UNIVERSITY OF HAWAI'I LIBRARY EVALUATION OF PARAMETERS INFUENCING OXYGEN TRANSFER EFFICIENCY IN A MEMBRANE BIOREACTOR A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HA WAI'I IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF. MASTER OF SCIENCE IN CIVIL & ENVIRONMENTAL ENGINEERING DECEMBER 2006 By JingHu Thesis Committee: Roger W. Babcock, Chairperson Albert S. Kim Chittaranjan Ray
112
Embed
EVALUATION OF PARAMETERS INFUENCING OXYGEN … › bitstream... · university of hawai'i library evaluation of parameters infuencing oxygen transfer efficiency in a membrane bioreactor
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UNIVERSITY OF HAWAI'I LIBRARY
EVALUATION OF PARAMETERS INFUENCING OXYGEN TRANSFER EFFICIENCY IN A MEMBRANE BIOREACTOR
A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HA WAI'I IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF.
MASTER OF SCIENCE
IN
CIVIL & ENVIRONMENTAL ENGINEERING
DECEMBER 2006
By JingHu
Thesis Committee:
Roger W. Babcock, Chairperson Albert S. Kim
Chittaranjan Ray
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 Masters of Science in Civil & Environmental
Engineering.
THESIS COMMIITEE
ii
ABSTRACT
Design of fine-bubble aeration systems for membrane bioreactors (MBRs) is
challenging due to high mixed liquor suspended solids (MLSS) concentrations that cause
changes in alpha value, which is the ratio of mass transfer rate under process conditions
to that under clean water conditions.
This study describes the results of pilot-scale fine-pore aeration testing to determine
a-values and influencing factors for MBRs. Clean water and process water aeration tests
were performed at the Honouliuli WWTP during the period of December 2005 and
October 2006. Three different 9-inch diameter fine-pore diffusers were tested.
Comprehensive analyses of the sludge properties were conducted.
Through this study, correlations were found to exist between a-value and oxygen
uptake rate (OUR), particle size distribution (PSD), MLSS and viscosity of activated
sludge, thereby providing better understanding and design guidance for MBR aeration
systems.
iii
TABLE OF CONTENTS
ABSTRACT .................................................................................................................... iii
LIST OF TABLES .......................................................................................................... vi
LIST OF FIGURES ...................................................................................................... viii
LIST OF ABBREVIATIONS & SYMBOLS .................................................................. x
CHAPTER I: INTRODUCTION .................................................................................... I
I. Comparisons of the Methods for OTE Measurement ...................................... 7
2. Factors Affecting Oxygen Transfer in Aeration Systems .............................. 10
3. Review of the Effect ofMLSS on Oxygen Transfer for MBRs .................... 13
4. Enviroquip 5MBR Pilot System Summary .................................................... 27
5. Chemicals Used in the Experiments .............................................................. 34
6. Comparison of a-value of 3 Diffusers ............................................................ 43
7. Clean Water Test (I) Raw Data ..................................................................... 63
8. In Situ OUR Test (1) Raw Data ..................................................................... 66
9. In Situ OUR Test (2) Raw Data ..................................................................... 67
10. In Situ OUR Test (3) Raw Data .................................................................... 68
II. In Situ OUR Test (4) Raw Data ..................................................................... 69
12. In Situ OUR Test (5) Raw Data ..................................................................... 70
13. In Situ OUR Test (6) Raw Data .................................................................... 71
14. In Situ OUR Test (7) Raw Data ..................................................................... 72
15. In Situ OUR Test (8) Raw Data ..................................................................... 73
16. Clean Water Test Results Summary ............................................................. 74
17. Off-gas Analysis (1) Data Summary ............................................................. 75
18. Off-gas Analysis (2) Data Summary ............................................................. 76
19. Off-gas Analysis (3) Data Summary ............................................................. 78
20. Off-gas Analysis (4) Data Summary ............................................................. 80
vi
21. Off-gas Analysis (5) Data Summary ............................................................. 82
22. Off-gas Analysis (6) Data Summary ............................................................. 84
23. Off-gas Analysis (7) Data Summary ............................................................. 86
24. Off-gas Analysis (8) Data Summary ............................................................. 87
25. Sludge Properties at Various MLSS Concentrations .................................... 89
vii
LIST OF FIGURES
Figure ~
1. Effect of SCOD Values on OTEzo ..................................................................... 10
2. Alpha-value as a function ofML VSS .............................................................. 11
3. Alpha-MLSS relationships for fine-bubble systems ........................................ 14
4. Specific oxygen transfer efficiency as a function ofMLSS ............................ IS
5. Alpha-Viscosity relationships for fine-bubble systems ................................... 16
6. Schematic of the Analyzer Structure ................................................................ 26
7. Anatomy of the Membrane Cartridge .............................................................. 28
8. Cutaway Illustration of Membrane Unit .......................................................... 28
9. Process Flow Diagram of the Enviroquip MBR .............................................. 29
10. Aeration Column Setup for Clean Water Testing ............................................ 36
11. Aeration Column Setup for Process Water Testing ......................................... 38
12. Average mass transfer coefficient for three fine pore diffusers ....................... 42
13. Average SOTE for three fine pore diffusers .................................................... 43
14. Alpha-value of the membrane diffuser at varied AFR ..................................... 45
15. Alpha-value of the ceramic diffuser at varied AFR ......................................... 44
16. Comparison of the a-value under different process conditions via MLSS ........................................................................... 46
17. Comparison of the a-value under different process conditions via MLVSS ........................................................................ 46
18. Correlation ofMLSS and a-value .................................................................... 47
19. Correlation of ML VSS and a-value ................................................................. 48
20. Correlation ofMLSS and viscosity .................................................................. 48
viii
21. Comparison of the a-value under different dit· .. .ty 49 process con IOns Via VlSCOSI ...................................................................... .
22. Correlation of viscosity and a-value ................................................................ 49
23. Correlation of OUR and MLSS ....................................................................... 50
24. Comparison of the a-value under different process conditions via OUR .............................................................. 51
25. Correlation of OUR and a-value ...................................................................... 51
26. Particle size distributions at various MLSS conc .............................................. 53
27. Correlation of particle size and MLSS ............................................................. 54
28. Comparison of the a-value under different process conditions via particle size ................................................... 55
29. Correlation of particle size and a-value ........................................................... 55
30. Relationship between SCOD and MLSS ......................................................... 56
31. Relationship between TDS and MLSS ............................................................ 57
32. Relationship between total SMP and MLSS .................................................... 57
33. Relationship between total EPS and MLSS ..................................................... 57
34. Correlation ofMLSS or MLVSS and VOTE ................................................... 58
35. Correlation of viscosity and VOTE ................................................................. 58
36. Correlation of OUR and VOTE ....................................................................... 59
37. Correlation of particle size and VOTE ............................................................. 59
ix
LIST OF ABBREVIATIONS & SYMBOLS
The following abbreviations and symbols are used in this paper:
ABS Acrylonitrile-Butadiene-Styrene
AECOR Aemtion Engineering Resources Corp.
