UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

RBC treatment of a municipal landfill leachate : a pilot scale evaluation Peddie, Craig Cameron 1986

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

Item Metadata


831-UBC_1986_A7 P32.pdf [ 20.75MB ]
JSON: 831-1.0062727.json
JSON-LD: 831-1.0062727-ld.json
RDF/XML (Pretty): 831-1.0062727-rdf.xml
RDF/JSON: 831-1.0062727-rdf.json
Turtle: 831-1.0062727-turtle.txt
N-Triples: 831-1.0062727-rdf-ntriples.txt
Original Record: 831-1.0062727-source.json
Full Text

Full Text

RBC TREATMENT O F A MUNICIPAL  LANDFILL LEACHATE: A  PILOT SCALE  EVALUATION  by CRAIG C A M E R O N  A THESIS SUBMITTED  PEDDIE  IN PARTIAL FULFILMENT  THE REQUIREMENTS  FOR THE DECREE  MASTER O F APPLIED  OF  SCIENCE  in  THE FACULTY OF GRADUATE STUDIES Department of Civil Engineering  W e accept this thesis as conforming to the required standard  THE UNIVERSITY  OF BRITISH  COLUMBIA  October 14, 1986  ®  Craig Cameron Peddie, 1986  OF  In presenting this thesis in partial fulfilment of  the  requirements  degree at the THE UNIVERSITY OF BRITISH COLUMBIA,  for an advanced  I agree that the  Library  shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of  my Department  or by  his or her representatives.  It  is understood  that  copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Department of Civil Engineering  THE UNIVERSITY OF BRITISH COLUMBIA 2075 Wesbrook Place Vancouver, Canada V6T 1W5  Date: October 14, 1986  ABSTRACT  This study evaluated the on-site treatment of a moderately low strength municipal landfill leachate with a Rotating Biological Contactor (RBC), at pilot scale (0.9  m dia.). The leachate generally had C O D and NH^-N concentrations of less  than 1000 mg/L and 50 mg/L respectively. A high treatment efficiency for both carbon removal and nitrification was achieved despite variable and intermittent loading conditions. The effluent filtrable BOD^ was generally less than 10 mg/L and the effluent  NH^-N concentration was usually less than 1.0 mg/L. This effluent  quality was achieved at mass loading levels comparable to those for sewage treatment (10.0  g B O D / m * d for carbon removal and 0.8 g N H - N / m * d 2  2  5  3  for  nitrification). The results demonstrated that long hydraulic retention times (HRT hrs.) can offset the effects of lower temperatures.  >4  Nitrification efficiency in particular  was shown to be HRT dependent. Limited heavy metal data indicated that heavy metals were removed at efficiencies and relative affinities comparable to those observed in activated sludge studies. An aside to this study showed that trace organics, some of which are on the EPA priority pollutant  list, were present in this  leachate and were effectively removed during passage through the RBC.  Keywords:  Leachate treatment, Rotating Biological Contactor (RBC), carbon removal, nitrification, loading rates, hydraulic retention time (HRT) effects, heavy metal removal, priority pollutants.  ii  Table of Contents  ABSTRACT  ii  ACKNOWLEDGEMENTS  x  1.  INTRODUCTION  1  2.  SITE DESCRIPTION  3  3.  RATIONALE  7  3.1  Purpose  7  3.2  Literature Review - Leachate Treatment  7  3.3  RBC  4.  5.  Treatment  23  EXPERIMENTAL P R O G R A M  45  4.1  Sampling  And Analysis  4.2  Sampling  Procedures  52  4.3  Analytical Procedures  54  LEACHATE  Program  47  QUALITY  58  5.1  Leachate Generation  58  5.2  Affect of Water Inputs on  5.3  Premier Landfill Leachate  67  5.4  Organics  85  5.5  VFA's  89  5.6  Nitrogen  94  5.7  Total Solids  5.8  Metals  5.9  Specific Trace Organics  Leachate Quality  and Specific Conductance  65  95 98 104  6.  PILOT PLANT  106  7.  RBC  112  OPERATION  7.1  Start-Up  112  7.2  The  116  Disruptions  iii  7.3 8.  9.  A  New  TREATMENT  Beginning  119  RESULTS  131  8.1  Carbon Removal  132  8.2  Nitrification  140  8.3  Suspended  8.4  Metals  147  8.5  Specific Trace Organics  150  Solids  147  DISCUSSION  156  9.1  Organic Removal  156  9.2  Nitrification  170  9.3  RBC  9.4  Metals and Trace Organics  185  9.5  Toxicity  186  9.6  Implications for Full Scale Treatment  186  9.7  Experimental Program and RBC  190  Response  to Variable and Intermittent Loading  Operation  183  10.  SUMMARY  193  11.  CONCLUSIONS  195  12.  R E C O M M E N D A T I O N S FOR  13.  REFERENCES  199  14.  APPENDIX 1  206  15.  APPENDIX 2  214  16.  APPENDIX 3  220  FURTHER  iv  RESEARCH  197  List of Tables 3.1  Summary of Leachate Treatment  4.1  Sampling Program  Studies  12  48  5.1 Variability of Leachate Composition  59  5.2  Premier Leachate Characteristics (Well #1)  69  5.3  Leachate Heavy Metal Levels (AA)  102  5.4  Leachate Metal Analyses (ICP)  103  5.5  Leachate Trace Organic Content  105  6.1  RBC Specifications  107  8.1  RBC Heavy Metal Removal (AA)  151  8.2  RBC Metal Removal (ICP)  152  8.3  RBC Biomass Metal Levels (AA)  153  8.4  RBC Biomass Metal Levels (ICP)  154  8.5  RBC Trace Organic Removal  155  9.1  Design Loadings for RBC Treatment  v  of a Municipal Wastewater  188  List of  Figures  2.1 Photo of New Section of Premier Landfill looking North-West (June 1983)  3  2.2 Premier Street Landfill - North Vancouver, B.C. Site Plan and Location  5  2.3 Photo of New Section of Premier Landfill from Top of Old Landfill (June 1985)  6  5.1A Premier Leachate Characteristics vs. Time 82/83  70  5.IB Premier Leachate Characteristics vs. Time 83/84  71  5.1C Premier Leachate Characteristics vs. Time 84/85  72  5.2 Leachate Flow and Constituent Mass Release  Premier Street Landfill  75  5.3A Leachate Carbon Content vs. Time 82/83  76  5.3B Leachate Carbon Content vs. Time 83/84  77  5.3C Leachate Carbon Content vs. Time 84/85  78  5.4A Leachate Nitrogen Content vs. Time 82/83  79  5.4B Leachate Nitrogen Content vs. Time 83/84  80  5.4C Leachate Nitrogen Content vs. Time 84/85  81  5.5A Leachate Total Solids and Sp. Conductance vs. Time 82/83  82  5.5B Leachate Total Solids and Sp. Conductance vs. Time 83/84  83  5.5C Leachate Total Solids and Sp. Conductance vs. Time 84/85 vi  84  5.6 T O C vs. C O D  86  5.7 B O D  5  vs. C O D  87  5.8 B O D  5  vs. B O D r / C O D Ratio  88  5.9 Leachate VFA conc'n vs. Time  90  5.10  COD(vfa) vs. Leachate C O D and B O D  5.11  C O D vs. VFA  92  5.12  BOD  vs. VFA  93  5.13  NH  3  vs. C O D  96  5.14  NH  3  vs. Sp. Cond.  97  5  5  5.15 Tot. Solids vs. Sp. Cond.  91  99  5.16  C O D vs. Sp. Cond.  100  5.17  BOD  101  6.1  5  vs. Sp. Cond.  Photo of RBC Prior to Start-up, Showing Disk Media and Influent  Pump  6.2  106  Photo of RBC Installed Adjacent to the North Leachate Lift Station  107  6.3 Section of North Leachate Lift Station Showing RBC Connections  108  6.4 Photo of RBC Pump Inlet Screen  109  7.1 Photo of creamy, taupe coloured, initial bacterial growth (June 1983) 7.2 Photo of mature biomass growth during start-up (late June 1983)  113 114  vii  7.3  Photo of Pump  tubing failure  7.4  Photo of Aftermath of 1  7.5  Photo of RBC  s t  115  Flood in the RBC  (November  1983)  after being raised 1m to avoid flooding  117  118  7.6A  RBC  Operational History: Influent Flowrate and Loading  82/83  120  7.6B  RBC  Operational History: Influent Flowrate and Loading  83/84  121  7.6C  RBC  Operational History: Influent Flowrate and Loading  84/85  122  7.7  Photo of heavy dark growth on RBC  during April-May 1984  7.8 Photo of single bellows leachate pump  (165  rpm) and nutrient pump  123  126  7.9 Photo of twin bellows leachate pump (50 rpm) and nutrient pump  127  7.10  Photo of healthy bacterial growth  129  7.11  Photo showing  average biomass  129  7.12  Photo showing  leachate foaming in RBC  8.1  thickness across the RBC  first stage  130  RBC  Effluent B O D  5  vs. Loading and Time  133  8.2 RBC  Effluent B O D  5  vs. Loading Rate  135  8.3  1  s t  8.4  1  s t  8.5  RBC  Effluent C O D  vs. Loading Rate  138  8.6  RBC  Effluent C O D  vs. Loading and Time  139  8.7A  and 4  t h  and 4  t h  RBC  Stage B O D  5  (settled) vs. Time  136  Stage B O D  5  (filtered) vs. Time  137  Nitrification Performance - part 1 viii  142  8.7B RBC Nitrification  Performance - part 2  8.8 Effluent NH^ and N 0  st  8.9B 1  st  8.10  1  st  8.11  RBC Effluent Suspended Solids  BOD  t h  and 4  t h  and 4  t h  Stage N H  vs. Loading Rate  8.9A 1  9.1  and 4  3  144  and N 0  3  - part 1  145  Stage NH^ and N 0  3  - part 2  146  3  Stage Suspended Solids  148  149  Removal versus Loading Rate  5  143  159  9.2 C O D Removal versus Loading Rate  160  9.3 B O D  163  9.4  5  Percent Removal versus Loading Rate  BODjj Removal versus Loading Rate Corrected for Temperature  9.5 B O D  5  Removal - Monod Kinetics Approach  164  168  9.6 N H  3  -N Removal versus Loading Rate  173  9.7 N H  3  -N Percent Removal versus Loading Rate  174  9.8  NH  3  -N Percent Removal versus Temperature  175  9.9 N H  3  -N Removal versus Loading Rate Corrected for Temperature  177  9.10  NH  9.11  NH-. -N Removal - Monod Kinetics Approach  3  -N Removal versus Hydraulic Retention Time (HRT)  ix  178  180  ACKNOWLEDGEMENTS I dedicate this thesis, and give my greatest love and thanks, to my wife Joan, w h o  patiently supported me in every way during the protracted process  of  producing it. I w o u l d also like to thank my parents and family, for their support and encouragement  throughout  my education. I also extend a special thanks to  Professor Jim Atwater, for his guidance and support in overseeing this study. The greatest benefit which I received as a result of the extended duration of this study was  that those people who  might otherwise have been  remembered  as staff or aquaintances, have become valued friends. With this in mind, I would like to extend my sincere thanks to Sue  Jasper, Environmental Lab Manager, Susan  Liptak and Paula Parkinson, Research Technicians, and Tim Ma, GC/MS Technician, for their expert help and advice. During the course of this study, 1 also considerable use of the facilities and expertise within the Civil Workshop.  Engineering  Again, with the above in mind, 1 would like to thank  Head Technician, and his staff, especially Art Brooks, and Guy like to thank the members  made  Dick  Postgate,  Kirsch. I would also  of the Environmental Engineering faculty, Dr. Oldham,  Dr. Mavinic, Dr. Hall, and Prof. Atwater, for their help and tutelage. Last, but not least, I would like to thank the many other members  of the staff, and fellow  students, whose help and interaction have been very valuable, especially, Fred Koch, Ann  Davern, Kelly Lamb, Carolyn Foo, Bruce Anderson, Yves Comeau, Troy Vassos,  Ken Johnson, M a n o  Ramarathan, Chris Town, and Paula Wentzell.  I w o u l d also like to acknowledge the generous assistance and cooperation of the District of North Vancouver, the staff of the Premier Street Landfill, and Equipment  Ltd. of Mississauga, Ontario. Funding was  Sciences and Engineering  Research Council of Canada  provided by the Natural (NSERC).  CMS  1. INTRODUCTION  Landfill leachates are wastewaters formed when water migrates through emplaced solid wastes and carries off or leaches, soluble matter, decomposition products, and fine solids. This water or leachate emerges from the solid wastes laden with organic and inorganic compounds, heavy metals, etc., with the potential to pollute the surface and groundwater environment unless control measures are taken. Efforts to control or prevent pollution caused by leachates have come relatively recently. Landfill leachates have only been acknowledged as environmentally significant for about the last 20 years and awareness at the regulatory and local operational levels has occured mainly over the last 10 years. Prior to this awareness, landfills were commonly established on cheap land, such as peat bogs, ravines, and abandoned gravel pits, with no particular site preparation. Today many of these sites are still in use. In retrospect, the choice of these types of sites to minimize disposal costs and to reclaim land was unfortunate, as their hydrological characteristics have often exacerbated the leachate problems at these landfills, necessitating expensive remedial action. Dykes, groundwater barriers, and leachate collection systems, have been installed at many of these landfills to prevent any further escape of leachate into the environment. Across North America large volumes of leachate are being collected from existing landfill sites. Many of these existing landfills are nearing capacity and so in several jurisdictions (such as Vancouver), an urgent evaluation of municipal waste disposal options is underway. Regardless of the types of solid waste handling and treatment methods employed, a residual component of the solid wastes remains for ultimate disposal, likely by landfilling. In addition, land disposal remains very competitive economically with other solid waste handling methods in many areas.  1  2  Thus landfilling will be an important component of most solid waste disposal schemes well into the foreseeable future. However, the recently developed public awareness of the pollution potential of landfills has made it very difficult to establish new landfill sites. One benefit of this awareness is the public insistance that new landfill sites include detailed measures to prevent pollution of the environment from leachate, either by in-situ attenuation in underlying soil layers, or by containment, collection, treatment, and disposal. The current trend in North America is towards full collection and treatment of landfill leachates before they are discharged into the environment. Therefore, increasing volumes of leachate can be expected from new landfills, carefully engineered to contain all the leachates they produce. Once these leachates have been collected, efficient methods of treatment and/or disposal must be found. As landfill leachate characteristics and site conditions (such as climate, proximity to sewers, etc.) are highly site specific, a variety of effective treatment and disposal schemes must be developed for these wastes. The Environmental Engineering Croup at the University of British Columbia has had, and continues to have, an extensive research interest in landfill leachate issues. Numerous studies have been conducted on leachate generation, characterization, toxicity, treatment methods, and treatment parameters. To further this effort, this study was initiated to evaluate the performance of a Rotating Biological Contactor (RBC) treating a relatively weak municipal landfill leachate. The leachate that was treated generally had a chemical oxygen demand (COD) of less than  1000  mg/L and a total Kjeldahl nitrogen (TKN) value of less than 50 mg/L. This study evaluated the capacity of the RBC to remove both the carbonaceous and nitrogenous oxygen demanding material from this leachate at pilot scale and under field conditions.  2. SITE DESCRIPTION  The Premier Street landfill is situated on a natural bench on the east  bank  of the lower Lynn Creek ravine in the District of North Vancouver. This bench lies within a steep-sided bowl, immediately downstream of the lower boundary of Canyon 6  Park. The bench is composed  m in depth, underlain by a dense  Lynn  of fluvial sand and gravel up to approx. glacial till which is contiguous  with the walls  of the bowl. The District of North Vancouver began 1959, and by agreement began  using  the property for a landfill in  accepting wastes from the District of  Vancouver, and the City of North Vancouver, in 1969 1981  and 1970  West  respectively. In  the then active landfilling area of the property was nearing capacity and the  decision was made to develop the final 10.5  ha section of the property for use  (see Figures 2.1,2.2,2.3).  Figure 2.1  Photo  of N e w  Section of P r e m i e r Landfill l o o k i n g (June 1983)  North-West  4  N.Vancouver y  Burrard"""ir"  -PREMIER STREEJJ LANDFILI  Inlet  Vancouver  SCALE =  100  Figure 2.2  200m  Premier Street Landfill • North Vancouver, Site Plan and Location  B.C.  5  Figure 2.3 P h o I o b T ^ e w ^ e c l i o n __  of Premier Landfill from Top of Old Landfill (June 1985)  6 This section is bounded by Lynn Creek to the north-west, the steep walls of the bowl to the east, and by a mountain of garbage, which was the previous active landfill area, to the south-west. A bentonite slurry trench and dyke were constructed along Lynn Creek, paralled by a perforated leachate collection pipe. An extension of the slurry trench runs along the base of the previously filled area to isolate the old and new sections. The slurry trench and underlying glacial till combine to form a relatively impermeable dish beneath the site to contain the leachate produced. A ditch at the base of the bowl slopes diverts storm water around the site into the creek. The leachate collection pipe terminates in the north lift station which transfers the leachate via another pipe to the municipal sewer system. These preparations were completed in January 1982  and this new site began  receiving wastes shortly thereafter. Up till now the landfilling activity has been restricted to an area adjacent to the old site, which is about 40% of the 10.5 ha available. Since drainage from the whole section is recovered by the leachate collection system, rainwater from the unfilled area has a diluent effect on the characteristics of the leachate received in the pumpwell. Although the site is at a low elevation (<70  m), its postion at the  base of the North Shore mountains attracts a relatively heavy annual rainfall of approximately 2000 mm.  3. RATIONALE  3.1  PURPOSE The purpose of this study was to evaluate the suitability and effectiveness of  a Rotating Biological Contactor (RBC) for the treatment of a municipal landfill leachate. This evaluation was conducted at pilot scale (0.9  m dia.), and under field  conditions at a landfill site in order to produce data which would closely approximate full scale expectations. The suitability of the RBC for the treatment of this leachate was evaluated primarily on the basis of three determinations: the capacity of the RBC to remove the carbonaceous component of the leachate; whether or not the leachate could be nitrified; and if so, the capacity of the RBC for nitrification. These capacities, for carbon removal and nitrification, would be defined as the maximum mass loading rate (g/m^*d of BOD^ or NH^) for which complete treatment was maintained (effluent BOD^s25 mg/L and/or N H ^ s i . O mg/L). Loading rates for leachate treatment could then be compared to those established for domestic sewage treatment.  3.2 LITERATURE REVIEW - LEACHATE TREATMENT A search of the literature in early 1983, prior to the start of experiments, failed to find any references concerning RBC, or other aerobic fixed growth process, treatment of landfill leachates. In the absence of directly comparable results, some literature concerning leachate generation and composition, leachate treatment (primarily by aerobic suspended growth systems), and RBC treatment of other types of wastes, was collected to provide background information for this study. This discussion will focus on the later two topics as leachate quality is discussed adequately in Section 5.0.  7  8 Landfill leachates are a relatively recent topic of environmental concern. Chian and DeWalle (11)  attribute one of the first studies to Merz, who investigated the  leachate from incinerator ash dumps in 1952  and went on to later study leachates  from municipal solid waste landfills. Over the next twenty years the extent of the pollution caused by leachates was documented and the emphasis of research shifted to studying the mechanisms of leachate generation and movement, and measures to control or treat the leachate after it has been produced. With respect to the treatment of leachates, two of the earliest, widely referenced studies are those by Boyle and Ham (5),  and Cook and Foree (15),  both published in the early 1970s.  Boyle and Ham looked at the treatability of landfill leachates by biological processes, both anaerobic and aerobic. In their anaerobic studies, which received the greater emphasis, Boyle and Ham achieved greater than 90 percent removals of both C O D and B O D , from influent C O D concentrations of 2,240 to 22,400 mg/L. 5  Loading rates ranged from 0.43  to 2.16  - 20 days, at an average temperature  kg COD/m^*d, and detention times were 5  of 23° C. Effluent quality was found to  improve with decreased loading rates and/or longer detention times. They also found that anaerobic system performance was very temperature removals dropping from 87.2  dependent, with C O D  % at 23° C, to only 22 % at 10° C.  Subsequent studies have repeatedly confirmed the capabilities of anaerobic treatment of landfill leachates. For example Bull et al. (7), realized a 96.8  percent  BODjj removal from an influent concentration of 5700 mg/l BOD^ at a detention time of 30 days. Although these two, and many other, studies have demonstrated a high percentage BOD^ removal, the effluent BOD^ values are typically greater than 100 mg/L (13).  Effluent ammonia levels are also usually very high because of the  minimal nitrogen requirements of anaerobic bacteria and the efficient conversion of organic nitrogen to ammonia. Therefore in most instances, anaerobic treatment can not be regarded as a complete treatment process and further treatment or effluent  9  polishing is required. The aerobic treatment studies conducted by Boyle and Ham were considered less successful. Three fill and draw reactors with a 5 day detention time were relatively heavily loaded (0.3  - 1.4  kg B O D / m * d ) with landfill leachate. Effluent 3  5  BODj- levels ranged from 160 to 1400 mg/L, and the units were plagued by foaming and solids separation problems which increased in severity at the higher loadings. However, the results indicated that for loading levels less than 0.48 BOD[-/m3*d and warm temperatures  kg  (23° C), that BOD,, removals of greater than  90 percent could be achieved, ln another segment of this study, Boyle and Ham also demonstrated that a landfill leachate (COD =  10,000 mg/L) could be combined  with domestic sewage up to 5 percent leachate by volume for co-treatment  in an  extended aeration process without impairing process performance. The most important  result of this study however was the demonstration that biological  treatment of landfill leachates was possible. Cook and Foree (15)  expanded upon the aerobic biological treatment studies  of Boyle and Ham, and also evaluated various physical-chemical treatment processes for landfill leachate treatment. Using fill and draw aerated batch reactors, with a 10 day detention time and loading rates between  1.58  and 7.9 kg COD/m^*d, they  were able to achieve BODj- removals in the order of 99.7 percent, from an influent BOD^ of 7100  mg/L, to effluent values of less than 26 mg/L. Nutrient  additions of nitrogen (N)  and phosphorus (P) did not improve the treatment  efficiency significantly despite the nutrient ratio of the leachate (100:2.33:0.23 BODj.:N:P) being far less than the generally accepted 100:5:1 ratio for healthy growth. Solids settleability was observed to be very good but foaming remained problematic requiring the periodic use of a defoaming agent. A theoretical  minimum  detention time of 5.3 days was calculated from kinetic considerations and confirmed by the failure of a 5 day detention time unit. This result indicated that the  10  detention time of the reactors used in the Boyle and Ham experiments were probably too short to achieve stable operation or efficient treatment. The evaluation of physical-chemical treatment involved chemical coagulation followed by activated carbon. Chemical coagulants were effective for suspended solids and colour removal but since the C O D of leachate is mainly soluble, C O D removal was minimal. Activated carbon proved fairly effective  at removing the soluble  C O D from the leachate, but given the high concentrations of organic material in many leachates, this treatment method would not be economical. As a polishing step for biologically treated effluents, activated carbon proved very effective  for  residual C O D and colour removal. Chian and DeWalle (11) later reviewed the experience with physical-chemical treatment of landfill leachates and came to a similar conclusion, that physical-chemical treatment is best suited for polishing biological treatment effluents, or treating old leachates, which have a low soluble organic content. Reverse osmosis was shown to be the most effective treatment method followed closely by activated carbon. It was also effective for treating raw leachate, except that rapid blinding of the membranes made such an application impractical. The studies by Boyle and Ham, and Cook and Foree, indicated that biological treatment, both anaerobic and aerobic, could effectively remove organic material from relatively strong leachates and that physical-chemical methods were much less effective  except for suspended solids removal. Subsequent studies of  leachate treatability have expanded upon these initial results and qualified the conditions under which the various treatment methods are applicable. The remainder of this discussion however will focus on aerobic leachate treatment as the RBC is primarily an aerobic treatment process. The articles concerning aerobic biological landfill leachate treatment reviewed for this study were intended to be a representative  sampling of previous treatment  11  experience. Since all of the studies involved aerobic suspended growth systems (activated sludge or aerated lagoons), and therefore were not directly comparable to this study, a more comprehensive review was unwarranted. The papers reviewed cover a fairly wide variety of different problems, and treatment  leachates, treatment  conditions, operational  related topics.  Table 3.1 summarizes various results and parameters from the  treatment  studies reviewed. A number of points are readily apparent from this table. Column 1 shows that the C O D and B O D , , values of the leachates used in these studies are, with the exception of  Palit and Quasim (56), w h o used diluted feed, moderate  to high in comparison to those of the Premier leachate used in this study. However the pointed out  BOD^/COD  ratios of these leachates are all relatively high (>0.5). As  by Chian and DeWalle (11), the organic strength of a leachate, as well  as its biodegradability, as exemplified by its B O D ^ / C O D ratio, reflects the degree of stabilization of the landfill, with high organic strength and degradability  being  associated with fairly new or young landfills. Therefore all of these treatment have dealt with high organic strength leachates from  studies  young landfills. The discrepancy  between organic strengths, as well as the lack of comparable leachates encountered in the literature,  indicates that the  such a low total organic strength the low strength  and other  Premier leachate is somewhat unique to have coming from  a young landfill. The reasons for  characteristics of the Premier leachate are explored in  Section 5.0. C o l u m n 2 shows what  is perhaps the most significant point, that all the  studies demonstrated very efficient  carbon removal from different  The results indicate that aerobic treatment  processes, operating within  conditions, are generally capable of complete treatment less than 25 mg/L, regardless of the  landfill leachates.  to effluent  limiting  B O D ^ values of  initial leachate strength. However effluent  levels are usually much higher (100 - 900 mg/L), largely due to refractory  COD  humic  Table  Column No.  3.1  #1 Influent  #2 Effluent  COD mg/L (BOD mg/L)  References  Summary  COD BOD  5028 (3035) [0.60]  ct  al.  18488 12468 [0.67]  (43)  (1984)  (22,24) (1984,1985)  Ehrig,  285-49900 (27-29975) [0.013-0.92 1]  H.J.  Wong  & Mavinic  (81)  (1982)  Zapf-Gilje (86) 1981)  & Mavinic  13000 (8090) [0.62]  19000 (13640) [0.71]  Treatment  % Rem % Rem  kg / m (F/M Ratio) 3  #5 Op. Prob. Temp. Foaming Effects or  (0.21 or less) |>I0 d)  #7  Heavy Metal  Nitrification  Antifoam Good Added Bulking @ SRTs5d rising sludge  95 99  939 1 18  #6  (0.12-0.32) Mostly BOD<25  148-888 7-188  0.0005-1.128  (0.1 1-0.405) (best @<0.16) <900 <97  >95 >99  0.96,2.14,3.21 (0.18-0.49) [6,9,20 d]  Poor  Minor  Good Poor  -SRTs 1-20 days, <5 erratic, worked well -very long SRTs required for nitrification  >10  Very Good  -full scale plant, 0.144 Mgd -influent NH3 conc*n toxic (1072 mg/L) air stripped. 3  Good  Minor  Very Poor  Excel, -full scale Idg. <20 g / m prod. eff. <25 mg/L - F / M <0.05 recommended to avoid filamentous bulking -complete nitrification when N/MLSSS0.03  Foam Poor Settling  «0.1) [10-70 d]  >93 >97  Comments  Settling Removal  Minor BOD<20 SRT>10 d  Studies (suspended growth)  #4  tSRT]  @ Keenan  Leachate  #3 Loading Rate  [BOD/COD Ratio]  Robinson & Maris (65,66) (1983,1985)  of  Control -nutrients 100:3.2:1.1, temp down to Reactor 5" C Only -nutrient levels have little effect, F/M deter, settling -temperatures  2 5,16,9' C  Column No.  #1 Influent  References  t  #2 Effluent  COD mg/L < 5 9^-) t30D  COD  m  #3 Loading Rate  % Rem kg / m % & \?AA Ratio) 3  ern  [BOD/COD Ratio]  Stegmann & Ehrig (74) (1980)  4000-16660 (750-11253) [0.19-0.67]  [SRT]  2 0 - 3 5 0 0 40-94 2 0 - 6 0 95-99.9  0.16-0.9 (0.02-0.1 1)  #4  #5 #6 #7 Op. Prob. Temp. Foaming Heavy NitriEffects or Metal fication  Comments  Settling Removal  Foam  Excel, -activated sludge and aerated lagoon studies -treatment dependent on BOD/COD ratio -complete nitrification when <1 kg BOD/m 3  Paid & Quasim (56) (1977)  365  29-55  -influent diluted 22-26 times -kinetic parameters evaluated  85-93 (0.226-0.436) [7-23.8 d]  Uloth & Mavinic (1977)  Cook & Foree (1974)  (79)  (15)  15800 (7 100) [0.45]  Minor Good  >98  48000 (36000) [0.75J  (0.06-0.22) [>20 d]  Good  SRTs 10,20,30,45,60 d, BOD inhibition obs. -best treatment at F/M <0.12 and SRT>20 d -mechanical mixing and low air controlled foaming  290-360 10-26 [10 d]  Boyle & Ham (5) (1974) Premier  Leachate  2700-9200 (1550-8000) [0.47-0.87] 86-4421 (44-3020) [0.25-0.75]  430-6720 160-7800  0.3-5.28 [5 d]  Foam  -F/M  > 1.5  for failed unit  14  and fulvic acids (11,12). Further C O D physical-chemical effluent polishing effective (as mentioned  with  (11,23,74), activated carbon being particularly  previously), but as indicated by Stegmann and Ehrig  the necessity of removing Column  removal has been demonstrated  this residual C O D  (74),  is a subject for debate.  3 gives the loading rates and/or solids retention times (SRT)  which the various  at  experiments were run. The conditions for which g o o d treatment  was achieved were very similar in all of the studies. It was  generally concluded that  a SRT  kg  S*  10 days and a F/M  ratio of less than 0.1  volumetric loading less than 0.1  - 0.15  kg BOD,-/m *d, was  and reliable treatment (43,66,79,81). These to those  - 0.15 3  operating  necessary  for efficient  describing the extended aeration variant of activated sludge  treatment indicate low rates of growth of the process reflected in the process  kinetics. Mavinic  carbon removal from a number  leachate, and those  (49) summarized  proportions  strength of the leachates. It was  indicated that old leachates, those  also found  and demonstrated that  material in the  indicated and increased with the  concluded that the SRT  required to effectively treat  of the leachate. Since the SRT is  ratio and loading rate, the maximum  as the leachate strength  biodegradability, would  be  the kinetic parameters for  of readily biodegradable  markedly with the strength  inversely proportional to the F/M decrease  for efficient leachate  increased with increasing leachate  leachates studied, inhibition of bacterial growth was  a leachate increases  (51).  of the kinetic coefficients determined for a  typical of domestic sewage,  strength. Therefore, despite high  fairly well  bacteria, which should  of leachate treatment studies  the difference between the values  MLVSS, or  conditions correspond  The long SRTs and low organic loading rates necessary  would  BODj/kg  loading rate  increased. Kinetic considerations  with a low BOD^/COD ratio and thus  also low  also require longer SRTs and lower loading rates. Mavinic  that the kinetic parameters were greatly influenced by cold temperatures.  15  The effects of the changes in kinetics with leachate strength the results of Stegmann and Ehrig (74). They reported aerated lagoon leachate treatment  can be seen in  on activated sludge and  from bench, pilot, and full scale studies. The  more extensive aerated lagoon results indicated that complete treatment  to BODj.  values less than 25 mg/L were possible at loading rates and detention  times  determined by the  BODr/COD ratio of the leachate. As reported  previously by  Chian and DeWalle (11), the BOD,-/COD ratio is a useful characteristic with which to catagorize leachates and evaluate their treatment leachates (>0.4  BODj/COD) could be treated at loading rates up to  BODr/m.3*d, but intermediate  results. High ratio (high  longer detention  ratios of  low BOD^/COD ratios BODg/m^*d, but  0.05 kg  times were also required. For leachates with  0.1 - 0.4, loading rates of  necessary but the detention  strength)  <0.01  kg BOD^/m^*d were  times required were relatively constant. Leachates with  <0.05 required very low organic loading rates, < 0.002 kg  the detention  times were also reduced reflecting the small fraction  of degradable material. In addition to  maintaining  reasonable SRTs and loading rates, top  treatment  efficiency was generally dependent on achieving a proper nutrient (N  +  in the process. Landfill leachate is generally found to have sufficient  nitrogen  present in the form  of ammonia (NH^) but  deficient. Therefore most leachate treatment phosphorous, to mentioned treatment  prevent  levels of phosphorous are usually studies have added nutrients,  particularly  nutrient deficiencies. C o o k and Foree (15) found, as  previously, that nutrient additions did not have a great effect on total efficiency, but  the effluent  quality was improved slightly when  were added. Stegmann and Ehrig (74) also found that P addition effect  P) balance  on effluent  quality. A lack of phosphorous inhibited  sludge unit but the unit was judged  reduction in soluble effluent  insignificant  growth  nutrients  had only a minor in one activated  B O D , , with P addition in another  compared to the overall removal. W o n g and Mavinic  16  (81), while investigating  the effects of sludge age  treatment, confirmed the previous showed a nutrient ratio of requirements  work  by Temion  and Mavinic  100:3.2:1.1 BOD[-:N:P was  (see  sufficient to satisfy the  nitrogen requirements P addition had  remained in the process  ammonia  nitrogen  effluent. The reduced  in  nutrient  for leachate treatment may help explain the small effect that  in the C o o k  lower effluent soluble  that  for leachate treatment were less  accepted ratio for sewage treatment when  excess of 100:3.6 B O D ^ N  growth  settling of the bacteria. Robinson and  also found that nutrient requirements  than the commonly  leachate 81), which  of the biomass for leachate treatment. They also found  phosphorous deficiencies resulted in poor Maris (65)  and temperature o n  BOD,-  and Foree, and Stegmann and Ehrig studies, but the levels achieved with P addition indicate addition  beneficial to attain high levels of treatment and improve  process  is-  reliability.  Column 4 shows that despite the indications of temperature sensitivity from kinetic considerations found  (49), temperature effects on treatment efficiency were  to be minor. Zapf-Cilje and Mavinic (86)  treatment efficiency with decreasing and Maris (65), and on  5 summarizes  of some of these  excessive  foaming  reported impaired solids settling at  to be the main adverse  the operational problems  frequently cannot instances,  effect.  encountered  during  studies. There were basically two types of  concern  the  problems;  of leachate treatment. The high aeration rates  required for treatment and the general use  diffusers, to avoid  temperatures  of the leachate, and poor settling of the bacteria. Excessive  formation is a c o m m o n usually  temperature down to 9° C. Similarly Robinson  all of these studies  lower temperatures, so this appears  course  a minimal loss of  Ehrig (22), reported insignificant effects of lower  B O D j removal. However  Column  observed  generally  plugging,  further aggravates the problem. Surface  be used because  heat loss)(33). The  of inefficient coarse bubble  of concerns  treatment studies  about foaming  aerators  (or in other  show a trend towards  increased  foam  17  foaming as the leachate strength increases. Foaming problems also increased with leachate loading levels. Uloth and Mavinic (79)  linked the foaming tendency of  leachate to metal concentrations and anticipated severe problems with the high strength leachate used in their study. However, the use of mechanical mixing and minimal aeration was very successful at avoiding excessive foam formation. In other studies, anti-foaming  chemicals were used effectively.  For full scale applications, the  results of Uloth and Mavinic indicate that mechanical mixing greatly reduces foaming problems, such that the use of submerged turbine-sparger combinations could reduce the need for chemical or physical foam control measures. The settling problems reported in some of these studies can be associated with either low temperatures,  as mentioned above, excessive loading levels, or  nutrient deficiencies, except for Keenan ef al. (43)  who had denitrification and  turbulence occurring in their clarifier. Both the low temperature  and excessive  loading conditions led to filamentous bacterial growth and sludge bulking conditions which impaired solid-liquid separation. Under more favourable operating conditions, excellent solids settleability  was generally observed. Good solids settleability was  frequently attributed to the inclusion of inorganic precipitates and adsorbed metals in the bacterial floes. As shown in column 6, heavy metal removals from the leachate during aerobic treatment was very good. Heavy metals are removed by various mechanisms but the two main ones are as inorganic precipitates, usually hydroxides or phosphates, and adsorbed or complexed with the biological solids (6,9,76). However the various metal species are not removed equally well. Studies of metal removal by activated sludge during domestic sewage treatment have established a fairly consistent order of metal removal efficiency. Iron, Zinc, Copper, Chromium, and Lead, are removed best while Nickel, Manganese, Calcium, and Magnesium, are removed least. Brown and Lester (6)  reported average removals of Fe (86%), Zn  1 8  (69%), Mn  Cr (66%),  (20%),  Pb (64%),  Hg  (63%), Al (51%)  (46%),  Ni  (33%),  landfill leachates often have higher heavy metal  the leachate treatment studies have demonstrated  with similar affinities for metal species  than  heavy metal  and similar or better removal rates.  removal rates during leachate treatment are probably attributable in part to  the generally longer sludge  ages employed  as metal removal is enhanced  sludge  ages (76). Very efficient heavy  during  anaerobic treatment of leachates (5,7). Under  heavy  Cd  and different relative concentrations between metal species  domestic sewage,  Higher  (66%),  and C a (6%). Although  concentrations  removals  Cu  metals precipitate as sulphides  metal removals  have also been  by  longer  observed  anaerobic conditions,  many  which, combined with the biological  complexing, yields the high removal rates. The efficient removal of soluble heavy  metals, by precipitation and biological  inactivation, explains the general lack of observed very high  concentrations of heavy  example, Uloth and Mavinic (79) the studies  metals in leachates undergoing had the highest  reviewed here, Fe 960  mg/L, yet a stable process and very high  was  toxic effects despite the often  mg/L, As  leachate metal concentrations  3.6 mg/L, Pb 1.44  maintained and achieved  metal removal rates. N o n e  treatment. For  >98  of the other studies  mg/L, and Zn percent C O D  nitrification performance. Studies  as one  223  removal,  reviewed indicated  any metal toxicity problems, except possibly for Jasper et al. (42), who metal accumulation in their sludge  of  proposed  explaination for a deterioration of  are ongoing  at UBC  to clarify heavy metal toxic  effects, particularly of Zinc, on nitrification/denitrification of leachate (18).  Although  leachate metal levels are generally non-toxic, Mavinic (49) indicates that they probably contribute to inhibiting bacterial growth rates and thus affect the kinetics. In most  cases however,  process  leachate metal concentrations are not considered  significantly impair biological leachate treatment, either aerobic or anaerobic  (11,33).  to  19  In addition to the high levels of carbonaceous oxygen demanding material found in landfill leachates, particularly young leachates, relatively high levels of nitrogenous oxygen demanding material, mostly in the form of ammonia (NH^ - N), are also generally present. Since, as will be discussed in Section 5.0, the high ammonia concentrations usually persist long after the organic strength has been reduced by the maturation of the landfill, there is an increasing interest and emphasis on ammonia removal from landfill leachates. Ammonia is removed biologically by conversion to organic nitrogen during bacterial cell synthesis, or oxidized to nitrate (NO^)  by nitrifying  bacteria. Depending upon site specific effluent  guidelines or goals, the nitrate could possibly be further treated by biological denitirfication (33)  to remove the nitrogen completely as nitrogen gas, although Henry  indicates that experience with denitrification of leachate is insufficient to predict  its reliability. Research emphasizing nitrification  or nitrogen removal from leachate is very  recent; thus, most of the studies reviewed looked at nitrification as a secondary topic to carbon removal. Given that, as mentioned previously, all these studies dealt primarily with high organic strength leachates, with C O D concentrations much greater than ammonia concentrations, this approach was not unreasonable. Fortunately, nitrification  the  results, where given, are sufficiently detailed to support a number of  conclusions. The nitrification column 7 of Table  results from the various studies are summarized in  3.1.  The first point demonstrated by the results of these studies is that the occurrance of nitrification  depends upon the BOD^NHg ratio of the leachate.  