. (1990). Flocculation of Suspended Fine-Grained Sediment in Rivers. NWRI Contribution 90-78. National Water Research Institute, Burlington, Ontario. Dugan, R. (1994). Axisvmetric Buoyant Jets in a Cross Flow with Shear: Transition and Mixing (PhD thesis). California Institute of Technology, Pasadena, California (260 PP)-Dwernychuk, L.W.(1989). The Receiving Environment of the Upper Fraser River: A Pilot Environmental Monitoring Program Examining Physical/Chemical/Biological Elements of the System Related to Pulpmill Effluents. Prepared by Hatfield Consultants Ltd.,Vancouver, British Columbia. Evans, W.J. (1996). Flocculation of Fraser River Sediments Due to Pulp M i l l Effluents (MASC Thesis), University of British Columbia, Vancouver, British Columbia (141 PP) Fischer, H.B., List, E.J., Koh, R . C , Imberger, J., and Brooks, N.H. (1979). Mixing in Inland and Coastal Waters. London, Academic Press. Gowda, T.P.H. (1981). Critical concentrations of toxic ammonia and chlorine in mixing zones of rivers. Water Pollution Research Journal of Canada, 15 (3), 255- 270. Hall, E.R (1993). Chemical Engineering 579 Course Notes. University of British Columbia, Vancouver, British Columbia. Hall, E.R. and Liver, S.F. (1996). Interactions of resin acids with aerobic and anaerobic biomass: II Partitioning on biosolids. Water Research, 30 (3), 672-678. Hall, K.J , Schreier, H. , and Brown, S.J. (1991). Water Quality in the Fraser River Basin. In Dorcey A.H.J, and Griggs J.R. (Ed.) Water in Sustainable Development: Exploring Our Common Future in the Fraser River Basin (pp 41-76). Westwater Research Centre, University of British Columbia, Vancouver, British Columbia. Harris, H.S., Kaufman, W.J., and Brone, R.B. (1966). Orthokinetic flocculation in water purification. A S C E Journal of the Sanitary Engineering Division, 92 (6), 95-111. Henderson, F . M . (1966). Open Channel Flow. Macmillan Publishing Company (522 pp). 208 Hodgson, J.E. (1991). Turbulent Jet Discharges in Rivers (PhD thesis). University of Alberta (385 pp). Holman, H.N. (1986). Particle Coagulation in a Turbulent Plume (PhD thesis). University of California, Berkeley. Jafvert, C.T., Westall, J .C , Grieder, E., and Schwarzenbach R.P. (1990). Distribution of hydrophobic ionogenic organic compounds between octanol and water: organic acids. Environmental Science and Technology, 24 (12), 1795-1803. Jirka, G.H. and Doneker, R.L. (1991). Hydrodynamic classification of submerged single-port discharges. A S C E Journal of Hydraulic Engineering, 117 (9). 1095-1112. Jirka, G.H., Doneker, R.L. and Hinton, S.W. (1996). User's Manual for CORMTX: A Hydrodynamic Mixing Zone Model and Decision Support System for Pollutant Dischages into Surface Waters. Prepared for the U.S. Environmental Protection Agency by DeFree Hydraulics Laboratory, School of Civi l and Environmental Engineering , Cornell University, Ithaca, New York. Johnston, A.J . , Nguyen, N . , and Volker, R.E. (1993). Round buoyant jet entering shallow water in motion. Journal of Hydraulic Engineering, 119 (12), 1364-1382. Johnston, A.J . , Nguyen, N . , and Volker, R.E. (1994). Round buoyant jet in shallow water: bed shear velocity. Journal of Hydraulic Research, 32 (2), 233-250. Karickhoff, R.W. (1981). Semi-empirical estimation of sorption of hydrophobic pollutants on natural sediments and soil. Chemosphere, 10 (8), 832-846. Keffer, J.F., and Baines, W.D. (1963). The round jet in a cross-wind. Journal of Fluid Mechanics, 15, 481-496. Keffer, J.F., and Baines, W.D. (1967). Profiles of the round turbulent jet in a cross flow. A S C E Journal of the Hydraulics Division. 92 (6), 53-64. Koh, R.C.Y. (1973). Hydaulic Tests of Discharge Ports. W . M . Keck Laboratory of Hydraulics and Water Resources Technical Memorandum 73-4. California Institute of Technology, Pasadena, California. Krishnappen, B.G. (1994). Fraser River study Field Data (fall 1994). National Water Research Institute, Burlington (Unpublished). Krishnappen, B.G. (1996). Fraser River study Field Data (fall 1996). National Water Research Institute, Burlington (Unpublished). 209 Krishnappan, B.G. , and Engel, P. (1994a). Critical Shear Stresses for Erosion and Deposition of Fine Suspended Sediments in the Fraser River. National Water Research Institute, Burlington. Krishnappan, B.G. , Stephens, R., Kraft, J.A., and Moore, B.H. (1994b). Size Distribution of Suspended Particles in the Athabasca River near Hinton. National Water Research Institute, Burlington. Labridis, C. (1989). Buoyant Jets in Shallow Water with a Crossflow (MASc Thesis). University of British Columbia, Vancouver (58 pp). Lau, Y . L . (1994). Temperature effect on settling velocity and deposition of cohesive sediments. Journal of Hydraulic Research, 32 (1), 41-51. Lee, J.H.W. (1993). An experimental study of jet dilution in crossflows: Discussion. Canadian Journal of Civi l Engineering, 20, 1073-1076. Lee, J.H.W. and Rodi, W. (1993). Numerical simulation of line puffs. In Davies, P.A. and Neves, M.J .V. (Eds.) Recent Research Advances in the Fluid Mechanics of Turbulent Jets and Plumes (pp 73-88). Nato Advanced Science Institutes Series, Klumer Academic Publishers. Lehtinen, K.J . (1992). Environmental effects of chlorine bleaching - facts neglected? Paperi Ja Puu - Paper and Timber. 74 (9), 715-719. Letterman, R.D. (1981). Flocculation. In Coagulation and Filtration: Back to the Basics (pp 23-44). Proceedings of the A W W A Seminar, June 7. Lick, W. and Huang, H . (1993). Flocculation and the physical properties of floes. In Mehta, A.J . (Ed.) Nearshore and Estuarine Cohesive Sediment Transport (pp 21-39). Washington DC, American Geophysical Union. List, E.J. (1982). Mechanics of turbulent buoyant jets and plumes. In Rodi, W. (Ed.), Turbulent Buoyant Jets and Plumes (pp 1-68). Pergamon Press, Oxford (184 pp). Liver, S.F. (1990). The Fate and Effect of Resin Acids in Aerobic and Anaerobic Biological Batch Assays (M.Eng. thesis). McMaster University, Hamilton. Luettich Jr., R.A., Wells, J.T., and Kin. S. (1993). Insitu variability of large aggregates: preliminary results on the effects of shear. In Mehta, A.J . (Ed.) Nearshore and Estuarine Cohesive Sediment Transport. Washington DC, American Geophysical Union. MacLatchy, M.R. (1993). Radial Spreading of Vertical Buoyant Jets in Shallow Water (MASC Thesis). University of British Columbia, Vancouver, British Columbia, (78 PP)-210 Marks, B.J. (1996). Intital Dilution of a Horizontal Jet in a Strong Current (MASC thesis). University of British Columbia, Vancouver. Mckinney, R.E. (1962). Microbiology for Sanitary Engineers. McGraw-Hill Book Company Inc. (293 pp) McConnachie, G.L. (1991). Turbulence intensity of mixing in relation to flocculation. A S C E Journal of Environmental Engineering, 92 (6), 731-750. McCorquodale J.A., Zhou, S.P., Moawad, A. , Yuen, E -M. , and Rajaratnam, N . (1992). Near Field Modelling in the Connecting Channels of the Great Lakes. For Ontario Ministry of the Environment; Great Lakes Branch. Industrial Research Institute and Department of Civil and Environmental Engineering, University of Windsor (96 pp). McLeay, D.J. and Associates (1987). Acute Toxicity of Pulp and Paper M i l l Effluent: A Review. Report 4/PF/l , Environment Canada, Environmental Protection Service. Miller, M . M . , Wasik, S.P., Huang, G., Shui, W., and Mackay, D. (1985). Relationships between octanol-water partition coefficient and aqueous solubility. Environmental Science and Technology, 19 (6), 522-529. Monenco Consultants Ltd (1991). Chlorinated Organics, Water Quality and Fisheries Survey in the Peace, Smoky and Slave Rivers, Alberta and Northwest Territories, Vol. 1. Prepared for Daishowa Canada Co. Ltd., PRP 8634-8/6352 (167 pp). Montgomery, J .M. Consulting Engineers Inc. (1985). Water Treatment Principles and Design. John Wiley and Sons (696 pp). Montaque, C.L., Paulic, M . , and Parchure, T .M. (1993). The stability of sediments containing microbial communities: initial experiments with varying light intensity. In Mehta, A.J . (Ed.) Nearshore and Estuarine Cohesive Sediment Transport (pp 348-359). Washington DC, American Geophysical Union. Moussa, Z . M . , Trischka, J.W., and Eskinazi, S. (1977). The near field mixing of a round jet with a cross-stream. Journal of Fluid Mechanics, 80, (1), 49-80. Muhle, K. (1993). Floe Stability in laminar and turbulent flow. In Dobias B. (Ed.) Coagulation and Flocculation Theory and Applications (pp 265-319). Marcel Dekker Inc, New York. Murphy, E .M. , Zachara, J .M., and Smith, S.C. (1990). Influence of mineral-bound humic substances on the sorption of hydrophobic organic compounds. Environmental Science and Technology, 24 (10). 1507-1516. 211 NCASI (1978). Characterization of Dispersed Residual Solids in Biologically Treated Pulp and Paper M i l l Effluents. Stream Improvement Technical Bulletin No. 303. National Council of the Paper Industry for Air and Stream Improvement, Inc., New York, New York, p 21. Northwest Hydraulic Consultants Ltd. (1994). Determination of Sediment Depostition Zones Fraser River Basin. DOE FRAP Report No. 1994-32. Environment Canada, Vancouver, British Columbia. Northcote, T.G. and Larkin P.A. (1989). The Fraser River: A major salmonine production system. In Dodge, D.P. (Ed.) Proceedings of the International Large River Symposium. Canadian Special Publication of Fisheries and Aquatic Sciences, 106, 172-204. Parker, D.S., Kaufman, M . , and Jenkins, D. (1972). Floe breakup in turbulent flocculation processes. A S C E Journal of the Sanitary Engineering Division, 98 (1), 79-99. Partheniades, E. (1993). Turbulence, flocculation and cohesive sediment dynamics. In Mehta, A.J . (Ed.) Nearshore and Estuarine Cohesive Sediment Transport (pp 40-59). Washington DC, American Geophysical Union. Pearson, H.J., Valioulis, I.A.