Mitigating Risks of Oil Sands Process Affected Water Using Hybrid

Clemson University TigerPrints

All Dissertations Dissertations

12-2016 Mitigating Risks of Oil Sands Process Affected Water Using Hybrid Constructed Wetland Treatment Systems Andrew David McQueen Clemson University, [email protected]

Follow this and additional works at: https://tigerprints.clemson.edu/all_dissertations

Recommended Citation McQueen, Andrew David, "Mitigating Risks of Oil Sands Process Affected Water Using Hybrid Constructed Wetland Treatment Systems" (2016). All Dissertations. 1827. https://tigerprints.clemson.edu/all_dissertations/1827

This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by an authorized administrator of TigerPrints. For more information, please contact [email protected].

MITIGATING RISKS OF OIL SANDS PROCESS AFFECTED WATER USING HYBRID CONSTRUCTED WETLAND TREATMENT SYSTEMS

A Dissertation Presented to the Graduate School of Forestry and Environmental Conservation Clemson University

In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy Wildlife and Fisheries Biology

by Andrew David McQueen December, 2016

Accepted by: Dr. John H. Rodgers Jr., Committee Chair Dr. James W. Castle Dr. Burton C. Suedel Dr. George M. Huddleston Dr. William C. Bridges

i ABSTRACT

Mining leases in the Athabasca Oil Sands (AOS) region produce extensive

volumes of oil sands process-affected water (OSPW) containing constituents that limit beneficial uses, including discharge into receiving aquatic systems. The aim of this

research is to provide a scalable approach using hybrid constructed wetland treatment

systems for the mitigation of problematic constituents in OSPW. In the first experiment in this dissertation, OSPW was characterized to identify constituents of concern (COCs) using chemical, physical, and toxicological analyses. Following identification of COCs, bench-scale manipulations (termed process-based manipulations [PBMs]) were used to remove or alter “classes” of COCs in an effort to eliminate toxicity to a sentinel aquatic invertebrate and discern treatment processes. COCs identified in OSPW included organics (naphthenic acids [NAs], oil and grease [O/G]), metals/metalloids, and suspended solids. Results from PBMs indicated that the organic fraction of OSPW was the primary source of toxicity, with oxidation (i.e. H2O2+UV254) and granular activated

charcoal treatments eliminating toxicity to Ceriodaphnia dubia (7-8 d), in terms of

mortality and reproduction. In the second experiment, photocatalytic degradation of

commercial (Fluka) NAs was evaluated using fixed-film titanium dioxide (TiO2)

irradiated with sunlight for 8 hours. Changes in NA concentrations by photocatalytic

degradation were confirmed analytically and with toxicity tests using sentinel fish and aquatic invertebrate species. The half-life for Fluka NAs achieved by photocatalytic degradation was approximately 2 hours, with toxicity eliminated for vertebrate and

invertebrate sentinel organisms (Pimephales promelas and Daphnia magna) by the 5th

ii hour of the sunlight exposure. In the third experiment, toxicity of the NA fraction to microbial populations was evaluated to discern adverse impacts to microbially driven processes within wetlands. Following exposures to a commercial NA, potential effects on sulfate-reducing bacteria (SRB), production of sulfides (as acid-volatile sulfides [AVS]), and precipitation of divalent metals (i.e. Cu, Ni, Zn [as aqueous and simultaneously extracted metals; SEM]) were evaluated. Extent of AVS production was sufficient in all

NA exposure concentrations tested to achieve ∑SEM:AVS <1, indicating conditions were conducive for treatment of divalent metals. In addition, no adverse effects to SRB (in terms of density, relative abundance, and diversity) were observed. The lines of evidence indicated that dissimilatory sulfate reduction and subsequent metal precipitation in wetlands will not be vulnerable to NA exposures. In the final experiment, a hybrid pilot- scale CWTS was designed to promote treatment processes to alter (transfer and transform) COCs using sequential reducing and oxidizing wetland reactors and a solar photocatalytic reactor using fixed film titanium dioxide (TiO2). Performance criteria were achieved as the CWTS decreased concentrations of NAs, O/G, suspended solids, and metals to an extent that eliminated toxicity to the aquatic invertebrate C. dubia. Results from this study provide proof-of-concept data to inform hybrid passive or semi-passive treatment approaches (i.e. constructed wetlands) that could mitigate COCs contained in

OSPWs. Data presented in this dissertation provide approaches to identify problematic constituents contained in complex energy derived waters (e.g. OSPW) and strategies for mitigating risks by altering exposures using passive (low-energy) treatment systems.

iii DEDICATION

I dedicate this dissertation to my wife Cindy, who continuously encourages me and to my sons Ethan and Samuel for helping me keep life in perspective.

iv ACKNOWLEDGMENTS

I would like to thank my major advisor Dr. John H. Rodgers Jr. for his devotion to

students and for providing me the tools needed to achieve my goals. I would also like to

thank my all my committee members, Dr. James W. Castle, Dr. Burton C. Suedel, Dr.

George M. Huddleston, and Dr. William C. Bridges for their support, wisdom, technical

reviews, and research guidance. I would also like to thank my fellow graduate students

who have provided an exceptional research team; Alyssa Calomeni, Ciera Kinley, Kyla

Iwinski, Maas Hendrikse, Tyler Geer, Daniel Gaspari, and Kayla Wardlaw. I also am grateful to the sponsors of this research for providing funding support. Finally, I thank

Dr. Wayne Chao for providing analytical support and always a smiling face. I offer my

extreme gratitude to all who were apart of this process.

v TABLE OF CONTENTS

Page

TITLE PAGE ...... i

ABSTRACT ...... ii

DEDICATION ...... iv

ACKNOWLEDGMENTS ...... v

LIST OF TABLES ...... viii

LIST OF FIGURES ...... x

CHAPTER

I. INTRODUCTION ...... 1

Background and Approach ...... 1 Organization of Dissertation ...... 11 References ...... 16

II. A RISK-BASED APPROACH FOR IDENTIFYING CONSTITUENTS OF CONCERN IN OILS SANDS PROCESS AFFECTED WATER FROM THE ATHABASCA OIL SANDS MINING AREA ...... 22

Abstract ...... 23 Introduction ...... 24 Materials and Methods ...... 28 Results and Discussion ...... 34 Conclusion ...... 44 References ...... 46

III. PHOTOCATALYSIS OF A COMMERCIAL NAPHTHENIC ACID IN ..... WATER USING FIXED-FILM TIO2 ...... 64

Abstract ...... 65 Introduction ...... 66 Materials and Methods ...... 69 Results and Discussion ...... 73 Conclusion ...... 77

vi Table of Contents (Continued)

Page

References ...... 79

IV. INFLUENCE OF COMMERCIAL (FLUKA) NAPHTHENIC ACIDS ON . ACID VOLATILE SULFIDE PRODUCTION AND DIVALENT ...... METAL PRECIPITATION ...... 93

Abstract ...... 94 Introduction ...... 95 Materials and Methods ...... 98 Results and Discussion ...... 104 Conclusion ...... 111 References ...... 113

V. PERFORMACE OF HYBRID PILOT-SCALE CONSTRUCTED ...... WETLAND SYSTEMS FOR TREATING OIL SANDS PROCESS ...... AFFECTED WATER FROM THE ATHABASCA OIL SANDS AREA ...... 129

Abstract ...... 130 Introduction ...... 131 Materials and Methods ...... 135 Results and Discussion ...... 141 Conclusion ...... 154 References ...... 155 Supplemental Material ...... 177

VI. CONCLUSIONS...... 185

Objectives ...... 185 References ...... 190

APPENDICES ...... 177

A: Chapter V: Supplemental Material ...... 177

vii LIST OF TABLES

Table Page

2.1 Analytical methods for site-specific OSPW...... 56

2.2 Manipulations and treatment objectives of process based manipulations ...... (PBMs) performed on OSPW...... 57

2.3 Comparison of constituents in OSPW to water quality guidelines (CCME ... 1999, USEPA 1999; ESRD 2014) and toxicity values for C. dubia (Cd), rainbow trout (Oncorhynchnus mykiss [Om]), and fathead minnow (P. .. promelas [Pp])...... 58

2.4 Comparison of inorganic constituents in OSPW to water quality guidelines . (CCME 1999; USEPA 1999; ESRD 2014) and toxicity values for C. dubia (Cd), D. magna (Dm), fathead minnow (P. promelas [Pp]), and rainbow trout (Oncorhynchnus mykiss [Om])...... 59

2.5 Bioassay data for OSPW using fish (P. promelas), aquatic invertebrates (C. . dubia and D. magna), and a macrophyte (T. latifolia)...... 60

3.1 Methods for water characteristics, light, and NA concentrations...... 83

3.2 Physical and chemical characteristics of Fluka naphthenic acids (Sigma- ...... Aldrich)a ...... 84

3.3 Summary of toxicity test conditions for P. promelas and D. magna (USEPA, 2002) ...... 85

3.4 Measurements of water characteristics during photocatalysis and photolysis treatments (and dark controls) and toxicity testing...... 86

3.5 Fluka naphthenic acid removal rate coefficients and extents for photocatalysis, photolysis, and dark control treatments ...... 87

3.6 Comparative rates and extents of removal for photocatalysis and photolysis of Fluka NA ...... 88

3.7 Summary of mean NA concentrations (mg/L) and 96-hr percent survival for microinvertebrates (D. magna) and fish (P. promelas) for photocatalysis, photolysis, and dark control treatments. NA concentrations with different letters are significantly different (p< 0.05) ...... 90

viii List of Tables (Continued)

Table Page

4.1 Analytical methods for experimental parameters...... 120

4.2 Chemical sources and nominal concentrations of constituents used for ...... aqueous exposures...... 121

4.3 Water and sediment characteristics measured during treatment periods 3-d to 21-d (n=21) ...... 122

5.1 Hybrid pilot-scale CWTS design features...... 166

5.2 Comparison of water quality characteristics and organic constituents in OSPW to water quality guidelines (USEPA 2007; CCME 2011, ESRD 2014) and toxicity values for C. dubia (Cd), rainbow trout (O. mykiss [Om]), and .. fathead minnow (P. promelas [Pp])...... 167

5.3 Comparison of metals and metalloids in OSPW to water quality guidelines . (USEPA 2007; CCME 2011, ESRD 2014) and toxicity values for C. dubia (Cd), D. magna (Dm), fathead minnow (P. promelas [Pp]), and rainbow trout (Oncorhynchnus mykiss [Om])...... 168

5.4 Targeted treatment processes, operational metrics, and measured ranges in . hybrid CWTS designed for OSPW ...... 169

5.5 Inflow concentrations, outflow concentrations, removal efficiencies, and .... removal rate coefficients of COCs for each CWTS replicate series. ... 170

5.1.A Treatment dates and conditions for hybrid pilot-scale CWTS...... 178

5.2.A Mean water characteristics (n=5) measured in hybrid pilot-scale CWTS..179

ix LIST OF FIGURES

Figure Page

1.1 Athabasca oil sands (AOS) region located in Alberta, Canada ...... 2

1.2 Conceptual model for context of experiments ...... 5

1.3 Conceptual model of treatment pathways for naphthenic acids ...... 6

2.1 Approach for identifying constituents of concern (COCs) in OSPW and ...... informing potential treatment processes to mitigate ecological risk .... 61

2.2 Hereroatom classes present in OSPW using high-resolution Orbitrap MS . 62

2.3 Responses of Ceriodaphnia dubia, in terms of survival (top) and reproduction (bottom), in 7-8 d static/renewal tests following process based ...... manipulations (PBMs)...... 63

3.1 Schematic of photocatalytic reaction chamber ...... 91

3.2 Concentrations of Fluka naphthenic acids (mg/L) with time in photocatalysis, photolysis, and dark control treatments ...... 92

4.1 Acid volatile sulfide (AVS) concentrations (µmol/g) measured during 21-day testing period (n=3). Error bars indicate standard deviations. Asterisks . indicate AVS concentrations significantly different (p<0.05; α=0.05) from untreated controls ...... 123

4.2 Comparison of rate of acid volatile sulfide (AVS) production (µmol/g ) in .. sulfate-reducing bacteria (SRB) biocide treatments, untreated control, and 80 mg/L naphthenic acid (NA) treatments (n=3) on days 3 through 15. . Error bars indicate standard deviations ...... 124

4.3 Simultaneously extracted metals (∑SEM; Cu, Ni, Zn; µmol/g) and acid ...... volatile sulfide (AVS) ratios among controls and highest NA treatment (80 mg NA/L) during 21-day testing period (n=3). Errors bars indicate ...... standard deviations ...... 125

4.4 Copper (top), nickel (middle), and zinc (bottom) removal extent (mg/L) ...... measured during 21- day treatment durations ...... 126

x List of Figures (Continued)

Figure Page

4.5 MPN of sulfate-reducing bacteria per gram of sample. Error bars indicate ... standard error of the average results of samples (n = 3) ...... 127

4.6 Percentage of known SRB in the bacterial community based on genetic ...... sequencing. Names of organisms are either genus (g) or family (f) level classification. Relative abundance (%) for each sample is the average .... across three replicates...... 128

5.1 A schematic diagram of the pilot-scale experiment ...... 171

5.2 O/G concentrations in inflow and reactor outflows for Series A (solid line) .. and Series B (dashed line) ...... 172

5.3 Total NA concentrations in wetland inflow and outflows for Series A (solid lines) and Series B (dashed lines), and inflow and outflow from ...... photocatalysis treatments ...... 173

5.4 Inorganic COC (Al, B, Cu, Ni, Se, Zn, and TSS) concentrations in inflow and reactor outflows for Series A (solid line) and Series B (dashed line) for . treatment periods 1-5...... 174

5.5 Survival (a) and reproduction (b) of Ceriodaphnia dubia exposed to inflow . (untreated) and wetland outflow samples of OSPW treated by pilot-scale CWTS. (n=20 per treatment)...... 176

5.1.A Mean plant shoot density measured in reducing reactors (top graph; n=4) and oxidizing reactors (bottom graph; n=10) during CWTS maturation and ... treatment periods for replicate wetland Series A and B...... 180

5.2.A Mean plant shoot height measured in reducing (n=6) and oxidizing (n=36) .. reactors during CWTS maturation and treatment periods...... 181

5.3.A Mean oxidation-reduction potential (ORP; mV) measured in reducing (n=4) and oxidizing reactors (n=10) during CWTS maturation and treatment .. periods...... 181

5.4.A Hereroatom classes present in untreated (inflow) OSPW using high-resolution Orbitrap MS...... 182

xi List of Figures (Continued)

Figure Page

5.5.A UV/visible transmittance (250 to 700 nm) at 1.0 cm water depth of untreated (inflow) OSPW and post-wetland treatment outflow...... 183

5.6.A Light extinction coefficient (Kd cm-1) of untreated OSPW ...... 184

xii

CHAPTER ONE

INTRODUCTION

1.1 Background and Approach

Oil sands are comprised of sands and silts permeated with a highly biodegraded and viscous form of hydrocarbon known as bitumen (Carrigy, 1963). Due to the continued global demand for oil, “unconventional” sources of petroleum (e.g. oil sands) have developed, with many sources outpacing conventional oil recovery. The Athabasca

Oil Sands (AOS) deposits in Alberta, Canada are the third largest proven oil reserve in the world, with ˜168 billion barrels recoverable and an estimated oil production rate of

1.9 million barrels/day (Alberta Energy, 2014). Current bitumen extraction processes include open pit mining and in situ recovery (i.e. steam assisted gravity drainage and cyclic steam stimulation; Schramm, 2000). Due to the close surface proximity of bitumen deposits in the AOS region, open pit mining represents the majority (approximately 51%) of extraction (Alberta Energy, 2014). Surface mining in the AOS requires the removal of overburden (i.e. vegetation and sediments), which disturbs the naturally occurring landscape. The AOS landscape is predominantly wetlands (>50%), with approximately

90% of those wetlands existing as peatlands (Vitt et al., 1996; Raab and Bailey, 2012). In addition to peat-dominated wetlands, areas mined to date consist of boreal forests, bogs, and fens (Bayley and Mewhort, 2004). Legislation requires operators to reclaim the mined landscape to “equivalent land capabilities” that existed pre-disturbances (CEMA,

2014; Province of Alberta, 2014).

Spatial extent of the Athabasca deposits encompasses approximately 140,200 km2, with 715km2 disturbed to date (Alberta Energy, 2014; Figure 1.1). Bitumen is commonly extracted from the oil sands by an alkaline warm water assisted extraction process, using water (50–80°C) and a conditioning agent (NaOH; Clark process;

Schramm, 2000). Approximately 1.6 barrels of fresh water are required for every barrel of petroleum produced from surface mining operations (Shell Canada Ltd., 2016). The resulting water is referred to as oil sands process-affected water (OSPW).

Figure 1.1 Athabasca oil sands (AOS) region located in Alberta, Canada.

(Source: Hein et al., 2013)

OSPW can contain complex mixtures of residual bitumen (petroleum

hydrocarbons), suspended solids, organic acids, metals, metalloids, and salts (Mackinnon

and Boerger, 1986; Allen, 2008a; Mahaffey and Dube, 2016). Ecological risks, in terms

of toxicity to aquatic biota, have been attributed to the organic acid fraction of OSPW,

referred to as naphthenic acids (NAs; e.g., Verbeek et al., 1994; Nero et al., 2006; Frank

et al., 2008; Armstrong et al., 2008; Kavanagh et al., 2012; Leclair et al., 2013;

Marentette et al., 2015). NAs are a natural component of bitumen and are transferred into

the process-affected water during extraction. Warm water extraction with sodium

hydroxide promotes retention of NA in the water column due to the alkaline pH (Rogers

et al., 2002; Headley and McMartin, 2004). NAs present in OSPW are compositionally

complex, comprised of over 3,000 individual acidic compounds (Ross et al., 2012),

including carboxylic acids fitting classical definitions of NAs (CnH2n+ZO2) in addition to dibasic, heteroatomic, aromatic, and diamondoid adamantine acids (Headley et al., 2011,

Rowland et al., 2011). Although the mechanisms of NA toxicity are not well known, it is

thought that general narcosis, membrane disruption, and osmotic stress are possible

factors in manifestation of adverse effects (Schramm, 2000). Recent research highlights

that composition of NAs (and not concentration alone) is critical to understanding and

predicting adverse effects (Morandi et al., 2015; Mahaffey and Dubé, 2016). Potency of

NAs is typically greater in more readily biodegradable and lower-molecular weight NA fractions (i.e. <180 dalton range; Holowenko et al., 2002, Lo et al., 2006). OSPW sourced from tailings in the AOS typically contains a greater fraction of “weathered” NAs (higher molecular weight NAs [typically in the 180-300 dalton range]; Headley et al., 2007). In

addition to NAs, there are a number of other constituents in OSPW that may require treatment, including trace metals/metalloids (e.g., Al, As, Cr, Cu, Fe, Pb, Ni, Se, Zn), unrecovered bitumen (oil and grease), major anions (chloride), and suspended solids

(Allen, 2008a; Mahaffey and Dubé, 2016).

To date, there has been a zero-discharge policy of OSPW in the AOS region

(Province of Alberta, 2014). However, to achieve reclamation goals for mining leases in the AOS, OSPW will need to be returned to the environment (Mahaffey and Dubé, 2016).

Due to the volume of OSPW currently stored in the AOS (~1 billion m3; Mahaffey and

Dubé 2016), it is clear that economically viable mitigation approaches are required, offering passive (low-energy input) solutions. Constructed wetland treatment systems

(CWTS) are plausible treatment options for renovating OSPW to achieve water return goals. However, for CWTS to be successful, a thorough design basis will need to be developed based on: 1) accurate characterization of problematic constituents contained in

OSPW that impede water return goals, 2) identification and evaluation of treatment processes to mitigate risks of OSPW, and 3) evaluation of treatment performance using model CWTS (Figure 1.2).

Figure 1.2 Conceptual model for context of experiments

1.1.1 OSPW Characterization

The composition of OSPW stored on the AOS landscape can vary spatially and

temporally (Frank et al., 2016); therefore, to properly assess ecological risks, accurate

characterization of constituents in OSPW is needed. This research focused on OSPW procured from Shell’s Muskeg River Mine External Tailings Facility (MRM-ETF).

MRM-ETF OSPW is produced from a surface mining operation in the AOS region (near

Fort McMurray, AB, Canada) and stored in a “fresh” tailings pond receiving ~94 million

m3 of total fluid fine tailings annually (Shell Canada Ltd., 2016).

1.1.2 Naphthenic Acid Treatment Pathways

Due to the contribution of toxicity from NAs contained in OSPW (Verbeek et al.,

1994, Morandi et al., 2015), a focus of this research is experimentally testing treatment processes for mitigating NAs as well as potential vulnerabilities of microbially mediated processes within wetland treatment systems to NA exposures. There are a number of NA treatment pathways that can be targeted or enhanced in wetland systems, including hydrolysis, oxidation, photolysis, volatilization, sorption, settling, and microbial degradation (Figure 1.2).

Figure 1.3 Conceptual model of treatment pathways for naphthenic acids.

Note: “oxidation” includes advanced oxidation processes (e.g. photocatalysis).

Photocatalysis is a type of advanced oxidation that has shown promising results as

a treatment option for degradation of NAs (Headley et al., 2009; Mishra et al., 2010;

Leshuk et al., 2016). Photocatalytic oxidation occurs when a catalyst absorbs photons,

and the corresponding energy increase causes electrons (i.e. energy) to be transferred. In

the case of metal oxides such as titanium dioxide (TiO2), the combination of excitation

energy and surface moisture produces hydroxyl radicals (Linsebigler, 1995).

Photocatalytic degradation has been used successfully in treating problematic organic

constituents in the petroleum industry, offering an innovative alternative to conventional

physical, chemical, and biological treatment strategies (Sain and Shahrezaei, 2011). To

date, laboratory-scale photocatalytic degradation of both commercial and OSPW NA

mixtures has been successful. Reported degradation rates for OSPW NA mixtures using

photocatalytic treatment are orders of magnitude greater than conventional

biodegradation techniques, with NA half-lives ranging between 1.55 and 4.50 h (Headley

et al., 2009; Mishra et al., 2010). To date, this technology has been evaluated by adding

the catalyst as a “slurry” (i.e. TiO2 nanoparticles), potentially prohibiting larger scale

application due to the cost of adding and recovering the catalyst. Part of this research

focused on fixing catalyst to a film (i.e. fixed-film), coupled with natural solar irradiance,

providing a passive treatment approach for NAs that can be translated to larger scale

applications.

Many transfer and transformation processes that occur in wetlands are mediated by microbial activity (Lovley, 1997; Kadlec and Wallace, 2009). Microbial activity can include dissimilatory sulfate reduction, nitrification/denitrification, and biodegradation of

organic matter (Newman et al., 1997; Kadlec and Wallace, 2009). Microbial degradation

pathways reported in the literature indicate some NAs fractions are biodegradable

(MacKinnon and Boerger, 1986; Scott et al., 2005; Del Rio et al., 2006); however, higher

molecular weight NAs (180-300 dalton range) are more recalcitrant (Quagraine et al.,

2007). NA biodegradation rates are more favorable in aerobic than in anaerobic conditions (Del Rio et al., 2006). Under aerobic conditions, monocyclic NA concentrations were decreased by 30% in 14 days (Del Rio et al., 2006), with anaerobic degradation half-lives of years (Quagraine et al., 2007). Residual consumable organic

matter (presumably residual hydrocarbons) in tailing ponds results in anaerobic

conditions in portions of the water column, promoting conditions that minimize or

eliminate degradation of NAs (Whitby, 2010). Half-lives for OSPW NAs in tailings

ponds range from 12.8-13.6 years (Han et al., 2009). At this rate, it would take decades to

decrease NA concentrations to non-toxic levels through in situ degradation.

In complex waters like OSPW, it is critical to understand the potential inhibitory

effects that fractions (e.g. metals, salts, organics) may pose to treatment processes

mediated by microbial activity. Due to potential toxicity of the organic acid fraction

(NAs) to microbial assemblages (Microtox EC50 values ranging from 12 to 65 mg/L NAs;

Rogers et al., 2002; Frank et al., 2008), efficiencies of targeted microbial treatment

pathways may be minimized or eliminated. To date, limited data are available regarding

effects of NAs on sulfate reducing bacteria (e.g. Desulfovibrio, Desulfotomaculum, and

Desulfomonas). In order to design efficient and effective wetland treatment systems, it is

necessary to determine the effects of potentially toxic organic acid fractions of the water

on biogeochemical pathways supporting treatment of constituents of concern.

1.1.3 Constructed Wetland Treatment Systems

To reclaim impacted waters, wet-landscape mitigation approaches (e.g. constructed wetland treatment systems) are among technologies considered for treatment

of problematic constituents (e.g. metals, organics; Allen, 2008b; Foote 2012; Toor et al.,

2013; Brown and Ulrich, 2015). Although CWTS have successfully treated many

constituents in petroleum related waters (Kadlec and Wallace, 2009), a robust system is likely needed to treat the variety of COCs that may be found in OSPW with specific respect to mitigating risks to downstream aquatic biota. To date, information regarding the viability of CWTS for treatment of OSPW in the AOS has been limited (Quagraine et al., 2005; Allen, 2008b; Brown and Ulrich, 2015). Bench-scale constructed wetlands have been tested using OSPW with limited degradation of the total NA fraction (Toor et al.,

2013); however, these systems were not explicitly designed using a focused process based approach. Assuming that NAs are a rate limiting constituent in OSPW, hybrid

CWTS offer the flexibility to add or enhance pathways (e.g. photocatalysis) that may otherwise be a rate limiting step in a “traditional” design. To test the viability of hybrid

CWTS, pilot-scale systems can be used to assess potential wetland performance for COC and toxicity reduction as they are easily manipulated to provide proof-of-concept data

(Rodgers and Castle, 2008; Kadlec and Wallace, 2009). Constituent removal rates and percentages can be estimated from pilot-scale studies to decrease uncertainties and confirm design features for future, field-scale CWTS (Huddleston et al., 2000; Murray-

Gulde et al., 2005; Johnson et al., 2008; Huddleston and Rodgers 2008; Kadlec and

Wallace, 2009). Carefully designed pilot-scale CWTS offer testable models that can be used to: 1) measure rates and extents of removal 2) measure bioavailability of COCs 3) measure effects of operational or environmental changes to performance 4) provide repetition and allowing greater confidence for meeting performance criteria, and 5) provide data for scaling.

To be sustainable and achieve performance goals (i.e. water return to receiving systems), treatment wetlands must be carefully designed and constructed with explicit consideration of OSPW composition and other materials such as hydrosoil and vegetation that are used to build them (Rodgers and Castle, 2008). Performance of these pilot-scale systems will be monitored not only analytically, but also using sentinel organisms. To accurately understand the rate and extent of treatment (i.e. removal of elements or compounds), changes in exposure of an element or compound can be monitored using aquatic organisms (e.g. aquatic invertebrates [Ceriodaphnia dubia]). In the case of organic acids in OSPW, due to limitations in quantifying the more than 3000 acidic compounds that co-occur with NAs (Ross et al., 2012), it is not sufficient to rely on analytical summation of NAs alone to confirm successful mitigation. Therefore, use of sensitive sentinel organisms will be necessary endpoints for demonstration of treatment performance.

The rational of this research is to provide a scalable approach using hybrid constructed wetland treatment systems for the mitigation of constituents in OSPW. This research has four main objectives:

1. Characterize the constituents of concern in oil sands process-affected water

(OSPW) and discern potential treatment pathways in constructed wetland

treatment systems (CWTS) (Chapter 2),

2. Determine the photocatalytic degradation rates and extents for a commercial

naphthenic acid using fixed film TiO2 (Chapter 3),

3. Measure the responses of sulfate reducing bacteria assemblages, acid volatile

sulfide, simultaneously extracted and aqueous metal concentrations following

exposures to a commercial naphthenic acid (Chapter 4), and

4. Measure the performance of a hybrid pilot-scale constructed wetland treatment

systems using OSPW sourced from the Athabasca oil sands (Chapter 5)

1.2 Organization of Dissertation

This dissertation consists of six chapters: an Introduction (Chapter One), four independent manuscripts (Chapters Two, Three, Four, and Five), and Summary and

Conclusions (Chapter Six). Chapters Three and Four have been accepted for publication in Water, Air, and Soil Pollution and Ecotoxicology and Environmental Safety, respectively. The manuscripts and their respective target journals are as follows:

. Chapter 2: A Risk-based Approach for Identifying Constituents of Concern in Oil

Sands Process-affected Water from the Athabasca Oil Sands – Chemosphere

. Chapter 3: Photocatalysis of a Commercial Naphthenic Acid in Water using

Fixed-film Titanium Dioxide (TiO2) – Water, Air, and Soil Pollution

. Chapter 4: Influence of Commercial (Fluka) Naphthenic Acids on Acid Volatile

Sulfide (AVS) Production and Divalent Metal Precipitation – Ecotoxicology and

Environmental Safety

. Chapter 5: Performance of a Hybrid Pilot-scale Constructed Wetland System for

Treating Oil Sands Process-affected Water from the Athabasca Oil Sands –

Ecological Engineering

Collectively, these manuscripts provide information on problematic constituents contained in OSPW and treatment techniques for mitigating risks associated with those respective COCs using hybrid constructed wetland treatment systems. Brief overviews of the overall scope and objectives of each body manuscript (Chapters 2, 3, 4, and 5) are outlined below.

Chapter 2: A Risk-based Approach for Identifying Constituents of Concern in Oil

Sands Process-affected Water from the Athabasca Oil Sands Region

The aim of this experiment was to identify COCs in OSPW sourced from an active settling basin with the goal of providing a sound rational for developing mitigation

strategies using CWTS. COCs were identified through several lines of evidence with the following specific objectives:

1) compare concentrations of chemical and physical constituents in OSPW with

numeric water quality guidelines (i.e. Alberta WQGs, CEQGs, and USEPA

WQC) and toxicity threshold values for fish (Pimephales promelas and

Oncorhynchus mykiss) and freshwater invertebrates (Ceriodaphnia dubia and

Daphnia magna),

2) measure toxicity in OSPW using fish (P. promelas), aquatic invertebrates (C.

dubia and D. magna), and seedlings from a rooted macrophyte (Typha latifolia),

3) conduct process-based manipulations (PBMs) on OSPW to alter toxicity, in terms

of mortality and reproduction, to the aquatic invertebrate C. dubia, and

4) discern potential treatment pathways to mitigate ecological risks of OSPW based

on identification of COCs, toxicological analyses, and PBM results.

Chapter 3: Photocatalysis of a Commercial Naphthenic Acid in Water using Fixed- film Titanium Dioxide (TiO2)

The overall objective of this study was to measure rates and extents of photolysis and photocatalytic degradation of a commercially available (Fluka) NA using bench-scale fixed-film TiO2, and confirm changes in NA concentrations using sensitive vertebrate

(fish = Pimephales promelas) and invertebrate (Daphnia magna) species. To achieve this overall objective, specific objectives were to:

1) measure the rates and extents of removal of commercial (Fluka) NAs throughout

an 8 hour duration of natural sunlight (“photolysis”) and natural sunlight in the

presence of fixed-film TiO2 (“photocatalysis”), and

2) measure changes in toxicity after photolysis and photocatalysis treatments (in

terms of mortality) with sentinel fish (P. promelas) and microinvertebrate (D.

magna) species in 96-hr static tests.

Chapter 4: Influence of Commercial (Fluka) Naphthenic Acids on Acid Volatile

Sulfide (AVS) Production and Divalent Metal Precipitation

The overall objective of this experiment was to measure responses of sulfate-reducing

bacterial assemblages and microbially mediated treatment pathways (e.g. acid volatile

sulfide concentrations) following a series of exposures to a commercial (Fluka) NA in

bench-scale reactors. To achieve this overall objective, specific objectives were to:

1) measure relationships of acid-volatile sulfide (AVS), simultaneously extractable

metal (SEM), and aqueous metal (copper, nickel, and zinc) concentrations

following 21-d exposures to NAs, and

2) measure responses of sulfate-reducing bacterial (SRB) assemblages in terms of

relative abundance, diversity, and density (most-probable number) in sediment to

21-d exposures of NAs.

Chapter 5: Performance of a Hybrid Pilot-scale Constructed Wetland System for

Treating Oil Sands Process-affected Water from the Athabasca Oil Sands

The overall aim of this study was to evaluate the performance of a specifically designed hybrid pilot-scale CWTS for treating OSPW. In order to achieve this overall objective, specific objectives were to:

1) characterize OSPW in terms of chemical composition and targeted constituents of

concern,

2) design and assemble a hybrid pilot-scale CWTS to treat target constituents in

OSPW,

3) measure performance of the pilot-scale CWTS for OSPW based on rates and

extents of constituent removal, and

4) measure the performance of the pilot-scale CWTS using toxicity testing with the

aquatic invertebrate Ceriodaphnia dubia.

1.3 References

Alberta Department of Energy. (2014). Alberta’s oil sands 2014. [online]. Accessed online January 2016: http://www.ceri.ca/images/stories/CDN_Economic_Impacts_of_New_and_Existi ng_Oil_Sands_Development_in_Alberta_-_November_2014_-_Final.pdf

Alberta Environment & Sustainable Resource Development (ESRD). (2014). Environmental Quality Guidelines for Alberta Surface Waters. Water Policy Branch, Policy Division. Edmonton, AB, CA 48 pp.

Allen, E.W., (2008a). Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Journal of Environmental Engineering Science 7:123-138.

Allen, E.W., (2008b). Process water treatment in Canada’s oil sands industry: II. A review of emerging technologies. Journal Environmental Engineering and Science 7:499-524.

Armstrong, S.A., Headley, J.V., Peru, K.M., Germida, J.J. (2008). Phytotoxicity of oil sands naphthenic acids and dissipation from systems planted with emergent aquatic macrophytes. Journal of Environmental Science and Health, Part A 43(1): 36-42.

Bayley SE and Mewhort RL. (2004). Plant community structure and functional differences between marshes and fens in the southern boreal region of Alberta, Canada. Wetlands 24:277-294.

Brown, L.D., Ulrich, A.C., (2015). Oil sands naphthenic acids: A review of properties, measurement, and treatment. Chemosphere 127:276-290.

Carrigy, M.A., (Ed.). (1963). Athabasca Oil Sands: A collection of papers on the Athabasca Oil Sands. Research Council of Alberta. Edmonton, Alberta.

Canadian Council of Minister of the Environment (CCME). (2007). Canadian water quality guidelines for the protection of aquatic life. Guidance on the site specific application of water quality guidelines in Canada. Pp. 146.

Cumulative Environmental Management Association (CEMA). (2014). Guidelines for wetlands establishments on reclaimed oil sands leases. 3rd Ed. Alberta. Fort McMurray, AB.

Del Rio, L.F., Hadwin, A.K.M., Pinto, L.J., MacKinnon, M.D., Moore M.M., (2006). Degradation of naphthenic acids by sediment micro-organisms. Journal of Applied Microbiology 101: 1049-1061.

Foote, L. (2012). Threshold considerations and wetland reclamation in Alberta's mineable oil sands. Ecology and Society 17(1):35.

Frank, R.A., Kavanagh, R., Burnison, B.K., Arsenault, G., Headley, J.V., Peru, K.M., Van Deer Kraak,, G., Solomon, K.R., (2008). Toxicity Assessment of Collected Fractions from an Extracted Naphthenic Acid Mixture. Chemosphere 72:1309- 1314

Frank, R.A., Milestone, C.B., Rowland, S.J., Headley, J.V., Kavanagh, R.J., Lengger, S.K., Scarlett, A.G., West, C.E., Peru, K.M. and Hewitt, L.M., (2016). Assessing spatial and temporal variability of acid-extractable organics in oil sands process- affected waters. Chemosphere 160: 303-313.

Han, X., MacKinnon, M.D., Martin, J.W., (2009). Estimating the in situ biodegradation of naphthenic acids in oil sands process waters. Chemosphere 76:63-70.

Headley J, Du J, Peru K, McMartin D. (2009). Electrospray ionization mass spectrometry of the photodegradation of naphthenic acids mixtures irradiated with titanium dioxide. Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering 44: 591-597.

Headley, J.V., Peru, K.M., Barrow, M.P., & Derrick, P.J. (2007). Characterization of naphthenic acids from Athabasca oil sands using electrospray ionization: The significant influence of solvents. Analytical chemistry 79(16): 6222-6229.

Headley, J.V., McMartin, D.W., (2004). A Review of the Occurrence and Fate of Naphthenic Acids in Aquatic Environments. Journal of Environmental Science and Health A 39: 1989-2010.

Hein, F. J., Leckie, D. A., Larter, S. R., & Suter, J. R. (Eds.). (2013). Heavy-oil and oil- sand petroleum systems in Alberta and beyond (pp. 23-102). American Association of Petroleum Geologists (AAPG). Danvers, Massachusetts.

Holowenko, F.M., MacKinnon, M.D., Fedorak, P.M., (2002). Characterization of naphthenic acids in oil sands wastewaters by gas chromatography-mass spectrometry. Water Resources 36: 2843-2855.

Huddleston GM, Gillespie WB, Rodgers Jr. JH. (2000). Using constructed wetlands to treat biochemical oxygen demand and ammonia associated with a refinery effluent. Ecotoxicology and Environmental Safety 45:188-193.

Huddleston, III G.M., Rodgers, Jr. J.H., (2008). Design of a constructed wetland system for treatment of copper-contaminated wastewater. Environmental Geosciences 15:9-19.

