Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Treatment of log yard run-off with a continuous attached growth bioreactor Liao, Charles 2006

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

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

Item Metadata

Download

Media
831-ubc_2006-0551.pdf [ 6.45MB ]
Metadata
JSON: 831-1.0058949.json
JSON-LD: 831-1.0058949-ld.json
RDF/XML (Pretty): 831-1.0058949-rdf.xml
RDF/JSON: 831-1.0058949-rdf.json
Turtle: 831-1.0058949-turtle.txt
N-Triples: 831-1.0058949-rdf-ntriples.txt
Original Record: 831-1.0058949-source.json
Full Text
831-1.0058949-fulltext.txt
Citation
831-1.0058949.ris

Full Text

TREATMENT OF L O G Y A R D RUN-OFF WITH A CONTINUOUS A T T A C H E D GROWTH BIOREACTOR by , CHARLES LIAO B.A.Sc , The University of British Columbia, 2004 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF APPLIED SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Chemical and Biological Engineering) THE UNIVERSITY OF BRITISH C O L U M B I A October 2006 © Charles Liao, 2006 A b s t r a c t Log yard run-off, which can be toxic and have high levels of biochemical oxygen demand (BOD), chemical oxygen demand (COD) and tannin and lignin (T&L), is a potential threat to the environment. Run-off is generated at log yards when precipitation comes into contact with logs, wood debris and equipment at outdoor wood processing, sorting and storage facilities. Log yards are generally located near bodies of water for the easy transport of logs to the sites; therefore, run-off is likely to enter nearby water sources. Treatment of run-off at log yards is not common. Five run-off samples were collected between September 2004 and July 2005, from two sawmills located in British Columbia (BC). The run-off samples collected had BOD ranging from 16 to 371 mg/L, COD from 230 to 2660 mg/L, tannin and lignin from 200 to 680 mg/L of tannic acid. Four run-off samples were acutely toxic according to the Microtox toxicity test. The purpose of this study was to evaluate the treatability of the run-off with a lab-scale continuous biological attached growth reactor. Biofilm was grown from a mixture of Kraft mill return activated sludge (RAS) and primary treated Kraft mill effluent. The bacteria from the RAS quickly colonized the plastic support material within the reactor. The reactor was used to treat the run-off at different operating conditions. Hydraulic retention time (HRT) and temperature were varied to determine their effect on treatment performance. The reactor showed it was capable of treating run-off with BOD removal ranging from 73.0% to 97.6%, COD removal ranging from 48.0% to 76.9%, and tannin and lignin removal ranging from 27.8% to 60.1%, respectively for different operating conditions. In general, the treatment performance increased with increase in temperature and HRT. There appears to be a transition in microbial culture as temperature decreased from 15°C to 10°C, which hindered the reactor performance. The reactor was able to remove all acute toxicity at 30°C, but at lower temperatures, treated run-off remained acutely toxic. T a b l e o f C o n t e n t s Abstract i i Table of Contents iv List of Tables ix List of Figures xi List of Abbreviations and Acronyms xiv Acknowledgements xv 1 Introduction 1 1.1 What is log yard run-off? 1 1.2 Structure of report 2 2 Literature Review 4 2.1 Characteristics of log yard run-off 4 2.1.1 Constituents of run-off 10 2.1.2 Possible sources of run-off toxicity 12 2.1.3 Chronic effects of log yard run-off 15 2.1.4 Microtox as a screening test 17 2.2 Treatment of log yard run-off. 19 2.2.1 Potential treatment technologies for run-off 21 2.3 Trickling film bioreactor 26 2.3.1 Biofilm startup phase 27 iv 2.3.2 Nutrient addition 30 2.3.3 Environmental Conditions 32 3 Research Objectives 34 4 Materials and Experimental Methods 35 4.1 Run-off sample collection 36 4.1.1 Sample sites 36 4.1.2 Sample procedure 37 4.2 Trickling film reactor design 39 4.3 Biofilm growth phase 41 4.4 Biological treatment of log yard run-off 42 4.4.1 Run-off treatability study 42 4.5 Effects of varying H R T and temperature on reactor 43 4.6 Generating synthetic run-off. 45 4.7 Shake flask method for determining biodegradation 46 4.8 Respirometry study 46 4.9 Determination of the amount of biomass on support material 48 4.10 Analytical Methods 49 4.10.1 Biochemical oxygen demand (BOD) 50 4.10.2 Chemical oxygen demand (COD) 50 4.10.3 Tannin and Lignin 51 4.10.4 Microtox toxicity test 51 v 4.10.5 Total suspended solids 52 4.10.6 Heavy metal analysis 53 4.10.7 Antisapstain chemical analysis 53 5 Results and Discussion 54 5.1 Run-off sample characteristic 54 5.2 Startup and operation of the reactor 58 5.2.1 Process upsets 59 5.3 Treatability of run-off. 60 5.3.1 Zinc removal by bioreactor treating run-off 64 5.3.2 Removal of Antisapstain chemicals during bioreactor operation 66 5.4 Effects of H R T and temperature on reactor performance 67 5.4.1 Effects of varying H R T with constant temperature 67 5.4.2 Reactor performance with varying H R T and temperature 69 5.4.2.1 Characteristics of Synthetic run-off 70 5.4.2.2 Results of factorial design 71 5.4.2.3 Interactive effects of H R T and temperature 75 5.4.2.4 Statistical analysis 79 5.4.2.5 Temperature effect at all temperatures 80 5.4.2.6 Temperature trial at 24 hr H R T 83 5.4.2.7 Overall summary of H R T and temperature trials 85 5.5 Estimating amount of biomass present in reactor 85 5.6 Degradation of run-off in batch trials 86 5.6.1 Degradation within reactor 86 5.6.2 Degradation within holding tank 89 5.7 Respirometry studies 93 6 Conclusions 95 6.1 Run-off characteristics 95 6.2 Attached growth in a lab-scale reactor 95 6.3 Run-off treatment phase 96 6.4 Biomass present in reactor 97 6.5 Batch test for run-off degradation 97 6.6 Run-off kinetic values 97 7 Recommendations for future work 98 7.1 Run-off characterization 98 7.2 Antisapstain chemical degradation 98 7.3 Zinc removal 99 7.4 Transition of microbial culture 99 7.5 Anaerobic treatment 99 7.6 Simulating real conditions 100 References 101 Appendix A Supplementary rainfall data 107 vii Appendix B Supplementary experimental data 108 B . l Heavy removal during treatment of run-off. 108 B.2 Temperature trial at 12 hr H R T 113 Appendix C Reactor startup phase 114 Appendix D Synthetic run-off generation 116 viii L i s t o f T a b l e s Table 2.1 Characteristics of log yard run-off from five Alberta log yards (McDougall, 2002) 5 Table 2.2 Characteristics of wood waste leachate from three types of wood species (Schermer et al., 1976) 6 Table 2.3 Run-off characteristics from Courtney Mill (Woodhouse, 2003) 7 Table 2.4 Water quality of leachates and run-off from log decks, chip piles, bark and sawdust (NCASI, 1992) 8 Table 2.5 Summary of leachate quality from Aspen log pile (Taylor, 1994) 9 Table 2.6 Removal of toxicity and T O C obtained with each technologies (Zenaitis et al., 1999) 22 Table 2.7 Nutrients available in woodwaste leachate (Masbough et al., 2005) 31 Table 2.8 Nutrient levels in run-off samples from five Alberta log yards (McDougall, 2002) 31 Table 2.9 Nutrient data from six log yards in Western Washington (Golding,2004)..... 32 Table 4.1 Details on run-off sample collection 38 Table 4.2 Factorial design for studying effects of H R T and temperature 44 Table 5.1 Run-off sample characteristics 55 Table 5.2 Rainfall data 56 Table 5.3 The strengths of different wastewaters 57 Table 5.4 Antisapstain chemical concentration in treated and untreated run-off 66 ix Table 5.5 Factorial design setup 70 Table 5.6 Characteristics of synthetic run-off. 70 Table 5.7 Results of factorial design 71 Table 5.8 A N O V A test on BOD removal 79 Table 5.9 A N O V A test on C O D removal 79 Table 5.10 A N O V A test on tannin and lignin removal 79 Table 5.11 Reactor performance at 24 hr H R T with varying temperatures 83 Table 5.12 Comparing results of current study to past study 84 Table 5.13 Biomass present within reactor 86 Table 5.14 Run-off kinetic data 94 Table A.1 Daily Rainfall Data 107 x L i s t o f F i g u r e s Figure 4.1 Summary of experimental work 35 Figure 4.2 Run-off pond at site B on November 1, 2005 37 Figure 4.3 Trickling film reactor system 39 Figure 4.4 Reactor system flow schematics 4.1 Figure 5.1 Run-off samples from batch sample #4 and #5 54 Figure 5.2 Support material before being inserted into the bioreactor 58 Figure 5.3 Support materials after six weeks of attached growth 59 Figure 5.4 B O D removal over the first 6 months of operation (error bars represent the standard deviation). Arrows represent points at which the feed was changed over from run-off sample #1 to run-off sample #2 (1) or from run-off sample #2 to sample #3 (2) 61 Figure 5.5 C O D removal over the first 6 months of operation (error bars represent the standard deviation). Arrows represent points at which the feed was changed over from run-off sample #1 to run-off sample #2 (1) or from run-off sample #2 to sample #3 (2) 62 Figure 5.6 Colour comparison between untreated and treated run-off (left beaker is untreated run-off and right beaker is treated run-off) 63 Figure 5.7 Zinc concentration of treated and untreated run-off 65 Figure 5.8 Percent BOD, C O D and tannin and lignin removal different H R T .. 68 Figure 5.9 Actual BOD removal compared to BOD removal estimated by equation 5.2 73 xi Figure 5.10 Actual C O D removal compared to C O D removal estimated by equation 5.3 74 Figure 5.11 Actual tannin and lignin removal compared to tannin and lignin removal estimated by equation 5.4 74 Figure 5.12 Interactive effects of H R T and temperature on B O D removal 75 Figure 5.13 Interactive effects of H R T and temperature on C O D removal 76 Figure 5.14 Interactive effects of H R T and temperature on T & L removal 77 Figure 5.15 Interactive effects of H R T and temperature on treated run-off Microtox EC50 values 78 Figure 5.16 Comparison between temperature trials at 8 hr and 4 hr H R T 80 Figure 5.17 Microtox EC50 values at 4 hr and 8 hr H R T with varying temperature 81 Figure 5.18 BOD profile with 1 s t order fit over 24 hr batch test with reactor biomass at 30°C 87 Figure 5.19 C O D profile with 1 s t order fit over 24 hr batch test with reactor biomass at 30°C 88 Figure 5.20 BOD and C O D profile over 24 hr batch trial with holding tank biomass at 5°C .....90 Figure 5.21 BOD and C O D profile over 24 hr batch trial with holding tank biomass at 15°C 91 Figure 5.22 BOD profile with 1 s t order fit over 24 hr batch trial with holding tank biomass at 30°C 91 xii Figure 5.23 C O D profile with 1 s t order fit over 24 hr batch trial with holding tank biomass at 30°C 92 Figure B . l Aluminum concentration of treated and untreated run-off. 108 Figure B.2 Calcium concentration of treated and untreated run-off. 109 Figure B.3 Copper concentration of treated and untreated run-off 109 Figure B.4 Potassium concentration of treated and untreated run-off. 110 Figure B.5 Magnesium concentration of treated and untreated run-off 110 Figure B.6 Manganese concentration of treated and untreated run-off 111 Figure B.7 Sodium concentration of treated and untreated run-off I l l Figure B.8 Nickel concentration of treated and untreated run-off 112 Figure B.9 Phosphorous concentration of treated and untreated run-off 112 Figure B.10 Temperature trial at 12 hr H R T 113 Figure C . l Percent BOD removal of primary treated Kraft effluent during reactor startup 114 Figure C.2 Percent C O D removal of primary treated Kraft effluent during reactor startup 115 Figure D . l BOD profile over 24 hr after initial insertion of wood chips into run-off sample 116 Figure D.2 C O D profile over 24 hr after initial insertion of wood chips into run-off sample 117 xiii L i s t o f A b b r e v i a t i o n s a n d A c r o n y m s A N O V A Analysis of variance BC British Columbia BOD Biochemical oxygen demand COD Chemical oxygen demand D D A C Didecyldimethylammonium chloride DO Dissolved oxygen EC50 Effective concentration F / M Food to micro-organism ratio HRT Hydraulic retention time IPBC 3-iodo-2-propynl butyl carbamate LC50 Lethal concentration NCASI National Council for Air and Stream Improvement RAS Return activated sludge T & L Tannin and lignin TIE Toxicity identification evaluation TMP Thermal mechanical pulp TOC Total organic carbon TSS Total suspended solids USEPA United States Environmental Protection Agency V F A Volatile fatty acids xiv Acknowledgements I would like to thank my supervisor, Dr. Sheldon Duff, for giving me the opportunity to do graduate school here at UBC. Thank you for your guidance and patience over the period of my project. I really appreciated the freedom you give to your students and I really enjoyed our talks about the Canucks. Thank you to my committee members, Dr. Richard Branion and Dr. Madjid Mohseni for their helpful suggestions in improving this project. Thank you to everyone in the environmental lab, Steve, Nicole, Allan and John for all your help, suggestions and advice along the way, all your helps are greatly appreciated. I would also like to thank the staff at the Department of Chemical and Biological Engineering and Tim from the Pulp and Paper Centre for their help. Funding for this project from NSERC is gratefully acknowledged. I would also like to acknowledge Tom from Envirochem Services Inc. for helping with sample collection and for sponsoring the antisapstain chemicals and heavy metal analysis. Thank you also to the staff at Hemmera, CANTEST and A L S Environmental for conducting the antisapstain chemicals and heavy metal analysis. Finally, I would like to thank my parents for their constant support and encouragement. xv 1 I n t r o d u c t i o n 1.1 W h a t is log y a r d run-off? A large number of raw logs are stored and processed at outdoor log yards where they can be exposed to precipitation. Water can be introduced into log yards by rainfall, snow melt, carryover from logs transported in water, equipment cleaning and sprinkling of logs to prevent fire. When water comes into contact with logs and wood debris, a run-off is generated. This is usually deep red-brown in colour and may be contaminated with various constituents such as wood extractives, suspended solids, oil and grease, trace metals and antisapstain chemicals. The volume and characteristics of run-off generated vary from site to site and are greatly affected by volume and species of logs handled at the site, amount of precipitation, site characteristics and practices, location and climate of the log yard (Orban, 2000; Woodhouse, 2003). A number of studies have shown that run-off quality frequently exceeds water quality parameters regulated for industrial effluent discharge and is potentially toxic (Bailey et al., 1999a and 1999b, McDougall, 1996 and 2002, Zenaitis and Duff, 2002a, NCASI, 2005). This is of great concern since most log yards in British Columbia are located along riversides or coastlines for easy transport of logs. Therefore, it is likely that run-off will flow into surrounding water sources. 1 A permit for discharging run-off is typically not required, but the discharge of antisapstain chemicals is regulated under the Environmental Management Act (2004). Therefore, the discharge of run-off is not regulated as long as the antisapstain chemical concentration does not exceed the regulated concentration. However, since run-off can be potentially detrimental to the surrounding environment; researchers, industries and regulators are starting to pay more attention to it. The purpose of this study is to evaluate a treatment process for run-off in the hope of eventually applying it at a commercial log yard. 1.2 Structure of report Section 2 is a review of literature related to log yard run-off and processes that are used, or have been proposed, to treat it. The section starts by summarizing past studies which characterized run-off, then the constituents within the run-off are discussed, followed by sources of run-off toxicity. Various treatment processes for treating run-off are then described along with their results. The section concludes by giving a description of the biological trickling filter process including background information on biofilms and the effects of nutrients and environmental conditions. Section 3 defines the objectives and scope of this project. Section 4 describes the methods and materials used in the laboratory for characterizing run-off and determining treatment performance. The section starts by providing details on run-off sample collection, including site and date of sample collection. A description 2 of the reactor system setup is then given, and the process for developing a biofilm is described. Detailed descriptions of the run-off treatment phase, followed by methods used for biomass quantification are given. The section concludes with analytical methods used for analyzing run-off characteristics and treatment performance. Section 5 shows and discusses the results from the laboratory work, starting with run-off characterization, followed by startup of the attached growth reactor. The treatability of run-off in a continuous reactor is first reported; then the results from trials in which HRT and temperature were varied are provided. An estimation of the amount of biomass present in the reactor is then given, followed by batch tests for determining run-off degradation rate. Respirometry, aimed at generating kinetic coefficients in run-off biodegradation, concludes the section. Section 6 summarizes the results of the project and section 7 proposes suggestions for future research with log yard run-off. 3 2 Literature Review 2.1 Characteristics of log yard run-off There have been several studies in the past that tried to characterize the strength of log yard run-off by conventional wastewater assays. However, the strength of the run-off can vary significantly between log yards and even between different locations and different sample times at a single log yard. This inconsistency makes the generalization of log yard run-off characteristics difficult. Based on reports by National Council for Air and Stream Improvement (NCASI) and British Columbia (BC) Ministry of Environment, McDougall (1996) found a wide range in run-off characteristics. Run-off biochemical oxygen demand (BOD) and chemical oxygen demand (COD) ranged from 6 to 4950 mg/L and 11 to 6530 mg/L, respectively. McDougall (1996) also surveyed 33 log yards in Alberta in 1995. Samples collected from three sites were used to characterize run-off strength, but results from only two sites were reported. Samples analyzed from two sites showed a BOD range from 4 to 465 mg/L, COD range from 75 to 1660 mg/L and total suspended solids (TSS) range from 7 to 812 mg/L. This study found that the strength of run-off was determined by many factors, which include: the amount of precipitation, types of wood species present, site location and how logs were stored. 