Studies in which this ratio was greater than or equal to roughly 100:3.6 had no nitrification  take place (65,66,81). The lack of nitrification under these conditions is  attributed to the greater  growth  rate of the heterotrophic bacteria which convert  ammonia to organic nitrogen required for cell sysnthesis (34,35,47,53). As mentioned  20  previously, the 100:3.6 B O D : N H 5  3  ratio represents the minimum  nutrient requirement  of the heterotrophic bacteria for cell growth. This ratio of organic strength  also  ensures a high growth rate of the heterotrophic bacteria at practical loading rates, since the heterotrophs follow a first order kinetic response  to substrate  concentration. Civen the relatively high organic loading rates employed in most the studies reviewed, the heterotrophs were likely growing maximum rate in most instances. O n zero order response thus essentially grow  of  at or near their  the other hand, nitrifying bacteria follow a  to substrate concentrations greater than about 0.5 mg/L  and  at a constant slow rate. Therefore, under nutrient limiting  conditions, the heterotrophs  consume  essentially all of the ammonia for their  nutrient requirements by virtue of their much higher growth rate. Results from the studies B O D ^ / N H j ratio was  less than  in which excess ammonia was  present such that the  100:3.6, generally indicate that efficient nitrification of  landfill leachate is possible. Stegmann and Ehrig (74) achieved complete nitrification in both lab scale activated sludge, and pilot scale aerated lagoon later case at organic loadings  of up to 1 kg B O D g / m * d . W h e n denitrification was 3  attempted in the activated sludge  studies, 9 9 %  removal of the influent nitrogen  achieved as an influent ammonia concentration of 973 mg/L was mg/L N H ^ than 9 9 %  and 25 mg/L  NO^  in the effluent. Keenan er al. (43)  nitrification of leachate in a full scale activated sludge  nitrification was  also observed  study, to which excess Mavinic (18)  studies, ln the  reduced to  reported greater plant. Efficient  recently, Dedhar  maintained efficient nitrification of an old, low organic  leachate. G o o d  8.2  in the control reactor of the W o n g and Mavinic  nutrients had been added. More  was  (81)  and  strength  denitrification performance was also achieved but complete  denitrification could not be maintained due to variable carbon loading to the anoxic zone. Therefore the results of these studies support the general view that efficient nitrification of landfill leachates is usually readily achieved (13,33).  21  However  the experience with nitrification of landfill leachates has not  as overwhelmingly  positive as that for carbon  reported that for a study no high  removal. Robinson  involving both a high and low organic  nitrification had occurred after a retention time of 20 days. strength leachate, influent ammonia  organic  of some of this organic  digestion of the sludge. The of the influent ammonia  strength  leachate,  In the case of the  20 days allowed for a gradual  nitrogen to nitrate, essentially by  low organic  (64,65,66)  levels were initially completely converted to  nitrogen but continued detention beyond  conversion  and Maris  aerobic  strength leachate had only a small portion  convert to organic nitrogen after 20 days retention,  reflecting the low BODj. removals, and ammonia by volatilization at the p H  losses were largely accounted  of approximately 9.3. Continued  detention beyond  days resulted in some nitrification but a total retention period of 70 days required to reduce ammonia  influent ammonia  and 100  series of experiments, Robinson  levels. N o  and Maris  appreciable nitrification was  of the micro-organisms  Therefore the SRT was  (65)  subjected  observed  in any of the  in excess  of the nutrient  remained in the effluent. This result was  not  of the relatively short SRT and lower temperature.  increased to 20 days and the units operated for a further  Most of the experimental units did not stabilize at the new SRT  while some nitrification occurred in units with excess unstable  (MLSS  units operating with a 10 day SRT at 10° C, to artificially elevated  entirely unexpected because  70 days.  solids  mg/L respectively).  units over 82 days of operation and influent ammonia requirements  20  levels to less than 1 mg/L. Very low suspended  and MLVSS were less than 200  activated sludge  for  was  levels in these units probably contributed to the poor ammonia conversion  In a previous  been  and incomplete. The incomplete conversion  and  influent ammonia, it was of ammonia to nitrite  (NC^),  and the accumulation of nitrite in the reactors, reduces the pH, as was  observed,  and inhibits the re-establishment  Maris  of stable nitrification (22). Robinson  and  22 concluded that their results did not show a fundamental reason why nitrification of leachates could not be achieved, but rather indicated that a greater degree of process control, particularly of pH, and much longer SRTs in some instances, are required to maintain a reliable nitrification process. Jasper et al. (42) also reported unsatisfactory  nitrification/denitrification  performance from their study. Initially, during the first eight weeks of the study, efficient nitrification was established in 10 and 15 day SRT units and ammonia removals in excess of 90% were achieved. However, as the study progressed, the nitrification performance deteriorated such that in the final three weeks (22 - 25), ammonia removals were just over 50% and nitrate levels in the aerobic zone were less than 10 mg/L. The denitrification performance of these units was also very poor. During the first eight week period, no denitrification took place despite the high levels of nitrate available. Following some operational changes at the end of week eight, denitrification was established and outperformed the nitrifiers, but it too deteriorated as the study progressed. On a percentage basis the denitrification rate increased towards the end of the study to over 90%, but this was due more to the reduction in nitrate levels than an increase in denitrification performance. The authors speculated that low effluent ammonia levels could be achieved at very long SRTs (>>30 days), as suggested by Robinson and Maris (65) above, but this would be due as much to other removal mechanisms (assimilation and stripping) as nitrification. It was also postulated, as mentioned previously, that the inability to establish a stable nitrification/denitrification  process was due to toxic inhibition,  possibly from accumulated heavy metals in the sludge. Not withstanding the difficulties with nitrification of landfill leachate experienced in a small number of studies, biological leachate treatment, especially by aerobic suspended growth systems, has proven to be a very effective and efficient treatment alternative for removing the majority of pollutants from landfill leachates.  23  Aerobic suspended growth treatment systems operated within limiting conditions have demonstrated complete removal of biodegradable substrates (BODj. <25 mg/L and NH  3  <1 mg/L), as well as efficient removal of suspended solids, heavy metals,  odours, and colour. Henry (33) generalizes these limiting conditions as SRTs of twice, and loading rates of half, those used for domestic sewage treatment, but it is the leachate characteristics in each case which dictate specific operating limits. The discussion above has also underlined the need for careful and detailed process monitoring and control in order to maintain process reliability and high levels of treatment, especially for reliable  nitrification/denitrification.  3.3 RBC TREATMENT Given the proven capabilities of aerobic suspended growth systems to effectively treat landfill leachates, one might question the need to evaluate other processes such as the Rotating Biological Contactor (RBC). However the aerobic fixed growth RBC process has operational characteristics and bacterial growth conditions which are distinct from other processes, particularly suspended growth systems, and claims to have various advantages over other processes, especially for nitrification. Given the wide variety of composition of landfill leachates, and the increasing emphasis on nitrifying leachates, the RBC could well prove advantageous for the treatment of at least some types of leachates. Therefore the evaluation of the RBC's performance with respect to landfill leachate treatment is worthwhile. The Rotating Biological Contactor is a very simple treatment device both mechanically and conceptually. Thin discs, or some other shaped media, which provide the surface area upon which the bacteria will grow, are mounted on a horizontal shaft over a trough containing the wastewater so that the disks or media are partially submerged in the wastewater. Generally the depth of submergence is such that 40 to 50 percent of the total surface area is underwater. The shaft is  24  then slowly rotated so that the media surface is alternately immersed in the wastewater and exposed to the atmosphere. A laminar film of water remains attached to the rotating  media and it is within this film that the bacteria grow and  affix themselves to the media surface. O n c e the media surface is covered by the bacterial growth, the surface roughness of this biomass helps determine thickness of the water film which attaches to  the  it.  During the time the thin water film is exposed to the air, gas transfer occurs across the air-water interface with for example, oxygen diffusing into the film, and gaseous products of bacterial activity such as C O 2 , diffusing out. W h e n the film of water is immersed in the wastewater, diffusion of soluble materials; substrates, products, and gases, occurs across the liquid-liquid  interface. For example,  this is considered the main mechanism by which the bulk liquid is aerated or stripped of dissolved gases. Some mixing also occurs between the bulk liquid and the attached water film but the extent  of this exchange and the thickness of  laminar film are uncertain, and depend upon other factors such as the  the  rotational  speed of the media, and the surface roughness of the attached growth (as mentioned above). The rotation of the wastewater in the trough  of the shaft and media also causes a gross mixing such that it is considered completely mixed, which  enhances mass transfer at the liquid-liquid  interface, and also contributes some  mechanical aeration of the bulk liquid, although this is considered minor and frequently  neglected (26). To complete the system, baffles are placed between  sections of disks on a shaft, or a number of troughs are connected in series, to provide a number of stages which prevent treatment  hydraulic short circuiting, and enhance  efficiency.  As RBC systems are generally staged, the process is analogous to a series of completely mixed reactors, with the important  difference that the bacteria are  retained in each reactor, and practically speaking, only the wastewater moves  25  between the different stages. The RBC  thus combines  many attributes of completely  mixed and plug flow reactor systems. These  attributes, along with characteristics  unique to the design  systems,  of the advantages One  and operation of RBC  and disadvantages  of the most obvious  the mechanics  and process  mechanical drive RBC drive components  can be used to explain many  claimed for the RBC  advantages  of the RBC  process. system is the simplicity of  itself, which results in economical operation. In a  system, operation and maintenance costs  and bearings  are simple and reliable, and the energy required to  turn the shaft is relatively small. Air driven RBCs similarly use technology, and the low air pressures  and volumes  result in even lower operating costs. Process biomass  are minimal as the  proven blower  required to rotate the shafts  control is also very simple as the  is self regulating. Control is limited to periodic monitoring of organic  hydraulic loading rates to ensure they are within design dissolved oxygen  and  limits, as well as checking  levels, pH, effluent quality, etc., to both monitor  process  performance and meet regulatory requirements. RBCs have been used widely for small package  plants because  of their economical and simple operation with minimal  operator skill and attention. Smith and Bandy RBC  technology  (1983),  2  (9290 m ) 2  averaging  2.6  hours/week,  MCD  3  One high  of the most  RBC  RBC  plants for capacities above approximately  but landfill leachate volumes  and are within the range where  ranged  kw for mechanical drive  kw for air driven units (1.2 rpm). Capital costs of  plants are higher than for activated sludge (3800 m /d)  plants  and that power requirements for  of standard density media were 3.6  units (1.6 rpm), and 2.93  1 MGD  in their state-of-the-art review of  found that hourly labour requirements for RBC  from 1 to 7 hours/week, 100,000 f t  (71)  are generally much less than 1  plants are economically viable.  important attributes claimed for the RBC  process  is its  performance for carbon removal and nitrification (84). RBCs operating within  design  limits have demonstrated their capability to produce  high quality effluents  26  with soluble B O D , . (SBOD,.) levels of less than 10 mg/L and ammonia concentrations of less than 1 mg/L (54). This effluent  quality is comparable to that  achieved by other high efficiency aerobic systems and represents a practical limit for biological treatment.  The mechanisms by which this treatment  RBC, and the factors which affect mass transfer/kinetic  models of  is achieved in the  it, are at present best explained in terms of  the  Famularo and Mueller, et al. (26,53).  It has been found, and is n o w generally accepted, that the biofilm develops up to three layers or zones of biological activity of varying thicknesses, depending upon loading conditions (48,53). The outermost growing  layer consists primarily of  rapidly  aerobic heterotrophs which utilize the carbonaceous substrates from  wastewater. W h e n conditions permit,  the  slow growing nitrifiers will predominate to  form  a second aerobic layer beneath the aerobic heterotrophs. A third innermost layer generally forms against the supporting media where anaerobic conditions can prevail as the oxygen is unable to penetrate to this depth in the biofilm. The activity in the anaerobic layer varies according to the penetration of other substrates but can include; acid fermentation available, denitrification  and methane production  if exogenous substrates are  if both carbon and nitrate are present, and endogenous  reduction of the biomass. Other autotrophs become active when conditions  such as sulphur bacteria can also  permit.  In stages receiving high loadings of organic carbon, usually the first few stages in a multi-stage very rapidly.  system, the heterotrophic  High influent  gradient for the diffusion  bacteria in the aerobic zone grow  substrate concentrations provide a strong  concentration  of soluble substrates into the biofilm. The availability of  substrate supports a large, actively growing biomass which exerts a high oxygen demand and similarly results in a strong gradient for the diffusion of oxygen into the biofilm. Growth is usually limited by the mass transfer rate of oxygen under these circumstances, rather than growth  kinetics or substrate availability, and therefore  27  oxygen  does  not penetrate beyond  the first layer. This limits the maximum  thickness  of the aerobic layer as well as the rate of substrate utilization. Though given the higher mass transfer driving forces present, the majority of the BOD,, the highest  BOD^  removal rates are commonly  observed  in these initial  (3,57,60,71). However, overloading conditions, as will be discussed  organic and inorganic are also effectively removed by adsorption and  sloughing  rate of excess  growth rate in the early stages  biomass  so that suspended  and  stages  later, interfere  with transfer rates and reduce removal efficiency. Influent suspended  on the thick biofilm. The high  removal  solids, both impingement  results in a high  solids levels are usually  highest  in these first stages. Development  of a nitrifier layer in heavily loaded stages  is generally inhibited  by the growth rate of the heterotrophs which outstrips that of the nitrifiers to the extent that oxygen  and/or ammonia are unavailable to them. Marsh e( al.  observed that nitrification does  not begin until BOD,-  levels in an RBC  reduced to 60 mg/L, and a stable nitrifier population only becomes when  BOD  5  levels approach  nitrifiers become  30 mg/L. At B O D  the dominant  similar findings and it is now  5  (47)  have  been  established  levels less than 10 mg/L the  population. The other studies reviewed all reported generally accepted that BOD,-  levels of less than 30  to 40 mg/L are required to establish a stable nitrifier population. In terms of loading rates, Kincannon  ef al. (44) found that nitrification did not begin until the  loading in a given stage was is frequently observed  reduced to 4.15  that TKN  g S B O D ^ / m * d or less. Therefore it 2  or ammonia removal in the first stages  of an  RBC  is limited to the nutrient requirements of the heterotrophs and nitrate production is minimal (34,39,44,53,58). The limited pentration of oxygen conditions prevalent in the early stages anaerobic layers. Depending  upon  into the biofilm under the heavy loading allows the formation of the thickest  the diffusion of organic substrates there may  be  28 some exogenous growth but most of the activity involves endogenous reduction of the biomass. Denitrification generally does not occur because of both the lack of nitrate production, mentioned above, and the lack of carbon penetration into the anaerobic layer due to its rate of utilization in the aerobic layer. Under very heavy loading, anaerobic conditions become stable enough to allow the growth of sulphur reducing bacteria. This generally leads to operational problems as will be discussed further below. In stages receiving moderate loadings of organic carbon, corresponding to intermediate stages of a multi-stage system, biological activity in all three layers generally interact in relation to the availability of the various substrates. The growth rate of the aerobic heterotrophs is usually substrate limited and reduced from that observed in the previous stages or at higher loading rates. This results in a thinner biomass and increased penetration of other substrates and oxygen into the biofilm, as mass transfer rates exceed the rate of utilization in the first layer, ln particular, oxygen and ammonia can diffuse into the biofilm to a depth at which the slow growing nitrifying bacteria can outgrow the heterotrophs to form their own layer. The greater diffusivity of ammonia over carbon substrates aids in this process (36). Nitrates produced in this layer then diffuse outwards in both directions in response to concentration gradients. Within the anaerobic layer, which underlies the nitrifier zone, considerable denitrification can occur with nitrates readily available and using carbon from residual substrate or endogenous respiration of the heterotrophs (1,48,53). The substrate limited growth rate of the aerobic heterotrophs in these intermediate stages derives from the reduced influent soluble B O D ^ concentrations. Low B O D j concentrations in the bulk liquid result in even lower concentrations in the biofilm because of the gradient required to drive the mass transfer. The lower mass transfer rates and heterotrophic growth rates result in lower B O D - removal  29  rates and reduced removal efficiency being observed  in the intermediate stages or  at lower loading rates (57,60). A contributing factor to the lower heterotrophic growth rates is that the soluble fraction of the organic load entering these later stages contains a higher proportion of compounds readily utilized substrates However  having  resistant to degradation, the  been preferentially consumed  in these later stages, many  slower growing  in the initial  stages.  and/or specialized bacterial  populations can compete more effectively to remove much of this material. A significant portion of the organic load to the intermediate stages from the suspended generally become  or sloughed  solids from the previous stages. These  re-attached to the media in subsequent  solids  stages (43) and the extra  retention time under substrate limited conditions encourages  the degradation of the  entrapped organic solids and endogenous reduction of the excess biomass. endogenous reduction of the biomass RBC  reduces the overall sludge  process, which is another advantage  production rate of 0.37  treating domestic sewage.  However the sludge  landfill leachate w o u l d increase somewhat precipitates as was  observed  The  production of the  claimed for the system. Kincannon  (44) determined a sludge  comes  kg solids/kg BOD,,  et al.  removed,  production for an RBC treating  due to the addition of inorganic  for suspended  growth  systems.  ln stages receiving very light organic loadings, corresponding to the last stage  of a multi-stage system, nitrifier activity predominates and the biomass  is very  thin reflecting their low growth rate. Heterotrophic activity is limited primarily to endogenous respiration of re-attached biomass  from previous stages. Denitrification is  also limited by the lack of carbon and the penetration of oxygen. efficiency of all substrates reduced because  Removal  under these severely substrate limited conditions is further  the low concentration gradients provide a very small driving force  with which to overcome the mass transfer resistances. This explains why it is not possible to reduce soluble substrate concentrations to zero.  30 From the discussion above it follows that efficient achieved with the  B O D , , removal can be  RBC process provided that the sizing, and number, of the stages  is adequate. Since B O D ^ removal rates increase with the applied loading, B O D ^ removal can be enhanced by increasing the proportion  of surface area in the first  stage compared to the subsequent stages. This factor is frequently RBC plant design; for example the pilot  incorporated  into  plant used in this study had twice as  much surface area in the first stage than in each subsequent stage. Conversely the decreasing B O D , , removal efficiency with successive stages, or diminishing means that the maximum number be  of stages for  return,  B O D ^ removal is generally taken  to  four. Configuring an RBC plant to  stages not  only improves the  performance  achieve maximum  BOD,- removals in the  B O D j removal efficiency but the  nitrification  as well. As discussed above, the onset of nitrification  upon the prior earlier in the  removal of most of the carbonaceous substrates and therefore  stages. In cases of high influent  concentrations, or the treatment  the  nitrification  B O D , , and/or ammonia  of substrates which are difficult to degrade (such  as phenols), additional stages may be required to high effluent  is dependent  process this occurs, the more surface area is available for  in the following  early  achieve complete treatment  quality. As will be discussed in Section 5.0, landfill leachates  or  frequently  have one or more of these characteristics. The attached growth nitrification  over suspended growth  as being sessile (22,24) their  growth.  of the RBC provides several advantages for systems. Nitrifying  and therefore  Crowing attached to the  long solids retention adequate time nitrification,  nature  times which  bacteria are generally described  the RBC provides a prefened environment RBC media also results in  allows the slow growing  to develop. Since only two  species of  nitrifiers  indeterminately more than  bacteria are involved in  Nitrosomonas and Nitrobacter, they have proven more sensitive to  for  31  various environmental conditions such as, temperature, pH, and inhibitory levels of substrates or other substances, than the more diverse and adaptable populations responsible for carbon removal or denitrification. Therefore, the relatively protected location of the nitrifiers, within an interior layer of the biofilm, usually in the later stages of the process, could significantly improve the stability of the nitrifier population. As will be discussed further below, the fact that the biomass is attached also greatly reduces the possibility of the slow growing nitrifiers being washed out of the system. When all these factors are considered, the fixed growth RBC system provides an environment far more conducive to the growth of a stable nitrifier population. Therefore a more reliable and efficient nitrification process should result. The attached nature of the biomass also greatly enhances the settling of the effluent suspended solids which are frequently mainly nitrifiers. It is generally reported that nitrifying bacteria from suspended growth systems are finely dispersed and settle very poorly (22,66), which aggravates the wash out problem in these systems. However in the RBC process, the effluent suspended solids are concentrated in chunks sloughed off of the media and these chunks generally settle well, even when they are composed mostly of nitrifiers. In the case of leachate treatment, effluent suspended solids settling is further improved by heavy inorganic precipitates. The mass transfer mechanisms can also be used to explain the observed reduction in BODj. removal efficiency as the size of the RBC is increased. Kincannon ef al. (44) found there were no scale-up effects for overall loadings of less than 4.9 to 7.3 g S B O D / m * d or first stage loadings of less than 12.2 g 2  5  SBOD[-/m *d. However, at loading rates higher than these levels, full scale RBC 2  units became oxygen limited and lost removal efficiency sooner than smaller scale units. Similarly Wilson, Murphy, and Stephenson (54,80) found that a 0.5 m diameter  32 RBC  unit achieved 15 percent higher C O D  removals than a 2.0 m diameter unit  and assumed that a further 10 percent reduction in performance w o u l d occur in 3.5 m  diameter units. Famularo et al. (26)  expained that since the peripheral velocity of  the media is kept constant in most studies, to limit shear forces, the rotational speed  of the shaft decreases  as the diameter of the media increases. Therefore the  period of rotation increases with diameter and an element of surface area is either immersed or exposed  to the air for increasing periods of time. W h e n this period  of rotation is t o o long for the rate of growth of the bacteria, the substrate concentration within the liquid film will be depleted and the biomass  will  become  inactive for part of each cycle, reducing the removal efficiency. Presumably,  oxygen  would be similarly depleted for some portion of the immersion cycle, further reducing the activity of the biomass. As  indicated by the results of  et al. (44), at lower loading rates the growth rate of the biomass  Kincannon is not sufficient  to cause this effect. It follows from the above discussion  that if rotational speed were kept  constant, scale-up effects would be reduced or eliminated. Results such as those Friedman et al. (29)  indicate that to some extent this is true. They found that as  the rotational speed, and thus that the maximum  of  peripheral velocity, of their RBC  units were increased  removal rate increased. Since their units were of the same  diameter (11.88 inches) this improvement would be primarily due to increased mass transfer rates caused  by the increased mixing and turbulence at the biofilm interface,  rather than overcoming  the substrate depletion decribed above. In a full scale  RBC,  both of these factors would improve performance but rotational speed is limited by the need to keep the peripheral velocity within acceptable limits so that the hydraulic shear doesn't  strip off the biomass.  More  or less out of tradition this  limit has been set at 0.3 m/s or 1 ft/s, but Friedman et al. demonstrated that up to a 50 percent increase in this value may prove practical. However  power  costs  33  also increase with the rotational speed. It has  been found  that nitrification within the RBC  proceeds  at a constant,  temperature dependent, rate per unit area of nitrifying bacteria and thus does not exhibit any scale-up  effects (44,53,54,80). Kincannon  constant nitrate production rate of 0.73 (54) calculated a constant TKN constant  et al. (44)  g NOym^d.  removal rate of 1.12  determined a  Similarly Murphy and  Wilson  g T K N - N / m * d at 20° C. The 2  reaction rate is due to the slow growth rate of the nitrifying bacteria and  their zero order response that substrate is the number  to substrate concentrations above  about  0.5 mg/L.  Given  utilization is reaction rate limited, the other factor limiting nitrification of nitrifying bacteria. It appears  stages in which it develops,  as though  the nitrifier layer, in those  grows to a fairly uniform thickness, limited by reaction  rates, endogenous decay, and to a lesser extent predacity, and mass transfer limits. This would explain the strong  areal relationship of nitrification in the  Another important advantage  claimed for the RBC  resistance to shock organic, hydraulic, or toxic loadings  process  RBC.  is excellent  (54,81). Of  these, the R B C s  resistance to hydraulic loading is the easiest to explain. As the bacteria are attached to the fixed media they are much a comparable usually do  suspended  growth  not occur. Upsets  less prone to wash-out  system  of the process  and therefore major losses of  in a suspended  growth  process  than in  biomass  typically result in  reduced solids settleability and high hydraulic flows would only increase the losses of biomass system.  into the effluent. Biomass losses are particularly serious  In a RBC  system  however,  process  performance can be impaired due to a  reduction in the hydraulic retention time (HRT). RBCs designed generally have  short  which is considered the conditions  in a nitrifying  for sewage treatment  HRTs, in the order of 0.5 - 2.0 hours at design another advantage  of the process  of a hydraulic surge, the system  value. Filion et al. (27)  HRT  flowrates,  itself (84). Therefore  under  could fall below some limiting  investigated the effects of variable hydraulic loading on an  34  RBC  and found that varying the HRT between 0.44  and 0.94  significantly affect carbon removal, but nitrification performance variations between much  more  1.12  and  3.36  hours.  However  hours did not did respond  nitrification performance  to  HRT  recovered  quickly from hydraulic loading fluctuations, than from changes in TKN  loading or influent concentration. Poon B O D j removal was  adversely  ef al. (60) found that for their system,  affected by a reduction of the HRT  hours, but low influent B O D j  from 0.73  0.42  concentrations were cited as a contributing factor  (lower reaction rates). Therefore, while the RBC  provides  biomass  performance, especially for  during  to  high hydraulic loadings, process  g o o d resistance to loss of  nitrification, may be reduced by lower hydraulic retention times. The  RBC's resistance to shock organic  mass of highly concentrated micro-organisms previously the first stage than subsequent  of an RBC  or toxic loads stems  from the large  resident on the fixed media. As  noted  is typically larger or has more surface area  stages to maximize the mass of bacteria in contact with influent  conditions. In the case of peak assimilative capacity necessary  organic loadings,  to absorb  With respect to toxic loadings,  this large biomass  provides  the  extra substrate over short term periods.  toxic effects become  manifest when the levels of  toxin exceed a critical ratio to the mass of bacteria, rather than reach a specific concentration. Therefore the greater the biomass concentration of toxin which suspended somewhat  growth more  present, the greater the  can be tolerated. Theoretically, a completely mixed  system with the same total biomass  effective at resisting shock loads  as an RBC  because  would  all of the bacteria would  be available to moderate the shock conditions. By the same reasoning would be more  effective than a plug flow suspended  growth system  higher proportion of biomass  near the influent end of the process.  pointed out above, the RBC  system  bacteria become  upset.  stressed  or  is much  be  an  RBC  due to a However,  less prone to losses of biomass  as if the  35  The results of Filion ef al. (27)  tend to confirm the greater responsiveness  of a completely mixed suspended growth system; a side by side comparison of comparable activated sludge and RBC systems indicated that the effects of loading peaks on effluent  quality were more pronounced and lasted longer in the RBC  system. As might be expected, the impact of peak TKN loadings on nitrification performance were roughly three times greater than the effects of peak TOC loadings on carbon removal. Peak loading rates of 24 - 27 g T O C / m * d and 6.0 2  7.2 g TKN/m *d obviously exceeded the assimilative capacity of their RBC system. 2  The recovery times of roughly one hour for carbon removal and three hours for nitrification indicate that longer HRTs could significantly increase the RBCs resistance to shock organic loads. Poon ef al. (60) from 4.3 to 15.4  found that for shock organic loadings  g SBOD^/m *d, representing 124 to 444 percent of the normal 2  applied load, that no adverse affect on the RBC unit performance was observed. -  They conclude that these loadings were within the assimilative capacity of the RBC since other studies had demonstrated removals of up to 17.0 g SBOD,., or 28.3 g total BOD,., per m * d . Therefore, the RBC has demonstrated good handling of 2  shock organic loads within its assimilative capacity and reasonable response to even higher loads. Again the main advantage may be the RBCs resistance to biomass losses, which often plague suspended growth systems during shock conditions. Resistance to shock organic loadings is very important with respect to leachate treatment because as will be discussed fully in Section 5.0, landfill leachate composition and flowrates can be highly variable, and peak organic and hydraulic loadings frequently occur coincidently. Landfills also often receive various types and amounts of toxic material, knowingly or otherwise, which can end up in the leachate. One of the main disadvantages of the RBC process is its sensitivity to low temperatures. The same large surface area and very thin water film in contact with  36 the atmosphere  which maximizes  gas  transfer also maximizes the potential for heat  transfer. Since as a general rule of thumb  the rate of biological reactions is  reduced  in temperature, the rapid cooling  by half for each 10° C  decrease  influent wastewater can significantly decrease  process  weather the heat transfer efficiency of the RBC  performance. During  warm  is beneficial as influent wastewaters  are generally cool. The heating effect of warm temperatures is moderated by evaporative cooling which Kincannon temperatures by 2 - 3 °C across  et al. (44) observed  an RBC  of  unit. During  could  somewhat  reduce  cold weather  conditions  however, the situation is deleterious as the RBC will efficiently lose heat to the atmosphere  and evaporative cooling further aggravates  the problem. Therefore, RBCs  installed in areas subject to low temperatures (less than 10° C) are usually fitted with insulated covers  to reduce heat loses. Under very cold conditions they  be covered to prevent icing The reduction in RBC well documented  problems. performance as water temperatures decrease has  (28,54,57,80,85). In most  the reaction rates of the RBC  have  equation coefficient 8. For carbon 1.11  Many of these studies above  showed  10 to 15° C, so  (28,54). Nitrification has  0 = 1.05  in terms of an  temperature ranges. Wilson  over the temperature range  from 1.03  for  of 5.5 to 13°  C.  corrections are not usually applied at higher  increase temperatures  to be much more sensitive to temperature  of temperatures. The temperature coefficients  determined for nitrification are therefore generally larger than for carbon alone, typically 1.09 factor of  0 = 1.09  to 1.11.  to  et al. (80)  that the carbon removal rate does not  effects and over a wider range  on  Arrhenius  removal, coefficient values ranging  been observed  been  instances the effects of temperature  been expressed  have been determined for various  example, used a value of  must  removal  Murphy et al. (54) for example, determined that a  applied for carbon  which no further correction was  removal with nitrification up to 20°  required. However,  C,  as pointed out by Forgie  above (28)  37 the use  of Arrhenius  coefficients to correct for temperature effects is not strictly  correct. Although  the Arrhenius  equation and coefficients are widely used  and  accepted for correcting reaction rates for temperature effects, Forgie recalls the fact that  6 itself is a function of temperature and that the use of a constant  an approximation. Therefore the use  of a constant  6 is only  6 over a relatively wide  temperature range could lead to a significant error. Secondly,  he pointed out that  the form of the equation is exponential, which implies that reaction rates increase continuously with temperature, and conversely, that temperature effects decrease the temperature decreases.  As  indicated by Forgie, a number  of studies have  that, in fact, the reaction rates drop off more sharply as the temperature  as  shown  decreases;  thus temperature effects increase with decreasing temperature, and also reaction rates level off at warm temperatures, rather than increase continuously. Therefore, he concludes  that the Arrhenius  coefficients provide a reasonable  actual temperature effects only when  approximation  used over small 4 to 5° C  of  temperature  ranges. Forgie presented an empirical curve fit equation from experimental data, which indicated a parabolic Experiments  shape.  conducted  results. The first was well at temperatures  by Forgie  also produced  a couple of other interesting  that an established nitrifier population could continue to nitrify as low as 1° C. However, these low temperature runs were  only maintained for short  periods  of one or two weeks  so as Forgie conceeded, it  is uncertain whether or not this performance could be maintained at this low temperature. The second  point illustrated by his results was that hydraulic retention  time had an influence on the temperature effects. Specifically, longer HRTs reduced the adverse  effects of low temperatures  and restored some of the  process  efficiency. This effect is atttributable to longer contact times between the wastewater and biomass, offsetting the reduced reaction rates. W u  ef al. (83,84) also  found  38 HRT to be a factor in RBC performance and therefore included it as a parameter in their empirical models of the RBC process. Another disadvantage of the RBC process is a history of operational problems, poor performance, and mechanical failures, which has made engineers wary of RBCs. Most of the mechanical problems can be attributed to early RBC installations in which poor design and fabrication of the units, resulting from inexperience with the weight of biomass which could accumulate and the forces involved, lead to numerous shaft and media failures. Although RBC design and manufacture are greatly improved, mechanical failures still occur periodically, usually in units which have been continuously overloaded and suffer fatigue failures. Other operational problems and poor treatment performance have also generally resulted from overloading conditions, frequently from underdesign. Some early designs were based on the performance of pilot scale studies by designers not cognizant of the scale-up effects which reduce performance. In other instances, RBC manufacturers have used overly optimistic design factors for competitive reasons. This underscores another disadvantage of the RBC process; that RBC design is still proprietary, which makes comparison and evaluation of RBC units and designs from different manufacturers difficult  (71).  A variety of operational problems have been observed in RBC units overloaded hydraulically or organically. As discussed previously when response to shock loading was considered, hydraulic overloading results in incomplete treatment and thus poor system performance. Organic overloading on the other hand can lead to a number of unpleasant conditions. Excessive organic loads, which occur quite frequently in the initial stages of RBC plants, cause overgrowth of the biomass, both on the media and eventually in suspension. The excessive growth  frequently  results in the biofilm bridging the gaps between the media surfaces which reduces the active surface area by restricting the access, and thus the transport, of oxygen  39  and substrates  (25). Bridging of the biomass  also reduces the ability of hydraulic  shear to control the biofilm thickness and the rate of biomass  sloughing  is greatly  reduced. The excessive growth rates drastically reduce the dissolved oxygen levels in the  bulk liquid which, combined with the bridging effects, allows most of the  biomass the  to become deeply anaerobic, severely reducing reaction rates in most  biomass  of  and thus it contributes very little to the removal performance of the  system. Within the anaerobic zone, sulphur and hydrogen and this encourages to  the growth of Beggiatoa  produce energy. These  micro-organisms  sulphide is often produced  bacteria which oxidize these  store sulphur in their cells giving them a  white milky appearance, which is characteristic of overloaded RBC addition to the reduced BOD,-  severe  odour  stages (25,34). In  removal indicated by the Beggiatoa growth, the  sulphuric acid they produce can lower the system pH nitrifying organisms.  products  and adversely affect the  Anaerobic conditions in one or more stages can also  cause  problems.  The excess  biomass  in overloaded stages can dramatically increase the weight  of the shaft and media, thus greatly increasing the stress in these structures. Since the  shaft is rotating, the stresses  are cycled continuously and the effects of fatigue  are multiplied, reducing the life expectancy of the shafts and media. The weight of the shaft also requires significantly more energy to turn it so costs  are increased while performance decreases. Therefore, overloading  generally result in much  a number  of RBC  that for plants in which first stage  BOD,-/m *d no 2  conditions  plants for operational problems  loading were less than 17.6  problems were reported, but for plants with first stage  greater than 43 g BOD,./m *d, problems 2  loadings  energy  higher operating, maintenance, and replacement costs.  Evans (25) surveyed found  increased  between these two extremes, no  and  g loadings  always occurred. At plants with first stage clear pattern was observed and the  40  occurrence of overgrowth conditions was attributed to other factors such as; wastewater characteristics, rotational speed, temperature,  tank design, and media  configuration. It was statistically determined from the survey results that a first stage loading of 35.6  g BODj/m^*d had a 50 percent probability of operational problems.  Various modifications of the standard mechanical drive RBC system have been employed to alleviate overloading problems associated with the initial stages and/or improve overall system performance. Step-feeding, de-staging, internal recycles, and supplementary air diffusers, have been used to reduce first stage loadings or prevent oxygen depletion. De-staging and step-feeding are two very similar methods of reducing the loading rate in the initial stages. They involve rearranging the process flow path so that the initial stages operate in parallel, or splitting the influent flow between the initial stages which still operate in series, respectively. Internal recycling of aerated wastewater from the last stage of the RBC back into the first stage both reduces the loading in the first stage by dilution of the influent and adds dissolved oxygen. However, internal recycles are only beneficial to process performance when used to alleviate an oxygen deficiency, otherwise the dilution of the influent reduces removal efficiency by decreasing mass transfer gradients and the HRT in each of the stages (3,55). Supplementary air diffusers placed in the initial stages both prevent oxygen deficiency, and produce additional turbulence in the wastewater which helps control biomass thickness and prevent bridging. One of the most effective developments in RBC technology has been the use of air-driven RBCs, which expand upon the benefits of supplemental aeration. Hynek and Chou (39)  conducted a comparison study of air and mechanical drive  RBC units and reported a number of advantages associated with air driven units. Diffused air introduced from the bottom of the tank bubbles up through the media and becomes trapped in cups on the periphery of the disks to cause the rotation. The diffused air both aerates the bulk liquid as well as causing increased mixing  41  and turbulence at the interface with the laminar water layer and biofilm. This increased turbulence and air contact allows aeration of the biofilm to occur the submerged  during  cycle and causes increased shear forces on the biofilm which  prevents excessive  buildup. The  result is a thinner, more active biomass.  