UNIVERSITY OF HAWAIII LIBRARY

AN INTEGRATED TREATMENT AND REUSE SYSTEM OF DAIRY WASTEWATER

- A CASE STUDY IN THE STATE OF HAWAI'I

A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE

UNIVERSITY OF HAWAI'IIN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

IN

BIOENGINEERING

AUGUST 2003

By
Liangjie Dong

Thesis Committee:

Ping-Yi Yang, Chairperson

Clark Liu
PingSun Leung
We certify that we have read this thesis and that, in our opinion, it is satisfactory in scope

and quality as a thesis for the degree of Master of Science in Bioengineering.

/

THESIS COMMITTEE

~c~a~rson~ ?
~~.• I-...-.~

-1

ii
 ACKNOWLEDGEMENTS

The author wishes to express his gratitude to his thesis committee, Dr. Ping-Yi Yang,

Ping Sun Leung and Clark Liu for their guidance, encouragement and professional advice.

He particularly wishes to express his appreciation to Dr. Yang for his support on new

challenge works during his research period.

Sincere thanks are also due to Mr. Charles Nelson for his help on manufacturing the

reactors used in the research. The author also appreciates Dr. C.N. Lee and Mr. David

Wong for his cooperation. USDAIHATCH fund proviged the support this research and the

author's scholarship.

Finally, the author is grateful to his family for the encouragement and endless support.

iii
 ABSTRACT

Milk. parlor wastewater in a dairy operation in Hawaii was investigated for its potential

treatment and reuse. Two earthen settling ponds were served as settling process and

another 10 acres of wetland were planted with California grass to receive the etlluent

from settling ponds. It was found that the present treatment systems still remained many

problems, .such as odor,potential groundwater contamination and surface water pollution.

The etlluent could not meet the requirements for either disposal or reuse. Improving and

integrating other cost effective biological treatment component(s) to the existing lagoons

is required in order to meet discharge or reuse standards. Several biological treatment

technologies were evaluated and investigated as the alternatives. In order to improve

these existed problems, a5-S approach (system existed, short HRT, small investment,

simple operation and maintenance, system thinking) is developed. Among the 5-S,

developmentofa higher SRT(Solids Retention Time) for the bioreactor is essential in

order to integrate this unit into the other 4-S. A bioreactor, which is the modification of

VASB (Vpflow Anaerobic Sludge Blanket) reactor, was investigated. It was found that

more than 70-75% ofTCOD (Total Chemical Oxygen Demand) removal efficiency and

biogas production of 1.51/1/day with CH4 content of 70-75% could be achievedat8-15g

COD/1/day. The unique operation of the developed bio-nest reactor is able to achieve a

higher COD removal efficiency with high organic loading rate with simple design and

operation. This unique bioreactor is able to provide better sludge distribution and less

dead zone compared to the conventional VASB. The process performance, design and

iv
operation criteria of this bioreactor were presented, discussed and compared with the

existing anaerobic filter reactor and UASB reactor for the treatment of milk parlor

wastewater. It was found that this bioreactor is able to operate in the temperature of 22± 2

°c compared to others (35 0q with high COD removal efficiency and biogas production.
Thus, it can be easily integrated into the existing lagoon system (popularly in the United

States) or integrated to the aerobic/intermittent aerobic treatment unit for further removal

of carbon and nitrogen in order to be able to dispose, discharge or reuse.

Based on the laboratory results and field investigation, a treatment and reuse systems is

proposed to improve one of the existing milk parlor wastewater management system.

v
 LIST OF TABLES

Table 1.1. Characteristics of Milk Parlor Wastewater 9

Table 2.2. The New Treatment Target Criteria ; 11

Table 1.3. Evaluation of Technology Alternatives 11

Table 4.4. Anaerobic Pretreatment Process (Rl+R2) Performance 12

Table 1.5. Aerobic Post Treatment (R3) Process Performance 13

Table 1.6. The Final EffluentQuality ; 14

Table 1.7. Facultative Lagoon Typical Design Parameters Compared With Existed Lagoon 15

Table 1.8.Total Cost Estimation of Treatment System with Byproduct Utilization 16

Table 2.1. Characteristics of Milk Parlor Wastewater 24

Table 2.2. Anaerobic Pretreatment Process (Rl +R2) performance 27

Table 2.3. Process Performance for EMMC Technology for Anaerobically Treated Effluent... 28

Table 3.1. Basic Parameters of the Bionest Reactors 38

Table 3.2. Surface-to-Volume Ratios of Selected Fixed Film Support Media .41

Table 3.3. Characteristics of milk parlor wastewater(average of sample analysis) 42

Table 3.4. Anaerobic PretreatmentProcess (R1+R2) performance ; 45

Table 3.5. The Relationship among SRT, HRT anq OLR 49

Table 3.6. Comparison ofVarious Types of Anaerobic Reactors 56

Table 3.7.. Comparison of Different Reactors Performance under Different Operation Condition56

vi
 LIST OF FIGURES

Figure 1.1. Current fann water and wastewater flowchart 10

Figure 1.2. Bioreactor Structure and Flow Chart 12

Figure 1.3. TCOD Removal Efficiency and Methane Content vs TCOD Loading Rate 13

Figure 1.4. Effluent TCOD and SCOD in Different Stage of Treatment 14

Figure 2.1 Current Water Practice and Treatment Problems 20

Figure 2.2. 5-S Approach and Their Relationship 22

Figure 2.3. Two-Stage Anaerobic Pretreatment Experiments ; 23

Figure 2.4. EMMC Reactor Structure and Flowchart 24

Figure 2.5. Existing System Problems and Potential Solution Evaluation Process 26

FigUre 2.6. TCOD Removal Efficiency and Methane Content vs TCOD Loading Rate 27

Figure 2.7. Proposed System Flow Chart 29

Figure 3.1. Two-Stage Bionest Reactors ; 38

Figure 3.2~ Bionest Reactor Appearances 39

Figure 3.3. Bionest Materials and Structure 39 .

Figure 3.4. SCOD Reduction during the Startup Period 44

Figure 3.5. TCOD Removal Efficiency and Methane Content vs TCOD Loading Rate 45

Figure 3.7. pH Changes inside the Reactor 1 with Different Loading Rates 47

Figure 3.8. pH Changes inside the Reactor 2 with Different Loading Rates .48

Figure 3.9. The Relationship between SRT and HRT 49

Figure 3.10. SludgeDistribution in Reactors with and Without Bionest Installation 51

Figure 3.11. Various CaSes of Mixing Related to Sludge Gravity Settling and Upflow Velocity. 54 .

vii
 TABLE OF CONTENT
Acmow ledgements...........................•................................................................................ iii
Abstract Iv
List ofTables ••••••••••••••••••••••••••.••••••••••••••••••..••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• vi
List of Figures •••••••••••••••••••••••••••••••.•••••••••••••..•..•..••.•••••••••••••••••••.•••••••••••••••••••••••••••••••••••••••••••••••••••• vii
PART I
Introduction....•..............................•.................................................................•............' t
Research Problem 2
Research Objectives 5
Organization of the Thesis 5
Reference: 5
PART II
ESSAY ONE
Evaluation of Potential Dairy Farm Wastewater Treatment and Reuse Systems in the
Tropics ..........................................................................•................................................................. 7
Abstract 7
Introduction 8
Pre-analysis of technology alternatives 11
Preliminary Anaerobic Treatment plus Aerobic Post Treatment ou 12
Integration 14
Cost analysis ; 15
Conclusion 16
Acknowledgement: 17
References 17
ESSAY TWO
Developing a 5-S Approach for a Milk Parlor Wastewater Treatment and Reuse System .. 18
Abstract 18
Introduction 19
Method /Approach ; 22
Results and Discussion 24
Conclusion 29
Reference: 30

viii
ESSAY THREE
Design and Analysis an Anaerobic "Bionest" Reactor for Milk Parlor Wastewater 34
Abstract 34
Introduction ; 34
Material and Method 37
Surface Area and Porosity: 40
Operation Procedure: 41
Analysis Method: 42
Wastewater Characteristic 42
Startup : 43
Results and Analysis 44
General Perfonnance 44
Process Stability 45
pH 46
SRT and HRT 48
Sludge Layer and Distribution 50
Comparison 55
Conclusion 57
Reference: ;; 57

ix
 PART I

Introduction

Animal waste generated by Concentrated Animal Feeding Operations (CAFOs) poses an

increasing threat to the health of America's waters. Dairy farms are one of the largest
wastewater producers among the CAFOs. In the past decades, wastewater management
has focused on the land application and land-based treatments (lagoon and wetland).
These management and technologies can not satisfy the new regulation requirement.
A dairy farm, located in the state of Hawaii, was investigated in this study. The main
purposes are to evaluate its current systems impact on the environment, develop/upgrade
water/wastewater management system and design a suitable technology for the land
limited application.
The investigation was made for IS months; both field and laboratory investigation were
carried out and the results are prepared into three papers as following:

1. Evaluation of Potential Dairy Farm Wastewater Treatment and Reuse Systems in

the Tropics. This paper was accepted and will be presented at The Ninth
International Symposium on Animal, Agricultural and Food Processing Wastes
(ISAAFPW 2003), October 12..15,2003, Durham, NC. The authors are Liangjie
Dong, P.Y. Yang, P.S. Leung and C. N. Lee.
2. Developing a 5-S Approach for a Milk Parlor Wastewater Treatment .and Reuse
System. The paper was presented at the 6th IWA Specialty Symposium on Strong
Nitrogenous and Agro-Wastewater, 11-13 June 2003 - Seoul, Korea. The authors
are LiangjieDong, Anthony Yang, and P.Y. Yang.
3. Design and Analysis of an Anaerobic Bionest Reactor for Milk Parlor Wastewater
Treatment. Part of this paper was presented in the National Summit ofAnaerobic
Digestion organized by Water Environment Federation and sponsored by USDA,
USEPA and USDOE. The full paper will be also presented at the Asian Waterqual

1
 2003, organized by the lAW (International Association of Water), Baughah,
Thailand, October 19-23,2003. The authors are Liangjie Dong and P. Y. Yang.

Research Problem
Milk parlor wastewater in a dairy operation in Hawaii was investigated for its potential

treatment and reuse. Currently, a lagoon system serves as a holding area with poor

treattnent performance with a 10 acre land planted with California grass serves as

receiving body for the lagoon effluent. This type of land based treatment is popular in the

u.s. mainland as well because of low cost. However, it has often been found that this
kind of treatment application caused many problems, such as odor, groundwater

contamination and surface water pollution. The effluent could not meet the requirements

for either disposal or reuse. With the increasing of dairy industry, the dairy farm and

other Animal Feeding Operations (AFOs) were widely recognized as a significant cause

of surface water impairment, air pollutions and groundwater contamination. National

water quality inventory (EPA 2002) shows· that, nearly 40 percent of the nation's assessed

waters show impairments from a wide range of sources. Improper management of

manure and wastewater from concentrated animal feeding operations (CAFOs) is among

the many contributors to remaining water quality problems. Improperly managed manure

has caused serious acute· and chronic water quality problems throughout the United States.

Environmental problems associated with animal manure and wastewaters are surface

water (e.g., lakes, streams, rivers, and reservoirs) and ground water quality degradation,

adverse effects on estuarine water quality and resources in coastal areas and effects on

soil and air quality. Unlike other industry, animal operation including dairy farms as

2
opened systems, may have several possible danger pollution threat the environment. They

are 1). Contaminated well (well water contaminated by bacteria and nitrates which leach

through soil). 2). Waste storage structure: Poisonous and explosive gases in structure. 3).

Animals in poorly ventilated building (ammonia and other gases) 4). Waste applied at

high rates (nitrate toxicity and other N-related leaching of N03 and microorganisms

through soil, fractured rock, and sinkholes.) 5). Discharging lagoon, runoff from open

feedlot, and cattlein creek, 6). Runoff from fields where livestock waste is spreadiand no

conservation practices on land (especially P and NH4 nutrients reach stream). 7).