AFR Air flow mte
ASCE American Society of Civil Engineers
ASP Activated sludge process
BOD Biological oxygen demand, mgIL
CASP Conventional activated sludge process
CST Capillary suction time
CER Cation exchange resin
DO Dissolved oxygen, mgIL
EMBR External membrane bioreactor
EPDM Ethylene Propylene Diene Monomer
EPS Extracellular polymeric substances, mgIL
HOPE High density polyethylene
HRT Hydmulic retention time, hr
MBR Membrane bioreactor
MCRT Mean cell residence time, hr
MOD Million gallons per day
MLSS Mixed liquor suspended solids, mgIL
"
MLVSS
OTE
OTR
OUR
PSD
PTFE
SCFM
SCOD
SMP
5MBR
SOTE
TDS
TSS
VOTE
VSS
a
C
Co
C' w
Mixed liquor volatile suspended solids, mgIL
Oxygen transfer efficiency in percent
Oxygen transfer rate, mg/L
Oxygen uptake rate, mg I L . hr
Particle size distribution
Polytetrafluoroethylene
Standard cubic foot per minute
Soluble chemical oxygen demand, mgIL
Soluble microbial products, mgIL
Submerged membrane bioreactor
Standard oxygen transfer efficiency in percent
Total dissolved solids, mg/L
Total suspended solids, mg/L
Volumetric oxygen transfer efficiency in percent
Volatile suspended solids, mg/L
Diffuser specific area, m 2 •
DO concentration, mg/L
DO concentration at time zero, mg/L
Equilibrium DO concentration at tested conditions, mg/L
Equilibrium DO concentration at 20'C, 1 atm and zero salinity. mg/L
Equilibrium DO concentration at the test temperature T, 1 atm and zero salinity. mg/L
xi
C· ST
D
MRog/i
ND
t
T
P,
v
Vo
Wo.
Tabular value of DO saturation concentration at 20°C, I atm and 100% relative humidity, mg/L
Tabular value of DO saturation concentration at the test temperature T, 1 atm and 100% relative humidity, mg/L
Particle diameter, J.Ull
Mass rate ofinerts, kg/s
Volumetric mass transfer coefficient, S-I
Volumetric mass transfer coefficient at 20·C, S-I
Molecular weights of oxygen
Molecular weights of inerts
Mole ratio of oxygen to inerts in the inlet stream
Mole ratio of oxygen to inerts in the off-gas stream
Total diffuser number
Time, s
Temperature, ·C
Barometric pressure during the test, psi
Standard atmospheric pressure 14.7 psi at 100% relative humidity
Total volumetric gas flow rates of inlet gas, m3 /s
Total volumetric gas flow rates of outlet gas, m3 /s
Liquid volume of water in the test tank, m3
Gas hold-up volume, m3
Mass flow of oxygen in air stream, kg/ s
Mole fractions of oxygen in the inlet gas
xii
Yco,(R)
z
a
aSOTE
TJr,40
p
T
n
Mole fractions of oxygen in the exit gas
Mole fractions of CO2 in the reference gas(R)
Mole fractions of CO2 in the off-gas (og)
Mole fractions of water vapor in the reference gas (R)
Mole fractions of water vapor in the off-gas (og)
Diffuser submergence, m
The ratio of the value of KLa measured in process water to the KLa measured in clean water
Oxygen transfer efficiency corrected for all process conditions such as DO, salinity, temperature and barometric pressure, except for a factor
Correction factor for salinity
Shear strain rate imposed on the sample, S-I
Shear stress, mPa
Yield stress, mPa
Absolute viscosity, mPa· s or cP
Absolute viscosity at a shear rate of 40 S·I, mPa· s
Empirical temperature correction factor
Absolute viscosity, mPa· s or cP
Density of oxygen at temperature and pressure of gas flow, kg/ m3
Temperature correction factor
Barometric pressure correction factor
xiii
1.1 Background
CHAPTERl INTRODUCTION
Statewide in Hawaii, approximately 16% (23 MOD) of municipal wastewater is
currently recycled. With the potential for compact, decentralized water reclamation
installations combined with very high quality permeate water, membrane bioreactors
(MBRs) promise to be an effective method for enhancing water recycling if they can be
shown to be reliable and cost effective (Babcock et al., 2003).
MBR systems can sustain higher biomass concentrations by replacing the secondary
clarifier with membrane filtration and allowing smaller aeration basins to be used. In an
MBR, the membranes create a physical barrier of solids and therefore the process is not
subject to gravity settling solids limitations. However, they are limited instead by fluid
dynamics of high solids mixed liquor, and by the effect of high solids on oxygen transfer.
Oxygen transfer is a major factor influencing the efficient and economic operation of
all aerobic bioprocesses, including MBRs. The overall volumetric mass transfer
coefficient, K La, is a parameter to characterize the rate of oxygen transfer in aeration
processes; where K L represents the mass transfer coefficient based on the liquid film
resistance and a, the interfacial area.
The value of alpha (11) is another parameter commonly used to describe the oxygen
transfer in biological aerated systems, which is the ratio of mass transfer rate under
process conditions (mixed liquor) to that under clean water conditions. This correction
factor quantifies the influence of mixed-liquor constituents on aeration capacity. It is an
important operating and design variable for aeration systems in activated sludge process
(ASP).
MBR systems utilize aeration in two forms. First, they utilize coarse-bubble (cross
flow) aeration for membrane scour to control permeability and fouling. Second, they use
fine-pore aeration for mass transfer of oxygen to meet biological requirements as well as
for mixing in the aeration tank. According to the report prepared for the Water
Environment Research Foundation (WERF, 2004), energy costs associated with aeration
in submerged MBR (SMBR) systems represent more than 90 percent of the total energy
cost. Therefore, potential oxygen limitations appear to be a serious cost issue for 5MBR
systems, and acquiring accurate oxygen transfer information on these systems is of great
necessity. Substantial savings in overall energy costs of WWTPs with MBRs can be
realized by using this information to improve the energy efficiency of the aeration
system. In addition, if designers understand the oxygen transfer in mixed liquor of an
aeration tank and factors that affect it, they can provide more optimized designs to treat
the wastewater to the required effluent quality.