4 In a report by Alberta Environment, McDougall (2002) monitored run-off characteristics from three log yards in Alberta between 1996 to 1998. Table 2.1 sums up the results. Table 2.1 Characteristics of log yard run-off from five Alberta log yards (McDougall, 2002). Parameter No. of Samples Median Minimum Maximum BOD (mg/L) 72 • 157 23 1800 COD (mg/L) 95 608 160 3500 TOC (mg/L) 46 264 62 1080 Tannin & lignin (mg/L) 56 30.3 3.8 345 Resin acids (ng/L) 24 540 <10 15265 Fatty acids 24 119 <10 1218 Rainbow trout 96hr L C 5 0 (%v/v) 17 70 >100 7.1 Microtox EC50 (% v/v) 59 46 >100 1.5 PH 95 7.3 6.2 9.1 TSS (mg/L) 89 85 <0.4 3015 Schermer and Phipps (1976) studied wood waste leachate from three types of wood -Douglas fir, Hemlock and Cedar. The authors found that COD, tannin and lignin concentration (measured as mg/L of tannic acid) and toxicity varied significantly between 5 different types of wood species and within the same wood species. Table 2.2 provides selected data of the waste leachate generated by the three different types of wood species. Table 2.2 Characteristics of wood waste leachate from three types of wood species (Schermer et al., 1976). Wood species COD Tannin and lignin 96 hr L C 5 0 (mg/L) (mg/L tannic acid) (%v/v) Douglas fir 3155 -8282 823 - 2436 30.6 to 5.6 Hemlock 1397-5172 359- 1817 100-19 Cedar 1911 -3677 1131 -2002 45-20 In a technical report published by Fisheries and Oceans Canada, Samis et al. (1999) studied wood residue leachate from hogfuel. The authors found a BOD and COD concentration of 550 mg/L and 1100 mg/L respectively. The leachate samples collected were generally acidic with pH ranging from 3.5 to 6.5. An audit report done by British Columbia Ministry of Environment, Lands and Parks (2001) studied the operations of 41 dryland log sorting sites located in the Vancouver Island region. The audit was done during spring of 1999. The report showed a BOD range of 14 to 725 mg/L from 16 samples collected from 16 sites. Samples collected from 26 sites had a TSS range from 8 to 26,400 mg/L. A total of 14 sites were sampled to analyze for toxicity; all 14 sites showed toxic response with Microtox EC50 ranging from 75.4% to 1.92%. Out of the 14 sites studied, 10 sites had a Microtox EC50 of below 6 20%. The report concluded that, in most cases, the flows are low enough that there is sufficient dilution in the receiving environment to prevent negative effects. Woodhouse (2003) studied run-off collected from a sawmill located in Courtenay, BC. Nine different samples were collected and analyzed. The strength of the run-off collected varied significantly between different samples. Table 2.3 summarizes the results of the study. Table 2.3 Run-off characteristics from Courtney mill (Woodhouse, 2003). Sample number BOD (mg/L) COD (mg/L) Tannin and lignin (mg/L) Microtox EC50 (%v/v) 1 325 (± 35) 1360 (±45) 435 (± 5) 13 2 395 (± 20) 1710 (±40) 600 (± 10) 7.11 3 - 1780 (±20) 470 (± 5) -4 660 (± 15) 2995 (± 75) 1210 (± 10) 4.35 5 0 ( ± 5 ) 50 (± 5) 10 (± 0) >100 6 35 (± 15) 260 (± 30) 65 (± 5) 91 7 625 (± 45) 2370 (± 25) 685 (± 5) 4.9 8 515 (±25) 2100 (± 10) 655 (± 10) 5.1 9 25 (±5) 40 (± 30) 5 ( ± 0 ) >100 7 The National Council for Air and Stream Improvement (1992) did a study on the water quality of leachates and run-off from different parts of a log yard. Table 2.4 summarizes the results of the study. Table 2.4 Water quality of leachates and run-off from log decks, chip piles, bark and sawdust (NCASI, 1992). BOD (mg/L) COD (mg/L) TSS (mg/L) Log yard 352 N / A 1580 Wet deck 11-52 83-115 76-440 Dry deck 110-300 590-11000 132-164 Bark pile N / A 8700 4800 Sawdust pile 210-339 240-440 1665-5400 Chip pile 117-630 280-4400 1100-4100 Redwood chip pile 840 460-2500 N / A Golding (2004) studied the run-off quality from six log yards in Western Washington. The study found BOD ranging from 34 to 630 mg/L with a median BOD of 129 mg/L and TSS ranging from 8 to 2310 mg/L with a median TSS of 392 mg/L. The study also showed that the run-off samples collected were slightly acidic with pH ranging from 4.5 to 5.7 with a median pH of 5.5. 8 In another study done by British Columbia Ministry of Environment, Lands and Parks, Taylor (1994) studied leachate from Aspen log stacks simulated in the field as they would be at an industrial log storage site. Table 2.5 summarizes the results of the study. Table 2.5 Summary of leachate quality from Aspen log pile (Taylor, 1994). Variables Maximum Median pH 5.0 6.1 BOD (mg/L) 4970 748 COD (mg/L) 6530 1695 Phenols (mg/L) 27.3 4.5 TOC (mg/L) 2230 620 Toxicity index 4.7 3.5 Bailey et al. (1999a) studied the toxicity of run-off from nine sawmills in British Columbia. A total of 58 samples were collected over a period of 23 months. Samples were taken from each mill up to 6 times and a maximum of 8 discharge points. The study showed a toxic response as shown by 96 hr LC 5 o < 100% to rainbow trout in 42 out of 58 samples (72%) and more than half of the samples (57%) caused 100% mortality. In another study done by Bailey et al. (1999b), run-off samples were collected from three Vancouver Island sawmills at different discharge points during a period of two years. A total of 27 samples were tested for acute toxicity to rainbow trout; 26 samples showed 9 toxic effects with L C 5 0 values < 100%. The L C 5 0 value ranged from 20% to 100%. A total of 22 samples showed 100% mortality within 96 hrs when tested at full strength. 2.1.1 Constituents of run-off A report published by National Council for Air and Stream Improvement (2005) summarized the contents of run-off. The main constituents of run-off are wood extractives, trace metal ions and wood debris. Minor components of run-off include antisapstain chemicals, heavy metals and oil and grease. Wood extractives is a general term for a group of chemicals that can be extracted from wood by water or an organic solvent. The three main types of extractives include phenolics such as tannin and lignin, terpenes/terpenoids such as resin acids and aliphatic compounds such as fats and sterols (NCASI, 2005). Phenolics are aromatic compounds usually with hydroxyl groups, and include as major components the likes of tannin and lignin. Tannins are water-soluble phenolic compounds that protect plants from herbivores and diseases. Lignin is a compound that forms parts of the cell walls of plants; lignin conducts water in plant stems and helps protect plants from diseases. According to Tao et al. (2005), who characterized leachate from woodwaste, tannin and lignin make up about 33% to 45% of total COD in run-off. Tannin and lignin along with humic substances appear to be the major cause for the colour of run-off (Tao et al., 2005). 10 Terpenes are a group of molecules composed of isoprene units (CsHg) and terpenoids are derivatives of terpenes. Terpenes serve as protection against herbivores and as signal compounds and growth regulators for plants (Breitmaier, 2006). Resin acids are part of the terpenes group. Some common resin acids are abietic acid, dehydroabietic acid, pimaric acid and isopimaric acid. Fatty acids are organic acids often with aliphatic tails. There are more than thirty saturated and unsaturated extractive fatty acids. Fatty acids account for about 16% to 34% of total COD in woodwaste leachate (Tao et al., 2005). Wood contains many types of inorganic components. Major inorganic components include calcium, magnesium, potassium, aluminum, manganese and silicone. Minor inorganic components include heavy metals such as zinc and nickel and trace elements such as barium. Inorganic components generally make up 1% or less of wood and 1% to 3% of hog fuel (NCASI, 2005). Antisapstain chemicals are preservatives that protect wood from microbial and fungi attacks. Antisapstain chemicals are usually sprayed onto freshly cut lumber at log storage sites. The most commonly applied antisapstain chemicals in British Columbia are didecyldimethylammonium chloride (DDAC) and 3-iodo-2-propynyl butyl carbamate (IPBC). Residue antisapstain chemicals can be carried away by run-off during rainstorms. D D A C and IPBC discharge are regulated in BC due to their potential toxic effect. The discharge limits are 700 and 120 ppb for D D A C and IPBC, respectively (Szenasy, 1999). 11 2.1.2 Possible sources of run-off toxicity Several studies have been conducted to determine the causes of toxicity from run-off. Researchers had found many potential toxicants such as: heavy metals, antisapstain chemicals, pH, wood extractives, suspended solids and oil and grease. Bailey et al. (1999a) followed the guidelines of toxicity identification evaluation (TIE) procedure setup by the USEPA and identified zinc as a primary source of run-off toxicity. The authors found that a possible source of zinc comes from galvanized roofs, which were installed to prevent release of antisapstain chemical during rain events. It was also found that zinc toxicity is greatly dependant on run-off hardness. The zinc LC50 at hardness of 5 mg/L is 14 pg/L and it increases to approximately 272 pg/L at a hardness of 100 mg/L. Hardness of run-off is generally low; Golding (2004) found the hardness to range from 24 to 226 mg/L with a mean hardness of 38 mg/L from six log yards in Western Washington. Bailey et al. (1999a) suggested that copper could also be a source of run-off toxicity. In the same study, the authors found that when zinc was not the toxicant, the toxicity of the run-off could be caused by wood extractives. In particular, tannin and lignin concentrations correlated with run-off toxicity. This suggests that toxicity is caused by either tannin and lignin or other organic acid wood extractives which co-varied with tannin and lignin. Temmink et al. (1989) found that toxicity to carp from Norway spruce bark extracts was caused by the presence of tannins. 12 Buchanan et al. (1976) found wood bark extractives from Sitka spruce were toxic to adult and larval pink shrimp and larval Dungeness crab, with the 96 hr LC50 values being 205 mg/L, 415 mg/L, and 530 mg/L, respectively. The same study also found bark extractives from Sitka spruce and western Hemlock were toxic to pink salmon fry with a 96 hr LC50 of 100-120 mg/L and 56 mg/L for Sitka spruce and western Hemlock respectively. A study done by Peters et al. (1976) found heartwood lignans, which are compounds that form the building blocks of plant cell walls, and bark extractives from western red cedar were moderately toxic, while terpenes and heartwood tropolones, which are aromatic compounds that protect trees from fungal attacks, were more toxic to coho salmon. The LC50 for terpenes and tropolones are 0.33 and 2.7 mg/L respectively to coho salmon fry. Samis et al. (1999) concluded that toxicity in run-off generated from Douglas fir, pine, spruce, and larch is due to their high concentration of resin acids. The main types of resin acids that were found to be toxic to fish are abietic, dehydroabietic, isopimaric, pimaric, neoabietic, palustric, and sandaracopimaric acids. Oikari et al. (1984) stated that resin acids are the main cause of toxicity in pulp mill effluents. Antisapstain chemicals which are widely applied across British Columbia are another possible cause of toxicity in run-off. Studies done by different research groups had raised concerns about the current regulatory standards for antisapstain chemicals in British Columbia (Wood et al., 1996; Farrell et al., 1998). Farrell et al. (1998) did a study on the acute toxicity of antisapstain chemicals Polyphase P-100 which contains 97% IPBC, and 13 another mixture of IPBC and D D A C at a ratio of 1:6.4. The study found a juvenile rainbow trout 96 hr L C 5 0 of 100 ppb from exposure to Polyphase P-100. The juvenile rainbow trout 96 hr L C 5 0 to the mixture of IPBC and D D A C is 460 ppb. The study also showed that the range of concentration between none acutely toxic to acutely toxic is very narrow. Bailey et al. (1999c) also did a study on the acute toxicity of IPBC, D D A C and a mixture of 1:10 ratio of IPBC to DDAC. The 96 hr L C 5 0 to juvenile rainbow trout is 537 pg/L and 67 pg/L for D D A C and IPBC, respectively. The 96hr LC50 to juvenile rainbow trout for the 1:10 ratio mixture is 482 pg/L of D D A C and 48 pg/L of IPBC. Different tree species can also affect the toxicity of the run-off generated. A study of seven log yards running at different operating conditions was done by Alberta Forest Products Association (1999) from 1996 to 1998. Four species of wood: aspen, pine, spruce, and fir were processed at the different sites. Sites processing aspen and pine produced the most toxic run-off and sites processing spruce had the least toxic run-off. Singer et al. (1997) studied the toxicity of distillates from five different types of wood, Douglas fir, red oak, eastern white pine, southern yellow pine, and yellow-poplar. A l l five species of wood produced toxic distillates. The ranking from most toxic to least toxic is as follows: eastern white pine, yellow-poplar, red oak, Douglas fir, and southern yellow pine. However, since distillates only contain volatile compounds, this ranking might not be the same for run-off, but run-off from these five species of wood may be potentially toxic. O'Connor et al. (1992) did a study on a series of simulated mechanical pulping effluents prepared from different wood species commonly used in Canada. The different effluents were tested for acute toxicity against fathead minnows and 14 Ceriodaphnia affinis. The ranking from the most toxic to the least toxic is as follows: white pine, balsam fir, hemlock, black spruce, and aspen. 2.1.3 Chronic effects of log yard run-off Potential chronic effect on organisms in the surrounding environment is also a concern." Chronic toxicity can have great affects on the surrounding environment in that it can cause abnormal behaviour changes, physiological changes and decrease in swimming performance in fish over a period of time. Past studies have shown that wood extractives such as resin acids, tannins and phenolics and antisapstain chemicals are potential chronic toxicants to fish. Wood et al. (1996) did a study on the sublethal effects of D D A C to rainbow trout. The study showed a decrease in swimming performance after 12 hr and 24 hr exposure to both concentrations of 0.2 mg/L and 0.4 mg/L D D A C . Rainbow trout exposed to 0.4 mg/L of D D A C over a period of 24 hours also had significant increase in blood plasma, lactate, glucose and Cortisol values. Farrell et al. (1998) discovered that rainbow trout had an elevated Cortisol response at a much lower concentration when D D A C is mixed with IPBC compared to exposure to D D A C alone. The study also showed bluegill sunfish exposed to 210 ppb of IPBC for 96 hours had darkened pigmentation and showed lethargic behaviour. A study done by Tatarazako et al. (2002) showed that D D A C inhibited the growth and biological functions of green algae, water flea (C. dubid) and two species of fish (Medaka and Zebrafish). 15 Sublethal concentrations of resin acids have also been shown to have serious affects on livers of trout exposed to the resin acids (Oikari et al., 1984). With a high concentration of resin acid, the ability of the liver to process and remove the toxicant could be insufficient, which could lead to death. With lower concentrations of resin acids, it requires the fish to use more energy for detoxication and could also lead to other physiological dysfunctions such as steroid imbalance or change in reproduction. Another study done by Oikari and Nakari (1982) showed that the level of UDP-glucuronyltransferase, which is a group of enzyme that catalyze the conjugation of glucuronic acid and toxins within the liver, can be reduced by roughly 30% from control values with exposure to a concentration of resin acids at 30% of LC50 value. The reduction of UDP-glucuronyltransferase can lead to many liver disorders. A study done by Tana (1988) showed that resin acids accumulate inside the liver, and that most of the hematological parameters measured were affected by resin acids. The affected parameters included plasma glucose, hemoglobin, lactate, and liver glycogen. The study also found significant changes in concentration of UDP-glucuronyltransferase within the liver. Temmink et al. (1989) did a study on the sublethal effects of tannins on carp. The study showed some behaviour changes including longer waiting time before eating after being fed, the fish exposed to higher concentrations of tannins started to rub their body against glass as if they were infected with skin parasites and the exposed fish tended to swim more often in the upper part of the aquarium. Microscopic examinations revealed 16 damages to the gills of fish exposed. There were some fusion of secondary lamellae and damaged cells within the epithelial layers. 