Johnson, B.M., Kanagy, L.E., Rodgers, Jr. J.H., Castle, J.W., (2008). Feasibility of a pilot-scale hybrid constructed wetland treatment system for formulated natural gas storage produced waters. Environmental Geosciences 15:91-104.

Kadlec, R.H., Wallace, S., (2008). Treatment wetlands. CRC press. Boca Raton, Fl.

Kavanagh, R.J., Frank, R.A., Oakes, K.D., Servos, M.R., Young, R. F., Fedorak, P.M., MacKinnon, M.D., Solomon, K.R., Dixon, D.G., Van Der Kraak, G., (2011). Fathead minnow (Pimephales promelas) reproduction is impaired in aged oil sands process-affected waters. Aquatic toxicology 101(1): 214-220.

Leclair, L.A., MacDonald, G.Z., Phalen, L.J., Köllner, B., Hogan, N.S., van den Heuvel, M.R., (2013). The immunological effects of oil sands surface waters and naphthenic acids on rainbow trout (Oncorhynchus mykiss). Aquatic Toxicology 142: 185-194.

Leshuk, T., Wong, T., Linley, S., Peru, K.M., Headley, J.V., Gu, F., (2016). Solar photocatalytic degradation of naphthenic acids in oil sands process-affected water. Chemosphere 144:1854-1861.

Linsebigler, A.L., Guangquan, L., Yates, J.T., (1995). Photocatalysis on TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chemical Reviews 95:735.

Lo, C.C., Brownlee, B.G., Bunce, N.J. (2006). Mass spectrometric and toxicological assays of Athabasca oil sands naphthenic acids. Water research 40(4): 655-664.

Lovley, D.R., (1993). Dissimilatory Metal Reduction. Annual Review of Microbiology. 47:269-290.

MacKinnon, M.D., Boerger, H., (1986). Description of two treatment methods for detoxifying oil sands tailings pond water. Water Pollution Research Journal of Canada 21:496–512.

Mahaffey, A., Dube, M., (2016). Review of the composition and toxicity of oil sands process-affected water. Environmental Review. doi: 10.1139/er-2015-0060

naphthenic acid mixtures, to fathead minnow (Pimephales promelas) embryos. Aquatic Toxicology 164:108-117.

Mishra, S., Meda, V., Dalai, A., McMartin, D., Headley, J., Peru, K., (2010). Photocatalysis of Naphthenic Acids in Water. Journal of Water Resource and Protection 2:644-650.

Morandi, G.D., Wiseman, S.B., Pereira, A., Mankidy, R., Gault, I. G., Martin, J. W., Giesy, J.P., (2015). Effects-Directed Analysis of Dissolved Organic Compounds in Oil Sands Process-Affected Water. Environmental Science & Technology 49(20): 12395-12404.

Murray-Gulde, C.L., Huddleston, III G.M., Garber, K.V., Rodgers, Jr. JH., (2005). Contributions of Schoenoplectus californicus in a constructed wetland treatment system receiving copper contaminated wastewater. Water, Air, and Soil Pollution 163:355-378.

Nero, V., Farwell, A., Lee, L.E., Van Meer, T., MacKinnon, M.D., Dixon, D.G., (2006). The effects of salinity on naphthenic acid toxicity to yellow perch: Gill and liver histopathology. Ecotoxicology and Environmental Safety 65:252-264.

Newman D. K., Kennedy, E. K., Coates, J. D., Ahmann, D., Ellis, D. J., Lovley, D. R., Morel, F. M. M., 1997. Dissimilatory arsenate and sulfate reduction in Desulfotomaculum auripigmentum sp. Archives of Microbiology 168: 380-388.

Province of Alberta. (2014). Environmental Protection and Enhancement Act. Revised Statues of Alberta 2000. Chapter E-12. Division 1, Section 2.

Quagraine, E.K., Headley, J.V., Peterson, H.G., (2005). Is biodegradation of bitumen a source of recalcitrant naphthenic acid mixtures in oil sands tailing pond waters?. Journal of Environmental Science and Health, 40(3), 671-684.

Raab, D., Bayley, S.E., (2012). A vegetation-based Index of Biotic Integrity to assess marsh reclamation success in the Alberta oil sands, Canada. Ecological Indicators 15:43-51.

Rodgers, Jr. J.H., Castle, J.W., (2008). Constructed wetland systems for efficient and effective treatment of contaminated waters for reuse. Environmental Geosciences 15:1-8.

Rogers, V.V., Liber, K., MacKinnon, M.D., 2002. Isolation and characterization of naphthenic acids from Athabasca oil sands tailings pond water. Chemosphere 48: 519-527.

Ross, M. S., Pereira, A. D. S., Fennell, J., Davies, M., Johnson, J., Sliva, L., & Martin, J. W. (2012). Quantitative and qualitative analysis of naphthenic acids in natural waters surrounding the Canadian oil sands industry. Environmental science & technology 46(23): 12796-12805.

Rowland, S.J., Scarlett, A.G., Jones, D., West, C.E., Frank, R.A., (2011). Diamonds in the rough: identification of individual naphthenic acids in oil sands process water. Environmental science & technology 45(7): 3154-3159.

Saien, J., Shahrezaei, F., (2012). Organic Pollutants Removal from Petroleum Refinery Wastewater with Nanotitania Photocatalyst and UV Light Emission. International Journal of Photoenergy 2012: 5p

Schramm LL. (Ed.). (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press. N.Y.

Scott, A.C., MacKinnon, M.D., Fedorak, P.M., (2005). Naphthenic acids in Athabasca oil sands tailing are less biodegradable than commercial naphthenic acids preparations. Environmental Science and Technology 39:8388-8394.

Shell Canada Limited. (2016). Oil Sands Performance Report 2015 (dated April 22, 2016). Accessed online June 2016: http://www.shell.ca/can/en_ca/energy-and- innovation/oil-sands/heavy-oil-overview.html#vanity- aHR0cDovL3d3dy5zaGVsbC5jYS9vaWxzYW5kcw

Shi, Y., Huang, C., Rocha, K. C., El-Din, M. G., Liu, Y., (2015). Treatment of oil sands process-affected water using moving bed biofilm reactors: with and without ozone pretreatment. Bioresource technology 192, 219-227.

Toor, N.S., Franz, E.D., Fedorak, P.M., MacKinnon, M.D., Liber, K., (2013). Degradation and aquatic toxicity of naphthenic acids in oil sands process-affected waters using formulated wetlands. Chemosphere 90:449-458.

United States Environmental Protection Agency (USEPA). (1991). Technical support document for water quality-based toxics control. EPA/505/2-90-001. Washington, D.C.

United States Environmental Protection Agency (USEPA). (2007). National Recommended Water Quality Criteria., Washington, D.C.

Verbeek, A.G., Mackay, W.C., MacKinnon, M.D., (1994). Acute toxicity of oil sands wastewater: A toxic balance. (cited in Schramm LL. (Ed.). (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press. N.Y.

Vitt DH, Halsey LA, Bauer IE, and Campbell C. (2000). Spatial and temporal trends in carbon storage of peatlands of continental western Canada through the Holocene. Canadian Journal of Earth Sciences 37:683-693.

Whitby, C., (2010). Microbial naphthenic acid degradation. Advances in Applied Microbiology 70:93-125.

CHAPTER TWO

A RISK-BASED APPROACH FOR IDENTIFYING CONSTITUENTS OF CONCERN

IN OIL SANDS PROCESS AFFECTED WATER FROM THE ATHABASCA OIL

SANDS MINING AREA

Andrew D. McQueen1,5; Ciera M. Kinley1; Maas Hendrikse1; Daniel P. Gaspari2; Alyssa

J. Calomeni 1; Kyla J. Iwinski1; James W. Castle2; Monique C. Haakensen3; Kerry M.

Peru4; John. V. Headley4; John H. Rodgers Jr.1

1 Department of Forestry and Environmental Conservation, 261 Lehotsky Hall, Clemson University, Clemson, SC 29634, USA

2Department of Environmental Engineering & Earth Sciences, 445 Brackett Hall, Clemson University, Clemson, SC 29634, USA

3Contango Strategies Limited, LFK Biotechnology Complex, 15-410 Downey Road, Saskatoon, SK S7N 4N1

4Water Science and Technology Directorate, Environment and Climate Change Canada, 11 Innovation Blvd, Saskatoon, SK., S7N 3H5

5Corresponding author: Andrew McQueen

Manuscript prepared for submission to: Chemosphere

2.1 Abstract

Mining leases in the Athabasca Oil Sands (AOS) region produce large volumes of

oil sands process-affected water (OSPW) containing constituents that limit beneficial

uses, including discharge into receiving systems. The aim of this research is to identify

constituents of concern (COCs) in OSPW sourced from an active settling basin with the

goal of providing a sound rational for developing mitigation strategies for using

constructed treatment wetlands for COCs contained in OSPW. COCs were identified

through several lines of evidence: 1) chemical and physical characterization of OSPW

and comparisons with numeric water quality guidelines and toxicity endpoints, 2)

measuring toxicity of OSPW using a taxonomic range of sentinel organisms (i.e. fish,

aquatic invertebrates, and a macrophyte), 3) conducting process-based manipulations

(PBMs) of OSPW to alter toxicity and inform treatment processes, and 4) discern

potential treatment pathways to mitigate ecological risks of OSPW based on

identification of COCs, toxicological analyses, and PBM results. COCs identified in

OSPW included organics (naphthenic acids [NAs], oil and grease [O/G]),

metals/metalloids, phosphorus, and suspended solids. In terms of species sensitivities to

undiluted OSPW, fish ≥ aquatic invertebrates > macrophytes. Bench-scale manipulations

of the organic fractions of OSPW via PBMs (i.e. H2O2+UV254 and granular activated charcoal treatments) eliminated toxicity to Ceriodaphnia dubia (7-8 d), in terms of mortality and reproduction. Results from this study provide critical information to inform mitigation strategies using passive or semi-passive treatment processes (e.g., constructed treatment wetlands) to mitigate ecological risks of OSPW to aquatic organisms.

2.2 Introduction

Mining leases in the Athabasca Oil Sands (AOS) region, in Alberta Canada,

produce large volumes of oil sands process affected water (OSPW) during the operational

process of bitumen extraction, with approximately 1.6 barrels of fresh water required for

every barrel of petroleum produced in surface mining operations (Shell Canada Ltd.,

2016). OSPW often contains problematic constituents that need to be treated prior to

beneficial uses (e.g., discharge into receiving aquatic systems; Allen, 2008a; Alberta

Department of Energy, 2014). Ecological risks, in terms of toxicity to aquatic and terrestrial biota have been attributed to the organic acid fraction of OSPW, referred to as naphthenic acids (NAs; e.g., Verbeek et al., 1994; Nero et al., 2006; Frank et al., 2008;

Armstrong et al., 2008; Kavanagh et al., 2011; Leclair et al., 2013; Marentette et al.,

2015). In addition to NAs, other constituents in OSPW may require treatment including

trace metals/ metalloids (e.g., Al, As, Cr, Cu, Fe, Pb, Ni, Se, Zn), unrecovered bitumen

+ - 2- - (e.g., oil and grease [O/G]), major cations and anions (e.g., Na , Cl , SO4 , HCO3 ), and suspended and dissolved solids (Mackinnon and Boerger, 1986; Allen, 2008a).

Identification of site-specific constituents of concern (COCs) is a crucial step in

developing and implementing treatment systems for the goal of mitigating site-specific

OSPW for discharge to aquatic receiving systems. For this purpose, COCs are defined as elements, compounds, or parameters measured in OSPW waters that can adversely affect receiving system biota. Wetlands designed to promote ecosystem services (e.g., habitat,

flood protection, carbon storage, etc.) are part of a comprehensive reclamation plan for

mined leases, restoring areas to “equivalent land capabilities” of pre-mined landscapes in

the AOS region (Allen, 2008b; CEMA, 2014; Province of Alberta 2014). Logically, constructed wetland treatment systems (CWTS) could be designed to actively treat problematic constituents in OSPW and transition to serve as passive (reclaimed) wetlands

following their operational lifespan. To be successful, CWTS must be designed with

explicit processes to alter (transfer or transform) COCs to concentrations or forms that

mitigate ecological risks for aquatic organisms. Physical, chemical, and toxicological

characterization of site-specific OSPW is necessary to provide a sound design basis for

selection and prioritization of treatment processes. OSPW composition varies temporally

and spatially (Frank et al., 2016), and with bitumen extraction processes (e.g., froth

treatment, process aides, upgrading, etc.). Therefore, defining site-specific COCs will be

necessary to confirm rates and extents of treatment required for process-specific OSPW.

This study focuses on OSPW sourced from an active settling basin located in the AOS

region. A risk-based approach was used to identify COCs, including: 1) chemical-specific

approach comparing concentrations of parameters to relevant water quality criteria

thresholds, 2) toxicological approach using sentinel aquatic species, and 3) manipulation

of chemical and physical characteristics of OSPW samples using treatment processes

relevant to wetland treatment systems, with the goal of decreasing toxicity to sentinel

aquatic species.

The initial step for identifying COCs includes comparisons of concentrations of

constituents in OSPW to surface water quality guideline limits (i.e. Canadian

Environmental Quality Guidelines for the protection of freshwater aquatic life [CEQG],

Alberta Environment Water Quality Guidelines [Alberta WQGs], and United States

Environmental Protection Agency [USEPA] aquatic life criteria; CCME, 2007; USEPA,

2007; ESRD, 2012, 2014). Further, comparisons of concentrations of constituents contained in OSPW to water quality thresholds was supported with acute and chronic toxicity endpoints for constituents reported in peer-reviewed literature for sentinel aquatic organisms (i.e. fish: Pimephales promelas Rafinesque [fathead minnow] and

Oncorhynchus mykiss Walbaum [rainbow trout]; and aquatic invertebrates: Ceriodaphnia dubia Richard and Daphnia magna Straus). These species were selected based on availability of toxicological data for constituents present in OSPW (i.e. metals, organics, cations, and anions). In addition, these species are commonly used for evaluation of whole effluent for effluent discharge permits and for determination of water quality guidelines for protection of aquatic biota (CCME, 2007; USEPA, 2007; ESRD, 2014).

The second step for characterization of COCs included a toxicological evaluation of OSPW. Excursions from numeric criteria offer an initial step for identifying COCs; however, a chemical-specific approach is not sufficient to identify or confirm the presence of potential adverse effects to receiving system biota (USEPA, 1991).

Toxicological characterization of OSPW was achieved using OSPW sourced from the

AOS region (corresponding samples with analytical data). Toxicity testing using sentinel aquatic organisms provided an integrated response measure of ecological risk, accounting for interactions of constituents in complex mixtures of whole effluent (e.g., OSPW) and presence of unknown toxicants (USEPA, 1991). The toxicological analysis included a taxonomic range of aquatic animals and plants (i.e. fish, invertebrates, and a macrophyte)

to discern diagnostic responses of biota (e.g., relative sensitivities among species and

potencies within species), which may implicate sources of toxicity.

The third step in characterization of OSPW included manipulations of the water in

an effort to decrease toxicity (if present) to sentinel aquatic species. Chemical and

physical treatments can be evaluated in laboratory-scale experiments to simulate transfer

or transformation processes that can be implemented in CWTS, with the goal of altering

constituents into less bioavailable or non-toxic forms. These process-based manipulations

(PBMs) are strategically selected based on the chemical-specific evaluation and

toxicological analysis to inform potential wetland treatment processes required to alter exposures of COCs and mitigate risk. Fractions of OSPW (e.g., organics, metals, suspended solids) can be altered using bench-scale manipulations. Chemical and physical manipulations selected to alter toxicity included: filtration (removal of suspended solids),

divalent metal chelation (removal of metals), activated charcoal (removal of non-polar organics), and coupled hydrogen peroxide (H2O2)/ UV254 treatments (removal of organic

compounds). Changes in toxicity (in terms of lethal and sub-lethal responses) were evaluated using the aquatic invertebrate C. dubia, based on their relative sensitivity to constituents contained in OSPW. PBMs used in this study are not meant for direct translation as treatment strategies for altering ecological risk of OSPW (e.g., additions of chelators, oxidants, etc.), but to inform treatment processes that can be promoted in passive or semi-passive constructed wetland treatment systems (e.g., metal complexation and precipitation, microbial degradation of organic constituents). A synthesis of the

information gained from the OSPW characterization (steps 1-3) provide critical information to inform CWTS processes necessary to alter exposures and mitigate risk.

The aim of this research is to identify COCs in OSPW with the goal of providing a sound rational for developing mitigation strategies using CWTS. COCs were identified through several lines of evidence with the following specific objectives: 1) compare concentrations of chemical and physical constituents in OSPW with numeric water quality guidelines (i.e. Alberta WQGs, CEQGs, and USEPA WQC) and toxicity threshold values for fish (P. promelas and O. mykiss) and freshwater invertebrates (C. dubia and D. magna), 2) measure toxicity of OSPW using fish (P. promelas), aquatic invertebrates (C. dubia and D. magna), and seedlings from a rooted macrophyte (T. latifolia), 3) conduct process-based manipulations (PBMs) on OSPW to alter toxicity, in terms of mortality and reproduction, to the aquatic invertebrate C. dubia, and 4) discern potential treatment pathways to mitigate ecological risks of OSPW based on identification of COCs, toxicological analyses, and PBM results.

2.3 Materials and Methods

2.3.1 Identification of Constituents of Concern

The OSPW used in this study was procured from the Muskeg River Mine

External Tailings Facility (MRM-ETF) operated by Shell Canada Limited. MRM-ETF

OSPW is produced from a surface mining operation in the AOS region (near Fort

McMurray, AB, Canada) using a Clark caustic warm water extraction process (NaOH assisted at water temperatures of 50-80°C) to separate bitumen from ore. The approach

for identifying COCs in OSPW used chemical and toxicological data (Figure 1). In the initial step, chemical or physical parameters that exceeded water quality criteria/guidance limits (i.e. Alberta WQGs, CEQGs, and USEPA WQC) or toxicity endpoint values (point estimates were chosen when available; e.g., LC50s) were identified as COCs. The toxicity testing species O. mykiss, P. promelas, C. dubia, and D. magna were chosen for this comparison because of their sensitivity to many constituents found in OSPW, availability of data, and use in developing and enforcing water quality criteria (USEPA,

2002; CCME, 2007). COCs were parameters that exceeded the most conservative values from the selected criteria (Equation 1).

COCs = OSPW Parameter > WQC or Toxicity Endpoint Equation 1

2.3.2 Analysis of OSPW

OSPW samples were transported from the MRM-ETF to Clemson University, SC

(USA) for physical, chemical, and toxicological analyses. Samples for elemental analysis were collected in acid-cleaned 50-ml centrifuge tubes and preserved with concentrated trace metal-grade nitric acid (1% v/v; Fisher Scientific). Concentrations of Ag, Al, As, B,

Ba, Cd, Cr, Cu, Co, Fe, Pb, Mn, Mo, Ni, Se, and Zn were measured with an Inductively

Coupled Plasma Atomic Emission Spectrometer (Spectro Flame Modula; ICP-AES) following EPA method 200.7 (USEPA, 2001). Dissolved oxygen and pH were measured using YSI (model 85) and Orion® (model 410A+) instruments, respectively. Alkalinity, hardness, conductivity, and total suspended and dissolved solids were determined according to Standard Methods (APHA, 2012; Table 2.1).

Methods for NAs derivatization and analysis using High Performance Liquid

Chromatography HPLC (Dionex, UltiMate-3000; Sunnyvale, CA) were based on Yen et al. (2004). Commercially available (Fluka) NAs (Sigma-Aldrich; St. Louis, Mo) were used to prepare stock solutions of NAs for developing standard curves. “Total” NA concentrations were quantified based on derivatization of carbonyl compounds amendable to derivatization and detected at 400 nm (Yen et al., 2004). The HPLC column was an Agilent LiChrospher 100 RP-18 (5 µm particle size, 125mm x 4 mm) with a guard column packed with 2 µm RP-18 solid phase material. Column temperature was maintained at 40° C with a sample injection volume of 60 µL mobilized with HPLC grade methanol (Fisher Scientific) at a flow rate of 60 µL/min. Results were reported with a range of 85-115% recovery. NA concentrations were reported as means of triplicate analyses.

The chemical composition and profile of NAs contained in OSPW were identified by use of direct injection linear ion trap-orbitrap mass spectrometer (Orbitrab MS;

Orbitrap Elite, Thermo Fisher Scientific, San Jose, CA, USA) in negative (ESI−) electrospray according to the method described by Headley et al. (2016) and Leshuk et al.

(2016). Solid phase extraction (SPE), as previously described by Headley et al. (2002) was used as a cleanup and concentration technique for test samples. Chemical species detected in each fraction were grouped according to heteroatom empirical formula classes in ESI− electrospray: Ox− (where x = 1–5), N−, NOx− (where x = 1–4), S/−, SOx−

(where x = 1–5), or NOxS− (where x = 2).

2.3.3 Toxicity Testing Procedures

Freshwater organisms (P. promelas and C. dubia) were cultured at Clemson

University’s Aquatic Animal Research Laboratory according to USEPA (2002), under protocols in compliance with Clemson University’s Institutional Animal Care and Use

Committee. Toxicity testing protocols for P. promelas, C. dubia, and D. magna were

based on Environment Canada protocols (1996; 2007; 2011). Toxicity tests for P.

promelas were conducted by exposing 30 organisms (< 24 h old larvae) per concentration

(10 organisms per replicate for 3 replicates) in 250 ml borosilicate beakers. During

exposures, fish were fed Artemia sp. once daily. Toxicity tests for C. dubia and D. magna

were conducted by exposing 20 organisms (< 24 h old neonates) per concentration (1

organism per replicate for 20 replicates) in 15 ml borosilicate vials. During exposures, C.

dubia and D. magna were fed 200 µL of a 1:1 mixture of Pseudokirchneriella

subcapitata and YCT (yeast, cerophyll, trout chow) once daily.

For phytotoxicity testing, mature T. latifolia inflorescences were collected in

August and September, 2015, from a wetland site at Clemson University, Clemson, SC

(34°40'7.12"N, 82°50'53.98"W) and seeds were separated from bristle hairs by placing in

a blender filled with NANOpure® water and blending. Seeds that sank to the bottom after

blending were considered viable and used for testing (Kinley et al., 2016). Viable seeds

were then added to a small volume (~1 ml) of moderately hard water and incubated for 2

days to induce germination. Toxicity experiments for T. latifolia were initiated by adding

10 germinated T. latifolia seedlings (2 d old) to each replicate 50mL beaker (three

replicates/concentration) under fluorescent lighting (1,500-3,000 Lux) with a 16 h light/8

h dark photoperiod at 24 ± 1°C. Exposure concentrations were pipetted into treatment

chambers and volumes were maintained as necessary. Control (untreated) exposures

were moderately hard water. After 7 days, seedlings were removed from exposures and

preserved in 70% ethanol until analysis. Root and shoot lengths (mm) of seedlings were

measured using a Leica® M80 Stereoscope and software (Leica Microsystems®).

Reference toxicity tests were conducted for all test species using copper sulfate

(CuSO4·5H20; Fisher Scientific) for intra- and inter-laboratory comparisons (USEPA,

1991; USEPA, 2002). Acid soluble copper concentrations (exposures) were confirmed

using flame atomic absorption spectroscopy and graphite atomic absorption spectroscopy

(Agilent PSD 120 atomic absorption spectrometer; APHA, 2012).

Water characteristics of exposures were measured at test initiation and completion of toxicity tests, with the exception of T. latifolia exposures, which were measured at test initiation. Dissolved oxygen, pH, and conductivity of exposure waters were measured using a YSI® Model 52 dissolved oxygen meter, Orion® Model 250A pH meter, and

Orion® Model 142 conductivity meter, respectively. Hardness and alkalinity of samples

were measured according to Standard Methods for Examination of Water and

Wastewater (APHA, 2012).

2.3.4 Process-based Manipulations (PBMs)

The purpose of the PBMs outlined below is to remove or alter “classes” of COCs

(i.e. nonpolar organics, divalent metals, oxidizable compounds) which could be targeted using constructed wetlands. Treatment processes were conducted using 100% OSPW in an effort to reduce toxicity to the aquatic invertebrate C. dubia. PBMs consist of: 1)

filtration, 2) chelation with ethylenediaminetetraacidic acid (EDTA) addition, 3) granular

activated charcoal (GAC) treatment, and 4) coupled hydrogen peroxide H2O2/UV254 treatments (Table 2.2).

All manipulations were conducted at initial OSPW pH (~8.2). For fractionation

experiments, OSPWs were stored in 1 liter HDPE Nalgene® bottles at 4 ± 1°C and

warmed to 25°C prior to manipulations. To obtain a filterable fraction, OSPW was passed

through a 0.45μm nitro cellulose membrane filter using a glass separatory funnel

applying vacuum suction (-700 mm Hg). To chelate the metal fraction (e.g., cationic

metals: Al3+, Ba2+, Cd2+, Cu2+, Fe3+, Pb2+, Mn2+, Ni2+, and Zn2+), 50 mL of 0.1 M EDTA was used per liter of OSPW (based on hardness of untreated OSPW). Non-polar organic fractions were treated using granular activated charcoal (GAC; coconut 6-14 mesh;

Fisher Scientific) at concentrations of 5 g/L. GAC was added to OSPW and allowed a 1-2 h contact time, after which the residual GAC was removed prior to toxicity testing. To remove compounds susceptible to oxidation and photolysis, 100 ug/L hydrogen peroxide

(H2O2; Fisher Scientific) was added to OSPW and exposed to UV254 (8W fluorescent

tube; Philips Ltd.) for a duration of 12 h.

No observable effect concentrations (NOECs) and lowest observable effect

concentrations (LOECs) of OSPW (as % of undiluted OSPW) were determined by

statistically significant differences relative to untreated controls using one way analysis

of variance (ANOVA) and Dunnett’s multiple range test (α = 0.05; JMP Pro V.11).

Median lethal effect concentrations (LC50s) were estimated using the Probit model using

JMP Pro V.11 (SAS Institute Inc., Cary, NC).

2.4 Results and Discussion

2.4.1 Chemical-specific Approach for Identifying COCs

2.4.1.1 Naphthenic acids

Using two analytical methods, HPLC and Orbitrap MS, OSPW NAs

concentrations ranged from 80 to 128 mg/L (HPLC; n=10) and 93 to 103 mg /L (Orbitrap

MS; n=2). In addition, speciation and quantification of NAs were conducted via pH-

dependent extractions and quantification using Orbitrap MS analysis to investigate the

abundance of “classical” NAs (NAs, (CnH2n+zO2), oxidized NAs (Ox-NAs), and nitrogen-

- and sulfur-containing NAs. The sum of O2 -NA species accounted for ~50% of the total

abundance of extracted organic matter (Figure 2.2). The abundance of O2--NAs in OSPW is consistent with distribution of NAs reported for “fresh” OSPW, containing a greater abundance of monocarboxylic acids (CnH2n+zO2; Marentette et al., 2015; Barrow et al.,

2016). Morandi and coauthors (2015) implicated “classical” NAs as the primary

contributor to toxicity to fathead minnows (P. promelas), with non-acidic organic species

as minor contributors to toxicity. NAs concentration and species present in OSPW are generally consistent with OSPW-NAs quantified using similar extraction techniques and analysis (Grewer et al., 2010; Jones et al., 2015; Kavanagh et al., 2011; Marentette et al.,

2015; Barrow et al., 2016; Frank et al., 2016). NAs were identified as COCs based on the

concentration and species present as compared to published toxicity data (Table 2.3).

2.4.1.2 Residual Petroleum Hydrocarbons

The OSPW samples had a slight hydrocarbon sheen, with oil and grease (O/G)

concentrations ranging from 8 to 13 mg/L (Table 2.3). Reported concentrations of O/G

(an aggregate measure of residual hydrocarbons) in OSPW exceed narrative criteria for discharge water (i.e. no visible film or sheen of oil present; Table 2.3). Interestingly, residual bitumen (O/G) concentrations measured in OSPW used in this study were lower than ranges in OSPW from other sources, with a range of 9 to 92 mg/L as O/G (Allen,

2008a). Organics present in OSPWs studied in the AOS region exceeding numeric criteria also include: benzene, toluene, ethylbenzene, xylene (BTEX), polycyclic aromatic hydrocarbons (PAHs), and phenols (Allen, 2008a; Rogers 2002; Galarneau et al., 2014); however, data on lower-molecular weight hydrocarbons in OSPW are generally limited in peer reviewed literature (Mahaffey and Dubé, 2016). In this study, petroleum hydrocarbons (as O/G) were identified as COCs.

2.4.1.3 Inorganics

OSPW was analyzed for 16 elements (Table 2.4). Elements not above the MDL included: cadmium (MDL = 0.0002 mg/L), chromium (MDL = 0.004), cobalt (MDL =

0.0002 mg/L), and silver (MDL = 0.0002 mg/L). Based on the chemical-specific characterization approach, COCs for metals/ metalloids in OSPW included: Al, B, Cu,

Fe, Pb, Ni, Se, and Zn (Table 2.4). These elements measured as “total” exceed conservative concentrations in water quality criteria or toxicity endpoints for sensitive sentinel aquatic biota measured in unconfounded laboratory studies. Although metal speciation and concentrations in tailings ponds differ due to geologic heterogeneities and recycling of tailing pond water, concentration ranges of elements contained in OSPW are generally consistent with other OSPWs from the AOS (MacKinnon and Boerger, 1986;

Siwik et al., 2000; Allen, 2008a).

Maximum total ammonia (N) and phosphorus concentrations in OSPW samples

(n=10) were 0.099 mg/L and 0.082 mg/L, respectively. Published concentrations of

ammonia in other OSPWs have been reported as high as 18.4 mg/L (± 1.2; Lai et al.,

1996), exceeding toxicity values or water quality criteria guidelines (CCME, 2007;

ESRD, 2014). Ammonia toxicity is dependent upon pH of a solution (Thurston and

Russo, 1981). At 14.1°C and pH of 8.29, the 96 h LC50 of ammonia to rainbow trout is

0.563 mg/L (Thurston and Russo, 1981), approximately 5.5x greater than ammonia

concentrations observed in the source OSPW. In this study, total phosphorus maximum

concertation (0.082 mg/L) was above CCME (2007) guidelines, and phosphorus was

identified as a COC (Table 2.4). However, in the context of passive treatment, nutrient

concentrations (e.g., TP) can influence microbially mediated wetland biogeochemical

processes (e.g., biodegradation of NAs [Herman et al., 1994; Lai et al., 1996];

dissimilatory sulfate-reduction).

2.4.1.4 Cations and Anions

Dissolved salts are found in OSPW due to the connate water (water that forms

part of the ore body), extraction processes, and recycling of process water (Allen, 2008a).

- + Ions in OSPW were predominantly composed of bicarbonate (HCO3 ), sodium (Na ), and chloride (Cl-). These ions are associated with the ore body that forms the regional

geology (Mikula, 2013) and become part of the OSPW during the warm water extraction

process. Aggregate measures of major ions in OSPW (i.e. total dissolved solids [TDS])

indicate the ionic strength is greater than “background” receiving aquatic systems

(Golder Associates, 2002), with TDS in OSPW ranging from 510 to 1100 mg/L. TDS and

conductivity (generic measures for salinity of water) are not reliable predictors of aquatic toxicity (Goodfellow et al., 2000), therefore, evaluation of the composition and strength of ions present in OSPW is necessary. The role of total dissolved ions in terms of the strength and balance (or imbalance) is important in determining potential risk to receiving

aquatic biota (Goodfellow et al., 2000). The concentrations and distribution of ions can

directly cause adverse effects or indirectly alter toxicity of co-occurring elements due to

competition for ions at active sites (Di Toro et al., 2001). The ionic balance (ratio of the

summation of cations to anions [as meq/L]) in OSPW is near neutral, ranging from 0.85

+ - 2+ - 2- to 1.0. Relative toxicity for major ions is K > HCO3 ~Mg > Cl > SO4 , and ion

deficiencies can be as detrimental to aquatic organisms as excessive concentrations

(Goodfellow et al., 2000). Sodium is introduced to OSPW by the use of NaOH in the

caustic hot water extraction process, and is the predominant cation in OSPW (Allen,

2008a). Sodium ion (Na+) concentration ranged from 150 to 364 mg/L; however, Na+

ions are generally not a major contributor to aquatic toxicity as compared to the

associated Cl- anion (Mount et al., 1997; Goodfellow et al., 2000). Chloride ion

concentration are ~240 mg/L in OSPW, which at the maximum concentrations observed,

exceed recommended water quality guidelines (CEQG guidance of 120 mg/L). Chloride

ions in OSPW are elevated as compared to regional background (Athabasca River concentrations range from 2 to 50 mg/L). Based on OSPW concentrations of major ions present in OSPW, chloride ions are at concentrations exceeding ambient water quality criteria and are identified as a constituent of concern (CCME, 2007; ESRD, 2014).

Verbeek et al. (1994) indicated that major cations and anions were not significantly

contributing to toxicity following bench-scale toxicity identification evaluations of

OSPW.

2.4.1.5 General water characteristics

In terms of general water characteristics, OSPW is well buffered (alkalinity range

- from 320 to 340 mg/L as CaCO3) due to the concentration of bicarbonates (HCO3 ; range

from 300 to 320 mg/L; Table 2.3). Based on presence of bicarbonates, pH of OSPW is

relatively stable, ranging from 7.91 to 8.45. OSPW hardness and conductivity range from

160 to 178 mg/L (as CaCO3) and 1791 to 1800 µS/cm, respectively. Total suspended

solids range from 130 to 400 mg/L in OSPW, therefore exceeding narrative WQC (e.g.,

based on the potential for increasing background receiver turbidity; CCME, 2007; ESRD,

2014).

2.4.2 Toxicity of OSPW

In this study, a fish (P. promelas), two aquatic invertebrates (C. dubia and D. magna), and seedlings from a rooted macrophyte (T. latifolia) were used to evaluate toxicity of OSPW (Table 2.5). In terms of species sensitivities, P. promelas and C. dubia were more sensitive to exposures of OSPW as compared to D. magna and T. latifolia.

Lowest observed effect concentrations (LOECs), expressed as percentage of OSPW, were

50% and 25% for larval P. promelas (7 d; biomass) and C. dubia (7-8 d; reproduction), respectively. In undiluted OSPW, mortality was not observed for 48 h D. magna. The freshwater aquatic macrophyte seedlings were not as sensitive to OSPW as fish and aquatic invertebrates. T. latifolia seedling root growth was not adversely affected in 7 d exposures (LOEC >100% OSPW).

Availability of exposure-response data for an array of taxonomic groups provides a unique opportunity to integrate information, allowing general inferences regarding

“diagnostic” effects implicating sources of toxicity. Direct comparisons should be made with caution, as differences in exposures (e.g., duration of tests, renewals, etc.) may complicate comparisons. Data from this study indicate that sensitivities of sentinel aquatic organisms to ABS-OSPW were: fish ≥ invertebrates > macrophytes. Relative sensitivities of organisms observed in this study imply that the toxicity is attributed to organic fractions (i.e. petroleum hydrocarbons, organic acids; Verbeek et al., 1994;

Morandi et al., 2015). In general, Cladocera (i.e. C. dubia and D. manga) are more sensitive to cationic metals as compared to fish (P. promelas; USEPA, 2002; Suter and

Tsao 1996). Interestingly, OSPW did not adversely affect D. magna (no mortality present in 100% OSPW), implying metals and metalloids are within the environmental tolerances of these aquatic invertebrates for chronic effects.

Although comparisons among other OSPWs from the AOS is challenging due to differences in OSPW composition (i.e. influenced by extraction processes, age, etc.), some general observations can be made from other studies. Relative species sensitivities to OSPW are generally consistent with observations among OSPWs produced in the AOS region. Other studies have indicated that fish are more sensitive to OSPW as compared to

D. magna (Verbeek et al., 1994; Zubot et al., 2012). Zubot and coauthors measured toxicity of rainbow trout (96 h toxicity tests) exposed to OSPWs (produced using caustic soda extraction process stored in settling basins) and observed 100% mortality for fish exposed to OSPW, with a LC50 of 35% (vol% of “raw” water). In addition, Verbeek et

al. (1994) observed that rainbow trout bioassays (96 h) were seven times more sensitive

(in terms of mortality endpoints) to OSPW as compared to D. magna (48 h exposures;

Verbeek et al., 1994). In this study, C. dubia were slightly more sensitive to constituents contained in OSPW as compared to fish (P. promelas). Sensitivity differences among

Cladocera (C. dubia and D. manga) to constituents in OSPWs is surprising. One possible explanation is differences in exposure durations (48 h [D. magna] as compared to 6-8 d

[C. dubia]). Based on results from the OSPW toxicity testing data, C. dubia were chosen as the sensitive species for conducting PBMs. C. dubia were selected based on their sensitivity to OSPW in terms of lethal and sub-lethal (fecundity) endpoints (Table 2.5).

2.4.3 Process-based Manipulations (PBMs)

The purpose of PBMs was to use bench-scale treatments to remove fractions of compounds analogous to biogeochemical processes used for designing and implementing processes in CWTS (Rodgers and Castle, 2008). In this study, PBMs were evaluated using C. dubia. Initial (pre-treated) OSPW resulted in 76% C. dubia mortality, and a statistically significant (p<0.005; α = 0.05) decrease in reproduction. Following H202 +

UV254 treatments (targeting oxidizable organics in OSPW), toxicity to C. dubia was eliminated (96% C. dubia survival, 7 d), with no statistically significant differences in reproduction as compared to a laboratory control (Figure 2.3). Additionally, GAC treatments eliminated toxicity to C. dubia, in terms of both mortality and reproduction (7 d survival was 100%; no statistically significant differences [p=0.68; α =0.05] in reproduction as compared to laboratory control). Filtered OSPW decreased mortality (7 d survival was 86%); however, reproduction was still impaired (<2 neonates per live adult

as compared to an average of 34 neonates per adult in untreated control). EDTA treated

OSPW did not exhibit any measurable change in toxicity to C. dubia as compared to

untreated water (Figure 2.3).