The aeration, the oxygen transfer and the biomass characteristics interrelate with and
impact on each other. When studying aeration operations in MBRs, we need to consider
both the effects of biomass characteristics on aeration efficiency, represented by the
oxygen transfer parameters (Le., KLa and a-value), and the effects of aeration (intensity
and type of diffusers) on biomass characteristics (Germain and Stephenson, 2005).
Fine-pore aeration system design requires the determination of an appropriate a
value. However, this design in MBRs is challenging at high MLSS due to changes in a
values. Currently there is a lack of good data on a-values at the high MLSS found in
2
MBRs. If the wrong value of alpha is used, the aeration system can be either over
designed or under-designed. Previous studies have obtained correlations of a-values at
high solids with MLSS and with viscosity. However, the existing data have considerable
variability in the solids range of interest in current MBR designs (Babcock, 2006).
This context highlights the need for further research on accurate detennination of
oxygen transfer efficiencies and a-values in MBR process, as well as exploration of
relationships between a-values and more parameters describing mixed liquor properties
besides MLSS and viscosity. Better correlations may be obtained with these parameters.
1.2 Characteristics of Activated Sludge in MBRs
Activated sludge is a complex and variable heterogeneous suspension containing both
feed water components and metabolites produced during the biological reactions as well
as the biomass itself (Chang et al., 2002). It is a mixture of particles, microorganisms,
colloids, organic polymers and cations, which all have different shapes, sizes and
densities.
MBRs and conventional activated sludge process (CASP) have many similarities,
particularly in tenns of microbial metabolism and kinetics. However, substituting
membrane separation for gravity sedimentation allows much higher MLSS
concentrations (8,000 to 20,000 mgIL) in MBRs with resulting high metabolic rates.
Mainly due to the high MLSS, the sludge characteristic differs from conventional
activated sludge.
The activated sludge suspension is a non-Newtonian liquid with pseudoplastic
properties (Dick and Ewing, 1967). For Newtonian fluids, shear stress is proportional to
shear rate, with the proportionality constant being the viscosity, while the viscosity of
3
non-Newtonian fluids changes as the shear rate is varied. A pseudoplastic liquid is a non
Newtonian fluid whose viscosity decreases as the applied shear rate increases. This type
of behavior is also called shear-thinning.
High biomass concentrations give rise to non-Newtonian behavior with high apparent
viscosities. This can largely be attributed to the fact that cross-linked filamentous
organisms and flocs are present in the sludge. These high apparent viscosities affect the
energy required for pumping and oxygen supply of the microorganisms. They may
impede oxygen transfer and the degree of mixing by influencing bubble coalescence
(Germain and Stephenson, 2005; WERF, 2004).
High biomass also forms a high production of soluble microbial products (SMP) and
extracellular polymeric substances (EPS) in the bioreactor, which are both important
components in describing biomass kinetics. EPS are a complex mixture of proteins,
carbohydrates, acid polysaccharides, DNA, lipids, and humic substances that surround
cells and form the matrix of microbial flocs and films (Liao et al., 2004). They are an
essential part of activated sludge. The importance of EPS in controlling membrane
fouling has been studied extensively while there is little information on the role of EPS in
affecting oxygen transfer. SMP are the pool of organic compounds that are released into
solution from substrate metabolism (usually with biomass growth) and biomass decay
(Barker and Stuckey, 1999). Like EPS, SMP are primarily composed of proteins, humic
compounds and polysaccharides. In order to be able to reach the active sites of the
bacterial cell membrane, the oxygen contained in the air bubbles needs to penetrate the
liquid film surrounding the floes (SMP) and then diffuse through the floc matrix (EPS)
4
(Gennain and Stephenson, 2005). Therefore both compounds are likely to affect the
oxygen transfer.
Particle size is another parameter characterizing the biomass. Compared to
conventional activated sludge, the average diameter of a particle in a MBR is
considerably smaller, because bacteria are not selected for their ability to aggregate to
large, settleable floes. Moreover, the high shear forces introduced, particularly by
pumping during cross-flow filtration, can break up floes. In conventional activated
sludge, floes may reach several 100 J.lm in size (Wisniewski et aI., 1999). Hydrodynamic
stress in MBRs reduces floc size to approximately 30--60 J.lm in immersed systems
(Zhang et aI., 1997; Song et aI. 2003) and 3.5 J.lm in sidestream MBRs (Cicek et aI.,
1999).
Oxygen uptake rate (OUR) is a measure of the rate of oxygen utilization of
microorganisms in a body of liquid. It is a good indicator of metabolic activity of the
biological system and has been traditionally used in aerobic processes to estimate on-line
the biomass activity (Oca et aI., 2004). In MBRs, higher MLSS concentrations and
accordingly higher possible sludge age affects the bacterial numbers and the composition
of the microbial community, and thus changes OUR (Drews et aI., 2005; Li et aI. 2006).
OUR is a potential oxygen transfer parameter and could be used to control the aeration
rate or the sludge recycling rate in wastewater treatment.
Other biomass characteristics in MBRs may also have an impact on oxygen transfer
and accordingly a-values, such as the soluble COD fraction (SCOD) and the total
dissolved salt content (TDS). It's necessary to comprehensively investigate the influences
of these parameters on the diffusion of oxygen in MBR process.
5
CHAPTER 2 LITERATURE REVIEW
2.1 Methods/or OTE or OTR measurement
A variety of techniques exist for estimating oxygen transfer efficiency (OTE) or
oxygen transfer rate (OTR) of an aeration system, which can be generally divided into
four categories (Stenstrom, 1997; Stenstrom, 2005):
• Adsorption or re-aeration method
It adopts the same procedures as in clean water testing and converts the results to
field rates with conversion factors. This method determines KLa and the
equilibrium dissolved oxygen concentration C: for process water by first
removing DO from then re-oxygenating the water to near the saturation level.
• In-situ oxygen uptake rate (OUR) method
It's process water testing using methods to account for the biological consumption
of oxygen during the transfer test. It determines KLa for a mixed aeration tank
under process conditions by measuring the in-situ OUR under steady-state
conditions. The in-situ OUR is measured by monitoring the DO concentration
after stopping aeration.
• Material balance methods
They attempt to determine difference in inputs and outputs of oxygen consuming
material.
• Off-gas method
This method estimates the oxygen transfer capability by a gas phase mass balance
over the aerated volume. It requires the capture of a representative sample of the
6
gas, which exits the aeration basin surface, and analysis of this gas for its
composition.