Air drive  RBCs have g o o d resistance to oxygen depletion in the initial stages at high rates and Hynek  and C h o u  filamentous micro-organisms.  observed The  greatly reduced growth of Beggiatoa and  in the shafts  was also enhanced  because  stages leaving more  and media due to the thinner biomass.  BOD^  removal was  benefit of air driven  possible  in covered units. The one  disadvantage  however, this may be controlled somewhat the various  the treatment capacity and design  of  stages.  RBCs is that heat recovery from the blower air is of air drive RBCs for landfill  leachate treatment could be that the diffused aeration would promote  discussed  and  Nitrification  achieved in a fewer number  surface area available for nitrification in the remaining  Another  Having  other  air drive RBCs also permit easy regulation of  rotational speed, rotate slower for a given removal rate, require less energy, develop less stress  loading  foaming;  by the media.  properties and characteristics of the RBC  performance  of the RBC  remains  system,  to be stated. Early  specifications for RBCs were in terms of hydraulic loading rates determined  from manufacturers efficiency. This  nomographs, using  design  approach  permit simple comparisons  the waste strength and desired removal  did not easily adapt to differing waste types  of loading levels. Kincannon  prompted to introduce the total organic  and Stover were  thus  loading concept in the early 1970s and it  has since gained wide acceptance (44,54). Therefore, RBC capacity are usually expressed  or  as mass of substrate  loading rates or treatment  applied or removed, per unit  area of media surface. Design specifications are also now from the previous  generally expressed  terms of organic  loading rates although  discussions,  hydraulic loading  rates, which are temperature dependent, should  in  maximum  be specified to  42  ensure an adequate  HRT  for efficient treatment. These  HRT limits would generally  only apply to low strength wastewaters or low temperature nitrification. From the studies reviewed, the capacity of the RBC  to remove  substrate ( B O D ^ ) and achieve complete treatment (effluent B O D , ^ 2 5 the order of 15 - 18 g  B O D / m * d , at temperatures 2  5  a g o o d effluent quality and removals  in excess  BOD,-/m *d and at 15° C. Kincannon  of 9 0 %  than 9.8 g S B O D / m * d , soluble effluent B O D 2  5  g SBODj./m *d, 2  e  mg/L) is in  C. Forgie (28)  at loadings  achieved  up to 15.2  g  ef al. (44) found that at loading rates less  2  loading of 18.3  >15  carbonaceous  5  <10  which corresponds  mg/L were achieved, but at a  to a total BOD,-  loading  considerably higher, the removal efficiency was only 53%. Paolini and Variali (58) found that their effluent quality deteriorated at loading rates greater than 19 g BOD,-/m *d. Poon 2  et al. (60) treating a clarified trickling filter effluent (tertiary  treatment) achieved an average effluent S B O D ^ of less than 15 mg/L at loadings  up  to 7.8 g S B O D i j / m * d . Murphy et al. (54) found that g o o d treatment efficiency 2  could be achieved up to a loading of 15 g B O D g / m * d after which some scatter 2  in the results occurred. This led them to recommend g  BODj/m *d 2  for temperatures of 15° C  a design  loading rate of  15  or higher, which compares favourably with  many other design  factors and practical experience, as Evans (25) found that all the  plants he surveyed  had loading rates  at <12  <19.5  g B O D g / m * d and most were operating 2  g BOD /m *d. 2  5  Some of the modifications or variations of the RBC removal capacity of the RBC  somewhat  beyond  system can increase the  these levels without reducing  effluent quality. In particular, air driven RBCs and the use of oxygen enriched systems have demonstrated  higher capacities. Hynek  and mechanical drive RBCs found carbon removal and 2 5 % recommended  and C h o u  that air drive units were 1 8 %  RBC  (39) comparing air more efficient for  more efficient for carbon removal with nitrification, but  designs with 7 and  5 percent higher loadings for each  mode  43  respectively. Huang and Bates (38) investigated the potential benefits of enriched oxygen environments with pressurized air and pure oxygen RBC systems. As indicated by Famularo et al. (26) earlier, since the RBC is typically limited by the mass transfer of oxygen, the RBC could benefit  appreciably from an oxygen enriched atmosphere  as predicted by their model. Greatly increased biomass thicknesses resulted from the oxygen enrichment, particularly with pure oxygen under pressure, but increased C O D removal was not consistently observed. The lack of improved C O D removal in the first stage was blamed on severe bridging of the biomass, but C O D removals in the second stage were increased by the oxygen enrichment. Nitrification was observed to be improved by pressurized air, but the use of pure oxygen, especially under pressure, resulted in inhibitory high dissolved oxygen levels. They concluded that oxygen enrichment would prove beneficial for C O D removal if the RBC was modified to prevent bridging. The RBC system is also capable of higher removal rates if complete treatment is not required. Mikula et al. (52) found that an RBC treating dairy wastewater was capable of 71.1%  C O D removal at a loading rate of 38.5 g  C O D / m * d (27.4 g C O D / m * d removed). At this high loading rate the fourth stage 2  2  accounted for 18 - 30 percent of the total removal. Poon et al. (60) reported that BOD,, removals ranging from 17 to 28.3  g BOD^/m *d had been found in the 2  literature. Higher removal rates are achieved at higher loadings by making more efficient use of the later stages for carbon removal. Since the BOD^ loadings to the later stages are increased, higher carbon removal rates are achieved at the expense of nitrification, which will be inhibited. As mentioned previously the capacity of the RBC for nitrification is determined by the surface area participating in nitrification. Nitrification then occurs at a fixed rate, established by Murphy et al. (54) to be approximately 1.2 g  44 T K N / m * d at 20° C. Their design 2  loading recommendations  carbon removal are presented in Table 9.1 As  mentioned at the beginning  of the  Discussion.  of this section, a review of the literature  prior to the start of the experimental program concerning RBC  for both nitrification and  failed to find any references  treatment of landfill leachate. However a second  review of the  literature, conducted after the protracted experimental phase, did yield a couple of studies on this topic, by Ehrig (22,24) and Coulter (16). Ehrig reported on the treatment of three different old, or methanogenic the RBC was  phase, leachates and found that  capable of almost complete nitrification of these leachates at loading  rates up to 2 g N/m *d. Coulter reported some results of a companion study to 2  this one and found that efficient carbon removal was achieved at loading rates of 9.6 and 18.3  g COD/m *d 2  (6.2 and 11.6  nitrification was also observed  g B O D / m * d ) . An interesting lack of 2  5  during this study. These  papers are discussed  fully, within the context of the results of this study, in the Discussion, The scarcity of studies concerning  RBC  more  Section  9.0.  treatment of landfill leachates was  confirmed by Chian et al. (13) when their review of the literature failed to find enough data to enable them to present ranges of expected treatment efficiency for aerobic fixed film processes.  However, as reported by Ishiguro (40), and Masuda  ef al. (48), the dearth of experience with RBC  treatment of landfill leachates does  not apply to the Japanese literature. Ishiguro noted that Japan has had extensive experience with RBCs, treating mostly industrial wastes, and as of 1983 than 1600  RBC  more  plants installed. He also reported that there were 135 plants treating  landfill leachate, the first having effort should  had  been installed in 1976. Therefore, it seems an  be made to benefit from their experience as translation is much  economical than research.  more  4. EXPERIMENTAL  The experimental program three main component  proposed  PROCRAM  to fulfill the purpose  parts. First among these was  of this study  the characterization of the  Premier Landfill leachate, both to determine the constituents of the RBC process  had  evaluation, and to provide a basis for comparison  influent for  of the treatment  experience from this study to other leachates and waste treatment situations.  Of  primary interest was the carbonaceous  and nitrogenous  content of the leachate, as  these are the main fractions removed  by biological treatment; however, several other  tests were conducted to determine typical levels of selected heavy metals and specific trace organic  compounds. In addition to the chemical analysis of the  leachate, physical properties such as total solids, specific conductance, pH, temperature were monitored The  second  some  and  regularly.  part of the experimental program  evaluation of the capacity of the RBC leachate. For simplicity, the RBC  was concerned with the  to effect carbonaceous  removal from this  would be operated under pseudo  conditions, for which the flow rate would  steady-state  be set and the influent strength,  temperature, etc., would be allowed to vary naturally. In order to determine the maximum was  capacity, or mass loading rate, of the RBC for carbonaceous  proposed  to begin  removal it  operation at a low mass loading rate (and therefore low  flowrate), and then increase the loading rate by increments until the effluent quality deteriorated, indicating an overloaded condition. The starting flow rate and size of the incremental flow increases would be determined by the leachate measured would  strength  prior to each change in flow. After each increase in loading the  be allowed to stabilize over a minimum  45  RBC  period of three weeks. A nutrient  46  solution of phosphoric acid (H^PO^), and when necessary, ammonium chloride (NH^CI), would be added to the first stage to maintain a nutrient level in excess of 100:5:1 BOD,-:N:P, so that no nutrient deficiences would limit growth (4,7,18,66). The performance of the RBC would be monitored by twice weekly sampling of the RBC influent, effluent, and operating parameters. The third part of this study concerned the evaluation of the capacity of the RBC to nitrify this leachate. It was intended to set the influent flow rate such that the carbonaceous loading rate was approximately 25% of the maximum capacity determined previously, and then to vary the ammonia (NH^-N) loading rate with additions of ammonium chloride (NH^CI) . The ammonia loading rate would be doubled during each increase until the effluent ammonia levels indicated overloading ^.conditions. Then the loading rate would be adjusted downwards to find the maximum capacity. Again the RBC unit would be allowed to stabilize at each loading level before the next change was imposed. The later two parts of the experimental program outlined above would probably have worked well to provide the data necessary to evaluate the performance of the RBC if it had proceeded as planned. However, mechanical problems, natural calamity, and variable leachate strength, imposed numerous upsets and operational changes such that no orderly progression of loading rates could be maintained. In practice, the experimental program involved operating the RBC as steadily as possible during the periods between upsets. Variation of the loading rate was accomplished largely by the natural variation of the leachate strength, although changes in the influent pumping rate were also made. A drastic reduction in the BOD^:NHj ratio of the leachate during the course of the study made additions of NH CI 4  unnecessary for both nutrient requirements and the evaluation of nitrification.  Efforts to evaluate the carbonaceous removal capacity of the RBC were exasperated by a decline in the carbon content of the leachate. Complete details about the  47  variation of leachate strength and the operation of the RBC  are given in Sections 5  and 7 respectively. In addition to the three main parts of this study, a number of smaller topics received a cursory investigation. These settleability of suspended  topics include: the generation  and  solids, the removal and fate of some heavy metals, the  presence of several trace organics, and the effects of variable and intermittent hydraulic and organic loading rates. Observations on  specific test results and in part based  4.1  SAMPLING A N D  ANALYSIS  on these topics were in part based  on general operating data.  PROGRAM  An extensive sampling program  was set up to characterize and monitor  various raw leachate parameters as well as monitor the performance of the (Table 4.1). The sampling program measurements  was  based upon  grab samples  RBC  and field  taken during twice or thrice weekly visits to the landfill. Since the  landfill is a 45-60 minute drive from the University, it was considered impractical to go  to the site on a daily basis. Automated sampling was ruled out because of a  lack of resources. This would have been an expensive alternative because the installation was the sampling  located beyond  any supervision and therefore a secure enclosure for  equipment would have been required to thwart vandalism, (for which  there was precedent). Since the period of the study was expected to be weeks, it was  many  assumed that the twice weekly samples would provide sufficient data  to evaluate the treatment efficiency of the The  RBC  RBC.  main parameters used to characterize the raw leachate were: chemical  oxygen demand  ( C O D ) , ammonia nitrogen (NH^-N), specific conductance  Cond.), total solids  (TS), and pH. Sampling  (Sp.  of these parameters had begun in Oct.  1982, when weekly grab samples were collected as part of another study.  During  48  Table  4.1  Sampling  Program  - Samples to be collected twice weekly: (Tues. & Fri.) - Twice Weekly Procedures: (each site visit) - Check & Record:  -  - Sample: - influent - 1  s t  influent flow rate nutrient flow rate influent temperature (raw leachate) 1 & 4 stage water temperature s t  t n  (raw leachate)  & 4  t h  stage  (raw) (settled) (filtered)  COD,TKN,NH ,Sp.Cond.,TS,pH 3  COD,TSS/TVSS,pH COD,TSS,TVSS,TKN,N H , N O 3  B  COD  - Once Weekly Procedures: (in addition to above) - Sample: - influent 1 ,. 2 , & 3 stage st  n d  (raw leachate) (raw)  BODj.,TOC,aIk. BOD  5  rd  - 4  t h  stage  (settled)  BOD ,TOC  (filtered)  BOD^alk.  (raw)  5  BODj  (settled)  BOD ,TOC  (filtered)  BOD ,P0 ,alk.  5  5  4  - Measure: - D.O. levels in all stages - depth of biological growth on all stages - Before each change in loading: - scrape off areal sample from each stage for biomass determination - collect biomass samples from each stage for nutrient (N,P), and heavy metal analysis - Periodic Samples: - collect samples of raw leachate, 1 & 4 stage liquid for metal analysis, attempt to sample at high, medium, and low leachate production rates s t  t h  49  the course  of this study, additional tests were performed at various times for:  biochemical oxygen demand C1-C3  (BOD^), total organic carbon (TOC), volatile fatty acids  (VFA), total Kjeldah! nitrogen (TKN),  combined with the previously mentioned constitute a fairly thorough While the sampling  and alkalinity (Alk.). These  heavy metal and trace organic  and analysis  performance did not proceed entirely as  of Table 4.1 were generally earned out  proposed; however, the remainder of the sampling less frequently, or discontinued. These by the operational problems  reductions  procedures  were either performed  in the sampling program  of these extra tests and samples,  reduced the time available to make these tests, as maintenance took precedence. However  the quantites which were measured  other circumstances, was Thus, the RBC  suspended  process  for most  solids  measurements  was  suspended  and  procedures  also  often  on the twice weekly  data, while desirable  for  under  monitored primarily by sampling liquid from the samples  of the same parameters  for: combined  whether raw, settled, or filtered, as the raw leachate. ln addition,  nitrate and nitrite nitrogen ( N O ^ + N O j  (TSS), and total volatile suspended  for the most  the first and fourth stages, thickness  and  not central to the goals of this study.  first and fourth stages of the unit. These  they were analysed  were  important with respect to evaluating the R B C s performance  carbon removal and nitrification. The other supporting  were analysed  as  alluded to earlier. The irregular operation of the  reduced the significance of many  basis were the most  analyses,  of the raw leachate proceeded as outlined  of the R B C s  planned. The twice weekly procedures  RBC  when  characterization of this landfill leachate.  in Table 4.1, the monitoring  caused  analyses,  part consisted and making  colour of the biomass,  solids (TVSS). Field  of recording the liquid temperature of  notes  on visible changes in such factors as;  foaming within the RBC,  settleability of the  solids, and effluent clarity. The only significant change  weekly routine of Table 4.1  was  -N), total  made to the twice  the twice weekly, rather than weekly,  measurement  50 of BODj. during the later half of the study. This change was made after it became apparent that BOD was a better indicator of process performance than C O D , because of the relatively high levels of refractory C O D which persisted in the RBC effluent. This refractory C O D could conceivably mask the breakthrough of degradeable C O D when the process became overloaded. The rest of the sampling and analysis program was carried out to varying extents in response to changing conditions and priorities. Samples from the intermediate  stages, stages two and three, were collected during the first three  weeks of the study and then stopped when operational problems developed. These samples, in combination with those from the first and last stages, were intended to monitor the progression of treatment through the RBC unit. The results of the first three tests, and subsequent data from sampling the first and fourth stages, indicated that most of the treatment was occurring in the first stage and that there were only slight changes in the liquid quality between the first and fourth stages. Therefore, it was decided to discontinue the intermediate  sampling until the data  from the first and fourth stages indicated that measureable changes in liquid quality would occur between the individual stages. The necessity of the  intermediate  sampling was not indicated during the rest of the study. A similar re-evaluation of priorities took place with respect to field measurements of dissolved oxygen (DO)  and pH. In the case of pH measurements,  an initial test was made using a laboratory pH meter which was taken to the site, but the lack of shelter and possibility of damaging the meter, ruled out its regular use. Since a reliable portable pH meter was not available, it was decided to measure the pH of the samples back at the laboratory. An initial measurement of DO levels in the RBC stages, made with a Yellow Springs Instruments  Ltd. (YSI)  portable dissolved oxygen meter, indicated levels  approaching saturation except for the first stage which was about 1 mg/L less.  51  These results were obtained under light organic loading conditions of approximately 4.1  g C O D / m - d . Measurements during subsequent weeks were interrupted by 2  operational problems. The DO measurements were then discontinued as one step to streamline the sampling program until steady operation could be achieved. While D O levels indicate the extent to which the oxygen transfer capability of the RBC is being used at a given loading level, and indicates inhibitory or limiting conditions, this information was of secondary importance in this study as effluent quality was used as the prime indicator of process performance. Therefore, it was decided that the measurement of DO levels in the RBC would be discontinued until limiting conditions were approached as indicated by the effluent quality or changes in the colour of the biomass. In practice, steady-state limiting conditions were not indicated, and no further DO measurements were made during the course of this study. The sampling program for T O C , alkalinity, and effluent orthophosphate (P0 ), 4  should also be elaborated upon. The raw leachate and RBC samples were analysed for TOC during the first half of the experiment in order to establish a correlation between this parameter and the C O D and BOD results. Once sufficient data had been collected to show whether or not such a correlation existed, the TOC analysis was discontinued as originally planned. Alkalinity on the other hand was only monitored during the second half of the study when greater emphasis was placed on evaluating the performance of the RBC for nitrification. The weekly checks of the effluent  orthophosphorus levels were done quantitatively during the early part of  the study, but this was later reduced to a qualitative check, and finally the frequency of these checks was reduced to approximately monthly. This reduction in sampling was justified on the basis of the previous results, which indicated that an excess of P O . was consistently maintained in the system.  52  Sampling of the biomass attached to the RBC disks consisted of one sample from all four stages, and two samples of just the first and fourth stages. This small number of samples reflects the practical limitations which were imposed upon the significance of the analysis on those samples. Once the biomass became established, it was quickly realized that measurements of the biomass thickness, or determining the weight of areal samples, would not yield good estimates of the total biomass, particularly in the first stage, which is the most important. This was because samples could only be taken off of the external fibreglass disks of each stage, and the growth on these disks differed from that of the internal mesh disks. During light loading conditions, the biomass appeared to grow preferentially  on the  fibreglass disks. Under heavier loading, the thick growth on the first stage was patchy on the external disks, and considerable bridging of the biomass occurred between  the internal disks. In addition, the rationale for determining the total  biomass was removed because the variable loading conditions made it impossible to relate the amount of growth to the availability of substrate. Therefore biomass determination  was discontinued after one sample. The other two samples were taken  for analysis of nutrients and heavy metal accumulation, checks which, for the purposes of this study, did not warrant further samples.  4.2 SAMPLING PROCEDURES  The sample collection and preservation procedures used during the course of this study generally followed those recommended by Standard Methods (72). During each sampling visit to the landfill site, grab samples of the raw leachate and liquid from various stages of the RBC were taken. A bucket and rope were used to hoist a quantity of the raw leachate from the bottom  of the North lift station wet well. Usually the temperature  of the  53  leachate was  determined with a thermometer, and then two samples  were taken from the bucket. A 2 L sample was taken for C O D , p H , and TS  analysis, and a 500  concentrated r ^ S O ^  for TKN, N H ^ , and T O C  made  during the course  taken from the influent line to the of the RBC  sampled with a 125 collected for C O D ,  for C O D ,  was  there was  from the wet well and  BOD,-, TSS, TVSS, and pH  measurements.  in a 1 L graduated  and TVSS tests; 125  5  Then 1 L was cylinder. The  mL preserved with H S 0 2  The temperature of the 1 biomass  s t  samples  disk and then scraping of each stage  scraped  mouth jar). Then  from the TSS  and BOD,,  and 4  for TKN,  t n  stage  tests was  +  2  NO^  liquid was  analysis.  used for the  as well as residual orthophosphate  a square  patch of biomass  of interest. A  3 inch (7.6  (P0 ). 4  also usually recorded.  into a previously acid washed  the samples  off one  were dried at 104°  of the biomass  biomass  for nutrient and metal analysis was  of the fibreglass  cm) wide metal paint scraper  from a 3 inch square  the dry weight sample  mL  were taken by momentarily stopping the rotation of the  used to remove the biomass was  4  500  and 50 mL were filtered with a test tube plunger type filter and  determination of soluble C O D  biomass  samples  off to provide the settled samples:  Filtrate of the raw and settled samples  was  no  liquid were collected by dipping from each stage to be  preserved with phenylmercuric acetate ( C H ^ C O O H g C g H ^ ) , for N 0  endplates  analysis.  mL plastic beaker. First 500 mL of the stage liquid was  B O D , TSS,  The  half of  RBC.  then carefully poured  N H ^ , and T O C ;  of  the second  of this study showed  withdrawn to be settled for 30 minutes supernatant  analysis. During  Sp.Cond.,  collected for volatile fatty acid (VFA)  detectable difference between leachate samples  Samples  BOD^,  mL sample was preserved with 1 mL  the experiment, a 60 mL sample was Periodic checks  of the leachate  (58.1  cm ) 2  area. The  and tared glass jar (8 oz. wide C and reweighed to determine  per unit disk area for each stage. Additional collected and dried in a similar  54  manner, and then ground  to a fine powder and stored for analysis.  The acid preserved samples RBC  of leachate, and settled first and fourth stage  liquid, from several dates were saved until the end of the study for heavy  metal analysis. Samples  were selected to be representative of high and low leachate  strength and leachate production rate. In addition, some  samples  specifically for metal analysis and were preserved with H N O ^ Standard On  as prescribed by  Methods. three occasions  samples  of leachate and RBC  effluent  were collected for a CC/MS scan of trace organic compounds, study. These  samples  4.3 ANALYTICAL  t n  stage)  as an aside to this  or  bubbles.  PROCEDURES  The analytical methods Standard Methods  (settled 4  were collected in clean, oven-dried glass vials with teflon caps.  The vials were filled completely leaving no head-space  COD  were taken  15  t h  used for this study were except as noted below from  ed.  - The potassium  dichromate reflux method as per the 1 3  of Standard Methods  was  used as this method has  t n  been  adopted as a lab standard at U.B.C..  TOC  - Acidified samples 915A  Carbon  ( p H < 2 ) were analysed using a  Analyser.  Beckman  ed.  55  BODj.  - The 5 day BOD was determined using the dissolved oxygen probe method. Probe calibration was by Winkler titration. Dilution water was seeded with settled RBC solids or unsettled RBC  TKN  effluent.  - Samples preserved with acid (pH<2) were analysed using a Technicon AutoAnalyser II and Technicon industrial method no. 325-74W.  NH^  - Samples preserved with H^SO^ (pH<2) were analysed using the automated  phenate method on a Technicon AutoAnalyser  II, a tentative standard method (15  NO2  +  NO3  tn  ed.)  - Samples preserved with phenylmercuric acetate were analysed using the automated cadmium reduction method on a Technicon AutoAnalyser II, a tentative standard method (15  tn  ed.)  Total Solids (TS)  - as per 1 5  Total Suspended  - as per 15™ ed., RBC samples in duplicate, Whatman  Solids (TSS)  glass microfibre  Total Volatile  as per 1 5  Suspended Solids (TVSS)  tn  th  ed., 80 mL leachate samples in triplicate  filters  ed.  934-AH  56  Specific  - measured using a Radiometer Model CDM3,  Conductance  Conductivity in /zS/cm  (Sp. Cond.) Alkalinity (Alk.)  - titration to pH = 4.5 as per 15 *  pH  - laboratory pH meter  Metals  - Total Metals samples prepared as per 1 5  1  on a Jan-ell Ash #810  1  ed.  tn  ed. and analysed  Atomic Absorption Spectrophotometer  (AA), flame method used except for lead (Pb),  (graphite  furnace). - Selected samples were sent to the Environmental Protection Service laboratory for an Inductively Coupled Plasma (ICP) scan  metal  57  Volatile Fatty Acids  - The analyses for Volatile Fatty Acids C2-C4 were performed  (VFA)  on a Hewlett-Packard  5750 Gas Chromatograph equipped with  a flame ionization detector and using helium as the carrier gas. A 6 ft. by 1/4"  O.D. and 1/8"  with 0.3% Carbowax/0.1% H P 0 3  4  I.D.  glass column packed  on 60/80 Carbopack C  (supplied by Supelco Inc.) was used. The column was conditioned as specified on the instructions supplied with the packing. Quantification was by the external standard method using reagent grade standards dissolved in 0.1% aqueous phophoric acid. Samples were stored in 60 mL plastic bottles and preserved by freezing.  Organic  - The analyses for specific volatile and semi-volatile trace  Compounds  organics were performed on a Hewlett-Packard 5985B Gas Chromatograph/Mass Spectrometer. A purge and trap method was employed in which the samples were purged with an inert gas (helium) and the volatiles then trapped onto an adsorbtive material (Tenax-GC/Chromosorb-101). The trap was then backflushed into the gas chromatograph column and the GC/MS analysis started. Samples were collected in 40 mL glass vials without headspace, and stored at 4° C until analysis, which was within 5 days.  5. LEACHATE QUALITY  5.1  LEACHATE GENERATION  Landfill leachates are complex wastewaters which reflect the unique circumstances of their formation in their varied chemical and physical properties. One of the first realizations of leachate researchers was that the composition of leachates varied widely from landfill to landfill such that it was impossible to describe a typical  leachate. Table 5.1  shows the ranges of observed values for  some leachate characteristics assembled from the literature by Pohland (59). Similar tables have been compiled by many other authors and generally also show a wide variation in leachate composition between sites. It was soon recognized that each landfill site had a different combination of the many factors which were thought to affect the nature of the leachate produced. Climate, types of wastes and their relative amounts, landfilling methods, compaction density, soil types, hydrology, site dimensions, collection system layout; these are just a few of the many parameters involved. In addition to the site to site variation of leachate characteristics attributable to physical differences, early comparisons of leachate data showed clearly that the nature of a landfill leachate changed with the increasing age of the landfill  (11).  Leachates from relatively new landfills (receiving wastes for less than 5 years), and research lysimeters, usually had very high concentrations of degradable organics (BOD), and high levels of ammonia (NH^-N), and heavy metals (relative to domestic sewage). These were labelled young  leachates. Landfills which had been in operation  for more than 10 years generally produced leachates with very low BOD  58  59  Table 5.1 Variability of Leachate Composition* * 1  pH Total hardness (mg/L as CaCO^)  4.9 30  8.4 13,100  Total alkalinity (mg/L as CaCO^)  100  20,805  Total iron (mg/L) Sodium (mg/L) Potassium (mg/L) Sulphate (mg/L) Nitrate nitrogen (mg/L as N) Ammonia nitrogen (mg/L as N) Chemical oxygen demand (mg/L) Biochemical oxygen demand (mg/L) Total volatile acids (mg/L C H j C O O H ) Total dissolved solids (mg/L)  (1) from table 1. of Pohland (ref.  2 85 28 24 5 0.2 246 5.9 <100  1000 1805 3770 1220 196 1106 750,000 720,000 10,000  1740  11,254  59)  concentrations, considerably higher ammonia levels, and variable heavy metal concentrations. These leachates were called old leachates. The reduced condition of leachate constituents coupled with the observed production of large volumes of methane gas (CH^), led to the conclusion that solid wastes within a landfill were stabilized over time primarily by anaerobic microbial processes. Therefore, landfills are now generally conceptualized as large anaerobic batch digesters in which the infiltrating precipitation provides both the transport phase for leaching and mobilizing contaminants, as well as the moisture necessary to promote biological activity. Numerous lysimeter studies have examined the nature of the decomposition process within landfills; the interaction of various physical parameters with the biological processes, and the resulting leachate quality (12,13,17,63,75,77,78). These controlled studies in conjunction with more detailed and long term observations of full scale landfills have resulted in a basic understanding of landfill evolution and some cause and effect relationships. It has been observed that a landfill generally progresses through 5 identifiable stages between first use and final stabilization  (13).  60 The two most the  important and dominant  phases with respect to leachate quality are  acid formation and methane fermentation phases. The acid formation phase  of a landfill's life becomes  established quite quickly  after the field capacity of the fill, or a zone of the fill, is exceeded and moisture begins become  to move  through the wastes. W i t h the onset of water movement,  conditions  ideal for the growth and spread of an anaerobic microbial culture. The first  group of bacteria to establish themselves bacteria degrade  are the facultative acid formers. These  the larger organic compounds,  or hydrolysed from the wastes, down  found dissolved in the pore water,  t o simple organic acids, hence their name.  Acetic acid ( C H ^ C O O H ) is the main catabolic end product of these microorganisms undergoes  during anaerobic fermentation. Some of the acetic acid then  condensation  reactions, or is combined by other bacteria, to produce the  other volatile fatty acids (VFAs) of higher order which are commonly found in leachate (such as propionic C 3 , and butyric C4) (11). These acids are produced in large quantity and their concentration in the leachate draining out of the wastes can range  to over 10,000 mg/L. At such high levels it is not surprising that the  leachate generally achieves its highest  organic strength during this phase, and that  the  VFAs normally account for a large proportion of the total organic strength of  the  leachate. In terms of total organic  carbon (TOC), the VFAs often represent  80-95% of the total value (32,64). The remaining fraction of the T O C is usually made  up largely of refractory humic and fulvic acids (11). Since the VFAs are  readily biodegradable under aerobic conditions, they exert a strong oxygen  demand  and generally also account for almost all of the B O D of the leachate. Production of these acids also reduces the pH of the pore water, or leachate, which increases the solubility of most heavy metals. Therefore metal levels in the  the leachate are usually highest growth  during this phase. Low pH conditions also inhibit  of other types of bacteria, notably the methanogens,  and thus the acid  61  formation phase tends to be self propagating. Lysimeter studies have identified a number of management  options or physical conditions which prolong the acid  formation phase within a landfill. Placement of the wastes in thick layers, shredding of the wastes, high compaction densities, and low moisture inputs, have  been  shown to promote acid formation (63,75). All these factors tend to reduce the movement  of leachate through  the wastes  and therefore would maintain the low  pH  conditions and inhibit the development of the methane bacteria. Stegmann ef al. (75)  observed  (65%  that in a limiting case in which there was no moisture  movement  moisture content), that an acid conservation effect, such as occurs in silage,  took place. Leachate recycle was  also observed to prolong the acid formation  phase  by maintaining high acid levels in the leachate moving through the wastes. However, leachate recycling also increased the rate of waste stabilization and intensified the activity of the methanogenic  phase which followed (13,64). Factors  which hasten the end of the acid formation phase production phase  are generally the converse  and the onset of the methane  of those mentioned above i.e., high  moisture inputs, etc.. Stegmann ef al. also found that the placement of an uncompacted and/or aerobically c o m p o s t e d onset  of methanogenesis.  unsaturated soil zone  Similarly Robinson  and Lucas (67) found that an  beneath the fill rapidly developed a population of  bacteria such that VFAs produced through  bottom layer significantly accelerated the  the layer. (This result was  methane  in the wastes were not observed to penetrate probably aided by the low leachate production  rate and therefore long detention time in both the fill and the soil zone at this site.) Thus, the duration of the acid formation phase  in a landfill is also a function  of all the site specific conditions mentioned earlier, and has been observed to vary from less than one year to more than 10 years. In addition to the organic carbon and heavy metal content of the leachate, high concentrations of nitrogen compounds  are usually present. During  the anaerobic  62 degradation of the organic material in the wastes, the organic nitrogen component is rapidly reduced to the ammonia form. Since the growth rate and therefore nutrient requirements of the anaerobic bacteria are relatively low, very little nitrogen is assimilated by the bacteria. Therefore, the ammonia passes readily through the wastes in the leachate. As it is the same degradation process which produced the high concentrations of VFAs, high concentrations of ammonia (over 1000 mg/L) can also be produced. The dissolved solids levels in the leachate during this phase are also generally very high. Since the acid formation phase is established rapidly with the onset of leachate migration, this leachate contains the first flush of soluble inorganic material from the wastes in addition to the dissolved organic matter. Straub and Lynch (77)  showed that the inorganic strength of the leachate decreases  exponentially as the cumulative volume of water passing through the wastes increases. They found that the inorganic strength was stabilized after approximately four moisture changes through the wastes. Therefore, the inorganic material would generally be flushed from the wastes while they are in the acid formation phase, adding to the dissolved metals and organic compounds to increase the total dissolved solids observed during this phase. Although the duration of the acid formation phase may vary, the end of this phase is initiated by its very beginning. The simple acids produced by the acid forming bacteria are the prefered substrate for various other bacteria, most importantly the methanogenic bacteria. While growth of these bacteria may be inhibited by the low pH conditions produced by the acid formers, gradually the population of methane bacteria establishes itself and eventually balances the activity of the acid formers. When the balance point is reached, the methane forming bacteria consume most or all of the organic acids produced by the acid formers and thus the VFA content of the leachate is drastically reduced. This marks the  63  establishment  of the methane  fermentation phase.  It is frequently observed  transition between the acid formation and methanogenic particularly where the methane  that the  phases is relatively abrupt,  leachate recycling is practiced. This would tend to indicate that  bacteria population develops  inhibitory conditions  moderate  and spreads,  but at reduced activity, until  enough to permit a rapid exploitation of the available  substrate. From the previous organic  discussion  it follows that with the virtual removal of the  acids from the leachate, the organic  strength of the leachate is drastically  reduced from that of the acid formation phase. established methanogenic mg/L), although  the C O D  phase typically have  Leachates from landfills with a well  a very low B O D g concentration  (<100  may remain significantly higher due to the refractory  compounds. The  establishment  of the methane  bacteria also affects the pH  conditions within the landfill and leachate. As to approach  neutral values. The  the acids are consumed,  gradually prior to the rapid growth  of the methane  of conditions favourable to the growth  These two conditions combine  conditions  the pH  values generally  bacteria and reflects the of these obligate  anaerobes.  encourage  stable complex The reduced  A possible  metals  levels reduce the solubility of the metals, while metals which are dissolved to precipitate as  sulphides. Therefore, heavy metal levels in the leachate are generally much during this phase.  rises  decrease  to greatly reduce the mobility of most heavy  during this phase. The higher p H the low ORP  ORP  leachate shifts from a volatile-acid buffered system,  to a predominately bicarbonate buffered system. ORP  development  and  lower  exception to this trend is lead (Pb), which forms a  with humic substances and thus remains  reduction in dissolved  concentration of dissolved  During the acid formation phase,  metals and organic  mobile  (32).  acids is reflected in the  solids and value of the specific conductance. heavy  metals and organic acids constitute a major  64  portion of the dissolved material. Therefore, the immobilization and removal of these materials during the methanogenic  phase  results in significantly lower concentrations  of both dissolved solids, and charged species which contribute to the conductance. While the concentration of almost every other constituent of the leachate is reduced markedly with the onset ammonia  of the methanogenic  phase, the concentration of  generally remains constant or even increases slightly. This reflects the fact  that degradation of the wastes  is continuing at similar rates as occurred during the  acid formation phase. The pathway over which the organic carbon leaves the landfill (as methane CH^)  may have changed, but the fate of the ammonia  remains the same.  If the rate of water movement  produced  has decreased by this time due  to the increased depth of the fill, or placement of the final cover, the concentration of the ammonia in the leachate may be observed to increase over time. This persistence of high ammonia concentrations in landfill leachates over very long periods of time (until the wastes numbers more  are fully stabilized), has led increasing  of researchers to conclude that the ammonia content of leachate is a  serious  and difficult problem than the organic carbon content (22,64).  