Results from PBMs implicate non-polar or slightly polar organics as contributing

to toxicity of OSPW. Decreased C. dubia toxicity following H2O2 + UV254 treatments in

this study parallels decreases in toxicity from other studies using advanced oxidation

treatments (i.e. ozonation and photocatalysis) to alter toxicity of OSPWs (Scott et al.,

2008; El-Din et al., 2011; Anderson et al., 2012; He et al., 2012; Leshuk et al., 2016).

GAC treatments used in this study support observations from the coupled H2O2 + UV254 treatments targeting organic fractions of OSPW. Other studies have used adsorbents to target the non-polar fraction of OSPW in attempts to alter exposures of organic constituents to mitigate toxicity (McTernan et al., 1986; Marr et al., 1996; Zubot et al.,

2012). Zubot et al. (2012) observed decreases in toxicity to rainbow trout and Microtox™ test (bioluminescent bacteria; Vibrio fischeri) following petroleum coke treatments of

OSPW, attributing reduction in toxicity to the adsorption of NAs (removing 91%; 75 mg

NA/L initial to 5.7 mg/L post-treatment).

Verbeek et al. (1994) identified non-polar compounds and surfactants (organic acids) contributed 100% of the observed toxicity in “fresh” tailings OSPW collected from the AOS region. Although these “fractions” of non-polar organic and surfactants contributing to toxicity contain a myriad of bitumen-derived constituents, more recent efforts have been made to identify species of these fractions using high resolution analytical techniques (Morandi et al., 2015). Morandi et al. (2015) identified NAs (i.e.

O2- species) as among the most toxic fractions of OSPW contained in an end-pit lake in

the AOS. Results from this study and others (Verbeek et al., 1994, Morandi et al., 2015) indicate minimal contributions from elemental fractions to the observed toxicity of

OSPWs. Although numeric exceedances of elemental constituents (i.e. Al, B, Cu, Fe, Pb,

Ni, Se, and Zn) were observed in this study, results from the PBMs indicate that

elemental constituents are in concentrations or forms that limit their bioavailability. Data

from PBMs in this study, in context with the identified COCs in OSPW, provide useful

insight to processes needed in passive and semi-passive systems that could be used to alter exposures and mitigate ecological risks.

2.4.4 Implications to Constructed Wetland Treatment System Design

The goal of this risk-based approach was to characterize a site-specific OSPW to identify specific COCs needing treatment and informing CWTS processes for altering risks to aquatic biota. Clearly, a robust treatment strategy is needed for OSPW based on physical, chemical, and toxicological characteristics of these complex waters. To be economically viable, CWTS must be passive (i.e. low energy demand) due to the volume of OSPWs currently stored in the AOS (estimated 975 million m3; Alberta Department of

Energy, 2013).

Based on the COCs and toxicity data for OSPW, clearly the mitigation strategy

should focus on the organic fraction (e.g., non-polar organics and NAs) to achieve

narrative discharge criteria (i.e. “no toxics in toxic amounts”). Treatment wetlands have

been successfully designed to mitigate a variety of waters containing organic constituents

(e.g., oil field produced water, natural gas storage produced water; Knight et al., 1999;

Pham et al., 2011; Horner et al., 2012). Organic constituents are microbially transformed in treatment wetlands due to the presence of organisms and enzymes capable of degrading organics (Knight et al., 1999; Pham et al., 2011; Horner et al., 2012).

Treatment wetlands designed to mitigate oil field produced waters decreased O/G concentrations by > 98% with initial concentrations between 10 and 100 mg/L (Horner et al., 2012). In addition, microbial degradation processes could be coupled with advanced oxidation to increase biodegradability of recalcitrant organic fractions (i.e. NAs; Metcalf and Eddy 2003; Martin et al., 2010; El-Din et al., 2011; Shi et al., 2015; Vaiopoulou et al., 2015). Advanced oxidation is a promising pathway that has demonstrated environmentally relevant rates and extents of degradation of NAs in OSPW (Martin et al.,

2010; Vaiopoulou et al., 2015) and has decreased toxicity of NAs extracted from OSPW

(Scott et al., 2008; Anderson et al., 2012; He et al., 2012; Shi et al., 2015). Hybrid wetland treatment approaches such as coupling advanced oxidation and biodegradation may offer greater rates of removal for recalcitrant organic fractions (i.e. NAs) contained in OSPW and flexibility (in terms of footprint) in wetland design and placement.

PBMs of OSPW did not indicate cationic metals were a source of toxicity; however, excursions in numeric criteria indicate potential for ecological risk. Therefore, wetland processes to reduce aqueous concentrations of inorganic COCs (i.e. Al, B, Cu,

Fe, Pb, Ni, Se, and Zn) need to be promoted. Treatment wetlands can be designed to provide pathways that enable transformation and/or transfer of metals and metalloids to stable chemical forms, limiting their mobility, bioavailability, and re-distribution

(solubility over time), thus enabling discharge of OSPW. Biogeochemical processes

targeted in wetland treatment systems for altering aqueous concentrations and bioavailability of elements include: sulfide sequestration (Huddleston et al., 2008;

Murray-Gulde et al., 2008; Rodgers and Castle, 2008), formation of Fe- and Mn- oxyhydroxide co-precipitates (Rodgers and Castle, 2008; Schwindaman et al., 2014),

microbial transformations (e.g., Se, As; Spacil et al., 2011; Lizma et al., 2011), and

complexation with organic ligands (e.g., Al complexation with fulvic and humic acids;

Driscol and Postek, 1995). These processes have been used successfully in treatment

wetlands to achieve performance goals (i.e. numeric and narrative discharge criteria) for

energy-derived waters (e.g., refinery effluents, oilfield produce waters, acid and neutral

mine drainage; Mooney and Murray-Gulde, 2008; Haakensen et al., 2015).

Microbially-driven processes in wetlands (as described above) depend on the availability of macro- and micro-nutrients; therefore, concentrations of nitrogen (as

ammonia; 0.13 to 0.99 mg/L) and phosphorus (as total phosphorus; 0.037 to 0.62 mg/L)

in OSPW are necessary for facilitating treatment of some COCs (e.g., NAs, O/G). In

addition, wetlands have been used extensively to treat excess nutrients (nitrogen [as

ammonia, nitrate, and nitrite], total phosphorus; Gersberg et al., 1986; Braskerud et al.,

2002; Kadlec and Wallace, 2008).

2.5 Conclusions

In this study, COCs were identified using a risk-based approach, including

chemical, physical and toxicological characterization. COCs identified in OSPW include

organics (NAs, O/G), metals/metalloids, and suspended solids. Toxicity testing confirmed that COCs were in sufficient forms and concentrations to have measurable

adverse effects on sentinel aquatic species. Sensitivities of aquatic organisms to OSPW indicated that fish ≥ aquatic invertebrates > macrophytes. The sensitivity distribution of

organisms to OSPW in addition to strategic bench-scale manipulations indicate organic

constituents are contributing to the observed toxicity. Alteration of the organic fraction of

OSPW (i.e. H2O2 + UV254 and GAC treatments) significantly increased survival and reproduction of C. dubia as compared to untreated OSPW. Based on multiple lines of

evidence, these data indicate that organic fractions (i.e. O/G, NAs) of OSPW are sources

of toxicity. In addition, metals and metalloids in OSPW exceeding numeric guidelines

indicates the need to decrease concentrations to achieve WQC thresholds. Results from

this study provide critical information to inform mitigation strategies using passive or

semi-passive treatment processes (i.e. constructed treatment wetlands) to mitigate

ecological risks of OSPW to aquatic organisms. The approach used in this study could be

applied to other site-specific and compositionally complex process-affected waters (i.e.

OSPWs and consolidated tailings) located in the AOS region.

2.6 Acknowledgements

Funding for this research was provided by Shell Canada Ltd. and Suncor Energy. We are

grateful to our colleagues at Shell Canada Ltd. for procuring source water for this

research. The authors are also grateful to Dr. Wayne Chao of Clemson University for

providing analytical support.

2.7 References

Adelman I.R., Smith Jr. L.L., (1976). Standard test fish development. Part I. Fathead minnows (Pimephales promelas) and goldfish (Carassius auratus) as standard fish in bioassays and their reaction to potential reference toxicants. EPA-600/3- 76-061A, USEPA, Duluth, MN. pp. 77.

Alberta Environment & Sustainable Resource Development (ESRD). (2012). Lower Athabasca Region Surface Water Quality Management Framework for the Lower Athabasca River. Water Policy Branch, Policy Division. Edmonton, AB, CA.

Alberta Environment & Sustainable Resource Development (ESRD). (2014). Environmental Quality Guidelines for Alberta Surface Waters. Water Policy Branch, Policy Division. Edmonton, AB, Canada, 4 pp. 48.

Allen, E.W., (2008a). Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Journal of Environmental Engineering Science 7:123-138.

Allen, E.W., (2008b). Process water treatment in Canada’s oil sands industry: II. A review of emerging technologies. Journal Environmental Engineering and Science 7:499-524.

Anderson, J.C., Wiseman, S.B., Wang, N., Moustafa, A., Perez-Estrada, L., Gamal El- Din, Giesy, J.P., (2011). Effectiveness of ozonation treatment in eliminating toxicity of oil sands process-affected water to Chironomus dilutus. Environmental science & technology 46(1): 486-493.

APHA, American Water Works Association, and Water Pollution Control Federation. (2012). Standard Methods for examination of water and wastewater, 20th ed. Washington, D.C.: American Public Health Association.

Armstrong, S.A., Headley, J.V., Peru, K.M., Germida, J.J., (2008). Phytotoxicity of oil sands naphthenic acids and dissipation from systems planted with emergent aquatic macrophytes. Journal of Environmental Science and Health, Part A 43(1): 36-42.

Barrow, P.A., Peru, K.M., McMartin, D.W., Headley, J.V. (2016). Effects of extraction pH on the fourier transform ion cyclotron resonance mass spectrometry profiles of Athabasca oil sands process water. Energy Fuels 30 (5):3615-3612.

Biesinger, K.E., Christensen, G.M., Fiandt, J.T., (1986). Effects of metal salt mixtures on Daphnia magna reproduction. Ecotoxicology and environmental safety 11(1): 9- 14.

Birge, W.J., (1977). Aquatic toxicology of trace elements of coal and fly ash. In Energy and Environmental Stress in Aquatic Systems, Selected Papers from a Symposium, held at Augusta, Georgia November 2-4, 1977. CONF-771114. pp. 219-240.

Boucher, A.M., Watzin, M.C., (1999). Toxicity identification evaluation of metal‐contaminated sediments using an artificial pore water containing dissolved organic carbons. Environmental Toxicology and Chemistry 18(3): 509-518.

Braskerud, B.C., (2002). Factors affecting phosphorus retention in small constructed wetlands treating agricultural non-point source pollution. Ecological Engineering 19(1): 41-61.

Brown, L.D., Ulrich, A.C., (2015). Oil sands naphthenic acids: A review of properties, measurement, and treatment. Chemosphere 127:276-290.

Canadian Council of Ministers of the Environment (CCME). (2011). Canadian water quality guidelines for the protection of aquatic life. Canadian environmental quality guidelines, 1999. Canadian Council of Ministers of the Environment, Winnipeg, MB, Canada.

Cumulative Environmental Management Association (CEMA). (2014). Guidelines for wetlands establishments on reclaimed oil sands leases. 3rd Ed. Alberta. Fort McMurray, AB.

Dominski, M., (2007). Surface mined oil sand: tailings practices, performance, and projections. Alberta Energy and Utilities Board. In Proceedings of the 3rd International Heavy Oil Conference, March 5–7, 2007, Calgary, Alta. Alberta Energy and Utilities Board.

Driscoll, C.T., Postek, K.M., (1995). The chemistry of aluminum in surface waters. The Environmental Chemistry of Aluminum. Lewis Publishers, Chelsea, MI, pp. 363- 418.

El-Din, M.G., Fu, H., Wang, N., Chelme-Ayala, P., Pérez-Estrada, L., Drzewicz, P., Smith, D.W., (2011). Naphthenic acids speciation and removal during petroleum- coke adsorption and ozonation of oil sands process-affected water. Science of the Total Environment 409(23): 5119-5125.

Environment Canada. (1996). Biological test method: acute lethality test using Daphnia spp. Report EPS 1/RM/11. Ottawa, Ontario, Canada: Environmental Science and Technology Centre, Science and Technology Branch.

Environment Canada. (2007). Biological test method: test of reproduction and survival using the cladoceran Ceriodaphnia dubia. Report EPS 1/RM/21. Ottawa, Ontario, Canada: Environmental Science and Technology Centre, Science and Technology Branch.

Environment Canada. (2011). Biological test method: test of larval growth and survival using fathead minnows. Report EPS 1/RM/22. Ottawa, Ontario, Canada: Environmental Science and Technology Centre, Science and Technology Branch.

Frank, R.A., Kavanagh R., Burnison B.K., Arsenault G., Headley J.V., Peru K.M., Van Der Kraak G., Solomon K.R., (2008). Toxicity Assessment of Collected Fractions from an Extracted Naphthenic Acid Mixture. Chemosphere 72: 1309-1314

Freeman, R.A., & Everhart, W.H., (1971). Toxicity of aluminum hydroxide complexes in neutral and basic media to rainbow trout. Transactions of the American Fisheries Society 100(4): 644-658.

Galarneau E., Hollebone B.P., Yang Z., Schuster J., (2014). Preliminary measurement- based estimates of PAH emissions from oil sands tailings ponds. Atmospheric Environment 97: 332-335.

Gamal El-Din M., Fu M., Wang N., Chelme-Ayala P., Perez-Estrada L., Drzewicz P., Martin J.W., Zubot W., and Smith D.W., (2011). Naphthenic acids speciation and removal during petroleum-coke adsorption and ozonation of oil sands process- affected water. Science of the Total Environment 409: 5119-5125.

Gersberg, R.M., Elkins, B.V., Lyon, S.R., & Goldman, C.R., (1986). Role of aquatic plants in wastewater treatment by artificial wetlands. Water Research 20(3): 363- 368.

Goodfellow, W.L., Ausley, L.W., Burton, D.T., Denton, D.L., Dorn, P.B., Grothe, D.R., Rodgers, Jr., J.H., (2000). Major ion toxicity in effluents: a review with permitting recommendations. Environmental Toxicology and Chemistry 19(1): 175-182.

Grewer, D.M., Young, R.F., Whittal, R.M., & Fedorak, P.M., (2010). Naphthenic acids and other acid-extractables in water samples from Alberta: what is being measured?. Science of the Total Environment 408(23): 5997-6010.

Haakensen, M., Pittet, V., Spacil, M.M., Castle, J.W., & Rodgers Jr, J.H., (2015). Key aspects for successful design and implementation of passive water treatment systems. Journal of Environmental Solutions for Oil, Gas, and Mining 1(1): 59- 81.

He, Y., Wiseman, S.B., Zhang, X., Hecker, M., Jones, P.D., El-Din, M.G., Giesy, J.P., (2010). Ozonation attenuates the steroidogenic disruptive effects of sediment free oil sands process water in the H295R cell line. Chemosphere 80(5): 578-584.

Headley, J.V., Peru, K.M., Barrow, M.P., (2015). Advances in mass spectrometric characterization of naphthenic acids fraction compounds in oil sands environmental samples and crude oil—a review. Mass spectrometry reviews 35(2): 311-328.

Herman, D.C., Fedorak, P.M., MacKinnon, M.D., & Costerton, J.W., (1994). Biodegradation of naphthenic acids by microbial populations indigenous to oil sands tailings. Canadian Journal of Microbiology 40(6): 467-477.

Horner, J.E., Castle, J.W., Rodgers Jr, J.H., Gulde, C.M., & Myers, J.E., (2012). Design and performance of pilot-scale constructed wetland treatment systems for treating oilfield produced water from Sub-Saharan Africa. Water, Air, & Soil Pollution 223(5): 1945-1957.

Holowenko, F.M., MacKinnon, M.D., Fedorak, P.M., (2002). Characterization of naphthenic acids in oil sands wastewaters by gas chromatography-mass spectrometry. Water Resources 36: 2843-2855.

Hoke, R.A., Gala, W.R., Drake, J.B., Giesy, J.P., Flegler, S., (1992). Bicarbonate as a potential confounding factor in cladoceran toxicity assessments of pore water from contaminated sediments. Canadian. Journal of Fisheries Aquatic Science 49: 1633-1640.

Huang, R., Sun, N., Chelme-Ayala, P., McPhedran, K. N., Changalov, M., & El-Din, M. G., (2015). Fractionation of oil sands-process affected water using pH-dependent

extractions: A study of dissociation constants for naphthenic acids species. Chemosphere 127:291-296.

Huddleston, III, G.M., & Rodgers Jr, J.H., (2008). Design of a constructed wetland system for treatment of copper-contaminated wastewater. Environmental Geosciences 15(1): 9-19.

Jones, D., West, C. E., Scarlett, A. G., Frank, R. A., & Rowland, S. J., (2012). Isolation and estimation of the ‘aromatic’ naphthenic acid content of an oil sands process- affected water extract. Journal of Chromatography A 1247: 171-175.

Kadlec, R.H., Wallace, S., (2008). Treatment wetlands. CRC press. Boca Raton, Fl.

Keithly, J., Brooker, J.A., Deforest, D.K., Wu, B.K., Brix, K.V., (2004). Acute and chronic toxicity of nickel to a cladoceran (Ceriodaphnia dubia) and an amphipod (Hyallela azteca). Environmental Toxicology and Chemistry 23: 691-696.

Kimball, G., (1978). The effects of lesser known metals and one organic to fathead minnows (Pimephales promelas) and Daphnia magna. Manuscript, Department of Entomology, Fisheries and Wildlife, University of Minnesota, Minneapolis, MN. pp. 88.

Kinley, C.M., McQueen, A.D., Rodgers, Jr., J.H., (2016). Comparative responses of freshwater organisms to exposures of a commercial naphthenic acid. Chemosphere 153: 170-178.

Knight, R.L., Kadlec, R.H., Ohlendorf, H.M., (1999). The use of treatment wetlands for petroleum industry effluents. Environmental Science and Technology 33: 973- 980.

Lai, J.W., Pinto, L.J., Bendell‐Young, L.I., Moore, M.M., & Kiehlmann, E., (1996). Factors that affect the degradation of naphthenic acids in oil sands wastewater by indigenous microbial communities. Environmental Toxicology and Chemistry 15(9): 1482-1491.

Leclair, L.A., MacDonald, G.Z., Phalen, L.J., Köllner, B., Hogan, N.S., & van den Heuvel, M.R., (2013). The immunological effects of oil sands surface waters and naphthenic acids on rainbow trout (Oncorhynchus mykiss). Aquatic Toxicology 142: 185-194.

Leshuk, T., Wong, T., Linley, S., Peru, K. M., Headley, J. V., Gu, F., (2016). Solar photocatalytic degradation of naphthenic acids in oil sands process-affected water. Chemosphere 144: 1854-1861.

Lizama, K., Fletcher, T.D., Sun, G., (2011). Removal processes for arsenic in constructed wetlands. Chemosphere 84(8): 1032-1043.

MacKinnon, M.D., Boerger, H., (1986). Description of two treatment methods for detoxifying oil sands tailings pond water. Water Pollution Research Journal of Canada 21:496–512.

Mahaffey, A., Dube, M., (2016). Review of the composition and toxicity of oil sands process-affected water. Environmental Review. doi: 10.1139/er-2015-0060

Marentette, J.R., Frank, R.A., Bartlett, A.J., Gillis, P.L., Hewitt, L.M., Peru, K.M., Parrott, J.L., (2015). Toxicity of naphthenic acid fraction components extracted from fresh and aged oil sands process-affected waters, and commercial naphthenic acid mixtures, to fathead minnow (Pimephales promelas) embryos. Aquatic Toxicology 164: 108-117.

Marr, J.B., Mohammed, M., MacKinnon, M., Rogers, M., (1996). Removal of naphthenic acid from water via pH adjustment, coagulation and sand filtration. In Proceedings of the International Water Conference, Engineer’s Society of Western Pennsylvania, Pittsburgh, Pa pp. 738-754

Martin, J.W., Barri, T., Han, X., Fedorak, P. M., El-Din, M. G., Perez, L., Jiang, J.T., (2010). Ozonation of oil sands process-affected water accelerates microbial bioremediation. Environmental science & technology 44(21): 8350-8356.

McTernan, W.F., Blanton, W.E., Boardman, G.D., Nolan, B.T., & Kocornik, D.J., (1986). Polymer flotation and activated carbon adsorption treatment for in situ tar sand process water. Environmental progress 5(3): 154-158.

McQueen, A.D., Kinley, C.M., Kiekhaefer, R.L., Calomeni, A.J., Rodgers Jr, J.H., Castle, J.W., (2016). Photocatalysis of a Commercial Naphthenic Acid in Water Using Fixed-Film TiO2. Water, Air, & Soil Pollution 227(5): 1-11.

Metcalf, E., (2003). Wastewater Engineering: Treatment, Disposal, Reuse, Metcalf & Eddy. Inc., McGraw-Hill, New York.

Mikula R., (2013). Trading water for oil: Tailings management and water use in surface- mined oil sands. In Hein FJ, Leckie D, Larter S, and Suter JR (eds.). Heavy-oil

and oil-sand petroleum systems in Alberta and beyond: AAPG Studies in Geology. 64:689-699.

Mooney, F.D., & Murray-Gulde, C., (2008). Constructed treatment wetlands for flue gas desulfurization waters: full-scale design, construction issues, and performance. Environmental Geosciences 15(3): 131-141.

Moore, M.T., Huggett, D.B., Huddleston III, G.M., Rodgers Jr., J.H., Cooper, C.M., (1999). Herbicide effects on Typha latifolia (Linneaus) germination and root and shoot development. Chemosphere 38: 3637-3647.

Mount, D.I., (1968). Chronic toxicity of copper to fathead minnows (Pimephales promelas, Rafinseque). Water Research 2: 215-223.

Mount, D., Gulley, D.D., Hockett, J.R., Garrison T.D., Evans J.M., (1997). Statistical models to predict the toxicity of major ions to Ceriodaphnia dubia, Daphnia magna, and Pimephales promelas (fathead minnows). Environmental Toxicology and Chemistry 16: 2009-2019.

Murray-Gulde, C.L., Bridges, W.C., Rodgers Jr, J.H., (2008). Evaluating performance of a constructed wetland treatment system designed to decrease bioavailable copper in a waste stream. Environmental Geosciences 15(1): 21-38.

Nero V., Farwell A., Lee L.J., Van Meer T., MacKinnon M.D., and Dixon D.G., (2006). The effects of salinity on naphthenic acid toxicity to yellow perch: Gill and liver histopathology. Ecotoxicology and Environmental Safety 65:252-264.

Oil sands regional aquatics monitoring program (RAMP) 2001. Vol. 1: Chemical and biological monitoring. Report prepared for the RAMP Steering Committee by Golder Associates Ltd., Fort McMurray, AB.

Pham, M.P.T., Castle, J.W., Rodgers Jr, J.H., (2011). Biogeochemical process approach to the design and construction of a pilot-scale wetland treatment system for an oil field-produced water. Environmental Geosciences 18(3): 157-168.

Pickering, Q.H., Henderson, C., (1966). The acute toxicity of some heavy metals to different species of warm water fishes. International Journal of Air and Water Pollution 10: 453-463.

Province of Alberta. (2014). Environmental Protection and Enhancement Act. Revised Statues of Alberta 2000. Chapter E-12. Division 1, Section 2.

Rodgers, Jr, J.H., & Castle, J.W., (2008). Constructed wetland systems for efficient and effective treatment of contaminated waters for reuse. Environmental Geosciences 15(1): 1-8.

Rogers, V.V., Wickstrom, M., Liber, K., MacKinnon, M.D., (2002). Acute and subchronic mammalian toxicity of naphthenic acids from oil sands tailings. Toxicological Sciences 66:347–355.

Schramm L.L., (Ed.). (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press. N.Y.

Schwindaman,, J.P., Castle, J.W., Rodgers, J H., (2014). Fate and distribution of arsenic in a process-designed pilot-scale constructed wetland treatment system. Ecological Engineering 68: 251-259.

Shi, Y., Huang, C., Rocha, K. C., El-Din, M. G., & Liu, Y., (2015). Treatment of oil sands process-affected water using moving bed biofilm reactors: with and without ozone pretreatment. Bioresource technology 192: 219-227.

Soucek, D.J., & Kennedy, A.J., (2005). Effects of hardness, chloride, and acclimation on the acute toxicity of sulfate to freshwater invertebrates. Environmental Toxicology and Chemistry 24(5): 1204-1210.

Spacil, M.M., Rodgers, Jr., J.H., Castle J.W., Murray-Gulde C.L., Myers, J.E., (2011). Treatment of selenium in simulated refinery effluent using a pilot-scale constructed wetland treatment system. Water, Air, & Soil Pollution 221(1-4): 301- 312.

Spacil M.M., Rodgers, Jr., J.H., Castle, J.W., Chao, W.Y., (2011). Performance of a pilot scale constructed wetland treatment system for selenium, arsenic, and low- molecular-weight organics in simulated fresh produced water. Environmental Geosciences 18:145-156.

Suedel, B.C., Rodgers J.H. Jr., Deaver, E., (1997). Experimental factors that may affect toxicity of cadmium to freshwater organisms, Archives of Environmental Contamination and Toxicology 33: 188-193.

Suter, G.W., Tsao, C L., (1996). Toxicological benchmarks for screening potential contaminants of concern for effects on aquatic biota: 1996 revision (No. ES/ER/TM--96/R2). Oak Ridge National Lab., TN (United States).

Thurston, R.V., Chakoumakos, C., Russo, R.C., (1981). Effect of fluctuating exposures on the acute toxicity of ammonia to rainbow trout (Salmo gairdneri) and cutthroat trout (S. clarki). Water Research, 15(7), 911-917.

United States Environmental Protection Agency (USEPA). (1991). Technical support document for water quality-based toxics control. EPA/505/2-90-001. Washington DC.

United States Environmental Protection Agency (USEPA). (2002). Short-term Methods for Estimating Chronic Toxicity of Effluents and Receiving Water to Freshwater Organisms, EPA-821-R-02-013. Washington, DC.

United States Environmental Protection Agency (USEPA). (1999). National recommended water quality criteria-Correction. Office of Water 4304. EPA 822- Z-99–001. Washington, DC.

United States Environmental Protection Agency (USEPA). (2001). EPA Method 200.7, Trace elements in water, solids, and biosolids by inductively coupled plasma- atomic emission spectrometry, Revision 5.0. EPA-821-R-01-010. Washington, DC.

United States Environmental Protection Agency (USEPA). (2002). Methods for measuring acute toxicity of effluents and receiving waters to freshwater and marine organisms. EPA-821-R-02-12. Washington, DC.

Yen, T.W., Marsh, W.P., MacKinnon, M.D., Fedorak, P.M., (2004). Measuring naphthenic acids concentrations in aqueous environmental samples by liquid chromatography. Journal of Chromatography 1033: 83-90.

Vaiopoulou, E., Misiti, T.M., & Pavlostathis, S.G., (2015). Removal and toxicity reduction of naphthenic acids by ozonation and combined ozonation-aerobic biodegradation. Bioresource technology 179: 339-347.

Verbeek, A.G., Mackay, W.C., MacKinnon, M.D., (1994). Acute toxicity of oil sands wastewater: A toxic balance. In Schramm LL. (Ed.). (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press. N.Y.

Zubot, W., MacKinnon, M.D., Chelme-Ayala, P., Smith, D.W., El-Din, M.G., (2012). Petroleum coke adsorption as a water management option for oil sands process- affected water. Science of the Total Environment 427: 364-372.

Table 2.1 Analytical methods for site-specific OSPW.

Method Detection Parameter Methods Limit Direct Instrumentation: Orion Model 420A pH (Standard Methods 4500-H+ B) (APHA, 0.01 SU 2012) Direct Instrumentation: YSI 30 (Standard Method 2510 B) (APHA, Conductivity 0.1 μS/cm 2012) Alkalinity Standard Methods: 2320 B (APHA, 2012) 2 mg/L as CaCO3 Hardness Standard Methods: 2340 B (APHA, 2012) 2 mg/L as CaCO3 TSS Standard Methods: 2540 D (APHA, 2012) 0.1 mg/L TDS Standard Methods: 2540 C (APHA, 2012) 0.1 mg/L USEPA Method 1664 A (Environmental Express StepSaver Oil and grease (O/G) ~2 mg/L Modification) (USEPA, 1999) Cations/ Anions Dionex ISC-2000 Ion Chromatograph (APHA, 2012) 2-7 mg/L Total Ammonia Standard Methods: 4500-NH3 (APHA, 2012) 0.01 mg/L Total Phosphorous Standard Methods: 4500-P (APHA, 2012) 0.01 mg/L Inductively Coupled Plasma-Atomic Emissions Spectrometry (ICP-AES): Metals and Metalloids (mg/L) 200.7 (USEPA, 2001); 0.045 Aluminum (Al) Iron (Fe) 0.006 a a Arsenic (As) 0.0006 Lead (Pb) 0.0002

Barium (Ba) 0.003 Manganese (Mn) 0.002

Boron (B) 0.006 Molybdenum (Mo) 0.012 a Cadmium (Cd) 0.0002 Nickel (Ni) 0.015 a a Chromium (Cr) 0.004 Selenium (Se) 0.002 a a Cobalt (Co) 0.002 Silver (Ag) 0.0002 a 0.003 Copper (Cu) Zinc (Zn) 0.002 Naphthenic Acids (NAs) HPLC; Derivatization based on Yen et al., 2004 5 mg/L Orbitrap MS; Headley et al., 2015; Leshuk et al., 2016 1 mg/L aMDL represents samples which were concentrated using heat (60°C) assisted evaporative loss

Table 2.2 Manipulations and treatment objectives of process based manipulations

(PBMs) performed on OSPW.

Process Based Analogous Passive or Semi-passive Manipulation Objective Manipulation (PBM) Treatment Process

Filtrationa Remove suspended solids Precipitation, settling, and sedimentation

Dissimilatory sulfate reduction and sulfide- EDTA chelation Remove cationic metals metal [MeS] precipitation Granular activated Remove non-polar organic Sorption; microbial biotransformation and charcoal (GAC)b compounds biodegradation c Hydrogen peroxide H2O2 Remove compounds susceptible Advanced oxidation (e.g., solar + UV254 to oxidation photocatalysis) a0.45 µm nitrocellulose filter bCharcoal, coconut (6 -14 mesh; Fisher Scientific) cAdvanced oxidation proposed for coupled "hybrid" semi-passive treatment options (Shi et al., 2015; Vaiopoulou et al., 2015; Leshuk et al., 2016; McQueen et al., 2016)

Table 2.3 Comparison of water quality characteristics and organic constituents in OSPW to water quality guidelines (CCME 2007, USEPA 2007; ESRD 2014) and toxicity values for C. dubia (Cd), rainbow trout (O. mykiss [Om]), and fathead minnow (P. promelas [Pp]). COCs (i.e. concentration > guideline) are bolded. Water Quality Guidelines (mg/L) Parameter mg/L OSPW (n=10) Reported Toxicity COC Alberta WQG CEQG USEPA Values (mg/L) (unless noted) Min Max Chronic Chronic Acute Chronic Yes/No pH (SU) 7.91 8.4 6.5-9.0 No

a Alkalinity (CaCO3) 320 340 20 No

Total Ammonia (N) 0.051 0.099 0.17-1.5b 2.9 0.26 LC50: 0.563 (96h Omc) No

Total Phosphorus (TP) 0.037 0.082 0.01-0.02 Yes

Total Suspended Solids 130 400 * * * Yes (TSS) Total Naphthenic Acids (NAs)

NAs (HPLC) 80 128 ** LC50: 51.8 (96h Ppd) Yes NAs (Orbitrap MS) 93 103 ** EC50: 7.5 (96h Ppe)

Oil and grease (O/G) 8 13 * * Yes

- f Bicarbonate (HCO3 ) 300 320 LC50: 995 (48h Cd ) No

Calcium (Ca+) 29 31 LC50: 4630 (96h Ppg) No

Chloride (Cl-) 240 245 120 120 860 230 LC50: 7341 (96h Pph) Yes

Sodium (Na+) 360 364 LC50: 1770 (48h Cdg) No

-2 i Sulfate (SO4 ) 150 175 LC50: 2,078 (96h Cd ) No aMinimum fHoke et al., 1992 bTemperature range from 5-20°C and pH of 8.0-8.5 gMount et al., 1997 cThurston and Russon, 1981; @ 14.4°C and pH of 8.3 hAdelman et al., 1976 dKavanagh et al., 2011; 5 d larvae; NAs quantified by ESI-MS iSoucek and Kennedy, 2005 eMarentette et al., 2015; "fresh" OSPW; embryos; NA quantified by LC/QToF *Narrative statement; no visible sheen or odor; or unreasonable turbidity or color **”No toxics in toxic amounts” Note: NAs quantified by different methods do not necessarily measure the same suite of compounds (Headley et al., 2015)

Table 2.4 Comparison of metals and metalloids in OSPW to water quality guidelines (CCME 2007; USEPA 2007; ESRD 2014) and toxicity values for C. dubia (Cd), D. magna (Dm), fathead minnow (P. promelas [Pp]), and rainbow trout (O. mykiss [Om]). Identified COCs (i.e. concentration > guideline) are bolded. Water Quality Guidelines (mg/L) Parameter mg/L OSPW (n=10) Reported Toxicity Values COC Alberta WQG CEQG USEPA (mg/L) (unless noted) Min Max Chronic Chronic Acute Chronic Yes/No Aluminum (Al) 0.36 10.9 0.1a 0.1a 0.75 0.087 IC42: 0.514 (45d Omd) Yes

Arsenic (As) 0.0006 0.0032 0.005 0.005 0.34 0.15 No

Barium (Ba) 0.21 0.33 EC16: 5.8 (21d Dme) No

Boron (B) 2.09 2.12 1.5 1.5 LC50: 53.2 (21d Dmf) Yes

Cadmium (Cd) <0.0002 <0.0002 0.00016b 0.00038b 0.002 0.0025 LC50: 0.1 (10d Cdg) No

Chromium (Cr) <0.004 <0.004 0.0089 0.0089 0.57 0.074 No

Cobalt (Co) <0.0002 <0.0002 LOEC: 0.009 (28d Dmh) No

Copper (Cu) 0.003 0.12 0.007 0.0024b 0.016b 0.011b LC50: 0.095 (96h Ppi) Yes

Iron (Fe) 1.2 1.5 0.3c 0.3c 1.0 LC50: 3.1 (96h Ppj) Yes

Lead (Pb) <0.0002 0.0014 0.0032b 0.00318b 0.065 0.0025 Yes

Manganese (Mn) 0.137 0.162 LC50: 9.1 (48h Cdk) No

Molybdenum (Mo) 0.041 0.062 0.073 0.073 LC50: 0.73 (28d Oml) No

Nickel (Ni) 0.0056 0.01 0.052 0.0096 LC50: 0.8 (48h Cdm) Yes

Selenium (Se) 0.002 0.004 0.001 0.001 0.0015 Yes

Silver (Ag) <0.0002 <0.0002 0.001 0.00025 No

Zinc (Zn) 0.014 0.113 0.03 0.03 0.12 0.12 LC50: 2.55 (96h Ppj) Yes apH: ≥6.5 hKimball 1978

bHardness: 100 mg/L as CaCO4 iMount, 1968

cDissolved fraction jPickering and Henderson, 1966

dFreeman and Everhart 1971; biomass kBoucher and Watzin, 1999

eBiesinger and Christensen, 1972 lBirge 1978

fOPP, 2000 mKeithly et al. 2004

gSuedel et al. 1997

Table 2.5 Bioassay data for OSPW using fish (P. promelas), aquatic invertebrates (C. dubia and D. magna), and a macrophyte (T. latifolia). Endpoint Estimates Test % Mortality Species Age (as % OSPW) Duration (in 100% OSPW) LC50 LOEC NOEC Fish Fathead minnow (P. promelas) 7 d larval (<24 h) 15 to 20 >100 50a 25a Aquatic Invertebrate Ceriodaphnia dubia 7-8 d neonate (<24 h) 10 to 80 >100 25b 12.5b Daphnia magna 48 h neonate (<24 h) 0 >100

Macrophyte Cattail (T. latifolia) 7 d 2 d seedlings - >100 >100c - aBiomass (µg/live organism) bReproduction (average neonate/adult per day) cNo statistical difference (α = 0.05) observed in root growth

Figure 2.1 Approach for identifying constituents of concern (COCs) in OSPW and informing potential treatment processes to mitigate ecological risk.

Note: “numeric” criteria refer to constituent concentration comparisons to water quality guidelines/criteria and “narrative” criteria refer to “no toxics in toxic amounts” (USEPA, 1991).

60 %

y t i

s 40 n e t n I

e v i

t 20 a l e R

2 S 3 S 4 S 5 S 6 7 S S O 2 O 3 O 4 O 5 O O O 2 O O O O 3 O N N Heteroatom Class

Figure 2.2 Hereroatom classes present in OSPW using high-resolution Orbitrap MS.

