OCCURRENCE AND FATE OF
PERFLUOROOCTANE SULFONATE (PFOS) AND
PERFLUOROOCTANOIC ACID (PFOA) IN WATER
AND WASTEWATER AND THEIR REMOVAL
USING A HYBRID PAC-MBR SYSTEM
YU JING
NATIONAL UNIVERSITY OF SINGAPORE
2010
brought to you by COREView metadata, citation and similar papers at core.ac.uk
provided by ScholarBank@NUS
OCCURRENCE AND FATE OF
PERFLUOROOCTANE SULFONATE (PFOS) AND
PERFLUOROOCTANOIC ACID (PFOA) IN WATER
AND WASTEWATER AND THEIR REMOVAL
USING A HYBRID PAC-MBR SYSTEM
YU JING
(M.Eng., Southeast University)
A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF CIVIL ENGINEERING
NATIONAL UNIVERSITY OF SINGAPORE
2010
i
ACKNOWLEDGEMENT
I would like to take this opportunity to acknowledge and thank all those who
have helped me along the way.
First and foremost, I would like to express my utmost appreciation from
bottom of my heart to my academic supervisor, Associate Professor Jiangyong
HU, for sticking with me through these many years. Without her professional
guidance, constructive advice and constant encouragement, this work could
not have been completed.
Acknowledgements are made to all former and current staff at Environmental
Laboratory in the Centre for Water Research for their kind support and
cooperation in various ways. Heartfelt thanks are also given to my former
Final Year Project students who have made contributions to this work in
whatever aspects.
At last but not least, I would like to thank my family for their vital support and
consistent encouragement. Their love is always the power source for me to
struggle all difficulties in my life.
The research scholarship provided by the National University of Singapore
throughout the whole period of candidature is greatly appreciated.
ii
TABLE OF CONTENTS
ACKNOWLEDGEMENT…………………………………………...……….i
TABLE OF CONTENTS…………………………………………...………..ii
SUMMARY…………………………………………………..…………….viii
NOMENCLATURE…………………………………………………...……xii
LIST OF TABLES…………………………………………………...……...xv
LIST OF FIGURES…………………………………………..…………...xvii
CHAPTER 1 INTRODUCTION……………………………………………1
1.1 Background………………………………………………………….1
1.2 Objective and Scope of Study………………………………………4
1.3 Outline of Thesis…………………………………………………….5
CHAPTER 2 LITERATURE REVIEW....................………………………7
2.1 Introduction…………………………………………………………7
2.1.1 Physico-chemical properties of PFOS and PFOA…………7
2.1.2 Persistence, bioaccumulation and toxicity of PFOS and
PFOA...............................................................................................10
2.1.2.1 Persistence………………………………………………...10
2.1.2.2 Bioaccumulation………………………………………….11
2.1.2.3 Toxicity……………………………………………………11
2.1.3 Preliminary regulations for PFOS and PFOA……………12
2.2 Analytical method for PFCs ............................................................12
2.2.1 Introduction of LC/MS/MS analysis for PFCs…………...12
2.2.2 LC/MS/MS method for water and wastewater…………...13
2.2.3 LC/MS/MS method for sludge and sediment……………..14
iii
2.2.4 Limitation of Electrospray Ionization (ESI)……………...15
2.2.5 Matrix interference………………………………………...16
2.2.6 Post extraction clean-up method for analysis of
environmental matrices………………………………………...18
2.3 Occurrence of PFOS/PFOA in the environment...........................19
2.3.1 Occurrence in the surface water…………………………..19
2.3.2 Occurrence in the drinking water…………………………24
2.3.3 Occurrence in the seawater………………………………..25
2.3.4 Occurrence in the sludge and sediment…………………...26
2.4 Fate and behavior in the sewage treatment plants .......................27
2.4.1 Occurrence in the wastewater……………………………..27
2.4.2 Mass flow and mass change………………………………..28
2.5 Removal Technologies .....................................................................29
2.5.1 Advanced oxidation process……………………………….29
2.5.2 RO/NF membrane………………………………………….31
2.5.3 Adsorption…………………………………………………..32
2.5.3.1 Activated carbon adsorption…………………………….32
2.5.3.2 Adsorption onto sediment and sludge…………………..33
2.5.4 Membrane biological reactor (MBR)……………………..34
2.5.4.1 Introduction………………………………………………34
2.5.4.2 Configuration and application…………………………..35
2.5.4.3 Technology benefits and problems……………………...36
2.5.4.4 Hybrid PAC-MBR system……………………………….38
2.6 Research statement ..........................................................................40
CHAPTER 3 MATERIALS AND METHODS ...........................................48
iv
3.1 Chemicals, materials and reagents .................................................48
3.1.1 Chemicals and reagents……………………………………48
3.1.2 Materials…………………………………………………….48
3.2 Water sample collection and preparation......................................50
3.2.1 Water sample collection……………………………………50
3.2.2 Water sample preparation (Basic SPE extraction)………52
3.3 Experiment on development of clean-up method for wastewater
and sludge sample ..................................................................................53
3.3.1 Silica cartridge clean-up procedure……………………….53
3.3.2 Application of clean-up method to sludge samples………53
3.3.3 Evaluation of matrix effect and recoveries……………….54
3.4 Wastewater and sludge sample collection and preparation .........55
3.4.1 Wastewater and sludge sample collection………………...55
3.4.2 Wastewater and sludge sample preparation……………...57
3.5 PAC-MBR experimental setup and operation ..............................58
3.5.1 MBR configuration…………………………………………58
3.5.2 Synthetic wastewater and operational conditions………..59
3.5.3 PFCs mass balance calculation……………………………62
3.5.4 Membrane resistance calculation…………………………62
3.6 Adsorption study on PAC and activated sludge ...........................63
3.6.1 Preparation of EfOM………………………………………63
3.6.2 EfOM characterization…………………………………….64
3.6.3 Equilibrium adsorption experiments……………………...64
3.6.4 Adsorption kinetics experiments…………………………..65
3.6.5 Mathematical modeling……………………………………66
v
3.7 Analysis method ...............................................................................67
3.7.1 COD, DOC and UV254 analysis…………………………..67
3.7.2 Carbohydrate and protein analysis……………………….67
3.7.3 MLSS and MLVSS…………………………………………67
3.7.4 EPS and SOUR analysis……………………………………67
3.8 LC/MS/MS analysis .........................................................................68
3.8.1 Optimization of LC/MS/MS detection method…………...68
3.8.2 Method validation and quantification…………………….70
3.9 Fractionation process.......................................................................72
3.10 Quality assurance and control ......................................................74
3.11 Statistical analysis ..........................................................................75
CHAPTER 4 OCCURRENCE of PFOS AND PFOA IN WATER AND
WASTEWATER ............................................................................................76
4.1 Introduction ......................................................................................76
4.2 Results and discussions ....................................................................78
4.2.1 PFOS/PFOA Concentration in surface water…………….78
4.2.2 PFOS/PFOA Concentration in wastewater………………84
4.2.3 PFOS/PFOA Concentration in coastal water…………….87
4.2.4 Seasonal variations of PFOS and PFOA………………….89
4.2.5 Correlations between PFOS and PFOA…………………..90
4.3 Summary ...........................................................................................93
CHAPTER 5 DEVELOPMENT OF POST EXTRACTION CLEAN-UP
METHOD FOR PFOS/PFOA DETERMINATION IN WASTEWATER
AND SLUDGE SAMPLES............................................................................96
5.1 Introduction ......................................................................................96
vi
5.2 Results and discussions ....................................................................99
5.2.1 Effect of clean-up procedures on matrix effect…………..99
5.2.2 Effect of internal standards on matrix effect……………101
5.2.3 Detection of PFOS and PFOA in water and sludge
samples…………………………………………………..………103
5.3 Summary .........................................................................................104
CHAPTER 6 BEHAVIOR OF PFOS AND PFOA IN SEWAGE
TREATMENT PLANTS .............................................................................106
6.1 Introduction ....................................................................................106
6.2 Results and Discussion ...................................................................108
6.2.1 PFOS/PFOA in wastewater ………………………………109
6.2.2 Seasonal variation of PFOS and PFOA………………….112
6.2.3 Mass flow in aqueous sample during treatment………...114
6.2.4 PFOS/PFOA in sludge…………………………………….118
6.3 Summary .........................................................................................121
CHAPTER 7 PFOS/PFOA REMOVAL BY HYBRID PAC-MBR
PROCESS .....................................................................................................124
7.1 Introduction ....................................................................................124
7.2 Results and Discussion ...................................................................128
7.2.1 Adsorption study on PAC and activated sludge………...128
7.2.1.1 Characterization of EfOM……………………………...128
7.2.1.2 PFOS and PFOA adsorption onto PAC……………….129
7.2.1.3 Effect of EfOM on the PFOS and PFOA adsorption onto
PAC………………………………………………………………132
7.2.1.4 PFOS and PFOA adsorption onto activated sludge…..134
vii
7.2.2 Performance of MBR and PAC-MBR systems at different
SRT ………………………………………………………………137
7.2.2.1 Overall performance of MBR and PAC-MBR system..137
7.2.2.2 SMP and DOM fraction characteristics……………….140
7.2.3 Removal of PFOS and PFOA in PAC-MBR and MBR
system……………………………………………………………142
7.2.3.1 Removal by adsorption onto activated sludge………...142
7.2.3.2 Removal by adsorption onto PAC……………………..143
7.2.3.3 Mass balance…………………………………………….145
7.2.3.4 Effect of SRT on PFOS and PFOA removal…………..150
7.2.3.5 Effect of PAC dosage on PFOS and PFOA removal….152
7.2.4 Membrane fouling………………………………………...153
7.2.4.1 Variations of TMP………………………………………153
7.2.4.2 Effect of PAC on TMP………………………………….154
7.3 Summary .........................................................................................155
CHAPTER 8 CONCLUSIONS ...................................................................160
8.1 Conclusions .....................................................................................160
8.2 Recommendations ..........................................................................168
REFERENCE ...............................................................................................170
APPENDIX: PUBLICAIONS .....................................................................193
viii
Summary
Perfluorinated compounds (PFCs), particularly perfluorooctane sulfonate
(PFOS, C8F17SO3-) and perfluorooctanoic acid (PFOA, C7F15COO-
), have
emerged as a new class of environmentally persistent pollutants, which have
been widely used in different applications. PFOS and PFOA, regarded as the
terminal breakdown end-products of PFCs, have been detected in a wide array
of environmental matrices including biota, water, air, sediment and sludge.
The primary objective of this thesis is to contribute towards establishment of
fundamental understanding of fate and behavior of PFCs in the aquatic
environment and sewage treatment plants (STPs) as well as development of a
hybrid PAC-MBR process to effectively remove these two trace organic
compounds. More than one hundred water samples from reservoirs,
rivers/canals, coastal waters and treated effluents of wastewater treatment
plants (WWTPs) were collected and analyzed to characterize the spatial
distribution and seasonal variation of PFOS and PFOA in the aquatic and
oceanic environment of Singapore. Coastal waters had lower concentrations of
PFOS and PFOA as compared to surface waters and wastewaters, while
highest concentration of PFOS and PFOA were observed in treated effluents
of two WWTPs. Our results suggest that coastal waters in the western area of
Singapore were more heavily contaminated compared to those in the middle
and eastern areas. Between dry and wet season, significant seasonal difference
(p=0.025) was observed in surface waters for PFOS only, while no discernable
ix
seasonal differences were found for both PFOS and PFOA in coastal waters
and wastewaters.
An efficient sample clean-up method was developed in this study to
significantly remove co-eluting matrix components by applying the SPE
extracts onto a silica cartridge. Internal standardization was used to further
compensate for the matrix effect, which was also proven to improve the signal
reproducibility. The clean-up method described in this study was applied to
different water samples (surface water and wastewater) and sludge samples to
evaluate the efficiency of silica clean-up and the influence of sample origin on
the matrix effect. Results showed that the method was robust and could be
applied to analyze PFOS and PFOA in different environmental matrices. In
water and sludge samples, matrix effect and recovery efficiency were in the
range of 91.8%-98.3% and 81.3%-98.0%, respectively, indicating that clean-
up method can effectively remove co-eluting matrix components in various
environmental matrices.
The behavior of PFOS and PFOA in the biological units of various full-scale
municipal sewage treatment plants was also investigated. Mass flow of PFOS
increased significantly (mean 62.2%) in conventional activated sludge process
(CAS) of plant B, while it remained consistent after the secondary treatment
in plant A. Mass flow of PFOA increased 82.9% (mean) in CAS of plants A
and B and 62.3% (mean) in membrane biological reactor (MBR), while it
remained unchanged after the treatment of liquid treatment module (LTM). In
terms of behavior pattern of PFOS and PFOA, our results suggest that there
x
was no significant difference between conventional activated sludge process
and membrane biological reactor operated at comparable sludge retention time
(SRT). However, mass flow of these two compounds remained consistent
after treatment of activated sludge process operating at short SRT. Seasonal
variations of PFOS in concentrations of raw sewage were found in plant A,
while PFOA did not have significant seasonal variation in both plants A and B.
The adsorption of PFOS and PFOA onto powdered activated carbon (PAC)
was investigated in the presence and absence of effluent organic matter
(EfOM) at low concentration range (0.1-500 µg/L). Adsorption of PFOS and
PFOA onto PAC fitted the Freundlich model well (r2>98%) and adsorption
capacity of PFOS (KF=17.55) and PFOA (KF=10.03) in the absence of EfOM
was more than one order of magnitude higher than that in the presence of
EfOM, indicating EfOM greatly reduce the adsorption capacity of PAC.
Moreover, the EfOM fraction of <1 k, which had greater effect on the
adsorption than that of >30 k fraction, was the major contributor to the
adsorption competition. Additionally, the estimated partition coefficient Kd
was 729 and 154 L/kg for PFOS and PFOA, respectively, suggesting PFOS
and PFOA, especially PFOA, have a low tendency to partition onto the
activated sludge.
The overall performance and removal efficiencies of PFCs were also
investigated in PAC-MBRs which were operated with different PAC dosages
and SRTs. On the one hand, the effect of PAC dosage on the removal of PFCs
in PAC-MBR was studied at the SRT of 30 d. Removal efficiency of PFCs
xi
increased with the increase of PAC dosage from 30 to 100 mg/L, suggesting
adsorption on PAC was the efficient and predominant process in the removal
of PFCs in activated sludge system. On the other hand, the effect of SRT on
removal of PFCs in PAC-MBR was studied. Removal efficiencies of PFCs
were >90% for PFOS and >84% for PFOA at different SRT studied,
suggesting that adsorption onto PAC could be dominant and removal
efficiencies may be not significantly affected by different operational SRTs.
With the increase of SRT, PFCs concentration on PAC decreased significantly,
indicating significant effect of SRT on the PAC adsorption capacity in PAC-
MBR due to different PAC concentrations at different SRTs.
Keywords: PFOS, PFOA, Aquatic environment, Sewage treatment
plant (STP), Hybrid powdered activated carbon-membrane biological
reactor (PAC-MBR), Fate and behavior
xii
NOMENCLATURE
AHS Acid Humic Substance
AMWD Apparent Molecular Weight Distribution
BAFs Bioaccumulation Factors
CAS Conventional Activated Sludge Treatment
COD Chemical Oxygen Demand
DOC Dissolved Organic Carbon
DOMs Dissolved Organic Matters
EDCs Endocrine Disruptors Compounds
EfOMs Effluent Organic Matters
EPS Extra Cellular Polymeric Substance
ESI Electrospray Ionization
GC Gas Chromatography
GAC Granular Activated Carbon
HDPE High Density Polyethylene
HPLC High-performance Liquid Chromatography
HRT Hydraulic Retention Time
IDL Instrumental Detection Limit
J Permeate Flux
Kd Partition Coefficient
Koc Organic Carbon Partition Coefficient
Kow Octanol-water Partition Coefficient
LOD Limit Of Detection
xiii
LOQ Limit Of Quantification
MBR Membrane Bioreactor
ME Matrix Effect
MF Microfiltraiton
MLSS Mixed Liquor Suspended Solids
MRM Multiple Reaction Monitoring
MS Mass Spectrometry
NMR Nuclear Magnetic Resonance
NOMs Natural Organic Matters
NPS Nonpoint Source Pollution
PAC Powdered Activated Carbon
pKa Acid Dissociation Constant
PFASs Perfluoroalkyl Sulfonates
PFCAs Perfluoroalkyl Carboxylates
PFCs Perfluorinated Compounds
PFDoA Perfluorododecanoic Acid
PFEES Pperfluoro (2-ethoxyethane) Sulfonic Acid
PFO Perfluorooctanoate
PFOS Perfluorooctane Sulfonate
PFOA Prfluorooctanoic Acid
POPs Persistent Organic Pollutants
RE Recovery Efficiency
Rm Intrinsic Membrane Resistance
Rr Reversible Resistance
Ri Irreversible Resistance
xiv
SMPs Soluble Microbial Products
SOUR Specific Oxygen Uptake Rate
SPE Solid Phase Extraction
SRT Sludge Retention Time
STPs Sewage Treatment Plants
TMP Transmembrane Pressure
UF Ultrafiltraton
VP Vapor Pressure
VSS Volatile Suspended Solids
WWTPs Wastewater Treatment Plants
WAS Waste Activated Sludge
xv
LIST OF TABLES
Table 2.1 Definition of acronyms and structures of PFCAs and PFSAs….8
Table 2.2 Physico-chemical properties of PFOS and PFOA…………......10
Table 2.3 Review of PFOS and PFOA concentrations in surface water
(ng/L), drinking water (ng/L), wastewater (ng/L) and sludge (ng/g)….....21
Table 2.4 Sorption and desorption coefficients of PFOS from various
matrices...........................................................................................................33
Table 3.1 Characteristics of powdered activated carbon ...........………...49
Table 3.2 characteristics of the MF hollow fiber membrane………….....50
Table 3.3 Wastewater treatment plants characteristics………………......52
Table 3.4 Composition and concentration of synthetic wastewater……..60
Table 3.5 PAC added at the startup of PAC-MBR system…………….....61
Table 3.6 MRM-transitions, compound-dependent parameters of the
analytes............................................................................................................69
Table 3.7 IDL, LOD and LOQ of PFOS and PFOA…………………..….72
Table 5.1 Influence of sample clean-up and internal standardization on
ME% and RE% (n=5)………………………………………………….…101
Table 5.2 Influence of sample origin on ME% and RE% with internal
standard (n=3)……………………………………..……………..………..103
Table 5.3 ME% and RE% with internal standard by application of clean-
up method on sludge samples (n=3)………………….………………..….104
Table 6.1 Mass flow (mg/d) of PFCs in influent, effluent and solid waste in
CAS1, CAS2, LTM and MBR……………………...……………………..116
xvi
Table 6.2 Calculated partition coefficient Kd in primary sludge and
activated sludge……………………………………………………...……..121
Table 7.1 Characteristics of EfOM solution obtained from the lab scale
MBR (n=5)…………………………………………………………………128
Table 7.2 Langmuir isotherm constants and Freundlich isotherm
constants for the adsorption of PFCs onto PAC at 25 oC…………….....131
Table 7.3 Freundlich iostherm parameters for the adsorption of PFCs on
PAC in EfOM fractions……………………………….…………………..134
Table 7.4 Linear isotherm parameters for PFCs onto activated sludge..135
Table 7.5 Measured PFCs concentrations in activated sludge of MBR at
different SRT……………………………..………………………………..137
Table 7.6 Estimated mass flows of PFCs in activated sludge of WAS in
PAC-MBR operated at different SRTs…………………………..………150
Table 7.7 Effect of SRT on the PFCs removal in PAC-MBR system with
PAC dosage of 100 mg/L (based on mass balance)…………………..…..152
Table 7.8 Effect of PAC dosage on the PFCs removal in PAC-MBR
system (based on mass balance)…………………..………………………153
Table 7.9 Resistances of membrane for the MBR and PAC-MBR
systems……………………………………………………………………...155
xvii
LIST OF FIGURES
Figure 1.1 Research scope and content…………………………..…………5
Figure 2.1 Structures of PFOS and PFOA……………………………...…..9
Figure 2.2 Average mass flow (mg/d) for PFOS and PFOA in sewage
treatment plant...............................................................................................29
Figure 2.3 Configuration of MBR systems: (a) Side-stream MBR, (b)
Suctioned- filtration submerged MBR, and (c) Gravitational-filtration
submerged MBR…………………………………………………………….35
Figure 3.1 Sampling locations for reservoir waters, river/canal waters,
effluents of WWTPs, coastal waters and location for outfalls of
WWTPs…………………………………………………………..………….51
Figure 3.2 Flow scheme of the sewage treatment plants A and B……......56
Figure 3.3 Schematic diagram of lab-scale PAC-MBR system………......59
Figure 3.4 Mass balances of PFCs in PAC-MBR or MBR system……….62
Figure 3.5 LC/MS/MS chromatograms of PFOS, PFOA and internal
standards PFEES and PFDoA……………………………………………...70
Figure 3.6 Procedure for fractionation of DOM……………………..…...74
Figure 4.1 Concentrations of PFOS and PFOA in surface waters,
wastewaters and coastal waters from western area of Singapore collected
by: 1. Oct 2006; 2. Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007…...80
Figure 4.2 Concentrations of PFOS and PFOA in surface waters,
wastewaters and coastal waters from middle area of Singapore collected
by: 1. Oct 2006; 2. Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007…...81
xviii
Figure 4.3 Concentrations of PFOS and PFOA in surface waters,
wastewaters and coastal waters from eastern area of Singapore collected
by: 1. Oct 2006; 2. Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007…....82
Figure 4.4 Total PFOS and PFOA concentrations in surface waters
summed up by 5 sampling campaigns.……………………………………83
Figure 4.5 Total PFOS and PFOA concentrations in wastewaters summed
up by 5 sampling campaigns……………………………………………….87
Figure 4.6 Total PFOS and PFOA concentrations in coastal waters
summed up by 5 sampling campaigns……………………………………..88
Figure 4.7 Correlations of PFOS and PFOA between coastal water C4
and wastewater W2........................................................................................89
Figure 4.8 Correlations between PFOS and PFOA concentrations in
surface waters.………………………………………………………………92
Figure 4.9 Correlations between PFOS and PFOA concentrations in
coastal waters…………….…….……………………………………………92
Figure 4.10 Correlations between PFOS and PFOA concentrations in
wastewaters……………………..…………………………………………...92
Figure 4.11 Correlations between PFOS and PFOA concentrations in the
effluents of (a) W2 and W4; (b) W5………………………...…………...93
Figure 5.1 LC-MS-MS chromatograms of PFOS and PFOA in the raw
sewage extracted by (a) HLB SPE and (b) HLB+silica…………..…….100
Figure 6.1 PFOS concentrations in wastewater of STP A (CAS1, LTM
and MBR) and STP B (CAS2)…………………………………………….109
Figure 6.2 PFOA concentrations in wastewater of STP A (CAS1, LTM
and MBR) and STP B (CAS2)…………….…………………...………….111
xix
Figure 6.3 Seasonal variations in influent concentrations of (a) PFOS and
(b) PFOA in STPA. 1: Oct 06 (CAS1), 2: Mar 07 (CAS1), 3: Sep 07
(CAS1), 4: Mar 07 (MBR), 5: Sep 07 (MBR), 6: Oct 06 (LTM), 7: Mar 07
(LTM), 8: Dec 06 (CAS1), 9: Dec 06 (LTM), 10: Dec 07 (CAS1), 11: Dec
07 (MBR)......................................................................................................113
Figure 6.4 Seasonal variations in influent concentrations of (a) PFOS and
(b) PFOA in STP B……………………………….………………………..114
Figure 6.5 Change of mass flow after primary treatment in (a) STP A and
(b) STP B……………….…………………………………………………..118
Figure 6.6 PFOS concentrations in sludge samples from STP A and STP
B………………………………………………………………………….....119
Figure 6.7 PFOA concentrations in sludge samples from STP A and STP
B…………………………………………………………………………….120
Figure 7.1 Adsorption isotherms of PFCs onto the PAC in the absence
and presence of EfOM: (a) PFOS; (b) PFOA. Experimental data fit to
Freundlich model (solid line)…………………...…………………………131
Figure 7.2 Adsorption of PFOS and PFOA onto PAC as a function of
contact time: (a) in the presence of EfOM; (b) in the Milli-Q water…...132
Figure 7.3 Log-log plot of PFCs adsorption isotherms in the presence and
absence of EfOM fractions: (a) PFOS and (b) PFOA…………………...133
Figure 7.4 Adsorption isotherms of PFCs onto the activated sludge…...136
Figure 7.5 COD removal in MBR and PAC-MBR systems with different
SRTs……………………………………………….………………..………138
Figure 7.6 DOC of supernatant and effluent in MBR and PAC-MBR
system with different SRTs………………………………………..………138
xx
Figure 7.7 MLSS/MLVSS in MBR and PAC-MBR systems with different
SRTs………………………..………………………………………………139
Figure 7.8 SOUR in MBR and PAC-MBR systems with different
SRTs………………………………………………………………………...140
Figure 7.9 AMWD of SMP in the supernatant of (a) MBR and (b) PAC-
MBR systems at different SRTs…………….…………………………….141
Figure 7.10 Hydrophobicity of DOM in the supernatant of (a) MBR and
(b) PAC-MBR systems at different SRTs……………….………..……...141
Figure 7.11 PFCs removal in MBR with different SRTs…………..…...143
Figure 7.12 PFCs removal in PAC-MBR system operated with different
PAC dosages…………………………………………………………….…144
Figure 7.13 PFCs removal in PAC-MBR system with PAC dosage of 100
mg/L at different SRTs………………………….……..……….…………145
Figure 7.14 Distribution of removed PFCs flow in MBR operated at
different SRT: (a) PFOS; (b) PFOA……………………..………..……..147
Figure 7.15 Estimated distributions of removed PFCs mass flow in waste
of PAC-MBR at SRT 30 d with different PAC dosage: (a) PFOS; (b)
PFOA………………………………………………………….……..……..148
Figure 7.16 Estimated distributions of removed PFCs mass flow in waste
of PAC-MBR operated at different SRTs: (a) PFOS; (b)
PFOA…………………………………………………………….…………149
Figure 7.17 Effect of SRT on the PFCs adsorption onto activated sludge in
MBR: (a) PFOS; (b) PFOA……………………………………….……...150
Figure 7.17 Long-term TMP profile for the MBR and PAC-MBR systems
at different SRTs……………………….…………………………………154
Chapter 1-Introduction
1
CHAPTER 1 INTRODUCTION
Perfluorinated compounds (PFCs) have been manufactured for over 50 years
and, due to their unique properties of repelling both water and oil, they have
been used as surfactants and surface protectors in carpets, leather, paper, food
containers, fabric, and upholstery and as performance chemicals in products
such as fire-fighting foams, floor polishes, and shampoos. Widespread use of
PFCs has led to ubiquitous occurrence of these chemicals in the environment
particularly Perfluorooctane sulfonate (PFOS, C8F17SO3-) and
perfluorooctanoic acid (PFOA, C7F15COO-), which are the final breakdown
products of PFCs. PFOS and PFOA are also well known for their application
in production of Teflon and other stain resistant materials.
1.1 Background
The occurrence of PFOS and PFOA have been reported in human blood,
biological tissues, water, air, sludge, sediment, and soil since 1968 that they
were first detected with nuclear magnetic resonance (NMR) spectroscopy
(Taves, 1968; Giesy et al., 2001; Taniyasu et al., 2003; Saito et al., 2003;
Martin et al., 2003; Higgins et al., 2005; Sinclair et al., 2006). Currently, high-
performance liquid chromatography (HPLC) with triple quadrupole mass
spectrometry in electrospray negative mode is the most promising and
extensively applied method for analyzing PFCs in various environmental and
biological matrices (Giesy et al., 2001; Taniyasu et al., 2003; Kannan et al.,
Chapter 1-Introduction
2
2004; Moody et al., 2001; Hansen et al., 2001; Martin et al., 2004b). Analysis
was accomplished by direct injection (Schultz et al., 2006) or preconcentration
on solid phase extraction (SPE) cartridges, followed by LC/MS/MS analysis
(Giesy et al., 2001; Tomy et al., 2004; Becker et al., 2008; Boulanger et al.,
2005; Sinclair et al., 2006).
It was reported that PFOS and PFOA were detected in surface waters (Hansen
et al., 2002; Boulanger et al., 2004; Loos et al., 2008), wastewaters (Boulanger
et al., 2005; Becker et al., 2008; Yu et al., 2009), drinking waters (Harada et
al.,2003), ground waters (Schultz et al., 2004) and coastal waters (So et al.,
2004; Saito et al., 2003; Yamashita et al.,2005) all over the world. As their
ubiquitous presence in the environment, PFOS and PFOA arouse great
concerns due to their impact on animal and human. They are known to cause
acute and subchronic toxicity effects in laboratory studies (Haughom et al.,
1992; Seacat et al., 2003). One main concern is their persistence and
bioaccumulativity on live tissue. PFOS and PFOA are readily absorbed by
mammals following oral and inhalation exposure. Once absorbed in the body,
they distribute mainly in the serum and the liver (Kudo et al., 2003; OECD,
2002; US EPA, 2003). However, there is no evidence of any metabolic
degradation of PFOS and PFOA (Kissa, 2001; Schultz et al., 2003).
Furthermore, both chemicals are poorly excreted in both urine and feces.
Biological half-life of PFOA in plasma of a few days for mice and rats and
approximately 4.4 years for humans are reported (Kudo et al., 2003). Half-life
of PFOS varies from 7.5 days in rats to 8.7 years in humans, estimated from
retired 3M production workers (3M, 1999; OECD, 2002; Thibodeaux et al.,
Chapter 1-Introduction
3
2003). Another main concern regarding their adverse effect on animals is
endocrine disruption. A well-known case, for example, is that some male
fishes that are exposed to these pollutants may undergo feminization. These
compounds can bind with the natural estrogen receptors (ER) in the organism
body, and consequently interfere with the normal binding of hormones
generated by the body with ER. So far, more and more evidences of
malfunction of organisms are considered to be related to estrogenic
compounds although direct evidences and clear mechanism of estrogenic
effect still need to be revealed (OECD, 2002; US EPA, 2003).
Due to the toxic and adverse estrogenic effects, investigations on the fate of
PFOS and PFOA have been extensively carried out. For example, the pathway
and distribution in aquatic environment such as river, lake and seawater have
been researched. The main contamination source resulting in their occurrence
in the environment could be the sewage treatment plants (STPs), which
receive industrial and domestic wastewater discharges and usually consist of
conventional activated sludge treatment (CAS). Even though the precursors
could be degraded and produce PFOS and PFOA in the atmosphere, STPs are
identified as the major contamination source, through which PFOS and PFOA
enter into the aquatic environment. These compounds are discharged into the
environment with increased mass flow as they are resistant to CAS. For
example, STPs played an important role in the release of these compounds
into the local environment in some cities in U.S.A, Europe and Japan
(Boulanger et al., 2005; Hansen et al., 2006; Moody et al., 2005). Also, it was
Chapter 1-Introduction
4
observed that mass flow of PFOS and PFOA increased after treatment of CAS
(Schultz et al., 2006).
As the STPs can not effectively remove PFOS and PFOA, these compounds
enter into the environment and occur in the drinking water at trace
concentrations. For example, Harada et al. (2003) observed that the mean
levels ranged from 0.1 to 40.0 ng/L for PFOA and from <0.1 to 12.0 ng/L for
PFOS in treated drinking water in Japan. Although the adverse effect under
such concentration is not clear till now, it is certain that long-term exposure
will cause unexpected adverse effect since they are persistent and easily
accumulated in biological tissue. Therefore, research on the removal
technologies is important and urgent. Currently, various biological and
physico-chemical treatment processes including adsorption, biological
treatment, advanced oxidation and membrane separation have also been
studied to remove these compounds. However, these processes cannot remove
these pollutants both technologically and cost-effectively. The removal of
these compounds is still a challenge, especially for the full-scale wastewater
treatment. Thus, new advanced processes and removal mechanism have to be
developed and studied to remove these PFCs compounds effectively at low
cost for wastewater treatment.
1.2 Objective and Scope of Study
The primary research objective is to contribute towards establishment of
understanding of fate and behavior of PFOS and PFOA in environment and
full-scale activated sludge treatment system as well as development of proper
Chapter 1-Introduction
5
removal technology for wastewater treatment. Figure 1.1 shows the detailed
research scope and content. The specific objectives are listed as follows:
і) Characterize the spatial distribution and seasonal variation of PFOS and
PFOA in the aquatic and oceanic environment of Singapore.
іі) Develop a novel post extraction clean-up method for the determination
of PFOS and PFOA in environmental matrices, such as wastewater and
sludge.
ііі) Investigate fate and behavior of PFOS and PFOA in full-scale activated
sludge treatment system.
іv) Study the adsorption of PFOS and PFOA onto powdered activated
carbon (PAC) and activated sludge as well as removal of PFOS and PFOA
by hybrid PAC-MBR process.
PFCs Concentrations (liquid phase)
i) Occurrence and fate in water
Rivers/Canals
Reservoirs/Lakes
Effluents from STPs
ii) Development of clean-up method
Wastewater Sample
(liquid phase)
Sludge Sample (solid phase)
Coastal Water
Solid Phase Extraction (SPE)
Post Extraction Clean-up method
Reduce Matrix Effect
Solid Phase Extraction (SPE) LC/MS/MS
iv) Removal in hybrid PAC-MBR process
liquid phase
iii) Behavior in STPs
Seasonal Variation
Mass Change
Effect of SRT
Partition Coefficient
CAS MBR LTM
solid phaseKinetic Study
Adsorption Experiment
Equili-brium Study
PAC&Sludge
PAC-MBR
MBR
Effect of PAC dosage
Effect of SRT
Adsorption by PAC and sludge
Overall Performance
Mass Balance
Figure 1.1 Research scope and content.
1.3 Outline of Thesis
Chapter 1-Introduction
6
This thesis provides an overview of the spatial and seasonal distribution of
PFOS and PFOA in the waters of Singapore, develops a novel post-extraction
clean-up method for the determination of these two compounds in
environmental matrices, investigates the effect of SRT on the behavior of
these two compounds in the activated sludge process and explores removal
strategy of hybrid PAC-MBR process. The background information and
literature review, which shows the necessity and importance of the study, are
presented in Chapter 1. Chapter 2 reviews current available literature on PFCs,
including their basic properties, analytical method, occurrence in the
environment, fate and behavior in STPs and removal technologies. Chapter 3
describes the detailed materials and methods used in this study. Spatial and
seasonal distribution of PFOS and PFOA in different water matrices in
Singapore are presented in Chapter 4. Chapter 5 discusses the development of
post-extraction clean-up for wastewater and sludge sample as well as its effect
on eliminating matrix interference in complicated environmental samples.
Chapter 6 compares the behavior of PFOS and PFOA in full-scale
conventional activated sludge processes and membrane biological reactor, as
well as in an activated sludge process operated with a short SRT. Chapter 7
explores overall removal performance and factors affecting PAC adsorption
capacity in hybrid PAC-MBR process. Conclusion from this study and
recommendations for improvements and future study directions are presented
in Chapter 8.
Chapter 2-Literature Review
7
CHAPTER 2 LITERATURE REVIEW
2.1 Introduction
Perfluorinated compounds (PFCs) include perfluoroalkyl carboxylates (PFCAs)
and sulfonates (PFASs) with variable chain-lengths usually between about 6
and 15 carbon atoms. In addition, they contain precursors, which may break
down to PFASs or PFCAs of different chain lengths. The final breakdown
products are the sulfonates and carboxylates like PFOS and PFOA. In
perfluorinated organic compounds or perfluorochemicals all hydrogen atoms
of the corresponding hydrocarbon compound are substituted for fluorine atoms.
The polar carbon-fluorine bond is the most stable bond in organic chemistry.
Therefore, PFCs are thermally and chemically more stable than the analogue
hydrocarbons. One important group of PFCs is the group of perfluorinated
surfactants. They consist of a hydrophilic end group, i.e., sulfonate or
carboxylate end group, and a hydrophobic perfluorinated carbon chain (Table
2.1). Perfluorinated alkylsulfonates and carboxylates occur in numerous
consumer products as active ingredients, impurities or as degradation products
of derivatives, e.g. in oil, water and stain repellents for paper, leather and
textiles or in fire fighting foams. They may be emitted to the aquatic
environment during production and application and also after waste disposal.
Among all PFCs, the most important key compounds are PFOS and PFOA.
2.1.1 Physico-chemical properties of PFOS and PFOA
Chapter 2-Literature Review
8
Structures of PFOS and PFOA are shown by Figure 2.1. The reported pKa
values of PFOA is 2-3 (Gilliland et al., 1992), indicating PFOA are present in
the environment. At pH 7, only 3-6 in 100,000 molecules are PFOA, with the
remaining being perfluorooctanoate (PFO). Physico-chemical properties of
PFOS and PFOA are summarized in Table 2.2. The pKa for PFOS has not
been measured but is expected to be negligible. A calculated pKa of -3.27 for
PFOS indicates that PFOS will be present in the environment completely in
the ionized form (OECD, 2002).
Table 2.1 Definition of acronyms and structures of PFCAs and PFSAs.
Chapter 2-Literature Review
9
Figure 2.1 Structures of PFOS and PFOA.
The vapor pressure (VP) of 3.31x10-4 Pa has been measured for the potassium
salt of PFOS, using the spinning rotor method (OECD, 2002). Vapor pressures
of PFOA and perfluorononanoic, -decanoic, -undecanoic, and –dodecanoic
acids have been measured at the temperature range of 59.25-190.80 oC (Kaiser
et al., 2005). Extrapolation of the Antoine equation to 25 oC for PFOA results
in an estimated VP of 4.2 Pa (Kissa, 2001; US EPA, 2003). The solubility of
PFOS in water is reported to be 519 mg/L at 20±0.5 oC, and 680 mg/L at 24-
25 oC (3M, 2003). The sharp increase of solubility with temperature is
qualitatively consistent with the reported Krafft point of PFOS. The Krafft
temperature is the limit at which compounds cease to be singly dispersed and
begin to form micelles. Above the Krafft point, the solubility increases
abruptly on account of the formation of micelles. The solubility of PFOA in
water has not been published, although it is expected to be less soluble than
PFOS. The aqueous solubility of PFOA could be determined in a concentrated
acid solution. The octanol-water partition coefficient (Kow) is often used to
estimate other properties such as bioconcentration factors and sorption
coefficients. The surface active properties of PFCs make a direct
determination of the Kow impossible. For example, PFO/PFOA is expected to
form multiple layers in octanol/water making determination of Kow extremely
Chapter 2-Literature Review
10
difficult (US EPA, 2003). In a preliminary study reported by 3M an
inseparable emulsion was formed. No measurements of the Henry’s law
constant (H) have been made for PFOS or PFOA. H is usually given by the
ratio of vapor pressure and water solubility. H for PFOS is expected to be very
low and H for PFOA is expected to be relatively high. 3M (2003) reported H
of 3.19x10-4 Pa·m3/mole for PFOS by calculated as the ratio of vapor pressure
and water solubility.