Eutrophic conditions caused excess algae and aquatic weeds 8). Leaching ofnutrients and

bacteria from poorly sealed lagoon.

Much effort focus on dairy wastewater treatment technology and nutrient application

were taken for a long period of time. Among the technology alternatives, most popular

alternatives are solids separator with liquid or slurry storage, anaerobic lagoons,

constructedwetland, anaerobic digesters,aerobic lagoon and activated sludge. In most

dairy farms, these treatments arecotnbined two or more as a system service as treatment

process. The majority dispose of their waste through land application (USDA APHIS,

1996b) as an end of application of nitrogen and phosphorus. However, the .1997

agriculture census data,analyzed by Kellogg (Kellogg, 2000), shows that more than

73% and 90% dairy farms(large operation more than 700 cows) do not have sufficient

land (or noland application) to apply the nitrogen and phosphorous, respectively.

Among the land application practices, constructed wetland caused more interest because

the crops and wastewater treatment can be combined together, and cost is relatively

effective. However, some potential problems, such as nitrate and nitrite increasing

3
 (Schaafsma et aI, 2000), potential runoff (Newman et al,2000) are occured, even wetland

suitability for the dairy wastewater were questioned (Geary and Moore, 1999).

A large number of treatment techniques are practiced in dairy fann wastewater and

manure management. However, many of them were transferred directly from the

municipal wastewater treatment field (Bicudo et aI, 2000), the challenge has been to

. develop a treatment system that perfonnswell regarding the targets of dairy wastewater

and waste management such as wastewater stabilization, groundwat~r protection, odor

control, surface water protection and nutrient management, etc.

One ofa large scale dairy farm (with 60 acres area, 1600 livestock) located on the island

of Oahu, Hawaii, U.S., was investigated for developing an appropriate wastewater

treatment and reuse system. This island dairy farm is located in a sub-watershed with

total area of 70 acres. It consists of 50 acres land available for nutrient Ilpplication, 10

acres for wastewater treatment, and 10 acres of use for milking facility for 1000 milk

cows. The fann annually produces 2.6xl06 gallons milk (9,841m3) with 1.82xlO.8 gallons

(47,556m3) water consumption, and generates 1.8xI08gallons/year(6.814xl0s m /yr)

3

wastewater. There is an average of 300,000 gallons I day (l,136m3/day) wastewater that

needs to be managed. Currently, local water company supplies the water for milk cows

drinking. Water for floor flushing, equipment washing and cow cooling is pumped from a

well in the farm. The wastewater generated from milking center flows to two uncovered

and unlined lagoons that each of them has a volume of 250x50x6 f{(length,width, depth,

n.6x15.2x1.83m3). A 10 acres ofland planted with California grass receive the lagoon

overflow.

4
Research Objectives
The main objectives of this research are to evaluate and investigate the current

technology alternatives in order to develop an environment sounds and cost effective

treatment unit to be able to integrate to the existing treatment system. For the milk

parlor wastewater, specific objectives for this study are .included as follows:

-To investigate the problems of the existing treatment systems and evaluate the

current technology alternatives applied in the dairy wastewater treatment.

-To evaluate the process performance of proposedbioreactor to .be able to

integrate to the existing treatment system.

-To establish the design and operation criteria for the developed bioreactor with a

cost analysis.

Organization of the Thesis

The following three essays in Part 2 of this thesis cover the general objectives as follows:

1. Evaluation of Potential Dairy Farm Wastewater Treatment and Reuse Systems in

the Tropics.

2. Developing a 5-S Approach for a Milk Parlor Wastewater Treatment and Reuse

System.

3. Design and Analysis of an Anaerobic BionestReactor for Milk Parlor Wastewater

Treatment.

Reference:
U.S. EPA (2001) National Management Measures to Control Nonpoint Source Pollution

from Agriculture, U.S. EPA, Washington DC.

5
USDA APHIS. 1996a. National Animal Health Monitoring System, Part I: Reference of

1996 dairy management practices. U.S. Department of Agriculture (USDA),

Animal and Plant Health Inspection Service (APHIS). Fort Collins, Colorado.

Kellogg, R.L., C. Lander, D. Moffitt, and N. Gollehon. (2000) Manure nutrients relative

to the capacity of cropland and pastureland to assimilate nutrients: Spatial and

temporal trends for the U.S. April 27, 2000.

Schaafsma Jennifer A., Baldwin Andrew H. and Streb Christopher A (2000), An

evaluation of a constructed wetland to treat wastewater from a dairy farm in

Maryland, USA. Ecological Engineering 14, 199-206

Bicudo, J. R., Westerman P. W.,and Qleszkiewicz 1. A. (2000) Anaerobic andaerobic

treatment ofanimal manure - A review. In: Proceedings of Animal Residuals

Management Conference - Issues and Solutions. Nov. 12-14, Kansas City,

MO. Water Environment Federation, Alexandria, VA.

Schaafsma Jennifer A., Baldwin Andrew H. and Streb Christopher A (2000), An

evaluation of a constructed wetland to treatwastewater from a dairy farm in

Maryland, USA. Ecological Engineering 14, 199-206

Newman Jana Majer, Clausen John C., and Neafsey Joseph A.,(2000) Seanal

performance of a wetland constructed to process dairy milkhouse wastewater in

Connecticut, Ecological Engineering 14, 181-198

Geary P. M..and Moore 1. A.(1999) Suitability ofa Treatment Wetland for Dairy

Wastewaters, Wat.Sci.and Tech., 40( 3), 179-18S

6
 PART II
ESSAY ONE:

Evaluation of Potential Dairy Farm Wastewater Treatment and

Reuse Systems in the Tropics
Abstract
Milk parlor wastewater in a dairy operation in Hawaii was investigated for its potential

treatment and reuse. Two earthen settling ponds serve as settling process and another 10

acres lagoon planted with California grass serves as a wetland. This land-based low cost

treatment is popular in the U.S. mainland as well. However, it has often been found that

this kind of treatment application caused many problems such as odor, potential

groundwater contamination and surface water pollution. Such effluent could not meet the

requirements for either disposal or reuse other than agriculture irrigation applications.

IInproving and integrating other cost effective biological treatment systems or

component(s) to the existing lagoons may be required to meet discharge or reuse

standards. Several biological treatment technologies were evaluated and investigated as

alternatives. Based on the laboratory results, it was found that the anaerobic-aerobic

pretreatment provided a total COD, and total nitrogen removal efficiency of 88% and

55%, respectively. The recommended system will solve major problems and make the

treated wastewater reusable for floor flushing. Cost analysis estimated for the new system.

The cost of new treatment system was· $482,430 initial investment. Annual operation &

maintenance cost will be $15,297.

7
KEYWORDS. Dairy wastewater, anaerobic and aerobic pretreatment, lagoon system

integration, cost analysis.

Introduction
Large-scale dairy production system generates great quantities of wastewater. Potential

pollutants from decomposing dairy manure and milk parlor are organic matter (proteins

and lipids), nutrients, various gases including methane anet ammonia. The major pollution

problems associated with these wastes are surface and groundwater contamination, and

air pollution caused by odors, dust, and ammonia.

According to EPA Feedlots Point Source Category Study (EPA, 1999), dairy farm

wastewater has average chemical oxygen demand (COD) concentrations of 4997IIlg!l,

and biochemical oxygen demand (BODs) of 1003mg/l, respectively. The COD

concentration varies in the range of 2000-700OmgIL depending on wastewater

management, climate, operation conditions and types of flushing. The higheOD

concentration is due to waste milk (produced bywashing milking equipments), detergent,

manure and waste feeds combined in the washing or flushing of holding pens and exit

alleys. According to Henze et al.(2002), 0.7 -1. 7 m3/ton milk wastewater generated

BOD7 concentration of 500-1500 mg/I.

A dairy farm with 1000 milking cows that employs flushing to clean milking center waS

studied. Two parallel settling ponds, each with a volume of 250x50x6 re (leIlgth, width,
depth, 72.6xI5.2x1.83m\ were installed to receive a 300,OOOgallons/day (l136m3/d)

wastewater alternately. This wastewater includes water used for drenching the cows at

the feed stanchions to alleviate heat stress in the open pens. The effluent from settling

ponds is received by 10 acres of lagoon (2 parts; primary and secondary overflow

8
components) planted with California grass. Both the solid settling ponds and overland

flow lagoons were without synthetic liners. While th~ settling ponds are 1.5 years old

while the receiving lagoons were over 30 years old. The objectives of this study were to

evaluate the characteristic of milk parlor wastewater and to design a treatment system

that would potentially allow the effluent reuse. The characteristics of wastewater

generated from the dairy milk farm in Hawaii was evaluated and presented in Table 1.1.

Table I· I Characteristics of Milk Parlor Wastewater

Source Washing Temperature TSS(g/L) VSS(g/L) TCOD SCOD TKN Phosphorus Main
times/day (mg/L) (mgll) component
of
(UC) (N (Pmgll)
washing rrioIIl
Milk Parlor 3 65 10.5 8.7 17500 3710 N/A NlA Milk
Walking 3 27 3.21 2.35 12000 2200 N/A NlA Manure and
way flushing foods
Manure 3 27 3 2.1 4180 1300 N/A NlA Manure
flushino
Average(26 28 4.2 3.1 5200 1900 200 20.1 Mixed
of samolesI
Note: Milking Cows are milked 3 times in this operation

The relation between BODs and COD was. tested using nine samples. This relationshipis

COD= 2.46 BODs+126, (R2 =0.95). It was found that the biodegradability for TCODand

SCOD was 88%-93% and 75%-80%, respectively.

Potential problems may include:

• odor generation,

• ground water contamination,

• surface water pollution,

• lagoon sludge clean-out and distribution channel clogging, and

9
 • poor treatment performance.

These problems are a typical for dairies throughout the mainland USA. They are

illustrated in Figure 1.

Milk Cows Well water storage Milking center

cooling r+-- tank -

,..----...;...-----1: Settling pond r--

r--
Lagoon for ~etland with grass
evaporation I+-------+----{
Settling pond •

Well House

Figure 1.1.. Currentfarm water and wastewaterflowchart.

1. Lagoons and overland without liner causedpollutant leach to groundwater. 2. Closed water
circle caused accumulation N03(N03-N is 13.8 mg/; COD is 69mgll). 3. Uncovered wastewater
Channel and lagoons generate odor. 4. Overland water consisted 680mgll COD and 85-125mgll
ammonia nitrogen is potential polluted runoffsource.

In orderto solve these problems and improve wastewater management, the farmer would

like to plan a new management system to solve the above listed problems,particularly,

potential reuse of the treated wastewater for floor flushing.• Although there is no strict

reuse standard for flushing in the US, Crook andSurampalli.(1996) suggested that the

treated wastewater ofTSSbe less than 5mgll, and BODS ofless than 20 mg/1 be

acceptable for flushing purpose.

10
The target quality criteria for the effluent to be reusable for flushing floor were developed

as shown in Table 2.2.

'table 22The N ew Treatment 1:argetC't .

nena
Category TCOD Nitrogen (TN) Phosphorous(P) SS Effluent pH
Target
Concentration 650·700 mg/I 70·80 mg/I 6·lOmgIJ 5·8mg/1 6.5·8.5
Target reduction
efficiency 85%-90% 60%-65% 50'%-70% 97%·99%

Pre-analysis of technology alteroatives

Based on the target criteria and existing problems, several treatment technologies

currently used in the dairy waster treatment were evaluated. They are 1) anaerobic lagoon

system; 2) natural lined wetland or constructed wetland; 3) activated sludg¢ treatment

system; 4)anaerobic treatment. These treatments are presented in Table 1.3..