2.1.4 Microtox as a screening test The Microtox test was developed by the Microbics Corporation (now Azur Environmental), using a luminescent bacteria, Vibrio fisheri, as its test organism. The test measures the reduction in light emission when the bacteria are exposed to toxic materials, the decrease in light emission is directly proportional to the toxicity of the sample. The reduction in light emission is caused by the inhibition of enzymes involved in the light-producing metabolic reactions. The assay measures the effective concentration, EC50, where the light emitted is reduced by 50%. The following section discusses how Microtox test can be an effective screening test for toxicity compared to the standard rainbow trout 96 hr LC50 bioassay. The Microtox test is widely used in Canadian pulp and paper mills as an in-house toxicity test, even though most Canadian discharge regulations are based on rainbow trout toxicity test. In Alberta, the Microtox test is recognized as a legal test for toxicity. The Microtox test is preferred for toxicity prediction and for monitoring historical data trends for process control as it is cheaper, and requires much shorter time to perform, compared to the rainbow trout bioassay (Jamieson, 1992). The Microtox toxicity test requires only 15 minutes as compared to a rainbow trout LC50 test which requires 96 hours. Other advantages of Microtox test are that no bacteria culture maintenance is required since the 17 bacteria is freeze dried until immediately prior to use, and only small amount of sample is required for the test (Ard and McDonough, 1996). Many researchers had done studies in the past to compare results from Microtox toxicity tests to other types of bioassays using different test organisms. There have being varying results regarding the sensitivity of Microtox test in analyzing different types of waste effluents. However, in general, most of the literature indicates that Microtox is an efficient indicator for forest industry effluent toxicity as compared to other bioassays. Firth and Backman (1990) did a study comparing the Microtox toxicity test to rainbow trout acute test and Ceriodaphnia chronic test for mill effluents. The authors collected 38 data points representing both treated and untreated effluents from Kraft and sulphite mills. The authors found that a linear regression of a log plot of rainbow trout LC50 versus Microtox EC50 to give the best prediction for rainbow trout acute toxicity from Microtox toxicity. The plot had a correlation coefficient of 0.90. Microtox results also correlated well linearly with Ceriodaphnia chronic test with a correlation coefficient of 0.84. The authors concluded that Microtox toxicity test is a good screening tool for both acute toxicity and chronic toxicity from pulp and paper mills. Another study done by Fraser (Jamieson, 1992) compared the Microtox test to rainbow trout and Daphnia magna tests; the results showed that the Microtox test correlated well with rainbow trout acute test, but not well with Daphnia magna test. 18 Codina et al. (1993) did a study on the effectiveness of the Microtox test for detecting metal toxicity. The authors found the Microtox test to be effective for detecting zinc, copper, and mercury toxicity, but not toxicity associated with cadmium, chromium and nickel. Another study done by Toussaint et al. (1995) compared the Microtox test to five other test organisms, including Daphnia magna, Ceriodaphnia dubia, green algae, fathead minnows and mysid shrimp. The authors found Microtox test to be less sensitive than the other five test organisms, however, the authors stated that the Microtox test is suitable for estimating toxicity. A review done by Munkittrick et al. (1991) also concluded that Microtox test is not as sensitive to metal toxicity as rainbow trout and Daphnia, but is more comparable to fathead minnows. In the review, the authors also found that as the complexity and toxicity of the waste effluent increased, the Microtox sensitivity and correlation to other types of test organisms also increased. Based on the literature sources mentioned, it was concluded that Microtox test is a suitable bioassay for analyzing log yard run-off. Therefore, the toxicity tests done in this study were based on Microtox tests. 2.2 Treatment of log yard run-off Based on published literature, most log yards do not aggressively treat the run-off generated. According to a survey done by Orban (2000) on log yards and dryland sort in British Columbia, of the 72 respondents to the survey, 64 sites had visible run-off and 19 only 35 of the sites collected their run-offs. Of the 35 sites that collected the run-off, 29 of them treated the run-off with devices like oil and water separators or sediment traps. McDougall (1996) surveyed log yard run-off treatment practices for 33 log yards in Alberta. Only two of the log yards actively treat their run-off by combining run-off with their wastewater, which then goes through biological wastewater treatment. Twelve sites relied on passive treatment, which directs the run-off from the site through natural vegetation or vegetative filter traps. Twelve log yards direct their run-off to ditches, while seven other log yards have retention ponds or dug outs for collecting run-off. The British Columbia Ministry of Environment, Lands and Parks (2001) did an audit report on 41 dryland log sorting sites located on Vancouver Island and found that 26 out of 41 sites have some form of treatment on site. The different treatment facilities include septic tanks, oil and water separators and settling ponds. The report also stated that the treatment facilities were usually retrofitted to an existing site and usually designed to fit the available space at the site rather than designed to meet specific discharge requirements. From section 2.1, samples collected from 14 sites were all acutely toxic, which indicates the treatment facilities are insufficient in treating run-off. Samples taken from the different sites indicated that many of the treatment facilities do not meet discharge qualities generally attainable with these types of works. It was observed that sites treating run-off with existing facilities have reduced ability to collect oil and hydrocarbons during peak flows. The audit also found many treatment facilities were full of solids, which compromised their effectiveness and contributed to poor discharge 20 quality. Constant clean out is necessary to remove the solids, however due to design limitations, many facilities require a period of dry weather prior to clean out. Based on the literature cited, it is essential that more sites treat their run-off and better treatment methods have to be implemented to improve the effectiveness of treating run-off. 2.2.1 Potential treatment technologies for run-off The following section is a review of the treatment methods that have been applied to log yard run-off and similar waste streams in the past. Zenaitis et al. (1999) tested four types of treatment technologies for treating run-off toxicity and removing total organic carbon (TOC). The four techniques used were: activated carbon, ozonation, organically tailored minerals, and coagulation and flocculation. Activated carbon and ozonation were found to be the most effective in removing run-off toxicity, while coagulation and flocculation and tailored minerals were less effective. Table 2.6 summarizes the results of the four techniques. 21 Table 2.6 Removal of toxicity and T O C obtained with each technologies (Zenaitis et al., 1999). Technology Toxicity Removal (%) TOC Removal (%) Tailored minerals 31 35 Coagulation/Flocculation 12 27 Activated Carbon 100 90 Ozonation 95 N / A However, the authors noted that activated carbon is only economical for wastewater that contains low dissolved organics. The typical cut-off value for activated carbon treatment is < 50 mg/L of TOC. Based on previous literature cited, the TOC value for run-off could potentially far exceed the cut-off value of 50 mg/L, therefore, it was concluded that activated carbon is not a treatment technology suitable for treating run-off. A further study was done by Zenaitis and Duff (2002a) to study the effectiveness of ozonation in treating log yard run-off. Run-off samples were collected from two sawmills located in British Columbia. The authors found that ozonation significantly reduced toxicity by 80 to 90% as measured by Microtox toxicity test. Ozonation was also efficient in reducing tannin and lignin and dehydroabietic acid. The removal efficiencies were 90% to 95% and 95% to 100% for tannin and lignin and dehydroabietic acid, respectively. The authors discovered that both toxicity removal and tannin and lignin removal improved slightly under neutral pH as compared to acidic condition, while the removal rate improved significantly for dehydroabietic acid under neutral pH. However, 22 the study showed that ozonation is not really efficient for reducing BOD and COD. The removals were 15 to 25% and 30 to 35% for BOD and COD, respectively. It was found that pH did not have an effect on the reduction of BOD and COD. The study also found that ozonation can achieve similar removal efficiency regardless of the type of wood species involved in the different run-off samples. A study combining ozonation and biological treatment on run-off was carried out by Zenaitis et al. (2002b). Run-off was treated in a batch biological reactor operating at 35°C for 48 hours. The batch biological treatment was efficient in removing BOD, COD and tannin and lignin with reductions of 99%, 80% and 90%, respectively. Toxicity was also reduced from an initial EC50 of 1.83% to an EC50 of 50.4%. The effects of using ozonation as pretreatment and post-treatment for biological treatment were also investigated. Tannin and lignin and toxicity were rapidly reduced during ozone pretreatment with a reduction of 70% and 71%, respectively. However, both BOD and COD concentrations were reduced by less than 10% during ozone pretreatment. Ozonation was able to further reduce tannin and lignin and COD after biological treatment, however, there is no further reduction in toxicity and BOD actually increased by 38% due to the conversion of COD to BOD. Doig (2005) treated run-off samples from three dry land sorts in British Columbia with a sand column filtration process using oxide-coated sand. When operated as a re-circulating filter for 24 hours the oxide-coated sand was able to reduce COD up to 86%, turbidity up to 92%. The oxide-coated sand was also able to remove suspended solids 23 and heavy metals. TSS reduction was close to 100%, while achieving 100% reduction for zinc, copper, and aluminum. The pH of the run-off samples also increased from an average of 4.5 to 6.5. The filtration process continued functioning after eight successful treatment experiments without saturation and decrease in performance when operated at low infiltration rates (0.1 m/hr). Therefore, regeneration of the column was not required. However, this treatment method showed limitations to COD loading rates, therefore requiring a pretreatment stage when dealing with high strength run-off. The author suggested that a biological pretreatment might be included to optimize the treatment performance of the system. Birkbeck examined the effectiveness of various biological treatment technologies in treating wood-residue leachate (Samis et al., 1999). Five different treatment methods were investigated, which were aerated lagoon, fluidized bed system, artificial wetland, flow-through and sequencing-batch activated sludge. The aerated lagoon was found to be ineffective in treating wood-residue leachate with BOD and COD removal of 0% and 8%, respectively; however, the acute toxicity of the leachate was completely removed. The fluidized bed system with a retention time of one day reduced BOD and COD by 31 % and 18%>, respectively, but the treated effluent remained toxic. A constructed artificial wetland was ineffective in removing toxicity with a 10 day retention time. The system reduced BOD by 43% and COD by 29%, however the author noted that the removal of BOD and COD was more effective if the initial levels of BOD and COD were kept relatively low with a concentration of 463 mg/L and 1024 mg/L, respectively. The treatment performance decreased significantly with a removal of only 4% and 19%, 24 respectively, for BOD and COD with initial values of 535 mg/L and 1726 mg/L. Activated sludge systems were very effective in treating wood-residue leachate with the sequencing-batch system removing 94% of BOD and 67% of COD, while the flow-through activated sludge system with six day retention time reduced BOD by 97% and COD by 69%. For both activated sludge systems, the acute toxicity was completely removed. However, the author noted that the performance of the activated sludge system could deteriorate during rainstorms i f the mixed-liquor suspended solids concentration fell below the range for effective treatment due to dilution from rain. Therefore, activated sludge systems are not suitable for the highly variable flows associated with log yard run-off. Masbough et al. (2005) studied the effectiveness of constructed wetland for treating woodwaste leachate. Four identical pilot-scale wetlands planted with cattail (Typha latifolia) were constructed. Woodwaste leachate was passed through the wetlands with 7 day hydraulic retention time. Nutrient addition and pH adjustments were made to improve treatment performance. The wetlands were able to remove in average 60%, 50%, 69% and 42% of BOD, COD, volatile fatty acids and tannin and lignin, respectively. Woodhouse (2003) studied the ability of a batch biological trickling film reactor in treating run-off. Treatment trials were run at three different temperatures with a retention time of 24 hours. The three different temperatures were 34°C, 24°C and 5°C. Significant reduction of BOD, COD and tannin and lignin were achieved for both 34°C and 24°C trials. At 34°C, a reduction of 94 to 100%, 86 to 93% and 91 to 97% were achieved for 25 BOD, COD and tannin and lignin, respectively. At 24°C, BOD, COD and tannin and lignin were reduced by 97%, 91% and 95%, respectively. The performance of the reactor decreased at 5°C with BOD, COD and tannin and lignin reduction of 76%, 64% and 67%, respectively. The system was also able to achieve near complete removal of toxicity and colour at 34°C. Based on the above literature, biological treatment seems to be the best treatment technology for log yard run-off or similar wastewater. A trickling filter reactor seems to be effective in treating run-off and is more resistant to fluctuations in load and flow than suspended growth systems. Therefore, the trickling filter will be the treatment method for this investigation. 2.3 Trickling film bioreactor According to a report published by the United Nations (2003), trickling filter is the most commonly used aerobic biological attached-growth treatment process for removing organic materials from wastewater. A trickling film reactor is a packed bed reactor consisting of highly permeable plastic support materials onto which organisms can grow to form a biofilm. Wastewater flows down over the biofilm-covered support material, while air flows upward through the void space between the packing materials (Gavrilescu and Macoveanu, 2000). The organic material within the wastewater is removed by adsorption onto the biofilm where the organic material is degraded by aerobic micro-organisms (United Nations, 2003). 26 There are several advantages of trickling filter reactors over conventional biological treatment like activated sludge. The diffusional barrier from the biofilm protects the micro-organisms by shielding them from shock loads and toxic loading (Gavrilescu and Macoveanu, 2000). Since the micro-organisms are not distributed throughout the wastewater, post-treatment solid removal might not be necessary. Fixed biofilm processes have increased stability and a smaller footprint and can handle high volumetric loading rate (Lazarova and Manem, 1996). Trickling filters are also less susceptible than activated sludge to dilution from rain water (Samis et a l , 1999). Even though it is difficult to compare different situations, Parker (1999) found numerous published examples where trickling filters were more economical than conventional activated sludge systems. 2.3.1 Biofilm startup phase Qureshi et al. (2005) describes the mechanism of biofilm formation onto support materials. The authors divided the formation of mature biofilms into four stages: initial attachment, irreversible attachment, early development and maturation of biofilm. The initial attachment is when free floating cells attach to the surface of the support material. Solid support material immersed in an aqueous environment will develop a surface charge which attracts inorganic solutes and highly polar organic molecules onto the support material. The concentration of cations, proteins and organic molecules at the surface provides a nutritious zone for bacteria compared to the bulk solution, which 27 attracts bacteria onto the support material. Fluid flow near the surface can be considered negligible, which also helps bacteria to approach the surface. When attracted to the surface, bacteria will form a temporary association with the surface or bacteria already present at the surface. Once initial association occurs, micro-organisms start to reproduce and colonize the surface, and they become irreversibly attached to the surface. This process involves the production of extracellular polymers, which forms a diffusive barrier that binds the micro-organisms and protects them from surrounding environment. The diffusive barrier also stores the necessary nutrients for microbial growth. The growth of the biofilm is greatly affected by the amount of water and nutrient available to the micro-organisms. Since the amount of water and nutrients which can diffuse into the interior of the biofilm is limited, water channels can develop within the biofilm, which give more pathways for water and nutrients to penetrate deeper into the biofilm. The final stage in biofilm formation is to form a balance of cell production with cell removal through detachment and death. Bacteria will leave the surface if there is an insufficient amount of nutrients available or due to other processes such as sloughing, abrasion and shear stress (Qureshi et al., 2005). Gavrilescu and Macoveanu (2000) developed an equation for the mass balance of cells in a biofilm: (rate of cell accumulation in biofilm) = (rate of cell transport in biofilm) * (sticking efficiency) + (cell growth) - (rate of cell death) - (rate of cell detachment) 28 The seed material for the startup of a biofilm is of great importance. Annachhatre and Bhamidimarri (1992) stated that by using a seed that is accustomed to the wastewater intended for treatment by the biofilm can greatly reduce the startup time of attached growth process. Such a seed can be taken from an existing plant treating similar types of wastewater or grown from a similar carbon source. A high seed concentration is also recommended for reducing startup time since a higher microbial concentration suspended in the bulk solution increases the concentration gradient between bulk solution and surface of support material. However, a high seed concentration will require high carbon loading to maintain a healthy food to micro-organism (F/M) ratio for the formation of biofiilm (Annachhatre and Bhamidimarri, 1992). There are several parameters that can affect the formation of a biofilm. Support material with rough surface tends to enhance cell attachment by providing larger surface areas. The type of material used for the support is also a major factor in biofilm formation; bacteria tend to attach to plastic much faster than to glass or metal. Providing a sufficient amount of nutrients is also a key to producing a healthy biofilm. Phosphorous is especially important since cells saturated with phosphorous tend to flocculate easier and stick together. Temperature also affects the formation of biofilm. High temperature increases microbial growth rate, extracellular polymers production rate and surface adhesion, which helps biofilm formation. The pH of the solution also has an effect in biofilm formation since it affects the charge (Qureshi et al., 2005; Gavrilescu and Macoveanu, 2000). 