Table 2.2 Physico-chemical properties of PFOS and PFOA.
Property PFOS PFOA Mocular weight 500a 414
Vapor pressure (Pa) 3.31 x 10-4 1.3 x 104 Kow N.A N.A
Henry's law constant (Pa·m3/mole) 3.19 x 10-4 1.52 x 103 Water solubility (g/L) 0.519 3.4
pKa -3.27b 2.5 Note: a. potassium salt; b. calculated
2.1.2 Persistence, bioaccumulation and toxicity of PFOS and PFOA
2.1.2.1 Persistence
PFCs are stable to acids, bases, oxidants, and reductants and are generally not
believed to undergo metabolic or other degradation in the environment
(Schultz et al., 2003; Kiss 2001). Hatfield (2001) reported that aqueous
photolytic degradation of PFOA showed rather long half-life times in natural
environment. PFOS also showed its resistance to advanced oxidation
processes including ozone, ozone/UV, ozone/H2O2 and Fenton reagent due to
very strong and stable carbon-fluorine bond (Hori et al., 2006; Moriwaki et al.,
2005). Biological half-life of PFOA in plasma of a few days for mice and rats
and approximately 4.4 years for humans were reported (Kudo et al., 2003).
Chapter 2-Literature Review
11
Half-life of PFOS varied from 7.5 days in rats through 200 days in
Cynomolgus monkeys to 8.7 years in humans, estimated from retired 3M
production workers (3M, 1999; OECD, 2002; Thibodeaux et al., 2003).
2.1.2.2 Bioaccumulation
Bioaccumulation factors (BAFs) represent accumulation potentials of organics
from environment to organisms. BAFs are calculated by dividing the average
concentrations in organism by the concentrations in water environment as
partition coefficient between octane and water phases for PFOS and PFOA are
not measurable (OECD, 2002; US EPA, 2002). Preliminary study showed
dietary BAFs of PFOS were 2796 in bluegill sunfish and 720 in carp (OECD,
2002). BAFs of PFOA were about 2 in fathead minnow and 3~8 in carp (US
EPA, 2002), which are much lower than PFOS.
2.1.2.3 Toxicity
PFCs are known to cause acute and subchronic toxicity effects in laboratory
studies. PFOA can cause peroxisome proliferation and affect mitochondrial,
microsomal, and cytosolic enzymes and proteins involved in lipid metabolism
(Kudo et al., 2003; Lau et al., 2003; Lau et al., 2004). Also PFOA reportedly
exerts other toxic effects, including accumulation of triglycerides in liver and
reduction of thyroid hormone in circulation (US EPA, 2003). PFOA produces
hepatomegaly, focal hepatocyte necrosis, hypolipidemia, alteration of hepatic
lipid metabolism, peroxisome proliferation, induction of the cytochrome P450
superfamily, and uncoupling of oxidative phosphorylation in laboratory-
exposed animals (Case et al., 2001). Exposure of rats and rabbits to PFOS and
Chapter 2-Literature Review
12
n-EtFOSA results in reduced body weight gain, feed consumption, litter size,
and fetal weight at doses >5 mg/kg∙d. There is lot of information on toxicity
and toxico-kinetics of perfluorinated chemicals in the literature.
2.1.3 Preliminary regulations for PFOS and PFOA
PFOS and PFOA were recently nominated as candidates for POPs by the
Stockholm Convention in May 2009. Exposure criteria of PFCs for human
health were still in debating and there was no agreement yet. Minnesota
Department of Health recommended 0.3 μg/L for PFOS and 0.5 μg/L for
PFOA in drinking water as the safe level for human health in 2007 (MDH,
2007). However, North California Division of Water Quality proposed 2 μg/L
of PFOA to be interim maximum allowable concentration (NC DWQ, 2006).
Rather high screening levels of PFOA was established by West Virginia of
USA (WV DEP, 2002), which were 150 μg/L for water environment and 1360
μg/L for aquatic life. On January 15, 2009 U.S. Environmental Protection
Agency (US EPA) set a "provisional health advisory" of 0.4 ppb for PFOA
and 0.2 ppb for PFOS as safe level in drinking water (US EPA, 2009).
However, the advisory is not meant to protect the public from long term
exposure but might protect individuals for a couple of years.
The European Parliament approved a new EU directive (2006/122/EU) on
restrictions of marketing and use of PFOS and PFOS-related substances,
which came into effect on June 27, 2008. The provisions imply a prohibition
to use PFOS and substances that could degrade to PFOS in chemical products
Chapter 2-Literature Review
13
and articles. Fire-fighting foams that have been placed on the market before 27
December 2006 can be used until 27 June 2011.
2.2 Analytical method for PFCs
2.2.1 Introduction of LC/MS/MS analysis for PFCs
More than three decades ago, Taves and co-workers first postulated that
perfluoroalkyl substances were widespread environmental contaminants
(Taves, 1968; Martin et al., 2004a). They used arduous, yet elegant, methods
to extract, clean up, and detect organic fluorine in human serum with nuclear
magnetic resonance (NMR) spectroscopy. These first studies revealed
compounds that resembled perfluorooctanoic acid (PFOA), but the inherent
ambiguity of the detection system prevented definitive identification. In
addition, the low concentration, lack of authentic standards, and unusual
physical and chemical properties of perfluoroalkyl chemicals made it difficult
to confirm their identity by traditional techniques, such as gas
chromatography/mass spectrometry (GC/MS).
Perfluorinated surfactants can be determined using derivatization techniques
coupled with gas chromatography followed by electron capture detection and
mass spectrometric detection (Jahnke, et al., 2006; Shoeib, et al., 2006). Since
PFOS has low volatility and its derivatives are unstable (Hekster, et al., 2002),
gas chromatography is not applicable for the determination of PFOS. It
implies liquid chromatography, which separates the analyte from other
molecules in the mixture based on differential partitioning between the mobile
and stationary phases, could be the suitable method to analyze PFCs. Ohya et
Chapter 2-Literature Review
14
al. (1998) applied high-performance liquid chromatography (HPLC) and
fluorescence detection to measure perfluorocarboxylic acid concentrations in
biological samples.
2.2.2 LC/MS/MS analytical method for water and wastewater
High-performance liquid chromatography (HPLC) with triple quadrupole
mass spectrometry in electrospray negative mode is the most promising and
extensively applied method for analyzing PFCs in various environmental and
biological matrices, including water, wastewater, sludge and sediment samples
(Giesy et al., 2001 ; Kannan et al., 2002; Tomy et al., 2004; Martin, et al.,
2004a; Higgins et al., 2005; Hansen et al., 2001; Tseng et al., 2006). Up to
date, internal standard is generally used for quantitation of perfluorinated
compounds in water and wastewater since internal standard compensates
matrix suppression. Sixteen short- and long-chain perfluorinated compounds
were quantified by internal standards in water sample (Taniyasu et al., 2005).
Seven perfluorinated compounds were detected at ppt level in seawater by
internal standard quantitation using LC/MS/MS (Yamashita et al., 2004). Six
precursors and PFOS were detected in lake water by internal standard
quantitation (Boulanger et al., 2005). In municipal wastewater, quantitative
determination of perfluorinated compounds were successfully conducted by
two internal standards (Higgins et al., 2005; Tseng et al., 2006).
External standard quantitation was applicable to detect surface water
(Boulanger et al., 2004), but not suitable for wastewater because matrix
interference caused low recovery. It was observed the matrix interference on
Chapter 2-Literature Review
15
PFOA and PFOS analysis caused low recovery (<35%) in influent of one STP
in Iowa (Boulanger et al., 2005).
2.2.3 LC/MS/MS analytical method for sludge and sediment
Quantitative determination of perfluorinated compounds in sludge and
sediment was achieved by three internal standards using HPLC with triple
quadrupole mass spectrometry in electrospray negative mode by internal
standard quantification (Higgins et al., 2005). Internal standard (surrogate
standard) was recommended and it compensated the loss due to matrix
interference. External standard was not available to quantify PFOS and PFOA
in sludge and sediment due to matrix suppression.
2.2.4 Limitation of Electrospray Ionization (ESI)
Electrospray ionization (ESI) is a method used to generate gaseous ionized
molecules from a liquid solution. This is done by creating a fine spray of
highly charged droplets in the presence of a strong electric field. The sample
solution is sprayed from a region of a strong electric field at the tip of a metal
nozzle maintained at approximately 4000 V. The highly charged droplets are
then electrostatically attracted to the mass spectrometer inlet. Either dry gas,
heat or both are applied to the droplets before they enter the vacuum of the
mass spectrometer, thus causing the solvent to evaporate from the surface. As
the droplet decreases in size, the electric field density on its surface increases.
The mutual repulsion between like charges on this surface becomes so great
that it exceeds the forces of surface tension, and ions begin to leave the droplet
Chapter 2-Literature Review
16
through what is known as a “Taylor cone”. The ions are directed into an
orifice through electrostatic lenses leading to the mass analyzer.
ESI is especially useful in producing ions from macromolecules because it
overcomes the propensity of these molecules to fragment when ionized. It is
currently indispensable for identifying and quantifying perfluorinated acids;
however, this method has some inherent limitations such as low salt tolerance,
low tolerance for mixtures and difficulty in cleaning overly contaminated
instrument due to high sensitivity for certain compounds. In particular, co-
eluting matrix components can either suppress or enhance ionization, which
must be controlled to achieve maximum accuracy. For example, Benijts et al.
(2004) observed a decrease of 66% and an increase of 72% in MS/MS
response for 4-t-Octylphenol and estriol, respectively. In addition, several
studies have shown that matrix effects resulting from co-eluting residual
matrix components enhanced or suppressed electrospray ionization of
perfluorinated analytes, leading to considerable inaccuracy (Boulanger et al.,
2005; Higgins et al., 2005).
2.2.5 Matrix interference
Matrix interference resulting from co-eluting residual matrix components
affects the ionization efficiency of target analytes and can lead to erroneous
results. It was reported that recoveries of STP influent are only 34% (PFOS)
and 16% (PFOA), while effluent was 74% (PFOS) and 80% (PFOA)
(Boulanger et al., 2005). This low recovery of influent is due to matrix
suppression of analyte signals, which is confirmed by standard addition to the
Chapter 2-Literature Review
17
final extracts of influent. It was also observed the matrix interference on
PFOA and PFOS analysis caused low recovery (<35%) in influent of one STP
in Iowa (Schultz et al., 2006a).
Matrix-matched standards are one possible control measure but become
impractical when an appropriate “clean” matrix cannot be found. Standard
addition quantitation, which involves spiking successive known quantities of a
standard into the sample and reanalyzing, is common in atomic absorption
spectroscopy and an acceptable technique to use when matrix effects are
unavoidable. Successive spiking has already been proven necessary for
perfluorinated acid quantitation by direct-injection MS analysis. Unfortunately,
standard addition quantitation can place further demands on instrument and
sample preparation time but should be used for accuracy when spike/recovery
experiments indicate a problem. Therefore, sample clean-up is desired to
eliminate matrix interference in complicated environmental and biological
samples (van Leeuwen et al., 2006; Szostek et al., 2004; Simcik et al., 2005;
van de Steene et al., 2006).
In order to rule out the matrix interference, internal standard (Isotopically
labeled chemical) is an effective tool. An important prerequisite, however, is
that analyte and internal standard have very similar characteristics, and
identical, or at least very close, retention times. Both compounds should be
affected by the co-eluted matrix to the same extent. In this respect isotopically
labeled internal standards offer the best solution. However some researchers
are still using external standard to quantify perfluoroalkyl substances by
Chapter 2-Literature Review
18
external calibration since the use of stable isotopes is generally very costly,
and commercial availability is often limited. PFOS and PFOA were detected at
ng/L level in lake water by external standard quantification (Boulanger et al.,
2004). For the determination of PFCs in complex environment samples,
external standard quantification is not applicable due to matrix interference.
2.2.6 Post extraction clean-up method for analysis of environmental
matrices
Analysis of complex environmental matrices such as sediment, sludge and
wastewater by electrospray LC/MS/MS can be significantly hampered by
ionization effects induced by co-eluting components present in the sample
extracts. Several studies have shown that matrix effects resulting from co-
eluting residual matrix components enhance or suppress electrospray
ionization of perfluorinated analytes, leading to considerable inaccuracy
(Boulanger et al., 2005; Sinclair et al., 2006; Higgins et al., 2005). Therefore,
it is very important to eliminate matrix effects when the LC/MS/MS method is
used to quantitatively determine the concentration of perfluorinated
compounds.
Post-extraction clean-up is desired to eliminate matrix interference in
complicated environmental and biological samples (Martin et al., 2004a, van
Leeuwen et al., 2006, Szostek et al., 2004, Simcik et al., 2005). Powley et al.
applied Envi-carb (graphitized carbon) and glacial acetic acid to purify the
crude extracts of biological matrices (blood, serum, live and plant tissue).
Szostek et al. (2004) used silica column to clean up fish tissues by eluting the
Chapter 2-Literature Review
19
lipids with dichloromethane, while the target compounds (PFCAs and PFSAs)
were eluted with acetone. For surface water samples, fluorous silica column
chromatography was used to clean up the SPE extracts and remove the
interfering compounds prior to LC/MS detection (Simcik et al., 2005).
Although the effect of these post-extraction clean-ups was assessed by the
improved recoveries for PFCs, matrix effect issue has not been sufficiently
studied and addressed. The assessment of matrix effect during development
and validation of LC/MS/MS method is necessary to ensure the precision,
selectivity, and sensitivity would not be compromised (Matuszewski et al.,
2003).
2.3 Occurrence of PFOS/PFOA in the environment
2.3.1 Occurrence in the surface water
PFOS and PFOA concentrations in surface waters are summarized in Table
2.3. Surface water in developed countries and industrialized areas were usually
highly polluted by PFCs, such as U.S.A (Hansen, et al., 2002; Takino, et al.,
2003), Japan (Saito, et al., 2003; 2004), Germany (Skutlarek, et al., 2006) and
coastal areas of China (So, et al., 2004). It was reported that concentrations of
PFOS and PFOA in the Great Lakes ranged from 21-70 and 27-50 ng/L,
respectively (Boulanger et al., 2004). Also, PFOS was detected in all of the
surface seawater samples collected from Tokyo Bay, at concentrations ranging
from 8 to 59 ng/L (mean of 26 ng/L) (Taniyasu et al, 2003). Several studies
reported on the occurrence of PFCAs and PFASs in surface waters in the USA,
Canada, Japan, Hong Kong, South China and Korea, both in freshwater and in
seawater. Elevated concentrations of PFOS (114±19 ng/L) and PFOA
Chapter 2-Literature Review
20
(394±128 ng/L) were detected downstream of the receiving water of the 3M
fluorochemical manufacturing facility at Decatur, USA (3M, 1999). Upstream,
the concentration of PFOS was 32±11 ng/L and there were no measurable
PFOA levels (<25 ng/L) (Hansen et al., 2002). A comprehensive study on the
occurrence of PFOS and PFOA at 78 sampling sites in Japanese rivers and
creeks demonstrated the widespread occurrence of these compounds. In
different districts geometric means between 0.97 and 21.2 ng/L were evaluated
for PFOA and between 0.89 and 5.7 ng/L for PFOS. Individual concentrations
comprised a range from 0.10 to 456 ng/L for PFOA and from 0.24 to 37.3
ng/L for PFOS. Systematic surveys revealed two highly contaminated sites, a
public-water-disposal site for PFOA and an airport for PFOS (Taniyasu et al.,
2003). Measurements in German rivers, predominantly located in the Rhine
River catchment area, demonstrated that PFCAs and PFOS also occurred in
comparable levels to those found in USA, Canada and Japan (Skutlarek et al.,
2006).
Table 2.3 Review of PFOS and PFOA concentrations in surface water (ng/L), drinking water (ng/L), wastewater (ng/L) and sludge (ng/g). Environmental
Matrix Internal Standard
LOQ Concentration Recovery Location Reference
PFOS PFOA PFOS PFOA
Surface water
0.7 13 21-70 27-50 56-176% USA Boulanger, et al., 2004 PFDoA 17 9 n.d-995,000 n.d-11,300 68-93% Canada Moody, et al., 2002
10 25 27-144 25-598 83-112% USA Hansen, et al., 2002 0.8 8 1.8-16.1 n.d-21.6 USA Sinclair et al., 2004 0.1 n.a 0.3-157 n.a 75-105% Japan Saito et al., 2003 0.1 0.7-157 n.a n.a Japan Harada et al., 2003 0.1 0.1 0.2-67,000 0.6-526 92-106% Japan Saito et al., 2004
13C-PFOA n.a n.a 0.8-1,090 10-173 70-130% USA Sinclair et al., 2006 0.05 0.05 3.4-14.5 2.4-12 69-83% Germany Weremiuk et al., 2006 0.005 0.03 n.d-99 0.85-260 94-105% China So et al., 2007 2 2 n.d-5,900 n.d-33,900 98-100% Germany Skutlarek et al., 2006
13C-PFOA 0.1 0.1 n.d-44.6 n.d-297.5 95-106% China Jin et al., 2009 13C-PFOA 4 1.1 n.d-35 n.d-19 92-106% Australia Clara et al., 2009 13C-PFOA 1.2 1.8 29-82 3.6-10.9 90-101% Switzerland Huset et al., 2008 13C-PFOA, 13C-PFOS
7.2-8.4 2.0-2.8 Italy Loos et al., 2007
PFDoA 0.5 2 4-79 113-181 78-81% Taiwan Tseng et al., 2006 Drinking
water 13C-PFOA 0.1 0.1 n.d-14.8 n.d-45.9 95-106% China Jin et al., 2009
2 2 n.d-22 n.d-519 98-100% Germany Skutlarek et al., 2006 0.1 0.3-59 n.a n.a Japan Harada et al., 2003
A C
hapter 2-Literature Review
21
Environmental Matrix
Internal Standard
LOQ Concentration Recovery Location
Reference
PFOS PFOA PFOS PFOA
Seawater
0.0008 0.0052
0.338-57.7 1.8-192
80-160%
Tokyo Bay
Yamashita et al., 2005
0.04-0.075 0.137-1.06 Offshore (Japan)
0.070-2.6 0.673-5.45 Coastal area of Hong Kong
0.023-9.68 0.243-15.3 Coastal area of China
0.039-2.53 0.239-1.135 Coastal area of Korea n.d-0.109 0.088-0.51 Sulu Sea (surface) n.d-0.024 0.076-0.117 Sulu Sea (deep)
0.008-0.113 0.16-0.42 South China Sea 0.054-0.078 0.136-0.142 Western Pacific Ocean 0.0011-0.02 0.015-0.062 Pacific Ocean
8.6-36 0.015-0.036 North Atlantic Ocean 0.037-0.073 0.1-0.439 Mid Atlantic Ocean
0.005 0.02 0.02-12 0.24-16
84-190% Hongkong and South
China Ocean So et al., 2004 0.035-755 0.22-345 South Korea
0.1 n.a 0.2-25.2 n.a 75-105% Japan Saito et al., 2003 PFDoA 0.5 2 60 270 78-81% Taiwan Tseng et al., 2006
A C
hapter 2-Literature Review
22
Environmental Matrix Internal Standard
LOQ Concentration Recovery Location
Reference PFOS PFOA PFOS PFOA
Wastewater
- - 48-454 41-674 - USA (2 STPs) 3M, 1999 - - 41-5290 67-2420 - USA (4 STPs) 3M, 1999 26 22 72-92% USA (1 STP) Boulanger et al., 2005
13C-PFOA, PFEES 0.5 0.5 1.1-400 1.7-65 88-97% USA (10 STPs) Schultz et al., 2006 13C-PFOA, PFBS 2.5 2.5 3-68 58-1,050 70-130% USA (6 STPs) Sinclair et al., 2006
13C-PFOA, PFBS - - 1.8-149 1-334 90% USA (2 STPs) Loganathan et al., 2007
n.a 0.2 0.07 4.1 5.5 85.5-91.2% China (1 STP) Zhao et al., 2007
PFDoA 0.5 2 21-79 36-170 78-81% Taiwan (2P) Tseng et al., 2006 n.a 0.5 0.5 3.4-67 49.1-548.4 40-70% Japan (2 STPs) Nozoe et al., 2008
13C-PFOA - - 14-336 14-41 76-109% Japan (1 STP) Murakami et al., 2009 Sludge 13C-PFOA, PFEES 0.9 1 14.4-2,610 n.d-13.3 71-87% USA (12 STPs) Higgins et al., 2005
13C-PFOA, PFEES 0.7-2.2 0.7-2.2 3.8-160 n.d-12 >70% USA (3 STPs) Schultz et al., 2006
13C-PFOA, PFBS 10 10 n.d-65 18-241 n.a USA (2 STPs) Sinclair et al., 2006
13C-PFOA, PFBS 2.5 2.5 n.d-990 7-219 37-89% USA (2 STPs) Loganathan et al., 2007
23
A C
hapter 2-Literature Review
Chapter 2-Literature Review
24
2.3.2 Occurrence in the drinking water
Some data are available on drinking water contamination with PFCAs and
PFASs from the USA and Japan (Harada et al., 2003; Skutlarek, et al., 2006).
The USA studies were conducted to obtain data about the presence of
fluorochemicals in drinking waters in the vicinity of fluoropolymer production
plants and where secondary manufacturers use these chemicals (3M, 2001). In
internal studies of tap water in 1984 in the vicinity of this works, DuPont
detected PFOA at concentrations of 1.5 μg/L in a store tap in Lubeck, at
concentrations of 1.0 and 1.2 μg/L in a store tap in Washington, and at
concentrations of 0.8 and 0.6 μg/L in Little Hocking (US EPA, 2002).
Recently, PFOA and PFOS contamination was reported in private
groundwater wells in Lake Elmo, Minnesota and in some of the Oakdale,
Minnesota municipal wells (Schultz et al., 2004). These contaminations
originated from several landfills, where PFCs were disposed by the 3M
Company decades ago. PFOA and PFOS were also detected in treated
drinking water and tap water in Columbus, Georgia, where several secondary
manufacturers were located, which produced non-wovens, household additives,
apparel, carpet, and home textiles. The PFOS concentrations ranged from 53
to 63 ng/L, and the PFOA concentrations ranged from 25 to 29 ng/L (3M,
2001). In a Japanese study, PFOA and PFOS had also been found in tap water
samples. The mean levels ranged from 0.12 ng/L to 40.0 ng/L for PFOA and
from <0.1 to 12.0 ng/L for PFOS (Harada et al., 2003). Occurrence of high
concentrations of PFCs in tap water indicated poor performances of current
water treatment processes to remove PFCs from surface water (Saito et al.,
2003; 2004; Skutlarek, et al., 2006).
Chapter 2-Literature Review
25
2.3.3 Occurrence in the seawater
Concentrations of PFOS and PFOA in open ocean water samples from the
Pacific and Atlantic Oceans, and from several coastal seawaters from Asian
countries, were shown in Table 2.3. PFOS and PFOA were found in 80% of
the surface seawater samples analyzed (Yamashita et al., 2005). The
similarities between PFCs composition in coastal and open ocean waters were
found in some regions, which suggests that tidal and/or water current
movements play a major role in the transport of these compounds from coastal
locations; therefore, information on oceanic currents appeared necessary to
explain the transport of PFCs from coastal waters to the open ocean.
Relatively high concentrations of PFOS and PFOA were detected in Tokyo
Bay waters. PFOA was the predominant fluorochemical detected, which was
in the range of 1800 to 192,000 pg/L, followed by PFOS (338–57,700 pg/L)
(Taniyasu et al., 2003). Concentrations of PFOS and PFOA in offshore waters
of the Pacific Ocean were approximately three orders of magnitude lower than
those in Tokyo Bay. Concentrations of all of the target fluorochemicals in
offshore waters were in the pg/L range (Yamashita et al., 2004; Yamashita et
al., 2005). In the offshore waters of Japan, PFOA was also the predominant
fluorochemical investigated, which was similar to what was observed for
coastal waters. All target PFCs in open-ocean water samples collected in the
mid-Atlantic Ocean were at pg/L levels (Table 2.3). PFOA and PFOS
concentrations were comparable to those in offshore waters collected in the
South China Sea and the Sulu Sea. These concentrations were one order of
magnitude lower than those found in offshore waters, and four orders of
Chapter 2-Literature Review
26
magnitude lower than the concentrations measured in Tokyo Bay waters (So et
al., 2004). It seems that these are the background values for remote marine
waters far from local sources.
2.3.4 Occurrence in the sludge and sediment
The binding of PFCs to sediment and sewage sludge is in general strong
and stable, which means a high potential for accumulation herein. In the so-
called multi-city study of the 3M Company, PFOS and PFOA
concentration ranges in sludge samples were: 58.9-2,980 ng/g dw (dry
weight) for PFOS and 0.297-173 ng/g dw for PFOA (3M, 2001).
Concentrations of PFOS and PFOA in sludge were in the range of 26-65
ng/g dw and 69-241 ng/g dw, respectively, in combined sludge of a STP in
New York State (Sinclair et al., 2006), 31-55 ng/g dw and <6-8.2 ng/g dw,
respectively, in activated sludge of a STP in U.S.A (Boulanger et al., 2005).
Higgins et al. (2005) reported PFOS and PFOA concentrations were in the
range of 14.4 – 2610.0 and n.d - 29.4 ng/g dw in the sludge samples of 8
STPs, respectively. Loganathan et al. (2007) observed that concentrations
of PFOS and PFOA in sludge were in the range of <2.5-77 ng/g dw and
7.0-130 ng/g dw in a STP of Kentucky, respectively.
In general sediment samples had lower levels of PFOS and PFOA than
those in sludge. Concentration of PFOS and PFOA in sediment was in the
range of n.d-3.07 ng/g dw and n.d-0.625 ng/g dw in 17 sites located in San
Francisco Bay area (Higgins et al., 2005), 0.09-0.14 ng/g dw and 0.84-1.1
ng/g dw in samples collected in Tojin river estuary, Japan (Taniyasu et al.,
2003).
Chapter 2-Literature Review
27
2.4 Fate and behavior in the sewage treatment plants
2.4.1 Occurrence in the wastewater
Schultz et al. (2006a) reported that PFOS and PFOA were ubiquitous in the
influent and effluent of ten STPs in U.S.A. In the effluents of those ten STPs,
PFOS concentrations were in the range of 1.1-130 ng/L, while PFOA
concentrations varied from 2.5-97 ng/L. It was reported that concentrations of
PFOA in effluents of the six WWTPs ranged from 58 to 1,050 ng/L, while a
much lower PFOS concentrations (3-68 ng/L) were observed in effluents of
these WWTPs (Sinclair et al., 2006). Loganathan et al. (2007) observed that
PFCs concentrations ranged from 1.8 to 22 ng/L for PFOS and from 1.0 ng/L
to 227 ng/L for PFOA in a STP of U.S.A, while higher PFOS (7.0-149 ng/L)
and PFOA (22-334 ng/L) concentrations were detected in another STP of
U.S.A. In the so-called multi-city study, elevated PFCs concentrations were
found in publicly owned treatment works effluent in the range of 48-4,980
ng/L for PFOS and 42-2,280 ng/L for PFOA (3M, 2001). The concentrations
were highly variable and differed much between the treated effluent of the so-
called supply-chain cities and the control cities.
2.4.2 Mass flow and mass change
Few articles published are available on the behavior of PFCs in STPs due to
the difficulty in determination of their concentrations in sludge and wastewater
samples. Past studies are still not enough to draw general and reliable
conclusions on PFC behavior in STPs. In most studies, only the influent and
effluent were analyzed to estimate performance of overall process. A survey in
ten US STPs showed no obvious removal of PFOS in ten plants except one
Chapter 2-Literature Review
28
STP with influent as high as 400 μg/L. PFOA in effluent of seven STPs was
increased by 10~100% of influent which contained 16~49 μg/L PFOS (Schultz
et al., 2006a). Surveys of STPs in Iowa of USA also showed no removal of
PFOS and PFOA, as well as other PFCs (Boulanger et al., 2005; Sinclair and
Kannan, 2006). Studies in a STP of Japan obtained similar results of poor or
even negative removal of PFOA and PFOS by activated sludge process
(Nozoe et al., 2006). These results implied that activated sludge process might
be ineffective to remove PFOS or PFOA, and certain amount of PFCs was
discharged from WWTPs to environment. PFCs precursors like telomer
alcohols, sulfonamides or esters were suspected to degrade to PFASs and
PFCAs during activated sludge process. Furthermore, Sinclair and Kannan
(2006) observed that mass flow of PFOS and PFOA in aqueous phase
increased significantly after secondary treatment in a sewage treatment plant
with industrial influence, while no increase in mass flow of PFOA was found
in another sewage treatment plant with no industrial influence.
Up to now, the first and only one study which estimated performances of
individual facilities in activated sludge process revealed the interesting vision
of PFC behavior inside STP (Schultz, et al., 2006b), as shown in Figure 2.2.
Mass flows of PFOA were nearly unchanged as a result of wastewater
treatment, which indicates that conventional wastewater treatment is not
effective for removal of this compound. A net increase in the mass flows for
PFOS occurred from trickling filtration and activated sludge treatment was
observed since the more highly substituted perfluorooctyl surfactants had been
biodegraded. The precursor compounds formed an additional source of PFOA
Chapter 2-Literature Review
29
and PFOS in the STP effluents. It seems that mass flow of PFOS or PFOA
either increased or remained consistent, indicating conventional activated
sludge process can not effectively remove these compounds.
2.5 Removal Technologies
PFOS and PFOA are not only metabolically but also photochemically inert,
resisting both biotic and abiotic degradation. PFOS and PFOA were
considered stable and persist in environment without natural degradations
(OECD, 2002; US EPA, 2002). Also, Schröder et al. (2003) observed that
PFOS was not degradable by activated sludge. Furthermore, past study on fate
and behavior of these pollutants in STPs implied that they can not be
effectively removed by biological treatment process. Thus, various physico-
chemical processes have been studied to remove PFOS and PFOA.
2.5.1 Advanced oxidation process
Figure 2.2 Average mass flow (mg/d) for PFOS and PFOA in sewage treatment
plant. PC-primary clarifier,TF -trickling filter, AS-activated sludge, SC-secondary clarifier, FC-final chlorination/dechlorination, TH-thickener, AD-anaerobic digester,
RAS-recycled activated sludge.
Chapter 2-Literature Review
30
Advanced oxidation processes (AOPs) involve the generation of hydroxyl
radicals in sufficient quantity to effect water purification. These common
processes include O3/H2O2, O3/UV, UV/ H2O2. UV/TiO2 process and Fenton
reagent are also effective to specific wastewater (Gottschalk, et al., 2000).
AOPs such as O3, O3/UV, O3/H2O2 and Fenton reagents were unable to
decompose PFOS in normal state, but able to degrade PFOS precursors and
partially fluorinated polymers effectively (Schröder and Meesters, 2005).
Some oxidation processes can decompose some PFCs completely in critical
conditions or coupled with catalysts. It was observed that PFOS was
completely oxidized by subcritical water oxidation, with catalyst of zerovalent
metals like iron. PFOS molecules were observed to be strongly adsorbed on
Fe3O4 sediments and further decomposed to carbon dioxide and fluorine ions
due to oxidation by molecular oxygen in subcritic water (Hori et al., 2006).
The author, however, did not consider that PFOA was able to be decomposed.
It could be due to that catalyst iron could not excite the oxidation of PFOA
and no cleavage of C-F bond occurred.
Other oxidation process has been studied to decompose environmental
contaminants, including PFCs. One of the most effective oxidation processes,
sonochemical reaction was applied to degrade PFOS and PFOA. Under
ultrasonic irradiation (20, 3 W/cm2), PFOS molecules were firstly transformed
to PFOA by releasing the sulfonate group, and the product of PFOA was
consequently degraded to short-chained PFCAs (Moriwaki, et al., 2005).
Unfortunately only partial PFOS (28%) and PFOA (63%) were decomposed
Chapter 2-Literature Review
31
although high energy was consumed. It seems that partial decomposition could
be resulted from unevenly and unfully irradiation in the reaction. Higher
energy and longer exposure irradiation may be needed to completely
decompose compounds.
2.5.2 RO/NF membrane
Membrane technology is one of the most promising technologies in water
reclamation and reuse. Reverse osmosis (RO) and nanofiltration (NF) are used
extensively in water and wastewater treatment. Both NF and RO are pressure
driven membrane processes, where an applied transmembrane pressure forces
water through the ‘pores’ and contaminants are retained due to charge and size
interactions. NF distinguishes itself from RO in that it only retains multivalent
ions, which makes it a very economic alternative where the retention of
monovalent salts is not required. RO and NF membranes are effective in
removing most organic (Kiso et al., 2001; Kimura et al., 2003; Schafer et al.,
2003) and inorganic compounds from water solutions. The main motivation to
use those processes in water and wastewater treatment is the removal of
micropollutants such as PPCPs. Because of the difficulty in effectively
removing trace organic compounds with low molecular weight from
wastewaters by conventional treatment process, membrane technology has
been investigated and applied to improve their removal. Nearly complete
retention of those micropollutants by RO and NF has been reported by many
researchers (Kimura et al., 2003; Schafer et al., 2003; Childress et al., 1996;
Kiso et al., 2001; Hu et al., 2007). Both size exclusion and adsorptive effects
appears to be essential in maintaining high retention of those micropollutants
Chapter 2-Literature Review
32
on a variety of NF and RO membranes over a range of solution conditions.
Tang et al. (2006) reported that RO membrane rejected 99% or more of the
PFOS with feed concentration ranging from 0.5 to 1500 ppm. Although the
author did not mention PFOA rejection by RO, it could be predicted that
PFOA would be significantly removed based on size exclusion mechanism.
Consequently, membrane separation appears to be an effective technology for
removal of PFCs from wastewater. However, RO or NF filtration is rarely
used in wastewater treatment because of high cost.
2.5.3 Adsorption
2.5.3.1 Activated carbon adsorption
Compared with AOPs, adsorption is a more common and widely used method
for removing organic contaminants from wastewater steams. In water and
wastewater treatment the most often used adsorbent is granular activated
carbon (GAC) and powdered activated carbon (PAC). Activated carbon
adsorption is one of the most promising methods to remove PFCs in aqueous
stream due to the effectiveness and low cost. Ochoa-Herrera et al. (2008)
reported that PFOS could be effectively removed by granular activated carbon
(GAC) and Freundlich isotherm was applicable at high and low equilibrium
concentrations. In contrast, Yu et al. (2009) studied the feasibility of using
powder activated carbon (PAC), granular activated carbon (GAC) and anion-
exchange resin (AI400) to remove PFOS and PFOA from water. It was
observed that adsorption isotherms of PFOS and PFOA fitted Langmuir
isotherms better than Freundlich isotherm. Qiu et al. (2006) also reported that
GAC was able to effectively remove PFOS and PFOA. In 4 hours, 93% PFOS
Chapter 2-Literature Review
33
and 99% PFOA in pure water at ppb level were adsorbed onto GAC.
Compared with AOPs, GAC is able to remove both PFOS and PFOA in
normal state effectively. Although PAC has similar adsorption capacity to
GAC, it has not yet found broader application in wastewater treatment as it is
not easily removed from the treated effluent.
2.5.3.2 Adsorption onto sediment and sludge
Sorption of the potassium salt of PFOS to three types of soil, sediment and
sludge from a domestic wastewater treatment plant has been measured using a
method based on OECD 106 (3M, 2003). Adsorption occurred rapidly in all
cases, and the concentrations remained fairly constant after 16 hours.
Desorption was also investigated which took place rapidly, and after 8 hours
the concentration in water did not vary significantly. Values for the sorption
and desorption coefficients were calculated and presented in Table 2.4.
Table 2.4 Sorption and desorption coefficients of PFOS from various matrices.
Matrix type Kd (L/kg) Kdes (L/kg) Mean Clay soil 18.3 47.1 32.7
Clay loam soil 9.72 15.8 12.8 Sandy loam soil 35.3 34.9 35.1 River sediment 7.42 10 8.7
STP sludge - - 1028 Note: mean values are mean of sorption and desorption coefficients. For sludge, value is the mean of the Freundlich coefficients for sorption and desorption, as direct values are only reported as limit values.
The occurrence of PFCs in sludge from STPs indicates an adsorption of these
compounds to the activated sludge during the treatment process. It was
reported that the measured log Koc value for PFOS and PFOA are 2.57 and
2.06, which all are in the range of 2.57-3.1 [log(L/kg)] for PFOS (3M, 2002)
Chapter 2-Literature Review
34
and 1.9-2.17 [log(L/kg)] for PFOA (Dupont, 2003). In activated sludge
treatment process, 100-400 gSS/m3 of sludge is usually produced. Therefore
removal by sorption onto sludge is generally relevant (>10%) only for
compounds with a Kd>300 L/kg. According to the reported data, Kd ranged
from 371 to 1,258 L/kg for PFOS and 79 to 148 L/kg for PFOA, indicating
<35% PFOS and <6% PFOA were adsorbed onto sludge. Therefore, it can be
expected that PFOS and PFOA did not adsorb significantly onto sludge and
sorption was not an important removal process in conventional wastewater
treatment system, which were proven by Figure 2.2.
2.5.4 Membrane biological reactor (MBR)
2.5.4.1 Introduction
Since research on membrane bioreactor (MBR) technology began over 30
years ago, several generations of MBR systems have evolved (Gander et al.,
2000). Up to this date, MBR systems have mostly been used to treat industrial
wastewater, domestic wastewater and specific municipal wastewater, where a
small footprint, water reuse, or stringent discharge standards were required. It
is expected, however, that MBR systems will increase in capacity and broaden
in application area due to future, more stringent regulations and water reuse
initiatives.
In the early 1990s, MBR installations were mostly constructed in external
configuration, in which case the membrane modules are outside the bioreactor
and biomass is re-circulated through a filtration loop. This limited wider
application in treatment of municipal wastewater in North America because of
Chapter 2-Literature Review
35
high power consumption. After the mid 1990s, with the development of
submerged MBR system, MBR applications in municipal wastewater extended
widely. In the past 10 years, MBR technology has been of increased interest
both for municipal and industrial wastewater treatment in North America.