„ and List, E.J. (1984). Monte-Carlo simulation of coagulation in discrete particle-size distributions. Part 1. Brownian motion and fluid shearing. Journal of Fluid Mechanics, 143, 367-385. Podoll, R.T., and Mabey, W.R. (1984). Factors to consider in conducting laboratory sorption/desorption tests. In Dickson, K .L , Maki, A.W., and Brungs, W.A. (Eds.) Fate and Effects of sediment Bound Chemicals in Aquatic Systems (pp 99-108). Proceedings of the sixth Pellston Workshop, Florissant, Colorado, Pergamon Press (449 pp). Potter, M . C , Wiggert, D.C. and Hondzo, M . (1997). Mechanics of Fluids. Prentice-Hall Inc. (752 pp). Prahacs, S.M (1994). An Evaluation of Leeches as In-Situ Biomonitors of Chlorinated Phenolic Compounds Discharged from Bleach Kraft Pulp Mills (MSc Thesis). Department of Resource Management and Environmental Studies, University of British Columbia, Vancouver, British Columbia (192 pp). Pratte, B.D., and Baines, W.D. (1967). Profiles of round turbulents jets in a cross flow. A S C E Journal of Hydraulics Division. 92 (3), 815-816. Prokopy, W.R. (1993). Phenol and Phenolic Materials Distilled Water Samples QuikChem Method 10-210-001-B. Lachat Instuments, Milwaukee, Wisconson. 212 Rajaratnam, N . (1976). Turbulent Jets. Elsevier Scientific Publishing Company. Amsterdam, Netherlands. Reynolds, T.D. (1982). Unit Operations and Processes in Environmental Engineering. PWS-K E N T Publishing Company, Boston, Massachusetts (576 pp). Rajaratnam, N . and Langat, J.K. (1995). Mixing region of circular turbulent wall jets in crossflows. Journal of Hydraulic Engineering, 121 (10), 694-698. Rohr, A .C . (1994). Application of Semipermeable Membrane Devices (SPMDs) to the Monitoring of Kraft M i l l Effluents with Emphasis on Potential Fish-Tainting Compounds (MASc Thesis). Department of Civil Engineering, University of British Columbia, Vancouver, British Columbia (187 pp). Roy, W.R., Krapac, I.G., Chou, S.F.J, and Griffin, R. A . (1991). Batch-Type Procedures for Estimating Soil Adsorption of Chemicals. Technical Resource Document EPA/530-SW-87-006-F. Illinois State Geological Survey, Champaign, Illinois. Russel, W.A., Saville, D.A., and Schowalter, W.R. (1989). Colloidal Dispersions. Cambridge University Press (525pp). Schellenberg, K., Leuenberger, C , and Schwarzenbach, R.P. (1984). Sorption of chlorinated phenols by natural sediments and aquifer materials. Environmental Science and Technology, 18 (9), 652-657. Schreier, H. , Brown, S.J., and Hall, K.J . (1991). The land-water interface in the Fraser River basin. In Dorcey A.H.J, and Griggs J.R. (Ed.) Water in Sustainable Development: Exploring Our Common Future in the Fraser River Basin (pp 77-116). Westwater Research Centre, University of British Columbia, Vancouver, British Columbia. Schroeder, E.D. (1977). Water and Wastewater Treatment. McGraw-Hill. Schwarzenbach, R.P. (1985). Sorption behavior of neutral and ionizable hydrophobic organic compounds. In Bjorseth, A . and Angeletti, G. (Ed) Organic Micropollutants in the Aquatic Environment Proceedings of the Fourth European Symposium (ppl68-177), Vienna, Austria. Schwarzenbach, R.P., Gschwend, P .M. , and Imboden, D . M . (1993). Environmental Organic Chemistry. John Wiley and Sons, Inc. Scorer, R.S. (1978). Environmental Aerodynamics. Ellis Horwood, Chichester, U.K. 213 Sekela, M . , Brewer, R., Baldazzi, C , and Moyle, G. (1994). Change in Contaminant Concentration in Fraser River Suspended Sediment and Water During the Onset of Freshet (Marguerite - 1993). DOE FRAP Report No. 1994-29. North Vancouver: Aquatic Section, Science Division, Environmental Conservation Branch, Pacific and Yukon Region, Environment Canada. Sekela, M . , Brewer, R., Baldazzi, C , Moyle, G., and Tuominen, T. (1995). Survey of Contaminants in Suspended Sediment and Water in the Fraser River. DOE FRAP Report No. 1995-21. North Vancouver: Aquatic Section, Science Division, Environmental Conservation Branch, Pacific and Yukon Region, Environment Canada. Shaw, D.P., and El-Shaarawi, A . H . (1995). Patterns in Water Quality at Selected Stations in the Fraser River Basin (1985-1991). DOE FRAP Report No. 1995-20. Rivers Research Branch, National Water Research Institute, Burlingtion, Ontario. Stoy, R.L. and Ben-Haim, Y . (1973). Turbulent jets in a confined crossflow. AS M E Journal of Fluids Engineering, 551-556. Strenge, K. (1993). Structure formation in disperse systems. In Dobias B. (Ed.) Coagulation and Flocculation Theory and Applications (pp 265-319). Marcel Dekker Inc, New York. Suntio ,L.R., Wan, Y.S., Mackay, D. (1988). A review of the nature and properties of chemicals present in pulp mill effluent. Chemosphere, 17 (7), 1249-1290. Tomar, P., and Allen, D.G. (1991). Removal of organochlorines from kraft pulp mill effluent by an aerated lagoon. Water Pollution Research Journal Canada, 26 (1), 101-117. Tritton, D.J. (1988). Physical Fluid Dynamics. Clarendon Press, Oxford. Turner, J.S. (1960). A comparison between buoyant vortex rings and pairs. Journal of Fluid Mechanics, 7, 419-432 Unkulvasapaul, M . (1994). Resin Acids Compounds in a Kraft Pulp and Paper M i l l Effluent and Their Distribution in Bottom Sediments Near the Outfall (PhD Thesis). The College of William and Mary in Virginia. Veith, G.D., Austin, N . M . and R.T. Morris (1979). A rapid method for estimating log P for organic chemicals. Water Research, 13, 43-47. Velz, C.J. (1934). Influence of temperature on coagulation. Civil Engineering, 4 (7), 345-349. 214 Vine, J. (1996). Randon Walk Model Applicable to Rivers (MASC Thesis). University of British Columbia, Vancouver (p 107). Voss, R. H. and Yunker M . B. (1983). A Study of Chlorinated Phenolics Discharged into Kraft M i l l Receiving Waters. Pulp and Paper Research Institute of Canada, Point Claire, Quebec, Canada. Voice, T.C., Rice, C P . , and Weber, W.J., (1983). Effects of solids concentration on the sorptive partitioning of hydrophobic pollutants in aquatic systems. Environmental Science and Technology, 17 (9), 513-518. Voss, R.H., Wearing, J.T., and Wong, A . (1981). A novel gas chromatographic method for the analysis of chlorinated phenolics in pulp mill effluents. In Keith, L . H . (Ed.) Advances in the Identification and Analysis of Organic Pollutants in Water, Volume 2 (pp 1059-1095). Ann Arbor Science. Wang, L., Govind, R., and Dobbs, R.A. (1993). Sorption of toxic organic compounds on wastewater solids: mechanism and modelling. Environmental Science and Technology. 27 (1), 152-158. Westall, J .C , Leuenberger, C , and Schwarzenbach, R.P. (1985). Influence of pH and ionic strength on the aqueous-nonaqueous distribution of chlorinated phenols. Environmental Science and Technology, 19 (2), 193-198. Wilson, D.G. and Holloran, M.F. (1992). Decrease of A O X with various external effluent treatments. Pulp and Paper Canada, 93 (12), 136-142. Winterwerp, J.C. (1998). A simple model for turbulence induce flocculation of cohesive sediment. Journal of Hydraulic Research, 36 (3), 309-326. Wood, I.R., Bell, R.G., and Wilkinson, D.L. (1993). Ocean Disposal of Wastewater. World Scientific Publishing Co., Singapore (425 pp). Wright, S.J. (1977). Mean behaviour of buoyant jets in a crossflow. A S C E Journal of Hydraulics Division, 103 (5), 499-513. Wright, S.J. (1978). Closure to discussion of "Mean behavior of buoyant jets in a crossflow." A S C E Journal of Hydraulics Division. 104 (9), 1359-1360. Xie, T .M. , Abrahamsson, K., Fogelqvist, E., and Josefsson, B. (1986). Distribution of chlorophenolics in a marine environment. Environmental Science and Technology, 20 (5), 457-463. 215 Xie, T .M. , Hulthe, B. , and Folestad, S. (1984). Determination of partition coefficients of chlorinated phenols, guaiacols and catechols by shake-flask GC and HPLC. Chemosphere, 13 (3), 445-459. Yotsukura, N and W. W. Sayre, 1976. Transverse mixing in natural channels. Water Resources Research, 12 (4), 695-704. Yu-Ping, C , and Weber W.J. Jr. (1989). Estimating the effects of dispersed organic polymers on the sorption of contaminants by natural solids. 1) A predictive thermodynamic humic substance-organic solute interaction model. Environmental Science and Technology, 23 (8), 978-984. Zanella, E., Conkey, J., and Tesmer, M . (1978). Nonsettleable solids: Characteristics and use by two aquatic food chain organisms. TAPPI, 61 (10), 61-65. Zrymiak, P. and Tassone, B. (1986). Sediment Station Analysis: Fraser River at Hansard. Environment Canada, Water Resources Branch, Inland Waters Directorate. 216 A P P E N D I X A H P L C Determination of K, The logarithm of the retention time of an organic compound in high pressure liquid chromatography is linearly related to the logarithm of the octanol-water partition coefficient (Veith et al., 1979; Xie et al., 1984)). A calibration curve can be made using reference compounds with known KoW 's (de Kock et al., 1987). From this, the coefficient for other compounds can be estimated from their retention time. This method was used to determine the Kow for three pulp mill pollutants. The six reference compounds used were: p ,p' DDT, napthalene, fluoranthene, toluene, benzoic acid and pentachlorophenol. The three unknown compounds used were 4 ,5. 6-trichloroguaiacol, tetrachloroguaiacol and dehydroabietic acid. Acetone was used to determine the unretained volume and was diluted by a factor of 1/100 for the HPLC. Acetone is easily detected by the U V detector and is chemically innocuous to the HPLC instrumentation. Each compound was made up to approximately 20 ppm in HPLC grade methanol. The mobile phase was 75% methanol: 25% buffered H2O. The buffer was 0.06 Molar ammonium phosphate. For ionizable compounds (DHA, guaiacols, acids and phenols) this buffer was acidified to a pH of 2 using phosphoric acid. Both buffers were filtered through a 0.45 micron filter. The ultraviolet spectrum of each compound and its corresponding mobile phase was determined on a Beckman DB-G U V spectrometer located in the Environmental Engineering Laboratory . This information is required to properly set the wavelength of the U V detector on the HPLC system. The Waters 625 L C system HPLC at Paprican was used. The column used was a UBondapak C18 column and the system was equipped with a U V detector set at varying wavelengths. The flow rate of the mobile phase was 1.0 ml/minute and the injection volume was 5 pi. A l l injections were carried out at room temperature, which was 21 °C. Prior to the determination of the retention time on the HPLC, the ultraviolet spectrum of each compound and its corresponding mobile phase was determined and is given in Table A . l . This information was required to properly set the wavelength of the U V detector on the HPLC system for each specific compound. 