The advantages and disadvantages of the above four methods presented in literature
studies (Capela et ai., 2004; Cornel P. et aI., 2003; and Krause S. et aI., 2003) are
summarized in Table 1.
Table 1. Comparisons of the Methods for OTE Measurement
Method Advantage Main Drawback
It is very difficult to accurately estimate the a-factor. Constant process conditions should be
Adsorption - maintained during the test duration. A minimum, incremental DO cone. of 2 mgIL should occur for implementing this method. It is difficult to accurately estimate oxygen consumption rate, especially in oxygen limiting conditions occurring in overloaded
OUR It's simple to perfonn. treatment plants. It requires that the aeration and mixing functions be dissociated to maintain the mixed liquor in suspension when aeration is stopped. It requires long-tenn knowledge of process operating conditions such as sludge wasting
Material rate. Balance - It's susceptible to error from sludge settling
in the aeration basin or stripping of volatile oxygen consuming compounds.
It is perfonned at real in-process It requires an accurate measurement of the conditions without requiring airflow rate and an estimate of the DO process modification or chemical saturation cone. in process water. addition to complete the test.
Off-gas The DO concentration or OUR does not interfere with the test procedure. It offers the advantage of differentiation in location and time.
7
Adopting the improved off-gas technique - mole fraction approach, the mass
transfer efficiency can be determined without measuring flow rates of gas entering and
exiting the fluid. This improvement overcomes the main drawback of traditional off-gas
method. By knowing the molar percents of the reacting or changing gas constituents
(oxygen, carbon dioxide, and water vapor), OTE can be calculated based on the gas phase
mass balance. In the meanwhile, there is no need to estimate the DO saturation
concentration in process water. Only equilibrium DO concentrations in clean water at
20°C and test temperature( C:20 and C:T ) need to be determined. Conclusively, the off
gas method is preferable to the other three methods for determining OTE in mixed liquor,
since it can be applied on a continuous mode at any particular location in the aeration
tank and under a wide range of process conditions.
Capela et aI. (2004) systematically compared the four measurement techniques under
process conditions in conventional WWTPs: the off-gas, hydrogen peroxide (HzOz), re
aeration, and in situ OUR methods. The HzOz method is based on the same principle as
the re-aeration method. It determines OTE by monitoring the DO concentration over the
time after adding HzOz. With the comparisons of pilot-scale and full-scale results of
oxygen-transfer coefficients obtained by different methods, the off-gas technique was
recommended for the analysis of diffused systems when applicable.
Krause et aI. (2003) conducted a comparison of the adsorption method with the off
gas method in full scale MBRs. Nevertheless with both methods comparable OTRs were
obtained, the varying results achieved by application of off-gas analysis revealed the big
advantage ofthls method, being able to record exactly the time-variation of the OTR. It's
8
also pointed out that the off-gas method only can be applied at DO concentration below
50% of the saturation concentration and at low air flow rates.
Currently off-gas method is widely accepted and has been the preferred method for
measuring OTE in operating aeration basins, because of its combination of reliability and
convenience. (Groves et al. 1992; lranpour et aI., 2000; Jranpour et aI., 2002; Mueller et
aI., 2000; and Stenstrom, 1989). The modern off-gas test was developed by Redmon et aI.
(1983) in conjunction with the U.S. EPA-sponsored ASCE Oxygen Transfer Standard
Committee. The technique measures actual OTE at process conditions assuming that inert
gases (nitrogen, argon) are conserved and can be used as a tracer. The fundamentals of
this method are described in detail in Chapter 4.
Brochrup (1983) and Babcock et aI. (1999) evaluated the precision and accuracy of
off-gas analysis. Brochrup estimated a coefficient of variation of 6% or less. Babcock et
aI. estimated an error of less than 10% for fine pore aeration systems and pointed out
these errors can be easily avoided by diligent maintenance of test equipment and by
maintaining good quality control practices. Therefore, if conducted carefully, with
maintained equipment, off -gas testing is a very accurate method for determining OTE
under process conditions.
2.2 Co"elation of OTE or a-value with Influencing Factors in Aeration
Due to complex mechanisms underlying oxygen transfer in aeration systems, OTE
and a-value could be affected by three groups of factors including equipment factors,
operation factors and wastewater conditions as listed in Table 2. The information was
obtained from U.s. EPA (1989) EPA/ASCE Design Manual on Fine Pore Aeration.
9
Table 2. Factors Affecting Oxygen Transfer in Aeration Systems
Equipment Factors Diffuser type Diffuser density Diffuser submergence Diffuser layout Diffuser age Flow regime Basin geometry
Operation Factors Solids retention time/nitrification Food to microorganisms ratio Airflow rate per diffuser Mixed liquor DO Diffuser fouling
Wastewater Conditions Wastewater characteristics Mixed liquor temperature
Some correlations linking oxygen transfer and aerobic biological system
characteristics are found in CASP in the literature, depending on the parameters
considered.
Mueller et aI. (2000) found that solution-phase COD (SCOD) concentration is a
principal factor influencing OTE and alpha for the plug-flow systems. In this study, OTE
was measured with the membrane diffusers utilizing the off-gas technique. The difference
in a-values was correlated with the impact of SCOD on the individual station OTE2o
values (tank weighted avemge process OTE at zero DO and 20°C) as shown in Fig. 1. An
Single-tank 5MBR (full scale). Not specified 7-17 Fine-bubb[e Clean water (adsorption). 0.7-0.4 MLSS& Wagner et aI .• Dual-tank 5MBR mixed liquor (adsorption) Viscosity 2002 (full scale)
Single-tank 5MBR (full scale). Not specified 7-17 Fine-bubble Clean water (adsorption), 0.7-0.4 MLSS. Cornel et aI., Dual-tank 5MBR mixed liquor (off-gas. Viscosity, 2003 (full scale) adsorption) Air f10wrate &
Surfactant conc.
Pilot scale. Several activated Not specified 8-28 Fine-bubble, Clean water (adsorption). 0.5-0.1 MLSS, ML VSS, Krampe and sludge types lajector mixed liquor (adsorption) Viscosity. EPS. Krauth, 2003
CST". Polymer contents &
Surface tension
a EMBR = external MBR b 5MBR = submerged MBR • CST = capillary suction time
13
1.0 ~--------------------,
Gl = e,o.017l ·MISS (Gunder, 2001) 0.8+-~~~
Gl = e,O.0446 ·MLSS (Muller et al,1995) ., ::J Oi 0.6
(Wagner et aI, 2002; Cornel et ai, 2003) a = e,O.046 ·MLSS %
.<:
~0.4 Gl = e ,0.082 . MLSS
.lGunder an~ Krauth,I9921n _ ~~~:::::~~~~~] a = e,O.08788 ·MLSS
0.2
(Krampe and Krauth,2003) 0.0 +---~-~~c.:...:.~-----~::":"~~--~------l
o 5 10 15 20
MLSS concentration (gil)
25 30
Fig. 3. Alpha-MLSS concentration relationships for fine-bubble systems in MBRs
All of the evidence exists that MBRs, which operate at high MLSS concentrations,
have suppressed a-values, and that a-value is inversely proportional to MLSS. However,
the decrease rate of a with MLSS varies among studies.