100 a i b u d . C , l

a 50 v

i *

v * r u S

0 l 4 5 C d A ro W 2 e T t P V A r n S G te D o O U il E C + F d 2 te 0 a 2 e H tr n U Control a

i Untreated OSPW b

u 40 H202 + UV254 d .

C GAC t l 30 Filtered u d EDTA A / s

e 20 t a n o e

n 10

e g

a * * * r

e 0 v l 4

A 5 C d A ro W 2 e T t P V A r n S G te D o O U il E C + F d 2 te 0 a 2 e H tr n U

Figure 2.3 Responses of Ceriodaphnia dubia, in terms of survival (top) and reproduction

(bottom), in 7-8 d static/renewal tests following process based manipulations (PBMs).

Error bars represent standard error of replicates (n=3); Asterisks indicate significant

differences (p<0.05; α=0.05) from controls.

CHAPTER THREE

PHOTOCATALYSIS OF A COMMERICAL NAPHTENIC ACID IN WATER USING

FIXED-FILM TIO2

Andrew D. McQueen13; Ciera M. Kinley1; Rebecca L. Kiekhaefer2, Alyssa J. Calomeni1; John H. Rodgers Jr.1 James W. Castle2

1Department of Forestry and Environmental Conservation, 261 Lehotsky Hall, Clemson University, Clemson, SC 29634, USA

2Department of Environmental Engineering & Earth Sciences, 445 Brackett Hall, Clemson University, Clemson, SC 29634, USA

3Corresponding author: Andrew McQueen

Manuscript accepted in: Water, Air, and Soil Pollution

3.1 Abstract Photolysis or photocatalysis may provide a process for mitigating ecological risks of naphthenic acids (NAs) contained in energy-derived waters such as refinery effluents and process waters. If effective, fixed-film TiO2 photocatalysis of NAs could decrease

operational expenses as well as capital costs for water treatment. The overall objective of

this study was to measure rates and extents of photolysis and photocatalytic degradation

of commercial NAs using bench-scale fixed-film TiO2 and confirm changes in NA

concentrations using sensitive vertebrate (fish = Pimephales promelas) and invertebrate

(Daphnia magna) species. Specific objectives were to: 1) measure rates and extents of

degradation of commercial (Fluka) NAs throughout an 8-hr duration of natural sunlight

(“photolysis”) and natural sunlight in the presence of fixed-film TiO2 (“photocatalysis”),

and 2) measure changes in toxicity in terms of mortality with sentinel fish and

microinvertebrate species. Bench-scale chambers using thin-film TiO2 irradiated with

natural sunlight were used to measure photocatalysis and HPLC was used to quantify

NAs. After 4-hr in photocatalysis treatments, >92% decline was observed with an

average removal rate of 15.5 mg/L hr-1 and half-live of 2 hours. After 5-hrs of

photocatalysis, there was no measurable NA toxicity for fish (P. promelas) or microinvertebrates (D. magna). Photocatalytic degradation achieved efficacious rates and extents of removal of Fluka NAs and eliminated acute toxicity to sentinel aquatic organisms, indicating the potential for application of this technology for mitigating ecological risks. Coupled with existing treatment processes (i.e. aerobic biodegradation), photocatalysis can augment rates and extents of NA removal from impacted waters.

3.2 Introduction

Naphthenic acids (NAs) are a complex group of organic acids associated with

crude oils (Seifert and Teeter, 1969; Tomcyzk et al., 2001) and energy derived process

waters (Dorn, 1992; Allen, 2008). NAs are generally described by the formula

CnH2n+ZO2, where n is the number of carbons and Z is either zero or a negative even

integer representing the hydrogen deficiency of the molecule due to rings or double

bonds (Holowenko et al., 2002; Clemente and Fedorak, 2005). NAs are sources of

toxicity in energy derived waters such as refinery effluents and oil sands process-affected waters (Dorn, 1992; Schramm, 2000), with adverse effects observed for fish, invertebrates, aquatic macrophytes, and microorganisms (Nero et al., 2006; Frank et al.,

2008; Armstrong et al., 2008; Kavanagh et al., 2012; Leclair et al., 2013; Swigert et al.,

2015). In addition, NAs are relatively persistent in water, with aerobic biodegradation half-lives for NAs in OSPW ranging from months to years (Scott et al., 2005; Han et al.,

2009; Headley et al., 2010). To mitigate risks due to NAs, exposures must be sufficiently altered to eliminate toxicity to aquatic organisms. To effectively alter exposures of NAs, potential transformation pathways with reasonable rates and extents of removal should be

thoroughly evaluated.

Naphthenic acids are susceptible to photolysis (USEPA, 2012; Headley et al.,

2009), and the rate of this transformation process can be enhanced using catalysts (e.g.

titanium dioxide [TiO2]). TiO2 is widely used as a photocatalyst due to long term photo-

stability, relative effectiveness, and stability in acidic and oxidative conditions (Bagheri

® ® et al., 2014). Degussa and Aeroxide P25 are commercial forms of TiO2 previously

used in photocatalytic studies due to their relatively large surface area (~50 m2/g) and high ratio (4:1) of anatase to rutile (Wold, 1993). Photocatalytic degradation of NA has been accomplished using TiO2 in aqueous suspensions (i.e. “slurries”) with both artificial ultraviolet (UV) irradiation (e.g. UV254 lamps) and natural sunlight (McMartin et al.,

2004; Headley et al., 2009; Mishra et al., 2010). Bench-scale studies have demonstrated that rates of NA degradation using photocatalysis are of practical significance (in terms of scalability), with half-lives achieved in hours (Headley et al., 2009; Mishra et al.,

2010). However, some photocatalytic design features may limit full-scale application.

Post-treatment recovery of aqueous suspensions of TiO2 particles may be challenging

(Kinsinger et al., 2015) and artificial UV is energy intensive (Parsons, 2004; Metcalf and

Eddy, 2004). Immobilizing TiO2 on a fixed-film eliminates the need to amend and recover catalyst offering greater flexibility in treatment design. In addition, natural sunlight could provide sufficient energy to accomplish photocatalytic degradation of NAs while decreasing operational costs of treatment.

To evaluate the feasibility of fixed-film photocatalysis irradiated by natural sunlight for achieving degradation of NAs, commercially available NAs were used for this study. To assess performance of photolysis and photocatalysis, rates and extents of degradation of Fluka NAs have been measured (McMartin et al., 2004; Headley et al.,

2009; Mishra et al., 2010), providing an opportunity to compare results from this study.

In addition, Fluka NAs have been well studied as a simplistic analogue to understand more compositionally complex mixtures of NAs present in energy-derived waters

(Headley and McMartin 2004; Barrow et al., 2004; Scott et al., 2005; Rudzinski et al.,

2002; Lo et al., 2006; Armstrong et al., 2008; Headley et al., 2010). Photolysis and photocatalytic oxidation can decrease concentrations and complexity of parent NA

compounds (USEPA, 2012; McMartin et al., 2004; Headley et al., 2009); however, the

question of post-treatment toxicity remains.

Decreased toxicity to sensitive species can confirm alteration of exposures of

Fluka NAs achieved by fixed-film photocatalytic degradation. Sentinel species such as fathead minnow (Pimephales promelas Rafinesque) and microinvertebrate (Daphnia magna Straus) are relatively sensitive to NAs (Swigert et al., 2015; Kinley et al., 2016).

Commercial NA 96-hr LC50 for P. promelas is 5.6 mg NA/L and 48-hr EC50

(immobilization) for D. magna is 20 mg NA/L (Swigert et al., 2015), demonstrating

mortality is informative to assess changes in NA concentrations. Elimination of toxicity to these sensitive species in post-photocatalytic degradation samples can confirm

mitigation of risks, supporting observations of NA degradation measured analytically.

The overall objective of this study was to measure rates and extents of photolysis

and photocatalytic degradation of a commercially available (Fluka) NA using bench-scale

fixed-film TiO2 and confirm changes in NA concentrations using sensitive vertebrate

(fish = Pimephales promelas) and invertebrate (Daphnia magna) species. To achieve this

overall objective, specific objectives were to: 1) measure the rates and extents of removal

of commercial (Fluka) NAs throughout an 8 hour duration of natural sunlight

(“photolysis”) and natural sunlight in the presence of fixed-film TiO2 (“photocatalysis”),

and 2) measure changes in toxicity after photolysis and photocatalysis treatments (in

terms of mortality) with sentinel fish (P. promelas) and microinvertebrate (D. magna)

species in 96-hr static tests.

3.3 Materials and Methods

3.3.1 Experimental Design

The experimental design included three treatments: 1) Fluka NAs irradiated by

natural sunlight in the presence of TiO2 film (“photocatalysis”), 2) Fluka NAs irradiated

by natural sunlight without TiO2 film (“photolysis”), and 3) Fluka NAs in the presence of

TiO2 film with no sunlight (“dark control”). Treatments were conducted outdoors near

Clemson, SC USA (34°40'6.14"N, 82°50'52.02"W) in November with clear, sunny

conditions. All treatments were conducted in 28x43 cm Sterilite® high-density

polyethylene (HDPE) containers. For photocatalytic and dark control treatments, a thin-

film of silicone caulk (DAP®; 100% silicone rubber sealant) was applied to a thickness

of <0.2 cm to 28x43 cm on the bottom of the containers (Figure 3.1). Immediately after

TM silicone application, TiO2 (Aeroxide P25; Fisher Scientific, Fairlawn NJ) particles

were added to the surface of the film and air dried for 24-hr. The primary particle size of

2 the TiO2 was approximately > 45 μm and the specific surface area was 35 - 65 m /g with

a crystalline composition of 10-20% rutile and 80-90% anatase. Water depths of <1.0 cm

were used for sufficient light penetration based on preliminary experimentation. Dark

control containers were shaded with opaque polypropylene covers. Aqueous samples were collected every hour through the 8-hr duration of the study in 60 mL amber glass vials for quantification of NAs and to conduct toxicity testing.

During the 8-hr treatments, in situ dissolved oxygen, pH, and conductivity of NA

amended water were measured every hour using a YSI Model 52 dissolved oxygen

meter, Orion Model 250A pH meter and Triode electrode, and Orion Model 142

conductivity meter, respectively (Table 3.1). Alkalinity and hardness of aqueous samples were determined according to Standard Methods for the Examination of Water and

Wastewater (APHA, 2012). Turbidity was measured using a Hach 2100AN Turbidimeter,

(V2.2; APHA, 2012) and light intensity (LUX) was measured using a VWR®

Traceable® dual-range light meter. Ultraviolet (UV) and visible irradiance were estimated using a methylene blue and peroxide actinometer (Alpert et al., 2010). A stock solution of 0.5 mg/L methylene blue with 15 µL of 30% hydrogen peroxide (Fisher

Scientific) was used to estimate UV and visible light irradiance using sealed 1-cm quartz cuvettes at water depth intervals of 0, 0.5, and 1.0 cm. Measurements of methylene blue were performed using a SpectraMax®M2 spectrophotometer (Molecular Devices Corp.

Sunnyvale, CA).

3.3.2 Fluka Naphthenic Acid Exposure Preparation and Analysis

Fluka NAs (Sigma-Aldrich; St. Louis, MO; Table 3.2) were used to prepare initial

test concentrations and stock solutions in reconstituted moderately hard water (pH 8.2 ±

0.5 SU, alkalinity 65 ± 8 mg/L as CaCO3, hardness 88 ± 10 mg/L as CaCO3, conductivity

350 ± 20 µs/cm) which was prepared using reverse osmosis filtered water and reagent grade chemicals based on recommended culture methods (USEPA, 2002). The formulated water contained 5 mg/L CaCO3, 102 mg/L NaHCO3, 48 mg/L MgSO4-7H2O,

33 mg/L CaSO4-2H2O, 65 mg/L CaCl2-2H2O, 2 mg/L KCl, 0.8 mg/L KNO3, 0.02 mg/L

K2PO4, and 0.002 mg/L of each Cu, Se, and Zn (from aqueous standards). All reagents

were obtained from Fisher Scientific® (Pittsburgh, PA). Fluka NAs were added to

moderately hard water in a 20 L HDPE Nalgene® container (initial nominal

concentrations of 65 mg/L) and mixed to prepare a modified water accommodated

fraction (WAF), where solutions were mixed with magnetic stir bars for 24 hours at a

speed sufficient to create a vortex which extended 30-50% of the solution depth (OECD,

2000). Methods for NA derivatization and analysis were based on Yen et al. (2004)

using high performance liquid chromatography (HPLC; Dionex, UltiMate-3000;

Sunnyvale, CA). The HPLC analytical column was an Agilent LiChrospher 100 RP-18 (5

µm particle size, 125mm x 4 mm) with a guard column packed with 2 µm RP-18 solid

phase material. Column temperature was maintained at 40° C with a sample injection

volume of 60 µL mobilized with HPLC grade methanol (Fisher Scientific) at a flow rate

of 60 µL per minute. The detection limit for this HPLC method is approximately 5 mg

NA/L.

3.3.3 Removal Efficiency and Rate Calculations

Removal efficiencies (equation 1) were estimated using the following equation:

(1)

Where, measured initial concentrations of NAs are designated as [C0] (mg/L) and

[C] (mg/L) is concentration of NAs at test completion. A linear relationship was

observed between changes in NA concentration with time; therefore, removal rates

(k=mg/L day-1) were calculated using zero order kinetics, as the inverse slope of the line

indicating change in concentration with change in time (hours). Correlation coefficients

for the zero-order model are provided for each treatment. Half-lives were estimated

using the following equation based on zero order kinetics:

T1/2 = [C0] / 2k (2)

Where, T1/2 is half-life (hours), [C0] (mg/L) is NA concentration at test initiation

and k is degradation rate (mg/L day-1).

3.3.4 Toxicity Testing

Photolysis and photocatalysis treatments were assessed for their ability to alter toxicity of Fluka NAs using sensitive sentinel species P. promelas and D. magna. Larval

(<24h old) fish (P. promelas) and D. magna (<24h old) were obtained from cultures at

Clemson University Aquatic Animal Research Laboratory (AARL). To conduct toxicity experiments, 50 mL aliquots of treatments were collected from reactors and organisms were exposed to treated waters in 30 mL HDPE chambers. Survival of P. promelas and

D. magna were evaluated in 96-hr static/non-renewal toxicity tests conducted following a

United States Environmental Protection Agency (USEPA) freshwater toxicity testing protocol with (n=30) organisms per exposure (USEPA, 2002; Table 3.3). Water characteristics (dissolved oxygen, pH, conductivity, alkalinity, and hardness) of test waters were measured at test initiation and completion using methods described in Table

3.1. Reference toxicity tests were conducted with copper sulfate (CuSO4-5H2O) to

confirm the health of test organisms for quality assurance (USEPA, 2002) using the same

toxicity testing procedures.

3.3.5 Statistical Analysis

Data were tested for normal distribution and homogeneity of variance using Chi-

square and Bartlett’s tests, respectively. Normally distributed, homogeneous data were

analyzed by one-way analysis of variance (ANOVA). Differences among treatments

were determined using follow-up pairwise comparisons and contrasts using linear

models. Differences were considered significant at p ≤ 0.05 (JMP v11; SAS Institute Inc.,

Cary, NC, USA).

3.4 Results and Discussion

3.4.1 Exposure conditions for photolysis and photocatalysis

Mean measured initial concentrations of NAs for aqueous samples were 63 (±9),

65 (±5), and 64 (±10) mg NA /L for photocatalysis, photolysis, and dark controls,

respectively. Over the 8-hr treatment period in direct sunlight (excluding the dark

control), measured light intensity ranged from 11,500 (at 8-hr) to 109,700 LUX (at 4-hr).

There was no measureable incident light in dark controls. UV/Visible irradiance ranged from 12.6 W/m2 at water surface to 6.27 W/m2 at 1.0 cm water depth, respectively,

indicating rapid light attenuation with water depth (attenuation coefficient [Kd] = 0.69

cm-1). Ambient air temperatures during the 8-hr experiment ranged from 7.5 to 21.6°C

and water temperatures in treatments and controls ranged from 13-19ºC. Water

containing NAs had no detectable turbidity (<0.1 NTU). pH in treatments and controls

ranged from 8.02-8.26. In situ water characteristics measured during the 8-hr treatment

durations (e.g. temperature, pH, DO, conductivity) did not differ among treatments

(Table 3.4).

3.4.2 Photocatalysis

After 4-hr in photocatalysis treatments, NA concentrations declined to below

detection limit (method detection limit = 5 mg/L), resulting in >92% removal (Figure

3.2). Data were fit to a zero-order model with a correlation coefficient (R2) of 0.9792, based on 5 data points from Time 0 (test initiation) to Time 5 (hour 4). The calculated

NA removal rate (k) and half-life for photocatalysis were 15.5 mg/L hr-1 and 2.0-hr,

respectively (Table 3.5). Results from this study are similar to results for Fluka NAs

using TiO2 “slurries” in photocatalysis studies irradiated with sunlight and UV254 bulbs

(Headley et al., 2009; Mishra et al., 2010; Table 3.6). For example, using sunlight,

Headley et al. (2009) achieved 75% Fluka NA removal in 8-hr (initial concentration of 46

mg/L NA) using TiO2 slurries with concentrations of 2 g TiO2/L. Mishra et al. (2010)

used aqueous suspensions of TiO2 (at 0.3 g TiO2/L concentrations) irradiated with UV254

(8W lamps) and achieved first order rate coefficients ranging from 0.05 to 0.34 hr-1 for

Fluka NAs in South Saskatchewan River water and deionized water, respectively. Under

the same laboratory conditions, Mishra et al. (2010) achieved relatively rapid degradation

of NAs derived from oil sands process affected water (OSPW), with half-lives of 1.55

and 4.8-hr for deionized water and South Saskatchewan River water, respectively.

Generally, results were similar among fixed-film and slurries of TiO2, irradiance from

sunlight and artificial UV254, and Fluka and OSPW NAs (Headley et al., 2009; Mishra et

al., 2010). Among these treatments, measured photocatalysis rates and extents of removal

are of practical significance with NA half-lives achieved in hours.

3.4.3 Photolysis and Dark Control

In photolysis treatments, 51% removal of the initial NA concentration (65 mg/L

±5 SD) was achieved after 8-hr (final concentration 32 mg NA/L ±11 SD; Figure 3.2;

Table 3.5). NA concentrations after 4-hr (α =0.05; p<0.001) were significantly different from test initiation, which corresponded with maximum measured light intensity during the experimental duration (109,700 LUX at 4-hr). Data were fit to a zero-order model with a correlation coefficient (R2) of 0.9122, based on 9 data points from Time 0 (test

initiation) to Time 9 (hour 8). The calculated NA removal rate (k) and half-life for

photolysis were 4 mg/L hr-1 and 8.1-hr, respectively (Table 3.5). Since one NA half-life

was achieved with this treatment (rather than reaching the detection limit), this removal

rate (4 mg/L hr-1) applies to the first measured half-life only, and application of this

estimated rate beyond the measured bounds in this study should be undertaken carefully.

These photolysis results using Fluka NAs in sunlight are similar to modeled half-lives

estimated for photo-degradation of commercial NAs (USEPA, 2012). Empirical models

indicate photo-degradation half-lives for 1- to 4-ring structures with molecular weights of

254–325 amu (as compared to 210-250 amu of the NA used in this study) range from 3 to

6.8-hr (USEPA, 2012). Headley et al. (2009) estimated < 3% removal of Fluka NAs

(initial concentration 46 mg/L) in 8-hr photolysis (no TiO2) experiments irradiated with sunlight in Milli-Q water. A number of factors influence photolysis of NAs, including structure of the molecule (Headley et al., 2009), and exposure conditions (i.e. sunlight intensity, water depth, turbidity; Kirk, 1994). In this study, conditions that may confound photolysis and photocatalysis such as water depth, pH, and turbidity were managed to minimize their effects.

Dark controls containing fixed-film TiO2 were used to account for sorption and volatilization of NAs, which may influence the measured concentration of NAs. Based on a one-way ANOVA, NA concentrations in dark controls did not differ from test initiation

(64 mg/L ±10 SD) to completion at 8-hr (56 mg/L ±4 SD; α =0.05; p= 0.8487). In the absence of light, minimal decrease in NA concentration was anticipated due to the limited volatility (1.1 x 10-7 to 7.1 x 10-6 mm Hg at 25°C) and sorption (log Kd <0.5 [quartz

sand]) of naphthenic acids at the experimental pH of 8.02-8.26 (Schramm, 2000; USEPA,

2012).

3.4.4 Toxicity testing to confirm changes in NA exposures

Toxicity tests using fish and microinvertebrates were used to confirm degradation

of NAs within treatments. For all treatments at test initiation, complete mortality (0%

survival) of both D. magna and P. promelas was observed in 96-hr exposures (NA

concentrations 63-65 mg/L). In photocatalysis treatments, toxicity was completely

eliminated (100% survival) for both D. magna and P. promelas after 5-hr (Table 3.7).

Both photolysis and dark control treatments did not have any measurable change in

toxicity after 8-hr durations (i.e. 100% mortality throughout treatments).

In this study, larval fish (P. promelas) were more sensitive to Fluka NA exposures

than the microcrustacean D. magna, with 0% and 57% survival observed for P. promelas

and D. magna, respectively, after 3-hr of photocatalytic treatment. Changes in percent

mortalities with time observed for D. magna and P. promelas in this study provided an

opportunity to compare with reported endpoint estimations (i.e. EC50s and LC50s) for

commercial NAs (Swigert et al., 2015; Kinley et al. 2016). In photocatalysis treatments,

D. magna mortality was 43% in Fluka NA concentrations of 15 mg/L NA (±5), similar to

the reported 48-hr EC50 (immobilization) of 20 mg/L (17-23 mg/L C.I.) for D. magna

exposed to Merichem NAs (Swigert et al., 2015). After photocatalysis treatment in this

study, mortality of P. promelas declined from 100% to 40% after 4-hr and 0% mortality

was observed after 5-hr with measured NA concentrations below the analytical detection

limit of 5 mg/L NA. Swigert et al. (2015) measured a 96-hr LC50 for juvenile P. promelas of 5.6 mg/L Merichem NAs. Comparatively, Kinley et al. (2016) observed a 7-day LC50

for larval P. promelas of 1.9 mg/L as Fluka NAs. In this study, coupling bioassay

response data with analytical quantification of NAs provided a robust approach for

discerning changes in NA exposures and mitigation of ecological risks.

3.5 Conclusions

Greater than 90% removal of Fluka NAs (initial concentration 63 mg/L) was

achieved in 4-hr with photocatalysis in fixed-film (TiO2) reactors in direct sunlight.

Photocatalysis also eliminated acute toxicity to sentinel species, with mortality

decreasing from 100% to 0% after 5-hr of photocatalytic treatment for fathead minnow

(Pimephales promelas) and after 4-hr for the freshwater invertebrate (Daphnia magna).

In this experiment, measuring responses of aquatic organisms concomitantly with analytical quantification of Fluka NAs over time confirmed alteration of exposures as well as mitigation of risk. Fixed-film TiO2 application may provide an alternative

solution for scaling the technology for larger applications. Photocatalytic degradation

using fixed-film TiO2 irradiated with sunlight achieved efficacious rates and extents of

removal of Fluka NAs, indicating the potential for application of this technology for mitigating ecological risks associated with NAs.

3.6 Acknowledgements

Funding support for this research was provided by Shell Canada Ltd. and Suncor Energy.

The authors are also grateful to Dr. Wayne Chao of Clemson University for providing analytical support.

3.7 References

Allen, E.W. (2008). Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Journal of Environmental Engineering Science, 7, 123-138.

Alpert, S.M. (2010). Modeling the UV/hydrogen peroxide advanced oxidation process using computational fluid dynamics. Water Research, 44, 1797-1808.

American Petroleum Institute (API). (2012). Naphthenic acids category analysis and hazard characterization. Petroleum HPV Testing Group Technical Report # 1100997.

American Public Health Association (APHA). (2012). Standard methods for the examination of water and wastewater, 21st Ed. American Public Health Association, Port City Press: Baltimore, MD, pp. 1368.

Armstrong, S. A., Headley, J. V., Peru, K. M., Germida, J. J. (2008). Phytotoxicity of oil sands naphthenic acids and dissipation from systems planted with emergent aquatic macrophytes. Journal of Environmental Science and Health, Part A, Toxic/hazardous substances & environmental engineering 43, 36-42.

Bagheri, S., Muhd Julkapli, N., & Bee Abd Hamid, S. (2014). Titanium Dioxide as a Catalyst Support in Heterogeneous Catalysis. The Scientific World Journal.

Barrow, M. P., Headley, J. V., Peru, K. M., & Derrick, P. J. (2004). Fourier transform ion cyclotron resonance mass spectrometry of principal components in oil sands naphthenic acids. Journal of Chromatography A, 1058, 51-59.

Brient, J. A., Wessner, P. J., Doyle, M. N. (1995). Naphthenic acids. Kirk-Othmer encyclopedia of chemical technology, 4th ed., 1017-1029.

Clemente, J. S., Fedorak, P. M. (2005). A review of the occurrence, analyses, toxicity, and biodegradation of naphthenic acids. Chemosphere, 60, 585-600.

Dorn, P. B. (1992). Case histories—the petroleum industry. Toxicity Reduction: Evaluation and Control. In: Ford, DL (Ed.), Water Quality Management Library 3, 183-223.

Frank, R.A., Fischer, K., Kavanagh, R., Burnison, B.K., Arsenault, G., Headley, J.V., Peru, K.M., Van Der Kraak, G., Solomon, K.R. (2008). Effect of carboxylic acid content on the acute toxicity of oil sands naphthenic acids. Environmental Science & Technology, 43, 266-271.

Han, X., MacKinnon, M.D., & Martin, J.W. (2009). Estimating the in situ biodegradation of naphthenic acids in oil sands process waters. Chemosphere, 76, 63-70.

Headley, J. V., & McMartin, D. W. (2004). A review of the occurrence and fate of naphthenic acids in aquatic environments. Journal of Environmental Science and Health, Part A : Toxic/Hazardous Substances and Environmental Engineering 39, 1989-2010.

Headley, J. V., Peru, K. M., Adenugba, A. A., Du, J. L., & Mcmartin, D. W. (2010). Dissipation of naphthenic acids mixtures by lake biofilms. Journal of Environmental Science and Health Part A 45, 1027-1036.

Holowenko, F.M., MacKinnon, M.D., & Fedorak, P.M. (2002). Characterization of naphthenic acids in oil sands wastewaters by gas chromatography-mass spectrometry. Water Research 36, 2843-2855.

Kavanagh, R. J., Frank, R. A., Burnison, B. K., Young, R. F., Fedorak, P. M., Solomon, K. R., & Van Der Kraak, G. (2012). Fathead minnow (Pimephales promelas) reproduction is impaired when exposed to a naphthenic acid extract. Aquatic toxicology 116, 34-42.

Kinley, C.M., McQueen, A.D., & Rodgers, Jr., J.H. (2016). Comparative responses of freshwater organisms to exposures of a commercial naphthenic acid. In press: Chemosphere.

Kinsinger, N., Honda, R., Keene, V., & Walker, S. L. (2015). Titanium dioxide nanoparticle removal in primary prefiltration stages of water treatment: role of coating, natural organic matter, source water, and solution chemistry. Environmental Engineering Science, 32, 1-9.

Leclair, L. A., MacDonald, G. Z., Phalen, L. J., Köllner, B., Hogan, N. S., & van den Heuvel, M. R. (2013). The immunological effects of oil sands surface waters and naphthenic acids on rainbow trout (Oncorhynchus mykiss). Aquatic toxicology, 142, 185-194.

Lo, C. C., Brownlee, B. G., & Bunce, N. J. (2006). Mass spectrometric and toxicological assays of Athabasca oil sands naphthenic acids. Water Research, 40, 655-664.

McMartin, D. W., Headley, J. V., Friesen, D. A., Peru, K. M., & Gillies, J. A. (2004). Photolysis of naphthenic acids in natural surface water. Journal of Environmental Science and Health, Part A 39, 1361-1383.

Metcalf & Eddy., Tchobanoglous, G., Burton, F. L. 1., & Stensel, H. D. (2004). Wastewater engineering: Treatment and reuse (4th ed.). McGraw-Hill. Boston, MA.

Mishra, S., Meda, V., Dalai, A., McMartin, D.W., Headley, J.V., & Peru, K.M. (2010). Photocatalysis of naphthenic acids in water. Journal of Water Resource and Protection, 2, 644-650.

Nero, V., Farwell, A., Lee, L. E. J., Van Meer, T., MacKinnon, M. D., & Dixon, D. G. (2006). The effects of salinity on naphthenic acid toxicity to yellow perch: Gill and liver histopathology. Ecotoxicology and Environmental Safety 65, 252-264.

Organization for Economic Cooperation and Development (OECD). (2000). Guidance document on aquatic toxicity testing of difficult substances and mixtures. OECD Series on Testing and Assessment No. 23, ENV/JM/MONO (2000), vol. 6. OECD, Paris, France.

Parsons, S.(ed.). (2004). Advanced oxidation processes for water and wastewater treatment. IWA Publishing. London, UK.

Rudzinski, W. E., Oehlers, L., Zhang, Y., & Najera, B. (2002). Tandem mass spectrometric characterization of commercial naphthenic acids and a Maya crude oil. Energy & Fuels 16, 1178-1185.

Schramm, L. L. (Ed.) (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press, Cambridge, UK.

Scott, A.C., Mackinnon, M.D., & Fedorak, P.M. (2005). Naphthenic acids in Athabasca oil sands tailings waters are less biodegradable than commercial naphthenic acids. Environmental Science and Technology, 39, 8388-8394.

Seifert, W. K., Teeter, R. M. (1969). Preparative thin-layer chromatography and high- resolution mass spectrometry of crude oil carboxylic acids. Analytical Chemistry 41, 786-795.

Sigma-Aldrich, (2015). MSDS, Naphthenic acid. St. Louis, Missouri.

Swigert, J. P., Lee, C., Wong, D. C., White, R., Scarlett, A. G., West, C. E., & Rowland, S. J. (2015). Aquatic hazard assessment of a commercial sample of naphthenic acids. Chemosphere, 124, 1-9.

Tomczyk, N. A., Winans, R. E., Shinn, J. H., & Robinson, R. C. (2001). On the nature and origin of acidic species in petroleum. 1. Detailed acid type distribution in a California crude oil. Energy & Fuels, 15, 1498-1504.

United States Environmental Protection Agency (USEPA). (2002). Methods for measuring acute toxicity of effluents and receiving waters to freshwater and marine organisms. Washington DC. EPA-821-R-02-12.

United States Environmental Protection Agency (USEPA). (2012). Screening-level hazard characterization: Reclaimed substances (Naphthenic acids category). Washington DC. EPA-821-R-02-12.

Wold, A. (1993). Photocatalytic properties of Ti02. Chemistry of materials 5, 280-283.

Yen, T. W., Marsh, W. P., MacKinnon, M. D., & Fedorak, P. M. (2004). Measuring naphthenic acids concentrations in aqueous environmental samples by liquid chromatography. Journal of Chromatography, 1033, 83-90.

Table 3.1 Methods for water characteristics, light, and NA concentrations.

Parameter Method Method Detection Limit Electrometric method 4500-H+ B: Orion model 420A pH 0.01 SU (APHA, 2007) Laboratory method 2550 B: Orion Model 420A Temperature 0.01 °C (APHA, 2007) Membrane electrode method 4500-O G: YSI Model Dissolved oxygen 0.1 mg/L 52 (APHA, 2007)

Conductivity Laboratory method 2510 B: YSI 30 (APHA, 2007) 0.1 μS/cm

Alkalinity Titration method 2320 B (APHA, 2007) 2 mg/L as CaCO3

Hardness EDTA Titrimetric Method 2340 C (APHA, 2007) 2 mg/L as CaCO3

Turbidity Nephelometric Method 2130 B (APHA, 2007) 0.1 NTU

Light Intensity VWR® Traceable® Light Meter 0.1 LUX UV/ Visible Methylene blue and peroxide actinometer (Alpert et 1 W/m2 Irradiance al., 2010) Naphthenic acid HPLC; Derivatization based on Yen et al. (2004) 5 mg/L Concentration

Table 3.2 Physical and chemical characteristics of Fluka naphthenic acids (Sigma-

Aldrich)a

Parameter General Characteristic Source Identification 1338-24-5 (CAS No) Sigma-Aldrich, 2015 Color Pale yellow, dark amber Sigma-Aldrich, 2015 Physical State Viscous liquid Sigma-Aldrich, 2015 Molecular weightb 210-250 amu Brient et al., 1995 Water solubility 88.1 mg/L at pH 7.5 API, 2012 1.1 x 10-7 to 7.1 x 10-6 mm Hg at Vapor pressure 25°C USEPA, 2012 Partition coefficient ˜4 at pH 1 Schramm, 2000 octanol/ water (Log Kow)c; pH ˜2.4 at pH 7 Schramm, 2000 dependent < 0.1 at pH 10 Schramm, 2000

Density 0.92 g/mL Sigma-Aldrich, 2015 Viscosity 22 mm2/s Sigma-Aldrich, 2015 pKa 5 to 6 Brient et al., 1995 aAlkylated cyclopentane carboxylic acids (mixture) bAverage molecular weight for refined naphthenic acids cWeathered naphthenic acid mixture; for OSPW NAs

Table 3.3 Summary of toxicity test conditions for P. promelas and D. magna (USEPA,

2002).

Test Parameter Description Test type static non-renewal Endpoint measureda mortality Test Duration (hours) 96 Test temperature (°C) 25 ± 1 Test chamber 30 mL HDPE chamber Formulated moderately hard Test water waterb Number of organisms/ exposure 30 Number of organisms/ chamberc 10 Age of organisms < 24 hd Photoperiod 16 h light, 8 h dark aImmobilization bUSEPA, 2002 c3 replicates dPost-hatch for P. promelas; post-brood for D. magna

Table 3.4 Measurements of water characteristics during photocatalysis and photolysis

treatments (and dark controls) and toxicity testing.

Temp. pH DO Conductivity Alkalinity Hardness Treatment (°C) (S.U.) (mg O2/L) (µS/cm) (mg CaCO3/L) (mg CaCO3/L) In Situa

Photocatalysis 13-19 8.02-8.22 8 ± 1 340-355 - - Photolysis 13-19 8.06-8.26 8 ± 1 345-355 - - Dark Control 13-19 8.08-8.23 8 ± 1 345-352 - - Toxicity Testingb

Photocatalysis P. promelas 25 ± 1 8.08-8.26 8 ± 1 335-362 76-80 152-160 D. magna 25 ± 1 8.05-8.26 8 ± 1 340-366 70-78 155-164 Photolysis

P. promelas 25 ± 1 8.05-8.22 8 ± 1 346-365 70-80 152-160 D. magna 25 ± 1 8.11-8.21 8 ± 1 344-353 76-84 155-160 Dark Control P. promelas 25 ± 1 8.12-8.28 8 ± 1 345-360 68-82 152-160 D. magna 25 ± 1 8.10-8.22 8 ± 1 344-364 70-85 155-160 ameasured at 1 hour sampling intervals (n=8) bmeasured at test completion (n=3)

Table 3.5 Fluka naphthenic acid removal rate coefficients and extents for photocatalysis, photolysis, and dark control treatments.

Treatments Parameter Photocatalysis Photolysis Dark Control Initial [NA] mg/L (±SD) 63 (± 9) 65 (± 5) 64 (± 10) Ending [NA] mg/L BDLa 32 (± 11) 56 (± 4) (±SD) Removal Efficiency, % >92 51 NAc Y = -15.502x + Y=-4.0076x + Rate Equationb NA 64.727 61.519 R2 0.9792 0.9122 NA k average (mg/L hour-1)b 15.5 4.0d NA d T1/2 (hour) 2.0 8.1 NA NA = Not applicable aBDL = Below Detection Limit, detection limit = 5 mg/L NA bZero Order Rate Equation; Linear plot [C] versus time (hours), slope = -k cEnding [NA] not statistically different from initial [NA] in dark control (α =0.05; p= 0.8487) dEstimation based on 1 observed half-life Note: Removal rate coefficient calculated from best fit using five data points for photocatalysis (Time 0- 4 on Figure 3.2) and nine data points for photolysis (Time 0-8 on Figure 3.2)

Table 3.6 Comparative rates and extents of removal for photocatalysis and photolysis of Fluka NAs. Initial NA Ending NA (mg/L), Removal Removal Estimated Half Citation Treatment (Condition) (mg/L) Duration Efficiency Rate* Life Values Mean (±SD) Mean (±SD) (%) (hours)

-1 Photocatalysis (sunlight; TiO2 film) 63 (± 9) BDL, 8-hr > 92 15.5 mg/L hr 2.0 Current study a ~ Photocatalysis (sunlight; TiO2 slurry ) 46 9, 8-hr 80 - - Headley et al., 2009

b c -1 Photocatalysis (UV ; TiO2 slurry ) 40-100 NA; 5-hr - 0.34 hr 2.04 Mishra et al., 2010 b c Photocatalysis (UV ; TiO2 slurry ; 40-100 NA; 5-hr - 0.05 hr-1 13.86 Mishra et al., 2010 river waterd)

Photolysis (sunlight) 65 (± 5) 32 (± 11), 8-hr 51 4.0 mg/L hr-1 8.0 Current study Photolysis (sunlight) 46 45, 8-hr < 3 - - Headley et al., 2009 Photolysis (simulated solar radiatione; 50 NA, 168-hr - - 8120 ± 100 McMartin et al., 2004 river waterf) Photolysis (UVb) 40-100 NA, 5-hr - 0.22 hr-1 3.15 Mishra et al., 2010 Photolysis (UVb; river waterd) 40-100 NA, 5-hr - 0.04 hr-1 17.33 Mishra et al., 2010 Photolysis (UVg; river waterf) 10 NA, 8-hr - - 1050 ± 20 McMartin et al., 2004 Photolysis (UVg; river waterf) 50 NA, 8-hr - - 1200 ± 40 McMartin et al., 2004 Photolysis (UVh; river waterf) 10 NA, 8-hr - - 50 ± 1 McMartin et al., 2004 Photolysis (UVh; river waterf) 50 NA, 8-hr - - 52 ± 1 McMartin et al., 2004

~ Dark Control (TiO2 film) 64 (± 10) 56 (± 4), 8-hr 2 - - Current study

a Dark Control (TiO2 slurry ) 46 45, 8-hr < 3 - - Headley et al., 2009 BDL = below detection limit (<5 mg/L) eSimulated full spectrum solar radiation with incandescent and fluorescent lamps NA = data not available fFluka NA concentrations prepared in unaltered Athabasca River Water a g TiO2 (Degussa P25) slurry concentration = 2 g/L Philips medium pressure black light bulb; 300-400 nm (15W) b h UV fluorescent tubes (8W) Low pressure Phillips® UV254 tube c Note: Comparison studies quantified Fluka NAs using electrospray ionization TiO2 (Degussa P25) slurry concentration = 0.3 g/L mass spectrometry (ESI-MS)

dNA mixtures prepared in South Saskatchewan River water (Saskatoon, SK)

Note: pH not reported in McMartin et al. (2004), Headley et al. (2009); Mishra *Zero and first order rates/coefficients not directly comparable et al. (2010)

Table 3.7 Summary of mean NA concentrations (mg/L) and 96-hr percent survival for microinvertebrates (D. magna) and fish

(P. promelas) for photocatalysis, photolysis, and dark control treatments. NA concentrations with different letters are significantly different (p< 0.05).