217 Table A - l Compound Wavelength nm P ,p' DDT 235 Napthalene 272 toluene 270 Fluoranthene 284 TeCG 212 4, 5, 6-TCG 212 PCP 220 benzoic acid 228 D H A 205 acetone 270 The capacity factors (k) of each compound was calculated from the retention time of the solute, tr, an the retention time of the unretained solute, t0, using: k = where tQ = 3.57 The retention times, capacity factors, log capacity factors, and log of the partition coefficient for the standard compounds are given in Table A.2 and the calibration curve is shown in Figure A. 1. 218 Figure A.l Determination of Octanol-Water Coefficient using HPLC 6 5 & 4 " "Ho 3 " ^ ^ ^ ^ — 2 ^^^^^^ 1 n u 1 -0.5 0 0.5 1 log(k) Table A-2 Compound tr k log/: log K o w p, p' DDT 23.83 5.675 0.754 6.19 (Viethetal., 1979) PCP 16.13 3.518 0.546 5.06 (Xie et al., 1984) fluoranthene 13.77 2.857 0.456 5.22 (Miller et al., 1985)) napthalene 6.77 0.896 -0.0477 3.37 (Viethetal., 1979) toluene 5.87 0.644 -0.191 2.69 (Chiou et al., 1977) benzoic acid 4.13 0.157 -0.804 1.87 (Chiou etal. 1977) Linear regression of this data yields the following linear relationship: log(Kow)=2.825(logk) + 3.731 Using this calibration equation, the following octanol-water partition coefficients were found for the three test compounds. D H A 4.55 4, 5, 6 - T C G 3.677 TeCG 4.753 The values determined for the two guaiacols compare relatively well to those in literature (Xie et al., 1984): 3.92 and 4.76 for 4, 5, 6-TCG and TeCG respectively. No previously documented octanokwater partition coefficients were found for DHA. 219 APPENDIX B Temperature Data acquisition Program 100 ' * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 120 '* Program used for temperature data acquisition-type T thermistors 130 •* Lesl ie Gomm, October 20, 1992 140 i * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ' STEP 1: In i t ia l i ze an integer array D%(15) to receive data DIM D%(15) '16 elements, one for each EXP-16 channel 'Also i n i t i a l i z e a corresponding real array to receive temperature delta DIM T(15) DIM V(15) DIM VC(15) DIM ST(15) DIM ST2(15) DIM SV(15) DIM SV2(15) DIM SVC(15) DIM SVC2(15) DIM LT%(15) COMMON SHARED D%(), LT%() DECLARE SUB das8 (mode%, BYVAL dummy%, FLAG%) SCREEN 0, 0, 0: KEY OFF: CLS : WIDTH 80 GOSUB 2000: CLS This example performs scanning and measurement of T type thermocouples connected to one EXP-16. The program can be expanded to handle multiple EXP-16's . Steps are:-1 - Dimension other arrays and provide set up information 2 - In i t ia l i ze DAS-8 and load thermocouple look up tables 3 - Measure temperature of connector block from CJC channel (CJC = cold junction compensation) 4 - Measure output voltages of thermocouples on EXP-16 5 - Convert, correct and linearize thermocouple outputs to degrees 6 - Display output Note that thermocouple routines J.BAS, K.BAS etc. are in ASCII form and can be MERGE'd with any program and edited in . For purposes of example the EXP-16 output channel should be connected to DAS-8 channel #0 and the CJC channel to DAS-8 channel #7. STEP 2: i n i t i a l i z e 220 380 CLEAR , 49152! 385 LOCATE 25, 1: COLOR 0, 7: PRINT "-PLEASE WAIT-"; : COLOR 7, 0: PRINT Loading DAS-8 I/O address and thermocouple lookup table data": LOCATE 1, 1 440 OPEN "DAS8.ADR" FOR INPUT AS #1 450 INPUT #1, BASADR% ' i n i t i a l i z e & declare CALL parameters 460 CLOSE #1 480 FLAG% = 0 490 MD% = 0 'Mode 0 = i n i t i a l i z a t i o n 500 CALL das8(MD%, VARPTR(BASADR%), FLAG%) 510 IF FLAG% <> 0 THEN PRINT "INSTALLATION ERROR" 520 OPEN "w4.dat" FOR OUTPUT AS #1 530 'Load thermocouple l inearizing look up data 540 GOSUB 50000 542 'Get gain setting of EXP-16 545 CLS : INPUT "EXP-16 Gain setting (100,200,1000 etc . ) : ", AV 547 CLS 590 ' 900 ' STEP 6: Display temperature data 910 LOCATE 1, 1 92 0 DO 921 GOSUB 1600 922 LOCATE 1, 1 923 FOR I = 0 TO 15 930 PRINT USING "Ch ## temp= ###.# C, volt.=#####.###,cjc=#####.###"; I; T(I); V(I); VC(I) 93 8 NEXT I 93 9 test$ = INKEY$ 940 PRINT 941 PRINT USING "Cold junction temperature (CJC) = ###.# deg. C . " ; CJC 9 42 LOOP UNTIL test$ = "s" 943 INPUT "postition of thermistors", P 944 FOR I = 0 TO 15 945 ST(I) = 0 946 ST2(I) = 0 947 SV(I) = 0 948 SV2(I) = 0 949 SVC(I) = 0 950 SVC2(I) = 0 951 NEXT I 952 FOR C = 1 TO 1000 953 GOSUB 1600 954 FOR I = 0 TO 15 955 ST(I) = ST(I) + T(I) 956 ST2(I) = ST2(I) + (T(I) * 2) 957 SV(I) = SV(I) + V(I) 958 SV2(I) = SV2(I) + (V(I) " 2) 959 SVC(I) = SVC(I) + VC(I) 960 SVC2(I) = SVC2(I) + (VC(I) ~ 2) 961 NEXT I 962 N = C 963 NEXT C 964 FOR I = 0 TO 15 965 WRITE #1, I, ST(I), ST2(I), SV(I), SV2(I), SVC(I), SVC2(I), CJC, N, P 967 NEXT I 968 PRINT "done"; P 969 GOTO 900 221 1000 ' Subroutine to convert EXP-16 channels to number of bits 1010 'Firs t lock DAS-8 on the one channel that EXP-16 is connected to. 1020 LT%(0) = CH%: LT%(1) = CH% : MD% = 1 1030 CALL das8(MD%, VARPTR(LT%(0)), FLAG%) 1040 IF FLAG% <> 0 THEN PRINT "ERROR IN SETTING CHANNEL": END 1050 'Next select each EXP-16 channel in turn and convert i t . 1060 'Digital outputs OP1-4 drive the EXP-16 sub-multiplexer address, so use 1070 'mode 14 to set up the sub-multiplexer channel. 1080 FOR MUX% = 0 TO 15 'note use of integer index MUX% 1090 MD% = 14 1100 CALL das8(MD%, VARPTR(MUX%), FLAG%) 'address set 1110 IF FLAG% <> 0 THEN PRINT "ERROR IN EXP-16 CHANNEL NUMBER": END 1120 'Now that channel is selected, perform A/D conversion using mode 4. 1130 'Transfer data to corresponding array element D%(MUX%) 1140 MD% =4 'do 1 A/D conversion 1150 CALL das8(MD%, VARPTR(D%(MUX%)), FLAG%) 1160 IF FLAG% <> 0 THEN PRINT "ERROR IN PERFORMING A/D CONVERSION" 1170 'Now repeat sequence for a l l other EXP-16 channels 118 0 NEXT MUX% 1190 ' A l l done - return fron subroutine 1200 RETURN 1210 ' 1600 ' STEP 3: Get cold junction compensation temperature 1610 'Output of CJC channel is scaled at 24.4mV/deg.C. This corresponds to 1620 '0.1 deg .C. /b i t . Dividing output in bits by 10 yields degrees C. 1630 ' 1640 'Lock DAS-8 to channel #7 (CJC channel selected) using mode 1 1650 MD% = 1: LT%(0) = 7: LT%(1) = 7 1660 CALL das8(MD%, VARPTR(LT%(0)), FLAG%) 1670 IF FLAG% <> 0 THEN PRINT "ERROR IN SETTING CJC CHANNEL": END 1680 'Next get CJC data from this channel using Mode 4 1690 MD% = 4: CJ% = 0 1700 CALL das8(MD%, VARPTR(CJ%), FLAG%) 1710 'Change output in bits to real temperature 1720 CJC = CJ% / 10 1730 ' 1740 ' STEP 4: Get the thermocouple data 1750 CH% = 0 1760 GOSUB 1000 1770 'This step is written as a subroutine so you can use i t in your own 1780 'programs by editing i t out. Entry parameters are:-1790 ' CH% - specifies DAS-8 channel that EXP-16 is connected to (0-7) . 1800 ' D%(15) - integer data array to receive data from channels. 1810 ' 1820 ' STEP 5: Convert data to volts and l inearize 1830 'AV = Gain setting on Dipswitch of EXP-16 (change to su i t ) . 1840 FOR I = 0 TO 15 1850 V = (D%(I) * 5) / (AV * 2048) 1860 GOSUB 51000 'perform look-up l inearization 1 8 7 0 T ( I ) = T C '= TF for degrees Fahrenheit 1875 VC(I) = VT 1876 V(I) = V * 1000 1880 NEXT I 1890 RETURN 222 2000 ' Subroutine to describe operation and connections (pre-amble) 2030 PRINT "with following: 2040 PRINT 2050 PRINT 1 2060 PRINT 2 2070 PRINT 3 2080 PRINT 4 2090 PRINT 2100 PRINT 2010 CLS 2020 PRINT " This program demonstrates the operation of thermocouples" Acquires the data" Linearizes and performs cold junction compensation' Displays data" The thermocouples should be attached to the HI & LO's of each" 2110 PRINT "channel on the EXP-16 starting with channel 0. Also you should" 2115 PRINT "connect LO to L . L . GND. on each channel. Alternatively, for a" 2120 PRINT "permanent instal lat ion, a better method is to f i l l the solder gaps" 2130 PRINT "on the back of the EXP-16 board behind the connector with solder" 2140 PRINT "so that the two semi-circular halves are shorted together. The" 2145 PRINT "thermocouples can then be connected to HI & LO only (no jumper is" 2147 PRINT "required to L . L . GND.)." 2150 PRINT " Set the OUTPUT CHANNEL jumper block to position 0 and the CJC" 2155 PRINT "CHANNEL to position 7." 2160 PRINT " On running the program, you w i l l be prompted for the gain" 2170 PRINT "setting of the EXP-16. Usually a gain of 100,200 or 1000 is best" 2180 PRINT "for thermocouples depending on their output and measuring range." 2185 PRINT " LIST this program for commented explanation of steps." 2190 PRINT : COLOR 0, 7: PRINT " - Press any key to continue - " ,- : COLOR 7, 0 2200 IF INKEY$ = "" GOTO 2200 2210 RETURN 50000 ' Table lookup data for T type thermocouple 50230 ' 51000 ' Interpolation routine to find T thermocouple temperature 51010 'Entry variables:-5102 0 ' CJC = cold junction compensator temperature in deg. C. 51030 ' V = thermocouple voltage in volts 51040 'Exit variables:-51050 ' TC = temperature in degrees Centigrade 51060 ' TF = temperature in degrees Fahrenheit 51070 'Execution time on std. IBM P.C. = 46 milliseconds 51080 'Perform CJC compensation for T type 51090 VT = 1000 * V + .992 + (CJC - 25) * .040667'VT in mV 51100 ' 51120 ' Calculate calibrated temperatures 51130 IF I = 0 THEN TC = VT * 25.245 - .8712 51140 IF I = 1 THEN TC = VT * 25.2304 - .7886 51150 IF I = 2 THEN TC = VT * 25.2966 - .7355 223 51160 IF I = 3 THEN TC = VT * 25.3329 - .744 51170 IF I = 4 THEN TC = VT * 25.159 - .7684 51180 IF I = 5 THEN TC = VT * 25.2675 - .7284 51190 IF I = 6 THEN TC = VT * 25.2103 - . 6851 51200 IF I = 7 THEN TC = VT * 25.3473 - . 6879 51210 IF I = 8 THEN TC = VT * 25.2499 - . 9541 51220 IF I = 9 THEN TC = VT * 25.1989 - . 