2.5.4.2 Configuration and application
MBR systems are characterized by two configurations: submerged (immersed
or integrated) MBRs and external (recirculated or side-stream) MBRs. Due to
the absence of a high-flow recirculation pump, submerged MBRs consume
much lower power than external MBRs. This was the primary driver for
propelling submerged MBRs into the purview of large-scale wastewater
treatment plants in a few dozens of countries around the world. External
MBRs were considered to be more suitable for wastewater streams
characterized by high temperature, high organic strength, extreme pH, high
toxicity and low filterability. In the case of an external MBR system, the
membrane device is independent of the bioreactor. Feedwater enters the
Figure 2.3 Configuration of MBR systems: (a) Side-stream MBR, (b) Suctioned- filtration submerged MBR, and (c) Gravitational-filtration
submerged MBR.
Chapter 2-Literature Review
36
bioreactor where organic matters are biodegraded by biomass. The mixed
liquor in the bioreactor is then pumped around a recirculation loop containing
a membrane unit where permeate is discharged and the retentate is returned
back to the bioreactor. The transmembrane pressure (TMP) and crossflow
velocity of the membrane device are both generated from a pump (Hillis, 2000;
Kim et al., 2001).
2.5.4.3 Technology benefits and problems
The technical benefits of MBR include high quality effluent, small footprint,
short start-up time and low operating and maintenance manpower requirement.
Of these, the prime ones are the excellent effluent quality, easy management,
high biomass concentration, and less sludge production (Xing et al., 2000;
Fleischer et al., 2005). MBR systems can provide high-quality effluents which
are free of solids and bacteria and can be directly reused for municipal
watering, toilet flushing, and car washing (Huang et al., 2001; Xing et al.,
2001). Since suspended solids are completely retained by membranes in MBR
systems, quality of effluent would no more be affected by the settling problem
caused by poor flocculation of microorganisms or proliferation of filamentous
bacteria (Bai and Leow, 2002). Consequently, it is much easier to operate and
maintain MBR systems as compared to conventional activated sludge systems.
The elimination of secondary settlement stage allows the use of high activated
sludge concentration in a small volume tank. For example, some authors have
investigated MBR system with MLSS ranging between 10,000 and 23,000
mg/L (Dijk and Roncken, 1997; Churchouse et al., 1998). Bouhabila et al.
(1998) studied critical fluxes for the operation of the MBR with MLSS
Chapter 2-Literature Review
37
concentration of up to 15,000 mg/L. High biomass concentration in the reactor
enabled MBR to produce high quality effluent at short hydraulic retention time
(Gunder, 2001). Furthermore, MBR systems can be operated at low organic
loading rates with the combination of high biomass concentrations and the
complete retention of biosolids. These characteristics promote the
development of slow growth bacteria, such as nitrifiers, and result in lower
sludge production as compared with conventional aerobic treatment processes
(Chang et al., 2002).
Despite the many advantages of MBR systems, it has been shown that
membrane fouling is the most serious problem affecting system performance
(Visvanathan et al., 2000; Le-Clech et al., 2003; Kim et al., 2001). It is
reported that the nature and extent of fouling are strongly influenced by three
factors: characteristics of mixed liquor, operating conditions, and membrane
properties (Chang and Lee, 1998; Shimizu et al., 1996; Bouhabila et al., 2001;
Ng et al., 2005). It has been shown that membrane fouling is the most serious
problem affecting system performance in some recent reviews covering
membrane applications to bioreactors (Visvanathan et al., 2000; Kim et al.,
2001). Though numerous investigations of membrane fouling have been
published, the diverse range of operating conditions and feedwater matrices
employed, and the limited information reported in most studies on the mixed
liquor composition, have made it difficult to establish any generic behavior
with respect to membrane fouling in MBR systems (Chang et al., 2002).
However, it is evident that the nature and extent of fouling are strongly
influenced by characteristics of mixed liquor, operating conditions, and
Chapter 2-Literature Review
38
membrane properties (Chang and Lee, 1998; Chang et al., 1999; Bouhabila et
al., 2001).
2.5.4.4 Hybrid PAC-MBR system
Membrane fouling in MBR results from the interaction between membrane
material and components in the activated sludge mixture. The latter includes
substrate components, cells, cell debris, and microbial metabolites such as
extracellular polymeric substances (EPS). Accordingly, the floc structure,
particle size distribution and EPS contents of activated sludge can all
contribute to membrane fouling. To prevent or mitigate membrane fouling in
MBR, various techniques have been adopted such as low-flux operation, high
shear slug flow aeration in a submerged configuration, periodical air or
permeate backflushing and intermittent suction operation. In recent years, the
addition of PAC to a MBR (referred to as hybrid PAC-MBR in this study) has
been applied for wastewater treatment. A few studies of hybrid PAC-MBR
process have been reported and results showed that the addition of PAC
improved the performance of MBR system (Munz et al., 2007; Ng et al., 2006;
Satyawali et al., 2009). On the one hand, some studies observed that
membrane flux was enhanced since PAC decreased the compressibility of
sludge flocs and increased the porosity of cake layer by acting as supporting
medium (Kim et al., 1998; Aquino et al., 2006). Li et al. (2005) further
identified that PAC addition significantly decreased membrane total resistance
by 44% for long term operation of submersible membrane bioreactor, which
resulted in extension of operation time by 1.8 times as compared to normal
MBR system. On the other hand, a few studies found that adding PAC into
Chapter 2-Literature Review
39
MBR could not only increase porosity of cake layer but also reduce the
accumulation of foulants on the membrane surface and change the
composition and permeability of the cake layer (Kim et al., 1998; Ng et al.,
2006). Ng et al. (2006) even pointed out that the primary role of the PAC was
to provide adsorptive removal of foulants rather than providing supporting
medium. Other benefits of PAC addition include increase in the removal of
organics, reduction in the impact of organic shocking loadings and increase in
the resistance to toxic substances (Aktas et al., 2007; Lesage et al., 2007).
Therefore, it is evident that hybrid PAC-MBR system shows better
performance than normal MBR system in terms of effluent quality, stability
and fouling rate due to PAC effects on the foulants, sludge flocs and
membrane filtration.
Hybrid PAC-MBR could be an effective technique to remove micropollutants
in wastewater since the bioreactor combines three individual process
operations, namely physical adsorption, biological degradation and membrane
filtration in a single unit. A few studies have been conducted to investigate the
removal mechanism of micropollutants in hybrid PAC-MBR. Dosoretz et al.
(2004) reported that an almost complete removal of phenanthrene was
observed in hybrid PAC-MBR due to the simultaneous adsorption and
biodegradation. Baumgarten et al. (2007) also found that combination of MBR
and PAC could effectively remove some micropollutants, such as antibiotics.
However, little information is available on the removal of PFCs in hybrid
PAC-MBR till now.
Chapter 2-Literature Review
40
2.6 Research statement
PFOS and PFOA, regarded as the terminal breakdown end-products of PFCs,
have been detected in the air, surface waters, wastewaters, drinking waters,
groundwaters, coastal waters, sediments as well as various biological tissues
all over the world. Their ubiquious presence in the environment could be due
to the worldwide use of PFCs and the high mobility of their precursors in the
air. A few studies have been conducted to identify the contamination source of
PFCs in the environment. Some researchers observed that effluents form the
STPs are the most important PFCs sources for the aquatic ecosystems (Sinclair
and Kannan, 2006; Loganathan et al., 2007). Zushi et al. (2008), however,
reported that loads of PFCs in rain runoff were about 2-11 folds greater than
those in STP effluents that were discharged into a river. It indicates that
nonpoint source of PFCs could be the most important source for the river
studied. In addition, Yamashita et al. (2004) reported that application of PFC-
containing products could also be an important source of aquatic environment.
It seems that effluents from STPs, nonpoint source from rain runoff and
application of PFC-containing products might be important sources and
determine the PFCs concentration levels in the aquatic environment. However,
these studies failed to prove that there are no other significant PFCs sources
such as atmospheric deposition or precipitation for the aquatic environment.
Kallenborn et al. (2004) and Scott et al. (2006) both reported relatively high
PFOA concentrations in the rainwater samples from Europe and North
America, which could be important PFCs sources. Therefore, further research
is needed to identify possible contamination sources and transportion
pathways of PFCs in environment. Furthermore, seasonal variations in the
Chapter 2-Literature Review
41
PFCs concentraions were investigated. So et al. (2004) observed PFCs
concentrations in the winter were higher than in the summer in coastal waters
of China. In wastewater of STPs, Loganathan et al. (2007) found that mass
flow of PFCs were higher in winter than in summer. The authors suggest that
there were less rain in winter than in summer, which resulted in dilution effect
on the coastal waters or wastewaters. However, limited data is available on the
comparison of PFCs concentrations between dry season and wet season in the
aquatic environment. Singapore is an island country and also a true city-state
with a tropical rainforest climate and no distinctive seasons. Especially its
climate is characterized by uniform temperature, pressure and abundant
rainfall in wet monsoon season (November and December). In a such an
unique island city, it could be an ideal place to identify seasonal variations of
PFCs concentrations between dry seasons and wet seasons by excluding other
factors, such as temperature and atmospheric pressure variation. To the best of
our knowledge, this study is the first study to identify the seasonal varitions of
PFCs in aquatic environment between dry and wet seasons.
Solid phase extraction (SPE) followed by high performance liquid
chromatography coupled with tandem mass spectrometry (HPLC/MS/MS) are
widely applied to quantitatively identify PFOS and PFOA. However, analysis
of complex environmental matrices such as sediment, sludge and wastewater
by electrospray LC/MS/MS can be significantly hampered by ionization
effects induced by co-eluting components present in the sample extracts.
Several studies have shown that matrix effects resulting from co-eluting
residual matrix components enhance or suppress electrospray ionization of
Chapter 2-Literature Review
42
perfluorinated analytes, leading to considerable inaccuracy (Boulanger et al.,
2005; Sinclair et al., 2006; Higgins et al., 2005). Therefore, post-extraction
clean-up is desired to eliminate matrix interference in complicated
environmental and biological samples (van Leeuwen et al., 2006; Szostek et a.,
2004; Simcik et al., 2005; van de Steene et al., 2006). A few studies applied
different methods to remove the interfering compounds prior to LC/MS
detection. For examples, Powley et al. (2005) applied Envi-carb (graphitized
carbon) and glacial acetic acid to purify the crude extracts of biological
matrices (blood, serum, live and plant tissue). Szostek et al. (2004) used silica
column to clean up fish tissues by eluting the lipids with dichloromethane. For
surface water samples, fluorous silica column chromatography was used to
clean up the SPE extracts and remove the interfering compounds prior to
LC/MS detection (Simcik et al., 2005). However, the above post-extraction
methods may not be applied to wastewater and sludge samples collected form
STPs, in which stronger matrix effect was observed in comparison with
surface water (Boulanger et al., 2005; Sinclair et al., 2006). Thus, it is
necessary to develop a novel post-extraction clean-up for different
environmental matrices, including wastewater and sludge samples. In addition,
limited data is available on the quantitive estimation of matrix effect and effect
of post-extraction clean-up on different environmental matrices. In order to
ensure the precision, selectivity, and sensitivity of extraction method, there is
also a need to quantitively investigate matrix effect during development and
validation of LC/MS/MS method.
Chapter 2-Literature Review
43
Due to widespread usage of PFCs in industrial and commercial applications,
various contamination levels were reported in the influent and effluent of
municipal STPs in Iowa City (Boulanger et al., 2005), in Kentucky and
Georgia (Loganathan et al., 2007), in 10 national wide municipal STPs in
U.S.A (Schultz et al., 2006a) and in the effluent of 6 U.S.A cities (Sinclair and
Kannan, 2006). It is evident that the discharge of municipal wastewater
effluent is one of the major routes for introducing PFOS and PFOA that are
used in domestic, commercial and industrial settings into aquatic environment.
A few researchers studied the fate and behavior of PFCs in STPs. Sinclair and
Kannan (2006) observed that mass flow of PFOS and PFOA in aqueous phase
increased significantly after secondary treatment in a STP with industrial
influence, while no increase in mass flow of PFOA was found in another STP
with no industrial influence. Furthermore, Schultz et al. (2006b) identified the
fate and behavior of these two compounds in both aqueous phase and solid
phase (sludge) during each step of municipal wastewater treatment plant. It
was observed that mass flow of PFOS or PFOA either increased or remained
consistent, indicating conventional activated sludge process can not effectively
remove these compounds. Unfortunately, these investigations were conducted
at different STPs with different influents. Different influent of STP would
significantly affect the behavior pattern of PFOS or PFOA since their
precursors in the influent could biodegraded to PFOS or PFOA in the activated
treatment processes. Therefore, it is desired to investigate behavior of PFCs in
various activated sludge treatment processes which receive the same raw
sewage. In addition, sludge retention time (SRT) could also be an important
factor affecting the fate of PFOS and PFOA in STP. Clara et al. (2005) found
Chapter 2-Literature Review
44
that the degradation of the micropollutants, such as endocrine disrupting
compounds and pharmaceuticals, was dependent on the SRT in the activated
sludge process since the SRT determines the enrichment of the microorganism
that is able to degrade the micropollutants. Therefore, behavior pattern of
PFCs may be different in the conventional activated sludge process operated
with different SRT. However, no data is available about the effect of SRT on
the behavior pattern of PFOS and PFOA in the activated sludge process. It is
desired to study the fate and behavior of PFOS and PFOA in full-scale STP
comprising of different activated sludge treatment processes with different
SRT, which treat the same raw sewage.
Although there is no maximum allowable concentration of PFCs in the
discharge of STPs, PFOS and PFOA, candidates for persistent organic
pollutants (POPs), are reported to have adverse effect on the human health.
Since PFOS and PFOA can not be effectively removed by conventional STPs
and drinking water treatment plants, it is urgent to develop a new technology
to remove these compounds effectively at low cost for the wastewater
treatment. The hybrid PAC-MBR technology integrates adsorption and
biodegradation of organic matters with membrane filtration in one unit, which
has been proved to be a simple and highly efficient way to remove compounds
in wastewater. In particular, PAC addition increases the removal of organic
matters with low molecular weight by adsorption; it also serves as a
supporting medium for attached bacterial growth (Kim et al., 1998). Even
though MBR may not be able to significantly remove PFOS and PFOA due to
similar biodegradation and adsorption behavior in activated sludge system,
Chapter 2-Literature Review
45
combination of MBR and PAC technologies could effectively remove these
compounds while adsorption onto PAC occurs. It was reported that PFCs were
effectively removed by adsorption onto the activated carbon at high and low
equilibrium concentrations (Ochoa-Herrera et al., 2008; Qiu et al., 2006).
However, these studies were conducted in the buffer solution without the
presence of dissolved organic matters (DOMs). In STPs, effluent from
biological wastewater treatment contains complex and heterogeneous soluble
organic matters, which are so called effluent organic matters (EfOM). The
composition of EfOM is a combination of those of natural organic matter
(NOM), soluble microbial products (SMPs), and trace harmful chemicals. It
was observed that PAC adsorption capacity would be reduced dramatically
when EfOM was present during activated carbon treatment of wastewater
containing micropollutants (Newcombe et al., 2002; Matsui et al., 2003). The
direct competition for the adsorption sites was found to be the most likely
competition between EfOM and target micropollutants (Newcomber et al.
2002; Kilduff et al. 1998; Matsui et al., 2003). However, limited data is
available on the effect of EfOM on the PFCs adsorption to the activated
carbon. Therefore, study on the EfOMs effect on PFCs adsorption is needed
for the better understanding of competitive effects caused by the presence of
EfOM. In addition, it is essential to study the adsorption capacity and kinetics
of PFCs onto PAC to understand the removal mechanism in hybrid PAC-MBR
process.
It is evident that operation parameters can affect PFCs removal in the hybrid
PAC-MBR system. On the one hand, SRT, a commonly used parameter for
Chapter 2-Literature Review
46
biological process design and operation, could be an important factor affecting
the removal of PFOS and PFOA. It was reported that SMPs was the dominant
DOMs in the supernatant and effluent of MBR (Lee et al., 2003; Barker et al.
1999). At different SRT, composition of SMPs could be different, which may
affect the PAC adsorption. For example, Liang et al. (2007) observed that
SMPs in MBR was significantly reduced as SRT was increased, indicating
reduced adsorption competition from DOMs. Thus, PAC adsorption capacity
may be significantly affected by characteristics of SMPs, which can be
influenced by SRT of MBR. However, no study is available on the adsorption
capacity of PAC in MBR operated at different SRT. Therefore, it is necessary
to study the effect of SRT on PFCs adsorption on the PAC. On the other hand,
it is generally accepted that PAC addition in the MBR can enhance membrane
flux and decrease fouling rate. Most studies focused on the effect of PAC on
membrane filtration and fouling. However, little data is available on the effect
of PAC dosage on micropollutant removal. It is necessary to explore the
optimum PAC dosage in order to achieve the desired PFCs’ removal in hybrid
PAC-MBR process.
In summary, this study aims to identify possible contamination sources and
transportion pathways of PFCs and seasonal variation of PFOS and PFOA in
the aquatic and oceanic environment; to develop a novel post extraction clean-
up method for the determination of PFOS and PFOA in environmental
matrices; to study the fate and behavior of PFOS and PFOA in full-scale STP
comprising of different activated sludge treatment processes with different
SRT; to study the EfOMs effect on PFCs adsorption for the better
Chapter 2-Literature Review
47
understanding of competitive effects at the presence of EfOM; and to study the
mechanism of PFCs removal by hybrid PAC-MBR process running with
different SRT and PAC dosage.
Chapter 3-Materials and Methods
48
CHAPTER 3 MATERIALS AND METHODS
3.1 Chemicals, materials and reagents
3.1.1 Chemicals and reagents
Standards of perfluorooctane sulfonate potassium salt (PFOS, ≥98%),
perfluorooctanoate acid (PFOA, 96%), methanol (99.8%) and ammonium
acetate (97%) were purchased from Sigma-Adrich (Singapore). Internal
standard perfluoro (2-ethoxyethane) sulfonic acid (PFEES, 97%) and
perfluorododecanoic acid (PFDoA, 95%) was purchased from Oakwood
Research Chemicals (West Columbia, USA) and Sigma-Adrich (Singapore),
respectively. Oasis HLB (500mg, 6 cc) and Sep-Pak plus silica (1g) solid
phase extraction (SPE) cartridges were from Waters (Milford, USA). Nylon
syringe filter (0.2 μm) was from Millipore (USA).Stock solutions were
prepared in methanol at a concentration of 1 mg/mL. From these stock
solutions working solutions were prepared by diluting with 70:30 (v/v)
methanol/aqueous ammonium hydroxide (0.01%) solution. Stock solutions
and working solutions were stored at -20 o
C.
3.1.2 Materials
YM (Millipore, USA) series UF membranes with nominal molecular weight
cutoffs of 1, 10 and 30 kDa were used in this study. High density polyethylene
(HDPE) bottles were used for all adsorption experiments. Supelite™ XAD-8
resin (SUPELCO, U.S.A.), AG MP-50 cation exchange resin (Bio Rad,
Chapter 3-Materials and Methods
49
U.S.A.) and Amberlite IRA-96 anion exchange resin (Rohm and Haas, France)
were used for DOM fractionation.
PAC obtained from Sigma-Adrich (Singapore) was used as adsorbents for this
study. The characteristics of PAC used are given in Table 3.1. Prior to
adsorption experiments, PAC was soaked in Mill-Q water for 24 h to release
potentially adsorbed organics, dried in the oven at 103 oC for 48 h and then
stored in a desiccator. PAC and activated sludge were used as adsorbents for
the adsorption experiment. The activated sludge was collected from the
laboratory-scale MBR immediately before adsorption experiment and sodium
azide (100 mg/L), a respiratory inhibitor, was added to prevent microbial
metabolism. Table 3.2 shows the characteristics of the MF hollow fiber
membrane.
Table 3.1 Characteristics of powdered activated carbon (PAC).
Specification Description/value
Type Wood based
Ash content ≤ 5%
Particle size (nominal) ≤40 μm (75%)
BET Surface area 763.8 m2/g
Mean pore diameter 12.724 Å
Pore volume 0.486 cm3/g
Chapter 3-Materials and Methods
50
Table 3.2 characteristics of the MF hollow fiber membrane.
Properties of Membrane Unit Description/value
Material - PVDF
Type - Hydrophobic
Total surface area m2 0.2
Permeability at 20oC L·m-2·h-1·bar-1 50
Mean pore diameter µm 0.2
MWCO kDa 300 kDa
Internal diameter mm 0.46
External diameter mm 0.95
Flux L/m2 ·h 10
3.2 Water sample collection and preparation
3.2.1 Water sample collection
Water samples from reservoirs, rivers, canals and effluents of STPs and
coastal waters around the island were collected in October 2006, December
2006, March 2007, September 2007 and December 2007. Due to the tropical
climate in Singapore, November and December are the wet monsoon season,
while other months are relatively dry seasons in Singapore (Ooi and Chia,
1974). Monthly rainfall totals are 86, 788, 169, 197 and 413 mm for October
2006, December 2006, March 2007, September 2007 and December 2007,
respectively (National Environment Agency, 2008). For the sampling in
December, it was performed during the second week while it was raining.
During sampling, outside temperature ranged from 21 to 32 oC.
Chapter 3-Materials and Methods
51
Figure 3.1 Sampling locations for reservoir waters, river/canal waters, effluents of WWTPs, coastal waters and location for outfalls of WWTPs.
Western area of Singapore is a heavily industrial influenced area, while
eastern and middle areas are predominantly commercial and light industrial
areas. Some samples were collected from urban regions, while others were
from industrial influenced districts (sampling sites illustrated in Figure 3.1).
Grab wastewater samples were collected from the secondary effluent of
conventional activated sludge processes (CAS) in the six WWTPs in
Singapore. Characteristics of WWTPs are listed in Table 3.3. According to the
composition of the influent, WWTPs are categorized to domestic WWTPs and
industrial WWTPs. Industrial WWTPs include W2 and W5, while the rest are
domestic WWTPs. For all WWTPs, raw sewage is collected by separate sewer
system which receives no stormwater runoff. Outfalls of W1 and W2, OW1
and OW2, are 0.4 km and 1 km away from the shore, by which their effluents
are partially discharged to the sea. Samples were collected and stored in high
Chapter 3-Materials and Methods
52
density polyethylene bottles. Sample bottles were kept on ice and brought
back to the laboratory within 6 h of collection.
Table 3.3 Wastewater treatment plants characteristics.
WWTP Treatment process
Flow (m3/d) Wastewater treated
W1 CAS, MBR 232000 90% domestic/commercial, 10% industrial
W2 CAS, MBR 151000 48% domestic/commercial, 52% industrial
W3 CAS, MBR 247000 90% domestic/commercial, 10% industrial
W4 CAS,
MBR, LTM 361000 95% domestic/commercial, 5% industrial
W5 CAS 205000 40% domestic, 60% industrial
W6 CAS 282000 95% domestic/commercial, 5-10% industrial
CAS: conventional activated sludge system; MBR: membrane biological reactor; LTM: liquid treatment module (an activated sludge process operated
with short sludge retention time).
3.2.2 Water sample preparation (Basic SPE extraction)
Immediately after the collection, water samples were filtrated by GF/B glass
filter (Whatman, USA) and stored at -20 oC until extraction. In most cases,
samples were extracted within 24 h after collection. Extraction procedure used
was similar to that described previously (So et al., 2004) with minor
modifications. Briefly, analytes were extracted using 500-mg hydrophilic-
lipophilic balance (500mg, 6 cc HLB) cartridges, which were sequentially
preconditioned with 5 mL of methanol and 10 mL of Milli-Q water. 1 L of
filtrated water sample was loaded on cartridges at a fixed flowrate of 10
mL/min, after which the cartridges were rinsed with 5 mL of 40% methanol in
Milli-Q water. Then the cartridges were completely dried by air on the
manifold using the vacuum pump. Finally, the cartridges were eluted with 2×2
Chapter 3-Materials and Methods
53
mL methanol into a polypropylene tube. The resulting extract was reduced to
1 mL under a gentle stream of pure nitrogen gas. Final extracts were filtrated
by 0.2 μm nylon syringe filters to remove fine particles and then transferred to
polypropylene vial for analysis.
3.3 Experiment on development of clean-up method for wastewater and
sludge samples
3.3.1 Silica cartridge clean-up procedure
An additional solid phase cartridge, Sep-Pak plus silica (1g) cartridge, was
applied to further remove the interfering matrix in the extracts from basic HLB
SPE procedure. After eluting the HLB cartridge with 2×2 mL methanol,
extracts were diluted with 6 mL of dichloromethane. The silica cartridge was
preconditioned with 5 mL of dichloromethane/methanol (60:40, v/v). Then the
diluted extracts were loaded onto the silica cartridge at 10 mL/min and
collected into a polypropylene tube. The extract was evaporated to dryness
under a gentle nitrogen stream and reconstituted in 0.5 mL of
methanol/aqueous ammonium hydroxide (0.01%) solution (70:30, v/v).
Particles that appeared in the final extracts were removed by 0.2 µm nylon
syringe filter.
3.3.2 Application of clean-up method to sludge samples
Sludge samples were extracted according to an established method (Higgins et
al., 2005) and further cleaned by silica clean-up method. 100 mg of freeze
dried sludge was transferred to a 50 mL polypropylene vial and washed by 7.5
mL of 1% acetic acid solution. Each vial was vortexed, placed in the 60 oC
Chapter 3-Materials and Methods
54
sonication bath, and sonicated for 15 min. After sonication, the vials were
centrifuged at 3000 rpm for 10 min, and the acetic acid solution was decanted
into a second 50 mL polypropylene vial. An aliquot of the methanol/acetic
acid extraction solvent mixture (1.7 mL) was then added to the original vial
and the vial was again vortexed and sonicated for 15 min at 60 o
The extracts of both wastewater and sludge sample were filtrated by 0.2 μm
nylon syringe filters to remove fine particles and stored at -20
C before
centrifuging and decanting the extract. This process of acetic acid washing
followed by methanol/acetic acid extraction was repeated 3 times, and a final
7.5 mL of acetic acid wash was performed. For each sludge sample, all washes
and extracts were combined. The resulting extracts of 35.1 mL were loaded
onto Oasis HLB cartridge (500 mg) at 1-2 mL/min, which was preconditioned
by 5 mL of methanol and 10 mL of 1% acetic acid. Then the cartridge was
eluted by 2×2 mL methanol after it was rinsed by 10 mL of Mill-Q water.
Instead of reducing matrix interference by dilution, the silica clean-up method
was used to remove the matrix effect. After silica cartridge clean-up, the
eluent was concentrated under nitrogen gas to complete dryness. The final
extracts were kept to 1 mL by 70:30 (v/v) methanol/aqueous ammonium
hydroxide (0.01%) solution.
o
C until analysis.
Immediately prior to LC/MS/MS analysis, the 0.5 mL aliquot of extract was
transferred to an autosampler vial for analysis.
3.3.3 Evaluation of matrix effect and recoveries
Matrix effect was evaluated in correspondence to the strategy applied by
Matuszewski et al. (2003) with some modification. As analytes occurred
Chapter 3-Materials and Methods
55
ubiquitously in the aquatic environment and were observed at the level of tens
ng/L in the collected raw sewage (data not shown), the principle of standard
addition was employed to modify the applied strategy. MS/MS peak areas of
known amount of working standards is defined as A, while those of raw
sewage extract as B. MS/MS peak areas of raw sewage extracts spiked with
the same amount of analytes after and before SPE extraction is defined as C
and D, respectively. The matrix effect (ME) is calculated by comparing
MS/MS area for known amount of analytes spiked after extraction of raw
sewage (C-B) with those of the same amount of working standards (A). The
comparison of MS/MS area for known amount of analytes spiked before
extraction of raw sewage (D-B) with those of same amount of working
standards (A) is defined as recovery efficiency (RE). ME and RE are
calculated as followed:
ME%=(C-B)/A×100 (eq. 3.1)
RE%=(D-B)/A×100 (eq. 3.2)
The absence of absolute matrix effect is denoted by a ME% value of 100%,
which implies that the response of standards is same as that of extracts. There
is matrix suppression if ME% is <100%, while ME% of >100% indicates
matrix enhancement. In case of calculations associated with internal standards,
area ratios (area of analyte/area of internal standard) were applied instead.
3.4 Wastewater and sludge sample collection and preparation
3.4.1 Wastewater and sludge sample collection
Samples were collected from two STPs, plant A and B in Singapore. Plant A,
the largest STP in Singapore, treats 361 MLD (million litre per day), which
Chapter 3-Materials and Methods
56
consists of 95% domestic wastewater and 5% industrial and commercial
wastewater. The plant comprises of conventional activated sludge process
lines (CAS1) in parallel with liquid treatment module (LTM) and MBR
(Figure 3.2). CAS1 is operated with a SRT of ~15 days and hydraulic retention
time (HRT) of 8 h, while LTM with SRT of ~3.5 d and HRT of 6 h,
respectively. For the MBR, SRT and HRT are ~20 d and 6 h. Plant B, which
only has conventional activated sludge process (CAS2), treats 205 MLD,
among which 60% is industrial wastewater and 40% is domestic wastewater. It
is run with a SRT of ~12 d and HRT of 10 h. Both CAS1 and CAS2 have
similar treatment process, mainly consisting of primary clarifier (PC), aeration
tank (AT) and secondary clarifier (SC). The produced sludge together with the
solids obtained from the primary clarifier is disposed by thickening centrifuges,
followed by anaerobic digester and dewatering centrifuges.
Screen Grit Removal
Primary Clarifier
Aeration Tank
Secondary Clarifier
Thickening Centrifuge
Anaerobic Digester
Dewatering Centrifuge
Raw Sewage
Inorganic Solids
Primary Sludge Secondary
SludgeReturned Activated Sludge
Effluent
Dewatered Sludge
Primary Clarifier
Liquid Treatment Module
Seconday Clarifier
Primary Clarifier MBR
Effluent
Effluent
CAS 1 (CAS2 in Plant B)
LTM
MBR
267 MLD (205 MLD)*
71 MLD
23 MLD
Grab sample
Returned Liquid
Returned Activated Sludge
Influent Primary effluent
Aeration effluent
Secondary effluent
Thickened sludge
Digestered sludge
LTM-in LTM-pc LTM-eff
MBR-pc MBR-eff
MBR-sup
* denote the flowrate of CAS2 (plant B)
Figure 3.2 Flow scheme of the sewage treatment plants A (CAS1, LTM and MBR) and B (CAS2).
Chapter 3-Materials and Methods
57
Sampling campaigns were conducted in the October 2006, December 2006,
March 2007, September 2007 and December 2007 in plants A and B.
December is the wet season, which has much more precipitations than other
months, and other months are dry seasons in Singapore. For the sampling in
December, it was performed after it had been raining for a week and when it
was raining. For the sampling in other than months, there was no significant
rainfall in the period. During the sampling campaigns, outside temperature
ranged from 21 to 32 o
C. Grab aqueous samples were taken in plants A and B
and sampling points were shown in Figure 3.2. Grab samples of primary,
activated, secondary, returned activated, thickened and anaerobically digested
sludges were also collected.
3.4.2 Wastewater and sludge sample preparation
Wastewater and sludge samples were extracted according to the developed
method (Silica clean-up method) described in section 3.3.1 and 3.3.2. The
extracts of both wastewater and sludge samples were filtered by 0.2 μm nylon
syringe filter to remove fine particles and then stored at -20 o
C until analysis.
Immediately prior to LC/MS/MS analysis, the 0.5 mL aliquot of extracts was
transferred to an autosampler vial and 50 μL of 50 μg/L aqueous internal
standard mixture containing perfluoro (2-ethoxyethane) sulfonic acid (PFEES)
and perfluorododecanoic acid (PFDoA) was added.
Chapter 3-Materials and Methods
58
3.5 PAC-MBR experimental setup and operation
3.5.1 MBR and PAC-MBR configuration
Experiments were performed in four identical lab-scale submerged MBR and
PAC-MBR systems as illustrated in Figure 3.3. Each MBR consisted of
regular tank with an operating volume of 16 L and a microfiltration (MF)
membrane module submerged in the tank. The membrane module was made
of polyvinylidene fluoride (PVDF) hollow fibre membrane with a pore size of
0.2 µm and filtration area of 0.4 m2, which was mounted between two baffle
plates located above an air diffuser in the MBR. Two baffle plates were
mounted above the air diffuser to optimize the contact between air bubbles and
the membrane surface. Compressed air (36 L/h) was supplied through the air
diffuser to provide good mixing of the activated sludge and cross flow action
for effective scouring of the membrane surface. The membrane flux was kept
constant at 10 L/m2·h and followed a suction cycle of 8 min on and 2 min off.
Two water level sensors were installed at the high and low water level
respectively to maintain a constant water level in the bioreactor. Both the
bioreactor and the storage tank were initially filled with the synthetic
wastewater. The storage tank with an effective volume of 200 L was reloaded
everyday with the fresh wastewater to ensure the continuous supply to the
MBR over the entire experimental period. To minimize the variation of
wastewater characteristics, the storage tank was thoroughly cleaned every two
days to reduce the growth of microorganisms.
Chapter 3-Materials and Methods
59
Magnetic stirrer
Synthetic wastewater storage tank
Feeding pump
PAC dosing pump
PAC tank
Suction pump
Pressure guage
Flow meter
Air
MF membrane module
Air diffuser
Baffles
Figure 3.3 Schematic diagram of lab-scale PAC-MBR system. (For MBR system, no PAC dosage system)
3.5.2 Synthetic wastewater and operational conditions
The composition of the synthetic wastewater used in this study is listed in
Table 3.4. The carbon source was mainly from sodium acetate which is simple
and readily biodegradable. The influent COD concentration was 600±20 mg/L
with the ratio of COD: N: P maintained at 100: 10: 1.
PAC dosage system
Chapter 3-Materials and Methods
60
Table 3.4 Composition and concentration of synthetic wastewater.
Components Molecular weight (Da) Concentration (mg/L)
CH3COONa 82 768.75
(NH4)2SO4 132.1 284
KH2PO4 136.1 26
CaCl2 ·2H2O 147 0.368
MgSO4 ·7H2O 246.5 5.07
MnCl2·4H2O 197.9 0.275
ZnSO4·7H2O 287.5 0.44
FeCl3 162.2 1.45
CuSO4·5H2O 249.7 0.391
CoCl2·6H2O 237.9 0.42
Na2MoO4·2H2O 242 1.26
Yeast extract -- 30
Seed sludge was obtained from the aeration tank of a local pilot MBR system
for municipal wastewater treatment. After transferring into the lab-scale MBR,
the sludge was allowed to acclimate to the synthetic wastewater for 35 d.
During the startup period, the MBR was operated at the same condition as that
used in the experimental period except no sludge wastage. The experiments
were performed in three phases according to the change of SRT in the order of
30, 16 and 5 d. Before transferring to a new phase, a period of at least two
times of the new SRT was provided for MBR stabilization. In each phase, a
steady state of four weeks was maintained, during which measurements were
evenly conducted for parameters of interest. PAC was dosed into the MBRs
Chapter 3-Materials and Methods
61
with the dosage 30, 80 and 100 mg/L. Table 3.5 shows the PAC dose added to
the MBR system.
Table 3.5 PAC added at the startup of PAC-MBR system.
SRT (d) PAC
dosage (mg/L)
PAC amount added (g)
PAC calculated concentration in PAC-MBR (g/L)
30
30 72 4.5
80 115.2 7.2
100 144 9
16 100 76.8 4.8
5 100 24 1.5
The hydraulic retention time (HRT) of 8 hours and DO concentration of
around 5 mg/L were maintained during the entire experimental period of 515 d.
The MBRs was operated under ambient temperature (28 ± 2 °C) and the pH
was controlled within a range of 6.8-7.5. Fouling development, indicated by
the increase in suction pressure, was monitored by pressure gauges. Membrane
cleaning was carried out in about 47-132 d when the suction pressure
increased beyond 26 kPa. Typically, the interval between two membrane
cleanings became shorter as SRT decreased indicating membrane fouling was
more serious at short SRTs. The membrane module was taken out of the MBR.
It was rigorously rinsed with tap water to remove the attached cake layer
followed by backwashing with 0.05% sodium hypochlorite solution for 2 h to
further remove the foulants adsorbed within membrane pores. The membrane
module was thoroughly cleaned again with tap water before it was mounted
back in the MBR.
Chapter 3-Materials and Methods
62
3.5.3 PFCs mass balance calculation
The mass balance in MBR or PAC-MBR was shown in Figure 3.4. In the
PAC-MBR system, WAS includes waste activated sludge and PAC.
PAC-MBR (MBR)
Q0, C0 Qe, Ce
(Q0-Qe), Ce
Cs
WAS
Aqueous phase
Solid phase
Figure 3.4 Mass balances of PFCs in PAC-MBR or MBR system. 1. Q0 and Qe: flow rate of influent and effluent; 2. C0 and Ce: PFCs concentration in influent and effluent; 3. Cs: PFCs concentration in wasted solids; 4. WAS:
waste activated sludge.
3.5.4 Membrane resistance calculation
The transmembrane pressure (TMP) increased with the increase of operation
time while flux was maintained constant. The resistance-in-series model was
applied to evaluate the fouling characteristics. The permeate flux of a
membrane is governed by the basic membrane filtration equation as follows:
tR
PJµ∆
= (eq. 3.3)
Where J is the permeate flux, ΔP is the transmembrane pressure (TMP), µ is
the permeate viscosity, Rt is the total membrane resistance. The total
membrane resistance, typically, includes three parts, i.e.,
irmt RRRR ++= (eq. 3.4)
where Rm is the intrinsic membrane resistance, Rr is the resistance due to
reversible fouling caused by the cake layer deposited over the membrane
Chapter 3-Materials and Methods
63
surface, and Ri is the resistance due to irreversible fouling caused by solute
adsorption into the membrane pores.
At the end of the experiment, the fouled membrane module was rigorously
rinsed three times with DI water. After physical cleaning, the TMP of
membrane (ΔP’) was measured by filtration of pure water. Based on the
experimental data, the values of Rm, Rr, and Ri can be determined as follows.
JP
Rm µ0∆
= (eq. 3.5)
mi RJPR −
∆=µ
' (eq. 3.6)
imf
r RRJP
R −−∆
=µ
(eq. 3.7)
where ΔP0 is the TMP measured by filtrating pure water with virgin
membrane, ΔP’ is the TMP measured by filtrating pure water with fouled-
membrane after physical cleaning, and ΔP f is the final TMP at the end of
experiment.
3.6 Adsorption study on PAC and activated sludge
3.6.1 Preparation of EfOM
EfOM solution was collected from the mixed liquor of the laboratory-scale
MBR. Immediately, the mixed liquor was centrifuged at 3000 rpm for 10 min
followed by filtration by GF/B glass filter (0.45 µm, Whatman, U.S.A) and
stored at 4 oC until adsorption experiments.