Treatments Treatment Photocatalysis Photolysis Dark Control Duration % Survival % Survival % Survival (hours) NAs (±SD) NAs (±SD) NAs (±SD) D. Magna P. Promelas D. Magna P. Promelas D. Magna P. Promelas 0 63 (±9) A 0 0 65 (±5) A 0 0 64 (±10) A 0 0 1 54 (±12) AB 0 0 57 (±15) AB 0 0 60 (±12) A 0 0 2 32 (±8) BC 0 0 55 (±10) AB 0 0 67 (±8) A 0 0 3 15 (±5) C 57 0 47 (±8) ABC 0 0 65 (±5) A 0 0 4 BDL 100 60 38 (±10) BC 0 0 63 (±5) A 0 0

5 BDL 100 100 44 (±5) BC 0 0 58 (±5) A 0 0

6 BDL 100 100 38 (±10) BC 0 0 58 (±2) A 0 0

7 BDL 100 100 33 (±12) C 0 0 54 (±7) A 0 0

8 BDL 100 100 32 (±11) C 0 0 56 (±4) A 0 0 BDL = below detection limit (<5 mg/L)

Figure 3.1 Schematic of photocatalytic reaction chamber.

Figure 3.2 Concentrations of Fluka naphthenic acids (mg/L) with time in photocatalysis, photolysis, and dark control treatments.

Note: Incident light intensity (LUX) on the secondary axis corresponds to photocatalysis and photolysis treatments.

CHAPTER FOUR

INFLUENCE OF COMMERICAL (FLUKA) NAPHTHENIC ACIDS ON ACID

VOLATILE SULFIDE PRODUCTION AND DIVALENT METAL PRECIPITATION

Andrew D. McQueen1,3; Ciera M. Kinley1; John H. Rodgers Jr.1 Vanessa Friesen2, Jordyn Bergsveinson2, Monique C. Haakensen2

1 Department of Forestry and Environmental Conservation, 261 Lehotsky Hall, Clemson University, Clemson, SC 29634, USA

2 Contango Strategies Limited, LFK Biotechnology Complex, 15-410 Downey Road, Saskatoon, SK S7N 4N1

3 Corresponding author: Andrew McQueen

Manuscript accepted in: Ecotoxicology and Environmental Safety

4.1 Abstract Energy-derived waters containing naphthenic acids (NAs) are complex mixtures

often comprising a suite of potentially problematic constituents (e.g. organics, metals,

and metalloids) that need treatment prior to beneficial use, including release to receiving

aquatic systems. It has been suggested previously that NAs can have biostatic or biocidal

properties that could inhibit microbially driven processes (e.g. dissimilatory sulfate

reduction) used to transfer or transform metals in passive treatment systems (i.e.

constructed wetlands). The overall objective of this study was to measure the effects of a

commercially available (Fluka) NA on sulfate-reducing bacteria (SRB), production of

sulfides (as acid-volatile sulfides [AVS]), and precipitation of divalent metals (i.e. Cu,

Ni, Zn). These endpoints were assessed following 21-d aqueous exposures of NAs using bench-scale reactors. After 21-days, AVS molar concentrations were not statistically

different (p<0.0001; α=0.05) among NA treatments (10, 20, 40, 60, and 80 mg NA/L)

and an untreated control (no NAs). Extent of AVS production was sufficient in all NA

treatments to achieve ∑SEM:AVS <1, indicating that conditions were conducive for

treatment of metals, with sulfide ligands in excess of SEM (Cu, Ni, and Zn). In addition,

no adverse effects to SRB (in terms of density, relative abundance, and diversity) were

measured following exposures of a commercial NA. In this bench-scale study,

dissimilatory sulfate reduction and subsequent metal precipitation were not vulnerable to

NAs, indicating passive treatment systems utilizing sulfide production (AVS) could be

used to treat metals occurring in NAs affected waters.

4.2 Introduction

Energy-derived waters containing naphthenic acids (NAs) are complex mixtures

often containing a suite of potentially problematic constituents (e.g. organics, metals, and

metalloids) that need treatment prior to beneficial use, including release to receiving

aquatic systems (Seifert and Teeter, 1969; Dorn 1992; Tomcyzk et al., 2001; Allen

2008a). NAs are a class of organic acids (generally described by the formula CnH2n+ZO2) that are sources of toxicity in these process-affected waters (Dorn 1992; Schramm, 2000), with adverse effects observed for fish, macroinvertebrates, macrophytes, and microorganisms (Nero et al., 2006; Frank et al., 2008; Armstrong et al., 2008; Jones et al., 2011; Kavanagh et al., 2012; Leclair et al., 2013; Swigert et al., 2015). To treat impacted waters, passive treatment approaches (e.g. constructed wetlands) are among technologies considered for transfer and transformation of problematic constituents (e.g. metals, organics; Allen 2008b; Foote 2012; Toor et al., 2013; Brown and Ulrich 2015).

Microbial activity is an important contributor to biogeochemical processes and elemental cycling that occur in constructed wetlands (Rodgers and Castle 2008; Haakensen et al.,

2015). Microbially mediated processes can be promoted in specifically designed constructed wetlands to treat problematic constituents contained in process-affected waters (Huddleston and Rodgers 2008; Spacil et al., 2011; Pham et al., 2011). Divalent metals (e.g. Cu, Cd, Pb, Ni, and Zn) can be sequestered as sulfide minerals with sulfides produced through dissimilatory sulfate reduction, altering the solubility and bioavailability of metals (Kanagy et al., 2008; Murray-Gulde et al., 2008; Rodgers and

Castle 2008). To effectively design and implement constructed wetlands, potential

adverse effects of co-occurring constituents on microbially mediated pathways (e.g.

sulfate reduction) must be understood. NAs occurring in process water can exhibit

adverse effects on a laboratory strain of Vibrio fischeri commonly used in toxicity testing

(i.e. Microtox EC50 values ranging from 12 to 65 mg/L NAs; Rogers et al., 2002; Frank et al., 2008). There is evidence that sulfate-reducing bacteria (SRB) and microbial sulfate

reduction occur in process waters and substrates (fine tailings) containing NAs

(Holowenko et al., 2000; Salloum et al., 2002; Quagraine et al., 2005; Ramos-Padron et al., 2011; Liu et al., 2015); however, there are limited data measuring effects of NAs on production of acid volatile sulfides (AVS) and subsequent metal precipitation. To effectively design and implement passive or semi-passive water renovation strategies utilizing microbially driven AVS production, potential adverse effects on SRB (e.g.

Desulfovibrio, Desulfosporosinus, Desulfobulbus) following exposures to NAs must be investigated.

Precipitation of metals via sulfides is a naturally occuring and microbially mediated pathway that has been sucessfully implemented in constructed wetlands for treatment of a variety of impaired waters (e.g. refinery effluents, oilfield produce waters, acid and neutral mine drainage; and urban and industrial stormwaters; Gillespie et al., 1999;

Murray-Gulde et al., 2003; Johnson et al., 2008; Horner et al., 2012; Haakensen et al.,

2015). Treatment is achieved by targeting the lithic biogeochemical sequestration of

divalent metals through sulfide (i.e. S2-, HS-) precipitation as mineralized species (e.g.

chalcocite [CuS], covellite [Cu2S]). These sulfide bound species are relatively insoluble

-16 (CuS; ksp=10 ; Stumm and Morgan 1996), and are transferred from the water column

into the hydrosoil as non-bioavailable fractions (Murray-Gulde et al., 2003; Huddleston

and Rodgers 2008). Anaerobic conditions with relatively low oxidation-reduction

potentials (ORP; -250 to -100 mV) are necessary for promoting anaerobic metabolisms in

bacteria which oxidize organic matter, producing electrons which reduce sulfate to

- hydrogen sulfide (H2S) and other reduced sulfide species (i.e. bisulfide ion (HS ), sulfide

ion [S2-]; Mitsch and Gosselink 2007). AVS is a commonly used operational

measurement of the amount of sulfide in sediments (Allen et al., 1993; Rickard and

Morse 2005). AVS is a measure of the reactive “pool” of sulfides available as ligands that

can be compared to the molar ratio of simultaneously extracted metals (SEM; Di Toro et al., 1992). SEM and AVS ratios as a predictive measure have been used extensively for estimating ecological risk of divalent metal exposures (Di Toro et al., 1992; Ankley et al.,

1996; Boothman et al. 2001; Burton et al. 2005). SEM:AVS ratios are useful for constructed wetland treatment design and monitoring, indicating the divalent metal treatment removal capacity which is likely to occur (Rodgers and Castle, 2008).

In this study, multiple lines of evidence were used to discern potential effects of NAs on SRBs and production of sulfides. Concentrations of AVS, SEM, and aqueous metals

(Cu, Ni, and Zn) following exposures of NAs were measured. In addition, microbial analyses provided valuable information to discern changes in population densities, relative abundance, and diversity (Haakensen et al., 2015). Changes in total quantity, relative abundance, and types of SRBs may implicate shifts in target treatment processes

(i.e. dissimilatory sulfate reduction). Commercially available (Fluka) NAs were used for this study to provide a reliable and reproducible source of NAs. Fluka NAs have been

well studied as a simplistic analogue to understand more compositionally complex mixtures of NAs present in energy-derived waters (Headley and McMartin 2004; Barrow et al., 2004; Scott et al., 2005; Rudzinski et al., 2002; Lo et al., 2006; Armstrong et al.,

2008; Headley et al., 2010). In general, commercial NAs are more potently toxic to sentinel organisms (i.e. macroinvertebrates, fish, plants; Swigert et al., 2015; Kinley et al., 2016) as compared to NAs in process-affected waters. Therefore, in this study, commercial NAs offer a conservative exposure to evaluate the vulnerabilities of SRB.

The overall objective of this research was to measure responses of sulfate-reducing bacterial assemblages and microbially mediated treatment pathways (e.g. acid volatile sulfide concentrations) following a series of exposures to a commercial (Fluka) NA in bench-scale reactors. To achieve this overall objective, specific objectives were to: 1) measure relationships of acid-volatile sulfide (AVS), simultaneously extractable metal

(SEM), and aqueous metal (copper, nickel, and zinc) concentrations following 21-d exposures to NAs, and 2) measure responses of sulfate-reducing bacterial (SRB) assemblages in terms of relative abundance, diversity, and density (most-probable number) in sediment to 21-d exposures of NAs.

4.3 Materials and Methods

4.3.1 Bench-scale exposure chambers

Bench-scale experiments were conducted in 1-L borosilicate glass jars with 500 g of sediment and 500 ml of reconstituted moderately hard water (pH 7.7 ± 0.5 SU, alkalinity

65 ± 8 mg/L as CaCO3, hardness 88 ± 10 mg/L as CaCO3, conductivity 350 ± 20 µs/cm;

USEPA 2002; Table 4.1). Surficial sediment samples from 0-10 cm depth were collected

from the Clemson Aquaculture Facility (34°40'6.14"N, 82°50'52.02"W) in a shallow

wetland (water depth <30 cm) dominated by cattail (Typha latifolia). Prior to experiment

initiation, sediment samples were added to the exposure chambers and amended with

organic matter (5% wheat hay [w/w]) to support dissimilatory sulfate reduction by

targeting a negative ORP (i.e. -100 to -250 mV; Brookins, 1988). Measurements of

sediment characteristics included particle size distribution, percent solids, organic matter

content, and cation exchange capacity (Table 4.1). Sediment used in this experiment was

mostly silt (41%) and clay (32%), with a sand content of approximately 27%. Organic

matter content, percent solids, and pH of the sediment were 10.8 %, 46 %, and 5.4 SU,

respectively. Major cations in the sediment were 788 mg Ca/kg, 86 mg Mg/kg, and 46 mg

Na/kg, with a cation exchange capacity of 8.2 mEq/100 g. Prior to initiation, divalent

metals measured in the un-amended sediment included 5.4 mg Cu/kg, 1.18 mg Ni/kg, and

35.1 mg Zn/kg. In addition, there was no detectible AVS (MDL = 0.002 µmol/g) in

sediments prior to test initiation. Response parameters (e.g. AVS, SEM, and aqueous metal concentrations) were measured at 3-d intervals for 21-days. All exposures were conducted at ambient room temperatures ranging from 22-24°C.

Experiments were designed as static/renewal systems to simulate the hydraulic retention time (HRT) of a constructed wetland treatment system. Every 3-days, water renewals were conducted for each treatment chamber by removing ~50% of the water volume with an electronic pipet (Fisher Scientific; Pittsburgh, PA) fitted with a 100 mL glass tip. Untreated controls received no amendments of NAs. Negative controls were

prepared using a SRB biocide, tetrakis (hydroxymethyl) phosphonium sulfate (THPS;

Sigma-Aldrich; St. Louis, Mi), which is commonly used to inhibit growth of SRBs in the

petroleum industry (Downward et al., 1997). THPS was added to negative controls

(termed “SRB biocide”) at nominal concentrations of 25 mg THPS/L at each water

renewal.

4.3.2 Naphthenic acid exposures and analyses

To achieve a series of nominal exposure concentrations of 10, 20, 40, 60, and 80

mg/L NA, a commercially available (Fluka) NA (Sigma-Aldrich; St. Louis, Mo) was

pipetted into deoxygenated laboratory formulated moderately hard water adjusted to a pH

of ~8.3 with 0.1 molar NaOH (Fisher Scientific; Pittsburgh, PA). The water

accommodated fraction (WAF) method was used to prepare NA stock solutions, where

NA solutions were mixed with magnetic stir bars for 24 hours at a speed sufficient to create a vortex which extended 30-50% of the solution depth (OECD, 2000; Swigert et

al., 2015; Kinley et al., 2016 ; McQueen et al., 2016). Prior to analysis, samples and

standards were adjusted to a pH of 8-10 SU using 3 M NaOH and added to 1.5 mL

borosilicate amber glass vials. Methods for NA derivatization and subsequent analysis

using high performance liquid chromatography (HPLC; Dionex, UltiMate-3000;

Sunnyvale, CA) were based on Yen et al. (2004). The HPLC column was an Agilent

LiChrospher 100 RP-18 (5 µm particle size, 125mm x 4 mm) with a guard column packed with 2 µm RP-18 solid phase material. Column temperature was maintained at

40° C with a sample injection volume of 60 µL mobilized with HPLC grade methanol

100

(Fisher Scientific) at a flow rate of 60 µL/min. Recovery results ranged from 85-115%.

Naphthenic acid concentrations were reported as mean of triplicate analyses.

4.3.3 Acid volatile sulfide (AVS) and simultaneously extracted metal (SEM) analyses

AVS and SEM were measured using the diffusion method (Leonard et al., 1996).

Sulfide ions trapped in sulfide anti-oxidant buffer (SAOB) were measured using an ion- selective electrode (ISE; Fisher Accumet 950 pH/ion meter) to determine the molar concentration of AVS. The SEM soil-acid extracts were vacuum-filtered through 0.45

µm nitrocellulose Millipore® membranes and metal concentrations were measured (as described above) and reported as molar concentrations. Quality assurance and quality

control for AVS analyses included replicate measurements of AVS concentrations in sulfide standards. Reported AVS measurements were within ± 10% of the sulfide standards.

4.3.4 Statistical Analysis

Data were tested for normal distribution and homogeneity of variance using Chi-

square and Bartlett’s tests, respectively. Normally distributed, homogeneous data were

analyzed by one-way analysis of variance (ANOVA). Differences among treatments

were determined using follow-up pairwise comparisons and contrasts using general linear

models. Differences were considered significant at p ≤ 0.05 (JMP v11; SAS Institute Inc.,

Cary, NC, USA).

4.3.5 Aqueous metal amendments and analyses

To achieve the nominal concentrations of divalent metals (i.e. Cu, Ni, and Zn),

reagent grade salts (Fisher Scientific; Table 4.2) were added to WAF solutions and

101

magnetically stirred for 24-hrs (as described above) prior to renewals. Metals

concentrations were measured using Inductively Coupled Plasma-Optical Emission

Spectrometry (ICP-OES; Perkin Elmer; APHA, 2012). Quality assurance and control for analytical methods included comparisons to standard curves, replicate analysis, matrix spike recovery, and blank spike recovery.

Concentrations of metals (Cu, Ni, Zn) following 3-day HRTs were compared to

United States Environmental Protection Agency (USEPA) and Environment Canada water quality guidelines (USEPA 1999; CCME 2005). Comparisons of metal concentrations to water quality guidelines provided context for achieving treatment performance in terms of extent of removal. Additionally, metal concentrations among treatments were compared to SRB biocide treatments to contrast potential confounding factors due to competing removal pathways (e.g. sorption to clay minerals and organic matter, microbial reduction, etc.).

4.3.6 Microbial Analyses

4.3.6.1 Most-probable number assay

Most-probable number (MPN) assays were performed for sulfate-reducing bacteria using a modified Sulfate API RP-38 broth with the following per liter distilled water: 4 mL 60% sodium lactate, 1 g yeast extract, 0.01 g K2HPO4, 0.2 g MgSO4*7H2O,

0.2 g Fe(NH4)2(SO4)2*6H2O, 0.1 g ascorbic acid, 10 mL 20% sodium acetate, and 0.1 g

thioglycolic acid. Total anaerobic heterotrophs were also enumerated using R2A medium

(HiMedia Labs, 2011). Sediment samples were diluted 1/100 with 0.1% peptone solution,

and further serially diluted from 1/1,000 to 1/1,000,000,000 along a sterile 96-microwell

102

round-bottom plate containing the respective growth media. Plates were sealed with

Breathe-Easy sealing membranes (Diversified Biotech) and placed in AnaeroPack

Rectangular Jars with AnaeroPack-Anaero anaerobic gas generators (MGC). Plates were incubated without light at 20°C and monitored for visible growth (formation of a pellet,

R2A media) and/or color change to black (SRB media) until no changes were observed for three subsequent days (34 days). The MPN of microbes was then calculated as described by Blodgett (2010).

4.3.6.2 Microbiome genetic sequencing and analyses

DNA was extracted from 0.25 g of sample using the MO BIO PowerLyzer

PowerSoil DNA Isolation Kit as per the manufacturer’s protocol. Targeted DNA sequencing was used to identify bacteria present in samples via polymerase chain reaction (PCR) amplification of the v3/v4 region of the 16S ribosomal RNA gene

(Klindworth et al., 2013). Library preparation and sequencing was performed as per the manufacturer’s instructions for MiSeq v3 paired-end 300 bp sequencing (Illumina).

Library preparation included positive and negative controls, with the former consisting of mock communities, and the latter where no DNA is added to the PCR step, and the sample is carried through to sequencing. After sequencing, the forward and reverse reads were merged using PANDAseq (Masella et al., 2012), and all sequences then filtered to remove primer sequences and discard: low quality reads (Q < 30), reads containing N

(unknown) bases, and reads shorter than 350 bp. Bioinformatics pipelines consisting of internally developed scripts and selected QIIME scripts (Caporaso et al., 2010; Edgar,

2010) were used to process the reads. Similar sequences were clustered into groups called

103

Operational Taxonomic Units (OTUs) using a 97% identity threshold and the QIIME pick_de_novo_otus.py script. All OTUs with less than 10 representative sequences in at least one sample were discarded. Taxonomic classification of OTUs was performed using the Greengenes database version 13_8 (DeSantis et al., 2006; McDonald et al., 2012).

4.4 Results and Discussion

4.4.1 Extent and rate of acid volatile sulfide (AVS) production

AVS concentrations were similar among NA treatments during the 21-day bench-

scale experiment. With the exception of AVS concentrations measured on days 12 and 15

in the highest NA treatment (80 mg NA/L), there were no statistical differences among

AVS concentrations in NA treatments compared to the untreated control (no NAs added;

Figure 4.1). From day 3 to day 15, average AVS concentrations increased from 5.2 to

44.4 µmol/g and 7.7 to 24.4 µmol/g in the untreated control and the highest NA treatment

(80 mg NA/L), respectively. There was a slight statistical difference in AVS

concentrations measured on day 12 (p=0.0436; α=0.05) and day 15 (p=0.0238; α=0.05) in

80 mg NA/L treatments compared to the untreated control. However, no statistical differences of AVS concentrations in 80 mg NA/L treatments (as compared to the untreated control) were observed on days 18 or 21 (Figure 4.1). In contrast, significantly lower concentrations of AVS in SRB biocide treatments (as compared to untreated controls; p<0.0001; α=0.05) confirmed AVS production in treatments and untreated controls were due to microbially mediated sulfate reduction. After 9 days, black precipitates (presumably sulfide precipitates) were observed in the surficial water-

104

sediment interface in all NA treatments and untreated controls, but were absent in the

surficial water-sediment interface in SRB biocide treatments. The rate of AVS increase

was an apparent first order reaction, with AVS concentrations increasing at a rate of

0.225 and 0.158 day-1 (measured from day 3 to day 12) for the untreated control and 80

mg NA/L treatment, respectively (Figure 4.2). Since AVS concentrations in NA

treatments 10, 20, 40 and 60 mg/L were similar to the untreated control, rates of AVS

production were not compared.

4.4.2 Simultaneously extracted metals (SEM) and acid volatile sulfide (AVS) ratios

In terms of ∑SEM:AVS ratios (∑SEM = Cu, Ni, Zn), all NA treatments and the

control (no NA added) achieved greater molar concentrations of AVS as compared to

simultaneously extracted metals (i.e. ∑SEM:AVS <1; Figure 4.3) during the 21-day experiment duration. In contrast, SRB biocide controls had greater molar concentrations of SEM (∑SEM ranged from 0.183 to 0.857 µmol/g) as compared to AVS, with

∑SEM:AVS ratios exceeding 1 following 12 days of the experiment. The ∑SEM:AVS results indicate NAs did not adversely affect the production of AVS to an extent that would alter metal binding and subsequent precipitation.

In specifically designed passive treatment systems (e.g. constructed wetlands),

AVS are produced by microbially-catalyzed processes that use the plant biomass.

Divalent metals, represented as M2+ in equation 1, react with bisulfide or hydrogen sulfide (i.e. AVS) and precipitate as insoluble metal sulfides (Kosolapov et al., 2004).

2+ + H2S + M → MS(solid) + 2H (Eq. 1)

105

These metal sulfides are relatively insoluble (CuS; ksp=10-16; NiS; ksp= 10-19; ZnS

ksp=10-25; Stumm and Morgan 1996). The amount of AVS necessary for achieving treatment goals is site-specific, based on treatment goals and inflow water characteristics

(e.g. M2+ concentrations). Comparison of molar ratios of SEM:AVS are useful

parameters for predicting treatment capabilities for divalent metals (e.g. Cd, Cu, Ni, Pb,

Zn, etc.) and can inform readiness and performance of constructed wetland treatment

systems (Rodgers and Castle 2008; Haakensen et al., 2015). The goal for achieving

transfer and transformation of divalent metals via dissimilatory sulfate reduction is to

maintain an excess of molar concentrations of AVS as compared to the summation of the

SEM (∑SEM). In this study, excess molar ratios of AVS were achieved in all NA

treatments tested; however, the question of the aqueous metal removal extent remains.

4.4.3. Copper, nickel, and zinc removal extent

In this study, Cu, Ni, and Zn exposure concentrations were renewed every 3 days

at nominal concentrations of 0.5, 0.5, 1.0 mg/L, respectively. Following amendments,

aqueous metal concentrations precipitously declined in untreated controls (no NA added)

and NA treatments, with >95% removal of Cu, Ni, and Zn achieved on days 12 through

21 (Figure 4.4). Aqueous concentrations of Cu, Ni, and Zn were compared to water

quality criteria established for the protection of aquatic life (USEPA 1999; CCME, 2005)

to provide context for the environmental significance of the treatment extent. Following

12 days, metal removal extents for Cu, Ni, and Zn in untreated controls (no NA added)

and all NA treatments were below targeted quality criteria for ambient water

concentrations (Figure 4.4). In contrast, aqueous metal concentrations for Cu, Ni, and Zn

106

were significantly higher (p<0.0001; α=0.05) in SRB biocide controls as compared to untreated controls, and always in excess of the targeted guideline concentrations.

In SRB biocide controls, sulfate reduction was inhibited and AVS production was minimal; therefore, higher aqueous concentrations of metals were anticipated based on

∑SEM:AVS ratios previously discussed (Figure 4.3). In SRB biocide treatments, there was some metal removal observed (ranging from ~40 to 80%), likely due to formation of oxides of manganese and iron (Brookins 1988), as well as sorption and complexation with organic matter (Besser et al., 2003) and clay (Hoss et al., 1997). However, sorption sites in detritus and sediment are limited, as available sites “fill” with time (Machemer and Wilderman 1992). For this reason, sulfide precipitation is the preferred treatment process in passive systems (as compared to sorption) for achieving long-term treatment goals (Machermer and Wilderman, 1992; Murray-Gulde 2003). In this study, the range of

NA concentrations tested had no apparent effect on dissimilatory sulfate production and precipitation of Cu, Ni, and Zn from the aqueous phase.

4.4.4 Water and Sediment Characteristics

Measured water and sediment characteristics were consistent among untreated controls and NA exposed treatments (Table 4.3). Exposures to NAs were confirmed analytically to be near nominal concentrations, with relative percent differences (RPD) ranging from 3.2 to 15.5%. NAs were added to moderately hard laboratory formulated water buffered with bicarbonate to minimize confounding exposure conditions that may influence solubility of NAs. Initial pH was approximately 8.15 at testing renewals, with overlying water pH declining slightly between water renewals (range 6.35 to 8.25).

107

Average in situ pH ranged from 7.57 to 7.62 during the 21-day experiment, thus pH was not likely a confounding treatment variable. During the 21 day experiment duration, average (n=6) sediment ORP among NA treatments ranged from -219 to -201 mV, indicating that reducing conditions were achieved to support the target biogeochemical processes. Average dissolved oxygen (DO) concentrations were slightly higher in biocide treatments (1.53 mg/L) as compared to NA treatments (0.61 mg/L [80 mg NA/L]) and untreated controls (0.93 mg/L; Table 4.3), supporting that microbial activity was decreased in biocide treatments (i.e. less biological oxygen demand). A slightly lower average (n=6) sediment ORP was observed in SRB biocide treatments (-184 mV), indicating a potential influence of THPS on microbial assemblages capable of oxidizing organic material and producing elections.

4.4.5 SRB density, diversity and relative abundance

Samples for microbial analyses were collected from all treatments on day 15, representing the treatment duration where the greatest differences of AVS concentrations among treatments were observed (i.e. RPD among control and 80 mg NA/L was 48%).

Growth-based analysis of SRB and total anaerobic heterotrophs (as most probable number/gram) demonstrate no difference in density among untreated controls and NA exposures, and no measurable relationship between the concentrations of Fluka NA and

SRB or total anaerobic heterotroph densities (Fig 5 and data not shown, respectively). In contrast, SRB biocide treatments impacted the abundance of SRB as well as total anaerobic heterotrophs, although some residual SRB and anaerobic heterotrophs were observed (Fig. 5 and data not shown). This may indicate that the SRB biocide treatment

108

was not completely biocidal, or was biostatic. Alternatively, it may be that the biocide

(THPS) did not effectively penetrate below the surficial sediment-water interface, and there was carry-over from “deeper” sediments during sampling. Nonetheless, this control exhibited substantially less SRB density relative to experimental samples, which is consistent with the higher aqueous concentrations of metals and ∑SEM:AVS ratios in the

SRB biocide control treatment.

Genetic sequencing was used to assess the community composition and diversity of bacterial communities in each NA treatment and control. Results demonstrate that NA exposure, regardless of concentration, did not notably alter the diversity or relative abundance of bacterial community members. Further, NA did not affect the presence, types, and relative abundance of SRB, with 5.6-6.2% of the bacterial community representing known SRB in the control and each NA concentration (Figure 4.6). In contrast, the biocide treatment resulted in a lower proportion of the bacterial community representing SRB (i.e. relative abundance was 1.5% in SRB biocide treatments as compared to 5.6-6.2% in NA treatments).

The dominant SRB in all samples including the SRB biocide treatment were

Desulfobulbus, Desulfovibrio, and Desulfosporosinus (Fig. 6). In addition to SRB, several genera known to produce sulfides through reduction of other sulfur compounds such as thiosulfate and sulfur (e.g., Fusibacter, Geobacter, Sulfurospirillum) were present in all NA and control treatment samples and to a lesser extent in the biocide treatment.

These organisms would be another source of sulfides for the precipitation of divalent metals (i.e. Cu, Ni, Zn), and were similarly not affected by the presence of NA.

109

Commercially available NAs contain compounds with relatively lower molecular weights (<< 500 Da) and numbers of ring structures than NAs found in energy-derived waters (Scott et al., 2005; Han et al., 2009; Headley et al., 2010), which can result in increased bioavailability and toxicity of commercial NAs to microorganisms (Holowenko et al., 2002; Clemente and Fedorak, 2005). Factors or conditions that could influence NA exposures, SRB densities, and AVS production (i.e. nutrient availability, aqueous sulfate and dissolved oxygen concentrations, pH, ORP, and temperature) were controlled in this study to focus on discerning the potential effects of NAs on SRB and consequent AVS production. Since no effects on SRB due to exposures of a commercial NA could be discerned within the constraints of this study, effects on SRB and subsequent AVS production due to exposures of more compositionally complex NA mixtures are not anticipated.

4.4.6 Implications to water renovation strategies

Lack of NA toxicity to SRB could be beneficial for water renovation strategies by passive or semi-passive methods such as constructed wetland treatment systems that incorporate transfer and transformation pathways for treating process-affected waters.

NAs have been a primary source of toxicity in oil sands process-affected waters

(Schramm, 2000; Clemente and Fedorak, 2005; Allen, 2008a) and refinery effluents

(Dorn, 1992), and therefore, questions arise regarding the potential effects of NA exposures on SRB and subsequent sulfide production. In this conservative laboratory toxicity experiment, no adverse effects on SRB could be discerned. In addition, rates of

AVS production were sufficient to achieve excess molar ratios of AVS to SEM,

110

indicating that available sulfide ligands were well in excess of aqueous metal

concentrations regardless of NA exposure concentration within the range tested (10-80

mg NA/L). Further, SRB were relatively insensitive to commercial NA exposures. Thus,

within the bounds of the experimental conditions in this study, effects on performance of

metal-sulfide precipitation pathways due to co-occurring exposures of NAs are not

anticipated in field situations.

4.5 Conclusions

In this controlled laboratory toxicity experiment using multiple lines of evidence

(i.e. [AVS], ∑SEM:AVS, aqueous metal removal extents, and microbial analyses), dissimilatory sulfate reduction and metal precipitation were not influenced by exposures of a commercial (Fluka) NA. Following exposures of NAs, extent of AVS production were sufficient to achieve ∑SEM:AVS <1, indicating that available sulfide ligands were

in excess of SEM (Cu, Ni, and Zn) concentrations regardless of NA exposure

concentration (10-80 mg NA/L). In addition, no adverse effects to SRB populations in terms of density, diversity, or relative abundance were measured following exposures of a commercial NA. Since SRB were insensitive to exposures of a relatively potent (in terms of aquatic toxicity) commercial NA, adverse effects to SRB (and SRB-mediated pathways in wetlands) from exposures of more compositionally complex NAs (i.e. derived from oil sands process affected waters) are not anticipated. Further, lack of toxicity to the overall microbial population and absence of effect on diversity and community profile is a positive finding, given that maintaining diversity of microbially-

111

mediated pathways could be beneficial for passive wetland treatment systems. Passive

systems that utilize these biogeochemical processes can be cost effective alternatives to

traditional technologies, and having a robust microbial community capable of performing these processes can improve the efficiency and success of the system (Johnson and

Hallberg, 2005; Nelson and Gladden, 2008; Haakensen et al., 2015). For effective design and operation of passive treatment systems (i.e. constructed wetlands) which utilize

biogeochemical processes, it is useful to discern adverse effects due to co-occurring

constituents on SRB which could alter the dissimilatory sulfate reduction and limit

treatment of metals. In this study, dissimilatory sulfate reduction and subsequent metal

precipitation were not vulnerable to NAs, indicating passive treatment systems could be

used to treat metals occurring in NA affected waters.

4.6Acknowledgments

Funding support for this research was provided by Shell Canada Ltd. and Suncor Energy.

The authors are also grateful to Dr. Wayne Chao of Clemson University for providing

analytical support and Jenny Liang and Ainsley Stewart of Contango Strategies Ltd. for

providing support for microbial assays.

112

4.7 References

Allen, E.W., 2008a. Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Environ. Eng. Sci. 7, 123-138.

Allen, E.W., 2008b. Process water treatment in Canada’s oil sands industry: II. A review of emerging technologies. Environ. Eng. Sci. 7, 499-524.

Allen, H. E., Fu, G., Deng, B., 1993. Analysis of acid‐volatile sulfide (AVS) and simultaneously extracted metals (SEM) for the estimation of potential toxicity in aquatic sediments. Environ. Toxicol. Chem. 12, 1441-1453.

American Public Health Association (APHA), 2012. Standard methods for the examination of water and wastewater, 21st Ed. American Public Health Association, Port City Press: Baltimore, MD, pp. 1368.

Ankley, G.T., Thomas, N.A., Di Toro, D.M., Hansen, D.J., Mahony, J.D., Berry, W.J., Zarba, C.S., 1994. Assessing potential bioavailability of metals in sediments: a proposed approach. Environ. Manage. 18, 331-337.

Armstrong, S.A., Headley, J.V., Peru, K.M., Germida, J.J., 2008. Phytotoxicity of oil sands naphthenic acids and dissipation from systems planted with emergent aquatic macrophytes. J. Environ. Sci. Heal. A. 43, 36-42.

Barrow, M.P., Headley, J.V., Peru, K.M., & Derrick, P.J., 2004. Fourier transform ion cyclotron resonance mass spectrometry of principal components in oil sands naphthenic acids. J. Chromatogr. A. 1058, 51-59.

Besser, J M., Brumbaugh, W.G., May, T.W., Ingersoll, C.G., 2003. Effects of organic amendments on the toxicity and bioavailability of cadmium and copper in spiked formulated sediments. Environ. Toxicol. Chem. 22, 805-815.

Blodgett, R., 2010. US Food and Drug Administration, Bacteriological analytical manual, Appendix 2: Most probable number from serial dilutions. Accessed online April 2016. http://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm109656.

Boothman, W.S., Hansen, D.J., Berry, W.J., Robson, D.L., Helmstetter, A., Corbin, J. M., Pratt, S. D., 2001. Biological response to variation of acid‐volatile sulfides and metals in field‐exposed spiked sediments. Environ. Toxicol. Chem. 20, 264-272.

Brookins, D.G., 1988. Eh-pH Diagrams for Geochemistry. Springer-Verlag, New York, US.

113

Brown, L.D., Ulrich, A.C., 2015. Oil sands naphthenic acids: A review of properties, measurement, and treatment. Chemosphere. 127, 276-290.

Burton, G.A., Nguyen, L.T., Janssen, C., Baudo, R., McWilliam, R., Bossuyt, B., Green, A., 2005. Field validation of sediment zinc toxicity. Environ. Toxicol. Chem. 24(3), 541-553.

Canadian Council of Ministers of the Environment (CCME)., 2005. Canadian Environmental Quality Guidelines. Canadian Council of Ministers of the Environment, Winnipeg, Man.

Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Knight, R., 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods. 7, 335-336.

Clemente, J.S., Fedorak, P.M., 2005. A review of the occurrence, analyses, toxicity, and biodegradation of naphthenic acids. Chemosphere. 60, 585-600.

DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Andersen, G.L., 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microb. 72, 5069-5072.

