rbc treatment of a municipal landfill leachate: a pilot scale

RBC TREATMENT OF A MUNICIPAL LANDFILL LEACHATE: A PILOT SCALE

EVALUATION

by

CRAIG CAMERON PEDDIE

A THESIS SUBMITTED IN PARTIAL FULFILMENT OF

THE REQUIREMENTS FOR THE DECREE OF

MASTER OF APPLIED SCIENCE

in

THE FACULTY OF GRADUATE STUDIES

Department of Civil Engineering

We accept this thesis as conforming

to the required standard

THE UNIVERSITY OF BRITISH COLUMBIA

October 14, 1986

® Craig Cameron Peddie, 1986

In presenting this thesis in partial fulfilment of the requirements for an advanced

degree at the THE UNIVERSITY OF BRITISH COLUMBIA, I agree that the Library

shall make it freely available for reference and study. I further agree that permission

for extensive copying of this thesis for scholarly purposes may be granted by the

Head of my Department or by his or her representatives. It is understood that

copying or publication of this thesis for financial gain shall not be allowed without

my written permission.

Department of 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 COD 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 5 / m 2 * d for carbon removal and 0.8 g NH 3 -N/m 2 *d for

nitrification). The results demonstrated that long hydraulic retention times (HRT >4

hrs.) can offset the effects of lower temperatures. 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 Treatment 23

4. EXPERIMENTAL PROGRAM 45

4.1 Sampling And Analysis Program 47

4.2 Sampling Procedures 52

4.3 Analytical Procedures 54

5. LEACHATE QUALITY 58

5.1 Leachate Generation 58

5.2 Affect of Water Inputs on Leachate Quality 65

5.3 Premier Landfill Leachate 67

5.4 Organics 85

5.5 VFA's 89

5.6 Nitrogen 94

5.7 Total Solids and Specific Conductance 95

5.8 Metals 98

5.9 Specific Trace Organics 104

6. PILOT PLANT 106

7. RBC OPERATION 112

7.1 Start-Up 112

7.2 The Disruptions 116

iii

7.3 A New Beginning 119

8. TREATMENT RESULTS 131

8.1 Carbon Removal 132

8.2 Nitrification 140

8.3 Suspended Solids 147

8.4 Metals 147

8.5 Specific Trace Organics 150

9. DISCUSSION 156

9.1 Organic Removal 156

9.2 Nitrification 170

9.3 RBC Response to Variable and Intermittent Loading 183

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 Operation 190

10. SUMMARY 193

11. CONCLUSIONS 195

12. RECOMMENDATIONS FOR FURTHER RESEARCH 197

13. REFERENCES 199

14. APPENDIX 1 206

15. APPENDIX 2 214

16. APPENDIX 3 220

iv

List of Tables

3.1 Summary of Leachate Treatment Studies 12

4.1 Sampling Program 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 of a Municipal Wastewater 188

v

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 84 vi

5.6 TOC vs. COD 86

5.7 B O D 5 vs. COD 87

5.8 B O D 5 vs. BODr/COD Ratio 88

5.9 Leachate VFA conc'n vs. Time 90

5.10 COD(vfa) vs. Leachate COD and B O D 5 91

5.11 COD vs. VFA 92

5.12 B O D 5 vs. VFA 93

5.13 N H 3 vs. COD 96

5.14 N H 3 vs. Sp. Cond. 97

5.15 Tot. Solids vs. Sp. Cond. 99

5.16 COD vs. Sp. Cond. 100

5.17 B O D 5 vs. Sp. Cond. 101

6.1 Photo of RBC Prior to Start-up, Showing Disk Media and Influent

Pump 106

6.2 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) 113 7.2 Photo of mature biomass growth during start-up (late June 1983) 114

vii

7.3 Photo of Pump tubing failure 115

7.4 Photo of Aftermath of 1 s t Flood in the RBC (November 1983) 117

7.5 Photo of RBC after being raised 1m to avoid flooding 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 123

7.8 Photo of single bellows leachate pump (165 rpm) and nutrient pump 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 thickness across the RBC 129

7.12 Photo showing leachate foaming in RBC first stage 130

8.1 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 and 4 t h Stage B O D 5 (settled) vs. Time 136

8.4 1 s t and 4 t h Stage B O D 5 (filtered) vs. Time 137

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 RBC Nitrification Performance - part 1 142

viii

8.7B RBC Nitrification Performance - part 2 143

8.8 Effluent NH^ and N 0 3 vs. Loading Rate 144

8.9A 1 s t and 4 t h Stage N H 3 and N 0 3 - part 1 145

8.9B 1 s t and 4 t h Stage NH^ and N 0 3 - part 2 146

8.10 1 s t and 4 t h Stage Suspended Solids 148

8.11 RBC Effluent Suspended Solids 149

9.1 B O D 5 Removal versus Loading Rate 159

9.2 COD Removal versus Loading Rate 160

9.3 B O D 5 Percent Removal versus Loading Rate 163

9.4 BODjj Removal versus Loading Rate Corrected for Temperature 164

9.5 B O D 5 Removal - Monod Kinetics Approach 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 N H 3 -N Percent Removal versus Temperature 175

9.9 N H 3 -N Removal versus Loading Rate Corrected for Temperature 177

9.10 N H 3 -N Removal versus Hydraulic Retention Time (HRT) 178

9.11 NH-. -N Removal - Monod Kinetics Approach 180

ix

ACKNOWLEDGEMENTS

I dedicate this thesis, and give my greatest love and thanks, to my wife

Joan, who patiently supported me in every way during the protracted process of

producing it. I would 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 made

considerable use of the facilities and expertise within the Civil Engineering

Workshop. Again, with the above in mind, 1 would like to thank Dick Postgate,

Head Technician, and his staff, especially Art Brooks, and Guy Kirsch. I would also

like to thank the members 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, Mano Ramarathan, Chris Town, and Paula Wentzell.

I would also like to acknowledge the generous assistance and cooperation of

the District of North Vancouver, the staff of the Premier Street Landfill, and CMS

Equipment Ltd. of Mississauga, Ontario. Funding was provided by the Natural

Sciences and Engineering Research Council of Canada (NSERC).

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 Lynn

Canyon Park. The bench is composed of fluvial sand and gravel up to approx.

6 m in depth, underlain by a dense glacial till which is contiguous with the walls

of the bowl.

The District of North Vancouver began using the property for a landfill in

1959, and by agreement began accepting wastes from the District of West

Vancouver, and the City of North Vancouver, in 1969 and 1970 respectively. In

1981 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 New Section of Premier Landfill look ing North-West (June 1983)

4

-PREMIER STREEJJ y LANDFILI

Burrard"""ir" Inlet

V a n c o u v e r

N.Vancouver

S C A L E = 100 200m

Figure 2.2 Premier Street Landfill • North Vancouver, B.C. Site Plan and Location

5

Figure 2.3 PhoIo _ _ bTêwêclion 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 NH^si .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 COD and B O D 5 , from influent COD concentrations of 2,240 to 22,400 mg/L.

Loading rates ranged from 0.43 to 2.16 kg COD/m^*d, and detention times were 5

- 20 days, at an average temperature 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 dependent, with COD

removals dropping from 87.2 % 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 BOD 5 /m 3 *d) with landfill leachate. Effluent

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 kg

BOD[-/m3*d and warm temperatures (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

1 0

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 COD of leachate is mainly soluble, COD

removal was minimal. Activated carbon proved fairly effective at removing the soluble

COD 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 COD 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 leachates, treatment conditions, operational

problems, and treatment 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 BOD, , values of the leachates used in these studies

are, with the exception of Palit and Quasim (56), who used diluted feed, moderate

to high in comparison to those of the Premier leachate used in this study.

However the B O D ^ / C O D ratios of these leachates are all relatively high (>0.5). As

pointed out 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 studies

have dealt with high organic strength leachates from young landfills. The discrepancy

between organic strengths, as well as the lack of comparable leachates encountered

in the literature, indicates that the Premier leachate is somewhat unique to have

such a low total organic strength coming from a young landfill. The reasons for

the low strength and other characteristics of the Premier leachate are explored in

Section 5.0.

Column 2 shows what is perhaps the most significant point, that all the

studies demonstrated very efficient carbon removal from different landfill leachates.

The results indicate that aerobic treatment processes, operating within limiting

conditions, are generally capable of complete treatment to effluent B O D ^ values of

less than 25 mg/L, regardless of the initial leachate strength. However effluent C O D

levels are usually much higher (100 - 900 mg/L), largely due to refractory humic

Table 3.1 Summary of Leachate Treatment Studies (suspended growth)

Column No.

References

#1 #2 #3 Influent Effluent Loading Rate

COD mg/L COD % Rem kg / m 3

(BOD mg/L) BOD % Rem (F/M Ratio)

#4 #5 #6 #7 Op.

Prob. Temp. Foaming Heavy Nitri-

Effects or Metal fication Comments

[BOD/COD Ratio] tSRT] Settling Removal

Robinson & Maris (65,66) (1983,1985)

Keenan ct al. (43)

(1984)

Ehr ig , H . J . (22,24) (1984,1985)

5028 (3035) [0.60]

18488 12468 [0.67]

BOD<20 (0.21 or less) @ SRT>10 d |>I0 d)

939 1 18

95 99

285-49900 (27-29975)

[0.013-0.92 1]

Mostly BOD<25

(0.12-0.32)

0.0005-1.128 « 0 . 1 )

[10-70 d]

Minor Antifoam Good Added Bulking

@ SRTs5d rising sludge

Foam Poor

Settling

Very -SRTs 1-20 days, <5 erratic, >10 Poor worked well

-very long SRTs required for nitrification

Very -full scale plant, 0.144 Mgd Good -influent NH3 conc*n toxic (1072

mg/L) air stripped.

Excel, -full scale Idg. <20 g /m 3 prod. eff. <25 mg/L -F /M <0.05 recommended to avoid filamentous bulking -complete nitrification when N/MLSSS0.03

W o n g & Mavinic (81) (1982)

13000 (8090) [0.62]

148-888 7-188

>93 >97

(0.1 1-0.405) (best @<0.16)

Minor

Poor

Good Control -nutrients 100:3.2:1.1, temp down to Reactor 5" C

Only -nutrient levels have little effect, F/M deter, settling

Zapf-Gilje & Mavinic 19000 <900 (86) (13640) <97 1981) [0.71]

>95 0.96,2.14,3.21 >99 (0.18-0.49)

[6,9,20 d]

Minor Good -temperatures 2 5,16,9' C

Poor

Column No. #1 #2 #3 #4 #5 #6 #7 Influent Effluent Loading Rate Op.

Prob. COD mg/L COD % Rem kg / m 3 Temp. Foaming Heavy Nitri-

References t< t 3 0 D5 m9^-) % &ern \?AA Ratio) Effects or Metal fication Comments

[BOD/COD Ratio] [SRT] Settling Removal

Stegmann & Ehrig (74) 4000-16660 20-3500 40-94 0.16-0.9 (1980) (750-11253) 20-60 95-99.9 (0.02-0.1 1)

[0.19-0.67]

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)

Uloth & Mavinic (79) (1977)

365

48000 (36000)

[0.75J

29-55 85-93

(0.226-0.436) [7-23.8 d]

>98

(0.06-0.22) [>20 d]

Minor Good

Good

-influent diluted 22-26 times -kinetic parameters evaluated

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

Cook & Foree (15) (1974)

Boyle & Ham (5) (1974)

Premier Leachate

15800 (7 100) [0.45]

2700-9200 (1550-8000) [0.47-0.87] 86-4421

(44-3020) [0.25-0.75]

290-360 10-26

430-6720 160-7800

[10 d]

0.3-5.28

[5 d]

Foam -F /M > 1.5 for failed unit

14

and fulvic acids (11,12). Further C O D removal has been demonstrated with

physical-chemical effluent polishing (11,23,74), activated carbon being particularly

effective (as mentioned previously), but as indicated by Stegmann and Ehrig (74),

the necessity of removing this residual C O D is a subject for debate.

Column 3 gives the loading rates and/or solids retention times (SRT) at

which the various experiments were run. The conditions for which good treatment

was achieved were very similar in all of the studies. It was generally concluded that

a SRT S* 10 days and a F/M ratio of less than 0.1 - 0.15 kg BODj/kg MLVSS, or

volumetric loading less than 0.1 - 0.15 kg BOD,-/m 3*d, was necessary for efficient

and reliable treatment (43,66,79,81). These operating conditions correspond fairly well

to those describing the extended aeration variant of activated sludge (51).

The long SRTs and low organic loading rates necessary for efficient leachate

treatment indicate low rates of growth of the process bacteria, which should be

reflected in the process kinetics. Mavinic (49) summarized the kinetic parameters for

carbon removal from a number of leachate treatment studies and demonstrated that

the difference between the values of the kinetic coefficients determined for a

leachate, and those typical of domestic sewage, increased with increasing leachate

strength. Therefore, despite high proportions of readily biodegradable material in the

leachates studied, inhibition of bacterial growth was indicated and increased with the

strength of the leachates. It was concluded that the SRT required to effectively treat

a leachate increases markedly with the strength of the leachate. Since the SRT is

inversely proportional to the F/M ratio and loading rate, the maximum loading rate

would decrease as the leachate strength increased. Kinetic considerations also

indicated that old leachates, those with a low BOD^/COD ratio and thus low

biodegradability, would also require longer SRTs and lower loading rates. Mavinic

also found that the kinetic parameters were greatly influenced by cold temperatures.

15

The effects of the changes in kinetics with leachate strength can be seen in

the results of Stegmann and Ehrig (74). They reported on activated sludge and

aerated lagoon leachate treatment 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 results. High ratio (high strength)

leachates (>0.4 BODj/COD) could be treated at loading rates up to 0.05 kg

BODr/m.3*d, but longer detention times were also required. For leachates with

intermediate ratios of 0.1 - 0.4, loading rates of <0.01 kg BOD^/m^*d were

necessary but the detention times required were relatively constant. Leachates with

low BOD^/COD ratios <0.05 required very low organic loading rates, < 0.002 kg

BODg/m^*d, but 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 + P) balance

in the process. Landfill leachate is generally found to have sufficient nitrogen

present in the form of ammonia (NH^) but levels of phosphorous are usually

deficient. Therefore most leachate treatment studies have added nutrients, particularly

phosphorous, to prevent nutrient deficiencies. Cook and Foree (15) found, as

mentioned previously, that nutrient additions did not have a great effect on total

treatment efficiency, but the effluent quality was improved slightly when nutrients

were added. Stegmann and Ehrig (74) also found that P addition had only a minor

effect on effluent quality. A lack of phosphorous inhibited growth in one activated

sludge unit but the reduction in soluble effluent BOD,, with P addition in another

unit was judged insignificant compared to the overall removal. Wong and Mavinic

1 6

(81), while investigating the effects of sludge age and temperature on leachate

treatment, confirmed the previous work by Temion and Mavinic (see 81), which

showed a nutrient ratio of 100:3.2:1.1 BOD[-:N:P was sufficient to satisfy the growth

requirements of the biomass for leachate treatment. They also found that

phosphorous deficiencies resulted in poor settling of the bacteria. Robinson and

Maris (65) also found that nutrient requirements for leachate treatment were less

than the commonly accepted ratio for sewage treatment when ammonia nitrogen in

excess of 100:3.6 B O D ^ N remained in the process effluent. The reduced nutrient

nitrogen requirements for leachate treatment may help explain the small effect that

P addition had in the Cook and Foree, and Stegmann and Ehrig studies, but the

lower effluent soluble BOD,- levels achieved with P addition indicate addition i s -

beneficial to attain high levels of treatment and improve process reliability.

Column 4 shows that despite the indications of temperature sensitivity from

kinetic considerations (49), temperature effects on treatment efficiency were generally

found to be minor. Zapf-Cilje and Mavinic (86) observed a minimal loss of

treatment efficiency with decreasing temperature down to 9° C. Similarly Robinson

and Maris (65), and Ehrig (22), reported insignificant effects of lower temperatures

on B O D j removal. However all of these studies reported impaired solids settling at

lower temperatures, so this appears to be the main adverse effect.

Column 5 summarizes the operational problems encountered during the

course of some of these studies. There were basically two types of problems;

excessive foaming of the leachate, and poor settling of the bacteria. Excessive foam

formation is a common concern of leachate treatment. The high aeration rates

usually required for treatment and the general use of inefficient coarse bubble

diffusers, to avoid plugging, further aggravates the problem. Surface aerators

frequently cannot be used because of concerns about foaming (or in other

instances, heat loss)(33). The treatment studies show a trend towards increased

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%), Cr (66%), Cu (66%), Pb (64%), Hg (63%), Al (51%) Cd (46%), Ni (33%),

Mn (20%), and Ca (6%). Although landfill leachates often have higher heavy metal

concentrations and different relative concentrations between metal species than

domestic sewage, the leachate treatment studies have demonstrated heavy metal

removals with similar affinities for metal species and similar or better removal rates.

Higher removal rates during leachate treatment are probably attributable in part to

the generally longer sludge ages employed as metal removal is enhanced by longer

sludge ages (76). Very efficient heavy metal removals have also been observed

during anaerobic treatment of leachates (5,7). Under anaerobic conditions, many

heavy metals precipitate as sulphides 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 toxic effects despite the often

very high concentrations of heavy metals in leachates undergoing treatment. For

example, Uloth and Mavinic (79) had the highest leachate metal concentrations of

the studies reviewed here, Fe 960 mg/L, As 3.6 mg/L, Pb 1.44 mg/L, and Zn 223

mg/L, yet a stable process was maintained and achieved >98 percent C O D removal,

and very high metal removal rates. None of the other studies reviewed indicated

any metal toxicity problems, except possibly for Jasper et al. (42), who proposed

metal accumulation in their sludge as one explaination for a deterioration of

nitrification performance. Studies 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 process

kinetics. In most cases however, leachate metal concentrations are not considered to

significantly impair biological leachate treatment, either aerobic or anaerobic (11,33).

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 to remove the nitrogen completely as nitrogen gas, although Henry

(33) 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 COD concentrations much greater

than ammonia concentrations, this approach was not unreasonable. Fortunately, the

nitrification results, where given, are sufficiently detailed to support a number of

conclusions. The nitrification results from the various studies are summarized in

column 7 of Table 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 5 : N H 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 of

the studies reviewed, the heterotrophs were likely growing at or near their

maximum rate in most instances. O n the other hand, nitrifying bacteria follow a

zero order response to substrate concentrations greater than about 0.5 mg/L and

thus essentially grow 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 in which excess ammonia was present such that the

BOD^/NHj ratio was less than 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 studies, ln the

later case at organic loadings of up to 1 kg BODg/m 3 *d. When denitrification was

attempted in the activated sludge studies, 99% removal of the influent nitrogen was

achieved as an influent ammonia concentration of 973 mg/L was reduced to 8.2

mg/L N H ^ and 25 mg/L N O ^ in the effluent. Keenan er al. (43) reported greater

than 99% nitrification of leachate in a full scale activated sludge plant. Efficient

nitrification was also observed in the control reactor of the Wong and Mavinic (81)

study, to which excess nutrients had been added. More recently, Dedhar and

Mavinic (18) maintained efficient nitrification of an old, low organic strength

leachate. Good 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 been

as overwhelmingly positive as that for carbon removal. Robinson and Maris (64,65,66)

reported that for a study involving both a high and low organic strength leachate,

no nitrification had occurred after a retention time of 20 days. In the case of the

high strength leachate, influent ammonia levels were initially completely converted to

organic nitrogen but continued detention beyond 20 days allowed for a gradual

conversion of some of this organic nitrogen to nitrate, essentially by aerobic

digestion of the sludge. The low organic strength leachate had only a small portion

of the influent ammonia convert to organic nitrogen after 20 days retention,

reflecting the low BODj. removals, and ammonia losses were largely accounted for

by volatilization at the pH of approximately 9.3. Continued detention beyond 20

days resulted in some nitrification but a total retention period of 70 days was

required to reduce ammonia levels to less than 1 mg/L. Very low suspended solids

levels in these units probably contributed to the poor ammonia conversion (MLSS

and MLVSS were less than 200 and 100 mg/L respectively).

In a previous series of experiments, Robinson and Maris (65) subjected

activated sludge units operating with a 10 day SRT at 10° C, to artificially elevated

influent ammonia levels. No appreciable nitrification was observed in any of the

units over 82 days of operation and influent ammonia in excess of the nutrient

requirements of the micro-organisms remained in the effluent. This result was not

entirely unexpected because of the relatively short SRT and lower temperature.

Therefore the SRT was increased to 20 days and the units operated for a further

70 days. Most of the experimental units did not stabilize at the new SRT and

while some nitrification occurred in units with excess influent ammonia, it was

unstable and incomplete. The incomplete conversion of ammonia to nitrite (NC^),

and the accumulation of nitrite in the reactors, reduces the pH, as was observed,

and inhibits the re-establishment of stable nitrification (22). Robinson and Maris

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

N H 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. Once the media surface is covered by the

bacterial growth, the surface roughness of this biomass helps determine the

thickness of the water film which attaches to 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. When

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 the

laminar film are uncertain, and depend upon other factors such as the rotational

speed of the media, and the surface roughness of the attached growth (as

mentioned above). The rotation of the shaft and media also causes a gross mixing

of the wastewater in the trough 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 hydraulic short circuiting, and enhance

treatment 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 and operation of RBC systems, can be used to explain many

of the advantages and disadvantages claimed for the RBC process.

One of the most obvious advantages of the RBC system is the simplicity of

the mechanics and process itself, which results in economical operation. In a

mechanical drive RBC system, operation and maintenance costs are minimal as the

drive components and bearings are simple and reliable, and the energy required to

turn the shaft is relatively small. Air driven RBCs similarly use proven blower

technology, and the low air pressures and volumes required to rotate the shafts

result in even lower operating costs. Process control is also very simple as the

biomass is self regulating. Control is limited to periodic monitoring of organic and

hydraulic loading rates to ensure they are within design limits, as well as checking

dissolved oxygen 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 (71) in their state-of-the-art review of

RBC technology (1983), found that hourly labour requirements for RBC plants ranged

from 1 to 7 hours/week, averaging 2.6 hours/week, and that power requirements for

100,000 f t 2 (9290 m 2 ) of standard density media were 3.6 kw for mechanical drive

units (1.6 rpm), and 2.93 kw for air driven units (1.2 rpm). Capital costs of RBC

plants are higher than for activated sludge plants for capacities above approximately

1 M G D (3800 m 3/d) but landfill leachate volumes are generally much less than 1

M C D and are within the range where RBC plants are economically viable.

One of the most important attributes claimed for the RBC process is its

high performance for carbon removal and nitrification (84). RBCs operating within

design limits have demonstrated their capability to produce high quality effluents

26

with soluble BOD, . (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 is achieved in the

RBC, and the factors which affect it, are at present best explained in terms of the

mass transfer/kinetic models of Famularo and Mueller, et al. (26,53).

It has been found, and is now 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 layer consists primarily of rapidly

growing aerobic heterotrophs which utilize the carbonaceous substrates from the

wastewater. When conditions permit, 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 and methane production if exogenous substrates are

available, denitrification if both carbon and nitrate are present, and endogenous

reduction of the biomass. Other autotrophs such as sulphur bacteria can also

become active when conditions permit.

In stages receiving high loadings of organic carbon, usually the first few

stages in a multi-stage system, the heterotrophic bacteria in the aerobic zone grow

very rapidly. High influent substrate concentrations provide a strong concentration

gradient for the diffusion 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,, removal and

the highest BOD^ removal rates are commonly observed in these initial stages

(3,57,60,71). However, overloading conditions, as will be discussed later, interfere

with transfer rates and reduce removal efficiency. Influent suspended solids, both

organic and inorganic are also effectively removed by adsorption and impingement

on the thick biofilm. The high growth rate in the early stages results in a high

sloughing rate of excess biomass so that suspended 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. (47)

observed that nitrification does not begin until BOD,- levels in an RBC have been

reduced to 60 mg/L, and a stable nitrifier population only becomes established

when B O D 5 levels approach 30 mg/L. At B O D 5 levels less than 10 mg/L the

nitrifiers become the dominant population. The other studies reviewed all reported

similar findings and it is now 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 reduced to 4.15 g SBOD^/m 2*d or less. Therefore it

is frequently observed that TKN 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 into the biofilm under the heavy loading

conditions prevalent in the early stages allows the formation of the thickest

anaerobic layers. Depending upon 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 BOD^ 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 BOD - 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 resistant to degradation, the

readily utilized substrates having been preferentially consumed in the initial stages.

However in these later stages, many slower growing 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 comes

from the suspended or sloughed solids from the previous stages. These solids

generally become re-attached to the media in subsequent 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. The

endogenous reduction of the biomass reduces the overall sludge production of the

RBC process, which is another advantage claimed for the system. Kincannon et al.

(44) determined a sludge production rate of 0.37 kg solids/kg BOD,, removed,

treating domestic sewage. However the sludge production for an RBC treating

landfill leachate would increase somewhat due to the addition of inorganic

precipitates as was observed 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. Removal

efficiency of all substrates under these severely substrate limited conditions is further

reduced because 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 BOD, , removal can be

achieved with the 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 incorporated into

RBC plant design; for example the pilot plant used in this study had twice as

much surface area in the first stage than in each subsequent stage. Conversely the

decreasing BOD, , removal efficiency with successive stages, or diminishing return,

means that the maximum number of stages for B O D ^ removal is generally taken to

be four.

Configuring an RBC plant to achieve maximum BOD,- removals in the early

stages not only improves the B O D j removal efficiency but the nitrification

performance as well. As discussed above, the onset of nitrification is dependent

upon the prior removal of most of the carbonaceous substrates and therefore the

earlier in the process this occurs, the more surface area is available for nitrification

in the following stages. In cases of high influent BOD, , and/or ammonia

concentrations, or the treatment of substrates which are difficult to degrade (such

as phenols), additional stages may be required to achieve complete treatment or

high effluent quality. As will be discussed in Section 5.0, landfill leachates frequently

have one or more of these characteristics.

The attached growth nature of the RBC provides several advantages for

nitrification over suspended growth systems. Nitrifying bacteria are generally described

as being sessile (22,24) and therefore the RBC provides a prefened environment for

their growth. Crowing attached to the RBC media also results in indeterminately

long solids retention times which allows the slow growing nitrifiers more than

adequate time to develop. Since only two species of bacteria are involved in

nitrification, Nitrosomonas and Nitrobacter, they have proven more sensitive to

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 SBOD 5 /m 2 *d or first stage loadings of less than 12.2 g

SBOD[-/m 2*d. However, at loading rates higher than these levels, full scale RBC

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 would 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. When this period

of rotation is too 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 Kincannon

et al. (44), at lower loading rates the growth rate of the biomass 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 of

Friedman et al. (29) indicate that to some extent this is true. They found that as

the rotational speed, and thus peripheral velocity, of their RBC units were increased

that the maximum 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 et al. (44) determined a

constant nitrate production rate of 0.73 g N O y m ^ d . Similarly Murphy and Wilson

(54) calculated a constant TKN removal rate of 1.12 g TKN-N/m 2 *d at 20° C. The

constant reaction rate is due to the slow growth rate of the nitrifying bacteria and

their zero order response to substrate concentrations above about 0.5 mg/L. Given

that substrate utilization is reaction rate limited, the other factor limiting nitrification

is the number of nitrifying bacteria. It appears as though the nitrifier layer, in those

stages in which it develops, 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 RBC.

Another important advantage claimed for the RBC process is excellent

resistance to shock organic, hydraulic, or toxic loadings (54,81). Of these, the RBCs

resistance to hydraulic loading is the easiest to explain. As the bacteria are attached

to the fixed media they are much less prone to wash-out of the process than in

a comparable suspended growth system and therefore major losses of biomass

usually do not occur. Upsets in a suspended growth process typically result in

reduced solids settleability and high hydraulic flows would only increase the losses

of biomass into the effluent. Biomass losses are particularly serious in a nitrifying

system. In a RBC system however, process performance can be impaired due to a

reduction in the hydraulic retention time (HRT). RBCs designed for sewage treatment

generally have short HRTs, in the order of 0.5 - 2.0 hours at design flowrates,

which is considered another advantage of the process itself (84). Therefore under

the conditions of a hydraulic surge, the system HRT could fall below some limiting

value. Filion et al. (27) investigated the effects of variable hydraulic loading on an

34

RBC and found that varying the HRT between 0.44 and 0.94 hours did not

significantly affect carbon removal, but nitrification performance did respond to HRT

variations between 1.12 and 3.36 hours. However nitrification performance recovered

much more quickly from hydraulic loading fluctuations, than from changes in TKN

loading or influent concentration. Poon ef al. (60) found that for their system,

B O D j removal was adversely affected by a reduction of the HRT from 0.73 to 0.42

hours, but low influent B O D j concentrations were cited as a contributing factor

(lower reaction rates). Therefore, while the RBC provides good resistance to loss of

biomass during high hydraulic loadings, process performance, especially for

nitrification, may be reduced by lower hydraulic retention times.