Unlike the acid production phase, the methanogenic  phase  does not  end  abruptly but rather fades out as the stabilization of degradable material is gradually completed. The methane fermentation phase formation phase  is also less stable than the acid  and subject to upset. Jasper ef al. (41) observed that for a landfill  with a short hydraulic retention time, and subject to large water inputs, that wash-out  of VFAs occurred periodically, coincident with major rainfall events, after  the methanogenic  phase  had  become established. This further supports the concept  of a landfill as being a large anaerobic digester subject to similar constraints such as hydraulic overloading. However the literature indicates that at most landfills conditions are more moderate, and once the methanogenic is usually quite stable and the breakthrough  of VFAs is not  phase is established, it observed.  65  The discussion thus far has described the affects on leachate quality of a shift from the acid formation phase to the methane fermentation phase within a landfill undergoing stabilization. Due to the numerous factors which affect the stabilization process, there are no specific parameter values which define these two phases but rather they show to varying extents the characteristic changes mentioned above.  5.2 AFFECT OF WATER INPUTS O N LEACHATE QUALITY  When the rainy season begins in the Fall, wastes placed during the Summer are rapidly soaked to their field capacity and the top layers of the landfill can become almost saturated with each new rainfall. Additional water inputs increase the hydraulic flux within these top layers, conceivably forcing the water to move faster through existing pathways in underlying unsaturated layers, as well as opening up new paths, saturating more wastes, and exposing more surface area to the water. In less dense wastes the former mechanism would probably predominate, leading to a heavy flush of pollutants, followed by reduced concentrations due to the reduced contact time with the wastes. Within dense wastes the later mechanism would dominate , leading to increased concentrations of pollutants as more wastes were exposed. Saturated zones below the watertable,  or perched higher in the landfill,  could also lead to higher concentrations of pollutants due to greater contact with the wastes. The residence, or contact time, of the water with the wastes affects the leachate strength by varying the length of time which chemical and biological processes have to concentrate soluble products in the passing water (dilution). Residence time can also affect the ability of other chemical and biological processes to remove soluble constituents from the leachate. Jasper ef al. (41)  observed that  66 the concentrations  of organic constituents, T O C ,  increased with increasing  BOD,  COD,  VFA, and  leachate production or water inputs. It was  these increased concentrations came about because the wastes, combined with a shortened  theorized that  the increased water contact with  leachate retention time, overloaded the  methane  bacteria and resulted in the wash-out of organic material. They  observed  that the nitrogen content, N H ^ & TKN, as well as TIC,  Sp.  C o n d . levels decreased  during  also  CI", Alk., and  peak leachate flows. These parameters  generally less affected by biological activity and more wastes  VSS  are  affected by the exposure  of  to the water and dilution. It was noted that the product of the leachate  flow and parameter concentration or value, increased with increasing flow, supporting the notion that greater contact of water with the wastes was remaining tested parameters, metals, pH, TC, TP, TSS, relatively independent  and TDS,  occurring. For the concentration  of the rate of water input. Results from the monitoring  another landfill assumed to have  a long leachate retention time (3 to 4  indicated that the levels of all parameters are relatively independent Other researchers  have  observed  was of  months),  of water inputs.  similar variability of leachate strength with  water input. Bull (7) also indicates that heavy rainfall may cause an increase in leachate strength However,  by reducing the residence time of leachate within the fill.  Raveh ef al. (63)  observed  of pollutants in the leachate was to 1100  mm  that for their lysimeter study, the concentration  independent  of the level of water application up  of water per year. They speculate that retention time was  not limiting  in their case which allowed pollutants to concentrate to their saturation level. Therefore, while the variation of concentrations respect to water inputs, it is now  of pollutants may be variable with  generally held that the amount, or mass, of  pollutants leached from a landfill increases with increasing water flow (17). Considerable  effort has  been applied towards  formulating a mathematical  model capable of simulating the production of landfill leachate. Such a model  would  67  be invaluble to help explain the interaction of the many physical, chemical, biological, and hydraulic influences on the concentrations of the various leachate constituents and to aid in the design of leachate control measures. So far, these efforts have resulted mainly in hydraulic models to estimate leachate volumes, and simplified empirical models which can be made to fit observed data by varying coefficient values. Such models are useful tools for the analysis of historical data and can help identify which mechanisms are important in the leaching process (78). The work of Straub ef al. (78)  is a case in point. Their model indicated that high  moisture flow rates increased the relative importance of water movement and decreased the importance of microbial activity, which agrees with the previously discussed notions of leachate flow and residence time. While empirical models can yield useful insights into the leaching process, a mechanistic model would be more useful for predicting leachate quality. However, given the number of variables which affect leachate quality, the formulation of such a model seems an impossible task.  5.3  PREMIER LANDFILL LEACHATE  A leachate sampling program was started in October 1982, just a few months after the new section of the landfill site was opened (recall Fig. 2.2),  to  provide data for this and other studies. The weekly samples, and later the leachate feed for the RBC, were taken from the lift station wet well and recall, were therefore  already diluted roughly 50% by drainage from the unfilled portion of the  site. This is one reason why this leachate would be described as weak compared to most others encountered in the literature. Column A of Table 5.2 shows the high and low values of various tested parameters for this leachate to date. A comparison of these values with the corresponding ranges of Table 5.1 shows clearly that this leachate has concentrations of the typical leachate constituents  68  nearer the l o w end of the given 5.2  because,  not  meaningful.  ranges.  Average  values were omitted from Table  due to the nature of the strength fluctuations described later, they are  The dilution of the leachate by drainage from the unfilled portion of the site demonstrates system  how  important physical site conditions such as the collection  layout are in determining leachate quality, ln this case the placement of the  collection pipe within the sand and gravel underlying the site promotes drainage  and collection of the water from beneath both the filled and unfilled  areas. Due wastes  rapid  to the lower hydraulic conductivity and extra thickness of the  however,  compacted  the drainage from the filled area would lag behind that of the  unfilled area. Therefore the dilution ratio would be variable. Rapid drainage probably means that the soil zone  below the wastes  time, ln other cases, the collection system leachate from areas in one  phase  maintaining a saturated zone collection system  design  as it does in this Another  of stabilization rather than another, or by  can greatly influence the quality of the leachate collected,  for the relatively low strength of the Premier leachate is the this landfill. The wastes were placed over the  fluvial gravel in comparatively thin lifts ( < 2 before the next lift was  started. Moderate  m), covering the whole area of the fill compaction densities were achieved  increases  of water through  placed over the wastes their exposure  daily. The above  large volumes  method  to precipitation and promotes  of  good  the wastes. W h e n subjected to the heavy annual rainfall  normally received at this site, the field capacity of the wastes and  using  compactor and/or a large bulldozer, and a thin layer of  permeable cover material was  drainage  may affect leachate quality by collecting  below or within the wastes. Suffice ft to say, the  quite high moisture flux through  placing the wastes  of the  case.  reason  a small B O W - M A C  is unsaturated most  also  of water drain relatively quickly through  is rapidly exceeded  the garbage.  The large  69 Table 5.2 Premier Leachate Characteristics (Well #1)  B< >  CW  2  low - high  C O D mg/L B O D mg/L  86 - 4421 44 - 3020  TKN-N mg/L N H - N mg/L  low - high  low - high  263 - 1527 161 - 1035<>  150 - 434 49 - 251  8.1 - 53.8 6.9 - 49.1  18.5 - 51.2 17.1 - 46.4  20.1 - 41.2 18.4 - 40.3  VFA mg/L (as acetic) T.S. mg/L Alk. mg/L (as C a C 0 )  1 - 1470 540 - 3595 288 - 782  48 - 888 764 - 2176 428 - 750  5 - 108 639 - 1238 350 - 673  Sp. Cond. /xS/cm pH  527 - 3567 5.6 - 7.4  1162 - 2594 6.3 - 7.0  1070 - 1890 6.4 - 6.8  5  3  3  (1) (2) (3) (4)  Data Period Data Period Data Period BODr value  (4)  4  A - October 22/82 to March 31/85. B - April 10/84 to July 24/84. C - January 18/85 to March 31/85. estimated from COD.  volume of the water and the resulting short contact or residence time within the wastes act to reduce the strength of the leachate produced at this site as previously discussed. Figures 5.1 A,B,C, show the variation in concentration of the primary leachate constituents from the start of monitoring in October 1982, until June 1985. These figures illustrate several interesting points about the variation of leachate strength at this landfill. First, note that the levels of all these constituents parallel each other very closely. This contrasts somewhat the results of Jasper et al. (41) as they found that the ammonia levels would decrease, and total solids levels would remain constant, during peak concentrations of organic strength and peak leachate flows. The reasons for these differences becomes clearer when one notes how the variation of pollutant  concentration relates to the pattern of rainfall or water inputs.  PREMIER LEACHATE CHARACTERISTICS VERSUS TIME A N D PRECIPITATION  1982  1983  PREMIER LEACHATE CHARACTERISTICS VERSUS TIME A N D PRECIPITATION  D  J  F Legend  1984 z  A  o <  COD  X B0D5  20  •  O  T. SOLIDS  H Sp. Cond.  UJ  X NH3-N  iL  MM 0 1983  N  D  J 1984  F  M  LJL  h i  M  J  LL o  PREMIER LEACHATE CHARACTERISTICS VERSUS TIME AND PRECIPITATION 60 50 D)  h40  30  • A  E <  K  20 O  Ho < 0 OCT 1984  NOV DEC  JAN 1985  FEB M A R A P R M A Y J U N  JUL  Legend A  COD  X BOD5 •  T.SOLIDS  H Sp. Cond.  i  OCT 1984  II  Ii  NOV DEC  K  • 1 JAN 1985  i.  11  1  .•  i1  1  i •1 1..  FEB M A R A P R M A Y J U N  i1 JUL  |  NH3-N  73 Figures  5.1  A,B,C,  clearly show that the pollutant concentrations are  during the wet Winter and Spring  months  Summer and early Fall period. U p o n increases  highest  and then decrease over the dryer  closer examination it can be seen that sharp  in leachate strength are generally preceeded by wet periods or major  rainfall events. This is most  noticeable during March  1984, and December  In general, it can be observed  and strength  1984.  responded  1983,  January  that leachate volumes  quickly to rainfall events. The increases in concentration  generally lag behind the rainfall peaks often correspond  1983, December  to sharp  lag between the drainage  dips  by a few days and so  these rainfall  peaks  in the concentration values. This reflects the time  from the unfilled and filled portion of the site. Water  from the unfilled portion of the site is collected more quickly than that from the filled area and therefore has in Figure  an initial diluent effect. This time lag can also be  5.2, from Jasper ef al.  (41), which shows the leachate production  seen  volumes  and the mass of major pollutants released over the same period for this site. The pollutant discharges  lag behind the peak leachate discharges  by several days. This  figure also clearly shows that the mass of pollutants discharged volume  of leachate produced. Therefore, the data from this site indicates that the  main mechanism wastes  increases with the  governing  and the water. As  area or volume These  phase. As  noted earlier, increased water inputs increase the surface  of waste in contact with the passing water.  figures  quality through  the leachate strength is the area of contact between the  also s h o w quite well the evolution of this site and its leachate  the acid formation phase  the leachate sampling  to the start of the methane  began just a few months  fermentation  after the first wastes  were placed into the new landfill area, and near the start of the first wet it appears  as though  season,  some of the first leachate to be produced from this section  was collected. This is indicated by the very low concentrations of the first few samples. The  leachate strength  as exemplified by C O D ,  rose rapidly from 64  mg/L  74  450-1 r36 PHASE  400  I  o  350 28 x m  X z o  JC  300 24«  t—  250 •20 c 200 • 16 150 • 12 100  8  •~.5.0-• 4 OCT NOV 82 OEC JAN 83 FEB  MAR  APR MAY  JUN JUL  AUG  SEP OCT  I8r  450T36 PHASE  400 [ 32  Z  350- 28 LEGEND  r  Leochott voL Mas*COO Most NH Mo** T.S. 3  300 24 OT 20 200 • 16 150-12  OCT NOV 83 OEC JAN 84 FEB  18  150 12  OCT NOV 84 DEC JAN 83 FEB  MAR  APR  MAY  JUN  JUL  100  8  50  4  AUG SEP OCT  from Jasper et al. (41),  Figure  5.2  Leachate  Flow  Premier  and  Constituent  Street  Landfill  Mass  Release  0  75 in  O c t o b e r 1982  to a high of 4421  acid formation phase.  During  mg/L  the Summer and Fall of 1983  tapered off gradually to approximately 1500 Although because  in April 1983, indicating the start of the  mg/L C O D  the leachate strength  due to dryer conditions.  the dryer conditions could be expected to increase leachate strength of increased residence time and less dilution from the rest of the site, the  opposite occurred, possibly due to a minimum groundwater flow beneath the site. Note that all the main leachate constituents decrease proportionally during this period. The leachate strength then rose slightly over the Winter of 1983 2000 mg/L C O D ,  which held steady through the January to March  to about  period of  1984.  After that, the leachate strength decreased steadily like the previous year, except that the C O D  decreased proportionately more than the other parameters. This  indicates the establishment of the methane fermentation phase after less than two years. As  mentioned previously, moderate VFA concentrations, pH, and high water  inputs encourage  the rapid development of the methanogenic bacteria. Therefore this  period from March  to O c t o b e r 1984  represents a transitional phase of leachate  quality (which will be mentioned again in later Moderate  rainfall during the Fall of 1984  vary between 150 capacity was  and  mg/L C O D ,  caused the leachate strength to  with a slight increasing trend as the field  re-established after a dry Summer. Then the leachate strength  sharply in response washout  350  discussions).  to a heavy week of rain in December  condition like that observed  1984, indicating a  by Jasper ef al. (41). However, the landfill  recovered very quickly once the normal hydraulic regime was resumed For convenience  and clarity the data for the various  presented separately in subsequent data from the analyses  increased  Figures  (5.3  (Figure 5.1C).  major leachate parameters are  - 5.5 A,B,C). In addition, the raw  of this leachate is included in Appendix  1.  Z2/Z2  *SA  luajuo^  uoqjC3 a i e i p e a i  v€'S a j n S i J  COD, B O D , & TOC (mg/L) 5  LEACHATE CARBON CONTENT vs. TIME and PRECIPITATION 2500 - i  2000  1500 H  1000 H  Legend  500  £ u  30  0 1983  N  z o «C  20  CL O  CC Q.  10  hi  JJU 0 1983  N  D  J 1984  F  M  A  M  J J  U J  A  S  O  A  COD  X  BODp;  •  TOC  LEACHATE CARBON CONTENT vs. TIME and PRECIPITATION 1500  c  era  —  -i  1000  LO  n  n o 3" 01  n  t cr o 3 o O 3 re  3 <  in H  3' 00 00  Q  O  CQ  66  500H  Legend  Q  O  CJ  E 30 o  z o  1 OCT 1984  1 NOV  1  DEC  1 JAN 1985  1  FEB  1 MAR  1 APR  1 MAY  i  JUN  A  COD  X  BODfi  i  JUL  < 20 UJ  VI  JL OCT 1984  NOV  DEC  JAN 1985  ±  I  • i l l --f,  i I  FEB  APR  MAR  i  •I 1. MAY  •  • I  JUN  JUL 00  LEACHATE NITROGEN CONTENT vs. TIME and PRECIPITATION 60  -i  1982  1983  LEACHATE NITROGEN CONTENT vs. TIME and PRECIPITATION 60 -i CD  E  BOH  ^  40 H  y£  30  <  20  Legend  i(H  X 1  E (J  30  z o  0 1983  r~  N  i  D  J 1984  r  F  o  M  M  u a.  10H  1  IXI  0 1983  N  D  J 1984  JUL  F  M  A  M  111  J  LL 0  TKN  LEACHATE NITROGEN CONTENT vs. TIME and PRECIPITATION  Legend  X X J  1  !  OCT 1984  !  NOV  !  ,  DEC  JAN 1985  j  !  FEB  ,  MAR  ,  APR  ,  MAY  Nr-h TKN  ,  JUN  JUL  1  r OCT 1984  -1 NOV  1  DEC  |  1  JAN 1985  •l  |  FEB  i. MAR  I  II APR  r  i  I  MAY  I  Il 1 i  JUN  JUL  1  . • ll r  r  -  Total Solids & Specific Conductance vs. TIME and PRECIPITATION E o co  a  4500-] 400036003000-  -6  c o u  Cu  CO <3 co  T> o CO  250020001600-  Legend  1000-  A Tot. Solids  600-  "TO  n u•  t—  x i  i  O 1982  i  N  D  n  r  O 1982  N  i  i  J  i  F  :  M  i  A  i  M  i  i  i  J  i  J  A  S  S  O  1983  z  o <t 20  u <r UJ CL  10  1  1  D  11 -1 1. • • I I i  1.  ^  J  F 1983  n—  M  A  M  —  J  r  i  J  — i A  O  Sp. Cond.  Total Solids & Specific Conductance vs. TIME and PRECIPITATION E  00  c  3000-1  o co 3 e3  2500  In 00  to tu  O  sr -4 O  CD  S  -6  2000H  c o U  CL  1500  Legend  c3 CO TJ  l/l ai 3  Q. C/l  13  o o 3  Q. C  n  a) 3  —  o  IOOOH  A Tot. Solids  CO  x  ro 500  J  3 0  1 0 1983  r-  N  Sp. Cond.  i  D  J 1984  F  0  M  M  z  o !<  20-1  o  fD  UI  3 CO OJ CO  JjJ-4  UJ UJ  5  0 1983  N  D  J 1984  i i F  M  J U L  M  LLI  o CO  Total Solids & Specific Conductance vs. TIME and PRECIPITATION  Legend A  Tot. Solids  X  Sp. Cond.  85  5.4 ORGANICS  Figures 5.3 A,B,C, present the leachate C O D , B O D , and TOC data. These 5  figures show more clearly how these related parameters parallel each other. The close linear correlation between these values is further demonstrated in Figures 5.6 and 5.7 which show TOC and BOD^, plotted against C O D respectively, along with their corresponding tables of linear regression results. The data was analysed in roughly six month intervals to indicate whether or not the relationship between the various parameters changed over the period of this study. In the case of the relationship between TOC and C O D , Figure 5.6 shows that the strong linear relationship appears steady over the period of this data. The regression analysis reveals that the slope of this curve is only slightly less than would be predicted by stiochiometric considerations (0.3320 vs. 0.3750), indicating that oxidation of organic carbon accounts for a large proportion of the C O D , as expected. A similar close correlation is apparent between BOD,- and C O D (Fig. 5.7). The regression analyses show a slight trend toward a decreasing slope, which one would expect with increasing time, with the exception of the last interval when the slope increases markedly. This unexpected increase in slope is probably due to the contribution of the ammonia oxidation in the BOD^ test becoming more significant with respect to the low total BOD,, values. Since ammonia was quite likely oxidized in the BOD test due to the use of an acclimatized nitrifying seed, but is not oxidized in the C O D test, this small difference can significantly affect the BOD/COD ratio. The interference of the ammonia oxidation can also be seen in Figure 5.8 which shows the BOD^ values plotted against the BOD/COD ratio a la Stegmann ef al. (74). A comparison of this data with that of Stegmann et al. (74), reveals that this data would lie below the results they found, but follows a similar trend of decreasing BOD/COD ratio with decreasing B O D - values. The primary reason for  86  TOC versus COD 1000-1  800-  O)  600-  E ^  400-1  Legend  r-  200-  i  i  i  i  500  1000  1500  2000  COD  2500  A  10/82 to 6/83  X  7/83 to 12/83  •  1/84 to 6/84  3000  (mg/L)  Figure 5.6 T O C vs. C O D  Linear Regression  Data  Group  10/82 to 6/83 7/83 t o 12/83 1/84 to 6/84 10/82 to 6/84  Results  Slope  Y intercept  Correlation Coefficient  No. of Data Points  0.2087 0.3614 0.3363 0.3320  336.5 24.08 70.90 70.36  0.9466 0.9855 0.9867 0.9823  5 40 42 87  87  B O D versus COD 5  1500n  A A  A A  1000-  CO E  A A  Q O  CO  Legend  A 500-  500  1000  COD  1500  2000  BOD.  vs.  Linear Regression  Data Croup  to to to to  i  X  1/84 to 6/84  I  •  7/84 to 12/84  |  B  1/85 to  12/83 6/84 12/84 6/85  6/85  2500  (mg/L)  Figure 5.7  7/83 1/84 7/84 1/85  7/83 to 12/83  A  COD  Results  Slope  Y intercept  Correlation Coefficient  No. of Data Points  0.6380 0.6374 0.6160 0.7504  -32.29 -4.966 8.558 -55.86  0.9180 0.9225 0.9043 0.9332  16 11 43 13  j  88  B0D  5  vs. BOD/COD Ratio  1500-.  A  A A  AA  1000m  £  A  LO  Legend  Q  2 CO  500  • 0.25  0.50  0.75  1  A  7/83  to  12/83  X  1/84  to  6/84  •  6/84  to  •  E  1/85  to  1.25  1.50  BOD/COD Ratio  Figure 5.8 B O D . vs. B O D . / C O D  Ratio  12/84 6/85  89  the difference between the two sets of data is the dilution of this leachate which reduces  the BOD,-  and C O D  values  by about  50%,  B O D / C O D ratio. A second  difference is the higher  BOD  are probably  concentrations. These  mentioned  above. The abnormally high  but would not alter the B O D / C O D ratios observed  attributable to the ammonia  oxidation  ratios skew the plot to the right at the  lower levels, which explains the otherwise unlikely results in which the ratio is >0.8,  let alone  >1.0.  Once  at low  BOD/COD  these two factors are considered, the  B O D / C O D data from this study compares  favourably with the results of Stegmann  et al.  5.5  VFA'S  The VFA  concentration is closely related to the C O D  vice versa. Figure  5.9  shows the variation in the VFA  and B O D ^  results  and  concentration over the period  for which they were monitored. This period covers the transistion to the methanogenic 1984  phase  to O c t o b e r  as indicated by the steady decline in concentration from  1984. The wash-out  1984-85 is also demonstrated.  Figure  March  of VFAs during the Fall and Winter of 5.10  shows even more  clearly the significant  contribution that the VFAs make to the organic strength of the leachate and the reduced acid levels after the transition period, with the exception of  wash-out  events. Correlation plots of the concentration of VFAs versus C O D  and  values  levels are due  (Figures  5.11  & 5.12), also show that high C O D  in large part to the VFA  contribution. The  the data (particularly for BOD,-, as might strong  regression  fate of the VFAs produced  results indicate some scatter in  be expected), but still show a reasonably  linearity. Therefore, this data conforms  which show that the organic  and B O D ^  BOD^  to the experience of other studies  strength of a leachate is largely determined by the during the decomposition  of the wastes  (32).  Volatile Fatty Acid Concentration vs. Time 10000q  100CH  10CH  Legend  M A 1984  M  J  J  A  S  D  J F 1985  M  A  A  ACETIC  X  PROPIONIC  •  BUTYRIC  B  Total VFA  VFA Theoretical COD vs. Leachate COD and BOD 5  2000 -1  1984  1985  92  COD versus VFA 2000  n  A A  Legend  500  1000 VFA (mg/L)  1500  Figure 5.11  vs. VFA  COD  Linear Regression  Data C r o u p  Slope  3/84 to 6/84 7/84 to 12/84 1/85 to 3/85  0.9788 1.8214 1.6303  Y  intercept  450.8 114.3 139.2  A  3/84 to  6/84  X  7/84 to  12/84  •  1/85 to  3/85  2000  Results  Correlation Coefficient  No. of Data Points  0.8672 0.9907 0.9643  24 48 22  93  B0D 600  versusVFA  5  n  A  A  A  A A  Legend  200  400  A  3/84 to 6/84  X  7/84 to 12/84  •  1/85 to 3/85  600  VFA (mg/L)  Figure 5.12 B O D  5  Linear Regression  Data Group  Slope  3/84 to 6/84 7/84 to 12/84 1/85 to 3/85  0.7158 1.2234 1.7396  Y intercept  172.6 75.43 20.19  vs. VFA  Results  Correlation Coefficient  No. of Data Points  0.8624 0.8539 0.9110  10 42 13  94 5.6 NITROGEN  Figures 5.4 A,B,C, show the variation of ammonia -N and TKN -N over the course of monitoring period. These figures show that, with the exception of a few early values, virtually all of the leachate nitrogen is in the ammonia form, as indicated by the very small difference between the total Kjeldahl and ammonia values. It is also readily apparent that the ammonia level of this leachate is quite variable within the narrow range of values recorded thus far. The ammonia concentration was generally between 10 and 50 mg/L. From Figures 5.3 A,B,C, there are two points to note about the nitrogen strength of this leachate. Firstly, that the ammonia concentration parallels that of the other constituents very closely, and secondly, that the ammonia concentration is much lower than the values of the other parameters, particularly during the first year. Proportionally, however, the ammonia level increases with respect to the other constituents over time. During the first eight month interval, the average COD/NH^ ratio was 79.5:1, but during the final six months, the ratio was 7.8:1, roughly ten times less. This reduction is attributable  to the decrease in the C O D concentration from an average of 2619, to  183 mg/L over the same period, rather than an increase in the ammonia concentration. Figure 5.13 shows the changing relationship between ammonia nitrogen and C O D levels graphically. This figure clearly shows how the ratio of N H ^ C O D shifts markedly during the 1/84 to 6/84 interval, which corresponds roughly to the transition phase between the acidification and methanogenic phases. A change of this magnitude in the N H ^ C O D ratio has important implications with respect to the treatment of such a leachate. Similarly, Figure 5.14 shows that the ammonia concentration is increasing with respect to the specific conductance, again due to a reduction in this later parameter. Therefore, the ammonia levels in this leachate are maintained over time, as has been the experience at most other  95  landfills.  5.7 TOTAL SOLIDS  A N D SPECIFIC  The results for Total Solids  CONDUCTANCE  and Specific Conductance were closely related to  each other and varied linearly with the other parameters (recall Fig. 5.1 Figure 5.15 and the associated  linear regression  A,B,C).  results, show more clearly the  correlation between these t w o parameters. The close correlation between total and specific conductance was due largely to the very low suspended of the leachate, typically less than 5 % ( < 7 5  solids  solids  content  mg/L), of which very little was volatile.  Therefore, the total solids residue was primarily made  up of previously  dissolved  material, including the ionic salts and organic  acids which are indirectly measured  specific conductance. Periodically, in response  to a sudden  change  by  in leachate flow,  large chunks of biological solids would slough off of the collector pipe and be washed  into the lift station wet well. These were the only incidents which  the leachate suspended  solids. The sandy soil layers beneath the wastes,  which the leachate must flow to reach to collector pipe, appear to filter suspended  solids  Figures  through most  out of the leachate.  5.5 A,B,C, show quite clearly that a change  of the leachate with the onset  of the methanogenic  the numerical value of T.S. and Specific Conductance Beginning  increased  in March  value until O c t o b e r  activity. Prior to March 1984, were almost  identical.  1984, the T.S. value decreased with respect to the Sp. Cond. 1984, w h e n  shown graphically in Figure figure and regression  takes place in the nature  a new steady  relationship is established. This is  5.15, and numerically by the linear regression  data. The  data show that the January to June period of 1984 was a  transition period in which the slope reduction in the total solids  of the relationship shifted downwards.  A  level can be attributed to the reduction in dissolved  96  NH  versus COD  3  D)  Legend CO  1000  2000  COD  3000  4000  NH„  vs.  Linear Regression  10/82 to 7/83 to 1/84 to 7/84 to 1/85 to  6/83 12/83 6/84 12/84 6/85  10/82  to  6/83  X  7/83 to  12/83  •  1/84 to  6/84  B  7/84 to  12/84  S  1/85 to  6/85  5000  (mg/L)  Figure 5.13  Data Croup  A  COD  Results  Slope  Y intercept  Correlation Coefficient  No. of Data Points  0.01264 0.01224 0.00685 0.01975 0.03363  0.217 3.703 26.50 16.26 19.63  0.9917 0.8906 0.5268 0.7082 0.4797  29 34 42 51 36  97  N H 3 vs. Specific Conductance 60 n  40-  E  Legend  CO  x Z  20-1  A  10/82 to  X  7/83 to  12/83  •  1/84 to  6/84  H  7/84 to  12/84  ffi 1/85 to 1000  2000  6/83  6/85  i  3000  4000  Specific Conductance (nS/cm)  Figure  5.14  Linear  Data C r o u p  10/82 7/83 1/84 7/84 1/85  to to to to to  6/83 12/83 6/84 12/84 6/85  NH  3  vs. Sp.  Regression  Cond.  Results  Slope  Y intercept  Correlation Coefficient  No. of Data Points  0.01711 0.01174 0.01835 0.02103 0.02464  -7.536 -0.185 -3.375 -6.174 -5.780  0.9827 0.8601 0.8828 0.9806 0.9523  30 32 36 46 35  98  species, both VFAs and heavy metals (the  pH increased from 6 to 7 over this  period). As shown in Figures 5.14  - 5.17,  there were fairly steady relationships  between the Specific Conductance and the other major leachate parameters. The relationships with the individual parameters changed during the transition phase as the landfill evolved, but apart from this brief period, the correlations were quite consistent. This raises the possibility that for situations where a similar correlation exists, the easily measured Specific Conductance values may be used for monitoring, and/or treatment process control, purposes (68).  5.8 METALS  Tables 5.3 and 5.4 summarize the results of the heavy metal analyses performed on this leachate. It is readily apparent that the metal concentrations in this leachate are, like the other parameters, moderate to low in comparison with other leachates (11).  Although the number of data points is small, it can be seen that the metal  levels appear to parallel the organic strength of the leachate. Therefore, these results tend to confirm the reduction in metal mobility with the onset of the methanogenic phase. As has been observed frequently by others (11,73), filtered and unfiltered samples of leachate gradually changed colour, from clear or pale yellow, to a rust brown colour, as ferrous ions were oxidized to the ferric form which precipitates as a hydroxide. This colour change was pronounced despite the low concentrations of iron in the leachate.  99  Total Solids vs. Specific Conductance 4 0 0 0 -i  X 3000-  A A  E ^  Legend  2000  O  if)  A  10/82  "co  X  7/83 to  12/83  •  1/84  to  6/84  Kl  7/84  to  12/84  ffi  1/85 to  n  1000  to  6/83  6/85  i  1000  2000  3000  Specific Conductance  Figure  5.15  Linear  Data  Group  10/82 to 6/83 7/83 to 12/83 1/84 to 6/84 7/84 to 12/84 1/85 to 6/85  4000  (i|S/cm)  Tot. Solids  Regression  vs.  Sp.  Cond.  Results  Slope  Y intercept  Correlation Coefficient  No. of Data Points  1.0847 0.9569 1.1241 0.7374 0.6314  -146.1 -58.05 -568.8 -137.7 -43.75  0.9965 0.9814 0.8747 0.9512 0.9572  34 25 31 46 35  100  COD vs. Specific Conductance 5000-1  Legend  1000  2000  3000  A  1 0 / 8 2 to 6 / 8 3  X  7/83 to 12/83  •  1/84 to 6 / 8 4  R  7 / 8 4 to 1 2 / 8 4  m  1/85 t o 6 / 8 5  4000  Specific Conductance (ujS/cm)  Figure 5.16 C O D vs. Sp. C o n d .  Linear Regression Results  Data C r o u p  Slope  Y intercept  Correlation Coefficient  No. of Data Points  10/82 to 6/83 7/83 to 12/83 1/84 to 6/84 7/84 to 12/84 1/85 to 6/85  1.3281 0.9397 1.0658 0.4050 0.2408  -541.6 -270.9 -986.6 -286.4 -127.8  0.9886 0.9407 0.6712 0.6402 0.6587  33 35 36 46 35  101  BODg vs. Specific Conductance 1500-1  A  A A 1000-  A  E  ^  A  LO  Q  O CQ  Legend 500  A  7/83 to 12/83  X  .1/84 to  •  7/84 to 1 2 / 8 4  6/84  1/85 to 6 / 8 5 500  1000  1500  2000  Specific Conductance  2500  diS/cm)  Figure 5.17 B O D . vs. Sp. C o n d .  Linear Regression Results  Data C r o u p  7/83 1/84 7/84 1/85  to to to to  12/83 6/84 12/84 6/85  Slope  Y intercept  Correlation Coefficient  No. of Data Points  0.6561 0.2573 0.1716 0.2128  -295.2 -90.17 -81.49 -180.2  0.9548 0.7294 0.5933 0.7291  16 11 42 12  Table  5.3 Leachate Heavy Metal Levels (AA)  Leachate Samples from the North Leachate Lift June 22 84  July 17 84  Oct 19 84  Nov 30 84  Dec 21 84  Jan 1 85  Feb 15 85  May 10 85 j .  3520  1527  460  377  138  264  1352  434  155  126  0.057 31.6 16.2 3.54  0.033 27.7 34.1 3.68  0.047 31.8 28.4  0.010 22.4  3.30  15.2 2.24  0.01 1 23.9 20.3 2.30  0.476  2.93  0.082  0.1 13  Ca  265 0.13 <0.02  Ni Pb Zn  St. Landfill  May 11 84  Cd Cr Cu Fe Mg Mn  (Well #1), Premier  June* ) 83 1  COD  Station  0.0068  0.4 185  0.026 60.7  0.013 63.2  0.023 53.5  49.0  37.3 5.98  27.0 5.18  27.9 4.28  1.77 0.01 13 0.0036  <0.02 0.420  0.019 18.6  0.420  0.124  0.293  0.218  0.1 10  (1) from Raina, 1984, (62) note: analyses by Atomic Absorption Spectroscopy (AA), all results in mg/L.  o  103 Table 5.4 Leachate Metal Analyses (ICP)  .CP. Metal Scan of Premier Leachate Samples Element (mg/L)  5/11/84  7/17/84  1/21/85  5/10/85  As B Ba Be Cd  <0.05 0.704 0.159 <0.001 < 0.002  <0.05 0.454 0.036 <0.001 <0.002  <0.05 0.461 0.069 <0.001 < 0.002  <0.05 0.342 0.138 < 0.001 <0.002  Co Cr Cu Mn Mo  0.168 0.014 0.018 4.83 <0.005  0.141 0.008 0.015 3.7 < 0.005  0.096 0.01 0.039 2.86 <0.005  0.058 <0.005 < 0.005 2.01 < 0.005  Ni P Pb Sb Se  <0.02 0.38 <0.02 <0.05 0.09  <0.02 0.05 <0.02 <0.05 0.08  <0.02 0.64 <0.02 <0.05 0.07  <0.02 0.12 <0.02 <0.05 0.05  Sn Sr Ti V Zn  <0.01 1.23 0.093 0.007 0.33  <0.01 0.777 0.028 0.006 0.236  <0.01 0.804 0.03 0.006 2.4  <0.01 0.528 0.026 < 0.005 0.092  Al Fe Ca Mg Na  I. 85 57.4 II. 7 291.0 37.4 109.0  0.08 48.5 8.2 138.0 28.0 84.4  0.15 32.9 8.1 162.0 28.4 83.9  <0.05 22.0 7.3 81.4 20.3 67.6  Hardness Ca, Mg Total  880 1000  460 556  522 592  287 287  Si  104  5.9 SPECIFIC TRACE ORGANICS  An interesting adjunct to this study were the results of two samples of Premier leachate which were analysed for some volatile, and semi-volatile, organic compounds. Table 5.5 presents the results of these analyses. A number of these compounds as indicated, are on the EPA list of priority pollutants. Most of the compounds indentified are found in solvents and paint products which often find their way into landfills. Harmsen (32)  conducted a more extensive analysis of organic  compounds found in two leachates taken from landfills in different phases of stabilization (acidification and methanogenic). He identified many different aliphatic, aromatic, and polar compounds; some similar to those found in the Premier leachate, and in particular also observed a strong toluene peak. Harmsen also found that the concentrations of these compounds were lower in the old leachate, but not to the same extent as the volatile fatty acids (VFAs).  105  Table 5.5 Leachate Trace Organic Content  Premier (Well #1)  Compounds (ppb)  * Benzene * Toluene * Ethylbenzene * Chlorobenzene * Dichlorobenzene m - Xylene o & p Xylenes 1 - methylethyl benzene n - propylbenzene 1,3,5 - Trimethyl benzene n - Butylbenzene  7/17/84  11/30/84  13.1 385.0 13.0 1.0 24.4 20.8 Trace 1.6 -  2.80 84.80 6.64 16.21 18.99  * Compounds on the EPA list of priority pollutants (-) = Not Detected Trace = <1 ppb  Trace Trace  6. PILOT PLANT  The RBC unit used for this study was a model S5 package plant manufactured by CMS  Equipment Limited of Mississauga, Ontario (see Figure 6.1).  Table 6.1 lists some specifications of this small unit which is rated for a maximum hydraulic load of 3400 L Id of domestic wastewater. The unit has three chambers: primary settlement, disk zone, and final settlement, of which only the disk zone section was used. The disk zone is divided into four stages, with the first having roughly twice the volume and disk area of the other three. Each set of disks consists of two outer fiberglass plates which support the interior disks made of thin plastic mesh (roughly 4 mm thick), with 10 mm square  openning.  The RBC unit was installed adjacent to the North leachate lift station at the Premier Street Landfill in May of 1983.  Figure 6.1 Photo of RBC  A small excavation was made to sink the  Prior to Start-up, Showing Disk Media and Influent Pump  I  106  107  Table 6.1 R B C  Make & Model Disk Diameter No. of Stages No. of Disks Disk Area Disk Zone Volume (net) Surface to Volume Ratio Rotational Speed Peripheral Speed  Figure  6.2 P h o t o  of R B C  Specifications  model S5, CMS Equipment Ltd. 0.9 m 4 36 (arranged 15,7,7,7) 47 m 245 L 190:1 m / m 6 rpm 0.29 m/s 2  2  3  Installed Adjacent to the North  Leachate  Lift  Station  108  Manhole Finished Ground  Lid Elev. 29.02  ink  M  Elev. 28.50  Check Valve RBC Influent A n d Effluent Return Lines  Transfer Hose From Pipe Tee to Bucket (3/4 in. ID)  Location ol Inlet Screen During Later Phase of Experiment (Inside 19 L Plastic Bucket M o u n t e d Beside Pipe Tee) Elev. 27.00  Lag Pump O n 25.60  Leachate Collection 25.00  All Pumps Off Sump Elev.  Check Valve  Pipe Inv.  Location o( Inlet Screen During Initial Phase of Experiment  24.70  24.40 Flygt Submersible Pumps  Leachate  Scale 1:31.6  Figure  6.3  Section  of  North  Leachate  Lift  Station  Showing  RBC  Connections  109  Figure 6.4 Photo of RBC Pump Inlet Screen  RBC to ground level so that the plumbing and electrical connections were below ground and out of harms way (see Figure 6.2). The electrical power line was run from the power panel for the lift station to the RBC though metal conduit. This supplied power for the 1/4 HP disk drive motor, as well as the leachate (and later chemical) feed pumps. Both the influent and effluent lines for the RBC were run into the lift station through a small hole punched through the wall of the wet well. To start with, the influent line consisted of 3/8 in. (.95 cm) OD plastic tubing run inside metal electrical conduit. Effluent from the RBC flowed by gravity back into the wet well via 1.5 in. (3.8 cm) dia. ABS plastic drain pipe. The effluent  110  return flows had no  significant effect on  the relatively tiny volume  pumped  the influent leachate characteristics due  through  the RBC, and the spacial separation  the inlet and return lines within the wet well (see During  the previous  the leachate was and 4000 mg/L.  few months  quite strong It was  maximum  flowrate of about 750  overload the RBC. The C O D / m * d or 23.3 2  with a no. 1717 the RBC  to pump  mL/min. (1080  would not average  L/day) would  g fitted  mL/min. flow, was installed in  approximately 3.4 m (see  Figure  Figure 6.3). An inlet  the end of the leachate influent line to help prevent  solids  6.4). others, is nutrient deficient (15,16), particularly of  phosphorus (P). Therefore, a solution of ammonium added to the first stage  added via a gravity fed drip system pump  1500  to  1  rated for up to 1680  This leachate, like most  4  be adequate  the leachate feed up from the lift station wet well (see  it (Figure  acid H ^ P 0 , was  less than  this basis a Masterflex " peristaltic pump  2  placed on  from plugging  varied between 2000  loading rate is approximately 34.5  g BOD/m *d. On  6.1). The required suction lift was screen was  study,  (the anticipated study period), so that a  corresponding  pumphead,  of  6.3).  5.1A). The C O D  assumed that the C O D  mg/L over the next 6 - 8 months  Figure  of planning and preparation for this  (recall Figure  to  was employed. Over  chloride N H ^ C l , and  of the RBC. The  solution was  phosphoric initially  from a constant head reservoir, but later a  the course  of the study, the concentration and flowrate  of the nutrient solution varied, (the N H C I addition was later stopped), but the 4  nutrient levels in the RBC  were maintained in excess  of the 100:5:1 ratio of  BOD[-:N:P which is generally accepted as adequate for g o o d bacterial growth. This level of nutrient addition is particularly generous with respect to nitrogen, in light of the findings the minimum  of other studies  here at UBC  (62,79,81,86), which  found  nutrient ratio for leachate treatment to be 100:3.2:1.1. A preliminary  Reg. TM, Cole-Parmer Illinois, 60648.  T  conducted  Instrument  Company,  7425 North Oak  Park Ave.,  Chicago,  111  jar test determined a phosporous demand of about 20 mg/L P due to precipitation with dissolved metals; however, this demand was accounted for by maintaining an effluent orthophosphate concentration of generally >0.5  mg/L.  The hook-up of the RBC and the mounting of its ancillary equipment was completed within three weeks and the RBC was ready for operation in early June of 1983. Various changes and modifications were made to the pilot plant and its support equipment during the course of this study, but these will be discussed in the following Section RBC Operation  .  7. RBC  7.1  OPERATION  START-UP  The RBC was filled with leachate and went into operation in early June of 1983. Seeding  of the RBC with bacteria was not considered necessary as a sample  of Premier leachate examined for a microbiology laboratory course had shown a very high bacterial count. This was soon borne out by the development of a bacterial film on the disks within two weeks. Warm  summer temperatures and the  relatively high organic strength of the leachate during this period (recall Figure 5.1A), doubtlessly contributed to this rapid growth. As observed elsewhere (57), the initial growth on the RBC  disks was  quickly supplanted by a more diverse bacterial  population. This transition is generally marked by a change in both the colour and texture of the biomass.  Figure 7.1  shows the light taupe colour of the short lived  initial growth. After a few more weeks, the growth on the first stage in particular, was much thicker, and the texture had changed from creamy smooth, to a spongy filamentous structure. As seen in Figure 7.2, the colour had also changed to a light rust colour which darkened with successive stages. The usual progressively darker brown colour of the biomass augmented  observed in sewage treatment (57), is generally  in the case of leachate treatment by the precipitation of iron oxides,  which explains the red tinge. Thus, within about six weeks, the biomass  had  developed to the extent permitted by the applied loading. The rapid development of the biomass  took place despite  numerous  interruptions of the leachate flow due to tubing failures in the Masterflex pump. During the start-up period, the affect of these interruptions was dampened  112  because  113  Figure 7.1  Photo of  creamy, taupe coloured,  initial bacterial growth (June  1983)  the first stage was connected to the large primary chamber, which acted as a reservoir. Tests showed there was considerable mixing between these two zones and the liquid was essentially homogenous. When the connection between the primary chamber and the first stage was closed off, and the leachate pumped directly into the first stage, then the flow stoppages became more problematic. This tubing problem was quite unexpected as this type of pump and silicone tubing has been used extensively at UBC without proir problems. The silicone tubing has a service life expectancy to 825 hrs. at 100 rpm according to the  114  I Figure 7.2  manufacturers  Photo of  mature biomass growth during start-up (late June  1983)  specifications . Yet in this instance, at about 60 rpm, the tubing often 1  failed to last the two to four day (48 - 96 hr.) period between site visits. Figure 7.3 shows a typical tubing failure (notice the dark leachate puddle below the head). Numerous using  adjustments such  as changing  a different type of tubing, and using  Installing two pumpheads  the tubing completely each visit,  a new pumphead, were  unsuccessful.  