Wagner et aI. (2002) and Cornel et aI. (2003) determined a-values for the fine-bubble
aeration systems in full-scale municipal MBRs. They indicate that a-value decreases
from 0.6 to 0.4 in the MLSS concentration range of 10 to 20 gIL. Daily a-value variations
in the range of ± 0.1 were detected, which were attributed to varying surfactant loading.
Additionally, the coarse bubble "cross-flow" aeration system was found indicating no
dependence of a-value and MLSS.
Cornel et aI. (2003) also obtained the relationship between MLSS and specific
oxygen transfer efficiency. Fig. 4 shows the specific oxygen transfer efficiency as a
function of MLSS, which means the oxygen transfer efficiency per unit depth of aeration
tank.
14
10
co 6 ui b o 4 !5 Q) Co en 2
------------ ----------------------~
TJ = 9.00-S.63x 10-4 MLSS+2.56xI0-s MLSS2
O+---,----,--~---,----r_--._--~--_r--~
o 2000 4000 6000 8000 10000 12000 14000 16000 18000
MLSS (mg/L)
Fig. 4. Specific oxygen transfer efficiency as a function of MLSS
Furthennore, some researchers evaluated the effect of mixed-liquor viscosity on ex-
value. GUnder (2001) and Krampe and Krauth (2003) formulated equations linking the a-
value to the viscosity at a shear rate of 40 S·1 in activated sludge with high MLSS
concentrations (Fig. 5). An increase in viscosity has been shown to have a negative
influence on the oxygen transfer. The same trend was also observed by Wagner et aI.
(2002) while the correlating equation wasn't given.
In these studies, a-value was correlated better with viscosity than with MLSS
concentration, which suggests that the effect of MLSS on a might be better explained in
terms of the influence of MLSS on viscosity. Explanations concerning this phenomenon
have been given in the literature (WERF, 2004). High viscosity may lower a by
increasing the rate of bubble coalescence, and thus reducing the interfacial area of oxygen
transfer. Additionally, the ability of bubbles to induce turbulence and mixing decreases
with viscosity.
15
1.2 -- - ------------------
1.0
~ 0.8
a = I1r.40 ·0456 (Krampe and Krauth, 2003)
:./ --- ------
~ IV 0.6 .£: Q.
;( 0.4 a= I1r.40 -OAS (Gunder, 2001)
I 0.2 ---- ---j 0.0
0 20 40 60 80 100 120
ll,40 [mPa . sl
Fig. 5. Alpha-Viscosity relationships for fine-bubble systems
So far some limited work has been done to observe the impact of other biomass
properties besides MLSS and viscosity on aeration efficiency in MBR process. Krampe
and Krause (2003) analyzed the sludge in regard to MLSS, viscosity, polymer contents,
EPS components and capillary suction time (CSn. However, only the solid contents and
the viscosity were found to be possible parameters to describe the relations. In this study,
all sludge types were gradually diluted in order to get a series of solids contents. This is
very different than growing or accumulating sludge in the aeration tank and obtaining the
desired solids contents in sequence gradually. The dilution method wouldn't be expected
to accurately capture sludge characteristics under different growth conditions.
In conclusion, although several practical experiences and data are available for MBR
aeration processes, no systematic and comprehensive investigation has been conducted so
far on other sludge properties such as OUR, PSD, TOS, SCOD, SMP and EPS in addition
to MLSS and viscosity. The influence of other sludge properties on oxygen transfer still
remains unclear. Therefore additional or further studies are needed to determine how
other variables affect the a-MLSS concentration relationship in MBRs.
16
CHAPTER 3 SCOPE AND OBJECTIVES OF WORK
The pilot-scale aeration study was conducted under existing operating conditions at
Honouliuli WWTP located at Ewa beach, Honolulu, Hawaii, where an investigation of
parallel pilot MBR systems is underway. A 20 ft tall pilot column had been constructed
for aeration testing and was filled with clean or process water. Clean water tests were
performed with 3 different 9-inch diameter fine-pore diffusers, ceramic, membrane, and
high density polyethylene (HDPE) types, to determine standard oxygen transfer
efficiency (SOTE) under specific airflow rates. This was followed by off-gas testing with
process water at varied MLSS concentrations (5, 7.5, 10, 12.5, IS, 17.5, and 20 gIL) to
determine OTE. The process water consisted of mixed liquor from a MBR pilot. An off-
gas analyzer was constructed to measure OTE under steady-state conditions. In addition,
the comprehensive analyses of the activated sludge were carried out to investigate the
relationship between the sludge properties and a-value.
The main goal of the study was to determine a-value and provide better
understanding for more efficient fine-bubble aeration system design of full-scale MBRs.
The specific objectives included:
1. To acquire good data on OTE and a-values at the high MLSS concentrations in a
MBRsystem.
2. To systematically examine the effects of aeration intensity, diffuser type, and
various sludge characteristics, such as TSS, VSS, viscosity, SMP, EPS, PDS,
OUR, SCOD and TDS, on oxygen transfer; identify potential factors affecting
OTE and a-value.