The RBC's resistance to shock organic or toxic loads stems from the large

mass of highly concentrated micro-organisms resident on the fixed media. As noted

previously the first stage of an RBC is typically larger or has more surface area

than subsequent stages to maximize the mass of bacteria in contact with influent

conditions. In the case of peak organic loadings, this large biomass provides the

assimilative capacity necessary to absorb extra substrate over short term periods.

With respect to toxic loadings, 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 present, the greater the

concentration of toxin which can be tolerated. Theoretically, a completely mixed

suspended growth system with the same total biomass as an RBC would be

somewhat more effective at resisting shock loads because all of the bacteria would

be available to moderate the shock conditions. By the same reasoning an RBC

would be more effective than a plug flow suspended growth system due to a

higher proportion of biomass near the influent end of the process. However, as

pointed out above, the RBC system is much less prone to losses of biomass if the

bacteria become stressed or upset.

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 TOC/m 2 *d and 6.0 -

7.2 g TKN/m 2*d obviously exceeded the assimilative capacity of their RBC system.

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) found that for shock organic loadings

from 4.3 to 15.4 g SBOD^/m 2*d, representing 124 to 444 percent of the normal

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 2 *d . Therefore, the RBC has demonstrated good handling of

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 by half for each 10° C decrease in temperature, the rapid cooling of

influent wastewater can significantly decrease process performance. During warm

weather the heat transfer efficiency of the RBC is beneficial as influent wastewaters

are generally cool. The heating effect of warm temperatures is moderated somewhat

by evaporative cooling which Kincannon et al. (44) observed could reduce

temperatures by 2 - 3 °C across an RBC unit. During 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 must

be covered to prevent icing problems.

The reduction in RBC performance as water temperatures decrease has been

well documented (28,54,57,80,85). In most instances the effects of temperature on

the reaction rates of the RBC have been expressed in terms of an Arrhenius

equation coefficient 8. For carbon removal, coefficient values ranging from 1.03 to

1.11 have been determined for various temperature ranges. Wilson et al. (80) for

example, used a value of 0 = 1.05 over the temperature range of 5.5 to 13° C.

Many of these studies showed that the carbon removal rate does not increase

above 10 to 15° C, so corrections are not usually applied at higher temperatures

(28,54). Nitrification has been observed to be much more sensitive to temperature

effects and over a wider range of temperatures. The temperature coefficients

determined for nitrification are therefore generally larger than for carbon removal

alone, typically 1.09 to 1.11. Murphy et al. (54) for example, determined that a

factor of 0 = 1.09 applied for carbon removal with nitrification up to 20° C, above

which no further correction was required. However, as pointed out by Forgie (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 6 is only

an approximation. Therefore the use of a constant 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 as

the temperature decreases. As indicated by Forgie, a number of studies have shown

that, in fact, the reaction rates drop off more sharply as the temperature 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 approximation of

actual temperature effects only when used over small 4 to 5° C temperature

ranges. Forgie presented an empirical curve fit equation from experimental data,

which indicated a parabolic shape.

Experiments conducted by Forgie also produced a couple of other interesting

results. The first was that an established nitrifier population could continue to nitrify

well at temperatures 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. Wu 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 to become deeply anaerobic, severely reducing reaction rates in most of

the biomass and thus it contributes very little to the removal performance of the

system.

Within the anaerobic zone, sulphur and hydrogen sulphide is often produced

and this encourages the growth of Beggiatoa bacteria which oxidize these products

to produce energy. These micro-organisms store sulphur in their cells giving them a

white milky appearance, which is characteristic of overloaded RBC stages (25,34). In

addition to the reduced BOD,- removal indicated by the Beggiatoa growth, the

sulphuric acid they produce can lower the system pH and adversely affect the

nitrifying organisms. Anaerobic conditions in one or more stages can also cause

severe odour 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 increased

weight of the shaft also requires significantly more energy to turn it so energy

costs are increased while performance decreases. Therefore, overloading conditions

generally result in much higher operating, maintenance, and replacement costs.

Evans (25) surveyed a number of RBC plants for operational problems and

found that for plants in which first stage loading were less than 17.6 g

BOD,-/m 2*d no problems were reported, but for plants with first stage loadings

greater than 43 g BOD,./m 2*d, problems always occurred. At plants with first stage

loadings between these two extremes, no 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 during

the submerged 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 good resistance to oxygen depletion in the initial stages at high loading

rates and Hynek and Chou observed greatly reduced growth of Beggiatoa and other

filamentous micro-organisms. The air drive RBCs also permit easy regulation of

rotational speed, rotate slower for a given removal rate, require less energy, and

develop less stress in the shafts and media due to the thinner biomass. Nitrification

was also enhanced because BOD^ removal was achieved in a fewer number of

stages leaving more surface area available for nitrification in the remaining stages.

Another benefit of air driven RBCs is that heat recovery from the blower air is

possible in covered units. The one disadvantage of air drive RBCs for landfill

leachate treatment could be that the diffused aeration would promote foaming;

however, this may be controlled somewhat by the media.

Having discussed the various properties and characteristics of the RBC system,

the treatment capacity and performance of the RBC remains to be stated. Early

design specifications for RBCs were in terms of hydraulic loading rates determined

from manufacturers nomographs, using the waste strength and desired removal

efficiency. This design approach did not easily adapt to differing waste types or

permit simple comparisons of loading levels. Kincannon and Stover were thus

prompted to introduce the total organic loading concept in the early 1970s and it

has since gained wide acceptance (44,54). Therefore, RBC loading rates or treatment

capacity are usually expressed as mass of substrate applied or removed, per unit

area of media surface. Design specifications are also now generally expressed in

terms of organic loading rates although from the previous discussions, maximum

hydraulic loading rates, which are temperature dependent, should 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 carbonaceous

substrate (BOD^) and achieve complete treatment (effluent BOD , ^ 2 5 mg/L) is in

the order of 15 - 18 g B O D 5 / m 2 * d , at temperatures > 15 e C. Forgie (28) achieved

a good effluent quality and removals in excess of 90% at loadings up to 15.2 g

BOD,-/m 2*d and at 15° C. Kincannon ef al. (44) found that at loading rates less

than 9.8 g SBOD 5 /m 2 * d , soluble effluent B O D 5 <10 mg/L were achieved, but at a

loading of 18.3 g SBODj./m 2*d, which corresponds 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 2*d. Poon et al. (60) treating a clarified trickling filter effluent (tertiary

treatment) achieved an average effluent SBOD^ of less than 15 mg/L at loadings up

to 7.8 g SBODi j/m 2 *d . Murphy et al. (54) found that good treatment efficiency

could be achieved up to a loading of 15 g BODg/m 2 *d after which some scatter

in the results occurred. This led them to recommend a design loading rate of 15

g B O D j / m 2 * d for temperatures of 15° C 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 <19.5 g BODg/m 2 *d and most were operating

at <12 g B O D 5 / m 2 * d .

Some of the modifications or variations of the RBC system can increase the

removal capacity of the RBC somewhat beyond these levels without reducing

effluent quality. In particular, air driven RBCs and the use of oxygen enriched RBC

systems have demonstrated higher capacities. Hynek and Chou (39) comparing air

and mechanical drive RBCs found that air drive units were 18% more efficient for

carbon removal and 25% more efficient for carbon removal with nitrification, but

recommended 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 COD

removal was not consistently observed. The lack of improved COD removal in the

first stage was blamed on severe bridging of the biomass, but COD 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 COD 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% COD removal at a loading rate of 38.5 g

COD/m 2 *d (27.4 g COD/m 2 *d removed). At this high loading rate the fourth stage

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 2*d had been found in the

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

TKN/m 2 *d at 20° C. Their design loading recommendations for both nitrification and

carbon removal are presented in Table 9.1 of the Discussion.

As mentioned at the beginning of this section, a review of the literature

prior to the start of the experimental program failed to find any references

concerning RBC 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 phase, leachates and found that

the RBC was capable of almost complete nitrification of these leachates at loading

rates up to 2 g N/m 2*d. Coulter reported some results of a companion study to

this one and found that efficient carbon removal was achieved at loading rates of

9.6 and 18.3 g COD/m 2 * d (6.2 and 11.6 g BOD 5 /m 2 *d ) . An interesting lack of

nitrification was also observed during this study. These papers are discussed more

fully, within the context of the results of this study, in the Discussion, Section 9.0.

The scarcity of studies concerning RBC 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 had more

than 1600 RBC plants installed. He also reported that there were 135 plants treating

landfill leachate, the first having been installed in 1976. Therefore, it seems an

effort should be made to benefit from their experience as translation is much more

economical than research.

4. EXPERIMENTAL PROCRAM

The experimental program proposed to fulfill the purpose of this study had

three main component parts. First among these was the characterization of the

Premier Landfill leachate, both to determine the constituents of the RBC influent for

process evaluation, and to provide a basis for comparison 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 some

specific trace organic compounds. In addition to the chemical analysis of the

leachate, physical properties such as total solids, specific conductance, pH, and

temperature were monitored regularly.

The second part of the experimental program was concerned with the

evaluation of the capacity of the RBC to effect carbonaceous removal from this

leachate. For simplicity, the RBC would be operated under pseudo steady-state

conditions, for which the flow rate would be set and the influent strength,

temperature, etc., would be allowed to vary naturally. In order to determine the

maximum capacity, or mass loading rate, of the RBC for carbonaceous removal it

was proposed to begin 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 strength

measured prior to each change in flow. After each increase in loading the RBC

would be allowed to stabilize over a minimum period of three weeks. A nutrient

45

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 4CI 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 topics include: the generation and

settleability of suspended 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 these topics were in part based

on specific test results and in part based on general operating data.

4.1 SAMPLING AND ANALYSIS PROGRAM

An extensive sampling program was set up to characterize and monitor

various raw leachate parameters as well as monitor the performance of the RBC

(Table 4.1). The sampling program was based upon grab samples and field

measurements 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 RBC

installation was located beyond any supervision and therefore a secure enclosure for

the sampling equipment would have been required to thwart vandalism, (for which

there was precedent). Since the period of the study was expected to be many

weeks, it was assumed that the twice weekly samples would provide sufficient data

to evaluate the treatment efficiency of the RBC.

The main parameters used to characterize the raw leachate were: chemical

oxygen demand (COD), ammonia nitrogen (NH^-N), specific conductance (Sp.

Cond.), total solids (TS), and pH. Sampling 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: - influent flow rate - nutrient flow rate - influent temperature (raw leachate) - 1 s t & 4 t n stage water temperature

- Sample: - influent (raw leachate) COD,TKN,NH3,Sp.Cond.,TS,pH

- 1 s t & 4 t h stage (raw) COD,TSS/TVSS,pH (settled) COD,TSS,TVSS,TKN,N H 3 , N O B

(filtered) COD

- Once Weekly Procedures: (in addition to above)

- Sample: - influent (raw leachate) BODj.,TOC,aIk.

1 s t , . 2 n d , (raw) B O D 5

& 3rd stage (settled) BOD 5 ,TOC

(filtered) BODâlk.

- 4 t h stage (raw) BODj (settled) BOD 5 ,TOC (filtered) BOD 5 ,P0 4 ,a lk.

- 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 s t & 4 t h stage liquid for metal analysis, attempt to sample at high, medium, and low leachate production rates

49

the course of this study, additional tests were performed at various times for:

biochemical oxygen demand (BOD^), total organic carbon (TOC), volatile fatty acids

C1-C3 (VFA), total Kjeldah! nitrogen (TKN), and alkalinity (Alk.). These analyses, when

combined with the previously mentioned heavy metal and trace organic analyses,

constitute a fairly thorough characterization of this landfill leachate.

While the sampling and analysis of the raw leachate proceeded as outlined

in Table 4.1, the monitoring of the RBCs performance did not proceed entirely as

planned. The twice weekly procedures of Table 4.1 were generally earned out as

proposed; however, the remainder of the sampling procedures were either performed

less frequently, or discontinued. These reductions in the sampling program were

caused by the operational problems alluded to earlier. The irregular operation of the

RBC reduced the significance of many of these extra tests and samples, and also

reduced the time available to make these tests, as maintenance procedures often

took precedence. However the quantites which were measured on the twice weekly

basis were the most important with respect to evaluating the RBCs performance for

carbon removal and nitrification. The other supporting data, while desirable under

other circumstances, was not central to the goals of this study.

Thus, the RBC process was monitored primarily by sampling liquid from the

first and fourth stages of the unit. These samples whether raw, settled, or filtered,

were analysed for most of the same parameters as the raw leachate. ln addition,

they were analysed for: combined nitrate and nitrite nitrogen (NO^ + N O j -N), total

suspended solids (TSS), and total volatile suspended solids (TVSS). Field

measurements for the most part consisted of recording the liquid temperature of

the first and fourth stages, and making notes on visible changes in such factors as;

thickness and colour of the biomass, foaming within the RBC, settleability of the

suspended solids, and effluent clarity. The only significant change made to the twice

weekly routine of Table 4.1 was 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 COD,

because of the relatively high levels of refractory COD which persisted in the RBC

effluent. This refractory COD could conceivably mask the breakthrough of

degradeable COD 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 COD/m 2 -d. Measurements during subsequent weeks were interrupted by

operational problems. The DO measurements were then discontinued as one step to

streamline the sampling program until steady operation could be achieved. While

DO 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 TOC, 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 COD 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 of the leachate

were taken from the bucket. A 2 L sample was taken for COD, BOD^, Sp.Cond.,

pH, and TS analysis, and a 500 mL sample was preserved with 1 mL of

concentrated r ^SO^ for TKN, NH^, and TOC analysis. During the second half of

the experiment, a 60 mL sample was collected for volatile fatty acid (VFA) analysis.

Periodic checks made during the course of this study showed there was no

detectable difference between leachate samples from the wet well and samples

taken from the influent line to the RBC.

Samples of the RBC liquid were collected by dipping from each stage to be

sampled with a 125 mL plastic beaker. First 500 mL of the stage liquid was

collected for COD, BOD,-, TSS, TVSS, and pH measurements. Then 1 L was

withdrawn to be settled for 30 minutes in a 1 L graduated cylinder. The

supernatant was then carefully poured off to provide the settled samples: 500 mL

for COD, B O D 5 , TSS, and TVSS tests; 125 mL preserved with H 2 S 0 4 for TKN,

NH^, and TOC; and 50 mL were filtered with a test tube plunger type filter and

preserved with phenylmercuric acetate (CH^COOHgCgH^), for N 0 2 + NO^ analysis.

Filtrate of the raw and settled samples from the TSS tests was used for the

determination of soluble C O D and BOD,, as well as residual orthophosphate (P0 4 ) .

The temperature of the 1 s t and 4 t n stage liquid was also usually recorded.

The biomass samples were taken by momentarily stopping the rotation of the

disk and then scraping a square patch of biomass off one of the fibreglass

endplates of each stage of interest. A 3 inch (7.6 cm) wide metal paint scraper

was used to remove the biomass from a 3 inch square (58.1 cm 2 ) area. The

biomass was scraped into a previously acid washed and tared glass jar (8 oz. wide

mouth jar). Then the samples were dried at 104° C and reweighed to determine

the dry weight of the biomass per unit disk area for each stage. Additional

biomass sample for nutrient and metal analysis was collected and dried in a similar

54

manner, and then ground to a fine powder and stored for analysis.

The acid preserved samples of leachate, and settled first and fourth stage

RBC 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 were taken

specifically for metal analysis and were preserved with H N O ^ as prescribed by

Standard Methods.

O n three occasions samples of leachate and RBC effluent (settled 4 t n stage)

were collected for a CC/MS scan of trace organic compounds, as an aside to this

study. These samples were collected in clean, oven-dried glass vials with teflon caps.

The vials were filled completely leaving no head-space or bubbles.

4.3 ANALYTICAL PROCEDURES

The analytical methods used for this study were except as noted below from

Standard Methods 1 5 t h ed.

C O D - The potassium dichromate reflux method as per the 1 3 t n ed.

of Standard Methods was used as this method has been

adopted as a lab standard at U.B.C..

TOC - Acidified samples (pH<2) were analysed using a Beckman

915A Carbon Analyser.

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 effluent.

TKN - 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 t n ed.)

N O 2 + N O 3 - Samples preserved with phenylmercuric acetate were analysed

using the automated cadmium reduction method on a

Technicon AutoAnalyser II, a tentative standard method (15 t n

ed.)

Total Solids (TS) - as per 15 t n ed., 80 mL leachate samples in triplicate

Total Suspended - as per 15™ ed., RBC samples in duplicate, Whatman 934-AH

Solids (TSS) glass microfibre filters

Total Volatile

Suspended Solids

(TVSS)

as per 15 t h ed.

56

Specific

Conductance

(Sp. Cond.)

Alkalinity (Alk.)

- measured using a Radiometer Model CDM3,

Conductivity in /zS/cm

- titration to pH = 4.5 as per 151*1 ed.

pH - laboratory pH meter

Metals - Total Metals samples prepared as per 15 t n ed. and analysed

on a Jan-ell Ash #810 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) metal

scan

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" I.D. glass column packed

with 0.3% Carbowax/0.1% H 3 P 0 4 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.

- The analyses for specific volatile and semi-volatile trace

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.

Organic

Compounds

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 4.9 8.4 Total hardness (mg/L as CaCO^) 30 13,100

Total alkalinity (mg/L as CaCO^) 100 20,805

Total iron (mg/L) 2 1000 Sodium (mg/L) 85 1805 Potassium (mg/L) 28 3770 Sulphate (mg/L) 24 1220 Nitrate nitrogen (mg/L as N) 5 196 Ammonia nitrogen (mg/L as N) 0.2 1106 Chemical oxygen demand (mg/L) 246 750,000 Biochemical oxygen demand (mg/L) 5.9 720,000 Total volatile acids (mg/L CHjCOOH) <100 10,000

Total dissolved solids (mg/L) 1740 11,254

(1) from table 1. of Pohland (ref. 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 important and dominant phases with respect to leachate quality are

the 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 to move through the wastes. With the onset of water movement, conditions

become ideal for the growth and spread of an anaerobic microbial culture. The first

group of bacteria to establish themselves are the facultative acid formers. These

bacteria degrade the larger organic compounds, found dissolved in the pore water,

or hydrolysed from the wastes, down to simple organic acids, hence their name.

Acetic acid (CH^COOH) is the main catabolic end product of these

microorganisms during anaerobic fermentation. Some of the acetic acid then

undergoes 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 C3, 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 TOC 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 BOD 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 leachate are usually highest during this phase. Low pH conditions also inhibit

the growth 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 that in a limiting case in which there was no moisture movement

(65% 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 and the onset of the methane

production phase are generally the converse of those mentioned above i.e., high

moisture inputs, etc.. Stegmann ef al. also found that the placement of an

uncompacted and/or aerobically composted bottom layer significantly accelerated the

onset of methanogenesis. Similarly Robinson and Lucas (67) found that an

unsaturated soil zone beneath the fill rapidly developed a population of methane

bacteria such that VFAs produced in the wastes were not observed to penetrate

through the layer. (This result was 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

6 3

establishment of the methane fermentation phase. It is frequently observed that the

transition between the acid formation and methanogenic phases is relatively abrupt,

particularly where leachate recycling is practiced. This would tend to indicate that

the methane bacteria population develops and spreads, but at reduced activity, until

inhibitory conditions moderate enough to permit a rapid exploitation of the available

substrate.

From the previous discussion it follows that with the virtual removal of the

organic acids from the leachate, the organic strength of the leachate is drastically

reduced from that of the acid formation phase. Leachates from landfills with a well

established methanogenic phase typically have a very low BODg concentration (<100

mg/L), although the C O D may remain significantly higher due to the refractory

compounds.

The establishment of the methane bacteria also affects the pH and ORP

conditions within the landfill and leachate. As the acids are consumed, the pH rises

to approach neutral values. The leachate shifts from a volatile-acid buffered system,

to a predominately bicarbonate buffered system. ORP values generally decrease

gradually prior to the rapid growth of the methane bacteria and reflects the

development of conditions favourable to the growth of these obligate anaerobes.

These two conditions combine to greatly reduce the mobility of most heavy metals

during this phase. The higher pH levels reduce the solubility of the metals, while

the low ORP conditions encourage metals which are dissolved to precipitate as

sulphides. Therefore, heavy metal levels in the leachate are generally much lower

during this phase. A possible exception to this trend is lead (Pb), which forms a

stable complex with humic substances and thus remains mobile (32).

The reduction in dissolved metals and organic acids is reflected in the

reduced concentration of dissolved solids and value of the specific conductance.

During the acid formation phase, 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 of the methanogenic phase, the concentration of

ammonia 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 produced

remains the same. If the rate of water movement 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 are fully stabilized), has led increasing

numbers of researchers to conclude that the ammonia content of leachate is a

more 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 is also less stable than the acid

formation phase 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 phase is established, it

is usually quite stable and the breakthrough of VFAs is not 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 ON 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, TOC, BOD, COD, VFA, and VSS

increased with increasing leachate production or water inputs. It was theorized that

these increased concentrations came about because the increased water contact with

the wastes, combined with a shortened leachate retention time, overloaded the

methane bacteria and resulted in the wash-out of organic material. They also

observed that the nitrogen content, NH^ & TKN, as well as TIC, CI", Alk., and

Sp. Cond. levels decreased during peak leachate flows. These parameters are

generally less affected by biological activity and more affected by the exposure of

wastes 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 occurring. For the

remaining tested parameters, metals, pH, TC, TP, TSS, and TDS, concentration was

relatively independent of the rate of water input. Results from the monitoring of

another landfill assumed to have a long leachate retention time (3 to 4 months),

indicated that the levels of all parameters are relatively independent of water inputs.

Other researchers have observed similar variability of leachate strength with

water input. Bull (7) also indicates that heavy rainfall may cause an increase in

leachate strength by reducing the residence time of leachate within the fill.

However, Raveh ef al. (63) observed that for their lysimeter study, the concentration

of pollutants in the leachate was independent of the level of water application up

to 1100 mm 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 of pollutants may be variable with

respect to water inputs, it is now 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 low end of the given ranges. Average values were omitted from Table

5.2 because, due to the nature of the strength fluctuations described later, they are

not meaningful.

The dilution of the leachate by drainage from the unfilled portion of the

site demonstrates how important physical site conditions such as the collection

system layout are in determining leachate quality, ln this case the placement of the

collection pipe within the sand and gravel underlying the site promotes rapid

drainage and collection of the water from beneath both the filled and unfilled

areas. Due to the lower hydraulic conductivity and extra thickness of the compacted

wastes however, the drainage from the filled area would lag behind that of the

unfilled area. Therefore the dilution ratio would be variable. Rapid drainage also

probably means that the soil zone below the wastes is unsaturated most of the

time, ln other cases, the collection system may affect leachate quality by collecting

leachate from areas in one phase of stabilization rather than another, or by

maintaining a saturated zone below or within the wastes. Suffice ft to say, the

collection system design can greatly influence the quality of the leachate collected,

as it does in this case.

Another reason for the relatively low strength of the Premier leachate is the

quite high moisture flux through this landfill. The wastes were placed over the

fluvial gravel in comparatively thin lifts (<2 m), covering the whole area of the fill

before the next lift was started. Moderate compaction densities were achieved using

a small BOW-MAC compactor and/or a large bulldozer, and a thin layer of

permeable cover material was placed over the wastes daily. The above method of

placing the wastes increases their exposure to precipitation and promotes good

drainage of water through the wastes. When subjected to the heavy annual rainfall

normally received at this site, the field capacity of the wastes is rapidly exceeded

and large volumes of water drain relatively quickly through the garbage. The large

69

Table 5.2 Premier Leachate Characteristics (Well #1)

low - high

B<2>

low - high

C W

low - high

COD mg/L 86 - 4421 263 - 1527 150 - 434 B O D 5 mg/L 44 - 3020 ( 4 ) 161 - 1035<4> 49 - 251 TKN-N mg/L 8.1 - 53.8 18.5 - 51.2 20.1 - 41.2 NH 3 -N mg/L 6.9 - 49.1 17.1 - 46.4 18.4 - 40.3 VFA mg/L (as acetic) 1 - 1470 48 - 888 5 - 108 T.S. mg/L 540 - 3595 764 - 2176 639 - 1238 Alk. mg/L (as CaC0 3 ) 288 - 782 428 - 750 350 - 673 Sp. Cond. /xS/cm 527 - 3567 1162 - 2594 1070 - 1890 pH 5.6 - 7.4 6.3 - 7.0 6.4 - 6.8

(1) Data Period A - October 22/82 to March 31/85. (2) Data Period B - April 10/84 to July 24/84. (3) Data Period C - January 18/85 to March 31/85. (4) BODr value 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 AND PRECIPITATION

1 9 8 2 1 9 8 3


z o < 20

O UJ

D J F 1 9 8 4

M M 0 N

1 9 8 3 D J F

1 9 8 4

i L L J L h i J LL M M o

Legend A COD

X B0D5

• T. SOLIDS

H Sp. Cond.

X NH3-N


• A K

O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L

O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L 1 9 8 4 1 9 8 5

60

50

h40

30

D) E

<

20 O

Ho <

0

1 9 8 4 1 9 8 5

i II Ii • 1 i . 1 1 1 . • i 1 1 i • 1 1 . . i 1 |

Legend A COD X BOD5 • T.SOLIDS H Sp. Cond. K NH3-N

73

Figures 5.1 A,B,C, clearly show that the pollutant concentrations are highest

during the wet Winter and Spring months and then decrease over the dryer

Summer and early Fall period. Upon closer examination it can be seen that sharp

increases in leachate strength are generally preceeded by wet periods or major

rainfall events. This is most noticeable during March 1983, December 1983, January

1984, and December 1984. In general, it can be observed that leachate volumes

and strength responded quickly to rainfall events. The increases in concentration

generally lag behind the rainfall peaks by a few days and so these rainfall peaks

often correspond to sharp dips in the concentration values. This reflects the time

lag between the drainage 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 an initial diluent effect. This time lag can also be seen

in Figure 5.2, from Jasper ef al. (41), which shows the leachate production 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 increases with the

volume of leachate produced. Therefore, the data from this site indicates that the

main mechanism governing the leachate strength is the area of contact between the

wastes and the water. As noted earlier, increased water inputs increase the surface

area or volume of waste in contact with the passing water.

These figures also show quite well the evolution of this site and its leachate

quality through the acid formation phase to the start of the methane fermentation

phase. As the leachate sampling began just a few months after the first wastes

were placed into the new landfill area, and near the start of the first wet season,

it appears as though 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 COD, rose rapidly from 64 mg/L

74

P H A S E I

450-1 r36 400

o 350 28 x m X 300 24« z t— o 250 •20 c JC 200 • 16 150 • 12 100 8 •~.5.0-• 4

OCT NOV 82 OEC JAN 83 FEB I8r

MAR APR MAY JUN JUL

P H A S E Z

AUG SEP OCT

450T36 400 [ 32 350-

LEGEND r 300 Leochott voL Mas*COO Most NH 3

Mo** T.S. 200 • 16 150-12

28 24 OT 20

OCT NOV 83 OEC JAN 84 FEB

18

150 100 50

OCT NOV 84 DEC JAN 83 FEB MAR APR MAY JUN JUL AUG SEP OCT

from Jasper et al. (41),

12 8 4 0

Figure 5.2 Leachate Flow and Constituent Mass Release Premier Street Landfill

75

in October 1982 to a high of 4421 mg/L in April 1983, indicating the start of the

acid formation phase. During the Summer and Fall of 1983 the leachate strength

tapered off gradually to approximately 1500 mg/L C O D due to dryer conditions.

Although the dryer conditions could be expected to increase leachate strength

because 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 to about

2000 mg/L COD, which held steady through the January to March 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 October 1984 represents a transitional phase of leachate

quality (which will be mentioned again in later discussions).

Moderate rainfall during the Fall of 1984 caused the leachate strength to

vary between 150 and 350 mg/L COD, with a slight increasing trend as the field

capacity was re-established after a dry Summer. Then the leachate strength increased

sharply in response to a heavy week of rain in December 1984, indicating a

washout condition like that observed by Jasper ef al. (41). However, the landfill

recovered very quickly once the normal hydraulic regime was resumed (Figure 5.1C).

For convenience and clarity the data for the various major leachate parameters are

presented separately in subsequent Figures (5.3 - 5.5 A,B,C). In addition, the raw

data from the analyses of this leachate is included in Appendix 1.

Z2/Z2 * S A l ua juo^ uoqjC3 a i e i p e a i v€'S ajnSiJ

COD, B O D 5 , & TOC (mg/L)

LEACHATE CARBON CONTENT vs. TIME and PRECIPITATION

2500 - i

2000

1500 H

1000 H

500

£ 30 u

z o

«C 20 CL O CC Q. 10

0 N 1983

0 N 1983

J J U

D J F 1 9 8 4

hi J U M A M J J A S O

Legend A C O D

X BODp;

• T O C

LEACHATE CARBON CONTENT vs. TIME and PRECIPITATION

1500 - i

e r a c

n

n o 3" 01

n t cr o 3 o O 3 re 3 < in H 3'

00 00 VI

— 1000 LO

Q O CQ 66

Q O CJ

5 0 0 H

E 30 o z o < 20

UJ

Legend A COD

X BODfi

1 1 1 1 1 1 1 1 i i

O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L 1 9 8 4 1 9 8 5

JL ± • i l l - , i I I i • I 1 . • • I -f O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L 1 9 8 4 1 9 8 5

00

LEACHATE NITROGEN CONTENT vs. TIME and PRECIPITATION

60 -i

1982 1 9 8 3


60 -i

CD E BOH

^ 40 H

y£ 30

< 20

i ( H

E 3 0 (J z o

u a.