Chapter 3-Materials and Methods
64
3.6.2 EfOM characterization
The fractionation method used in this study was basically based on the
procedure developed by Barker et al., (1999a) with minor modification. The
apparent molecular weight distribution (AMWD) of the EfOM was
determined using ultrafiltraton (UF) membrane in a stirred and pressurized cell
(Model 8200, Amicon, USA), operated in dead end mode. The filtrate
permeating through each YM membrane was collected and DOC
concentration was measured. Nitrogen gas regulated at 30 psi pressure was
used as a driving force for filtration. Gentle turbulence was created at the
membrane surface using a magnetic stirrer to minimize the build-up of a dense
macromolecular layer at the membrane surface. The percentage of organic
matters for each fraction was calculated in terms of DOC based on the mass
balance. The <1 kDa and >30 kDa fractions were obtained and used for the
adsorption experiments. Sodium chloride solution (0.013M), which is of same
ionic strength as MBR effluent, was added to the >30 kDa fraction to achieve
a same DOC concentration as that of <1 kDa fraction.
3.6.3 Equilibrium adsorption experiments
PAC equilibrium adsorption experiments were conducted in duplicate in
EfOM free solution (Mill-Q water), EfOM raw solution and EfOM fractions
(<1 kDa and >300 kDa fractions). PFOS or PFOA stock was spiked to the
solutions and initial single adsorbate concentration ranged from 0.1 to 500
µg/L. Different amount of PAC was added at the appropriate dosage and 1
mM phosphate buffer (0.5mM Na2HPO4 and 0.5 mM NaH2PO4) was spiked to
maintain pH at 7.2.
Chapter 3-Materials and Methods
65
Sludge equilibrium adsorption experiments were conducted in duplicated with
activated sludge at the concentration of 2,000-5,000 mg/L. 1mM phosphate
buffer was spiked to maintain pH at 7.2. PFOS or PFOA stock was spiked to
the sludge solution at the concentration of 50-400 µg/L. Sodium azide (100
mg/L), a respiratory inhibitor, was added to prevent microbial metabolism.
All equilibrium adsorption batch experiments were carried out in an incubator
shaker (CMR, USA) at 25 oC with shaking speed of 120 rpm. Bottles was
sealed and agitated on the shaker for 5 d to reach adsorption equilibrium. Then
PAC particles or sludge were separated by GF/B glass filter (0.45 µm,
Whatman, USA) for the analysis of remaining PFCs concentration in liquid
phase.
3.6.4 Adsorption kinetics experiments
Batch kinetics experiment was conducted in duplicate to determine the
kinetics parameters that describe the rate of removal of the target
perfluorinated compounds by PAC. The initial PFOS or PFOA concentration
for kinetics experiments was 100 µg/L and 1mM phosphate buffer (0.5mM
Na2HPO4 and 0.5 mM NaH2PO4) was spiked to maintain pH at 7.2. PFOS or
PFOA stock was added to 1 L of EfOM solution stirred in a 1-L HDPE bottle
with magnetic stirrer. After 20 min mixing, PAC was added and samples were
collected at predetermined intervals over 6 h. Samples were then filtrated
through GF/B glass filter (0.45 µm, Whatman, USA) to remove PAC.
Chapter 3-Materials and Methods
66
3.6.5 Mathematical modeling
The most frequently used two isotherm models, Langmuir and Freundlich
equations were applied to fit the experimental data to determine the adsorption
capacity of PAC and sludge. These equations describe the non-linear
equilibrium between adsorbed organic compounds on the solid surface and
organic compounds in solution at a constant temperature. The Langmuir
equation which is valid for monolayer adsorption onto a surface with a finite
number of identical sites is given by
CebCebaCs⋅+⋅⋅
=1
(eq. 3.8)
where Cs is the concentration of the solute in the solid phase, Ce is the
equilibrium concentration of the solute in solution; a and b are Langmuir
constants related to maximum adsorption capacity (monolayer capacity) and
bonding energy of adsorption, respectively. The Langmuir equation is used for
homogeneous surfaces. The Freundlich equation assumes neither
homogeneous site energies nor limited levels of adsorption. The Freundlich
equation is defined by
nF CeKCs /1⋅= (eq. 3.9)
where KF and n are the Freundlich constants in relation to adsorption capacity
and adsorption intensity, respectively. The KF
However, the relationship between equilibrium concentrations of organic
compounds in liquid and solid phase could be linear and defined by simple
partition coefficients. For n=1, the partition between the two phase is
independent of the concentration and isotherms becomes linear Freundlich
value corresponds to the
adsorption capacity (ug adsorbate/mg carbon) at an equilibrium concentration
of 1.0 µg/L.
Chapter 3-Materials and Methods
67
euqation. In this case, the experimental data are fitted to linear adsorption
isotherm defined by
CeKCs d ⋅= (eq. 3.10)
where Kd
is the partition coefficient.
3.7 Analysis method
3.7.1 COD and DOC analysis
Chemical oxygen demand (COD) was determined in accordance with
Standard Methods (APHA-AWWA-WEF, 1998). Dissolved organic carbon
(DOC) was measured by 1010 Total Organic Carbon Analyzer (O.I.Analytical,
USA).
3.7.2 Carbohydrate and protein analysis
Carbohydrate and protein were determined according the method of Dubois et
al. (1956) and Lowry et al., (1951), respectively. The phenol-sulfuric acid
method (Dubois et al., 1956) was used to measure the content of carbohydrate
in DOM with glucose as the standard reference, whereas the modified Lowry
method (Lowry et al., 1951; Hartree, 1972) was used for protein determination
with bovine serum albumin (BSA) as the standard reference.
3.7.3 MLSS and MLVSS
Sludge concentration was measured as mixed liquor suspended solids (MLSS)
and volatile suspended solids (VSS) in accordance with Standard Methods
(APHA-AWWA-WEF, 1998).
Chapter 3-Materials and Methods
68
3.7.4 EPS and SOUR analysis
EPS content in biomass was extracted and determined using the established
procedure (Frølund et al. 1996; Ng et al., 2005). First, 200 mL biomass sample
was centrifuged at 2000 g at room temperature for 15 min and the supernatant
decanted. The centrifuged biomass was resuspended back to 200 mL with a
fresh phosphate buffer (526 mg/L NaCl, 74.56 mg/L KCl, 760.2 mg/L Na3PO4
and 552 mg/L NaH2PO4). Then cation exchange resin (DOWEX Marathon C,
Fluka Cat No. 91973F) was added to resuspended sample which was
transferred to a closed container, at 90 g/gVSS. The mixture was stirred at 600
rpm for 2 h in an ice water bath and then centrifuged at 12,000 g for 30 min to
remove the resin and the microorganisms. The supernatant was then analyzed
for carbohydrates and proteins using the analyzing method above (section
3.7.2). Specific oxygen uptake rate (SOUR) was measured in accordance with
the Standard Methods (APHA-AWWA-WEF, 1998).
3.8 LC/MS/MS analysis
3.8.1 Optimization of LC/MS/MS method
High performance liquid chromatograph, composed of a HP100 liquid
chromatograph (Aligent Technologies, U.S.A) interfaced with a triple
quadrupole MS/MS (Applied Biosystems, U.S.A) was applied to detect
samples in the electrospray negative ionization mode. Separation of
compounds was performed on a 150×2.1 mm (5 μm) Zobax Extend C18
column (Aligent Technologies, U.S.A). A 10-μL aliquot of the sample extract
was injected into a guard column (XDB-C8, 2.1 mm i.d.×12.5mm, 5μm;
Agilent Technologies, U.S.A) connected sequentially to Zorbax Extend C18
Chapter 3-Materials and Methods
69
column with 2 mM ammonium acetate aqueous solution (solvent A) and
methanol (solvent B) as mobile phases, starting at 3% of solvent B. The flow
rate was set at 300 uL/min. The gradient was held until 0.50 min, increased to
95% B until 6.00 min, held until 8.50 min, reverted to original conditions at
8.51 min and was held 3% B until 12.00 min. The column temperature was
kept constant at 30 o
C.
Table 3.6 MRM-transitions, compound-dependent parameters of the analytes.
Compound MRM-
transitions (amu)
Decluster Potential
(V)
Focusing Potential
(V)
Entrance Potential
(V)
Collision Energy (V)
Collision Cell Exit Potential
(V) PFOS 499 99 -31 -310 -3.5 -12 -10
PFOA 413 369 -51 -350 -9.5 -60 -6
PFEES 314.5 135 -41 -70 -9 -28 -6
PFDoA 613 569 -26 -150 -3 -16 -18
MS/MS was operated under multiple reaction monitoring (MRM) mode. The
mass spectrometer was operated using the TurboIonsprayTM (TIS) source in
the negative mode. The ionization source-specific parameters were: curtain
gas (CUR), 30 psi; collision gas (CAD), 6 psi; ionspray voltage, -4500V;
temperature of the turbo heater gas, 450 o
C; nebuliser gas (GS1), 40 psi.; turbo
gas (GS2), 90 psi. Nitrogen was used as the curtain gas, nebuliser gas and
turbo gas. Other analyte-dependent parameters were optimised for each
compound (Table 3.6). Methanol was run between the water samples to
prevent carryover effect.
Chapter 3-Materials and Methods
70
3.8.2 Method validation and quantification
The analytical characteristics of the method, such as linear response range,
reproducibility and limits of quantification, were investigated to evaluate the
efficiency of the method and the possibility of the method application to
various water samples. Retention Time was 6.98 min (PFOS) and 6.80 min
(PFOA) in Figure 3.5, which showed a chromatogram of PFOS and PFOS in
spiked Mill-Q water at 1 ng/L.
Figure 3.5 LC/MS/MS chromatograms of PFOS, PFOA and internal standards PFEES and PFDoA.
Seven calibration curve points bracketing the concentrations in samples were
prepared routinely, to check for linearity. Quantification was based on the
response of the external standards that bracketed the concentrations found in
samples. The curve covered a range equivalent to the concentration of the
analytes in 1000 mL water sample after the extract was concentrated to 1.0 ml
(approx. 1000-fold concentration). A calibration curve containing 0.25, 0.5, 1,
XIC of -MRM (4 pairs): 412.4/368.5 amu from Sample 8 (6) of DataSET1.wiff (Turbo Spray) Max. 1320.0 cps.
1 2 3 4 5 6 7 8 9 10 11 12Time, min
0.0
5.0e4
Inte
nsity
, cps
XIC of -MRM (4 pairs): 412.4/368.5 amu from Sample 8 (6) of DataSET1.wiff (Turbo Spray) Max. 1320.0 cps.
1 2 3 4 5 6 7 8 9 10 11 12Time, min
0
1000
Inte
nsity
, cps
9.52
2.21 9.69 12.777.780.57 4.131.77 2.51 10.295.35 10.537.915.55 11.940.68 4.43 6.02 6.60 11.302.62 8.701.360.37 9.083.803.19 4.76 6.75 7.14
XIC of -MRM (4 pairs): 498.2/79.9 amu from Sample 8 (6) of DataSET1.wiff (Turbo Spray) Max. 5.5e4 cps.
1 2 3 4 5 6 7 8 9 10 11 12Time, min
0.0
5.0e4
Inte
nsity
, cps
9.68
XIC of -MRM (4 pairs): 314.5/134.9 amu from Sample 8 (6) of DataSET1.wiff (Turbo Spray) Max. 1.7e4 cps.
1 2 3 4 5 6 7 8 9 10 11 12Time, min
0.0
1.0e4
1.7e4
Inte
nsity
, cps
8.93
XIC of -MRM (4 pairs): 612.0/568.1 amu from Sample 8 (6) of DataSET1.wiff (Turbo Spray) Max. 2760.0 cps.
1 2 3 4 5 6 7 8 9 10 11 12Time, min
0
20002760
Inte
nsi
ty, c
ps
10.19
4.91 10.650.13 6.97 12.078.053.100.83 2.49 5.613.59 6.62 7.34 11.821.89 12.713.711.25 4.52 6.40 9.42 9.578.377.46 9.190.74
PFOS
PFOA
PFOS
PFEES
PFDoA
Chapter 3-Materials and Methods
71
5, 10, 50, 100, and 250 μg/L standard was used. The correlation coefficients
(r2
) exceeded 0.995. The instrumental detection limits (IDL) were 1-5 pg, as
summarized in Table 3.7. The limit of detection (LOD), defined as the
concentration that yielded an S/N ratio of higher than or equal to 3, and the
limit of quantification (LOQ), defined as the concentration that yielded a S/N
ratio of higher than or equal to 10, were determined by the SPE extraction of
spiked Mill-Q water samples.
Procedural recovery was evaluated by spiking mixture of external standards
(100 ng/mL, 100 µL) to Milli-Q water. Procedural recoveries for PFOS and
PFOA were in the range of 95-103% (mean: 98.4%, n=3) and 90-98% (mean:
93.8%, n=3), respectively. During the analysis of samples, procedural and
instrumental blanks were analyzed. They were below the detection limit,
indicating no contamination occur in sampling and analysis. Sample extracts
with concentrations exceeding the range of calibration curve were
appropriately diluted by methanol and reinjected again. Furthermore, spiked
additions were applied to identify the matrix suppression on the ion signals for
each batch of samples based on the standard additoin method. They were
prepared by spiking mixture of external standards (100 ng/mL, 100 µL) into
the SPE extracts. Mean recoveries of spiked additions for PFOS and PFOA
were 84-91% and 70-84%, respectively. Sufficient recoveries achieved for
procedural blanks and spiked additions proved the reliability and efficiency of
analysis method.
Chapter 3-Materials and Methods
72
Table 3.7 IDL, LOD and LOQ of PFOS and PFOA.
Analyte IDL (pg)
Water and Wastewater Sludge Recovery
LOD (ng/L)
LOQ (ng/L)
LOD (ng/g)
LOQ (ng/g) Water
Waste-water Sludge
PFOS 1 0.1 0.25 1 2 88% 91% 84% PFOA 5 0.5 1.25 5 8 84% 82% 70%
3.9 Fractionation process
The fractionation method used in this study was basically based on the
procedure developed by Leenheer (1981) and Thurman (1985) except that the
anion exchange resin Duolite A-7 was substituted by Amberlite IRA-96, since
this type of resin was also suggested for fractionation process by Chang et al.
(2002) and it was readily available. Resins used (XAD-8, AG MP-50and IRA-
96 exchange resin) were pre-purified using the Soxhlet extraction method
described by Leenheer (1981).
Prior to the fractionation process, the columns (i.d = 25mm x 100mm),
endpieces and the accompanying frits for uniform water distribution were
washed with HCl acid (~0.3M) to remove trace carbon. The service flow rate
used for XAD8 resin was about 15 BV/h; while the service flow rates used for
ion exchange resins were about 30 BV/h. After removal of suspended solids
by MF (microfiltration) and adjustment of pH to 7 by HCl, 100-300 L
(according to resin capacity) of water sample was introduced and passed
through three types of resins (Figure 3.6). The compounds adsorbed by the
first XAD-8 resin column were eluted using 100 ml 0.1N HCl, defined as
Hypho-B. The filtrate was acidified to pH 2 with 2M HCl and then re-
introduced into another XAD-8 column. The organic matters adsorbed by the
Chapter 3-Materials and Methods
73
second XAD-8 resin column were eluted using 100 mL of 0.1N NaOH as a
brownish solution defined as AHS (acid humic substance, also called Hypho-
A) containing HA (humic acids) and FA (fulvic acids). Then the second XAD-
8 resin column was dried at 60 °C and the residual matters were washed out by
methanol (50 mL) to get the Hypho-N. A vacuum concentration instrument
(BÜCHI Rotavapor R-124, Switzerland) combined with high purity nitrogen
gas was used to concentrate this solution. The Rotavapor was operated under a
vacuum pressure around 900 mbar and at a temperature of 62 °C with a
rotation speed of 50 rpm, and the whole process lasted for 20-30 min. The
portion that passed through the second XAD-8 resin column, which contained
only hydrophilic solutes, was pumped through the AG-MP-50 cation-
exchange resin column. Hyphi-B retained on this cantion-exchange resin, was
eluted by 100 ml of 2M HCl. The filtrate was pumped through the IRA-96
anion-exchange resin column and the Hyphi-A absorbed on this resin was
eluted with 100 ml of 1M NaOH. The final effluent, which passed through
three types of resins, was defined as Hyphi-N.
Chapter 3-Materials and Methods
74
Figure 3.6 Procedure for fractionation of DOM.
3.10 Quality assurance and control
Because of the presence of the fluoropolymer in some laborotary equiments,
precautions were taken to minimize the possible contamination during the
analysis (Yamashita et al., 2004). For example, teflon bottles, Teflon-lined
caps, and any suspected fluoropolymer materials were not utilized throughout
the analysis. In order to ensure the quality of the sampling, Milli-Q water was
used as field blanks to evaluate the possible contamination during the
transportation for each batch of samples. For each field blank, PFOS and
Methanol
Hyphi-N
pH =2
XAD-8 resin
0.1N NaOH
AG-MP-50 cation resin
IRA-96 anion resin
Hypho-N 2 N HCl
AHS
1N NaOH
Hyphi-B
Hyphi-A
MBR supernatant
MF filter
XAD-8 resin
0.1 N HCl
Hypho-B
(Hypho-A)
Chapter 3-Materials and Methods
75
PFOA were below the detection limit, indicating that no discernable
contamination occurred during sampling.
Spiked additions were applied to identify the matrix suppression on the ion
signals for each batch of samples based on the standard addition method. They
were prepared by spiking mixtures of external standards (100 ng/mL, 100 µL)
into the SPE extracts of the effluents obtained from W4. Recoveries of spiked
additions for PFOS and PFOA were in the range of 80-93% (mean: 87.8%,
n=5) and 78-90% (mean: 83.9%, n=5), respectively. Sufficient recoveries
achieved for spiked additions demonstrated the reliability and efficiency of the
analysis method.
3.11 Statistical analysis
Statistical software Minitab (Minitab Inc, USA) was used to calculate the
correlation between PFOS and PFOA as well as the correlations of
concentrations between dry season and wet season. Statistical significance was
accepted at p<0.05.
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
76
CHAPTER 4 OCCURRENCE OF PFOS AND
PFOA IN WATER AND WASTEWATER
4.1 Introduction
PFOS and PFOA are ubiquitous in the environment because of their high
persistence, resulting from their exceptionally thermal and chemical stability.
Surveys have been conducted to monitor the extent of PFOS and PFOA
contamination in surface waters (Hansen et al., 2002; Boulanger et al., 2004;
Loos et al., 2008), wastewaters (Boulanger et al., 2005; Becker et al., 2008),
drinking waters (Harada et al.,2003), groundwaters (Schultz et al., 2004) and
coastal waters (So et al., 2004; Saito et al., 2003; Yamashita et al.,2005). The
pathways of PFCs to aquatic environment could include (a) discharge of
effluents from STPs, (b) direct discharge of wastewater from manufacture and
use of PFCs to the aquatic environment, (c) rain runoff moving PFCs
pollutants on ground (such as oil, fire-fighting foam) to the aquatic
environment, and (d) atmospheric transport of PFCs and subsequent
atmospheric loading of PFCs to surface waters (Prevedouros et al., 2006;
Zushi et al., 2008). A few studies have been conducted to identify the
contamination source of PFCs in the environment. Some researchers observed
that effluents from the STPs are the most important PFCs sources for the
aquatic ecosystems (Sinclair and Kannan , 2006; Loganathan et al., 2007).
Zushi et al. (2008), however, reported that loads of PFCs in rain runoff were
about 2-11 folds greater than those in STP effluents that were discharged into
a river, indicating that nonpoint source of PFCs could be the most important
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
77
source for the river studied. In addition, Yamashita et al. (2004) reported that
application of PFC-containing products could also be an important source of
contamination for aquatic environment. It seems that effluents from STPs,
nonpoint source from rain runoff and application of PFC-containing products
might be important sources and determine the PFCs concentration levels in the
aquatic environment. However, there could be other significant PFCs sources
such as atmospheric deposition or precipitation for the aquatic environment.
Therefore, further research is needed to identify possible contamination
sources and transportion pathways of PFCs in aquatic environment.
Furthermore, seasonal variations in the PFCs concentraions were investigated.
So et al (2004) observed that PFCs concentrations in the winter were higher
than those in the summer in coastal waters of China. In wastewater of STPs,
Loganathan et al. (2007) found that mass flow of PFCs were higher in winter
than in summer. The authors suggest that there were less rain in winter than in
summer, which resluted in dilution effect on the coastal waters or wastewaters
in summer. However, no data is available on the comparison of PFCs
concentrations between dry season and wet season in the aquatic environment.
Singapore is an island coutry and also a true city-state with a tropical
rainforest climate and no distinctive seasons. Especially its climate is
characterized by uniform temperature, pressure and abundant rainfall in wet
monsoon season (November and December). In a such an unique isoland city,
it could be an ideal place to identify seasonal variations of PFCs
concentrations between dry seasons and wet seasons by excluding other
factors, such as temperature and atmospheric pressure variation. To the best of
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
78
our knowledge, this study is the first study to identify the seasonal variations
of PFCs in aquatic environment between dry season and wet season.
In order to investigate the distribution of PFOS and PFOA in different water
matrices in Singapore, 138 water samples were collected from reservoirs,
rivers/canals, wastewater treatment plants and coastal waters around the island
over a year. The purpose of this study was to determine the magnitude and
extent of PFCs’ contamination and to provide an overview of the spatial
distribution of PFOS and PFOA in the waters of Singapore. Moreover, surface
water samples in the industrial districts and wastewater from all six WWTPs
in Singapore as well as coastal water samples were collected and analyzed in
an attempt to locate possible contamination sources within the island. In
addition, seasonal variations between dry season and wet season were studied.
The results of this study would identify the sources and transport pathways of
PFCs in the aquatic and oceanic environment of Singapore.
4.2 Results and discussions
4.2.1 PFOS/PFOA concentration in surface water
Concentrations of PFOS and PFOA in surface waters, wastewaters and coastal
waters from five batches of sampling campaigns are summarized in Figures
4.1-4.3, which show the spatial distribution of those two compounds in
western, middle and eastern areas of Singapore. Overall, PFOS and PFOA
concentrations in all samples were in the range of 2.2-532.1 ng/L and 2.4-
1,057.1 ng/L, respectively.
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
79
The concentrations of PFOS and PFOA in surface waters ranged from 2.2-
87.3 ng/L and from 5.7-91.5 ng/L, respectively. This is comparable to but
slightly higher than those observed in the Great Lakes (USA) (PFOS: 21-70
ng/L, PFOA: 27-50 ng/L) (Boulanger et al., 2004). Comparable PFOS
concentration range was also observed in Guangzhou (0.9-99 ng/L) (So et al.,
2007), one of most industrialized areas in China. The highest concentration of
PFOS (87.3 ng/L) in surface waters was detected at S5, eastern area subjected
to light industrial influence. This indicated potential PFCs contamination
sources nearby. In comparison to other studies, however, the highest PFOS
concentration was approximately half of that reported in Tama river in Japan
(157 ng/L) (Saito et al., 2003) and in downstream of discharge of 3M
fluorochemical manufacturing facility (144 ng/L) (Hansen et al., 2002). The
concentration of PFOS detected at S5 was also about 7 times lower than the
highest concentration (651 ng/L) measured in Lake Shihwa (South Korea),
which is heavily influenced by the industrial effluent from the Shihwa
industrial district (Rostkowski et al., 2004).
Compared to this study, lower PFOA concentration range was observed in
Guangzhou area (0.85-13 ng/L) (So et. al., 2007) and Pearl River Delta (0.24-
16 ng/L) (So et al., 2004), both of which are heavily associated with industrial
and urban activities. In particular, the highest PFOA concentration (91.5 ng/L)
in surface water was observed at S7, which was collected downstream of a
canal that flows along the edge of an airport. It suggests that the airport may
be a potential PFCs contamination source. In contrast, the PFOA level in this
study was approximately 2 and 3 times lower than those observed in Tokyo
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
80
Figure 4.1 Concentrations of PFOS and PFOA in surface waters, wastewaters and coastal waters from western area of Singapore collected by: 1. Oct 2006; 2.
Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007; n.a: not available.
Bay in Japan (Yamashita et al., 2004) and Yangtze Rive in China (So et al.,
2007).
PFOS
PFOA
n.a
n.a
n.a
n.a
n.a
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
81
Figure 4.2 Concentrations of PFOS and PFOA in surface waters, wastewaters and coastal waters from middle area of Singapore collected by: 1. Oct 2006; 2.
Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007; n.a: not available.
The total PFCs (i.e., PFOS and PFOA) concentrations from 5 sampling
campaigns for all surface waters are summarized in Figure 4.4. It can be seen
that S9, located at the western area, had the highest total PFCs concentration,
which suggests the presence of potential PFCs contamination source in the
surrounding area. However, S7 had the next highest total PFCs concentration
among all surface waters even though its location is in the eastern area (urban
region). This may be due to the leakage of perfluorinated surfactants, such as
PFOS
PFOA
n.a
n.a
n.a
n.a
n.a n.a n.a
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
82
Figure 4.3 Concentrations of PFOS and PFOA in surface waters, wastewaters and coastal waters from eastern area of Singapore collected by: 1. Oct 2006; 2.
Dec 2006; 3. Mar 2007; 4. Sep 2007; 5. Dec 2007; n.a: not available.
aqueous fire-fighting foams, gasoline, oil and lubricants (Moody et al., 2000),
from the adjacent airport. Similarly, the highest total PFCs concentration in
reservoir waters was detected in R8, which is the downstream of the S9.
Furthermore, R2, R3 and R4 which are in the nature reserve area (middle area)
had lower concentrations compared to other reservoirs which are in either
industrial or commercial influenced areas. In contrast, the higher
concentrations were observed in R7, R8 and R9 which are in the industrial
area (western area). It suggests that factories, such as petrochemical, paints,
coatings and surfactants manufacturing plants in the western region, may be
the potential PFCs sources, thus causing this area to be the most contaminated
area in Singapore. It is, however, worthy to point out that, even though the
PFOS
PFOA
n.a
n.a n.a
n.a
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
83
R1 R2 R9R3 R4 R5 R6 R7 R8 S1 S2 S3 S4 S8S6 S7S5 S9 S10S11S12
51.7%
64.5%61.4%
54.6%
56.6%
61.9%
69.2%
61.1%
49.9%
63.4%60.8%
61.2%
55.3%
29.5%
64.3%
55.4%
56.8%
59.9%
74.6%67.7%
69.6%
0.0
50.0
100.0
150.0
200.0
250.0
300.0
350.0
400.0
450.0To
toal
PFO
S a
nd P
FOA
con
cent
ratio
n (n
g/L) PFOA
PFOS
Figure 4.4 Total PFOS and PFOA concentrations in surface waters summed up
by 5 sampling campaigns. The value in pencentage above the each column indicates PFOA pencentage in total PFOS and PFOA concentration.
highest concentrations of PFCs in reservoirs were observed in R8 and R9,
their PFOS and PFOA concentrations were a few order of magnitude lower
than the lifetime drinking water health advisory for PFOS (1,000 ng/L)
(Hansen et al., 2002) and PFOA (150,000 ng/L) (Psoulsen et al., 2005).
Overall, the concentrations of PFOS and PFOA in reservoirs (mean: 17.6 ng/L
for PFOS, 24.9 ng/L for PFOA) were comparable to those of river water
samples (mean: 20.2 ng/L for PFOS, 28.4 ng/L for PFOA). Reservoir waters,
however, had relatively lower variation in concentrations for both PFOS and
PFOA in comparison with river/canal waters.
PFOA percentage in the total PFCs concentration varied from 29.5%-74.6% in
surface waters, indicating different contamination sources of perfluorinated
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
84
compounds (Figure 4.4). PFOA contribution to the PFCs, which were greater
than 50% in all surface water samples except for R9 and S5, indicated that
PFOA was predominant in most of surface waters. Predominance of PFOA
over other PFCs was also observed in the water samples collected along the
Yangtze River, the largest river in China (So et al., 2007). The lowest PFOA
percentage observed in S5 suggests there may be significant PFOS
contamination source upstream or in the surrounding area, which greatly
increase PFOS contribution to the PFCs. In addition, even though location of
R9 is geographically close to R8, the lower PFOA percentage at R9 suggests
different contamination sources occur in its catchment area or its tributaries.
4.2.2 PFOS/PFOA concentration in wastewater
PFOS concentrations were observed in the range of 5.8-532.1 ng/L in the
effluents of WWTPs. This is much higher than those of 10 municipal WWTPs
(1.1-130 ng/L) in USA (Schultz et al., 2006). The highest PFOS concentration
(532.1 ng/L) was detected in the effluent of W1, one of the domestic WWTPs.
This is comparable to those measured in a WWTP in Cleveland with no
known fluorochemical sources (417-454 ng/L) (3M, 2001). With the exception
of W1, PFOS concentrations in effluents of domestic WWTPs, in the range of
5.8-35.3 ng/L, were lower than those of 10 municipal WWTPs in USA
(Schultz et al., 2006). The higher PFOS concentration in W1 as compared to
other domestic WWTPs suggested the occurrence of significant PFOS
contamination sources (such as airport etc.) in the served area. High PFOS
concentration (461.7 ng/L) was also observed in the effluent of W5, which
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
85
receives 60% industrial wastewater with known fluorochemical sources such
as petrochemicals industry.
PFOA concentration in the effluents of WWTPs was observed in the range of
7.9-1,057.1 ng/L. PFOA concentrations in industrial WWTPs (22.6-1,057.1
ng/L) were much higher than those from domestic WWTPs (7.9-157.3 ng/L).
In the effluents of domestic WWTPs, PFOA concentrations were higher than
those of 10 WWTPs in USA (PFOA: 2.5-97 ng/L) (Schultz et al., 2006) and
comparable to 2 WWTPs in Kentucky and Georgia (PFOA: 6.7-183 ng/L). In
addition, the highest PFOA concentration (1,057.1 ng/L) was also observed in
the effluent of W5. This is significantly higher than those from Cleveland
(665-674 ng/L) but about two times lower than those from Decatur (2,140-
2,420 ng/L) with influence from fluorochemical manufacturing facilities (3M,
2001). The industries in the service area of W5 include petrochemical
intermediates, petroleum refining, paints, coatings and surfactants
manufacturing. Our results suggest that discharges from these fluorochemical
related plants may contain a large amount of PFOS and PFOA or their
precursors, thus leading to high PFCs concentrations in the effluent of
WWTPs. It is evident that discharge of effluents from WWTPs is an important
pathway through which PFCs enter the environment.
Figure 4.5 shows that W5 had the highest total PFCs mass load which was
calculated by multiplying WWTP’s daily flow rate with its total PFCs
concentration. Even though daily flow rate of W5 was comparable to that of
other WWTPs, its total PFCs’ mass load was 2.3-7.0 times of that of other
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
86
WWTP. It indicates that industrial influent significantly affect the PFCs’ mass
load which was discharged into the aquatic environment. However, much
higher total PFCs mass load occurred in W1, one of the domestic WWTPs,
than that of W2 (industrial WWTP). Compared to W5, there is fewer known
fluorochemical source in the served area of W2. It suggests that the
composition of the industrial influent entering W2 may be different and not
closely related to PFCs’ contamination. Moreover, PFOA was predominant
over PFOS in the WWTPs’ effluents except for W1. The lowest PFOA
percentage observed (22.5%) in the effluent of W1 indicates that there could
be potential contamination sources of PFOS in its service area. It is known that
PFOS is used widely in multiple photolithographic chemicals, such as
photoacid generators (PAGs) and anti-reflective coatings (ARCs) in the
semiconductor industry. Due to its health concerns, PFOS has been phased out
in the European Union semiconductor industry since 2008. However, there is
no restriction on the applications of PFOS in local semiconductor
manufacturing. Many semiconductor manufacturers are located in the service
area of W1, which would lead to high PFOS concentrations and predominance
of PFOS in the effluents of W1.
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
87
W6W5W4W3W2W1
22.5%
65% 83.3% 81.1%
57.3%
70.2%
050
100150200250300350400450500550600650
Toto
al P
FCs
mas
s lo
ad (g
/d) Total PFOA load
Total PFOS load
Figure 4.5 Total PFOS and PFOA mass load in the effluent of WWTP
summed up by 5 sampling campaigns. The value in pencentage above the each column indicates PFOA pencentage in total PFOS and PFOA mass load.
4.2.3 PFOS/PFOA concentration in coastal water
PFOS and PFOA were detected in all coastal waters and were in the range of
1.9-8.9 ng/L and 2.4-17.8 ng/L, respectively. This is relatively higher than
those detected in Hongkong coastal waters (PFOS: 0.09-3.1 ng/L, PFOA: 0.7-
5.5) (So et al., 2004), but much lower than those detected in Tokyo Bay
(PFOS: 0.3-57.7 ng/L, PFOA: 1.8-192.0 ng/L) (Yamashita et al., 2005). The
highest PFOS and PFOA concentration as well as total PFCs concentration in
coastal waters were detected at C4 (Figure 4.6). C4 is near the causeway
connecting the Singapore Island and the Malay Peninsula across the Johor
Strait. It suggests that Johor Straits is more heavily contaminated than the
southern and eastern coastal waters. Industries in the northwestern area may be
the significant contamination sources for Johor Straits since W2 discharges its
effluent nearby. The next highest total concentration occurred in C2, which
was collected from the confluence of rivers S2, S3 and S4 flowing through the
commercial areas. PFOS and PFOA concentrations in S2, S3 and S4 were in
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
88
the range of 4.8-28.3 ng/L and 7.4-36.6 ng/L respectively, and were expected
to be the major contributors for high level of PFOS and PFOA to C2. The
lowest total PFCs concentration was observed at the location of C1, suggesting
that eastern coastal waters is cleaner than other coastal waters in terms of
PFCs contamination. It was observed that PFOA was predominant over PFOS
in all coastal waters, which was indicated by the PFOA percentage. Such an
interesting observation was also reported by other studies (So et al., 2004;
Yamashita et al., 2005). The highest PFOA percentage (69.2%) was observed
at the site of C2, suggesting that commercial activity may lead to high PFOA
composition in waters.
C1 C2 C3 C4
60.6%
66.3%
69.2%
57.3%
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
Toto
al P
FOS
and
PFO
A c
once
ntra
tion
(ng/
L) PFOAPFOS
Figure 4.6 Total PFOS and PFOA concentrations in coastal waters summed up
by 5 sampling campaigns. The value in pencentage above the each column indicates PFOA pencentage in total PFOS and PFOA concentration.
As the OW1 is close to the location of C4, the correlation between coastal
water C4 and wastewater W2 was investigated (Figure 4.7). For both PFOS
and PFOA, water samples at C4 could be significantly correlated to effluent of
W2. Furthermore, PFOA percentage in C4 (60.6%) was in agreement with that
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
89
of W2 (65%). This suggests that discharge of WWTPs may be the major
source of perfluorinated compounds in the environment.
R2 = 0.8764
R2 = 0.5289
0.0
5.0
10.0
15.0
20.0
0.0 50.0 100.0 150.0PFCs concentration in W2 (ng/L)
PFC
s con
cent
ratio
n in
C4
(ng/
L)
PFOS
PFOA
Figure 4.7 Correlations of PFOS and PFOA between
coastal water C4 and wastewater W2.
4.2.4 Seasonal variations in concentration of PFOS/PFOA
For surface waters, significant seasonal difference (p=0.025) was observed for
PFOS between dry and wet seasons, while no significant difference (p=0.616)
was observed for PFOA. It seems that PFOS concentrations noticeably
decreased during wet season, while there was no discernable decrease in
PFOA concentrations in surface waters during wet season. The presence of
PFCs in rainfall indicates rainfall significantly affect their concentrations in
surface water (Prevedouros et al., 2006). PFOS have been observed at a low
concentration (0.59 ng/L) in the precipitation, while significant higher PFOA
concentrations were reported in rainwater. Kallenborn et al. (2004) reported
that PFOA was the predominant PFCs measured in rainwater samples from
Sweden and Finland with the greatest concentrations (11 ng/L and 17 ng/L,
respectively). Scott et al. (2006) also reported relatively high PFOA
concentrations (<0.1-89 ng/L) in the rainwater samples from U.S.A and
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
90
Canada. Based on the limited data, it seems that PFOS concentration in
rainwater is lower than that of PFOA, which leads to their different seasonal
variations in surface water. Furthermore, runoff could also be the potential
PFCs sources during rainy weather. Rainfall may pick up and carry away
PFCs pollutants (such as oil, fire-fighting foam) when it moves over and
through the ground, which leads to the occurrence of nonpoint source
pollution (NPS) of PFCs. Zushi et al. (2008) also observed that some PFCs
concentrations in a river did not decrease with the increase of river flow rate
during the rainy weather due to the nonpoint source pollution of PFCs. The
predominance of PFOA over PFOS in most of surface waters indicated that
PFOS is used less widely than PFOA in the island. Therefore, runoff contains
lower amount of PFOS, which dilutes the surface waters. Unlike the surface
waters, no discernable seasonal differences were found for both PFOS and
PFOA in coastal waters and wastewaters. NPS of PFCs occurring during wet
season may contribute to the indiscernible variations in PFCs concentrations
from coastal waters and wastewaters. However, it is possible that industrial
activities lead to high concentration variations in the wastewaters, which
override the seasonal differences between dry seasons and wet seasons. Since
wastewaters are discharged into the coastal water, the same effect could
subsequently apply to coastal waters. Similarly, no large-magnitude seasonal
variations in concentrations of PFCs were found among spring, summer, fall
and winter seasons in two municipal sewage treatment plants (Loganathan et
al., 2007).
4.2.5 Correlations between PFOS and PFOA
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
91
Correlations of PFOS and PFOA in the surface waters, coastal waters and
effluents of WWTPs were examined. It was found that PFOS and PFOA in the
coastal waters could be significantly correlated (R2=0.568, p=0.001), while
weak positive correlations were observed in surface waters (R2=0.197,
p=0.001) and wastewater (R2=0.282, p=0.003), as shown in Figures 4.8, 4.9
and 4.10. This suggests that the possibility of a common contamination source
for these two compounds in coastal waters may be higher than those of surface
waters and wastewaters. Such correlation between those two PFCs could be
attributed to the production and application of related products as well as their
subsequent release into the environment. Similarly, So et al. (2007) observed
strong positive correlations between PFOS and PFOA in coastal waters
collected from Hongkong and South China. In addition, correlations between
PFOS and PFOA in effluents of individual STPs were investigated. It was
found that concentrations of PFOS were significantly correlated to PFOA in
effluent of W4 (R2=0.739, p=0.005) and W5 (R2=0.629, p=0.002), while
medium positive correlations were observed in the effluent of W2 (R2=0.389,
p=0.001) (Figure 4.11). However, PFOS and PFOA concentrations were
weekly correlated in the effluent of W1 (R2=0.151, p=0.001), W3 (R2=0.053,
p=0.001) and W6 (R2=0.025, p=0.001). It seems that the correlations between
PFOS and PFOA are not determined by influent composition of STPs since
significant and weak correlations were observed in both domestic and
industrial STPs. Similarly, Loganathan et al. (2007) found that PFOS were
significantly correlated to PFOA in the wastewaters of one WWTP (R2=0.772),
while weak positive correlations were observed in those of another WWTP
studied (R2=0.084).