8309 51230 IF I = 10 THEN TC = VT * 25.2166 - 1.0064 51240 IF I = 11 THEN TC = VT * 25.2754 - .9362 51250 IF I = 12 THEN TC = VT * 25.2124 - .9163 51260 IF I = 13 THEN TC = VT * 25.2281 - .9007 51270 IF I = 14 THEN TC = VT * 25.2156 - .9283 51280 IF I = 15 THEN TC = VT * 25.209 - 1 51290 TF = TC * 9 / 5 + 321 Fahrenheit 51300 RETURN 224 APPENDIX C Thermistor Calibration Data and Curves Thermister Number 0 Calibration Data Linear Regression Fit T=25.2450(CV)-0.8712 where T = temperature in degrees C CV= cjc corrected millivolts slope = 25.2450 intercept =-0.8712 C J C m V Temp °C C J C m V Temp C 0.0911 1.0500 0.9034 22.2000 0.1501 2.2500 0.9092 23.2000 0.1452 2.7000 0.9301 23.3000 0.1796 3.5000 0.9484 23.7000 0.1834 3.6000 0.9732 24.2000 0.2333 4.8000 1.0109 24.9500 0.2445 5.1000 1.0326 25.5000 0.2841 6.1000 1.0491 25.9000 0.3067 6.7000 1.0729 26.4000 0.3600 8.0500 1.1197 27.5000 0.3580 8.1000 1.1344 27.9000 0.3887 8.7000 1.1616 28.5000 0.4162 9.5000 1.1938 29.4000 0.4229 9.8000 1.2462 30.5000 0.4550 10.5000 1.2737 31.2000 0.4697 11.1000 1.2896 31.8000 0.5077 11.8000 1.3000 32.5000 0.5146 12.1000 1.3464 33.4000 0.5342 12.6500 1.3929 34.4000 0.5441 12.9000 1.4232 35.0000 0.5804 13.8000 1.4403 35.6000 0.6188 14.7000 1.4802 36.4000 0.6412 15.2000 1.5268 37.4000 0.6648 15.7500 1.5380 38.0000 0.6981 16.6000 1.5618 38.5000 0.7222 17.3000 1.6066 39.4000 0.7348 17.9000 1.6434 40.3000 0.7618 18.4000 1.6633 41.3000 0.7964 19.2500 1.7122 42.3000 0.8178 19.9000 1.7583 43.4000 0.8436 20.5000 1.7980 44.3000 0.8623 20.8500 1.8363 45.3000 0.8668 21.2500 1.8672 46.0000 0.8924 21.9000 1.9189 47.0000 1.9653 47.9000 Millivolts (CJC) 226 Thermister Number 1 Calibration Data Linear Regression Fit T=25.2304(CV)-0.7886 where T=temperature in degrees C C V = cjc corrected millivolts slope = 25.2304 intercept = -0.7886 C J C m V Temp C C J C m V Temp C 0.0862 1.0500 0.9071 22.2000 0.1489 2.2500 0.9165 23.2000 0.1379 2.7000 0.9264 23.3000 0.1759 3.5000 0.9472 23.7000 0.1773 3.6000 0.9708 24.2000 0.2284 4.8000 1.0060 24.9500 0.2408 5.1000 1.0339 25.5000 0.2805 6.1000 1.0430 25.9000 0.3043 6.7000 1.0705 26.4000 0.3576 8.0500 1.1136 27.5000 0.3531 8.1000 1.1344 27.9000 0.3863 8.7000 1.1616 28.5000 0.4137 9.5000 1.1889 29.4000 0.4229 9.8000 1.2414 30.5000 0.4501 10.5000 1.2676 31.2000 0.4709 11.1000 1.2871 31.8000 0.5041 11.8000 1.3024 32.5000 0.5073 12.1000 1.3439 33.4000 0.5269 12.6500 1.3905 34.4000 0.5393 12.9000 1.4208 35.0000 0.5828 13.8000 1.4342 35.6000 0.6176 14.7000 1.4778 36.4000 0.6412 15.2000 1.5256 37.4000 0.6575 15.7500 1.5380 38.0000 0.6969 16.6000 1.5557 38.5000 0.7246 17.3000 1.6029 39.4000 0.7397 17.9000 1.6409 40.3000 0.7618 18.4000 1.6609 41.3000 0.7964 19.2500 1.7085 42.3000 0.8105 19.9000 1.7498 43.4000 0.8400 20.5000 1.7931 44.3000 0.8538 20.8500 1.8314 45.3000 0.8607 21.2500 1.8660 46.0000 0.8876 21.9000 1.9237 47.0000 1.9653 47.9000 227 228 Thermister Number 2 Calibration Data Linear Regression Fit T=25.2966(CV)-0.7355 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.2966 intercept = -0.7355 C J C m V Temp C C J C m V Temp C 0.0862 1.0500 0.8998 22.2000 0.1464 2.2500 0.9079 23.2000 0.1391 2.7000 0.9264 23.3000 0.1771 3.5000 0.9411 23.7000 0.1749 3.6000 0.9696 24.2000 0.2211 4.8000 0.9962 24.9500 0.2384 5.1000 1.0265 25.5000 0.2829 6.1000 1.0406 25.9000 0.3043 6.7000 1.0656 26.4000 0.3564 8.0500 1.1112 27.5000 0.3555 8.1000 1.1270 27.9000 0.3802 8.7000 1.1519 28.5000 0.4137 9.5000 1.1840 29.4000 0.4131 9.8000 1.2389 30.5000 0.4550 10.5000 1.2615 31.2000 0.4733 11.1000 1.2762 31.8000 0.4943 11.8000 1.2951 32.5000 0.5085 12.1000 1.3329 33.4000 0.5232 12.6500 1.3881 34.4000 0.5429 12.9000 1.4110 35.0000 0.5767 13.8000 1.4330 35.6000 0.6078 14.7000 1.4704 36.4000 0.6278 15.2000 1.5231 37.4000 0.6550 15.7500 1.5319 38.0000 0.6884 16.6000 1.5545 38.5000 0.7160 17.3000 1.5980 39.4000 0.7335 17.9000 1.6348 40.3000 0.7545 18.4000 1.6572 41.3000 0.7940 19.2500 1.6999 42.3000 0.8056 19.9000 1.7461 43.4000 0.8363 20.5000 1.7870 44.3000 0.8465 20.8500 1.8253 45.3000 0.8534 21.2500 1.8635 46.0000 0.8827 21.9000 1.9164 47.0000 1.9576 47.9000 229 Thermister No. 2 0 0.5 1 1.5 2 Millivolts (CJC) 230 Thermister Number 3 Calibration Data Linear Regression Fit T=25.3329(CV)-0.7440 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.3329 intercept = -0.7440 C J C m V Temp C CJC mV Temp C 0.0801 1.0500 0.8924 22.2000 0.1440 2.2500 0.9043 23.2000 0.1415 2.7000 0.9264 23.3000 0.1759 3.5000 0.9399 23.7000 0.1786 3.6000 0.9671 24.2000 0.2296 4.8000 0.9999 24.9500 0.2408 5.1000 1.0290 25.5000 0.2829 6.1000 1.0418 25.9000 0.3031 6.7000 1.0705 26.4000 0.3527 8.0500 1.1124 27.5000 0.3568 8.1000 1.1246 27.9000 0.3826 8.7000 1.1543 28.5000 0.4064 9.5000 1.1877 29.4000 0.4168 9.8000 1.2353 30.5000 0.4489 10.5000 1.2566 31.2000 0.4672 11.1000 1.2749 31.8000 0.4919 11.8000 1.3012 32.5000 0.5037 12.1000 1.3341 33.4000 0.5232 12.6500 1.3832 34.4000 0.5380 12.9000 1.4159 35.0000 0.5767 13.8000 1.4330 35.6000 0.6115 14.7000 1.4656 36.4000 0.6314 15.2000 1.5146 37.4000 0.6526 15.7500 1.5307 38.0000 0.6859 16.6000 1.5459 38.5000 0.7173 17.3000 1.5956 39.4000 0.7372 17.9000 1.6275 40.3000 0.7533 18.4000 1.6536 41.3000 0.7927 19.2500 1.6951 42.3000 0.8129 19.9000 1.7461 43.4000 0.8314 20.5000 1.7821 44.3000 0.8501 20.8500 1.8277 45.3000 0.8522 21.2500 1.8611 46.0000 0.8766 21.9000 1.9127 47.0000 1.9563 47.9000 231 Thermister No. 3 0 0.5 1 1.5 2 Millivolts (CJC) 232 Thermister Number 4 Calibration Data Linear Regression Fit T=25.1590(CV)-0.7684 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.1590 intercept = -0.7684 C J C m V Temp C C J C m V Temp C 0.0850 1.0500 0.9022 22.2000 0.1476 2.2500 0.9177 23.2000 0.1379 2.7000 0.9325 23.3000 0.1796 3.5000 0.9508 23.7000 0.1786 3.6000 0.9769 24.2000 0.2260 4.8000 1.0072 24.9500 0.2420 5.1000 1.0302 25.5000 0.2829 6.1000 1.0454 25.9000 0.3055 6.7000 1.0705 26.4000 0.3551 8.0500 1.1221 27.5000 0.3555 8.1000 1.1392 27.9000 0.3875 8.7000 1.1677 28.5000 0.4101 9.5000 1.1962 29.4000 0.4217 9.8000 1.2426 30.5000 0.4538 10.5000 1.2725 31.2000 0.4697 11.1000 1.2932 31.8000 0.5028 11.8000 1.3073 32.5000 0.5098 12.1000 1.3500 33.4000 0.5317 12.6500 1.3954 34.4000 0.5393 12.9000 1.4232 35.0000 0.5828 13.8000 1.4440 35.6000 0.6090 14.7000 1.4778 36.4000 0.6375 15.2000 1.5280 37.4000 0.6587 15.7500 1.5429 38.0000 0.6920 16.6000 1.5618 38.5000 0.7185 17.3000 1.6066 39.4000 0.7384 17.9000 1.6458 40.3000 0.7630 18.4000 1.6694 41.3000 0.7976 19.2500 1.7122 42.3000 0.8178 19.9000 1.7571 43.4000 0.8436 20.5000 1.7956 44.3000 0.8550 20.8500 1.8375 45.3000 0.8607 21.2500 1.8672 46.0000 0.8912 21.9000 1.9189 47.0000 1.9640 47.9000 233 Thermister No. 4 0 0.5 1 1.5 2 Millivolts (CJC) 234 Thermister Number 5 Calibration Data Linear Regression Fit T=25.2675(CV)-0.7256 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.2675 intercept = -0.7256 C J C m V Temp C CJC mV Temp C 0.0825 1.0500 0.9010 22.2000 0.1464 2.2500 0.9140 23.2000 0.1440 2.7000 0.9240 23.3000 0.1747 3.5000 0.9460 23.7000 0.1798 3.6000 0.9708 24.2000 0.2211 4.8000 0.9987 24.9500 0.2335 5.1000 1.0314 25.5000 0.2768 6.1000 1.0430 25.9000 0.3006 6.7000 1.0692 26.4000 0.3539 8.0500 1.1185 27.5000 0.3519 8.1000 1.1270 27.9000 0.3826 8.7000 1.1604 28.5000 0.4125 9.5000 1.1962 29.4000 0.4180 9.8000 1.2377 30.5000 0.4501 10.5000 1.2640 31.2000 0.4697 11.1000 1.2798 31.8000 0.4955 11.8000 1.2975 32.5000 0.5073 12.1000 1.3378 33.4000 0.5244 12.6500 1.3881 34.4000 0.5344 12.9000 1.4110 35.0000 0.5706 13.8000 1.4306 35.6000 0.6127 14.7000 1.4680 36.4000 0.6302 15.2000 1.5170 37.4000 0.6538 15.7500 1.5307 38.0000 0.6896 16.6000 1.5533 38.5000 0.7185 17.3000 1.5992 39.4000 0.7348 17.9000 1.6361 40.3000 0.7569 18.4000 1.6585 41.3000 0.7915 19.2500 1.6999 42.3000 0.8129 19.9000 1.7437 43.4000 0.8351 20.5000 1.7895 44.3000 0.8538 20.8500 1.8228 45.3000 0.8497 21.2500 1.8672 46.0000 0.8851 21.9000 1.9152 47.0000 1.9601 47.9000 235 0 0.5 1 1.5 2 Millivolts (CJC) 236 Thermister Number 6 Calibration Data Linear Regression Fit T=25.2103(CV)-0.6851 where T=temperature deg. C CV=cjc corrected millivolts slope = 25.2103 intercept = -0.6851 C J C m V Temp C C J C m V Temp C 0.0850 1.0500 0.8998 22.2000 0.1476 2.2500 0.9067 23.2000 0.1354 2.7000 0.9289 23.3000 0.1735 3.5000 0.9472 23.7000 0.1749 3.6000 0.9696 24.2000 0.2235 4.8000 1.0011 24.9500 0.2396 5.1000 1.0277 25.5000 0.2793 6.1000 1.0430 25.9000 0.3018 6.7000 1.0705 26.4000 0.3490 8.0500 1.1112 27.5000 0.3568 8.1000 1.1295 27.9000 0.3802 8.7000 1.1519 28.5000 0.4088 9.5000 1.1925 29.4000 0.4192 9.8000 1.2426 30.5000 0.4514 10.5000 1.2615 31.2000 0.4685 11.1000 1.2786 31.8000 0.4955 11.8000 1.3012 32.5000 0.5122 12.1000 1.3329 33.4000 0.5220 12.6500 1.3868 34.4000 0.5344 12.9000 1.4220 35.0000 0.5804 13.8000 1.4330 35.6000 0.6103 14.7000 1.4680 36.4000 0.6314 15.2000 1.5170 37.4000 0.6538 15.7500 1.5307 38.0000 0.6859 16.6000 1.5533 38.5000 0.7173 17.3000 1.5943 39.4000 0.7384 17.9000 1.6312 40.3000 0.7533 18.4000 1.6548 41.3000 0.7964 19.2500 1.7097 42.3000 0.8092 19.9000 1.7657 43.4000 0.8302 20.5000 1.8163 44.3000 0.8489 20.8500 1.8289 45.3000 0.8534 21.2500 1.8623 46.0000 0.8876 21.9000 1.9164 47.0000 1.9576 47.9000 0 0.5 1 1.5 2 Millivolts (CJC) 238 Thermister Number 7 Calibration Data Linear Regression Fit T=25.3473(CV)-0.6879 where T=temperature in deg.C CV=cjc corrected millivolts slope = 25.3473 intercept = -0.6879 C J C m V Temp C C J C m V Temp C 0.0788 1.0500 0.8876 22.2000 0.1415 2.2500 0.9043 23.2000 0.1379 2.7000 0.9277 23.3000 0.1759 3.5000 0.9423 23.7000 0.1737 3.6000 0.9671 24.2000 0.2162 4.8000 0.9999 24.9500 0.2384 5.1000 1.0290 25.5000 0.2768 6.1000 1.0357 25.9000 0.3018 6.7000 1.0680 26.4000 0.3466 8.0500 1.1075 27.5000 0.3531 8.1000 1.1234 27.9000 0.3777 8.7000 1.1531 28.5000 0.4015 9.5000 1.1828 29.4000 0.4131 9.8000 1.2316 30.5000 0.4465 10.5000 1.2542 31.2000 0.4672 11.1000 1.2713 31.8000 0.4931 11.8000 1.2963 32.5000 0.5037 12.1000 1.3256 33.4000 0.5232 12.6500 1.3807 34.4000 0.5368 12.9000 1.4074 35.0000 0.5730 13.8000 1.4294 35.6000 0.6066 14.7000 1.4680 36.4000 0.6253 15.2000 1.5073 37.4000 0.6538 15.7500 1.5270 38.0000 0.6872 16.6000 1.5423 38.5000 0.7124 17.3000 1.5931 39.4000 0.7335 