IXI

1 0 H

1 r~ 0 N

1983

0 N 1983

i r

D J F 1 9 8 4

M M

1 D J F

1 9 8 4

JUL 111 J LL

M A M

o

0

Legend

X TKN


Legend X Nr-h

X TKN

J ! ! ! , j ! , , , ,

O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L 1 1 9 8 4 1 9 8 5

1

r -1 1 | 1 • l| i . I I I r i I I I l i 1 1 . • lrl r -O C T N O V D E C J A N F E B M A R A P R M A Y J U N J U L 1 9 8 4 1 9 8 5

E 4500-] o

co 4000-

a a 3600-

3000-

-6 c 2500-o u u 2000-Cu

CO <3 1600-co T> 1000-o

CO 600-" T O

n t— u •

z o <t 20

u UJ <r CL 10

Total Solids & Specific Conductance vs. TIME and PRECIPITATION

i i i i i i : i i i i i i O N D J F M A M J J A S O

1982 1 9 8 3

n r 1 1 ^ 1. 11 -1 1. • • I I i O N D J F

1982 1 9 8 3

n— — r i — i M A M J J A S O

Legend A Tot. Solids

x Sp. Cond.

0 0

c

In 00

to tu O sr

-4 O

l/l

ai 3 Q. C/l

13

o o 3 Q. C n a) 3 o fD

3 CO OJ CO

E o co 3

3000-1

e3 2500

CD S

-6 c o U CL

c3 CO

T J — I O O O H o CO ro

2000H

1500

500

J 3 0

z o !< 20-1

UI

UJ UJ 5

1 r-

0 N 1983

0 N 1983


D J F 1 9 8 4

M M

JjJ-4 i i J U L LLI D J F

1 9 8 4 M M


x Sp. Cond. i

0

o CO



X Sp. Cond.

85

5.4 ORGANICS

Figures 5.3 A,B,C, present the leachate COD, BOD 5 , and TOC data. These

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 COD 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 COD, 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 COD, as expected. A similar close

correlation is apparent between BOD,- and COD (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

COD 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 BOD- values. The primary reason for

86

TOC versus COD

1000-1

800-

O) 600-E

^ 400-1 r -

200-

Legend A 10/82 to 6/83

X 7/83 to 12/83

• 1/84 to 6/84 i i i i

500 1000 1500 2000 COD (mg/L)

2500 3000

Figure 5.6 TOC vs. COD

Linear Regression Results

Data Group Slope Y intercept Correlation Coefficient

No. of Data Points

10/82 to 6/83 7/83 to 12/83 1/84 to 6/84

10/82 to 6/84

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 5 versus COD 1 5 0 0 n

1 0 0 0 -CO E

Q O CO 5 0 0 -

A A

A A

A

A A

5 0 0 1 0 0 0 1500 2 0 0 0 COD (mg/L)

Figure 5.7 B O D . vs. C O D


Legend A 7/83 to 12/83 i

j X 1/84 to 6/84 I • 7/84 to 12/84 |

B 1/85 to 6/85 2 5 0 0

Data Croup Slope Y intercept Correlation Coefficient

No. of Data Points

7/83 to 12/83 1/84 to 6/84

7/84 to 12/84 1/85 to 6/85

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

88

1 5 0 0 - .

1 0 0 0 -

£

LO Q 2 5 0 0 CO

B0D5 vs. BOD/COD Ratio

A A A

A A

m A

• •

0.25 0 .50 0.75 1 BOD/COD Ratio

1.25

Legend A 7/83 to 12 /83

X 1/84 to 6 / 8 4

• 6 / 8 4 to 12 /84

E 1/85 to 6 / 8 5

1.50

Figure 5.8 B O D . vs. B O D . / C O D Ratio

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%, but would not alter the

BOD/COD ratio. A second difference is the higher BOD/COD ratios observed at low

BOD concentrations. These are probably attributable to the ammonia oxidation

mentioned above. The abnormally high ratios skew the plot to the right at the

lower levels, which explains the otherwise unlikely results in which the BOD/COD

ratio is >0.8, let alone >1.0. Once these two factors are considered, the

BOD/COD 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 and BOD^ results and

vice versa. Figure 5.9 shows the variation in the VFA concentration over the period

for which they were monitored. This period covers the transistion to the

methanogenic phase as indicated by the steady decline in concentration from March

1984 to October 1984. The wash-out of VFAs during the Fall and Winter of

1984-85 is also demonstrated. Figure 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 BOD^

values (Figures 5.11 & 5.12), also show that high C O D and BOD^ levels are due

in large part to the VFA contribution. The regression results indicate some scatter in

the data (particularly for BOD,-, as might be expected), but still show a reasonably

strong linearity. Therefore, this data conforms to the experience of other studies

which show that the organic strength of a leachate is largely determined by the

fate of the VFAs produced during the decomposition of the wastes (32).

10000q

100CH

10CH

Volatile Fatty Acid Concentration vs. Time

M A M J J A S 1984

D J F M A 1985

Legend A ACETIC

X PROPIONIC

• BUTYRIC

B Total VFA

VFA Theoretical COD vs. Leachate COD and BOD 5

2000 -1

1984 1985

92

2000 n

COD versus VFA

A A

500 1000 1500 VFA (mg/L)

Figure 5.11 C O D vs. VFA



X 7/84 to 12/84

• 1/85 to 3/85

2000


No. of Data Points

3/84 to 6/84 0.9788 450.8 0.8672 24 7/84 to 12/84 1.8214 114.3 0.9907 48 1/85 to 3/85 1.6303 139.2 0.9643 22

93

600 n

B 0 D 5 v e r s u s V F A

A A

A

A

200 400 VFA (mg/L)

A


X 7/84 to 12/84

• 1/85 to 3/85

600

Figure 5.12 B O D 5 vs. VFA


Data Group Slope Y intercept Correlation No. of Data Coefficient Points

3/84 to 6/84 0.7158 172.6 0.8624 10 7/84 to 12/84 1.2234 75.43 0.8539 42 1/85 to 3/85 1.7396 20.19 0.9110 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 COD 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 COD 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 AND SPECIFIC CONDUCTANCE

The results for Total Solids and Specific Conductance were closely related to

each other and varied linearly with the other parameters (recall Fig. 5.1 A,B,C).

Figure 5.15 and the associated linear regression results, show more clearly the

correlation between these two parameters. The close correlation between total solids

and specific conductance was due largely to the very low suspended solids content

of the leachate, typically less than 5% (<75 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 by

specific conductance. Periodically, in response to a sudden change 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 increased

the leachate suspended solids. The sandy soil layers beneath the wastes, through

which the leachate must flow to reach to collector pipe, appear to filter most

suspended solids out of the leachate.

Figures 5.5 A,B,C, show quite clearly that a change takes place in the nature

of the leachate with the onset of the methanogenic activity. Prior to March 1984,

the numerical value of T.S. and Specific Conductance were almost identical.

Beginning in March 1984, the T.S. value decreased with respect to the Sp. Cond.

value until October 1984, when a new steady relationship is established. This is

shown graphically in Figure 5.15, and numerically by the linear regression data. The

figure and regression data show that the January to June period of 1984 was a

transition period in which the slope of the relationship shifted downwards. A

reduction in the total solids level can be attributed to the reduction in dissolved

96

N H 3 versus COD

D)

CO

1000 2000 COD

3000 (mg/L)


X 7/83 to 12/83

• 1/84 to 6/84

B 7/84 to 12/84

S 1/85 to 6/85

4000 5000

Figure 5.13 NH„ vs. C O D



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

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 CO x

Z 20-1


X 7/83 to 12/83

• 1/84 to 6/84

H 7/84 to 12/84

ffi 1/85 to 6/85

1000 2000 3000 Specific Conductance (nS/cm)

i 4000

Figure 5.14 N H 3 vs. Sp. Cond.



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

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. Specif ic Conductance

4 0 0 0 -i

3 0 0 0 -

E

^ 2 0 0 0

O if)

"co n 1000

X

A A

Legend A 10/82 to 6 / 8 3

X 7/83 to 12 /83

• 1/84 to 6 / 8 4

Kl 7 /84 to 12 /84

ffi 1/85 to 6 / 8 5

1000 2 0 0 0 3 0 0 0

Specif ic Conductance (i|S/cm)

i 4 0 0 0

Figure 5.15 Tot. Solids vs. Sp. Cond.


Data Group Slope Y intercept Correlation Coefficient

No. of Data Points

10/82 to 6/83 1.0847 7/83 to 12/83 0.9569 1/84 to 6/84 1.1241

7/84 to 12/84 0.7374 1/85 to 6/85 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

1 0 0

COD vs. Specific Conductance 5 0 0 0-1

1 0 0 0 2 0 0 0 3 0 0 0

Specific Conductance (ujS/cm)

Legend A 10 /82 to 6 / 8 3

X 7/83 to 12/83

• 1/84 to 6 / 8 4

R 7 /84 to 12/84

m 1/85 to 6 / 8 5

4 0 0 0

Figure 5.16 C O D vs. Sp. Cond.



No. of Data Points

10/82 to 6/83 1.3281 7/83 to 12/83 0.9397 1/84 to 6/84 1.0658

7/84 to 12/84 0.4050 1/85 to 6/85 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

1 0 0 0 -

E

LO Q O 5 0 0 CQ

A A

A ^ A

5 0 0 1 0 0 0 1500 2 0 0 0

Specific Conductance diS/cm)

A

Legend A 7 /83 to 12 /83

X .1/84 to 6 / 8 4

• 7 /84 to 1 2 / 8 4

1/85 to 6 / 8 5

2 5 0 0

Figure 5.17 B O D . vs. Sp. Cond.


Data Croup Slope Y intercept Correlation No. of Data Coefficient Points

7/83 to 12/83 0.6561 -295.2 0.9548 16 1/84 to 6/84 0.2573 -90.17 0.7294 11

7/84 to 12/84 0.1716 -81.49 0.5933 42 1/85 to 6/85 0.2128 -180.2 0.7291 12

Table 5.3 Leachate Heavy Metal Levels (AA)

Leachate Samples from the North Leachate Lift Station (Well #1), Premier St. Landfill

June*1) May 11 June 22 July 17 Oct 19 Nov 30 Dec 21 Jan 1 Feb 15 May 10 83 84 84 84 84 84 84 85 85 85 j .

COD 3520 1527 460 377 138 264 1352 434 155 126

Ca 265 Cd 0.13 Cr <0.02 0.0068

Cu 0.4 0.026 0.013 0.023 0.057 0.019 0.033 0.047 0.010 0.01 1

Fe 185 60.7 63.2 53.5 31.6 18.6 27.7 31.8 22.4 23.9

Mg 49.0 37.3 27.0 27.9 16.2 34.1 28.4 15.2 20.3

Mn 5.98 5.18 4.28 3.54 1.77 3.68 3.30 2.24 2.30

Ni 0.01 13 Pb <0.02 0.0036 Zn 0.420 0.420 0.124 0.293 0.218 0.1 10 0.476 2.93 0.082 0.1 13

(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)

Element (mg/L)

.CP. Metal Scan of Premier Leachate Samples

5/11/84 7/17/84 1/21/85 5/10/85

As B Ba Be Cd

Co Cr Cu Mn Mo

Ni P Pb Sb Se

Sn Sr Ti V Zn

Al Fe Si Ca Mg Na

Hardness Ca, Mg Total

<0.05 0.704 0.159 <0.001 < 0.002

0.168 0.014 0.018 4.83 <0.005

<0.02 0.38 <0.02 <0.05 0.09

<0.01 1.23 0.093 0.007 0.33

I. 85 57.4 II. 7 291.0 37.4 109.0

880 1000

<0.05 0.454 0.036 <0.001 <0.002

0.141 0.008 0.015 3.7

< 0.005

<0.02 0.05 <0.02 <0.05 0.08

<0.01 0.777 0.028 0.006 0.236

0.08 48.5 8.2 138.0 28.0 84.4

460 556

<0.05 0.461 0.069 <0.001 < 0.002

0.096 0.01 0.039 2.86 <0.005 <0.02 0.64 <0.02 <0.05 0.07

<0.01 0.804 0.03 0.006 2.4

0.15 32.9 8.1 162.0 28.4 83.9

522 592

<0.05 0.342 0.138 < 0.001 <0.002

0.058 <0.005 < 0.005 2.01

< 0.005

<0.02 0.12 <0.02 <0.05 0.05

<0.01 0.528 0.026 < 0.005 0.092

<0.05 22.0 7.3 81.4 20.3 67.6

287 287

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).

Table 5.5 Leachate Trace Organic Content

105

Premier (Well #1)

Compounds (ppb) 7/17/84 11/30/84

* Benzene 13.1 2.80 * Toluene 385.0 84.80 * Ethylbenzene 13.0 6.64 * Chlorobenzene 1.0 -* Dichlorobenzene - -m - Xylene 24.4 16.21 o & p Xylenes 20.8 18.99 1 - methylethyl benzene -n - propylbenzene Trace Trace 1,3,5 - Trimethyl benzene 1.6 n - Butylbenzene - Trace

* Compounds on the EPA list of priority pollutants (-) = Not Detected Trace = <1 ppb

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. A small excavation was made to sink the

Figure 6.1 Photo of RBC Prior to Start-up, Showing Disk Media and Influent Pump

I

106

107

Table 6.1 RBC Specifications

Make & Model model S5, CMS Equipment Ltd. Disk Diameter 0.9 m No. of Stages 4 No. of Disks 36 (arranged 15,7,7,7) Disk Area 47 m 2

Disk Zone Volume (net) 245 L Surface to Volume Ratio 190:1 m 2 /m 3

Rotational Speed 6 rpm Peripheral Speed 0.29 m/s

Figure 6.2 Photo of RBC Installed Adjacent to the North Leachate Lift Station

108

Finished Ground Elev. 28.50 M ink

RBC Influent And Effluent Return Lines

Lag Pump O n 25.60

Leachate Col lect ion Pipe Inv. 25.00

Manhole Lid Elev. 29.02

All Pumps Off 24.70

Sump Elev. 24.40

Check Valve

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 Mounted Beside Pipe Tee) Elev. 27.00

Check Valve

Location o( Inlet Screen During Initial Phase of Experiment

Flygt Submersible Leachate Pumps

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 influent leachate characteristics due to

the relatively tiny volume pumped through the RBC, and the spacial separation of

the inlet and return lines within the wet well (see Figure 6.3).

During the previous few months of planning and preparation for this study,

the leachate was quite strong (recall Figure 5.1A). The C O D varied between 2000

and 4000 mg/L. It was assumed that the C O D would not average less than 1500

mg/L over the next 6 - 8 months (the anticipated study period), so that a

maximum flowrate of about 750 mL/min. (1080 L/day) would be adequate to

overload the RBC. The corresponding loading rate is approximately 34.5 g

COD/m 2 * d or 23.3 g BOD/m 2 *d. On this basis a Masterflex 1" peristaltic pump fitted

with a no. 1717 pumphead, rated for up to 1680 mL/min. flow, was installed in

the RBC to pump the leachate feed up from the lift station wet well (see Figure

6.1). The required suction lift was approximately 3.4 m (see Figure 6.3). An inlet

screen was placed on the end of the leachate influent line to help prevent solids

from plugging it (Figure 6.4).

This leachate, like most others, is nutrient deficient (15,16), particularly of

phosphorus (P). Therefore, a solution of ammonium chloride NH^Cl, and phosphoric

acid H ^ P 0 4 , was added to the first stage of the RBC. The solution was initially

added via a gravity fed drip system from a constant head reservoir, but later a

pump was employed. Over the course of the study, the concentration and flowrate

of the nutrient solution varied, (the NH 4 C I addition was later stopped), but the

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 good bacterial growth. This

level of nutrient addition is particularly generous with respect to nitrogen, in light

of the findings of other studies conducted here at UBC (62,79,81,86), which found

the minimum nutrient ratio for leachate treatment to be 100:3.2:1.1. A preliminary

T Reg. TM, Cole-Parmer Instrument Company, 7425 North Oak Park Ave., Chicago, Illinois, 60648.

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 OPERATION

7.1 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 observed in sewage treatment (57), is generally

augmented 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 because

112

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 Photo of mature biomass growth during start-up (late June 1983)

manufacturers specifications 1. Yet in this instance, at about 60 rpm, the tubing often

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 pump

head). Numerous adjustments such as changing the tubing completely each visit,

using a different type of tubing, and using a new pumphead, were unsuccessful.

Installing two pumpheads in parallel to reduce the rotational speed only doubled

the frequency of tubing failures, and caused a second problem when the tubing

became knotted up inside the pumphead and jammed the pump.

The cause of these problems remains unclear. According to the tubing

compatability data provided by the manufacturer, silicone tubing is sensitive to some

substances found in the leachate such a toluene, but these materials are present in

only trace amounts. This would also fail to explain the problems with other types

T 1985 - 86 Catalog, Cole-Parmer Instrument Company, 7425 North Oak Park Ave., Chicago, Illinois, 60648.

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 is rated for a maximum flow of 1730 mL/min. and was installed in the RBC

on November 10, 1983. A much smaller no. 1713 pumphead was then mounted on

the Masterflex pump, which was relegated to dispensing the nutrient solution.

Initially the bellows pump also had problems with broken valves, and a collapsed

inlet line due to suction. These problems 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 on the inlet line. With these modifications the bellows pump

performed very well. The valve springs in particular should be recommended for use

whenever these pumps are used with the applicable poppet-valves.

7.2 THE DISRUPTIONS

Scarcely a week after the new bellows pump was installed, the first mishap

of what was to be a series of three major interruptions occurred. An unusually

heavy rainstorm during the week of November 18, 1983, completely overwhelmed

the leachate lift station and the resulting pond flooded out the RBC. Figure 7.4

shows part of the gooey aftermath of this flood. (Notice the high tide level of

mud on the electrical cord). The high water level was about 16 in. (40.6 cm)

above the normal water level in the RBC, and just short of the disk drive motor.

As the drive motor did not stop, oil washed out of the oil bath for the drive

chain was whipped up into a frothy grey emulsion, which along with the

considerable amount of mud washed into the RBC, coated everything. Both the

bellows leachate pump and the Masterflex chemical pump were stopped, but the

Masterflex sustained the most serious damage as both the speed controller and the

motor windings were shorted out. The one bright spot of this event was that the

shaded-pole type motor of the bellows pump was not damaged by the dunking

and only required a thorough 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 Photo of Aftermath of 1 s t F lood in the RBC (November 1983)

I

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 Photo of RBC after being raised 1m to avoid f lood ing

I

119

One of the vent covers was pried off and the drive chain derailed, which stopped

the disk, and 60% of the biomass was partially dried out. The remaining 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. By the beginning of April however, the RBC

was 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 throughout

the disruptions, (October 1983 through March 1984). The observed influent flow

rate was that measured upon 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 operated continuously

except for one minor interruption of the leachate flow, caused by a fouled

check-valve. During the first six weeks of this period the biomass re-established

itself. The new biomass grew very rapidly over the dried mat of previous growth

and in the first stage particularly, the new growth was very thick and shaggy. This

heavy regrowth of the biomass was no doubt encouraged by the relatively high

organic loading applied (averaging 14.5 g COD/m 2), and the warmer spring

temperatures. The rough growth periodically sloughed off in large chunks, giving the

RBC Operational History: Influent Rowrate and Loading

800-1

600 H

2 400 H

200 H

a o o

3 10 17 24 31 7 14 OCTOBER NOVEMBER

1983

i ) < ! » < ) < X i ani 21 28 S 12 19 26 DECEMBER

2 9 JANUARY

1984 FEBRUARY

rr 20 27 6 12 MARCH

Legend X Observed Rate

$ Reset Rate

• Average Rate

CM E » CD +-» (0

CC

•o

36 30 25 20-16-10-5-0 i " i

Legend • COD Ldg.

r — r - i 1

RBC Operational History: Influent Flowrate and Loading

2000

1600-

2 I O O O -

500 4> X X X

x- r u X

A / #

x x X

4 11

APRIL 1984

-1 1 1 r 18 28 2 0

MAY

T

18 23 30

JUNE

iX i 1 r 8 13 20 27

T—

JULY

X X X 18 28 1 8

AUGUST

— i 1 r IB 22 29 8 12

SEPTEMBER

.XX)00O<

Legend X- Observed Rate $ Reset Rate • Average Rate

18 28 3 OCTOBER

C M

20-

16-

® 10 co

CC

•a D EiB

Legend • COD Ldg. E 2 BOD Ldg. •1 NH 3 Ldg.

RBC Operational History: Influent Flowrate and Loading

1600

C E

E 1000 H

© CO

o ^ 5 0 0 c CD 3

Legend X Observed Rate

$ Reset Rate

• Average Rate

2 9 16 OCTOBER

1984

23 30 0 13 20 27 4 11 18 28 1 8 18 22 29 8 12 NOVEMBER DECEMBER JANUARY FEBRUARY

1985

19 28 8 12 MARCH

19 28 APRIL

CM .E

CD CO cc ti> •a

35-30-25-20 H 15-10 5-0 - i r -i r

Legend CD C O D Ldg. E2 BOD Ldg. •I N H 3 Ldg.

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 of

biomass and finally, the complete sloughing of the rough growth. It is generally

viewed that one mechanism for biomass sloughing is reduced adhesion between

cells in the anaerobic layer. In this case it appears as though the anaerobic layer

gradually broke down until ultimately it came unglued completely. The dark colour

of the biomass during this period was probably due in part to depressed dissolved

oxygen levels, because of the heavy organic loading on the first stage and the

greater oxygen demand of the thick biomass, and in part from the dark colour of

the thick anaerobic layer showing through.

The biomass which replaced the rough growth was much thinner, but of

uniform thickness 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, 1984.

This was indicated by poor floe settleability, and thus higher effluent suspended

solids, due to the presence of fluffy filamentous floe particles. The settleability

problem cleared itself up by June 19 and the RBC operated extremely well through

July 24, 1984.

This period of continous operation took place during the previously

mentioned transition phase of the leachate quality and therefore the organic strength

was decreasing. Column B of Table 5.2 shows the range of leachate composition

over this period. Although the leachate pumping rate was increased, the organic

loading rate decreased steadily from the 14.5 g COD/m 2 *d of April - May, to an

average of 7.7 g COD/m 2 * d during July. This is shown graphically in Figure 7.6B. It

can be seen that the influent flowrate was maintained much more consistently than

during the previous period. The declining leachate strength, while frustrating the

desired increases 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 unmanageable, such that

on three successive site visits the inlet line was choked off completely. In response

to this intermittent flow and loading, a large proportion of the biomass was ejected

from the disks. Following this loss of solids the RBC operated fairly steadily from

August 8 to September 4, 1984. Despite flow rates around 1 L/min., the loading

rates were less than 5 g COD/m 2 * d and would only support a relatively thin, but

healthy biomass. After September 4, another series of minor pump problems and

periodic fouling caused numerous stoppages.

This biological fouling problem, which did not appear during the previous

year, was observed to increase from May onwards, as the leachate strength

declined. It appears this problem arose because as the leachate strength declined,

the leachate in the wet well, and particularly in the intermediate bucket, was able

to become increasingly aerobic. Aerobic conditions, as well as 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 visit, when both the inlet screen and inlet lines were

thoroughly cleaned. Aside from the rate of growth, aerobic conditions were also

indicated by the light rust brown colour of the growth on the screen, which

appeared very similar to that of the first stage growth on the RBC. This colour

indicates that metal precipitates (mostly iron oxides) were also adding to the fouling

problem. A single sample of inlet line deposits analysed for metal content was

found to be 38.9% iron on a dry weight basis. While constant cleaning of the

inlet line was bothersome, 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 COD 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 recorded loading

occurred on December 21, 1984. At a loading of 32.7 g COD/m 2 * d , it was

observed that there was considerable foaming in the first stage and a thin white

growth covered the surface of the biomass in the first and second stages. Although

the 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 in large chunks. This growth gradually

thinned out over the next two weeks, but it provided a thick, healthy, biomass for

the start of the next period of relatively stable operation.

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 7.6c

shows the influent flowrate variation over this period. Since the C O D of the

leachate was generally less than 270 mg/L, the carbon loading rate was quite low

(less than 10 g COD/m 2 * d or 5.0 g BODg/m 2*d). However, since the ammonia

levels in the leachate and the influent flowrates remained high, the ammonia loading

rates were highest during this period. The effective loading rates for nitrification

were possibly even higher when temperature effects were taken into account. This

aspect will be discussed further in Section 9.2. Therefore, the nitrification

performance of the RBC during this period is of particular interest.

The biomass was fairly thick and healthy looking during this period, as it

had been for most of the study. Figures 7.10 and 7.11 show the colour colour

gradation, and thickness, of typical healthy growth. Foaming was never a problem

with this leachate. Figure 7.12 shows an above average foaming condition. The

heavier foaming incidents during high loading had, at most, 6 in. of foam in the

first stage, with much 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 RBC

unit was two periods of two or three months continuous operation, and 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. Appendix 2 contains

a listing of the RBC operational history and field observations. For the most part,

the RBC 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 good 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 from a sample of the

fourth stage liquid settled for 30 minutes in a 1 L graduated cylinder. Effluent from

the final clarifier zone of the RBC was not representative of the RBCs performance,

as settled solids were not removed, and considerable resolubilization of organics

occurred in this zone.

The main sampling program occurred from May 25, 1984, to March, 1985.

Samples of the influent leachate, and first and fourth stage liquid, were taken on a

two or three times a week basis, except during December, when samples were

taken only once a week. During the previous period of operation, from July 1,

1983, to May 25, 1984, the data is less complete because the operation of the

RBC was not considered stable enough to warrant a complete sampling and analysis

program (recall section 7 RBC Operation ). Appendix 3 contains the raw data from

the analysis of the RBC process samples.

Loading rates were calculated from the measured flow rate and leachate

COD, or NHg-N, etc., concentration, at the time the sample was collected

(observed rate). This assumes that the flowrate and leachate strength were constant

over the previous detention time period of the RBC. At a flowrate of 1000

mL/min., the theoretical detention time is 245 minutes. In most cases, the leachate

strength probably did not vary significantly over this period of time, particularly since

most of the data was collected over the later half of the study when the short

term variability of the leachate strength was reduced. As the flowrate of the pump

131

132

was consistent, when it was running, these assumptions seem reasonable.

The main sampling program extended over both of the longer periods of

steady operation, as well as many shorter periods of continuous operation. Results

from the two longer periods of continuous operation demonstrate the abilities of

the RBC 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 B O D ^/m 2 * d , and their results are also

significant. A study by Filion ef al. (27), found that a RBC would recover to

steady-state conditions within about 1 hour for carbon removal alone, and 3 hours

for carbon removal with nitrification, in response to a step increase in influent

loading. Since 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 essentially at steady-state during these short periods also.

8.1 CARBON REMOVAL

The carbon removal efficiency of the RBC treating this leachate was very

good. An effluent soluble BOD^ of less than 10 mg/L was maintained for all but a

few samples and the settled effluent B O D ^ was generally less than 25 mg/L. Those

few samples which did have a soluble BODj- greater than 10 mg/L occurred after

the loading rate had more than doubled over the previous few days. However, not

every sharp change in loading was followed by a significant reduction in treatment.

On a percentage basis, the soluble BOD^ removal was usually above 95%. Most of

the lower removal percentages were caused by influent BODj. values so low that

the effluent BOD^ values were relatively large in comparison. Figure 8.1 shows the

variation of effluent BOD^ over the main sampling period. This figure also presents

the loading rates and percent removals calculated for this data.

1000q

CD E

LO Q O CO

100-

o L U cc 10 Q O 03

10-

100-,

90

80-

70

RBC EFFLUENT B0D 5 VARIATION with LOADING RATE and TIME

JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR 1984 1985

i r

JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR 1984 1985

r10

* CN

D)

-5 U J

!< DC CD

Q < O

Legend A INF BOD

X SET BOD

• FIL BOD

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 and

four filtered samples 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 BOD^ removal was

occurring primarily in the first stage, and that generally less than 10 mg/L of

additional BOD^ was removed across the remaining three stages.

The C O D results paralleled those of the BOD^ test with a few minor

differences. Effluent soluble C O D values 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 values

closely. This refractory component reduces the significance of the C O D removal

percentages because it accounted for up to 60% of the influent C O D when the

leachate was weak. Secondly, the range between the C O D values for the settled

and filtered samples was, in some cases, considerably greater than the corresponding

range of BOD^ values. These differences occurred when the effluent suspended

solids were relatively high and the volatile component of the solids was low (less

than 30%). This indicates that the settled effluent C O D contained a large refractory

component from the highly endogenous suspended solids. Figure 8.6 shows the

variation of effluent C O D with loading rate over time, and also presents the highly

variable percent removal data.