29 2.3.2 Nutrient addition Sufficient nutrient supply to biological treatment processes is essential to the treatment performance. Micro-organisms require six macronutrients for metabolic processes, the nutrients are carbon, oxygen, hydrogen, nitrogen, sulfur and phosphorous. Out of the six nutrients, carbon, nitrogen and phosphorous are the most essential to micro-organisms (Burgess et al., 1999a). A lack of nitrogen and phosphorous can be the limiting factor for the degradation of hydrocarbons and can potentially create a culture with poor diversity, which can lead to increase in hydraulic retention time required for sufficient treatment even for readily biodegradable substances (Burgess et al., 1999a). Therefore, i f the nutrients present in the run-off is not sufficient, then additional nutrients must be supplied. Burgess et al. (1999b) concluded that the amount of nutrient required will be unique to the process. Forgie (1988) suggested a BOD to nitrogen to phosphorous ratio of 100:5:1 for aerobic biological treatment of leachate. Masbough et al. (2005) found the nutrient level in hog fuel leachate to be very low. Table 2.7 summarizes the results. 30 Table 2.7 Nutrients available in woodwaste leachate (Masbough et al., 2005). Parameter Concentration (mg/L) Ammonia 0.3 Nitrate + nitrite 0.07-0.21 Ortho-phosphate 0.75-1.19 BOD 1702-3465 COD 3221 -3980 McDougall (2002) also found the concentration of nitrogen and phosphorous to be low from log yard run-off. Table 2.8 Nutrient levels in run-off samples from five Alberta log yards (McDougall, 2002). Parameter No. of samples Median concentration (mg/L) Range (mg/L) Ammonia-N 29 0.09 <0.05-1.28 Phosphorous 16 0.82 0.15-3.0 BOD 72 157 23 - 1800 COD 95 608 160-3500 Golding (2004) studied run-off from six log yards in Western Washington and found the concentration of nutrients to be low. 31 Table 2.9 Nutrient data from six log yards in Western Washington (Golding, 2004). Parameter Median (mg/L) Range (mg/L) Ammonia 0.22 0.017-0.36 Nitrate + nitrite 0.05 <0.02 - 0.089 phosphorous 1.06 0.157-2.53 BOD 129 34-630 Fikart (2002) compared the run-off quality from coastal and interior log yards in British Columbia and found the phosphorous levels to be low in both cases. The mean phosphorous concentrations were 0.7 and 0.9 mg/L for coastal and interior log yards, respectively, while the average BOD were 134 and 326 mg/L for coastal and interior log yards, respectively. Based on the literature cited, the nutrient levels within log yard run-off are typically below the ideal BOD to nitrogen to phosphorous ratio. Therefore, nitrogen and phosphorous were added in this investigation to ensure optimal removal conditions. 2.3.3 Environmental Conditions Environmental conditions are also very important factors in biological treatment. Physical conditions such as pH and temperature can be the limiting factors in biological degradation of organic waste (Langwaldt and Puhakka, 2000). 32 The pH of the wastewater to be treated is an important factor. Most biological treatment processes function at, or near, neutral conditions. The pH of the wastewater affects the bioavailability of nutrients and substrates and growth rate of cells. The degradation rate of compounds within the wastewater differs at different pH due to their solubility (Burgess et al, 1999a). Annachhatre and Bhamidimarri (1992) suggested that the pH of the wastewater should be adjusted with NaOH or Ca(OH)2. The operating temperature can greatly affect the treatment performance of a biological process. Mesophilic temperatures (15°C - 40°C) are generally considered to be the best condition for aerobic biological treatment (Burgess et al, 1999a). Higher temperatures increase the rate of biodegradation, but not above 45°C. Temperature also affects the end products from biodegradation and the composition of the biofilm (Burgess et al., 1999a). However, for a biological trickling film reactor to be actually implemented at a log yard in B C , it must be able to treat run-off at low temperatures. In BC, most of the precipitation events occur at winter times, when the temperature is just above or around zero degrees. Research has shown that low temperature can be the limiting factor for degrading some compounds and that reaction rates decrease with decrease in temperature. However, there are some cases where efficient treatment were achieved at low temperatures. Bodik et al. (2002) reported efficient treatment of municipal wastewater with an anaerobic-aerobic baffled filter reactor at 5.9°C. Kalyuzhnyi and Gladchenko (2004) reported efficient treatment of landfill leachates with a sequenced anaerobic-aerobic treatment process under psychrophilic conditions (10±2°C). 33 3 R e s e a r c h O b j e c t i v e s Log yard run-off has the potential to be toxic and be detrimental to the receiving environment. Many treatment processes have being studied in the past. Zenaitis et al. (2002b) determined that log yard run-off is readily biodegradable. Woodhouse (2003) showed that biological trickling filter reactor is effective in treating run-off with 24 hour batch tests and Masbough (2005) successfully treated run-off with construction wetlands. The main purpose for this study is to determine the ability of a continuous biological trickling film reactor in treating run-off. The following sub-objectives are encompassed within the overall objective: 1) Collect and analyze run-off from two mills: site A located in Delta, BC, and site B located on Mitchell Island, BC. 2) Determine treatability of run-off with a continuous trickling filter reactor. 3) Determine if reactor can remove antisapstain chemicals. 4) Determine if there is any removal of dissolved metal ions by the reactor. 5) Study the effects of different hydraulic retention time (HRT) and temperature have on the treatment efficiency. 6) Quantify amount of biomass in reactor. 34 4 Materials and Experimental Methods This project involved several different components of laboratory work. The sequence of these components is shown in Figure 4.1. Reactor design Biofilm growth phase Run-off sample collection and characterization Run-off treatment phase Biomass quantification Run-off treatability study Trial on effects of varying reactor HRT with constant temperature Trial on effects of varying reactor temperature with constant HRT Determine amount of degradation in holding tank Trial to determine fraction of run-off that is readily biodegradable Figure 4.1 Summary of experimental work. 35 A l l of the experiments for this project were performed in the Environmental lab at the Pulp and Paper Centre and the lab at the Clean Energy Research Centre at the University of British Columbia. 4.1 Run-off sample collection 4.1.1 Sample sites Run-off effluent samples studied in this project were collected from two different sites. The names of the sites would be kept confidential and will be denoted site A and site B. The log-yard at site A is located right beside the Fraser River, and the run-off is discharged directly into the Fraser River. The log-yard at site B, shown in Figure 4.2, discharges its run-off to a municipal storm sewer located just outside the log-yard. 36 Figure 4.2 Run-off pond at site B on November 1,2005. Site A processes about 180 million board feet of Douglas fir and hemlock annually. Site B processes primarily Western Red Cedar and the mill processes around 182 million board feet of logs per year. 4.1.2 Sample procedure The site for, and date of sample collection are shown in Table 4.1. 37 Table 4.1 Details on run-off sample collection. Sample Number Sample site Date 1 Site A Sept 24, 2004 2 Site A Nov 6, 2004 3 Site A Jan 20, 2005 4 Site A May 22, 2005 5 SiteB July 6, 2005 Samples #1,3 and 4 were collected directly from the discharge pipe at site A , while sample #2 was collected from a pond inside the site A log-yard. Sample #2 had to be collected from the pond because at high tide the discharge pipe was submerged below water level. Sample #5 was collected from a pond within site B. The pond is shown in figure 4.2. Samples #1, 2 and 3 were collected during rain events, while sample #4 and 5 were collected a day after rain events. A l l samples were collected in 20 L plastic pails purchased from Great Western Containers, Delta, BC, and Home Hardware, Vancouver, BC. Samples #2, 3, 4 and 5 were transferred to the laboratory right after collection, while sample #1 was transferred to the laboratory one day after collection. A l l samples were stored in a cold room within the Pulp and Paper Centre at a temperature of roughly 5°C to minimize sample degradation. 38 4.2 Trickling film reactor design Figure 4.3 Trickling film reactor system. The trickling film reactor system, shown in Figure 4.3, had two main vessels: a trickling film vessel and a holding vessel. The system was designed to run continuously. The trickling film reactor was a 2 L glass vessel with an inside diameter of 11 cm and a height of 21 cm made by Canadian Scientific Glass located in Richmond, BC. A bed consisting of roughly 1 L of plastic biofilm support material rested on top of a plastic stand roughly 5 cm from the bottom of the vessel. The plastic support material is round in shape with a rough surface. The support material has a diameter of 2.1 ± 0.15 cm and a void space of 39 86%. The support material was provided by Hydroxyl Systems Inc. located in Sydney, BC. The 1 L Pyrex holding tank was set on a ring stand at approximately 20 cm below the bottom of the biomass bed of the trickling film reactor. Air was constantly pumped into the holding tank with an aquarium air pump Mk-602 (cULus, Taiwan). The temperature in the trickling film reactor and holding tank was controlled by a V W R Scientific Model 1162 water bath (the water bath only allowed temperature control to within ± 2° Celsius). A Masterflex peristaltic pump (model 7553-80 Cole-Parmer) was used to pump run-off from a refrigerator, where the feed was kept at roughly 5°C to the holding tank through Masterflex Norprene size 25 tubing. Another Masterflex pump Model 7553-80 pump was used to pump the run-off from inside the holding tank to the discharge through Norprene size 25 tubing. From the holding tank, another peristaltic pump was used to continuously pump the run-off to the trickling film reactor through Masterflex Norprene size 18 tubing. After the effluent flowed through the support materials within the trickling film reactor, it was pumped back into the holding tank. Both the fiowrate from the holding tank to reactor and from the reactor back to the holding tank were held constant at around 2.9 mL/s (10 L/hr) throughout the whole project. A ChronTrol X T timer controller was used to control the rate of effluent feed to the holding tank and the rate of discharge, depending on the desired HRT. Figure 4.4 shows the flow schematic of the reactor system. 40 feed discharge feed Figure 4.4 Reactor system flow schematic. 4.3 Biofilm growth phase To start attached growth of microorganisms onto the support materials, a mixture of return activated sludge (RAS) and primary treated Kraft pulp mill effluent was used. RAS was chosen as the seed material for this project since pulp mill effluent most resembles log yard run-off Using seed material from a similar wastewater process will decrease the reactor startup time (Annachhatre and Bhamidimarri, 1992). Also a study done by Nishihara et al. (2000) showed that bacteria from activated sludge has the potential to treat antisapstain chemical D D A C . Both the RAS and primary treated effluent were from the Western Pulp Kraft mill located in Squamish, BC. The ratio of primary treated effluent to RAS was 5:1. For this experiment, 250 ml of RAS and 41 1250 mL of primary treated effluent were used. The reactor temperature was controlled at 30°C with a water bath. A n aquarium air pump was used to constantly aerate the reactor during growth phase. The mixture of RAS and primary treated effluent were circulated through the reactor for 3 days before primary treated effluent was constantly fed to the reactor with a HRT of 12 hours for roughly 6 weeks, while discharge from the reactor were collected and analyzed during the 6 week period to ensure that there was steady removal of BOD, and COD (Appendix C). Primary treated effluent has an average BOD and COD of 260 ± 45 mg/L and 1970 ± 450 mg/L, respectively. No additional nutrients were added to the reactor during growth phase, since the primary treated Kraft mill effluent was collected from a point after nutrient addition. No run-off was added to the reactor during the initial growth phase, until the discharge from the reactor showed a consistent 95% and 50% removal of BOD and COD, respectively, from the Kraft mill effluent. After the reactor's performance stabilized, the feed was changed from primary treated Kraft mill effluent to log-yard run-off. 4.4 Biological treatment of log yard run-off 4.4.1 Run-off treatability study After the reactor had shown 95% removal of BOD and 50% removal of COD from primary treated Kraft effluent, the feed was changed to run-off. During the initial run-off treatability study, the reactor was kept at 30°C and with a HRT of 12 hours. Ammonium 42 hydroxide and potassium phosphate were added to the run-off sample to provide essential nutrients nitrogen and phosphorous to ensure bacteria growth. The ratio of BOD:N:P was 100:5:1 as suggested by Forgie (1988). The run-off samples were generally acidic with pH ranging from 4.5 to 7, therefore the pH of the run-off samples were adjusted to roughly pH 7 (± 0.1) with sodium hydroxide prior to use. Treated run-off samples were collected from the reactor on a regular basis, and BOD and COD tests were performed to analyze the performance of the reactor. The treatability study lasted for roughly six months until the reactor was able to remove 90% and 60% of BOD and COD, respectively. 4.5 Effects of varying HRT and temperature on reactor performance After the reactor has shown that it was capable of treating the run-off effluent, it was interesting to see how the reactor would perform at different operating conditions. The two parameters that were studied intensely were hydraulic retention time (HRT) and temperature. The retention time of the reactor was varied to try to determine the effect of retention time have on the treatment efficiency while holding the temperature constant. The different retention times used were 12, 8, and 4 hours. The retention time was changed by changing the discharge flow rate from the holding tank. Both 8 hour and 4 hour retention times were run for around five weeks, samples were collected along the way and analyzed. Tests run were BOD, COD, Microtox toxicity test, and tannin and lignin. 43 Another important factor in determining the efficiency of a biological treatment is temperature. Therefore, the effect of temperature on the treatment efficiency for the reactor was also studied. The range of temperatures used was from 30°C to 5°C, which represents the seasonal temperatures of Vancouver. The study started at 30°C and slowly went down to 5°C with a decrement of 3°C each time. The reactor ran for about one week for each different temperature and samples were collected each week and analyzed. Tests run were BOD, COD, Microtox toxicity test, and tannin and lignin analysis. To better understand the effects of HRT and temperature including interactive effects; a 3x3 factorial design was used. The factorial design does not include the HRT trial previously performed with 12 hr, 8 hr and 4 hr retention times at 30°C. Table 4.2 shows the factorial design setup. Table 4.2 Factorial design for studying effects of H R T and temperature. Temperature (°C) HRT (hr) 5 4 8 12 17.5 4 8 12 30 4 8 12 For each condition within the factorial design, the reactor was run for around three weeks and multiple samples were collected and analyzed. The factorial design was ran in order of 12 hr HRT from 30°C to 5°C, then 8 hr HRT from 30°C to 5°C and finally at 4 hr HRT from 30°C to 5°C. The factorial design was carried out in sequence instead of 44 randomized run because of the constraints of the reactor. As mentioned before, the temperature of the reactor was reduced 3°C at a time to prevent a sudden shock to the biomass, which might potentially die from major changes in operating conditions. A temperature study was also done for retention time of 24 hours to give a better comparison to the batch tests done by Woodhouse (2003). 