3. To obtain better correlations of these identified factors with a-value in MBRs.
17
CHAPTER 4 FUNDAMENTALS OF OXYGEN TRANSFER TESTS
4.1 Fundamentals of the non-steady state method
Interfacial oxygen transfer involves transport from the bulk gas phase to the interface,
and then from the interface into the liquid. For sparingly soluble gases, such as oxygen,
mass transfer on the gas side of the interface is much quicker; therefore, transfer on the
liquid side is expected to control oxygen transfer at the interface. Then oxygen transfer
can be described by the following two-resistance mass transfer model, which is most
commonly used to predict oxygen transfer in water (Aeration A Wastewater Treatment
Process, 1988; ASeE, 2000):
(4.1)
where:
e = DO concentration, mg I L
e: = DO saturation concentration, mg I L
KLa = apparent volumetric mass transfer coefficient, S-I
t = time, s
The method for determining KLa and e: in clean water is the unsteady adsorption
method, which involves first removing dissolved oxygen (DO) from the water volume by
the addition of a chemical reductant (normally sodium sulfite) and then re-oxygenating
the water to near the saturation level using the specific aeration device. The KLa and e: values are estimated by regression analysis of the measured DO data using the integrated
form of Equation (4.1), which is shown in Equation (4.2):
(4.2)
18
where:
Co = DO concentration at time zero, mg I L
The values of KLa and C: are dependent on water temperature and the barometric
pressure (under process conditions) and they are adjusted to standard conditions. The
standard oxygen transfer efficiency (SOTE) is obtained as follows:
(4.3)
where:
C:ao = equilibrium DO concentration at 20 °C, I atm and zero salinity, mg/L
KLaao = apparent volumetric mass transfer coefficient at 20 °C, 5-1
V = liquid volume of water in the test tan k, m3
W 02 = mass flow of oxygen in air stream, kg I s
The equilibrium DO concentration C:ao and apparent volumetric mass transfer
coefficient KLa aO are calculated as follows:
where:
• • ( I ) C"ao = CooT to
K a - K a·e(20-T) L 20 - L
(4.4)
(4.5)
C:T = equilibrium DO concentration at temperature T,I atm and zero salinity, mg/L
KL a = apparent volumetric mass transfer coefficient at the test temperature, S-1
t = temperature correction factor
o = barometric pressure correction factor
e = empirical temperature correction factor
19
T = temperature, °C
The pressure correction factor n accounts for the effect of non-standard barometric
pressures. It is calculated as follows for basins less than 6.1 m (20 ft) deep:
where:
n = Ph P,
Pb = barometric pressure during the test, psi
(4.6)
P, = standard atmosphere pressure 14.7 psi at I 00% relative humidity
The influence of temperature on the oxygen transfer coefficient and oxygen saturation
value can be expressed in terms of the factors e and 1", defined by:
e(T -20) = KL aT
K La 20
(4.7)
(4.8)
The influence of temperature on the various oxygen saturation concentrations will be
similar. Therefore, 1" can be calculated based on published DO surface saturation values:
where:
C' ST
(4.9)
= tabular value of DO saturation conc. at 20°C, I atm and 100% relative humidity, mg/L
= tabular value of DO saturation conc. at the test temperature, 1 atm and 100% relative humidity, mg/L
Values of e reported in the literature have ranged from 1.008 to 1.047 and are
influenced by geometry, turbulence level, and type of aeration device. The clean water
20
test standard recommends that the value of e be taken to1.024 unless experimental data
for the particular aeration system indicate conclusively that the value is significantly
different from 1.024 (Aeration A Wastewater Treatment Process, 1988).
4.2 Fundamentals of the off-gas method
4.2.1 Theory of analysis
The off-gas method is based on a gas-phase mass balance, which measures the change
in oxygen content of the air entering and exiting an aeration tank. By comparing the
composition of the off-gas to that of the gas entering an aeration tank, it is possible to
calculate the oxygen transfer occurring within the tank. If the flow rates of gas entering
an exiting the fluid are known, then the following mass balance can be made (Stenstrom
1997; Stenstrom 2005):
(4.10)
where:
p = density of oxygen at temperature and pressure of gas flow, kg! m 3
q;,qo = total volumetric gas flow rates of inlet and outlet gasses, m3 !s
YR, YOg = mole fractions (equivalent to volumetric fractions) of oxygen in
the inlet and exit gasses
KLa = volumetric oxygen transfer coefficient, 5-1
C: = equilibrium DO concentration in the test liquid at the given conditions, mgIL
C = oxygen concentration, mgIL
v = liquid volume, m3
21
Vo = gas hold-up volume, m3
At steady state the equation reduces to:
(4.11)
Since it is often difficult to measure the entering gas flow rate to an aeration system, a
procedure which does not rely on gas flow rates is needed. If one assumes that the inert
portions of the entering gas stream do not change, a mole fraction approach can be
developed which does not require gas flow rate. This assumption means that the
nitrogen, argon, and inert trace gases do not change as they pass through the aeration
system. The new technique (Redmon et aI., 1983) relies upon this assumption to
calculate oxygen transfer efficiency (aTE). It must be further assumed that the transfer at
the fluid surface and the atmosphere is negligible when compared to the transfer caused
by the aeration system, and that steady state conditions exist during the test. Both
assumptions are very good for the wastewater treatment systems.
aTE expressed as a fraction, can be derived as follows:
where:
OTE = mass O2 in - mass O2 out mass O2 in
MRo/i -MRog/i