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
92
y = 0.466 x + 18.077R2 = 0.197
0102030405060708090
100
0 20 40 60 80 100PFOS concentration (ng/L)
PFO
A co
nent
ratio
n (n
g/L)
Figure 4.8 Correlations between PFOS and PFOA concentrations
in surface waters.
y = 1.434 x + 1.479R2 = 0.568
0
4
8
12
16
20
0 2 4 6 8 10PFOS concentration (ng/L)
PFO
A c
once
ntra
tion
(ng/
L)
Figure 4.9 Correlations between PFOS and PFOA concentrations
in coastal waters.
y = 0.739 x + 40.357R2 = 0.282
0
200
400
600
800
1000
1200
0 100 200 300 400 500 600PFOS concentration (ng/L)
PFO
A c
once
ntra
tion
(ng/
L)
Figure 4.10 Correlations between PFOS and PFOA concentrations
in wastewaters.
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
93
y = 1.5589x + 11.092R2 = 0.389
y = 12.744x - 99.97R2 = 0.7386
0
50
100
150
200
0 20 40 60 80PFOS concentration (ng/L)
PFO
A c
once
ntra
tion
(ng/
L)
W2
W4
(a) (b)
Figure 4.11 Correlations between PFOS and PFOA concentrations in the effluents of (a) W2 and W4; (b) W5.
4.3 Summary
PFOS and PFOA were detectable in all 138 water samples from reservoirs,
rivers/canals, coastal waters and treated effluents from WWTPs around the
island. Ranges of PFOS concentrations in surface waters, wastewaters and
coastal waters were 2.2-87.3 ng/L, 5.8-532.1 ng/L and 1.9-8.9 ng/L,
respectively, while those of PFOA concentrations were 5.7-91.5 ng/L, 7.9-
1057.1 ng/L, 2.4-17.8 ng/L, respectively. Overall, coastal waters had lower
concentrations of PFOS and PFOA, compared with surface waters and
wastewaters.
In surface waters, the highest total concentrations of PFOS and PFOA were
observed in the western area because of the high levels of industrial activities
in that area. This region was noted to be the most highly contaminated by
PFCs. The next highest total concentration was observed at the location that is
adjacent to the airport, indicating that leakages of perfluorinated surfactants,
such as aqueous fire-fighting foams, gasoline, oil and lubricants (Moody et al.,
2000) from the airport may be potential contamination sources.
y = 2.1864x - 201.09R2 = 0.6286
0
200
400
600
800
1000
1200
0 100 200 300 400 500PFOS concentration (ng/L)
PFO
A c
once
ntra
tion
(ng/
L)
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
94
In wastewaters, the highest total PFCs mass load and PFOA concentration
(1057.1 ng/L) were observed in W5, suggesting discharges of fluorochemical
related factories in the service area of W5 may contain a large amount of
PFOS and PFOA, thus resulting in high concentrations in the WWTPs
effluents. The highest PFOS concentration (532.1 ng/L) was detected in the
effluent of W1 treating mainly domestic and commercial wastewater. This
indicates the presence of potential PFOS contamination sources in its service
area. Compared with surface waters and coastal waters, much higher PFCs
concentrations in wastewaters indicate that discharge of effluents of WWTPs
is an important pathway by which PFCs enter the environment.
In coastal water, the high PFOS and PFOA concentrations at C4 suggest that
Johor Straits is more heavily contaminated than the southern and eastern
coastal waters. The high levels of industrial activities in the western area may
be the significant contamination sources for Johor Straits. In addition, for both
PFOS and PFOA water samples at C4 were significantly correlated with
effluents of W2.
Between dry and wet seasons, significant seasonal variation was observed in
surface waters for PFOS only, while no discernable seasonal differences were
found for both PFOS and PFOA in coastal waters and wastewaters. In addition,
PFOS and PFOA were significantly correlated in the coastal waters, while
weak positive correlations were observed in surface waters and wastewaters. It
suggests that the possibility of a common contamination source for these two
Chapter 4-Occurrence of PFOS and PFOA in Water and Wastewater
95
compounds in coastal waters is higher than those of surface waters and
wastewaters.
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
96
CHAPTER 5 DEVELOPMENT OF POST
EXTRACTION CLEAN-UP METHOD FOR
PFOS/PFOA DETERMINATION IN
WASTEWATER AND SLUDGE SAMPLES
5.1 Introduction
High-performance liquid chromatography (HPLC) with triple quadrupole
mass spectrometry in electrospray negative mode is the most promising and
extensively applied method for analyzing PFCs in various environmental and
biological matrices (Giesy et al., 2001; Tomy et al., 2004; Becker et al., 2008;
Boulanger et al., 2005 ; Sinclair et al., 2006; Higgins et al., 2005 ; Taniyasu et
al., 2003; Kannan et al., 2004; Moody et al., 2001; Hansen et al., 2001;
Martin et al., 2004b). Analysis can be accomplished by direct injection
(Schultz et al., 2006) or preconcentration on solid phase extraction (SPE)
cartridges, followed by LC/MS/MS analysis (Giesy et al., 2001; Tomy et al.,
2004; Becker et al., 2008; Boulanger et al., 2005 ; Sinclair et al., 2006).
However, analysis of complex environmental matrices such as sediment,
sludge and wastewater by electrospray LC/MS/MS can be significantly
hampered by ionization effects induced by co-eluting components present in
the sample extracts. Several studies have shown that matrix effects resulting
from co-eluting residual matrix components enhance or suppress electrospray
ionization of perfluorinated analytes, leading to considerable inaccuracy
(Boulanger et al., 2005; Sinclair et al., 2006; Higgins et al., 2005). Therefore,
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
97
it is very important to eliminate matrix effects when the LC/MS/MS method is
used to quantitatively determinate the concentration of perfluorinated
compounds.
Structural analogues of analytes as internal standards are effective techniques,
which show similar behavior in the samples, to compensate for the matrix
effects. Standard addition quantitation, which involves spiking successive
known quantities of a standard into the sample and reanalyzing, is an
acceptable technique to use when matrix effects are unavoidable.
Unfortunately, standard addition quantitation places further demands on
instrument and sample preparation time. Structural analogues of analytes as
internal standards are valuable alternatives, which show similar behavior in
the samples and would compensate for the matrix effects (Petrovic et al., 2005;
Benijts et al., 2004; Matuszewski et al., 2006). An important prerequisite is
that analyte and internal standard have very similar characteristics, and
identical, or at least very close, retention times. Both compounds would
therefore be affected by the co-eluted matrix to the same extent. Structural
analogues of perfluoroalkyl such as PFDoA and PFEES have been used as
internal standards to determine perfluorinated compounds in water and
biological tissue samples with acceptable recovery (Higgins et al., 2005;
Moody et al., 2001; Benijts et al., 2004; Tseng et a., 2006). However,
isotopically labeled internal standard such as 1, 2-13C PFOA is either
expensive or limited by its commercial availability (Martin et al., 2004a).
Although structural analogues of analytes can serve as internal standards, the
potential for ionization enhancement or suppression remains high in complex
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
98
environmental and biological samples. On the other hand, standard addition
quantitation, which involves spiking successive known quantities of a standard
into the sample and reanalyzing, is a valuable alternative to use when matrix
effects are unavoidable. Without using any internal standard for compensation,
standard addition quantitation could be the most accurate analysis method for
the determination of PFCs in the water (Weremiuk et al., 2006; Furdui et al.,
2008). Unfortunately, standard addition quantitation places further demands
on instrument and sample preparation time. Therefore, post-extraction clean-
up is desired to eliminate matrix interference in complicated environmental
and biological samples (van Leeuwen et al., 2006; Szostek et a., 2004; Simcik
et al., 2005; van de Steene et al., 2006). Powley et al. (2005) applied Envi-carb
(graphitized carbon) and glacial acetic acid to purify the crude extracts of
biological matrices (blood, serum, live and plant tissue). Szostek et al. (2004)
used silica column to clean up fish tissues by eluting the lipids with
dichloromethane, while the target compounds (PFCAs and PFSAs) were
eluted with acetone. For surface water samples, fluorous silica column
chromatography was used to clean up the SPE extracts and remove the
interfering compounds prior to LC/MS detection (Simcik et al., 2005).
However, the above post-extraction methods may not be applicable to
wastewater and sludge samples collected form STPs, in which stronger matrix
effect was observed in comparison with surface water (Boulanger et al., 2005;
Sinclair et al., 2006). Furthermore, the developed method of this study is more
time efficient in comparison with other post-extraction clean-up method
(Simcik et al., 2005) as SPE extract was dried by gentle nitrogen stream only
once during the sample preparation. In addition, although the effect of these
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
99
post-extraction clean-ups was assessed by the improved recoveries for PFCs,
matrix effect issue has not been sufficiently studied and addressed. The
assessment of matrix effect during development and validation of LC/MS/MS
method is necessary to ensure the precision, selectivity, and sensitivity would
not be compromised. However, limited data is available on the quantitive
estimation of matrix effect and effect of post-extraction clean-up on different
environmental matrices.
The objective of this study was to develop a new post-extraction clean-up
method for the determination of PFOS and PFOA in environmental matrices.
The influence of different environmental matrices on the electrospray
ionization efficiency was assessed by comparing MS response of post-
extraction spiked sample and that of the standard. In addition, the developed
clean-up method was applied to sludge samples to further remove interfering
components after solid phase extraction.
5.2 Results and discussions
5.2.1 Effect of clean-up procedures on matrix effect
Figure 5.1 shows LC-MS-MS chromatograms of PFOS and PFOA in the raw
sewage extracted using the procedure of HLB SPE and HLB+silica. After
silica cartridge clean-up, the intensity of MS increased significantly by 57.6%
for PFOA and 60.4% for PFOS, respectively. It indicates that significant
enhancement in the MS responses of PFOS and PFOA could be due to the
efficient removal of co-eluting interfering compounds by silica cartridge
clean-up. The matrix effect and the extent of ionization suppression for the
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
100
basic HLB SPE procedure and HLB+silica procedure were evaluated by
comparing MS peak areas of the analyte standards and standards spiked after
extraction of raw sewages (Table 5.1). It can be seen that ME% for both PFOS
and PFOA were below 50%, indicating solid phase extraction alone was
insufficient to remove matrix components. During HLB SPE procedure, 40%
methanol in Milli-Q was applied to wash the cartridge and remove the matrix
components. However, this simple washing cannot effectively remove the
interfering compounds. Without further clean-up, the co-eluting matrix
constituents would lead to strong suppression of electrospray ionization and
result in large inaccuracies, which were observed by other studies on
perfluorinated compounds in environmental matrices (Boulanger et al., 2005;
van Leeuwen et al., 2006; Berger et al., 2005). Moreover, recoveries (RE%:
<50%) were significantly affected by the matrix effect in water samples due to
the ionization suppression even though this HLB SPE procedure can achieve
more than 90% recoveries (98.4% for PFOS and 93.8% for PFOA) for PFCs
spiked Mill-Q water (data not shown).
(a) (b)
Figure 5.1 LC-MS-MS chromatograms of PFOS and PFOA in the raw sewage extracted by (a) HLB SPE and (b) HLB+silica.
1 2 3 4 5 6 7 8 9 10 11Time min
0
500
1000
15009.60
9.865.672.54 11.00.06 5.133.79 6.936.31 8.557.34 8.09 8.724.040.48 10.301.781.46 3.542.21
1 2 3 4 5 6 7 8 9 10 110
500
1000
1500
2000
25009.96
9.89
9.75
8.58 8.952.03 8.010.17 2.71 7.12 10.362.920.46 1.65 11.16.505.60 5.780.88 3.97 4.85 5.483.63 4.28
1 2 3 4 5 6 7 8 9 10 1
1 2 3 4 5 6 7 8 9 10 1Time, min
10
1000
2000
3000
4000
9.95
9.758.63 9.19 10.598.11 8.467.796.930.50 4.862.961.64 5.334.641.76 4.201.27 3.582.36 6.36
10
500
1000
1500
2000
2490 9.61
9.41 15.90 6.571.35 5.332.92 8.664.77 10.434.02 4.622.56 8.127.31 10.606.69 7.392.301.720.76 3.09
Max.2750 cps.
Max.2490 cps.
Max.4410 cps.
PFOA
PFOS
PFOA
PFOS
9.60 min 9.61 min
Max.1580 cps.
9.95 min i
9.96 min iIn
tens
ity(c
ps.)
Time (min) Time (min)
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
101
In order to further clean up the extracts, silica cartridge was applied to reduce
the co-eluting interfering compounds. As can be seen, ME% (>70%) and RE%
(>67%) increased significantly for both PFOS and PFOA. It suggests that the
partition of analytes (PFOS and PFOA) and interfering compounds are
different between dichloromethane/methanol mixture and silica column.
Consequently, substantial amount of interfering compounds would be retained
by silica cartridge, while PFOS and PFOA would be eluted by mixture of
dichloromethane/methanol (Benijts et al., 2004; Powley et al., 2005). After
silica cartridge clean-up, the coefficient of variation (CV) decreased more than
44% (PFOS) and 34% (PFOA), indicating precision of the analysis increased
due to the reduced matrix effect. However, matrix suppression still existed
(ME%<90%) even though the additional silica cartridge clean-up had been
applied. Therefore, internal standardization was used to compensate for the
remaining matrix suppression.
Table 5.1 Influence of sample clean-up and internal standardization on ME% and RE% (n=5).
Analyte HLB SPE HLB+silica HLB+silica (Internal Standard)
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
PFOS 48.1
(8.4)
47.1
(9.1)
78.3
(4.1)
74.6
(4.7)
96.3
(2.9)
97.0
(3.2)
PFOA 37.8
(9.5)
36.3
(10.3)
71.2
(6.2)
67.1
(6.6)
93.2
(2.6)
90.3
(4.3)
5.2.2 Effect of internal standards on matrix effects
Internal standards have been shown to be an effective tool to compensate for
the matrix effect (Petrovic et al., 2005; Benijts et al., 2004; Matuszewski et al.,
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
102
2006). Ideally, isotopically labeled perfluoroalkyl internal standards are
prefered for negating ionization effects because they will have the same
retention times as their natural analogues. Moreover, both analytes and
internal standards are affected by the co-eluting matrix to the same extent.
Therefore, isotopically labeled perfluoroalkyl internal standards offer the best
solution. However, the use of stable isotopes of perfluoroalkyl is cost
prohibitive and commercial availability is often limited. For example,
perfluorooctanoic acid (1, 2-13C) is much expensive, while the required stable
radioactive perfluorinated acid standards (e.g., 14C PFOS) for toxicological
and environmental fate studies are not available. In many cases, structural
analogues such as PFDoA and PFEES which show similar behavior in the
source to compensate for matrix effect, have been used as internal standards to
determine the concentration of perfluorinated compounds in various
environmental samples with acceptable recovery (Higgins et al., 2005; Moody
et al., 2001; Benijts et al., 2004; Tseng et a., 2006). In this study, PFDoA and
PFEES served as internal standards for PFOA and PFOS, respectively. As can
be seen, they enhanced the ME% from 78.3% to 96.3% for PFOS and from
71.2% to 93.2% for PFOA (Table 5.1). The coefficient of variation (CV) was
greatly decreased by applying internal standardization with the value below
5%. In addition, a higher recovery (>90%) was achieved compared to that of
around 70% without internal standardization. The compensation effect of
internal standards was also observed in nine pharmaceuticals analysis, by
which ME% was increased by 20-78%. Also, RE% of those nine
pharmaceuticals were brought close to 100% (van de Steene et al., 2006). It
indicates similar behaviors of internal standards during the analysis
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
103
compensate for the matrix effect, resulting in minimized matrix effect and
maximized recovery.
5.2.3 Detection of PFOS and PFOA in water and sludge samples
Various types of water samples, including reservoir water, river water, treated
effluent and influent of WWTP B were collected and analyzed to evaluate the
matrix effect on the LC-ESI-MS/MS method (Table 5.2). As can be seen,
ME% was close to 100% for all water samples, indicating no noticeable
matrix effect was observed. Although influent of WWTP B (raw sewage)
showed a slightly lower ME% (more suppression) in comparison with other
three types of water samples, sample origin had limited impact on ME%. It
suggests that HLB together with silica cartridge method can effectively
remove most of the interfering matters which cause matrix suppression on the
detection of PFOS and PFOA by LC/MS/MS. In contrast, without a clean-up
procedure Boulanger et al. (2005) observed significant matrix suppression in
the influent of a wastewater treatment plant, while no signal suppression or
enhancement was observed in surface water and treated effluent. This
confirms that raw sewage has a very strong matrix effect and additional clean-
up procedure is necessary for the detection of PFOS and PFOA by LC/MS/MS.
Table 5.2 Influence of sample origin on ME% and RE% with internal standard (n=3).
Analyte Reservoir Water River Water Treated Effluent Influent of WWTP B
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
PFOS 96.2
(2.5)
97.0
(2.8)
98.3
(2.3)
98.0
(3.4)
95.8
(2.3)
95.7
(3.5)
92.8
(3.3)
93.2
(3.1)
PFOA 95.8
(2.6)
92.7
(2.9)
97.5
(2.2)
95.8
(2.8)
95.1
(2.3)
92.8
(2.4)
91.9
(3.6)
89.2
(4.8)
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
104
The developed silica cartridge clean-up method was applied to various sludge
samples collected from wastewater treatment plant. It was observed that ME%
was slightly lower than those of surface water samples, indicating stronger
matrix effects occurred in sludge samples (Table 5.3). The recovery was in the
range of 85.4-96.6% for PFOS and 81.3-83.8% for PFOA, which were higher
than those reported by Higgins et al. (2005). It suggests that the co-extraction
of lipids and other interfering matters in the extracts of various sludge samples
can be effectively removed by silica cartridge. Simcik et al. (2005) also found
that fluorous silica column could remove the interfering compounds to clean
up the SPE extracts. Moreover, this clean-up method can greatly improve the
detection limit (up to 10 times) as the extracts are concentrated instead of
being diluted after basic solid phase extraction.
Table 5.3 ME% and RE% with internal standard by application of clean-up method on sludge samples (n=3).
Analyte Digested Sludge Activated Sludge Primary Sludge ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
ME% (CV)
RE% (CV)
PFOS 92.2
(3.3)
85.4
(4.8)
93.1
(2.8)
96.6
(3.3)
95.4
(2.9)
88.1
(2.1)
PFOA 91.8
(3.0)
81.3
(2.9)
92.5
(3.3)
83.8
(6.2)
93.8
(3.0)
83.6
(4.8)
5.3 Summary
High performance liquid chromatography (HPLC) with tandem mass
spectrometry (MS/MS) is considered the method of choice for the quantitative
determination of perfluorinated compounds in environmental matrices.
Chapter 5-Development of Post Extraction Clean-up Method For PFOS/PFOA Determination in Wastewater and Sludge Samples
105
However, co-eluting matrix components may reduce or enhance the ion
intensity of the analytes and affect the reproducibility and accuracy of the
LC/MS/MS analyses. This study evaluated matrix effect on PFOS and PFOA
in raw sewage by comparing MS responses of standards and those of the same
known amount of analytes in post-extraction spiked samples. Strong matrix
suppression (ME%<49% and RE%<48% for raw sewage) confirmed that
further extracts clean-up after basic solid phase extraction was necessary. A
silica cartridge clean-up method was successfully developed to remove
remaining co-eluting interfering compounds in raw sewage, by which ME%
and RE% were increased to >71% and >67%, respectively. The application of
internal standards further compensated for matrix effect and brought the ME%
and RE% close to 100%, indicating minimal matrix effect was achieved
without significant loss of analytes. Moreover, internal standards improved
reproducibility by significantly decreasing coefficient of variation. The
developed LC/MS/MS detection method was applied to different water
samples and sludge samples. For sludge samples, the recovery was in the
range of 85.4-96.6% for PFOS and 81.3-83.8% for PFOA, respectively.
Results showed that this silica cartridge clean-up method can effectively
remove co-eluting matrix components in various environmental matrices.
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
106
CHAPTER 6 BEHAVIOR OF PFOS AND
PFOA IN SEWAGE TREATMENT PLANTS
6.1 Introduction
Effluent from wastewater treatment plants is one of the important routes for
the introduction of certain organic contaminants into aquatic ecosystems.
Several studies have described effluents of sewage treatment plants (STPs) as
an important source for metals, chlorinated organic compounds to aquatic
environments (Irvine et al., 1993; Loganathan et al., 1997). Perfluorinated
compounds (PFCs), a class of emerging environmental pollutants, have been
widely used for the last 50 years in industrial and commercial applications,
such as coatings, shampoos, electroplating, fire-fighting foams, stain repellants
for furniture and carpets. Widespread usage of these compounds has been
attributed to their contamination in wastewaters. Occurrence of two PFCs,
perfluorooctane sulfonate (PFOS, C8F17SO3-) and perfluorooctanoic acid
(PFOA, C7F15COO-) has been reported in STPs (Kissa, 2001). Various
contamination levels were observed in the influent and effluent of municipal
STPs in Iowa City (Boulanger et al., 2005), in Kentucky and Georgia,
respectively (Loganathan et al., 2007), in 10 national wide municipal STPs in
U.S.A (Schultz et al., 2006a) and in the effluent of 6 U.S.A cities (Sinclair and
Kannan, 2006). The discharge of municipal wastewater effluent is therefore
one of the major routes for introducing PFOS and PFOA that are used in
domestic, commercial and industrial settings into aquatic environment.
Recently, there has been increasing concern about the fate and behavior of
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
107
PFOS and PFOA in sewage treatment plant due to their biotic and abiotic
persistence and chronic toxicity (Prevedouros et al., 2006; Hof et al., 2004;
3M, 2003). Sinclair and Kannan (2006) observed that mass flow of PFOS and
PFOA in aqueous phase increased significantly after secondary treatment in a
sewage treatment plant with industrial influence, while no increase in mass
flow of PFOA was found in another sewage treatment plant with no industrial
influence. Furthermore, Schultz et al. (2006b) identified the fate and behavior
of these two compounds in both aqueous phase and solid phase (sludge)
during each step of municipal wastewater treatment plant. It was observed that
mass flow of PFOS or PFOA either increased or remained consistent,
indicating conventional activated sludge process can not effectively remove
these compounds. However, these investigations were conducted at different
STPs with different influents. Different influent of STP would significantly
affect the behavior pattern of PFOS or PFOA since their precursors in the
influent could be biodegraded to PFOS or PFOA in the activated sludge
treatment processes. Therefore, it is desirable to investigate behavior of PFCs
in various activated sludge treatment processes which receive the same raw
sewage.
Even though PFOS or PFOA is not biodegradable, their precursors could be
biodegraded and thus affect the mass flow of PFOS or PFOA in STP. For
example, it was found that secondary treatment by activated sludge
significantly increased the mass flows of PFOS and PFOA, probably resulting
from biodegradation of precursor compounds such as fluorotelomer alcohols
(Sinclair and Kannan, 2006). Currently, conflicting information is obtained
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
108
about the behavior patterns of PFCs in STPs. Some studies reported increase
in mass flow of PFCs in biological processes, while other studies observed
unchanged PFCs’ mass flow in the activated sludge treatment. Except for
effect of influent of STP, sludge retention time (SRT), a commonly used
parameter for sewage treatment plant design and operation, could also be an
important factor affecting the behavior of PFOS and PFOA in STPs. Clara et
al. (2005) found that the degradation of the micropollutants, such as endocrine
disrupting compounds and pharmaceuticals, was dependent on the SRT in the
activated sludge process since the SRT determines the enrichment of the
microorganism that is able to degrade the micropollutants. Therefore, behavior
pattern of perfluorinated compounds may be different in the conventional
activated sludge process operated with different SRT. However, no data is
available about the effect of SRT on the behavior pattern of PFOS and PFOA
in the activated sludge process.
The objective of this part of study was to compare the behavior of PFOS and
PFOA in full-scale conventional activated sludge processes and membrane
biological reactor, as well as in an activated sludge process operated with a
short SRT. This is the first study to investigate the effect of SRT on the
behavior of these two compounds in the activated sludge process. Furthermore,
seasonal variation in the concentrations of PFOS and PFOA in sewage
treatment plants was studied. In order to achieve these, aqueous and solid
samples were taken from each treatment unit of STPs, A and B.
6.2 Results and Discussion
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
109
6.2.1 PFOS/PFOA in wastewater
Sampling strategy can affect the concentrations of the analytes measured in
sewage treatment plants. The 24-h composite sample is appropriate to
represent average concentrations over 24 h in the wastewater streams of STPs.
Grab samples collected in this study, which could have been collected at high
or low flow period, may increase the variation in concentrations. Therefore,
care must be taken when concentrations of PFOS and PFOA measured were
compared.
1 2 3 4
16151312 14109 11
5 6 7 8PEAT
SE
Inf 0.0
100.0
200.0
300.0
400.0
500.0
600.0
PFO
S co
ncen
tratio
ns in
was
tew
ater
(ng/
L)
1.Dec 06-CAS1 2.Oct 06-CAS1 3.Mar 07-CAS1 4.Sep 07-CAS15.Dec 07-CAS1 6.Oct 06-LTM 7.Dec 06-LTM 8.Mar 07-LTM9.Mar 07-MBR 10.Sep 07-MBR 11.Dec 07-MBR 12.Oct 06-CAS213.Dec 06-CAS2 14.Mar 07-CAS2 15.Sep 07-CAS2 16.Dec 07-CAS2
Figure 6.1 PFOS concentrations in wastewater of STP A (CAS1, LTM and MBR) and STP B (CAS2). Inf: influent; AT: aeration tank; PE: primary clarifier effluent; SE: secondary clarifier effluent.
The measured PFOS and PFOA concentrations in wastewater samples from
STPs A and B are shown in Figures 6.1 and 6.2. PFOS and PFOA were
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
110
detected in all wastewater samples collected from STP A and B. PFOS was
observed at 5.3 - 29.8 ng/L in STP A, which are comparable to those measured
in the effluent of 4 STPs receiving domestic and commercial sewage (Sinclair
et al., 2006). However, much higher concentration of PFOS (48.1 - 560.9 ng/L)
was detected in STP B receiving 60% industrial wastewater. These measured
PFOS concentrations are also much higher than those in the influents and
effluents of 10 STPs mainly receiving domestic and commercial sewage in
USA (Schultz et al., 2006a), but much lower than those in the effluents of
Decatur which receives influent from fluorochemical manufacture or industry.
Nevertheless, the comparable concentration was observed in a wastewater
treatment plant in Cleveland (3M, 2001), which has no known fluorochemical
sources. It suggests that industrial sewage can contain a large amount of PFOS
in comparison with domestic and commercial sewage even though there is no
known source of fluorochemical exposure.
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
111
16151312 14109 115 6 7 8
1 2 3 4PEAT
SE
Inf 0.0
200.0
400.0
600.0
800.0
1000.0
1200.0
PFO
A c
once
ntra
tions
in w
aste
wat
er (n
g/L)
1.Dec 06-CAS1 2.Oct 06-CAS1 3.Mar 07-CAS1 4.Sep 07-CAS15.Dec 07-CAS1 6.Oct 06-LTM 7.Dec 06-LTM 8.Mar 07-LTM9.Mar 07-MBR 10.Sep 07-MBR 11.Dec 07-MBR 12.Oct 06-CAS213.Dec 06-CAS2 14.Mar 07-CAS2 15.Sep 07-CAS2 16.Dec 07-CAS2
Figure 6.2 PFOA concentrations in wastewater of STP A (CAS1, LTM and MBR) and STP B (CAS2). Inf: influent; AT: aeration tank; PE: primary clarifier effluent; SE: secondary clarifier effluent.
PFOA was the predominant contaminant in STP A, which was measured at
11.2 - 138.7 ng/L. Slight lower and comparable PFOA concentration was
reported in the influents and effluents of 10 STPs in USA (Schultz et al.,
2006a). However, Sinclair et al. (2006) observed much higher concentration in
4 STPs receiving domestic and commercial sewage. This suggests that
commercial sewage could be a significant source of PFOA, which includes a
wide range of sources (hospitals, shopping malls and so on) and provides more
variable amount of PFOA. In addition, the predominance of PFOA over other
perfluorinated compounds was observed in other STPs (Sinclair et al., 2006;
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
112
Loganathan et al., 2007). In STP B, PFOA concentration was detected in the
range of 31.8 – 1,057.1 ng/L, which is much higher than those of STP A and
those in a sewage treatment plant with similarly 60% industrial influent
(Sinclair et al., 2006). It suggests the effect of industrial influent on PFOA
concentration is dependent on the composition of the sewage that enters the
sewage treatment plants. Moreover, in this study, higher variation in
concentrations of both PFOS and PFOA was observed in STP B than those in
STP A, indicating industrial influent can result in high concentration variation.
6.2.2 Seasonal variation
In STP A, PFOS concentration in influent of dry season showed statistically
significant difference from the wet season (p=0.003), while PFOA had no
such significant difference (p=0.157) (Figure 6.3). It seems that PFOS
concentrations noticeably decreased during wet season, while there was no
discernable decrease in PFOA concentrations in surface waters. The presence
of PFCs in rainfall indicates rainfall significantly affect their concentrations in
surface water. PFOS have been observed at a low concentration (0.59 ng/L) in
the precipitation, while significant higher PFOA concentrations were reported
in rainwater. Kallenborn et al. (2004) reported that PFOA was the
predominant PFCs measured in rainwater samples from Sweden Finland with
the greatest concentrations (11 ng/L and 17 ng/L, respectively). Scott et al.
(2006) also reported relatively high PFOA concentrations (<0.1-89 ng/L) in
the rainwater samples from U.S.A and Canada. Based on the limited data, it
seems that PFOS concentration in rainwater is lower than that of PFOA, which
leads to their different seasonal variations in surface water. Furthermore,
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
113
0
5
10
15
20
25
30
35
1 2 3 4 5 6 7 8 9 10 11
PFOS Concentration (ng/L)
(a) (b)
Figure 6.3 Seasonal variations in influent concentrations of (a) PFOS and (b) PFOA in STPA. 1: Oct 06 (CAS1), 2: Mar 07 (CAS1), 3: Sep 07 (CAS1), 4:
Mar 07 (MBR), 5: Sep 07 (MBR), 6: Oct 06 (LTM), 7: Mar 07 (LTM), 8: Dec 06 (CAS1), 9: Dec 06 (LTM), 10: Dec 07 (CAS1), 11: Dec 07 (MBR)
runoff could also be the potential PFCs sources during rainy weather. Rainfall
may pick up and carry away PFCs pollutants (such as oil, fire-fighting foam)
when it moves over and through the ground, which leads to the occurrence of
nonpoint source pollution (NPS) of PFCs. Zushi et al. (2008) observed that
some PFCs concentrations in a river did not decrease with the increase of river
flow rate during the rainy weather possibly due to the NPS of PFCs. As a
result, the decreased PFOS concentrations in surface water may result in their
decrease in wastewater correspondently after surface water is treated by water
treatment plants and then subsequently utilized by various consumers.
However, in STP B no significant difference between dry season and wet
season for both PFOS (p=0.520) and PFOA (p=0.274) was observed despite
slightly lower concentration was observed in wet season (Figure 6.4). It is
likely that high concentration of industrial influent override the effect of
dilution by rainstorm. In comparison, no large-magnitude seasonal variation in
concentrations of perfluorinated compounds was found among spring, summer,
0
20
40
60
80
100
120
1 2 3 4 5 6 7 8 9 10 11
PFOA Concentration (ng/L)
Dry season Wet
season Wet season
Dry season
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
114
fall and winter seasons in two municipal sewage treatment plants (Loganathan
et al., 2007).
Dec07
Dec06
Sep07
Mar07
Oct06
0
50100
150200
250
300350
400450
500
PFOS
Con
cent
rati
on (
ng/L
)
(a) (b)
Figure 6.4 Seasonal variations in influent concentrations of (a) PFOS and (b) PFOA in STP B.
6.2.3 Mass flow in aqueous sample during treatment
The average mass flow was calculated by multiplying average PFOS and
PFOA concentrations in aqueous and solid phase by the daily average flow of
each treatment unit (Table 6.1). Total solid waste is the daily mass of PFOS or
PFOA associated with primary sludge and waste activated sludge. Related
information on the wastewater and solid stream was obtained from individual
STPs. It is worthy to note that sampling strategy can affect the concentrations
of the analytes measured in sewage treatment plants. Specially, grab sample,
which could have been collected at high or low flow period, may increase the
variation in concentration. As the concentration was based on grab samples, it
would result in additional variation in mass flow besides the error of
measurement. Therefore, only change of more than 30% in mass flow would
be taken into consideration in this study (Loganathan et al., 2007).
Dec07
Dec06
Sep07
Mar07
Oct06
0
100
200
300
400
500
600
700
800
900
PFOA
Con
cent
rati
on (
ng/L
)
Wet season
Dry season Wet
season
Dry season
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
115
CAS1, MBR and LTM, which are different treatment processes, receive same
raw sewage, while CAS2 receives different raw sewage. No significant change
in mass flow of PFOS (-24.5%-16.0%) was observed in CAS1, MBR and
LTM. It is known that PFOS or PFOA can not be biodegraded by activated
sludge process (Lange, 2002). A reduction in mass flow of PFOS or PFOA is
neither expected nor observed (Schultz et al., 2006b; Sinclair et al., 2006)
since biodegradation of precursors such as fluorotelomer alcohols (FTOHs),
perfluoroalkyl phosphates (PAPS), or fluorotelomer sulfonates (FTSs) during
activated sludge treatment process are likely sources of increase of PFOS and
PFOA. Specially, it is known that 2-(N-ethyl-perfluorooctanesulfonamido)
ethanol (N-EtFOSE alcohol) and 2-(N-ethyl perfluorooctane sulfonamido)
acetic acid (N-EtFOSAA) can be biotransformed to PFOS during activated
sludge treatment (Boulanger et al., 2005; Lange, 2000 and 2002). Our result
suggests that either raw sewage of STP A did not introduce the precursors of
PFOS or no significant biotransformation occurred during these processes.
However, significant increase in mass flow of PFOS (mean 94.6%) was
observed in CAS2, indicating biodegradation of precursors may occur during
the secondary treatment. As CAS1 is running with the similar operational
parameters (e.g. SRT and HRT) compared to CAS2, the results suggest that no
precursors of PFOS be likely contained in the raw sewage of STP A.
Table 6.1 Mass flow (mg/d) of PFCs in influent, effluent and solid waste in CAS1, CAS2, LTM and MBR.
Site Date
PFOS PFOA
influent (aqueous)
influent (particu-
late) Effluent
Solid waste (total)
Mass change
Mass change
(%)
influent (aqueous)
influent (particu-
late)
Effluent (aqueous)
Solid waste (total)
Mass change
Mass change
(%)
CAS 1 Oct 06 3722 487 3372 1014 -177 -4.2% 4355 ND 6496 292 -2433 -55.9% Dec 06 3782 348 3540 534 57 1.4% 6782 ND 7565 403 -1186 -17.5% Mar 07 3396 207 3009 607 -13 -0.4% 6584 ND 9465 501 -3382 -51.4%
MBR Mar 07 321 16 261 104 -29 -8.6% 451 ND 739 58 -345 -76.6%
LTM Oct 06 1076 130 1125 376 -295 -24.5% 1457 ND 1221 164 72 4.9% Dec 06 1002 100 729 240 133 12.1% 1950 ND 2346 165 -561 -28.8% Mar 07 1361 165 1022 260 244 16.0% 1378 ND 1544 180 -346 -25.1%
CAS 2 Oct 06 39834 3862 66318 7870 -30492 -69.8% 11583 375 15867 800 -4709 -39.4% Dec 06 20550 3395 34894 6200 -17150 -71.6% 11607 646 28543 1351 -17641 -144.0% Mar 07 16072 2214 32308 12028 -26050 -142.5% 39360 1784 47191 1213 -7260 -17.6%
ND: not detectable; Mass change=Influent (aqueous)+Influent (particulate)-Effluent-Solid waste (total); Mass change (%)=Mass change/[Influent (aqueous)+Influent (particulate)]
116
A C
hapter 6-Behavior of PFO
S and PFOA
in Sewage Treatm
ent Plants
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
117
Mean mass flow of PFOA increased by 41.6% (17.5%-55.9%) and 76.6% in
CAS1 and MBR, respectively, while PFOA mass flow remained unchanged
after the treatment of LTM with a SRT of 3.5 d. During activated sludge
treatment some precursors, especially 8:2 FTOH, have been shown to
biotransform into PFOA (Dinglasan et al., 2004; Wang et al., 2005). This
suggests that no noticeable biodegradation of PFOA precursors can occur in
LTM though their presence in the raw sewage has been demonstrated by mass
increase in CAS1 and MBR. Similarly, Clara et al. (2005) found that no
biodegradation of micropollutants, such as endocrine disruptors compounds
(EDCs) or pharmaceuticals could occur when the activated sludge treatment
system (CAS or MBR) was operated with a SRT, which was lower than a
critical SRT (e.g. approx. 10 days for estrogens, 17b-estradiole, estrone and
bisphenol-A). Only at a higher SRT which is more than the critical one, the
microorganisms that biodegrade certain micropollutants are able to be
detained and enriched in the system. It seems that the SRT of LTM is lower
than the critical one, resulting in no biodegradation of precursors. Furthermore,
mass flow of PFOA and PFOS increased 17.6-144.0% and 69.8-142.5% after
the secondary treatment of CAS2, respectively. It suggests that the precursors
of PFOS and PFOA be biodegraded at a SRT of ~12 days, which may be
higher than the critical SRT. Our results confirm that change in mass flow of
PFOS and PFOA may be determined by both the presence of precursors and
operating SRT of the activated sludge system. SRT could thus be an important
operational parameter that affects the behavior pattern of PFOS and PFOA.
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
118
PFOS and PFOA mass change after the treatment of primary clarifier in STP
A and B are shown in Figure 6.5. As can be seen, mass flow change was in the
range of -27.3%-6.7% for PFOS and -35.7%-12.5% for PFOA, respectively. It
suggests that there is no discernable mass change after the treatment of
primary clarifier for both PFOS and PFOA. In addition, their mass flow in the
inflow and outflow of primary clarifier were equivalent at the 95% CI
(confidential interval). It seems primary clarifier has no noticeable effect on
the mass flow of PFOS and PFOA. Similarly, Schultz et al (2006b) observed
that only 10% (PFOS) and 0.1% (PFOA) reduction in mass flow occurred due
to their sorption onto primary sludge.