Table 1.3. Evaluation of Technology Alternatives

TeOD Nitrogen p 88 Effiue Odor Ground Remark of disadvantage and
removal (TKN) remova remova ntpH problem Water
(0/0) removal 1 I protectiolj limitation
(0/0) (0/0) (0/0)
Requiremen 70·80 40-50 50·60 80·85 6.5· Fair Good
t 8.5
Activated Expensive construction cost
sludge .j 7 ? .j .j .j .j and 0& M, complexity in
operation
Strong odor, high risk of
Wetland .j 7 ? .j .j X Risky groundwater and surface
water pollution, land
limitation
Anaerobic
lagoon .j X X 7 .j 7 ? Poor effluent for reuse, land
limitation
Anaerobic }>oor effluent, need post
treatment .j 7 X 7 .j .j .j treatment for reuse

Note: X' no/satisfied; ./: Satisfied; 1: questionable.

From Table 1.3.,itisapparent that any single treatment willnot able to achieve the target

criteria proposed. A combination or integrating multiple processes is recommended. For

the particular case listed in Hawaii, an integration ofthe anaerobic bioreactor and aerobic

11
post treatment with the existing settling ponds and lagoons system may be appropriate. A

preliminary experiment for the anaerobic pretreatment process, thus, was investigated.

Preliminary Anaerobic Treatment plus Aerobic Post Treatment

A two-stage upflow bioreactor (IOL each) and a 5L aerobic reactor as shown in Figure 2

was designed and operated under room temperature (22 ± 2°C) for 150 days. The

performance of the anaerobic pretreatment process and aerobic post treatment were

presented in Tables 4 and 5, respectively.

Gasbag

BiooeSI Emaut

Feeding
Tank
.........__:;> Media

R2
R1 R3

Figure 1.2. Bioreactor Structure and Flow Chart

Special media named bio-nest were installed to support microorganism growth

Table 4.4. Anaerobic Pretreatment Process (R1+R2) Performance

TCDO Bloges
Rl"R2 removal Gas r-SS removal TKN Methane
Period Loading HRT Production pH changing
efficiency Quality efficiency Removal Gas Yield
Rate rete
TCOD SCaD Methane LJg COD
Units g/Uday Hrs
(%) (%) (%) (%) (%)
UUday R1 R2 IN
removal
1 2.68 60 75.4 45.6 76.0 88 51.0 1.33 0.310 7.3 7.3 ~.8-8.1
2 3.2 50 80.1 59.3 74.0 80 50.5 1.91 0.290 7.2 7.3
3 8.3 25 77.5 56.5 72.0 79 49.2 2.14 0.259 7.1 7.3
4 17 15 72.7 54.4 70.0 72 43.2 1.70 0.180 6.9 7.1
5 25 15 69.4 31.6 69.0 71 36.1 1.73 0.166 6.4 6.9
6 32 11.5 67.4 36.0 63.0 69 29.2 2.22 0.139 6.3 6.6
7 36 10 52.0 28.9 57.0 62 21.5 2.04 0.116 4.5 6.4
8 15.5 15 70.3 57.2 70.1 76 45.3 1.63 0.200 7.1 7.3
12
Note: In period 8, the treatment process was operated with various TCOD influents.

Table 1.5. Aerobic Post Treatment (R3) Process Performance

Period TeOD HRT TeOD seOD TSS TN P
Loading rate
Unit g/Uday Hours Removal Removal Removal Removal Removal
efficiency efficiency efficiency efficiency efficiency
(%) (%) (%) (%)
1 2.51 16 71.1 62.1 62.1 59.0 65.0
2 2.42 13 68.2 51.3 55.2 52.0 63.0
3 7.10 7 55.1 37.3 46.1 48.0 54.1
4 17.64 4 45.2 21.8 42.0 45.0 50.0
5 29.07 4 35.3 18.9 N/A N/A N/A
6 39.64 3 26.8 15.1 32.0 30.1 30.0
7 65.66 3 18.9 11.1 N/A NA NA
8 17.49 4 52.7 41.1 45.0 46.0 52.2

90 0% ~----------------------.- 8CW%

80.0% 70.0%

700%
eo.OOA.

~ 800%
500% ..,
g
4>
5::
<3 500% C
~
QI
"iij
40.0% g
~ 400% -+-TeaD removal efficiency 41
5;
~
-.Q-M ethane co ntenl
300% 1
a: 30.0% ~

20.0%

100% n.O%

00% +-------.---------,..-----r------+ 00%

o '0 20 30 40
TC 0010 ading rate (gJlIday)

Figure 1.3. rCOD Removal Efficiency and Methane Content vs rCOD Loading Rate.
In Table 4 and Figure 3, a TCOD removal efficiency of 70-75% can be achieved at the

TCOD loading rate of 10-15 gil/day. At this loading rate, a TSS reduction of 72-78% and

methane gas content of 70% can also be achieved.

For further aerobic post-treatment, a TCOD reduction efficiency of 45%-55%, TSS

reduction 43%-47%, TN reduction 40%-45%, P reduction 50%-55% can also be achieved

13
respectively. With the combination of anaerobic and aerobic treatment, the final effluent

quality is presented in Table 6. The reduction of COD from different stages of treatment

is shown in Figure 4.

Table 1.6. The Final Effluent Quality

TCOD SCOD Nitrogen (TN) p SS EffiuentpH
mg/L mg/L mg/L mg/L mg/L
650-750 350-400 91 9.06 50 8.0-8.5
Note: The TBOD j is in range of /5-22mglL.
8000

7000
• rCOD Remains in Different Reactor's Effluent
o SCOD Remains in Different Reactor Effluent
~6000
g
fsooo
c
Ii
i
rr: 4000
o
o
o
tJ)

.
!! 3000 1
o
o
~ 2000

1000

o
In Rl Reactors R2 R3

Figure 1.4. Effluent TCOD and SCOD in Different Stage ofTreatment

Integration
These pretreatment processes provided the required effluent quality that could be further

polished by incorporating the existing lagoon system. As analyzed, the existing settling

ponds may serve as the facultative lagoons. The criteria are shown in Table 7.

The size of the existing lagoon is closed to a typical facultative stabilization lagoon.

(Metcalf and Eddy 1991).

14
Table 1.7. Facultative Lagoon Typical Design Parameters Compared With Existed
L agoon
BODS
Pond Detention BODS Emuent
Flow Depth Temperature Optimum Loading
SIze Operation time pH conversion SS
Parameters regIme fl range OC temperature rate
acres (days) (%) mgIL
Ib/acre/d

Mixed
Typical Series or 6.5-
surface 50-180 85-90% 80-160
Lagoon 2-10 parallel 5-30 4-8 8.5 0·30 20
area

Plug 6.8-
Existed 4-7'
flow 0.6 Parallel 3.75 6 8.5 18-28 21(average) 125 85%
Lagoon

*The Influent after R3 treated to the eXIsted lagoon only contains about 50mgILSS. the SS In the final effiuent IS
expected to be 4-7mgIL.
The integration of anaerobic treatment plus aerobic treatment with the existing system

produced effluent quality: TCOD of 400-550mg/l, BODs of 8-14mg/l, TN of 30-40mg/l,

P of3-4mg/l, SS of 4-7mg/l, pH 7.5-8.5, which is suitable for flushing reuse.

Cost analysis
Based on the experiment and operation parameters, a sketch design was developed and

evaluated. Because of local topographic limitation, the concrete reactors are not suitable.

Suitable reactors are suggested to use are plastic reactors available commercially.

According to Cost Methodology Report for Beefand Dairy Animal Feeding Operations

(EPA, 2001) and Cost Model Reportfor Beefand Dairy Operations Developedfor the

Effluent Limitations Guidelines (EPA, 2000), total capital and O&M cost were estimated

by using 2002 price. The total capital cost and total 0 &M direct cost (yearly) are

$482,430 and $52,497, respectively. After the offset of byproducts sale, the total O&M

cost is $15,297/year. As suggested in Animal Waste Management Field Handbook

(USDA, 2000), the waste sludge both from anaerobic treatment process and aerobic

treatment process can be composted, which is an on-going process operated in this farm.

15
This also will improve the solid manure composting process and increase the value of

fertilizer. The details of Operation and Maintenance cost is presented in Table 1.8.

Tabl e I 8 TIC ota ost E' stImatIOn 0 fT reatment Ssystem WI'thBsyproduct UT

tllzatlon.
Total cost estimation of treatment system with byproduct utilization $ (2002)
Annual operation balance
Gas production for hot water ($) 16,000
Fertilizer value of sludge ($) 2,200
Land saved value ($) 18,000
Reuse of water ($) 1,000
O&M cost ($) -52,297
Balance (gas +Iand saved value+ Fertilizer +Reuse-O&M) -15,297
Although there are many successful cases of electnclty generatIOn by usmg methane gas

in the US mainland, the electricity conversion facility was not recommended because of

the limitation of electricity sale policy in Hawaii. Methane gas was calculated only for

the replacement of the current gas supply for hot water use. In addition, this farm has 10

acres of land that can be used for farming or lease for rental activity. Unfortunately, the

farmer cannot make use of this land because the wastewater treatment system was not

incorporated. If the treatment process is incorporated with the recommended system, the

land can be used for crops to generate further income.

Conclusion
Integrating an anaerobic treatment process and a simple aerobic process in dairy

wastewater is essential for the enhancement of environmental quality, especially for the

land limitation condition. With this integrated process, the final effluent quality for

TCOD, BODs TN, P, and SS were at 400-550mg/L, 8-14mgIL, 30-40mg/L, 3-4mgIL, and

4-7mg/l, respectively. The system provides the reduction of energy consumption cost and

promotes land usage. More importantly, the treated wastewater can be reused for floor

16
flushing. The cost analyses suggest that long tenn investment for such a system can be

reasonable and cost effective for dairy operations.

Acknowledgement:
The authors wish to thank the farm participation. This research was supported by

USDA/HATCH

References
1. Crook James, Surampalli Rao Y, (1996) Water Reclamation and Reuse Criteria in the

US. Water Science Technology, 33(10-11),451-462

2. EPA, (2001) Cost Methodology Report for Beef and Dairy Animal Feeding

Operations, U.S. Environmental Protection Agency, Washington, DC.

3. EPA,(l999) Feedlots Point Source Category Study, U.S. Environmental Protection

Agency, Washington, DC.

4. EPA, (2000) Cost Model Report for Beef and Dairy Operations Developed for the

Effluent Limitations Guidelines. Office of Water, U.S. Environmental Protection

Agency, Washington, DC.

5. Mogens Henze, Poul Harremoes, Jes la Cour Jansen, Eric Arvin, (2002). Wastewater

Treatment: Biological and Chemical Process, Springer, New York.

6. Metcalf & Eddy, Inc. (1991) Wastewater Engineering: Treatment, Disposal, and

Reuse, 3td ed,

McGraw-Hill, New York.

7. USDA, (2000) Animal Waste Management Field Handbook, Washington DC.

17
 ESSAY TWO:

Developing a 5-S Approach for a Milk Parlor Wastewater

Treatment and Reuse System

Abstract
A case study of dairy fann has problems include odor generation, sludge cleaning from

exiting lagoon system, potential groundwater and surface water pollution, etc. In order to

correct or improve these existed problems, a 5-S approach (system existed, short HRT,

small investment, simple operation and maintenance, system thinking) is developed.

Among the 5-S, effective development to achieve a higher SRT (Solids Retention Time)

from the bioreactor is essential in order to be able to integrate this into the other 4-S. A

bioreactor, which is the modification of UASB (Upflow Anaerobic Sludge Blanket)

reactor, was investigated. It was found that more than 70-75% of TCOD (Total Chemical

Oxygen Demand) removal efficiency and biogas production of 1.5UUday and CH4

content of 70-75% can be achieve at 8-15 g COD/l/day. An EMMC (entrapped mixed

microbial cell) operated with intennittent aeration was followed as post treatment. At a

HRT range of 24-48 hours with one-hour aeration and two-hour without aeration, further

removal of residual COD of 50-55% and of total N~-N of 92-98% from the effluent of

anaerobic bioreactor can be achieved. This will allow for the potential reuse of treated

wastewater on fann without creating negative impact towards the environment.