in parallel to reduce the rotational speed only  the frequency of tubing failures, and caused a second became  knotted up inside the pumphead The cause of these problems  doubled  problem when the tubing  and jammed the  pump.  remains unclear. According to the tubing  compatability data provided by the manufacturer, silicone tubing is sensitive to substances  found in the leachate such  only trace amounts.  some  a toluene, but these materials are present in  This would also fail to explain the problems with other types  1985 - 86 Catalog, Cole-Parmer Instrument Chicago, Illinois, 60648. T  pump  Company,  7425 North Oak Park Ave.,  115  I  Figure 7.3  Photo of Pump tubing failure  of tubing which have different sensitivities. Chemical compatibility also fails to explain why similar pumps used to pump the same leachate in lab scale experiments back at the university did not experience similar problems, unless some volatile component was responsible. Another possible cause is abrassion from particles and precipitates in the inlet line. Since the speed of the pump in this case was higher than most previous lab scale uses required, this may have pinched material beneath the rollers, which did not occur in previous experience. In any event, this experience suggests that the use of Masterflex pumps (or similar tubing pumps) for pumping leachate, particularly at speeds above 20 rpm, may be inappropriate  in some instances.  After a couple of months of trying to establish a reliable pumping regime using the Masterflex pump without success, it was decided to replace it with a Cormann Rupp Industries (CRI)'* bellows pump. This small positive displacement T  Gormann-Rupp Industries, Bellville, Ohio, 44813  116  pump on  is rated for a maximum  November  flow of 1730 mL/min. and was installed in the RBC  10, 1983. A much  smaller no. 1713 pumphead  the Masterflex pump, which was relegated to dispensing Initially the bellows pump  also had problems  inlet line due to suction. These  problems  was then mounted  the nutrient solution.  with broken valves, and a collapsed  were remedied by installing valve springs  to relieve the strain on the elastic valve stems,  and by installing a thicker walled  tubing and check-valve o n the inlet line. W i t h these modifications the bellows performed very well. The valve springs in particular should whenever  on  these pumps are used with the applicable  be recommended  pump  for use  poppet-valves.  7.2 THE DISRUPTIONS  Scarcely a week  after the new bellows pump  was installed, the first  of what was t o be a series of three major interruptions occurred. An heavy  rainstorm during the week  of November  the leachate lift station and the resulting p o n d  18, 1983, completely  mishap  unusually  overwhelmed  flooded out the RBC. Figure 7.4  shows part of the gooey aftermath of this flood. (Notice the high tide level of mud above As  on the electrical cord). The high water level was about  16 in. (40.6 cm)  the normal water level in the RBC, and just short of the disk drive motor.  the drive motor did not stop, oil washed  chain was whipped considerable amount  up into a frothy grey emulsion, which along with the of mud washed  bellows leachate pump Masterflex sustained motor windings shaded-pole  out of the oil bath for the drive  into the RBC, coated everything. Both the  and the Masterflex chemical pump were stopped, but the  the most  were shorted  serious  damage  controller and the  out. The one bright spot of this event was that the  type motor of the bellows pump  and only required a thorough  as both the speed  was not damaged  by the dunking  cleaning. Therefore the bellows pump  and a  117  borrowed Masterflex  pump were reinstalled in the RBC just a week later. This time  however, the pumps and electrical wiring were mounted on a platform above the high tide mark within the RBC. During the last six weeks of 1983, the RBC limped along, as minor problems such as the aforementioned broken valves, collapsing feed lines, and icing due to a December cold spell, caused interruptions. Then on New Years Day  1984,  another unusually heavy rainstorm caused a second major flood, which again stopped the pumps, and this time also stopped the disk drive, although the motor was not damaged. Once again the bellows leachate pump only required a good cleaning, so a second bellows pump of lower capacity was ordered to replace the shorted-out Masterflex  pump (the  repaired original pump had been re-installed just  10 days earlier), for dispensing the nutrient solution. It was then decided to dig up the RBC and raise it 1 m, to avoid the possibility of further flooding. Figure 7.5 shows the RBC in its new position. The location of the inlet screen in the lift  Figure 7.4 P h o t o of Aftermath of 1 I  s t  F l o o d in the R B C ( N o v e m b e r 1983)  118  station wet well was also raised to avoid increasing the suction lift (recall Figure 6.3).  It took only three weeks to move and reconnect the RBC but, in the  process, the RBC was drained and the biomass dried up. Therefore when the RBC was restarted on January 20, 1984, the biomass had to be re-established before the study could be continued. During February 1984, the previously mentioned poppet-valve springs and a check-valve for the inlet line arrived and were installed. To prevent further collapsing of the feed line due to the pump suction, the inlet tubing was replaced by a heavier walled 3/8 in. ID tubing. This was connected to the metal conduit such that the leachate now flowed through this conduit, and was therefore in contact with the metal. These modifications greatly increased the reliability of the leachate pumping. Also during this period, the biomass was regrowing quite rapidly despite the cool winter temperatures.  However, during the first week of March 1984, the  RBC was vandalized, which was the third major interruption to befall this study.  Figure 7.5  P h o t o of RBC after b e i n g raised  I  1m to avoid  flooding  119  One  of the vent covers was  the disk, and  60%  pried off and the drive chain derailed, which  of the biomass  was  partially dried out. The remaining  stopped biomass  grew anaerobically into a thick shaggy black mass. This vandalizism sparked a string of mishaps over the following few weeks, resulting in a burnt out drive motor  and  the drying of the rest of the biomass.  RBC  was  By the beginning of April however, the  back in operation with a rapidly growing  biomass  and the disruptions were  coming to an end. Figure 7.6A  shows graphically the erratic operation of the RBC  the disruptions, (October 1983 rate was that measured  upon  through  March  throughout  1984). The observed influent flow  arrival at the site during each visit. This value was  used to calculate the loading rates which prevailed at the time of sampling. The reset influent rate was that measured  at the end of each site visit after  maintenance procedures were performed, or the flow rate otherwise varied. An average rate for the preceeding period (between site visits), was calculated from the observed  and previous  reset rate values. With these terms explained, one can see  that the influent flowrate was frequently interrupted during this period.  7.3 A NEW  BEGINNING  From April 10, 1984, to July 24, 1984, the RBC except for one  operated  continuously  minor interruption of the leachate flow, caused by a fouled  check-valve. During  the first six weeks  itself. The new biomass and in the first stage  re-established  grew very rapidly over the dried mat of previous  growth  particularly, the new growth was very thick and shaggy. This  heavy regrowth of the biomass  was  organic loading applied (averaging temperatures. The  of this period the biomass  rough growth  no doubt encouraged  14.5  by the relatively high  g C O D / m ) , and the warmer 2  periodically sloughed  spring  off in large chunks, giving the  RBC Operational History: Influent Rowrate and Loading 800-1  a  o o  600 H  2  400 H  Legend  200 H  i )<!»<)<  3  10 17 24 31  OCTOBER 1983  7 14  NOVEMBER  21  28  S  X  i ani  12 19 26  DECEMBER  2  9  JANUARY 1984  FEBRUARY  20 27  X  Observed Rate  $  Reset Rate  •  Average Rate  rr 6 12  MARCH  36  CM  E ^» CD  +-»  (0  CC  •o  30 25 20-  Legend  16-  •  105-  0  i  " i  r—r  -i  1  COD Ldg.  RBC Operational History: Influent Flowrate and Loading 2000  1600-  A  u  I O O O -  2  x500 4>  XX  X  Legend  xX  11  APRIL 1984  -1  18  1  28  1  2  MAY  r 0  T  18  23  30  iX i  8  JUNE  13  1  20  r  27  XXX  T—  JULY  18  28  1  #  r  X 4  /  8  AUGUST  —i IB  1  22  x  r 29  .XX)00O< 8  12  SEPTEMBER  18  28  X-  Observed Rate  $  Reset Rate  •  Average Rate  3  OCTOBER  20CM  16® co CC  •a  Legend •  COD Ldg. BOD Ldg. • 1 N H Ldg.  10  E 2  3  D EiB  RBC Operational History: Influent Flowrate and Loading 1600  C  E  1000  E  H  © CO  o ^  Legend  500  c  CD 3  2  9  OCTOBER 1984  16  23  30  0  13  NOVEMBER  20  27  4  11  18  DECEMBER  28  1  8  18  JANUARY 1985  22  29  8  12  FEBRUARY  19  28  8  MARCH  12  19  X  Observed Rate  $  Reset Rate  •  Average Rate  28  APRIL  35CM  .E CD CO  cc ti>  •a  30-  Legend  2520  H  CD C O D L d g . E2 B O D L d g . •I N H L d g .  1510  3  50  -i  r  -i  r  123  disks a patchy appearance, and the colour of the growth was observed to be much darker brown than usual. Figure 7.7 shows the heavy patchy growth on the first stage during this time. This type of heavy growth continued until mid May, when almost all the rough growth quite suddenly sloughed off the disks and was washed out of the RBC as suspended solids. It appears as though this sudden general loss of the biomass and its dark colour, were at least partially caused by the underlying mat of residual biomass left over from the previous vandalism episode. Since the disks had rotated intermittently during this problem period, the biomass had not dryed out completely. A dry surface layer formed which probably protected deeper layers from moisture loss. When the normal RBC operation resumed and the new growth started, it appears as though this old anaerobic layer was revitalized. This produced an anaerobic layer which was much thicker than is normally developed. The extra  Figure 7.7 Photo of heavy dark growth on RBC during April-May 1984  I  124  thickness of anaerobic growth probably caused the large patchy sloughing biomass  and finally, the complete sloughing  viewed that one  mechanism for biomass  of the rough growth. It is generally  sloughing  is reduced adhesion  cells in the anaerobic layer. In this case it appears as though gradually broke down of the biomass  of  between  the anaerobic layer  until ultimately it came unglued completely. The dark colour  during this period was  oxygen levels, because  probably due in part to depressed  of the heavy organic loading o n the first stage  dissolved  and the  greater oxygen demand of the thick biomass, and in part from the dark colour of the thick anaerobic layer showing The biomass uniform thickness  through.  which replaced the rough growth was much thinner, but of  over the disk area, and small patches of distinct bacteria cultures  could be seen. Within another week the first stage biomass  had regained its light  rust colour and the bacteria were more homogenous. The biomass  continued to  evolve during the following two and onehalf weeks from June 1 to June 19, This was  indicated by poor floe settleability, and thus higher effluent  1984.  suspended  solids, due to the presence of fluffy filamentous floe particles. The settleability problem cleared itself up by June 19 and the RBC July 24,  operated extremely well  through  1984.  This  period of continous  mentioned transition phase was decreasing. Column  operation took place during the previously  of the leachate quality and therefore the organic  B of Table 5.2  over this period. Although  shows the range of leachate composition  the leachate pumping  loading rate decreased steadily from the 14.5 average of 7.7 can be seen  g COD/m *d 2  desired increases  rate was  increased, the organic  g C O D / m * d of April - May, to an 2  during July. This is shown graphically in Figure 7.6B.  that the influent flowrate was  during the previous  strength  It  maintained much more consistently than  period. The declining leachate strength, while frustrating the  in loading, also gave rise to a new maintenance problem. It was  125  observed  that biological growth  oh the inlet screen, and within the inlet line, was  increasing. Towards  the end of July the fouling rate became  on three successive  site visits the inlet line was  choked  unmanageable,  such that  off completely. In  to this intermittent flow and loading, a large proportion of the biomass from the disks. Following this loss of solids the RBC August 8 to September  4, 1984.  Despite  flow rates around  1 L/min., the loading  2  After September  periodic fouling caused  numerous  a relatively thin, but  4, another series of minor pump  problems  observed  declined. It appears  and  stoppages.  This biological fouling problem, which did not appear during the year, was  was ejected  operated fairly steadily from  rates were less than 5 g C O D / m * d and would only support healthy biomass.  response  to increase from May this problem arose  onwards,  because  previous  as the leachate strength  as the leachate strength declined,  the leachate in the wet well, and particularly in the intermediate bucket, was to become  increasingly  aerobic. Aerobic conditions, as well as  able  increasing  temperatures, greatly accelerated the rate of growth on the screen and in the lines. On  one  occassion  in particular, the inlet screen was  caked with a 0.5 in. (1.3  cm)  layer of bacterial solids, which had closed off the screen to the extent that it had partially collapsed under the suction of the pump. This growth occurred within the three days since the previous thoroughly  cleaned. Aside  visit, when  both the inlet screen and inlet lines were  from the rate of growth, aerobic conditions were  indicated by the light rust brown  colour of the growth on the screen, which  appeared very similar to that of the first stage indicates that metal precipitates (mostly problem. A single found  to be 38.9%  inlet line was  sample  bothersome,  growth  on the RBC. This  colour  iron oxides) were also adding to the fouling  of inlet line deposits  iron on  also  analysed for metal content  was  a dry weight basis. While constant cleaning of the  the problem only became serious  when the leachate  126  strength declined to very low levels (less than 250 mg/L COD). Once the leachate strength rose above 250 mg/L C O D in the late fall, the fouling rate decreased sharply, and became manageable with regular maintenance. The original bellows pump which turned at 165 rpm, the highest speed available for this type of pump, wore out a crank bearing by September 14, after approximately 10 months of continuous operation. A new bearing was easily fabricated but it appeared that this was an inherent weakness of this pump. The teflon bushing could not stand up to the continuous wear at this speed. Therefore, a twin bellows pump which turned at 50 rpm was installed on November 23 . Figures 7.8 and 7.9 show the original single, and later twin, 1.5 in. leachate bellows pumps respectively, as well as the smaller 0.5 in. dia. pump used for nutrient addition. By the time the pump problems had been ironed out, the wet fall weather had restarted the leachate flows. During December there were three mini floods  Figure 7.8 Photo of single bellows leachate pump (165 rpm) and nutrient pump I  127  Figure 7.9  Photo of twin bellows leachate pump (50  rpm)  and  nutrient pump  during which the leachate level in the lift station wet well rose high enough  above  normal levels to float the reservoir bucket and tip out the inlet screen. When these flows receeded, the inlet screen was  left high and dry, thus interrupting the  leachate flow. However, these heavy leachate flows also caused the washout  of  organic material mentioned previously (Section 5.3), so the organic loading of the RBC  increased dramatically between flow interruptions. The highest  occurred on  December  observed that there was  21, 1984. At a loading of 32.7 g C O D / m * d ,  it was  2  considerable foaming in the first stage  growth covered the surface of the biomass the  recorded loading  and a thin white  in the first and second  stages.  Although  loading rate fell sharply during the following week, this event caused a  noticeable increase in growth on the later stages, while the first stage  growth  became very thick, shaggy, and sloughed  gradually  in large chunks. This growth  thinned out over the next two weeks, but it provided a thick, healthy, biomass the  start of the next period of relatively stable operation.  for  1 2 8  A second  period of stable operation (two minor interruptions due to fouling)  occurred between January  18, and the end of March, 1985. The leachate  characteristics during this period are given  in column  C  of Table 5.2. Figure  shows the influent flowrate variation over this period. Since the C O D leachate was  generally less than 270  (less than 10 g C O D / m * d or 5.0 2  of the  mg/L, the carbon loading rate was g B O D g / m * d ) . However,  since the  2  quite low ammonia  levels in the leachate and the influent flowrates remained high, the ammonia rates were highest were possibly  during  this period. The  even higher when  temperature effects were taken into account. This  further in Section 9.2. Therefore, the nitrification  performance  during  The biomass had been for most  was  this period is of particular interest.  fairly thick and healthy looking during this period, as it  of the study. Figures  7.10  and 7.11  show the colour colour  gradation, and thickness, of typical healthy growth. Foaming with this leachate. Figure heavier foaming  7.12  shows an above  incidents during  first stage, with much  average  was never a problem  foaming condition. The  high loading had, at most, 6 in. of foam in the  less in following  stages.  Collection of treatment data from the RBC  was  suspended  on April 10,  1985. The end result of nearly two years of operating experience with this unit was  two periods  loading  effective loading rates for nitrification  aspect will be discussed of the RBC  7.6c  of two or three months  continuous  operation, and  RBC  numerous  shorter periods of a few days or weeks. While this operational history is less than was  hoped  for, the data collected is interesting none-the-less.  a listing of the RBC the RBC  operational history and field observations.  Appendix  2 contains  For the most part,  operated very well under difficult circumstances. The interruptions in  leachate feed and loading fluctuations generally had only a minor effect on the biomass  or effluent quality.  Figure 7.10 Photo of healthy bacterial growth  Figure 7.12  Photo showing leachate foaming in RBC first stage  8. TREATMENT RESULTS  The  RBC  unit performed remarkably well in treating this leachate under  difficult operating  conditions. A  g o o d effluent quality was  maintained  throughout  most of the fluctuations in hydraulic and organic loading. Recall (from Section 4 ) that the effluent for this study was taken as the supernatant fourth stage  liquid settled for 30 minutes  the final clarifier zone of the RBC  was  from a sample of the  in a 1 L graduated  cylinder. Effluent from  not representative of the R B C s  as settled solids were not removed, and considerable resolubilization of occurred in this  program  two or three times a week taken only once  RBC  was  program  basis, except during  a week. During  25, 1984,  the previous  of the RBC  to March,  process  1985.  liquid, were taken on  December, when  samples  a  were  period of operation, from July 1, the operation of the  stable enough to warrant a complete sampling  (recall section 7 RBC Operation  Loading  ). Appendix  and  analysis  3 contains the raw data from  samples.  rates were calculated from the measured  flow rate and leachate  or N H g - N , etc., concentration, at the time the sample was collected  (observed  rate). This assumes that the flowrate and leachate strength were  over the previous  detention time period of the RBC. At a flowrate of  mL/min., the theoretical detention time is 245 strength probably most  25, 1984,  the data is less complete because  not considered  the analysis  COD,  occurred from May  of the influent leachate, and first and fourth stage  1983, to May  organics  zone.  The main sampling Samples  performance,  minutes. In most  constant  1000  cases, the leachate  did not vary significantly over this period of time, particularly since  of the data was  collected over the later half of the study when the short  term variability of the leachate strength was  131  reduced. As the flowrate of the  pump  132  was  consistent, when The  it was  main sampling  running, these assumptions  program  seem  reasonable.  extended over both of the longer periods  steady operation, as well as many shorter periods of continuous from the two longer periods of continuous the RBC  of  operation. Results  operation demonstrate the abilities of  for carbon removal and nitrification of this leachate at low to moderate  loading rates. Some of the shorter periods of stable operation occurred during higher loading  rates of up to 22 g  significant. A study  2  by Filion ef al. (27), found that a RBC  steady-state conditions within about for carbon  B O D ^ / m * d , and their results are also  1 hour for carbon removal alone, and 3 hours  removal with nitrification, in response  loading. Since  would recover to  to a step increase in influent  these recovery times are shorter than the hydraulic retention time, or  the time required for substantial changes in leachate quality, it would be expected that the RBC  was  8.1  REMOVAL  CARBON  The good. An  essentially at steady-state during these short periods  carbon  removal efficiency of the RBC  effluent soluble  BOD^  and the settled effluent B O D ^ was  few samples  which did have  the loading rate had every sharp On  change  a percentage  a soluble  BODj-  in loading was  basis, the soluble  maintained for all but a  generally less than 25 mg/L. Those  greater than 10 mg/L occurred after not  followed by a significant reduction in treatment. BOD^  removal was usually above 95%. Most of by influent BODj. values so low that  values were relatively large in comparison. Figure 8.1 shows the  variation of effluent B O D ^ the loading  very  more than doubled over the previous few days. However,  the lower removal percentages were caused the effluent B O D ^  treating this leachate was  of less than 10 mg/L was  few samples  also.  over the main sampling period. This figure also  rates and percent removals  calculated for this data.  presents  RBC EFFLUENT B0D VARIATION with LOADING RATE and TIME 5  1000q  r10  *  CN  D)  100-  CD  E  U J  -5 LO  Q O  !< DC  CD  10-  CO  Q < O  100-,  o  JUN 1984  JUL AUG SEP O C T NOV DEC J A N FEB MAR 1985  APR  Legend  90  A  INF BOD  X  SET BOD  •  FIL BOD  L U  cc  10  Q O  03  80-  70  i  JUN 1984  r  JUL AUG SEP O C T NOV DEC J A N FEB MAR 1985  APR  134  Figure 8.2 presents the same effluent BOD,-  values versus the  corresponding  loading rates as a scattered plot. This figure shows that only 7 settled samples four filtered samples  and  exceed 25 and 10 mg/L, respectively, over this period. The  high settled values generally correspond  to higher suspended  solids levels, while the  high filtered values were preceeded by sharp loading increases or brief interruptions of the leachate flow. Figures  8.3 and 8.4 show that the B O D ^  removal was  occurring primarily in the first stage, and that generally less than 10 mg/L of additional B O D ^ The C O D  was  removed across  the remaining three  results paralleled those of the B O D ^  differences. Effluent soluble C O D  values  stages.  test with a few minor  generally ranged from 40 to 100 mg/L, as  show in Figure 8.5, indicating a relatively consistent refractory component. The size of this refractory component  did not appear to reflect the influent C O D  closely. This refractory component percentages  because  reduces the significance of the C O D  it accounted for up to 6 0 %  leachate was weak. Secondly, the range and filtered samples range  of B O D ^  component  of the influent C O D  between the C O D  removal when the  values for the settled  was, in some cases, considerably greater than the  values. These  solids were relatively high than 30%). This  values  differences occurred when the effluent  and the volatile component  indicates that the settled effluent C O D  from the highly endogenous suspended  variation of effluent C O D  corresponding suspended  of the solids was  low (less  contained a large refractory  solids. Figure 8.6 shows the  with loading rate over time, and also presents the highly  variable percent removal data. The  COD  and B O D  5  data shows that the RBC  carbon removal at loading rates up to .roughly  could maintain efficient  15 g C O D / m * d , 2  or 9 g  B O D / m * d , treating this leachate. This rate of loading would be considered only 2  5  moderate for sewage treatment. C O D the capacity of the RBC  results for a couple of samples  indicated that  is probably considerably higher but there were no  BOD  5  135  EFFLUENT F30D  vs. LOADING RATE  5  40-i  X  X CD E  o  30  X  IO Q  S  X 20  v  LTw  LU  Ei! UU _J  10J |  X X , X  o  • UJ  ,  X  X 1—1  X 2 X J  D  B  2  4 5  +  cn  Ul  ;J  UJ  u  x8a  #  UJ  -i  £  X  «P  f j i ^ X ^ +  B0D  o  + °  •  6  •  ft  Legend  ° X  • +  DATA PERIOD B OTHER DATA  •  •  DATA PERIOD C  10  LOADING RATE (g/m^d)  Figure 8.2 R B C Effluent B O D . vs. Loading  Rate  1  ST  & 4 ™ STAGE BOD  5  1000  100H  1  LO  Q O  CQ  }•1 10  / v.  f/  < 3 i r  Legend  1-1  JUN 1984  1  JUL  1  AUG  1  SEP  r  OCT  NOV  DEC  JAN 1985  FEB  MAR  APR  A  INF BOD  X  #4 SETTLED  B  #1 SETTLED  1  s  t  & 4 ™  STAGE  B O D  5  138  E F F L U E N T  C O D  v s . L O A D I N G  R A T E  X  200 -i  + o Ul -i  a cr  Ul  Legend  —I  10  1 20  r-  ^0  o  DATA PERIOD B  X •  OTHER DATA DATA PERIOD C  40  COD LOADING (g/m *d)  Figure 8.5 RBC Effluent C O D vs. Loading Rate  RBC EFFLUENT COD VARIATION with LOADING RATE and TIME lOOOOq  100CH  E Q  O  100d  CJ  10  J  100-.  < >  o UJ  cc  8060-  Q  O O  40-  N  D  J 1985  F  140  results for comparison. The BODj. results proved to be a better indicator of the treatment efficiency because increases in effluent  C O D could not otherwise be  attributed to an increase in biodegradeable, or non-biodegradeable, material. There was only one sample which indicated a possible overloading condition for the RBC. This sample occurred on December 21, 1984, during the organic wash-out event described in Section 5.3. The C O D of the leachate rose briefly to about 1350 mg/L which produced a loading rate of 32.7 g COD/m^*d. This loading resulted in an effluent filtrable  C O D of 152 mg/L, about twice the normal level, (see Figure 8.5)  which probably contained an increased biodegradable fraction. Unfortunately,  there  was no BODj. value for confirmation as this occurred over the Christmas holiday. The response of the RBC to this loading event will be discussed further in the following chapter.  8.2 NITRIFICATION  The nitrification efficiency of the RBC when treating this leachate was also very good. Generally the effluent  ammonia nitrogen (NH^-N) and total Kjeldhal  nitrogen (TKN), were less than 1.0 and 10.0 mg/L respectively. A large portion of the effluent TKN was presumably from the suspended solids in the settled samples. Nitrification was established during the first week on August 1983, after 2 months of operation. A month later, NH^CI was added to the nutrient solution and the nitrification capacity of the RBC was exceeded, as evidenced by high effluent NHj-N  levels. The effluent  ammonia levels varied between  30 and 60 mg/L, while  the corresponding nitrate levels were 70 to 90 mg/L. Unfortunately, the  nitrification  performance of the RBC could not be quantified over this period because the leachate flow was too intermittent, and the flow rate of the gravity fed nutrient solution was unsteady.  141  W h e n the RBC disruptions), it took  was  restarted at the end of March, 1984  another 2 months  until May  Complete nitrification was  re-established by May  been stopped, as it was  no  22, 1984,  (after the  to start nitrifying again.  28, 1984. The  NH^Cl  addition had  longer required to maintain the B O D ^ N  ratio below  20:1. Figures  8.7 A,B, show that nitrification of the ammonia nitrogen (NH^-N) was  very efficient over the main sampling cases, an effluent NH^-N loading rates, or system number  of samples  greater than 5.0  period except for a few samples.  concentration of greater than 1.0  mg/L  In  resulted from  from January to March, 1985,  had effluent ammonia  levels  mg/L, which indicated that the nitrification capacity of the RBC 2  temperatures below  days in December  1984,  when  8.8 presents  versus the NH^-N  g  level was  not observed  across  briefly on  a  During  to increase.  Discussion.  the effluent ammonia  (NH^-N) and nitrate (NO^-N) levels  Figures  1.0  levels with increasing loading rate. Effluent  mg/L generally occurred at loading rates greater than  8.9 A,B, show that as with B O D ^  nitrification occurred in the first stage, with less than 10.0 removed  stopped  loading rate as a scattered plot. This figure shows a trend  levels above  NH^-N/m^d.  at water  the water temperature fell below 5° C.  further in the  towards increasing effluent ammonia ammonia  greater. Nitrification was  however, the effluent ammonia  This point will be addressed Figure  was  10° C. Thus the effective loading rates, corrected for  temperature, were probably somewhat  these periods  high  upset due to an interruption of the leachate feed. A  being exceeded. The loading rates were greater than 0.7 g NH.j-N/m *d  few  most  the last three stages,  This will be discussed  removal, most of the mg/L  of ammonia  except during cold weather or heavy  further in Section  9.2.  1.0  being loading.  RBC NITRIFICATION PERFORMANCE E i  CO  O  + CO  10 17 24 31  16 22 29 6 13 2 0 27  APR MAY 1984  -o  E  JUL  JUN  14 21 28  SEP  AUG  x z  CO  Uj  0.6  < CC  O  z Q <  1  \ 1 1 1 1 1 1 1 1 1 1 1 1  13 2 0 27 4  APR MAY 1984  11 18 25 1  8  JUN  1  1  ••(•  - > Y  |  1  1—f  15 22 29 6 13 20 27 3 10 17 24 31 7 14 21 28  JUL  AUG  SEP  RBC NITRIFICATION PERFORMANCE 60  -i  5 12 19 26 2  OCT 1984  NOV  9 16 23 30 7 14 21 28 4  DEC  JAN 1985  11 18 25 1 8 15 22  FEB  1  8 15 22 29 5 12  MAR  144  EFFLUENT NHo & NOq -N vs. LOADING RATE 40-i  — 30H  X X X X  o  X  m  §(  xo<  X  X> 0<p X  X  x*<  9x • x  m  °  3 x  <  2 D  X  3  o x •  + •  1 -N  <  D  ~l  NH  z o  LOADING RATE  ^  Legend DATA PERIOD 6 OTHER DATA DATA PERIOD C  I  1.2  1.4  (g/m^d)  Figure 8.8 Effluent N H L a n d N O - vs. Loading  Rate  APRIL 1984  MAY  JUNE  JULY  AUGUST  SEPTEMBER  1°  1  & 4  1 n  STAGE N H  3  & N 0  3  -N  8.3 SUSPENDED  SOLIDS  In general, the suspended mg/L. Figure 8.10  solids levels in the RBC were quite low, <200  shows the variation of the suspended  solids levels in the first  and fourth stage. It can be seen that in most cases, suspended  solids levels  substantially above  200 mg/L followed loading, or leachate feed, interruptions.  Effluent suspended  solids, as shown in Figure 8.11, were less than 25 mg/L during  stable operation, and usually less than 100 mg/L during upset. The suspended were generally concentrated in clumps  solids  of biomass which settled rapidly. Although  the solids separation achieved over 30 minutes in the graduated cylinder was quite good, even better results could be expected from a properly designed The suspended  solids level in the RBC disk zone was observed to fluctuate  more during the periods of low organic loading ( < 3 conditions, much  clarifier.  g BOD^/m *d). Under these 2  of the biomass, particularly in the later stages, was  highly  endogenous and easily sloughed  off with the changes in organic and hydraulic  loading. The volatile component  of the solids decreased to less than 3 0 %  during  these periods. In fact, the volatile proportion of the solids appeared to be a good indicator of the general health of the biomass, varying from a low of 2 0 % , to a high of just over 8 0 % , depending upon the organic loading rate.  8.4  METALS  The results of the metal analyses, Tables 8.1 generally removed well as 5 0 %  over 8 0 %  of the iron (Fe), manganese (Mn), and zinc (Zn), as  of the copper (Cu), and lead (Pb), and lesser amounts  metals. Results from the analysis of a few samples RBC  and 8.2, indicate that the RBC  disks, Tables 8.3  of biomass  of other  scraped from the  and 8.4, indicates that the removed metals are concentrated  f  1  & 4  x  n  STAGE  SUSPENDED  SOLIDS  100003  Q  _|  MAY 1984  nun ll u r i H l l l l H u n n u i i n u  JUN  JUL  |  in m i nu nrj H  AUG  SEP  II n H nuiln |  OCT  f  nillin  NOV  Uipil  DEC  H  (  n  JAN 1985  n null nil I in u nil M J I I I M IIII nil lin |in nu  FEB  MAR  APR  ,  MAY  RBC EFFLUENT SUSPENDED SOLIDS _  IOOOO3  0  J  1 1 M J 1984  1 J  1 A  1 S  1 O  1 N  1 D  1 J 1985  1  1 F  r  1 M  A  M  150  in the biomass. The result from the one sample of the inlet line deposits tends to show that, particularly for iron, precipitation is a major removal mechanism. These precipitated metals are presumably then adsorbed onto the biomass, while further removal is affected by other mechanisms such as absorption, and chelation (6,9). The results from these few samples are not conclusive as to the metal removal efficiency of the RBC, but the observed relative affinities of the metals for biological removal, and the removal rates, are similar to those found for other biological processes (6,9).  8.5 SPECIFIC TRACE ORGANICS  An interesting adjunct to this study were the results of a few samples of Premier leachate and RBC effluent which were analysed for volatile and semi-volatile organics. Table 8.5 shows the results of these analyses. The results indicate that the RBC removed these compounds very effectively. However, it is not known how much of these compounds was removed by bacterial degradation, and how much was volatilized into the atmosphere. A number of the compounds indentified are on the EPA list of priority pollutants (see Table).  151  Table 8.1 RBC Heavy Metal Removal (AA)  Cu  Mn  Mg  Fe  Zn  May 11,1984 Stage Stage Removal  0.0263 0.0495 0.0727 -176.4  5.98 3.02 2.76 53.9  37.24 26.54 29.21 21.6  60.7 54.1 60.0 1.1  0.420 0.310 0.333 20.7  Well #1 June 22,1984 1 Stage 4 Stage % Removal  0.0130 0.0256 0.0088 32.3  5.18 2.19 1.61 68.9  27.0 23.93 23.68 14.0  63.2 3.65 1.56 97.5  0.1235 0.0628 0.0354 71.3  Well #1 July 17,1984 1 Stage 4 Stage % Removal  0.0225 0.0085 0.0119 47.1  4.28 1.29 0.711 83.4  27.88 22.49 24.53 12.0  53.5 6.25 3.62 93.2  0.293 0.392 0.0398 86.4  Well #1 1 4 %  Oct. 19,1984 Stage Stage Removal  0.0571 0.0546 0.0284 50.3  3.54 0.848 0.252 92.9  16.18 15.66 15.18 6.2  31.64 5.09 2.32 92.7  0.2184 0.0531 0.0514 76.5  Well #1 Dec. 21,1984 1 Stage (set) 4 Stage (set) % Removal  0.0333 0.0988 0.1099 -230  3.68 3.81 2.36 35.9  34.05 26.60 25.52 25.1  27.73 10.59 8.82 68.2  0.476 0.140 0.108 77.3  Well #1 1 4 %  Feb. 15,1985 Stage (set) Stage (set) Removal  0.0104 0.0248 0.0263 -153  2.24 0.808 0.244 89.1  15.18 13.80 11.55 23.9  22.35 4.68 2.03 90.9  0.0820 0.0568 0.0478 41.7  Well #1 May 10,1985 1 Stage (set) 4 Stage (set) % Removal  0.0906 0.0131 0.0126 86.1  2.57 0.75 1.13 56.0  20.92 12.44 12.64 39.6  25.0 5.07 5.92 76.3  0.1662 0.024 0.0745 55.2  53.9  83.6  27.5  93.6  77.9  Sample (mg/L)  Well #1 1 4 %  s t  t h  s t  t h  s t  t h  s t  t h  s t  t h  s t  t h  s t  t h  Average of 4 best % Removals  note: Lead (Pb) levels for all samplies <10. ppb  152 Table 8.2 RBC Metal Removal (ICP)  I.CP. Metal Scan of Leachate and RBC Samples Element (mg/L)  Well #1 7/17/84  1 Stage  4 Stage  As B Ba Be Cd  <0.05 0.454 0.036 <0.001 <0.002  <0.05 0.357 0.037 <0.001 < 0.002  <0.05 0.371 0.039 <0.001 <0.002  Co Cr Cu Mn Mo  0.141 0.008 0.015 3.70 <0.005  0.022 < 0.005 0.009 1.39 <0.005  0.011 <0.005 0.008 0.689 <0.005  Ni P Pb Sb Se  <0.02 0.05 <0.02 <0.05 0.08  <0.02 1.10 <0.02 <0.05 <0.05  <0.02 0.73 <0.02 <0.05 <0.05  Sn Sr Ti V Zn  <0.01 0.777 0.028 0.006 0.236  <0.01 0.61 0.022 < 0.005 0.02  <0.01 0.598 0.024 < 0.005 0.02  Al Fe Si Ca Mg Na  0.08 48.5 8.2 138.0 28.0 84.4  <0.05 6.96 6.0 114.0 23.4 72.9  0.14 3.51 5.9 111.0 23.5 75.2  Hardness Ca,Mg Total  460 556  382 398  374 383  st  th  % Well #1 Removal 5/10/85  18.3  92.2 46.7 81.4  <0.05 0.342 0.138 < 0.001 <0.002  1 Stage st  4 Stage th  <0.05 <0.05 0.258 0.255 0.183 0.098 < 0.001 < 0.001 < 0.002 <0.002  0.027 0.058 0.023 < 0.005 < 0.005 <0.005 0.008 < 0.005 0.008 1.54 1.23 2.01 <0.005 < 0.005 < 0.005 <0.02 0.12 <0.02 <0.05 <0.05  <0.02 3.71 <0.02 <0.05 <0.05  <0.02 3.58 <0.02 <0.05 <0.05  91.5  <0.01 0.528 0.026 <0.005 0.092  <0.01 0.35 0.028 < 0.005 0.022  <0.01 0.348 0.030 < 0.005 0.024  92.8 28.1 19.6 16.1 10.9  <0.05 22.0 7.3 81.4 20.3 67.6  0.15 7.35 6.3 58.4 12.1 51.3  0.17 8.44 6.5 59.6 12.3 51.3  287 331  196 212  199 219  23.0  % Removal  24.6 29.0  53.5 23.4  34.1 73.9 61.6 11.0 26.8 39.4 23.4  153  Table  Oct. 5/84  (AA)  Pb  Zn  <10.0  184.5  793.6  388780  11.3  96.5  Stage  27.9 37.3 5.0 13.2  1021.3 1824.9 1316.7 1154.2  5230.8 14552.1 22083.3 17121.9  255729 178821 116667 57961  21.3 42.6 24.0 18.3  418.5 1077.3 596.7 433.0  Stage Stage  22.7 10.0  849.4 869.7  5696.2 19326.9  159893 65478  4.3 5.7  2901.4 4148.6  1 Stage t h Stage  20.3 45.7  1584.2 1780.8  20342.1 54206.8  212554 108746  24.6 30.4  302.2 651.8  th  1 4  s t  t h  s t  4  Levels  Fe  Stage  1  Metal  Mg  2nd Stage 3rd Stage  May 10/85  Biomass  Mn  Inlet Line Deposits  4  RBC  Cu  Sample (Mg/g)  July 17/84  8.3  154  Table  8.4  RBC  Biomass  Metal  Levels  (ICP)  I.CP. Scan of Selected Biomass Samples Element (Mg/g)  1st Stage 7/17/84  4th Stage 7/17/84  1st Stage 5/10/85  4th Stage 5/10/85  As Ba Be Cd Co  <80. 767. <2. 8. 24.  <80. 1350. <2. <3. <8.  <80. 1260. <2. <3. 31.  <80. 1380. <2. 5. 53.  Cr Cu Mn  16. 18. 4940. <8. <30.  32. 32. 13000. <8. <30.  21. 24. 18000. <8. <30.  34. 30. 13300. <8. <30.  29200. 90. <20. 613. 45. 56. 403. 2060. 280000. <200. 43200. 2400. 1100.  41700. 80. <20. 885. 111. 50. 647. 4990. 186000. <200. 63900. 4100. 1500.  60300. 90. <20. 1020. 101. 61. 292. 3610. 289000. <200. 55600. 3700. 1300.  42200. 100. 20. 900. 114. 55. 655. 5070. 186000. <200. 63200. 4100. 1500.  Mo Ni P Pb Sn Sr Ti  V  Zn Al Fe Si Ca  Mg Na  155  Table 8.5  RBC Trace Organic Removal  Premier (Well #1)  RBC Effluent  Compounds (ppb)  * Benzene * Toluene * Ethylbenzene * Chlorobenzene * Dichlorobenzene m - Xylene 0 & p Xylenes 1 - Methylethyl benzene n - Propylbenzene 1,3,5 - Trimetyl benzene n - Butylbenzene  7/17/84  11/30/84  7/17/84  7/20/84  11/30/84  13.1 385.0 13.0 1.0  2.80 84.80 6.64  -  -  1.24  24.4 20.8  16.21 18.99  Trace 1.6  Trace  * Compounds on the EPA list of priority (-) = Not Detected Trace = <1 ppb  .  Trace  pollutants  -  .  Trace -  .  9. DISCUSSION  The results of this study, presented in the previous section, showed RBC could effectively treat this landfill leachate to remove degradable and nitrogenous treatment was  that an  carbonaceous  material, heavy metals, and some trace organic compounds.  This  achieved despite difficult operating conditions of variable and  intermittent loading. While these basic results are encouraging, there are a number of points or observations which warrant further discussion. Also, there are implications with respect to extrapolating this data to a full scale application. Finally, there are a number of practical aspects of the experimental program and pilot plant operation which require further comment.  9.1  ORGANIC  As BODj,  REMOVAL  presented previously, the carbon removal efficiency, as represented by  was very good.  Settled and filtered effluent BOD^ values were generally less  than 25 and 10 mg/L respectively. This effluent quality exceeds the most provincial requirement of 30 mg/L at organic loading rates ranging  BOD^ (19). These  up to 8.6  stringent  effluent values were achieved  g BODj/m^*d;  however, the majority of  the data relates to loading rates less than 6 g BODj/m^*d.  At these relatively low  organic loading rates, this effluent quality could reasonably be expected. The  organic  loading rates for sewage treatment are generally two to three times as much. Evans (25), surveyed  RBC plants for operational problems  plants were operating at loadings maximum response  of 19.5  between 10 and  g BOD^/m^*d. Paolini et al.  of R B C s , tried loadings  and presents data showing 15 g BOD^/m^*d, to a  (58), while investigating kinetic  between 8.7 and 39.5 g BODr/m *d 2  156  most  and  found  157  the effluent quality started to deteriorate above Wilson  19 g B O D g / m * d . Murphy  (54), operated a 2.0 m diameter pilot scale plant at between 4 and 36  B O D ^ / m * d loading. They found a fairly linear response  between B O D  2  loading up to a loading of about tended towards  2  <30  5  study to this one  with a B O D g  of 850  loading of 11.3 mg/L, or 9 7 %  stage The  at temperatures  above  were for a maximum  design  15° C, to achieve an effluent  leachate treatment study which could be found  conducted  mg/L and ammonia  g BODj/m *d,  profile of BODj-  levels averaged  reports the results of a  levels of 60 mg/L. At their highest  removal. At this level, first stage  oxygen (DO)  effluent B O D j  of  3.0 mg/L with a 1:1  19  recycle from the fourth  employed to reduce the oxygen depletion in the first stage. removal across  the RBC  indicated that extra capacity remained  if the loading could be increased further without  affecting first stage  removals.  adversely  The relatively low organic loading rates achieved during most of this were insufficient to accurately determine the maximum removal. During  greater than 800 2  capacity of the RBC  produced  flowrates. However,  7, continually postponed  for  the planned  was  loading rates greater than 30  the operational problems  g  recounted in  increases in loading, as it was  to establish steady operation at moderate loading first. By the  half of the study, when  study  the first half of the study, the leachate B O D ^  mg/L, which could have  B O D [ j / m * d at maximum  undertaken  organic  loading was nearing capacity as  in the later stages,  carbonaceous  in  in Montreal. They were treating a leachate  they achieved an average  2  already being  Section  and  2  the literature (with respect to carbon removal), Coulter (16)  dissolved  removal  mg/L.  