The C O D and B O D 5 data shows that the RBC could maintain efficient

carbon removal at loading rates up to .roughly 15 g COD/m 2 *d, or 9 g

B O D 5 / m 2 * d , treating this leachate. This rate of loading would be considered only

moderate for sewage treatment. C O D results for a couple of samples indicated that

the capacity of the RBC is probably considerably higher but there were no B O D 5

135

E F F L U E N T F 3 0 D 5 v s . L O A D I N G R A T E

40- i

CD 30 E

IO Q S 20

LU

U J

X X

o

o • X o

X X X , UJ UJ

X , -i cn v X £ U J

L T w X 1—1 X Ul ; J

«P u Legend Ei! 10J f j i ^ X 2 x 8 a ° • DATA PERIOD B U _ | X ^ D + B X J # + + ° • • X + OTHER DATA

• • DATA PERIOD C

2 4 6 ft 10 B0D5 LOADING RATE (g/m^d)

Figure 8.2 RBC Effluent B O D . vs. Loading Rate

1 S T & 4 ™ STAGE BOD5

1000

100H

1 }• LO Q O CQ

10

1

/ v. r

< 3 i f /

1-1 1 1 1 r JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR 1984 1985

APR

Legend A INF BOD X #4 SETTLED B #1 SETTLED

1 s t & 4 ™ S T A G E 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

200 - i X

+

—I 1 r-10 20 ^0

COD LOADING (g/m *d)

o a Ul Ul - i cr

o X •

40

Legend DATA PERIOD B OTHER DATA DATA PERIOD C

Figure 8.5 RBC Effluent COD vs. Loading Rate

RBC EFFLUENT COD VARIATION with LOADING RATE and TIME

lOOOOq

100CH

E

Q O CJ 100d

10 J

100-.

< > 80-

o UJ cc 60-Q O O 40-

N D J F 1985

140

results for comparison. The BODj. results proved to be a better indicator of the

treatment efficiency because increases in effluent COD 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 COD 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 COD 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

When the RBC was restarted at the end of March, 1984 (after the

disruptions), it took another 2 months until May 22, 1984, to start nitrifying again.

Complete nitrification was re-established by May 28, 1984. The NH^Cl addition had

been stopped, as it was no 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 period except for a few samples. In most

cases, an effluent NH^-N concentration of greater than 1.0 mg/L resulted from high

loading rates, or system upset due to an interruption of the leachate feed. A

number of samples from January to March, 1985, had effluent ammonia levels

greater than 5.0 mg/L, which indicated that the nitrification capacity of the RBC was

being exceeded. The loading rates were greater than 0.7 g NH.j-N/m 2*d at water

temperatures below 10° C. Thus the effective loading rates, corrected for

temperature, were probably somewhat greater. Nitrification was stopped briefly on a

few days in December 1984, when the water temperature fell below 5° C. During

these periods however, the effluent ammonia level was not observed to increase.

This point will be addressed further in the Discussion.

Figure 8.8 presents the effluent ammonia (NH^-N) and nitrate (NO^-N) levels

versus the NH^-N loading rate as a scattered plot. This figure shows a trend

towards increasing effluent ammonia levels with increasing loading rate. Effluent

ammonia levels above 1.0 mg/L generally occurred at loading rates greater than 1.0

g NH^-N/m^d. Figures 8.9 A,B, show that as with BOD^ removal, most of the

nitrification occurred in the first stage, with less than 10.0 mg/L of ammonia being

removed across the last three stages, except during cold weather or heavy loading.

This will be discussed further in Section 9.2.

RBC NITRIFICATION PERFORMANCE

E i CO O

+ CO

16 22 29 6 13 20 27 10 17 24 31 14 21 28

-o

E CO x z Uj 0.6 < CC O

z Q <

APR MAY 1984

JUN JUL AUG SEP

1 \ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 • • ( • - > Y | 1 1 — f 13 20 27 4 11 18 25 1 8 15 22 29 6 13 20 27 3 10 17 24 31 7 14 21 28

APR MAY JUN JUL AUG SEP 1984

RBC NITRIFICATION PERFORMANCE 6 0 - i

5 12 19 26 2 9 16 23 30 7 14 21 28 4 11 18 25 1 8 15 22 1 8 15 22 29 5 12 OCT NOV DEC J A N FEB M A R 1984 1985

144

EFFLUENT NHo & NOq -N vs. LOADING RATE

40 - i

— 30H X X X

X o X m § ( xo< °

X

X X>X0<p

x*< 9x 3

• x D X D

x m + •

<

o x •

~l ^ I 1 1.2 1.4

N H 3 -N LOADING RATE (g/m^d)

z o 2

< Legend DATA PERIOD 6

OTHER DATA

DATA PERIOD C

Figure 8.8 Effluent NHL and N O - vs. Loading Rate

APRIL MAY JUNE JULY AUGUST SEPTEMBER 1984

1 ° 1 & 4 1 n S T A G E N H 3 & N 0 3 -N

8.3 SUSPENDED SOLIDS

In general, the suspended solids levels in the RBC were quite low, <200

mg/L. Figure 8.10 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 solids

were generally concentrated in clumps 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 clarifier.

The suspended solids level in the RBC disk zone was observed to fluctuate

more during the periods of low organic loading (<3 g BOD^/m 2*d). Under these

conditions, much 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 30% 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 20%, to a

high of just over 80%, depending upon the organic loading rate.

8.4 METALS

The results of the metal analyses, Tables 8.1 and 8.2, indicate that the RBC

generally removed over 80% of the iron (Fe), manganese (Mn), and zinc (Zn), as

well as 50% of the copper (Cu), and lead (Pb), and lesser amounts of other

metals. Results from the analysis of a few samples of biomass scraped from the

RBC disks, Tables 8.3 and 8.4, indicates that the removed metals are concentrated

f 1 & 4 x n S T A G E S U S P E N D E D S O L I D S

100003

Q _ | n u n l l u r i H l l l l H u n n u i i n u | in m i nu nrj H II | n H nuiln f nil l in U ip i l H ( n n null nil I in u nil M J I I I M IIII nil lin |in nu ,

MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY 1984 1985

RBC EFFLUENT SUSPENDED SOLIDS _ I O O O O 3

0 J 1 1 1 1 1 1 1 1 1 1 1 1 r M J J A S O N D J F M A M

1984 1985

1 5 0

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)

Sample (mg/L) Cu Mn Mg Fe Zn

Well #1 May 11,1984 0.0263 5.98 37.24 60.7 0.420 1 s t Stage 0.0495 3.02 26.54 54.1 0.310 4 t h Stage 0.0727 2.76 29.21 60.0 0.333 % Removal -176.4 53.9 21.6 1.1 20.7

Well #1 June 22,1984 0.0130 5.18 27.0 63.2 0.1235 1 s t Stage 0.0256 2.19 23.93 3.65 0.0628 4 t h Stage 0.0088 1.61 23.68 1.56 0.0354 % Removal 32.3 68.9 14.0 97.5 71.3

Well #1 July 17,1984 0.0225 4.28 27.88 53.5 0.293 1 s t Stage 0.0085 1.29 22.49 6.25 0.392 4 t h Stage 0.0119 0.711 24.53 3.62 0.0398 % Removal 47.1 83.4 12.0 93.2 86.4

Well #1 Oct. 19,1984 0.0571 3.54 16.18 31.64 0.2184 1 s t Stage 0.0546 0.848 15.66 5.09 0.0531 4 t h Stage 0.0284 0.252 15.18 2.32 0.0514 % Removal 50.3 92.9 6.2 92.7 76.5

Well #1 Dec. 21,1984 0.0333 3.68 34.05 27.73 0.476 1 s t Stage (set) 0.0988 3.81 26.60 10.59 0.140 4 t h Stage (set) 0.1099 2.36 25.52 8.82 0.108 % Removal -230 35.9 25.1 68.2 77.3

Well #1 Feb. 15,1985 0.0104 2.24 15.18 22.35 0.0820 1 s t Stage (set) 0.0248 0.808 13.80 4.68 0.0568 4 t h Stage (set) 0.0263 0.244 11.55 2.03 0.0478 % Removal -153 89.1 23.9 90.9 41.7

Well #1 May 10,1985 0.0906 2.57 20.92 25.0 0.1662 1 s t Stage (set) 0.0131 0.75 12.44 5.07 0.024 4 t h Stage (set) 0.0126 1.13 12.64 5.92 0.0745 % Removal 86.1 56.0 39.6 76.3 55.2

Average of 4 best 53.9 83.6 27.5 93.6 77.9 % Removals

note: Lead (Pb) levels for all sampli es <10. ppb

152

Table 8.2 RBC Metal Removal (ICP)

I.CP. Metal Scan of Leachate and RBC Samples

Element Well #1 1st 4th % Well #1 1st 4th % (mg/L) 7/17/84 Stage Stage Removal 5/10/85 Stage Stage Removal

As <0.05 <0.05 <0.05 B 0.454 0.357 0.371 Ba 0.036 0.037 0.039 Be <0.001 <0.001 <0.001 Cd <0.002 < 0.002 <0.002

Co 0.141 0.022 0.011 Cr 0.008 < 0.005 <0.005 Cu 0.015 0.009 0.008 Mn 3.70 1.39 0.689 Mo <0.005 <0.005 <0.005

Ni <0.02 <0.02 <0.02 P 0.05 1.10 0.73 Pb <0.02 <0.02 <0.02 Sb <0.05 <0.05 <0.05 Se 0.08 <0.05 <0.05

Sn <0.01 <0.01 <0.01 Sr 0.777 0.61 0.598 Ti 0.028 0.022 0.024 V 0.006 < 0.005 < 0.005 Zn 0.236 0.02 0.02

Al 0.08 <0.05 0.14 Fe 48.5 6.96 3.51 Si 8.2 6.0 5.9 Ca 138.0 114.0 111.0 Mg 28.0 23.4 23.5 Na 84.4 72.9 75.2

Hardness Ca,Mg 460 382 374 Total 556 398 383

<0.05 <0.05 <0.05 18.3 0.342 0.255 0.258 24.6

0.138 0.183 0.098 29.0 < 0.001 < 0.001 < 0.001 <0.002 < 0.002 <0.002

92.2 0.058 0.023 0.027 53.5 < 0.005 < 0.005 <0.005

46.7 < 0.005 0.008 0.008 81.4 2.01 1.23 1.54 23.4

<0.005 < 0.005 < 0.005

<0.02 <0.02 <0.02 0.12 3.71 3.58 <0.02 <0.02 <0.02 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05

<0.01 <0.01 <0.01 23.0 0.528 0.35 0.348 34.1

0.026 0.028 0.030 <0.005 < 0.005 < 0.005

91.5 0.092 0.022 0.024 73.9

<0.05 0.15 0.17 92.8 22.0 7.35 8.44 61.6 28.1 7.3 6.3 6.5 11.0 19.6 81.4 58.4 59.6 26.8 16.1 20.3 12.1 12.3 39.4 10.9 67.6 51.3 51.3 23.4

287 196 199 331 212 219

153

Table 8.3 RBC Biomass Metal Levels (AA)

Sample (Mg/g) Cu Mn Mg Fe Pb Zn

<10.0 184.5 793.6 388780 11.3 96.5

27.9 1021.3 5230.8 255729 21.3 418.5 37.3 1824.9 14552.1 178821 42.6 1077.3 5.0 1316.7 22083.3 116667 24.0 596.7 13.2 1154.2 17121.9 57961 18.3 433.0

22.7 849.4 5696.2 159893 4.3 2901.4 10.0 869.7 19326.9 65478 5.7 4148.6

20.3 1584.2 20342.1 212554 24.6 302.2 45.7 1780.8 54206.8 108746 30.4 651.8

Inlet Line Deposits

July 17/84 2nd 3rd 4 t h

Stage Stage Stage Stage

Oct. 5/84 1 s t Stage 4 t h Stage

May 10/85 1 s t

1

4 t h Stage Stage

154

Table 8.4 RBC Biomass Metal Levels (ICP)

I.CP. Scan of Selected Biomass Samples

Element (Mg/g) 1st Stage 4th Stage 1st Stage 4th Stage 7/17/84 7/17/84 5/10/85 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 Mo Ni

16. 18.

4940. <8. <30.

32. 32.

13000. <8. <30.

21. 24.

18000. <8. <30.

34. 30.

13300. <8. <30.

P Pb Sn Sr Ti V Zn Al Fe Si Ca Mg Na

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.

155

Table 8.5 RBC Trace Organic Removal

Premier (Well #1) RBC Effluent Compounds (ppb)

7/17/84 11/30/84 7/17/84 7/20/84 11/30/84

* Benzene * Toluene * Ethylbenzene * Chlorobenzene * Dichlorobenzene m - Xylene 0 & p Xylenes 1 - Methylethyl benzene n - Propylbenzene 1,3,5 - Trimetyl benzene n - Butylbenzene

13.1 2.80 - - -385.0 84.80 - - 1.24 13.0 6.64 1.0 . . .

24.4 16.21 20.8 18.99

Trace Trace - Trace 1.6 - -

Trace

* Compounds on the EPA list of priority pollutants (-) = Not Detected Trace = <1 ppb

9. DISCUSSION

The results of this study, presented in the previous section, showed that an

RBC could effectively treat this landfill leachate to remove degradable carbonaceous

and nitrogenous material, heavy metals, and some trace organic compounds. This

treatment was 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 REMOVAL

As presented previously, the carbon removal efficiency, as represented by

BODj, 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 stringent

provincial requirement of 30 mg/L BOD^ (19). These effluent values were achieved

at organic loading rates ranging up to 8.6 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 and presents data showing most

plants were operating at loadings between 10 and 15 g BOD^/m^*d, to a

maximum of 19.5 g BOD^/m^*d. Paolini et al. (58), while investigating kinetic

response of RBCs, tried loadings between 8.7 and 39.5 g BODr/m 2*d and found

156

157

the effluent quality started to deteriorate above 19 g BODg/m 2 *d. Murphy and

Wilson (54), operated a 2.0 m diameter pilot scale plant at between 4 and 36 g

BOD^/m 2 *d loading. They found a fairly linear response between BOD removal and

loading up to a loading of about 15 g BODg/m 2 *d, after which the removal rate

tended towards a maximum. Their recommendations were for a maximum design

loading of 17 g BOD,j/m 2 *d, at temperatures above 15° C, to achieve an effluent

B O D 5 <30 mg/L.

In the only comparable leachate treatment study which could be found in

the literature (with respect to carbon removal), Coulter (16) reports the results of a

companion study to this one conducted in Montreal. They were treating a leachate

with a BODg of 850 mg/L and ammonia levels of 60 mg/L. At their highest organic

loading of 11.3 g B O D j / m 2 * d , they achieved an average effluent B O D j of 19

mg/L, or 97% removal. At this level, first stage loading was nearing capacity as

dissolved oxygen (DO) levels averaged 3.0 mg/L with a 1:1 recycle from the fourth

stage already being employed to reduce the oxygen depletion in the first stage.

The profile of BODj- removal across the RBC indicated that extra capacity remained

in the later stages, if the loading could be increased further without adversely

affecting first stage removals.

The relatively low organic loading rates achieved during most of this study

were insufficient to accurately determine the maximum capacity of the RBC for

carbonaceous removal. During the first half of the study, the leachate BOD^ was

greater than 800 mg/L, which could have produced loading rates greater than 30 g

BOD[j/m 2 *d at maximum flowrates. However, the operational problems recounted in

Section 7, continually postponed the planned increases in loading, as it was

undertaken to establish steady operation at moderate loading first. By the second

half of the study, when a reliable mechanical configuration had been achieved and

the other calamities overcome, the transition phase 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 5 loading of 22 g BODj-/m 2*d was achieved. Although no BODj.

results are available for this brief period, because the BOD testing had been

suspended over the Christmas holidays, the COD 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 COD 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 2*d, which is

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

BOD 5 REMOVAL vs. UNCORR. LOADING Temperature Effects

T3 *

CM

2 O LLJ r r LO Q O CD

10-1

8-

6-

4-

2

A A

•

X

A D

B O D 5 LOADING (g/m2*d)

Legend A Temp. >12C X Temp. 8-12C • Temp. <8C

"i 1 1 1 4 6 „ 8 10

Figure 9.1 B O D - Removal versus Loading Rate



Temp. >12°C 0.9974 -0.1078 0.9978 28 Temp. 8-12°C 0.9922 -0.1497 0.9988 9 Temp. <8°C 0.9507 -0.0223 0.9993 14

Overall 0.9842 -0.0886 0.9982 51

160

COD REMOVAL vs. LOADING Temperature Effects

30-i •

20 COD LOADING (g/m Z#d)

Legend A Temp. >12C X Temp. 8-12C • Temp. <8C

40

Figure 9.2 COD Removal versus Loading Rate



No. of Data Points

Temp. > 1 2 ° C 0.9547 Temp. 8-12°C 0.8435 Temp. < 8 ° C 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 2 *d. As pointed out by Murphy and Wilson

(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 2*d, thus under light loading. This

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 COD 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), who observed a very slight temperature effect between 5 and

15° C. Coulter (16), 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

B O D 5 % REMOVAL vs. LOADING

o

LU CC

100-.

95 A

90 A

LO Q O OQ

85

80-^

75

A X * ^XMAV A " • A Q

AX • • •

X

X

X

2 4 6 8 B0D5 L O A D I N G (g/m z«d)

Legend A Temp. >12C

X Temp. 8-12C

• Temp. <8C

10

Figure 9.3 BOD_ Percent Removal versus Loading Rate

164

BOD 5 REMOVAL vs. CORR. LOADING Temperature Effects

10 -i

•o CN* 8

O

UJ 4-

O 2

CO

•

4 6 8 „ 10 B O D 5 LOADING (g/rri *d)

X

•

Legend A Temp. >T2C

X Temp. 8-12C

• Temp. <8C

i 12

Figure 9.4 BOD,. Removal versus Loading Rate Corrected for Temperature



Temp. 8-12°C 0.8876 -0.1498 0.9983 9 Temp. < 8 ° C 0.7156 0.0098 0.9981 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, number of stages, rotational speed, etc., to

effluent 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 are generally adapted from those used to describe

suspended-growth cultures. One approach which has met with popular success is

the application of Monod type kinetics (44,57,58). Equation 1 shows the RBC form

of the relationship.

The greatest difficulty encountered with the application of this and other

kinetic approaches has been the estimation of the amount 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 value

and then multiplying by the disk area. Estimates of the active thickness (Z) for

BOD removal have ranged from 21 to 200 /im (26,58), depending upon the

substrate loading. Therefore, arriving at an estimate of the active biomass for a

fixed-film process is even more uncertain than for a suspended-growth process.

Kincannon et al. (44), used a simpler analogue of the Monod equation

which avoids the evaluation of biomass amounts. Equation 2 presents 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 D were determined to be 199.5 and max D

212.2 g BODj-/m 2 *d respectively. These results are equal to 40.9 and 43.3 lbs.

BOD^/1000 f t 2 * d respectively, which are roughly 4 times the values determined by

Kincannon ef al. for sewage treatment ( U m a x = 10.0 and K g = 10.4 lbs BODj/1000

f t 2 * d ) . The significantly higher values of U m a x and Kg indicate that this leachate is

more readily degraded than the sewage used in the other study. This finding fits

nicely with the observation that most of the BOD of the leachate was in the form

of readily degradable volatile fatty acids. These results also indicate that there was

no apparent inhibition of the heterotrophic activity by any characteristic of the

leachate composition.

167

U A X. Q(S Q-S)

Z = 4s J - + R S (D

A = disk area (cm 2 ) S Q,S= substrate concentration (mg/L)

X, = concentration of active biomass D i

( m g / c n r r )

U = specific substrate utilization rate (1/day)

k= M max. rate substrate utilization per unit active weight of bugs (1/day) Y

Q = flow rate (Ud) Z = active biomass thickness (cm)

K = Monod 1/2 velocity coefficient (mg/L)

— = — B — * — — + — — (2) U U QS/A IT

max ^ max U = substrate utilization rate K R = Saturation Constant

(g BOD 5 /m 2 * d ) (g BOD 5 /m 2 *d )

U = Maximum Removal Rate QS/A= Applied Loading (g BOD,-/m 2*d) (g BOD t ; /m 2 *d)

168

1/BOD REMOVAL vs. 1/BOD LOADING Monod Kinetics Approach

UJ DC

3-i

2.5-

$ 2 J

O

1.5-Q O m H

0.5-

A

•

1/BOD LOADING

•

Legend A Temp. >12C

P X Temp. 8-12C

• Temp. <8C

i 1 1 1 1

0.5 1 1.5 2 2.5

Figure 9.5 BOD_ Removal - Monod Kinetics Approach



Overall 1.0586 0.0050 0.9976 51

U m a x = 1 9 9 5 (8 B O D 5 / m 2 * d ) K R = 211.2 (g B O D 5 / m 2 * d )

169

While the kinetic models have proven to be fairly successful at modelling

the RBC process, their application is limited to conditions under which the kinetics

control the bacterial growth. As pointed out by Famularo ef al. (26), in

suspended-growth systems, kinetics almost always governs because mass transfer

resistances are negligible across the relatively small dimensions of a bacterial floe

particle. This is one reason for the success of kinetic models in suspended-growth

systems. However in fixed-growth systems, biofilms can be quite thick, and

substrates and products must move 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 system. There is

however considerable work still to be done to further refine these models (31). The

numerous kinetic and diffusion coefficients used require further verification, and the

effects of temperature and hydraulic retention time observed in this study (and

elsewhere) could be incorporated. As presented in these papers, the 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 2 *d.

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 be no mass transfer limitations, and thus the nitrifiers

should get a portion of the substrates available. Hockenbury (35), conducted a

series of tests which tended to show there was no good reason for nitrification to

be inhibited by high BOD concentrations. However in practice, the inhibition of

nitrification is observed.

Recall from Section 8.2, that for a few days in December 1984, nitrification

was reduced to zero as indicated by the effluent nitrate levels. However, since the

effluent ammonia levels remained essentially zero, the lack of nitrification was not

due to the low water temperatures ( < 5 ° C). It was then realized that this apparent

anomally corresponded with the organic wash-out event. When the organic loading

rate of 22.0 g BODtj/m 2*d is compared to the ammonia loading rate of 1.08 g

NH 3 -N/m 2 * d , the ratio of BOD 5 :N is 20.4:1, which matches that of the nutrient

requirements of the heterotrophic bacteria. Therefore, despite the presence of an

established nitrifier population, the hetertrophs apparently consumed all of the

available nitrogen for their growth requirements, which underscores the inhibition

discussed above. The issue is clouded somewhat by the low temperatures, which

have been found to reduce the activity of nitrifiers more than heterotrophs;

however, 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 for

organic removal, and as will be discussed further below.

In this study, considerable nitrification (generally >50%) occurred in the first

stage. This nitrifier activity reflects the the low rates of organic loading and long

hydraulic retention times, resulting in high first stage BOD removals and low residual

BODj. concentrations. Therefore, it is anticipated that there was a strong relationship

between surface area and nitrification rate for this study; however, the data is

172

insufficient to be conclusive in this area.

Figure 9.6 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 loading achieved in this study, 1.3 g NH.j-N/m2*d. The

regression results indicate an overall average removal of 80%, but when the outlying

points are ignored (see groomed results), the majority of the data indicates a better

than 9 5 % removal efficiency. From Figure 9.6, the causative effects of the outlying

points are not clear. The scatter of the data at the top end of the graph (above

loadings of 1.0 g NH^-N/m^d) could be due to temperature effects, the 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 9.7 and 9.8 respectively. Figure 9.7

indicates a trend towards 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 serve to clairify the

issue much as there is again considerable scatter of the data and the lowest

removal efficiencies do 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° C.

However, temperature effects on nitrification were much less than is

commonly observed or assumed for RBC treatment, which parallels the organic

removal results discussed earlier. Murphy and Wilson (54,80) determined that an

Arrhenius temperature coefficient of 0 = 1.09 applied to nitrification for temperatures

below 20° C. Coefficients of this magnitude have been determined or used in

many RBC studies. Figure 9.9 shows that, as with the organic removal example,

173

N H 3 -N REMOVAL vs. UNCORR. LOADING Temperature Effects

1.2

CM

LU <

LU CC

0.8

< 0.6H > O

0.4

' 0.2-CO X

• • • x •

• •

Legend A Temp. >12C

X Temp. 8-12C

• Temp. <8C

0.2 0.4 0.6 0.8 1 J . 2 1.4

N H 3 -N LOADING RATE (g/rrT*d)

Figure 9.6 NHL^ -N Removal versus Loading Rate



No. of Data Points

Temp. > 1 2 ° C 0.8006 0.0678 0.9235 29 Temp. 8-12°C 0.7999 0.0531 0.9694 17 Temp. < 8 ° C 0.7518 0.0849 0.9208 19

Overall 0.7718 0.0750 0.9433 65

Temp. > 1 2 ° 0.9822 0.0034 0.9994 26 (groomed)

Temp. 8-12° 0.9461 0.0095 0.9945 41 (groomed)

174

NH 3 -N % REMOVAL vs. LOADING

o

100-1

90 H

80

70

CO X

60 H

50

X •

• • •

• X •

•

0.2 0 .4 0.6 0.8 1 J.2 N H 3 -N LOADING RATE (g/rri *d)

L e g e n d A Temp. >12C

X Temp. 8-12C

• Temp. <8C

-1 1.4

Figure 9.7 NH„ -N Percent Removal versus Loading Rate

N H 3 - N % R E M O V A L v s . T E M P E R A T U R E

- x 1 0 0 - i

o LU rr

9 0 H

8 0

Z 70H

CO

6 0

5 0

x

X X X

X

X

X X

X

X

X

5 1 0 1 5 2 0

Temperature C 2 5

Figure 9.8 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 3 -N REMOVAL vs. CORR. LOADING Temperature Effects

* CM

LU

1.2 n

H

rr

o UJ rr

co x

0.6H

0.4 H

0.2 x • •

Ox

• X •

• • LTJJ X * •

•

• •

•

Legend A Temp. >12C

X Temp. 8-12C

• Temp. <8C

NH 3 -N LOADING RATE (g/m2#d)

Figure 9.9 NH, -N Removal versus Loading Rate Corrected for Temperature

178

AMMONIA % REMOVAL vs. RETENTION TIME

o LU rr

100-1

90-

80

70-

60

50

A

A n A

5 10 15 HYDRAULIC RETENTION TIME hrs.

20

Legend A TEMP. >12 C

X TEMP. 8-12 C

• TEMP. <8 C

Figure 9.10 N H , -N Removal versus Hydraul ic 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 process. The models for

nitrification in RBCs parallel those discussed earlier for organic carbon removal, so

there is no need to repeat those 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 analyses the parameters U m a x and Kg were evaluated to be 4.69 and

4.54 g N/m 2*d, or 0.96 and 0.93 lbs N/1000 f t 2 * d respectively. These 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 growth

rate is probably a major reason for the observed retention time effect. 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 treating this leachate compares very

favourably to results from RBC 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 (54,80), found that the maximum loading rate for

complete nitrification was between 1 and 1.2 g TKN-N/m 2*d, 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 3 REMOVAL vs. 1/NH 3 LOADING Monod Kinetics Approach

O

U J

cc

10 - i

8-

6-

CO X

4-

2 -

• A

k

A Legend A Temp. >12C

X Temp. 8-12C

• Temp. <8C _, , ! ( ! 2 4 6 8 10

VNH3 -N LOADING

Figure 9.11 NHL -N Removal - M o n o d Kinetics Approach


Data Group Slope Y intercept Correlation No. of Data Coefficient Points

Overall 0.967 0.213 0.991 65

U m a x = 4 6 9 <B NH 3 -N/m 2 *d) K B = 4.54 (g NH 3 -N/m 2 *d)

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 for

low temperature conditions may not be 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 90% N H ^ -N conversion, and

typically produce effluents with less than 10 mg/L N H ^ -N. However, there have

been problems encountered while trying to nitrify landfill leachates. Many of these

problems have resulted from the greater sensitivity of the nitrification process to

upset. For example, Keenan et al. (43), found it necessary to reduce influent

ammonia levels of roughly 1000 mg/L, by 50 to 60 % 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 20

days while treating an old leachate, and that an SRT of 70 days was required to

reduce effluent ammonia levels to less than 1 mg/L. Their lack of success may have

been due in large part to their inability to maintain adequate solids concentrations.

The MLVSS of their reactors were typically <100 mg/L. Jasper et al. (42), failed to

maintain consistent 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 removal.