Dec07
Dec06
Sep07
Mar07
Oct06
-40%
-30%
-20%
-10%
0%
10%
20%
30%
Change of mass flow (%)
PFOSPFOA
(a) (b) Figuure 6.5 Change of mass flow after primary treatment in (a) STP A and (b) STP B
6.2.4 PFOS/PFOA in sludge
PFOS and PFOA were detected in all sludge samples except for one sample
from STP A which was below LOQ of PFOA (Figure 6.6 and 6.7). PFOS was
observed at 13.1 - 46.0 ng/g dw in STP A, while 3.2 - 53.6 fold higher
concentration (145.1 - 702.2 ng/g dw) was observed in STP B. Similarly,
while higher PFOA concentration (18.0 - 69.0 ng/g dw) in STP B was
detected.lower PFOA concentration (<5.0 - 44.2 ng/g dw) was observed in
Dec07
Dec06
Sep07
Mar07
Oct06
-30%
-20%
-10%
0%
10%
20%
30%
Change of mass flow (%) PFOS
PFOA
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
119
6 7 8 9 10
1 2 3 4 5
ASPS
SS
Inf
DS
0
100
200
300
400
500
600
700
800
PFO
S co
ncen
traio
ns in
slud
ge (n
g/g
dw)
1.Dec 06-CAS1 2.Oct 06-CAS1 3.Mar 07-CAS1 4.Oct 06-LTM (AS)5.Dec 06-LTM (AS) 6.Mar 07-LTM (AS) 7.Mar 07-MBR (AS) 8.Oct 06-CAS29. Dec 06-CAS2 10. Mar 07-CAS2
Figure 6.6 PFOS concentrations in sludge samples from STP A and STP B. Inf: influent particulate; PS: primary sludge; AS: activated sludge; SS: secondary clarifier sludge; DS: digester sludge.
STP A. In comparison, Higgins et al. (2005) reported PFOS and PFOA
concentrations were in the range of 14.4 – 2,610.0 and n.d - 29.4 ng/g dw in
the sludge samples of 8 STPs, respectively. In addition, our results suggest
that high concentration in wastewater lead to high concentration in sludge
which is due to the partition between aqueous and solid phases. Nevertheless,
due to its higher partition coefficient in comparison with PFOA (Higgins et al.,
2006), PFOS was dominant in sludge samples of both STPs A and B even
though PFOA was dominant in aqueous samples of STP A.
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
120
In terms of PFOS or PFOA concentrations, there was no noticeable difference
among activated sludge samples collected from LTM, MBR and CAS1. It
seems their concentrations in sludge are more relevant to the aqueous
concentration than sludge characteristics, which are affected by the SRT (Liao
et al., 2001; Ng et al., 2005).
6 7 8 9 10
1 2 3 4 5
ASPS
SS
Inf
DS
0
10
20
30
40
50
60
70
PFO
A c
once
ntra
tions
in sl
udge
(ng/
g dw
) 1.Dec 06-CAS1 2.Oct 06-CAS1 3.Mar 07-CAS1 4.Oct 06-LTM (AS)5.Dec 06-LTM (AS) 6.Mar 07-LTM (AS) 7.Mar 07-MBR (AS) 8.Oct 06-CAS29. Dec 06-CAS2 10. Mar 07-CAS2
Figure 6.7 PFOA concentrations in sludge samples from STP A and STP B. Inf: influent particulate; PS: primary sludge; AS: activated sludge; SS: secondary clarifier sludge; DS: digester sludge.
The partition coefficient Kd for primary sludge and activated are estimated
based on the data obtained by dividing PFCs concentrations in primary sludge
or secondary sludge by their aqueous concentration in primary effluent or
secondary effluent (Table 6.2). Kd value of PFOS was in the range of 894-
2,237 L/kg (primary sludge) and 720-2,324 L/kg (activated sludge), while
significant lower Kd value of PFOA was observed at 188-597 L/kg (primary
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
121
sludge) and 201-513 L/kg (activated sludge), respectively. The mean Kd value
of PFOS was more than 3 times higher than that of PFOA, indicating much
more amount of PFOS adsorbed onto sludge as compared to PFOA. High
variations in Kd value for PFOS and PFOA may be caused by different
retention time of aqueous and solid streams in primary and secondary
clarifiers. In addition, it seems that there is no significant difference between
Kd values in primary sludge and activated sludge. Based on the data on
organic carbon ƒoc
Table 6.2 Calculated partition coefficient Kd
, calculated activated sludge log Koc values (partition
coefficient for the compound onto a hypothetical pure organic carbon) were
2.98-3.49 for PFOS and 2.43-2.83 for PFOA, respectively. In contrast, lower
organic carbon-normalized log Koc values were reported by 3M (2000) (2.57-
3.1 for PFOS) and DuPont (2003) (1.9-2.17 for PFOA) determined on
sediments.
in primary sludge and activated sludge.
Compound Sludge type Kd (L/kg ) Range Mean (±SD)
PFOS Primary sludge 894-2237 1408 (±481) Activated sludge 720-2324 1645 (±511)
PFOA Primary sludge 188-597 405 (±149) Activated sludge 201-513 368 (±106)
6.3 Summary
PFOS and PFOA were detected in all aqueous samples collected from STP A
and STP B, ranging from 5.3 - 560.9 ng/L and 11.2 - 1057.1 ng/L, respectively.
In sludge of STPs A and B, PFOS and PFOA concentrations were in the range
of 13.1 – 702.2 ng/g dw and <5.0 - 69.0 ng/g dw, respectively. Due to
industrial influence, PFOS and PFOA were observed at higher concentration
in aqueous and sludge samples in STP B than that of STP A.
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
122
Significant increase in mass flow of PFOS (mean 94.6%) was observed in
CAS2, while it remained consistent after secondary treatment in CAS1. This is
likely due to no occurrence of PFOS precursors in the raw sewage of STP A.
Mean mass flow of PFOA increased 41.6% in CAS1, 67.0% in CAS2 and
76.6% in MBR, while it remained unchanged after the treatment of LTM.
Different behavior pattern of these two compounds were found in LTM, an
activated sludge process operated at a relatively short SRT. The findings
suggest that change in mass flow of PFOS and PFOA in secondary sludge
treatment may be determined by the absence/presence of precursors and
operating SRT of the activated sludge system.
Compared with STP A, higher concentrations of PFOS and PFOA were
detected in STP B receiving 60% industrial wastewater. It suggests that
industrial sewage contain a large amount of PFOS and PFOA in comparison
with domestic sewage even though there was no known source of
fluorochemical exposure. Furthermore, industrial influent caused little
seasonal variation in concentrations of PFOS and PFOA. Between dry and wet
seasons, seasonal variation of PFOS was observed in STP A, while PFOA had
no significant difference in both STP A and STP B. PFOS concentration in
rainwater observed by other studies was lower than that of PFOA, which could
lead to their different seasonal variations in surface water. It is also likely that
NPS of PFCs occurred in wet season, which would contribute to consistent
PFOA concentrations in surface waters and subsequently resulted in
Chapter 6-Behavior of PFOS and PFOA in Sewage Treatment Plants
123
indiscernible variation in PFOA concentrations in the wastewaters between
dry and wet seasons.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
124
CHAPTER 7 PFOS/PFOA REMOVAL BY
HYBRID PAC-MBR PROCESS
7.1 Introduction
The discharge of municipal wastewater effluent is one of the major routes for
introducing PFOS and PFOA that are used in domestic, commercial and
industrial settings into aquatic environment. They were detected in the influent
and effluent of municipal WWTPs in Iowa City (Boulanger et al., 2005), in 10
national wide municipal WWTPs in U.S.A (Schultz et al., 2006a) and in the
effluent of 6 U.S.A cities (Sinclair et al., 2006). High PFCs concentrations
were observed in the effluent of fluorochemical manufacture or related
industries (3M, 2001). For an example, Tang et al. (2006) reported PFOS
concentration of 1650 mg/L in the effluent of semiconductor manufacturing.
They were considered stable and persistent in environment without natural
degradations (Prevedouros et al., 2006; 3M, 2003). Also, Lange (2002)
observed that PFOS and PFOA were not degradable by activated sludge.
Studies on fate and behavior of these pollutants in WWTPs implied that they
can not be effectively removed by biological treatment process (Sinclair et al.,
2006; Schultz et al., 2006b).
Although there is no maximum allowable concentration of PFCs in the
discharge of STPs, PFOS and PFOA, candidates for persistent organic
pollutants (POPs), are reported to have adverse effect on the human health.
Since PFOS and PFOA can not be effectively removed by conventional STPs
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
125
and drinking water treatment plants, it is urgent to develop a new technology
to remove these compounds effectively at low cost for the wastewater
treatment. Various physico-chemical treatment processes including adsorption
(Ochoa-Herrera et al., 2008), sonochemical treatment (Moriwaki et al., 2005),
reduction with zero-valent iron in subcritical water (Hori et al., 2006) and
membrane filtration (Tang et al., 2006) have been studied to remove these
compounds. Activated carbon adsorption is one of the most promising
methods to remove PFCs in aqueous stream due to the effectiveness and low
cost. It was reported that PFCs were effectively removed by adsorption onto
the activated carbon at high and low equilibrium concentrations (Ochoa-
Herrera et al., 2008; Qiu et al., 2006). Ochoa-Herrera et al. (2008) reported
that PFOS could be effectively removed by granular activated carbon (GAC)
and Freundlich isotherm was applicable at high and low equilibrium
concentrations. In contrast, Yu et al. (2009) studied the feasibility of using
powder activated carbon (PAC), granular activated carbon (GAC) and anion-
exchange resin (AI400) to remove PFOS and PFOA from water. It was
observed that adsorption isotherms of PFOS and PFOA fitted Langmuir
isotherms better than Freundlich isotherm. Qiu et al. (2006) also reported that
GAC was able to effectively remove PFOS and PFOA. In 4 hours 93% PFOS
and 99% PFOA in pure water at ppb level were adsorbed onto GAC. Based on
the information available, it seems that PFCs compounds can be effectively
removed by adsorption onto the activated carbon in water solution without the
presence of dissolved organic matters (DOMs).
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
126
The hybrid PAC-MBR technology integrates adsorption and biodegradation of
organic matter with membrane filtration in one unit, which has been proved to
be a simple and highly efficient way to remove compounds in wastewater. In
particular, PAC addition increases the removal of organic matters with low
molecular weight by adsorption; it also serves as a supporting medium for
attached bacterial growth (Kim et al., 1998). Even though MBR may not be
able to significantly remove PFOS and PFOA due to similar biodegradation
and adsorption behavior in activated sludge system, combination of MBR and
PAC technologies could effectively remove these compounds while adsorption
onto PAC occurs. However, there is no data available on the removal of PFCs
in the hybrid PAC-MBR process till now.
Effluent from biological wastewater treatment contains complex and
heterogeneous soluble organic matter, which is so called effluent organic
matter (EfOM). EfOM is highly heterogengeneous, containing molecular of
various molecular weight ranging from the simple compounds such as acetic
acid to very complex polymers. The composition of EfOM is a combination of
those of natural organic matter (NOM), soluble microbial products (SMPs),
and trace harmful chemicals. Most of the NOM originates from drinking water,
which is one of major components in wastewater, while SMPs come from
biological treatment with the wastewater treatment plant (WWTP) and non-
biodegradable organic matter (Shon et al., 2006). The SMP are organic
compounds that are biologically derived from substrate metabolism during
biomass growth (utilization associated products, UAP) and that are released
from cell lysis during biomass decay (biomass associated products, BAP)
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
127
(Grady et al., 1999; Barker et al., 1999). SMP has been found to constitute the
majority of soluble organic matter in wastewater effluent from biological
treatment system (Barker et al., 1999). It is known that natural organic matter
(NOM) adversely affected the adsorption of micropollutants onto, such as
pesticides, onto the activated carbon (Newcombe et al., 2002, Quinlivan et al.,
2005; Matsui et al., 2003). When background NOM is present during activated
carbon treatment of water containing micropollutants, a competition will occur
between the target compound and the compounds composing NOM. As a
consequence the adsorption of micropollutant will usually be reduced,
sometimes dramatically (Newcombe et al., 2002; Matsui et al., 2003). The
direct competition for the adsorption sites was found to be the most likely
competition between EfOM and target micropollutants (Newcomber et al.
2002; Kilduff et al. 1998; Matsui et al., 2003). However, limited data is
available on the effect of EfOM on the PFCs adsorption to the activated
carbon.
The objective of this chapter was to investigate the effect of EfOM on the
adsorption of micropollutants PFOS and PFOA onto the powdered activated
carbon. The EfOM was characterized and fractionated to study the adsorption
competition between PFCs and EfOM. Our results would contribute to better
understanding of competitive effects caused by presence of EfOM. In addition,
the performance and removal efficiencies of PFCs were investigated in a
hybrid PAC-MBR process which operated with different PAC dosage and
SRTs. The effect of SRT and PAC dosage was also studied for the better
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
128
understanding of removal mechanism of PFCs in the hybrid PAC-MBR
process.
7.2 Results and Discussion
7.2.1 Adsorption study on PAC and activated sludge
7.2.1.1 Characterization of EfOM
Four fractions of nominal molecular weights were obtained by ultrafiltration
of EfOM solution (Table 7.1). It can be seen that supernatant of MBR had a
broad spectrum of molecular weight. The fractions of smallest molecular
weight (<1 kDa) accounted for 27.8%, while the fraction of largest molecular
weight (>30 kDa) was the largest fraction, accounting for 31.2%. Other
fractions, 1-10 kDa and 10-30 kDa accounted for 29.9% and 12.9%,
respectively. Ultrafiltration is a size exclusion method of fractionation, some
factors, such as molecular structure and charge as well as solution chemistry
(pH and ionic strength) strongly affected the actual molecular weight of the
fractions (Kuchler et al., 1994). Pelekani et al. (1999) observed that
ultrafiltration of fractionation overestimated the actual molecular weight
distributions of NOM based on the membrane nominal molecular weight
cutoff values.
Table 7.1 Characteristics of EfOM solution obtained from the lab scale MBR (n=5).
Fraction <1 kDa 1–10 kDa
10-30 kDa
>30 kDa EfOM
% EfOM 27.8% 29.9% 12.9% 31.5% 100% SD 4.3 3.3 1.7 4.8 -
SD: standard deviation.
7.2.1.2 PFOS and PFOA adsorption onto PAC
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
129
The effects of EfOM on the PFCs adsorption onto the activated carbon were
investigated by conducting single solute isotherm experiments for PFOS and
PFOA in the absence and presence of EfOM using PAC. Figure 7.1 shows the
adsorption isotherms of PFOS and PFOA in EfOM free and EfOM raw. It can
be seen that adsorption capacity in EfOM free (Mill-Q water) was much
higher than that of EfOM raw for both PFOS and PFOA, suggesting EfOM
significantly decreased the adsorption capacity of PFCs onto PAC. The
isotherm experiment data were fitted to Langmuir and Freundlich models and
constants determined were listed in Table 7.2. Adsorption of PFCs to PAC
fitted the Freundlich model better (r2>98%) than Langmuir model (r2<90%) in
the absence and presence of EfOM within the studied concentration range
(0.1-500 µg/L), indicating PFCs were adsorbed to the heterogeneous sites with
different affinities for the solutes. Moreover, the adsorption capacity of PAC
tended to increase as the equilibrium concentration increased, which suggests
the possibility of more than just one monomolecular layer of coverage. The
application of Freundlich model is also extensively used to describe the
adsorption of organic solutes, such as Polychlorinated biphenyls (PCBs) and
polyacromatic hydrocarbons (PAHs) onto the activated carbon (Ahn et al.,
2005; Newcombe et al., 2002, Matsui et al., 2003).
The adsorption isotherms of PFOS and PFOA obtained in this study showed a
consistency within the concentration range of 0.1-500 µg/L, while Ochoa-
Herrera et al. (2008) observed significant difference in adsorption capacity of
activated carbon between high concentration range (15-150 mg/L) and low
concentration range (50-500 µg/L). Due to the different adsorbents used and
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
130
experiment conditions, the results may not be comparable. In addition, the
PFOS adsorption capacity was higher than that of PFOA, which is in
agreement with the data of other study (Ochoa-Herrera et al, 2008). This
observation is expected since PFOA has higher solubility and degree of
attraction by water molecules could be higher, which tends to prevent species
being bound by the carbon surface (Cooney, 1998).
According to Freundlich model, adsorption capacity of PFOS (KF=17.55) and
PFOA (KF=10.03) onto the PAC in pure water is more than one order of
magnitude higher than that of EfOM raw (KF=0.66 for PFOS, KF=0.20 for
PFOA), indicating the presence of EfOM greatly reduced the adsorption
capacity of PAC. Similarly, much lower adsorption capacity (e.g. Freundlich
constant KF) was observed for the simultaneous adsorption of organic
compounds and NOM onto the activated carbon in comparison with those
adsorption isotherms in pure water (Newcombe et al., 2002, Quinlivan et al.,
2005; Matsui et al., 2003). Moreover, it seems that the presence of EfOM
resulted in a linear Freundlich isotherm (n≈1) for both PFOS and PFOA.
Linear isotherm is the simplest expression of equilibrium adsorption, which is
valid for dissolved species that is present at concentrations less than one-half
of its solubility (Schwarzenbach et al., 2003). The effect of adsorption of
effluent organic matter on the efficiency of activated carbon for the removal of
PFCs is significant and would require high carbon dosage to effectively
remove PFCs in wastewater and attain desired water quality.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
131
0
50
100
150
200
250
300
0 100 200 300 400Ce (µg/L)
Cs (
µg/m
g) `
EfOM freeEfOM
(a) (b)
Figure 7.1 Adsorption isotherms of PFCs onto the PAC in the absence and presence of EfOM: (a) PFOS; (b) PFOA. Experimental data fit to Freundlich
model (solid line).
Table 7.2 Langmuir isotherm constants and Freundlich isotherm constants for the adsorption of PFCs onto PAC at 25 oC.
Adsorbate Solution
Langmuir isotherm Freundlich isotherm
a (µg/mg) b (L/µg) r2 KF
[(µg/mg) (L/µg)1/n]
1/n r2
PFOS EfOM
free
232.6 0.0545 0.881 17.5469 0.479 0.988
PFOA 200.0 0.0322 0.813 10.03 0.5369 0.981
PFOS EfOM
raw
232.6 0.003 0.345 0.6593 0.9321 0.98
PFOA 0.27 2.359 0.883 0.2043 1.1083 0.984
The adsorption kinetics of PFCs onto PAC was investigated in the presence
and absence of EfOM with initial concentration of 100 µg/L (Figure 7.2).
Equilibrium was observed at contact time of 72 h for both PFOS and PFOA,
while less contact time (4 h) was needed for the adsorption of PFOS or PFOA
onto PAC in Mill-Q water to reach steady state. It was observed that majority
of PFCs were adsorbed in 4 h, which suggests a rapid initial adsorption rate
for both PFOS and PFOA. Furthermore, it seems contact time of 8 h is
sufficient for the most of the PFCs that could be removed by PAC and be
0
50
100
150
200
250
0 100 200 300 400Ce (µg/L)
Cs (
µg/m
g) `
EfOM freeEfOM
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
132
adsorbed onto the carbon surface from the EfOM solution even though
equilibrium has not reached.
00.10.20.30.40.50.60.70.80.9
1
0 8 16 24 32 40 48 56 64 72Time (hour)
C/C
o
PFOSPFOA
(a) (b)
Figure 7.2 Adsorption of PFOS and PFOA onto PAC as a function of contact time: (a) in the presence of EfOM; (b) in the Milli-Q water.
7.2.1.3 Effect of EfOM on the PFOS and PFOA adsorption onto PAC
Figure 7.3 shows partial adsorption isotherms of PFCs onto PAC in the
presence of 3 type of EfOM fractions as well as absence of EfOM. It can be
seen that the adsorption capacity for PFCs onto PAC was in the following
order: EfOM free>30 k fraction>1 k fraction>EfOM raw. The adsorption
capacity of <1 k fraction was close to that of EfOM raw, especially for PFOA,
while >30 k fraction was close to EfOM free. The fraction of <1 k has greater
effect on the PFCs adsorption than >30 k fraction, indicating direct site
competition between PFCs and the <1 k fraction. It seems that larger
molecular size fraction, which absorb mainly in the larger pores, may not
compete directly for these adsorption sites. As smaller molecules, such as < 1
k fraction, diffuse faster than larger molecules (>30 k EfOM), larger
molecules may be still diffusing through pore structures after the PFCs have
been adsorbed, thereby causing no hindrance to PFCs adsorption. However,
larger molecular weight fraction, >30 k fraction, significantly reduced the
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6 7 8
Time (hour)
C/Co
PFOS
PFOA
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
133
PFCs adsorption capacity compared to EfOM free. This competition effect
could be caused by the presence of the low molecular weight compounds in
the >30 k EfOM fraction. Since UF characterization is influenced by a variety
of factors, including membrane pore size distribution, solution ionic strength,
as well as molecule size and shape (Logan and Jiang, 1990), Newcomber et al.
(2002) found that a certain amount of compounds with small molecular weight
(<500 Da) appeared in the fraction of high molecular weight (>30000 Da).
The presence of the low molecular weight compounds in the >30 k EfOM
fraction could cause decrease in PFCs adsorption by competition effect.
Therefore, the low molecular weight compounds, which have similar
molecular size of PFCs, are the major contributors to the competition. The
direct site competition between target micropollutants and low molecular
weight compounds of similar molecular size has been observed to be the
dominant mechanism by which NOM significantly reduced micropollutants
adsorption capacity onto activated carbon (Newcomb et al., 2002; Kilduff et
al., 1998; Matsui et al., 2003).
0
0.5
1
1.5
2
2.5
3
-1 0 1 2 3log Ce (µg/L)
log
Cs (
µg/m
g) <1K>30KEfOM rawEfOM free
(a) (b)
Figure 7.3 Log-log plot of PFCs adsorption isotherms in the presence and absence of EfOM fractions: (a) PFOS and (b) PFOA.
0
0.5
1
1.5
2
2.5
-1 0 1 2 3
log Ce (µg/L)
log Cs (µg/mg) <1K
>30KEfOM rawEfOM free
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
134
Freundlich iostherm parameters for the adsorption of PFCs on PAC in EfOM
fractions are shown in Table 7.3. The Freundlich constant 1/n for PFCs
increased with the decreasing molecular weight of the EfOM. It suggests that
EfOM occupied the high energy adsorption sites, which resulted in a decrease
in site heterogeneity. Therefore, the decease in adsorption of PFCs is due to
the decrease in suitable adsorption sites. Moreover, Freundlich constants (1/n
and KF) of >30 k fraction is much more closer to those of EfOM free than the
other samples with background of EfOM, indicating much less competition for
adsorption sites in the >30 k fraction than in EfOM raw or <1 k fraction. The
small molecular weight compounds may be present in the >30 k fraction and
cause the difference in adsorption between >30 k fraction and EfOM free.
Table 7.3 Freundlich iostherm parameters for the adsorption of PFCs on PAC in EfOM fractions.
Compund Parameter EfOM
raw <1 k >30 k EfOM free
PFOS
KF
[(µg/mg)(L/µg)1/n] 0.6593
1.052
2 7.4029 17.5469
1/n 0.9321 1.006 0.746 0.479
PFOA
KF
[(µg/mg)(L/µg)1/n] 0.2043
0.508
5 3.3045 10.03
1/n 1.1083 0.906
9 0.7292 0.5369
7.2.1.4 PFOS and PFOA adsorption onto activated sludge
Biosorption of PFOS and PFOA onto activated sludge were studied and
adsorption isotherms were shown in Figure 7.4. It can be seen that
experimental data fitted linear isotherms well (r2>0.9) for both PFOS and
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
135
PFOA, indicating that the partitions were independent of the concentrations.
According to the linear isotherm, partition coefficient Kd was 729 L/kg for
PFOS and 154 L/kg for PFOA, respectively (Table 7.4). In comparison,
relatively higher Kd values (720–2,324 L/kg for PFOS, 201–513 L/kg for
PFOA) were observed for the activated sludge in the WWTPs (Yu et al., 2009).
Based on the data on organic carbon ƒoc, calculated activated sludge log Koc
values (partition coefficient for the compound onto a hypothetical pure organic
carbon) were 2.86 for PFOS and 2.19 for PFOA, respectively. The log Koc
value for PFOS of this study is within the range measured by 3M Co. (log Koc
=2.57-3.1), while lower log Koc value (2.57) was reported by Higgins et al.
(2006). For PFOA, log Koc value of this study is slightly higher than those (log
Koc =1.9-2.17) observed by DuPont (2003) and that (log Koc =2.06) reported
by Higgins et al. (2006). In addition, no discernable difference in PFCs
biosorption onto activated sludge of different SRT was observed (Table 7.5),
indicating no effect of SRT on the PFCs biosorption. Some studies reported
the effect of SRT on the sludge characteristics (e.g surface charge, contact
angle) (Liao et al., 2000; Ng et al., 2005; Masse et al., 2006), which could
affect sludge biosorption capacity of organic matters. However, the effect of
SRT on the biosorption of micropollutants was not observed in study.
Table 7.4 Linear isotherm parameters for PFCs onto activated sludge.
Adsorbate Linear isotherm
Kd (L/kg) r2
PFOS 729 0.93
PFOA 154 0.90
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
136
The observed partition coefficients Kd for PFOS and PFOA are several of
orders of magnitude lower than those of bioaccumulative organic compounds
such as polychlorinated biphenyls and organochlorine pesticides (Katsoyiannis
et al., 2005). It suggests that PFCs have a lower tendency to partition onto the
sludge and sorption onto the activated sludge has no significant effect on the
removal of PFOS and PFOA in activated sludge treatment. Schultz et al.
(2006b) found that about less than 5% PFOS and PFOA were adsorbed onto
the activated sludge in the aeration tank of wastewater treatment plant with
conventional activated sludge treatment system. In addition, the adsorption
capacity of PFOS was more than 3 times higher than that of PFOA, suggesting
more PFOS could be adsorbed onto the activated sludge in wastewater
treatment process as compared to PFOA.
0
0.02
0.04
0.06
0.08
0.1
0 100 200 300Ce (µg/L)
Cs (
µg/m
g) `
PFOSPFOA
Figure 7.4 Adsorption isotherms of PFCs onto the activated sludge.
Experimental data fit to linear isotherm (solid line).
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
137
Table 7.5 Measured PFCs concentrations in activated sludge of MBR at different SRT.
SRT (d) 5 16 30
Compound PFOS PFOA PFOS PFOA PFOS PFOA
PFCs concentration
in sludge (µg/g) 106±12 22±2 116±13 27±2 111±10 23±3
7.2.2 Performance of MBR and PAC-MBR systems at different SRT
7.2.2.1 Overall performance of MBR and PAC-MBR system
The overall performance of the MBR in terms of COD and DOC in the
supernatant and effluent at different SRTs is summarized in Figures 7.5 and
7.6. The COD removal efficiencies were excellent and stable with an average
of over 95% at all investigated SRTs. Our results are generally consistent with
those reported in the literature (Huang et al., 2001; Lee et al., 2003). Also, it
can be seen that more than 30% DOC was rejected by membrane for both
MBR and PAC-MBR, indicating membrane separation play an important role
in maintaining satisfactory organic removal of MBR/PAC-MBR systems. In
addition, organic removal efficiencies of PAC-MBR at all studied SRTs were
a little higher than those of MBR. It suggests that PAC adsorption of organic
matters improved the overall performance in comparison with MBR.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
138
30 day16 day5 day60%
65%
70%
75%
80%
85%
90%
95%
100%
SRT (d)
CO
D re
mov
al (%
) `
MBRPAC-MBR
Figure 7.5 COD removal in MBR and PAC-MBR systems with different SRTs.
5 d 16 d 30 d
0.6
2.6
4.6
6.6
8.6
10.6
12.6
14.6
16.6
18.6
SRT (d)
DO
C (m
g/L) Supernatant (MBR)
Effluent (MBR)Supernatant (PAC-MBR)Effluent (PAC-MBR)
Figure 7.6 DOC of supernatant and effluent in MBR and
PAC-MBR systems with different SRTs. Figure 7.7 shows sludge concentrations in terms of MLSS and MLVSS in the
MBR and PAC-MBR system at different SRTs. As can be seen, average
MLSS concentration decreased accordingly with the decrease of SRT.
However, the ratios of VSS/SS were almost independent of SRT with an
average value over 0.95, indicating no considerable accumulation of inorganic
matter in the MBR system since synthetic wastewater was used as feed rather
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
139
than real wastewater. Furthermore, it was noted that the metabolic activity of
sludge, characterized by SOUR, slightly decreased as SRT lengthened (Figure
7.8). It could be attributed to the increase of inert biomass (i.e., metabolic
products mainly from endogenous respiration) at long SRTs and possibly to
the potential inhibition effect of soluble microbial products as observed by
Huang et al (2000). At different SRT, the MLVSS of PAC-MBR was found to
be slightly lower than that of MBR, while MLSS of PAC-MBR was
significantly higher than that of MBR. The increase in MLSS of PAC-MBR
could be due to the addition of a certain amount of PAC to the reactor, which
is confirmed by the comparable MLVSS between MBR and PAC-MBR.
Furthermore, the SOUR of PAC-MBR was close to that of MBR at different
SRTs, suggesting no discernable difference in metabolic activity of sludge was
observed between these two systems.
30 d16 d5 d02468
101214161820
SRT (d)
ML
SS/M
LV
SS (g
/L) `
MLSS(MBR)MLVSS(MBR)MLSS (PAC-MBR)MLVSS (PAC-MBR)
Figure 7.7 MLSS/MLVSS in MBR and PAC-MBR
systems with different SRTs.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
140
16 d 30 d5 d0
2
4
6
8
10
12
14
SRT (day)
SOUR (mgO2/gVSS h)
SOUR (MBR)
SOUR (PAC-MBR)
Figure 7.8 SOUR in MBR and PAC-MBR systems with different SRTs.
7.2.2.2 SMP and DOM fraction characteristics
Figure 7.9 shows the apparent molecular weight distributions (AMWD) of
DOM in the MBR and PAC-MBR at different SRTs. It can be seen that DOM
in the MBR systems had a broad spectrum of molecular weight. The majority
of DOM, accounting for around 53%, had molecular weight of less than 10
kDa, whereas the components with molecule weights between 10kDa and 30
kDa formed the smallest fraction, constituting 6.1-7.3% of DOM. The fraction
with molecule weights > 30 kDa account for 29-42% of DOM In addition, it
was noted that >30 kDa fraction increased with the increase of SRT, even
though the concentrations of DOM were significantly different. The results are
consistent with those reported in conventional biological treatment systems
where the AMWD of DOM have been found to be greatly affected by SRT
with high molecular weight components becoming more evident at long SRTs
(Barker and Stuckey, 1999).
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
141
>30K10K-30K1K-10K<1K0%
10%
20%
30%
40%
50%
Molecular weight (Da)
Perc
enta
ge (%
) SRT 5SRT 16SRT 30
(a) (b)
Figure 7.9 AMWD of SMP in the supernatant of (a) MBR and (b) PAC-MBR systems at different SRTs.
The DOM fractionations are shown in Figure 7.10. It can be seen that
hydrophilic HiA were the most abundant fraction of DOM, though their
proportion significantly increased in the MBR or decreased in the PAC-MBR
with the increase of SRTs. AHS accounted for the second largest fraction in
MBR and PAC-MBR systems, probably consisting of humic and fulvic acids.
In addition, it was noted that the proportion of AHS in total DOM gradually
decreased as SRT was lengthened, suggesting that DOM generated at long
SRTs tend to be more hydrophilic. As shown in Figure 7.10, HiB components
constituted the smallest fraction of in the MBR. In addition, proportions of
HoN and HoB were relatively stable and independent of SRT.
HiNHiBHiAHoNHoBAHS0%
10%
20%
30%
40%
50%
60%
DOM fraction
Perc
enta
ge (%
) SRT 5SRT 16SRT 30
Figure 7.10 Hydrophobicity of DOM in the supernatant of (a) MBR and
(b) PAC-MBR systems at different SRTs.
>30K10K-30K1K-10K<1K0%
10%
20%
30%
40%
50%
60%
Molecular weight (Da)
Perc
enta
ge (%
) SRT 5SRT 16SRT 30
HiNHiBHiAHoNHoBAHS0%
10%20%30%40%50%60%70%80%
DOM fraction
Perc
enta
ge (%
) SRT 5SRT 16SRT 30
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
142
7.2.3 Removal of PFOS and PFOA in PAC-MBR and MBR
7.2.3.1 Removal by adsorption onto activated sludge
Figure 7.11 shows the removal efficiency of PFCs in the MBR system
operated at different SRTs. The highest removal efficiency for both PFOS and
PFOA was observed in MBR with shortest SRT (5 d), while MBR with
longest SRT had lowest removal efficiency. Removal efficiencies of these two
compounds seem to decrease with the increase of SRT, implying no
improvement of biodegradation for these PFCs compounds at longer SRT. It
was reported that some micropollutants, such as endocrine disruptors
compounds (EDCs) or pharmaceuticals could be biodegraded when the
activated sludge treatment system (e.g MBR) was operated with longer SRT
(Clara et al. 2005a; Clara et al, 2005b). Some studies reported increase in
biodegradation of toxic or recalcitrant organic compounds at longer SRT due
to the acclimation and enrichment of certain microorganism (Kimura et al,
2007). However, this study confirmed that these two PFCs compounds can not
be biodegraded in activated sludge system. Furthermore, removal efficiencies
were in the range of 6-14.8% for PFOS and 1.4-3.8% for PFOA at the studied
SRT. As PFOS and PFOA can not be biodegraded, these two compounds can
only be removed by adsorption onto activated sludge or membrane. Filtration
experiment showed that removal efficiency for PFCs was negligible by
membrane (data not shown), indicating MF membrane can not significantly
remove PFCs. It suggests that adsorption onto the sludge would be the major
mechanism for PFCs removal in MBR system. However, low removal
efficiency of PFCs in MBR indicates PFCs can not be efficiently removed by
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
143
activated sludge system, which is also confirmed by some studies on fate and
behavior of PFCs in WWTPs (Sinclair et al., 2006; Schultz et al., 2006b; Yu et
al., 2009).
30 day16 day5 day0%
5%
10%
15%
20%
SRT (day)
PFC
s re
mov
al in
MB
R (%
)
PFOSPFOA
Figure 7.11 PFCs removal in MBR with different SRTs.
7.2.3.2 Removal by adsorption onto PAC
In PAC-MBR system, PFOS and PFOA could be effectively removed at
appropriate PAC dosage. Figure 7.12 shows the PFCs removal efficiency in
the PAC-MBR system operated at SRT of 30 d with PAC dosage varied from
30 to 100 mg/L. With the increase of PAC dosage, the removal efficiency
increased from 77.4% to 94.8% for PFOS and 67.7% to 90.6% for PFOA. In
contrast, negligible removal efficiencies for these two compounds were
observed in MBR with the same SRT (30 d), which suggest that adsorption of
PFCs onto PAC could play an important role in their removal in the PAC-
MBR system, instead of biosorption onto the activated sludge. Furthermore,
more PFCs were removed by the PAC-MBR at PAC dosage of 100 mg/L in
comparison with that of 30 mg/L, indicating the removal efficiency of PFCs
depend on the PAC dosage.
PFC
s rem
oval
in M
BR
(%)
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
144
100 mg/L80 mg/L30 mg/L0%
10%20%30%40%50%60%70%80%90%
100%
PAC dosage
PFCs
rem
oval
inPA
C-M
BR (%
)
PFOSPFOA
Figure 7.12 PFCs removal in PAC-MBR system
operated with different PAC dosages
PFCs removal in PAC-MBR system with PAC dosage 100 mg/L was studied
at different SRTs. It can be seen that the removal efficiencies were >90% for
PFOS and >84% for PFOA at different SRT (Figure 7.13). It suggests that
adsorption onto PAC was dominant and removal efficiencies may be not
significantly affected by different operational SRTs. Compared to those of
SRT at 16 d and 30 d, removal efficiencies at SRT of 5 d were slightly lower.
It seems PAC concentration in the reactor would affect the PFCs’ removal
efficiency as the there was the lowest PAC concentration in the reactor at SRT
of 5 d.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
145
30 day16 day5 day0%10%20%30%40%50%60%70%80%90%
100%
SRT (day)
PFC
s rem
oval
in P
AC
-MB
R (%
)
PFOSPFOA
Figure 7.13 PFCs removal in PAC-MBR system with
PAC dosage of 100 mg/L at different SRTs.
7.2.3.3 Mass balance
The mass balance of PFOS and PFOA in MBR system was established by
measuring PFCs concentration in aqueous and solid phases of inflow and
outflow. Mass flows of removed PFCs in MBR operated at different SRT are
shown in Figure 7.14. It can be seen that mass flow of PFOS or PFOA in
WAS accounted for more than 82.5% of its total removed amount. PFOS and
PFOA are not biodegraded in the activated sludge process due to their
exceptionally thermal and chemical stability. Since SPE extraction and other
analysis errors would lead to experimental errors, distribution of removed
PFCs mass flow suggests adsorption onto activated sludge could be the only
mechanism that removed PFCs in activated sludge system. In addition, more
PFCs were removed at shorter SRT since mass flow of PFCs in both liquid
and solid phases increased with the decrease of SRT. It seems that more
activated sludge (including solid and liquid phases) wasted out of the reactor
at shorter SRT result in more removed PFCs. Furthermore, mass flow of PFOS
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
146
or PFOA in the solid phase of WAS decreased with the increase of SRT. The
amount of PFCs in solid phase of WAS in MBR was determined by its
concentration on the sludge surface and mass flow of sludge in WAS. Since no
discernable effect of SRT on the PFCs adsorption on the sludge was found in
this study, decrease in WAS mass flow led to less sludge mass flow
discharged from the MBR with the increase of SRT, which could result in the
reduction of adsorbed PFCs mass flow in WAS. For PFOS, majority of
removed PFOS was adsorbed onto sludge and discharged with WAS at
different SRTs. In contrast, majority of removed PFOA was discharged from
the MBR system in the aqueous phase of WAS at SRT of 5 and 16 d,
indicating different behavior of PFOA in MBR at short SRT in comparison
with PFOS. Based on this study (section 7.2.1.4), adsorption capacity of PFOS
(Kd: 729 L/kg) was more than 3 times higher than that of PFOA (Kd: 154
L/kg). As can be seen, mass flow of PFOS on sludge of WAS was more than
3.5 times of that of PFOA at the same SRT. It suggests that more PFOS was
adsorbed onto the activated sludge, which could result in different behavior in
comparison with PFOA. In addition, mass flow of PFOA in the solid phase of
WAS at SRT of 30 d was more than that in the aqueous phase of WAS since
higher MLSS (avg 7.8 g/L) was observed at SRT of 30 d in comparison with
SRT of 5 and 16 d (3.5 g/L and 5.7 g/L, respectively).