In the only comparable  companion  g  15 g B O D g / m * d , after which the removal rate  a maximum. Their recommendations  loading of 17 g B O D , j / m * d , BOD  and  2  second  a reliable mechanical configuration had been achieved  the other calamities overcome, the transition phase  and  of leachate quality had begun  158 and the high influent BODj. levels were gone. At the lower BODj- values (OOO mg/L), the influent pump did not have the capacity to achieve the desired loadings. An exception occurred during the brief organic wash-out event. During this event, an estimated B O D  loading of 22 g BODj-/m *d was achieved. Although no BODj. 2  5  results are available for this brief period, because the BOD testing had been suspended over the Christmas holidays, the C O D results indicate effective removals at this rate, as will be discussed shortly. Figure 9.1 shows the plot of BOD^ removal rates versus the BOD^ loading rates. It is readily apparent that the BODj. removal rate was very linearly related to the loading rate over the range of this study. This follows the experience of many others (44,54). It can also be seen that temperature effects were apparently absent. This will be discussed further below. The regression results for this figure indicate that 98.4% of the applied BOD^ was very consistently removed. Figure 9.2 is a similar plot of C O D removal versus loading. This figure also shows a high degree of linearity, and most interrestingly, this relationship appears to extend right up to the highest loading point, which occurred during the wash-out event. This later data point indicates a BOD^ mass removal estimated to be 18.2 g BODj7m *d, which is 2  within the realm of possibility. A single point is not conclusive however, especially considering the large gap between it and the other data. The regression analyses indicates that an average of 91.5% of the applied COD was removed. Observations of the operation of the RBC during this period indicates that at this loading level the maximum capacity of the RBC was being approached. The darker colour of the biomass, and appearance of the white Beggiatoa sulphur-oxidizing bacteria (recall Section 7.3) are indicative of limiting conditions (25). Effluent COD values also indicated a somewhat reduced effluent quality. Unfortunately, this loading level did not persist long enough to assertain whether or not the results represent a steady-state condition. Since the loading rate jumped so  159  B O D REMOVAL vs. UNCORR. LOADING Temperature Effects 5  10-1  T3 *  2 O  LLJ rr LO Q  O  X  8-  CM  6-  •  A  A  AD  4-  Legend  2  CD "i 4  BOD  1  6  1  A  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  1  „ 8  10  LOADING (g/m *d) 2  5  Figure 9.1 B O D - Removal versus Loading Rate  Linear Regression Results  Data Croup  Slope  Y intercept  Correlation Coefficient  No. of Data Points  Temp. >12°C Temp. 8-12°C Temp. <8°C  0.9974 0.9922 0.9507  -0.1078 -0.1497 -0.0223  0.9978 0.9988 0.9993  28 9 14  Overall  0.9842  -0.0886  0.9982  51  160  30-i  COD REMOVAL vs. LOADING Temperature Effects •  Legend  20  9.2 C O D  Linear  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  40  COD LOADING (g/m  Figure  A  Z#  d)  R e m o v a l versus Loading Rate  Regression Results  Data Croup  Slope  Y intercept  Correlation Coefficient  No. of Data Points  Temp. > 1 2 ° C Temp. 8-12°C Temp. < 8 ° C  0.9547 0.8435 0.9119  -0.8687 -0.8937 -1.1268  0.9778 0.9661 0.9978  34 20 18  Overall  0.9145  -0.9258  0.9871  72  161 dramatically after a series of loading interruptions, it is likely that the RBC would require more than three days to increase its biomass a comparable amount, especially at temperatures around 6° C. Therefore, from the experience of this and other studies, the maximum capacity of this RBC unit would probably have been in the range of 15 - 18  g BOD^/m *d. As pointed out by Murphy and Wilson 2  (54,80), this would lead to design loading rates approximately 15% less to account for scale-up effects. The literature suggests that numerous modifications or variations of the RBC process can improve performance at higher loading rates. Coulter (16), reported the use of a 1:1 recycle from the fourth to the first stage, in order to reduce the influent concentrations in the first stage and thereby reduce oxygen depletion. In other cases, step feeding, destaging, and/or supplementary air diffusers, have been used in the first few or all stages to increase the performance of the RBC. Perhaps the best approach, especially for leachate treatment in which influent concentrations may be very high, is the use of air-driven RBCs. Studies have shown (39) improved perforemance for both organic carbon and ammonia removal with air-drive RBC units. These units have demonstrated greater resistance to oxygen depletion, a thinner, more active biomass, prevention of biomass overgrowth, and lower maintenance and energy costs. Figure 9.3 shows the BODj. % removal versus loading. This figure raises a number of interesting points. The first point is that the removal efficiency was generally very good, >95%. Of the results less than 95% however, all of these occurred at loading rates less than 5 g BODj./m *d, thus under light loading. This 2  seemingly incongruous result has a number of possible explainations. Many of these low removal percentages result from very low influent BOD^ values, which make the effluent BOD- relatively more significant. This reason was given earlier and is especially true for the lowest % removal values. A contributing factor however, may  162  be an effect observed in other studies (57,60), in which the removal percentage decreased with decreasing influent concentration. Poon et al. (60),  explain this effect  in terms of mass transfer rates, which would be reduced because of the substrate limiting conditions (low driving force). The final point to note is the effect of temperature  on the treatment  efficiency. From Figure 9.3 it can be seen, despite the scatter, that the best removal efficiencies in each temperature  group decrease slightly with decreasing  temperature. This small effect is also indicated by the subtle decreasing trend of the slopes in the regression results of Figure 9.1. The C O D data is more scattered and inconclusive. However, these slight temperature effects are much less than is normally observed for RBCs treating sewage. Murphy and Wilson (54,80) determined that an Arrhenius temperature  coefficient of 0 = 1.05  applied to carbon removal for  temperatures less than 13° C, for their sewage treatment study. Figure 9.4 illustrates the unnecessary distortion that this factor imparts to the data from this study. For this study, the linearity of Figure 9.1  indicates that essentially no temperature  correction is required down to 5° C. This result is supported by a similar finding by Forgie (28), 15°  who observed a very slight temperature  C. Coulter (16),  effect between 5 and  observed a similar lack of temperature effects in that study.  He found support in the literature for the notion that the reduced activity of the bacteria at lower temperatures can be offset by longer hydraulic retention times and/or high degrees of treatment, which increase the contact time of the wastewater.  This notion seems eminently reasonable, and as will be discussed  shortly, is especially apparent in the nitrification results. The absence of temperature effects on the treatment of concentrated wastes such as landfill leachates at long detention times, (>4  hrs.), has important design implications as required surface  areas are frequently increased over 50% to account for low temperature effects.  163  BOD 100-.  % REMOVAL vs. LOADING  5  A * ^X V " MA  A  X  X  •  95 A  o  LU CC LO  90 A  A Q  AX •  •  •  X  85  Legend  Q O  OQ  80-^  X  A  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  75 2  B0D  4 5  LOADING  Figure 9.3 BOD_ Percent  6  8  10  (g/m «d) z  Removal versus Loading Rate  164  B O D REMOVAL vs. CORR. LOADING Temperature Effects 5  10 -i  •o CN*  X  8  • •  O UJ  O  4-  Legend  2  CO  4  BOD  Figure  9.4  BOD,.  5  8  6  LOADING  Removal versus  Linear  Data C r o u p  Slope  Temp. 8-12°C Temp. < 8 ° C  0.8876 0.7156  (g/rri  Loading  Regression  Y  intercept  -0.1498 0.0098  „  Temp. >T2C  X  Temp. 8-12C  •  Temp. <8C  i  10 *d)  Rate  A  12  Corrected  for Temperature  Results  Correlation Coefficient  0.9983 0.9981  No.  of Data Points  9 14  165  One anticipated effect which did not occur during this study was the interference  of effluent  ammonia concentrations on the soluble BOD^ results. Since  RBC settled solids, or fourth stage liquid, was used as seed for the BOD test, it was anticipated that effluent  BOD^ values would reflect the effluent  ammonia levels.  However, except for one data point, there was no apparent correlation between effluent  ammonia levels and effluent  BODj..  Although the investigation of mathematical models of the RBC process was purposely excluded from the scope of this study, some brief comments on work in this area wouldn't  hurt. Researchers have taken a variety of different approaches  towards mathematically  modelling RBC process performance. Some, like Wu ef al.  (83,84), have developed empirical models which relate parameters such as flowrate, substrate concentration, temperature, effluent  number of stages, rotational speed, etc., to  quality using coefficients determined from fitting curves to historical data. A  well developed empirical model can predict RBC performance very effectively for conditions similar to those used to evaluate the coefficients, but the range of parameter values for which it remains valid is likely quite narrow. This is particularly true for the effects of different waste types. Therefore, a small-scale test run should be carried out to re-establish the coefficient values whenever a new type of waste is encountered. This disadvantage is not unique to this type of model however, as all models require calibration to new situations. An advantage of this type of model is its ease of use. The parameters used are generally those normally monitored, (temperature,  flowrate,  influent and effluent  concentrations), and fixed values of  system geometry, (no. of stages, rotational speed). A disadvantage of this type of model is that it gives little insight into what is happening within the process, so that when the model fails, there is no indication of what caused the problem. In an effort to develop a more theoretical model which would reflect the RBC process dynamics, various researchers have applied several kinetic approaches.  166  The process  kinetics employed  suspended-growth  cultures. O n e  are generally adapted from those used to describe approach which has met with popular success is  the application of M o n o d type kinetics (44,57,58). Equation  1 shows the RBC  form  of the relationship. The greatest difficulty encountered with the application of this and kinetic approaches  has been the estimation of the amount  other  of active biomass  for the  substrate in question. Since the biofilm is generally conceived to be layered with the different types  of bacteria (heterotrophs, autotrophs, etc.) concentrated in  specific layers, the biomass  is generally estimated by arriving at a thickness  and then multiplying by the disk area. Estimates BOD  removal have  substrate  ranged  from 21 to 200  of the active thickness  /im (26,58), depending  loading. Therefore, arriving at an estimate of the active  fixed-film process  is even more  Kincannon which avoids  et al. (44), used a simpler analogue  the evaluation of biomass  amounts.  (Z) for  upon  the  biomass  uncertain than for a suspended-growth of the M o n o d  Equation 2 presents  value  for a  process. equation this simplified  relation. The simplicity of this relationship was compelling enough to result in Figure 9.5, from which the values  of U  and K max  212.2  g BODj-/m *d 2  were determined to be 199.5  D  respectively. These  results are equal to 40.9  BOD^/1000 f t * d  respectively, which are roughly  Kincannon  for sewage treatment ( U  2  ef al.  f t * d ) . The significantly higher values 2  more  readily degraded  no  apparent  leachate  of U  m  a  x  lbs.  4 times the values determined by  = 10.0  m a x  and 43.3  and  and K = 10.4 g  Kg  lbs BODj/1000  indicate that this leachate is  than the sewage used in the other study. This finding fits  nicely with the observation of readily degradable  and  D  that most  of the B O D  volatile fatty acids. These  of the leachate was  in the form  results also indicate that there was  inhibition of the heterotrophic activity by any characteristic of the  composition.  167  A X. Z Q(S -S)  U  Q  = 4sR  A = disk area ( c m ) S , S = substrate concentration  J  -  S  Q= Z=  2  Q  X, = D  concentration  of active  i  (D  +  (mg/L) biomass  K =  flow rate (Ud) active biomass thickness  (cm)  M o n o d 1/2 velocity coefficient (mg/L)  (mg/cnrr)  U= k=  specific substrate utilization rate (1/day) M max. rate substrate utilization per unit active weight Y  — = U  U=  — B — * — — U QS/A max ^ substrate utilization rate (g B O D / m * d ) 2  5  U  = Maximum Removal Rate (g B O D / m * d ) 2  t;  +  of bugs (1/day)  (2)  — —  IT max K = R  Saturation Constant (g B O D / m * d ) 2  5  QS/A=  Applied  Loading (g  BOD,-/m *d) 2  168  1/BOD REMOVAL vs. 1/BOD LOADING M o n o d Kinetics A p p r o a c h 3-i  •  2.5$  2  J  O UJ DC  1.5-  A  •  Q O  Legend  m  H P  0.5-  1  1  1  2  2.5  1/BOD LOADING  BOD_  1.0586  Removal  Y  BOD /m *d) 2  9  5  (8  Temp. <8C  1.5  Overall 9  • 1  Slope  1  Temp. 8-12C  1  Data C r o u p  =  X  i  - M o n o d Kinetics  Linear Regression  max  Temp. >12C  0.5  Figure 9.5  U  A  5  intercept  0.0050 K = R  Approach  Results  Correlation Coefficient  No. of Data Points  0.9976  51  211.2 (g B O D / m * d ) 2  5  169  While the kinetic models the RBC  have proven to be fairly successful at modelling  process, their application is limited to conditions under which the kinetics  control the bacterial growth. As suspended-growth  systems,  kinetics almost always governs because  resistances are negligible across particle. This is one systems.  pointed out by Famularo ef al. (26), in  the relatively small dimensions  reason for the success of kinetic models  However in fixed-growth systems,  substrates  and products must  move  mass transfer  of a bacterial floe in  suspended-growth  biofilms can be quite thick, and  in and out of the biofilm generally from only  one direction. Under these conditions, mass transfer effects become  much  more  important and often determine reaction rates. It is quite likely that in many instances where kinetic coefficients have been evaluated for RBCs, they are in essence  macroscopic approximations of many mass transfer effects. Mass transfer models, such as developed by Famularo and Mueller ef al.  (26,53), incorporate both mass transfer and biological reaction kinetics into a comprehensive, albeit complicated, process  model. These models  have the  advantage  of being applicable over a wider range of operating conditions, as well as  having  the flexibility to predict the interaction between different groups of bacteria; carbon oxidizers, nitrifiers, denitrifiers, etc. This later capability has yielded some valuble insights  into the factors which affect the performance of the RBC  however considerable work numerous  still to be done  system. There is  to further refine these models (31). The  kinetic and diffusion coefficients used require further verification, and the  effects of temperature and hydraulic retention time observed in this study elsewhere) could be incorporated. As presented in these papers, the  (and  models  accounted for temperature by simply applying an Arrhenius coefficient, which as discussed  earlier, is not always appropriate.  170  9.2 NITRIFICATION  The nitrification efficiency of the RBC treating this leachate was also very good. Recall that the effluent  ammonia nitrogen (NH^ -N), and total Kjeldhal  nitrogen (TKN -N), were usually less than 1.0 and 10.0 mg/L respectively. This effluent quality was maintained at loading levels ranging up to 1.0  g NH^ - N / m * d . 2  The literature generally agrees that nitrification proceeds at a zero order reaction rate and therefore  depends only on the number of nitrifying organisms  (36,54,80). As the nitrifiers are considered to be concentrated within a relatively discrete layer on the RBC disks (53),  their number is proportional to the surface  area of the disks. In situations where nitrification occurs in all the stages, the nitrifier population would be proportional to the total disk surface area. This explains the strong relationship between the nitrification rate and total surface area observed in various studies (22,80). However, in other studies nitrification only occurs in the later stages of the process (57).  In these instances, there is no close  relationship between the rate of nitrification and disk area. The deferred onset of nitrification is associated with higher organic loading rates, which result in residual soluble BOD^ levels in excess of 30 g/L in the early stages. It has been a general observation in both fixed and suspended growth systems that nitrificaton is inhibited when residual BOD,- concentrations of this order exist. The reasons for this inhibition however are not well understood. Presumably in the fixed-growth  systems, the higher BOD^ concentrations cause higher growth  rates in the heterotrophic bacteria, which then concentrate in the outer layers, and thus reduce or prevent the penetration of oxygen and/or ammonia into the nitrifier layer. It follows that the nitrifiers would concentrate in a discrete inner layer, where they can compete against substrate limited heterotrophs, rather than be dispersed evenly throughout the aerobic biomass. The situation in suspended-growth systems is  171  less clear. Since the bacteria are completely mixed and in intimate contact with the substrates, there should should  be no  mass transfer limitations, and thus the nitrifiers  get a portion of the substrates  available. Hockenbury  series of tests which tended to show there was be inhibited by high B O D nitrification is  (35), conducted a  no g o o d reason  concentrations. However  in practice, the inhibition of  observed.  Recall from Section 8.2, that for a few days in December was  for nitrification to  1984, nitrification  reduced to zero as indicated by the effluent nitrate levels. However, since the  effluent ammonia  levels remained essentially zero, the lack of nitrification was  due to the low water temperatures ( < 5 ° anomally corresponded rate of 22.0  g  C). It was then realized that this  apparent  with the organic wash-out event. When the organic  BODtj/m *d is compared 2  to the ammonia  not  loading  loading rate of 1.08  g  N H - N / m * d , the ratio of B O D : N is 20.4:1, which matches that of the nutrient 2  3  5  requirements  of the heterotrophic bacteria. Therefore, despite the presence  established nitrifier population, the hetertrophs  apparently consumed all of the  available nitrogen for their growth requirements, which underscores discussed have  above. The  been found  however,  of an  issue is clouded somewhat  the inhibition  by the low temperatures, which  to reduce the activity of nitrifiers more than  heterotrophs;  as shown by Forgie (28), an established nitrifier population can continue  to nitrify down to temperatures  as low as 1° C. Also, the long  hydraulic retention  times which occurred probably reduced the temperature effects, as observed organic  removal, and as will be discussed  for  further below.  In this study, considerable nitrification (generally  >50%)  stage. This nitrifier activity reflects the the low rates of organic hydraulic retention times, resulting in high first stage BOD  occurred in the first loading and  removals  BODj. concentrations. Therefore, it is anticipated that there was  long  and low residual  a strong  relationship  between surface area and nitrification rate for this study; however, the data is  172  insufficient to be conclusive Figure 9.6  in this area.  shows the ammonia  nitrogen removal rate versus loading rate  relationship. It is readily apparent that the removal rates are linearly related to the loading up to the maximum regression points  results indicate an overall average  are ignored  than 9 5 % points  loading achieved in this study, 1.3  (see groomed  g NH.j-N/m2*d.  removal of 80%, but when  the outlying  results), the majority of the data indicates a better  removal efficiency. From  Figure 9.6, the causative effects of the outlying  are not clear. The scatter of the data at the top end of the graph  loadings  of 1.0  g  The  N H ^ - N / m ^ d ) could be due to temperature effects, the  (above RBC  nearing it's capacity for complete nitrification, or some other factors. Explanations  for  the few scattered points at lower loading rates were also not immediately apparent. To further investigate the factors affecting the nitrification performance of the RBC  in this study, the percent nitrogen removal was plotted against loading rate  and temperature. The results are Figures indicates a trend towards  9.7 and 9.8 respectively. Figure  reduced removal efficiencies at higher loading rates, as  noted earlier in Figure 8.9. However,  the data is again quite scattered and the  effect of temperature is difficult to assess. Figure 9.8 doesn't issue  much  as there is again considerable  removal efficiencies do  9.7  serve to clairify the  scatter of the data and the lowest  not occur at the extremes of loading or temperature. This  figure does tend to show a slight trend towards reduced removal efficiencies at temperatures less than 10° However, commonly  temperature effects o n  observed  or assumed for RBC  removal results discussed Arrhenius below many  C. nitrification were much less than is treatment, which parallels the  earlier. Murphy and Wilson  temperature coefficient of  0 = 1.09  20° C. Coefficients of this magnitude RBC  organic  (54,80) determined that an  applied to nitrification for temperatures have been determined or used in  studies. Figure 9.9 shows that, as with the organic removal example,  173  NH  3  - N REMOVAL vs. UNCORR. LOADING Temperature Effects  1.2  • • •  CM  •  0.8  LU  < < >  x  •  •  0.6H  O LU CC ' CO X  Legend  0.4  0.2-  0.2  NH  3  0.4  0.6  0.8  1  J.2  A  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  1.4  -N LOADING RATE (g/rrT*d)  Figure 9.6 NHL^ -N Removal versus Loading Rate  Linear Regression  Data C r o u p  Slope  Temp. > 1 2 ° C Temp. 8-12°C Temp. < 8 ° C Overall  0.8006 0.7999 0.7518 0.7718  Temp. > 1 2 ° (groomed) Temp. 8-12° (groomed)  Y  intercept  Results  Correlation Coefficient  No. of Data Points  0.0678 0.0531 0.0849 0.0750  0.9235 0.9694 0.9208 0.9433  29 17 19 65  0.9822  0.0034  0.9994  26  0.9461  0.0095  0.9945  41  174  NH -N % REMOVAL vs. LOADING 3  100-1  90 H  •  o  80  •  • X  • •  70  CO X  X  60 H  Legend  •  •  50 0.2  NH  3  0.4  -N  0.6  0.8  LOADING RATE  1  (g/rri  J.2  A  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  -1 1.4  *d)  Figure 9.7 N H „ -N Percent Removal versus Loading  Rate  N H  1  0  - N  3  0  %  R E M O V A L  v s .  T E M P E R A T U R E  -i  - ^x x 9  0  H  X  o LU  8  Z  70H  X X  0  X  rr  X  X  X  X  CO  X 6  0  X 5  0  5  1  0  1 5  Temperature  Figure 9.8  2  0  2  5  C  NH~ -N Percent Removal versus Temperature  176  such a correction (to 20°  C) is excessive. The reduced effect of temperature is  probably due to the long hydraulic retention periods, as discussed previously for organic removal. It was then decided to check the influence of hydraulic retention time (HRT), given that it had been an important factor in organic removal and that the scatter at high loading rates corresponded with higher influent flow rates. Once checks were made, it was observed that the other outlying points from Figure 9.6 also corresponded to high influent flow rates. Figure 9.10 shows the rather definitive relationship between hydraulic retention time and nitrification efficiency for this study. This relationship appears to be relatively independent of temperature effects, although the slight scatter at the corner of the graph seems to be temperature related. These results indicate that the nitrification efficiency is reduced sharply at hydraulic retention times less than about four hours. There is little indication in the literature surveyed that the effects of hydraulic retention time on nitrification have been investigated. Aside from the few studies found by Coulter, which compared temperature and retention time effects, retention time has generally been discounted as an important process parameter in RBCs. As pointed out by Wu et al. (83),  regression analyses have shown retention  time to be much less important than other parameters, but this may be because RBCs have generally operated over a narrow range of retention times. Since the ammonia levels in sewage are about the same as they were in this leachate, RBCs which have been operated to achieve complete nitrification have probably maintained hydraulic detention times of over four hours and therefore this effect may have gone unnoticed. Similarly, RBCs operated primarily for organic carbon removal generally have short hydraulic retention times, in the order of 0.5 to 2.0 hours, which would also fail to exihibit this effect. However, there is some other evidence of this effect as Mikula et al. (52),  attributed the loss of nitrification, while treating  177  N H - N REMOVAL vs. CORR. LOADING Temperature Effects 3  1.2 n  *  CM  •  H X  LU  •  •  rr  X  *  •  •  LTJJ  • •  •  0.6H  o UJ  rr co x  0.4 0.2  H  x • •  Ox  •  Legend A  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  NH -N LOADING RATE (g/m d) 2#  3  Figure 9.9 NH, -N Removal versus Loading Rate Corrected for Temperature  178  AMMONIA % REMOVAL vs. RETENTION TIME 100-1  An  A  90-  o  80  LU  rr  70-  Legend 60 A  A  TEMP. >12 C  X  TEMP. 8-12 C  •  TEMP. < 8 C  50 5  10  15  20  HYDRAULIC RETENTION TIME hrs.  Figure  9.10 N H , -N Removal  versus  H y d r a u l i c Retention Time (HRT)  179  cheese processing wastewater, to a drop in HRT from 16 to 9.5 hours. Therefore, further research should  be encouraged  to define the interrelationship of temperature  and hydraulic retention time, especially with respect to nitrification. The relationship between hydraulic retention time and nitrification efficiency is probably rooted largely in the growth kinetics and mass transfer rates which control the nitrification process.  However, retention time itself is rarely included as a  parameter in mathematical models  of the RBC  nitrification in RBCs parallel those discussed there is no need to repeat those  nitrification process. The models for  earlier for organic carbon removal, so  comments, except to present the results of the  Kincannon ef al. (44) approach as applied to the nitrification performance. Figure 9.11  shows the plot of 1/NH^-N removal versus 1/NH^-N loading. From the  regression 4.54  analyses the parameters U  g N/m *d, or 0.96 2  and 0.93  m  a  x  and Kg were evaluated to be 4.69  lbs N/1000 f t * d respectively. These 2  and  results show  that the activity of the nitrifiers is roughly 43 times less than the heterotrophs, which reflects the lower growth rate of the nitrifying organisms. The low rate is probably a major reason  for the observed  retention time effect.  growth  However,  since ammonia removal rates were comparable with those achieved in many other studies, there is no evidence that the growth rate of the nitrfiers in this study was inhibited or lower than normally observed  for sewage treatment.  The nitrification performance of the RBC favourably to results from RBC  treating this leachate compares  very  treatment of sewage and general aerobic treatment  of other landfill leachates. Removal efficiencies and loading rates correspond very well to those established for complete nitrification in sewage treatment applications. For example, Murphy and Wilson complete nitrification was  (54,80), found that the maximum  between 1 and 1.2  g TKN-N/m *d, 2  loading rate for  which relates very well  to the results of this study (recall Figure 9.6). The results of Murphy and  Wilson  are very typical for sewage treatment. Therefore the results of this study indicate  180  1/NH REMOVAL vs. 1/NH LOADING Monod Kinetics Approach 3  3  10 - i  8-  O U  6-  •  J  cc A  CO X  4-  Legend  A k  2-  _,  2  ,  !  4 VNH3 -N  Figure  9.11  6 LOADING  NHL  -N Removal  (  8  10  - Monod  Kinetics  Linear Regression  Temp. >12C  X  Temp. 8-12C  •  Temp. <8C  Approach  Results  Data Group  Slope  Y intercept  Correlation Coefficient  No. of Data Points  Overall  0.967  0.213  0.991  65  NH -N/m *d) 2  U  !  A  max=  4  6  9  <B  3  K = B  4.54 (g NH -N/m *d) 2  3  181  that the design nitrogen loading rates used for sewage treatment are applicable to this landfill leachate (considering  the prevailing organic  loading rates). As  indicated  previously for organic removal however, the loading reductions recommended low temperature conditions  may not be  for  necessary.  The results from other landfill leachate studies  generally indicate that  nitrification is readily achieved and maintained. Chian et al. (13) summarized that aerobic treatment processes are usually capable of 9 0 % typically produce been problems problems  effluents with less than 10 mg/L  -N conversion,  and  N H ^ -N. However, there have  encountered while trying to nitrify landfill leachates. Many of these  have resulted from the greater sensitivity of the nitrification process  upset. For example, Keenan ammonia  NH^  levels of roughly  et al. (43), found 1000  it necessary  mg/L, by 50 to 60 %  to  to reduce influent  with air-stripping, to avoid  inhibition of the nitrifying organisms. Robinson and Maris (66), found that nitrate production did not occur until the solids retention time (SRT), was greater than days while treating an old reduce effluent ammonia  leachate, and that an SRT  of 70 days was required to  levels to less than 1 mg/L. Their lack of success may  been due in large part to their inability to maintain adequate solids The MLVSS of their reactors were typically <100 maintain consistent  20  have  concentrations.  mg/L. Jasper et al. (42), failed to  rates of nitrification after it was  initially established and they  speculated that the fade in nitrification performance was due to toxic effects of accumulated metals, especially zinc (Zn). Therefore, while these cases may be exceptions to the rule, they demonstrate  that nitrification of landfill leachates  requires greater control and is less certain than organic carbon Although more  one  removal.  of the advantages ascribed to RBCs is that they provide a  stable environment  for nitrification (22), the few results concerning leachate  treatment are inconclusive. The results presented this study, as he found  by Ehrig (22) support the results of  efficient nitrification of three different leachates from  182  methanogenic  phase  (old) landfills. These  leachates had much  concentrations than the Premier leachate, ranging from 206 contrast, the study reported on by Coulter (16), observed  higher  to 1346  ammonia mg/L.  In  an almost complete lack  of nitrification. Effluent ammonia levels were in the order of 38 mg/L, while effluent nitrate levels were limited to 0.5  - 1.0  mg/L. The reasons for the lack of  nitrification in this case were not determined conclusively. Coulter speculates that if nitrification was  established during the first run, which was at a light BODj. loading  rate and coincident with warm water temperatures, (a fact that was not established analytically), it was then upset and lost because at the start of run #2. because  of the doubling of the loading rate  Nitrification w o u l d have then been difficult to re-establish  of the low wastewater temperatures ( < 1 1 °  inhibition of the nitrifiers , by something  C) during run #2.  within the 1 0 %  accepted by the Montreal landfill, was proposed  Toxic  of industrial waste  as a contributing factor.  The results of this study w o u l d tend to support the theory that some toxic effect was  responsible for the lack of nitrification in the Montreal study. Leachates  used in the two studies were quite comparable except that the Montreal leachate had mercury (Hg) Vancouver  levels of 0.5 mg/L, which is much higher than levels observed  in  area landfills. A more extensive analysis of the Montreal leachate's  composition may  have found other toxic and/or inhibitory compounds,  and organic. In the absence  both  inorganic  of toxic effects, given the experience of this study, it  seems implausible that nitrification would not have become established during the three month warm summer totally upset  period of run #1,  under the prevailing light loading conditions  temperatures. O n c e  and  established, the nitrifiers would not likely be  by just a doubling of the loading rate. During  rates were highly variable and doubled on various occasions  this study, the loading without even a loss of  nitrification efficiency, let alone loss of the process. Since hydraulic retention times were similarly long during the Montreal study, the loading and temperature effects  183  were likely moderated for nitrification just as they were for organic removal, and retention time would not be limiting. Therefore, in the absence of further information, toxic inhibition seems the most plausible reason for the lack of nitrification observed in the Montreal study. This would then underline the importance of leachate quality in determining treatment feasibility and performance.  9.3  RBC RESPONSE TO VARIABLE AND INTERMITTENT LOADING  As presented in Section 7 RBC Operation,  the RBC operated under difficult  conditions of variable and intermittent hydraulic and organic loading at times during this study. Overall, these conditions did not impair the process performance or adversely affect the biomass. Organic carbon removal and nitrification were observed to be relatively unaffected by the variability of the substrate loading, consistently maintaining a good effluent  quality. The resistance of the RBC to the effects of  variable loading were no doubt enhanced by the relatively gradual nature of the changes and the long hydraulic retention periods within the unit, ln the case of carbon removal, the low range of the organic loading rates was also a contributing factor. As pointed out earlier, Filion et al. (27) found that an RBC recovered in less than three hours to an instantaneous increase in loading. Therefore, given the more gradual changes in loading, and hydraulic retention times significantly greater than three hours, the RBC appears quite capable of responding to the loading variability observed during this study. However, for both carbon and nitrogen removal, a four fold increase in the loading rate over a four day period resulted in slight increases in effluent  BODj. and NH^ values. This indicates that larger, or  more rapid increases in mass loading rates would probably exceed the RBCs capacity to respond, without at least a temporary loss of effluent  quality.  Temperature would also affect the RBCs response time to increases in mass loading  184  rates. The variable organic loading was affect on the biomass above  discussion  or suspended  the biomass  generally observed to have only a minor  solids levels in the RBC. As implied in the  growth, and thus process  performance, was able to  adjust to the changing  loading conditions. Agian, the low range  generally avoided many  problems such  as oxygen  of organic  depletion, Beggiatoa  loading  growth, and  substrate inhibition, associated with heavy loading conditions. Suspended  solids levels within the RBC  periods of steady operation. It was solids  accumulated in the RBC  stoppages  of one  or two days, the RBC  normal rate, and then accumulate because ef al.  suspended  during interruptions of flow. During brief solids generally continue to slough  at a  of the lack of flow through the unit.  (54), observed that a flow of 1 0 %  wash out the sloughed  were usually lower during  observed in this study and elsewhere (54), that  suspended  Murphy  stages  of average flow was sufficient to  solids. This accumulation affect explains the higher  solids levels during periods of unsteady operation. Therefore, process  performance immediately after an interruption of flow depends  mainly upon the  ability of the final clarifier to handle this additional solids loading. Although it is expected that total solids production would increase slightly with loading levels, the data from this study was too scattered to establish sludge production rates. On sloughing  t w o occassions, and major loss  interruptions caused  such  an interruption in the leachate flow resulted in a general of the biomass.  It is uncertain why these two  a large loss of biomass  while many others did not. In  the first instance, in August 1984, warm temperatures may have increased the rate of endogenous November upsets  1984,  decay which w o u l d weaken the biomass. The second  instance, in  may have been a culmination of the various effects of  and declining temperatures. A  previous two weeks  previous  sharp decline in the loading rate over the  may also have been a contributing factor. Aside from these  185  two events however, the biomass was retained on the disks. During periods of very low organic loading, the biomass became endogenous, (the volatile component dropped to below 30%),  but was retained on the disks ready to assimilate higher  loads. Therefore, this study was able to demonstrate the good resistance of the RBC to any adverse effects of variable loading.  9.4 METALS AND TRACE ORCANICS  The determination  of some heavy metal and trace organic concentrations in  the Premier leachate and RBC effluent was supplementary to the main topic of this study. The small amount of data collected does not support conclusions beyond the general results presented in the previous section; however, some additional comment is possible; For metal removals, the removal rates and relative affinity of the various metal species for removal were very similar to results observed for activated sludge systems. The removed metals were concentrated in the biomass to levels comparable to, or higher, than observed in suspended-growth leachate studies (82,83), with no apparent adverse effects. The trace organic results indicate that an assortment of compounds are finding their way into municipal landfills and that these compounds are quite mobile and readily enter the leachate. This raises the question of whether or not greater control over the disposal of these types of materials is necessary. The RBC effluent samples indicated that these compounds were effectively removed during treatment but further research will be required to determine the fate of these compounds. If volatilization  or stripping into the atmosphere is the major removal mechanism, there  may be a potential for a localized health hazard where leachates are treated.  186  9.5  TOXICITY  A  number  leachate and RBC  of attempts were made to determine the toxicity of the Premier effluent using the Daphnia  ef al. (2). However,  problems  the dilution water blanks observed  procedure outlined in Atwater  bioassay  were encountered with the survival of the Daphnia  in  and therefore no reliable results were produced. It was  qualitatively that the Premier leachate was fairly toxic, which one  expect given the ammonia  concentrations alone. The  RBC  effluent samples  other hand were apparently non-toxic. This was indicated by the Daphnia  would on  the  growing  better in the effluent than either the stock culture or dilution water. Therefore, it was  indicated, but not conclusively, that the RBC  was capable of producing  a  non-toxic effluent.  9.6  IMPLICATIONS FOR  FULL  SCALE TREATMENT  There are many factors to be considered when encouraging  extrapolating from the  results of this and other studies, to a full scale application for the  treatment of the Premier or other landfill leachates. Of primary concern are the chemical and physical properties of the leachate to be treated. The Premier leachate used in this study was demonstrates  the RBC.  on  landfill; this aptly  leachate quality. There were no  indications that this  inhibitory to either the heterotrophic or autotrophic bacterial growth  on  Experience at this university and elsewhere (16,18,42,43,66), with other  leachates, particularly strong substrate  from a young  the effects that specific site conditions, such as climate, drainage  patterns, etc., can have leachate was  quite weak, despite coming  leachates, have shown that biological inhibition due to  concentration, heavy  metals, and other compounds, is quite  common.  Fortunately, in most cases the inhibition results in reduced reaction rates rather than  187  process failure. Nitrification has proven especially prone to inhibition. In many cases, some form of pretreatment  of the leachate was required before a stable biological  process could be established. Therefore, the loading levels and treatment efficiencies achieved in this study may not be as readily attainable with other leachates, especially much stronger ones. The results of this study, as well as those of Coulter (16),  and Ehrig (22)  for organic removal and nitrification respectively, tend to show that the design mass loading rates proposed by Murphy and Wilson (54,80) for RBC treatment of domestic sewage, apply equally well to the treatment of some landfill leachates. Their design loadings are presented in Table 9.1. The aforementioned studies indicate that these loadings levels may be applicable to relatively weak young leachates, as well as most old leachates for which nitrification governs the loading rate. Further research is required to both confirm the initial results of these few studies, as well as determine the ability of the RBC to treat high organic strength leachates. To reiterate, these loading levels are probably not universally applicable to leachate treatment, but only more experience will determine over what range of leachate quality they are valid. Therefore in the mean time, these design guidelines should be confirmed by pilot scale studies of the particular leachate to be treated. Aside from the site to site variation of landfill leachate quality, the changes which occur over time as a landfill stabilizes must also be accounted for in a treatment design. Organic carbon removal will usually govern the design of a treatment process when a landfill is young, or in the acid formation phase, but after the transition to the old, or methanogenic phase, nitrification will govern the design. Therefore, the treatment design should incorporate a high degree of flexibility of operation to permit adjustment to changing conditions. The modular design of RBCs has the potential to permit the movement of units between sites depending upon demand, as well as the simple rearrangement  of the staging or  188  Table  RBC  9.1  Design  Loadings  System  for  RBC  Treatment  Design Objective (Ave. Value mg/L)  of  a  Municipal  Wastewater^  Design Loading (g / m * d )  Parameters  2  15°C  BODr  removal  B  O  BOD  removal  B  O  5  B O D removal plus nitrification 5  - assumes primary mg/L filtrable - provides factors TKN removal (1) From Wilson  5 TSS S 20 5 ^ TSS S 30 TKN «s 3 D  s  2  0  B  3 0  D  5 (total) B O D Load (total) TKN Load (filtrable) O  D  L  o  a  d  9  7  5  10°C  7  6  5°C  6  0  15  12  9.3  0.60  0.39  0.25  clarified wastewater feed to RBC with 180 mg/L BODr, and 30 TKN, of 1.25 and 1.35 for BOD^ removal and combined BODj. plus to correct for diurnal flow variation.  e( al.  (80).  