4.6 Generating synthetic run-off During the HRT and temperature factorial design study, synthetic run-off was used as the reactor feed. The reason for this is twofold. First, the actual run-off samples collected were quite low in strength, and therefore the effects of varying reactor operating parameters might be masked. Secondly, the variable nature of the actual run-off sample would confound the examination of HRT and reactor temperature. To generate artificial run-off, 600 g of Western Red Cedar chips collected from site B were added to a 20 L pail of run-off sample and stirred for 24 hours with a stir bar and a magnetic stir plate. The decision to soak the wood chips for 24 hours was made after several trials, where the strength of the synthetic run-off was monitored by measuring their BOD and COD at different times after insertion of wood chips into run-off samples. Result from one trial is shown in Appendix D. After 24 hours, wood chips were removed from the run-off sample, and then the run-off sample was fed through a filter which removes the wood debris and other solids. 45 4.7 Shake flask method for determining biodegradation To determine the amount of degradation that goes on inside the holding tank, four 50mL samples were removed from the holding tank, centrifuged in a Damon/IEC Division C U -5000 Centrifuge for 5 minutes at 2253xg. After centrifuging, the liquid was discarded while the solids were re-suspended in 50 mL of untreated synthetic run-off inside a 125mL Erlenmeyer shake flask. The four samples are then incubated in a New Brunswick Scientific innOva 4230 Refrigerated Incubator Shaker, at 180 rpm. Samples were taken from the four flasks at 2, 4, 6 and 24 hour after incubation, and analyzed for BOD and COD. Holding tank degradation trials were ran at 5°C, 15°C and 30°C. Samples from holding tank were taken when the reactor was running at the three particular temperature to eliminate the acclimation time required for bacteria adjustment. The shake flask method was also used to determine the amount of degradation done by the reactor. The process was the same as holding tank trials except micro-organisms from the reactor were used instead of samples from holding tank. 4.8 Respirometry study The respirometric method developed by Cech et al. (1985) was used to determine the portion of run-off sample that is readily biodegradable and to determine the kinetic constants. The respirometer is a jacketed glass vessel with ports for dissolved oxygen (DO) measurement, aeration and sample injection. The respirometer was made by 46 Canadian Scientific Glass in Richmond, BC. The temperature of the respirometer is controlled by a V W R Scientific Model 1166 water bath. Mixing within the respirometer was provided by a YSI Inc. Model 5905 DO probe along with a stir bar and a magnetic stir plate. An aquarium air pump (model Elite802) was used to aerate the respirometer. For respirometry, biomass from the reactor was added to the respirometer and aerated for 15 minutes to allow the DO concentration to reach saturation and to remove any residual substrate. After aeration, the DO probe was inserted into the respirometer and the probe was allowed time to stabilize before sample addition. The DO level was measured by an YSI Inc. Model 59 DO meter. The endogenous respiration rate was measured for at least two minutes. Then a known volume of run-off sample was injected into the respirometer and the change in oxygen uptake rate (AOUR) as reflected by changes in DO concentration was monitored. Once the OUR returned to the original endogenous rate, the data collection was stopped. Then the DO probe was removed from the respirometer. The respirometer was aerated again until the DO concentration reached 6 to 8 mg/L before another sample was injected. The data from the DO meter were imported into an Excel spreadsheet where the maximum OUR, AOUR and oxygen consumed were calculated. The amount of biomass used was measured at the end of the trials by the total suspended solids (TSS) method mentioned in section 4.10.5. 47 From the respirometry data collected, the substrate uptake rate can be determined using Monod kinetics and Tessier kinetics. Monod: q = q m a x S (4.1) S + Ks Tessier: q = qmax(\-e~ks) (4.2) q is substrate uptake rate q m a x is maximum substrate uptake rate S is substrate concentration K s and k are constants 4.9 Determination of the amount of biomass on support material To determine the biomass concentration of the fixed film reactor, three samples, each consisting of five pieces of support material, were removed from the reactor. The solids that dislodged when removing the support material were also taken into consideration. The dislodged solids were rinsed off glass beakers, which were used to collect support material from reactor, with distilled water, then the solids were separated from distilled water by filtering through a Whatman Glass Microfibre Filters 934-AH™ with 47 mm diameter. The amount of biomass present was quantified by the TSS method described in section 4.10.5. 48 The support material were dried overnight at 105°C and then cooled to room temperature. Then the mass of the support material along with the biofilm were measured. The mass of the dried biofilm was then calculated by subtracting the mass of an average clean support material piece. A total of 20 pieces of clean support material were used to determine the average mass. Then the mass of the biofilm was added to the mass of the dislodged solids to determine the total mass of the biofilm removed from the reactor. Then the total mass was divided by five to give the biomass for one piece of support material. The biomass for one piece of support material was then multiplied by the number of support material pieces within the reactor to give an estimate of the dried mass of the biofilm present inside the reactor. 4.10 Analytical Methods The following sections describe the methods used for analyzing run-off samples. A l l chemicals used in the tests were purchased from Sigma-Aldrich and Fisher Scientific, unless otherwise stated. 49 4.10.1 Biochemical oxygen demand (BOD) BOD was measured in triplicate samples according to Standard Method 521 OB, 5-day BOD test (APHA, 1992). The dilution water for the test was aerated with an air pump for at least 1 hour before usage. The run-off samples were diluted down to 0.67 to 16.7% v/v, depending on the estimated strength of the sample. The seed material used was return activated sludge (RAS) supplied by Western Pulp located in Squamish, BC. The seed concentration used in each BOD bottle was 1 mL. Samples were incubated at 20°C for 5 days inside a Fisher Scientific low temperature incubator Model 307. The concentration of dissolved oxygen in the samples was measured before and after incubation with a YSI Inc. Model 59 dissolved oxygen meter and Model 5905 DO probe. 4.10.2 Chemical oxygen demand (COD) COD were measured in duplicates according to Standard Method 5220D, Closed Reflux, Colorimetric Method (APHA, 1992). Samples were diluted to 10 to 100% v/v of original strength by distilled water, depending on estimated strength. A blank sample was also prepared. A H A C H COD reactor was used to heat the samples for two hours. A H A C H DR/2000 Direct Reading Spectrophotometer was used to measure absorbance at a wavelength of 600 nm. Potassium hydrogen phthalate (KHP) was used to create a calibration curve ranging from 50 to 1000 mg O 2 / L . 50 4.10.3 Tannin and Lignin Tannin and lignin concentration were measured according to Standard Methods 5550B, Colorimetric Method (APHA, 1992). Samples tested ranged from 25 to 100% v/v of original strength. A l l dilutions were done with distilled water. A blank sample was also prepared. A H A C H DR/2000 Direct Reading Spectrophotometer was used to measure absorbance at a wavelength of 700 nm. Tannic acid was used to produce a calibration curve with concentrations ranging from 0 to 1000 mg/L of tannic acid. 4.10.4 Microtox toxicity test Microtox toxicity tests were performed according to Basic Protocol (Microbics Corp., 1992). Prior to analysis, all samples were centrifuged at 12400xg for 4 minutes with Fisher Scientific micro centrifuge Model 23 5 C to remove the solids. The Microtox test is very sensitive to solids concentration, which could dramatically change the outcome of the tests. Light intensity was measured with a Microbics M500 analyzer. From the observed light intensity, EC50 can be calculated based on the following equations: (4.3) Rt is Correction Factor I is light intensity t 0,samples ^t,samples -1 (4.4) 51 T t is ratio of corrected light lost over light remaining Then a graphical method is used to calculate EC50 by plotting log of concentration versus log of gamma. The following equation is the linear form of the plot: logC = M o g r + loga (4.5) C is concentration of sample (%v/v) b is slope of line a is y-intercept For EC50, r = 1, therefore log C = log a E C 5 0 = 1 0 a (4.6) The Microtox solutions and reagent (Vibrio fisheri) were purchased from Strategic Diagnostics Inc. located in Dewark, Delaware. 4.10.5 Total suspended solids Suspended solids were measured according to Standard Methods 2450D, TSS dried at 103-105°C (APHA, 1992). Fifty mL of sample were used for each test. Run-off samples were filtered through Whatman Glass Microfibre Filters 934-AH™ with 47 mm diameter. Then the filter was dried overnight at 105°C in a V W R Model 1350FM oven. After drying in the oven, the filter was cooled inside a SUNPLATEC Corp. dessicatpr with anhydrous calcium sulfate dessicant until room temperature. A n OHAUS Analytical Plus balance was used to measure the weight of the filter prior to and after drying. 52 4.10.6 Heavy metal analysis Heavy metal analyses were done by A L S Environmental in Vancouver, BC. Heavy metals were analyzed by procedures from Standard Methods, and test methods from USEPA. Samples were collected in 250 mL plastic bottles and stored in a refrigerator before sending off to be analyzed. Samples were sent to A L S Environmental by overnight courier and stored in a cooler box containing ice packs. 4.10.7 Antisapstain chemical analysis Antisapstain chemical analyses were done by Hemmera located in Vancouver, B C and CANTEST located in Vancouver, BC. Antisapstain chemicals were analyzed according to procedures outlined in the British Columbia Environmental Laboratory Manual (2000 edition) -DDACX364 and IPBCX371. Samples were collected in 1 L glass bottles and hydrochloric acid and Rexonic N25-7 were added as preservatives. The preservatives were provided by Hemmera. Samples collected were kept in a refrigerator before being sent off for analysis. Samples were transferred by overnight courier and stored in a cooler box containing ice packs. Antisapstain chemical analysis is a very sensitive test and the result can be greatly affected by the time between sample collection and analysis. To produce the most reliable result, the time between sample collection and analysis shouldn't be more than three days. 53 5 Results and Discussion 5.1 Run-off sample characteristic The five run-off samples collected all differed in strength and physical appearance. Figure 5.1 shows the difference in colour between sample #4 and #5. Run-off sample colours ranged from light brown to dark brown, and the colour appeared to correlate with increase in strength as measured by COD. The stronger samples were also more turbid. A l l run-off samples had a woody odour. Figure 5.1 Run-off samples from batch sample #4 and #5. 54 BOD, COD, Microtox toxicity, and tannin and lignin tests were used to characterize the strength of each run-off sample. Table 5.1 shows a summary of the different run-off samples. T a b l e 5.1 R u n - o f f s a m p l e c h a r a c t e r i s t i c s . Sample number BOD COD Microtox toxicity EC50 Tannin and Lignin pH (mg/L) (mg/L) (% v/v) (mg/L) 1 300 (±53) 2200 (± 650) 9.0 n/a 3.5 (±0.5) 2 90 (±43) 690 (± 155) >100 n/a 5.0 (± 0.5) 3 92 (±4) 600(±250) 70 350 5.5 (±0.5) 4 50(±30) 400(±230) 95 (±34) 450 5.8 (± 0.5) 5 82 (±11) 1400(±160) . 15 (±1.8) 600 5.0 (± 0.5) Tannin and lignin tests were not done for samples 1 and 2. The characteristics of different run-off samples varied significantly based on the time of sample collection. Table 5.2 summarizes the amount of rainfall accumulated in a week up to the day of sample collection. Daily rainfall data are recorded in Appendix A . 55 Table 5.2 Rainfa l l data. Sample number Amount of precipitation accumulated in a week before sampling (mm)1 1 105.6 2 60.4 3 166 4 32.8 5 26.8 (Environment Canada) Sample #3 was collected on the fifth day of heavy rain fall, and sample #4 was taken a day after a heavy rain, which significantly decreased the strength of the run-off sample. Sample #2 was scooped off the ground where run-off was trapped; however the pond also trapped large amounts of rain water, which would also decrease the strength of the run-off sample. Sample #5 was collected after a short period of rain fall. The COD of sample #5 was greater than those of sample #2, 3 and 4, and the Microtox EC50 was also lower; however, the BOD was similar to those from samples #2, 3 and 4. The five samples collected during this project were compared to the samples Woodhouse (2003) collected from a sawmill in Courtenay, BC (table 2.3). In general, the run-off samples collected in this investigation were weaker in strength. This shows the strength of the run-off is greatly dependent on the time of sampling, location of sampling, types of wood processed at the site and the amount of precipitation. 56 To give a better understanding of the strength of the run-off, it is best to compare it to other types of wastewaters. Table 5.3 shows typical strength of municipal wastewater, wastewater from thermo mechanical pulp mill (TMP), wastewater from dairy industries and past run-off studies. The run-off was weaker than wastewaters from the thermo-mechanical pulp mill and dairy industries, but was comparable in strength to municipal wastewater. Table 5.3 The strengths of different wastewaters. Wastewater type BOD COD Microtox EC50 (mg/L) (mg/L) (%v/v) Municipal 1 120-300 260 - 580 -TMP 11502 33402 4.5 - 5.4J Dairy 4 1670-2200 1420 - 4730 -Run-off 0-660 40 - 2995 4.35 ->100 (Woodhouse)5 Run-off 23 - 1800 160-3500 1.5 ->100 (McDougall)6 i — = — 1 (Greater Vancouver Regional District (GVRD), 2004);1 (Magnus et al., 2000a);3 (Magnus et al., 2000b); 4 (Rusten et al., 1992);5 (Woodhouse, 2003);6 (McDougall, 2002) 57 5.2 Startup and operation of the reactor The biomass used to seed the attached growth reactor started to colonize the support material within a few days. Figure 5.2 shows the support material before they were inserted into the reactor. Figure 5.2 Support material before being inserted into the bioreactor. The first visible attached growth onto the support material was observed after two days of circulating the mixture of RAS and primary treated Kraft pulp mill effluent through the reactor. A thin white layer of biofilm covered the surface of the support material. After a few more days, the colour of the biofilm started to change and eventually became a dark brown/black colour. The size of the biofilm started to increase significantly as it grew abundantly on the support material and the reactor walls, as well as inside the connective tubing. Growth inside tubing is not desired, and can lead to clogging inside the tubing, therefore constant cleaning of the tubing is required. The biofilm was allowed to grow continuously for about six weeks after initial colonization by continuously feeding 58 primary treated Kraft pulp mill effluent to the reactor. Figure 5.3 shows the support material after the six week growth period. Figure 5.3 Support material after six weeks of attached growth. The decision to start feeding run-off to the reactor was made after the reactor consistently removed 95% and 50% of BOD and COD, respectively, and the amount of attached growth in the reactor was estimated to be sufficient by the naked eye. The biofilm may have continued to develop after switching to run-off. The startup phase for biofilm can often take up to several months as it is a complex process which involves several different stages as described in section 2.3.1. 5.2.1 Process upsets During the startup of the reactor, as well as during the run-off treatment phase, several upsets occurred within the reactor system. These included plugging in tubing, 59 malfunction of pumps for pumping liquid and air pumps for aerating the system. A l l these upsets either caused the reactor to not function at all or significantly affected the reactor performance. However, the biomass was able to recover within a few days after process upsets and reactor performance returned to normal. Data collected during process upsets were discarded from analysis as they did not represent the reactor performance accurately. 