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
147
545 184 100
1193 666459
686 μg/d943 μg/d1966 μg/d
0%
20%
40%
60%
80%
100%
5 16 30
SRT (d)
Dis
tribu
tion
of re
mov
ed P
FOS
mas
s in
MB
R (%
) errorsludgeaqueous
(a) (b)
Figure 7.14 Distribution of removed PFCs flow in MBR operated at different SRT: (a) PFOS; (b) PFOA. The value on the top of column represents the total mass flow (μg/d) removed in the MBR system; the value in columns indicates
the mass flow of PFCs (μg/d) in aqueous and solid phases.
As the PFCs concentrations in PAC surface can not be measured, their mass
balances in the PAC-MBR system were established by calculations.
Distributions of removed PFCs mass flow in the PAC-MBR at SRT of 30 d
with different PAC dosages were estimated and shown in Figure 7.15. With
the increase of PAC dosage, more PFOS or PFOA was removed by adsorption
on the PAC and activated sludge. However, mass flow in the solid phase of
WAS only increased by 22% for PFOS and 33% for PFOA even though PAC
dosage increased from 30 to 100 mg/L. Based on the PAC mass balance, PAC
concentrations were 2.7, 7.2 and 9.0 g/L in MBR. It seems adsorption capacity
of PAC decreased significantly as PAC concentration in MBR increased
greatly. Furthermore, it can be seen that more than 98% of removed PFCs was
in the solid phase (including activated sludge and PAC) of WAS. Compared to
MBR with the same SRT, most of the PFCs in the solid phase of WAS seems
to be adsorbed onto the PAC instead of activated sludge. For example, 459
mg/d of PFOS and 120 mg/d of PFOA were removed by adsorption onto the
activated sludge of the MBR, while mass flows in solid phase of WAS of the
616195 105
247
155120
240 μg/d426 μg/d980 μg/d
0%
20%
40%
60%
80%
100%
5 16 30
SRT (d)
Dis
tribu
tion
of re
mov
ed P
FOA
mas
s in
MB
R (%
) differencesludgeaqueous
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
148
PAC-MBR with PAC dosage of 30 mg/L were 7,430 mg/d for PFOS and
6,499 mg/d for PFOA at same SRT (30 d). It suggests adsorption on PAC was
an efficient and predominant process in the removal of PFCs in activated
sludge system. PAC adsorption would be much more effective than
biosorption for the removal of PFCs in the wastewater treatment even though
its adsorption capacity was significantly reduced by EfOM.
100 mg/L80 mg/L30 mg/L
9106 µg/d8608 µg/d7455 µg/d
0%
20%
40%
60%
80%
100%
PAC dosage (mg/L)
Dis
tribu
tion
of re
mov
ed P
FOS
mas
s in
PAC
-MB
R (%
)
solid WAS(PAC+sludge)aqueousWAS
(a) (b)
Figure 7.15 Estimated distributions of removed PFCs mass flow in waste of PAC-MBR at SRT 30 d with different PAC dosage: (a) PFOS; (b) PFOA. The
value on the top of column represents the total mass flow removed in the PAC-MBR system.
Figure 7.16 shows the estimated distributions of removed PFCs mass in PAC-
MBR operated at different SRT with a PAC dosage of 100 mg/L. The total
removal mass flow of PFOS or PFOA was comparable at different SRT,
indicating insignificant effect of SRT on the PFCs removal with the presence
of PAC. It seems that the effect of SRT on the PFCs’ removal could be
overridden by the effect of PAC adsorption. Furthermore, even though PAC-
MBR was operated at different SRT, mass flow of PFCs in the solid phase was
more than 98% of the total removed PFCs mass flow. Compared to the MBR
with the same SRT, most of PFCs in the solid phase of WAS seemed to be
adsorbed onto the PAC instead of activated sludge. For example, 1,193 mg/d
6534 µg/d 8100 µg/d 8706 µg/d
100 mg/L80 mg/L30 mg/L0%
20%
40%
60%
80%
100%
PAC dosage (mg/L)
Dis
tribu
tion
of re
mov
ed P
FOA
mas
s in
PAC
-MB
R (%
)
solid WAS(PAC+sludge)aqueousWAS
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
149
of PFOS was removed by adsorption onto the activated sludge of MBR, while
mass flow of PFOS in solid phase of WAS was 8,650 mg/d at same SRT (5 d)
of PAC-MBR. According to this study (section 7.2), PAC adsorption was
much more than biosorption. With presence of PAC, most of the PFCs is
expected to adsorb onto the PAC in the PAC-MBR. It was estimated about
171 mg/d of PFOS, instead of 1193 mg/d, was adsorbed onto the activated
sludge in WAS based on the partition coefficient of PFOS (Table 7.6). Table
7.6 shows estimated mass flows of PFCs in activated sludge of WAS in the
PAC-MBR operated at different SRTs. Biosorption accounted for <2% of total
removed PFCs amount at different SRT, indicating PFCs removal due to
biosorption was negligible in the PAC-MBR. It also confirmed that adsorption
on PAC is the predominant process in the removal of PFCs in activated sludge
system at appropriate PAC dosage, which would not be significantly affected
by SRT.
5 d 16 d 30 d
8713 µg/d 9139 µg/d 9106 µg/d
0%
20%
40%
60%
80%
100%
SRT (d)
Dis
tribu
tion
of re
mov
ed P
FOS
mas
s in
PAC
-MB
R (%
) solid WAS(PAC+sludge)aqueous WAS
(a) (b)
Figure 7.16 Estimated distributions of removed PFCs mass flow in waste of PAC-MBR operated at different SRTs: (a) PFOS; (b) PFOA. The value on the
top of column represents the total mass flow removed in the MBR system.
8703 µg/d8735 µg/d8820µg/d
5 d 16 d 30 d0%
20%
40%
60%
80%
100%
SRT (d)
Dis
tribu
tion
of re
mov
ed P
FOA
mas
s in
PAC
-MB
R (%
) solid WAS(PAC+sludge)aqueous WAS
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
150
Table 7.6 Estimated mass flows of PFCs in activated sludge of WAS in PAC-MBR operated at different SRTs.
SRT (day) 5 16 30 PFOS PFOA PFOS PFOA PFOS PFOA
PFCs concentration in sludge (µg/g)
14.43 4.74 7.14 2.83 7.58 2.91
PFCs mass flow in sludge (µg/d) 171.36 56.31 39.72 15.75 30.53 11.72
Total removed PFCs (µg/d) 8712.96 8220.16 9139.4 8735.2 9106.35 8702.88
Percentage in WAS (%) 1.98% 0.69% 0.44% 0.18% 0.34% 0.13%
7.2.3.4 Effect of SRT on PFOS and PFOA removal
Figure 7.17 indicates that PFCs concentration in sludge were slightly different,
varying from 106 to 116 µg/g (PFOS) and 22 to 27 µg/g (PFOA). Furthermore,
calculated PFCs concentrations in sludge were estimated by dividing mass
flow of PFCs in solid phase of WAS by the amount of activated sludge
discharged from MBR. Calculated PFCs concentrations on the sludge surface
were consistent with the measured values. It seems that sludge adsorption
capacity was consistent at different SRTs, indicating SRT had no significant
effect on the PFCs adsorption onto activated sludge.
5 d 16 d 30 d0
20
40
60
80
100
120
140
160
SRT (day)
PFO
S co
ncen
tratio
n on
slud
ge su
rface
(u
g/g)
measuredcalculated
(a) (b)
Figure 7.17 Effect of SRT on the PFCs adsorption onto activated sludge in MBR: (a) PFOS; (b) PFOA.
30 d16 d5 d0
10
20
30
40
SRT (day)
PFO
A c
once
ntra
tion
on sl
udge
surfa
ce
(ug/
g)
measuredcalculated
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
151
The effect of SRT on the adsorption of PFCs onto PAC in the PAC-MBR
system is shown in Table 7.7. As the mass flow of PFCs in aqueous phase of
the WAS were negligible, the normalization of PAC adsorption was calculated
by dividing total removed PFCs mass flow by the mass flow of PAC in the
WAS. Expected PAC adsorption capacity was predicted by the partition
coefficient Kd of this study (see section 7.2.1.4). It can be seen that PFCs
concentrations on PAC at SRT of 5 d were 5 times more than those at SRT of
30 d. With the increase of SRT, PFCs concentration on PAC decreased
significantly, indicating significant effect of SRT on the PAC adsorption
capacity in the PAC-MBR due to different PAC concentrations at different
SRTs. Furthermore, PAC adsorption capacity was not fully utilized at different
SRT in comparison with expected adsorption capacity when PAC was dosed
at 100 mg/L in the MBR. With the increase of SRT, utilized PAC adsorption
capacity decreased from 54.1% to 17.3% (PFOS) and 65.5% to 19.8% (PFOA).
It seems that PAC adsorption capacity could decrease significantly with the
increase of SRT. Therefore, PAC could have highest adsorption capacity in
the PAC-MBR at shortest SRT, which suggests fouling of PAC may
deteriorate and result in significant reduction in its adsorption capacity (Lee et
al., 2005; Ng et al., 2006). In addition, PFOA concentrations on PAC at
different SRT were comparable to those of PFOS even though PAC adsorption
capacity of PFOS was higher than that of PFOA with the presence of EfOM. It
may be due to the high PAC dosage added in the system (100 mg/L), which
overrided the difference in their adsorption capacity.
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
152
Table 7.7 Effect of SRT on the PFCs removal in PAC-MBR system with PAC dosage of 100 mg/L (based on mass balance).
SRT (day) 5 16 30 PFOS PFOA PFOS PFOA PFOS PFOA
Inflow mass flow (µg/d) 9600 9600 9600 9600 9600 9600
Outflow mass flow (µg/d) 887.0 1379.8 460.6 864.8 493.6 897.1
Total removed mass (µg/d) 8713 8220 9139 8735 9106 8703
Mass flow in aqueous WAS (µg/d)
63.36 98.56 9.80 18.40 5.55 10.08
Mass flow in solid WAS (PAC+sludge) (µg/d)
8649.60 8121.60 9129.60 8716.80 9100.80 8692.80
PFCs concentrations on PAC (µg/g)
5766.40 5414.40 1902.00 1816.00 1011.20 965.87
Expected PAC adsorption capacity (µg/g)
10666.84 8265.42 5538.17 4738.42 5853.55 4877.63
Utilized PAC adsorption capacity (%)
54.1% 65.5% 34.3% 38.3% 17.3% 19.8%
7.2.3.5 Effect of PAC dosage on PFOS and PFOA removal
The effect of PAC dosage on the adsorption of PFCs in PAC-MBR system is
shown in Table 7.8. As PAC dosage was increased from 30 to 100 mg/L,
PFCs concentrations on PAC decreased from 2,750 µg/g to 1,011 µg/g,
indicating significant effect of PAC dosage on PAC adsorption capacity for
PFCs in the PAC-MBR. According to the PAC adsorption study, PFCs
adsorption on PAC fitted Freundlich isotherms with the presence of EfOM,
which predicted that PAC would have lower adsorption capacity at higher
PAC dosage. Furthermore, utilized PAC adsorption capacity varied from
11.9% to 17.3% (PFOS) and 13.1% to 19.8% (PFOA) even though PAC
dosage tripled. The comparable utilized PAC capacity at different PAC
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
153
dosages indicates that fouling effect on the PAC could be similar at the same
SRT. In addition, PFOA concentrations on PAC at different PAC dosages
were slightly lower than those of PFOS even though PAC adsorption capacity
of PFOS was much higher than that of PFOA. It is possible that fouling effect
on the PAC could significantly reduce the difference in PFCs adsorption onto
PAC.
Table 7.8 Effect of PAC dosage on the PFCs removal in PAC-MBR system (based on mass balance).
PAC dosage (mg/L)
30 80 100 PFOS PFOA PFOS PFOA PFOS PFOA
Inflow mass flow (µg/d) 9600 9600 9600 9600 9600 9600
Outflow mass flow (µg/d) 2145.5 3066.3 992.1 1499.9 493.7 897.1
Total removed mass flow (µg/d)
7454.5 6533.7 8607.9 8100.1 9106.3 8702.9
Mass flow in aqueous WAS (µg/d)
23.96 34.24 11.08 16.75 5.51 10.02
Mass flow in solid WAS (PAC+sludge) (µg/d)
7430.55 6499.42 8596.87 8083.31 9100.83 8692.86
PFCs concentrations on PAC (µg/g)
2750.34 2405.69 1194.01 1122.68 1011.20 965.87
Expected PAC adsorption capacity (µg/g)
23021.46 18394.22 11218.12 8497.52 5853.55 4877.63
Utilized PAC adsorption capacity (%)
11.9% 13.1% 10.6% 13.2% 17.3% 19.8%
7.2.4 Membrane fouling
7.2.4.1 Variations of TMP
Figure 7.17 shows the long-term TMP profile for MBR and PAC-MBR system
at different SRT. For MBR system, noticeable membrane fouling was first
observed on day 38, 57 and 61 in the MBR operated at SRT of 5, 16 and 30
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
154
day, respectively. Subsequently, the TMP increased rapidly until day 47, 74
and 79 when membrane was removed for chemical cleaning. For PAC-MBR
system, noticeable fouling was detected after 67 d of operation for SRT of 5 d,
which was 1.76 time longer than the MBR system without PAC addition. It
seems that PAC addition would decrease the TMP of PAC-MBR at the same
operation condition as that of MBR, thus allowing the PAC-MBR system to
operate for a longer time to reach maximum total membrane resistance caused
by cake layer formation and solute adsorption on the membrane.
0
1
2
3
4
5
6
7
8
9
10
11
0 20 40 60 80 100 120 140days
Nor
mal
ized
TM
P ΔP/
ΔP0
SRT 5d (MBR)
SRT 16d (MBRC)
SRT 30d (MBR)
SRT 5d (PAC-MBR)
SRT 16d (PAC-MBR)
SRT 30d (PAC-MBR)
Figure 7.17 Long-term TMP profile for the MBR and
PAC-MBR systems at different SRTs.
7.2.4.2 Effect of PAC on TMP
Resistances of membrane for MBR and PAC-MBR system are summarized in
Table 7.9. It can be seen that total resistance (Rt) and intrinsic resistance (Rm)
are nearly same for MBR and PAC-MBR system. Compared to MBR system,
reversible resistance (Rr) decreased form 2.11 to 1.95 (1012·m-1) in PAC-
MBR system, while irreversible resistance (Ri) increased from 0.18 to 0.33
(1012 ·m-1). It suggests that PAC addition could reduce the cake layer
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
155
resistance and increase the percentage of Ri in the total membrane resistance.
Thus it allows the PAC-MBR system to operate for a longer time to reach
maximum irreversible fouling resistance caused by solute adsorption on the
membrane. Li et al. (2005) confirmed that Rr of PAC-MBR was 17.9% lower
than that of MBR, while Ri was 25.5% higher than that of MBR. In addition,
Rt of PAC-MBR was significantly lower than that of MBR when they were
operated at the same condition before membrane fouling occurred, which was
found to be due to the significant reduction in reversible resistance (Rr). It
suggests that PAC play an important role in reducing cake resistance and
changing an overall particle size distribution to a greater size range. (Li et al.,
2005; Munz et al., 2007).
Table 7.9 Resistances of membrane for the MBR and PAC-MBR systems
Resistances MBR (1012·m-1) PAC-MBR
(1012·m-1)
Rm 0.31 0.32
Rr 2.11 1.95
Ri 0.18 0.33
Rt 2.6 2.6
7.3 Summary
The simultaneous adsorption of EfOM and PFOS or PFOA onto PAC was
investigated in this study. The presence of EfOM significantly decreased the
adsorption capacity of PFCs onto PAC in comparison with that in the absence
of EfOM. Adsorption of PFCs to PAC fitted the Freundlich model well
(r2>98%) in the absence and presence of EfOM within the studied
concentration range (0.1-500 µg/L). According to Freundlich model,
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
156
adsorption capacity of PFOS (KF=17.55) and PFOA (KF=10.03) onto the PAC
in pure water was more than one order of magnitude higher than that of EfOM
solution (KF=0.66 for PFOS, KF=0.20 for PFOA), indicating that presence of
EfOM greatly reduce the adsorption capacity of PAC. The adsorption kinetics
of PFCs was investigated in the presence of EfOM with initial concentration
of 100 µg/L. A rapid initial adsorption rate was observed in 4 h for both PFOS
and PFOA. It seems the contact time of 8 h was sufficient for the PFCs to be
adsorbed onto the carbon surface from the EfOM solution even though
equilibrium had not been reached.
EfOM solution was characterized by ultrafiltration and four EfOM fractions
were obtained to investigate their effects on the PFCs adsorption. The
adsorption capacity for PFCs onto PAC was in the following order: EfOM
free> 30 k fraction>1 k fraction>EfOM solution. It seems that larger
molecular size fraction, which was absorbed mainly in the larger pores, may
not compete directly for these adsorption sites. However, the smaller
molecular weight compounds, which had the similar molecular size of PFCs,
were the major contributors to the competition. The direct site competition
between target PFCs and low molecular weight compounds of similar
molecular size seems to be the dominant mechanism by which EfOM
significantly reduced PFCs adsorption capacity onto activated carbon.
Adsorption of PFCs to activated sludge fitted linear isotherms (r2>0.9) within
concentration range of 50-400 µg/L, which indicated that the partitions were
independent of the concentrations for both PFOS and PFOA. Based on our
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
157
data, the estimated partition coefficient Kd was 729 L/kg for PFOS and 154
L/kg for PFOA, respectively. It suggests that PFOS and PFOA, especially
PFOA, have a low tendency to partition onto the sludge, indicating sorption
onto the activated sludge has insignificant effect on the removal of PFOS and
PFOA in activated sludge treatment process.
Removal efficiencies of PFCs in MBR were investigated at different SRT,
which were in the range of 6-14.8% for PFOS and 1.4-3.8% for PFOA. PFCs
low removal efficiency (<15%) in MBR indicates PFCs can not be efficiently
removed by activated sludge system. Distribution of removed PFCs mass flow
suggests that adsorption onto activated sludge could be the only mechanism
that removed PFCs in activated sludge system. More PFCs was removed at
shorter SRT since mass flow of PFCs in both liquid and solid phases increased
with the decrease of SRT. Furthermore, PFCs mass flow in the solid phase of
WAS decreased with the increase of SRT. It is possibly attributed to the
decrease in sludge mass flow discharged from the MBR, which could result in
the reduction of adsorbed PFCs mass flow in WAS. In addition, PFCs
concentrations in sludge were slightly different at different SRT, varying from
106 to 116 µg/g (PFOS) and 22 to 27 µg/g (PFOA). It seems that sludge
adsorption capacity was consistent at different SRTs, indicating SRT had no
significant effect on the PFCs adsorption onto activated sludge.
The overall performance and removal efficiencies of PFCs were investigated
on PAC-MBRs which operated with different PAC dosages and SRTs. The
effect of PAC dosage on the removal of PFCs in PAC-MBR was studied at the
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
158
SRT 30 d. Removal efficiency increased from 77.4% to 94.8% for PFOS and
67.7% to 90.6% for PFOA with the increase of PAC dosage from 30 to 100
mg/L. Based on the established mass balance, it suggests that adsorption on
PAC was the efficient and predominant process in the removal of PFCs in
activated sludge system. PAC adsorption would be much more effective than
biosorption for the removal of PFCs in the wastewater treatment even though
its adsorption capacity was significantly reduced by EfOM. As PAC dosage
increased from 30 mg/L to 100 mg/L, PFCs concentrations on PAC decreased
from 2,750 to 1,011 µg/g, indicating the significant effect of PAC dosage on
PAC adsorption capacity for PFCs in PAC-MBR. However, utilized PAC
adsorption capacity was relatively consistent in the range of 11.9% to 17.3%
(PFOS) and 13.1% to 19.8% (PFOA) even though PAC dosage was
significantly increased. The comparable utilized PAC capacity at different
PAC dosage indicates that fouling effect on the PAC could be similar at the
same SRT.
The effect of SRT on removal of PFCs in PAC-MBR was further investigated.
Removal efficiencies were >90% for PFOS and >84% for PFOA at different
SRT, suggesting that adsorption onto PAC could be dominant and removal
efficiencies may be not significantly affected by different operational SRT.
With the increase of SRT, PFCs concentration on PAC decreased significantly,
indicating significant effect of SRT on the PAC adsorption capacity in PAC-
MBR due to different PAC concentrations at different SRTs. In addition,
utilized PAC adsorption capacity decreased from 54.1% to 17.3% (PFOS) and
65.5% to 19.8% (PFOA) when SRT was increased from 5 to 30 d. It seems
Chapter 7-PFOS/PFOA Removal by Hybrid PAC-MBR Process
159
that PAC adsorption capacity could decrease significantly with the increase of
SRT, which was possibly due to deteriorating fouling of PAC.
TMP profile for MBR and PAC-MBR system at different SRT observed in
this study suggests that PAC could significantly extend the operation time of
PAC-MBR. PAC addition could reduce the cake layer resistance and increase
the percentage of irreversible fouling resistance in the total membrane
resistance, thus allowing the PAC-MBR system to operate for a longer time to
reach the maximum irreversible fouling resistance caused by solute adsorption
on the membrane.
Chapter 8-Conclusions
160
CHAPTER 8 CONCLUSIONS
8.1 Conclusions
This study investigated the occurrence and fate of PFOS and PFOA in water
and wastewater as well as explored removal strategy of hybrid PAC-MBR
process. For the first time, it provided data on spatial and seasonal occurrence
and distribution of PFOS and PFOA in Singapore water environment,
including rivers, reservoirs and lakes and sea water around the island. PFOS
and PFOA were detected in all collected samples in the range of 1.9~532.1
ng/L (PFOS) and 2.4~1,057.1 ng/L (PFOA). Seawater had lower concentration
of PFOS and PFOA, compared with surface waters and treated effluents. In
surface waters, the highest total concentrations of PFOS and PFOA were
observed in the western area because of the high levels of industrial activities
in that area. This region was noted to be the most highly contaminated by
PFCs. In wastewaters, the highest total PFCs mass load and PFOA
concentration (1,057.1 ng/L) were observed in W5, suggesting discharges of
fluorochemical related factories in the service area of W5 may contain a large
amount of PFOS and PFOA, thus resulting in high concentrations in the
WWTPs effluents. The highest PFOS concentration (532.1 ng/L) was detected
in the effluent of W1 treating mainly domestic and commercial wastewater.
This indicates the presence of potential PFOS contamination sources in its
service area. Compared with surface waters and coastal waters, much higher
PFCs concentrations in wastewaters indicate that discharge of effluents of
WWTPs is an important pathway by which PFCs enter the environment. In
Chapter 8-Conclusions
161
coastal water, the high PFOS and PFOA concentrations at C4 suggest that
Johor Straits is more heavily contaminated than the southern and eastern
coastal waters. The high levels of industrial activities in the western area may
be the significant contamination sources for Johor Straits. Furthermore,
significant seasonal variation between dry seasons and wet seasons was
observed in surface waters for PFOS only, while no discernable seasonal
differences were found for both PFOS and PFOA in coastal waters and
wastewaters. In addition, PFOS and PFOA were significantly correlated in the
coastal waters, while weak positive correlations were observed in surface
waters and wastewaters. It suggests that the possibility of a common
contamination source for these two compounds in coastal waters is higher than
those of surface waters and wastewaters.
An efficient sample clean-up method was developed in this study to
significantly remove co-eluting matrix components by applying the SPE
extracts onto a silica cartridge after dilution with dichloromethane. Matrix
effect on PFOS and PFOA were evaluated by comparing MS responses of
standards and those of the same known amount of analytes in post-extraction
spiked samples. It was found that that ME% for both PFOS and PFOA were
below 50%, indicating SPE alone was insufficient to remove matrix
components. Also, recoveries (RE%: <50%) were significantly affected by the
matrix effect due to the ionization suppression even though this HLB SPE
procedure can achieve more than 90% recoveries (98.4% for PFOS and 93.8%
for PFOA) for PFCs spiked Mill-Q water. Therefore, silica cartridge was
applied to reduce the co-eluting interfering compounds and ME% (>70%) and
Chapter 8-Conclusions
162
RE% (>67%) were increased significantly for both PFOS and PFOA. It
suggests that substantial amount of interfering compounds were retained by
silica cartridge, while PFOS and PFOA were eluted by mixture of
dichloromethane/methanol (60:40, v/v). After silica cartridge clean-up, the
coefficient of variation (CV) decreased more than 44% (PFOS) and 34%
(PFOA) for ME% and RE%, indicating precision of the analysis increased due
to the reduced matrix effect. The application of internal standards further
compensated for matrix effect and brought the ME% and RE% close to 100%,
indicating minimal matrix effect was achieved without significant loss of
analytes. CV was greatly decreased by applying internal standardization with
its value below 5%. In addition, a higher recovery (>90%) was achieved
compared to that of around 70% without internal standardization. The
developed LC-MS-MS detection method was applied to different water and
sludge samples. Results showed that this silica cartridge clean-up method can
effectively remove co-eluting matrix components in various environmental
matrices with ME% >95% for water samples and >90% for sludge samples.
The behavior of PFOS and PFOA in the biological units of various full-scale
municipal wastewater treatment plants was studied. Samples of influent,
primary effluent, aeration tank effluent, final effluent and grab samples of
primary, activated, secondary and anaerobically digested sludge were
collected by 5 sampling events over one year. The two sewage treatment
plants (STPs) selected for this study included plant A receiving 95% domestic
wastewater and plant B receiving 60% industrial wastewater and 40%
domestic wastewater. PFOS and PFOA were detected in all aqueous samples
Chapter 8-Conclusions
163
collected from STP A and B, ranging from 5.3 - 560.9 ng/L and 11.2 – 1,057.1
ng/L, respectively. In sludge of STPs A and B, PFOS and PFOA
concentrations were in the range of 13.1 – 702.2 ng/g dw and <5.0 - 69.0 ng/g
dw, respectively. It is noted that PFOS and PFOA were observed at higher
concentration in aqueous and sludge samples in STP B than those of STP A,
indicating that industrial sewage contain a larger amount of PFCs in
comparison with domestic sewage. Significant increase in mass flow of PFOS
(mean 94.6%) was observed in CAS2, while it remained consistent after
secondary treatment in CAS1. This is likely due to no occurrence of PFOS
precursors in the raw sewage. Mean mass flow of PFOA increased 41.6% in
CAS1, 67.0% in CAS2 and 76.6% in MBR, while it remained unchanged after
the treatment of LTM. Different behavior pattern of these two compounds
were found in LTM, an activated sludge process operated at a relatively short
SRT. The findings suggest that change in mass flow of PFOS and PFOA in
secondary sludge treatment may be determined by the presence of precursors
and operating SRT of the activated sludge system. Furthermore, between dry
and wet seasons, seasonal variation of PFOS was observed in STP A, while
PFOA had no significant difference in both STP A and STP B. PFOS
concentration in rainwater observed by other studies was lower than that of
PFOA, which could lead to their different seasonal variations in surface water.
It is also likely that NPS of PFCs occurred in wet season, which would
contribute to consistent PFOA concentrations in surface waters and
subsequently resulted in indiscernible variation in PFOA concentrations in the
wastewaters between dry and wet seasons.
Chapter 8-Conclusions
164
The adsorption of PFOS and PFOA onto powdered activated carbon (PAC)
was investigated in the presence and absence of EfOM at low concentration
range (0.1-500 µg/L). Adsorption of PFOS and PFOA to PAC fitted the
Freundlich model well (r2>98%) and adsorption capacity of PFOS (KF=17.55)
and PFOA (KF=10.03) in the absence of EfOM was more than one order of
magnitude higher than those in the presence of EfOM (KF=0.66 for PFOS,
KF=0.20 for PFOA), indicating EfOM greatly reduced the adsorption capacity
of PAC. The adsorption kinetics of PFCs was investigated in the presence of
EfOM with initial concentration of 100 µg/L. A rapid initial adsorption rate
was observed in 4 h for both PFOS and PFOA. It seems the contact time of 8 h
was sufficient for the PFCs to be adsorbed onto the carbon surface from the
EfOM solution even though equilibrium had not been reached. Moreover,
EfOM was characterized by ultrafiltration and fractions of nominal molecular
weights were obtained to investigate their effect on the PFOS and PFOA
adsorption. The fraction of <1 k had greater effect on the adsorption than >30
k fraction, indicating the similar molecular size of target compounds, were the
major contributors to the adsorption competition. The direct site competition
between target PFCs and low molecular weight compounds of similar
molecular size seems to be the dominant mechanism by which EfOM
significantly reduced PFCs adsorption capacity onto activated carbon.
Additionally, biosorption of PFOS and PFOA to the activated sludge fitted the
Linear isotherm (r2>0.9) within concentration range of 50-400 µg/L. Based on
our data, the estimated partition coefficient Kd was 729 L/kg for PFOS and
154 L/kg for PFOA, suggesting PFOS and PFOA, especially PFOA, have a
low tendency to partition onto sludge.
Chapter 8-Conclusions
165
This study explored overall removal performance and factors affecting PFCs’
adsorption onto activated sludge and PAC in MBR and hybrid PAC-MBR
processes. Laboratory-scale MBR and PAC-MBR were operated in parallel at
SRT of 5, 16, and 30 days for treatment of readily biodegradable synthetic
wastewater. Removal efficiencies of PFCs in MBR were in the range of 6-
14.8% for PFOS and 1.4-3.8% for PFOA at different SRT studied. PFCs low
removal efficiency (<15%) in MBR indicates PFCs can not be efficiently
removed by activated sludge system. Distribution of removed PFCs mass flow
suggests adsorption onto activated sludge could be the only mechanism that
removed PFCs in activated sludge system. More PFCs was removed at shorter
SRT since mass flow of PFCs in both liquid and solid phases increased with
the decrease of SRT. In addition, PFCs concentrations in sludge were slightly
different at different SRT, varying from 106 to 116 µg/g (PFOS) and 22 to 27
µg/g (PFOA). It seems that sludge adsorption capacity was consistent at
different SRTs, indicating SRT had no significant effect on the PFCs
adsorption onto activated sludge.
The overall performance and removal efficiencies of PFCs were also
investigated in PAC-MBRs which operated with different PAC dosage and
SRTs. On the one hand, the effect of PAC dosage on the removal of PFCs in
PAC-MBR was studied at the SRT of 30 d. Removal efficiency increased
from 77.4 to 94.8% for PFOS and 67.7 to 90.6% for PFOA with the increase
of PAC dosage from 30 to 100 mg/L. Based on the established mass balance,
it suggests adsorption on PAC was the efficient and predominant process in
Chapter 8-Conclusions
166
the removal of PFCs in activated sludge system. PAC adsorption would be
much more effective than biosorption for the removal of PFCs in the
wastewater treatment even though its adsorption capacity was significantly
reduced by EfOM. As PAC dosage was increased from 30 to 100 mg/L, PFCs
concentrations on PAC decreased from 2,750 to 1,011 µg/g, indicating
significant effect of PAC dosage on PAC adsorption capacity for PFCs in
PAC-MBR. However, utilized PAC adsorption capacity was relatively
consistent in the range of 11.9 to 17.3% (PFOS) and 13.1 to 19.8% (PFOA)
even though PAC dosage significantly increased. The comparable utilized
PAC capacity at different PAC dosage indicates that biofouling effect on the
PAC could be similar at the same SRT. On the other hand, the effect of SRT
on removal of PFCs in PAC-MBR was studied. Removal efficiencies of PFCs
were >90% for PFOS and >84% for PFOA at different SRT studied,
suggesting that adsorption onto PAC could be dominant and removal
efficiencies may be not significantly affected by different operational SRT.
With the increase of SRT, PFCs concentration on PAC decreased significantly,
indicating significant effect of SRT on the PAC adsorption capacity in PAC-
MBR due to different PAC concentrations at different SRTs. In addition,
utilized PAC adsorption capacity decreased from 54.1 to 17.3% (PFOS) and
65.5 to 19.8% (PFOA) when SRT was increased from 5 to 30 d. It seems that
PAC adsorption capacity could decrease significantly with the increase of SRT,
which was possibly due to deteriorating fouling of PAC.
TMP profiles for MBR and PAC-MBR system at different SRTs observed in
this study suggests that PAC could significantly extend the operation time of
Chapter 8-Conclusions
167
PAC-MBR. PAC addition could reduce the cake layer resistance and increase
the percentage of irreversible fouling resistance in the total membrane
resistance. Overall, PAC addition would decrease the TMP of PAC-MBR at
the same operation condition as that of MBR, thus allowing the PAC-MBR
system to operate for a longer time to reach the maximum total membrane
resistance caused by cake layer formation and solute adsorption on the
membrane.
Contributions of this study would provide a better understanding of occurrence
and fate of PFOS and PFOA in aquatic environment as well as behavior in
sewage treatment plants. The seasonal varitions of PFCs in aquatic
environment were explored between dry and wet seasons in an ideal island,
where other climate factors were excluded from this study. Moreover, the
developed post extraction cleanup method should contribute to higher
accuracy for detection of wastewater and sludge samples. Also, it should be
noted that the effect of SRT on the PFCs mass change would deepen the
understanding of their behavior patterns in STPs. To our best of knowledge, it
is the first study to examine the effect of SRT on the PFCs’ behavior in
activated sludge treatment process. On the other hand, the study on
simultaneous adsorption of EfOM and PFOS or PFOA onto PAC investigated
in this study would provide valuable new insights into the characteristics of
PAC adsorption in the MBR and consequently further advance our knowledge
on the removal of PFOS and PFOA in the hybrid PAC-MBR process as well
as activated sludge process.
Chapter 8-Conclusions
168
8.2 Recommendations
The transport pathways of PFCs to aquatic environment could include
discharge of effluents from STPs, direct discharge of wastewater from
manufacture and use of PFCs, rain runoff and atmospheric transport of PFCs
and subsequent atmospheric loading of PFCs to surface waters. It is of note
that research with respect to fate of PFCs in the aquatic environment is far
from complete and much work is needed to fully understand this important
issue. PFCs concentrations in the air, drinking water, ground water, rainwater
and rain runoff should be investigated to identify possible contamination
sources and transport pathways of PFCs in environment.
The developed silica cartridge clean-up can effectively remove interfering
components and significantly improve the accuracy of the LC/MS/MS
analyses. In this study the developed method is limited to the analysis of PFOS
and PFOA. As other PFCs compounds have similar physico-chemical
properties, the developed clean-up method could be applied to analysis of
other PFCs compounds. It would significantly enlarge the contribution of
developed clean-up method. Therefore, further research should be conducted
to explore the possibility of extending the developed clean-up method to other
PFCs compounds analysis in environmental matrices.
In order to achieve better understanding of behavior of studied compounds in
sewage treatment plants, this study is restricted to investigate compounds
PFOS and PFOA only and there is no intention to identify their precursors and
product compounds. Future research should identify the related precursors
Chapter 8-Conclusions
169
and explore the mass transfer between aqueous and solid phases along the
treatment processes. Degradation of precursors can lead to occurrence and
variation of studied compounds in concentration and mass flow in water and
wastewater. Relationship between precursors and studied compounds should
be studied to identify the contribution in mass increases from their precursors.
Furthermore, mass balance in STPs in this study was not attempted because
only grab samples, instead of composite samples, were collected from STPs.
Composite samples should be collected and analyzed to establish the mass
balance of studied compounds in the whole STP. Then behavior of studied
compounds may be completely and accurately understood by identifying mass
flow increase due to degradation of their precursors and mass flow decrease
due to sludge adsorption.
It was found that PAC adsorption could be the removal mechanism of PFCs
for the hybrid PAC-MBR process. Site competition was suggested to be the
adsorption mechanism of PFCs onto the PAC in the presence of EfOM. It
should be noted that other adsorption mechanism such as pore blockage could
override site competition and be the dominant mechanism. In order to exclude
the pore blockage mechanism, a few adsorbents with different pore size
distributions should be studied to investigate the effect of pore size
distribution on the adsorption in the presence of EfOM. Moreover, it would
provide the knowledge on the characteristics of optimal adsorbent for removal
of PFCs in the hybrid PAC-MBR process.
Reference
170
REFERENCE
3M (1999) Fluorochemical Use, Distribution and Release Overview. US EPA
Administrative Record AR226-0550. U.S. Environmental Protection Agency,
Washington, DC.
3M (2000) Soil Adsorption/Desorption Study of Potassium Perfluorooctane
Sulfonate (PFOS). US EPA Docket AR 226-1030a 030. U.S. Environmental
Protection Agency, Washington, DC.
3M (2001) Environmental Monitoring-Multi-City Study (Water, Sludge,
Sediment, POTW Effluent and Landfill Leachate Samples). Docket AR-226-
1030a; Office of Pollution and Prevention and Toxic Substances, U.S.
Environmental Protection Agency, Washington, DC.
3M (2003) Health and environmental assessment of perfluorooctane sulfonic
acid and its salts. US EPA docket no. AR-226-1486. US Environmental
Protection Agency, Washington, DC.
Ahn, S., Werner, D., Karapanagioti, H.K., Zare, R.N., Luthy, R.G. (2005)
Phenanthrene and pyrene sorption and intraparticle diffusion in
polyoxymethylene, coke, and activated carbon. Environ. Sci. Technol. 39,
6516-6526.
Aktas, O., Cecen, F. (2007) Bioregeneration of activated carbon: A review. Int.
Biodeter. & Biodegr. 59, 257-272.
APHA-AWWA-WEF. Standard Methods for Examination of Water and
Wastewater, 20th ed; APHA-AWWA-WEF, Washington, DC, 1998.
Reference
171
Aquino, S.F., Hu, A.Y., Akram, A., Stuckey, D.C. (2006) Characterization of
dissolved compounds in submerged anaerobic membrane bioreactors
(SAMBRs). J. Chem. Technol. Biotechnol. 81, 1894-1904.
Austin, M.E., Kasturi, S.B., Barber, M., Kannan, K., MohanKumar, P.S. and
MohanKumar, S.M.J. (2003) Neuroendocrine effects of perfluorooctane
sulfonate in rats. Environ. Health Persp. 111, 1485-1489.
Bai, R. and Leow, H.F. (2002) Microfiltration of activated sludge wastewater-
the effect of system operation parameters. Sep. Purif. Technol. 29, 189-198.
Barker, D.J., Mannucchi, G.A., Salvi, S.M.L. and Stuckey, D.C. (1999a)
Characterization of soluble residual chemical oxygen demand (COD) in
anaerobic wastewater treatment effluents. Water Res. 33, 2499-2510.
Barker, D.J., Stuckey, D.C. (1999b) A review of soluble microbial products
(SMP) in wastewater treatment systems. Water Res. 33, 3063-3082.