Keywords: Cost effective; dairy wastewater; integration; reuse; SRT, treatment

18
Introduction
The dairy fam1 and other Animal Feeding Operations (AFOs) have been widely

reorganized as significant cause of surface water impaiffi1ent, air pollutions and

groundwater contamination. Improper management of manure and wastewater from

concentrated animal feeding operations (CAPOs) is among many contributors to

remaining water quality problems (EPA, 2002).

Environmental problems associated with animal manure and wastewaters are the

degradation of surface water (e.g., lakes, streams, rivers, and reservoirs) and ground

water quality degradation, adverse effects on estuarine water quality and resources in

coastal areas and effects on soil and air quality. According to USDA (2000), dairy

faffi1 operations in the United States typically may have following problems. They are

1). Contaminated well (well water contaminated by bacteria and nitrates because of

leaching through soil). 2). Waste storage structure: Poisonous and explosive gases in

structure. 3). Animals in poorly ventilated building (ammonia and other gases) 4).

Waste applied at high rates (nitrate toxicity and other N-related leaching of N03 and

microorganisms through soil, fractured rock, and sinkholes.) 5). Discharging lagoon,

runoff from open feedlot, and cattle in creek, 6). Runoff from fields where livestock

waste is spread and no conservation practices on land (especially P and NH4 nutrients

reach stream). 7). Eutrophic conditions caused excess algae and aquatic weeds 8).

Leaching of nutrients and bacteria from poorly sealed lagoon.

A large scale dairy farm (with 60 acres area, 1600 livestock) located on the island of

Oahu, Hawaii, U.S., was investigated for developing an appropriate wastewater

treatment and reuse system. The farm is located in a sub-watershed with total area of

19
70 acres. It consists 50 acres land available for nutrient application, 10 acres for

wastewater treatment, and 10 acres of use for milking facility for 1000 milk cows. The

farnl annually produces 2.6x106 gallons milk (9,841m3) with 1.82x10 8 gallons
3
(47,556m) water consumption, and generates 1.8xlO8gallons/year (6.814xlO 5 m 3/yr)

wastewater. There is average 300,000 gallons / day (l,136m 3) wastewater needed to be

managed. Currently, local water company supplies the water for milk cows drinking.

Water for floor flushing, equipment washing and cow cooling is pumped from a well.

The wastewater generated from milking center flows to two uncovered and unlined

lagoons that each of them has a volume of 250x50x6 ft 3 (length, width, depth,

n.6xI5.2x1.83m 3). A 10 acres land planted with California grass is receiving the

lagoon overflow. Current water practice and wastewater treatment are illustrated in

Figure 2.1.

300.000 gallons of
wastewater need to be
treated

451ect
Well
House

ttt<i' 14r...
_.V
Water

-~--':"....L~---------- Aquifer water level

Figure 2.1 Current Water Practice and Treatment Problems

They are

• odor generation and spread to society community closed to the farm,

• ground water contamination,

• surface water pollution and potential nutrient runoff from overland,

• lagoon management.
20
Many efforts focused on dairy wastewater treatment technology and nutrient

application in the U. S. were taken for a period of time. Among the technology

alternatives applied are solids separator with liquid or slurry storage, anaerobic

lagoons, constructed wetland, anaerobic digesters, and aerobic lagoon and activated

sludge. In addition to the above technologies, direct land application is practiced

(USDA APHIS, 1996b). However, according to Kellogg (2000), most of dairy

farms(large operation more than 700 cows) do not have sufficient land (or no land

application) to apply.

Wetland is combined as an alternative for land application. However, increasing

problems of nitrate and nitrite were reported (Cronk 1996, Geary and Moore 1999,

Schaafsma et a1. 2000, Newman et a1. 2000).

Anaerobic lagoons have been popularly installed for dairy wastewater treatment

because of their advantages, such as cost effectiveness, simple management, and large

quantity storage volume of wastewater. However, their disadvantages, especially the

threatens of environment, were widely reported (Drommerhausen et aI., 1995, Davis

and Burgoa, 1995).

A large number of treatment techniques practiced in dairy farm wastewater and

manure management were transferred directly from the municipal wastewater

treatment field (Bicudo et aI, 2000) that may not be appropriate for the application in

the field of dairy wastewater management system. Also, as suggested by Hopkins et a1.

(200 I), a design engineer should attempt to design a process with optimal trade-off

between capital and operation costs. Apparently, the land based treatment (including

lagoon system, and constructed wetland), which are easy to achieve minimizing the

21
capital cost, but can not satisfy the feasibility of water practice, controllability, safety

and reliability that required by the more strict regulation for dairy farm operation.

Therefore a treatment /reuse system for dairy wastewater needs to be developed as a

more appropriate one for land limited condition, i.e., island application.

Method IApproach
5-8 approach: Development of a dairy wastewater management system should allow

to include potential integrating of existing system, simple operation / maintenance,

small investment, short hydraulic retention time(HRT) and system thinking. A local

dairy wastewater management system in the state of Hawaii was proposed to include

this 5-8 approach. This 5-8 approach presented in Figure 2.2. demonstrate their inler

relationship.

Figure 2.2. 5-8 Approach and Their Relationship

In this study, a local dairy farm owns a series of two lagoons (72.6xI5.2x1.83m3 each)

to receive the wastewater discharged from the milk parlor operation. As presented in

our previous study (WRRC 2003), the lagoon system is under deigned, i.e, is

overloaded. It is, therefore, a 5-8 approach is proposed to be able to integrate the

existing treatment system in order to achieve the odor elimination, biogas energy

recovery, treated wastewater reuse and improvement of water quality with cost
22
effective approach. Among these 5-S component, the development of an effective, low

operation/maintenance treatment system with low HRT (hydraulic retention time)/high

SRT (solid retention time) is required to achieve the 5-S approach proposed.

Anaerobic treatment: A two-stage bioreactors were installed and investigated as

presented in Figure 3. A void volume of 8.6 liter each was operated by an intennittent

feeding schedule of 1 hour-on and 1 hour-off. This is an once through (without

practice of sludge recycle) operation system which is maintained in the ambient

temperature of 22 ± 2°C. Input and out COD (chemical oxygen demand), suspended

solid (SS), nitrogen contents were measured.

Gasbag

810 est

F«ding B~ CSt
Tnk

Effluent to EMMC
R1 R2
Figure 2.3. Two-Stage Anaerobic Pretreatment Experiments

EMMC (Entrapped Mixed Microbial Cell) reactor: Two EMMC reactors with

effective volume of 5 liter each were installed and operated. Their experimental set-up

is presented in Figure 4. This unit was operated in a two-aeration schedule, one is

continuous air supply and another is an intennittent aeration schedule. The

investigation of using EMMC technology for simultaneous removal of carbon and

nitrogen were previously reported (Yang et al 1997, Yang et al 2002a, Yang et al

2002b). Integrating this technology for further treatment of the anaerobic reactor

effluent may be required depending on the reuse/disposal scheme to be followed.

23
Especially, further removal of COD and nitrogen for further reuse of flushing the

milk-parlor floor requires a better water quality in order to prevent the accumulation

nitrogen and COD in both the production and treatment systems. This EMMC unit

was operated at HRT of 24 hour and 48 hours under the room temperature of 22 ±2 °c.

EffiUCDl

Feeding tank l-- L...- Air_ supply

RI R2

Figure 2.4. EMMC Reactor Structure and Flowchart

Results and Discussion

Wastewater characteristics The chemical physical characteristic of wastewater from

milking center was summarized and is presented in Table 1.

Table 2.1. Characteristics of Milk Parlor Wastewater

Temperature TSS VSS TeOD seOD TKN Phosphorus

Source of Washing (gil) (giL) (mgIL) (mg/l) (N Main
washing times/day mgl1) (Pmgll) component
Milk
Parlor 3 65 10.5 8.7 17500 3710 Milk
Walking
way Manure
flushing 3 27 3.21 2.35 12000 2200 and foods
Manure
flushing 3 27 3 2.1 4180 1300 Manure
Average
(26 of
samples) 28 4.2 3.1 5200 1900 200 20.1 Mixed
Note: Milking cows are milked 3 times in the tropics as compared to only 2 times in most semitropic.

The biodegradability for TCOD and SCaD was investigated and found as 88%-93%

and 75%-80%, respectively. The correlation between COD and BODS (fresh

wastewater) can be expressed by a simple linear regression equation:

24
COD=a x BOD5 + b. For this dairy farm wastewater: COD= 2.46 BOD5+126. R2

=0.95

For anaerobic unit effluent, COD and BOD5 correlation can be estimated as COD=

7.67 BOD5+382, R2=0.89. Based on the last equation, if the residual COD is less

than 382, BOD5 is not detectable.

Evaluation and technology alternatives: the crucial problems of the existing system

and potential solution were evaluated from economic, ecologic, environmental view.

The alternated process and solution was summarized and is presented in Figure 5.

25
 r-~L-~---- -;..----:::.~~;;::E~:~:~--·LNo-~:-- --1
* ~ Transfer to Next ProCess ••• ••••••• i •••••.•••• ,

Priority for
newsystern

Future treatment and reuse systems

Figure 2.5. Existing System Problems and Potential Solution Evaluation Process

As presented in Figure 5, development of a pretreated process to be integrated with the

existing lagoon system is acceptable. This will provide potential solution of the

existing problems of odor, treated wastewater reuse, improvement of water quality,

recovery of bioenergy as methane gas and protecting groundwater contamination. As

listed, a single biological treatment unit process will not be able to provide the solution

for the above listed problems. Therefore, an integrating two or more unit processes is

required in order to provide the solution of the listed problems for the local dairy farm.

26
Anaerobic Process Results: As presented in the Table 2 and Figure 6, the TCOD

(total COD) loading rate changed from 2.68 to 36 g/l/d were investigated for the

combination Of Rl and R2.Removal efficiency of TCOD and biogas production rate

is 52%-80% and 1.33-2.22 Ill/day, respectively. These results are comparable with

other studies for the dairy wastewater (Ince, 1998, Gavala, et ai, 1999). It is very

useful to use this result to determine whether further treatment is required if

reuseldisposal is decided.

Table 2.2. Anaerobic Pretreatment Process (Rl+R2) performance

Methane
Gas Vield pH changing
TooD SeOD Methane LJgeOD
Units g/LJday HIS (%) (%) (%1 (%) (%) LJl,lday removal R1 R2 IN
6.8-
1 2.68 60 75.4 45.6 76.0 88 51.0 1.33 0.310 7.3 7.3 8.1
2 3.2 50 80.1 59.3 74.0 80 50.5 1.91 0.290 7.2 7.3
3 8.3 25 77.5 56.5 72.0 79 49.2 2.14 0.259 7.1 7.3
4 17 15 72.7 54.4 70.0 72 43.2 1.70 0.180 6.9 7.1
5 25 15 69.4 31.6 69.0 71 36.1 1.73 0.166 6.4 6.9
6 32 11.5 67.4 36.0 63.0 69 29.2 2.22 0.139 6.3 6.6
7 36 10 52.0 28.9 57.0 62 21.5 2.04 0.116 4.5 6.4
8 15.5 15 70.3 57.2 70.1 76 45.3 1.63 0.200 7.1 7.3
Note: In period 8, the treatment process was operated with various reOD influents.

90.0% ..--~-------~----...,. SO.O%

so.o% 70.0%

70.0%
60.0%

..
60.0%

~..
50.0%
~50'0%1
1 30.0%

20.0%
-+- TeOD retnoval efficiency
-b--M ethane content
30.0%

20.0%

"0.0%

0.0% +-----.----..,.---__- - r - - - - - ! - 0.0%

o "0 TeOD IOadln~Orate(glLJday) 30 40

Figure 2.6. TCOD Removal Efficiency and Methane Content vsTCODLoading Rate.