Although one of the advantages ascribed to RBCs is that they provide a

more stable environment for nitrification (22), the few results concerning leachate

treatment are inconclusive. The results presented by Ehrig (22) support the results of

this study, as he found efficient nitrification of three different leachates from

182

methanogenic phase (old) landfills. These leachates had much higher ammonia

concentrations than the Premier leachate, ranging from 206 to 1346 mg/L. In

contrast, the study reported on by Coulter (16), observed 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 of the doubling of the loading rate

at the start of run #2. Nitrification would have then been difficult to re-establish

because of the low wastewater temperatures ( < 1 1 ° C) during run #2. Toxic

inhibition of the nitrifiers , by something within the 10% of industrial waste

accepted by the Montreal landfill, was proposed as a contributing factor.

The results of this study would 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) levels of 0.5 mg/L, which is much higher than levels observed in

Vancouver area landfills. A more extensive analysis of the Montreal leachate's

composition may have found other toxic and/or inhibitory compounds, both inorganic

and organic. In the absence of toxic effects, given the experience of this study, it

seems implausible that nitrification would not have become established during the

three month period of run #1, under the prevailing light loading conditions and

warm summer temperatures. Once established, the nitrifiers would not likely be

totally upset by just a doubling of the loading rate. During this study, the loading

rates were highly variable and doubled on various occasions 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

1 8 3

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 generally observed to have only a minor

affect on the biomass or suspended solids levels in the RBC. As implied in the

above discussion the biomass growth, and thus process performance, was able to

adjust to the changing loading conditions. Agian, the low range of organic loading

generally avoided many problems such as oxygen depletion, Beggiatoa growth, and

substrate inhibition, associated with heavy loading conditions.

Suspended solids levels within the RBC stages were usually lower during

periods of steady operation. It was observed in this study and elsewhere (54), that

suspended solids accumulated in the RBC during interruptions of flow. During brief

stoppages of one or two days, the RBC solids generally continue to slough at a

normal rate, and then accumulate because of the lack of flow through the unit.

Murphy ef al. (54), observed that a flow of 10% of average flow was sufficient to

wash out the sloughed solids. This accumulation affect explains the higher

suspended 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 two occassions, an interruption in the leachate flow resulted in a general

sloughing and major loss of the biomass. It is uncertain why these two

interruptions caused such 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 decay which would weaken the biomass. The second instance, in

November 1984, may have been a culmination of the various effects of previous

upsets and declining temperatures. A sharp decline in the loading rate over the

previous two weeks 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 of attempts were made to determine the toxicity of the Premier

leachate and RBC effluent using the Daphnia bioassay procedure outlined in Atwater

ef al. (2). However, problems were encountered with the survival of the Daphnia in

the dilution water blanks and therefore no reliable results were produced. It was

observed qualitatively that the Premier leachate was fairly toxic, which one would

expect given the ammonia concentrations alone. The RBC effluent samples on the

other hand were apparently non-toxic. This was indicated by the Daphnia 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 extrapolating from the

encouraging 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 quite weak, despite coming from a young landfill; this aptly

demonstrates the effects that specific site conditions, such as climate, drainage

patterns, etc., can have on leachate quality. There were no indications that this

leachate was inhibitory to either the heterotrophic or autotrophic bacterial growth on

the RBC. Experience at this university and elsewhere (16,18,42,43,66), with other

leachates, particularly strong leachates, have shown that biological inhibition due to

substrate 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 9.1 Design Loadings for RBC Treatment of a Municipal Was tewater^

Design Loading RBC System Design Objective Parameters (g /m 2*d)

(Ave. Value mg/L) 15°C 10°C 5°C

BODr removal B O D 5 s 2 0 B O D 5 L o a d 9 7 7 6 6 0

TSS S 20 (total) B O D 5 removal B O D 5 ^ 3 0 B O D 5 Load 15 12 9.3

TSS S 30 (total) B O D 5 removal TKN «s 3 TKN Load 0.60 0.39 0.25

plus nitrification (filtrable)

- assumes primary clarified wastewater feed to RBC with 180 mg/L BODr, and 30 mg/L filtrable TKN,

- provides factors of 1.25 and 1.35 for BOD^ removal and combined BODj. plus TKN removal to correct for diurnal flow variation.

(1) From Wilson 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 organic loading.

Another 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 hydraulic flow through the

fill, so that the hydraulic and organic loading tend to increase together. During this

study, the mass of C O D released from the landfill was observed to increase eight

fold over four days, after a prolonged period of low flows, in a full scale plant,

the biomass would likely be unable to assimilate so much additional substrate that

quickly. The results of this study however 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 differential of up to 4° C was observed

between the first and fourth stage liquid.

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, which roughly corresponds to an extended

aeration mode 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 suspended-growth

systems, and these have been inconclusive. None have been conducted for leachate

treatment. One problem with comparing the two systems is relating the loading

rates in the two systems, as again, the estimates of active biomass are determined

in different ways and are quite subjective. Further research and comparison studies,

with particular emphasis on the nitrification performance of the two types of

treatment systems is required to determine if there is a difference. Therefore, until

one type of treatment system proves superior, designers will continue to chose on

the basis of economics and personal experience.

9.7 EXPERIMENTAL PROCRAM AND RBC OPERATION

The experimental program as proposed could not be evaluated because it

was not carried out, but it still seems to be a valid approach. However, there are

a couple of changes to the sampling and analysis procedures which would have

been beneficial in hindsight. Firstly, it would 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 would have more clearly defined the residual carbonaceous and

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 oxygen

values during December 1984, when the organic wash-out occurred, to confirm the

other indications of oxygen depletion.

Other changes which were not implemented in this study were the use of

load cells on the shaft to give a measure of total biomass, and the use of an

autosampler to collect process and leachate samples. Load cells at either end of

the media shaft have been used in other studies with good success, to easily

determine a relative measure of the total biomass (47). For reasons discussed earlier

in Section 4, the use of areal biomass 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 would 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 of the RBC to loading fluctuations. However,

autosamplers should be used judiciously to test specific notions; because, while the

sample collection is relatively effortless, the analysis of those samples is not.

With respect to the RBC 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 of this

study but one useful modification would 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 mounts for the shaft and disk drive motor, a sludge removal

mechanism in the clarifier 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. SUMMARY

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 suspended

solids were less than 25 mg/L during periods of steady operation, and usually less

than 100 mg/L during upsets. Sharp changes in loading or interruptions of the

leachate flow were first reflected by increases in the suspended solids. Overall, the

RBC 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 limitations. In most cases, the

B O D j loading was less than 6 g BOD,-/m 2*d. However, a few samples had higher

loading rates, ranging up to 18 g BOD j/m 2 * d , and still produced a high quality

effluent. These results indicated that the carbon removal capacity of the RBC

treating this leachate was comparable to its capacity to treat domestic sewage.

Efficient nitrification of this leachate was also maintained throughout variable

conditions. Effluent NH-j -N and TKN -N were usually less than 1.0 and 10.0 mg/L

respectively. Nitrification was observed to stop under high organic loading conditions.

The average nitrogen loading rate during the study was approximately 0.6 g

N/m 2 *d. Results 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. RECOMMENDATIONS FOR FURTHER RESEARCH

1. Given the general lack of experience with RBC treatment of landfill leachates,

further studies should be undertaken to conclusively establish the capacity of

the RBC to treat landfill leachates of varying strengths and compositions, with

special emphasis on nitrification.

2. Side by side comparison studies of RBCs and Activated Sludge treatment of

leachate should be undertaken to evaluate advantages or disadvantages of the

two systems, especially for nitrification.

3. The relationship between temperature and hydraulic retention time effects should

be investigated more fully. Possibly the volume to surface area ratios of RBC

design could also be used as a factor to change hydraulic retention time and

reduce temperature effects.

4. The types and concentrations of trace organic compounds in leachate should be

investigated in more cases, and the major mechanisms of their removal during

treatment determined.

5. The denitrification of landfill leachate could be investigated using a submerged

RBC. If, in fact, the nitrification process is more stable in the RBC, the nitrite

accumulation observed by Ehrig (49) may permit the short-circuit denitrification

investigated by Sam Turk (PhD thesis, UBC, 1986), to occur more reliably.

6. Ishiguro (56) indicated in his paper that Japan has had considerable experience

treating landfill leachates with RBCs since 1976. When this paper was

presented in 1983, there were apparently 135 RBC plants treating landfill

wastes. This indicates that a review of the Japanese literature may provide the

answers to many questions concerning RBC treatment of leachate.

197

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 on 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 Design for Wastewater Treatment", Prentice-Hall, Inc., New Jersey, 1980.

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., "Metal 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. EE2, April, 1976, pp. 4 1 1 - 4 3 1 .

12. Chian, E.S.K., "Stability of Organic Matter in Landfill Leachates", Water Res., Vol. 11, 1977, pp. 225 - 232.

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., "Aerobic 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., "Ammonia 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 Stephenson, 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 Woods, 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 Model for Sanitary Landfill Leachate", Water Res., Vol. 16, 1982, pp. 699 -705.

31. Grady, C.P.LJr., "Model ing 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., New 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., "New 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 Grady, C.P.L., "Factors Affecting Nitrification", J. Environ. Eng. Div., Proc'Am. Soc. Civ. Eng., Vol. 103, No. EE1, February, 1977, pp. 9 - 1 9 .

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., "Some Experiences with Fixed-Film Biological Processes", "Fixed-Film Biological Processes for Wastewater Treatment", Ed. Wu, 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 Chou, 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., "Chemical Inhibition of Biological Nutrient Removal Processes", Water S.A., Vol. 6, No. 4, October, 1980, pp. 171 - 179.

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 Pollut. Control Fed., Vol. 53, No. 10, October, 1981, pp. 1469 - 1480.

48. Masuda, S., Watanabe, Y., and Ishiguro, M., "Simultaneous Nitrification 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 Collection Design for Containment Landfills", /. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, No. EE1, February, 1982, pp. 204 - 209.

51. Metcalf and Eddy, Inc., "Wastewater Engineering: Collection, Treatment, Reuse", 2nd. Ed., McGraw-Hill, New York, N.Y., 1979.

52. Mikula, W.J., Reynolds, J.H., Middlebrooks, E.J., and George, D., "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 Qasim, S.R., "Biological Treatment Kinetics of Landfill Leachate", /. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 103, No. EE2, April, 1977, pp. 353 - 366.

57. Pano, A., and Middlebrooks, E.J., "Kinetics of Carbon and Ammonia Nitrogen Removal in RBCs" /. Water Pollut. Control Fed., Vol. 55, No. 7, July, 1983, pp. 956 - 965.

58. Paolini, A.E., and Variali, G., "Kinetic Considerations on the Performance of Activated Sludge Reactors and Rotating Biological Contactors", Water Res., Vol. 16, 1982, pp. 155 - 160.

59. Pohland, 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.

60. Poon, C P . C , Chin, H.K., Smith, E.D., and Mikucki, W.J., "Upgrading With Rotating Biological Contactors for BOD Removal", J. Water Pollut. Control Fed., Vol. 53, No. 4, April, 1981, pp. 474 - 481.

61. Pretorius, W.A., "Nitrification on the Rotating Disc Unit", Progress Water Technol., Vol. 7, No. 2, 1975, pp. 243 - 249.

62. Raina, S., "Combined Treatment of Landfill Leachate and Domestic Sewage", Masters Thesis, University of British Columbia, Vancouver, B.C., 1984.

63. Raveh, A., and Avnimelech, Y., "Leaching of Pollutants From Sanitary Landfill Models", /. Water Pollut. Control Fed., Vol. 51, No. 11, November, 1979, pp. 2705 - 2716.

64. Robinson, H.D., 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.

65. Robinson, H.D., and Maris, P.J., "The Treatment of Leachates From Domestic Wastes in Landfills - 1. Aerobic Biological Treatment of a Medium-Strength 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. Robinson, H.D., "Treatment of Domestic Waste Leachate in a Full-Scale Automated Plant", 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. 166 - 184.

204

69. Salvato, J.A., Wilkie, W . C , and Mead, B.E., "Sanitary Landfill Leaching Prevention and Control ", /. 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. "Standard Methods for Examination of Water and Wastewater", American Public Health Association Inc., 13th, 14th, & 15th editions, Washington, D.C, 1971, 1975, 1980.

73. Stanforth, R., Ham, R., Anderson, M., and Stegmann, R., "Development of a Synthetic Municipal Landfill Leachate", /. Water Pollut. Control Fed., Vol. 51, No. 7, July, 1979, pp. 1965 - 1975.

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., "Models of Landfill Leaching: Moisture Flow and Inorganic Strength", J. Environ. Eng. Div., Proc. Am. Soc. Civ. Eng., Vol. 108, No. EE2, April, 1982, pp. 231 - 250.

78. Straub, W.A., and Lynch, D.R., "Models 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. EE4, August, 1977, pp. 647 - 661.

80. Wilson, R.W., and Murphy, K.L., and Stephenson, J.P., "Scaleup in Rotating Biological Contactor Design", /. Water Pollut. Control Fed., Vol. 52, No. 3, March, 1980 , pp. 610 - 621.

81. Wong, 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. Wood, J.M., and Wang, H.K., "Microbial Resistance to Heavy Metals", ). Environ. Sci. Technol., Vol. 17, No. 12, 1983, pp. 582A - 590A.

205

83. Wu, Y . C , Smith, E.D., and 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. Wu, 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. Wu, Y . C , and 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 Analyses

2 0 6

u b c e n g i n e e r i n g

207

o_ r | o

V<!

O i — In

Oo! o-i tV O Lni lfi' 1 v?

V'l A • ll'l

l/> r-l/i

0-u\

vS> l/>

r- r -ir, Ln

(X lfi

0 v9 | E r c>

LO

Li f r i')

cr v5 vS>

- i q c 

o

"3-

O 0c v9

on r r-

r-<>

<si

In

i o (V) r.1

r - r— T r - i - j

ol "J" vs

O cr

(N

r~ v5> V5 ol

r -rx VTl

r~ T VP

rvi 1— ( 1 01

e~ OJ

1

ro

(V

r— rO ^~ r i

tr-

"3" fVl

<r-r-(M

f>, o" CXl ol

> [Y| 1* (VI

r-V9

V>

t « r-TP tYl

VA O-J tyi rvi

u (N U\ (VI

\ t

vSJon o|o <ri ui

O

CV)

ol (3 0 rvi

v9

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

h

!Y1

(S ffl

o l/l H •3-

l/I

r~ VA

(VI

<T VS

o l o r~

o 6 l/>

r -v9 Oi rs

o r-rs

Do

rflj — t>'| < r»j r-1—| VA

H > o| v9 fN

c r 5=

V\ 0-v» ?r

vS

V9

r-ol

Vl rs r-01

<r Ol v9 o|

mpji

0.1! r-olj lJ1

O ro (T in N

Vl > tr 0)

> d Cr J r0

ol > v9 (Yl

O

(VI

1-

6 Vi

tvi

vc > tr v» (Yl

(So

I~

r

k IO

V9 V> ty>

0

6 fl—

(V\

<r r-

01 (Yl

In ol T vfl

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

> r~

i

i 1

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

VP

h

t 01

O Z

i i I i

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

VP

h

t 01

O Z

i

|

.i

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

L

X Z

z r -

i in

vS ui 1 O

r" LA lA

\n 6* ri

t r

VA v9 CVl

fr H

Ol

o~ ro

00 0

v9 ci 1A

i

- ! • ? IA'| 6 T Ui

1

ol •3-fVl

0Q

rv>

O c -tr

ro ^~ U

O

ui

r-6^ ZT

Lr,

ui r 6 Vi

c r In

v5

^

\r\ ol CM

\

c / ) Q

_ J

0 in

?

z LU

o

(-—r ^.

L

X Z

z r -

i

i 1 i i

Ol £

v^

Z o CD ac cr u

o

o

h-

i i i 1

i

8 v9

Ol £

v^

Z o CD ac cr u

in O o CD

Ol £

v^

Z o CD ac cr u a

0 u

CN

v9 cr

to

o-

0 rS

q r~

Oo

i/i i —

to

Lfi 1 1 O

cr / \

•3-

r-00

tr T~

l/l !—

L O

cr v9 cr rv

cr

fv 0 (VI

O r-r

lo

v-0 N

0 er r -

(V) («

U

r -K>

(^ 0

ol Ln

ol

r l m rs ra

0 tr r

rs Ul M 0 3-

tr

(V)

0 ° V9 cr (Y)

r-<r m ^ (V>

r-

CX)

0 O rr M

L« v9 H

OJ

a.

LO \

d - 4 -

a

of

oi rs

C

—

r-

~t) '0

U\

0 "0

ti o Z

rr"

t> o

V?

6 o z

(VI

J

Cx

0

w

r~ <j c

; ^

(Yl

0

r i

-T r< s:

Cr=>

i d r

3-

£>

U-Q)

LL.

^ ,

- j »<

Li.

ID

J> s 6 I

V 0

/_ 6

l?

c

(Yl V 6

If.

. i

ol CN

Cr fN w

r ^

v9

(3 2

52

c T

0

(S

jD

t~ H M

•J ;

O

(V >

r

rj

a) i s


208

z r ' o_

o O vS

q v'j

u\ 0 ; _ . | >v\ Oj C*.; <?j r~ ^ c , v9i U'\j U l

I— o _

O

q vi

0

v9 v5 q q "I

v3 v5 In v i

0 vr, O v9 v i

<r 0 ^r iw

r t (VI

zr r>l

v9 cy> <w ti r"

(VI' i . - i

0' f> <r c> 0

vP O r J c-J

r~ ZJ

0 r-rsi

(5~ ol 0 r i

tn in cr

cr o -

<r

O

r— cv c r 0

L9 0-

o--r< O i s

O 0

m i.'j r -

r (.•••

1

r-r~ zr

z r 0 r r

c>" fi~ t>-

r -

ZT O

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

</> r-

ro

C>i O ro H

rS O \f< rN

K i

<Y| vi> (VI

rs

3-

o" r KI f»

v.9 o~

c<>

v_o O

TT

6 (<n

r~

o r--> O r l

0-0

Kn 0 bo

O —

r -

<r r -

cr

0 0 CV v5

O i'-.

r (

Lo f ri

Lo zr r~

(V) vS O

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

OO

> r-

^n! O

! m

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

u>

h

1 ro

O Z

r— r- r-

\r>

1

In!

j

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

u>

h

1 ro

O Z

v«i -

cr |

T i

j i

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

i c<>

X Z

z .Y

t-

6 zr

O

H

0 c p i

r~

fits

In q ff"

q 6 ts

rn V?

cr

ol

\*

r \

i

t

1

r ; cr> rs

Do

rv<

3-rs

q

cs

q

TM

C» r«i rx

rs

q

zr

6 rS

o< cs

0

a1

cr

ZT

Lo r -

Lri

1.1 • c r

cr

/— "H \

i

in

Q

_ i 0 CP

£

Z O! \3 O a: I-—7 *£-

i c<>

X Z

z .Y

t-

6 zr

zr

K l

tr r-

rl

er (VO

*zr

K>

q r"

c> q K I K. e>--

ej cV

zr

e» (S r-r /

zr zr 0

Kl

00

(vi r-In rs

i zr! o° LA! \ $ rSl rS

I v_9 1^ rv

r In rs

L0

\fi rs

cr ro rS

Zr

M rs

(NI

LO N

(VI

rs rS

vP rs

r-; CN CS

v5

c>> —

in 50

zr cr

in cr \ CP

\ i

ro cs

O

*C \

£ v_

Z o CD Ct (£ U

0 0 k

bo cN cr

r>t v9 cr

i ' & CP c°

Lo r \ t>..

O ZT

L O

Ln

In N

m

0 O v9

o o

In vS r-

o z r

Lr> O r~

to >--)

v5

vi> rs n i

VP

in

0 c>

n l A

r-(>Q

In (/< LO

c< L"J

rs in

LO 0 ° U0

cr LO rs \f,

r— L O

m

r~

zr

Lrv

z r rs rS

O

6 rS

*C \

£ v_

Z o CD Ct (£ U

o CO

H M

O '

(*>

<J>

H O i~

0

(T

r—

c c J

0 cr

r~ r "

r— <r

rr^ rs Lf> 0

in

cr L I

cr

r 0

zr

?

*C \

£ v_

Z o CD Ct (£ U Q

0 u

zr v9 v3 <S

r -

CT H r l

>-Tl V Kl r j

m (VI Ln C N

rsl r H rs

vS v3 o

cr vS zr

In rj

<v\ L.-i ^9 vf

zr Co \s\

ol v3 er

CV)

v3

zr zr £-=

r -v9

Cx-

(•'J

fits 1/)

cn 0 0 zr

Vrj 0 (VI

t/i

zr O

vS O 0 I T

in

zr

L0 cr 3"

0 rs rS

lo 0° r4

Ln 0" 1)"

0 c> cr -

c-O lr>

r s vS

QJ

CL E

\ -+-

tf O

H 

a> ; m

Ln

i ? h

i ?

h

in

' <

n

e~

i 5

h

, s •> —•

r~

j

Ln

V"

•2

cr

i 6 3

X

H

) j

sT

• 6

s Z3 <Z

'JI

) 6

<?

f J

U-0>

1-1

f™

(V

—«-a/

t ' i

- r ai

- f

0

10

0

r

"5 0

H

0

zr

0

1—

0

cr

- r

0

,-s

0

00

0

6 0

—

O

0


209

tr v i

tr-; lo

rS

v9 \S v i

—\ • 1ft _ !

VJ>

H vO

Lo r o

V9 v i

rs V S

on

V 9

LA cO

v 9 v i ' v i

tr v i

i rr v 9 v i

r>-» v i

I'll L /

rvj ro v i ! v i

CO

v i

O l

v i

c 0

(J

0 "

F -v5 Lf i

c* r fx/

0 rr CO

0 0 vS lo f r o~

1 C T l f T -v y -H v<D

V ? Ln r J rvi

rj-r>) fv'

cy cr cr

(V)

o l

f M

O t> t>

Ln cr r/» o(

rvi O

rs

rS tr tr I S

Lo O

Lo

CN

O

tr o l

ro rs r~ cs

tr r -

Oa O Lo cs

vS r-ro cs

Co L, Cyo

0 0

cr v9

Lr

cr

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

CO

h r-tn

cr 0

Lo o |

co

0 rr In tr

r-

0 T r >

J

f> r> v9 v9

O tr

v<?

(>-

H

rs

0 o l

0 ov

c r LA

r s

r~ c»

rs

H n -1 o i l

> tr o<

0 LA H rs

LO O tr

o4

v3

ro O cs <S

i i L3

tr

CXD

(Yl tr ffl

£

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

to tn >

• i i {

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

u. >

i 01

0 z

!

i 1 j

i

0 r-rr

rr

v 9

tr cr O

ro vj>

Vn rj-Ln

C~ Vn LT)

0-rr 10

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

u. >

i 01

0 z ' i

I

(

j 1

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

1

tp rr z

Z

Y-

! Q i <s

jLo joc

er

O

CM

(vi

Oo .-s

r~ rr o l

r—

rs 6 r s

rA rr

r o c-l

0 o l

co 6 o l

r r rs cr rv> rs

LT. v i

CO

v9

CA

( O rsl

OA 0* r s

0= c rO

O ea

r o

Ln

c~ tr-

£ S

r S <3 r— o l

rA S3 Pt

j

i WA Lo" 3-

\i\ 0 •3-

u \

CVJ

r o tr

IJT rs

• S cr 3-

LO

_ r o

rr (VA rs

LA

LA

Ln

? r o

Lf)

£ 0

0

o< 0-1

}a: LO LO! cr OSI CS

O c s ; ro

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

1

tp rr z

Z

Y-

LO

\ i vS>

r~i <N

" 3 -

6 _cs

rs 6 Ol

cr CS vS> Cr-CS. c S

Lf) rf cs

°? rs|

vA

0* N

5 !

vS

N

Q

5r

rvi 0-r A ui

t>i

00 •3-r A

0 LA

<vi -3-

00

•3-

i i

-3-

cr i o r-•3-

(> o i LO

0 0 6

lo ct- 5?

q

5-

r o

rr

-cvS ro

0

h3 ro

Civ9 K i — r s j i

O l LO C»lo<" o(| f O

ffl £

t/7 a _ J

0 LP

? ffl f

•—'

z L U

\ 0

0 ca t-—7 ^~

1

tp rr z

Z

Y-

LO

\ i vS>

r~i <N

" 3 -

6 _cs

rs 6 Ol

cr CS vS> Cr-CS. c S

Lf) rf cs

°? rs|

°? r-(S

6 cS

vA

0* N

5 !

vS

N O ZJ-

Q

5r

rvi 0-r A ui

t>i

00 •3-r A

0 LA

<vi -3-

00

•3-r o

•3-

Lo - j 1

-3-

cr i o r-•3-

(> o i LO

0 0 6

lo ct- 5?

q

5-

r o

rr

-cvS ro

0

h3 ro

Civ9 K i — r s j i

O l LO C»lo<" o(| f O

N cn £

z 0 CO or cr u

O O

O r

0 to O'l L i ' .

r~ o l tN v9

cr m Lo

r~ rr

errs rr

Lf) r S LO

r -

r r O rv> L A

(VA V v 0 " )

r S L A

cr r l t~

c r r-v9

O

vS

to r~ Lf)

rS -J

r

v 5 rr

r—

rr Ln r~

rO

v S

LO

r— r S

V9

v9 t r V9

P-v9 vS

cS rr

vS CT rr

O Lo,

_V5

f O

v P

i

<""irr > S | c r Lnlco

1

i vS; v s — ; c r rr.rr

N cn £

z 0 CO or cr u

i/i O

O CO

j

i 1

N cn £

z 0 CO or cr u O

0 0

O t r

Oo O rr

of r 0

T v9

r r r r

O

rf)

ri O c s

v5

r r

vf) to ro

tyi v5> Lf i

0 °

0

vP CV)

O t -

(.,-.

c~ rv'l

~T"

cy> 0 ty>

C ~

cr O <v> v 9

r r O cr

r r r~ v9

tA L O

v9

r~ v9 r~

0 cr 0

vS O 0= v9

<r o l

O VA c r

— rr O

1— rr H O

c x

E tf

-+-

0')

•3-

0

r

3

r

_o

L'i o<

6 0 z

f S

6 0

OJ

O

cr

J Ql

O

O

a

v9

J a' O

0 r s

rvi

N

(J »

0

L> Q)

O

r

v3

(

f> -•

V

0

s

h

r -

>

(VI

_D a '

L i

O

S>

U -

r -

(U

ol

_ D 

r 0 E

0 0 1

0 r

rs

0 X

O (V)

6

X

L5

' > 0 SZ

O

>. <3Z

*>•

r~

- s-

1

>


-*

210

re x i -J v i v9 I 1 1 rr

- • "5" v i

- •

vS v5 rr VI rs

v i

r -v9

VI r~ v9

r-v i

cr v9

r-v9 q c r

v i

cr vs

60 v i

Cr v5> v i

0 r-'

LA cr V Q

cr v9 V5

rx, vs

L1 i A

v9 vS

u ° t VS

J C

onoJ

.

l/l In r l r l

T cT

CN rvi cr

O rr rS rs

rr tr LO F <

cr vJ3

V/> vS OO

v5 r-r-

In rr \ S

•J' r 0 r<

LA r 0 0 °

r r r 1

cr

vJl 0° cr

vS rs cr

vS cr

tr LA 7\

r s v i

IT C r~

rr v5»

Cr rr v9

r-vS 0o r-

V S

O O

o l (y L A

CS tr rr

r-v9 ~>

rA vS tr

tr VS

vS O tr-

•*v tn

£

CO Q - J 0 to

1/)

h i* rA cr

ro LT)

Ln In vS

vJ r

0° CN r~

--cyi

r4 CT f>

\A 0 lA

r~

£2 O cr (VI

iA 1A rvi

cr Q ^ H lA

tr vS r-

r -cr

O cr 0

r- L i cs

0 LA L A

U I

cr cr

P tr

cr vS O tr

— -

CT V.P

•*v tn

£

CO Q - J 0 to

• - ... — - —- 6

r -(vi 0° Ln

O

r~

cv*i CP

vS rS r—

O v9 r-

cx O r-

O O r-

( S 0° r-

00

tr

tr tr \S

O vS

oO CS v5

VA cS vS

VA V\ v5

O c« v9

Cr* vn Lo In

0 CO Ln

VA OA Ln

CO rr LTl

0 Ln

•*v tn

£

CO Q - J 0 to a

u. >

rs lA v9

LA L A

tr v i V9, rj-

cr v9 rr

0 0

vi) •

r ~ 0=

rj~ O •rr

rvi Lf)

O tr

r ^ rv>

t f cv rH

rA

N.

r ~ In

rs v9

lA rs:

vS CA vA >

>

ol cS

ol O

iA 0 0

tr 0 0 rr 1—

Cr-rS

rs rs

CA 0 !

? <*>

£

2 Ui tD O aa h-Z

i «

O 2

....

._. -

.._ - -

? <*>

£

2 Ui tD O aa h-Z

2 i <o

X Z

o rvi

VS

CO

Vn r f rn

0'

Vri CM

m

O V-A

_ J O

in rr rO

0 0 IN'l

"in rr

&

V ?

po

O

!•>

v i t r rs

cS

rr r-rA rr oo to

CO

**v?

C N

CM

cr-r-J

O

d rfl

•a-s c-m

-3 r v rO

P-rw 00 r — rs

r-J

rO C*> O CO

c-<_

CS rf

2 0 0

to

0 LO* po

CO ro vT (S_ rs I N I

CA CO -3-V "

r* rr

v£

LA r~

Ln C O cv in cs ""0

v5

c s

Ln

q tr

r— "T cs.