5.3 Treatability of run-off The first phase in run-off treatment trial was to determine i f the run-off was treatable by a continuous biological attached growth reactor. Zenaitis et al. (2002b) and Woodhouse (2003) showed that run-off is readily biodegradable in a batch biological reactor; however, a continuous process was not tested. The treatability study lasted roughly for a period of six months. Run-off samples used for this study were batch samples 1, 2 and 3. Samples were taken before and after treatment on a consistent basis and analyzed. BOD and COD removals were used to monitor the performance of the reactor. Figure 5.4 shows BOD removal versus time while figure 5.5 shows COD removal versus time. 60 BOD removal 100.0 CO 3 0 0 -° 20.0 -10.0 -0.0 -I 1 1 , , , , , , 08-Oct-04 28-Oct-04 17-Nov-04 07-Dec-04 27-Dec-04 16-Jan-05 05-Feb-05 25-Feb-05 17-Mar-0S time (month) Figure 5.4 B O D removal over the first 6 months of operation (error bars represent the standard deviation). Arrows represent points at which the feed was changed over from run-off sample #1 to run-off sample #2 (1) or from run-off sample #2 to sample #3 (2). The reactor made a smooth transition from primary treated Kraft pulp mill effluent to run-off samples in turns of BOD removal. The reactor removed 86% ± 5.6% of BOD from the run-off sample right after transition, which was a slight decrease from the 95% BOD removal from Kraft pulp mill effluent. After approximately three weeks of treating run-off, the BOD removal reached 96% ± 2.3%.' The sudden decrease in BOD removal after six weeks (arrow 1) was caused by switching over from run-off batch sample #1 to batch sample #2. The switch was again followed by approximately two weeks of adaptation before the BOD removal reached 90%. The BOD removal stayed relatively 61 consistent from middle of December 2004 to early March 2005. The BOD removal decreased by about 6.0% from the middle of March to early April 2005 as the run-off sample was switched from batch sample #2 to batch sample #3 (arrow 2). The overall BOD removal throughout the treatability study was 88.1% ± 1.8%. COD removal 100.0 T 28-Oct-04 17-Nov-04 07-Dec-04 27-Dec-04 16-Jan-05 05-Feb-05 25-Feb-05 17-Mar-05 06-Apr-05 26-Apr-05 time (month) Figure 5.5 C O D removal over the first 6 months of operation (error bars represent the standard deviation). Arrows represent points at which the feed was changed over from run-off sample #1 to run-off sample #2 (1) or from run-off sample #2 to sample #3 (2). The reactor was only able to remove about 10% of COD from the run-off sample right after switching over from primary treated Kraft pulp mill effluent. The reactor took about three weeks to adjust to the run-off, over which time the COD removal improved 62 until it reached about 80%. The drastic drop in COD removal around the middle of November 2004 (arrow 1) coincided with a switch of the reactor feed from batch sample #1 to batch sample #2. After approximately two weeks, COD removal came back up to 78.3% ± 3.5%. Consistent COD removal of 79.0% ± 4.7% was achieved between the middle of December 2004 to early March 2005. COD removal decreased to 63.5% ± 0.8% on March 18 2005, when the reactor feed was switched from batch sample #2 to batch sample #3. Overall COD removal throughout the study was 61.3% ± 25.2%. The pH of the treated run-off was generally higher than untreated run-off, with an average increase from 7.0 to 7.5 (an increase of 0.5 ± 0.2). The increase in pH is probably caused by the degradation of low molecular weight volatile fatty acids (VFA). VFAs are well known to be readily biodegradable (Tao et al., 2005; Forgie, 1988). The increase in pH also made the colour of the treated run-off darker (figure 5.6). Figure 5.6 Colour comparison between untreated and treated run-off (left beaker is untreated run-off and right beaker is treated run-off). 63 From this study, it can be concluded that the treatment performance is sensitive to changes in run-off characteristics as indicated by the decrease in both BOD and COD removal and the time it took to adjust when changing run-off samples. COD is also more sensitive to change than BOD as indicated by the bigger changes in removal percentage. 5.3.1 Zinc removal by bioreactor treating run-off Run-off samples were collected before and after treatment and analyzed for heavy metal removal as part of the run-off treatability study. The analyses performed were for a set of metal ions, however, only zinc is discussed in this section. The results for the rest of the metal ions (shown in Appendix B. l ) , showed a general decrease in concentration during treatment. The ability of the reactor to deal with zinc is of great interest since Bailey et al. (1999a) found zinc to be a potential cause of run-off toxicity. Four samples were collected and analyzed and the results are shown in figure 5.7. 64 0.3 0.25 0.2 £ 0.15 o 2 a> u c o O 0.05 'untreated Dtreated ba t ch s a m p l e n u m b e r Figure 5.7 Zinc concentration of treated and untreated run-off. The concentration of zinc is lower in the treated run-off compared to untreated run-off for all four samples collected; however, the cause for the reduction in zinc concentration is unknown. Two possible explanations for the reduction in zinc concentration are that zinc might be metabolized by the biofilm, or zinc might be adsorbed onto the biofilm or support material. Zinc is one of the trace elements that biomass require during metabolism at a concentration less than 1 mg/L (Burgess et al., 1999a). Since the zinc concentration in all run-off samples was well below 1 mg/L, the reduction in zinc concentration is likely due to biofilm metabolism. Metal ions are also known to adsorb onto extracellular ligands present in the biofilm at pH between 6 and 8 (Burgess et al., 1999a). 65 5.3.2 Removal of antisapstain chemicals during bioreactor operation Five samples were collected and analyzed for antisapstain chemicals. Three out of five samples were only analyzed for D D A C concentrations, while two samples were analyzed for both D D A C and IPBC concentrations. Table 5.4 shows the results of antisapstain chemical removal. Table 5.4 Antisapstain chemical concentration in treated and untreated run-off. Date D D A C of untreated run-off D D A C of treated run-off IPBC of untreated run-off IPBC of treated run-off ppb ppb ppb ppb Oct 22 <20 <20 n/a n/a Oct 27 <20 <20 n/a n/a Nov 18 <20 <20 n/a n/a Feb 3 76 <20 < .10 < 10 May 27 <20 <20 <10 <10 A l l samples except one had D D A C concentrations below the detection limit, while all samples of IPBC were below detection limit. The one sample that had detectable concentration of D D A C was reduced from 76 ppb to below 20 ppb after treatment. However, the reason for the reduction in D D A C is unknown. D D A C might be biodegraded by the biofilm as micro-organisms from activated sludge have the potential to biodegrade D D A C (Nishihara et al., 2000), or the results were affected by the time 66 elapsed during treatment and transport of sample for analysis. As mentioned in section 4.10.7, antisapstain chemicals analysis is time sensitive. Based on the results, antisapstain chemical concentrations in the samples collected were well below the 96 hr LC50 value of 537 and 67 ppb for D D A C and IPBC, respectively (Bailey et al., 1999c). The D D A C value was lower than the 200 ppb concentration where a decrease in swimming performance was observed for rainbow trout (Wood et al., 1996). The antisapstain chemical concentrations are also much lower than the regulated values in BC of 700 and 120 ppb for D D A C and IPBC, respectively (Szenasy, 1999). Therefore it was concluded that antisapstain chemicals are not a threat at the sites where run-off samples were collected. 5.4 Effects of H R T and temperature on reactor performance 5.4.1 Effects of varying HRT with constant temperature To highlight the effect of varying HRT, a single HRT trial was conducted that is separate from the factorial design discussed in section 5.4.2. For this trial, HRT of 4, 8 and 12 hours were used while keeping the temperature constant at 30°C. A l l acute toxicity was removed from the run-off sample as shown by microtox EC50 > 100% v/v. Figure 5.8 shows the percent removal of BOD, COD and tannin and lignin at different HRT. 67 -BOD removal - • — COD removal -*—T&L t 80 H R T (hr) Figure 5.8 Percent BOD, C O D and tannin and lignin removal different H R T . Both the BOD and COD removals decreased with decrease in HRT at a rate close to linear. Percent BOD removal went from an average value of 95.6% at 12 hr HRT to 77.2% at 4 hr HRT, while the percent COD removal decreased from 76.5% at 12 hr HRT to 53.6% at 4 hr HRT. There was no significant difference in tannin and lignin removal between 12 hr HRT and 8 hr HRT, however, the percent tannin and lignin removal decreased significantly from 8 hr HRT to 4 hr HRT. The percent tannin and lignin removal went from 57.4% at 8 hr HRT to 38.4% at 4hr HRT. The tannins present in run-off are made up mostly of condensed tannin, which is abundant in wood bark and are hard to biodegrade (Bhat et al., 1998). Studies done by Masbough et al. (2005) showed that tannin and lignin are resistant to biodegradation in constructed wetland with an 68 average removal of 42% over a period of seven days. The result, which showed no further improvement in tannin and lignin removal from 8 hr HRT to 12 hr HRT is comparable to results obtained by Zenaitis et al. (2002b). Zenaitis et al. (2002b) found the concentration of tannin and lignin decreased steadily in the first 8 hrs of treatment in a batch biological reactor running at 35°C, however, the tannin and lignin concentration stayed about the same afterwards until the end of the trial at 48 hrs. This result implies that the remaining tannin and lignin is probably non-biodegradable or requires very long retention times. 5.4.2 Reactor performance with varying HRT and temperature The following section discusses the results from the factorial design described in section 4.5. However, due to the low run-off strength in samples #4 and 5, synthetic run-off was used for 4 hr, 8 hr and 24 hr HRT and temperature studies. The weak run-off might mask the effects of varying operating conditions. Synthetic run-off samples provide stronger strength and similarity in run-off sample characteristics between trials to give a better comparison between them. Therefore, only results from 8 hr and 4 hr are used for the factorial design; results from 12 hr HRT are presented in Appendix B.2. Table 5.5 presents the setup for the final factorial design. 69 Table 5.5 Factorial design setup. Temperature (°C) HRT (hr) 5.0 (LI) 4 (LI) 8(L2) 17.5 (L2) 4 (LI) 8(L2) 30 (L3) 4 (LI) 8 (L2) 5.4.2.1 Characteristics of synthetic run-off Synthetic run-off was generated by the process described in section 4.6. In general, synthetic run-off had a strength similar to that of primary treated Kraft mill effluent in terms of BOD and COD. Synthetic run-off was also quite toxic as indicated by the Microtox EC50 values; table 5.6 summarizes the characteristics of synthetic run-off. Table 5.6 Characteristics of synthetic run-off. Parameter Average value BOD (mg/L) 250 (± 45) COD (mg/L) 1790 (±374) Tannin and lignin (mg/L tannic acid) 640 (± 109) Microtox 5min E C 5 0 (% v/v) 7.12 (±4.21) Microtox 15min E C 5 0 (% v/v) 5.52 (± 3.62) 70 5.4.2.2 Results of factorial design Results of factorial design are summarized in table 5.7. Table 5.7 Results of factorial design. 4 hr H R T 8 hr H R T T ( ° C ) B O D C O D T & L Microtox B O D C O D T & L Microtox removal removal removal E C 5 0 (% removal removal removal E C 5 0 (% (%) (%) (%) v/v) (%) (%) (%) v/v) 5 73.0 ± 48.0 ± 27.8 ± 23.38 86.9 ± 58.5 ± 38.4 ± 75.03 1.1 2.6 2.9 4.3 3.3 5.4 17.5 89.9 ± 58.9 ± 32.3 ± 39.22 89.7 ± 66.4 ± 47.7 ± > 100 0.4 0.6 3.6 7.9 0.3 3.6 30 90.3 ± 64.6 ± 36.1 ± > 100 94.1 ± 76.9 ± 60.1 ± > 100 2.2 0.8 3.8 0.5 0.3 2.9 In general, the reactor performance decreased with decrease in HRT and temperature. From table 5.7, the majority of the BOD was removed for 8 hr HRT at all temperatures. Substantial amounts of COD were removed for 8 hr HRT at all temperatures, however, the variation in COD removal at different temperatures is greater compared to BOD. The reactor removed 18.4% less COD at 5°C as compared to 30°C. Tannin and lignin removal also varied significantly at different temperatures with 8 hr HRT. Only 38.4% ± 5.4% tannin and lignin were removed at 5°C, while 60.1% ± 2.9% were removed at 30°C The reactor was able to completely remove the acute toxicity from the treated run-off for 71 30°C and 17.5°C with 8 hr HRT; however, some acute toxicity remained for treated run-off at 5°C and 8 hr HRT as shown by Microtox EC50 value of 75.03. For 4 hr HRT, around 90% of the BOD was removed at 30°C and 17.5°C, however, BOD removal decreased significantly to 73.0% ± 1 . 1 % at 5°C. COD removal decreased with decrease in temperature as expected. Tannin and lignin removal at all three temperatures for 4 hr HRT were much less compared to 8 hr HRT. At 4 hr HRT, temperature does not affect the removal of tannin and lignin as much since the difference in removal between 30°C and 5°C is only 8.3%. The reactor was able to remove all acute toxicity from run-off at 30°C, but treated run-off at 17.5°C and 5°C remained acutely toxic. Multiple linear regression of the following form y = ao + a iXi + a 2 X 2 + a 3 X 3 + + a;Xi (5.1) was used to fit the results. Polymath 6.0 was used to generate the following equations for estimating BOD, COD and tannin and lignin removal within the range of the operating conditions of the factorial design: BOD removal - 1.09T + 3.47HRT - 0.10T*HRT + 58.40 (5.2) COD removal = 0.596T + 2.21HRT + 0.018T*HRT + 36.64 (5.3) T & L removal = -0.204T + 1.82HRT + 0.134T*HRT + 19.04 (5.4) Where T is temperature (°C) HRT is hydraulic retention time (hr) Equations for COD and tannin and lignin removal fit the results extremely well with r values of 0.988 and 0.997 for COD removal and tannin and lignin removal, respectively. 72 The x1 value for BOD removal is 0.835. Microtox results were also fitted by non-linear regression, however, the results did not fit the model well with r 2 of 0.37, therefore, the model for Microtox is not presented. Figures comparing values predicted by equations 5.2, 5.3 and 5.4 to actual results are shown in figure 5.9 to figure 5.11. Multiple Linear Graph, Solution £14 o BOD calc • BOD expr -—^ data point Figure 5.9 Actual BOD removal compared to BOD removal estimated by equation 5.2. 73 Multiple Linear Graph, Solution #25 3.00 data point Figure 5.10 Actual C O D removal compared to C O D removal estimated by equation 5.3. Multiple Linear Graph. Solution #16 data point 5.00 Figure 5.11 Actual tannin and lignin removal compared to tannin and lignin removal estimated by equation 5.4. 74 5.4.2.3 Interactive effects of H R T and temperature To better understand the effects of HRT and temperature on reactor performance, interactive plots were generated by inputting data from the factorial design into a statistical program JMP In 4.0.3. This section discusses the results obtained. Levels shown in this section correspond to the levels in table 5.6. The interactive effects for BOD removal are shown in figure 5.12. 92.8571-BOD removal 85.7143-78.5714-T 1 7 . 5 ° C / A / 5 ° C 71.4286-92.8571-B O D removal 85.7143-bhr /-""~"~4hr HRT 78.5714- / B 71.4286-i i i 5°C 30°C i i 4hr 8hr 17.5°C T HRT Figure 5.12 Interactive effects of H R T and temperature on BOD removal. From box A, it can be concluded that HRT has a significant effect on BOD removal at 5°C as shown by the steep slope. The effect of HRT is not as apparent at temperatures of 17.5°C and 30°C as their slopes are flatter as compared to 5°C. From box B, temperature appears to have significant effect on BOD removal at 4 hr HRT over the range of 5°C to 75 17.5°C, however, the effect is minimal from 17.5°C to 30°C as shown by the flat slope. At 8 hr HRT, temperature effect is less significant as the slope is less steep. Both temperature and HRT appear to be important factors at the lower end of their ranges; when reactor is under stress. Effects of HRT and temperature on COD removal are shown in figure 5.13. COD removfi COD removal 50-T 30°C A / - ^ U . 5 ° C 8hr B / HRT HRT 5°C 30°C 4hr 17.5X 8hr Figure 5.13 Interactive effects of H R T and temperature on C O D removal. From box A , HRT has a positive effect on COD removaFat all temperatures as all three lines have similar slopes. From box B, it seems that temperature had a greater effect on COD removal between 5°C and 17.5°C and less effect between 17.5°C and 30°C for 4 hr HRT. For 8 hr HRT, temperature seems to have a linear effect on COD removal. From figure 5.13, it appears both temperature and HRT are important factors for COD removal at all operating conditions. 76 Figure 5.14 shows the results for tannin and lignin removal with varying temperature and HRT. T&L removal T&L removal 28.5714-J T A } ° ° C / J 7 . 5 ° ( B § h r -^^_3hr HRT 5°C 30°C 4hr HRT 8hr 17.5°C Figure 5.14 Interactive effects of H R T and temperature on T & L removal. From box A , HRT has a positive effect on tannin and lignin removal as shown by the positive slopes. HRT is especially important for tannin and lignin removal at higher temperature as the slope of the lines increase with increase in temperature. From box B, it can be see that temperature effect at 4 hr HRT is minimal as the slope of the line is flat, while the effect of temperature for 8 hr HRT is greater since the slope of the line is steeper. Bssed on figure 5.14, HRT is important in tannin and lignin removal, while temperature is only important at higher HRT. This implies that HRT is the limiting factor in tannin and lignin removal. Figure 5.15 shows the effect of temperature and HRT on toxicity removal. 77 5min E C 28.5714-c . , . ,414.286-5min EC50 T 30°C A / > 7 - 5 ° ( /4hr H R T H R T 5°C 30°C 4hr 8hr 17.5°C Figure 5.15 Interactive effects of H R T and temperature on treated run-off Microtox E C 5 0 values. From box B, temperature did not affect the EC50 value of treated run-off at 4 hr HRT between 5°C and 17.5°C. However, for temperature from 17.5°C to 30°C, it had a significant effect as shown by the slope of the line. For 8 hr HRT, temperature has an effect in removing toxicity between 5°C and 17.5°C, but no effect between 17.5°C to 30°C as all the acute toxicity was removed. From box A, HRT also had significant effect on toxicity removal at 5°C and 17.5°C, as shown by the slope of the lines. However, HRT had no effect on toxicity removal at 30°C when all the acute toxicity was removed from run-off during treatment. From the results, it appears that both temperature and HRT significantly affect toxicity removal at lower end of their operating ranges. 78 5.4.2.4 Statistical analysis A N O V A method was used to determine i f the results obtained from the factorial design were statistically significant. A N O V A tests were done on BOD, COD and tannin and lignin removal with 95% confidence interval. The results from A N O V A tests on BOD, COD and tannin and lignin removal are shown in table 5.8, table 5.9 and table 5.10, respectively. Table 5.8 A N O V A test on BOD removal. MS F P-value Fcritical T 85.50 3.58 0.036 3.18 HRT 174.22 7.30 0.024 5.12 Table 5.9 A N O V A test on C O D removal. MS F P-value Fcritical T 113.23 39.50 0.00051 5.05 HRT 336.70 117.44 0.00012 6.61 Table 5.10 A N O V A test on tannin and lignin removal. MS F P-value Fcritical T 97.25 6.04 0.0030 2.82 HRT 1621.68 100.80 7.1E-7 4.84 79 Based on the results from table 5.8, 5.9 and 5.10, the effect of both temperature and HRT are significant in BOD, COD and tannin and lignin removal at 95% confidence interval since all F-ratio values are greater than critical F values. 