27
 EMMC Technology As shown in Table 3, both the results of continuous and

intermittent aeration is presented. The removal efficiency of TCOD and total nitrogen

is about 41-50% and 25.7-51.5%, respectively. For the combination of these two

bioreactors, since the main purpose of this study was focused on the reuse of treated

wastewater for floor flushing and recovery of biogas for the water heating, it is,

therefore, that only one EMMC reactor may be required. The high content ofeffluent

N03-N (e.g. 159mgll) can be used for floor flushing which will· be automatically

introduced to the anaerobic treatJ:l1ent and lagoon system. This may provide further

nitrogen removal through denitrification process, which requires further study. This is

l>ased on our previous study (Yang, et al.,2002) for simultaneous removalof carbon

and nitrogen research. A similar result waS conducted by Barber and Stuckey's study

(2000).

Table 2.3. Process Performance for EMMC Technology for Anaerobically Treated
Effluent
Influent Influent Effluent Effluent R2 R2 Effluent Overall Overall Overall
Operations Influent NH4 Effluent TN TeDO
TCOD SeOD NH4* N03 effluent frOm Rl SeOD
frOm Rl frOm Rl NH4 N03 TooD/SCOD nsmovel nsmovel removel
mgn NH4·N NH4·N N03·N NH4·N N03·N
mgn mgll mg/l mgll mgn mgn mg/l % % %
eontinuous
HRT"24hr 906 613 210±20 6.5 160-175 4.5 159 5281300 25.7% 41.6 58.4
1 hauronl
lhourall 900 620 210±15 6.3 150-175 6 97 507/405 51.5% 50.0 49.2
HRT=24hr
1 hour ani 2
hour 011 816 642 9O±15 NlA N/A 5 48.6 420/328 40.4% 48.8 47.4
HRT=48hr
Note: influent for EMMC was operated by following up the optimal TCOO loading range of anaerobic treatment unit.

Integration_Based on the results generated from this study, an integration of the exiting

system with the·proposed treatJ:l1ent component can be d~veloped as presented in Figure 7.

The existing two-lagoon system can be served as the stabilization portdsas reported

(Metcalf and Eddy, 1993). The anaerobic units proposed in this system can be operated

with a 10aditlg rate of 15-20kglm3/day (HRTI0-15hours) and EMMC can be operated

with HRT of9-12 hour. The effluent from these pretreatment units contains COD of less

28
than 400mg/1 (BODs < 5mg/l), SS less than 5 mg/l, nitrate-N of 130-150mg/l, P of 5-

6mg/l, pH of 8-8.5. According to US EPA report (Crook and Surampalli, 1996), the

present water quality is suitable to be reused for floor washing/flushing.

I--------------~
; - - - - -- -- - - - - --- - - -- -- -- - - - - -
I
~ Biogas for hot water
1------- -------
!+---------:.
I
1
I
Inlet Anaerobic
I Reactor 2
I
I
I
I
I
I
I
I
I
Solid. compost Sludge compost Equalization
I.';"
Tank

Figure 2.7.· Proposed System Flow Chart

Based on the design and operation criteria proposed, the capital cost was estimated as

about $761,600 and annual operation/maintenance cost at $79,800 with byproduct

recovery (methane gas, land savage, fertilizer utilization). This is based on life

expectancy of 15 years.

Based on this cost estimation, it is apparent that the exploration of more biogas utilization

is important to reduce the cost of operation. The current cost estimation is only based on

one fifth of the utilizationofbiogas for water heating.

Conclusion
Based on experiment and . evaluation results, the high strength and complex milk parlor

wastewater, both COD and SS can be biologically reduced to an acceptabl~ level for

reuse by integrating an anaerobic pretreatment and EMMC aerobic post treatment ifit is

necessary. By using a 5-S approach, existed problems in this large dairy farm can be
29
resolved by the proposed system. Especially, by making treated wastewater reuse for

floor flushing, environmental impact, such as groundwater contamination, odor problem,

bioenergy recovery and improvement of water quality, can be reduced or avoided.

Economically, the proposed system capital cost was estimated at an affordable level, and

the O&M cost can be reduced by energy recovery to achieve low O&M cost. The 5-S

approach provide an effective system thinking can be integrated with the existing

treatment system to achieve a low cost, low operation! maintenance for protection of

e.Ilvironmental quality

Acknowledgment: This work was supported by USDA-Hatch 550 grant and College of

Tropical Agriculture and Human Resource, University of Hawaii at Manoa. Assistances

from Dr.C.N. Lee (Dept. of Human Nutrition, Food and Animal Science) and Mr. David

Wong, Dairy farmer are greatly appreciated.

Reference:
APHA, AWWA and WEF. (1992) Standard Methods for the Examination of Water and

Wastewater, 18th ed. American Public Health Association, Washington,. DC.

Barber William P. and Stuckey David C. (2000) Nitrogen removal in a modified

anaerobic baffled reactor (ABR): 1, denitrification. Wat. Res. 34(9), 2413-2422

Bicudo, J. R., Westerman P. W., and 01eszkiewicz J. A. (2000) Anaerobic and aerobic

treatment ofanimal manure'" A review. In: Proceedings of Animal Residuals

Management Conference - Issues and Solutions. Nov. 12-14, Kansas City,

MO. Water Environment Federation, Alexandria, VA.

30
Crook James and Surampalli Rao Y. (1996) Water reclamation and reuse criteria in the

U.S. Wat.Sci.and Tech., 33(10-11),451-462

Davis J.G. and Burgoa B (1995) Runoff and leaching of crop nutrients from soil in tilted

beds as influenced by three rates of dairy lagoon effluent, Journal ofEnvironmental

Science and Health, 830 (2) 243-264

Donkin Michael J. and Russell John M. (1997) Treatment of a milk powderlbutter

wastewater using the AAO activated sludge configuration, Wat.Sci.and Tech.,

36(10), 79-86

DrOll1l11erhausen D.l; Radcliffe,-D.E.;Brune,-D.E.;Gunter,-H.D. (1995)

Ele.ctromagnetic conductivity surveys of dairies for groundwater nitrate, Journal of

Environment Quality, 24 (6) 1083-1091

Gavala H. N., Kopsinis H., Skiadas 1. V., Stamatelatou:k. and Lyberatos G. (1999)

Treatment of Dairy Wastewater Using an Upflow Anaerobic Sludge Blanket

Reactor, Journal ofAgricultural Engineering Research 73(1), 59-63

Gavrilescu M., 2002 Engineering concerns and new developments in anaerobic waste-

water treatment, Clean Techn Environ Policy 3:.346-362

Geary P. M.and Moore J. A.(1999) Suitability ofa Treatment Wetland for Dairy

Wastewaters, Wat.Sci.and Tech., 40( 3), 179-185

lnce O. (1998) Performance of a two-phase anaerobic digestion system when treating

dairy wastewater, Wat. Res. 32(9),2707-2713

Kellogg, R.L., C. Lander, D. Moffitt, and N. Gollehon. (2000) Manure nutrients

relative to the capacity of cropland and pastureland to assimilate nutrients: Spatial and

temporal trends for the U.S. April 27, 2000.

31
Lettinga G., Field 1., Lier 1. van, Zeeman G. and HulshoffPol L. W. (1997) Advanced

anaerobic wastewater treatment in the near future. Wat. Sci. and Tech., 35(10),5-12

Meals Donald W., (1996), Watershed-scale response to agricultural diffuse pollution

control programs in Vermont, USA Wat.Sci.and Tech., 33(4-5), 197-204

Metcalf & Eddy, Inc. (1991) Wastewater Engineering: Treatment, Disposal, and Reuse,

3td ed, McGraw-Hill, New York.

Nadais H., Capela I., Arroja L. and Duarte A (2001) Effects of organic, hydraulic and fat

shocks on the performance ofUASB reactors with intennittent operation

Wat.Sci.and Tech., 44(4), 49-56

Newman Jana Majer, Clausen John C., and Neafsey Joseph A.,(2000) Seanal

performance of a wetland constructed to process dairy milkhouse wastewater in

Connecticut, Ecological Engineering 14, 181-198

Qian,X, Yang P.Y. and Maekawa T., (2001) Evaluation of direct removal of nitrate with

entrapped mixed microbial cell technology using ethanol as the carbon source

Water Environment Research 73(5), 584-589

Schaafsma Jennifer A., Baldwin Andrew H.and Streb Christopher A(2000), An

evaluation of a constructed wetland to treat wastewater from a dairy farm in

Maryland, USA. Ecological Engineering 14, 199-206

U.S. EPA (2001) National Man'!gement Measures to Control Nonpoint Source Pollution

from Agriculture, U.S. EPA, Washington DC.

U.S. EPA,(2002) National Pollutant Discharge Elimination System Permit Regulation

and EjjluentLimitation Guidelines and Standards/or Concentrated Animal Feeding

Operations, Washington DC, 2002

32
USDA APHIS. 1996a. National Animal Health Monitoring System, Part I: Reference of

1996 dairy managementpractices. U.S. Department of Agriculture (USDA),

Animal and Plant Health Inspection Service (APHIS). Fort Collins, Colorado.

USDA, 2000 Agricultural Waste Management Field Handbook, ppll-24, U.S.

Department of Agriculture (USDA), Washington DC

Willers H. C., Karamanlis X. N. and Schulte D. D.(1999) Potential of closed water

systems on. dairy fanus. Wat.Sci.and Tech., 39(5), 113-119

WRRC (2003) Conference of Scientific, Regulatory and Caltered Facts, Influencing

Water and Environmental Issued in Tropic Pacific Island, January 15-16,2003,

Honolulu, Hawaii, 2003

Yang, P.Y., Zhang, Z.Q., and Jeong, B.G. (1997) Simultaneous removal of carbon and

nitrogen using an entrapped-mixed-microbial-cell process Wat. Res. 31(10), 2617-

2625

Zeeman Grietje, Sanders Wendy T. M., Wang Kaijun Y. and Lettinga Gatze (1997)

Anaerobic treatment ofcomplex wastewater and waste activated sludge -

application of an upflow anaerobic solid removal (UASR) reactor for the removal

and pre-hydrolysis of suspended COD. Waf.Sci.and Tech., 35(10), 121-1128

33
 ESSAY THREE:
Design and Analysis of an Anaerobic Bionest Reactor for Milk
Parlor Wastewater Treatment
Abstract
Milk parlor wastewater contains highfat and lipids. These contents make the

conventionalllpflow anaerobic sludge blanket (UASB) reactor to be difficult for the

granulation and cause sludge flotation and washing·out. A two-stage anaerobic reactor

(Bionest) filled with special media was designed and investigated for 6 months at the

ambient temperature of 22±2 °c. A range of total chemical oxygen demand(TCOD)

loading rate at of2.68-36g1l/d was investigated for its performance and stability. It was

found that the TCOD removal efficiency could be achieved at 520/0- 75% with.n1ethane

content of 56-76%. At the TCOD loading rate of 15-20 gll/d, the system is stable for

both COD removal (71%-75%) and gas production at(2.2l/lId-2.5Vld). Compared with

conventional UASB and anaerobic filter, the "bionest" reactor is able to provide a better

sludge distribution, less dead zone and higher TCOD loading rate for the treatment of

milk parlor wastewaters or other organic wastewater with high lipid/fat COntent.

Keyword: Anaerobic reactor, COD removal, biogasproduction,design, bionest, sludge

distribution, milk parlor wastewater.