17.9000 1.6275 40.3000 0.7569 18.4000 1.6536 41.3000 0.7903 19.2500 1.6951 42.3000 0.8080 19.9000 1.7437 43.4000 0.8265 20.5000 1.7834 44.3000 0.8501 20.8500 1.8167 45.3000 0.8485 21.2500 1.8550 46.0000 0.8863 21.9000 1.9054 47.0000 1.9511 47.9000 239 0 0.5 1 1.5 2 Millivolts (CJC) 240 Thermister Number 8 Calibration Data Linear Regression Fit T=25.2499(CV)-0.9541 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.2499 intercept = -0.9541 C J C m V Temp C C J C m V Temp C 0.0911 1.0500 0.9120 22.2000 0.1550 2.2500 0.9226 23.2000 0.1440 2.7000 0.9362 23.3000 0.1820 3.5000 0.9472 23.7000 0.1834 3.6000 0.9781 24.2000 0.2296 4.8000 1.0157 24.9500 0.2469 5.1000 1.0375 25.5000 0.2878 6.1000 1.0479 25.9000 0.3079 6.7000 1.0790 26.4000 0.3600 8.0500 1.1258 27.5000 0.3653 8.1000 1.1368 27.9000 0.3948 8.7000 1.1689 28.5000 0.4149 9.5000 1.2023 29.4000 0.4326 9.8000 1.2426 30.5000 0.4611 10.5000 1.2737 31.2000 0.4770 11.1000 1.2884 31.8000 0.5028 11.8000 1.3110 32.5000 0.5183 12.1000 1.3500 33.4000 0.5354 12.6500 1.3978 34.4000 0.5490 12.9000 1.4318 35.0000 0.5828 13.8000 1.4440 35.6000 0.6188 14.7000 1.4814 36.4000 0.6448 15.2000 1.5292 37.4000 0.6623 15.7500 1.5478 38.0000 0.7030 16.6000 1.5642 38.5000 0.7258 17.3000 1.6151 39.4000 0.7458 17.9000 1.6483 40.3000 0.7618 18.4000 1.6694 41.3000 0.8025 19.2500 1.7122 42.3000 0.8214 19.9000 1.7376 43.4000 0.8448 20.5000 1.7809 44.3000 0.8648 20.8500 1.8399 45.3000 0.8693 21.2500 1.8770 46.0000 0.8985 21.9000 1.9237 47.0000 1.9704 47.9000 241 0 0.5 1 1.5 2 Millivolts Thermister Number 9 Calibration Data Linear Regression Fit T=25.1989(CV)-0.8309 where T=temperature in deg. C. CV=cjc corrected millivolt slope = 25.1989 intercept =-0.8309 C J C m V Temp C C J C m V Temp C 0.0886 1.0500 0.9034 22.2000 0.1464 2.2500 0.9153 23.2000 0.1379 2.7000 0.9325 23.3000 0.1808 3.5000 0.9460 23.7000 0.1822 3.6000 0.9732 24.2000 0.2296 4.8000 1.0096 24.9500 0.2445 5.1000 1.0339 25.5000 0.2817 6.1000 1.0393 25.9000 0.3104 6.7000 1.0766 26.4000 0.3576 8.0500 1.1173 27.5000 0.3580 8.1000 1.1356 27.9000 0.3875 8.7000 1.1628 28.5000 0.4198 9.5000 1.1974 29.4000 0.4253 9.8000 1.2475 30.5000 0.4599 10.5000 1.2688 31.2000 0.4794 11.1000 1.2835 31.8000 0.4992 11.8000 1.3024 32.5000 0.5146 12.1000 1.3451 33.4000 0.5317 12.6500 1.3942 34.4000 0.5429 12.9000 1.4281 35.0000 0.5791 13.8000 1.4403 35.6000 0.6176 14.7000 1.4790 36.4000 0.6412 15.2000 1.5256 37.4000 0.6599 15.7500 1.5404 38.0000 0.6957 16.6000 1.5569 38.5000 0.7246 17.3000 1.6066 39.4000 0.7433 17.9000 1.6470 40.3000 0.7630 18.4000 1.6670 41.3000 0.7989 19.2500 1.7061 42.3000 0.8178 19.9000 1.7559 43.4000 0.8375 20.5000 1.7968 44.3000 0.8611 20.8500 1.8363 45.3000 0.8631 21.2500 1.8757 46.0000 0.8973 21.9000 1.9225 47.0000 1.9807 47.9000 243 Thermister No. 9 Millivolts Thermister Number 10 Calibration Data Linear Regression Fit T=25.2166(CV)-1.0064 where T=temperature in degrees C CV=cjc corrected millivolts slope = 25.2166 intercept =-1.0064 C J C m V Temp C CJC mV Temp C 0.0923 1.0500 0.9071 22.2000 0.1550 2.2500 0.9214 23.2000 0.1501 2.7000 0.9374 23.3000 0.1857 3.5000 0.9533 23.7000 0.1859 3.6000 0.9757 24.2000 0.2357 4.8000 1.0096 24.9500 0.2506 5.1000 1.0326 25.5000 0.2939 6.1000 1.0515 25.9000 0.3153 6.7000 1.0815 26.4000 0.3625 8.0500 1.1246 27.5000 0.3665 8.1000 1.1441 27.9000 0.3960 8.7000 1.1738 28.5000 0.4211 9.5000 1.1999 29.4000 0.4339 9.8000 1.2475 30.5000 0.4562 10.5000 1.2762 31.2000 0.4831 11.1000 1.2969 31.8000 0.5175 11.8000 1.3085 32.5000 0.5134 12.1000 1.3488 33.4000 0.5317 12.6500 1.4051 34.4000 0.5563 12.9000 1.4318 35.0000 0.5865 13.8000 1.4489 35.6000 0.6127 14.7000 1.4851 36.4000 0.6461 15.2000 1.5317 37.4000 0.6684 15.7500 1.5502 38.0000 0.7043 16.6000 1.5667 38.5000 0.7331 17.3000 1.6175 39.4000 0.7580 17.9000 1.6507 40.3000 0.7728 18.4000 1.6743 41.3000 0.8074 19.2500 1.7146 42.3000 0.8251 19.9000 1.7669 43.4000 0.8522 20.5000 1.8029 44.3000 0.8660 20.8500 1.8399 45.3000 0.8693 21.2500 1.8770 46.0000 0.9047 21.9000 1.9262 47.0000 1.9730 47.9000 245 Millivolts (CJC) 246 Thermister Number 11 Calibration Data Linear Regression Fit T=25.2754(CV.)-0.9362 where T=temperature in degrees C C V = cjc corrected millivolts slope = 25.2754 intercept = -0.9362 C J C m V Temp C C J C m V Temp C 0.0935 1.0500 0.9083 22.2000 0.1562 2.2500 0.9092 23.2000 0.1464 2.7000 0.9301 23.3000 0.1808 3.5000 0.9460 23.7000 0.1822 3.6000 0.9757 24.2000 0.2345 4.8000 1.0035 24.9500 0.2493 5.1000 1.0326 25.5000 0.2951 6.1000 1.0430 25.9000 0.3116 6.7000 1.0766 26.4000 0.3612 8.0500 1.1209 27.5000 0.3665 8.1000 1.1368 27.9000 0.3924 8.7000 1.1628 28.5000 0.4211 9.5000 1.1999 29.4000 0.4241 9.8000 1.2450 30.5000 0.4587 10.5000 1.2688 31.2000 0.4794 11.1000 1.2810 31.8000 0.5065 11.8000 1.3024 32.5000 0.5110 12.1000 1.3500 33.4000 0.5330 12.6500 1.3990 34.4000 0.5490 12.9000 1.4220 35.0000 0.5865 13.8000 1.4367 35.6000 0.6188 14.7000 1.4753 36.4000 0.6375 15.2000 1.5231 37.4000 0.6623 15.7500 1.5441 38.0000 0.6969 16.6000 1.5594 38.5000 0.7258 17.3000 1.6078 39.4000 0.7397 17.9000 1.6434 40.3000 0.7618 18.4000 1.6682 41.3000 0.8025 19.2500 1.7085 42.3000 0.8263 19.9000 1.7657 43.4000 0.8461 20.5000 1.7956 44.3000 0.8599 20.8500 1.8326 45.3000 0.8595 21.2500 1.8745 46.0000 0.8949 21.9000 1.9274 47.0000 1.9704 47.9000 247 248 Thermister Number 12 Calibration Data Linear Regression Fit T=25.2124(CV)-0.9163 where T=temperature in degrees C CV=cjc corrected millivolts slope = 25.2124 intercept =-0.9163 CJCmv Temp C CJC mv Temp C 0.0911 1.0500 0.9120 22.2000 0.1476 2.2500 0.9128 23.2000 0.1403 2.7000 0.9325 23.3000 0.1869 3.5000 0.9472 23.7000 0.1859 3.6000 0.9744 24.2000 0.2345 4.8000 1.0109 24.9500 0.2542 5.1000 1.0326 25.5000 0.2866 6.1000 1.0442 25.9000 0.3067 6.7000 1.0766 26.4000 0.3539 8.0500 1.1295 27.5000 0.3604 8.1000 1.1368 27.9000 0.3960 8.7000 1.1726 28.5000 0.4198 9.5000 1.2035 29.4000 0.4326 9.8000 1.2438 30.5000 0.4538 10.5000 1.2701 31.2000 0.4709 11.1000 1.2920 31.8000 0.5089 11.8000 1.3073 32.5000 0.5171 12.1000 1.3488 33.4000 0.5366 12.6500 1.4003 34.4000 0.5551 12.9000 1.4281 35.0000 0.5865 13.8000 1.4403 35.6000 0.6164 14.7000 1.4802 36.4000 0.6448 15.2000 1.5280 37.4000 0.6636 15.7500 1.5404 38.0000 0.6969 16.6000 1.5655 38.5000 0.7209 17.3000 1.6127 39.4000 0.7470 17.9000 1.6446 40.3000 0.7643 18.4000 1.6719 41.3000 0.8037 19.2500 1.7048 42.3000 0.8239 19.9000 1.7571 43.4000 0.8473 20.5000 1.8053 44.3000 0.8611 20.8500 1.8411 45.3000 0.8705 21.2500 1.8770 46.0000 0.8973 21.9000 1.9225 47.0000 1.9756 47.9000 249 Thermister No. 12 Millivolts 250 Thermister Number 13 Calibration Data Linear Regression Fit T=25.2281(CV)-0.9007 where T=Temperature in deg. C CV=cjc corrected millivolts slope = 25.2281 intercept = -0.9007 C J C m V Temp C CJC mV Temp C 0.0874 1.0500 0.9034 22.2000 0.1513 2.2500 0.9104 23.2000 0.1391 2.7000 0.9313 23.3000 0.1820 3.5000 0.9521 23.7000 0.1834 3.6000 0.9732 24.2000 0.2296 4.8000 1.0023 24.9500 0.2506 5.1000 1.0326 25.5000 0.2866 6.1000 1.0393 25.9000 0.3116 6.7000 1.0778 26.4000 0.3600 8.0500 1.1197 27.5000 0.3629 8.1000 1.1392 27.9000 0.3887 8.7000 1.1653 28.5000 0.4198 9.5000 1.1938 29.4000 0.4314 9.8000 1.2438 30.5000 0.4550 10.5000 1.2737 31.2000 0.4721 11.1000 1.2896 31.8000 0.5089 11.8000 1.3097 32.5000 0.5171 12.1000 1.3512 33.4000 0.5330 12.6500 1.4003 34.4000 0.5502 12.9000 1.4245 35.0000 0.5852 13.8000 1.4428 35.6000 0.6176 14.7000 1.4790 36.4000 0.6424 15.2000 1.5305 37.4000 0.6648 15.7500 1.5392 38.0000 0.6957 16.6000 1.5618 38.5000 0.7185 17.3000 1.6066 39.4000 0.7482 17.9000 1.6434 40.3000 0.7655 18.4000 1.6682 41.3000 0.8050 19.2500 1.7073 42.3000 0.8288 19.9000 1.7596 43.4000 0.8497 20.5000 1.8041 44.3000 0.8574 20.8500 1.8375 45.3000 0.8705 21.2500 1.8757 46.0000 0.8937 21.9000 1.9225 47.0000 1.9678 47.9000 251 252 Thermister Number 14 Calibration Data Linear Regression Fit T=25.2156(CV)-0.9283 where T=temperature in Deg. C CV=cjc corrected millivolts slope = 25.2156 intercept =-0.9283 CJCmV Temp C CJC mV Temp C 0.0911 1.0500 0.9095 22.2000 0.1537 2.2500 0.9128 23.2000 0.1488 2.7000 0.9264 23.3000 0.1845 3.5000 0.9484 23.7000 0.1847 3.6000 0.9720 24.2000 0.2333 4.8000 1.0011 24.9500 0.2481 5.1000 1.0351 25.5000 0.2902 6.1000 1.0430 25.9000 0.3092 6.7000 1.0790 26.4000 0.3625 8.0500 1.1221 27.5000 0.3653 8.1000 1.1405 27.9000 0.3924 8.7000 1.1665 28.5000 0.4211 9.5000 1.1974 29.4000 0.4290 9.8000 1.2475 30.5000 0.4550 10.5000 1.2762 31.2000 0.4819 11.1000 1.2847 31.8000 0.5053 11.8000 1.3097 32.5000 0.5122 12.1000 1.3488 33.4000 0.5317 12.6500 1.4027 34.4000 0.5539 12.9000 1.4269 35.0000 0.5901 13.8000 1.4452 35.6000 0.6200 14.7000 1.4741 36.4000 0.6412 15.2000 1.5341 37.4000 0.6599 15.7500 1.5441 38.0000 0.6981 16.6000 1.5642 38.5000 0.7258 17.3000 1.6090 39.4000 0.7433 17.9000 1.6458 40.3000 0.7606 18.4000 1.6694 41.3000 0.8086 19.2500 1.7122 42.3000 0.8324 19.9000 1.7620 43.4000 0.8473 20.5000 1.8005 44.3000 0.8648 20.8500 1.8436 45.3000 0.8754 21.2500 1.8770 46.0000 0.8949 21.9000 1.9274 47.0000 1.9756 47.9000 253 Millivolts 254 Thermister Number 15 Calibration Data Linear Regression Fit T=25.2090(CV)-1.0000 where T=temperature in deg. C CV=cjc corrected millivolts slope = 25.2090 intercept = -1.0000 C J C m V Temp C 0.0972 1.0500 C J C m V Temp C 0.1525 2.2500 0.9169 22.2000 0.1525 2.7000 0.9153 23.2000 0.1857 3.5000 0.9325 23.3000 0.1859 3.6000 0.9533 23.7000 0.2369 4.8000 0.9720 24.2000 0.2481 5.1000 1.0109 24.9500 0.2963 6.1000 1.0302 25.5000 0.3165 6.7000 1.0491 25.9000 0.3661 8.0500 1.0778 26.4000 0.3678 8.1000 1.1282 27.5000 0.3924 8.7000 1.1478 27.9000 0.4198 9.5000 1.1738 28.5000 0.4351 9.8000 1.2023 29.4000 0.4611 10.5000 1.2475 30.5000 0.4819 11.1000 1.2786 31.2000 0.5138 11.8000 1.2896 31.8000 0.5183 12.1000 1.3097 32.5000 0.5378 12.6500 1.3537 33.4000 0.5527 12.9000 1.4112 34.4000 0.5938 13.8000 1.4281 35.0000 0.6151 14.7000 1.4440 35.6000 0.6473 15.2000 1.4851 36.4000 0.6611 15.7500 1.5317 37.4000 0.7030 16.6000 1.5465 38.0000 0.7270 17.3000 1.5716 38.5000 0.7531 17.9000 1.6127 39.4000 0.7630 18.4000 1.6507 40.3000 0.8037 19.2500 1.6816 41.3000 0.8288 19.9000 1.7158 42.3000 0.8485 20.5000 1.7608 43.4000 0.8623 20.8500 1.8029 44.3000 0.8778 21.2500 1.8424 45.3000 0.8998 21.9000 1.8818 46.0000 1.9298 47.0000 1.9756 47.9000 255 Thermister No. 15 0 0.5 1 1.5 2 Millivolts 256 APPENDIX D Chemical Analytical Procedures Analytical Method for Chloroguaiacols 1. Typical test volume was 25 ml. This was transferred to a 50 ml Erlenmeyer flask and made up to 50 ml using distilled water. 