MRo/i (4.12)
= mass rate of inerts, which is constant (by assumption) in both the inlet and off-gas streams, kg I s
= molecular weights of oxygen and inerts, respectively
22
MRo1;, MRogii = mole ratio of oxygen to inerts in the inlet and off-gas streams
The mole ratio of oxygen to inerts is calculated by subtracting the mole fractions of
oxygen, carbon dioxide and water vapor, as follows:
Testing Date: 8/17/06 MLSS = 11967 mgIL Initial T __ = 32.9 °C Final T ..... = 30.9 °C
Time Time DO TIme Time DO
(sec) (miD) (mg/L) (see) (miD) (mg/L)
0 0.00 6.13 230 3.S3 3.44
10 0.17 6.0S 240 4.00 3.30
20 0.33 6.02 250 4.17 3.16
30 0.50 5.97 260 4.33 3.02
40 0.67 5.S7 270 4.50 2.88
50 0.83 5.7S 280 4.67 2.74
60 1.00 5.67 290 4.S3 2.60
70 1.17 5.56 300 5.00 2.46
80 1.33 5.44 310 5.17 2.32
90 1.50 5.31 320 5.33 2.IS
100 1.67 5.1S 330 5.50 2.04
liO 1.83 5.06 340 5.67 1.90
120 2.00 4.93 350 5.S3 1.75
130 2.17 4.S1 360 6.00 1.62
140 2.33 4.67 370 6.17 1.47
ISO 2.50 4.54 380 6.33 1.33
160 2.67 4.39 390 6.50 1.19
170 2.83 427 400 6.67 1.05
ISO 3.00 4.12
190 3.17 3.9S
200 3.33 3.S5
210 3.50 3.71
220 3.67 3.57
71
Table 14. In Situ OUR Test (7) Raw Data
Testing Date: 8131106 MLSS = 14760 mgIL Initial T .... (test 7.1) =32.2 °C Final T __ (test 7.1) = 30.S °C Initial T ..... (test 7.2) = 33.S °C Fina\T __ (test7.2)= 32.loC
Test7.! Time Time DO Time Time DO Test 7·2 Time TIme DO Time Time DO
12 i2 27.148 43.906 I 2.927 Ceramic I 10.556 10.499 52.886 3.526 I I I I I Ceramic 1.5 10.551 20.419 67.035 4.469 I I I I I Ceramic 2 10.116 17.423 41.225 2.748 HOPE 1.5 10.536 14.738 47.954 3.197 HOPE 3 10.419 25.404 41.747 2.783
Note: Probe I is the upper DO probe in the aeration column and probe 2 is the lower one. The di1fuser submergence is 15 Il
74
Testing Date: May 26, 2006 Air Temp ("F): 92 Barom Pres (in Hg): 29.99
DiflUser Airflow Test Rate No. (SCFM)
1 Memb. 1
2 Memb. 1
3 Memb. 1
4 Memb. 2
5 Memb. 2
6 Memb. 2
7 Memb. 2
8 Ceramic 1
9 Ceramic I
10 Ceramic 1
11 Ceramic 1.5
12 Ceramic 1.5
13 Ceramic 1.5
14 Ceramic 2
15 Ceramic 2
16 Ceramic 2
17 HOPE 1.5
18 HOPE 1.5
19 HOPE 1.5
20 HOPE 1.5
Water 00 Temp. (mgIL) ("C)
30.0 1.11
30.1 636
30.1 3.76
30.0 5.06
29.9 7.04
28.4 5.84
TABLE 17. Off-gas Analysis (1) Data Summary
MLSS = 3082 mgIL Theta = 1.024
Ref-gas Off-gas M Ampere Ampere Fra<lion
(pA) (pA) Off-gas
557 527 0.1982
563 528 0.1%5
565 528 0.1958
567 539 0.1992
564 542 0.2013
557 533 0.2005
545 524 0.2014
525 516 0.2059
513 508 0.2075
51\ 508 0.2083
491 472 0.2014
490 471 0.2014
489 471 0.2018
480 466 0.2034
477 469 0.2060
478 470 0.2060
534 518 0.2032
538 516 0.2009
538 518 0.2017
539 517 02009
75
YR =0.2095 MR oIi = 0.2650
MRatio C.T Off-gas (mgIL)
0.2472
0.2445 8.40
0.2434
0.2487
0.2521 8.53
0.2S07
0.2522
0.2593
0.2618 8.75
0.2631
0.2522
0.2522 8.76
0.2528
0.2553
0.2594 8.41
0.2594
0.2551
0.2515 8.99
0.2527
0.2515
C." (mgIL)
10.11
10.29
10.56
10.55
10.12
10.54
aTE OTE aSOTE SaTE Alpha (%) a>g. avg. avg. avg.
("10) ("10) (%)
6.72
7.74 7.53 8.34 25.75 032
8.14
6.17
4.88 532 20.66 35.87 0.58
539
4.83
2.16
123 \38 2.34 52.89 0.04
0.74
4.85
4.86 4.77 10.99 67.04 0.16
4.61
3.66
2.11 2.63 16.36 4123 0.40
2.11
3.76
5.12 4.66 13.15 47.95 0.27
4.66
5.11
Testing Date: June 9, 2006 Air Temp ("F): 95 Barom Pres (in Hg): 30.01
Test Diffuser Airflow No. Rate
(SCFM)
1 Memb. 1
2 Memb. 1
3 Memb. 1
4 Memb. 1
5 Memb. 2
6 Memb. 2
7 Memb. 2
8 Memb. 2
9 Memb. 2
10 Commie 1
11 Commie 1
12 Commie 1
13 Commie 1
14 Ceramic 1.5
15 Commie 1.5
16 Commie 1.5
17 Cemmic 1.5
Water 00
~~. (mg/L)
30.3 4.43
30.5 6.45
31.0 4.56
31.1 5.14
- ..
TABLE 18. Off-gas Analysis (2) Data Summary
MLSS = 3867 mg/L Theta = 1.024
Ref_ 00_
~;;e ~e 634 600
634 603
638 604
639 604
603 588
596 578
579 567
576 562
570 557
524 510
530 51!
531 511
531 510
520 501
522 502
521 503
522 503
M Fraction 00_ 0.1983
0.1993
0.1983
0.1980
02043
02032
02052
02044
02047
02039
02020
02016
02012
02018
02015
02023
02019
76
VR =02095 MR"" =02650
MRalio C'T 00_ (mg/L)
02473
02488 8.36
0.2474
02469
02567
025SO
02581 8.47
02569
02574
02561
02531 8.61
02525
02519
02529
02523 8.60
02535
02529
C.,. (mg/L)
10.11
10.29
10.56
10.55
OTE Avg. oSOTE SOTE Alpha ("10) OTE ("10) (%)
('Vol
6.69
6.11 6.57 13.53 25.75 0.53
6.65
6.83
3.13
3.79
2.61 3.09 12.81 35.87 0.36
3.06
2.87
3.36
4.49 4.38 8.99 52.89 0.17
4.72
4.95
4.58
4.80 4.57 10.98 67.04 0.16
4.33
4.56
TABLE 18. (Continued) Off-gas Analysis (2) Data Summary
Test Airflow Water 00 Ref-gas Off-gas M MRatio C'T C." OTE Avg. aSOTE SOTE No. Diffuser Rate ~e~r (1llfIL) Ampere Ampere Fradion Off-gas (1llfIL) (mg/L) ("/0) OTE ("/0) ("/0)
Aeration A Wastewater Treatment Process. (1988). WPCF-Manual of Practice No. 13, ASCE Manuals and Report on Engineering Practice No. 68, American Society of Civil Engineers and Water Pollution Control Federation.
APHA, Standard Methods for the Examination of Water and Wastewater, 18th ed., American Public Health Association, Washington DC, 1992.
ASCE (2000). ASCE Standard: Measurement of Oxygen Transfer in Clean Water. ISBN 0-87626-430-7, New York, NY.
Babcock, R. W. Jr., and Stenstrom, M. K. (1993). Precision and Accuracy of Offgas Testing for Aeration Cost Reduction. Proc. 66th annual conference of the Water Environment Federation, Oct. 3-7, Anaheim, California.
Babcock, R. W. Jr., Chun, W., and Strom, H. (2003). The Honouliuli Membrane Bioreactor Pilot Study. Presented at the 25th Annual Hawaii Water Environment Association Conference, Honolulu, Hawaii.
Babcock, R. W. Jr. (2006). Fifth Periodic Progress Report: Honolulu Membrane Bioreactor Pilot Study. Subrnitted to WateReuse Foundation.
Barker, D. J., and Stuckey, D. C. (1999). A Review of Soluble Microbial Products (SMP) in Wastewater Treatment Systems. Water Research, 33 (14), 3063-3082.