Introduction
Approximately, 1400 upflow anaerobic sludge blanket (UASB) plants were built

worldwide for the full-scale. application (Frankin, 2001). The UASB high rate reactors are

based on the difference of solid and liquid retention time by applying the granule blanket

34
(Lettinga et al., 1980; Marin et al., 1999). However, the UASB granule blanket reactors

have some limitations because the granulation formulation and property must depend on

wastewater types rather than reactor design and operation (Jhung J.K. and Choi E. 1995,

Batstones and Keller, 2001). For instance, the dairy wastewater, which contains high

concentration offat and lipids, the granule, is difficult to form (Hawkes et at, 1995).

There is only a few UASB application experience in dairy wastewater treatment. Instead

ofsuccess, several reports (Hawkes et al., 1995, Perle et al. 1995, Petruy and Lettinga

1997, Vidal et al., 2000,) published the failure or poor results in treating dairy wastewater

by usingUASB reactors.

The main reason caused the poor performance ofUASBis the lipids attached to the

granules which causes the granule broken down, floating and washing out (Perleet al.

1995; Petruy and Lettinga, 1997). As Alphenaarand Lettinga (1993) concluded that

UASB reactors were not suitable to be applied if lipids contribute 50% 01' more of the

chemical oxygen demand (COD) of wastewater. If lipid loading rates exceeded 2-3 kg

COD m-3 day"l, granule would be broken down and total sludge wash-out occ\lITed. Even

at lower loading rates(less than 1.5g1Vd), the system was still unreliable because of

unpredictable sludge flotation.

The milk fat and lipids also caused the inhibition of methanogenic activity (Perleet al.,

1995). Thisinhibition made the fat and lipid hard to be biodegraded anaerobically. It was

suggestedthat anaerobic treatment need to be staged up to prolong the methanogenic

process. Single stage UASB, is questionable ifdairy wastewater content is more than

100mgli of lipids. Under thermophilic condition, lipids and long chain fats were

35
relatively easier to be biodegraded than under mesophilic condition, but the economic

cost is very high·if external heat source was not available (Petruy and Lettinga, 1997).

Another limitation of applying UASB is the dead zone or shortcut of wastewater flow in

the conventional UASB (Lens et aI, 1998). For the medium and high strength wastewater

(COD>2500mg/l ), although the high organic loading rate(OLR) of more than 20 kg

3
COD/m /day could be achieved in the laboratory scale, however, for full sale application,

UASH were generally designed not exceeding 15kg/m3/day at an optimum temperature

35°C (Driessen and Yspeert, 1999). Frankin (2001) reported thatthe averageOLRwas

about 10 kg COD/m3/day. Regardingthis concern, were the mass transfer limitations d.ue

to sub-optimal mixing condition. inside the reactor and high upflow velocity and gas

production causing granules wash-out were investigated by (Driessen and Yspeert,. 1999).

The increasing of the upflow velocity may cause· too· little contact time between the

granule and soluble substrate. Also, a dead zone where containing less sludge or poor

mixing inside the reactor was reported (Lens et aI, 1998, Chernicharo and Cardoso, 1999).

Another problem is the poor reliability of operating the biologicaltreatment systems for

dairy wastewater. In most cases, UASB reactors were operated by using continuous

feeding. It was reported that scum layer and sludge layer on the top of reactor were

subsequently washed out (Rinzema, 1988; Yang, 1994). The solids content was

accumulated in the sludge bed, which led the poor reliability of operational performance.

In order to solve these problems,severalresearchers (Sayed, 1987; Lettinga and Hulshoff

Pol 1991) suggested using the flocculants sludge instead of granules. Also, it was

36
reported that the intennittent feeding operation could improve system performance

compared with continuous feeding operation (Nadais et aI., 2001). Thus, several VASB

modifications, such as upflow anaerobic buffer reactors (UABR) and upflow anaerobic

filter reactor results were reported for the application for treating milk wastewater

(Hawkes et. aI., 1995; Cordobaet. aI., 1995; Vartak et. aI., 1998). A total chemical

oxygen demand (ICOD) loading rate of 6...,8 gllId and 66...-69% TCOD removal

efficiency were achieved at HRT of22 hours (Hawkes et aI. 1995). Instead of granule

blanket sludge, the floes sludge was applied in these modifided reactors. The main

advantage of these reactors were maintained high sludge concentration and anti-sludge

washing out by using filters, media and baffles.

In order to correct the above weaknesses provided by the operation of conventional

VASB and take the advantages of anaerobic filter reactor, a "bionest" reactor which was

initiated and developed for investigating potential treatment of milk parlor wastewater.

Material and Method

A two-stage bioreactor (10 liter each) as shown in Figure 1, was designed and filled with

special media - "bio-nest". They were filled with two layers ofmedium providing a 98%

ofvoid volume in the reactor. The reactor was made of transparent PVC in order to be

able to observe the sludge formation and accumulations. Three sample outlets were

designed to test the sludge distribution at the different height of 15 cm, 65 cm arid 90 cm

respectively. A special puncture needle was designed to take samples from the bionest.

Table 3.1. shows the main parameters of the two-stage bionest rellctors and bionest

content, Figure 3.1,3.2., and 3.3. show the reactor structure, appearance and bionest

structures, respectively.

37
Table 3.1. Basic Parameters of the Bionest Reactors
Reactor 1 Reactor 2
(Rl) (R2) Bionest layer
Filled with 2 layers Filled with 2
bionest layers bionest
Black PVC clear tubing was
into 0.02mm-0.05mm PVC
Material PVC PVC
String and was placed into a
certain layer
Weight( kg) 3.86 3.86 0.35 kg
Diameter(in/cm) 4.5/11.4 4.5/11.4
Height(ftlm) 3.3611.024 3.3611.024
Height/Diameter Ratio 8.9 8.9
Total volume(L) 10 10
Effective Volume(L) 8.6 8.6 0.12
Void Volume (L) 8.48 8.48
Void Volume percentage of
effective volume (%) 98.6 98.6

Gasbag

Effluent

Feeding
Tank
.- Snmp I'109
Oullet

R2
R1

Figure 3.1. Two-Stage Bionest Reactors

38
Figure 3.2. Bionest Reactor Appearances.

Figure 3.3. Bionest Materials and Structure.

39
Surface Area and Porosity:
The bionest was made of waste material of black PVC tube cutting. The long string shape

material was twisted and pressed in order to increase surface area.

The area of the reactor wall can be calculated by the following equation:

Where Areactor is the area of the reactor wall;

d is the diameter of the reactor;

H is the height of the reactor.

The surface area of the bionest can be estimated by using the following equation:

A nest =2w·l
Where w is the width of the PVC leaf; I is the length of the leaf. Because of the nest provided

two sides for the attachment of microorganism, the surface area calculated need be doubled.

For the specific area per weight ofbionest can be calculated according to the parameters

provided in Table 1:

The specific surface areas the twisted bionest provided can be calculated by using the following

equation:

rrt
-Vs =33,000-3
ill

The Voidage can be calculated as:

Voidage = VI.10<J!Jo
V2

Where V 1 is media actual volume;

V 2 is the nest replaced the water volume.

40
 The bionest provide 33,OOO-80,OOOm2/m3 specific area which depend on the nest density or

the leaf thickness. Comparison was presented with traditional media in the Table 3.2..

Table 3.2. Surface-to-Volume Ratios of Selected Fixed Film Support Media

Media Void age SN Source:
Process/Carrier Type (%) (m2/m3)
Malina 1992
Downflow Fixed Film 50-95% 70-100
Malina 1992
Upflow Modular Medium 90-95% 85-100
Malina 1992
Upflow Random Medium(9-15cm) 90-95% 90-200
Malina 1992
Expanded Bed(0.3mm) 45-55% 9000-11000
Malina 1992
Fluidized Bed(0.3mm) 50-300% 4000-10000
33,000- Present study
Bionest >700% 80,000

Operation Procedure:
Wastewater was sampled from milk parlor in a local farm and stored in a refrigerator (0-

40Q. It was fed into reactor directly from the refrigerator. The seeded sludge was

collected from an existing anaerobic reactor operated by batch.

In order to test the sludge distribution and compare different operation schedule, sludge

was sampled both at the start of feeding period and at the end of feeding period. Sludge

sampling was collected from the sampling outlets of the different height. The sludge from

the bionest was taken by using long puncture needle which connected with vacuum ball.

The system was operated for 6 months in a room with temperature range of 22±2 DC.

To compare the sludge distribution, a controlled single stage reactor filled with same

sludge content was tested. Another aerated batch reactor was installed and operated for

30 days to measure the wastewater biodegradability.

41
By using different TCOD concentration of wastewater and controlling the flowrate, the

system was operated by gradually increase of the overall TOCD loading rates from

2.68g/l/d to 36 gil/d. At each loading rate, it was operated to achieve steady-state for 3

times ofHRT or 5 days (ifHRT was less than 24 hours) applied.

Analysis Method:
Wastewater was fed intermittently by using timer controller setting at one hour on/off. In

each feeding operation, samples were collected from the input (feed) and output (digested

effluent) for the chemical analysis ofTS, TVS, TCOD, SCaD, ammonium nitrogen and

total phosphorus by following the procedure provided by the Standard Methods (APHA,

AWWA, and WPCF, 1989). Chemical analysis was conducted daily after sampling. The

pH values were measured immediately by using an Orion ion analyzer (Model 501).

Biogas production was measured by a field gas meter. Methane content was analyzed

weekly by using Shimadzu Model GC-3BT gas chromatography connected to a

Hewlette-Packard 3390A integrator.

Wastewater Characteristic
Characteristics of wastewater generated from a local milk farm in Hawaii characterized

and are presented in Table 3.3 ..

Table 3.3. Characteristics of milk parlor wastewater (average of sample analysis)

Temperature TSS(g/L) VSS(g/L) TKN Phosphorus

Source of Washing TCOD SCOD (N Main
washing times/day (oC) (mg/L) (mg/I) mg/I) (Pmg/I) component
Milk Parlor 3 65 10.5 8.7 17500 3710 Milk
Manure.
Walking way
3 27 3.21 2.35 12000 2200 milk and
flushing
foods
Manure flushing
from the cow 3 27 3 2.1 4180 1300 Manure
waiting area
Average(26 of
28 4.2 3.1 5200 1900 200 20.1 Mixed
samples)

42
The wastewater generated from washing milking equipment was about 8% of the

volumetric percentage, but it contributed about 17~22% of the total COD in the mixed

wastewater. The average lipids content is 446mg/I which result in an average of 512 mg

TCOD/L (9.8% of total COD). The biodegradability ofSCOD was 75%-80% based on an

aerobic treatment process using a batch operation.

Startup
Each reactor was fed 2 I of anaerobic sludge with TVS/TSS at 32/40 and 6.2 I raw

wastewater with a SCOD/TCOD of 1280/5460mg/L. The effluent from R2 was

constantly recycled at flowrate of 2 literslhour for stabilizing seed sludge. Gas was

collected by using airbag and measured by the field gas meter.

After 2 days operation, sludge started to attach the up-side leaf; after 7 days, biomass

started to grow at backward leaf and the up-side sludge became thicker; after 12 days

operation, up-side leaf was completely covered by the sludge. Additionally, II cm sludge

layer was formed in the bottom of reactor. As shown in Figure 4, about 50% SCOD has

been removed after 2 weeks of starting-up. Application of effluent recycling was stopped

and planned experiments were started with various organic loading rates. The process

performance with various loading rate was started to monitor.

43
 Start up SCOD discline

1400

1200

1000
c

.Ii
:ll
~
u
800

FSi8lfup sceD diSCiffi8J

8u 600
Ul

II!
400

200

0
0 4 10 12 14 16
Operation days

Figure 3.4. SCOD Reduction during the Startup Period.