treatment flowpath to adjust for changing  conditions. A given number of RBC  stages  in a flowpath also tends to be self-regulating with respect to allocating surface area to carbon removal or nitrification, although  the later always defers to the former,  which may reduce nitrification performance  at high  Another  organic  loading.  difficulty with the application of biological treatment processes  to  leachate treatment is the large variation in hydraulic and organic loading which  can  occur over a short period of time at some landfills. Frequently, the mass of pollutants leached from a landfill increase with increasing fill, so that the hydraulic and organic study, the mass of C O D  the biomass  loading tend to increase together. During  released from the landfill was  fold over four days, after a prolonged  hydraulic flow through  observed  to increase  the this  eight  period of low flows, in a full scale plant,  would likely be unable to assimilate so  quickly. The results of this study however  much additional substrate that  did show that the RBC was very resistant  to less severe variations in loading and interruptions in leachate flow. Recall that a  189  four fold increase in organic loading over a four day period resulted in only a slight increase in effluent BOD^. In most full scale applications, some form of equalization, to modulate the loading peaks would probably be necessary for any treatment scheme. Where possible, recirculation of some of the leachate back onto the landfill is an attractive method as it can both hold-over flows until dryer periods, and reduce the pollutant load due to in-situ stabilization of the leachate. The  results of the RBC leachate treatment studies indicate that the RBC is  particularly well suited for leachate treatment. Other studies indicate that air-driven RBCs may be even more so (recall Section 3). The fixed-growth of the RBC provides much better resistance to variable hydraulic, and to a lesser extent, organic loading than suspended-growth systems. Air-driven RBCs would provide a high degree of operational flexibility, as well as permit the RBC to accept organic loadings in the first stages which would be problematic in a mechanical-drive unit. Such air-drive units more closely approximate the ability of completely mixed, or tappered-aeration plug flow, suspended-growth units to accept peak organic loadings. The  staging of an RBC process train potentially provides a protected environment in  the later stages for the nitrifying organisms, which would be further protected by their location in an interior layer of the biomass. Given the predisposition of the nitrifying organisms to attached growth, these factors probably contribute to a more stable nitrification process in the RBC as opposed to suspended-growth systems. While RBCs are more sensitive to ambient temperatures than suspended-growth systems, air-driven RBCs in particular can use warm air from the blowers, retained within the insulated covers, to ameliorate temperature effects. It was observed during this study that there was relatively efficient heat transfer between the liquid and the surrounding air, as a temperature between the first and fourth stage liquid.  differential of up to 4° C was observed  190  While the above discussion  has  proposed  numerous  advantages  of the  RBC  for leachate treatment, and especially for the air-driven RBCs, there are few apparent disadvantages,  there is no conclusive evidence that RBCs perform better than  suspended-growth  systems.  Recall from Section 3.2 that Henry (33) generalized  suspended-growth  leachate treatment as requiring SRTs of twice, and loading rates of  half, those used for domestic sewage, aeration mode  which roughly corresponds  to an extended  of operation. While the results of this study indicate that an  RBC  can treat landfill leachates at loading levels which are the same as those used for sewage  treatment, this does  not necessarily indicate an advantage for RBCs since  RBCs are generally considered to relate more closely to an extended aeration process. There have been few side by side tests of RBCs and systems,  and these have been inconclusive. None  treatment. O n e  have been conducted for leachate  problem with comparing the two systems  rates in the two systems,  suspended-growth  is relating the loading  as again, the estimates of active biomass  are determined  in different ways and are quite subjective. Further research and comparison with particular emphasis treatment systems one  the nitrification performance of the two types of  is required to determine if there is a difference. Therefore, until  type of treatment system  the basis  9.7  on  of economics  proves  EXPERIMENTAL P R O C R A M A N D  RBC  will continue to chose  on  to the sampling  been beneficial in hindsight.  OPERATION  as proposed  not carried out, but it still seems  a couple of changes  superior, designers  and personal experience.  The experimental program was  studies,  could not be evaluated because it  to be a valid approach. However, there are and analysis procedures which would have  Firstly, it w o u l d have been helpful to have determined  both a total and nitrification inhibited B O D  r  value for the raw leachate and filtered  191  effluent. This w o u l d have more clearly defined the residual carbonaceous nitrogenous  BOD,  as the total results did not reflect the ammonia levels on a  regular basis. Secondly, it would have been nice to have some dissolved values during  and  December  1984, when  the organic wash-out  oxygen  occurred, to confirm the  other indications of oxygen depletion. Other changes which were not implemented in this study were the use of load cells o n the shaft to give a measure autosampler to collect process  of total biomass, and the use of an  and leachate samples.  Load cells at either end of  the media shaft have been used in other studies with g o o d success, to easily determine a relative measure in Section 4, the use  of the total biomass  of areal biomass  (47). For reasons discussed earlier  determinations was not satisfactory during  this study. The use of load cells appears to becoming  more popular, judging  from  more recent studies, and it certainly has the advantage  of simplicity. Use of an  autosampler during this study w o u l d have permitted a characterization of short term fluctuations in leachate quality, as well as the collection of more process  data  during the periods of stable operation. An autosampler may also have proved useful to more closely study the response autosamplers  should  of the RBC  to loading fluctuations. However,  be used judiciously to test specific notions; because, while the  sample collection is relatively effortless, the analysis of those samples With respect to the RBC  is not.  pilot plant, its ancilliary equipment, and operation,  the extensive and varied experience gained during this study gives rise to a number of recommendations. The pilot plant itself was adequate for the purposes study but one  of this  useful modification w o u l d see the top cover fit over the lip of the  bottom section, rather than inside it; this would then prevent rainwater from entering the unit and affecting the hydraulic loading. Other modifications which are desirable are; stronger mechanism  mounts  in the clarifier  for the shaft and disk drive motor, a sludge  removal  to permit its use as a clarifier. and uniform rigid media  192  to help prevent the biomass bridging which occurred between  the flexible mesh  media. The two biggest operational problems aside from the natural calamities were the pump failures, and the biological fouling. Over the course of this study, the pump problems were more or less sorted out, and the Cormann-Rupp bellows pumps, with the valve springs installed, proved to be adequately reliable and easily serviced. The biological fouling problem was never adequately resolved. Susequent consideration of this problem had led to the suggestion that a relatively high capacity submersible pump (approx.  20 L/m) should have been used to lift the  leachate from the wet well into a short retention  time reservoir mounted in the  RBC. Feed for the RBC would then be pumped from this reservoir, through very short delivery lines, which would reduce fouling and facilitate easier cleaning. Excess flows would overflow the reservoir and return to the wet well. The line from the submersible pump to the reservoir would be much less likely to plug up because of high flow velocities and positive pump pressure conditions. Plugging of the pump inlet screen would also be less likely because of the higher flows and the use of a coarser screen. With these changes, hopefully many of the problems encountered in this study could be avoided.  10. S U M M A R Y  The results from this pilot scale study of RBC  treatment of a landfill leachate  indicate that efficient treatment can be maintained even under difficult operating conditions. Settled and filtered effluent samples  had BOD,-  values generally less than  25 and 10 mg/L, respectively. This effluent quality was maintained despite variable loading and frequent interruptions of the leachate supply. Settled effluent solids were less than 25 mg/L than 100  during periods of steady operation, and usually less  mg/L during upsets. Sharp changes in loading or interruptions of the  leachate flow were first reflected by increases in the suspended RBC  suspended  solids. Overall, the  demonstrated a remarkable resistance to fluctuations and interruptions in organic  and hydraulic loading. The RBC  operated under low carbon loading conditions for much of the test  period, due to declining leachate strength and pump BODj  loading was  less than 6 g BOD,-/m *d. However, a few samples had  loading rates, ranging effluent. These  limitations. In most cases, the  2  up to 18 g B O D j / m * d , 2  higher  and still produced a high quality  results indicated that the carbon removal capacity of the  treating this leachate was comparable to its capacity to treat domestic  RBC sewage.  Efficient nitrification of this leachate was also maintained throughout variable conditions. Effluent NH-j -N and TKN respectively. Nitrification was The average  and 10.0  mg/L  observed to stop under high organic loading conditions.  nitrogen loading rate during the study was approximately 0.6 g  N / m * d . Results 2  -N were usually less than 1.0  and loading rates for nitrification compare very well with those  found for sewage treatment. Temperature effects for both carbon removal and nitrification were offset by long hydraulic retention times, and for nitrification in particular, retention time  193  194 appeared to be a controlling factor. This result indicates that for concentrated wastes like landfill leachates, for which hydraulic retention times exceed four hours, reductions in loading rates at lower temperatures will be much less than normally applied for sewage treatment. Therefore, the design loading rates for nitrification and carbon removal developed for sewage treatment at moderate temperatures could be applied to the treatment of some landfill leachates over a wider range of temperatures. This study also indicated, to varying extents, that the RBC was capable of removing heavy metals and specific organic compounds, and produce a non-toxic effluent when treating this leachate. Overall, this study showed that the RBC is a viable process choice for leachate treatment and possibly has advantages over other systems, especially for nitrification.  11. CONCLUSIONS  1. This study indicates, although not conclusively, that the capacity of an RBC for carbon removal from this and similar leachates is comparable to it's capacity to treat domestic sewage. The design loading rates recommended by Murphy ef al. (54),  for BODj. removal from domestic sewage should therefore apply  equally well for the treatment of many moderate to low strength landfill leachates. 2. This study showed more conclusively that the capacity of an RBC for nitrification of this and some other leachates, is comparable to its capacity to nitrify domestic sewage. The design loading rates recommended by Murphy ef al. (54),  for complete nitrification of domestic sewage should therefore apply  equally well for the treatment of landfill leachates, except possibly in instances of toxic effects. 3. This study demonstrated conclusively that hydraulic retention time is an important parameter with respect to RBC treatment efficiency. For nitrification especially, hydraulic retention time appeared to be a controlling factor. The results also showed that hydraulic retention times of greater than four hours could effectively offset the temperature  effects which have been frequently observed  at lower retention times, for both carbon removal and nitrification. This result indicates that, for situations where sufficiently long hydraulic retention times are maintained, that loading rates need not be reduced in response to lower temperatures. 4. This study showed that the RBC process is remarkably resistant to fluctuations and interruptions of organic and hydraulic loading. Process effluent  quality was  not impaired by these variations, likely due to the moderating effects of the  195  196  long hydraulic retention time. Sharp changes in loading or interruptions of the leachate flow were first reflected by increases in the suspended solids, indicating that solids separation may be the controlling factor for  effluent  quality in these instances. 5. This study indicated that heavy metals are removed from the leachate and concentrated in the RBC biomass at similar rates and affinities for metal species as observed in suspended-growth systems. 6. This study showed that various specific organic compounds are present in this leachate and are effectively removed during passage through the RBC. Some of these compounds are on the EPA list of priority pollutants. The mechanism of their removal was not determined  however.  12. R E C O M M E N D A T I O N S FOR  FURTHER RESEARCH  1. Given the general lack of experience with RBC further studies should the RBC  be undertaken  to conclusively establish the capacity of  to treat landfill leachates of varying strengths  special emphasis  on  leachate should two systems,  and compositions, with  nitrification.  2. Side by side comparison  3. The  treatment of landfill leachates,  studies  of R B C s and Activated Sludge treatment of  be undertaken to evaluate advantages  or disadvantages  especially for nitrification.  relationship between temperature and hydraulic retention time effects  be investigated more design  of the  should  fully. Possibly the volume to surface area ratios of  could also be used as a factor to change  RBC  hydraulic retention time  and  reduce temperature effects. 4. The types and concentrations of trace organic investigated in more  cases,  compounds  and the major mechanisms  in leachate should of their removal  be  during  treatment determined. 5. The  denitrification of landfill leachate could be investigated using a  RBC.  If, in fact, the nitrification process  accumulation observed  by Ehrig (49)  investigated by Sam Turk 6. Ishiguro (56)  (PhD  is more stable in the RBC, the nitrite  may permit the short-circuit denitrification  thesis, UBC,  1986), to occur more reliably.  indicated in his paper that Japan has had considerable  treating landfill leachates with R B C s since 1976. When this paper presented  in 1983, there were apparently 135  wastes. This  RBC  indicates that a review of the Japanese  answers to many  questions  submerged  concerning  197  RBC  experience was  plants treating landfill literature may provide the  treatment of leachate.  Although unrelated to this study, it would be interesting to investigate feasibility and performance of a sequencing batch RBC for biological phosphorus removal.  13. REFERENCES  1. Atwater, J.W., and Bradshaw, G., "Operational Performance of an RBC Unit in Conjunction With a Septic Tank", Can. /. Civ. Eng., Vol. 8, No. 4, 1981, pp. 477 - 483. 2. Atwater, J.W., Jasper, S.E., Mavinic, D.S., and Koch, F., "Experiments Using Daphinia to Measure Landfill Leachate Toxicity", Water Research, Vol. 17, No. 12, 1983, pp. 1855-1861. 3. Ball, R.O., "The Effects of Hydraulic Variation o n Fixed-Film Reactor Performance", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y.C., and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 174 - 205. 4. Benefield, L.D., and Randall, C.W., "Biological Process Treatment", Prentice-Hall, Inc., New Jersey, 1980.  Design for Wastewater  5. Boyle, W . C , and Ham, R.K., "Biological Treatability of Landfill Leachate", J. Water Pollut. Control Fed., Vol. 46, No. 5, May 1974, pp. 860-872. 6. Brown, M.J., and Lester, J.N., " M e t a l Removal in Activated Sludge: The Role of Bacterial Extracellular Polymers", Water Res., Vol. 13, 1979, pp. 817 - 837. 7. Bull, P.S., Evans, J.V., Wechsler, R.M., and Cleland, K.J., "Biological Technology of the Treatment of Leachate from Sanitary Landfills", Water Res., Vol. 17, No. 11, 1983, pp. 1473 - 1481. 8. Cameron, R.D., and Koch, F.A., "Toxicity of Landfill Leachates", J. Water Pollut. Control Fed., Vol. 52, No. 4, April, 1980, pp. 760 - 769. 9. Cheng, M . H . , Patterson, J.W., and Minear, R.A., "Heavy Metal Uptake by Activated Sludge", J. Water Pollut. Control Fed., Vol. 47, No. 2, February, 1975, pp. 362 - 376. 10. Chian, E.S.K., and DeWalle, F.B., "Sequential Substrate Removal in Activated Sludge Systems Fed with Naturally Occuring Wastewater", Prog. Water Technol., Vol. 17, No. 2, 1975, pp. 235 - 241. 11. Chian, E.S.K., and DeWalle, F.B., "Sanitary Landfill Leachates and Their Treatment", /. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 102, No. April, 1976, pp. 4 1 1 - 4 3 1 . 12. Chian, E.S.K., "Stability of Organic 11, 1977, pp. 225 - 232.  EE2,  Matter in Landfill Leachates", Water Res., Vol.  13. Chian, E.S.K., Pohland, F.G., Chang, K.C., and Harper, S.R., "Leachate Generation and Control at Landfill Disposal Sites", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 14 - 30. 14. Clark, J.H., Moseng, E.M., and Asano, T., "Performance of a Rotating Biological Contactor Under Varying Wastewater Flow", /. Water Pollut. Control Fed., Vol. 50, No. 5, May, 1978, pp. 896 - 911.  199  200  15. Cook, E.N., and Foree, E.G., " A e r o b i c Biostabilization of Sanitary Landfill Leachate", J. Water Pollut. Control Fed., Vol. 46, No. 2, February, 1974, pp. 380 - 392. 16. Coulter, R.G., "Pilot Plant Study of Landfill Leachate Treatability by Rotating Biological Contactors", Proc. Second International Conf. on Fixed-Film Biological Processes, Arlington, Virginia, July 10 - 12, 1984, pp. 777 - 813. 17. Dass, P., Tamke, G.R., and Stoffel, C M . , "Leachate Production at Sanitary Landfill Sites", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 103, No. EE6, December, 1977, pp. 981 - 988. 18. Dedhar, S., and Mavinic, D.S., " A m m o n i a Removal From a Landfill Leachate by Nitrification and Denitrification", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 502 - 516. 19.  Department  of  Lands, Forests, and  Water  Resources, "Report  on  Pollution  Control Objectives for Municipal Type Waste Discharges in British Columbia", Water Resources Service, Victoria, British Columbia, September, 1975. 20. DeWalle, F.B., and Chian, E.S.K., "Kinetics of Formation of Humic Substances in Activated Sludge Systems and Their Effect on Flocculation", Biotechnology and Bioengineering, Vol. 17, 1974, pp. 739 - 755. 21. Duggan, J.B., and Cleasby, J.L., "Effect of Variable Loading on Oxygen Uptake", J. Water Pollut. Control Fed., Vol. 48, No. 3, March, 1976, pp. 540 - 550. 22. Ehrig, H.J., "Treatment of Sanitary Landfill Leachate: Biological Treatment", Waste Management and Research, Vol. 2, 1984, pp. 131 - 152. 23. Ehrig, H.J., "Laboratory and Full-Scale Experiments on Physical-Chemical Treatment of Sanitary Landfill Leachate", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 200 - 214. 24. Ehrig, H.J., "Biological Treatment of Sanitary Landfill Leachate with Special Aspects on the High Ammonia Concentration", Proceedings of the New Directions and Research in Wasfe Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 232 - 248. 25. Evans, F.L., "Consideration of First-Stage Organic Overloading in Rotating Biological Contactor Design", /. Water Pollut. Control Fed., Vol. 57, No. 11, November, 1985, pp. 1094 - 1098. 26. Famularo, J., Mueller, J.A., and Mulligan, T., "Application of Mass Transfer to Rotating Biological Contactors", }. Water Pollut Control Fed., Vol. 50, No. 4, April, 1978, pp. 653 - 671. 27.  Filion,  M.P.,  Murphy,  K.L.,  and  S t e p h e n s o n , J.P.,  "Performance  of  a  Rotating  Biological Contactor Under Transient Loading Conditions", j. Water Pollut. Control Fed., Vol. 51, No. 7, July, 1979, pp. 1924 - 1933. 28. Forgie, D.J.L., "Effect of Temperature on Rotating Biological Contactor Treatment Efficiency", Proc. Second International Conf. on Fixed-Film Biological Processes, Arlington, Virginia, July 10 - 12, 1984, pp. 730 - 753.  201  29. Friedman, A.A., Robbins, L.E., and W o o d s , R.C, "Effect of Disk Rotational Speed on Biological Contactor Efficiency", J. Water Pollut. Control Fed., Vol. 51, No. 11, November, 1979, pp. 2678 - 2690. 30. Gibs, )., Schoenberger, R.J., and Suffet, I.H., " A Simplified Buffer Capacity M o d e l for Sanitary Landfill Leachate", Water Res., Vol. 16, 1982, pp. 699 705. 31. Grady, C.P.LJr., " M o d e l i n g of Biological Fixed-Films A State-of-the-Art Review", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y . C , and Smith, E.D., N e w Jersey, Noyes Data Corporation, 1983, pp. 75 - 134. 32. Harmsen, J., "Identification of Organic Compounds in Leachate From a Waste Tip", Wafer Res., Vol. 17, No. 6, 1983, pp. 699 - 705. 33. Henry, J.G., " N e w Developments in Landfill Leachate Treatment", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 139 - 149. 34. Hitdlebaugh, J.A., and Miller, R.D., "Operational Problems with Rotating Biological Contactors", /. Water Pollut. Control Fed., Vol. 53, No. 8, August, 1981, pp. 1283 - 1293. 35.  Hockenbury,  M.R.,  Daigger, G.T.,  and  Nitrification", J. Environ. Eng. Div., EE1, February, 1977, pp. 9 - 1 9 .  Grady, C.P.L., "Factors  Proc'Am.  Soc.  Civ.  Affecting  Eng., Vol. 103,  No.  36. Huang, C.S., "Nitrification Kinetics and its RBC Application", /. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, No. EE3, June, 1982, pp. 473 487. 37. Huang, C.S., " S o m e Experiences with Fixed-Film Biological Processes", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. W u , Y.C, and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 422 - 449. 38. Huang, Ju-Chang, and Bates, V.T., "Comparative Performance of Rotating Biological Contactors Using Air and Pure Oxygen", J. Water Pollut. Control Fed., Vol. 52, No. 11, November, 1980, pp. 2686 - 2703. 39. Hynek, R.J., and C h o u , C C , "Development and Performance of Air-Driven Rotating Biological Contactors", Proc. 34th Industrial Waste Conf. Purdue University, Lafayette, Indiana, May 8 - 10, 1979, Ann Arbor Science Pub. Inc., Ann Arbor, Mich., pp. 805 - 815. 40. Ishiguro, M., "Current Status and Future Trends of Rotating Biological Contactors in Japan", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y . C , and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 466 - 484. 41.  Jasper, S.E.,  Atwater, J.W.,  and  Mavinic,  D.S.,  "Leachate  Production  and  Characteristics as a Function of Water Input and Landfill Configuration", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 64 - 83.  202  42. Jasper, S.E., Mavinic, D.S., and Atwater, J.W., "Influent Constraints on Treatment and Biological Nitrification of Municipal Landfill Leachate", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 843 - 863. 43. Keenan, J.D., Steiner, R.L, and Fungaroli, A.A., "Landfill Leachate Treatment", J. Water Pollut. Control Fed., Vol. 56, No. 1, January, 1984, pp. 27 - 33. 44.  Kincannon,  D.F.,  Stover, E.L.,  Emrie,  D.,  and  Jankovsky, M.,  "Comparison  of  Pilot Scale and Full Scale RBC Design and Operation", Proc. Second International Conf. on Fixed-Film Biological Processes, Arlington, Virginia, July 10 - 12, 1984, pp. 1840 - 1870. 45.  Knoetze,  C,  Davies, T.R.,  and  Wiechers, S.G.,  Biological Nutrient Removal Processes", 1980, pp. 171 - 179.  "Chemical  Inhibition  of  Water S.A., Vol. 6, No. 4, October,  46. Maris, P.J., Harrington, D.W., and Mosey, F.E., "Treatment of Landfill Leachate; Management Options", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference, University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 280 - 296. 47.  Marsh,  D.,  Benefield,  L.,  Bennett,  E.,  Linstedt, D.,  and  Hartman, R.,  "Coupled  Trickling Filter-Rotating Biological Contactor Nitrification Process", /. Water Control Fed., Vol. 53, No. 10, October, 1981, pp. 1469 - 1480. 48.  Masuda,  S.,  Watanabe, Y.,  and  Ishiguro, M.,  "Simultaneous  Nitrification  Pollut.  and  Denitrification in a Rotating Biological Contactor", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y.C., and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 227 - 246. 49. Mavinic, D.S., "Kinetics of Carbon Utilization in Treatment of Leachate", Water Res., Vol. 18, No. 10, 1984, pp. 1279 - 1284. 50.  McBean,  E.A.,  Poland,  R.,  Rovers, F.A.,  and  Crutcher, A.J., "Leachate  Design for Containment Landfills", /. Environ. Eng. Div., Proc. Am. Eng., Vol. 108, No. EE1, February, 1982, pp. 204 - 209.  51. Metcalf and Eddy, Inc., "Wastewater Engineering: Collection, Treatment, 2nd. Ed., McGraw-Hill, New York, N.Y., 1979. 52.  Mikula, W.J.,  Reynolds, J.H., M i d d l e b r o o k s ,  E.J., and  G e o r g e , D.,  Collection  Soc.  Civ.  Reuse",  "Performance  Characteristics of a Rotating Biological Contactor (RBC) Treating a Cheeze Processing Wastewater", Proc. 34th Industrial Waste Conf. Purdue University, Lafayette, Indiana, May 8 - 10, 1979, Ann Arbor Science Pub. Inc., Ann Arbor, Mich., pp. 335 - 353. 53. Mueller, J.A., Paquin, P., and Famularo, J., "Nitrification in Rotating Biological Contactors", J. Water Pollut. Control Fed., Vol. 52, No. 4, April, 1980, pp. 688 - 710. 54. Murphy, K.L., and Wilson, R.W., "Pilot Plant Studies of Rotating Biological Contactors Treating Municipal Wastewater", Report SCAT 2, Central Mortgage and Housing Corporation, Ottawa, Ont., 1980.  203 55.  Oleszkiewicz,  J.A., "Efficiency of Plastic Media Trickling Filters Operating Under Extreme Organic Loading", Progress Water Technol. Vol. 7, No. 3/4, 1975, pp. 607 - 615.  56.  Palit, T.,  and  Environ.  Qasim,  Eng. Div.,  S.R.,  Proc.  pp. 353 - 366. 57.  Pano, A.,  58.  Paolini, A.E.,  59.  Pohland,  60.  Poon,  "Biological Treatment Kinetics of Landfill Leachate", /. Soc. Civ. Eng., Vol. 103, No. EE2, April, 1977,  Am.  and M i d d l e b r o o k s , E.J., "Kinetics of Carbon and Ammonia Nitrogen Removal in RBCs" /. Water Pollut. Control Fed., Vol. 55, No. 7, July, 1983, pp. 956 - 965.  and Variali, G., "Kinetic Considerations on the Performance of Activated Sludge Reactors and Rotating Biological Contactors", Water Res., Vol. 16, 1982, pp. 155 - 160. F.G., "Accelerated Solid Waste Stabilization and Leachate Treatment by Leachate Recycle Through Sanitary Landfills", Progress Water Technol., Vol. 7, No. 3/4, 1975, pp. 753 - 765. CP.C,  Chin,  H.K.,  Smith,  E.D.,  and  Mikucki, W.J.,  "Upgrading With  Removal", J. Water Pollut. Control Fed.,  Rotating Biological Contactors for BOD  Vol. 53, No. 4, April, 1981, pp. 474 - 481. 61.  Pretorius, W.A.,  Technol.,  62.  Raina,  63.  Raveh, A.,  S.,  Masters  "Nitrification on the Rotating Disc Unit", Progress Water  Vol. 7, No.  2, 1975, pp. 243  - 249.  "Combined Treatment of Landfill Leachate and Domestic Sewage", Thesis, University of British Columbia, Vancouver, B.C., 1984. and  A v n i m e l e c h , Y.,  Models", /. Water Pollut.  2705 - 2716.  "Leaching of Pollutants From Sanitary Landfill  Control  Fed., Vol. 51, No. 11, November, 1979, pp.  Barber, C , and Maris, P.J., "Generation and Treatment of Leachate from Domestic Wastes in Landfills", j. Water Pollut. Control, Vol. 81, 1982, pp. 465 - 478.  64. R o b i n s o n , H.D.,  65.  H.D., and Maris, P.J., "The Treatment of Leachates From Domestic Wastes in Landfills - 1. Aerobic Biological Treatment of a Medium-Strength  Robinson,  Leachate", Water Res., Vol. 17, No. 11, 1983, pp. 1537 - 1548.  66.  Robinson,  H.D., and Maris, P.J., "The Treatment of Leachates From Domestic Waste in Landfill Sites", /. Water Pollut. Control Fed., Vol. 57, No. 1, January, 1985, pp. 30 - 38.  67.  Robinson,  H.D., and Lucas, J.L., "The Behaviour and Attenuation of Leachate from Domestic Wastes in a Landfill with a Designed and Engineered Unsaturated Zone, instrumentation and Research at Stangate East Landfill, in Kent, England", Proceedings of the New Directions and Research in Waste  Treatment and Residuals  Management  Conference,  University of British  Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 31 - 49. 68.  H.D., "Treatment of Domestic Waste Leachate in a Full-Scale Automated Plant", Proceedings of the New Directions and Research in Waste  Robinson,  Treatment and Residuals  Management  Conference, University of British  Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 166 - 184.  204  69. Salvato, J.A., Wilkie, W . C , and M e a d , B.E., "Sanitary Landfill Leaching Prevention and C o n t r o l " , /. Water Pollut. Control Fed., Vol. 43, No. 10, October, 1971, pp. 2084 - 2100. 70. Sawyer, C.N., and McCarty, P.L., "Chemistry for Environmental Engineering", 3rd. Ed., McGraw-Hill, New York, N.Y., 1978. 71. Smith, E.D., and Bandy, J.T., "State of Knowledge of Rotating Biological Contactor Technology", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y.C., and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 450 - 465. 72.  73.  "Standard  Methods  Public  Health  1971,  1975,  1980.  Stanforth, R.,  Ham,  for  Association  R.,  Examination  Inc.,  of  Water  and  Wastewater",  American  13th, 14th, & 15th editions, Washington,  Anderson,  M.,  and  Stegmann, R.,  Synthetic Municipal Landfill Leachate", /. Water Pollut. No. 7, July, 1979, pp. 1965 - 1975.  "Development  Control  D.C,  of  a  Fed., Vol. 51,  74. Stegmann, R., and Ehrig, H.J., "Operation and Design of Biological Leachate Treatment Plants", Progress Water Technol., Vol. 12, 1980, pp. 919 - 947. 75. Stegmann, R., and Spendlin, H., "Research Activities on Enhancement of Biochemical Processes in Sanitary Landfills", Proceedings of the New Directions and Research in Waste Treatment and Residuals Management Conference; ; University of British Columbia, Vancouver, B.C., June 23 - 28, 1985, pp. 84 107. 76. Sterritt, R.M., and Lester, J.N., "The Influence of Sludge Age on Heavy Metal Removal in the Activated Sludge Process", Water Res., Vol. 15, 1981, pp. 59 65. 77. Straub, W.A., and Lynch, D.R., Inorganic Strength", J. Environ. No. EE2, April, 1982, pp. 231  " M o d e l s of Landfill Leaching: Moisture Flow and Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, - 250.  78. Straub, W.A., and Lynch, D.R., " M o d e l s of Landfill Leaching: Organic Strength", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, No. EE2, April, 1982, pp. 251 - 268. 79. Uloth, V . C , and Mavinic, D.S., "Aerobic Bio-Treatment of a High-Strength Leachate", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 103, No. August, 1977, pp. 647 - 661. 80.  Wilson,  R.W.,  and  Murphy,  K.L.,  and  Stephenson, J.P.,  Biological Contactor Design", /. Water Pollut. Control March, 1980 , pp. 610 - 621.  "Scaleup in  EE4,  Rotating  Fed., Vol. 52, No. 3,  81. W o n g , P.T., and Mavinic, D.S., "Treatment of a Municipal Leachate Under Multi-Variable Conditions", Water Pollut. Res. /. Canada, Vol. 17, 1982, pp. 135 - 148. 82. W o o d , J.M., and Wang, H.K., "Microbial Resistance to Heavy Metals", ). Environ. Sci. Technol., Vol. 17, No. 12, 1983, pp. 582A - 590A.  205  83. W u , Y . C , Smith, E.D., a n d Cratz, J., "Prediction of RBC Performance for Nitrification", /. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 107, No. EE4, August, 1981, pp. 635 - 652. 84. W u , Y . C , and Smith, E.D., "Rotating Biological Contactor System Design", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, No. EE3, June, 1982, pp. 578 - 588. 85. W u , Y . C , a n d Smith, E.D., "Temperature Effect on RBC Scale-Up", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, Y . C , and Smith, E.D., New Jersey, Noyes Data Corporation, 1983, pp. 287 - 304. 86. Zapf-Cilje, R., and Mavinic, D.S., "Temperature Effects on Biostabilization of Leachate", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 107, No. EE4, August, 1981, pp. 653 - 663.  14. APPENDIX 1  Raw  Data of Premier Leachate  2 0 6  Analyses  u b c  e n g i n e e r i n g 207  r |  o  o_ -iq  <  c  U  J  O i—  In  V<!  o  O 0c "3- v9  In  >  !Y1 o  l/I  on r r  <si  o  tV ^1  V'l l/>  O A • rv? ll'l l/i  i  r- <>  In (V)  (VI  o  rr.1  u\  h  r-  (X 0 ir, Ln lfi v9  O r~ r - r~ rvi ro T "J" cr v5> rx T 1— V5 VTl VP (^1 e~ r - i vs 01 OJ (V (N ol -j  r—  1  ol  rflj — H  r-  vS> rl/>  0-  t>'| <^ > v9 o r~ ol (S H Oi r- Do r»j r- o | <T o 1—| VA v9 ffl •3- VA VS r~ l/> rs rs fN  6  l/l  \  Oo! o-i Lni lfi'  cr  r— tr- <rrrO ^~ "3" r i fVl (M  |E  f r cr  LO  f>, r- t « o" > V9 r[Y| CXl 1* V> TP ol tYl (VI  >  Li  c>  r  v5 vS>  i') VA  \ t  u  ol (3  vSJon O O-J (N tyi U\ rvi (VI  o|o <ri ui  CV)  0  rvi  ol mpji O 1 - vc (So 0 <r k ro Vl d rs Ol 6 > rI O (T > 00 . 1 ! rCr tr I~ O r- r - v9 V9 Vi 01 v9 tr o| olj lJ1 in J vS v» V> fl— 01 ol N 0) r0 (Yl (VI tvi (Yl r ty> (Yl (V\  5= v» V9 Vl <r  V\ ?r  >  6  v9  In ol  T vfl  i c/) > Q r~  i  1  _J  0  i i I i  in VP h  ?  t  i  01  |  O  .i  Z  z  i  L  vS  LU X  in ui  Z  1  i  O LA lA 6* t r r" \n ri VA  H  CVl  v9  fr  Ol ro  o  (—r ^.  o~ 00 0  v9  ci  £ v^  Z  6  1A T 1  z  6  cr  v5  \r\  ^  ol CM  In  Vi  i i 1 i  r-  Ol  ol 0Q O ro O r- Lr, ^~ c- U •36^ r Ui fVl rv> tr ui ZT ui  - ! • ? IA'|  i  i i  o o  8  i  1  h-  v9  i  in O o CD  o CD  ac a cr  uu  0  CN c r o-  v9  to  0  rS  q  •3-  Oo  i/i r~ i —  to Lfi  1 1  O r-  a.  LO \  oi rs  a  r-  U\  ~t)  0 "0  rr" V? (VI 0  r~  0  <j r ti t> 6 J w o o o Cx c i Z  z  ;^  —  !—  /\  C  -4-  LO  lo 0 O er0 rv-0 r cr (VI r N (V) fv  rv  U  (« r K>  ol (^ 0  Ln ol  rl  0  m tr rs r ra  0  L« Ul tr 0 ° r- ^ rrs M O v9 V9 <r rr H 0 cr m (V> CX) M 3- (V) (Y)  (Yl  '0  d  T~  c r 00 l/l  of OJ  cr cr v9  tr  -T  r<  Cr=> 3-  s: i £> d  r  U-  ^,  -j Q) »< LL.  Li.  l?  ID J>  s 6  I  V  0 /_  6  (Yl V  c  6  If.  .i  ol Cr v9 CN fN  52  w  c  r  ^  (3  2  T  0 (S  t~ H  M •J  ; jD  O  rj  (V a) >  r  i  s  u b c  e n g i n e e r i n g 208  o  zr ' o_  q  O  rt  ro  ti  (VI'  o  o"  <Y|  O \f<  vi>  rN  rs  (VI  r  q q  _  q  O  v i v9 v5  0  In  "I  v.9  TT  o~ v_o O c<>  6  0  O ZJ rJ c-J  (<n  o  <r r— (5~ tn cv r- ol i n c r O cr 0 rsi 0 cr o  o--  -  O  r~  L9  0-  ri  Kn  r  0 O  r--> 0-  0  -  bo  r^l  O  r  m  CV^  r - z r c>" r  r  O i'-.  v5  00  v9 v i -  r< O i.'j (.••• r~ 0 fi~ ZT O 0 r- 1 r r t>- O zr i s  cr  -  <r r  vr,  0  v3 v5 v i  vP r~ 0  f>  i.-i <r  0'c>  r"  3-  Ki  rS  H  i  ; _ . | >v\ — I  r>l  C>i O ro  "H </> \ r-  cy> v9 <w  zr  iw  /—  0  O j C*.; <?j r~ vS v'j ^ c , v 9i U'\j U l  (VI  <r 0 ^r  u\  (  Lo Lo (V) f zr vS ri r~ O  —  KI  f»  in Q  OO  ^n!  > r-  !  O m  _i  0  1  CP u>  r—  In!  r-  r-  \r>  h  j  -  v«i 1  ro  cr |  j  T  O Z  £  i  i c<> X O!  O  Z  c  H  \3 Z O a: I-  In  fits  pi  zr t r r-  zr Kl  £  k  6  ts  rn cr V? o l  ii  \*  r ; 3-  rl *zr  q  er  r"  (VO  c>  q KI  K>  bo r>t cN v9 c r cr i '  Lo CP r \  & c°  LO  O  q  zr cr> rs rs TM rs  r\  Do q  C»  r«i  q  6  ZT r -  0  o< a  1  rS c s  1.1 c r •  cr Lo Lri  cr  rv< c s rx  1  t-  *C \  q  t  z —7 *£6 .Y  0 0  q ff"  r~  0  vP in in ro i zr z r 0 00 rv_9 r L 0 cr Zr (NI (VI rs v5 50 zr! o° c r cs K . e>-- cV e» (S r- zr \fi ro M LO rs c>> \ (vi In LA! \ $ 1^ In ej r / rS r-; — zr rv rs rs rS rs N Kl rs CP O rSl rS CN cr CS \i In  0 O v9  N  t>.. ZT Ln m  o  I  In o Lr> to >--) vi> VP rs vS z r O v5 n i in rr~  o  0  n  c>  r(>Q  lA  In  LO (/< c< r s 0 ° cr LO L"J in U0  Lrv O LO r— r~ rs L O zr \f, m zr r s rS rS  6  v_^ Z  o  o  H M  (£  Q  0  u QJ  c  0 cr  J  r  in  zr  LI  cr cr 0  0  0  z r r >-Tl m rsl r vS <v\ L.-i zr ol cr In v9 CT V (VI v3 vS rj ^9 Co v3 CV) v3 H K l Ln H o zr v f \s\ er v3 -  <S r l  CN  0'  E h  -+tf O  r~ r— rr^ r " <r r s Lf>  r—  0  ?  H CL <U  \  O  O ' <J> i ~ (T c  CD CO Ct  U  (*> H  rj  CN  <s  O Ln p">  1  a>  ;  mi?  -  rs  in e~ , s r~ i ? ' <i  5  •> —•  j  h h n  h  Ln cr H  V"  i  6  3  •2 X  )  'JI  f J •  j  sT  0  lo Ln czr vS O in L0 z r r - Cx- fi- cn Vrj ts 0 0 0 (VI O cr rs 0° 0" c> O r s zr v9 0 IT (•'J 1/) zr t/i zr 3" rS r4 1)" cr lr> vS £-=  6  s  ) 6  Z3  U0>  <Z <? 1-1  f™  (V  —«-  -r  a/  ai  t'i  r  0  "5 00  -f  0  H  10  zr  1—  cr  ,-s  00  O  6  -r  0  0  0  0  0  —  0  0  u b c  e n g i n e e r i n g 209  tr tr-; rS v i lo v9  0  F - Lfi  fx/  ffl  0 0 vS  CO  r- c r Lo o| 0 tn co  h  lo  a  1  f  H  o~  r  rr In  0 T  tr r-  v y -  rs on  V9  vi  VS  V 9  cy  (V)  cr  ol  O Ln rvi rS t> cr O tr tr t>^ r/»  Ln rj- cr r J  r>)  H  v<D  rvi fv'  J  r>  f>  v9  O tr  v9  r>  tr  Lo  ro  V?  CTlfT-  0  £  t/7  _ !  VJ> vO  (J  CO  1ft  •  vi  c* 0 v5 r rr  0 "  c  —\ \S  fM  cr  v<?  0  (>-  ol  LA  0 rs ov r s H  LA cO v 9  o(  vi'vi  rs  c»  Lo  CN  O  O  ro  -  tr o<  oil  I'll  o4 ro O v3  rs  cs <S  L/  CO  Ol  vi  vi  0 0  Lr  rvj ro vi!  vi  v S Co  r-  ro cs cs  0 LO LA O H tr  >  vi  tr r O Oa Lo  rs  ol  -1  to tn >  vi  tr r~ cs  Lo  n  rs  r>-»  v9  IS  H  r~  rr  i  vi  L,  cr  cr  Cyo v 9  ii  CXD  L3  (Yl  tr  tr  i i  •  {  _J  0  !  LP  u.  1  >  j  tr ro Vn C~ 0cr vj> rj- Vn rr O Ln LT) 10  rr  0  rrr  i  v 9  i I  ?  i 01  (  ffl 0  f z  '  z  !Q  1  i  tp  rr z  L U \0  0  —7 ^~  <s er  LO  Z  \ i  Y-  r~i  "3-  6 _cs vS>  rs  <N  6 Ol  0 O O tor~ N O r O'l cn Li'.  £  z  (vi  r~  rA  rr  r—  Oo rr .-s o l rs jLo O joc CM  ca  t-  j1  i  •—'  0 ol  rr rs  co cr 6 rs  ro c-l  6 ol  rv>  LT. v i CO  ea  (O rsl OA  0* rs v9 r s  0=  £ S  rS  j  r<3— o l rA  rO  °? rf °? rrs|  Lf)  CS. c S  cs  (S  er- Lf) cr m r~ rs r S Lo rr rr LO  6 cS  r rr -  vA  N  vS  0*  5!  N  O rv> LA  (VA Vv  rS  0")  LA  O ZJ-  0r A ui  cr  cr O  rl  t~  Lo"  3-  00 r o 00 •3- <vi rA -3- •3- •3-  t>i  to  r-  rS  r~ -J  Lf)  v9 vS  r  0  •3-  ro  tr-  IJT  rs  LO  •S  _ro  cr rr 3-  (VA  LA  v5  LA  ro  Lo -j  1  -3-  io  v9  P-  r-  cr  rS vS  •3-  (> oi LO  tr  r— V9 V9  v9 vS  0 lo ct0 6 vS  -  cS CT _V5  rr rr  fO  cvS ro  ro  0  Ol  h3 ro  C»lo<"  vP  fO  1  vs  —;cr  rr.rr  Lnlco j  i 1  or cr  Oo  of  O O r t r rr 0  T rr v9 r r  ri v5 O rf)c s r r  fS  cr  O  0') •3-  cx  E  0  tf  -+-  o(|  i  i/i  0 0  — LO  <""irr vS;  O  O  r sij i  K  i  0 O CO CO  u  ro Civ9  > S | c r  r—  O  cs;  o< 0-1  5-  5? rr O Lo,  cr  LO! OSI C S  £ 0  0  rs q  rr rO LO  rr Ln r~  Ln Lf) }a: LO  ?  i i  LA  rvi  \i\  i WA  S3 Pt 0  5r  Cr-  v9  CVJ  tr-  Q  vS>  tN  ro  c^ Ln  cr CS  ol  u\  O c~  CA  r 3  r  _o  L'i  o<  6 6  0  J  0  OJ  Ql  z  O  O  v9  O  a  J  a'  O  tyi 0° to v5> ro L f i 0  vf)  0  rvi  rs  N  (J »  vP  CV)  0 f> v3 -• L> Q) O  (  V  0 s h  O t-  r -  >  c~ cy> (.,-.  (VI  rv'l ~T"  O  0  ty>  r  O  cr <v>  -  _D S> a' Li- U -  C~  (U  ol  r-l  v 9  Co rs  _D _ D  j5  <u U-  LL.  u.  r r r r tA r~ 0 vS O <r O — 1— O r~ L O v9 cr 0= o l VA rr rr cr O H cr v9 v 9 r~ 0 v9  0  cs vS H  (!)D  r  01  •> 0  2.  r  0  O O rs (V) L 5  0 0  6  ' > >. *>• 0  E r X X SZ <3Z  O  r~ -^ s-  r  1  >  u b c  e n g i n e e r i n g 210 -*  re  -J  I  1  1  rr  vi  v9  In  T  cr  rl rl  cT  rr CN O tr rvi rr cr rS LO  i* •*v 1/) rA cr  ro LT)  vJ  0°  J ConoJ.  xi  l/l  tn  h  rs  vJ3  F<  Ln In  r  vS  "5"  - •  - • vi vS v5  CN  r~  rr  VI  rs  •  -  Q  ...  —  r- cr r-  q  cr  cr  1  --  r4 \A  LA 0 cr cr  60 C r  v i vs v i v5> v i r-'  •J' V/> v5 r r r vJl vS vS tr r s IT vS r- In r LA 0 r 0° rs cr LA v i C 0 r~ OO r- rr 0 ° cr cr cr 7\ \ S r<  rr Cr rr rv5» v9  0  r~ O iA cr Q tr r - O r- L i cr 1A vS cr cr cs LA ^^ (VI rvi 0 H lA r-  0 cyi f> lA CT  v9 V5 vs  V Q  vS V S 0o O  r-  £2  L1  rx,  o l CS  (y  LA  O  tr  rr  UI P  cr  LA  u  i A  v9 ° t vS VS  r- rA tr vS v9 vS VS ~>  tr  O tr-  t r vS — - V.P O CT tr  cr  cr  vn 0 VA CO oO VA VA O 6 (vi O cv*i O cx O ( S 00 tr rS v9 O O 0° tr O CS cS V\ c« Cr* Lo CO OA rr 0 CP r - 0° Ln r~ vS r— r- r- r- r- tr \ S vS v5 vS v5 v9 In Ln Ln LTl Ln  —-  -  VI  r~ v i v9 v9 v i v9 v9  £ CO  r-  -J  0 to a  u.  >  rs v i cr lA LA tr V9, v9 rj- rr v9 L A  vi)  r~ 0=  i  ? « <*>  ._.  2  2 o  VS 0'  i <o rvi CO  X  2 Ui  rn  tD  O  traa 2 rv>  vA  rA r ~ rs lA vS CA >  cv In v9 rs: rH N.  rv>  O  0  V ?  vi  -A 0  V  _JO  m  rO  tr  IN'l po  rs  r-  rr  O  &  rr  cS rA  oo  !•>  to  r r vS (J* •a* cn  CO  **v?  crr-J  CN O  CM  d rfl  s  •a- cm  r-J c-<_ P-  u>  Z  rw  rO  2  -3 C*> CS 0 0 0 vT (S_ r v 00 O LO* rO r — CO t o po rs  rf  rs O vj> rs  vS> •3-  INI  fV\  rr- CS  rn Vi oi <yi rS.  0° 0 CP  CVl  VA CO O r-l Vo 1— VA •or LA ro" -3" r f v- c s CO •3" 0 cs. ro f A f A C X rs rs cs cs in Ln 0 Q O rA Lo 0 CS LA cs r— rS vS 3- C v j i> rs r~ ""0 "T cs. ("A r o r-s c s cS cs rA o< tc»s 0o LTJ C N c s CS O s  V"  r* rr  ro  O 0  (V* ro  rf  LCl  rf  r> CS  («  ol  O  ro  O  V?  A A  CNA  3"  co co  a  CC  ua  0 u  1  > m r~ co vS» rS r r *r  cr  a* cr  L/l  O rr  rs rvi  tr, cx O  rj*3H  CL  £ tf to  C  v.  rH  0  r-  r(  CS  tn  X  0  O a  r X  vn  Lf) t» Hi  h h  5  CS  r- ro m  &  tr  1—  rS  v9 vi>  m r<  C S  rA  rf rrrf  ol cJ  "H  VA  VA  Vri  CS. o l O l  OQ Vn rS Ln vS rLn O rA V_9 Vo rs cr r Lf> rj- LTi rA rfi  H  rH cr H  £  Hi  c  OJ  eJ %  3 h  v 5 err4 rs  o> a  v5  O  1  S "i r h  1  fV>  iI 5  L )  C S  ;-\  ) CS  -  CY)  r CT* -  O H H C rr rS H H  vJ>  rv) oJ  3  cr tr  (v'l v 3 r A V 9  r- O  '  6  h  0  Vi  v9 cS C  8  v ^ v9  c  T>  C5 c _T 0 0  Q.4 3Z X  0  vS c*» r r <n CO c x > cr 0° rs 1— rs  cs- LA  to  tj- 0 0  17" rf  V? rs  rd o< CT&o rr 0 vS> vn in rr r" cn U7 trm ro m cr rs- XT Ln m rr v>. £ rs r  rr O  0)  v9  oJ  ! 5"  5  "3  M  h  cv CSl  h CT  -4tf  o  r  -  cr v5 LA rvA CO VA O LA VA O rA C/A § )-T) 8m VAon rrOr L*n LArA ro rrrs rs VA LO  t>°  <5 (VA rf  1  rS v i  rf  rf  l—  o O  cr  1— Cr- rs CA rS rs 0 !  q tr  o cn tr rk r - 00 *A tr rr *3-  rr  -  CA Ln v5 CO v£ C O ro -3cv c s  CO  ro  tv\  en 0 a~  00  >  -  Cr rO tr rs m «"A c-s "o « o C° rs 0. hO "3 r~ lO 00 r- •o ,0 CO* tr zr in r r rr rA Z ra zr -3-  ol ol i A 0 0 cS O tr  .._  Vn Vri in "in r f CM rr rr  Z  tf  r^  -  ....  O  £  rj~ O rvi O •rr Lf) tr  •  00  « ~c~j  3 -|  \ —  O 6\  u b c  e n g i n e e r i n g  * r- u l/> CXT .,J» o_ \ S vS v i rOQ  ~s ^? c r f> • c v9 0 ° o  T  vj  LA  o  vS bo  O tr 0 °  °?  e  In CO  00 00 r °? 00 r-v9 v9 vi v S vi  rrs"  Csl vS \ S r- O O Oo r0 6^ cr oo o 0 r r— S5 0~-tfi O Oo <y r- v i  I- cr  tn  I  ff  in * J  a  O  m  O  OA  m  m  Oa  —  tr  t r VA -M 0 ° LO ks  _i 0 a Ul  > 1  -  -•  £ Oz  vS  0  z  2 i  «S  \i\ 0 O 0 tr tr or 60 -rs zr tr tj- t r z r  ro t r  on m  cs  vS  r~  Cs cs:  tj- C  V?  S &° r* \ S 1—  6  Ln  OT  1  vi  rs 0 0 irv  VA  0  VA  vc v9  vS  CS cS  tr °°. rn c \  fr  O O en  00  vS trot  tr- t r O CP t r  CO  CL  <x  ua  0 u  <u CX  £  \  m  vS cn t r fA  CS  O  1r  vi  ~r 00 cr v i v9 vs r~  tN O O 1A Vn LA Ul 0 rr 0 Cr vi r v5 v5 cn O O tS r cs O 0 LO  SOc%  v5 t r t r O 0 v5» O Do oC ol O 00 r- LO CTLD ti- t r tr vS ro rn Ln  r-  rr O CO v5 CS r— er v9 =r  tr  vi>  tr in 0- v5  ro  q cr  v5 CCi cs rri rn cr v i  vi  cs  rr  cs  07  cs  \.'\  cs  CS  0 in rcs rt  r  in tr In V A va Oo vS 1-7 * CO cn t-ri r f Lfj rr— c rs r- r— c-v  cr \A O r£ tr  c-v O 0? 