Baumgarten, S., Schroeder, H., Charwath C., Lange M., Beier S., Pinnekamp,
J. (2007) Evaluation of advanced treatment technologies for the elimination of
pharmaceutical compounds. Water Sci. Technol. 56, 1-8.
Becker, A.M., Gerstmann, S. and Frank, H. (2008) Perfluorooctane surfactants
in waste waters, the major source of river pollution. Chemosphere 72, 115-121.
Benijts, T., Dams, R., Lambert, W. and De Leenheer, A. (2004) Countering
matrix effects in environmental liquid chromatography-electrospray ionization
tandem mass spectrometry water analysis for endocrine disrupting chemicals.
J. Chromatogr. A 1029, 153-159.
Berger, U. and Haukas, M. (2005) Validation of a screening method based on
liquid chromatography coupled to high-resolution mass spectrometry for
Reference
172
analysis of perfluoroalkylated substances in biota. J. Chromatogr. A 1081,
210-217.
Boudreau, T.M., Sibley, P.K., Mabury, S.A., Muir, D.C.G. and Solomon, K.R.
(2003) Laboratory evaluation of the toxicity of perfluorooctane sulfonate
(PFOS) on Selenastrum capricornutum, Chlorella vulgaris, Lemna gibba,
Daphnia magna and Daphnia pulicaria. Arch. Environ. Contam. Toxicol. 44,
307-313.
Bouhabila, E.H., Aїm, R.B. and Buisson, H. (2001) Fouling characterisation in
membrane bioreactors. Sep. Purif. Technol. 22-3, 123-132.
Boulanger, B., Vargo, J., Schnoor, J.L., Hornbuckle, K.C. (2004) Detection of
perfluorooctane surfactants in Great Lakes water. Environ. Sci. Technol. 38,
4064-4070.
Boulanger, B., Vargo, J.D., Schnoor, J.L. and Hornbuckle, K.C. (2005)
Evaluation of perfluorooctane surfactants in a wastewater treatment system
and in a commercial surface protection product. Environ. Sci. Technol. 39,
5524-5530.
Bouhabila, E. H., Aim, R. B., Buisson, H. (2001) Fouling characteristics in
membrane bioreactors. Sep. Purif. Technol. 22, 123-132.
Case, M.T., York, R.G., Christian, M.S. (2001) Rat and rabbit oral
developmental toxicology studies with two perfluorinated compounds. Int. J.
Toxicol. 20, 101-109.
Chang, I.S., Lee, C.H. (1998) Membrane filtration characteristics in
membrane-coupled activated sludge systems-The effect of physiological states
of activated sludge on membrane fouling. Desalination 120, 221-223.
Reference
173
Chang, I.S., Lee, C.H. and Ahn, K.H. (1999) Membrane filtration
characteristics in membrane coupled activated sludge system: The effect of
floc structure of activated sludge on membrane fouling. Sep. Sci. Technol. 34,
1743-1758.
Chang, I.S., Clech, P.L., Jefferson, B. and Judd, S. (2002) Membrane fouling
in membrane bioreactors for wastewater treatment. J. Environ. Eng.-ASCE
128, 1018-1029.
Childress, A.E., Elimelech, M. (1996) Effect of solution chemistry on the
surface charge of polymeric reverse osmosis and nanofiltration membranes. J.
Membr. Sci. 119, 253.
Churchouse, S.J. (1998) Kubota Maintains Popular Trend. Water and Environ.
International 26-27.
Clara, M., Kreuzinger, N., Strenn, B., Gans, O. and Kroiss, H. (2005a) The
solids retention time - a suitable design parameter to evaluate the capacity of
wastewater treatment plants to remove micropollutants. Water Res. 39, 97-106.
Clara, M., Strenn, B., Gans, O., Martinez, E., Kreuzinger, N.,and Kroiss, H.
(2005b) Removal of selected pharmaceuticals, fragrances and endocrine
disrupting compounds in a membrane bioreactor and conventional wastewater
treatment plants. Water Res. 39, 4797-4807.
Cooney, D.O. (1998) Adsorption design for wastewater treatment. Lewis
Publishers, NJ.
Deng, H., Peng, P., Huang, W., Song, J. (2004) Distribution and loadings of
polycyclic aromatic hydrocarbons in the Xijiang River in Guangdong, South
China. Chemosphere 64, 1401-1411.
Reference
174
Dinglasan, M.J.A., Ye, Y., Edwards, E.A. and Mabury, S.A. (2004)
Fluorotelomer alcohol biodegradation yields poly- and perfluorinated acids.
Environ. Sci. Technol. 38, 2857-2864.
Dosoretz C.G., Boddeker K.W. (2004) Removal of trace organics from water
using a pumped bed-membrane bioreactor with powdered activated carbon. J.
Membr. Sci. 239, 81-90.
DuPont (2003) Adsorption/Desorption of Ammonium Perfluorooctanoateto
Soil (OECD 106); E.I. du Pont de Nemours and Company: Wilmington, DE,
US EPA, Docket OPPT-2003-0012-040. U.S. Environmental Protection
Agency, Washington, DC.
Fleischer, E.J., Broderick, T.A., Daigger, G.T., Fonseca, A.D, Holbrook, R.D.,
Murthy, S.N. (2005) Evaluation of membrane bioreactor process capabilities
to meet stringent effluent nutrient discharge requirements. Water Environ. Res.
77, 162-178.
Gander, M., Jefferson, B., Judd, S. (2000) Aerobic MBRs for domestic
wastewater treatment: a review with cost considerations. Sep. Purif. Technol. 8,
119-130.
Giesy, J.P. and Kannan, K. (2001) Global distribution of perfluorooctane
sulfonate in wildlife. Environ. Sci. Technol. 35, 1339-1342.
Giesy, J.P. (2006) Perfluorinated compounds in streams of the Shihwa
industrial zone and lake Shihwa, South Korea. Environ. Toxicol. Chem. 25,
2374-2380.
Gilliland, F. (1992) Fluorochemicals and human health: Studies in an
occupational cohort. Doctoral thesis, Division of Environmental and
Occupational Health, University of Minnesota. US EPA, AR226-0473.
Reference
175
Grady Jr., C.P.L., Daigger, G.T., Lim, H.C. (1999) Biological Wastewater
Treatment, second ed. Marcel Dekker, New York.
Gunder, B. (2001) The membrane-coupled activated sludge process in
municipal wastewater treatment. Technomic, Lancaster, PA, USA.
Hansen, K.J., Clemen, L.A., Ellefson, M.E. and Johnson, H.O. (2001)
Compound-specific, quantitative characterization of organic: Fluorochemicals
in biological matrices. Environ. Sci. Technol.35, 766-770.
Hansen, K.J., Johnson, H.O., Eldridge, J.S., Butenho., J.L., Dick, L.A. (2002)
Quantitative characterization of trace levels of PFOS and PFOA in the
Tennessee River. Environ. Sci. Technol. 36, 1681-1685.
Harada, K., Saito, N., Sasaki, K., Inoue, K., Koizumi, A. (2003)
Perfluorooctane sulfonate contamination of drinking water in the Tama River,
Japan: estimated effects on resident serum levels. Bull. Environ. Contam.
Toxicol. 17, 31-36.
Hatfield, T.L. (2001) Screening studies on the aqueous photolytic degradation
of perfluorooctanoic acid (PFOA), 3M Environmental Laboratory, USEPA
docket: EPA-HQ-OPPT-2002-0051-0023. U.S. Environmental Protection
Agency, Washington, DC.
Haughom, B., Spydevold, O. (1992) The mechanism underlying the
hypolipemic effect of perfluorooctanoic acid (PFOA), perfluorooctane
sulphonic acid (PFOSA) and clofibric acid. Biochim. Biophys. Acta. 1128, 65-
72.
Hekster, F., Pijnenburg, J.R.W, Laane, P.M. and Voogt, P.d. (2002)
Perfluoroalkylated substances: Aquatic environmental assessment, ministry of
Reference
176
transport, public works and water management, 99 pp, Report RIKZ 2002.043,
Hague, Netherlands.
Higgins, C.P., Field, J.A., Criddle, C.S., Luthy, R.G. (2005) Quantitative
determination of perfluorochemicals in sediments and domestic sludge.
Environ. Sci. Technol. 39, 3946-3956.
Higgins, C.P. and Luthy, R.G. (2006) Sorption of perfluorinated surfactants on
sediments. Environ. Sci. Technol.40(23), 7251-7256.
Hof, P. T., Scheirs, J., van de Vijver, K., van Dongen, W., Esmans, E. L.,
Blust, R., de Coen, W. (2004) Biochemical Effect Evaluation of
Perfluorooctane Sulfonic Acid-Contaminated Wood Mice (Apodemus
sylvaticus). Environ. Health Perspect. 112, 681-686.
Houde, M., Martin, J.W., Letcher, R.J., Solomon, K.R., Muir, D.C.G. (2006)
Biological monitoring of polyfluoroalkyl substances: a review. Environ. Sci.
Technol. 40, 3463-3473.
Hu, J.Y., Jin, X, Ong, S.L. (2007) Rejection of estrone by nanofiltration:
Influence of solution chemistry. J. Membr. Sci. 302, 188-196.
Huang, X., Liu, R. and Qian, Y. (2000) Behaviour of soluble microbial
products in a membrane bioreactor. Process Biochem. 36, 401-406.
Huang, X., Gui, P. and Qian, Y. (2001) Effect of sludge retention time on
microbial behaviour in a submerged membrane bioreactor. Process Biochem.
36, 1001-1006.
Huset, C.A., Chiaia, A.C., Barofsky, D.F., Jonkers, N., Kohler, H.E., Ort, C.,
Giger, D.W., Field, J.A. (2008) Occurrence and mass flows of fluorochemicals
in the Glatt Valley watershed, Switzerland. Environ. Sci. Technol. 42, 6369-
6377.
Reference
177
Irvine, K.N., Loganathan, B.G., Pratt, E.J. and Sikka, H.C. (1993) Calibration
of PCSWMM to estimate metals, PCBs, and HCB in CSOs from an Industrial
sewershed. In. New Techniques for Modelling the Management of Stormwater
Quality Impacts. (Ed. W. James). Lewis Publ. pp 215-242.
Iversen V, Mehrez R, Horng RY. (2009) Fouling mitigation through
flocculants and adsorbents addition in membrane bioreactors: Comparing lab
and pilot studies. J. Membr. Sci. 345, 21-30.
Jahnke, A., L. Ahrens, R. Ebinghaus and C. Temme (2007). Urban versus
Remote Air Concentrations of Fluorotelomer Alcohols and Other
Polyfluorinated Alkyl Substances in Germany. Environ. Sci.Technol. 41, 745-
752.
Jin, YH, Liu, W, Sato, I, Nakayama SF, Sasaki K, Saito N, Tsuda S . (2008)
PFOS and PFOA in environmental and tap water in China. Chemosphere 77,
605-611.
Kallenborn, R., Berger, U., Järnberg, U. (2004) Perfluorinated Alkylated
Substances (PFAS) in the Nordic Environment, Norwegian Institute for Air
Research, Oslo, Norway, 107 pp.
Kannan, K., Newsted, J.L., Halbrook, R.S., Giesy, J.P. (2002) Perfluorooctane
sulfonate and Related Fluorinated Hydrocarbons in Mink and River Otters
from the United States. Environ. Sci. Technol. 36, 2566-2571.
Kannan, K., Corsolini, S., Falandysz, J., Fillmann, G., Kumar, K.S.,
Loganathan, B.G., Mohd, M.A., Olivero, J., Van Wouwe, N., Yang, J.H. and
Aldous, K.M. (2004) Perfluorooctanesulfonate and related fluorochemicals in
human blood from several countries. Environ. Sci. Technol. 38, 4489-4495.
Kaiser, M.A., Larsen, B.S., Kao, C.-P.C., Buck, R.C. (2005) Vapor Pressures
Reference
178
of Perfluoro -Octanoic, -Nonanoic, -Decanoic, -Undecanoic, and -Dodecanoic
acids. J. Chem. Eng. Data published online Aug. 26,
http://dx.doi.org/10.1021/je050070r.
Katsoyiannis, A. and Samara, C. (2005) Persistent organic pollutants (POPs)
in the conventional activated sludge treatment process: fate and mass balance,
Environ. Res. 97, 245-257.
Key, B.D., Howell, R.D. and Criddle, C.S. (1997) Fluorinated organics in the
biosphere. Environ. Sci. Technol. 31, 2445-2454.
Kilduff, J.E., Karanfil, T. and Weber, Jr. W.J. (1998) TCE adsorption by GAC
preloaded with humic substances, J. Am. Water Works Assoc. 90, 76-89.
Kim, J.S., Lee, C.H. and Chang, H.D. (1998) Comparison of ultrafiltration
characteristics between activated sludge and BAC sludge. Water Res. 32,
3443-3451.
Kim, J.S., Lee, C.H. and Chang, I.S. (2001) Effect of pump shear on the
performance of a crossflow membrane bioreactor. Water Res. 35, 2137-2144.
Kimura, K., Amy, G., Drewes, J. E., Heberer, T., Kim, T. U., Watanabe, Y.
(2003) Rejection of organic micropollutants (disinfection by-products,
endocrine disrupting compounds, and pharmaceutically active compounds) by
NF/RO membranes. J. Membr. Sci. 227, 113.
Kimura, K., Hara, H., Watanabe, Y. (2007) Elimination of selected acidic
pharmaceuticals from municipal wastewater by an activated sludge system and
membrane bioreactors. Environ. Sci. Technol. 41, 3708-3714.
Kiso, Y., Sugiura, Y., Kitao, T., Nishimura, K. (2001) Effects of
hydrophobicity and molecular size on rejection of aromatic pesticides with
nanofiltration membranes. J. Membr. Sci. 192, 1-9.
Reference
179
Kissa, E. (2001) Fluorinated surfactants and repellents, second ed. Marcel
Dekker, New York.
Kuchler, I.L. and Miekeley, N. (1994) Ultrafiltration of humic compounds
through low molecular mass cut-off level membranes. Sci. Total Environ. 154,
23-28.
Kudo, N. and Y. Kawashima (2003). Toxicity and toxicokinetics of
perfluorooctanoic acid in humans and animals. J. Toxicol. Sci. 2, 49-57.
Lange, C.C. (2000) 3M Environmental Laboratory. The aerobic
biodegradation of N-EtFOSE alcohol by the microbial activity present in
municipal wastewater treatment sludge; Report CA058, Docket AR-226-
1030a078; U.S. Environmental Protection Agency: Washington, DC.
Lange, C.C. (2002) 3M Environmental Laboratory. Biodegradation screen
study for telomer type alcohols; Docket AR226-1149; U.S. Environmental
Protection Agency: Washington, DC.
Lau, C., Thibodeaux, J.R., Hanson, R.G., Rogers, J.M., Grey, B.E., Stanton,
M.E., Butenhoff, J.L. and Stevenson L.A. (2003) Exposure to perfluorooctane
sulfonate during pregnancy in rat and mouse. II: Postnatal evaluation, Toxicol.
Sci. 74, 382-392.
Lau, C., Butenhoff, J.L. and Rogers, J.M. (2004) The developmental toxicity
of perfluoroalkyl acids and their derivatives. Toxicol. Appl. Pharmacol. 198,
231-241.
Le-Clech, P., Jefferson B., Chang, I.S., Judd, S.J. (2003) Critical flux
determination by the flux-step method in a submerged membrane bioreactor. J.
Membr. Sci. 227, 81-93.
Lee, K.M., Lim, P.E. (2005) Bioregeneration of powdered activated carbon in
Reference
180
the treatment of alkyl-substituted phenolic compounds in simultaneous
adsorption and biodegradation process. Chemosphere 58, 407-416.
Lesage, N., Sperandio, M., Cabassud, C. (2007) Study of a hybrid process:
adsorption on activated carbon/membrane bioreactor for the treatment of an
industrial wastewater. Chem. Eng. Process 47, 303-307.
Li, Y.Z., He, Y.L., Liu, Y.H., Yang, S.C., Zhang, G.J. (2005) Comparison of
the filtration characteristics between biological powdered activated carbon
sludge and activated sludge in submerged membrane bioreactors. Desalination
174, 305-314.
Liao, B.Q., Allen, D.G., Droppo, I.G., Leppard, G.G. and Liss, S.N. (2001)
Surface properties of sludge and their role in bioflocculation and settleability.
Water Res. 35, 339-350.
Loewen, M., Halldorson, T., Wang, F., Tomy, G. (2005) Fluorotelomer
Carboxylic Acids and PFOS in Rainwater from an Urban Center in Canada.
Environ. Sci. Technol. 39, 2944-2951.
Löfstrand, K., Jörundsdóttir, H., Tomy, G., Svavarsson, J., Pál Weihe, Torgeir
Nygård, Åke Bergman. (2008) Spatial trends of polyfluorinated compounds in
guillemot (Uria aalge) eggs from North-Western Europe. Chemosphere 72,
1475-1480
Logan, B.E. and Jiang, Q. (1990) Molecular size distributions of dissolved
organic matter. J. Environ. Eng. Div. ASCE 116, 1046-1062.
Loganathan, B.G., Irvine, K.N., Kannan, K. Pragatheeswaran, V. and Sajwan,
K.S. (1997) Distribution of selected PCB congeners in the Babcock Street
Sewer District: A multimedia approach to identify PCB sources in combined
Reference
181
sewer overflows (CSOs) discharging to the Buffalo River, New York. Arch.
Environ. Contam. Toxicol. 33, 130-140.
Loganathan, B.G., Sajwan, K.S., Sinclair, E., Kumar, K.S. and Kannan, K.
(2007) Perfluoroalkyl sulfonates and perfluorocarboxylates in two wastewater
treatment facilities in Kentucky and Georgia. Water Res. 41, 4611-4620.
Loos, R., Wollgast, J., Huber, T., Hanke, G. (2007) Polar herbicides,
pharmaceutical products, perfluorooctanesulfonate (PFOS),
perfluorooctanoate (PFOA), and nonylphenol and its carboxylates and
ethoxylates in surface and tap waters around Lake Maggiore in Northern Italy.
Anal. Bioanal. Chem. 387, 1469-1478.
Loos, R., Locoro, G., Huber, T., Wollgast, J., Christoph, E.H., De Jager A,
Gawlik, B.M., Hanke, G., Umlauf, G., Zaldivar, J.M. (2008) Analysis of
perfluorooctanoate (PFOA) and other perfluorinated compounds (PFCs) in the
River Po watershed in N-Italy. Chemosphere 71, 306-313.
Martin, JW, Mabury, SA, Solomon, KR and Muir, DCG (2003a). Dietary
accumulation of perfluorinated acids in juvenile rainbow trout (Oncorhynchus
mykiss). Environ. Toxicol. Chem. 22, 189-195.
Martin, JW, Mabury, SA, Solomon, KR and Muir, DCG (2003b).
Bioconcentration and tissue distribution of perfluorinated acids in rainbow
trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 22, 196-204.
Martin, J.W., Kannan, K., Berger, U., De Voogt, P., Field, J., Franklin, J.,
Giesy, J.P., Harner, T., Muir, D.C.G., Scott, B., Kaiser, M., Jarnberg, U.,
Jones, K.C., Mabury, S.A., Schroeder, H., Simcik, M., Sottani, C., Van Bavel,
B., Karrman, A., Lindstrom, G. and Van Leeuwen, S. (2004a) Analytical
Reference
182
challenges hamper perfluoroalky research. Environ. Sci. Technol. 38, 248A-
255A.
Martin, J.W., Whittle, D.M., Muir, D.C.G. and Mabury, S.A. (2004b)
Perfluoroalkyl contaminants in a food web from lake Ontario. Environ. Sci.
Technol. 38, 5379-5385.
Masse, A., Sperandio, M., Cabassud, C. (2006) Comparison of sludge
characteristics and performance of a submerged membrane bioreactor and an
activated sludge process at high solids retention time. Water Res. 40, 2405-
2415.
Matsui, Y., Fukuda, Y., Inoue, T., Matsushita, T. (2003) Effect of natural
organic matter on powdered activated carbon adsorption of trace contaminants:
characteristics and mechanism of competitive adsorption. Water Res. 37,
4413-4424.
Matuszewski, B.K. (2006) Standard line slopes as a measure of a relative
matrix effect in quantitative HPLC-MS bioanalysis. J. Chromatogr. B 830,
293-300.
Matuszewski, B.K., Constanzer, M.L. and Chavez-Eng, C.M. (2003)
Strategies for the assessment of matrix effect in quantitative bioanalytical
methods based on HPLC-MS/MS. Anal. Chem. 75, 3019-3030.
MDH (2007) Health Based Values for Perfluorooctanoic Acid (PFOA).
http://www.health.state.mn.us/divs/eh/hazardous/topics/pfcs/pfoamemo0307.p
df. U.S. Environmental Protection Agency, Washington, DC.
Moody, C.A., Field, J.A. (2000) Perfluorinated surfactants and the
environmental implications of their use in fire-fighting foams. Environ. Sci.
Technol. 34, 3864-3870.
Reference
183
Moody, C.A., Kwan, W.C., Martin, J.W., Muir, D.C.G. and Mabury, S.A.
(2001) Determination of perfluorinated surfactants in surface water samples
by two independent analytical techniques: Liquid chromatography/tandem
mass spectrometry and F-19 NMR. Anal. Chem. 73, 2200-2206.
Moriwaki, H., Takagi, Y., Tanaka, M., Tsuruho, K., Okitsu, K., Maeda, Y.
(2005) Sonochemical decomposition of perfluorooctane sulfonate and
perfluorooctanoic acid. Environ. Sci. Technol. 39, 3388-3392.
Munz, G., Gori, R., Mori, G., Lubello, C. (2007) Powdered activated carbon
and membrane bioreactors (MBR-PAC) for tannery wastewater treatment:
long term effect on biological and filtration process
performances. Desalination 207, 349-360.
Murakami, M., Shinohara, H., Takada, H. (2009) Evaluation of wastewater
and street runoff as sources of perfluorinated surfactants (PFSs). Chem. 74,
487-493.
National Environmental Agency (2008) Daily rainfall at 5 standard stations for
the period, January 2006 to December 2007. Ministry of the Environment and
Water Resources, Singapore.
Newcombe, G., Morrison, J., Hepplewhite, C., Knappe, D.R.U. (2002)
Simultaneous adsorption of MIB and NOM onto activated carbon: II.
Competitive effects. Carbon 40, 2147-2156.
North California Division of Water Quality (2006), Recommended interim
maximum allowable concentration for perfluorooctanoic acid. U.S.
Environmental Protection Agency, Washington, DC.
Reference
184
Ng, C. A. , Sun, D. and Fane, A. G. (2006) Operation of Membrane Bioreactor
with Powdered Activated Carbon addition. Separation Sci. and Tech. 41,
1447-1466.
Ng, H.Y., Hermanowicz, S.W. (2005) Membrane bioreactor operation at short
solids retention times: Performance and biomass characteristics. Water Res.,
39, 981-992.
Nozoe, M., S. Fujii, S. Tanaka, H. Tanaka and N. Yamashita (2006).
Investigation of PFOS and PFOA in a wastewater treatment plant. Environ.
Eng. Res. 43, 105-111.
Ochoa-Herrera,V., Sierra-Alvarez, R. (2008) Removal of perfluorinated
surfactants by sorption onto granular activated carbon, zeolite and sludge.
Chemosphere 72, 1588-1593.
OECD (2002). Hazard Assessment of Perfluorooctane Sulfonate (PFOS) and
its Salts.
ENV/JM/RD(2002)17/FINAL. Organisation for Economic Co-operation and
Development, Paris, France.
Ohya, T., N. Kudo, E. Suzuki and Y. Kawashima (1998). Determination of
perfluorinated carboxylic acids in biological samples by high-performance
liquid chromatography. J. Chromatogr. B 720, 1-7.
Ooi, J.B. and Chia, L.S. (1974) The climate of west Malaysia and Singapore.
Oxford University Press.
Pelekani, C., Newcombe, G., Snoeyink, V., Hepplewhite, C., Assemi, S.,
Beckett, R. (1999) Characterization of natural organic matter using high
performance size exclusion chromatography. Environ Sci Technol. 33, 2807-
22813.
Reference
185
Poulsen, P.B., Jensen, A.A., Wallström, E. (2005) More environmentally
friendly alternatives to PFOS-compounds and PFOA. Environmental
Protection Agency, Danish Ministry of the Environment.
Petrovic, M., Hernando, M.D., Diaz-Cruz, M.S. and Barcelo, D. (2005) Liquid
chromatography-tandem mass spectrometry for the analysis of pharmaceutical
residues in environmental samples: a review. J. Chromatogr. A 1067, 1-14.
Powley, C.R., George, S.W., Ryan, T.W. and Buck, R.C. (2005) Matrix effect-
free analytical methods for determination of perfluorinated carboxylic acids in
environmental matrixes. Anal. Chem. 77, 6353-6358.
Prevedouros, K., Cousins, I.T., Buck, R.C., Korzeniowski, S.H. (2006)
Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 40,
32-44.
Qiu, Y., Fujii, J., Tanaka, S., Lien, N.P.H., Koizumi, A., (2005). Removal
characteristics of perfluorooctane sulfonate and perfluorooctane acid by
granular activated carbon adsorption. Proc. of Joint 14th KU-KAIST-NTU-
NUS Sympo. in Env. Eng., 63-72.
Quinlivan, P.A, Li, L., Knappe, D.R.U. (2005) Effects of activated carbon
characteristics on the simultaneous adsorption of aqueous organic
micropollutants and natural organic matter. Water Res. 39, 1663-1673
Saito, N., Sasaki, K., Nakatome, K., Harada, K., Yoshinaga, T., Koizumi, A.
(2003) Perfluorooctane sulfonate concentrations in surface water in Japan.
Arch. Environ. Contam. Toxicol. 45, 149-158.
Satyawali, Y. and Balakrishnan, M. (2009) Effect of PAC addition on sludge
properties in an MBR treating high strength wastewater. Water Res. 43, 1577-
1588.
Reference
186
Schafer, A. I., Nghiem, L. D., Waite, T. D. (2003) Removal of the natural
hormone estrone from aqueous solutions using nanofiltration and reverse
osmosis. Environ. Sci. Technol. 37, 182.
Schultz, M.M., Barofsky D.F. and Field J.A. (2003). Fluorinated alkyl
surfactants. Environ. Eng. Sci. 20, 487-501.
Schultz, M.M., Barofsky, D.F., Field, J.A. (2004) Quantitative determination
of fluorotelomer sulfonates in groundwater by LC MS/MS. Environ. Sci.
Technol. 38, 1828-1835.
Schultz, M.M., Barofsky, D.F. and Field, J.A. (2006a) Quantitative
determination of fluorinated alkyl substances by large-volume-injection liquid
chromatography tandem mass spectrometry - Characterization of municipal
wastewaters. Environ. Sci. Technol. 40, 289-295.
Schultz, M.M., Higgins, C.P., Huset, C.A., Luthy, R.G., Barofsky, D.F. and
Field, J.A. (2006b) Fluorochemical mass flows in a municipal wastewater
treatment facility. Environ. Sci. Technol. 40, 7350-7357.
Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M. (2003) Environmental
Organic Chemistry, 2nd ed.; Wiley: Hoboken, NJ.
Scott, B.F., Spencer, C., Mabury, S.A., Muir, D.C.G. (2006) Poly and
perfluorinated carboxylates in North American precipitation. Environ. Sci.
Technol. 40, 7167-7174.
Seacat, A.M., Thomford, P.J., Hansen, K.J. Clemen, L.A., Eldridge, S.R.,
Elcombe, C.R., and Butenhoff J.L. (2003) Sub-chronic dietary toxicity of
potassium perfluorooctanesulfonate in rats. Toxicology 183, 117-131.
Reference
187
Shimizu, Y., Okuno, Y.I., Uryu, K., Ohtsubo S., Watanabe, A. (1996)
Filtration characteristics of hollow fiber microfiltration membranes used in
membrane bioreactor for domestic wastewater. Water Res. 30, 2385-2392.
Shon, H.K., Vigneswaran, S., Snyder, S.A. (2006) Effluent organic matter
(EfOM) in wastewater: Constituents, effects, and treatment. Crit. Rev. Env. Sci.
Tec. 36, 327-374.
Skutlarek, D., M. Exner and H. Färber (2006). Perfluorinated surfactants in
surface and drinking waters, Environ. Sci. Pollut. Res. Int. 13, 299-307.
Shoeib, M., T. Harner and P. Vlahos (2006). Perfluorinated Chemicals in the
Arctic Atmosphere. Environ. Sci. Technol. 40, 7577-7583.
Simcik, M.F. and Dorweiler, K.J. (2005) Ratio of perfluorochemical
concentrations as a tracer of atmospheric deposition to surface waters. Environ.
Sci. Technol. 39, 8678-8683.
Sinclair, E. and Kannan, K. (2006a) Mass loading and fate of perfluoroalkyl
surfactants in wastewater treatment plants. Environ. Sci. Technol. 40, 1408-
1414.
Sinclair, E., Mayack, D.T., Roblee, K., Yamashita, N., Kannan, K. (2006b)
Occurrence of perfluoroalkyl surfactants in water, fish, and birds from New
York State. Arch. Environ. Contam. Toxicol. 50, 398-410.
So, M.K., Taniyasu, S., Yamashita, N., Giesy, J.P., Zheng, J., Fang, Z., Im,
S.H. and Lam, P.K.S. (2004) Perfluorinated compounds in coastal waters of
Hong Kong, South China, and Korea. Environ. Sci. Technol. 38, 4056-4063.
So, M. K., Miyake, Y., Yeung, W. Y, Ho, Y. M, Taniyasu, S., Rostkowski, P.,
Yamashita, N, Zhou, B. S., Shi, X. J., Wang, J. X., Giesy, J. P, Yu, H., Lam, P.
Reference
188
K. S. (2007) Perfluorinated compounds in the Pearl River and Yangtze River
of China. Chemosphere 68, 2085-2095.
Stock, N.L., Lau, F.K., Ellis, D.A., Martin, J.W., Muir, D.C.G., Mabury, S.A.
(2004) Polyfluorinated telomer alcohols and sulfonamides in the north
American troposphere. Environ. Sci. Technol. 38, 991-996.
Szostek, B. and Prickett, K.B. (2004) Determination of 8:2 fluorotelomer
alcohol in animal plasma and tissues by gas chromatography-mass
spectrometry. J. Chromatogr. B 813, 313-321.
Tang, C.Y., Fu, Q.S., Robertson, A.P., Criddle, C.S., Leckie, J.O. (2006) Use
of reverse osmosis membranes to remove perfluorooctane sulfonate (PFOS)
from semiconductor wastewater. Environ. Sci. Technol. 40, 7342-7349.
Taniyasu, S., Kannan, K., Horii, Y., Hanari, N. and Yamashita, N. (2003) A
survey of perfluorooctane sulfonate and related perfluorinated organic
compounds in water, fish, birds, and humans from Japan. Environ. Sci.
Technol. 37, 2634-2639.
Taniyasu, S., Kannan, K., So, M.K., Gulkowska, A., Sinclair, E., Okazawa, T.
and Yamashita, N. (2005) Analysis of fluorotelomer alcohols, fluorotelorner
acids, and short- and long-chain perfluorinated acids in water and biota. J.
Chromatogr. A 1093, 89-97.
Taves, D. (1968) Evidence that there are two forms of fluoride in human
serum, Nature 217, 1050-1051.
Thibodeaux, J.R., Hanson, R.G., Rogers, J.M., Grey, B.E., Barbee, B.D.,
Richards, J.H., Butenhoff, J.L, Stevenson, L.A. and Lau, C. (2003). Exposure
to perfluorooctane sulfonate during pregnancy in rat andmouse. I: Maternal
and prenatal evaluations. Toxicol. Sci. 74, 369-381.
Reference
189
Tomy, G.T., Budakowski, W., Halldorson, T., Helm, P.A., Stern, G.A.,
Friesen, K., Pepper, K., Tittlemier, S.A. and Fisk, A.T. (2004) Fluorinated
organic compounds in an eastern Arctic marine food web. Environ. Sci.
Technol. 38, 6475-6481.
Tseng, C.L., Liu, L.L., Chen, C.M. and Ding, W.H. (2006) Analysis of
perfluorooctanesulfonate and related fluorochemicals in water and biological
tissue samples by liquid chromatography-ion trap mass spectrometry. J.
Chromatogr. A 1105, 119-126.
US EPA (2002) Draft hazard assessment of perfluorooctanoic acid and its salts,
February 20, 2002. U.S. Environmental Protection Agency, Washington, DC.
US EPA (2003) Preliminary risk assessment of the developmental toxicity
associated with exposure to perfluorooctanoic acid and its salts. Office of
Pollution Prevention and Toxics, Risk Assessment Division. U.S.
Environmental Protection Agency, Washington, DC.
US EPA (2005) Preliminary Risk Assessment of the Developmental Toxicity
Associated with Exposure to Perfluorooctanoic Acid and its Salts. Office of
Pollution Prevention and Toxics, Risk Assessment Division, U.S.
Environmental Protection Agency, Washington, DC.
US EPA (2009) Provisional Health Advisories for Perfluorooctanoic Acid
(PFOA) and Perfluorooctane Sulfonate(PFOS).
http://www.epa.gov/waterscience/criteria/drinking
/pha-PFOA_PFOS.pdf. U.S. Environmental Protection Agency, Washington,
DC.
van De Steene, J.C., Mortier, K.A. and Lambert, W.E. (2006) Tackling matrix
effects during development of a liquid chromatographic-electrospray
Reference
190
ionisation tandem mass spectrometric analysis of nine basic pharmaceuticals
in aqueous environmental samples. J. Chromatogr. A 1123, 71-81.
van Leeuwen, S.P.J., Karrman, A., Van Bavel, B., De Boer, J. and Lindstrom,
G. (2006) Struggle for quality in determination of perfluorinated contaminants
in environmental and human samples. Environ. Sci. Technol. 40, 7854-7860.
van Leeuwen, S.P.J. and de Boer, J. (2007) Extraction and clean-up strategies
for the analysis of poly- and perfluoroalkyl substances in environmental and
human matrices. J. Chromatogr. A 1153, 172-185.
Vigneswaran, S., Kwon, D.Y., Ngo, H.H., Hu, J.Y. (2000) Improvement of
microfiltration performance in water treatment: is critical flux a viable solution?
Water Sci. Technol. 41, 309-315.
Visvanathan, C., Aim, R. B. and Parameshwaran, K. (2000) Membrane
separation
bioreactors for wastewater treatment. Crit. Rev. Environ. Sci. Technol. 30, 1-
48.
Wang, N., Szostek, B., Folsom, P.W., Sulecki, L.M., Capka, V., Buck, R.C.,
Berti, W.R. and Gannon, J.T. (2005) Aerobic biotransformation of C-14-
labeled 8-2 telomer B alcohol by activated sludge from a domestic sewage
treatment plant. Environ. Sci. Technol. 39, 531-538.
Weremiuk, A.M., Gerstmann, S., Frank, H. (2006) Quantitative determination
of perfluorinated surfactants in water by LC-ESI-MS/MS. J. Sep. Sci. 29,
2251-2255.
West Virginia Department of Environmental Protection (2002). Final
ammonium perfluorooctanoate (C8) Assessment of Toxicity Team (CATT)
report. U.S. Environmental Protection Agency, Washington, DC.
Reference
191
Xing, C.H., Tardieu, E., Qian, Y. and Wen, X.H. (2000) Ultrafiltration
membrane bioreactor for urban wastewater reclamation. J. Membr. Sci. 177,
73-82.
Xing, C.H., Wen, X.H., Qian, Y. and Tardieu, E. (2001) Microfiltration-
membrane coupled bioreactor for urban wastewater reclamation. Desalination
141, 63-73.
Yamashita, N., K. Kannan, S. Taniyasu, Y. Horii, T. Okazawa, G. Petrick and
T. Gamo (2004). Analysis of perfluorinated acids at parts-per-quadrillion
levels in seawater using liquid chromatography-tandem mass spectrometry,
Environ. Sci. Technol., 38, 5522-5528.
Yamashita, N., Kannan, K., Taniyasu, S., Horii, Y., Petrick, G., Gamo, T.
(2005) A global survey of perfluorinated acids in oceans. Mar. Pollut. Bull. 51,
658-668.
Yu, J., Hu, J.Y., Tanaka, S., Fujii, S. (2009a) Perfluorooctane sulfonate (PFOS)
and perfluorooctanoic acid (PFOA) in sewage treatment plants. Water Res. 43,
2399-2408.
Yu Q, Zhang R, Deng S, Huang J, Yu G. (2009b) Sorption of perfluorooctane
sulfonate and perfluorooctanoate on activated carbons and resin: kinetic and
isotherm study. Water Res. 43, 1150-1158.
Zhao, X., Li, J., Shi, Y., Cai, Y., Mou S. and Jiang G. (2007). Determination
of perfluorinated compounds in wastewater and river water samples by mixed
hemimicelle-based solid-phase extraction before liquid chromatography-
electrospray tandem mass spectrometry detection. J. Chromatogr. A 1154, 52-
59.
Reference
192
Zushi, Y., Takeda, T., Masunaga, S. (2008) Existence of nonpoint source of
erfluorinated compounds and their loads in the Tsurumi River basin, Japan.
Chemosphere 71, 1566-1573.
Appendix: Publications
193
APPENDIX: PUBLICATIONS
Journal Paper:
1. Yu, J., Hu, J.Y., Tanaka, S., Fujii, S. (2009) Perfluorooctane sulfonate
(PFOS) and perfluorooctanoic acid (PFOA) in sewage treatment plants. Water
Res. 43, 2399-2408.
2. Hu, J.Y. and Yu, J. (2010) Development and Validation of a LC-MS-MS
method for the Determination of perfluorinated compounds in environmental
matrices. Chromatographia. 72, 411-416.
3. Hu, J.Y., Yu, J., Tanaka, S., Fujii, S. (2011) Perfluorooctane sulfonate
(PFOS) and perfluorooctanoic acid (PFOA) in water environment of
Singapore. Water Air Soil Pollut. 216,179-191.
4. Yu, J. and Hu, J.Y. (2010) Adsorption of perfluorinated compounds onto
activated carbon and activated sludge. (submitted to J Environ. Eng.)
Conference Paper:
1. Yu, J. and Hu, J.Y. (2007) Occurrence of pharmaceuticals in treated sewage
of local WRPs. The 16th Joint KAIST-KYOTO-NTU-NUS Symposium on
Environmental Engineering, Taiwan, pp.206-215.
2. Hu, J.Y. and Yu, J. (2007) Occurrence of perfluorooctane sulfonate (PFOS)
and perfluorooctanoic acid (PFOA) in water and wastewater of Singapore. The
16th Joint KAIST-KYOTO-NTU-NUS Symposium on Environmental
Engineering, Taiwan, pp.129-139.