Results and Analysis

General Performance
As presented in the Table 4 and Figure 6, the TCOD (total COD) loading rates of 2.68

to 36 g/l/d were investigated for the combination ofR! and R2. Removal efficiency of

TCOD and methane gas production rate are 52%-80% and 0.63-3.0 lIl/day,

respectively. These results are comparable with other studies for the anaerobic

treatment of dairy wastewater (Ince, 1998, Gavala, et aI, 1999). Both methane gas

yield and gas content decline when TCOD loading rates increase. This result may

indicate methanogenic process may not complete when TCOD loading rates increased

more than 17 g/l/d and the HRT was less than 15 hours. At the TCOD loading rates of

10-15 g/lId, methane gas yield of 0.20-0.22 l/g COD and TCOD and SCOD removal

efficiency are in the range of71-75%, 54-55%, respectively.

44
Table 3.4. Anaerobic Pretreatment Process (Rl +R2) performance
Methane
TCOD TSS gas
Loading R1+R2 removal Gas removal Production Melhane
Period Rate HRT efficiency Quality efficiency rale Gas Yield Effluent pH changing
TCOD SCOD Methane UgCOD
Units glUday Hrs ('Yo) ('Yo) (%) ('Yo) UUday removal R1 R2 IN
6.8-
1 2.68 60 75.4 45.6 76.0 88 0.63 0.310 7.3 7.3 8.1
2 3.2 50 80.1 59.3 74.0 80 0.74 0.290 7.2 7.3
3 8.3 25 77.5 56.5 72.0 79 1.67 0.259 7.1 7.3
4 17 15 72.7 54.4 70.0 72 2.22 0.180 6.9 7.1
5 25 15 69.4 31.6 69.0 71 2.88 0.166 6.4 6.9
6 32 11.5 67.4 36.0 63.0 69 3.00 0.139 6.3 6.6
7 36 10 52.0 28.9 57.0 62 2.17 0.116 4.5 6.4

90.0% 80.0%

~
80.0% 70.0%

70.0%
60.0%

r~ 50.0% \: -+- TeOD removal efficiency

50.0%

40.0%
40.0%
30.0%
30.0% --{:r- M elhane co nlenl

20.0%
20.0%

1).0% 1).0%

0.0% 0.0%
0 1) 20 30 40
TCOD loading rale(g/Uday)

Figure 3.5. rCOD Removal Efficiency and Methane Content vs rCOD Loading Rate.

Process Stability
In order to observe process stability, the system was fed by the wastewater with

different TCOD and SCOD concentration and was operated at a constant HRT of 15

hours. As shown in Figure 3.6., it is found that although the influent TCOD is

fluctuated in the range of 5300~12000 mg/l, the effluent TCOD from RI and R2 were

very stable.

45
 50

__ R1 TCOD Loading rale

45
__ R2 TCOD Loading rate
__ Over all TCOD Loading rate
40

>; 35
m
"0
::J
:9 30
.l!l
l!!
g> 25
'ti
m
0
~ 20
0
()
I- 15

10

0
110 115 120 125 130 135 140
Operating days

Figure 3.6. TCOD Tolerances with Different Influent Concentration in the Period 8

pH
The changes of pH at different height of R] and R2 for various organic loading rates are

presented in Figure 3.7. and 3.8., respectively. It can be found that design of the two

stages of "Bionest" anaerobic reactor is required. This is to be able to accommodate the

higher organic loading rate and to provide a "buffer" zone for achieving the formation of

methane gas in the R2 under TCOD loading rate of 25 gil/d. Both TCOD loading rates of

32 and 36 gilld are not well enough to achieve the methane gas production, although the

two-stage bioreactor was installed. Apparently, more stages ofbioreactor or bioreactor

with higher SRT are required to accouter the TCOD loading rate of32 gil/d.

46
 7.5
-lI!-OLR=8
__ OLR=17

7 -lI-OLR=25
~OLR=32
__ OLR=36

6.5

...
:J:

5.5

4.5

o 10 20 30 40 50 60 70 80 90
Reactor 1 Height (em)

Figure 3.7. pH Changes inside the Reactor 1 with Different Loading Rates.

47
 R2 pH changing vs different loading rate at different height

7.5

5.5
o -36

5
o 10 20 30 40 50 60 70 80 90
Reator Height (em)

Figure 3.8. pH Changes inside the Reactor 2 with Different Loading Rates.

SRT and HRT

The SRT (RI +R2) value was estimated by using the mass balance of nitrogen input and

output. The following equation was used to estimate SRT.

SRT= (TNinfluent - TNejJ1uenrTNsoluble)/ (TNejJ1uenrTNsoluble)/day

Where TN_influent is the concentration oftotal nitrogen of influent, N-mg/l;

TNe./fluent is the concentration oftotal nitrogen in the effluent, mg/l;

TNsoluble is the soluble nitrogen in the reactor, mg/l;

(TNejJ1uenrTNsolubIJ is the daily discharged nitrogen in the biomass, mg/l/day.

48
Tabl e 35The R eIaf Ions h'IP among SRT, HRT andOLR
Period HRT(hours) OlR(g/l/d) SRT(days)
1 60 2.68 114
2 50 3.2 110
3 25 8.3 106
4 15 17 103.2
5 15 25 104
6 11.5 32 98
7 10 36 89
The SRT was calculated by uSing N balance:

120

100
SRT vs HRT(R1 +R2)
,,
80 ,,,
, ,
,,
i I

~ ,,
60 I

en I
I

,,
I

40
,,,

20

o +--------.--~-~_______,_
o 10 20 30 40 50 60 70
HRT(hours)

Figure 3.9. The Relationship between SRT and HRT

As shown in the Figure3.9., when the system is operated at HRT> 15 hours, SRTs are

relatively independent from the HRT. While the system is operated at HRT <15 hours,
49
SRT is affected by the HRT seriously. The dash line indicates that sludge would never be

washing out. For this design and this high milk lipid content wastewater, more than 15

hours ofHRT is needed achieve the stable COD removal and methane gas production.

For high fat and lipids content wastewater, Miron (2000) reported, at 25°C ambient

operation condition, SRT~8 days resulted in acidogenic conditions with negligible

biogas production, whereas an SRT>8 days resulted in methanogenic conditions. The

lipids and carbohydrate hydrolysis process increased with increasing SRT, whereas the

protein hydrolysis only occurred under methanogenic conditions. Apparently, at

mesophilic condition, the way to get high lipids conversion is to increase the SRT and/or

increase stages ofbioreactors.

Sludge Layer and Distribution

After startup period, the sludge layer at the bottom portion was gradually increased along

with the operation time. After 72 days, the sludge layer in R1 was fully developed from

the bottom to the top of the reactor; after 81 days of operation, the sludge layers in R2

was fully developed. The sludge layers in both Rl and R2 were clearly distinguished in

three layers by color differences with the bionest installing positions. Densities measured

by using concentrations of TSS and TVS confirmed the differences. At height of 0-19cm,

the sludge layer had TSS concentration of 43-47g/l; at the height of 40~ 65cm, sludge

layers had a concentration of 23-32g/l; at the height of 70-85 cm, the sludge layers had a

TSS concentration of 18-27g/1. Inside the bionest, the sludge concentration was in the

range of 26-30g/1. The average VS/TSS ratios in the three layers (from bottom to the top)

were 84.1 %, 80.3% and 78.0%, respectively. During the whole operation period, no

50
clogging and foaming occurred. As shown in Figure 3.10., the sludge distribution in the

reactor I is compared with that in a same size of reactor without "Bionest" installation.