Di Toro, D.M., Mahony, J.D., Hansen, D.J., Scott, K.J., Carlson, A.R., & Ankley, G.T., 1992. Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments. Environ. Sci. Technol. 26, 96-101.

Dorn, P.B., 1992. Case histories—the petroleum industry. Toxicity Reduction: Evaluation and Control. In: Ford, DL (Ed.), Technomic Publishing Company, Inc., Lancaster, PA, 183-223.

Downward, B.L., Talbot, R.E., Haack, T.K., 1997. Tetrakishydroxymethylphosphonium sulfate (THPS), a new industrial biocide with low environmental toxicity, paper no. 401. In Corrosion/97. NACE International, Houston, TX.

Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 26, 2460-2461.

Faulkner, S.P., Patrick, W.H., Gambrell, R.P., 1989. Field techniques for measuring wetland soil parameters. Soil Sci. Soc. Am. J. 53, 883-890.

Foote, L., 2012. Threshold considerations and wetland reclamation in Alberta's mineable oil sands. Ecol. Soc. 17, 35.

114

Frank R.A., Kavanagh R., Burnison B.K., Arsenault G., Headley J.V., Peru K.M., Kraak G.V.D., Solomon K.R., 2008. Toxicity Assessment of Collected Fractions from an Extracted Naphthenic Acid Mixture. Chemosphere. 72, 1309-1314.

Gee, G.W., Bauder. J.W., 1986. Particle-size analysis. In A. Klute (ed.), Methods of Soil Analysis, Part 1, 2nd Edition. ASA and SSSA Publ., Madison, WI, pp. 383– 441.

Gillespie, W.B. Jr., Hawkins, W.B., Rodgers Jr., J.H., Cano, M.L., Dorn, P.B., 1999. Transfers and transformations of zinc in flow-through wetland microcosms. Ecotox. Environ. Safe. 43, 126–32.

Haakensen, M., Pittet, V., Spacil, M.M., Castle, J.W., Rodgers Jr., J.H., 2015. Key aspects for successful design and implementation of passive water treatment systems. J. Environ. Sol. Oil Gas Min. 1, 59-81.

Han, X., MacKinnon, M.D., Martin, J.W., 2009. Estimating the in situ biodegradation of naphthenic acids in oil sands process waters. Chemosphere. 76, 63-70.

Headley, J.V., McMartin, D.W., 2004. A review of the occurrence and fate of naphthenic acids in aquatic environments. J. Environ. Sci. Heal. A. 39, 1989-2010.

Headley, J.V., Peru, K.M., Adenugba, A.A., Du, J.L., McMartin, D.W., 2010. Dissipation of naphthenic acids mixtures by lake biofilms. J. Environ. Sci. Heal. A. 45, 1027- 1036.

Holowenko, F.M., MacKinnon, M.D., Fedorak, P.M., 2000. Methanogens and sulfate- reducing bacteria in oil sands fine tailings waste. Can. J. Microbiol. 46, 927-937.

Holowenko, F.M., MacKinnon, M.D., Fedorak, P.M., 2002. Characterization of naphthenic acids in oil sands wastewaters by gas chromatography-mass spectrometry. Water Res. 36, 2843-2855.

Horner, J.E., Castle, J.W., Rodgers Jr., J.H., Murray-Gulde, C.M., Myers, J. E., 2012. Design and performance of pilot-scale constructed wetland treatment systems for treating oilfield produced water from Sub-Saharan Africa. Water Air Soil Poll. 223, 1945-1957.

Hoss, H., Haitzer, M., Traunspurger, W., Gratzer, H., Ahlf, W., Stienberg, C., 1997. Influence of particle size distribution and content of organic matter on the toxicity of copper in sediment bioassays using Caenorhabditis elegans (Nematoda). Water Air Soil Poll. 99, 689-695.

Huddleston, G.M. III, Rodgers Jr., J.H., 2008. Design of a constructed wetland for treatment of copper-contaminated wastewater. Environ. Geosci.15, 9-19.

115

Johnson B.M., Kanagy L.E., Rodgers Jr., J.H., Castle, J.W., 2008. Feasibility of a pilot- scale hybrid constructed wetland treatment system for simulated natural gas storage produced waters. Environ. Geosci. 15, 91-104.

Johnson, D.B., Hallberg, K B., 2005. Acid mine drainage remediation options: a review. Sci. Total Environ. 338, 3-14.

Jones, D., Scarlett, A.G., West, C.E., Rowland, S.J., 2011. Toxicity of individual naphthenic acids to Vibrio fischeri. Environ. Sci. Technol. 45, 9776-9782.

Kanagy, L.E., Johnson, B.M., Castle, J.W., Rodgers Jr., J.H., 2008. Hydrosoil conditions in a pilot-scale constructed wetland treatment system for natural gas storage produced waters. Environ. Geosci. 15, 105-113.

Kavanagh, R.J., Frank, R.A., Burnison, B.K., Young, R.F., Fedorak, P.M., Solomon, K.R., Van Der Kraak, G., 2012. Fathead minnow (Pimephales promelas) reproduction is impaired when exposed to a naphthenic acid extract. Aquat. Toxciol. 116, 34-42.

Kinley, C.M., McQueen, A.D., Rodgers, Jr., J.H., 2016. Comparative responses of freshwater organisms to exposures of a commercial naphthenic acid. Ecotox. Environ. Safe. 153: 170-178.

Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., & Glöckner, F. O., 2012. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic acids Res. 41, 1-11.

Kosolapov, D.B., Kuschk, P., Vainshtein, M.B., Vatsourina, A.V., Wiessner, A., Kastner, M., Muller, R.A., 2004. Microbial processes of heavy metal removal from carbon- deficient effluents in constructed wetlands. Eng. Life. Sci. 4, 403–411.

Leclair, L.A., MacDonald, G.Z., Phalen, L.J., Köllner, B., Hogan, N.S., & van den Heuvel, M. R., 2013. The immunological effects of oil sands surface waters and naphthenic acids on rainbow trout (Oncorhynchus mykiss). Aquat. Toxciol. 142, 185-194.

Leonard, E.N., Ankley, G.T., Hoke, R.A., 1996. Evaluation of metals in marine and freshwater surficial sediments from the environmental monitoring and assessment program relative to proposed sediment quality criteria for metals. Environ. Toxicol. Chem. 15, 2221-2232.

116

Liu, H., Yu, T., & Liu, Y., 2015. Sulfate reducing bacteria and their activities in oil sands process-affected water biofilm. Sci. Total Environ. 536, 116-122.

Lo, C.C., Brownlee, B.G., Bunce, N.J., 2006. Mass spectrometric and toxicological assays of Athabasca oil sands naphthenic acids. Water Res. 40, 655-664.

Machemer, S.D., Wildeman, T.R., 1992. Adsorption compared with sulfide precipitation as metal removal processes from acid mine drainage in a constructed wetland. J. Contam. Hydrol. 9, 115-131.

Masella, A.P., Bartram, A K., Truszkowski, J.M., Brown, D.G., Neufeld, J.D., 2012. PANDAseq: PAired-eND Assembler for Illumina sequences. BMC Bioinformatics. 13, 31.

McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Hugenholtz, P., 2012. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J. 6, 610-618.

McQueen, A.D., Kinley C.M., Kiekhaefer, R.L., Calomeni A.J., Rodgers Jr., J.H., Castle, J.W., 2016. Photocatalysis of a Commercial Naphthenic Acid in Water Using Fixed-film TiO2. Submitted to Water Air Soil Poll. 227:132.

Mitsch, W.J., Gosselink, J.G., 2007. Wetlands, 5th ed., John Wiley & Sons, Inc., New York.

Murray-Gulde, C.L., Bridges, W.C., Rodgers Jr., J.H., 2008. Evaluating performance of a constructed wetland treatment system designed to decrease bioavailable copper in a waste stream. Environ. Geosci. 15, 21-38.

Murray-Gulde, C.L., Heatley, J.E., Karanfil, T., Rodgers Jr., J.H., Myers, J.E., 2003. Performance of a hybrid reverse osmosis-constructed wetland treatment system for brackish water. Water Res. 37, 705-713.

Nelson, E.A., Gladden J.B., 2008. Full-scale treatment wetlands for metal removal from industrial wastewater. Environ. Geosci. 15, 39-48.

Nelson, D.W., Sommers, L., 1982. Total carbon, organic carbon, and organic matter. Methods of soil analysis. Part 2. Chemical and microbiological properties, (methodsofsoilan2), American Society of Agronomy, Inc. Madison, Wi.

Nero, V., Farwell, A., Lee, L.E., Van Meer, T., MacKinnon, M.D., Dixon, D.G., 2006. The effects of salinity on naphthenic acid toxicity to yellow perch: Gill and liver histopathology. Ecotox. Environ. Safe. 65, 252-264.

117

Organization for Economic Cooperation and Development (OECD), 2000. Guidance document on aquatic toxicity testing of difficult substances and mixtures. OECD Series on Testing and Assessment No. 23, vol. 6. OECD, Paris, France.

Pham, M., Castle, J.W., Rodgers Jr., J.H., 2011. Biogeochemical process approach to the design and construction of a pilot-scale wetland treatment system for an oil field- produced water. Environ. Geosci. 18, 157-168.

Quagraine, E.K., Peterson, H.G., Headley, J.V., 2005. In situ bioremediation of naphthenic acids contaminated tailing pond waters in the Athabasca oil sands region—demonstrated field studies and plausible options: a review. J. Environ. Sci. Heal. A. 40, 685-722.

Ramos-Padrón, E., Bordenave, S., Lin, S., Bhaskar, I M., Dong, X., Sensen, C.W., Gieg, L.M., 2011. Carbon and sulfur cycling by microbial communities in a gypsum- treated oil sands tailings pond. Environ. Sci. Technol. 45, 439-446.

Rickard, D., Morse, J.W., 2005. Acid volatile sulfide (AVS). Mar. Chem. 97, 141-197.

Rodgers, Jr., J.H., Castle, J.W., 2008. Constructed wetland systems for efficient and effective treatment of contaminated waters for reuse. Environ. Geosci. 15, 1-8.

Rogers, V.V., Liber, K., MacKinnon, M.D., 2002. Isolation and characterization of naphthenic acids from Athabasca oil sands tailings pond water. Chemosphere. 48, 519-527.

Rudzinski, W.E., Oehlers, L., Zhang, Y., Najera, B., 2002. Tandem mass spectrometric characterization of commercial naphthenic acids and a Maya crude oil. Energ. Fuel. 16, 1178-1185.

Salloum, M.J., Dudas, M.J., Fedorak, P.M., 2002. Microbial reduction of amended sulfate in anaerobic mature fine tailings from oil sand. Waste Manage. Res. 20, 162-171.

Schramm, L. L. (Ed.), 2000. Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press, Cambridge, UK.

Scott, A.C., Mackinnon, M.D., Fedorak, P.M., 2005. Naphthenic acids in Athabasca oil sands tailings waters are less biodegradable than commercial naphthenic acids. Environ. Sci. Technol. 39, 8388-8394.

Seifert, W.K., Teeter, R.M., 1969. Preparative thin-layer chromatography and high- resolution mass spectrometry of crude oil carboxylic acids. Anal Chem. 41, 786- 795.

118

Sigma-Aldrich, 2015. MSDS, Naphthenic acid. St. Louis, Missouri.

Spacil, M.M., Rodgers Jr., J.H., Castle, J.W., Murray Gulde, C., Myers, J., 2011. Treatment of selenium in simulated refinery effluent using a pilot-scale constructed wetland treatment system. Water Air Soil Poll. 221, 301-312.

Stumm, W., Morgan J.J., 1996. Aquatic chemistry, chemical equilibria and rates in natural waters. John Wiley & Sons, Inc. New York, NY, USA.

Swigert, J.P., Lee, C., Wong, D.C., White, R., Scarlett, A.G., West, C.E., Rowland, S.J., 2015. Aquatic hazard assessment of a commercial sample of naphthenic acids. Chemosphere. 124, 1-9.

Tomczyk, N.A., Winans, R.E., Shinn, J.H., & Robinson, R.C., 2001. On the nature and origin of acidic species in petroleum. 1. Detailed acid type distribution in a California crude oil. Energ. Fuel. 15, 1498-1504.

Toor N.S., Franz E.D., Fedorak P.M., MacKinnon M.D., Liber K., 2013. Degradation and aquatic toxicity of naphthenic acids in oil sands process-affected waters using simulated wetlands. Chemosphere. 90, 449-458.

United States Environmental Protection Agency (USEPA), 1999. National recommended water quality criteria-Correction. Report Office of Water 4304, Washington DC. 822-Z-99–001.

United States Environmental Protection Agency (USEPA), 2001. EPA Method 200.7, Trace elements in water, solids, and biosolids by inductively coupled plasma- atomic emission spectrometry, Revision 5.0, Washington DC. EPA-821-R-01- 010.

United States Environmental Protection Agency (USEPA), 2002. Methods for measuring acute toxicity of effluents and receiving waters to freshwater and marine organisms. Washington DC. EPA-821-R-02-12.

Yen, T.W., Marsh, W.P., MacKinnon, M.D., Fedorak, P.M., 2004. Measuring naphthenic acids concentrations in aqueous environmental samples by liquid chromatography. J. Chromatogr. 1033, 83-90.

119

Table 4.1 Analytical methods for experimental parameters.

Method Detection Parameter Methods Limit Direct Instrumentation: Orion Model 420A (Standard pHa,b 0.01 SU Methods 4500-H+ B) (APHA, 2012) Temperaturea Direct Instrumentation: Orion Model 420A 0.01 °C Dissolved Oxygena Direct Instrumentation: YSI Model 52 0.1 mg/L Direct Instrumentation: YSI 30 (Standard Method 2510 Conductivitya 0.1 μS/cm B) (APHA, 2012) a Alkalinity Standard Methods: 2320 B (APHA, 2012) 2 mg/L as CaCO3 a Hardness Standard Methods: 2340 B (APHA, 2012) 2 mg/L as CaCO3 Particle sizeb Hydrometer method (Gee and Bauder, 1986) 2% Percent Solids %b Dry matter content: Standard Methods: 2540C 0.0001 g Organic matter contentb Loss-on-ignition at 450°C (Nelson and Sommers, 1996) 0.0001 g Cation exchange capacity Standard Methods: 2340 B (APHA, 2012) 0.1 meq/ 100g (CEC)b Oxidation-Reduction Modified standard method 2580B: Accumet ®calomel Potential (ORP)b reference electrode, Fluke® 77 III voltage meter 10 mV (Faulkner et al., 1989) AVS and SEMb,c Modified diffusion method (Leonard et al., 1996) 0.001 µmol sulfide/ g Inductively Coupled Plasma-Atomic Emissions Cd, Cu, Pb, Ni, Zna,b,c 0.002-0.042 mg/L Spectrometry (ICP-AES): 200.7 (USEPA, 2001) Sulfate-Reducing Bacteria Most probable number (MPN; Blodgett, 2010) 202 MPN/g c (SRB) Bacterial community profiling by 16S ribosomal RNA 0.002 – 0.0008 % sequencing aOverlying water measurement bSediment measurement cSurficial sediment depth (<1cm)

120

Table 4.2 Chemical sources and nominal concentrations of constituents used for aqueous exposures.

Constituent Chemical Source Nominal Concentration(s), mg/L Alkylated cyclopentane carboxylic Commercial NA 10, 20, 40, 60, 80 acids (mixture)a Copper CuSO4·5H2O 0.5 Nickel NiCl2·6H2O 0.5 Zinc ZnCl2 1 aCAS # 1338-24-5; Sigma-Aldrich (2015)

121

Table 4.3 Water and sediment characteristics measured during treatment periods 3-d to 21-d (n=21).

Conductivity Alkalinity Hardness pH (S.U.) DO (mg/L) ORP (mV)b Treatmenta (mS/cm) (mg/L as CaCO3) (mg/L as CaCO3) Mean (Range) Mean (Range) Mean (Range) Mean (Range) Mean (Range) Mean Control 7.58 (6.71-8.25) 0.93 (0.25-3.2) 1.45 (1.17-1.98) 194 (175-205) 110 (90-120) -218 10 mg NA/L 7.58 (6.57-8.15) 0.92 (0.05-2.87) 1.45 (1.21-1.82) 200 (178-212) 92 (90-95) -219 20 mg NA/L 7.55 (6.55-8.10) 0.83 (0.05-2.42) 1.49 (1.03-1.94) 196 (182-198) 105 (95-112) -201 40 mg NA/L 7.62 (6.58-8.20) 0.73 (0.25-2.47) 1.44 (1.1-1.95) 190 (185-202) 98 (92-110) -217 60 mg NA/L 7.59 (6.48-8.01) 0.76 (0.32-2.03) 1.53 (1.17-2.01) 188 (182-193) 105 (95-112) -210 80 mg NA/L 7.58 (6.4-8.18) 0.61 (0.16-1.78) 1.45 (1.23-1.92) 195 (185-200) 100 (88-115) -211 SRB Biocide 7.57 (6.39-8.11) 1.53 (1.05-2.43) 1.98 (1.65-2.05) 192 (182-198) 110 (95-118) -184 aTemperature range (21.5-24.6°C) bOxidation reduction potential (ORP) measured in situ, 2-3 cm below sediment water interface (n=6; collected at 12 and 21 day sampling periods)

122

60 SRB Biocide g /

l Control

o 40

m 10 mg NA/L

µ *

, * 20 mg NA/L S

V 20 40 mg NA/L A 60 mg NA/L * * * * 0 * * 80 mg NA/L 3 6 9 12 15 18 21 Time (days)

Figure 4.1 Acid volatile sulfide (AVS) concentrations (µmol/g) measured during 21-day testing period (n=3). Error bars indicate standard deviations. Asterisks indicate AVS concentrations significantly different (p<0.05; α=0.05) from untreated controls.

123

50 SRB Biocide Control 40 80 mg NA/L g / l

o 30 m µ ,

S 20 V A 10

0 3 6 9 12 15 Time (days)

Figure 4.2 Comparison of rate of acid volatile sulfide (AVS) production (µmol/g ) in sulfate-reducing bacteria (SRB) biocide treatments, untreated control, and 80 mg/L naphthenic acid (NA) treatments (n=3) on days 3 through 15. Error bars indicate standard deviations.

Note: Treatments 10, 20, 40, and 60 mg NA/L were similar to the untreated control.

124

SRB Biocide Control 80 mg NA/L 1000

100

S 10 V A :

M 1 E S

Σ 0.1

0.01

0.001 3 6 9 12 15 18 21 Time (days)

Figure 4.3 Simultaneously extracted metals (∑SEM; Cu, Ni, Zn; µmol/g) and acid volatile sulfide (AVS) ratios among controls and highest NA treatment (80 mg NA/L) during 21-day testing period (n=3). Errors bars indicate standard deviations.

Note: Values below dashed line indicate “excess” molar ratios of AVS as compared to

SEM. Treatments 10, 20, 40, and 60 mg NA/L were similar to the untreated control.

125

SRB Biocide Control Target WQC

10 mg NA/L 20 mg NA/L 40 mg NA/L 60 mg NA/L 80 mg NA/L

0.2 L / g m

, r e p

p 0.1 o C

0.0 3 6 9 12 15 18 21 Time (days) 0.3 L / 0.2 g m

, l e k c i 0.1 N

0.0 3 6 9 12 15 18 21 Time (days) 0.8

0.6 L / g m

, 0.4 c n i Z 0.2

0.0 3 6 9 12 15 18 21 Time (days)

Figure 4.4 Copper (top), nickel (middle), and zinc (bottom) removal extent (mg/L)

measured during 21- day treatment durations.

Note: Dashed line represents target removal goals based on water quality criteria (WQC;

Cu = 0.011 mg/L; Ni = 0.052 mg/L; Zn = 0.120 mg/L; USEPA 1999; CCME 2005).

126

Figure 4.5 MPN of sulfate-reducing bacteria per gram of sample.

Error bars indicate standard error of the average results of samples (n = 3).

127

Figure 4.6 Percentage of known SRB in the bacterial community based on genetic sequencing. Names of organisms are either genus (g) or family (f) level classification.

Relative abundance (%) for each sample is the average across three replicates. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

128

CHAPTER FIVE

PERFORMACE OF HYBRID PILOT-SCALE CONSTRUCTED WETLAND

SYSTEMS FOR TREATING OIL SANDS PROCESS AFFECTED WATER FROM

THE ATHABASCA OIL SANDS AREA

Andrew D. McQueen1,3; Maas Hendrikse1, Daniel P. Gaspari2, Ciera M. Kinley1; John H.

Rodgers Jr.1, James W. Castle2

1 Department of Forestry and Environmental Conservation, 261 Lehotsky Hall, Clemson

University, Clemson, SC 29634, USA

2Department of Environmental Engineering & Earth Sciences, 455 Brackett Hall,

Clemson University, Clemson, SC 29634, USA

3Corresponding author: Andrew McQueen

Manuscript prepared for submission to: Ecological Engineering

129

5.1 Abstract Mining leases in the Athabasca oil sands (AOS; near Ft. McMurray, Canada)

produce large volumes of oil sands process-affected water (OSPW) that contain potentially problematic constituents requiring treatment prior to surface water discharge into receiving aquatic systems. The aim of this research was to identify constituents of concern (COCs) in OSPW sourced from an AOS external tailings facility and design a hybrid pilot-scale constructed wetland treatment systems (CWTS) to decrease concentrations of COCs and subsequently mitigate risks. COCs were identified based on comparisons to ambient water quality thresholds for the protection of aquatic life (i.e.

Canadian Environmental Quality Guidelines [CEQGs], Alberta Environment Water

Quality Guidelines [Alberta WQGs], and United States Environmental Protection Agency

Water Quality Criteria [USEPA WQC]) and toxicity endpoints. Performance of the hybrid pilot-scale CWTS was evaluated by rate and extent of COC removal and change in toxicity as measured by an aquatic invertebrate Ceriodaphnia dubia. Following characterization of OSPW, specific COCs were identified as: naphthenic acids (NAs), oil and grease (O/G), metals/ metalloids (Al, B, Cu, Ni, Se, and Zn), chloride, and total suspended solids (TSS). A hybrid pilot-scale CWTS was designed to promote treatment processes to alter (transfer and transform) COCs using sequential reducing and oxidizing wetland reactors and a solar photocatalytic treatment reactor using fixed film titanium dioxide (TiO2). Performance criteria were achieved as the CWTS decreased

concentrations of NAs, O/G, and metals below protective thresholds and decreased

toxicity to C. dubia. Results from this study provide proof-of-concept data to inform

130

hybrid passive or semi-passive treatment approaches (i.e. constructed wetlands) that

could be used to mitigate COCs contained in OSPWs.

5.2 Introduction

Oil sands process-affected water (OSPW) is produced during the extraction of

bitumen, and in the Athabasca oil sands (AOS) presents considerable economic and

environmental challenges due to the accumulation and composition of these waters

(Mahaffey and Dubé 2016). OSPWs can contain a mixture of problematic constituents

(e.g., naphthenic acids [NAs], metals, metalloids, and salts) that need to be treated prior to surface water discharge into receiving aquatic systems (Allen 2008a; McQueen et al.,

2016a). A compositionally complex class of carboxylic acids (NAs; generally described by the formula CnH2n+ZO2) is present in many OSPWs that have been identified as

sources of toxicity (Verbeek et al., 1994; Morandi et al., 2015), with adverse effects

observed for fish (Nero et al., 2006; Kavanagh et al., 2012; Leclair et al., 2013;

Marentette et al., 2015a,b), macroinvertebrates (Verbeek et al., 1994; Zubot et al., 2012;

Anderson et al., 2012), macrophytes (Armstrong et al., 2008), and microorganisms

(Frank et al., 2008). In addition to NAs, there are a number of other constituents in

OSPW that may require treatment including trace metals/metalloids (Al, As, Cr, Cu, Fe,

Pb, Ni, Zn), unrecovered bitumen (e.g., oil and grease [O/G]), cations and anions, and

suspended and dissolved solids (Mackinnon and Boerger 1986; Allen 2008a; Zubot et al.,

2012; McQueen et al., 2016a). To reclaim impacted waters, wet-landscape mitigation

approaches (e.g., constructed wetlands) are among technologies considered for treatment

131

of problematic constituents (e.g., metals, organics; Allen 2008b; Foote 2012; Toor et al.,

2013; Brown and Ulrich 2015). To achieve reclamation goals in the AOS, OSPW will

need to be treated and returned to aquatic receiving systems in addition to restoring

disturbed areas to “equivalent land capabilities” of the pre-mined landscape (Allen

2008b; CEMA 2014; Province of Alberta 2014). Constructed wetlands may provide

unique opportunities for achieving these AOS restoration goals. A constructed wetland

treatment system (CWTS) could be designed to actively treat OSPW (achieving water

return goals) and transition to serve as passive wetlands (achieving reclaimed wetland

goals) following their operational lifespan. To be effective, CWTSs must be carefully

designed and constructed with explicit consideration of OSPW composition and

compatibility of materials (hydrosoil and vegetation) to promote desired treatment processes in the AOS region.

Accurate characterization of OSPWs and identification of constituents of concern

(COCs) is an initial step crucial for design and construction of successful wetlands used for treatment (McQueen et al., 2016a). In this study, COCs are elements, compounds, or parameters that interfere with the goal of releasing OSPW to freshwater receiving aquatic systems. OSPW can vary spatially, temporally, and with the extraction process employed

(Mikula 2013; Frank et al., 2016). Thus the design basis will range widely with the constituents of OSPWs and will need to be robust (e.g., incorporate hybrid components) to ensure that wetlands can perform for a range of OSPWs. The treatment design approach taken here is based on an OSPW sourced from an external tailings facility in the

AOS near Ft. McMurray, Canada.

132

CWTSs have been used successfully to treat constituents in a variety of

problematic energy-derived process waters, (Knight et al., 1999; Gillespie et al., 2000;

Eggert et al., 2008; Murray-Gulde et al., 2008), and typically cost less to construct,

operate, and maintain than conventional water treatment systems (Halverson 2004;

Mooney and Murray-Gulde 2008). To be successful for OSPWs stored in the AOS,

CWTS will need to be passive (or semi-passive), requiring low energy inputs due to the

volumes of water currently stored and requiring treatment. CWTS are designed to

incorporate features (vegetation, hydrosoil, hydroperiods) facilitating treatment processes

specifically designed to alter (transfer or transform) COCs contained in impacted waters

(Rodgers and Castle 2008). Wetland systems are unique in offering a variety of

transformation processes, including: photolysis, hydrolysis, speciation/ionization,

oxidation, reduction, and biotransformation (Rodgers and Castle 2008). By choosing

appropriate wetland macrofeatures (e.g., water depth, sediment type, plant species) that

control biogeochemical conditions, modular reactors were designed to promote specific

pathways for the transfer and transformation of organic (i.e. NAs and O/G) and inorganic

(Al, B, Cu, Ni, Se, Zn, suspended solids) COCs. More recently, benefits of incorporating

non-traditional “hybrid” components in CWTS in efforts to add or enhance treatment

processes have been recognized (Murray-Gulde et al., 2003; Johnson et al., 2008;

Haakensen et al., 2015).

Hybrid components (e.g., ozonation, solar catalysts) coupled with CWTS could

augment performance of complex waters containing petroleum-derived organic constituents (e.g., NAs; Shi et al., 2015; Vaiopoulou et al., 2015; Leshuk et al., 2016;

133

McQueen et al., 2016b) potentially increasing the rate and extent of treatment. In the

case of solar catalysts such as titanium dioxide (TiO2), the combination of excitation

energy and surface moisture produces hydroxyl- and superoxide radicals (Linsebigler

1995). To date, laboratory-scale photocatalytic degradation of both commercial and

OSPW NA mixtures has been successful (Leshuk et al., 2016; McQueen et al., 2016b);

however, larger-scale studies (i.e. pilot- or full-scale) are needed. Pilot-scale CWTSs offer the ability to experimentally test hypotheses under a variety of conditions prior to investing in full-scale CWTSs (Rodgers and Castle 2008). Factors such as treatment rates

and extents can be evaluated to determine critical full-scale design parameters (e.g.,

loading rate, wetland area, and hydraulic retention time). In addition, the use of pilot

scale CWTSs can decrease full-scale performance uncertainties.

This study uses multiple lines of evidence to measure performance of a pilot-scale

CWTS, including analytical measurements of rates and extents of COC removal (e.g.,

NAs, metals) and changes in toxicity to sentinel aquatic species. A critical step for demonstrating performance of wetland systems will include the use of bioassays to monitor alterations in exposures, providing insight to changes in constituent bioavailability and mixture interactions (e.g., synergy, additivity, antagonism). The aquatic invertebrate (Ceriodaphnia dubia Richard) used in this study is sensitive to constituents (e.g., NAs, metals) contained in OSPW (Zubot et al., 2012; McQueen et al.,

2016a). C. dubia are ecologically important to freshwater ecosystems (Pennak 1978;

Carpenter et al. 1985) and routinely used by regulatory agencies for whole effluent

134

toxicity testing as a part of discharge permit requirements (USEPA 1992; Environment

Canada 2007).

The overall purpose of this research was to evaluate the performance of a

specifically designed hybrid pilot-scale CWTS for treating OSPW. In order to achieve

this overall objective, specific objectives were to: 1) characterize OSPW in terms of

chemical composition and targeted constituents of concern, 2) design and assemble a

hybrid pilot-scale CWTS to treat target constituents in OSPW, 3) measure performance

of the pilot-scale CWTS for OSPW based on rates and extents of constituent removal,

and 4) measure the performance of the pilot-scale CWTS using toxicity testing with the

aquatic invertebrate Ceriodaphnia dubia.

5.3 Materials and Methods

5.3.1 Characterization of oil sands process waters (OSPW)

The OSPW used in this study was procured from Shell Canada Ltd.’s Muskeg

River Mine External Tailings Facility (MRM-ETF; Shell Canada Limited 2016). MRM-

ETF OSPW is produced from a surface mining operation in the AOS region (near Fort

McMurray, AB, Canada) using a Clark caustic warm water floatation process to separate

bitumen from ore in which NaOH is added to a heated (50-80°C) mixture of recycled

OSPW and river water. OSPW at the MRM-ETF facility also receives: groundwater from depressurizing aquifers associated with the oil sands ore layer, surface water and shallow aquifer water from dewatering activities, precipitation and surface water runoff, water

collected from dyke seepage control systems, and connate water.

135

NAs were derivatized and analyzed using high performance liquid

chromatography (HPLC; Dionex, UltiMate-3000; Sunnyvale, CA) according to Yen et al.

(2004) to quantify total concentrations of NA. Inductively couple plasma atomic emission spectrometer (ICP-AES; Spectro Flame Modula) was used to measure element concentrations: Ag, Al, As, B, Ba, Cd, Cr, Cu, Co, Fe, Pb, Mn, Mo, Ni, Se, and Zn following USEPA method 200.7 (USEPA 2001). General water chemistry parameters, including: pH, alkalinity, hardness, dissolved oxygen (DO), conductivity, total suspended solids (TSS) and dissolved solids (TDS), total ammonia, total phosphorus, and chemical oxygen demand (COD) were analyzed following methods described in Standard Methods for the Examination of Water and Wastewater (APHA 2012). O/G concentrations were measured using gravimetric methods with n-hexane extraction according to USEPA method 1664 revision A (USEPA 1999). Cations and anions were measured using a

Dionex ISC-2000 ion chromatograph following APHA (2012) methods.

Identification of COCs was achieved by comparing chemical or physical parameters to water quality criteria limits (i.e. CEQGs, Alberta Environment WQG, and

USEPA WQC; USEPA 2007; CCME 2011; ESRD 2014) or toxicity endpoint values

(point estimates were chosen when available; e.g., LC50s). The toxicity testing species

Ceriodaphnia dubia, Daphnia magna, rainbow trout (Oncorhynchnus mykiss), and fathead minnow (Pimephales promelas) were chosen for this comparison because of their sensitivity to many constituents found in OSPW, availability of data, and use in developing and enforcing water quality criteria standards (USEPA 2007; CCME 2011).

136

COCs are parameters that exceeded the most conservative water quality standards or toxicity endpoint values selected (Equation 1).

COCs = OSPW Parameter > WQC or Toxicity Endpoint Equation 1

5.3.2 Design and Construction of a Hybrid Pilot-scale CWTS

The hybrid pilot-scale system included a 3780 L detention tank, five wetland reactors in replicate series, and a photocatalytic reactor, (Figure 1). The pilot-scale

CWTS was built and housed in a greenhouse with natural (i.e., solar) photoperiod and regulated temperature. The photocatalytic reactor was located outside, using natural

(solar) light as the energy source. OSPW was continually mixed in a detention basin with a 0.56 kW (¾ hp) sump pump for treatment in the hybrid pilot-scale CWTS. The flow of

OSPW from the detention basis was maintained by FMI® piston pumps (Fluid Metering,

Inc., NY) calibrated to a flow rate of 34 mL/min to achieve a nominal hydraulic retention time (HRT) of 13.5 days per wetland series. Wetland outflow was stored in a 1890 L detention tank prior to inflow to the photocatalytic reactor to allow for intermittent release during targeted solar conditions.

5.3.2.1 Wetland Reactors

Each wetland reactor was constructed using HDPE containers containing hydrosoil, plants, and treatment water connected by PVC pipe (Figure 1; Table 1).

Reactors were gravity-fed following the first reactor receiving metered inflow water.

Replicate treatment systems were defined as “Series A” and “Series B”. Two types of wetland reactors were designed to achieve “bulk” oxidative or reductive conditions and are operationally defined as “oxidizing” and “reducing” wetland reactors. These wetland

137

mesocosms incorporate design features that achieve criteria supportive of transfer and

transformations targeting COCs present in OSPW and are compatible within the AOS

region. The contributions of these design features are discussed in detail in section 5.4.2.

5.3.2.2 Photocatalytic reactor

The photocatalytic (PC) reactor was constructed using stainless steel sheets placed

in a 60 cm wide by 182 cm long fiberglass container (Table 1). The reactor was a

shallow depth (water depth of ~1 cm) flow-through reactor designed to achieve nominal

HRTs during hours of direct sunlight. A metered FMI® piston pump was calibrated to achieve a HRT of 12 hours of direct sunlight using a timer switch. Epoxy resin and hardener (105 Epoxy Resin; 206 Hardener; West Marine Products, Inc., Watsonville,

California) were applied as a <0.2 mm film on the stainless steel sheets followed by

TM application of TiO2 (Aeroxide P25; Fisher Scientific, Fairlawn NJ). TiO2 used a

mixture of 10-20% rutile and 80-90% anatase, with a mean particle diameter of

approximately 21 nm and a specific surface area of 35-65 m2/g. Ultraviolet (UV [250-

400nm]; Apogee SU-100, Logan, UT) and visible irradiance (HOBO Pendant®

Temperature/Light 64K Data Logger, Bourne, MA) were measured continually

throughout the treatments (supplementary material).

5.3.3 Performance Monitoring

The following parameters were measured to monitor performance of the hybrid

pilot-scale wetland for removing COCs (e.g. organics, divalent metals, suspended solids).

To monitor COC removal, as well as the rate of removal given different retention times, influent from the detention basin and outflows from reactors were sampled. In situ pH,

138

temperature, conductivity, and dissolved oxygen were measured bi-weekly in the water column of each wetland reactor (APHA 2012). Alkalinity and hardness were measured every two weeks in samples from wetland reactors (APHA 2012). Sediment oxidation- reduction potential (Eh [mV]) was measured every two weeks in the sediment at 2.5 cm depth in the middle of each wetland reactor using in situ platinum-tipped electrodes and an Accumet® calomel reference electrode with a Gardner Bender® GDT-311 voltmeter

(Faulkner et al., 1989). Acid volatile sulfides (AVS) were measured monthly from surficial hydrosoil samples using the diffusion method (Leonard et al., 1996), where sulfide ions trapped in sulfide anti-oxidant buffer (SAOB) were measured using an ion- selective electrode (ISE; Fisher Accumet 950 pH/ion meter) to determine the molar concentration of AVS.

To assess treatment performance of the pilot-scale systems, removal efficiency and rate coefficients were estimated using the following equations:

Equation 2

Where, initial concentration of COCs are denoted [C]0 (mg/L), and [C] (mg/L) is concentration of COCs in the outflow of the system. Removal rate coefficients (k, day-1) were estimated using first order rate kinetics described by the following equation:

Removal Rate Coefficient Equation 3 Where, t (days) is time, which for this purpose is at set intervals in the duration of the

HRT. The removal rate coefficient (day-1) represents the slope of the line by plotting – ln([C]/[C]0) versus time assuming first-order kinetics.

139

Performance was assessed over five “treatment periods” (treatment dates are provided in supplementary material). With the exception of the PC reactor, each

“treatment period” is equivalent to the HRT of the wetland system (13.5 days).

Performance was monitored by collecting samples following a plug flow to minimize potential influences of inflow variability.

5.3.4 Toxicity Testing

Toxicity experiments were performed with C. dubia exposed to untreated (inflow) and treated (outflow) OSPW. C. dubia were cultured based on USEPA methods (2002) at

Clemson University’s Aquatic Animal Research Laboratory. Test organisms were ≤ 24 h old at the initiation of each experiment. C. dubia (n=20) were exposed to untreated

(inflow) and treated (outflow) OSPW in 7-8 day static/renewal toxicity tests conducted following a USEPA and Environment and Climate Change Canada freshwater toxicity testing protocol (USEPA 2002; Environment Canada 2007). Toxicity to C. dubia was evaluated by measuring survival and reproduction. Statistical differences for mortality and reproduction in treatments relative to untreated controls were determined using one way analysis of variance (ANOVA) and Dunnett’s multiple range test (α = 0.05; JMP Pro

V.11; SAS Institute Inc., Cary, NC, USA).