00 cs e~ O t r v 9 c r cs  CP  q  0  O  CO O H Vn ~Q cvi O t r tr rf cr 0 cs rs cs CP  Q in =r K }  H  OJ  v3  Ln  OV  <r 0 tr  0° I-  0"  o s  r- 0 ° C P cr cs; CP cr  r  LO  cr  TT"  *-  cr v9 rA  S  %  O  ro  tr  v9  t r bo N rH ol  m cs- cs cr cf cr 0 if"  cr> 1— LO  6  'i  >-cj »tf rG .3 3 <E <JE  ox  Lrv  3-  •3- r-ti  v/i  • s rs rs  % •/>  r  o~  v S rty> 0 °  tr tr  r- 0  cs Lo 1  -> Si S>. • V ft! QJ • 0  IV)  1  0  rto t r  -  c< 0°  m cr  cr •rJ  -f -r  O C  H  r  -V  0 0  0  O  cr  ol vS 0 CA o l cr tr fA r~ Lf) CP t r r- CS ol CS  tr O >j\v-C r rN rs cs t r vS tr v3 v S v5 0° CS cn ot o i rn CS CS r r -  rr  v^  CS rs  O  ol O  6  cr  v3 0 M r- 0 t r 0 H —•  _& 0 jO 0 _ c c "2 2 72 0 O 0 O  O •3 0 c T2  a  -!  o  0°  -  1e 10  u CXi  co o( v3 t r CN r- on t r Ln rs" r- O CM oi in If) LO v^ 0° v9 r— r- in 0 ° vS v J O  C r c  v; r-'  rs c^  t r o r< t r 0 r>l O v_9 t r cs tf P- t r rs Cr of v ^ rS cs cs rn rs cs cs 2S vj o  O  <d Cr  z  z a o O  O  --  VA  c4  z  0  r-  vi  tv\ Wi M rCT v5 rs cS O O jbo cr  ttr tr c° rr t r t r  —  vi  <r  Oo 0b >  I" 0Q  q rS cr rJ  r-  ....  Z  O at  I A in Q 09 <s9 v5 v3  r  vi vi  VA  ...  o o o cs q I — 0 Vv \ A rs r c^ C-xl cs "tf rf •a CS of r~ cs INT. cs rs erf Ul a: <-< r\D  2  \9  O  ol M  r  Cr  O H CO r- in t r Nj> lA 0 0  VN  r to  1  r  vS  v i vS  -  3  1  ii  u b c  e n g i n e e r i n g  —  ir ZT vS vS tr zr vJJ v i v i v i v i v i v i "vS v i v i v i vi  o  TS C  In  00  r-  O r  CO  X h v3  in O 0 O O r- cx cs cr- l^ 0 ° X ^^ 1-1 tr  in  zr  r.0  TT  tf  ro  tr c<  ry>  ZT  cr  i \J  Q cr  O  0 0 0 IA N r- rCA O 0  e">  Ln in  rA O cn in cs" 0 r- 00 0 cs LA zr LA O cr O 01 cn rm T zr O v5 0~ r Cx> rv) sS O r cr r CT" r~ cs. (P cT C» CP r~ r~ [~ v9 v3 D«  cr  r  Ln Ln Oo m M O Cio vS u~-O P 00 0 Oo r~ O r- CV in vS ZT Ln zr cn Ln zr -J v j m in  00 tfi  r-  r—  cr  or  , :  6o cN  SS  212  r-  lA  vi s i vi vi  O  zr  I  0 v3 0- 0 oC 0 Lo  1 CX  v3 0  CA sS zr  cr  J  VA cv  tr ej zr  *  v i v i v i vi v3  Vn 0 O O O 0 VA 10 0 O sS zr cr O io In > »A rr O cr r0 v 9 bo Cr vS C LT) r tr tr LO cr 00 t » cr-vS  tr 00  -  r~  O s9 v3 s9  CP cn 00 cy> ol Zr r( zr ro Lri Ln  _J  0  CP  ^V  a: LO  rA  O r rs- ro D° r— cr in zr OA vS O fA r v3 s9 O in rA ZT on zr v^  0° c x  r 00 ei Ln s9 -  LL oi >  1O r-  zr Ln zr 0 °  hi  0  N  r-  v3  m o< 0  VA  v9  tr e(  et  rO ol  z 1  m  O  *—' z  zz  z  I  co X  vn rs cx  in w> 0 r cr tr vrz cs 5> cn = r -  Oz  Z Z .y  tr  <-n  cs rv-> cvi  tr  o  r-^  CS. <-s ex J-_ 0O -r c s ox- r- vS 06 en c S c s C-S  1  Ln r- 0  m rH in 0  .5 CS m  r s CS c s rs. c s x/\ cs  NS  r O  \-  U-) rs r —  tr-  8?  e; 0  cs  0  O  lA t r 0^ IV) r^i  8  0 - cr Ln tr 0° rs c-s  8 s? 0^ C'  in  88 cs  Cxi r rx r s  &  cr  Oo CS O rr cs c x tr rr> r A  ir  r-  rS  FA  r— rs  v.»  tr vi f^ (S> d fsA cs'  VA c n  CO (JO ct> rs  c^ rs  rTi  tr  rS  O  rz  rn  O ; r~ rs  r~  ^A  cr vn  rS  PA rs  m  c£  Oj rvA N  O  cv  rs. c s  rS  W\  LA. cvi CH  1  rsj rs rS rA  0'  O {£  rs 0  rS  of cr cs rA  cr  cs  o  «c \ 0 cn £ \-  00 LO  rcr oC cr Ln zr  in  Z O  00  oO CD ac  v? vJ  CO  u a er-  u  C"  r  tr  J  0  CD  u  a  O  1  o<  vj>  0.  I  i  6  r> h h  >  b r K 2  1  \* 0  tr  CS  Si _o cS a/ Q)  u_  Li.  cn cr  CX  u-  6 J> "0 QJ Q)  U- U_  H c" r l rs  cs  zr r~  0 H>n  on  j6  a;  )j- L i -  sj  0  X X  15  L0  Ol  v5 Lo rc v5 n  0  •>  y V/l  V?  el d  -> 0 D  vi> c  Co O ol  l-\ v5  In rr H  rs Lf\ C  ^')  Lu  tf Q  Al  v -  in  a a  \ -f-  r~  tr CA cr t r <s zr O vS cs O cs- Ln t r r- r~- r~ \9 Ln LO 1— vi> tr tr m tr 0 m Ln 0 co v3 v9 ( P in vS <A es rr cs- CH Ln ot CS o( N CN ol  0 - 0° 0 .  rs ei  0°  £  trr es  1  1  tr ~CL  LA  s  cr  0  Ln 0cr O rr rA zr cs.  r~ zr  v3 0  0  ^1 ^D D  X  r~ r~  cr ol In 0  6  X  V  CJ  >  r—  0  c^  =1 1 5  u b c  e n g i n e e r i n g 213  1  CY) ro Lo v3 vS v9 N tr- r A vi vi vi vS V5» vi v i vi vi vi vS v3 v i  or Cu  c c/»  l o IT > O 03 cr  0  N  U  cr CN 0"  tjfV-  > r-  rv  v>\A Cr CM rr 8  rs 0 O VA zr  r-  i  v3 r- c\n Vn r- rt r In O tr tr v. h r- Ln v5 t— v5 v3 r- LA en LO  CX)  O  SO  vS  cr N V) t r O 0 0  cr  tr tr O 1A  er  O \n V)  tr  vS vS cr  cr cr cr cr r- r cr cr CT LA iA O  tr  rS  0  rrv9 0O- 0 r- tr  60 cr cr CT  rA tr vS VA 0° tr tr r~ cT O O cv) tr ro fA > rs Lo ty r~ r- r- C)0 cA OA tr CA cr $ tr cr i A i A LA O r r VS vs I A in io LO t r vS> vs vS  r  S iA in  r r  CO.  r- rr  r  iA  £  to CO  LO  r-  _J  O If)  -  >  Q  cr  >  1 v9 T  V  O  1  N  1 1 j  \J  z  i 1  m  at I  Z  I-  O V\ t r vS 0 i-1 Lr~A i LA c r vi c r en rs vS rs rs sr'  z Ui  ar  \D  Z  O  cS  CY  z  Z  -  i  O  LA  rs  & rs  ro  1  0  O rs  t r vS cr (V) H rS r- r- vS rz Oo rr cs cS —  CA c~  rS  d  LA  LA  \A  rv  rS cr •r VA O rs O Lr csl rcs vn L A t f rr 2* rr Oo vS r r~ 6 C S ^3 cs rS  t r 00 00  r  CA  ev-  CN  rS  1-  o « \f 0 CT)  r-  E  u>  O O  Z  CO  o CD  a: Q  cr  o  0  u  CD  o  CM  m  CS  £T  CL.  £  cs v9 <rs vS Cr • \ A CN W = cr cs O  c*r vi! V  &  .0 a:  ce  \  - V)  <2 o  -4-  c?  F  (•  cs  CJ  0  0-  CH  °  1. 1-  i  r  VQJ  rs a'  0'  h' tA  •3-  "3-  r  r~  er  O fS  1— C^A t r r- t r cr O tf v 3 r- v9  v9 /-Nl  a t  n r  rS cr  r  0  a  i  O  (VI  O  m  rxi  (Nl  c r~ TS  61 CJ!  CJ1  5 -5  cr <Z.  r^  rA tr O er O rs r- t<>CP r- 0  CA  cs.  -t— Ci  L/v 0) CA  V  4)  LA  0 "0  Lf)  0 rr  CM  rv  OO L«  3  -r  0 O O 0  8  ! 1  i  3  rs O  2: r  <s. r  V)  r- O O  r  r-  0  •5 ) 5 > - , ! 5 vi • •aC3 _ C 0  0  cr  rr  1  1.  1 1  1  r  1  e  I  J  15. APPENDIX 2  Listing of R B C s  214  Operational  History  Appendix 2 RBC Operational  Date  Inf. Q  Reset Q  Ave. Q  mL/min  mL/min  mL/min  #1  T. "C  #4  T. 'C  History  COD Ldg. g/m  2  BOD Ldg.  Observations/Comments  2 g/m z  •  Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Nov. Nov. Nov. Nov. Nov. Nov. Dec. Dec. Dec. Dec. Dec. Dec.  3 /83 5 7 12 14 17 19 21 28 4 10 14 18 25 30 2 6 9 13 16 20  Dec. 23 Dec. 30 Jan. 3 /84  52 88 56 33 0 52 0 0 0 0 0 320  90 150 50 150 160 150 140 0 0 0 200 320  0 0 0 0 345 0 310 125  510 0 405 520 345 370 310 125  0  0  89.0 103.0 41.5 75.0 106.0 75.0 70.0 0.0 0.0 0.0 260.0  14.0 13.5 13.0 13.0 12.5 1 1.8 12.5 1 1.0  14.5 13.8 13.6 13.2 12.5 1 1.8 12.5 1 1.5  9.0 9.0 10.0 10.0 RBC FLOODED  2.346 4.002 2.352 1.460 0.000 1.903 0.000 0.000 0.000 0.000 0.000 3.850 0.000 0.000 0.000 0.000 13.559 0.000 13.820 5.085  255.0 202.0 432.5 172.5 340.5 2 17.5 62.5  0.000 RBC PARTLY FLOODED RBC FLOODED  Foaming 1 Observered 2 * pumphead added tor influent tubing split, cont. with one head st  nt  pump tubing split pump tubing split pump tubing jammed in pump pump tubing jammed, no fuse restarted with one pumphead GRI bellows pump installed pumps knocked out, disc motor OK restarted disc, covered in mud and oi poppet valve broke, no replacement rain guage frozen, snowing lightly loss of suction, inst. 3/8in inlet line feed line collapsed, growth reappearing poppet valve broke, discharge side line partly collapsed, rebuilt inst. inlet almost frozen solid  Masterflex  flooded Jan. 1, both pumps and disc stopped  Jan. 20  0  475  0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 30.396 0.000  Jan. Jan. Jan. Feb. Feb. Feb. Feb. Feb. Feb. Mar. Mar. Mar. Mar.  24 27 31 3 10 17 21 24 28 2 6 7 9  0 0 0 0 0 0 0 0 610 0  0 560 0 0 300 465 275 640 610 500  0  0  Mar. Mar. Mar. Mar. Apr. Apr.  16 20 23 27 4 6  0 565 565  600 610 550  0 0  505 0  202.0  Apr. Apr. Apr. Apr. Apr. Apr. May May May May May May May  10 13 17 20 24 27 1 4 8 11 15 18 25  0 470 460 450 430 425 420 385 385 250 355 340 325  470 470 460 460 430 425 420 385 385 385 380 385 400  470.0 465.0 455.0 445.0 427.5 422.5 402.5 385.0 317.5 370.0 360.0 355.0  May  28  380  415  390.0  150.0 232.5 137.5 625.0 305.0  9.0 9.3 RBC VANDALIZED  0.000 MOTOR BURNT OUT 0.000 18.476 22.306  582.5 587.5 SPROCKET FELL OFF 11.2 11.2 13.2 13.2  14.0  14.0  14.1 14.2  14.4 14.5  14.0 14.0  14.0 14.0  0.000 0.000 0.000 14.678 17.209 18.414 14.487 12.457 1 1.945 13.987 16.078 1 1.453 8.786 10.506 7.527 10.123  surface leachate black, foam in pumpwell inf. pump lost prime poppet valve broke, growth reappearing poppet valve broke, no replacement installed new valves and valve springs pump lost prime pump lost prime, inf. checkvalve inst. checkvalve fouled growth coming along check valve and inlet line plugged drive chain knocked off restarted unbalanced disc drive motor pulled off mounts and jammed new motor and elec. breaker installed rapid regrowth cleaned inlet screen and checkvalve disc left stopped bellows nutrient pump inst. Mar. 30 feed pump stopped, removed for servicing rapid regrowth  cleaned inlet screen feed line to bucket partly plugged new  line to bucket sump installed  inlet screen plugged, very high SS very heavy suspended solids solids washed out, thinner remains ammonia addition slopped  growth  10.032  Jun. 1  400  420  407.5  Jun. Jun. Jun. Jun. Jun. Jun. Jun. Jun.  440 0 360 340 410 590 570 450  405 410 400 410 600 610 600 595  430.0 202.5 385.0 370.0 410.0 595.0 590.0 525.0  15.0 16.0 19.0 21.0 19.5  15.0 16.0 19.0 21.0 19.5  17.0 15.0  17.5 15.0  3 6 10 13 17 20 24 26 31  460 610 660 620 650 830 820 10 0  600 610 650 650 990 1030 990 1070 1050  527.5 605.0 635.0 635.0 650.0 910.0 925.0 500.0 535.0  20.0 16.0 17.0 18.0 21.0 17.0 19.0  20.5 16.0 17.0 18.0 21.0 17.0 20.0  25.0  25.0  6.969 8.272 8.4 15 7.366 7.352 8.914 6.470 0.068 0.000  Aug. 3  0  940  525.0  21.5  21.5  0.000  Aug. 8  520  925  730.0  20.0  21.0  2.964  Jul. Jul. Jul. Jul. Jul. Jul. Jul. Jul. Jul.  Aug. Aug. Aug. Aug. Aug. Aug. Aug. Sept. Sept. Sept.  5 8 12 15 19 22 26 29  10 14 17 21 24 28 31 4 7 1 1  7.894 0.000 6.793 6.273 6.777 8.142 9.200 4.293  6.732 SS have filamentous floes, normal colouration returns 4.739 SS settle poorly, inlet line cleaned 0.000 checkvalve plugged, inlet lines cleaned 4.039 fluffy filamentous floe remains 3.366 settlability improving 3.641 settlability vastly improved 5.222 settlability good 7.097 solids increased but settle well 2.862 settlability good, inlet screen very plugged 4.375 inlet screen and valves cleaned 6.277 cleaned inlet screen 5.287 4.873 inlet screen cleaned 4.739 inlet screen lost down pumpwell 4.880 inlet valves cleaned 3.961 inlet valves cleaned 0.059 inlet valves cleaned 0.000 cleaned checkvalve, discs lost solids SS 0.000 checkvalve plugged, installed new inlet screen 2.699 inlet screen very plugged, SS mostly washed out, 1 stage lost most interior growth 2.480 2.318 cleaned inlet screen, slow regrowth 1.228 inlet screen cleaned 3.371 inlet screen cleaned 1.747 inlet screen cleaned 0.517 inlet screen cleaned 1.086 inlet screen cleaned inlet screen cleaned 0.000 replaced valves 0.000 inlet screen plugged, pump bearings st  870 920 930 1060 260 2 10 385 1520 0 0  930 920 1220 1 120 1500 1430 1480 1520 1 150 860  897.5 925.0 925.0 1 140.0 690.0 855.0 907.5 1500 760.0 575.0  18.0 18.0 17.0 16.5 18.0 17.0 17.0 14.0 17.0 15.5  19.0 19.0 19.0 18.0 19.0 18.0 18.0 15.0 17.0 15.5  4.541 4.306 4.938 4.706 1.552 0.939 1.444 5.472 0.000 0.000  Sept.  14  Sept. 15 Sepl. 18 Sepi. 2 1  0  0  430.0  17.0  17.0  0.000  0 0 0  500 850 690  250.0 425.0  20.0 17.0  20.0 17.5  0.000 0.000 0.000  16.0 13.5 17.0 14.5 15.0  17.0 14.0 18.0 15.5 16.0  0.000 1.867 0.000 1.200 1.290  12.0 12.0  12.0 13.0  0.000 6.825  10.5 10.0  11.0 1 1.0  5.299 4.034  10.0 5.0 8.0 10.0 10.0 9.0 9.5 9.0  10.0 5.0 8.0 10.0 10.0 9.0 9.5 9.0  3.993 0.000 13.607 5.856 6.029 2.853 0.000  6.0 8.0 6.5 5.0 6.0 6.0  7.0 8.0 7.0 5.0 6.0 6.0  7.998 5.623 13.110 7.156 0.000 0.000  0 610 0 465 230 0 0 1250 0 1280 1 130 0 1 100  345.0 835.0 375.0 682.5 427.5. 0.0 640.0 1280.0 0.0 1290.0 1205.0 0.0 1 150.0 PUMP BROKE 0.0 1320 1240.0 1200 1005.0 1 1 10 840 940.0 612.5 460 0 230.0 0.0 980 976.5 500 980 605.0 965.0 980 852.5 725 362.5 960 950 480.0 1060 750 900 625 0 1280 1310 0 1300 1280 0 1200 1020  Sept. Sepl. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. Nov. Nov. Nov. Nov. Nov. Nov. Nov. Nov. Nov. Dec. Dec. Dec. Dec.  25 28 2 5 9 10 12 16 17 19 23 25 26 30 2 6 9 13 16 20 23 27 30 4 7 1 t 14  Dec.  18  0  960  475.0  1.5  1.5  0.000  Dec. 21  805  920  882.5  6.0  6.0  32.651  0 1 160 810 770 385 0 0 973 7 10 950 725 0 0  0.000 pump would not start, removed for servicing 0.000 reinstalled pump 0.000 pump stopped, restarted 0.000 pump lost prime, crud in the poppet valve 0.000 changed inlet screens and pump cam 0.641 inlet screen very plugged 0.000 inlet screen fully plugged 0.879 inlet screen changed 0.945 inlet screen changed, fully plugged pump restarted 0.000 pump prime lost, gummed up valves 4.388 assistant couldn't restart pump pump restarted 2.342 growth reappearing on 2 stage 1.254 pump wouldn't restart pump restarted 1.551 changed inlet screen pump removed for repairs 0.000 pump reinstalled 8.387 2.819 inlet bucket sump not staying full 4.343 heavy foaming in 1 stage 1.617 growth getting thicker on 2 stage 0.000 pump stopped new twin bellows pump installed 4.991 3.238 «8.664 *4.676 pumpwell flooded, couldn't clean screen 0.000 inlet screen left high+dry after flood 0.000 inlet bucket had again been flooded out 0.000 bucket tipped again by high water levels '22.049 lots of foam, white growth on 1 +2 stages n d  st  n d  st  n d  ^ CO  Dec  28  350  875  635.0  7.077  Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Feb.  4 /85 8 1 1 16 18 22 25 28 1  0 570 250 0 1 15 1000 670 1093 750  820 770 250 890 1 145 1 120 1 170 1 160 1 180  437.5 695.0 510.0 125.0 502.5 1072.5 895.0 1 13 1.5 955.0  4.5 7.0 6.0 6.5 8.0 6.0 7.0 6.5 6.0  5.0 7.5 6.0 7.5 8.5 6.0 7.0 7.0 7.0  0.000 0.000 1.868 0.000 0.731 13.020 4.844 6.558 3.510  Feb. Feb. Feb. Feb. Feb. Feb. Feb. Feb. Mar. Mar. Mar. Mar. Mar. Mar. Mar. Mar. Mar. Mar. Apr. Apr. Apr. Apr. Apr. Apr.  4 8 12 15 19 22 23 26 2 5 8 12 15 16 19 22 26 29 2 5 9 12 19 26  1 100 635 1 145 775 545 225 0 1 1 70 1 170 1 160 1 105 1000 630 0 1200 1 160 1 130 840 700 1000 11 10 1 130 0 50  1 120 1 145 1 175 1015 580 0 1200 1 180 1 180 1 185 1 125 1 100 0 1 160 1220 1 180 1 180 1010 1 160 1 160 1 130 1 130 1260 1230  1 140.0 877.5 1 145.0 975.0 780.0 402.5 0.0 1 185.0 1 185.0 1 175.0 1 163.5 1062.5 865.0 0.0 1 180.0 1 190.0 1 155.0 985.0 855.0 1080.0 1 150.0 1 150.0 565.0 655.0  5.0 6.0 4.0 7.0 6.0 8.5  5.0 6.0 4.0 8.0 7.0 9.0  5.016 2.858 6.939 3.604 3.826 1.465  6.0 7.5 7.0 8.0 6.5 9.0  6.5 7.5 7.0 8.5 6.5 10.0  9.372 6.529 10.544 8.420 6.510 2.986  8.0 7.5 8.0 8.0 9.5 10.0 1 1.0 10.0 1 1.0  9.0 8.0 8.0 8.5 10.0 1 1.0 12.5  5.940 5.429 4.170 4.234 3.150 3.960 5.062 2.712 0.000 0.159  • BOD estimated  1 i.o 1 1.0  from COD values  •4.725 heavy growth in 1 stage, growth spreading, high SS 0.000 0.000 1.260 0.000 pump throughly cleaned 0.393 installed full stroke on front pump 7.530 2.111 heavy, healthy growth on 1 stage 3.246 0.743 effluent line plugged, almost flooded out RBC 1.617 good growth on all stages 0.991 inlet line silted up, cleaned sl  st  1.372 overhauled pump 2.322 1.330 inlet line plugged cleaned inlet line thoroughly •6.037 •4.107  pump and inlet line gummed up cleaned and rebuilt pump and inlet good growth  on 1 + 2 s l  n d  stages  Ii  16. APPENDIX 3  Raw Data of RBC Process Sample Analyses  220  u b c  e n g i n e e r i n g 221  o_  r-  m rr r i— r- r-  O  c"  3  o  <n $  co  v^  (><  tr r  r  r- r-  r- Cr-'. rr-  0  Ln  lo r  O  N  T  t)  cr- Ln cx> rr r/^J rvl  cn  i  I;  cr  (•'•  O  rs  Ln 0 °  r  r — r  r-  m  Ln  H  r~ r  oi cs-  01 rs  l  LO  vS cr  r-  Tl' C! 1.,. (T o c Ln r? n'i v>  tS  (VI  h  H  o Ln vi>  rs rs  VJ J  (-•  Mr!  C 0' r  LA  rr! v_P s9 rvi r^ r — r  l A i ^  re  r  r"  O  1 !  i-i  1  rl o  1  H  v5 X r  .  r  c/>  Q r~  X  — -  H  |  r~  rJ  I i  !\  iO  ;rs  Ln  i VP  r-  rS  _J  0  rr  rs  tr O" rs  •  rs  i  rs  r- r  rs  <-l rvi <v Lo  0 ° rs LO Ln  1  rr  rJ  Ln \  cn  T  tO  cn  s5  i — a~  \  rr Oo  o|rJ  m Ln'  oo'rr  rs; c l  rs r>l rr  -  rs! ( l v  i  LO i  |  If)  O  i .  z  z> i  1  c  rr Z  £  Z LL!  VD  i  J-  r-z.  £  6*  i  O  i  i —i—  o o r-  rs r>''  vS  Lei rs rs  ...er o  cr  m m r~ cr Ln rS (V) r o rs Lo rs rr  Ln  Lo  rr cr  CP  rs Lo  CO  r-  vi  Lo  croc"  r  rs cn O X  v.?  vJ  rS  r<"> i~  Ln  rr  v9 c r vS  cr  cr  0  q e  VC  CS 6  rr  i  z  c£ r-  VJ?  V.0  m  CY)  VB  fA  r r Al  i i —  LA r- rv  ir)o  [V)  1 !  u>  t—  Z O  v5> ry>  O  o o  Ln  u  v.9  rr  rs cr  v.?  r4  cr-  f\  CO CO or cr  u  Q o  cr  vS rs r0 Ln vS CT  u  rS  tr  OJ  £  i. ?  I  cl h  r<  I  55  —f-  Q  V& _  al if  r>  0;  tS ~ZCU. 6 s. W -r  <j  u La  1  .W  oJ.  Lo rv") O  rr  . ) 5 ' J) y i r "i i  r> h  r  rS  QC  It*  VJ  1-  cu. _£  r^  vn rsl  r~  O rr cr r~ O rs ^=  rS r* rs nt rs 5s  10 cr  0  UJ  O •ar-  O rs <r  o  5 ) 3, 5) 3  "5  cr  i n vS  Ln Lo  o rr  Q.  r- Oo  errs  O  eS  rr  5i 3 5  Ln Ln r  ' (5  r>  rV  5  v>  VJ ex.  o>  rs  5  V  U  0  5 <£. s2  d.  % a~ 6 iii -V UJ  (A  r  6' 6 ><5 . r:  6  rfa -  r  v9 LA  O  rJ cr; <r  r"  o»  <f  r rs  r>> '  a)  cf  o u.  -V  V  O  c  a-  L V,.'  7^  ii-'  C d  -r  Or. v9  6°  v5  cr r-  10 \S  r-  VJ> VJP -r  v9  rr  N  Lo LA  V? c rs — —r <J>, r ej ' c —«' 3 "5 "5 ~5 —1  rS rA 1  rs  r  »' a/ Vr V/i  Vi <y 0 V') VA  cT  6  6  -2  <z <zVA o>  zx- 6cS -V  &  VA  <j  w 1"  o~  cr a>c5 LU -r i''  z.  a?  VT  I  a*  cX3 -  6 JJu>  LA  -r cy a?  cl  u b c  e n g i n e e r i n g 222  0  tr Cu  v9i tritHrHoo  v i v i ! vi!  r"  !""  c<q C v ' l  r O  0  r-i f V  * i  rr LA cx  j rr  rl H rv rsi oJ  CM  fN  (A  rr-  I  LA  rv9  v9  Ol  rS  Lo LO  H  f  vv  to CO LO D >  0  rA r 1  r~  o  LT) LO  r  CS OJ  H  ro O  rv  io i-  :  ro  CO ro v9 o4 oo q ol ol rr rr v9 rr VS v5 rr Lo q q VA q 0°; c VA v i cr t> vr> 00 6 ol CO rr do vi rr cr LA oJ r~ r r~ r I r- LV r~ r~ r~ i — r |- o= O r 0 ° 0" £ .*'• j 6 °  7t~  O* r-  ZZ  LI  v9  o  q  X rr r Lo" r- r v£) O LU  V.  cr  cr  rr rN  CA rr CO  rv  0° Ol ol  CA  0c  6 Cr  LA  Ln  o rr oi v9 rr 0o rr VO d ol rr ryi t> cr rr rr ol Vn vo  c<  Z r~  co o! N q q ro CP o r- rn co rr- oi q q LO LA rv5 o Ln" rLA rr rr L O rr C^ r v i .•Yl rr 6 ~ k v9 rr! C A ol o! Ol cs ro Ln LTl b° V ? LO vS  O O  O r- LO  i  .y rCJ)  E  O  r~  In Cw N  o  V-  c>o (Vc rr ro ol ol ro  ol  (VI  r—rr ni ol  OA  UI  oJ ro rv|  VA LA  —  CP v5  ?2  r  r-; L-)  LO" oi  r~ LA rr  rV) CV) o— ol of r>i LT)  i  — i  oi rr rs cr rs  -  «i  rvi  °. cr co v9  o  IS  CO.Z  \  rs  6 ro 6  i  Z I<n oJ r  0  Lo  -  -  Z  Z  >jlr~  rs  lm orr Ln rfO  q rvi  LO  rl cs  \ Lo  ro  ? i  rO r~ v9  (VI  q O O in vs rr Lo \ in" Ln o. co 6 vi> c-i vi  r-  _i  -—'  Ln  ol 0" r r  CO cr LA T  /o Z  v i v9 vS  O ro J  tn T L0  rO rA cr  r- cr  v i V5 vsj v i  cv  o- rA  LO h  i r-  v9 ro Ln!LA oil cr v9 (Vi 0- ^ ! r-l rJ ol tvij A1  c"  vS r~ v9 v i  IO  v i ' r-  vS  cs  o  :  Lo  z o O K) VA r rvi  ol  CO CO  cn cr  0  u  Q  <j  rA -rvi o'' 0  o>  Ol  (V'i  -5 " €  "Q.  i  a'  £  v/> VO \A  d  0>  a  0  £  3  cv  LA  6  -r  -t-  Q  rv> o| rr VA I A O ol o  q r- rr  o» O  (S  o-  r~  iS  r r r-  ro rs  vf  1  L'»  e  1-0A J .  f  a* u. 1U  LA  cv -*~-  A'  . Si  T3 0 0  n lo  8  VA  r r  -/**  -ti -<r  0  VA »;  ft' 0' Cf 0 0 ! O' O 1) c C 6 0 0 r O -r - f -v LA V", CJv. LA "? e  rr  •L  "£  rr H  0"  Ol  <s G s ^ _  rr  oJ 'l  -1-  ij  KJ  r r —V  3  0 0 0 o  o  1 a< V f- eS c cc 6 C —V - r r I" 1.  t  cy 0  r~  <r cr  ol  ol CV) cs e£ ot rs c° LA  o (VI  O cr  Qi  \  —-  rr 0 o ol q ol O  rr 0 (V) r— v9 cr  Ln VA rA  ol  rr rr  oJ  rr cy cr rA  •V -f D u ~T> _c O cy u O o Oo O O 0O - f  u IV  i>  c  8 C5'csc  To  n rr  2 I"  6 •b i. o» ro  ?•  O  rr rr  rr  7.  / s ample  Ucvre  Dec. \i Cec. IC  V* ii  I  91  -  If  V  j*  l»  ?i  •*  ii ••  1*"  »  I'  U  "I"  M  5  CARBON COD BODs  l»  tl '»  l«  VI 11  tl l\ II  ic  Dec Dec. 2 0 bee 20 D«. 23 Dec. 2 3  \12-H \\Uo  120. C 102.1  230  ns V' n mi  »W M  ?2 TM 120/I" 501 / "10  N ITR06 E N  kN  T 32.7  2G.3  !H.l 17 T  11.3  n.7  26.S 2S.9 MO. HO.  IrJO Gl  C o no!.  pH  . 0.56 . 0 . 2</ O. 3 2  o.n  o.n 0. 3 2 O.ST O. ISC ' / 1  <o.  52 n.s 23.0  <o.i  30.4 /31.02^.3/22.15 2JL2_ to  20.5/20.5  fa/.o  S'5 101 1H  S3-I II..5 IO?.! 117.0 in.o 101.«/  1 103  52.0 55.2  ! 13  n.sli o/ 31.8 35.? 21.0 /25.0  n  O s A O TS  ST.? /HI  >03  20sIG5  N  SOLIDS TSS TVSS  22- / •2 . ?  ?Xo  H<oO  ! NO3-  720  /zn  Feb. 21 27? FeW. 21 270/201 230/1?0 112 few. 2/y FeV>. 25 T.6 F-'T. 2^: M a / . 2. Hov. 1 IS? Ma*'. 23. 313 / 20I He/. 23 212. fipv'.l C 230 250 2C0 10? KJfy.'i 13 113 1  TOC  q 5Tl  FeW. n  1  O9/-O  1  0.07.  O.02 r  0  : 20 1  'IT?  196  \H5  2T! 2^./ 3t9.3 36.6  ro  0.03 0.02  u b c  e n g i n e e r i n g 224 I A VA  PA  CL. r- r-t  TJ «/> ' C  U°  ...  o- Oo vs  0  Oo  .....  —  r—  ....  lA 0  r; r^  r  r~ D~  vS -  (~  ...  —  • ....  -  -  -  ••- —  CS O  ...  CO  Q  >  -  h-  ?  Z  rs) H i O O  A  o £ O  z  0  ....  —  Ln  -O  0 o  • -6 O  9  -  ...  m .. O  -••  0 O  6 6  2  z  LU zr  Z  ol  hz Z V r-  o o CT  —  vj" _r> Vv VA 0<=> rJ S 9 C» O r- ts- cr (VI t r xr t r t r t r xr c s c s o CS° tj» o S S r- c r o xr xf LO r- i— oo tr tr ba x  S  o C»  ro r— vn xr  Ui  vP  O  o=>  r—  6  O  b°  fA  •  -  0  •  —  C-"> v9 i—  ro  CA  rri  r  O •xr  xr  xr cn  0  CA cs  c s rs  •5  rs t r  rA c<  c r In  *  r^ r—  X r—  5  o- r Oc 6°  O r~  rs rS  1  r— rA  r-'  X  vS  r^ t f  V.  m  =  ry\  1  r  cs  LQ Ln  i ir *  ...  rs  a~ in v5 o H LA lA <r. E; OA <rr  15  r  CS  r-t  f © -r  sr'  'iTL8<  1  X  s  vi» cr L0 0 CV) O 0° rs rs H X _1.  fSJ  > 5  i  c*  H  v  r  5  X 2Z  tr  x  XX  •£ tr  tx  -o T _at  c c5 xc xr 6 6 a;  X  zr  t_  i •' X 7  \?  (V3  cs:  t r AA r~ cr rJ  Ln 0° LA  1—  rs  r  V.'  th 1  •i T  X  m  2:  r  in (/>  ?1 t r  0 0  H  LA LA LA Lr> t r Cr cr. cr H ri ri rs H r s r s rtr* *-v' 1 r 1 5p ^ » 5• 5i ? ) S )5 JT> a 6 C 6 _c t> 6  i5  3  Lo  Ld CA nr.  60° crcr  L0  rA  rA  «J  t r T Cy 0°  > 5  -  T>  —»-  <r vj> cs" r~ vS r^ * xn SS c< t r & rJ  t r LA r r-  H vj> vS LA  f>  LL  r  T  La La  -o -•  •3-  r  £ '£ tfa: to «' \ d  O  c5  r— rs rs  5xr  bo •xr  vn LA vS VA XT t r s9 v S  OO v5 Lfl r LO oA t r r»"> r r Ln  r  •tf  V\  00  q  0  Z o CO o co err a cs cO tr 0 v3 u u  *>  OA r-  cs  £2  r"tr c^  r-  c r 1  X  vi  10  CL  r-  0A  m  LA Ln  r-  £  Ln r-  tr  c£  O  6  cs  rs Lr.  rs  -  tr  rs o  t r xr  vs in  tr rA  o Vv  xr rv\ t r ^~  <»  ^S vS  i  LA fA  0 sS rr tr 6 rS. \s  0  <9  c r rS co  O  V\  i  \D O  r-  vS  0 O  LA SS  cr  -.  L/> L/>  tr t r  r- >« tr  ol tr  vi  _J  0  0  ...  —  -  OA rLn cX  Ln In  ...  £ CO CO  in r-  VA r < 03  C>  tr  /^s  i/>  0'  c 0  1  1 ^_  u b c  e n g i n e e r i n g 225  LA  zr  rr- U  -  cu r-  VA  cs CS ....  «  C>5 Oo  r- r-"  iA  -  -5  u  V)  in  cr  rT 60'  cr  bo  Oo  60*  LA  (V>  -  <y  r-t DQ  0°  -  'c  tfs >  ....  r-  ^< Sv  CX  £  CO CO  >  co  a  _J 0 v/> vV> rZ  rO t r r-j Lr>  O  O  O rv>  tr  tr -<-i VA CN  f— vS  XT rS ~l  rS . . . S9 OA — '  -  CVi  Z  O rr  rjr  OA  O  LrV  s  O  rs  O  0  <S  r>J  °i r S  Tcs  O  'CN  SJ>  xr  .... :  "fA  vn  OQ  O  0°  5  rr  s  CN  O rr  OS  |  O  <y>  \  OA Ln r- VS>  r— CN  rliv  r-  rr~  0  rS  s~ s5  VA L A  LA CS  O r<  CA  r-  O r r~  b  tn XT •O»  JZ  Z 1t» Ui x t v9 VD Z O O <5 ad {-  i*  Z  V  r  6  0  O r  O rr Ln  CO CO  LO  r [NO cs O oA a  o~CV) t r Ln XI  v_!> In r~ cS. cS  :  FX  H  •i i i  ! xr  rr r 1  cr  rsl c S r—  O  vi  s 0  • -  LA  OA  r  S9 b »  ~r  1  7>  >°  -XL  •J  r  1  cr in  -  v^ L-i cS  H  rr  OA  cr  cvi  bo  0—  rs rS  yo cS  ti-i  6  r;  fs  rr,  H  0  LA  cs"  r-  O 0 r- cr LO cr  tr sS  "I  1  vti r—  cr~cn  s? S 3  xr VA  -p r> xr  "6  Q- _ y x r  -  0)  y  x s  cV 6 VA  ft"  A rJ .  6  v/i  T> 3*  1  £x^ rr  rr in  rr  rr  0>  V^  V  iy  s.  0-1  rs  O  vS v 5 v5 IT)  r4  rJ  t  SJ?  1— c r  lS s^  tr r~ I n O r  r VJ9  -  cr  i£ 0  vJ  Vi  xr  g c~ cr~  0-)  -0  Q'  % 6  XT' h  a) vy> V  <y  cf C5  *1 r r 5>  vJ9  r-' ci  v5 Lo r S  v9 r vs  cr  rS  s rcs o l s^ O (V.  Ln  r cn r rr  cr r  rs  v5  CS  er-  cs  CS  rs  CV  XB, a?...  vn  Oo 0 O 0  0  rS  r — oa s9 r -  cr  'xO  0/  T>  tf  V5  rl  t>  1— 0  <5  v 5 r—  00 O  Xr-  0  Lo r r tr  vi  O  in  (V\ v5  rs cs rS  Lri v i  vS Ol  r-  rr cN  r~  r~ 0  v/»2~ *xr rr  7>  r r rr  8 tOOr  rS cr c-i  cn r-  (U  0  0  vS  6  O  0 0  c s oj q tr s 9 O  O rr  CS vS  cr  <s  on  r >  rA rs  IA VI  cr  S3  vS VA v i rn  —  rr 0 o° 0  v3  i-s" rr OA  S9  O  r—  —D  •O •rA  LA a)  0/  n  r-  d •6  0  OA CO.*  s9 s9  rr  LLT  <u  r— LT>  SS  VT) rv  0-  cr  rr  cc  "tf  O"  rr  q  tr tr  u>  Z O O O  \  O Of  sS  O  E tf  vS 0  LO  -2  cn O E f-  0  F-  oi OO  cr 0 0 0  .CD  Vr,  OA,  of  r-  vn x r . r— O cri O  •30  rs  vn  6  O  I — fA  (A O  r  s3  OA L A  rr  C-N  r  OA  OA CV1 Cv-1  0  tr  Lfi s i  r~  LA  r  r  rs  tr  r-  <"S Ln^v£  r-  «v5 Ln rr t r rr  La Lr> rj- tr  O) \> rz  OA ry>  O  rr rr cr CA  v5 O  tr tr  0 r csri rs r S  v>°  s9  1 0 OA  O  \ rsi rr sS LT> rS CP O  .V9 CS CS  s9  CS H  0  rs  L A VA  O rr rr tr rr  1 0  rr  VA O OA  ....  1/5 s v9  W.of. CN  to CN O  £  rr  O- 0 ° Tr rr Oft  0  <N  OA CO  -  r r tr  Oo  1  1—  - C » V.0  rr  1^  cv  ^' 5" xr 11 tr  >  ~r> »•  b  a*J  a'  xz  XJ? 6) <D V VA  1  -s  l  x z  »'  -t> t>  h  »  5.  CV  to  n  a.  3? cu  V?  L* tr  IA  * xrH  e  rJ -r"  XT  L  * tr  i-  xr  +- • r ^  rr  u b c  e n g i n e e r i n g 226 SA  c4  c/V  Oo  0  tr  rr  0> CA  vS  CO  CA  tA  cr  is  Q  rA  _i  V9  tr  cs  tr  cr (VI  (VI  r-  tr.  o  cvi  Cr  vs Ln cH  O  O r o  r s ;  s  v9  o.  sS  e r r  <s "cvi VQ  "3-  -LC\  V5  rs  cvi  (-1  CS  I  00 0/  Ln  Z Ui  r? cs  lV>  cr  O at  v9 tr  00  cr  cr  rvi  V?  \ CT)  tr  t r rr Vi^ tr  KJ  PA  CH  cr u  C>°  tr  cs;  rA in  O  £  (A  a/  H  he 3:  a)  VA  V  0)  iA  in  tr cN  LA  rvi  Cj Vi  iO in  Ln  1  h  rA  h  1  rs.  J  V.A  UA  tri  tr  5  S  3  1  Si  rz 0  "OJ LA  0 —  OA  IS  5>  95  C/Ci  O  C  -+- ,3  tf  v£>  r-  c?  Lo  3)  - iy)  IA  CA  CyO  VTJS-/  CP c r LO  ft h  C>A  0-1  cs  rA iA tr  tr  o"  v 9 1A  lA  Q.  q  errs  cr vs  O v5>  E  Z O CD  o  rr  LO  cr cs  LD  LA  a/  £ 6  far VA  ~6  h CD  LA  vA l/  •X  i  LA  1  ^  ft;  ^  u b c  e n g i n e e r i n g  227 tc o_  Vi cr c r  --  O  r~ r—  ~s ^7 • ct o o U  \  to > Q r_J  0  0)  cr  rr tr ror .o cs <V>  pl  to  o 0°  v5 O Cr o tr -  CC  £  o-  r — CS3 cs  § rri/> CS r?  vS>  k  vS>  o<  CS  v2  r•3Vr>  CT rJ  6o  So  0°  r cy  rr  t r  u\  V)  I  «» Ui rr  t r  Tr  "Tr-  cS  VA  -  vn  rr cy  CA  O  0°  r  Ln r"  O  CA  r\A  cs  rcr  VA  Ln 6  ...  Cr oC  CN  rA rS  cs.  v9 v5  O.  OP  rA  cn r A r~ r~  d  Ln CA  ccsr cr rr Vn  vs  rr  CO  i—  o  (VI  o -  |— L n ' V?  r- O  L0  co"  r~ r- c s  rs  CA  o  P-  rA  o  ol  tr  sS  r—  VA  rr;  Of  rvi  oJ  r  vS v5 (SA  ro  LO  oJ q t~ cr-  <s cs  J—  cr ro  r  6  O  rs  &  rvi  OS  rr cr  c^ Ln cvi r  rS  rvi  Cr  0 rc i—  CP  o  CN  rr cr l A  Vn* CA  r-  o  r~  0  cs  cs  bO  \  cS  cs-  O  r" (T 6  vii d  AA  r S  e<  o- rO O o  v5  oo r— r A  O  CA  rr cr  rr cr rvS  r~ rr  r  O  o l o l  r  r r  V9  o.  tr ro  vS  O  1A cS  V5  rr CV) CO  oC rr cr C A cS  O o°  v^  -  •or" CJ^ o rr rj  OS.  O v5 cr C S  O d  cs tr C O rr  r-  o  cr  v9  tr  r—  r-  o<  r r  <Q 6  -  tr  vS  ra-iV/v  o4  O f>  VA  cr  cr" in O o  vO Z O  to  o) VO CS  rw  CS  VS3  cS r(V) tr  -  cs r- o  o O  rS  <o vh ri r r rr  VI  z  v9  tr O v9 r r cvi  O vS  o  o V3 rrr tr bo trr T r  o  tr tr  r- rr rr  1  (0 O  Z  cr  r~ (X)  vi  rr  i  Z  o-  vS  VA  s  CN  i—  Z  z  ...  —  •  /—  O o°  CV)  *  rr r o 1  o o £  k in  Z £>  cs  cs  rr  o O O O CD co  ?"-}  o zr fA r•-.9 u 0 cr r.' O u cr  £  \ V  -+tf O  CL CS CT> A  CN  T> te I  vo  r r CO  O  r^-  rrr  00 r—-  cH  cr D  L')  '3  O  rS  rA r  •  L'l  cs  s  LA  r-'i  rr fA  (V)  R  CA  cr v.. <-~s i— o O LO CO cs v./ o^, er- a; i— rA cr ^>rtr rr ro rCA cr r rri— r- LO o r— o olLo O 0" r cr cr r .—'  V>  PA  i A  > N  rcss a  to  C5  ti  d  rr  75  V  ri  cr  -D  ~CJ  a c^ 1  I  rl-  r  s iC  a)  V^  f f"  VA  cf  rs  vt5  if'  rr L  *—\ ^—1  vn  CS>  *•  v>  ^_  v3  rs  cr  C-°  r~  o<  ry  S  rr v5  0°  cr v5  rvi  to  Vy  /"^ I—  3>  (V  'H  i  V . '  6  \ A  rr  rr r r  r s  rv'  1  v3  CO  t> 3t  \"  1.1 3?"  •  r~  r r  rS  o4  v9  CO  t ~  a;  d  tO  O'  o  ty>  CJ  rs  &  '  s  ^ s  vS  n  7f  cs ^ i rs  vo  i _ .  S  (/>  LO  CA  rA  $ -r  for- o  r  s  CL  4) Q.  v9 vS \ o l  Li  >-  a: CJ  c0f  ~U -J£ o)  lo / tr tr  rL  VA  £| rr.  1h_  -  u b c  e n g i n e e r i n g 228  rXT  -  -•  60  o  cS  ...  v> r-  r-  o Cr  CsJ  0o  q  Oo  r~  r-  r~  r  cs Co Ln o  o O  sS  o  r-  -  TS t/> ' C  tf  ......  s3 0r 0°  «  f  -•  cs CO  cS C S  CO cr  .....  —  O-  m  <N O  O  vS Cxi CS rr r rn  -  s9 LA Ln  CS"  r-  L A  6o bo  L A  £  Sy  Cr CO Xr 0° CO \  CO  1-  _l  -  \n  >  Q  0=  o  0° O  Z i  -  (O  O  z  s^  2.  V\  <  -  S9  0°  vS  iA  q  ....  6  ...  •  ^  J—  q  r—  o r~ o°  LA  s9 Vx  OA  <o  vi) in  |  <>  cr x f  o  o 6o  TT  O  r  <r  er-  o rr o tr  rs  o  oo Cr  o  A r>c« V oo OA cs "o CV)  0O  r-X-  0°  r- LA r- rc S ol  cn  r— LA  O o  bo"  0° r~ Oo  r r L \ CvO  to  on -  ty>  LA  r r CA Cr <0 ot OA  Cr Ln  In  c d  6  O 0  o  oi r—  xr  cr cS CS  cs  m'  s  r cr  cr cr  Lh  d d  CS  r— 6o  OA AA  O O  vS  rv\  rr vS  -  O «r-  T—  r- vs. o  <rs  Ln  of cs cs  rn OA  —  CA s9 r-t oo rQ on  in CS  q  S9  V> rs cs  CA  s9 xr q rJ Ln  rf  O  r-  on s5 i n LA  LA  Lr\ r — rfS •xr" 0° o  cS*  o  OA rr r~ LA" fri  q  U\  -•  r-  rA  X3-  SS  O  rA  <6  cr>  cs  rJ CP  OA  vi --  X3-  0o ~r r f LA r  s r-  •c^ 00  OA o xr  cr XT  rS  an TS  OA  LA CV)  bo  vS o=  -  o eO  6  q  O cn  r^v  O  xr H  QC  z-  OA r r  Oo O OJ r A LA  O Z  r_  •r-  o O 0" s i rs vS  O  OS  Z 4 Ul X Z  si  O cr O rr O rs  rS  v9  q  vS  <S T-  in :  rS  rf  O  ss  O CD — I  rr  oA  *n 0 cr xr Xn CA tn cn r— tr rOA •s  rr  cn  q  OA  rA  CN)  Ln  q d  q  rr rr  o  xr tA  a on  rA  OO  --  0  -  o  ......  •-  £ V10  in  oO CO CO or:  cr  o  u  0 u  2 S>  cs cS in XT  s9 |  rs ot  v3 LA  V9  ol  a>  i  "V .0)  XT  D \n  j?"  xr-  ri  q  s5i s5>  s9 v5>  rs  xr CP  37  o- s9 rr  oi  rr  i.  i  •y  a) • Q>  5"  L ZT T T  T- sS>  £N)  3  OA  LO  v_ • O xr  ' o "o  OA xr xr OA Ln XT xr <r-  xr  00 rs LA S ?  itrLA  cn m  m  Cy> O LA LO  5  8  (5 ©  - i  X  LA  l-"i  a)  t>  VA  s-  rA  c*  v^ rr  r-  CP  r s9  r-  ai  v>  01  xr r r  s-  ~T>  O cr  in  Ln  ol  LO in  xrr  <zz  LA  v9  c>r r—  CS  -xr  xr  0°  CYl r s  v5  S  .....  /•-—>  s9 in rr-  i— ->^/ cn 6S  r r e>s  r  v3 s 5 •  /•—s  rA  £ tf sz  "tf  .•~\  cr r*  CL e  s9 Lr>  cr  'LA xr  C  V  4)  \  rr  rr  Z CD x r X3^ rr  XZ  6  LA  xr  rL r r  3.-  C5  IA  1  if 7 £" i/i xr rr  cn  LA  CT  -tr  ~T>  a?  0'  lV  0)  LA  -i  &  £~  4--  rr  1 lA  of  T> 0'  5  tf & CA  L^  u b c  e n g i n e e r i n g 229  a : ir  I 4  •-  ...  --  r-;  r; °T r  cr r r  q  do 1  1  Vi r~  r~  fS  -  03  ' C«/> s  tf  0 \  8  cs  I V  :  vS bo  rr  cr  —  £  sv  i—  CO  rr rf x> to tn ro > >- r ~ Q r- ttjxr _J CO  0  cn v/> vo  r- r A rr 2 ro Ln  L A  r—  q  r-  o O  rr  xr  d  .ss  xr C O r~ ln r r tf  f-  Z  rA  •  fO  O  w" r- ti-  in  O  r-  CS.  •  in  _ «  2 I «o  Z  a:  Ul  VD  Z  O  r-  Z  Z  L A  J  o  b  66O o~ rn  r-  rf  rr  -5  L A  O  cn r>0  cr  V)  tA CN  -  o Q  cv> rs  CO  0°  S3  :  OA  O  -•  -  r<  c  cs cs  •  0  I>A  10  O rO Z O rs CO  o  Lfl  O co  O r rcS xf rf  r~  fA  O  CA  v i in c> r- r~ cv>  cs  r~  rs  &Q  r- t*l  PA  r -  V A  Si r S  C»  O  r J  cn .rr  • »  r^  C A  -  cs  X3-  6.  Ol  6  6  r ;  CD  rs  CP  cs  OA  OA  -  CA  Ln  s5  rr Lri rr cf Ln  IO  LA  Ln  d  r  rr rr  |SJ  i— -  «  r-  cr s S rf O s9 rr rr Lri sS> r r  cf  LA  cr CP r-  s3  cs  6  t£  O  o  q  rS  Ln  o  6  a  cS  cr\ cs  oo  O  CV) CO  OA  X3-  rr  r- &>  r-  in cr n' c s cs  6  rr  ^  9  r- <v> ry> Ln  Ln s9 O q  rs  xr  OA  1A xr rS rs  r  Cr on rS  r r xr xf  q  r -  v3 csl  O ry)  V  -  V-  £  rs"  C A  rr  r-  rA  In  Ln cS  O O  01  "o r— rf  CS  Ov  <VJ  rv\ ?  In  0 6  e  C S  ro s  v9 xr CP CP  H  00  C S  ...  _.  a  •O.  o  CA  O  Cf  0° O  ...  sS  L P  I  evj  O cs  CA  O of) o o fry cs cs  CA  o  rr r¥  vS  vs  O r~ cs v 5  O cr  ol  r -  o"  er o  C A  cS est  z  o  CA  vi  r -  b° ro S9 r—  i  o  O A  TP  Cs/ o( O C  rr- t r  In rr  *  C A  s  cr cr  N  r-  CP XV  rr  3  ol  8  o  .xr rr tyi rs rs  'in  rS  V/v  In v ;  cr CS  LO rC A  in lA  O  cs rvS  Ln  r  'N  •s,  rs  Ln O rs  73  rr  ;  8  ^s xr> L I Ln o CV) CO vi> m r r 5-c s oO vE)xr rrC S x r , rr a> CA ^y tT) v^> I i i r- ol L-v v^ cr 0 0^ t*> V? rr L7) r- r r- LA s9 CP rs r- r~CS s S \9 rLn IS) rrs <r4 Cv> rr u u cr- t— o= LO LD LA LO Ln rr Ln s ? sS s S ^xr 0Q s S SS u\ L A vS rJ <r ircr croV V cs CO cc  i n  CL  E tf  ts  CO -+tf Q  g -r  net)  J  1  LV  a  -s  Q)  V/  t—  I—  W  ft'  i •  Vy  •f Jf rr  0  C -  •V~  v.  rr  rr  7  o cv  1  ^=  0  - r L^  xx  a)  r— XT",  V'  ar  *t  1  rr 5_  j"  I—  •i  -D  o  rr  $  6  <J)  L^  J  ~v  o  -o a'  O  a>  i  O J  V  T>  x±-  0 3  |  cr  .—  »  tt^  "5j_ •£  =1  LA  rr  rr  —j  -J  t»  |  L-A  *'  ^ .  ~b  xz  ,cP  s 0  v/>  o  V  §  xl rr  V/  V  rr  •  -  u b c  e n g i n e e r i n g  — IA  O ro  Q_ cr r- r-  q  ...  iA - •  VA  CM  ...  r~  0Q  230  Cr  Mr  r  rt  >A  0 r*  O r~ (¥>  °?  ~T~sS> 0°  O rr LA rr  CP  r  1  r-  0  -  cy5  -rs  -c o U  •  b* tr- in ot CN  0»  0 vs  £  <E m  T— r-  q r-  * CO CS «A CO t/i > on rA  Q  _J 0 fl  r- errs to  c x cs  rr  rs  VP  rO  Z  K  t  rt  £ O  U I  z  r  1  <f (P  1-  —  r-  O  OA 1—  \A CjO  cs  rs  rr  ...  i«  r-  •rr •S;  cs  CA  in  cs O d O  O O O  rs  rr  CN  r^  O  rA  0Q Cr CO* O  O 6  d O  CX  cs  r- rr  rA  —3  cc  cr  Q  0  VA v9  CN lA  r s cs c4  rv  v„'  3  r10 cs?  1/1  \  CA  bo iA  I - CA  CA rvj rf r- cvi tvA U v9 rr VA 0 -  r-  ...  rr-  rjj rv»  CA sO  VS  >•  CO bo  CA  Vo c— \A  rf  L"> cr s5 ro O CA  ra-  CA' d  r>i O  T77 V^ CS O > VS  tr-  -A  t—  tr  VA  •6  co"  ....  r-o  rr" O s i rf  (V) r r -  c s c<  \N  in  cr  3  >  6  VA  rr 5^ rr  -  1  rt  rr  V  3  t> Lf)  "Tr  rs  £  I i ! i  "c? r-rr r  rr vs s9  cS  0  'r- " r n  in  rr  8crr  6  fA  c  _o  cr VA ro  bo ri  VA  rr  •=r  rZ vi  $ •r  rr<  cr  CS  -y  v>  cu  J  s  \S  LA  O  O r-  1—  rr  rr t.  rA  cr  rr-  cr  0  *-  IP  r  •c srS  O  CP  cS s9  r~  m _c  v^  0;  s9  OA  a; c?  8  CA CN  A O  c<v rr rs  sS  3  VA J  CA  po L">  _o  VA  V IT  CNl -J  cs  cr  O  2 s" t •3? V~  r r -  r r tn r-  v_9 s  31  s  v?  ro CO  ra  t — v5> CN CN  LA v9  ro  8  rs  s&  6 s5 cy-  •  "tf Q  T3- o c x CS  In  cx  E  d  _  00 CO  <u  O O 0 VS CN CS <3 O rA rr- iA r-r- r - VA rA  vs  r- 0  V9  -  to  U  —  r-  lA  rr?.O r\ n rcr  -r-  r~ 0 O cs CN V9 OA <N.  (vA O r1 & S9 OA CO rr  VA  cn  rD O  0  r-  r  -  r^ C«  r-  u  r-  cs  rc  PA  r s rs  cy=  (VJ •0 0  —  rr O LQ c s Vn CA in CS.  o o z o  0  0 0 v9 rr r s CS O O r~ vi v9  0 0 O PA ---  r?  CA O AA • rr rrO VA r s rc -s rr rA rr c < c^ c s |V> v-c O V A O v9 c s VA cs v< r i --r CA \ S rs r s LA rA  IA rA  cs  O 1—  6  (vA  r-  rA c M  CA rs  rr oo 0 0  0  & O  rr cr r-  r-  c s rs  rs c s r—  rA  r-  CP  ro vr> 0 Vo CA •3-  Tr  (VI CH Q  cr  $  0~_ O •q O  OO  r-  OA rA  O  O  <N CH  T<  -  LA <s  rr r s  VJ  t-r rvi O  rr rj-  IA  -  vi  CA  cs  Vc  1—  H  <H co  I <a X  -  N\  r-  vO z O oC  z  S  o  1ft  to r-7 0 t£ JA c s <D  CO  0?  \A  z  --  rr- rrj rA r r— CA  ©  z  —  .....  «./  cv vo  1)  c5l  0  -r\P  rr y rr 1.  CC  Ql IA>  V  IT  VA  O  -v Kr  it  7 rr  D a l e/ S a m p l eC O DC A R BBOOND tpTir  1** S k x c *  U3U.20M u — /  L  Nou.23  5  3 7S" <l7 "2,44 '! 2 ;o S 3 (551 ' ? , <«7 C60  0^9/-O TOC  NITROGEN T Vi N i NH*-N N 0 - N 3  2.3 2.0  0. 1 /0./5" 0.0<3  2.4 /5.Z O.I5/0./5 2.7 /5.7 0.015/0.11  47.0 47.o  61-3/76.0  (35, i sc  11 &3) (.ci)  74  T  MnS  I'  2.9  l/.f>  4.5  3.2 2.1  0.13/0.1?  n.is  4* 4^  (b.l 5 10.0  n.*.  /(o.O  bT.l  ASO  1.1 5.0  /2.I5  5.S  lQ/c%  7.1  II  243  p  S b e « Dec .2?, SeWed "  (14,0/  2G2  123  I up 20.0/2^.7  10.s  Stc^e Dec XI  +  c cr n  1-22  7.7  CD D CQ  II?  7/£ 7.6<f  CD CD 2  2 ^ 0  IM  i-zo  7.7  13.3 16.7 3.15/2.4$  -  1"  4G / F4 2-0 26 1026 /qos I020/H6«, 3.W/3C2  34-O  e  L  Conol,  2 i c /24</  :  470 <•! (5"S-) 103(55) (3) 21 31 *?5 1'* H o o f rJo-J- 21 qrt. „ Z5 4 4 C60) OO 23 Go C60) ( ? ) 'Co 24 12 20 5*? 5c; C4G; CO IS qn i. (4) 7 SIO^P p <:.'-l 4"  SOLIDS i>9/.0 TSS TVSS  12.2/12.0 l.vr/\.<is 0.06/0.025" IO.fi  o.o o.oz  i<2.. I14.3 5.3 16.0 4/. zlq o.i 17.3/17.3 14.1/14.7  ^  27  in  7.3 /6.0 q?.i/M4.o 23.5/2^.0 $.7 / i l . 1 /S.l n o / K G 56 /s?,7 lg.l/2%0 I4.0/M.7 I°I3 /262 HI. N G / l i o io4 / | 4 0 21.Z/2V.1  3 CD  2 ?  V  3 2 0 ? 2 0  iM 7.7  7.4 7.7  7 6 . 0 2 . 0 6 G. 0 / ^ . 0 ro  u b c  e n g i n e e r i n  IA  rr r- f -  Cu  C Vv r- r-  r A I/!  sS O  vS> O o) r-  r-' r  r—  r-  ^7 c to 0  U  r\ •  CSJ rrr  cr  £  o tr  LA  _CN  to >  tr  Q r_l  0  05 t o  h  v5> t r C~ (V  t r c> O co cr  rr  OS  <r  o  rs cS  6 tr  z  r  <?  «rr5s Cs s> O o  rr-4 cr r>o  c>-  <— IA  o  oC  r f vS <S oc  cO in  iA CO  r~ o  r  •Tr  tf  rr  cS  -  OA  in  r  CA  r~ rO  o rs  rs  rr  0  CSl  fO  CA  n  q  rf  o  1— (O vi  rO  rs  LO Ln J— r m cr  in LA rS  q  rs  Ln  vi  O  (— r  s ^ VA'  d  In  O  if  "io  Ln Ln cs cS  o  cf cn  LO LO  o r  ro r 6Q P r f Lo  LA  cS.  Co  J— in cs cf CO ro tv\ r f J A ISA  o  in cs  ro  Oo r o O tr  S3  oS  cj rr  O  o  >cS  o  Z  O  o< o  ccl  VA  Cy=  {>£. v_9 ?  Do  U l 3C  o  vS  o  o  c» rf  r-  CN •XT  1—  r f CX  CO CX  cf  cs  tr  O  tr  l/v LA  \A Ln  in  o  CO r o rS t i -  o  Oc. o~  tr-  i —  Ln  cc  vS r s LA'  cr  CP Lo tn  rr,  Cn  ro  10 L i  o- q o  v 5 cs  z  /-> *C \ 01  r  cs  OA  cs  zz c"  Z  cs  -S  V S S3  r-  L0  r r v3  1  <rS vS  t r rS tr  to t o  cf r cx  q q  rS  o v— r-  CN  rs  (VI N  rs  o  6 s S rr Ol  r r CS vS  to  o  rm rs  r-  o OA r~-  sS  cf r r  CP  ro O  h;  ^? o  oo  vS  JvA  «r rs Cv> ty>  1  vj  cr  O o  ro  a-  cf tr  Cr  r s O l rr r— lo  Z  O  rA to  ol  O v/> CS.  t r v-< K. rr O 0° tr O l  to  z  sS r r  v9  to  i/)  O  rr rr r S o4  *H  o  ff  m  r~,o  rr tr  cs  t ~  O  0 h  ir>  C4 r — 0 ° u")  L")  rS r —  cr  o o  rr cs0 cr r o  CS  u cy> d 4) cx  Q  t r I*  vS  cr  r  -  r rS  O rA O  rr  rr  rs  r~  ol Co  CA  cS  o-  AO r xr  b  cr 0 °  cr  cy  •O  c5  V  0  ~v  —f  V  rr  1  a'  if  P  * V rr  I/  * rr  rr  o  CS 0 CP r-  h°  ».  CA m O  v./ CA  CP C P  on cr i—  s  c  •±  ni  U  ^_  •£  u0'  5  6  V-  *  x r r 2— r r l  Ln  r O  ft-  l  6  v  D  s  ±.  rr  rr  T)  a-  r5 y V.A  -r L^>  1  fL r r  (JO  SS  o  T a "ti  c  rj;  60  1  rrr* c L^  s3  'r-  r r IV) vf)  o  c  ]  rr:  rre  0'  a' tr 0  \A  TT  ~t'  r-n c s  "3  'rf  l»  o  "T>  -—V  r- S3  m  r  Ln  ol r -  s9  vS do  cr  r,  :> c5 T> o  L  i — rS  Ln  s> r—  oJ cs rS v J 0° 0 °  tr O  LA  v9 r- D° in  o-  cf  \  Cx  1A  s  cr  of ro  v5>  E  LT)  r r r r cv  v3  ol cr  j—  z O oO CD CO or  rr  H rr  tr  u b c  e n g i n e e r i n  CL.  233  CTPO  rs  Cr  v9 r—  r-  vS>r~  O O  0=>O 00 6r<0  O rr 0o PA CX rs  rr bo  OA  r  r~ r-  rA  v9 oo cr Ln cx r<  rr r r  r r-  iA  r  r  1  TJ ^7 C </> O  vj /-~  S-<  ^£  £  rrs  N  E  \  V-'"'  oor  tn >  O  a _J 0  01  r- ©  rA  cr  r- r— rA  o  vi  si  rr  -rrOr q  r~  rA  r-  r  Al  r  PA  r- o q rs  oqd •rr rr z Vn  rA  rs  rs  r-  cr s9  1 c"  Ui X  rr  0°  s i rA rA  "rr rr  r s rs I—  -  si  I-  rA  o q  O rs O  of  rr  sS  r  8 •SCP -SO rA  r s rr  cr  LA  O  z z'2  rs  O  O  m  OA  r s r s s9  Ui  o  r—  rr  ~~A  1  CN  r-  rA  LA i A  rs  s9  o rLA r=9s 3  v3  s  O  q o  rrv->  rs  rf  L A CS  fA  r—  rr-  Pjo"  r-  r-  (Vl  L"  rA  r-_  vA  VA'  rr  si  O  si  o n r-i r ~ s9 i-n O  L 1Cw ss> Ln •o  r~ o>  z zz a: r-  Vi  r*  VA  cr  c  IT)  rA  r r~rC  C^ rA  ri  rs s5  -  r"  rr  r—  r-i  r  rvi  o  r-  r> &  iA O  vi  Cxi CO  rr  iA  rf  r  rs!  q  LA  V  CNA  q  si  tr  M  LO LA  cs  -  Do  rA  Ln  01  IS)  O  o-t r  rcr r i  rr  LT  s9  r  1  LA  rr  q  LD  CY1  r\9  rs cs r s ro  Cr-cr  LA  O  0 fsA (vA  LA  rs  O O  iA  rr o r o r rr o r s q r—i rv i r>, <sl q o„ q  rr  ISA  rr  CN  r — r—  O  VA  rr. r—; -o  L/?  rA  iyi  (V) o  PA  rr rr r~  r-  r- 6  s 3 r s rs  cr jrr- rr  rs  O rr rr cs H  oo  t~  VA PA  rr  rs  PA  fS  <?  VA rS  rf  cr  CS, cs  Ln  1-  ex  O 01  E  a  0  r-  z oO a: cr  rJ rs rs  0  CO tA  r~ O  r;  po  rr  vn  0  s9  u  s9 Lf) *s_  s5>  1  1  s_/  ua  r ^  rr O  vA  •Hi o rr Cr§s^ rr O  r-  rs s9  V . " LA  si> s9  E CV 0> tf U - U_ ~b r5 0* V>-  0  tf  Q  I  s  5s  iIx  rs  VA  c°  s9  s3 LA  sj» La  rr  0'  5: 6• if  0 IA  0'  6 C  rr I n rs V3 O  -r" L  •3' o  o  .£ rr  rr  r_  f  •i  rr V zr  vn  o  ^\ il-  rr"  6  v/»  m  r— Crcr s S  rl? <r  O  t-•ZJ m  rs  _Q  •>  LA  C5  r  "T-  _Z  rr ? L rr  si  v5 rs  a»  a>  (5  c- 0 °  (i> C D  LA  V  /J  cr LA  m  vn  tl  CD  CA  rS  -D  V^  rj- hA  s 5 rS 3 03 r*  r~  ai  61  LO o rr(.-  r  iSL  Lu  VA  4  O  rn cs  jci  LV  d  ro r o rs;  LT\  r - sS s 3 LA  V  \ "Tr a. _o  *•  1  "s  <u  L/~)  5  CD  LO  CO  A j  O  a> (? 0  cu  v/^  -f  c  4) VA  i.n  if  rr  -r"  V  •£  rr  rr  r:  rr  7_  rr  u b c  e n g i n e e r i n g  *Zl  Ln. CP r-^  LA  r- r~  Cu  .  tr  lo r-  r-  _ _ _  vO  234  >  r- 1—  c  U°  £ Ol  to to  >  a 0  cn  on  ryi  PA  tr  to  p° on LA  0°  rv)  CS  fro o h- c r r~  Z i  jo  c» rf  rS  o  X  O  o r~r rv)  r; •o  rv ir»  -  00 q O  r- o o  _^A  rs  rn rf  rr o fO v i >o ro O r- CO cfrf rf  fA  rv, ^/-\ i A  OA  VA OA  rO r>A q  c~  cf  tf  Oo  rr  LO  r rr  0°  cr  rS  v3 t y  0o  cr  rf tf  o r;  cvi  vi rs  ro q to r s OA r s  rS LO  vi  r~ r;  q  sS  r o r r s Cr rf  «  q  <5  o Q •o CO  Cr  (O CO  rS> O  o rr O i n cs  (V)  r~- r-  r- r-; on vj> tf CM  $  —  CA  rf  kA r-M  t—  (O  (VA  O  rs  O  Ln  LA  c f T~  cf  ro  cr cf  LA  ro  q r— o rvj  rrf CO  rsi  ii  if  ro  o ~r o  CS  C»a  (vo  CO  if  o rf  r~ ro r f c f r-  CS  in r r  q s9 6  c r L0  :cS  O 0^  V?  Co  rA  "to  q  tr tf  cf r f  r  r^f^ q JV\  S3 ryi  c r  to  O  r o LA  r~ rrfs r.  O  rs  cf  ro  q rr  q  CX>  cf  LA-  to Oz  cs  LT3  al  r/} t r cr o  r-  6  rs  z r-  q  S VCL  E  cf-  tf  K>  Z  CO  vi  rS Lo  i  Ui  o  vi  O q  z  z  q  to  If)  O  f  rs rs  CA  fO  o  o o o tr rf r\ o 6 O  cT  co  _J  t r cr  s9 rrr on  h  N  OO rr 00  rr  <v to  cc? cS  cy?  l O  rr  H r-<6  cr  i  tr  rs  6 6  cf  cf  rs!  r  </; r\  cS  (f  c- v5  d  6  rS  °? d oi  to  d  —  ^s  L'l  in  C  r~  6  C  in  Z  O  o  0  CO CO  a  1  CL  C P U a r- H s_>  0  u  O rs  0"  —  <u n "a. _C E tf  to  \  o  LA  <-s  q  rv)  cr  rr  ~V ai  c  o;  _  LA  i  tr  Oo C« r  t>  H  i  32  a)  X <£_  s  rl  b LA  5- r r  s  *a  •£  Si. r r  c>  tr  O  if  •  cr  cr r~  r-  r r  v3  rv S  _D  21  -c  y  g  ai  •  CV  0" &  § aJ  rr  vS  c r  rS  io rr  f-  'cr  vVJl  v5 w  o  5  is Xn  ^1 r- cr v3 La  o 0~  X  -Tj  <v  Lo  if '7 rr  O rr  o  0  J  a> rE  in  ri  o  0" 6 0 •£  r?  LA  v3  uv  i/  tr  vf3  tr-cr  X  0  QJ z  >  r-  ~>  0 r  r  c r  3 La r  C r  •3-  -t- X"  Q  /  ^rx  Cr  d  a  C  r— I~  r-  5-  1  V* 7  rr  0)  L/  ZT  u b c  e n g i n e e r i n g  235 — =c  •  ex.  c U° I/)  h £ 00 C/7  m Q  r-  (>•>'  0 to • A  r~_ vs  V/)  r\ri  PA  r-  CA  0/>  CP CsA  O  vi  ooo  1  VA flA r s  On wi  <s i — tA tr  r A CA  r-- cs C £  cs  z  oo  r~- C A o q o v; V ) s i  r- Cvi Cvi q vs rcs cs r y i r- rr vr ~i;  o  \- v> o /"S  rA r r CA  r> s i c>o  si  r-  _J  O o r-  or-  cv  Cvi iA  ve t>  q PA  cs  rvi  vi  rr  rS  1 ro  rs  O  £ •~—'z  Z  Vi  1"  tu 3 T VO  O h-  Z  r  rV\  1  tr  CP  z  v3 tr-  tr- vS V A l A  O  0°  O O <5 6 V  •j-  z  tr  72}  CS  A/J  1-  CT)  cr O Cr o> CV) 60 v9 r-  tr rr  0  CN  rr rA cr  o  0 o  v9  rs vS  (V  Q  6  CN  rr  co &o °? 00 v9 0" r  r-  £ h  rr  r~  r r fA  S>  z  O0  o  rl  O  rs  CV)  to r r  x  CD CO  Ct cr  o- in  • —  u o cs L A 0 r~ L A u CD  LA u i  r—  rr LA  S<5  r-  r~  r-  sl  -CA "a  r— v9  \3  rA  v5 in  V 9  H  o  s  V Si.  2  5; rA v3  b  i.  >.'  :  t»r  0  v/>  ~Y l/>  o> T  t> ~*c  Ot  \f *r  P  cy <?  0' tr-  r  \P  :£ tr  i  '1  ' c  <JX  t>  o 1  SS,  'A  r-  "&  si  %  r-—  cr  on -+tf  v  CS  CL  E tf  O 0°  P~  v9 v_/ —. (SO V/i  t  8  rr  a)  V>  V>A  tr 7  T>  rr  TT  to  T  0. •3= o> 6  V  sT tr  ^. ajr  i  <-  r~  0 a» 0 0  r*r  (i  -V  lA I- " 3  i  \  0)  e) 3) o> 0" Cf 0; A" 6 0 O 6'A 0 -r i d  t> i.  ri  7)  P A tr  zr  i  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items