CO 0 ro

rS ("A

O CO f A

0 Q

r s ro

•3" f A

rA rf rA

r-l C X

0 i > cS

Vo ro" r s

O rf cs

V A 1— -3" rs 17" rf o<

VA

r f cs

vS

rf cs

•or v-r rf cs Lo 3-r-s

LA cs cs. CS C v j c s

? <*>

£

2 Ui tD O aa h-Z

2

r-

tr-rv>

o •o tr

rO

« r~ ,0

-3-m lO zr ra

rr

O

vS •a* «"A

"3 rA tv\

(J* cn tr "o CO* zr

c-s

in

Cr rs C° rs

rr 0. 00 r r f V \

vS> rn

O r o

•3-Vi <yi

vj>

rs

rr-oi rS.

CS CVl

0° CP ro

0 O (V* ro

0 rf

LCl r> CS («

o l O ro

O <5 C N A

V ? (VA

t>° rf A A

0) v9

r r O CS.

o l Vri o l

t » rf cs c J VA O l

rr-LTJ CS

0o C N Os " H VA C S

en

Z o co CC cr u

o 0 k

a~ r -*A

cn 0 0 tr

tr rr

r-*3-§ l—

1

r S L O

v i )-T)

8 m

cr 3"

v5 VA on

LA r r rO

L*n rvA LA rA

CO V? r s

VA O ro

LA VA rr rs

O rs

rA C/A VA

-en

Z o co CC cr u

u>

O co

rd m rr

o<

v>. Ln

CT U7 m

&o rr

£ a

rr r" ro

0 cn m

vS> cr

vn tr-rs-

in

XT

CS

&

r-m

ro tr m

1—

v9 r<

rS vi> C S

rA

cS V i C v9 cr 0

tr CY) r -

CT*

en

Z o co CC cr u a

0 u

> vS»

m rS

r~ r r a*

co *r cr

rs1

cr r

tr,

L/l rs cx

O rvi O

rr 0 0

rs 1— r-

cx> vS CO

c*» cr tn

r r 0° v ^

<n rs v9

vn Ln Lf>

O V_9 rj-

OQ rA LTi rA

Vn

8 rS Vo

Ln rs

vS cr

r-r rfi

L )

;-\ )

(v'l V 9

C S -

v3 H r S

rA H H

C H

O rr \ —

CL

£ tf

to

-4-tf

o

r j -

C

« ~c~j 3

* 3 -H

v. Q

r-H

.4 3 Z

C5 X

tj-

T> c

0 0

_T 0 X

to

0 0

cs-r(

O r

LA CS

a X

Lf)

h

t»

c

h 5

Hi

£

Hi

rH

c

cr

OJ

H H

eJ % 3

h

v5 r4

o>

errs

a1

S r

"i h

v5 O

1 i

h 5

fV>

I 3

r-

6

O rv) oJ

' 5

"3

vJ> o J

h

M

! 5"

h

CSl

cv

CT

- |

O

6\


* CXT

o_ .,J»

\S OQ

vS

r-v i

u r-vS

l/> °? 00 v i r-

00 00 -v9 v9

r vi

°? vS 1

v S

v i r vS

Cr \ 9

r -

v i

r v i

q r-

rS cr v i

rJ r-

0 O u CXi v i

v i

VA cr vs

~r r~

00 v i v9

0°

~s • c o vj

^? cr v9 T

f> 0 ° LA

o tr

bo O 0°

e In CO

r to rs"

VN r-

O t r

H Nj> 0

CO r-lA O

in 0

<r v5 rs cS

M CT O O

tv\ r-jbo

Wi

cr 1 O r

tN 0 O

O Cr O

O vi tS

1A r r

Vn

v5 cs

LA v5 O

Ul cn 0

0 r r LO

tn

I

in a _i 0 Ul

I- 0 cr ol M 0~-tfi

Csl Oo O

vS r0 Oo

\S o <y

oo r-cr v i

r-O r

O r—

6^ S5

«S

v S

I A 09 <s9

in

v5

Q

v3 S 1—

r~ &° r*

Cs

\S

cs: t r If)

Ln LO

co rs" v^ 0° v9

v3 oi r—

in r-

t r

in

CN

0°

r-r-vS

on O v J

o( CM O tn

I

in a _i 0 Ul

ff * J

O t r m VA LO

O

-M ks

m 0°

Oa

tj-

\i\ or t r

0 zr

0 -rs O zr

0 t r

t r 60

t r I"

Oo 0Q

0b

c°

>

rr

tj-ttr t r

C tr t r

V ?

r C c t r

vS cn fA S O c%

v5 ol tr

t r r-

t r O ti t r

O CT-vS

Do oC LD

O 00 ro

0 r-rn

v5» LO Ln

tn

I

in a _i 0 Ul a

> O A m

— t r ro t r on m cs vS — 6 Ln O T

m C S t r v5 CS

r r r— er O v9

CO =r

vi> tr 0-

in v5

£

2 Ul \D

O at

z

z 1

O

z

-• - . . .

. . . .

--

£

2 Ul \D

O at

z

2 i •a

a: Z

V A

CS <-< c4

o

r-vS CS

o c^ cs

cS

o C-xl INT.

cs cs cs

q rs

I — erf rf

rs of

0 r~

Vv r1 \ A

"tf v i rs 0

irv

0 vc

0

v9

V A

<d O Cr

ro v i

q cr cs

v5 rri rn

CCi cr v i

q rr cs

cs 07

cs CS vS

rs

\ . ' \

rt

r-

0 r

* r—

in r-cs

£

2 Ul \D

O at

z

2 i •a

a: Z

V A

CS <-< c4

o

r-vS CS

o c^ cs

cS

o C-xl INT.

cs cs cs

q rs

I — erf rf

rs of

0 r~

Vv r1 \ A

"tf v i rs 0

irv

0 vc

0

v9

VA

rs

c^ v;

cs r-' CO

in cn t-ri

tr rf

In Lfj

V A r-r—

va 1-7 c c-v

Oo

cs CS vS

rs

\ . ' \

rt

r-

0 r

* r—

in r-cs

£

2 Ul \D

O at

z

2 i •a

a: Z

V A

CS <-< c4

o

r-vS CS

o c^ cs

cS

o C-xl INT.

cs cs cs

q rs

I — erf v i

rs 0 irv

0 vc

0

v9

VA

rs

c^ v;

cs r-' CO

in cn t-ri

tr rf

In Lfj

V A r-r—

va 1-7 c c-v

Oo

cs CS vS

rs

\ . ' \

rt

r-

0 r

* r—

in r-cs

£

2 Ul \D

O at

z z

P- t r rS

v j

cs

o t f cs

°°. rn rn

t r c \ rs

t r cs cs

o cs

r< 2S

t r fr

0 rs

r>l Cr

O of

v_9 v^

t r C P 00 v9 cr

O tr cs

cr O c-v cs

r £ 0? e~

\A

O

t r

0 Q in t r

=r tr

K }

rf cr H cs

O J cvi

O rs

H O cs

Vn CP

O CO

~Q 0

en e

z o 00 CL <x u

O O 1-

-en e

z o 00 CL <x u

10 a O CO

tr-CP

t r t r

vS O

trot O V 0°

I-

o s 0" LO

<r 0

S

t r cr

cr TT" v9

*-rA

t r o~

ro v9

r- 0 r

r - r-to t r cr r- CS t r

ol vS

CP 0 t r

CA o l fA CS

r~ o l Lf)

en e

z o 00 CL <x u a

0 u

v3 Ln

r-r 0°

CP C P

cr cr

cs; % O cr> 1— LO cr m cf

cs-cr

cs 0 if"

vS ty>

r-0 ° H IV)

c< 0 ° m cr rN

>j\ rs

v-C cs CS

rr cn

t r ot

vS oi

r -

tr 0 °

v5 rn CS

tr v3 CS O v S r r

-cj

t r rH

»tf . 3 <E

bo ol

rG 3

<JE

•3- r-

ox - t i

v/i

3-

%

•/>

• s

Si • V

Lrv rs

S>. ft!

rs

QJ

cs1

-> • 0 O

Lo

-f

C

cr

-r 1 0

•rJ

-V

0 0

c r

0 O

CS

0

v^ rs

O O

O

0

ol

_&

O

6 0

"2

cr

jO 0 72

v3

—• _ c

0

c 2

M 0

r-H

O 0

T2

0

•3 c

-! 1

t r

ii

u b c e n g i n e e r i n g — ir , : 212

or cr r— v i vi

r-vi

ZT

vi vS vi v i

vS v i

tr vi

zr "vS vi

vJJ v i v i

SS

s i r-vi v i

lA v i

VA

vi

cv

vi tr vi

ej v3

zr *

TS C tf

o TT 00

In r.0 tr c<

in zr ro t r

in r-

O cx 0 °

0 cs X

O cr-^

O l^ 1-1

0 IA CA

0 N

O r-O

0 r-0 e">

Ln in I

zr v3 oC

O 0-0

0 0 Lo

Vn O In

0 sS

O >

O zr »A

O rr

0 cr O

VA O

10 io cr

0

r-

X i \J

tfi Q _J 0 CP

CO h

O r v3

ry> ZT cr

rA m (P

O T cT

cn zr

in O c r

v5 r r

0 O CT"

r-01 r~

00 cn cs.

0 r-C»

cs

CP

LA cr r~

zr 0~ r~

LA r [~

O Cx> v9

cr rv) v3

cs" sS D« O

r 1 CA

CXJ

v3 sS

0 zr cr

r0 LT)

v9 r -

cr bo LO 00

Cr

t»

vS tr cr-

C tr vS r~

tr O v3

00 s9 s9

r-

X i \J

tfi Q _J 0 CP

cr 00 6o cN

Oo Oo m

m r~ vj

M

O in

O r--J

Cio CV ZT

vS in Ln

u~-vS zr

O cn

P Ln

Ln 00 zr

Ln 0 (

CP Zr zr

cn r-ro

00

Lri

cy> ol Ln

r-

X i \J

tfi Q _J 0 CP a:

LL >

LO oi

rA r -

ei

0° 00 Ln

cx

s9

O fA

r r

rs-v3

ro s9

D° O in r—

rA cr in zr OA

ZT vS on

O zr v^ 1 O

r-zr zr

Ln 0 ° hi 0 N r- v3 m o< 0

et rO ol

^V

*—'

z

O \-Z

z 1 m

O z

^V

*—'

z

O \-Z

zz I co

X

z

Z .y r-

vn rs cx

in tr cs

w> rz v5>

0 cr cn

r--=r

rs tr-r s

U-) r — CS

N S CS.

tr

cs

<-s

m in rs. x/\ cs r- ex

rH

0 c s

8? c s

Ln r-

.5 0

CS m e; 0

0 lA IV)

O t r rî i n cs Cxi rx

0^

8 r-rs

0-t r rs

s? cr

cr 0 °

Ln c-s

v.» cs' f^

vi (S>

VA

d

rS

rn

v9 tr fsA

tr cs

e( rs 0

O { £

LA. cvi CH

^V

*—'

z

O \-Z

zz I co

X

z

Z .y r-

vn rs cx

in tr cs

w> rz v5>

0 cr cn

r--=r

rs tr-r s

U-) r — CS

N S CS.

tr

cs

<-s

m in rs. x/\ cs r- ex

rH

0 c s

8? c s

Ln r-

.5 0

CS m e; 0

8 &

88

O t r rî i n cs Cxi rx

0^

8 r-rs

0-t r rs

s? cr

0^ C' r —

r s

CO (JO ct> r s

VA c ^ r s

c n

of rTi cr1

c s

v.» cs' f^

vi (S>

VA

d

rS

rn

v9 tr fsA

tr cs

e( rs 0

O { £

LA. cvi CH

^V

*—'

z

O \-Z

zz I co

X

z

Z .y r-

r <-n

O rv->

c s cvi tr

o 1 J-_ oxen

0O r-cS

-r vS cs

c s 06 C-S

Oo c s cx

CS

tr

O

rr>

ir

rr

rA

rS FA

cr 0'

rS O W\

tr rz rS rs.

rA

rS

O cv c s

rsj rA

rs O ; r~

r~ rs

Â

m

cr PA

vn rs

rS

c£ Oj rvA N

«c \ cn £

Z o CD ac cr u

o 0 \- 00

LO

«c \ cn £

Z o CD ac cr u

in O O CO

00 v?

r~ vJ

zr cs.

Ln O cr

rr rA 0 -zr LA 1 cr

Ln oC zr

r-cr

s

trr es r~

«c \ cn £

Z o CD ac cr u a

0 u

errs C"

ei 0-

Ln

0° Ln ot

0. LO

tr 1— vS

CA vi> <A

cr tr

tr tr e s

<s zr m rr

tr cs-

O 0 CH

vS m

cs1

Ln O in

cs-0 CS

Ln t r co o(

r-

N

r~-v3 CN

r~ v9 ol

\9 ( P

A l

0 on

zr >n

H

r~

O l L0

l-\ v5

v5 Lo rc

Co v5

O n

ol

~CL £

\ - f -

tf Q

tr 0°

r a

15

J a

tr

0

CD u

0.

a 1

O

in

I r> h

vj>

i 6

h

>

r2

o Q) U_

H r l

c" rs

j6 a;

)j-

c s

L i -

rs

sj 0

X

')

Lf\

X

vi>

C

-> 0

c V/l

D

v3

In

0

0

rr

•>

y

0

H e l

1 D

X

V? d

-D

r~ r~

cr 0

6 X 0 c

ol

V CJ

= 1 1

In

>

r—

u b c e n g i n e e r i n g 1 2 1 3

or Cu

CY)

vi vi ro vi

Lo vS

v3 V5» vi v i vi

vS vi

v9

vi N vS

tr-v3

rA

vi CX)

S O

O vS

c 0

U c/»

> N

O cr

lo 03 CN

IT c r 0"

t j -fV-

> r-rv i

rs 0 zr

O r- v> VA CM

\A Cr rr 8 N V) cr

t r O

cr

0 0

CO.

r-O rr

\n cr cr

V) cr cr

cr CT

tr cr LA

r-iA

vS r O

vS cr tr rS 0

r-v9 O

0-60

0

cr

r-r-cr

tr

CT

v. en £

CO

Q _ J O If)

LO

h t r r-

r-LA

c\n t r r-

Vn In Ln

r-v5

r-

t— O v5

v3 tr v3

er r

t r O r

t r 1A in S

i A iA

rA r~ I A

tr r-in

vS r-

VA C)0 vS

0° cA iA

tr

iA

tr Lo LA

r~ ty O

cT

rr

O OA VS

O tr vs

cv) CA

r tr cr io

ro tr LO

r cr

fA rs tr

> vS> $

vs v. en £

CO

Q _ J O If)

to LO

>

r--

v. en £

CO

Q _ J O If) cr

> v9 T

V O N

\J

1 1 1 1 j

at I

z Ui

\ D

O CY

Z

z m

O Z

i 1 i I-1

-at I

z Ui

\ D

O CY

Z

i

en ar Z

rs

O vS

V\ LA

t r cr

vS vi rs

0 cr rs

i-1

sr'

cS

LA r~

& rs

0 rS

d r s

O LA CN

CA

\A

rv

c~ r-—

t r r-

vS vS

cr

rz (V)

Oo H rr cs

LA

rS cS

t r r

00 r

00

r~ CA

6 cs

vn rS

rS Oo

cr rr

•r cs CS

ev

VA r-3 2*

O LA

rs tf

O rr

Lr csl

vS

at I

z Ui

\ D

O CY

Z z 1-

rs LA ro

rS

«f \ CT)

E

Z o CD

a: cr o

o 0 r -

«f \ CT)

E

Z o CD

a: cr o

u> O O C O

«f \ CT)

E

Z o CD

a: cr o

Q 0 u

CM

m o cs

cs v9 O

<rs vS CN

Cr W=

• \ A

cr rr r-rs O

O O r V)

"3- r r~ cr

1— O

C A tf

t r v3

r-r-

t r v9

rA tr r-

O t<>

er CP

O r-rs 0 0

CM

Lf)

rr

CD

C L .

£

\ -4-c?

£ T

&

ce

F

3

CS

. 0 a :

cr

c*r

V

- V)

<2

o CJ

vi!

vi

<s. (• 0 cs

•5 • •a

CH

0

) 5 C3

r

r-

> - ,

_ C

0-

1. 1-i

! 5 0 2: 0

° !

i

r

tA •3-

r V Q J

8

a'

h'

rs 0'

1

O fS a t

n

1

r

er

i

r

1 .

v9 /-Nl

r

rS

a

1 1

cr

0

cr

1

O

61 5

<Z.

(VI (Nl

CJ!

-5

O rxi

CJ1

r 1

L/v

e

m

-t— Ci 0)

CA

c TS

r~ cs.

V

4) L A

0 0

"0 O

CA

O

rv

-r 0

I

OO

J

L«

3

15. APPENDIX 2

Listing of RBCs Operational History

214

Appendix 2 RBC Operational History

Date Inf. Q Reset Q Ave. Q #1 T. #4 T. COD Ldg. BOD Ldg. Observations/Comments

mL/min mL/min mL/min "C ' C g /m 2 2 g / m z

Oct. 3 /83 52 90 •

2.346 Foaming 1 s t Observered

Oct. 5 88 150 89.0 14.0 14.5 4.002 2 n t* pumphead added tor influent Oct. 7 56 50 103.0 13.5 13.8 2.352 tubing split, cont. with one head Oct. 12 33 150 41.5 13.0 13.6 1.460 Oct. 14 0 160 75.0 13.0 13.2 0.000 pump tubing split Oct. 17 52 150 106.0 12.5 12.5 1.903 pump tubing split Oct. 19 0 140 75.0 1 1.8 1 1.8 0.000 pump tubing jammed in pump Oct. 2 1 0 0 70.0 12.5 12.5 0.000 pump tubing jammed, no fuse Oct. 28 0 0 0.0 1 1.0 1 1.5 0.000 restarted with one pumphead Nov. 4 0 0 0.0 0.000 Nov. 10 0 200 0.0 9.0 9.0 0.000 GRI bellows pump installed Nov. 14 320 320 260.0 10.0 10.0 3.850 Nov. 18 RBC FLOODED pumps knocked out, disc motor OK Nov. 25 0 510 0.000 restarted disc, covered in mud and oi Nov. 30 0 0 255.0 0.000 poppet valve broke, no replacement Dec. 2 0 405 0.000 rain guage frozen, snowing lightly Dec. 6 0 520 202.0 0.000 loss of suction, inst. 3/8in inlet line Dec. 9 345 345 432.5 13.559 feed line collapsed, growth reappearing Dec. 13 0 370 172.5 0.000 poppet valve broke, discharge side Dec. 16 310 310 340.5 13.820 Dec. 20 125 125 2 17.5 5.085 line partly collapsed, rebuilt Masterflex

inst. inlet almost frozen solid Dec. 23 0 0 62.5 0.000

line partly collapsed, rebuilt Masterflex inst. inlet almost frozen solid

Dec. 30 RBC PARTLY FLOODED Jan. 3 /84 RBC FLOODED flooded Jan. 1, both pumps and disc

stopped

Jan. 20 0 475

Jan. 24 0 0 Jan. 2 7 0 560 Jan. 31 0 0 Feb. 3 0 0 Feb. 10 0 300 Feb. 17 0 465 150.0 Feb. 2 1 0 275 232.5 Feb. 24 0 640 137.5 Feb. 28 610 610 625.0 Mar. 2 0 500 305.0 9.0 9.3 Mar. 6 RBC VANDALIZED Mar. 7 0 0 Mar. 9 MOTOR BURNT OUT

Mar. 16 0 600 Mar. 20 565 610 582.5 Mar. 23 565 550 587.5 Mar. 2 7 SPROCKET FELL OFF Apr. 4 0 505 11.2 11.2 Apr. 6 0 0 202.0 13.2 13.2

Apr. 10 0 470 Apr. 13 470 470 470.0 Apr. 17 460 460 465.0 Apr. 20 450 460 455.0 Apr. 24 430 430 445.0 Apr. 27 425 425 427.5 14.0 14.0 May 1 420 420 422.5 May 4 385 385 402.5 14.1 14.4 May 8 385 385 385.0 14.2 14.5 May 11 250 385 317.5 May 15 355 380 370.0 14.0 14.0 May 18 340 385 360.0 14.0 14.0 May 25 325 400 355.0

May 28 380 415 390.0

0.000

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

30.396 0.000

0.000

0.000 18.476 22.306

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 growth remains ammonia addition slopped

Jun. 1 400 420 407.5

Jun. 5 440 405 430.0 15.0 Jun. 8 0 410 202.5 16.0 Jun. 12 360 400 385.0 19.0 Jun. 15 340 410 370.0 21.0 Jun. 19 410 600 410.0 19.5 Jun. 22 590 610 595.0 Jun. 26 570 600 590.0 17.0 Jun. 29 450 595 525.0 15.0

Jul. 3 460 600 527.5 20.0 Jul. 6 610 610 605.0 16.0 Jul. 10 660 650 635.0 17.0 Jul. 13 620 650 635.0 18.0 Jul. 17 650 990 650.0 21.0 Jul. 20 830 1030 910.0 17.0 Jul. 24 820 990 925.0 19.0 Jul. 26 10 1070 500.0 Jul. 31 0 1050 535.0 25.0

Aug. 3 0 940 525.0 21.5

Aug. 8 520 925 730.0 20.0

Aug. 10 870 930 897.5 18.0 Aug. 14 920 920 925.0 18.0 Aug. 17 930 1220 925.0 17.0 Aug. 2 1 1060 1 120 1 140.0 16.5 Aug. 24 260 1500 690.0 18.0 Aug. 28 2 10 1430 855.0 17.0 Aug. 3 1 385 1480 907.5 17.0 Sept. 4 1520 1520 1500 14.0 Sept. 7 0 1 150 760.0 17.0 Sept. 1 1 0 860 575.0 15.5

10.032 6.732 SS have filamentous floes, normal colouration returns

15.0 7.894 4.739 SS settle poorly, inlet line cleaned 16.0 0.000 0.000 checkvalve plugged, inlet lines cleaned 19.0 6.793 4.039 fluffy filamentous floe remains 21.0 6.273 3.366 settlability improving 19.5 6.777 3.641 settlability vastly improved

8.142 5.222 settlability good 17.5 9.200 7.097 solids increased but settle well 15.0 4.293 2.862 settlability good, inlet screen very

plugged 20.5 6.969 4.375 inlet screen and valves cleaned 16.0 8.272 6.277 cleaned inlet screen 17.0 8.4 15 5.287 18.0 7.366 4.873 inlet screen cleaned 21.0 7.352 4.739 inlet screen lost down pumpwell 17.0 8.914 4.880 inlet valves cleaned 20.0 6.470 3.961 inlet valves cleaned

0.068 0.059 inlet valves cleaned 25.0 0.000 0.000 cleaned checkvalve, discs lost solids

SS 21.5 0.000 0.000 checkvalve plugged, installed new inlet

screen 21.0 2.964 2.699 inlet screen very plugged, SS mostly

washed out, 1 s t stage lost most interior growth

19.0 4.541 2.480 19.0 4.306 2.318 cleaned inlet screen, slow regrowth 19.0 4.938 1.228 inlet screen cleaned 18.0 4.706 3.371 inlet screen cleaned 19.0 1.552 1.747 inlet screen cleaned 18.0 0.939 0.517 inlet screen cleaned 18.0 1.444 1.086 inlet screen cleaned 15.0 5.472 inlet screen cleaned 17.0 0.000 0.000 replaced valves 15.5 0.000 0.000 inlet screen plugged, pump bearings

Sept. 14 0 0 430.0 17.0

Sept. 15 0 500 Sepl. 18 0 850 250.0 20.0 Sepi. 2 1 0 690 425.0 17.0

Sept. 25 0 1060 345.0 16.0 Sepl. 28 610 750 835.0 13.5 Oct. 2 0 900 375.0 17.0 Oct. 5 465 625 682.5 14.5 Oct. 9 230 0 427.5. 15.0 Oct. 10 0 1280 0.0 Oct. 12 0 1310 640.0 12.0 Oct. 16 1250 0 1280.0 12.0 Oct. 17 0 1300 0.0 Oct. 19 1280 1280 1290.0 10.5 Oct. 23 1 130 0 1205.0 10.0 Oct. 25 0 1200 0.0 Oct. 26 1 100 1020 1 150.0 10.0 Oct. 30 PUMP BROKE 5.0 Nov. 2 0 1320 0.0 8.0 Nov. 6 1 160 1200 1240.0 10.0 Nov. 9 810 1 1 10 1005.0 10.0 Nov. 13 770 840 940.0 9.0 Nov. 16 385 460 612.5 9.5 Nov. 20 0 0 230.0 9.0 Nov. 23 0 980 0.0 Nov. 27 973 500 976.5 6.0 Nov. 30 7 10 980 605.0 8.0 Dec. 4 950 980 965.0 6.5 Dec. 7 725 725 852.5 5.0 Dec. 1 t 0 960 362.5 6.0 Dec. 14 0 950 480.0 6.0

Dec. 18 0 960 475.0 1.5

Dec. 21 805 920 882.5 6.0

17.0 0.000 0.000 pump would not start, removed for servicing

0.000 0.000 reinstalled pump 20.0 0.000 0.000 pump stopped, restarted 17.5 0.000 0.000 pump lost prime, crud in the poppet

valve 17.0 0.000 0.000 changed inlet screens and pump cam 14.0 1.867 0.641 inlet screen very plugged 18.0 0.000 0.000 inlet screen fully plugged 15.5 1.200 0.879 inlet screen changed 16.0 1.290 0.945 inlet screen changed, fully plugged

pump restarted 12.0 0.000 0.000 pump prime lost, gummed up valves 13.0 6.825 4.388 assistant couldn't restart pump

pump restarted 11.0 5.299 2.342 growth reappearing on 2 n d stage 1 1.0 4.034 1.254 pump wouldn't restart

pump restarted 10.0 3.993 1.551 changed inlet screen 5.0 pump removed for repairs 8.0 0.000 0.000 pump reinstalled

10.0 13.607 8.387 10.0 5.856 2.819 inlet bucket sump not staying full 9.0 6.029 4.343 heavy foaming in 1 s t stage 9.5 2.853 1.617 growth getting thicker on 2 n d stage 9.0 0.000 0.000 pump stopped

new twin bellows pump installed 7.0 7.998 4.991 8.0 5.623 3.238 7.0 13.110 «8.664 5.0 7.156 *4.676 pumpwell flooded, couldn't clean screen 6.0 0.000 0.000 inlet screen left high+dry after flood 6.0 0.000 0.000 inlet bucket had again been flooded

out 1.5 0.000 0.000 bucket tipped again by high water

levels 6.0 32.651 '22.049 lots of foam, white growth on

1 s t + 2 n d stages ^ CO

Dec 28 350 875 635.0 7.077 •4.725 heavy growth in 1 s l stage, growth spreading, high SS

Jan. 4 /85 0 820 437.5 4.5 5.0 0.000 0.000 Jan. 8 570 770 695.0 7.0 7.5 0.000 0.000 Jan. 1 1 250 250 510.0 6.0 6.0 1.868 1.260 Jan. 16 0 890 125.0 6.5 7.5 0.000 0.000 pump throughly cleaned Jan. 18 1 15 1 145 502.5 8.0 8.5 0.731 0.393 installed full stroke on front pump Jan. 22 1000 1 120 1072.5 6.0 6.0 13.020 7.530 Jan. 25 670 1 170 895.0 7.0 7.0 4.844 2.111 heavy, healthy growth on 1 s t stage Jan. 28 1093 1 160 1 13 1.5 6.5 7.0 6.558 3.246 Feb. 1 750 1 180 955.0 6.0 7.0 3.510 0.743 effluent line plugged, almost flooded

out RBC Feb. 4 1 100 1 120 1 140.0 5.0 5.0 5.016 1.617 good growth on all stages Feb. 8 635 1 145 877.5 6.0 6.0 2.858 0.991 inlet line silted up, cleaned Feb. 12 1 145 1 175 1 145.0 4.0 4.0 6.939 Feb. 15 775 1015 975.0 7.0 8.0 3.604 1.372 overhauled pump Feb. 19 545 580 780.0 6.0 7.0 3.826 2.322 Feb. 22 225 0 402.5 8.5 9.0 1.465 1.330 inlet line plugged Feb. 23 0 1200 0.0 cleaned inlet line thoroughly Feb. 26 1 1 70 1 180 1 185.0 6.0 6.5 9.372 •6.037 Mar. 2 1 170 1 180 1 185.0 7.5 7.5 6.529 •4.107 Mar. 5 1 160 1 185 1 175.0 7.0 7.0 10.544 Mar. 8 1 105 1 125 1 163.5 8.0 8.5 8.420 Mar. 12 1000 1 100 1062.5 6.5 6.5 6.510 Mar. 15 630 0 865.0 9.0 10.0 2.986 pump and inlet line gummed up Mar. 16 0 1 160 0.0 cleaned and rebuilt pump and inlet Ii Mar. 19 1200 1220 1 180.0 8.0 9.0 5.940 Mar. 22 1 160 1 180 1 190.0 7.5 8.0 5.429 good growth on 1 s l + 2 n d stages Mar. 26 1 130 1 180 1 155.0 8.0 8.0 4.170 Mar. 29 840 1010 985.0 8.0 8.5 4.234 Apr. 2 700 1 160 855.0 9.5 10.0 3.150 Apr. 5 1000 1 160 1080.0 10.0 1 1.0 3.960 Apr. 9 11 10 1 130 1 150.0 1 1.0 12.5 5.062 Apr. 12 1 130 1 130 1 150.0 10.0 1 i.o 2.712 Apr. 19 0 1260 565.0 1 1.0 1 1.0 0.000 Apr. 26 50 1230 655.0 0.159

• BOD estimated from COD values

16. APPENDIX 3

Raw Data of RBC Process Sample Analyses

220


221

o_ r-m r

r-i— r-

r r-

l A i ^

M r ! rr! v_P

r s9 r

rvi r—

LA r-

tr r r r-

c r

r- r- Cr-'. r

r

r-Ln r

0° r—

rs r-

o 3 VJ J

O

rs rs

c" C o

tS (-• 0 ' Ln

H r vi> v^ (><

(T Ln

l

Tl' o r?