5.4.2.5 Temperature effect at all temperatures During temperature trials, samples were taken at temperatures other than the three used for factorial design and analyzed for their characteristics. Figure 5.16 shows the results of BOD, COD and tannin and lignin removal at different temperatures for 4 hr and 8 hr HRT. -»-4hrBOD - Q - 8 h r B O D -»-4hrCOD - a - 8 h r C O D - " -4h rT& L -p -8hrT&L| — — 1 0 0 30 25 20 15 10 5 T ( C ) Figure 5.16 Comparison between temperature trials at 8 hr and 4 hr H R T 80 BOD, COD and tannin and lignin removal stayed relatively consistent for temperatures above 20°C (figure 5.13). The decrease in BOD, COD and tannin and lignin removal starting at temperatures below 17.5°C is most likely due to the fact that mesophilic bacteria were operating outside their optimum temperature range of 20°C to 40°C. BOD, COD and tannin and lignin removal all reached their lowest point between 10°C to 15°C, then their removal started to improve again below those temperatures. This is most likely caused by a transition between a reactor microbial community dominated by mesophilic bacteria to one in which psychrophilic bacteria dominate. The time lag before recovery is likely due to the slow growth of psychrophilic bacteria. Microtox EC50 results of 4 hr and 8 hr HRT at different temperatures are shown in figure 5.17. 30 25 20 15 10 5 T(C) Figure 5.17 Microtox E C 5 0 values at 4 hr and 8 hr H R T with varying temperature. 81 At an 8 hr HRT, complete toxicity removal was achieved as the reactor operating temperature decreased from 30°C to 17.5°C. It was only below 15°C that the treated run-off remained acutely toxic. Toxicity removal was lowest at 10°C with 5 min EC50 of 42.5% v/v. Toxicity removal improved at 5°C with 5 min EC50 value of 75.0% v/v. For a 4 hr HRT, the reactor was able to remove all acute toxicity at 30°C. However, toxicity removal started to decline at lower temperatures. Treated run-off samples at 21°C had a microtox 5 min EC50 of 54.7% v/v. Toxicity removal continued to decline and reached a low point at 10°C with E C 5 0 value of 19.6% v/v. Treated run-off at 5°C had 5 min E C 5 0 of 23.4% v/v. The decrease in toxicity removal correlates well with the observed reduction in BOD, COD and tannin and lignin removal. The transition between mesophilic bacteria and psychrophilic bacteria was also apparent for toxicity removal at 8 hr HRT as the EC50 value of run-off treated at 5°C is much higher than that of run-off treated at 10°C. However, for 4 hr HRT, there does not seem to be any significant change in toxicity removal between 10°C and 5°C. This implies that toxicity removal at low temperatures is possibility limited by HRT. Based on the results of this trial, the bioreactor is capable of treating run-off at temperatures from 30°C to below 10°C if enough time is given for the bacteria to adjust to the temperature change and sufficient HRT is given. If the bioreactor is to be applied in real log yards, bacteria should have sufficient time to adjust to temperature changes since temperature changes in nature occur slowly. 82 5.4.2.6 Temperature trial at 24 hr H R T In addition to the 4 hr, 8 hr and 12 hr HRT trials, a 24 hr HRT trial was done to compare with results from past studies. Samples were only collected and analyzed at 30°C, 15°C and 5°C for 24 hr HRT for comparison purposes. Table 5.11 summarizes the results. Table 5.11 Reactor performance at 24 hr H R T with varying temperatures. Parameters 5°C 15°C 30°C BOD removal (%) 95.0 ±3 .6 97.6± 1.1 97.3 ± 1.1 COD removal (%) 64.6 ± 1.1 62.0 ± 1.9 69.6 ± 0.4 T & L removal (%) 48.8 ±5.5 53.7 ±2.5 54.1 ±2 .5 Microtox EC50 (%v/v) 100 100 100 Based on the results, high BOD removals were achieved for all three temperatures with maximum removal of over 97%. COD removal was highest at 30°C with 69.6% ± 0.4% removal. COD removal at 15°C was lower than 5°C; this could have been caused by the low activity of the mesophilic bacteria at that temperature. There is no significant difference between tannin and lignin removal at all three temperatures sampled; this implies that most of the biodegradable tannin and lignin that is degraded at 24 hr HRT, hence there is no significant difference between temperatures. At all three temperatures, the reactor was able to completely remove acute toxicity. 83 The results from the 24 hr HRT trials were compared with results obtained by Woodhouse (2003) and shown in table 5.12. Table 5.12 Comparing results of current study to past study. Temperature (°C) Reactor performance for current study Woodhouse1 BOD (%) COD (%) T & L (%) BOD (%) COD (%) T & L (%) 30 (342) 97.3 ± 1.1 69.6 ± 0.4 54.1 ±2.5 94 ± 6.4 91 ±0 .7 96 ± 0.5 5 i 95.0 ±3 .6 64.6 ± 1.1 48.8 ±5.5 76 ±5 .6 64 ±3 .3 67 ± 0.2 (Woodhouse, 2003) 2 Woodhouse did her study at 34°C In general, the BOD removals from the two studies were similar at high temperature, but Woodhouse (2003) also achieved much higher COD and tannin and lignin removal at high temperature. At the low end of the temperature range, BOD removal from the current investigation was significantly higher, and the COD removals between the two trials are comparable. Tannin and lignin removal observed in the current study was considerably less. However, direct comparison between two different studies should be done with caution. There are many factors that could affect the outcome of the comparison such as: the amount of biomass present inside the reactor, and the type of logs processed at the log yards. The Courtenay mill processed hemlock, while site B processed Western Red 84 Cedar. Age of the run-off is also an important factor; fresh run-off samples contain more biodegradable material compared to old run-off (Tao et al., 2005). 5.4.2.7 Overall summary of H R T and temperature trials Based on the results from all the temperature and HRT trials, in general removal efficiency increases with an increase in HRT and temperature. A l l acute toxicity was removed at 30°C regardless of HRT, however, at 17.5°C and 5°C, the treated run-off from 4 hr and 8 hr HRT remained acutely toxic. Therefore, i f attached growth bioreactor is to be applied in log yards, longer retention time would be required during winter months to remove the acute toxicity. Alternatively, another treatment method like ozone treatment should be installed in sequence (Zenaitis et al., 2002b). 5.5 Estimating amount of biomass present in reactor Three sets of support material, each consisting of five pieces, were withdrawn from the reactor and analyzed according to the method described in section 4.9. The weight of the biomass attached to a single piece of support material was multiplied by the total number pieces of support material within the reactor to come up with an estimation of the amount of biomass present within the reactor. The total number of pieces of support material within the reactor was 114. Table 5.13 summarizes the results. 85 Table 5.13 Biomass present within reactor. Biomass weight (mg) Weight of biomass attached to a single piece of support material 133.3 ±18.6 Estimation of total biomass within reactor 15200 ±2125 The volume of support material within the reactor is approximately 1 L , and therefore the biomass concentration inside the reactor is 15,200 ±2125 mg/L. 5.6 Degradation of run-off in batch trials A batch degradation test was done to determine the amount of biodegradation that goes on inside the reactor and holding tank. The method for the batch test is outlined in section 4.7. 5.6.1 Degradation within reactor Biomass from the reactor was submerged into untreated synthetic run-off to form a suspension with TSS concentration of 6,090 mg/L, which was roughly 40% of the biomass concentration within the reactor. Figure 5.18 and figure 5.19 show the BOD and COD profile over 24 hours. 86 200 time (hr) Figure 5.18 B O D profile with 1 s t order fit over 24 hr batch test with reactor biomass at 30°C. 87 1800 900 A 800 •] 1 , 1 1 , 1 0 4 8 12 16 20 24 time (hr) Figure 5.19 C O D profile with 1 s t order fit over 24 hr batch test with reactor biomass at 30°C. After 24 hours of batch treatment, 93.5% of the BOD was removed; however only 37.0% of COD was removed. For BOD removal,'55.1% was removed after the first two hours, 56.0% after four hours and 70% after six hours. The low COD removal suggests that the batch test was only able to remove the biodegradable material. The BOD and COD profile were modeled as a first order reaction of the following form: = kC (5.4) C = C0e~k' (5.5) 88 Where C is concentration of BOD or COD t is treatment time k is first order rate constant BOD and COD data were fitted to equation 5.5 with nonlinear regression in Polymath 6.0 The first order rate constant k for BOD and COD were 0.243 hr"1 and 0.0247 hr"1, respectively. The r 2 for BOD and COD removal were 0.913 and 0.523, respectively. To give an estimate of the rate of BOD and COD degradation within the reactor, the two rate constants were multiplied by a factor of 2.5, since the biomass concentration in the batch test was about 40% of biomass concentration in the reactor. The estimated rate constants for the reactor were 0.608 hr"1 and 0.0618 hr"1 for BOD and COD removal, respectively. However, these results might not represent the real reactor well because undiluted run-off samples were used for the batch test, while the run-off feed to the reactor would be much more dilute and it is also unlikely that all the biomass inside the reactor will be in contact with run-off all at once. Therefore, the actual rate of degradation inside the reactor would most likely be lower compared to the estimated rate of degradation from the batch trials. 5.6.2 Degradation within holding tank Batch tests were also done with biomass from the holding tank to determine i f degradation within the holding tank contributes to the overall treatment performance of the reactor system. Holding tank degradation trials were done at 5°C, 15°C and 30°C to give a better understanding of the degradation that goes on inside the holding tank. The average biomass concentration for. the batch tests was 641 mg/L, which is the same as 89 inside the holding tank. The BOD and COD profiles for 5°C, 15°C batch trials are shown in figure 5.20 and figure 5.21. The BOD and COD profiles for 30°C trials are shown in figure 5.22 and figure 5.23. | - « -BOD profile - * - C O D profile! 8 12 16 20 24 time (hr) Figure 5.20 BOD and C O D profile over 24 hr batch trial with holding tank biomass at 5°C 90 | - » - B O D profile - » - C O D profile | time (hr) Figure 5.21 BOD and C O D profile over 24 hr batch trial with holding tank biomass at 15°C. 200 Figure 5.22 BOD profile with 1 s t order fit over 24 hr batch trial with holding tank biomass at 30°C. 91 Q 400 • O 200 -0 J , , , _ . _, , 0 4 8 12 16 20 time (hr) Figure 5.23 C O D profile with 1 s t order fit over 24 hr batch trial with holding tank biomass at 30°C. There was no significant removal of BOD and COD within the holding tank for 5°C and 15°C. The amount of BOD and COD removed was actually higher at 5° than 15°C. This was possibly caused by the low activity of the mesophilic bacteria at 15°C. The BOD and COD removal at 30°C were surprisingly high with 90.6% and 54.9%, respectively. The BOD and COD data for 30°C were fitted as a first order reaction and compared to rate constants obtained for the reactor. The rate constants were found to be 0.083 hr"1 and 0.0327 hr"1 for BOD and COD removal, respectively. The r 2 values were 0.985 and 0.997 for BOD and COD removal, respectively. However, the results from these batch tests might not represent the actual degradation within the holding tank since undiluted run-off was used in the batch test while the run-off within the holding tank will be much more 92 dilute. These results suggest that at lower temperatures, degradation within the holding tank most likely did not affect the overall reactor treatment performance much. At 30°C, the holding tank might improve the overall treatment performance of the reactor system; however, from the rate constants obtained, the BOD removal rate for the reactor is 7.3 times greater than the holding tank and the COD removal rate for the reactor is 1.89 times greater than the holding tank. As previously discussed in section 5.6.1, the rate constants estimated for the reactor may be higher than the actual values; therefore, the actual ratio of degradation in the reactor to degradation in the holding tank may be lower, but it is unlikely that the actual degradation rate in the reactor would be lower than the degradation rate in the holding tank due to the difference in biomass concentration. Therefore, most of the run-off degradation likely occurred inside the reactor. 5.7 Respirometry studies Two respirometry tests were done at the end of the treatment phase according to methods described section 4.8. Respirometry data were fitted with Monod and Tessier kinetics using nonlinear regression with Polymath 6.0. Results obtained are summarized in table 5.14. 93 Table 5.14 Run-off kinetic data. Monod Tessier qmax (mg BOD/mg VSS*min) K s (mg/L) qmax (mg BOD/mg VSS*min) k(mg/L) 0.00029 2.50 0.00022 0.42 The maximum substrate uptake rate for both Monod and Tessier kinetics were much smaller compared to those found by Zenaitis et al. (2002b), which implies the current run-off samples are less readily biodegradable. Zenaitis et al. (2002b) found qm ax of 0.0038 mg BOD/mg VSS min and K s of 1.4 mg/L for Monod kinetics and q m a x of 0.0033 mg BOD/mg VSS min and k of 0.48 mg/L for Tessier kinetics. Some possible explanations for the difference in substrate uptake rate include: type of log processed at the log yard, strength of the run-off and the age of the run-off. Western Red Cedar was processed at site B, the site for which run-off was chosen for respirometry studies, while for the previous study hemlock and Douglas fir were used. The BOD of run-off sample used for the respirometry test in the current study was 180 mg/L, while the BOD for the samples Zenaitis et al. (2002b) used was 1540 mg/L. The higher BOD implies there is more biodegradable material within the run-off. As mentioned in section 5.4.2.5, the amount of biodegradable material decreases with an increase in run-off sample storage time. The run-off sample used for the respirometry test was in storage for over one year. Old run-off sample was used to carry out the respirometry tests because it was thought that they best represent the run-off samples that were used during the treatment phase. 94 6 C o n c l u s i o n s Five run-off samples were collected from two log yards in BC between September 2004 and July 2005 for this project. The run-off samples were first characterized with standard wastewater parameters. A biological attached growth reactor was used to treat the run-off samples. The performance of the reactor was evaluated using standard wastewater parameters and by monitoring removal of metal ions and antisapstain chemicals. The effects of varying HRT and temperature were studied using a factorial experiment. Finally the amount of biomass present within the reactor was estimated. 6.1 Run-off characteristics Run-off sample colours ranged from light brown to dark brown with a distinctive woody odour. The five run-off samples collected varied significantly in strength. The BOD ranged from 16 to 371 mg/L, COD from 230 to 2660 mg/L, tannin and lignin from 200 to 680 mg/L of tannic acid. Based on Microtox toxicity, four run-off samples were toxic, with the lowest EC50 value of 9.0% v/v. Run-off sample pH ranged from 3.0 to 6.3. 6.2 Attached growth in a lab-scale reactor RAS was initially mixed with primary treated Kraft mill effluent inside the reactor; then Kraft mill effluent was continuously fed to the reactor for six weeks. Micro-organisms rapidly colonized the support material in the reactor. 95 Several upsets occurred during the startup of the reactor, as well as during the treatment phase which hindered reactor performance. The micro-organisms were able to recover from the upsets and return to normal after a few days. 6.3 Run-off treatment phase 1. The initial run-off treatability trial lasted for a period of 6 months at 12 hr HRT and 30°C. Average BOD and COD removal were 88.1% and 61.3%, respectively. 2. The reactor was able to reduce the concentration of the antisapstain chemical D D A C in one run-off sample. A l l four samples analyzed showed a reduction in zinc concentration. 3. The reactor's ability to treat run-off is greatly affected by HRT. From 12 hr HRT to 4 hr HRT, BOD removal went from 95.6% to 77.2%, COD removal decreased from 76.5%) to 53.6%, tannin and lignin removal went from 56.9% to 38.4%. 4. Reactor performance is also affected by the operating temperature. Treatment efficiency decreased significantly with decreasing temperature. At 8 hr HRT, BOD removal were 94.1% and 86.9%, COD removal were 76.9% and 58.5%, and tannin and lignin removal were 60.1% and 38.4%, respectively for 30°C and 5°C. At 4 hr HRT, BOD removal were 90.3% and 73.0%, COD removal were 64.6% and 48.0%), and tannin and lignin removal were 36.1% and 27.8%, respectively for 30°C and 5°C. Treated run-off samples at low temperatures and short HRT remained acutely toxic. The microbial community in the reactor seemed to go through a transitional phase as temperature was decreased from 15°C to 10°C. It 96 was hypothesized that this coincided with a transition from a microbial culture dominated by mesophilic bacteria to one dominated by psychrophilic bacteria. 6.4 Biomass present in reactor Total concentration of biomass present in reactor was estimated to be 15200 mg/L. 6.5 Batch test for run-off degradation Batch tests were done at 30°C for 24 hours with a biomass concentration of 6090 mg/L. After 24 hours incubation, BOD and COD removal were 93.5% and 37.0%, respectively. 