2. The pH was adjusted to 7 using NaOH. 3. The sample was then transferred to a 125 ml separatory funnel. 4. A surrogate (10 pi of 100 ppm 2,4,6-tribromophenol) was added to all samples including the blank. Surrogate was made up in methanol. 5. A quantity of 5 M K2CO3 solution was added as a buffer such that the final concentration was 0.1 M . 6. The samples were acetylated by adding 1.2 ml of acetic anhydride then shaken vigorously for 2 minutes, venting frequently. The samples were then let to stand for 15 minutes. 7. Samples were extracted with 10 ml of hexane by shaking for 3 minutes. 8. The samples, blanks, and standards were then concentrated to approximately 5 ml using the Roto vac. 9. Approximately 2 ml of each were then transferred to the gas chromatograph vials and an internal standard added (100 ul of lppm 2,6-dibromophenol). This internal standard was made up in hexane. Temperature Program for Chlorinated Guaiacols 75°C - 3 minutes/15°C/min to 120°C/120°C - 0.1 min/3°C/min to 200°C/°- 5 min/5°C/min to 220°C/220°C - 0.1 min/20°C/min to 265°C GC Setup for Chlorinated Guaiacols DB-5 Column - 30 m and 0.32 mm internal diameter Electron capture detector - 310°C Injection port temperature - 250°C Helium carrier gas - 30-35 cm/s Nitrogen detector makeup gas - 60-70 ml/min 257 Analytical Method for Dehyroabietic Acid Extraction 1. The pH was adjusted to approximately 8 or 9 with dilute K O H which was previously extracted. 2. 25 ml of the sample was transferred to a 250 ml separatory funnel. Sufficient amount of the surrogate (heneicosanoic acid) was added such that the final concentration was similar to that of the DHA. In most cases this was 500 pi of the 50 ppm surrogate solution. 3. 25 ml of methyl-t-butyl ether (MTBE) was added and the samples shaken for 2 minutes. 4. After the layers settled, the bottom aqueous layer was transferred into a 100 ml flask and the M T B E extract was collected into a 250 ml round bottom flask. 5. Extraction steps 3 through 4 were repeated with an additional portion of M T B E . Any water from the bottom of the round flask was removed using a pipette. 6. The extract was concentrated using the rotary evaporator to approximately 5 ml. The extract was then transferred to graduated centrifuge vials, rinsing with di-ethyl ether. 7. The extract was further concentrated to 0.3 ml under a gentle stream of N 2 . Then 1 ml of di-ethyl ether and 0.1 ml methanol was added. Derivatization 1. 16 ml of di-ethyl ether and 4 ml of 2 -(2-ethosyethoxy)ethanol (carbitol) was added to the glass generator tube along with 0.4 grams of diazald. 2. Behind the explosion shield, 4 ml of K O H (600 g/1) was added. 3. The methylation aparatus was then attached, turning on the N 2 such that ther is a gentle bubbling in the diazomethane generating solution. 4. At the sample end, a clean Pasteur pipette was attached, allowing the diazomethane to bubble through the sample until a pale yellow colour persisted. The sample was adequately derivatized when this colour persisted for 5 minutes. On charge of diazald should have lasted 1- to 14 samples. 5. The generating solution was neutralized with a slow addition of glacial acetic acid. 6. The derivatized samples were then concentrated down to 0.3 ml and made up to 1 ml with isooctane. They were then transferred to the GC vials and the internal standard 258 (oleic acid methyl ester) added such that the final concentration is similar to that of the DHA, approximately 100 pi of 500 ppm solution. Temperature Program for Dehydroabietic Acid 45°C - 2 min/15°C/min tol90°C/190°C - 1 min/l°C/min to 215°C/215°C - 0.5 min/15°C/min to 290°C/290°C - 5 min. GC Setup for Dehydroabietic Acid DB-1 Column - 30 m and 0.32 mm internal diameter Flame ionization detector - 290°C Injection port temperature - 275°C Helium carrier gas - 20 cm/s Helium (20 ml/min), hydrogen (30ml/rnin) and air (400 ml/min) detector makeup gas 259 APPENDIX E Details of Cross-section Video Analysis Cross sectional video images were obtained to provide qualitative information on the onset of bifurcation at a given distance downstream. The images were recorded by a video camera directed upstream, focussing oh a section of the flow which was illuminated, by a vertical light sheet. Figure E . l shows details of this configuration. The camera.Was! set up at a fixed location , 96 cm from the light sheet at an angle of 62 degrees from the vertical. The light sheet was 32 cm from the jet exit. 1 As the camera looked obliquely at the flow cross section, the images produced were skewed vertically. This was due to both light refraction at the water surface and the width of the light sheet (1 mm). This did not affect the ability to detect whether bifurcation was occurring or not. Light Sheet 1 a = 62° ^ ^ 96 cm 32 cm Figure E - l Setup for Cross Sectional Videos 260 APPENDIX F Experimental Data - Mixing Experiments Temperature and trajectory data are given in the following section. For trajectory, z c is measured from the top of the jet exit. Temperature data is given in terms of the maximum centreline excess temperature (Tm-Ta). The lateral temperature measurements (Experiments 16 through 19) are given in actual flow temperature. A complete summary of the experimental conditions for each run is given in Table F - l . 261 e ' 1 \3 B C v o C N ON wo SO oo o ON p oo C N C N p —< ro —i C N © ON T t C N ro —• Fo ©' wo C N C N C N C N C N o C N C N SO C N ro WO wo C O C N ro so cn ON od vd C N ON T t C O ON O O ON T t C N 00° O T t T t o O T t o T t o T t O T t o T t o o T J -O T t p T t o T t o T t o T t o T t o T t o T t o T t O T t O T t o T t O WO © T t © T t © T t © T t © T t © Xi N s pq co pq ON 0 0 w ro pq 0 0 ro pq C O © w ON p W WO W 0 0 so W wo ON W T t o W 0 0 W 0 0 o-w wo W wo SO W so C N W ON WO W C N ro pq WO ON pq o oo W o -—i W 0 0 C N W ON oo pq ON 0 0 pq © pq VO C N w oo © pq f - H SO ro vd C N wo t~-' co T t wo C N CM' C O T t ro vd C O W 0 T t ~* f-' ON ro T t wo E N s ro O ro ro o T t ro o W0 C N o wo o o ro C N o wo ro o ON C N o C O o ro T t O C N o 0 0 o 0 0 cs o ro o C O T t o oo o C N ro O r -C N O oo cn o O O C N o 0 0 C N o O O C N © wo C O o ro © wo © oo © wo C N © ©' o ©' ©' ©' o o o O o o o o o o o O O o O o ©' ©' ©' ©' ©' © E 0 0 o ro VO wo C N oo wo C N ro o C O wo ON ro C N WO C N o C N ro ON SO cn C -0 0 0 0 oo C N o ro i~H C O T t o ro 0 0 C N C N C N C N C N ro cn r--C N C N o T t © VO © VO ro C N ON C N 0 0 ON © C N C N 0 0 SO f - H ©' ©' o ©' o o o o O o o o o O o o o O O o o ©' ©' ©' ©' ©' E o O o o o o o o o o o o o o o o o o o o o © © © © © © ©' ©' ©' ©' ©' o o o o o o O ' o o o o o o o o o ©' © ©' ©' ©' © VO © .50E-05 wo o wo o SO o VO o wo o wo o wo p wo p VO o SO o wo o wo o wo o SD O VO O so p wo p W0 p VO p © wo © wo © so © SO © so © PQ C/3 -* a w f- .50E-05 pq w T t f - H w ON T t w o T t w ro w ro w C N wo W SO W o VO W C N WO w C N ro w W0 W T t o W C N C N pq 0 0 w ro so pq T t pq o o pq C N pq © VO pq wo pq T t pq T t pq C O ON W W0 wo *-< '-' r~-' rt 0 0 0 0 —' C N WO ON ON —' ro —' T t WO od CM-CO ON O ON T t C N ON 0 0 so ro t > wo T t T t O WO o O ro wo ro © WO © © ON ON ON ON ON ©' ©' ©' ©' O o O o o o o o o o o O o O O o O ©' ©' ©' ©' ©' ©' wo O © W0 © wo © SO O W0 o W0 o wo o T t p T t O wo o wo p o T t o T t o W0 o wo p wo p T t p wo p wo p wo © wo © T t © W0 © wo © WO © s c/a a pq o 0 0 pq r>-C N W W0 ro w WO ro w SO C N W o W ro T t W so ON w T t ro w ON r> W 0 0 w 0 0 T t W p W ON W T t T t W ON ON W ON O W ON O pq T t C N W T t T t pq T t T t pq T t T t pq ro T t pq C N pq C N pq ON ON pq SJD rt ~* 0 0 od —1 C O —1 T t T t C N —' SO VO —' vd vd VO C - ' —' —' ro CA a wo O • W WO .26E-04 T t o T t o wo p wo p wo p T t p T t o T t o wo o wo o T t o T t O T t O wo o wo o wo o T t o wo o wo © wo © wo o T t © W0 © wo © wo © a .26E-04 pq fN O pq C N © PJ SO < N W wo C N W so so W o o W o ro W o WO W o wo w o wo pq C N w o wo w wo p i o o pq wo r>-W wo r -W wo C N o o pq © © pq © © pq VO VO pq OS pq © © pq © © 0 0 T t —' —' —' vd ON —' ~ t>-' —' —' wo od od -•' ON ON' ON' ON T t wo' SO VD ro T t <N W 0 0 0 C N 0 0 o 0 0 C O 0 0 wo VO ro so T t ro wo t>-< N ON ON C N ON ro wo T t 0 0 C N o wo wo ro O WO vO C O VO ro VO cn ON T t © SO © C N wo C N wo ro ©' O ©' ©' o o o O o o o o O o o o o O o o ©' ©' ©' ©' ©' ©' ©' C N r~-ON W0 © C N T t T t o wo T t wo o ro 0 0 ON T t vq so T t oo C N ro oo ro SO W0 0 0 o T t ro T t T t T t r~-T t W 0 r~-oo © C O ro so T t CM rt C N ro C N o C N ro C N C O ro' rt C N co ro »—( C N C N ro C N C N C N ro' C O CM P/'H ( N Os C N ON ON 0 0 C N T t C N T t C N T t VO so VO SO o wo o wo o wo o wo T t T t C N ON C N ON C N ON C N ON C N ON wo T t W0 T t W0 T t wo T t wo T t P/'H SO VO wo 0 0 wo wo wo t>- r '^ r--' od od od 0 0 W0 wo vd vd vd SD VO SO vd vd SO SO B o O O o- wo — ro cn ro ro o o o o wo wo O O o © © T t T t T t T t T t B ©' ©' oo od 0 0 od O o o O C N < N C N C N od 0 0 O O o" ©' © ' ON ON ON ON cK t o g wo ON NO O SO ro wo C N r - VO so so 0 0 VO o so ro wo so T t o T t 0 0 ro 0 0 ro wo ro wo ON T t ON VO ON VO ON C N O C N © C N © 0 0 0 0 r--0 0 » • O O t > -oo a 1 0 0 ON ON ON ON ON 0 0 ON ON 0 0 ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON ON 00° ON ON od ON ON od ON ON 00° ON ON ON ON ON ON ON ON ON' ON ON od ON ON od ON ON od ON ON od ON ON 0 0 ON OS H C J 6 0 •o T t ON 0 0 C N ro SO wo ro T t ro ro 0 0 T t C N C N C O O -—i cn T t 0 0 O SO ON C N wo ON SO O O ON C O ON O WO © WO © wo © © © © © © © © © © wo wo r>-' od ON o o O oi C N ro C N cn wo so' wo WO wo wo WO t - ' f-' r>-' CN g t~-0 0 ro ON r~-o so C N SO o so ON C N C N ON SO 0 0 wo 0 0 ro 0 0 C N T t O W0 wo T t C O