C!

n'i

1.,. c v>

0 (•'•

Ln O lo T

r N O t)

cr-r-rvl

Ln cx> rr /^J

m

01 H rs

Ln

oi

r~ r cs-

<n $

h

(VI vS c r

LO

r" r

.

re O

1 ! 1

1 i-i

cn r l o

i I; H v5 X r

co Q _J 0

LO

£

Z LL! VD O c£ J -

z.

c/>

r~ i

r •

rs X

rs r-t r rs

r O"

rs rs

— -

r-rS

! \ i O ; rs

T

r r

cn rr

s5

Oo

Ln rJ

\ \

o | r J

co Q _J 0

LO

£

Z LL! VD O c£ J -

z.

VP H

i If)

O z

i I i 1 |

rJ

0 ° LO

rs Ln

<-l <v

r~ rvi

Ln cn

Lo rr

i —

tO

a~

m Ln'

rs r>l rr

oo'rr rs; c l

rs! (vl

-

co Q _J 0

LO

£

Z LL! VD O c£ J -

z.

VP H

i If)

O z

| i

i .

co Q _J 0

LO

£

Z LL! VD O c£ J -

z.

z> i c rr

Z

z r-

1

i

V.0 VJ? rr

rs

CS r>''

6 VC q

e

co Q _J 0

LO

£

Z LL! VD O c£ J -

z.

z> i c rr

Z

z r-

i

i i

i — i —

6* Lei rs

vS

rs cr

... er r<">

o i~

rS

Ln r r r-rs Lo

vi CO Lo

croc"

m VB fA

CY) rr Al

r--

£

Z o CO or

cr

u

o o r-

Ln Ln Lo

m rS rs

m (V) rr

r~ r

cr o rs

Lo CP

rr cr

v9 vS

cr cr 0

cr r

rs O

cn X

v.? v J r-

LA rv

i i — ir)o 1 !

[V)

r--

£

Z o CO or

cr

u

u> O o CO

t— O Ln

v5> ry>

v.9 rr

u v.? rs cr

cr- r4 f\

O vn r~

rsl r>'> O v9 LA Or. v9 r-

rr

r--

£

Z o CO or

cr

u Q 0 u

o vS Ln

rs vS

cr r-CT

r-errs

Oo i n o

vS O

cr rs r<

O <r

•a-r-rS

Lo rv") O

rr O O

rr rs ^=

cr r~

rs

r r

(A 0 rs

r 6° 10 \S

v5 cr r- VJ> - r VJ

v9 P Lo

N LA

Q .

£

—f-

Q

tr

o rr OJ

i.

?

UJ V-&_

5

I ) 3 "5

rS , 5

cl

Ln

) 3

h

Lo I 5

r

s. <j oJ. QC

10 5

al if 0

~Z

u

. W

cr

i r>

CU.

La

It*

) 5

h

0; tS 6

- r

1

r* rs

' J

r>

W

VJ

nt ) y "i 1-

cu. _£

VJ

rS rs

i r

r

rV <f

5s . 5 5

r>

ex.

i 3

v>

r r eS

rf 6

Ln

' (5

o» a -iii o>

Ln

5

% -V

r

6

d.

a~ UJ

U

r "

V cf o -V

V

6'

u.

O

rJ

6 5

<£.

a) c

cr;

><5 5

s2

a -L

V,.' 7 ^

<r

. r:

ii-'

C d

-r

<J> 6

-2

zx-

&

V? rS , ej1 3 <z

o> cS 6 -V

VA

1"

rA

6

<j w

c

' c <z

cr c5

- r i''

z.

rs "5 VA

o~ a>-LU

a?

—1

rs

"5 a/

Vr

I -r

— r ~5 V/i

a *

cy a?

r—

»'

VT cX 6 LA

r —«' Vi 0

V')

3 -JJ-u>

cT

<y VA

cl


222

tr Cu

0 r"

v9i v i

tritHrHoo v i ! v i ! v i ' r-

I i O r-

vS v9

r~ v i

LA v i V5

r- cr vsj v i

rO v i

rA v9

cr vS

c<q !""

C v ' l

* i

r O r-i

c"

0 f V

v9 v9 r-l

ro

rl

Ln! (Vi rJ H

LA 0-ol rv

oil ^ ! tvij rsi I

cr A1 oJ

rr cx j C M

LA rr f N

tn L0 H

r-T v9

O L n Lo O l

ro v9 LO rS

J CS rl

OJ cs

rO (VI LO rs

r~

>jl v9

r~

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

LO h

o-CO

rA cr LA

T

cv ol r

0" r

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

to LO >

(A rr-r-

rA r 1

q vs \ o. vi>

O rr co c-i

O

in" 6

in Lo Ln v i

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

LO H

o

ro

r~ r lm o rr \ Lo fO

Ln r-io i- :

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

Z i ro O Z

q rvi O*

rv r-

7t~

rr O

v9 cr r

o4 LA 0°

oo c oJ 0"

q 00 r~

ol vr> r~

ol VA r~

rr v i i —

CO cr r -

ro t> |-

q CO o=

q VA rr

q do

0°; ro v i

£ .*'• j 6 ° i

rr r~

v9

r

rr ol r~

VS

r v5

I -

rr 6 r-

Lo

LV

Lo 6

rs ro 6

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

ZZ I <n X

Z

Z .y r-

oJ rr r-

r

r L I

r v9 Lo"

q cr V.

o rr

cr r-N

CA

rv rr Ol

CO 0° ol

6 Ln

CA Cr

0c

LA

o t> Vn

rr cr vo

oi

rr v9 d rr

rr ol ol

0o rr

rr ryi

VO o °.

co IS

v9 cr

UI

f vv

CO D _ i 0 LT)

/o ? - — '

Z LU v£) O CO.

Z

ZZ I <n X

Z

Z .y r-

c< r~ O r-O LO

O rvi

ro «i

CP v9

o r-6

rn rr o!

co C Ol

rr-rr cs

oi ro

q rr Ln

q LO LTl

LO r -

b°

LA v i V?

r-r-LO

co Ln"

o! N ~ k rr! C A

i i

q LA ol

q rr

vS ?2

v5 .•Yl

o (VI oi rs rs

rr cr

LO" r-; L-)

oi

\ CJ) E z o CO cn cr u

0 O V-

r~ In Cw o

N ol

c>o (Vc ol rr ol ro ro r—

ni rr ol OA

oJ ro

VA rv|

LA CP v5 — — i

r o— ol of

rV) r>i

CV) LT)

r~ LA vS

rr o :

cs \ CJ) E z o CO cn cr u

Lo O CO

K) VA r rvi ol rr rr ol

\ CJ) E z o CO cn cr u Q 0

<j

Ln VA rA rr r—

0 v9

(V) cr

rr —-0 q

o ol

ol O rv> o| VA rr

I A ol O o q r- rr

o- ( S

o» O

r~ r r iS r-

ro rs

vf1

L'» O

(VI

o cr

c°

ol e£

CV) ot

cs rs LA

Q i

" Q .

£

\ -t-

Q

d a £

v/>

e

rA Ol

-5 VO

0> 0

f

-rvi

" € a'

\A

cv 6 -r 1-0

o'' o>

i

LA

A J .

a* u. 1U

0

(V'i

3 LA ft' O O V"

cv -*~-. Si

VA

1) , CJ

e

rr

0

0' c -v v. •L

A '

T3 0

Cf C

- f

LA

0 6 -r

n

8 0 0 LA

rr

lo

! O' 0

"£ H

VA

-/**

0

»;

r "? 0"

r

-ti G

s

r -<r <s

^ _

O l

V c C

- r r

oJ

'l

o c c —V I"

-1-0

1 f-

1.

KJ

0

a< eS 6

rr

ij 0

0

t

—V

o

cy c 8

r

3

C5 To n

r

-f O

i> ' c cs

rr

r~

o

I"

O

6 i. o»

<r

•V

o

•b ro

cr

O

o l

~T>

0

?•

oJ

- f

u O

rr

rr

_c

rr

cy

D O

2

rr

cr

u cy O

rA

u IV

O

7.

/ s ample C A R B O N O 9 / - O N I T R 0 6 E N ! S O L I D S O s A O C o no!. pH Ucvre / s ample C O D B O D s T O C T k N N O 3 - N T S S T V S S T S pH V * Dec. \i 32.7 2 G . 3 . 0 . 5 6 .

Cec. IC q 5- 11.3 ! H . l 0 . 2</ ii D e c ic T l n.7 17 T O. 3 2

I 9 1 - Dec. 20 \12-H ? 2 26.S 22- / If b e e 2 0 \\Uo TM 2S.9 •2 .? V D « . 2 3 120. C 1 2 0 / I " MO. S T . ?

j * l» Dec. 2 3 102.1 501 / "10 HO. /HI o.n

o.n ?i FeW. n 720

Feb. 21 27? /zn 0. 32 • * FeW. 21 270/201 ?Xo O.ST

230 /1?0 O. ISii few. 2/y 112 >03 C ' /

•• FeV>. 25 230 <o. 1 F-'T. 2 : T . 6 H<oO 52 <o.i

1 * " » M a / . 2. n.s Hov. 1 IS? 20s- 23.0

I5' U Ma*'. 23. 313 / 20I IG5 30.4 /31.0 2^.3/22.15 l» He / . 23 IrJO 2JL2_ to 20.5/20.5

"I" M fipv'.l C 212. G l fa/.o 230 S'5 S3-I

tl 250 101 II..5 IO1?.! '» 2C0 1 H 117.0 in.o 101.«/

10? 1 n 0.07. KJfy.'i 13 113 103 52.0 55.2 O . 0 2

l« ns n.sliro/0

VI V1'1 n mi ! 13 31.8 35.? 21.0 /25.0 11 » W M tl : 2 0 1 2T !

' I T ? 2^./ l\ 196 3t9.3 0.03 II

\H5 36.6 0.02

ro

u b c e n g i n e e r i n g 224

CL. PA r- r-t

IA VA

r— ... — ..... ....

r r ; r^

lA C > r~

0

D~ VA r< 03

in r-

TJ ' C U°

«/>

o-vs Oo

Oo 0

vS

(~ — ...

-

£

CO

Q _J 0

• - - - .... ••- — ...

...

CS

Ln

O

In

OA Ln t r

r-cX t r

£

CO

Q _J 0

CO CO > - — ... 0

cr

ol tr

r-

r-v i

>« LA

i tr SS

rs Lr.

0 LA

O fA

cs £2

£

CO

Q _J 0 L/>

L/> h-

- —

- .

. . . ....

-

0 vS

^ S v S

<9

vs in

0 6 rs r— "tr

0 rr rS. rs r-c^

sS tr \s

rs

q

0

CA

00 rA c< xr cn cs

? £

z LU \D O ol h-Z

Z i A

O z

rs)

O

o

H O

9

Ln 0 •

O

o

-6 - O

6 O 6

-••

0

m

O

. . cr co

O

rS

6

t r rA cs

t r Ln

* r-

- c r

6 cs

In fA rs

- • • ? £

z LU \D O ol h-Z

2 i <»

zr Z 0<=> S9

t r

rJ C» t r

O t r xr

vj" r-(VI

x_r> ts-tr

Vv cr xr

VA

S — xr t r

t r

V\ rv\ ^~ m

xr

o rs

Vv o t r 0 A

X

r-1

OA

c r

r-V \

5-

O xr

r-

•5 b°

CA

0

C-"> i— ro

v9

—

? £

z LU \D O ol h-Z

z V r -

cs S S LO o

U i

cs r-

o cr t r

CS°

t r

vn XT

O LA t r

tj» o r-

vS s9

i—

c£ VA vS

o xr oo xf

ba

O o=> LA vi

Ln bo c5 r—

•xr r^ r—

rs

X t r xr

O •xr

r r i

r 1

r^

r— rA

X t f

r-'

vS r-t

? £

z LU \D O ol h-Z

z V r -

cs S S LO o

U i

C» ro vn r— xr

vP vn XT

O LA t r

tj» o r-

vS s9

i—

c£ VA vS

o xr oo xf

ba

O o=> LA vi

Ln bo c5 r—

•xr r^ r—

rs

X t r xr

O •xr

r r i

r 1

r^

r— rA

X t f

r-'

vS r-t

CT

£

Z o co err tr u

o o r -

OO LO

v5 oA

Lfl t r r»"> rr r

5

Ln r — O r~

o-Oc

r -6°

La V.

La f>

H vS

vj> LA tr

r -LA r-

0

rA CT

£

Z o co err tr u

10

o CO

-• — » -

LL

-o T> m

«J

Ld

CS Lo rA

0 0 cs:

CT

£

Z o co err tr u a 0 u

cs v3

cO *

c< vS t r

r^ & rJ

<r vj> xn

cs" SS

r~ = cs <r-a~ <r.

in

E;

r y \

OA

rs v5 o

LA r

H lA

cr

CA

6 cr

nr.

0° 0° rs

X

vi» CV) _1.

cr O

L0 0 rs H

m \? (V3

t r r~ r J

AA cr rs

Ln 1—

0° LA H

CL £ tf to

\ *>

• t f

O

r

rs

' £

a: «' d

r rS

...

T

•3-

1 ?1 t r

t r

r

T

r

t r

Cy

> 5

i *

0°

x

15

i r sr'

XX

r

f ©

- r

•£ t r

LQ

'i X

Ln

< 8

TL

3

tx

1 s L0

i

zr

> 5

fSJ

i 5

X

v 5 r

c* H

H

» 5 X

X

ri ri • 5 6 2Z

-o _at

xc

in

t_

rs i ? _c

T c

xr a;

(/> i •' X

LA H ) S t>

r c5 6

7

LA r s

)5 6

6

th •i T

LA r s

2:

Lr> rt 1 5 a

r V . '

1 t r

t r r*

p ^ 6

0' c 0

Cr *-v'

C

cr.

^ _

cr

1 r JT>

/ s i/>

1 1


zr cu

L A

r- - rr-«

r-

U r-"

cs C>5

VA CS Oo

.... -i A

rT 60' cr in

bo V) cr 60*

u Oo (V>

0 °

L A <y r-t DQ

-

-5 ' c tf

^< Sv

£

co a _J 0

s> .... -

-5 ' c tf

^< Sv

£

co a _J 0

CX -r-C» rr

1— V.0 tr

-OA O-Tr

CO 0 ° rr

. . . . O rr rjr 1

O r tr

r-rr rr

CN 10 tr

s9 OA tr

v5 La rj-

O Lr> tr

«v5 Ln tr

rr rr

C-N

O LO

O rr

-5 ' c tf

^< Sv

£

co a _J 0

CO CO >

O

O rv>

rO r-j

O

t r Lr>

rr

rS S9 — '

. . .

Oft 1/5

OA L A rr

s

VA

v9

rs

0

0 rsi LT>

CS H

rr rS

v>° \ sS CP

O csri

O rr OS

0 rs r-OA OA r-

r rS

ry> Ln VS>

O) Ln \ L A

\> <"S

^v£ r-

rz

liv

r~ rr

r-

OA CV1 Cv-1

OA r cS

O i-s"

rsl r—

S9 rr

v3

b » OA

0

S3

on

o° cr 0 0

0

<s 0

-5 ' c tf

^< Sv

£

co a _J 0 v/>

vV> r-

tr <N

X T rS ~ l

0 -<-i f— W.

tr VA CN

vS of. CN

O T-cs r s

r>J

O A LrV .V9 O

°i SJ>

s CS

0 s9 rS

O O CS <S 0°

O 'CN

OQ

O 5 s

| rr cr

<y>

rr CA

O

O rs t r

r— CN

r rS

r-

r 0

r~

r-0

( A O

r-vi

I — rs

s

fA — S9

r I A

v i

OA

vS rn

O rA

VA

rr rs

0 r >

q rs

8 c s

tr O cs

t r OO oj s 9 r S

£

Z Ui VD O ad {-

Z

Z 1 to O Z

Oo CN -

VA

OA

CVi

O xr "fA vn

.... : Lfi

OA V A

rr s i

L A L A

s~ LA

s5 CS

t r

s3

O r<

CA

r-O r

r

r~

• - 0 LA CS

Lri

VI

vS

vS v i v i cr

- Lo cvi

r r t r £

Z Ui VD O ad {-

Z

JZ 1 t»

x t

Z v9 O <5

V r 0

• 3 -0 6

. vn r— cri

xr O O

Vr,

F-O" 0

b 6

0

tn XT vn

.CD vS d

• » O

cr O •6

bo

vS

6

Ol O

0— ti-i

r; rr,

rs rS 6

fs <5

yo 1—

Oo

O

cS 0

vn 0 0

£

Z Ui VD O ad {-

Z i*

r-

OA,

oi o f OO O

sS r

O tr

Of tr

q rr cr

-2 0-

rr OA

cN CO.*

r-L-i

v ^ cS

v5 tf

r— r S

rS r—

0 r-'

cr c-i vJ9 ci

cn E

Z O CO cc cr 0

O O f-

rr O S S

rr Ln rr

r — LT>

s 9 s9 OA

H rr

r— s9 oa r-

0-1 t

r4 r J

vS v5

v5 IT) cn

E

Z O CO cc cr 0

u> O O CO VT)

L O rv [NO cs

oA

r O a

~ r

1 LLT

o~ t r

rr

CV) Ln

rr

—D

XI (U

Xr-

cn r-H

r-

in

s S

00 O tr

cr

rs

v J

O

i£

SJ? tr

0

r~ O In r -

lS s ^

1— r

s

c r VJ9

cn E

Z O CO cc cr 0 0

0 0

v_!>

cS

In cS.

r~ cr rr1

FX H

r r~ •O

0 •rA r 1

cr 0

in (V\ 0 cs"

v5 L A

O r-

0 cr LO cr

r~ 1 vti r— s?

"I cr~ S 3

cn rr 0-)

Vi c~ v5

g cr~ Lo r S

r cr xr vs v9

r S V 5

cs

(V.

o l O Ln

r-

cr

s ^

r r r

cn rr rs



•i -XL

LA a)

i i

> °

• J v/» 7>

2~ xr

T>

xr VA

y

*-

-p x r -

0)

rr

t>

r>

cV 6 6

ft"

rr

Q-x r

V A

in

"6 _y x s

v/i

rr

r l

A

rJ

1 x^

. ...

£-rr

XB, 0>

V^

*1

T> 3* a?-V iy

r r

' x O

0/ s.

<y cf C5

5> rr

- 0

a) vy>

1^

Q'

XT'

V

5" xr

cr

% 6 h

cv

11 ^' tr

~r> »• xz <D

VA

a' XJ? 6)

V

t r

CS CS

>

1 L* IA *

x r H

a* -s

»' l n

J x z

XT

v5 cs CV

b h CV

L * tr

-t> »

5. t o

t> a.

3 ? cu V?

i -xr

errs

e rJ -r"

+- • r ^ rr


SA c/V c4 Oo 0 s

0> rr t r

t r

c s cr (VI

tr (VI

CO Q _ i O

CA vS CA

cr tA is rA Cr V9

o

v9

cvi

r-tr.

O r o r s ;

vs Ln cH

"3 -rs

e r r

<s

VQ "cvi -LC\ V5

s S

(-1

o .

cvi

I CS

0 0

0 /

Z Ui

O at

lV> r? cs cr

Ln

LD o o"

v9 tr cr cs

00 LO

cr rvi

cr errs

r r

q 0-1

tr CA C>A

\ CT) E

Z O CD CH cr u

V? tr O

v 5 >

cr vs 1A v9

KJ PA

t r r r C>°

tr Vi^ tr in

cs; O rA in

iA r -CyO

iO in VTJS-/

v£> Cj rvi Vi1 Ln

IA LA

Q.

£

V -+-tf

l A

rA i A

ft h he

a/ 3 : a)

VA

0) ,3

t r CP LO

c r

(A

H - iy)

3)

h 1 "OJ LA

95

tr cN

C

V.A

Lo

5>

h

J UA

Si rz

0 LA a/

£ 6

tri t r

cs

c? rA

C/Ci

1 far VA

O I S

S

LA

OA

5

3

h CD vA

l / 1

0 —

rs.

i

~6

•X

LA

^ ft; ^


tc o_ - - cr

r~ Vi c r r—

O O o°

. . . o-r~ (X)

c r 6o

CT r

rJ cy

0° So -

~s • c o U

7 t o

—

/—

\

£

to Q _J 0 0)

•

CC -

v5 Cr O o tr o cr

0° V A

v i

vS tr o tr -

O cS tr

vS r-(V)

VS3 O r r

tr v9 cvi rr

C S

v^ CV)

O tr

o° V5 ro

rr C A

oC cr cS /—

\

£

to Q _J 0 0)

t o > r-

pl o-r—

c s

i—

CS3 tr cs

CN

rr o r r

.o <V>

s

o r

V3 tr t r r T r

rr-bo

to v S

v9 cs tr o <

r w

r-r—

T r r-

V A o O f> o

"Tr-

vn

C A Ln

rr cy O r"

vS

r

r r C O

cr cs

rr cr rr

r

C O Vn

v5 rs c S cs

O

0

o-O \ rr Cr

r-o

cr

b O C N

O S

/—

\

£

to Q _J 0 0) i/>

k

§ CS

o< CS

rr-r? 1

vS> r-vS>

r-

v2

•3-Vr>

rr

rr o rr

r S o) CS

t r

o4

o VO

ra-v9 VA

tr

O

iV/v

V A

cS cr O cr

CA

O S .

v5 C S r \ A

O

•o o . o c s

r-

0° r" rr rj

r -

1A c S

CJ^ o l o l

O V9 O vs

C A

o r S

O i —

o -

oo A A

|— V ?

r— (VI

Ln'

C A

r A P -

o r-

o r A

Vn* C A

rr l A vS ro

V A

cr r v i oJ v5 ( S A

Of

r

z Ui vO O

Z

Z i (0 O z

V I

rr V )

t r

u\

<o r r

vh r i rr

. . .

o C

Cr r A r S

Ln cr 6 C N

rr L0

cr e<

r-co"

O cs-

0 c

r-i —

LO oJ t~ <s

q cr-cs

z Ui vO O

Z

I «» rr Z

o in <Q

cr" O 6

r r

O d

cr c s .

O. rA d

v9 v5 O P

vii d

r~ r-(T r~

c s

r" 6

O

o

cs o l

c s S cvi

tr Ln r— r r

J — cr O

ro 6

z Ui vO O

Z Z

r-- cs

C O

tr rr

- cn r~

r A r~ Ln r~

CA

rr rr cr rvS o r v i r S

CP r v i

&

rr; r s

* CV)

rr1 r o

£

Z o CD CL cr u

o o k £

Z o CD CL cr u

in £> O co

c s O rr O cs

s

o l vS

v9 \ r

fo CN

r-

vo

o L O

cs ^ i r s

r^-

r r C O

O r rr 00 r—-

cr

cH

' 3 .—'

L ' ) O

rA

cr r S

r • L ' l v..

D

s rr (V) fA

r-'i

C A

'

cs

& R v9 r s

r s V . '

LA o4

rr i

CO r r

r S

'H V y

v3

t o

£

Z o CD CL cr u

o 0 u

zr cr O

fA •-.9 r.' ?"-} r- CL

C S

CT>

A

O cs

<-~s i— LO v./ r CA

v S

cr rA P A cr r > N

V >

rr-cr > cr r-tr i— rr

^ s o, i A O O '

0"

err s

-o a; i— o s r— o ro ol

CO

Lo r

C O

r-t ~

LO

r rvi

/ " ^ I—

vt5

cr v5

CS> rr 0 °

*—\ vn v5

rs C-°

—1

v3

cr r~

4) Q. £

\ V

-+-tf O

r A

S

$ -r (/>

I

T> te vo

CA

n

CJ C5 d

t O

7f

i _ .

rr

to

75 3?"

ty>

cs rs a ti d

1.1 V

\ "

rl-

•

o

ri

I r

s

i C

-D

V

f "

f

r r

cr

c f

rr

~CJ a;

3 t a)

V A

if' rr

r~

t>

a1

c

L

1 (V

\ A

v>

*•

rv'

S

^ _

3> ry Li

/ tr

o <

6

a: C J cf 0

l o

rL

>-

tr

~U

-J£ o)

VA

1h_

£| rr.


rXT - 60 -• . . . r-

cS o v> r-

o Cr r~

CsJ

Oo 0o r~

o r-r

q r- -

TS ' C tf

t/>

...... — .....

f

CO £ Sy

CO Q _l 0 tn

«

cr -•

s3 0r cS

0° C S

cs CO O m

- -vS CS r r

Cxi r-rn

O s9 LA

<N Ln CS"

O r -

cs Ln

Co o o

6o O bo

sS L A L A

f

CO £ Sy

CO Q _l 0 tn

CO CO >

1 -

Xr

\

0°

Cr 0°

\n O

-

0=

o oA

CN)

rr

rr

rr

r f

rA

O v9

CD

OA rS I— vS

q ss

q s i

o

cn

eO LA CV)

O

O A

cr

OA xr rA

XT

o <6 X3-

r-rf

r f o o

0o LA

on in

~r r

s5 LA

r-

S 9

O

<r

oo

errs Cr

o o

o

r r t r

O «r-

r cr

T—

s oi

f

CO £ Sy

CO Q _l 0 tn

cn r-

cr xr CA r—

*n Xn

tr OA

O <S

O rs

O T-

OA S 9

cr •r-r r vS

0" vS

0° vS

o s i

i A o=

O rs

O

r_

6 bo r S

•c^

00 OA

r-s rA r~ cr>

L A s9

OA LA"

Vx <o rr fr i

r~ o° OA CA r - t on

| s9 oo

V> r s cs

rQ

— o c« OA 0O r-X-

VA oo cs rr OA

<rs r>-C V )

"o vS AA

r-

r— 6 o

rr

vs. rv\

cr L \ to

o xr cr CvO on

r — bo" 0° r~ Oo ty>

•s

s ^

Z i (O

O z - :

in rS Oo OJ

O O rA

-TS

q an cs

--

q v i

q S S

X3-

r— q in L A

CS Ln r- LA - Lh

cS cr CS

- LA r r Cr ot

CA

<0 OA

•s

s ^

Z i (O

O z - :

in rS Oo OJ

O O rA

-TS

q an cs

--

q v i

q S S

X3-

r— q <> CA

0° r-c S

r-ol

Lh

cS cr CS

- LA r r Cr ot

CA

<0 OA

Z Ul

O QC

Z

2. 4

X Z

LA OS

xr

< V\ H

6 . . . .

... J— U\ Lr\

r-•xr"

r— 0°

fS

o o vi) in

6 q o

o O

cn O o

r— O d

LA O d

Cr O c

Ln 0 d

In

6

Ln q d

Z Ul

O QC

Z

2. 4

X Z

LA OS

xr

< V\ H

6 . . . .

...

O

Lr\ r-•xr"

r— 0°

fS

o o vi) in

6 q o

o O

cn O o

r— O d

LA O d

Cr O c

Ln 0 d

In

6

Ln q d

Z Ul

O QC

Z z-

r -

-O cn

• ^ rJ CP

cS*

-• s9 Ln

xr cr

q xf

rJ TT rn

r OA

of cs cs

vS CS o m'

cs rA

q r r a

on r r OO

^ v

£

Z

o CO or: cr u

0 o V-

- ......