Sudge Distribution: Blonest Reactor vs Batch Reactor

50

~~~-----~---
45 • BionestReactor Sludge distribution
m Batch R~~ctor Sludge Distribution
40

35

30

TSS(g/Lj 25

20

15

10

o
o 19 40 46 65 70 85
Height(cm)

Figure 3.10. Sludge Distribution in Reactors with and Without Bionest Installation.

It can be seen that the bionest installation provides a great advantages for biomass growth

and even distribution, good sludge quality and easy maintenance. The sludge quality is

comparable with the USAB granule which usually has 45-57 gil (Bastone and Keller,

2000). Also, the even distribution of the sludge in the reactor does contribute more space

for mixing and reaction. Compared with the batch reactor (without bionest), the total
51
sludge inside the bionest reactor is 5.7 times more than batch reactor without bionest

installation does. The bionest reactor only has a 7% dead zone compared with DASB has

38-42% of dead zone (Lens et aI., 1998).

Instead of granule, flocs biomass was presented in the bionest reactor. Since the flocs

have poorer settling property than the granule, appropriate operation is very important to

maintain high concentration of biomass and even distribution of sludge. Therefore, in

addition to the HRT, SRT, biomass concentration (or aggregation), from the view of

operation, other operation criteria may need to be also considered. These criteria are

listed as follow:

Effective Mixing Time (EMT): Biomass effectively mixed with substrate including

soluble and suspended forms based on the hydraulic retention time.

Effective Biomass Reaction Process (EBRP): the reaction process and reactor

performance highly and directly depend on the effective biological reaction process

rather than SRT itself.

The mixing condition is a more critical control criteria than SRT. Vieira and Garcia

(1992) indicated that biomass gravity settling velocity and upflow velocity were more

appropriate than HRT. Figure 11 illustrates different cases of mixing condition related to

sludge gravity settling and upflow velocity.

The effective mixing time prevails that no matter what kind of growth rate and sludge

type, the reaction is determined by EMT rather than HRT or SRT. The high rate reactor

needs to provide sufficient EMT and biomass substrate efficiency mixing. In this study,

sludge gravity was about 1.7~2.2m1hour in steady water, in upflow velocity ofO.6m1hour

and gravity settling velocity decreased to O.9~ 1.1 mlhour because of the gas content. So

52
the intermittent feeding operation with 1 hour on and 1 hour off was suitable for

preventing biomass washing out.

AAI\AA
..

Casel (batch reactor completely mixed Case 2

when feeding) In flowing water, the gravity
In steady water, the gravity settling settling velocity and settling time:
velocity and settling time:
t -lL
11 = ..!!.... 2 - v2
VI
h is the effective height,
V J is the gravity settling velocity.

53
 ~~:!!:
_+ ~ 1
·, , ..

I
. . .
... ; !
. h
z
j ~ ~ 1 ]
: : : : :

Case 5 VASB
Case 4 Bionest
( z =..l!... while
t 2 -.l!-
Case 3
v, - V2 ' I, -+0.
If t, <0, biomass will be tHRT~(z ~HRT Ift2 ~O, which means that the solids can not be
washed out. (z =..l!...
v, Safety control and the operation moved up a little height, the EMF == !p- ;hz,
process optimization. V, is sludge
That meats V2 has sludge height; v, influent liquid velocity.
settling velocity in the up-flowing
reactor.
This case explains why VASB granular layer
determines the performance as wen as
bottleneck limitation of increasing whole
performance.

Figure 3. 11. Various Cases ofMixing Related to Sludge Gravity Settling and Upflow
Velocity.
Thus, the sludge distribution comparison between pre-feeding and post-feeding was

compared for RI by the feeding operation of Ihour on/off. As shown in Figure 12, the

sludge distribution based on total biomass estimation are not much different between pre-

feeding and post-feeding. Apparently, intermittent feeding operation can periodically

maintain the sludge moving up and down. As shown in the Figure 12, the total sludge

mass were only 2~3.2% difference between the start feeding point and the end of the

feeding. The two hours feeding makes a 5% difference between the post-feeding and pre-

feeding for sludge washing out.

54
 45 j
I

40 ___ R1 pre-feeding
- & - R1 post feeding

35

~
5 30
~
5jl: 25

8
III
til 20
11
iii
15

10

o +---~-,-_ .._._~--~---_._~-_.~~--~---~~.-._-~-_._~~~_._,
0 10 20 30 40 50 60 70 80 90
Reactor Height(cm)

Figure 3.12. Comparison ofSludge Distribution with 1 Hour On/OffFeeding Operation.

Accordingly, because of the sludge movement in the bionest unit during the feeding

period, flow shortcut will not occurred with intermittent operation. The formation of

shortcut and channeling using upflow anaerobic filter was reported by Cordoba et al.

(1994) and Nadias et al. (2001).

Comparison
Hawkes et. al.(1995) had compared different performances of different types ofreactor

for the anaerobic treatment of ice cream wastewater. They concluded that anaerobic filter

for the treatment of ice cream wastewater performed the best, i e. high organic loading

rate without other operational problems involved compared to anaerobic contact,

fluidized bed and VASB as shown in Table 6. Since the present anaerobic bionest reactor

was designed to combine the advantages of both anaerobic filter and VASB for the

55
treatment of dairy wastewater, their performance were compared and is presented in

Table 3.7.. Apparently, the "Bionest" reactor performs the best regarding high organic

loading rate, lower temperature (20 DC) and reasonable COD removal rate and methane

yield.

Table 3.6. Comparison of Various Types ofAnaerobic Reactors (Hawkes et. ai., 1995).
Reactor type Loading rate HRT(days) TCOD removal Methane Crucial
(kgCOD/m3/d) efficiency % content % problems
Anaerobic filter 6.38 0.93 66.9 75.3
Contact process 1.05 5.51 81.8 76.9 Poor
performance of
settling
Fluidized Bed 4.2 1.47 55.7 70.1 Difficulty of
GACbreakup
UASB 2.19 1.62 49.0 69.6 Granule never
form
Note: Reactors were operated at 35°C and using dairy ice cream wastewater as substrate.

Table 3.7. Comparison of Different Reactors Performance under Different Operation

Condition

Organic
loading TCOD
rate Removal
Reactor Wastewater Temperature (COD HRT efficiency
Types (COD gil) (oC) gil/d) (days) (%) Source Problem

UASB Dairy farm

hybrid 8.4 30 8 1 90.1 a
Granulation
UASB 0.4-20 35 2.19 1.62 49 b failed

Dairy Granule
UASB 5.0-7.0 35 3.5-5 2 90 c support

Dairy
UASB 50-60 30 3.84 83 d Digester

Anaerobic
Filter Dairy 4.9 35 6.38 0.93 66.9 b

Bionest Dairy Present

reactor 5.0-13.5 20±2 15 0.63 70-75 study

a, Cordoba et at, (1995); b Hawkes Freda (1995) R., Donnelly T and Anderson G. K. (1995); c 1nce

(1998); d. Vartak et at., (1997).

56
Conclusion
In general, the unique design of "Bionest" anaerobic reactor provides an effective tool for

the biological treatment of high lipid/fat contact of milk parlor wastewater at lower

temperature of 22°C. The advantages of operating the "Bionest" anaerobic reactor

include high SRT, i.e., high biomass content, best application ofEMT and EBRP, less

dead zone, providing stable operational condition regarding for COD removal and

methane gas production, combining the advantage ofUASB and anaerobic filter design

/operation and make-up the disadvantage ofUASB and anaerobic filter process,

providing the absorption of shock loading and a stable operational condition. The current

result was based on a six months of operational investigation. It is apparent that a long

term operational investigation is required in order to assure the operational stability of the

reactor and to obtain necessary operation/maintenance requirement for operation this type

of reactor.

Reference:

Alphenaar Arjen RinzemaAme and Lettinga G., (1993) Anaerobic digestion of long-

chain fatty acids in UASB and expanded granular sludge bed reactors, Process

Biochemistry, 28(8), 527-537.

APHA, AWWA and WEF. (1989) Standard Methods for the Examination of Water and

Wastewater, 17th ed. American Public Health Association, Washington, DC.

57
Batstone D. J. and Keller J. (2001) Variation of bulk properties of anaerobic granules

with wastewater type, Water Research, 35(7), 1723-1729.

Bellouti M., Alves M. M., Novais J. M. and Mota M. (1997) Floes vs granules:

differentiation by fractafdimension, Water Research, 31(5), 1227-1231.

Chernicharo Carlos Augusto L. and Cardoso MarcHio dos Reis (1999) Development and

Evaluation of a Partitioned Upflow Anaerobic Sludge Blanket (UASB) Reactor for

the Treatment of Domestic Sewage from Small Villages Wat. Sci. and Tech., 40(8),

107-113.

Cordoba Pedro R., Francese Alejandro P. and Sineriz Faustino (1995) Improved

Performance of a Hybrid Design Over an Anaerobic Filter for the Treatment of

Dairy Industry Wastewater at Laboratory Scale, Journal ofFermentation and

Bioengineering, 79( 3), 270-272

Cordoba, Pedro R., Francese Alejandro P. and Sineriz Faustino (1995) Improved

Performance of a Hybrid Design Over an Anaerobic Filter for the Treatment of

Dairy Industry Wastewater at Laboratory Scale, Journal ofFermentation and

Bioengeering 79(3), 270-272.

Driessen, W. and Yspeert, P. (1999) Anaerobic Treatment of Low, Medium and High

Strength Effluent in the Agro-industry Wat.Sci.and Tech.,40(8), 221-228.

Fernandez, Jose M., Francisco Omil, Ramon Mendez and Juan M. Lema (2001)

Anaerobic treatment of fibreboard manufacturing wastewaters in a pilot scale

hybrid usbfreactor, Wat..Res., 35(17), 4150-4158.

Frankin, RJ., (2001) Full-scale Experiences with Anaerobic Treatment ofIndustrial

Wastewater Wat. Sci. Tech. 44(8), 1-6.

58
Gavala, H. N., Kopsinis, H., Skiadas, 1. V., Stamatelatou, K. and Lyberatos, G. (1999)

Treatment of Dairy Wastewater Using an Upflow Anaerobic Sludge Blanket

Reactor, Journal of Agricultural Engineering Research 73(1),59-63.

Gavrilescu, M., 2002 Engineering concerns and new developments in anaerobic waste-

water treatment, Clean Techn Environ Policy 3,346-362.

Habets L. H. A, Engelaar A J. H. H. and Groeneveld N. Anaerobic treatment ofinuline

effluent in an internal circulation reactor Water Science and Technology, 35 (10),

189-197.

Hawkes, Freda R., Donnelly T. and Anderson, G. K. (1995) Comparative performance of

anaerobic digesters operating on ice-cream wastewater, Wat. Res. 29 (2),525-

533.

Hills DJ. and Nakano K. ,(1984), Effect ofpartic1e size on anaerobic digestion of

tomato solid waste. Agric. Wastes 10,285-295.

Ince O. (1998) Performance of a two-phase anaerobic digestion system when treating

dairy wastewater, Wat. Res. 32(9),2707-2713
Jhung J.K. and Choi E.(1995) A comparative study ofUASB and anaerobic fixed film

reactors with development of sludge granulation, Wat. Res., 29(1), 271-277.

Lens, P.N. L., Bosch, M. C. van den, Pol L. W. Hulshoff and Lettinga, G. (1998) Effect

of staging on volatile fatty acid degradation in a sulfidogenic granular sludge

reactor, Water Research, Volume 32, Issue 4, April 1998, Pages 1178-1192.

Lettinga Gatze, Rebac Salih and Zeeman Grietje (2001) Challenge of psychrophilic

anaerobic wastewater treatment, Trends in Biotechnology, 19( 9), 363-370.

59
Lettinga, G, van Velsen. A.F. M., Hobma, S.W., de Zeeuw, W. and Klapwijk, A.(1980)

Use of the Upflow Sludge Blanket (USB) Reactor concept for Biological

Wastewater Treatment. Biotechnology and Bioengineering 22, 699-734.

Lettinga, G., Field, J., van, Lier J., Zeeman ,G. and Hulshoff, Pol L. W. (1997)

Advanced anaerobic wastewater treatment in the near future. Wat. Sci. and Tech.,

35(10),5-12.

Malina Joseph F., Pohland Jr. Frederick G. ( 1992), Design of anaerobic process for the

treatment of industry and municipal wastes, 85.

Marin, P., Alkalay, D., Guerrero, L., Chamy, R. and Schiappacasse, M.e. (1999) Design

and Startup of an Anaerobic Fluidized Bed Reactor Wat. Sci. Tech. 40(8), 63-70

McGraw-Hill, New York.

Metcalf & Eddy, Inc. (1991 ) Wastewater Engineering: Treatment, Disposal, and Reuse,

3td ed,

Miron Yehuda, Zeeman Grietje, van Lier Jules B., and Lettinga, Gatze, (2000)The role

of sludge residence time in the hydrolysis of lipids, carbohydrates and proteins

during the anaerobic treatment of domestic sewage. Water Res. 34 (5), 1705-

1713.

Nadais, H. , Capela, I., Arroja, L. and Duarte, A. (2001) Effects of organic, hydraulic and

fat shocks on the performance ofUASB reactors with intermittent operation

Water Sci. Technol 44 ( 4) 49-56.

Pereboom 1. H. F. and Vereijken T. L. F. M. Methanogenic granule development in full

scale internal circulation reactors Water Science and Technology, 30 (8), 9-21

60
Pereboom 1.H.F. (1994)Size distribution model for methanogenic granules from full scale

UASB and IC reactors Water Science and Technology, 30(12),211-221.

Petruy R. and Lettinga G. (1997) Digestion of a milk-fat emulsion, Bioresource

Technology, 61(2), 141-149.

Uyanik S., Sallis P. J. and Anderson G. K. (2002)The effect of polymer addition on

granulation in an anaerobic baffled reactor (ABR). Part I: process performance,

Wat. Res., 36(4), 933-943.

van Langerak E. P. A., G. Gonzalez-Gil, A. van Aelst, 1. B. van Lier, H. V. M. Hamelers

and G. Lettinga (1998). Effects of high calcium concentrations on the

development ofmethanogenic sludge in upflow anaerobic sludge bed (UASB)

reactors. Wat. Res., 32(4), 1255-1263.

Vartak D. R., Engler C. R, McFarland M. 1. and Ricke S. C.(l997) Attached-film media

performance in psychrophilic anaerobic treatment of dairy cattle wastewater,

Bioresource Technology, 62( 3), 79-84.

Vidal ,G., Carvalho A., Mendez R. and Lema 1. M. (2000) Influence of the content in fats

and proteins on the anaerobic biodegradability of dairy wastewaters Bioresource

Technology 74(3), 231-239.

Vieira, S.M.M. and Carcia, A.D. Jr (1992) Sewage treatment by UASB reactor:

Operation and recommendations for design and utilization. Water Sci. Technol

25(7) 143-158.

Wiegant W.M. (2001) Experiences and potential of anaerobic wastewater treatment in

tropical regions. Water Sci. Technol, 44 (8), 107-113.

61
Wirtz Randall A. and Dague Richard R. Laboratory studies on enhancement of

granulation in the anaerobic sequencing batch reactor. Water Science and

Technology, 36(4), 279-286.

Yang, Y.P. (1994). Treatment of milkpowder wastewater with lipids. Internal Report.

Agricultural University ofWageningen, Wageningen, the Netherlands.

Yu H. Q., Tay J. H. and Fang Herbert H. P.(2001) The roles of calcium in sludge

granulation during uasb reactor start-up, Wat. Res., 35(4), 1052-1060.

Yu, H. Q., Fang, H. H. P. and Tay, J. H. (2001)Enhanced sludge granulation in upflow

anaerobic sludge blanket (UASB) reactors by aluminum chloride. Chemosphere,

44(1), 31-36.

Zeeman G. and G. Lettinga ,(1999) The role of anaerobic digestion of domestic sewage

in closing the water and nutrient cycle at community level. Water Sci. Technol.

39 (5), pp. 187-194.

Zeeman Grietje, Sanders Wendy T. M., Wang Kaijun Y. and Lettinga Gatze (1997)

Anaerobic treatment of complex wastewater and waste activated sludge -

application of an upflow anaerobic solid removal (UASR) reactor for the removal
...
and pre-hydrolysis of suspended COD. Wat.Sci.and Tech., 35(10), 121-1128.

62