ro wo C N wo f -ro T t r> © wo © wo © © © © © a -X r--' 0 0 ON vd 0 0 ON 0 0 ON 0 0 ON r~-' 0 0 ON r~' 0 0 ON 0 0 ON od oo ON t>-' 0 0 ON f-' 0 0 ON r-~ 0 0 ON f-' 0 0 ON 0 0 ON 0 0 ON r '^ 0 0 ON 0 0 0 0 ON 0 0 0 0 ON r-' oo ON r-' 0 0 ON f-' 0 0 ON wo" ON ON 0 0 ON ON SO 0 0 ON vd 0 0 ON vd oo ON vd 0 0 ON SO 0 0 OS H C J ao T t T t © wo O CM O O O O ro o VO VO VO C O 0 0 ON O ro SO C N wo ro T t VO C N ro -—i T t 0 0 wo ON VO O TT O 0 0 © wo © wo © © © © © © © © © © ©' W0 C N wo C N wo C N WO wo wo O WO ON T t o wo wo o wo O W0 wo WO wo ON T t ON T t O WO WO o wo o C O od ro wo ro wo ro' wo ro wo ro wo 1 O O T t O © WO O O WO 0 0 o o ON T t ON SD VO ON T t C O o ro ON r~ ON WO ON WO wo ON 0 0 ro ON C N ON ro O O ON ro SO ON so VO ON VO T t ON ON vo f -VO vo ON VO t> wo T t ON 0 0 ro 0 0 ON ro f -T t wo ©' ©' ©' o O o O -—i o o o o O O O o © © ©' ©' ©' ©' ©' ~g 0 0 ro ro 0 0 ro ro SO C N ro SO ro T t T t C N T t T t C N T t T t C N C O C O ro ro ro ro ro ro ro so ro ro vo ro cn so ro ro SO ro T t T * C N T t T t C N O ON C N O ON C N o ON C N © ON C N © ON C N T t T t C N T t T t C N T t T t C N T t T t C N T t T t Cvl O ©' ©' ©' o o o o O O O O O O o o o O O o O O ©' ©' ©' © ©' C N ro wo SO oo ON o C N ro wo SD 0 0 ON o C N C N C N C N cn C N T t C N WO C N SO C N u B B Ito Experiment No. 1 Trajectory Dilution x (cm) Vc(cm) x(cm) AT=Tm-T a (°C) 1 1.5000 0.30 33.84 2 2.1300 1.30 16.81 3 2.4200 2.30 16.98 4 2.6300 3.30 12.88 5 2.6300 4.30 10.62 6 2.8800 5.30 9.15 7 3.0400 6.30 8.08 8 2.7100 7.30 7.65 9 3.0400 8.30 6.99 10 3.0000 9.30 6.29 11 3.0400 10.30 5.94 12 3.1300 11.30 5.72 13 3.2500 12.30 5.56 14 3.2100 13.30 5.18 15 3.2100 14.30 4.85 16 2.9600 15.30 4.65 17 3.0400 16.30 4.52 18 3.4600 17.30 4.58 19 3.5400 18.30 4.27 20 3.1300 19.30 4.20 21 3.2100 20.30 4.13 22 2.9600 25.30 3.68 30.30 3.29 35.30 3.13 40.40 3.70 Experiment No. 2 Trajectory Dilution x (cm) yc(cm) x(cm) AT=T m -T a CC) 1 3.0900 0.40 29.00 2 3.9500 1.40 17.88 3 4.5900 2.40 14.67 4 5.1800 3.40 11.48 5 5.4500 4.40 10.50 6 5.4500 5.40 8.98 7 6.0000 6.40 8.24 8 6.0900 7.40 7.47 9 6.5000 8.40 6.96 10 6.4100 9.40 6.80 11 6.7700 10.40 6.43 12 6.7300 . 11.40 5.97 13 6.8600 12.40 5.68 14 6.6400 13.40 5.36 15 7.1400 14.40 5.17 16 7.3600 15.40 5.14 17 7.6800 16.40 4.82 18 7.6800 17.40 4.64 19 7.7700 18.40 4.48 20 8.1800 19.40 4.72 21 8.1800 20.40 4.81 22 8.0000 25.40 4.37 23 8.3200 30.40 4.36 24 8.2700 Experiment No. 3 Trajectory Dilution x (cm) yc (cm) x(cm) AT=Tm-T a (°C) 1 2.7400 0.40 30.39 2 3.6800 1.40 16.81 3 4.2600 2.40 12.56 4 4.6800 3.40 11.38 5 4.8900 4.40 9.68 6 5.0500 5.40 8.68 7 5.4700 6.40 8.12 8 5.2600 7.40 7.57 9 5.5800 8.40 7.29 10 5.8400 9.40 6.69 11 6.5300 10.40 6.44 12 6.5300 12.40 6.40 13 6.3700 14.40 6.33 14 6.6300 16.40 6.24 15 6.6300 18.40 5.85 16 6.5800 20.40 5.42 17 6.5800 22.40 5.40 18 6.5300 24.40 5.07 19 6.5300 26.40 5.00 20 6.4700 28.40 4.90 21 6.5300 30.40 4.48 22 6.5300 35.40 4.14 23 6.5300 40.40 3.70 24 6.5300 25 6.5300 Experiment No. 4 Trajectory Dilution x (cm) Vc(cm) x(cm) AT=Tm-T a (°C) 1 2.1200 0.40 33.17 2 2.7600 1.40 21.33 3 3.3500 2.40 16.54 4 3.5900 3.40 12.66 5 3.7100 4.40 10.96 6 3.9400 5.40 9.33 7 4.1800 6.40 8.57 8 4.1800 7.40 7.63 9 4.5300 8.40 7.17 10 4.5300 9.40 6.66 11 4.4100 10.40 6.26 12 4.5300 12.40 5.55 13 4.2900 14.40 5.06 14 4.7600 16.40 4.57 15 5.0000 18.40 4.33 16 4.7600 20.40 4.04 17 4.4100 22.40 3.77 18 4.8800 24.40 3.47 19 5.0000 26.40 3.59 20 4.8800 28.40 3.14 21 4.8800 30.40 3.07 22 4.6500 35.40 2.83 23 4.6500 40.40 2.63 24 5.0000 25 4.8800 26 5.4100 27 5.6500 28 5.3500 29 5.4100 30 5.3500 31 5.2400 Experiment No. 5 Trajectory Dilution x (cm) Vc(cm) x(cm) AT=Tm-T a (°C) 1 0.8890 0.50 34.65 2 1.0600 1.00 34.31 3 1.1700 2.00 24.15 4 1.2200 3.00 17.17 5 1.3300 4.00 15.42 6 0.0000 5.00 11.67 7 0.0000 7.00 8.43 8 0.0000 9.00 6.39 9 1.0000 11.00 5.48 10 2.1700 13.00 4.56 11 1.8900 15.00 4.06 12 2.0600 17.00 3.58 13 1.6700 22.00 2.78 14 0.8890 27.00 2.32 15 1.9400 32.00 2.14 16 2.4400 37.00 1.81 17 2.3300 42.00 1.56 18 3.0600 19 2.0000 20 2.3900 21 2.6100 22 0.1110 23 2.3300 24 2.0000 25 2.0000 26 2.3900 27 1.5000 28 0.1110 29 2.2200 30 0.1670 -31 0.0000 Experiment No. 6 Trajectory Dilution x (cm) yc(cm) x(cm) AT=Tm-T a (°C) 1 2.6700 0.50 39.42 2 3.4400 1.00 30.22 3 3.7800 2.00 21.62 4 4.3900 3.00 16.23 5 4.5600 4.00 13.55 6 4.7800 5.00 11.45 7 4.8900 7.00 9.25 8 5.0000 9.00 7.87 9 5.1700 11.00 6.78 10 5.2800 13.00 5.96 11 5.3900 15.00 5.70 12 5.4400 17.00 5.96 13 5.5000 22.00 5.23 14 5.5000 27.00 4.88 15 5.5000 32.00 4.65 16 5.5000 37.00 4.11 17 5.6100 42.00 3.64 18 5.6100 19 5.6100 20 5.6100 21 5.6700 22 5.6100 23 5.6100 24 5.6100 25 5.6700 26 5.7200 27 5.6700 28 5.6700 29 5.7200 30 5.7200 31 5.7200 Experiment No. 7 Trajectory Dilution x (cm) yc(cm) x(cm) AT=Tm-T a (°C) 1 3.8900 0.50 37.26 2 4.5800 1.00 29.01 3 5.1600 2.00 20.72 4 5.5800 3.00 15.27 5 5.7900 4.00 13.62 6 5.8400 5.00 11.23 7 5.8900 7.00 9.98 8 5.9500 8.00 10.81 9 5.9500 9.00 10.51 10 5.9500 11.00 9.36 11 5.9500 13.00 8.52 12 5.9500 15.00 7.82 13 5.9500 17.00 6.96 14 5.8900 22.00 5.75 15 5.9500 27.00 4.43 16 5.9500 32.00 4.04 17 6.0000 37,00 3.12 18 5.9500 42.00 2.86 19 5.8900 20 5.9500 21 5.9500 22 5.8900 23 6.0000 24 6.0000 25 6.0000 26 6.0000 27 6.0000 28 6.0000 29 6.0000 Experiment No. 8 Trajectory Dilution x (cm) Vc(cm) x(cm) AT=T m -T a (°C) 1 2.7059 0.50 39.10 2 3.5294 1.00 34.07 3 4.0588 2.00 24.68 4 4.4706 3.00 16.81 5 4.7059 4.00 14.17 6 5.0000 5.00 11.56 7 5.0588 7.00 9.50 8 5.2941 9.00 8.10 9 5.2353 11.00 7.25 10 5.5882 13.00 6.56 11 5.7647 15.00 6.12 12 5.7059 17.00 5.64 13 5.9412 22.00 4.87 14 5.7059 27.00 4.34 15 5.9412 32.00 3.93 16 5.9412 37.00 3.67 17 6.0000 42.00 3.56 18 6.1176 19 5.9412 20 6.1176 21 6.4706 22 6.7647 23 6.8824 24 6.8824 25 6.5882 26 6.8824 27 6.4706 28 6.3529 29 8.5294 30 8.5294 31 8.4706 32 7.2353 33 7.3529 Experiment No. 9 Trajectory Dilution x (cm) yc(cm) x(cm) AT=Tm-T a (°C) 1 3.1176 0.50 37.22 2 4.2353 1.00 35.50 3 4.7059 2.00 21.01 4 5.1765 3.00 15.47 5 5.5882 4.00 13.39 6 5.7059 5.00 11.19 7 5.9412 7.00 8.59 8 6.1176 9.00 7.39 9 6.5882 11.00 6.65 10 6.4706 13.00 6.10 11 6.7647 15.00 5.59 12 6.7059 17.00 5.22 13 7.7647 22.00 4.66 14 8.1765 27.00 4.42 15 8.1765 32.00 4.27 16 8.2941 37.00 4.08 17 7.8824 42.00 3.56 18 7.8235 19 7.7647 20 7.7647 21 8.2941 22 8.2941 23 8.2941 24 8.2941 25 8.2941 26 8.4118 27 8.2941 28 8.3529 29 8.2941 30 8.4118 31 8.2941 32 7.8235 33 7.6471 Experiment No. 10 Trajectory Dilution x (cm) y c(cm) x(cm) AT=T m -T a , (°C) 1 3.9412 0.50 34.73 2 4.8824 1.00 26.96 3 5.6471 2.00 20.14 4 6.1765 3.00 14.98 5 6.6471 4.00 12.35 6 7.0588 5.00 10.54 7 7.5294 7.00 8.43 8 7.7647 9.00 7.37 9 7.6471 11.00 6.95 10 7.7647 13.00 7.06 11 7.7647 15.00 6.63 12 7.8824 17.00 6.38 13 7.8824 22.00 5.73 14 7.7647 27.00 4.57 15 7.8824 32.00 4.22 16 7.8824 37.00 4.00 17 7.7647 42.00 3.69 18 7.8824 19 7.7647 20 7.8824 21 7.8824 22 7.7647 23 7.8824 24 7.8824 25 7.7647 26 7.8824 27 7.7647 28 7.8824 29 8.1176 30 7.8824 31 8.0000 32 7.6471 33 7.6471 272 Experiment No. 11 Trajectory Dilution x (cm) y c(cm) x(cm) AT=T m -T a (°C) 1 2.1176 0.50 34.73 2 2.7647 1.00 26.96 3 3.2353 2.00 20.14 4 3.5294 3.00 14.98 5 3.8235 4.00 12.35 6 3.9412 5.00 10.54 7 4.1765 7.00 8.43 8 4.1765 9.00 7.37 9 4.1765 11.00 6.95 10 4.1765 13.00 7.06 11 4.4706 15.00 6.63 12 4.4118 17.00 6.38 13 4.8824 22.00 5.73 14 5.0588 27.00 4.57 15 5.0588 32.00 4.22 16 4.7647 37.00 4.00 17 4.7059 42.00 3.69 18 5.5882 19 5.4118 20 5.4706 21 5.2941 22 5.5882 23 5.1765 24 5.1765 25 5.4118 26 5.2941 27 5.5882 28 5.2941 29 5.8235 30 5.8235 31 6.2941 32 6.6471 33 6.5882 273 Experiment No. 12 Trajectory Dilution x (cm) yc(cm) x(cm) AT=Tm-T a (°C) 1 1.8800 0.50 36.90 2 2.4700 1.00 34.22 3 2.7600 2.00 26.17 4 3.1200 3.00 17.57 5 3.2400 4.00 15.50 6 3.3500 5.00 11.89 7 3.6500 7.00 9.38 8 3.5300 9.00 7.45 9 3.6500 11.00 6.48 10 3.8800 13.00 5.81 11 3.8800 15.00 5.17 12 3.7600 17.00 4.60 13 3.7100 22.00 3.84 14 3.6500 27.00 3.28 15 4.0000 32.00 2.81 16 4.3500 37.00 2.48 17 4.0000 42.00 2.32 18 3.9400 19 3.8800 20 3.8800 21 4.0000 22 4.2400 23 4.6500 24 5.1200 25 4.2400 26 4.0000 27 4.1200 28 4.1200 29 4.2900 30 4.1200 31 4.2400 32 4.0000 33 5.1200 274 Experiment No. 13 Trajectory Dilution x (cm) yc(cm) x(cm) AT=Tm-T a CC) 1 2.4100 0.50 38.64 2 3.2900 1.00 33.23 3 3.7100 2.00 26.11 4 3.9400 3.00 16.96 5 4.2900 4.00 14.58 6 4.5300 5.00 12.01 7 4.5900 7.00 9.33 8 4.8200 9.00 7.71 9 4.7100 11.00 6.94 10 4.9400 13.00 6.08 11 5.2900 15.00 5.46 12 5.2900 17.00 4.83 13 4.9400 22.00 4.12 14 5.2400 27.00 3.52 15 5.4100 32.00 3.10 16 5.5900 37.00 2.83 17 5.4700 42.00 2.59 18 5.5300 19 5.3500 20 5.5300 21 5.4700 22 5.9400 23 5.7100 24 5.8200 25 5.5900 26 5.8200 27 5.5900 28 5.5900 29 5.8200 30 5.8200 31 . 