--

•-

^ v

£

Z

o CO or: cr u

10

CD O CO

in x r

2

X3^ r r

in

rr

XT V

cs cS

r r

'LA C

cr

xr • . . . . .

v3 s5 s9

S

Lr>

rs

v5 CS ol

CYl

37

rs

r r

in

01 Ln c*

O

3

cr rA £N)

LO v_ •

o T- sS>

OA

xr

v^

tA

^ v

£

Z

o CO or: cr u o

0 u

S> rs

s9 ot |

cr

v3 LA

r r r*

cn e>s

i— 6S

. • ~ \ V9 s9 -> /

s9

/ •—s 0° in

ol v5>

rr-c>r r—

/•-—> xr v9

s5i s5>

r i q

xr CP

LO

-xr

oi in

o-s9 r r

xr

00 LA

r r

rs S ?

OA xr

' o xr

it

"o xr

r 5

LA

OA cn

Ln m

O XT

Cy> LA

xr xr

O LO

<r-m r-

r

rr

CP s9

4) CL £ tf

\

"tf

e

sz a>

D

\n

r "V .0)

XT

j?" xr-

rA

i <zz

i. LA

L ZT

~T>

a) v>

s-

5" TT

xrr

ai

VA

• Q>

8 xr

• y

s -r-

i

(5 ©

rL rr

t>

XZ 0)

LA

3.- if r r

- i a) L A

lV CT 6

-tr LA

7

1

£" xr

LA

i/i £~

X l-"i

0' cn C5

IA

&

r r

~T>

4--

a?

1 l A

T> of 0'

5

-i

tf &

CA

L


a: ir •- I 4 -- ... r-; q do

r 1 cr r 1

r; r ~ °T r

Vi r~

fS 03

-

' C tf

«/> s

\

£ s v

tn Q _J 0 cn

0 8 I V :

cs cr rr —

vS In

bo rr ol

rr-O vS

tr cr Cs/ O Ov

o( C

CS

C A

rr O ry)

v3 on

csl Cr rS

\

£ s v

tn Q _J 0 cn

C O

to > r-

rr t>-tj-

CO rf x> ro r ~ xr

i — TP

r -

rr

O A

r-2

o vi

r A

ro

C A

r-

q

w" tier rr

o -5 o r¥

r -o" vs

O o cs

C A

V) r ~ v 5

evj tA CN

cv>

rs

...

O o o Q

of) f ry

o cs cs r-

r~ vi

r A

L f l

O co

O cS

C A

r xf

r -

rf r-

6 OA

<v> Ln

oo L A

ry>

IO

Ln

O 6 q 6

\

£ s v

tn Q _J 0 cn

C O

to > r-

rr t>-tj-

CO rf x> ro r ~ xr

i — TP

r -

rr

O A

r-2

o vi

r A

ro

C A

r-

q

w" tier rr

o -5 o r¥

r -o" vs

O o cs

C A

V) r ~ v 5

evj tA CN

cv>

rs

...

O o o Q

• O . 00 C S

rs" cs

r-

r~ vi

f A

in O c>

O cS

C A r xr

r xf

r-

6 OA

<v> Ln

oo L A

ry>

IO

Ln

O 6 q 6

\

£ s v

tn Q _J 0 cn v/>

vo f -

b° S9 ro r— r—

rr L A

xr ln

o L n

C O

rr

O xr .ss r~ tf

r-d

O

C A

L A

r>0

O cn C A

cr

O 0 °

O

C O

S3 cs s S

C A

O A ...

o C S

ro "o rf rv\

s r—

?

<VJ r ~ & Q

PA V A C»

r-

r-

Si

r~ r s

cv> r -

q xr r ;

rs

O A

rS CD C P

OA 1A rs

X3-cS

-

O A

C A

CV)

rS rr a

CO

rr

r t £

|SJ i —

\

£ s v

tn Q _J 0 cn v/>

vo f -

b° S9 ro r— r—

rr L A

xr ln

o L n

C O

rr

O xr .ss r~ tf

r-d

O

C A

L A

r>0

O cn C A

cr

O 0 °

O

C O

S3 cs s S

C A

O A ...

o C S

ro "o rf rv\

s r—

?

<VJ r ~ & Q

PA V A C»

r-

r-

Si t*l r -r S

r r

q xr r ;

rs

O A

rS CD C P

OA 1A rs

X3-cS

-

O A

C A

CV)

rS rr a

CO

rr

r t £

|SJ i —

Z U l

VD O r-Z

Z i fO

O z

r A

• cS

in _ «

est

r-•

C S .

in :

-L P In

Cf _ .

Ln cS cn

In r J

.rr O • »

C A

cs -Ln O

s9 q rr

Ln «

r-

- cr rf

s S O

Z U l

VD O r-Z

2 I «o

a: Z

L A

6 b

6 O

o J 0° O a

H e

O

v9 0 6

xr •

CP 0

6 CP Ol 6.

X3-

xr

6 9

cr\ s5 rs

^

cs o

o d

Ln O

rr rr cf

Lri

Ln

s9 rr sS> cf

rr Lri

r r L A

Z U l

VD O r-Z

Z

r-

o~ rn rf

rr I -• -

r< cs

c cs n'

in cs

cr cs

r- r-cs

&> rr cs

s3 V

cr CP r-

• -

01

£

Z o CO cc cr u

O O V-

-01

£

Z o CO cc cr u

10 O O CO

rO rs N

I>A

s * cr cr

C A

rr

CP XV

3 ol 8

r- .xr

'in

r s tyi

rr rs

o rS

V/v

In v ;

CS O

r

cr Ln

in lA

' N • s , ;

cs r-vS

LO C A

r-

r s rr rs

Ln

73

O

8

01

£

Z o CO cc cr u

o 0 u

CV) 0^

CO t*>

i n

V?

Cv> LA

I—

rr LO

rr C S

L7) r r

rr Ln

5-rr

cs CP

oO rs vE) rs s ?

t— xr

<r4 s S

r-L-v v^

r — XT",

o= SS

cr x r ,

IS) u\

r-r- r r-

LO

I—

a> LA LD

t T )

s9 CA

. —

L I

Ln Ln

Ln v> r-

I i i

\9 r-

xr> vi> y cr-s S

m

t— xr

r-r- ol r~

^ s O J

CS 0Q

§ s S s S

V CL E tf

CO

-+-tf

Q

L A

cs

ts Q)

V/1

LV

g -r

i •

•f rr

J

ft' Vy

Jf xx

oV

-s net) W

0 0 C

- r L ^

•V~ v.

-

rr

a) ^=

rr

cr-a |

ar rr

1

5_

i r -

o cv 0

V'1

*t

7

3

rr

-D

x±-

i

J

cr •i

o

"5j_ •£

T> a>

<J) L^

rr

rJ

V

O

$ 6

v/> ^ .

»

-o a ' tt^

rr

vS

~v o

s LA

,cP 0

V

L-A

xl *' rr

<r

o

|

V

~b t»

xz V /

rr

V CL E tf

CO

-+-tf

Q

L A

cs

ts Q)

V/1

LV

g -r

i •

•f rr

J

ft' Vy

Jf xx

oV

-s net) W

0 0 C

- r L ^

•V~ v.

-

rr

a) ^=

rr

cr-a |

ar rr

1

5_ j "

7

3

rr

-D

x±-

i

J

cr •i

o

"5j_ =1 rr

rJ

V

O

$ 6

v/> ^ .

»

-o a ' tt^

rr —j -J

V

L-A

xl *' rr rr

u b c e n g i n e e r i n g — 230

Q_ IA cr r - r -

O ro . . . q V A CM

0Q . . .

r~ iA - • Mr

r r t Cr r1

CP r~

O (¥>

°? r -

0

cy5 -

-rs - c o

U ..... —

£

t/i Q _J 0 f l

•

<E 0 m

0»

vs b* trot

in CN

--o 1ft

VJ

— rr OA

r s rA r- >A 0

CA

CP r* rs

~T~

rA

sS> 0 ° c M

O LA r s

rr rr rs

£

t/i Q _J 0 f l

CO CO > r-

T— CS «A on err s

r-*

rA

q rr-r c x

r-rrj

r— cs

CO rA

CA N\

to JA

r-7 t£ c s

0 <D

O r-

Tr r-ro Vo

vr> CA rr

0 •3- O

\A 0 v9

1— CjO c s 0

rr CS

r? r-r s

0 cy= (vA r 1

PA

O &

r-

r-S9

0 V9 O VS

O cs OA

r~ CN <N.

r-V9

r-

£

t/i Q _J 0 f l

CO CO > r-

T— CS «A on err s

r-*

rA

q rr-r c x

r-rrj

r— cs

CO rA

CA N\ - 1—

vi LA <s

O

$ (VI cr

CH r-

Q r-

oo (vA

0 0 6 rs

O \A 0 v9

1— CjO c s 0

rr CS

r? r-r s

0 cy= (vA r 1

PA

O &

r-

r-S9

0 V9 O VS

O

CN 0 d CS

r-V9

r-

£

t/i Q _J 0 f l VP

to

©

rr rs

r-O

S

rr

H CA T< rj-

t-r <N CH O r s c s

rvi

•q

r—

O O 0~_ O

•

OA 1— 0 c s rs

---

0 0 O PA rr

O v i r c CA VA v-c

O

O r s rr

r ~ v9 c s AA r -c s

— OA CO bo rr vs iA CA c v i rr

<3

r-r-

rvj tvA

O rr-r -rA CA VS

rA iA VA

sO

>•

r-rjj rv» CA

£

t/i Q _J 0 f l VP

to

©

rr rs

r-O

S

rr

H CA T< rj-

t-r <N CH O r s c s

rvi

•q

r—

O O 0~_ O

• rr c <

LA c^

rA O c s

--- (VJ r-|V>

•0 rr-r

0 rA

O v i r c CA VA v-c

O

O r s rr

r ~ v9 c s AA r -c s

— I--rf

CA r-v9

rr vs iA CA c v i rr

<3

r-r-

rvj tvA

O rr-r -rA CA VS

rA iA VA

sO

>•

r-rjj rv» CA

£

z U I

vO O oC

z

Z t rt

O z

K

\A r-

0? -

OO rA

rr r-

& O

cr . . .

c s VA cs --r

O v< rs

O r i rs — v9

CA VA \ S -

r-U r-0 . . .

VA CO

0 b o

Vo c— r f

\A

£

z U I

vO O oC

z

Z t rt

O z

K

\A r-

-

OO rA

rr r-

& O

cr . . .

r^ •rr

C« lA •S;

O v< rs

O r i rs — v9

CA VA \ S -

r-U r-0 . . .

VA CO

0 b o

Vo c— r f

\A

£

z U I

vO O oC

z

I <a

X

z

<H

Vc r 1

co c s

cs

IA r r LQ

rA O c s O

CA O O

O

cs c s d

in O O rr r^

O

L"> cr ro

s5 O ra-

tr-VA •6

V

>

T77 CS O

£

z U I

vO O oC

z

I <a

X

z

<H

Vc r 1

co c s

cs Vn in

CA CS. d

CA O O

O rs O

CN

6 0Q CO*

rA Cr O

CA t — CA' d

-A tr r>i O

V

> VS co"

£

z U I

vO O oC

z z

1-- <f

(P

IA - r -

i« r-c n

rr?. r-\n CX r-

VA O

rcr

rr c s

r-o \N

(V) c s

r r -c<

. . . .

rr" s i O rf in

cr cr VA ro

b o r i

VA

z o 00 cc cr u

o o r -

-_

z o 00 cc cr u

to rD O CO —-

rA

3 r s c4

v„'

CN cs

lA rv

T3-c x

CA

o CS

r s t — CN

v5> CN

s &

s r r - "Tr

ra

rs I i ! i CNl -J

r-r<

ro c s CO

s po \ S

L"> A O

CA rA

rr

8 CA CN

•=r cr z o 00 cc cr u Q

0 U 0 VA

v9 3 r-10 cs? 6 s5

c y -

In

LA

v9

v?

r r

r r v_9

tn

t> Lf)

r-rt

£

0 s9

"c? r-r rr vs

r r cS

in rr 'r-

8 " r n

cr r O O r-

LA

1— s S

c<v rs rr cr

rr-

cr

s9

IP r~

r O CP

•c srS cS s9



•

rr

3

VA

5^ rr

1 VA

-

2 •3? V V~

V

s

s"

3

IT

O fA

t $

•r J

rZ vi

rr

_o

CS

VA

-y v>

rr

6 c cu J

t.

3 0; v

*-

cr

_o

a; c? 0

rr

cv vo

y rr

OA

1) c5l 0

-r-\P

1.

CC Ql

IA>

m _c

V IT O -v Kr

7 it rr

«./

VA

Dale /Sample C A R B O N 0^9/-O N I T R O G E N S O L I D S i>9/.0 Conol, Dale /Sample C O D B O D 5 T O C T Vi N i NH*-N N 0 3 - N T S S T V S S

Conol,

U3U.20M 3 7S" <l7 t p T i r u — / - L "2,44 ' : ! 2 2 i c /24</

S 3 (551 ' ? , ; o 2.3 0. 1 /0./5" 47.0 1-22 7.7 <«7 C60 2.0 0.0<3 4 7 . o 4G / F4 2-0 26 i-zo 7.7

1** S k x c * N o u . 2 3 1026 /qos 470 I020/H6«, 3.W/3C2

<•! (5"S-) 2.4 /5.Z O.I5/0./5 34-O 7 / £ 103(55) (3) 21 2.7 /5.7 0.015/0.11 I up I I ? 7.6<f

1'* H o o f rJo-J- 21 *?5 31 61-3/76.0 20.0/2^.7 qrt. „ - Z 5 13.3 16.7

4 4 C60) OO 23 3.15/2.4$ 10.s 2 ^ 0 IM Go C60) ( ? ) 'Co 2.9 l/.f> i<2.. I14.3 5.3 16.0 2 7 ^

12 24 4/. zlq o.i 17.3/17.3 5*? 2 0 14.1/14.7

5c; C4G; CO IS 3.2 4.5 qn i. (4) 7 2.1 0.13/0.1? n.is 7.3 /6.0 in SIO^P pe<:.'-l q?.i/M4.o 23.5/2^.0 4 " L - $.7 / i l . 1

1" (35, n.*. /(o.O (b.l 5 iM bT.l A S O 10.0 / S . l 2 ? V 7 . 7

i sc no / K G 56 /s?,7 lg.l/2%0

11 &3) 1.1 /2.I5 5.S 3 2 0 7.4 7 4 (.ci) 5.0 lQ/c% 7 .1 21.Z/2V.1 I4.0/M.7 ? 2 0 7 . 7

T + Stc e Dec XI I°I3 /262 HI. MnS I' II N G / l i o io4 / | 4 0

243 (14,0/ 12.2/12.0 l.vr/\.<is o . o IO.fi 0.06/0.025" o . o z 7 6 . 0 2 . 0 6 G. 0 / ^ . 0

4 * S b e « Dec .2?, 2 G 2 4 ^ p SeWed " 123

c cr n

CD D CQ

CD CD 2 3 CD

ro

http://IO.fi

u b c e n g i n e e r i n

Cu r-r r f - r-'

IA

r C r-

Vv r-

rA r —

I/! r-

c 0

U

^ 7 to

r\ *H

£

i/) Q

_ l 0 05

•

c r

CSJ r r r S

rrr r r o4

O cr JvA

v j vS

(VI

oo r -N

sS r s m

O r-rs

vS> o) OA

O r -rs

r\ *H

£

i/) Q

_ l 0 05

t o t o > r -

o v9

_ C N

O CS.

t r

t r v/>

sS LA

rs lo

r r

O l

o l Cr

rr r—

rA cf

? cf

to a-1

t r o

r r

r o

<? r s S

O «r Cv>

o r5s

o rs ty> to «r-Cs

o rS

r r s

OA <rS r~

r~-vS

o •Tr

q c s

r cs

q cs cS.

J— cf

Co in

CA in

OA r

O i f

O

r~ O

0 v i

r-

fO Ln

CA

r\ *H

£

i/) Q

_ l 0 05 to

t o h

v5> C~ tr K. rr tr

t r (V v-< O 0° O l

O

<r

r r CP

t r co OS

rs cS CS

c> cr

o vS

r r

sS

r-4 cr <— 6 rr

r-r>o

IA

O l r-

s> c>-

o cf cx

O

to r 1

VS

o CN o t o LA S3

o -S r 6Q cS

v— t r

o P Lo

r-rS tr ro r f r r

o r f

o t f

CO tv\ J A r s ^ r r

r-ISA "io VA'

(— o r s

r s

r CS l

1— v i r s

q rf

(O

n o

rO

z Ul

O ccl h;

Z

Z ro

O z

6 t r

r r r f <S

v3

vS oc

oC

cO LO

L0

in LO

in cs cf

o cn -

Ln cs In

Ln cS

d cs

in

q

LA rS

vS

LO J— m

Ln r cr

z Ul

O ccl h;

Z

zz c" 3C

Z

z

r-

iA CO Do

{>£. v_9

Oo O o<

? o r o tr o

r o

O S 3 O

Cy= o o oS o o

cj r r i — >cS

VA

c » tr-r f

v5

Cn

Oc. i n

cs rr, o

o~ O

o-o q

t r tr

CO rS c f

r o t i cs

z Ul

O ccl h;

Z

zz c" 3C

Z

z

r-CN •XT

10

1 —

L i Lo CO r f

tn CX CX

r~, r r

o t r

r o o CP

l/v cr

LA \A vS rs

Ln LA'

L n cc o f f

z Ul

O ccl h;

Z

zz c" 3C

Z

z

r-CN •XT

10

1 —

L i Lo CO r f

tn CX CX

r~, r r

o t r c s

m

t ~

/-> *C \ 01

z o CD or cr o

O 0 h

/-> *C \ 01

z o CD or cr o

ir> O O CO L")

rS

j —

r —

C4 0°

r — u")

ol

s cr

rr o f ro

v3

r s LA s >

cr O Ln

t r

o-

i — rS

cS Ln

CA

r ,

cr

- — V r-n r cs

m O

m

' r -v./

CA r-

s S3

3

/-> *C \ 01

z o CD or cr o o

0 u

cs-c r cy>

r d r r o

CS r r Cx t r

r r

I *

cv

vS c r

1A r -

r rS

rA oJ

O O cs

rr

rS 0°

r-

v J 0°

D° o l

in r — v9

Co

r~

vS do

r xr AO

cr

b s9 0 °

ol

o r -o

'rf

CS CP

"3 0 r-

r r CP

IV) CP

vf)

on i—

cr (JO

CA rj;

r

SS

Ln O

4) cx E c f

LT) \

Q

o-

cr •O

L TT

T> o

\A

r r

a' tr 0

—f V

1 if P

V

* r r

v5>

0' c5 0

~v V *

r r

"T>

rre

r r

:> c5

cy

I/ 1

fL •± r r

~t'

rr: a'

ni U

o

] ^ _ •£

».

rrr* c

L^

s

r r

l»

c h°

c c

x l2— * r r

T a1

"ti V-

±. r r

60 o

5 6 s

rr

ft-

l v

r r

u-

0' 6 D

- r L>

H rr

T) a-

r5 y

V.A

t r

u b c e n g i n e e r i n 233

CL.

CT v9

PO r—

rs

r-Cr vS> r~ r

OA

r~ r-rr r bo r-

rr r r

iA r 1 r

TJ C O vj

^7 </>

S - <

/-~

\

£

tn a _J 0 01

^£

E O N V-'"'

O r-rs

0=> 00 O 6 0 r<

O 0o CX

rr PA rs

v3 r s

rA CN

v9 cr cx oo Ln r<

rr cs

O rr H CN i A fsA

rr 0 (vA

O CY1 rr

O CNA

/-~

\

£

tn a _J 0 01

LA >

o r=9

r-s3 o

-rr-

r o r-s9

r-v i

© rA cr r-

s i r— rA

r~ o r r

q o r-s3

r-r s

r-6 rs

r— rr

r— o r-

\9

r— s i

q

/-~

\

£

tn a _J 0 01

LA > O O

rr r q rA

r— o rs

r-0 °

r s i

r - A l PA

r r s

(V) rf

o rr- rv-> L A

iyi VA CS

O r rr o o

r rs

r-i iA &

r r>

rvi vi

/-~

\

£

tn a _J 0 01

LA Ui

i A

o •rr rr

rA ~~A q d r s

r-rs

rs

o r s rs rA cr

q s9

r-s9

OA

rs

rr -S rr O

I-

-8 rA of

rA

O r s

o •S O s

q CP O rs

fA

cr r-

r— Pjo" jrr-(Vl

L"

r-rr rA vA ISA

rr. -o r-_

r—;

VA' rr

q

q rs

r—i v i <sl o„ cs

r-r>, q rs ro

O

r f CO iA tr

o Cxi

M

r-

rr

z Ui

O a: r-

z

z 1 m O z

Vn O

LA s S

cr s i

rr rA rA

O P A

s i

L/? r~ O

o> s i

rr r s

LA

Cr-cr LO r -

Ln LA cs rA

IS)

Do 01

O

z Ui

O a: r-

z

'2 1 c"

X

z z 1-

"rr rr

rs I —

rs rA rr r~ rr on

r ~ i-n

r-i s9 O

L 1

Ln

LA LD

Cw ss> •o

rs!

LA q

q V

cr LT rr oo

r-r i

o-rr s9 LA

t r r 1 z

Ui

O a: r-

z

'2 1 c"

X

z z 1-

V i r* rs

VA

s5

cr r i

c -

C rA rA

IT)

r~ r rC r"

t~ rs

VA PA

PA rr fS

<? VA rf

rS cr

z Ui

O a: r-

z

'2 1 c"

X

z z 1-

r" t~ rs

VA PA

PA rr fS

<? VA rf

rS cr

01 E

z o CO

a: cr u

O 0 r -

ex a O CD

A j 5 01

E

z o CO

a: cr u

LO

O CO

0 tA

rs rs

rJ

vn s_/

r; po r~ O

1 1 1 I 5 s ro ro rs;

rn cs

rr

rr

I n

(.-rj- rr-hA

LO

cr

o

LA

CA rl? CS, cs

Ln

01 E

z o CO

a: cr u a

0 u rr s9

s9 *s_

s5> V."

Lf)

LA

o r-

rr r r Cr§

r ^ O vA s^

rs s 9

rr si> O s9

"s

•Hi

V A s3

s LA

r -s3

sS LA

sj»

rs LT\

s9 La

O c°

rs V3

O r~ vn

s 5 03

rS r*

3

m o c-vn 0 ° r— cr

s i

Cr-sS

m O t-

m

<r •ZJ LA

 U_

4

~b

r 5 0* V>-

i Ix LV 6 d

• if 0

.£ rr

0'

5 :

IA

L u

0' 6 C

- r " L / J

rr

V j c i iSL

ai 6

r_

1

rr

-D

CD

V ^

f V

•i zr

r

o o V

rr"

tl •3'

rr

rS ^\ il-

a> (5 6

LA

? L rr

a» _Z

v/»

if rr

v5 r s

_ Q

(i> C D

-r" V •£

rr

"T-

cu v/

rr

rs •> C5

r

a> (? 0

- f i.n r :

rr

c

4) V A

7_ rr

u b c e n g i n e e r i n g *Zl . _ _ _ 234

Cu r-LA r~

Ln. r-

CP t r r-

lo r-

vO r-> 1—

c

U°

<v N £

Ol

a _ J

0 cn

to h rrr

rr s9 on

O t r rr

O cr 00

<v N £

Ol

a _ J

0 cn

to to >

on

cT co

PA ryi

tr o o

o j o o t r

q vi rS

o r f r \

o vi r-t f

CO cf-kA r-; vj>

O rf r-M

on CM

CA r~-

rr v i

(V)

r-

o >o

rn r f

fO ro

00 _ A OA VA

q rf

o •o

O tf

Q CO

r-r s

r r s

o <5

o Cr

o s S

r rf

q

r o

O O

LA

r-

to r o

r-r f

q rvj

r— cf

r

o r f

o ~r

tr t f

Co (vo

<v N £

Ol

a _ J

0 cn to

h-

p° LA CS

on 0°

o rv)

6 t o c»

O

$ o rv)

r; •o

r~r rv, /-\ rS>

fA

iA O

r-r f

rO cf

CO cf-(O r>A

O rf CO q

OA Cr c~ t f

o r; vi rs

cvi r~ r~

t — r f rs

(O

O O

f

JV\ rsi i f CS

q

"to

CO i i r~ r f

ro

CS rA c f

o C»a

r o r-

CO i f

o r f

<v N £

Ol

a _ J

0 cn to

h- fro c r

o r~ r s

r s

O rf

q rv ir»

o rv)

r; •o

r~r rv, /-\ rS>

fA

iA O

r-r f

rO cf

CO cf-(O r>A

O rf CO q

OA Cr c~ t f

ro to OA

q r s r s

r; S3

r. r s ryi

(VA

(O

O O

f

JV\ rsi i f CS

q

"to

CO i i r~ r f

ro

CS rA c f

o C»a

r o r-

CO i f

o r f

f

z Ui to O al r-Z

Z i fO O z

CA If)

—

-

o r r in

O cs cr

Oo 0o LO

r r r

« r r r S

Ln

c f T~

LA

cf

q in

c r

r r

L0 q s9

6 :cS f

z Ui to O al r-Z

i K> X

z

z r-

rS

LA-cs LT3

Lo tf

q vi 0 ° cr

LO c r

LA ro cr

cf V? rs

O cf

0^ q cf

CX>

f

z Ui to O al r-Z

i K> X

z

z r-

r/} cr

t r o 6

rs

cc? cS

cy? r S rr

v3 ty rs

t r cf cf

</; r \ cS

ro (f

q r r to

°? d

r S oi

E

Z o CO a CL U

S CL V-q l i

O cr 6 6 H r-

<6 rs! r d

c-6 v5 d —

^s C

L'l r~ C

in 6

E

Z o CO a CL U

in O 0 CO

E

Z o CO a CL U a

0 u

r-O 0"

H r s —

C P s_>

o rv)

r-LA

1 <-s q

r— I~

rr C r

Cr cr

^rx / c r

3 r r -

La >

tr-

vf3

cr r? v3

O C«

o r

LA O in

rr

r

o

b

LA 0~

r-v3

cr La

^1 vVJl

v5

'cr •

cr w

cr r~ vS

r r

f -

r-v3



X

o; 0" 0

a>

6 i /

if r r

J rE Lo

'7 s

uv

o

X

r l

i

<£_

t> <v 32 a)

5- r r

0 H

i Xn

Si.

s

•£ r r

-Tj

i f

5

r r

ri

o X

tr O io

rr

*a c>

s o rS

21

ai

§

r r

-c

aJ

5-

c r r v S

_D gy

CV 0" &

V* 7

0) L/1

•

ZT

u b c e n g i n e e r i n g 235 —

=c ex.

•

c U°

£

m Q

_ J 0 to

I/)

h £

m Q

_ J 0 to

0 0 C/7

r -

r-(>•>' o-

r-s i

O r>

r\ri

o s i

o

r-c>o

o 0/>

r r

r A

rA C A

CA CA

r~-o C A v; i — t r

o V) <s t A

q s i

On w i

O v i

£

m Q

_ J 0 to • A

V/)

\-

r~_ v s

v> cs

o r-o

PA

r y i r--

VA flA r-cs

q r s cs r-C £

r-vs

r CP

Cvi

CsA

Cvi r-1

cs r ~ ; v i cs

o cv

q v i

o ve PA

o t>

r S

rs

Cvi i A r v i r r

/ " S

£ •~—'

Z tu VO O h -

Z

z 1 ro O

z

/ " S

£ •~—'

Z tu VO O h -

Z

1 1"

3 T z

Vi rV\ r

C P

tr v3 O

tr- tr-

0° vS VA O

<5

l A O 6 V

/ " S

£ •~—'

Z tu VO O h -

Z z

1-

• j - tr C S 72}

A/J t r r r

0 r r

Cr o> r~ CN

CV)

rs 60 rr

v9 r-

6 cr O

Q CN

CT)

£ S >

z O CD Ct cr u

0 o h

rA cr o

r -

v9 co r r

(V

&o fA

°? rl

vS 00 r s

rr v9 0"

O r to

CV)

r r CT)

£ S >

z O CD Ct cr u

0 CO

o x

CT)

£ S >

z O CD Ct cr u o 0 u

cs r~

L A L A

• — o-LA

r r r—

in u i

L A O 0°

v9 v_/

S<5

r-

P ~ v—.

r-

(SO

r~ V/i

r-— SS,

' A r-

sl

V 9

r — v9

rA \3 -CA v3

v5

"a 2

5 ;

r A in

CD CL E tf on

-+-tf

o

C S

s i

%

1

>.' :

t> t»r o> v/>

P T

"&

0 ~Y l/>

0' t r -

Ot \f *r

r :£ t r

c r

V Si.

<JX

cy <? t>

~*c \P

tr

r r

V>A

7 rr

H

s

8 t

T T

T> a)

V>

to

o

b

T

o>

V

i .

0. •3=

6

sT t r

i '1

' c

^. j r

a -

<-

0 a» r*r

(i

0

0 -V

lA

I-3"

r~

i 6 -r

\

' A i d t> i. ri

0) 0" 0

7) P A

e) Cf 6

tr

i

3) 0; 0

o> A" O

zr i

rbc treatment of a municipal landfill leachate: a pilot scale

Documents