Capela. S., Gillot, S., and Heduit A. (2004). Comparison of Oxygen-Transfer Measurement Methods Under Process Conditions. Water Environment Research, 76 (2), 183-188.
Chang, 1. S., et al. (2002). Membrane Fouling in Membrane Bioreactors for Wastewater Treatment. Journal of Environmental Engineering. 128(11), 1018-1029.
Cicek, N., Franco, J. P., and Suidan, M. T. (1999). Characterization and Comparison of a Membrane Bioreactor and a Conventional Activated Sludge System in the Treatment of Wastewater Containing High Molecular Weight Compounds. Water Environment Research, 71 (1),64-70.
Cornel, P., Wagner, M., and Krause, S. (2003). Investigation of Oxygen Transfer Rates in Full Scale Membrane Bioreactors. Water Science and Technology, 47 (11), 313-319.
Dick, R. J., and Ewing, B. B. (1967). The Rheology of Activated Sludge. Journal of Water Pollution Control Federation, 39 (4), 543-560.
96
Drews, A., Evenblij, H., and Rosenberger, S. (2005). Potential and Drawbacks of Microbiology-Membrane Interaction in Membrane Bioreactors. Environmental Progress, 24 (4), 426-433.
Dubois, M., et al. (1956). Colorimetric Method for Determination of Sugars and Related Substances. Analytical Chemistry, 28 (3), 350-356.
Frolund, B., Palmgren, R., and Nielsen, P.H. (1996). Extraction of Extracellular Polymers from Activated Sludge Using a Cation Exchange Resin, Water Research, 30, 1749-1758.
Gea T., et al. (2004). Monitoring the Biological Activity of the Composting Process: Oxygen Uptake Rate (OUR), Respirometric Index (RI), and Respiratory Quotient (RQ). Biotechnology and Bioengineering, 88 (4), 520-527.
Germain, E., and Stephenson, T. (2005). Biomass Characteristics, Aeration and Oxygen Transfer in Membrane Bioreactors: Their Interrelations Explained by a Review of Aerobic Biological Process. Reviews in Environmental Science and BiolTechnology, 4, 223-233.
Groves, K. P., et al. (1992). Evaluation of Oxygen Transfer Efficiency and AlphaFactor on a Variety of Diffused Aeration Systems. Water Environment Research, 64 (5),691-698.
GOnder, B., and Krauth, K. (1999). Replacement of Secondary Clarification by Membrane Separation-Results with Tubular, Plate and Hollow Fiber Modules. Water Science and Technology, 40, 311-320.
GOnder, B. (2001). The Membrane Coupled-Activated Sludge Process in Municipal Wastewater Treatment. Technomic Publishing Company, Inc., U.S.A.
Iranpour, R., et al. (2000). Assessment of Aeration Basin Performance Efficiency: Sampling Methods and Tank Coverage. Water Research, 34 (12), 3137-3152.
Iranpour, R., et al. (2002). Case Study of Aeration Performance under Changing Process Conditions. Journal of Environmental Engineering, 128(6),562-569.
Krampe, J., and Krauth, K. (2003). Oxygen Transfer into Activated Sludge with High MLSS Concentrations. Water Science and Technology, 47 (11),297-303.
Krause, S., Cornel, P., and Wagner, M. (2003). Comparisons of Different Oxygen Transfer Testing Procedures in Full-Scale Membrane Bioreactors. Water Science and Technology, 47 (12), 169-176.
97
Lee, W., Kang, S., and Shin, H. (2003). Sludge Characteristics and Their Contribution to Microfiltration in Submerged Membrane Bioreactors. Journal of Membrane Science, 216,217-227.
Li, F., Chen, J., and Deng, C. (2006). The Kinetics of Crossflow Dynamic Membrane Bioreactor. Water SA. 32 (2), 199-203. http://www.wrc.org.za
Liao, B. Q., et al. (2004). A Review of Biofouling and its Control in Membrane Separation Bioreactors. Water Environment Research, 76 (5), 425-436.
Liu, D. (1983). Resazurin Reduction Method for Activated Sludge Process Control. Environmental Science and Technology, 17 (7), 407-411.
Lowry, O. H., and Rosebrough, N. J. (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry, 193,265-275.
Mueller, J., et al. (2000) Full-Scale Demonstration of Improvement in Aeration Efficiency. Journal of Environmental Engineering, 126 (6), 549-555.
Redmon, D. T., Boyle, W. C., and Ewing, L. (1983) Oxygen Transfer Efficiency Measurements in Mixed Liquor Using Off-gas Techniques, Journal of Water Pollution Control Federation, 55, 1338-1347.
Rosenberger, S., et al. (2000). Operation of Different Membrane Bioreactors: Experimental Results and Physiological State of the Micro-organisms. Water Science and Technology, 41, 269-277.
Song, K. G., et al. (2003). Performance of Membrane Bioreactor System with Sludge Ozonation Process for Minimization of Excess Sludge Production. Desalination, 157,353-359.
Stenstrom M. K. (1989). Upgrading Existing Activated Sludge Treatment Plants with Fine Pore Aeration Systems. Presented at the IA WPRC conference on Upgrading of Wastewater Treatment Plants, Mnich, West Germany, Sept. 3-7, 1989.
Stenstrom M. K. (1997). Clean Water Column Test and Process Tests Report for the Ashland, NH. Wastewater Treatment Facilities Performed on June 19-23, 1995.
Stenstrom M. K. (2005). Report on Oxygen Transfer Testing on Point Lorna WWTP biological aerated filter pilot plants.
U. S. EPA. (1989). EPAIASCE Design Manual on Fine Pore Aeration. Cincinnati, Ohio.
98
Wagner, M., Cornel, P., and Krause, S. (2002). Efficiency of Different Aeration Systems in Full Scale Membrane Bioreactors. Proceedings of the 75th annual conference of the Water Environment Federation, Chicago, lllinois U.S.A.
WERF Project #00-CTS-8a: MBR Website Strategic Research. (2004). Water Environment Research Foundation, Alexandria, Virginia U.S.A. http://www.werf.org
Wisniewski, C., Leon, Cruz A., and Grasmick, A. (1999). Kinetics of Organic Carbon Removal by a Mixed Culture in a Membrane Bioreactor. Biochemical Engineering Journal, 3 (I), 61-69.
Zhang, B., et al. (1997). Floc Size Distribution and Bacterial Activities in Membrane Separation Activated Sludge Processes for Small-Scale Wastewater TreatmentIReclamation. Water Science and Technology, 35 (6), 37-44.