6.3500 32 6.4700 Experiment No. 14 Trajectory Dilution x (cm) y c(cm) x(cm) AT=T m -T a (°C) 1 3.2900 0.50 35.59 2 4.2400 1.00 28.86 3 4.8800 2.00 20.45 4 5.2400 3.00 15.43 5 5.8200 4.00 12.97 6 5.8200 5.00 10.93 7 6.1200 7.00 8.76 8 6.2900 9.00 7.50 9 6.5300 11.00 6.52 10 6.4100 13.00 5.96 11 6.7600 15.00 5.53 12 6.7600 17.00 5.15 13 7.2400 22.00 4.47 14 7.3500 27.00 4.05 15 7.0000 32.00 3.46 16 7.4700 37.00 3.28 17 7.3500 42.00 2.97 18 7.3500 19 7.8200 20 7.9400 21 7.8800 22 7.7100 23 7.7100 24 7.8800 25 8.0600 26 8.0000 27 8.4100 28 9.0600 29 9.1800 30 9.0600 31 9.7600 32 9.7100 276 Experiment No. 15 Trajectory Dilution x (cm) Vc(cm) x(cm) AT=Tm-T a (°C) 1 4.1200 1.00 36.62 2 5.2400 2.00 16.99 3 5.9400 3.00 14.49 4 6.5300 4.00 11.84 5 6.9400 5.00 10.25 6 7.1800 7.00 8.08 7 7.4100 9.00 7.13 8 7.6500 11.00 6.42 9 7.9400 13.00 5.87 10 8.0600 15.00 5.45 11 8.4700 17.00 5.13 12 8.7600 22.00 4.74 13 8.9400 27.00 4.41 14 9.0600 32.00 4.03 15 9.1800 37.00 3.64 16 9.0600 42.00 3.43 17 9.0600 18 9.5900 19 9.5300 20 9.6500 21 9.1800 22 9.7100 23 9.7100 24 9.6500 25 9.7100 26 9.6500 27 9.7100 28 9.7600 29 9.6500 30 9.6500 31 9.7600 32 9.7100 277 Experiment No. 16 Trajectory x (cm) Vc(cm) 1 2.0600 2 2.7600 3 3.1200 4 3.4700 5 3.7100 6 3.8800 7 3.9400 8 4.0600 9 4.1200 10 4.2400 11 4.1800 12 4.2900 13 4.4700 14 4.7100 15 4.6500 16 4.5900 17 4.4100 18 4.5300 19 4.7100 20 4.5900 21 5.0000 22 5.0000 23 5.0600 24 5.0600 25 5.0000 26 6.5900 27 5.2900 28 6.1200 29 6.5300 30 6.5900 31 6.6500 32 6.7100 33 6.1200 Experiment No. 16 Lateral Temperature Data (°C) x=5 cm y (cm) 12 9 6 3 0 3 6 9 12 z(cm) 8.0 15.93 15.82 15.91 15.83 16.06 16.13 15.72 16.08 15.92 6.2 15.93 15.80 15.93 15.79 20.93 16.09 15.67 16.04 15.90 4.4 15.98 15.80 15.96 16.11 23.88 16.42 15.69 16.04 15.91 2.8 15.95 15.81 15.94 15.86 18.98 16.36 15.68 16.03 15.88 x=10 8.0 15.98 15.84 15.95 15.76 16.35 16.09 15.58 15.95 15.84 6.2 15.98 15.84 15.97 16.53 21.53 16.62 15.63 16.00 15.86 4.4 15.96 15.86 15.95 17.46 21.03 17.57 15.66 16.02 15.89 2.8 15.95 15.85 15.96 16.26 17.87 16.58 15.68 16.04 15.90 x=20 8.0 15.99 15.86 15.96 16.69 18.10 16.67 15.56 15.96 15.86 6.2 16.01 15.87 16.00 18.06 19.43 17.46 15.54 15.93 15.82 4.4 16.03 15.89 16.03 17.73 18.87 17.87 15.53 15.96 15.83 2.8 16.04 15.91 16.01 16.14 16.92 16.67 15.54 15.93 15.81 x=30 8.0 16.07 15.90 16.18 17.78 18.65 17.55 15.45 15.83 15.78 6.2 16.04 15.88 16.14 17.47 18.71 17.63 15.50 15.85 15.78 4.4 16.07 15.91 16.07 17.03 17.83 17.35 15.61 15.93 15.81 2.8 16.03 15.91 16.03 16.26 16.68 16.55 15.54 15.95 15.81 x=40 8.0 16.08 15.96 17.09 18.23 18.61 18.06 15.95 15.87 15.77 6.2 16.06 15.89 16.20 17.44 18.15 17.51 15.64 15.82 15.75 4.4 16.08 15.89 16.12 16.85 17.42 17.04 15.54 15.80 15.71 2.8 16.06 15.86 15.94 15.90 16.32 16.41 15.36 15.72 15.66 Experiment No. 17 Trajectory x (cm) yc(cm) 1 2.9400 2 3.8200 3 4.4100 4 4.7100 5 5.0000 6 5.2900 7 5.4700 8 5.6500 9 6.1200 10 6.2400 11 6.2400 12 6.2900 13 6.2400 14 6.2400 15 6.2400 16 6.2400 17 6.5300 18 6.2400 19 6.2900 20 6.3500 21 6.3500 22 6.2900 23 6.2900 24 6.3500 25 6.2900 26 6.2400 27 6.2400 28 6.2400 29 6.2400 30 6.2400 31 6.5900 32 6.6500 33 6.1200 Experiment No. 17 Lateral Temperature Data (°C) x=5 cm y (cm) 12 9 6 3 0 3 6 9 12 z(cm) 8.0 16.26 16.06 16.17 15.84 18.92 16.11 15.85 16.19 16.06 6.2 16.23 16.00 16.15 16.83 24.81 17.01 15.87 16.24 16.13 4.4 16.24 16.03 16.14 16.64 21.83 17.48 15.80 16.15 16.08 2.8 16.21 15.99 16.11 15.88 17.62 16.26 15.74 16.15 15.99 x=10 8.0 16.20 15.99 16.08 18.30 21.44 18.12 15.78 16.16 16.04 6.2 16.25 16.00 16.09 18.65 21.48 18.90 15.70 16.03 15.96 4.4 16.24 16.03 16.13 17.16 18.66 17.51 15.73 16.07 16.01 2.8 16.26 16.03 16.13 15.88 16.44 16.40 15.75 16.08 16.02 x=20 8.0 16.22 16.01 17.90 20.09 20.66 20.46 17.01 16.07 15.98 6.2 16.20 15.92 16.54 19.35 19.11 18.60 15.74 16.07 15.94 4.4 16.25 16.01 16.27 17.06 17.35 17.10 15.65 15.99 15.86 2.8 16.25 16.01 16.13 15.83 16.22 16.29 15.59 15.98 15.89 x=30 8.0 16.19 16.73 19.21 19.24 19.40 19.62 18.17 16.21 15.77 6.2 16.20 15.97 17.31 18.45 18.04 18.59 16.16 15.92 15.80 4.4 16.20 15.95 16.42 16.67 16.82 17.03 15.87 15.93 15.81 2.8 16.21 15.96 16.13 15.91 16.21 16.24 15.64 15.99 15.86 x=40 8.0 16.43 17.62 19.11 18.62 18.70 19.11 18.60 17.33 16.12 6.2 16.22 16.09 17.78 17.67 17.48 18.00 16.66 15.88 15.74 4.4 16.20 15.94 16.39 16.46 16.49 16.75 15.78 15.80 15.74 2.8 16.22 15.92 16.03 15.61 15.90 16.05 15.45 15.77 15.70 Experiment No. 18 Trajectory x (cm) yc(cm) 1 2.5300 2 3.2400 3 3.7100 4 4.1200 5 4.2400 6 4.5900 7 4.6500 8 4.9400 9 5.1200 10 5.3500 11 5.3500 12 5.2900 13 5.4700 14 5.3500 15 5.7100 16 5.9400 17 5.3500 18 5.3500 19 5.5900 20 5.7100 21 5.8200 22 6.0000 23 . 5.9400 24 6.1200 25 5.8200 26 6.0000 27 6.1200 28 6.2400 29 6.1800 30 6.4100 31 6.7100 Experiment No. 18 Lateral Temperature Data (°C) x=5 cm y (cm) 12 9 6 3 0 3 6 9 12 z(cm) 9.5 16.22 15.97 16.08 15.51 15.67 15.77 15.68 16.07 16.01 7.7 16.20 15.99 16.06 15.49 16.38 15.78 15.66 16.04 15.99 5.9 16.23 15.97 16.09 15.61 25.47 15.99 15.68 16.06 15.99 4.1 16.25 16.00 16.11 16.19 21.79 16.64 15.66 16.06 15.99 2.3 16.26 16.03 16.10 15.58 17.67 15.93 15.69 16.08 16.01 x=10 9.5 16.27 16.00 16.10 15.46 15.67 15.76 15.65 16.03 15.98 7.7 16.26 16.01 16.09 15.64 19.55 15.97 15.66 16.04 15.99 5.9 16.26 16.01 16.07 17.25 22.05 17.31 15.59 16.09 16.01 4.1 16.28 16.03 16.12 16.84 19.60 17.40 15.69 16.09 16.02 2.3 16.27 16.04 16.12 15.79 16.61 16.07 15.70 16.11 15.99 x=20 9.5 16.23 15.99 16.07 15.83 17.14 16.01 15.61 16.03 15.94 7.7 16.23 15.99 16.09 16.86 19.36 17.24 15.61 16.02 15.95 5.9 16.22 15.97 16.11 17.58 19.39 17.92 15.61 15.99 15.92 4.1 16.21 15.96 16.06 16.76 17.47 17.14 15.59 15.95 15.90 2.3 16.24 15.99 16.07 15.76 16.27 16.28 15.52 15.95 15.93 x=30 9.5 16.26 15.96 16.06 16.13 17.43 16.60 15.46 15.80 15.77 7.7 16.30 16.00 16.21 17.07 18.37 17.47 15.60 15.88 15.84 5.9 16.26 15.97 16.36 17.15 18.22 17.39 15.70 15.90 15.85 4.1 16.28 15.99 16.23 16.52 17.13 16.95 15.67 15.95 15.92 2.3 16.25 15.98 16.06 15.78 16.10 16.08 15.56 15.94 15.89 x=40 9.5 16.23 15.94 16.37 16.79 17.79 17.07 15.56 15.81 15.75 7.7 16.27 15.96 16.24 16.68 17.84 17.59 15.70 15.81 15.79 5.9 16.24 15.93 16.42 17.00 17.60 17.12 15.61 15.74 15.72 4.1 16.25 15.93 16.20 16.12 16.66 16.64 15.55 15.72 15.67 2.3 16.24 15.91 15.99 15.45 15.82 15.93 15.30 15.68 15.69 Experiment No. 19 Trajectory x (cm) Vc(cm) 1 3.1800 2 4.0600 3 4.7100 4 5.1200 5 5.4700 6 5.8200 7 5.9400 8 6.1200 9 6.3500 10 6.4100 11 6.7100 12 6.9400 13 6.7100 14 6.7100 15 7.3500 16 8.0000 17 8.1200 18 8.0000 19 8.1200 20 8.1200 21 8.0000 22 8.1800 23 8.1800 24 8.1200 25 8.0600 26 8.0000 27 8.1200 28 8.0600 29 8.1200 30 8.0000 31 8.0000 32 7.5900 2 8 4 Experiment No. 19 Lateral Temperature Data (°C) x=5 cm y (cm) 12 9 6 3 0 3 6 9 12 z(cm) 9.5 16.37 16.11 16.20 15.49 15.99 15.79 > 15.82 16.18 16.13 7.7 16.39 16.11 16.22 15.84 23.97 16.01 15.84 16.20 16.15 5.9 16.40 16.12 16.23 16.73 23.93 17.66 15.86 16.25 16.17 4.1 16.42 16.16 16.25 16.30 19.91 17.16 15.87 16.26 16.20 2.3 16.43 16.14 16.28 15.63 16.52 15.93 15.91 16.25 16.23 x=10 9.5 16.43 16.15 16.27 16.70 19.96 17.13 15.91 16.27 16.22 7.7 16.46 16.13 16.28 18.32 22.09 18.51 15.90 16.27 16.21 5.9 16.43 16.16 16.29 18.41 20.46 18.27 15.92 16.29 16.24 4.1 16.43 16.14 16.28 16.72 17.61 16.92 15.93 16.29 16.23 2.3 16.42 16.13 16.27 15.67 16.06 15.98 15.89 16.25 16.21 x=20 9.5 16.46 16.19 17.24 19.55 20.41 19.17 16.33 16.31 16.23 7.7 16.47 16.20 16.55 18.87 19.79 19.09 16.07 16.32 16.28 5.9 16.51 16.20 16.58 17.61 18.35 18.17 16.14 16.29 16.24 4.1 16.53 16.20 16.43 16.34 16.71 16.79 15.96 16.28 16.22 2.3 16.51 16.22 16.32 15.68 15.92 16.06 15.85 16.27 16.19 x=30 9.5 16.51 16.51 18.88 19.37 19.57 19.61 18.01 16.29 16.08 7.7 16.53 16.20 17.26 18.57 18.67 18.50 16.44 16.13 16.10 5.9 16.55 16.26 16.73 17.09 17.44 17.60 16.30 16.19 16.15 4.1 16.51 16.21 16.49 16.24 16.48 16.61 16.06 16.22 16.17 2.3 16.55 16.23 16.38 15.72 15.95 16.08 15.88 16.27 16.23 x=40 9.5 16.53 17.49 19.30 19.02 19.00 19.21 18.38 16.71 16.10 7.7 16.51 16.27 17.37 17.87 18.30 18.37 17.00 16.22 16.07 5.9 16.48 16.18 16.74 16.82 17.06 17.16 16.08 16.09 15.99 4.1 16.46 16.16 16.37 15.93 16.20 16.47 15.81 16.01 15.96 2.3 16.40 16.12 16.17 15.46 15.75 15.97 15.54 15.90 15.91 Experiment No. 20 Trajectory x (cm) yc(cm) 1 2.5000 2 3.3125 3 3.7500 4 4.1250 5 4.3750 6 4.5625 7 4.8750 8 5.0625 9 5.1250 10 5.1875 11 5.5000 12 5.5000 13 5.5000 14 5.1875 15 5.1875 16 5.2500 17 5.4375 18 5.3125 19 5.6875 20 5.8125 21 6.1250 22 6.0000 23 6.1875 24 5.4375 25 6.0000 26 6.0000 27 5.6875 28 6.5000 29 6.6250 30 6.7500 31 7.0625 32 7.5625 33 8.1250 34 7.9375 35 8.1250 36 8.1250 Experiment No. 21 Trajectory x (cm) yc(cm) . 1 2.5625 2 3.2500 3 3.8125 4 4.1250 5 4.3125 6 4.5625 7 4.6250 8 4.7500 9 4.8750 10 5.0000 11 5.3750 12 5.5000 13 5.3750 14 5.0000 15 5.6250 16 5.3125 17 5.5625 18 ' 5.7500 19 5.9375 20 5.5000 21 5.7500 22 5.8750 23 5.7500 24 5.8125 25 5.6250 26 5.8125 27 5.9375 28 6.1875 29 8.0000 30 8.0000 31 8.0000 32 8.0000 33 8.0000 34 8.0625 35 8.1250 36 8.0000 Experiment No. 22 Trajectory x (cm) yc(cm) 1 2.4375 2 3.2500 3 3.6250 4 4.0000 5 4.0625 6 4.3750 7 4.5625 8 4.5625 9 4.9375 10 4.8750 11 5.1875 12 5.3125 13 5.0000 14 4.8750 15 5.2500 16 5.3750 17 5.3750 18 5.3750 19 5.2500 20 5.5000 21 5.5625 22 5.5000 23 5.3750 24 5.3750 25 5.3750 26 5.5000 27 5.5000 28 6.0625 29 6.4375 30 6.5000 31 6.1250 32 6.3125 33 6.4375 34 8.0000 35 7.8750 36 7.6875 APPENDIX G Details of Calculation of Buoyancy Stabilizing Ratios The two buoyancy stabilizing ratios, W and W , . 3 , given in Table G-1 were / H(u y / w(u ) calculated for Experiment 5 and 15 based on the following. 1. Reynolds number is given by Re = UAr, where n, is the hydraulic radius, Ua the ambient velocity and v the kinematic viscosity. 2. The friction factor f is determined from the Moody Diagram (Potter et al., 1997) based on Reynolds number and the relative roughness ^ r = 0.001. 3. The shear velocity was then calculated using = ^y^f (Fischer et al., 1979). Table G-1 Expt. 5 Expt. 15 Ua (m/s) 0.244 0.363 H(m) 0.081 0.12 B 1.49 x IO 6 2.04 x 10"5 v (m2/s) 1.36 x IO 6 1.14 x 10"6 n (m) 0.06 0.08 Re 4.30 x 10 4 1.03 x 105 f 0.025 0.023 u (m/s) 0.014 0.019 B/ /H(u*f 6.6 24.8 W /w(uf 1.1 5.9 287 APPENDIX H Details of Analysis of Total and Dissolved Contaminant Levels in CANFOR Effluent 1. Six 25 ml samples of well mixed CANFOR whole mill effluent were taken. Three were filtered through 0.45 micron glass fibre filter and the filtrate collected for analysis of chlorinated guaiacols in the dissolved portion of the effluent. The remaining three were analyzed for the concentration of these compounds in whole mill effluent. The CANFOR effluent sample was collected by the B.C. Ministry of Environment on April 25, 1994. 2. Each sample, along with a blank and standard solutions, were analyzed according to the method laid out in Appendix D. 3. The results of this analysis are given in Table H - l Table H-l Sample 3,4,5 T C V (ppb) 3,4,5 T C G (ppb) D l - dissolved 0.607 1.17 D2 - dissolved 0.547 0.98 D3 - dissolved destroyed destroyed T l - total 1.64 1.96 T2 - total 1.38 1.66 T3 - total 1.28 1.38 4. Based on the above preliminary analysis, approximately 60 % of the 3,4,5 -trichloroveratol and 35 % of the 3,4,5 - trichloroguaiacol was associated with the larger than 0.45 micron fraction of the effluent. 288 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0228836/manifest

Comment

Related Items