140

5.4 Results and Discussion

5.4.1 Characterization of OSPW

5.4.1.1 Organics

OSPW evaluated in this study had a slight hydrocarbon sheen, with oil and grease

(O/G) concentrations ranging from 8 to 13 mg/L (Table 2). Concentrations of O/G (an aggregate measure of residual hydrocarbons) in OSPW exceeded narrative criteria for discharge water (i.e. no visible film or sheen of oil present; Table 2). OSPW NAs were measured at concentrations ranging from 80 to 128 mg NA/L (HPLC; n=10). Although there are currently no established numeric water quality criteria or guidance for NAs contained in OSPW, based on the concentrations and species of NAs present in OSPW as compared to reported toxicity endpoints, there is evidence that NAs may pose risk to aquatic biota, supporting the designation of NAs as a COC for this study (Table 2).

5.4.1.2 Water characteristics (e.g. pH, suspended solids) and nutrients (ammonia, TP)

OSPW is well buffered (alkalinity ranges from 320 to 340 mg/L as CaCO3) due to

- bicarbonate concentrations (HCO3 ; ranges from 300 to 320 mg/L) and as a result pH of

OSPW is relatively stable, ranging from 7.91 to 8.45 (Table 2). OSPW hardness and

conductivity range from 160 to 178 mg/L (as CaCO3) and 1791 to 1800 µS/cm,

respectively. Total suspended solids concentrations range from 130 to 400 mg/L in

OSPW, and exceed narrative WQC, which are based on the potential for increasing

background turbidity in receiving systems (CCME 2011). Water characteristic parameters

observed for this source OSPW are generally representative of other OSPWs located in

the AOS (Allen 2008a; Mahaffey and Dubé 2016).

141

Based on maximum concentrations of ammonia present in OSPW in this study

(0.099 mg/L), ammonia was not identified as a COC. Published concentrations of ammonia in other OSPWs have been reported as high as 18.4 mg/L (± 1.2; Lai et al.,

1996), exceeding toxicity values or water quality criteria guidelines (CCME 2011; ESRD

2014). Ammonia toxicity is dependent upon pH and temperature (Thurston and Russo

1981). At 14.1°C and pH of 8.29, the 96 h LC50 of ammonia to rainbow trout is 0.563 mg/L (Thurston and Russo 1981), approximately 5.5x greater than ammonia concentrations observed in the source OSPW. In this study, the maximum total phosphorus concertation (0.082 mg/L) observed in OSPW was above CCME (2011) guidelines (Table 4). However, in the context of passive treatment, nutrient concentrations (e.g., TP) can promote microbial activity and enhance microbially mediated wetland biogeochemical processes (e.g., biodegradation of NAs [Herman et al.,

1994; Lai et al., 1996]; dissimilatory sulfate-reduction). Thus, in the context of measuring treatment performance of CWTS, total phosphorus was not carried forward in monitoring rate and extent of removal.

5.4.1.3 Cations/ Anions

- + Predominant ions in OSPW are: bicarbonate (HCO3 ), sodium (Na ), and chloride

(Cl-). The ionic strength and balance (or imbalance) of impaired waters has implications for risks to receiving system biota (Goodfellow et al., 2000). The ionic balance, which is the ratio of the sum of cations to anions, of OSPW is near neutral, ranging from 0.85 to

1.0. Chloride concentrations in OSPW are elevated (~240 mg/L) in comparison to regional background (Athabasca River concentrations range from 2 to 50 mg Cl-/L;

142

RAMP 2001), exceed ambient water quality criteria, and were thus identified as a

constituent of concern (CCME 2011; ESRD 2014). Sodium ions (Na+) were ~360 mg/L in

OSPW (Table 2); however, the contributions of Na+ to aquatic toxicity are minor in

- comparison to the Cl anion, with C. dubia median effects thresholds (48 h LC50)

occurring at concentrations of 1770 mg Na+/L (Mount et al., 1997; Goodfellow et al.,

2000).

5.4.1.4 Inorganics

OSPW was analyzed for 16 elements (Table 4). Elements that were not measured

above the method detection limit (MDL), and therefore were not identified as COCs,

included: cadmium (MDL = 0.0002 mg/L), chromium (MDL = 0.004), cobalt (MDL =

0.0002 mg/L), and silver (MDL = 0.0002 mg/L). Based on the chemical-specific characterization approach, COCs for metals/ metalloids in OSPW included: Al, B, Cu,

Fe, Pb, Ni, Se, and Zn (Table 3). These elements measured as “total” exceed conservative concentrations in water quality criteria or toxicity endpoints for sensitive sentinel aquatic biota. Although metal speciation and concentrations in tailings ponds differ due to geologic heterogeneities and recycling of tailing pond water, concentration ranges of elements contained in OSPW are generally consistent with other OSPWs from the AOS

(MacKinnon and Boerger 1986; Siwik et al., 2000; Allen 2008a).

5.4.2 Design of Hybrid Pilot-scale CWTS for OSPW

5.4.2.1 Design Basis

The design basis was accomplished using COC fate processes and removal rate

data (e.g., Horner et al., 2013; Del Rio et al., 2005; Hwang et al., 2013; McKenzie et al.,

143

2014; Kinley et al., 2016), biogeochemical modeling (e.g., Brookins 1988; Kirk 2004), and batch reactor trials (unpublished data). For this pilot-scale study, two types of wetland reactors were chosen to achieve “bulk” oxidative or reductive conditions and are operationally defined as “oxidizing” and “reducing” wetland reactors. An advanced oxidation process followed the wetland reactors using a solar catalyst (TiO2) to target degradation of “residual” organic compounds (i.e. NAs).

5.4.2.2 Treatment Processes for Organic Constituents (e.g. petroleum hydrocarbons and

NAs)

Microbial transformation is a primary removal process in wetlands for organic constituents (e.g., NAs). Microbial transformation in wetlands is facilitated by organisms present in or on substrate (plants, detritus, hydrosoil) capable to degrading organic constituents (Bishay 1996; Knight et al., 1999; Pham et al., 2011; Horner et al., 2012;

Toor et al., 2013). NAs contained in OSPWs are putatively the primary contributors to toxicity (Verbeek et al., 1994; Headley and McMartin 2004; Marinette et al., 2015;

Morandi et al., 2015; McQueen et al., 2016a). Therefore, targeting treatment of NAs is critical for achieving narrative performance goals (e.g., “no toxics in toxic amounts”;

USEPA 2007). NA biodegradation rates are more favorable in aerobic than in anaerobic conditions (Del Rio 2006), where β-oxidation is a primary pathway by which aerobic microorganisms degrade carboxylic acids (Taylor et al., 1978; Quagraine et al., 2005;

Han et al., 2008). To achieve β-oxidation, molecular oxygen is necessary to facilitate microbial transformation of aliphatic and alicyclic carboxylic acids (Taylor and Trudgill

1978; Trudgill et al., 1984). Promoting aerobic conditions (e.g., 4-8 mg DO/L) in

144

wetlands is achieved by selecting hydrosoil with low organic matter content and high

porosity (Rodgers and Castle 2008), minimizing water depth, and selecting plants which

can translocate diatomic oxygen through rhizosphere radial oxygen loss (ROL). Cattail

~ -1 -1 (T. latifolia) provides 0.3 nmol O2 g root dry weight s ROL (Inoue and Tsuchiya

2008). Additionally, the presence of cattail aid in altering the structural composition of

NAs contained in OSPW presumably due to transformation of NAs by microbial

communities on roots or by cometabolic properties of root exudates (e.g., sugars,

enzymes, inorganic ions; Armstrong et al., 2009).

In this study, “oxidizing” wetland reactors were designed using cattail (Typha

latifolia; initial density ~30 shoots/ m2), quartz sand hydrosoil containing <1% organic

matter content, and water depths ~20 cm. These combined features in the pilot-scale

system provided bulk aerobic conditions (DO ~3-4 mg/L) in the initial oxidizing wetland

reactors and in reactors 2-5 (DO> ~4 mg/L; Table 4). Bulk hydrosoil redox indicated that aerobic conditions were maintained in the oxidizing reactors (>50 mV; supplementary

material).

For residual petroleum hydrocarbons (e.g., O/G), transfer processes in wetlands

include sorption, volatilization, precipitation, and bioconcentration in plants (Rodgers and

Castle 2008). Phase-separation and sorption can be an important process for higher

molecular weight hydrophobic chemicals (O/G). Although sorption of hydrocarbons

offers an incomplete removal mechanism, it allows additional contact time for

transformation processes (e.g., microbial degradation). In this pilot-scale design, sorption

145

was promoted by the target flow rate (<1 cm/s) coupled with macrophyte densities (T.

latifolia mean shoot density 48 shoots/ m2; n=90; supplementary material).

Solar photocatalysis was used as an advanced oxidation step following outflows

from wetland reactors. Wetland reactors provide treatment of suspended solids (and

turbidity) which can influence UV attenuation (supplementary data) and therefore

efficiency of photocatalysis. The placement of the PC reactor followed wetlands reactors

to minimize the influence of suspended particles (inflow TSS ranged from 130 to 400

mg/L). Photocatalysis of NAs can be achieved within environmentally relevant rates

(half-lives hours to days) and extents (achieving [NA] <5 mg/L), and is effective at

decreasing toxicity of commercial and OSPW-extracted NAs (Mishra et al., 2010;

Leshuk et al., 2016; McQueen et al., 2016b). Paring fixed-film catalyst (e.g., TiO2) irradiated with natural sunlight would eliminate the need for an energy source and recovery of catalyst, thus decreasing construction and operating costs. Based on reported

NA degradation rates of solar-driven photocatalysis using settled or fixed-film TiO2

(Leshuk et al., 2016; McQueen et al., 2016b), the design basis for the photocatalytic

reactor targeted cumulative solar UV (250-400 nm) of 1.0 to 3.0 MJ/m2 or insolation

(400-800 nm) of 30 to 40 MJ/m2 (photoperiod ranging from 12-36 h; Leshuk et al., 2016;

McQueen et al., 2016b; Table 4; supplementary material).

5.4.3 Treatment Processes for Inorganic Constituents (e.g., metals and metalloids, and

suspended particles)

The treatment pathway targeted for divalent metals in OSPW is precipitation with

sulfide forming relatively insoluble and non-bioavailable forms (Brookins 1988).

146

Hydrosoil conditions with redox in the range of -50 to -250 mV can facilitate

microbially-mediated dissimilatory sulfate reduction, creating reduced sulfide species

(e.g., bisulfide ions (HS-), sulfide ions [S2-]) for complexation with divalent metals that

are precipitated as sulfide minerals (e.g., CuS and ZnS; Murray-Gulde et al., 2003).

-16 These sulfide-bound species are relatively insoluble (e.g., CuS; ksp=10 ; Stumm and

Morgan 1996) and are transferred from the water column to the hydrosoil as non- bioavailable fractions (Murray-Gulde et al., 2003; Huddleston and Rodgers 2008). The initial “reducing” wetland reactors were designed with features to promote anoxic conditions and precipitation of divalent metals as sulfides. To date, there is no evidence that NAs adversely affect sulfate-reducing bacteria (and hence dissimilatory sulfate reduction); therefore, the sequence (or placement) of reducing reactors within the hybrid

CWTS can be flexible (McQueen et al., 2016a).

To maintain bulk reducing conditions, reactors were designed with greater water depths (34 cm) as compared to the oxidizing reactors. In addition, hydrosoil was amended with 2-3% organic matter (vol/vol; as wheat hay), and 0.5-1% (vol/vol) pelletized gypsum (CaSO4·2H20) was added as an additional sulfate source. Softstem

bulrush (Schoenoplectus tabernaemontani) was the vegetation selected for the “reducing”

reactors based on the ability of Schoenoplectus spp. to tolerate and maintain reducing hydrosoil conditions (Kantrud 1989; Murray-Gulde et al., 2005).

Targeted ranges of ORP, DO, and pH were maintained throughout the duration of the 10-week experiment in the “reducing” reactors (Table 4 and supplemental data).

Surficial sediment AVS concentrations in the reducing reactors ranged from 0.54 to 13.6

147

µmol/g. Wetland reactors were designed to maintain flow velocities <1cm/s to allow

suspended sediments and precipitated minerals to settle from the water column to the

hydrosoil. Shoot density in the reducing reactors reached >100 shoots/m2 (S.

tabernaemontani), further promoting settling of suspended (inorganic) solids

(Karathanasis et al., 2003).

5.4.3.1 Organic COCs

Organic COCs identified in OSPW included residual petroleum hydrocarbons (as

O/G) and NAs. Inflow water had a visible but slight hydrocarbon sheen. Following a 4.5-

day HRT in the wetland reactors, hydrocarbon sheens were no longer visible. Inflow O/G concentrations ranged from 6 to 15 mg/L and declined to non-detect (MDL = 2 mg/L)

within 1.5 to 4.5 day HRTs for all five treatment periods evaluated (Figure 2). Residual

hydrocarbons (as O/G) have been treated successfully by CWTS, with inflow

concentrations of 20 mg O/G /L declining to <1.4 mg/L following a 1 to 2 day HRT in

free-water surface wetlands planted with T. latifolia (Horner et al., 2012). Spacil et al.

(2011) achieved treatment of low molecular weight petroleum hydrocarbons in pilot-scale

-1 ~ CWTS planted with T. latifolia with removal rate coefficients of 0.8 day (T1/2 = 0.9 days). OSPWs can contain O/G concentrations as high as 92 mg/L (Allen 2008a). Oil- water separators prior to inflows into CWTS may be warranted for waters containing higher concentrations of hydrocarbons (e.g., >50 mg/L O/G; Pardue et al., 2014); however, based on the inflow O/G concentrations in OSPW (6 to 15 mg/L), concentrations declined to achieve narrative guidelines within 1.5 to 3.5 days of treatment.

148

“Total” NA concentrations (quantified by HPLC) did not significantly decrease in

outflows from wetland reactors after 13.5 days (Figure 3). Other studies using unplanted

wetland bench-scale microcosms have demonstrated an average 40% decrease in NAs

(mean inflow concentrations of 80 mg NA/L) following a 40-day HRT (Toor et al.,

2013). Additionally, Toor et al. (2013) observed complete removal of acute toxicity

(100% to 0% mortality) to rainbow trout (96 h bioassays) following a 40-day HRT in the

bench-scale wetland microcosms.

The rate and extent of biodegradation of naphthenic acids is highly dependent on

molecular weight and structure (Han et al., 2008, Holowenko et al., 2002, Scott et al.,

2005). Holowenko et al. (2002) found that NAs with a carbon number ≥22 are less

degradable than those with a carbon number <22. Han et al. (2008) demonstrated the

importance of molecular structures of NAs in predicting their biodegradability. Increased

cyclicity and alkyl branching (typical of NAs in weathered OSPW) decrease

degradability of NAs by microbial β-oxidation. Following wetland reactors,

photocatalysis provided an additional process for degrading “recalcitrant” organic

fractions (NAs).

Following photocatalysis, NA concentrations decreased from initial

concentrations of 75 to 122 mg/L to outflow concentrations of 8 to 65 mg/L (47 to 93%

removal efficiencies). Flow-through solar photocatalysis experiments indicate that NAs

in OSPW are degrading at environmentally relevant rates, with ~50% removal achieved

within ~12h of direct sunlight exposures (~1.3 to 2.3 MJ/m2 cumulative UV; Figure 3).

Mishra et al. (2010) achieved degradation of NAs extracted from OSPW using TiO2 and

149

artificial UV254, with half-lives of 1.55 and 4.8 h for deionized water and South

Saskatchewan River water, respectively. Leshuk et al. (2016) degraded acid extractable

organics (NAs) in OSPW using natural sunlight to an extent of <1 mg/L (initial [NA] of

40 mg/L) following 30 MJ/m2 (400-700 nm) insolation (2-3 days of sunlight exposure).

In this study, degradation of NAs were achieved to an extent of 10 to 16 mg/L (week 4

treatment) following an average daily cumulative radiant exposure of 1.65 MJ/m2.

5.4.3.2 Inorganic COCs

Concentrations of Al, Cu, Ni, and Zn following a 13.5 day HRT in wetland reactors significantly declined from inflow to outflow (Figure 4). With the exceptions of

B and Se, extents of removal (i.e. outflow concentrations) reached targeted performance goals for all metals and metalloids identified as COCs (Figure 4). Average removal efficiency (n=5 treatment periods) for Al and Cu ranged from 88 to 90%. Removal extents for Al and Cu ranged from 0.019 to 0.043 mg/L and 0.005 to 0.011 mg/L, respectively. OSPW inflow concentrations of Ni ranged from 0.01 to 0.015 (n=5), slightly above the numeric water quality guideline of 0.0096 mg/L (CEQGs; CCME

2011). Following treatment in wetland reactors, Ni concentrations were below the

CEQGs, with an average removal efficiency of 49 and 44% for wetland Series A and B, respectively. Inflow concentrations of Zn ranged from non-detect (MDL = 0.002 mg/L) to 0.064 mg/L. Outflow concentrations of Zn were below the target removal extent of

0.03 mg/L (based on CEQGs), with average removal efficacies of 65 and 58% (n=5) for

Series A and B, respectively.

150

Inflow Se concentrations ranged from non-detect (MDL = 0.0002 mg/L) to

0.0066 mg/L, above regional background (Athabasca River; range <0.001 to 0.0014;

RAMP 2001; Athabasca River upstream of Muskeg River) and guidelines of 0.001 mg/L

(CEQGs) and 0.0015 mg/L (USEPA WQC; USEPA 2007; CCME 2011). Following 13.5

day HRT in wetland reactors, Se concentrations declined, with outflow concentrations

ranging from 0.0004 to 0.0045 mg/L. Average Se removal efficiencies for Series A and B

were 42 and 85%, respectively. However, outflow concentrations of Se following

wetland reactors were above the CEQGs target of 0.001 mg/L (Figure 4). Development

of site- or regionally-specific Se criteria are forthcoming, which may incorporate

exposure modifying factors (e.g. sulfate) to the interpretation and use of current

guidelines. Exposure modifying factors are crucial to understanding bioavailability and

ecological risk of Se (Chapman et al., 2009). Nonetheless, Se removal efficiencies can be

enhanced in CWTS by including longer retention times in reducing biogeochemical

conditions (Spacil et al., 2011). Se removal in CWTS is primarily achieved by microbial

-2 reduction from oxidized forms of Se (SeO3 ) to elemental selenium (Spacil et al. 2011).

Spacil et al. (2011) reported Se removal efficiencies measured in a pilot-scale system

-1 from 86-99%, with removal rate coefficients (k) ranging from 0.16 to 0.99 day by

promoting reducing conditions (e.g. < 2 mg DO/L and ORP < -50mV). In this study,

boron concentrations did not decline from inflow to outflow, with outflow concentrations

ranging from 1.8 to 2.7 mg B/L, greater than the target numeric criterion of 1.5 mg/L

(CCME 2011). However, boron concentrations in OSPW indicate concentrations are

below toxicity thresholds to sentinel aquatic species (T. latifolia 7-day NOEC [seedling

151

root growth] = 10.4 mg B/L; C. dubia 7-day NOEC [reproduction] = 16.4 mg B/L

[Damiri 2009]; rainbow trout 28-day LC50 = 79 mg/L [Birge and Black 1977]).

Al and Zn had the greatest removal rate coefficients of the elemental COCs, with

-1 ~ mean rate coefficients of 0.35 day or (T1/2 = 2 days; n =5; Series A and B) and 0.37

~ (T1/2 = 2 days; n =5; Series A), respectively. Removal rates were slower for Cu and Ni,

-1 ~ with mean rate coefficients ranging from 0.20 day (Cu, T1/2 = 3.5 days Series A and B)

~ -1 to 0.07 (Ni; T1/2 = 10 days; Series A and B). Se mean rate coefficient was 0.17 day

(n=5 treatment periods) in wetland Series A; however, Series B concentrations increased

slightly (e.g. 0.001 to 0.004 mg/L) from 7.5 to 13.5 days, therefore rates were not

estimated.

Rates of removal observed in this study parallel rates of removal measured in sequential reducing and oxidizing wetlands reactors for Al, Cu, Ni and Zn contained in

other energy-derived waters (Johnson et al., 2008; Murray-Gulde et al., 2008; Eggert et al., 2008; Spacil et al., 2011; Horner et al., 2012). In CWTS studies where sulfide or iron- manganese co-precipitation is the targeted biogeochemical removal process, aqueous removal of Al, Cu, Ni, and Zn (to the extent of achieving target water quality guidelines) in CWTS is achieved on the order of days (i.e. k = 0.1 to 1.0 day-1; Johnson et al., 2008;

Murray-Gulde et al., 2008; Eggert et al., 2008; Spacil et al., 2011; Horner et al., 2012).

Removal rates and extents are essential for scaling similar system designs to other

locations (Huddleston and Rodgers 2008). In this study, pilot-scale wetland reactor

experiments were conducted in a covered greenhouse, with no influence of dilution by

152

rainfall; therefore, rates and extents of removal reported for all constituents are conservative.

5.4.3.3 Toxicity Experiments

Survival and reproduction of C. dubia in 7-8 day exposures were used to evaluate the ability of the pilot-scale CWTS to mitigate risks associated with OSPW. C. dubia mortality ranged from 18 to 80% (n=5 treatment periods) in untreated (inflow) OSPW.

Reproduction of C. dubia was impaired in all but one (treatment period 3) of the inflows tested as compared to a laboratory control. Following 13.5 day HRT in the wetland reactors, C. dubia toxicity was eliminated (Figure 5). For treatment periods 1 and 2, reproduction increased from inflow to outflow of wetland reactors with a mean of 27 and

30 neonates/live adult (no statistical difference from laboratory control; p= 0.56; α =

0.05). Toor et al. (2013) achieved decreases in rainbow trout mortalities to OSPW following treatment in unplanted simulated wetland hydrosoils (wetland sediment and aerated microcosms with input of OSPW collected from Mildred Lake settling basin).

Presumably, toxicity in OSPW is influenced by “classical” NAs (Verbeek et al., 1994;

Morandi et al., 2015). In this study, total NAs concentrations (as quantified by HPLC) did not appreciably decline in the wetland reactors following 13.5 day HRT; however, toxicity was eliminated (Figure 5). These observations support that adverse effects of

OSPW (and specifically the organic acid fraction) are influenced by composition of NAs and not concentration alone (Brown and Ulrich 2015, Mahaffey and Dubé 2016). Results from this study also support the hypothesis that wetlands can be effective for removing

153

acute OSPW toxicity (Armstrong et al., 2009; Toor et al., 2013) within environmentally relevant timeframes (e.g., hydraulic retention times of ~2 weeks).

5.5 Conclusions

Results from this hybrid pilot-scale study provided evidence that problematic constituents contained in OSPW could be treated to extents necessary to remove toxicity to a sentinel organism (C. dubia). Based on characterization of an OSPW from an active settling basin (OSPW), COCs needing treatment were identified as: NAs, petroleum hydrocarbons (as O/G), Al, B, Cu, Ni, Se, Zn, and TSS. The hybrid pilot-scale CWTS used passive (i.e. low energy) biogeochemical and advance oxidation processes to transfer or transform COCs to achieve numeric and narrative treatment goals. Results from this study provide proof-of-concept data to inform hybrid passive or semi-passive treatment approaches (i.e. constructed wetlands) that could be used to mitigate COCs contained in OSPWs.

5.6 Acknowledgements

Funding support for this research was provided by Shell Canada Ltd. and Suncor Energy.

The authors are also grateful to Dr. Wayne Chao of Clemson University and Dr. John

Headley and Kerry Peru of Environment and Climate Change Canada for providing analytical support.

154

5.7 References

Alberta Environment & Sustainable Resource Development (ESRD). (2014). Environmental Quality Guidelines for Alberta Surface Waters. Water Policy Branch, Policy Division. Edmonton, AB, CA 48 pp.

Allen, E.W., (2008a). Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Journal of Environmental Engineering Science 7:123-138.

Allen, E.W., (2008b). Process water treatment in Canada’s oil sands industry: II. A review of emerging technologies. Journal Environmental Engineering and Science 7:499-524.

American Public Health Association (APHA). (2012). Standard methods for the examination of water and wastewater 21st Ed. American Public Health Association, Port City Press: Baltimore, MD, pp. 1368.

Anderson, J.C., Wiseman, S.B., Wang,N., Moustafa, A., Perez-Estrada, L., Gamal El- Din, M., Martin, J.B., Liber, K., Giesy, J. P. (2011). Effectiveness of ozonation treatment in eliminating toxicity of oil sands process-affected water to Chironomus dilutus. Environmental science & technology 46(1): 486-493.

Armstrong, S.A., Headley, J.V., Peru, K.M., Germida, J.J. (2009). Differences in Phytotoxicity and Dissipation between ionized and nonionized oil sands naphthenic acids in wetland plants. Environmental Toxicology and Chemistry 28: 2167-2174.

Biesinger, K.E., Christensen, G.M., Fiandt, J.T., (1986). Effects of metal salt mixtures on Daphnia magna reproduction. Ecotoxicology and environmental safety 11(1): 9- 14.

Birge, W. J., & Black, J. A. (1977). Sensitivity of vertebrate embryos to boron compounds: US Environmental Protection Agency. EPA-560/1-75-008.

155

Birge, W. J. (1977). Aquatic toxicology of trace elements of coal and fly ash. In Energy and Environmental Stress in Aquatic Systems, Selected Papers from a Symposium, held at Augusta, Georgia November 2-4, 1977. CONF-771114. pp. 219-240

Bishay, F. S. (1998). The use of constructed wetlands to treat oil sands wastewater. A thesis presented to the University of Alberta, Edmonton, Alberta.

Boucher, A.M., & Watzin, M.C., (1999). Toxicity identification evaluation of metal‐contaminated sediments using an artificial pore water containing dissolved organic carbons. Environmental Toxicology and Chemistry 18(3): 509-518.

Brookins, D.G., (1988). Eh, pH diagram for geochemistry. Berlin, Springer. 1-14.

Brown, L.D., Ulrich, A.C., (2015). Oil sands naphthenic acids: A review of properties, measurement, and treatment. Chemosphere 127: 276-290.

Carpenter, S.R., Kitchell, J.F., Hodgson, J.R., (1985). Cascading trophic interactions and lake productivity. BioScience 35(10): 634-639.

Chapman, P.M,. Adams, W.J., Brooks, M.L., Delos, C.G., Luoma, S.N., Maher, W.A., Ohlendorf, H.M., Presser, T.S., Shaw, D.P. (2009). Ecological assessment of selenium in the aquatic environment: Summary of a SETAC Pellston Workshop. Pensacola FL (USA): Society of Environmental Toxicology and Chemistry (SETAC).

Cumulative Environmental Management Association (CEMA). (2014). Guidelines for wetlands establishments on reclaimed oil sands leases. 3rd Ed. Alberta. Fort McMurray, AB.

Damiri, B., (2007). "Risk characterization for boron and aquatic plants and animals". All Theses. Paper 235.http://tigerprints.clemson.edu/all_theses/235

Del Rio, L.F., Hadwin, A.K.M., Pinto, L.J., MacKinnon, M.D., Moore M.M., (2006). Degradation of naphthenic acids by sediment micro-organisms. Journal of Applied Microbiology 101: 1049-1061.

156

Di Toro, D.M, Mahony, J.D., Hansen, D.J., Scott, K.J., Carlson, A.R., Ankley, G.T., (1992). Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments. Environmental Science and Technology 26: 96-101.

Eggert, D.A., Rodgers, Jr. J.H., Huddleston, G.M., Hensman, C.E., (2008). Performance of pilot-scale constructed wetland treatment systems for flue gas desulfurization waters. Environmental Geosciences 15:115-129.

Environment Canada (2007). Biological Test Method: Test of Reproduction and Survival using the Cladoceran Ceriodaphnia dubia. EPS 1/RM/21(2nd Ed.). Science and Technology Branch.

Faulkner, S.P., Patrick, W.H., Jr, Gambrell, R.P., (1989). Field techniques for measuring wetland soil parameters. Soil Science Society of America 53: 883-890.

Foote, L. (2012). Threshold considerations and wetland reclamation in Alberta's mineable oil sands. Ecology and Society 17(1): 35.

Frank, R.A., Kavanagh, R., Burnison, B.K., Arsenault, G., Headley, J.V., Peru, K.M., Van der Kraak,, G., Solomon, K.R., (2008). Toxicity Assessment of Collected Fractions from an Extracted Naphthenic Acid Mixture. Chemosphere 72:1309- 1314

Freeman, R.A., & Everhart, W.H., (1971). Toxicity of aluminum hydroxide complexes in neutral and basic media to rainbow trout. Transactions of the American Fisheries Society 100(4): 644-658.

Gillespie, W.B.J., Hawkins, W.B., Rodgers, Jr. J.H., Cano, M.L., Dorn, P.B., (2000). Transfers and transformations of zinc in constructed wetlands: Mitigation of a refinery effluent. Ecological Engineering 14:279-292.

Haakensen, M., Pittet, V., Spacil, M.M., Castle, J.W., & Rodgers Jr, J.H., (2015). Key aspects for successful design and implementation of passive water treatment

157

systems. Journal of Environmental Solutions for Oil, Gas, and Mining, 1(1), 59- 81.

Halverson, N.V., (2004). Review of constructed subsurface flow vs. surface flow wetlands. Westinghouse Savannah River Company, Aiken, SC. Prepared for the US Dept. of Energy. WSRC-TR-2004-00509.

Han, X., MacKinnon, M.D., Martin, J.W., (2009). Estimating the in situ biodegradation of naphthenic acids in oil sands process waters. Chemosphere 76:63-70.

Headley, J.V., McMartin, D.W., (2004). A Review of the Occurrence and Fate of Naphthenic Acids in Aquatic Environments. Journal of Environmental Science and Health A39: 1989-2010.

Herman, D.C., Fedorak, P.M., McKinnon, M.D., Costerton, J.W., (1994). Biodegradation of naphthenic acids by microbial populations indigenous to oil sands tailings. Canadian Journal of Microbiology 40:467-477

Holowenko, F.M., MacKinnon, M.D., Fedorak, P.M., (2002). Characterization of naphthenic acids in oil sands wastewaters by gas chromatography-mass spectrometry. Water Resources 36: 2843-2855.

Hoke, R.A., Gala, W.R., Drake, J.B., Giesy, J.P., Flegler, S., (1992). Bicarbonate as a potential confounding factor in cladoceran toxicity assessments of pore water from contaminated sediments. Canadian Journal of Fisheries and Aquatic Science 49: 1633-1640.

Horner, J.E., Castle, J.W., Rodgers, Jr. J.H., Murray-Gulde, C.L., Myers, J.E., (2012). Design and performance of pilot-scale constructed wetland treatment systems for treating oilfield produced water from Sub-Saharan Africa. Water, Air, & Soil Pollution 223:1945-1957.

Huddleston, III G.M., Rodgers, Jr. J.H., (2008). Design of a constructed wetland system for treatment of copper-contaminated wastewater. Environmental Geosciences 15:9-19.

Hwang, G., Dong, T., Islam, M.S., Sheng, Z., Pérez-Estrada, L.A., Liu, Y., El-Din, M.G., (2013). The impacts of ozonation on oil sands process-affected water

158

biodegradability and biofilm formation characteristics in bioreactors. Bioresource technology 130: 269-277.

Inoue, T.M., & Tsuchiya, T., (2008). Interspecific differences in radial oxygen loss from the roots of three Typha species. Limnology 9(3): 207-211.

Kantrud, H.A., Krapu, G.L., Swanson, G.A., (1989). Prairie basin wetlands of the Dakotas: A community profile. US Fish and Wildlife Service Biological Report 85:7- 29.

Karathanasis, A.D., Johnson, C.M., (2003). Metal removal potential by three aquatic plants in an acid mine drainage wetland. Mine Water and the Environment 22(1): 22-30.

Kavanagh, R.J., Frank, R.A., Oakes, K.D., Servos, M.R., Young, R. F., Fedorak, P.M., MacKinnon, M.D., Solomon, K.R., Dixon, D.G., Van der Kraak, G., (2011). Fathead minnow (Pimephales promelas) reproduction is impaired in aged oil sands process-affected waters. Aquatic toxicology 101(1): 214-220.

Kinley, C.M., Gaspari, D.P., McQueen, A.D., Rodgers, J.H., Castle, J.W., Friesen, V., Haakensen, M., (2016). Effects of environmental conditions on aerobic degradation of a commercial naphthenic acid. Chemosphere 161: 491-500.

Kirk, G., (2004). Biogeochemistry of Submerged Soils. New York: John Wiley and Sons.

Knight, R.L., Kadlec, R.H., Ohlendorf, H.M., (1999). The use of treatment wetlands for petroleum industry effluents. Environmental Science and Technology 33: 973- 980.

Lai, J.W., Pinto, L.J., Bendell‐Young, L.I., Moore, M.M., Kiehlmann, E., (1996). Factors that affect the degradation of naphthenic acids in oil sands wastewater by

159

indigenous microbial communities. Environmental Toxicology and Chemistry 15(9): 1482-1491.

Leonard, E.N., Ankley, G.T., Hoke, R.A., (1996). Evaluation of metals in marine and freshwater surficial sediments from the environmental monitoring and assessment program relative to proposed sediment quality criteria for metals. Environmental Toxicology and Chemistry 15: 2221-2232.

Leshuk, T., Wong, T., Linley, S., Peru, K.M., Headley, J.V., Gu, F., (2016). Solar photocatalytic degradation of naphthenic acids in oil sands process-affected water. Chemosphere 144: 1854-1861.

Li, S., Lissner, J., Mendelssohn, I.A., Brix, H., Lorenzen, B., McKee, K.L., Miao, S. (2009). Nutrient and growth responses of cattail (Typha domingensis) to redox intensity and phosphate availability. Annals of botany pp. 1-10

Linsebigler, A.L., Guangquan, L., Yates, J.T., (1995). Photocatalysis on TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chemical Reviews 95: pp. 735.

MacKinnon, M.D., Boerger, H., (1986). Description of two treatment methods for detoxifying oil sands tailings pond water. Water Pollution Research Journal of Canada 21:496–512.

Mahaffey, A., Dube, M., (2016). Review of the composition and toxicity of oil sands process-affected water. Environmental Review. doi: 10.1139/er-2015-0060

Marentette, J.R., Frank, R.A., Bartlett, A.J., Gillis, P.L., Hewitt, L.M., Peru, K.M., Parrott, J.L., (2015a). Toxicity of naphthenic acid fraction components extracted from fresh and aged oil sands process-affected waters, and commercial naphthenic acid mixtures, to fathead minnow (Pimephales promelas) embryos. Aquatic Toxicology 164: 108-117.

Marentette, J.R., Frank, R.A., Hewitt, L.M., Gillis, P.L., Bartlett, A.J., Brunswick, P., Shang, D., Parrott, J.L., (2015b). Sensitivity of walleye (Sander vitreus) and fathead minnow (Pimephales promelas) early-life stages to naphthenic acid fraction components extracted from fresh oil sands process-affected waters. Environmental Pollution 207: 59-67.

160

McKenzie, N., Yue, S., Liu, X., Ramsay, B.A., Ramsay, J.A., (2014). Biodegradation of naphthenic acids in oils sands process waters in an immobilized soil/sediment bioreactor. Chemosphere 109: 164-172.

McQueen, A.D., Kinley, C.M., Hendriskse, M., Gaspari, D.P., Calomeni, A.J., Iwinski, K.J., Castle, J.W., Haakensen, M.C., Peru, K.M., Headley, J.V., Rodgers Jr, J.H., (2016a [in review]). A Risk-Based Approach for Identifying Constituents of Concern in Oil Sands Process-affected Water from the Athabasca Oil Sands Region. Chemosphere (In Review)

McQueen, A.D., Kinley, C.M., Kiekhaefer, R.L., Calomeni, A.J., Rodgers Jr, J.H., Castle, J.W., (2016b). Photocatalysis of a Commercial Naphthenic Acid in Water Using Fixed-Film TiO2. Water, Air, & Soil Pollution 227(5): 1-11.

Mikula, R., (2013). Trading water for oil: Tailings management and water use in surface- mined oil sands. In Hein FJ, Leckie D, Larter S, and Suter JR (eds.). Heavy-oil and oil-sand petroleum systems in Alberta and beyond: AAPG Studies in Geology. 64:689-699.

Mishra, S., Meda, V., Dalai, A., McMartin, D., Headley, J., Peru, K., (2010). Photocatalysis of Naphthenic Acids in Water. Journal of Water Resource and Protection 2: 644-650.

Mooney, F.D., Murray-Gulde, C.M., (2008). Constructed treatment wetlands for flue gas desulfurization waters: Full-scale design, construction issues, and performance. Environmental Geosciences 15: 131-141.

Mount, D.I., (1968). Chronic toxicity of copper to fathead minnows (Pimephales promelas, Rafinseque). Water Research 2: 215-223.

Murray-Gulde, C., Heatley, J.E., Karanfil, T., Rodgers, Jr. J.H., Myers, J.E., (2003). Performance of a hybrid reverse osmosis-constructed wetland treatment system for brackish oil field produced water. Water Resources 37: 705-713.

161

Murray-Gulde, C.L., Bridges, W.C., Rodgers, Jr. J.H., (2008). Evaluating performance of a constructed wetland treatment system designed to decrease bioavailable copper in a waste stream. Environmental Geosciences 15: 21-38.