6.6 Run-off kinetic values Respirometry data were fitted using Monod kinetics and Tessier kinetics. For Monod kinetics, qm ax was 0.00029 mg BOD/mg VSS min and K s was 2.50 mg/L. For Tessier kinetics, q m a x was 0.00022 mg BOD/mg VSS min and k was 0.42 mg/L. These values were significantly lower than observed previously in literature. 97 7 Recommendations for Future Work 7.1 Run-off characterization Any future run-off treatment should continue to characterize run-off. Nutrient levels within the run-off sample should be analyzed to determine i f sufficient amounts of nitrogen and phosphorous are present in run-off. This may eliminate the need for nutrient addition before run-off treatment. Total organic carbon (TOC) can also be added to the run-off characterization assays. TOC assays provide a better idea of the amount of organic material present in run-off samples. Respirometry tests should also be done on fresh run-off to determine the biodegradability of the run-off and to determine the portion that is readily biodegradable. 7.2 Antisapstain chemical degradation More samples should be analyzed to get a better understanding of whether the reactor is capable of treating antisapstain chemicals. To get the best results, run-off samples collected need to be treated right away, then the treated samples along with the untreated samples should be analyzed immediately, since antisapstain chemical analysis is time sensitive. To determine i f the reactor is capable of degrading antisapstain chemicals, batch tests should be done with micro-organisms from the reactor with known concentrations of D D A C and IPBC solution. 98 7.3 Zinc removal To determine i f zinc removal within the reactor is due to metabolism by biomass or adsorption onto support material; support material along with biomass attached to the support material should be analyzed for zinc concentration. 7.4 Transition of microbial culture More studies should be done on the reactor operating between 15°C and 10°C, where there is an apparent change of microbial culture that dominates within the reactor. The transitional phase should be monitored closely to determine the time required for switching from a stable reactor dominated by mesophilic bacteria to a stable reactor dominated by psychrophilic bacteria. Methods for measuring substrate uptake rate, such as respirometry, could be used to monitor the activity of the biomass present. 7.5 Anaerobic treatment The potential of anaerobic treatment of run-off should be tested. Published literature has shown anaerobic treatment to be effective in treating industrial and municipal wastewater. The results can be compared to aerobic treatment to determine which treatment method is better suited for treating run-off. 99 7.6 Simulating real conditions Before a biological reactor can be applied in real log yards, it is important to determine i f it can function under real life conditions. An important factor to determine is i f the reactor can function after long periods of not getting any substrate. In BC, rainfall events usually do not occur during summer months, therefore, it is important for the reactor to stay active during that period and be ready to treat run-off when the rainfall season starts. In a lab trial, this could be mimicked by discontinuing the reactor feed for two months, then start feeding it with fresh run-off. After startup, the reactor should be monitored closely to see i f it is capable of treating the run-off and the time it requires to recover. 100 R e f e r e n c e s Alberta Forest Products Association (AFPA 1999) Characterization of Surface Water Run-off from Log Yard Sites in Alberta American Public Health Association (APHA, 1992) Greenberg, A.E. , Clesceri, L.S., Eaton, A .D . Standard Methods for the Examination of Water and Wastewater 18 t h edition. American Water Works Association, Water Environment Federation, Washington, DC Annachhatre, A.P., Bhamidimarri, S.M.R. (1992) Microbial Attachment and Growth in Fixed-Film Reactors: Process Startup Considerations. Biotechnology Advances 10,69-91 Ard, T.A., McDonough, T.J. (1996) Toxicity Assays in the Pulp and Paper Industry - A Review and Analysis. TAPPI International Environmental Conference & Exhibit Proceedings, 901-913 Bailey, H.C., Elphick, J.R., Potter, A. , Chao, E., Konasewich, D., Zak, B. (1999a) Causes of Toxicity in Stormwater Runoff from Sawmills. Environmental Toxicology and Chemistry 18(7), 1485-1491 Bailey, H.C., Elphick, J.R, Potter, A . , Zak, B. (1999b) Zinc Toxicity in Stormwater Runoff from Sawmills in British Columbia. Water Research 33(11), 2721-2725 Bailey, H.C., Elphick, J.R., Potter, A. , Chao, E., Zak, B. (1999c) Acute Toxicity of the Antisapstain Chemicals D D A C and IPBC, Alone and in Combination to Rainbow Trout. Water Research 33(10), 2410-2414 Bhat, T., Singh, B., Sharma, O. (1998) Microbial Degradation of Tannins - A Current Perspective. Biodegradation 9, 343-357 Bodik, I., Kratochvil, K. , Herdova, B., Tapia, G., Gasparikova, E. (2002) Municipal Wastewater Treatment in the Anaerobic-Aerobic Baffled Filter Reactor at Ambient Temperature. Water Science and Technology 46(8), 127-135 Breitmaier, E. (2006) Terpenes. W I L E Y - V C H Verlag GmbH & Co: KGaA, Weinheim British Columbia Ministry of Environment, Lands and Parks (2001) Environmental Audit Report for Dryland Log Sorting Operations in the Vancouver Island Region 101 Buchanan, D.V., Tate, P.S., Moring, J.R. (1976) Acute Toxicities and Hemlock Bark Extracts to some Estuarine Organisms in Southeastern Alaska. Journal of Fisheries Research Board of Canada 33, 1188-1192 Burgess, J., Quarmby, J., Stephenson, T. (1999a) Role of Micronutrients in Activated Sludge-based Biotreatment of Industrial Effluents. Biotechnology Advances 17, 49-70 Burgess, J., Quarmby, J., Stephenson, T. (1999b) Micronutrient Supplements for Optimisation of the Treatment of Industrial Wastewater Using Activated Sludge. Water Research 33(18), 3707-3714 Cech, J., Chudoba, J., Grau, P. (1985) Determination of Kinetic Constants of Activated Sludge Microorganisms. Water Science and Technology 17, 259-272 Codina, J . C , Perez-Garcia, A. , Romero, P., de Vicente, A. (1993) A Comparison of Microbial Bioassays for the Detection of Metal Toxicity. Archives of Environmental Contamination and Toxicology 25, 250-254 Doig, P. (2005) Characterization of Forestry Dry Land Sort Runoff and Treatment Using a Sand Filtration Process. M.Sc. thesis, The University of British Columbia, Vancouver, BC. Environment Canada. Daily Data Report, http://www.climate.weatheroffice.ec.gc.ca/climateData/dailydata_e.html7timefra me=2&Prov=XX&StationID=889&Year=2004&Month=9&Day=7, last accessed September 8, 2006 Environmental Management Act (2004) Antisapstain Chemical Waste Control Regulation, http://www.qp.gov.be.ca/statreg/reg/E/EnvMgmt/300_90.htm, British Columbia Ministry of Environment Farrell, A.P., Stockner, E., Kennedy, C.J. (1998) A Study of the Lethal and Sublethal Toxicity of Polyphase P-100, an Antisapstain Fungicide Containing 3-Iodo-2-Propynyl Butyl Carbamate (IPBC), on Fish and Aquatic Invertebrates. Archives of Environmental Contamination and Toxicology 35, 472-478 Fikart, A . (2002) A Comparative Assessment of Stormwater runoff From a Coastal and Interior Log Yard. M.Sc. thesis, The University of British Columbia, Vancouver, BC. Firth, B.K. , Backman, C.J. (1990) Comparison of Microtox Testing with Rainbow Trout (Acute) and Ceriodaphnia (Chronic) Bioassays in M i l l Wastewaters. TAPPI Journal 1?>(\2\ 169-174 102 Forgie, D.J.L. (1988) Selection of the Most Appropriate Leachate Treatment Mothods Part 1: A Review of Potential Biological Leachate Treatment Methods. Water Pollution Research Journal of Canada 23(2), 308-328 Gavrilescu, M . , Macoveanu, M . (2000) Attached-growth Process Engineering in Wastewater Treatment. Bioprocess Engineering 23, 95-106 Golding, S. (2004) Stormwater Quality Survey of Western Washington Log Yards. Washington Depart of Ecology Greater Vancouver Regional District (2004) Quality Control Annual Report for Greater Vancouver Sewage & Drainage District Jamieson, S. (1992) In Environmental Monitoring, What You Don't Know Can Hurt You. Pulp&Paper Canada 93(2), 9-10 Kalyuzhnyi, S.V., Gladchenko, M . A . (2004) Sequenced Anaerobic-Aerobic Treatment of High Strength, Strong Nitrogenous Landfill Leachates. Water Science and Technology 49(5-6), 301-312 Langwaldt, J.H., Puhakka, J.A. (2000) On-site Biological Remediation of Contaminated Groundwater: A Review. Environmental Pollution 107, 187-197 Lazarova, V. , Manem, J. (1996) A n Innovative Process for Waste Water Treatment: The Circulating Floating Bed Reactor. Water Science and Technology 34(9), 89-99 Magnus, E., Carlberg, G., Hoel, H. (2000a) TMP Wastewater Treatment, Including a Biological High-efficiency Compact Reactor: Removal and Characterisation of Organic Compounds. Nordic Pulp and Paper Research Journal 15, 29-36 Magnus, E., Carlberg, G., Hoel, H. (2000b) TMP Wastewater Treatment, Including a Biological High-efficiency Compact Reactor: Toxicity Reduction and Removal of Extractives. Nordic Pulp and Paper Research Journal 15, 37-45 Masbough, A. , Frankowski, K., Hall, K. , Duff, S. (2005) The Effectiveness of Constructed Wetland for Treatment of Woodwaste Leachate. Ecological Engineering 25, 552-566 McDougall, S. (1996) Assessment of Log Yard Runoff in Alberta. Preliminary Evaluation. Alberta Environment McDougall, S. (2002) Assessment of Log Yard Runoff in Alberta. Results of Monitoring Program 1996 - 1998. Alberta Environment 103 Microbics Corporation (1992) Microtox Manual. A Toxicity Testing Handbook, Volume IL Detailed Protocols, Carlsbad, C A Munkittrick, K.R., Power, E.A., Sergy, G.A. (1991) The Relative Sensitivity of Microtox, Daphnid, Rainbow Trout, and Fathead Minnow Acute Lethality Tests. Environmental Toxicology and Water Quality: An International Journal 6 , 35-62 National Council for Air and Stream Improvement, Inc. (1992) A Literature Review and Case Study. Technical Bulletin No.637 National Council for Air and Stream Improvement, Inc. (2005) Characterization, Toxicology, and Management and Treatment Options for Wood Pile Leachates and Runoff: Part I - Literature Review. Technical Bulletin No.911 Nishihara, T., Okamoto, T., Nishiyama, N . (2000) Biodegradation of Didecyldimethylammonium Chloride by Pseudomonas Fluorescens TN4 Isolated from Activated Sludge. Journal of Applied Microbiology 88, 641-647 O'Connor, B., Kovacs, T., Voss, R. (1992) The Effect of Wood Species Composition on the Toxicity of Simulated Mechanical Pulping Effluents. Environmental Toxicology and Chemistry 11, 1259-1270 Oikari, A. , Anas, E., Kruzynski, G., Holmbom, B. (1984) Free and Conjugated Resin Acids in the Bile of Rainbow Trout, Salmo gairdneri. Bulletin of Environmental Contamination and Toxicology 33, 233-240 Oikari, A . , Nakari, T. (1982) Kraft Pulp M i l l Effluent Components Cause Liver Dysfunction in Trout. Bulletin of Environmental Contamination and Toxicology 28, 266-270 Orban, J.L. (2000) The Extent, Causes and Environmental Risk of Log yard/Dryland Sort Run-off in British Columbia, M.Sc. thesis, The University of British Columbia, Vancouver, BC. Parker, D. (1999) Trickling Filter Mythology. Journal of Environmental Engineering 125(7), 618-625 Peters, G.B., Dawson, H.J., Hrutfiord, B.F., Whitney, R.R. (1976) Aqueous Leachate from Western Red Cedar: Effects on Some Aquatic Organisms. Journal of the Fisheries Research Board of Canada 33, 2703-2709 Qureshi, N . , Annous, B., Ezeji, T., Karcher, P., Maddox, I. (2005) Biofilm Reactors for Industrial Bioconversion Processes: Employing Potential of Enhanced Reaction Rates. Microbial Cell Factories 4(24) 104 Rusten, B., Odegaard, H. , Lundar, A . (1992) Treatment of Dairy Wastewater in a Novel Moving Bed Biofilm Reactor. Water Science and Technology 26(3-4), 703-711 Samis, S .C, Liu, S.D., Wernick, B.G. , Nassichuk, M.D. (1999) Mitigation of Fisheries Impacts from the Use and Disposal of Wood Residue in British Columbia and the Yukon. Canadian Technical Report of Fisheries and Aquatic Sciences 2296 Schermer, E.D., Phipps, J.B. (1976) A Study of Woodwaste Leachate. Washington State Department of Ecology Singer, J.T., Rice, R.W., Zibilske, L . M . , Helyar, L. (1997) Investigation of Acute Toxicity of Distillates from Five Species of Wood for Fathead Minnows. Forest Products Journal 47(3), 96-99 Szenasy, E. (1999) Assessing the Potential Impact of the Antisapstain Chemicals, D D A C and IPBC, in the Fraser River. http://www.rem.sfu.ca/FRAP/9807.pdf. Environment Canada Tao, W., Hall, K. , Masbough, A. , Frankowski, K. , Duff, S. (2005) Characterization of Leachate from a Woodwaste Pile. Water Quality Research Journal of Canada 40(4), 476-483 Tana, J. (1988) Sublethal Effects of Chlorinated Phenols and Resin Acids on Rainbow Trout (Salmo Gairdneri). Water Science and Technology 20(2), 77-85 Tatarazako, N . , Yamamoto, K. , Iwasaki, K. (2002) Subacute Toxicity of Wood Preservatives, D D A C and B A A C , in Several Aquatic Organisms. Journal of Health Science 48(4), 359-365 Taylor, B.R. (1994) Toxicity of Aspen Wood Leachate to Aquatic Life. Part II: Field Study. Ministry of Environment, Lands and Parks of British Columbia Temmink, J., Field, J., van Haastrecht, J., Merkelbach, R. (1989) Acute and Sub-Acute Toxicity of Bark Tannins in Carp (Cyprinus carpio L.). Water Research 23(3), 341-344 Toussaint, M.W., Shedd, T.R., van der Schalies, W.H. (1995) A Comparison of Standard Acute Toxicity Tests with Rapid-Screening Toxicity Tests. Environmental Toxicology and Chemistry 14(5), 907-915 United Nations (2003) Waste-water Treatment Technologies: A General Review. http://www.escwa.org.lb/information/publications/edit/upload/sdpd-03-6.pdf Wood, A. , Johnston, B., Farrell, P., Kennedy, C. (1996) Effects of Didecyldimethylammonium Chloride (DDAC) on the Swimming Performance, Gi l l Morphology, Disease Resistance, and Biochemistry of Rainbow Trout 105 (Oncorhynchus myskiss). Canadian Journal of Fisheries and Aquatic Sciences 5 3 , 2424-2432 Woodhouse, C. (2003) Attached Growth Biological Treatment of Stormwater Run-off from Log Yards. M.A.Sc thesis, The University of British Columbia, Vancouver, BC. Zenaitis, M . , Duff, S J .B. (2002a) Ozone for Removal of Acute Toxicity from Log yard Run-off. Ozone Science & Engineering 24, 83-90 Zenaitis, M . , Franko wski, K. , Hall, K. , Duff, S. (1999) Treatment of Run-off and Leachate from Wood Processing Operations. The Sustainable Forest Management Network Conference, Science and Practice: Sustaining the Boreal Forest, Edmonton, Alberta Zenaitis, M . , Harinder, S., Duff, S. (2002b) Combined Biological and Ozone Treatment of Log Yard Run-off. Water Research 3 6 , 2053-2061 106 A p p e n d i x A S u p p l e m e n t a r y r a i n f a l l d a t a Table A . l Daily Rainfall Data Date Total rainfall (mm)1 Sept 17, 2004 4.6 Sept 18,2004 91.6 Sept 19, 2004 3.4 Sept 20, 2004 0.0 Sept 21, 2004 0.4 Sept 22,2004 5.2 Sept 23, 2004 0.4 Oct 28, 2004 2.4 Oct 29, 2004 5.4 Oct 30, 2004 0.0 Oct 31, 2004 0.4 Nov 1,2004 35.0 Nov 2, 2004 17.2 Nov 3, 2004 0.0 Jan 14, 2005 0.0 Jan 15,2005 4.4 Jan 16, 2005 14.4 Jan 17, 2005 56.6 Jan 18, 2005 44.2 Jan 19, 2005 28.4 Jan 20, 2005 18.0 May 16, 2005 2.4 May 17, 2005 0.0 May 18, 2005 9.2 May 19, 2005 8.6 May 20, 2005 1.0 May 21, 2005 3.8 May 22, 2005 7.8 June 30, 2005 3.8 July 1, 2005 0.0 July 2, 2005 0.2 July 3, 2005 0.0 July 4, 2005 0.2 July 5, 2005 19.6 July 6, 2005 3.0 ' ( E n v i r o n m e n t C a n a d a ) 107 A p p e n d i x B S u p p l e m e n t a r y e x p e r i m e n t a l d a t a B.l Heavy metal removal during treatment of run-off The removal of other heavy metal ions excluding zinc during treatment is shown in figure B. 1 through figure B.9. The concentration of most metal ions decreased during treatment. 0.3 0.25 4 1 2 3 4 batch sample number Figure B . l Aluminum concentration of treated and untreated run-off 108 £ 8 c o +3 CC 6 c <D U C 4 o o •untreated ntreated ba t ch s a m p l e n u m b e r Figure B.2 Calcium concentration of treated and untreated run-off 0.03 H < 0.025 E c o re u c o o 0.02 1 0.005 •untreated Dtreated ba t ch s a m p l e n u m b e r Figure B . 3 Copper concentration of treated and untreated run-off 109 1 2 3 4 batch sample number Figure B.4 Potassium concentration of treated and untreated run-off batch sample number Figure B.5 Magnesium concentration of treated and untreated run-off 110 2.5 1 2 3 4 batch sample number Figure B.6 Manganese concentration of treated and untreated run-off 70 1 2 3 4 batch sample number Figure B.7 Sodium concentration of treated and untreated run-off 111 1 2 3 batch sample number Figure B.8 Nickel concentration of treated and untreated run-off 1 2 3 4 batch sample number Figure B.9 Phosphorous concentration of treated and untreated run-off 112 B.2 Temperature trial at 12 hr HRT The results for the temperature trial at 12 hr HRT showed a similar trend compared to the temperature trials at 8 hr and 4 hr HRT with a transition period between 15°C and 10°C. Results are shown in figure B. 10. T (degree C) 20 10 60 > o - • - B O D E - • - C O D 50 5 S Figure B.10 Temperature trial at 12 hr H R T 113 Appendix C Reactor startup phase The BOD and COD removal from primary treated Kraft mill effluent during reactor startup are shown in figure C l and figure C.2. 08-Aug-04 18-Aug-04 28-Aug-04 07-Sep-04 17-Sep-04 time Figure C l Percent BOD removal of primary treated Kraft effluent during reactor startup 114 60 50 time Figure C.2 Percent C O D removal of primary treated Kraft effluent during reactor startup Appendix D Synthetic run-off generation This section shows one of the trials conducted for analyzing characteristics of synthetic run-off. Figure D . l and D.2 show the BOD and COD with time after initial insertion of wood chips into run-off sample. time (hr) F i g u r e D . l B O D prof i le over 24 h r after in i t ia l insert ion o f w o o d chips into r u n -of f sample 1 1 6 1600 1400 1200 -J- 1000 _ 800 Q O 400 -200 • 0 -I 1 1 1 , 1 0 4 8 12 16 20 24 time (hr) Figure D.2 C O D profile over 24 hr after initial insertion of wood chips into run-off sample 117 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0058949/manifest

Comment

Related Items