Oil sands regional aquatics monitoring program (RAMP) (2001) In. Golder Associates Ltd., (2002). Vol. 1: Chemical and biological monitoring. Report prepared for the RAMP Steering Committee, Fort McMurray, AB.

Pardue, M.J., Castle, J.W., Rodgers, J.H., Huddleston, G.M., (2014). Treatment of oil and grease in produced water by a pilot-scale constructed wetland system using biogeochemical processes. Chemosphere 103: 67-73.

Pennak, R.W., (1978). Fresh-water Invertebrates of the United States, 2nd ed. John Wiley & Sons, New York, NY

Pham, M.P.T., Castle, J.W., & Rodgers Jr, J.H., (2011). Biogeochemical process approach to the design and construction of a pilot-scale wetland treatment system for an oil field-produced water. Environmental Geosciences 18(3): 157-168.

Pickering, Q.H., Henderson, C., (1966). The acute toxicity of some heavy metals to different species of warm water fishes. International Journal of Air and Water Pollution 10: 453-463.

Province of Alberta. (2014). Environmental Protection and Enhancement Act. Revised Statues of Alberta 2000. Chapter E-12. Division 1, Section 2.

Rodgers, Jr. J.H., Castle, J.W., (2008). Constructed wetland systems for efficient and effective treatment of contaminated waters for reuse. Environmental Geosciences 15: 1-8.

SAS Institute (2010). Statistical Analysis System. JMP Pro version 11, Cary, NC: SAS Institute.

162

Siwik, P.L., Van Meer, T., MacKinnon, M.D., & Paszkowski, C.A., (2000). Growth of fathead minnows in oil sand‐processed wastewater in laboratory and field. Environmental Toxicology and Chemistry 19(7): 1837-1845.

Swigert, J.P., Lee, C., Wong, D.C., White, R., Scarlett, A.G., West, C.E., Rowland, S J. (2015). Aquatic hazard assessment of a commercial sample of naphthenic acids. Chemosphere 124: 1-9.

Taylor, D.G., Trudgill, P.W., (1978). Metabolism of cyclohexane carboxylic acid by Alcaligenes strain W1. Journal of bacteriology 134(2): 401-411.

Trudgill, P.W., (1984). Microbial Degradation of the Alicyclic Ring. In: Gibson, D.T. (Ed.), Microbial Degradation of Organic Compounds. Marcel Dekker, New York, pp. 131-180.

163

United States Environmental Protection Agency (USEPA). (1991). Technical support document for water quality-based toxics control. EPA/505/2-90-001. Washington, D.C.

United States Environmental Protection Agency (USEPA). (1999). Revision A: n-hexane extractable material (HEM; oil and grease) and silica gel treated n-hexane extractable material (SGTHEM; non-polar material) by extraction and gravimetry, Method 1664. Washington, D.C.

United States Environmental Protection Agency (USEPA). (2007). National Recommended Water Quality Criteria., Washington, D.C.

Whitby, C., (2010). Microbial naphthenic acid degradation. Advances in Applied Microbiology 70: 93-125.

Yen, T.W., Marsh, W.P., McKinnon, M.D., Feorak, P.M., (2004). Measuring naphthenic acids concentrations in aqueous environmental samples by liquid chromatography. Journal of Chromatography 1033: 83-90.

Quagraine, E.K., Headley, J.V., Peterson, H.G., (2005). Is biodegradation of bitumen a source of recalcitrant naphthenic acid mixtures in oil sands tailing pond waters? Journal of Environmental Science and Health 40(3): 671-684.

164

165

Table 5.1 Hybrid pilot-scale CWTS design features.

Vol HRT Flow-rate Reactors Reactor Reactor Type . (time/reacto (mL/min) per Dimensions Hydrosoilc,d Vegetation (L)a r) b series (cm) 30 cm medium to coarse Schoenoplectus Reducing Wetland 26 radius, 73 grained quartz sand; 2% v/v tabernaemontani (softstem 74 1.5d 34 1 Reactors height organic mattere, 1% v/v bulrush); Initial density 2 gypsum (CaSO4·2H20) 20-30 shoots/m T ypha latifolia (broadleaf Oxidizing Wetland 61 height, 123 30 cm medium to coarse 148 3d 34 4 cattail); Initial density 20- Reactors length, 64 width grained quartz sand 30 shoots/m2 Fixed -film TiO2 15 height, 182 Solar Photocatalysis 7 12h(f) 10 1 - - length, 60 width Reactors avolume by direct measure bflow-rate maintained by metered piston pump (Fluid Metering Inc. [FMI®]) cslow release fertilizer (Osmocote®) added to all reactors on initiation dhydrosoil bulk porosity = 0.25 eorganic matter = wheat hay fHRT based on hours of direct sunlight

166

Table 5.2 Comparison of water quality characteristics and organic constituents in OSPW to water quality guidelines (USEPA 2007; CCME 2011, ESRD 2014) and toxicity values for C. dubia (Cd), rainbow trout (O. mykiss [Om]), and fathead minnow (P. promelas [Pp]). COCs (i.e. concentration > guideline) are bolded.

Water Quality Guidelines (mg/L) Parameter mg/L OSPW (n=10) Reported Toxicity COC Alberta WQG CEQG USEPA Values (mg/L) (unless noted) Min Max Chronic Chronic Acute Chronic Yes/No pH (SU) 7.91 8.4 6.5-9.0 No

a Alkalinity (CaCO3) 320 340 20 No

Total Ammonia (N) 0.051 0.099 0.17-1.5b 2.9 0.26 LC50: 0.563 (96h Omc) No

Total Phosphorus (TP) 0.037 0.082 0.01-0.02 Yes

Total Suspended Solids 130 400 * * * Yes (TSS)

Total Naphthenic Acids (NAs)

NAs (HPLC) 80 128 ** LC50: 51.8 (96h Ppd) Yes NAs (Orbitrap MS) 93 103 ** EC50: 7.5 (96h Ppe)

Oil and grease (O/G) 8 13 * * Yes

- f Bicarbonate (HCO3 ) 300 320 LC50: 995 (48h Cd ) No

Calcium (Ca+) 29 31 LC50: 4630 (96h Ppg) No

Chloride (Cl-) 240 245 120 120 860 230 LC50: 7341 (96h Pph) Yes

Sodium (Na+) 360 364 LC50: 1770 (48h Cdg) No

-2 i Sulfate (SO4 ) 150 175 LC50: 2,078 (96h Cd ) No aMinimum fHoke et al. 1992 bTemperature range from 5-20°C and pH of 8.0-8.5 gMount et al. 1997 cThurston and Russon 1981; @ 14.4°C and pH of 8.3 hAdelman et al. 1976 dKavanagh et al. 2011; 5 d larvae; NAs quantified by ESI-MS iSoucek and Kennedy 2005 eMarentette et al. 2015; "fresh" OSPW; embryos; NA quantified by LC/QToF *Narrative statement; no visible sheen, or unreasonable turbidity (no more than >25mg/L TSS than background in a 24-hr period)

**”No toxics in toxic amounts” Note: NAs quantified by different methods do not necessarily measure the same suite of compounds (Headley et al., 2015)

167

Table 5.3 Comparison of metals and metalloids in OSPW to water quality guidelines (USEPA 2007; CCME 2011; ESRD 2014) and toxicity values for C. dubia (Cd), D. magna (Dm), fathead minnow (P. promelas [Pp]), and rainbow trout (O. mykiss [Om]). Identified COCs (i.e. concentration > guideline) are bolded.

Water Quality Guidelines (mg/L) Parameter mg/L OSPW (n=10) Reported Toxicity Values COC Alberta WQG CEQG USEPA (mg/L) (unless noted) Min Max Chronic Chronic Acute Chronic Yes/No Aluminum (Al) 0.36 10.9 0.1a 0.1a 0.75 0.087 IC42: 0.514 (45d Omd) Yes

Arsenic (As) 0.0006 0.0032 0.005 0.005 0.34 0.15 No

Barium (Ba) 0.21 0.33 EC16: 5.8 (21d Dme) No

Boron (B) 2.09 2.12 1.5 1.5 LC50: 53.2 (21d Dmf) Yes

Cadmium (Cd) <0.0002 <0.0002 0.00016b 0.00038b 0.002 0.0025 LC50: 0.1 (10d Cdg) No

Chromium (Cr) <0.004 <0.004 0.0089 0.0089 0.57 0.074 No

Cobalt (Co) <0.0002 <0.0002 LOEC: 0.009 (28d Dmh) No

Copper (Cu) 0.003 0.12 0.007 0.0024b 0.016b 0.011b LC50: 0.095 (96h Ppi) Yes

Iron (Fe) 1.2 1.5 0.3c 0.3c 1.0 LC50: 3.1 (96h Ppj) Yes

Lead (Pb) <0.0002 0.0014 0.0032b 0.00318b 0.065 0.0025 No

Manganese (Mn) 0.137 0.162 LC50: 9.1 (48h Cdk) No

Molybdenum (Mo) 0.041 0.062 0.073 0.073 LC50: 0.73 (28d Oml) No

Nickel (Ni) 0.0056 0.01 0.052 0.0096 LC50: 0.8 (48h Cdm) Yes

Selenium (Se) 0.002 0.004 0.001 0.001 0.0015 Yes

Silver (Ag) <0.0002 <0.0002 0.001 0.00025 No

Zinc (Zn) 0.014 0.113 0.03 0.03 0.12 0.12 LC50: 2.55 (96h Ppj) Yes apH: ≥6.5 hKimball 1978

bHardness: 100 mg/L as CaCO4 iMount, 1968

cDissolved fraction jPickering and Henderson 1966

dFreeman and Everhart 1971; biomass kBoucher and Watzin 1999

eBiesinger and Christensen 1972 lBirge 1977

fOPP, 2000 mKeithly et al. 2004

gSuedel et al. 1997

168

Table 5.4 Targeted treatment processes, operational metrics, and measured ranges in hybrid

CWTS designed for OSPW.

Measured Target Treatment Process Operational Metric Target Range Range(a) Sorption/ Settling shoot density and root mass(b,c) >100 shoots m2 (bulrush) 165 shoots/m2 >30 shoots m2 (cattail) 48 shoots/m2

HRT low flow rate (<10 cm/s)(b) <1 cm/s

Reduction ORP ORP -150 to -250 mV(d) ORP <-150 mV excess molar ratio of AVS AVS:SEM, 1.5 to acid volatile sulfides (AVS) to SEM (AVS:SEM>1)(e) 4.6

Oxidation ORP ORP >-50 mV ORP > -28 mV plant density (estimation of >30 shoots·m2 48 shoots/m2 rhizome radial oxygen loss) aqueous DO concentration >2.0 mg/L 2.65-6.8 mg/L

Aerobic biodegradation ORP ORP >-50 mV ORP > -28 mV aqueous DO concentration >2.0 mg/L 2.15-6.32 mg/L (f) (g) SOD5-day >100 mg/L O2 333-546 mg/L shoot density (indicator of >30 shoots/m2 48 shoots/m2 rhizome radial oxygen loss) Advanced oxidation daily cumulative UV (250- >1.0 MJ/m2(h) 0.86 to 1.5 MJ/m2 (photocatalysis) 400nm) ORP = oxidation reduction potential (mV); surficial sediment (2-6 cm) dBrookins 1988 HRT = hydraulic retention time eDi Toro et al., 1992 f SOD5-day = 5-day sediment oxygen demand Indicator of biological activity in sediment-water interface aConditions measured during experiment duration (mean [n=5] or range) gTarget SOD based on glucose+glutamic standard with no added nutrients bRodgers and Castle 2008 hbased on batch reactor fixed-film photocatalysis trials cgreen shoots with a healthy and vibrant core using OSPW (unpublished) Note: shoot density can be an indicator of root mass (typically 1:1 biomass ratio; Li et al., 2010)

169

Table 5.5 Inflow concentrations, outflow concentrations, removal efficiencies, and removal rate coefficients of COCs for each CWTS replicate series.

[Inflow] Removal extent (mg/L) Removal efficiency (%) Rate coefficient (day-1), (r2) Treatment Period mg/L Series A Series B Series A Series B Series A Series B Aluminum (Al) Period 1 0.879 0.023 0.021 97 98 0.33 (0.92) 0.37 (0.78) Period 2 0.102 0.031 0.036 70 65 0.17 (0.94) 0.15 (0.87) Period 3 0.363 0.022 0.026 94 93 0.25 (0.85) 0.25 (0.78) Period 4 0.524 0.019 0.022 96 96 0.39 (0.65) 0.22 (0.49) Period 5 0.702 0.037 0.043 95 94 0.41 (0.77) 0.26 (0.75) Average (n=5) 90 89 0.31 0.25 Boron (B) Period 1 2.05 2.02 1.80 1 12 - - Period 2 2.04 2.20 1.99 -8 3 - - Period 3 2.10 2.16 2.00 -3 5 - - Period 4 2.12 2.54 2.19 -20 -3 - - Period 5 2.16 2.77 2.60 -29 -21 - - Average (n=5) -12 -1 - - Copper (Cu) Period 1 0.13 0.007 0.008 94 94 0.29 (0.98) 0.27 (0.94) Period 2 0.058 0.007 0.007 88 88 0.22 (0.99) 0.19 (0.97) Period 3 0.062 0.006 0.005 90 92 0.20 (0.87) 0.24 (0.92) Period 4 0.123 0.011 0.005 91 93 0.16 (0.64) 0.17 (0.68) Period 5 0.043 0.006 0.007 79 79 0.12 (0.85) 0.15 (0.85) Average (n=5) 88 89 0.20 0.21 Nickel (Ni) Period 1 0.014 0.0056 0.008 60 43 0.08 (0.71) 0.08 (0.90) Period 2 0.010 0.0052 0.0048 48 52 0.06 (0.72) 0.03 (0.97) Period 3 0.015 0.0092 0.0094 35 36 0.06 (0.78) 0.07 (0.89) Period 4 0.015 0.0060 0.0068 60 55 0.22 (0.69) 0.06 (0.55) Period 5 0.011 0.0062 0.0070 44 36 0.06 (0.76) 0.06 (0.86) Average (n=5) 49 44 0.09 0.06 Selenium (Se ) Period 1 0.0066 0.0045 0.0004 32 94 0.36 (0.88) 0.08 (0.74) Period 2 0.005 0.0032 0.0015 36 70 - - Period 3 0.0058 0.0025 0.0006 57 90 0.32 (0.93) - Period 4 BDLa BDL BDL - - - - Period 5 BDL BDL BDL - - - - Average (n=3) 42 85 0.34 (n=2) - Zinc (Zn) Period 1 0.051 0.002 0.002 96 96 0.64 (0.68) 0.41 (0.79) Period 2 BDLb BDL BDL - - - - Period 3 BDL BDL BDL - - - - Period 4 0.064 0.018 0.027 72 58 0.22 (0.70) 0.22 (0.98) Period 5 0.042 0.02 0.021 28 20 0.07 (0.67) 0.17 (0.71) Average (n=3) 65 58 0.31 0.27 BDL = below method detection limit Treatment periods = 13.5 day HRT aSelenium analytical method detection limit = 0.0002 mg/L bZinc analytical method detection limit = 0.002 mg/L

170

Fig 5.1 A schematic diagram of the pilot-scale experiment.

171

Series A Series B 20

Period 1

15 Period 2

Period 3

10 Period 4 Period 5 O /Gm , g/L 5

0 1.5 4.5 7.5 10.5 13.5

Inflow Hydraulic retention time, days

Fig 5.2 O/G concentrations in inflow and reactor outflows for Series A (solid line) and

Series B (dashed line).

172

Wetland Reactors Photocatalysis

Wetland Reactors

Series A Series B 150 Period 1 Period 4

125 Period 2 Period 5

Period 3 100 Photocatalysis 75 W eek 1 W eek 4

N Am , g/L 50 W eek 2 W eek 5 W eek 3 25

0 1.5 4.5 7.5 10.5 13.5

Inflow

PC Inflow PC Outflow Hydraulic retention time, days

Fig 5.3 Total NA concentrations in wetland inflow and outflows for Series A (solid lines) and Series B (dashed lines), and inflow and outflow from photocatalysis treatments.

Note: Photocatalysis (PC) represents inflow and outflow from thin-film TiO2 reactors conducted outdoors during the months of

May, June, and July. Photocatalysis nominal HRT = 12h. Average daily cumulative UV (200-400 nm) ranged from 0.86 to 1.5

MJ/m2 (supplementary material). PC treatments are presented as mean (n=3) NA concentration with error bars representing standard deviation.

173

1.00 Series A Series B

0.75 Al Period 1 Period 4

0.50 Period 2 Period 5 Al, m g/L

0.25 Period 3 WQC Target

0.00 1.5 4.5 7.5 10.5 13.5

Inflow Hydraulic Retention Time, days

4.0 0.15

3.5 B Cu 3.0 0.10

2.5 B ,m g/L 2.0 mg/L Cu, 0.05

1.5

1.0 0.00 1.5 4.5 7.5 10.5 13.5 1.5 4.5 7.5 10.5 13.5

Inflow Inflow Hydraulic retention time, days Hydraulic retention time, days

0.020 0.008

0.015 Ni 0.006 Se

0.010 0.004 Ni, m g/L S e, m g/L

0.005 0.002

0.000 0.000 1.5 4.5 7.5 10.5 13.5 1.5 4.5 7.5 10.5 13.5

Inflow Inflow Hydraulic retention time, days Hydraulic retention time, days

0.08 400 0.06 Zn TSS 300

0.04 200 Zn, mg/L Zn, TS Sm , g/L 0.02 100

0.00 0 1.5 4.5 7.5 10.5 13.5 1.5 4.5 7.5 10.5 13.5

Inflow Inflow Hydraulic retention time, days Hydraulic retention time, days

Fig 5.4 Inorganic COC (Al, B, Cu, Ni, Se, Zn, and TSS) concentrations in inflow and reactor outflows for Series A (solid line) and Series B (dashed line) for treatment periods 1-5. Note: “Water Quality Criteria Target” represented by dashed line: Al = 0.087 mg/L (USEPA WQC) B = 1.5 mg/L (CEQG) Cu = 0.007 mg/L (Alberta WQG)

174

Ni = 0.0096 mg/L (CEQG) Se = 0.001 mg/L (CEQG) Zn = 0.03 mg/L (CEQG)

175

a) Inflow Outflow

Period 1 Period 2 Period 3 Period 4 Period 5 100

80 * 60 Survival, % Survival,

40 *

dubia * . 20 C

Inflow Inflow Inflow Inflow Inflow

O utflowO A utflow B O utflowO A utflow B OutflowO A utflow B O utflowO A utflow B O utflowOutflow A B b)

40 Period 1 Period 2 Period 3 Period 4 Period 5

30 *

20 *

10 * *

Neonates/ Surviving Adult Surviving Neonates/ 0

Inflow Inflow Inflow Inflow Inflow

O utflowO A utflow B OutflowOutflow A B OutflowO A utflow B O utflowOutflow A B OutflowO A utflow B

Fig 5.5 Survival (a) and reproduction (b) of Ceriodaphnia dubia exposed to inflow

(untreated) and wetland outflow samples of OSPW treated by pilot-scale CWTS. (n=20 per treatment). Note: 7-8 day durations with static/renewals. Bars with asterisks are significantly different (α = 0.05; p<0.05) as compared with control. Error bars represent standard deviation.

Note: Treatment period timelines and conditions are provided in supplementary data.

176

APPENDIX A

CHAPTER 5

Supplemental Data

177

Table 5.1.A. Treatment dates and conditions for hybrid pilot-scale CWTS.

Average Weekly Daily Evaporation Treatmenta Reactor(s) Location Treatment Dates Rainfall Ambient Air Temperaturee Cumulative Rate (mm h-1) (mm)c UV (MJ/m2)b Average Daily Average Daily Start End Low (°C) High(°C) Period 1 Wetland Greenhouse 7-Mar-16 21-Mar-16 - - 0.2 to 2.0d 11 23 Period 2 Wetland Greenhouse 23-Mar-16 6-Apr-16 - - 0.2 to 2.0d 13 26 Period 3 Wetland Greenhouse 10-Apr-16 24-Apr-16 - - 0.2 to 2.0d 15 20 Period 4 Wetland Greenhouse 25-Apr-16 9-May-16 - - 0.2 to 2.0d 12 22 Period 5 Wetland Greenhouse 10-May-16 24-May-16 - - 0.2 to 2.0d 15 23 Week 1 Photocatalysis Outdoor 11-May-16 18-May-16 1.36 5.00 0.3 to 0.5e 17 30 Week 2 Photocatalysis Outdoor 1-Jun-16 7-Jun-16 1.45 9.14 0.3 to 0.5e 21 31 Week 3 Photocatalysis Outdoor 13-Jun-16 20-Jun-16 0.85 25.40 0.3 to 0.5e 19 32 Week 4 Photocatalysis Outdoor 11-Jul-16 18-Jul-16 1.65 9.91 0.3 to 0.5e 22 36 Week 5 Photocatalysis Outdoor 20-Jul-16 27-Jul-16 1.33 22.86 0.3 to 0.5e 21 35 aEach treatment period represents a 13.5 HRT bUV = 250-400 nm; Apogee SU-100 and HOBO Pendant® Temperature/Light 64K Data Logger cData collected from outdoor weather station accessed for Clemson, SC; accessed via: https://www.wunderground.com/history/ dReported range of evapotranspiration coefficients for outdoor pilot-scale reactors planted with Typha latifolia (Beebe, 2013) eRange of average daily evaporation rates measured in the photocatalytic reactor fTemperature data measured via HOBO Pendant® Temperature/Light 64K Data Logger, Bourne, MA References: Beebe, Donald, "Renovation of Ammonia Contaminated Produced Water Using Constructed Wetlands" (2013). All Dissertations. Paper 1187.

178

Table 5.2.A Mean water characteristics (n=5) measured in hybrid pilot-scale CWTS.

(Ranges are in parenthesis)

Temperature pH DO Alkalinity Hardness Conductivity Sample (mg/L (mg/L (°C) (SU) (mg/L) (uS/cm) CaCO3) CaCO3) Wetland Series 20.2 (15.4- 8.30 (7.95- 8.16 (7.98- 2006 (1791- Inflow 364 (340-394) 169 (160-178) 23.8) 8.65) 8.38) 2210) Series A 20.1 (15.1- 7.72( 7.45- 2.29 (1.99- 1892 (1782- A1 344 (280-394) 139 (120-152) 23.5) 7.98) 2.61) 1987) 20.2 (15.3- 7.66 (7.65- 3.47 (3.01- 1939 (1790- A2 344 (248-405) 152 (145-160) 23.4) 7.69) 3.68) 2061) 20.0 (15.5- 7.67 (7.55- 4.21 (3.25- 1973 (1801- A3 397 (264-488) 141 (112-184) 23.4) 7.75) 4.65) 2163) 19.9 (15.2- 7.69 (7.62- 5.10 (4.45- 2078 (1901- A4 368 (228-440) 148 (130-180) 23.5) 7.75) 5.55) 2204) 20.2 (15.3- 7.82 (7.62- 6.11 (5.88- 2245 (1910- A5 374 (236-448) 167 (142-184) 23.5) 8.15) 6.25) 2678) Series B 20.3 (15.6- 7.73 (7.56- 2.29 (1.97- 1945 (1787- B1 357 (288-424) 168 (152-176) 23.8) 7.84) 2.6) 2059) 20.2 (15.5- 7.70 (7.66- 3.32 (2.65- 1939 (1801- B2 374 (272-430) 161 (146-178) 23.6) 7.76) 4.51) 2104) 20.2 (15.4- 7.80 (7.72- 4.33 (3.32- 1957 (1845- B3 376 (240-488) 154 (132-178) 23.4) 7.92) 5.21) 2083) 20.2 (15.5- 7.76 (7.65- 5.64 (3.74- 2030 (1910- B4 367 (232-450) 166 (138-200) 23.5) 7.82) 6.35) 2181) 20.2 (15.6- 7.79 (7.74- 6.32 (5.55- 2129 (1979- B5 368 (256-440) 176 (128-224) 23.6) 7.91) 6.8) 2301) Photocatalysis Treatment 8.15 (7.78- 3312 (3198- Inflow - 7 ± 1 403 (380-420) 149 (120-170) 8.25) 3557) Week 1 - Outflow - 8.35 8 ± 1 6810 570 182 Week 2 - Outflow - 8.52 8 ± 1 4421 490 174 Week 3 - Outflow - 8.34 8 ± 1 2476 385 112 Week 4 - Outflow - 8.35 8 ± 1 4177 484 160 Week 5 - Outflow - 8.5 8 ± 1 5256 520 178

179

Series A Series B ) 2

300 Softstem Bulrush

200

Treatment Initiation 100

0 0 50 100 150 200 250 300 Green shoot density (shoots/m density Greenshoot Maturation Day ) 2

100 Broadleaf Cattail Treatment Initiation 80

0 0 50 100 150 200 250 300 Green shoot density (shoots/m density Greenshoot Maturation Day

Figure 5.1.A Mean plant shoot density measured in reducing reactors (top graph; n=4) and oxidizing reactors (bottom graph; n=10) during CWTS maturation and treatment periods for replicate wetland Series A and B. Error bars indicate standard deviations.

Note: Reducing reactors = softstem bulrush S. tabernaemontani; oxidizing reactors = broadleaf cattail (T. latifolia)

180

Reducing Reactors Oxidizing Reactors

250

200

150

100

Treatment Initiation 50 Mean shoot height (cm) height Meanshoot 0 0 50 100 150 200 250 Maturation Day

Figure 5.2.A Mean plant shoot height measured in reducing (n=6) and oxidizing (n=36) reactors during CWTS maturation and treatment periods. Error bars indicate standard deviations. Note: Reducing reactors = softstem bulrush S. tabernaemontani; oxidizing reactors = broadleaf cattail (T. latifolia)

Reducing Reactors Oxidizing Reactors

400 Treatment Initiation 300

200

100

-100 O R P(m V ) -200

-300

-400 0 50 100 150 200 250 300 Maturation Day

Figure 5.3.A Mean oxidation-reduction potential (ORP; mV) measured in reducing (n=4) and oxidizing reactors (n=10) during CWTS maturation and treatment periods. Error bars indicate standard deviations. Note: Reducing reactors = softstem bulrush S. tabernaemontani; oxidizing reactors = broadleaf cattail (T. latifolia)

181

20 Relative Intensity % Intensity Relative

O2 O 3 O4 O5 O6 O7 O2S O3S O4S O5S N3OSNO2S Heteroatom Class

Figure 5.4.A Hereroatom classes present in untreated (inflow) OSPW using high- resolution Orbitrap MS.

Note: The chemical composition and profile of NAs contained in OSPW were identified by use of direct injection linear ion trap-orbitrap mass spectrometer (Orbitrab MS; Orbitrap Elite, Thermo Fisher Scientific, San Jose, CA, USA) in negative (ESI−) electrospray according to the method described by Headley et al. (2016).

182

1.0

0.8

0.6

0.4 Distilled Water OSPW (untreated) Transmittance 0.2 OSPW (wetland outflow)

0.0 200 300 400 500 600 700 Wavelength (nm)

Figure 5.5.A. UV/visible transmittance (250 to 700 nm) at 1.0 cm water depth of

untreated (inflow) OSPW and post-wetland treatment outflow. Distilled water included as a reference point.

Note: UV/visible light data were collected using a Spectofluorometer (SpectraMax M2; microplate reader [Molecular Devices Corp. Sunnyvale, Ca])

183

3000 -1 Kd (cm )= 0.019 2500 2 R = 0.962 2000 50% attenuation depth = 36 cm

1500

1000 Irradiance (LUX) Irradiance 500

0 0 10 20 30 40 Water Depth (cm)

Figure 5.6.A. Light extinction coefficient (Kd cm-1) of untreated OSPW.

Note: Light intensity (250-700 nm), both incident and submersed, was recorded at each

water depth (n=20) using a LI-1400 data logger with a LI-190 photometric sensor

(incident light) and a LI-192 submersible sensor (LICOR Biosciences, Lincoln, NE,

USA).

184

CHAPTER SIX

CONCLUSIONS

6.1 Objectives

The rational of this research was to provide a viable approach using hybrid constructed wetland treatment systems for the mitigation of constituents in OSPW. Four major objectives were completed and are presented in Chapters 2 through 5 of this dissertation:

6.2 Identify Constituents of Concern in OSPW and Discern Potential Treatment

Processes

The second chapter of this dissertation focused on the identification of specific

COCs in OSPW that need to be targeted to effectively mitigate ecological risk. The goal of this risk-based approach was to characterize a site-specific OSPW for the purpose of identifying specific COCs needing treatment and informing CWTS processes necessary for altering risks to aquatic biota. COCs identified in OSPW include organics (NAs,

O/G), metals/metalloids, and suspended solids. Toxicity testing confirmed that COCs were in sufficient forms and concentrations to have measurable adverse effects on sentinel aquatic species. Sensitivities of aquatic organisms to OSPW indicated that fish ≥ aquatic invertebrates > macrophytes. The sensitivity distribution of organisms to OSPW in addition to strategic bench-scale manipulations indicate organic constituents are contributing to the observed toxicity. Alteration of the organic fraction of OSPW (i.e.

H2O2 + UV254 and GAC treatments) significantly increased survival and reproduction of

C. dubia as compared to untreated OSPW.

185

Based on multiple lines of evidence, these data indicate that organic fractions (i.e.

O/G, NAs) of OSPW are sources of toxicity. In addition, numeric exceedances of metals/

metalloids indicate the need to decrease concentrations to achieve WQC thresholds.

Results from this study provide critical information to inform mitigation strategies using

passive or semi-passive treatment processes (i.e. constructed treatment wetlands) to

mitigate ecological risks of OSPW to aquatic organisms.

6.3 Photocatalysis of a Commercial Naphthenic Acid using Fixed-film TiO2

The third chapter of this dissertation focused on measuring performance of fixed- film photocatalysis for degradation of a commercial NA. Greater than 90% removal of

Fluka NAs (initial concentration 63 mg/L) was achieved in 4-hr with photocatalysis in fixed-film (TiO2) reactors in direct sunlight. Photocatalysis also eliminated acute toxicity to sentinel species, with mortality decreasing from 100% to 0% after 5-hr of photocatalytic treatment for fathead minnow (Pimephales promelas) and after 4-hr for the freshwater invertebrate (Daphnia magna). In this experiment, measuring responses of aquatic organisms concomitantly with analytical quantification of Fluka NAs over time confirmed alteration of exposures as well as mitigation of risk. Fixed-film TiO2

application may provide an alternative solution for scaling the technology for larger

treatment systems. Photocatalytic degradation using fixed-film TiO2 irradiated with

sunlight achieved efficacious rates and extents of removal of Fluka NAs, indicating the potential for application of this technology for mitigating ecological risks associated with

NAs.

186

6.4 Influence of Commercial Naphthenic Acids on Acid Volatile Sulfide (AVS)

Production and Divalent Metal Precipitation

The fourth chapter of this dissertation focused on potential vulnerabilities of a

microbially mediated treatment process (dissimilatory sulfate reduction). Lack of NA toxicity to SRBs could be beneficial for the passive or semi-passive renovation of process-affected waters using constructed wetland treatment systems. Following exposures of NAs, extent of AVS production were sufficient to achieve ∑SEM:AVS <1, indicating that available sulfide ligands were in excess of SEM (Cu, Ni, and Zn) concentrations regardless of NA exposure concentration (10-80 mg NA/L). In addition, no adverse effects to SRB populations in terms of density, diversity, or relative abundance were measured following exposures of a commercial NA. Since SRB were insensitive to exposures of a relatively potent (in terms of aquatic toxicity) commercial

NA, adverse effects to SRB (and SRB-mediated pathways in wetlands) from exposures of more compositionally complex NAs (i.e. derived from oil sands process affected waters) are not anticipated. Further, lack of toxicity to the overall microbial population, and absence of effect on diversity and community profile is a positive finding, given that maintaining diversity of microbially-mediated pathways could be beneficial for passive wetland treatment systems. Passive systems that utilize these biogeochemical processes can be cost effective alternatives to traditional technologies, and having a robust microbial community capable of performing these processes can improve the efficiency and success of the system (Johnson and Hallberg, 2005; Nelson and Gladden, 2008;

Haakensen et al., 2015). In this study, dissimilatory sulfate reduction and subsequent

187

metal precipitation were not vulnerable to NAs, indicating passive treatment systems

could be used to treat metals occurring in NA affected waters.

6.5 Performance of a Hybrid Pilot-scale Constructed Wetland System for Treating

Oil Sands Process-affected Water from the Athabasca Oil Sands

Using the information gained from the preceding experiments (Chapters 2-4), a process-based design approach was used to design and construct a hybrid pilot-scale

CWTS and facilitate conditions conducive for mitigating risks associated with OSPW.

Results from this experiment provided evidence that problematic constituents contained in OSPW could be treated to an extent necessary to remove toxicity to a sentinel organism (C. dubia). Based on characterization of an OSPW from an active settling basin, COCs needing treatment were identified as: NAs, petroleum hydrocarbons (as

O/G), Al, B, Cu, Ni, Se, Zn, and TSS. The hybrid pilot-scale CWTS used in this study provided passive (i.e. low energy) biogeochemical and advance oxidation processes to transfer or transform COCs to achieve numeric and narrative treatment goals. Results from this study provide proof-of-concept data to inform hybrid passive or semi-passive treatment approaches (i.e. constructed wetlands) that could be used to mitigate COCs contained in OSPWs.

6.6 Conclusion

Results from this study demonstrate that specifically designed hybrid CWTS are a viable treatment option for OSPWs located in the AOS. Sulfate-reducing bacteria, and hence metal treatment via the production and precipitation of acid volatile sulfides are not likely to be adversely influenced by the low molecular weight NA fraction of OSPWs

188

(i.e. presumably the most potent fraction). Fixed-film photocatalysis was effective at decreasing concentrations of commercial NAs and subsequently eliminating toxicity in environmentally relevant rates (i.e. hours). Results from this research provide approaches to identify problematic constituents contained in complex energy derived waters (e.g.

OSPW) and strategies for mitigating risks by altering exposures using passive (low-

energy) treatment systems. Results from this research provide proof-of-concept data to inform hybrid passive or semi-passive treatment approaches (i.e. constructed wetlands)

that could be used to mitigate COCs contained in OSPWs.

189

6.7 References

Allen, E.W., (2008a). Process water treatment in Canada’s oil sands industry: I. Target pollutants and treatment objectives. Journal of Environmental Engineering Science 7:123-138.

Allen, E.W., (2008b). Process water treatment in Canada’s oil sands industry: II. A review of emerging technologies. Journal Environmental Engineering and Science 7:499-524.

Braskerud, B.C., (2002). Factors affecting phosphorus retention in small constructed wetlands treating agricultural non-point source pollution. Ecological Engineering 19(1): 41-61.

Brown, L.D., Ulrich, A.C., (2015). Oil sands naphthenic acids: A review of properties, measurement, and treatment. Chemosphere 127:276-290.

Clemente, J. S., Fedorak, P. M. (2005). A review of the occurrence, analyses, toxicity, and biodegradation of naphthenic acids. Chemosphere 60: 585-600.

Driscoll, C.T., Postek, K.M., (1995). The chemistry of aluminum in surface waters. The Environmental Chemistry of Aluminum. Lewis Publishers, Chelsea, MI, 363-418.

Dorn, P. B. (1992). Case histories—the petroleum industry. Toxicity Reduction: Evaluation and Control. In: Ford, DL (Ed.), Water Quality Management Library 3: 183-223.

190

Gersberg, R.M., Elkins, B.V., Lyon, S.R., & Goldman, C.R., (1986). Role of aquatic plants in wastewater treatment by artificial wetlands. Water Research 20(3): 363- 368.

Huddleston, III G.M., Rodgers, Jr. J.H., (2008). Design of a constructed wetland system for treatment of copper-contaminated wastewater. Environmental Geosciences 15:9-19.

Johnson, D.B., Hallberg, K B., 2005. Acid mine drainage remediation options: a review. Science in the Total Environment 338: 3-14.

Kadlec, R.H., Wallace, S., (2008). Treatment wetlands. CRC press. Boca Raton, Fl.

Knight, R.L., Kadlec, R.H., Ohlendorf, H.M., (1999). The use of treatment wetlands for petroleum industry effluents. Environmental Science and Technology 33:973-980.

Lizama, K., Fletcher, T.D., Sun, G., (2011). Removal processes for arsenic in constructed wetlands. Chemosphere 84(8): 1032-1043.

Martin, J.W., Barri, T., Han, X., Fedorak, P. M., El-Din, M. G., Perez, L., Jiang, J. T. (2010). Ozonation of oil sands process-affected water accelerates microbial bioremediation. Environmental science & technology 44(21): 8350-8356.

Metcalf, E. (2003). Wastewater Engineering: Treatment, Disposal, Reuse, Metcalf & Eddy. Inc., McGraw-Hill, New York.

191

Nelson, E.A., Gladden J.B., 2008. Full-scale treatment wetlands for metal removal from industrial wastewater. Environmental Geosciences 15: 39-48.

Rodgers, Jr. J.H., Castle, J.W., (2008). Constructed wetland systems for efficient and effective treatment of contaminated waters for reuse. Environmental Geosciences 15:1-8.

Schramm LL. (Ed.). (2000). Surfactants: fundamentals and applications in the petroleum industry. Cambridge University Press. N.Y.

Schwindaman,, J.P., Castle, J.W., Rodgers, J H. (2014). Fate and distribution of arsenic in a process-designed pilot-scale constructed wetland treatment system. Ecological Engineering 68:251-259.

192