ABSTRACT
Title of dissertation: THE FATE AND BEHAVIOR OF OCTYL- AND 
NONYLPHENOL ETHOXYLATES AND THEIR 
DERIVATIVES IN THREE AMERICAN 
WASTEWATER TREATMENT PLANTS AND THE 
BACK RIVER, MARYLAND
Jorge Eduardo Loyo-Rosales, Doctor of Philosophy, 2006
Dissertation directed by: Professor Alba Torrents
Department of Civil and Environmental Engineering
and
Dr. Clifford P. Rice
U. S. Department of Agriculture
The octyl- and nonylphenol ethoxylates (collectively known as alkylphenol 
ethoxylates, APEOs) are a family of widely used surfactants  in industrial processes and 
as detergents in both industrial and household applications.  After being used, the APEOs 
are transformed into more toxic and endocrine disrupting products, such as short-chain 
APEOs, nonylphenol (NP), and octylphenol (OP).  The main objective of the present 
work was to study the fate of the APEOs and transformation products (APEs) in three 
American wastewater treatment plants (WWTPs) and in Back River, an estuary of the 
Chesapeake Bay that receives treated effluent from one of the plants.  In order to 
accomplish this, analytical methods were developed based on solid-phase extraction for 
water, accelerated solvent extraction for solids, and isotope dilution liquid 
chromatography/tandem mass spectrometry for quantitation.  Analysis in the WWTPs 
showed that influent wastewater had similar APEs concentrations, whereas effluent 
concentrations were only similar when samples from the same season (fall or winter) 
were compared, with concentrations being several times higher in winter than in fall.  
Sorption to particulate was approximately 1.6 times higher for nonylphenolic compounds 
than for their octylphenolic counterparts, in agreement with their difference in K
ow
values.  Effluent concentrations and APEO removal rates?the latter averaging 99% in 
summer and 94% in winter for the NPEOs?were strongly correlated to water 
temperature, and no correlation was found with suspended solids or organic carbon 
removal.  In Back River the most abundant of the APEs were the carboxylated 
transformation products (APECs, > 95% on mass basis), followed by NP in September 
and October, and NP1-2EO in March.  NP concentrations found, 0.087 ? 0.69 ?g/L, were 
below acute toxicity thresholds, and generally below recently proposed water quality 
criteria by the US EPA.  Total NPE concentrations in the Back River seemed to vary in 
accordance to the concentrations in the WWTP effluent, especially in the case of the 
APECs.  However, a closer analysis of the data suggested that in the fall sampling events, 
when rain occurred, the ethoxylates present in the particulate matter originated in the 
river?s tributaries rather than the WWTP.
THE FATE AND BEHAVIOR OF OCTYL- AND NONYLPHENOL ETHOXYLATES 
AND THEIR DERIVATIVES IN THREE AMERICAN WASTEWATER 
TREATMENT PLANTS AND THE BACK RIVER, MARYLAND
by
Jorge Eduardo Loyo-Rosales
Dissertation submitted to the Faculty of the Graduate School of the
University of Maryland, College Park in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
2006
Advisory Committee:
Professor Alba Torrents, Chair/Advisor
Professor Joel E. Baker
Professor Oliver J. Hao
Professor Alice Mignerey, Dean Representative
Dr. Clifford P. Rice, Co-Advisor
ii
To the memory of
Hugo Sousa
... parce que la terre
n?est pas
une vall?e des larmes
el cielo muri? en sus ojos
y su rostro se detuvo
en un rapto visceral
poco envidi? al sue?o
envuelto en m?tica coraza
y tampoco exento de dolor
las llamas dejaron l?grimas
hijas del suelo
voluntad innegable
rodeada de llanto y soledad
deja la tierra roja
esta sopa primigenia
sangre borrada
del pecado original
el viento se rompi? en pedazos
y un trozo olvid? tu nombre
en la tersa soledad del recuerdo
el destino forj? un suspiro
que apag? el color del tiempo
doblen la esquina las aguas turbulentas
me invade el horror de su silencio
iii
ACKNOWLEDGMENTS
No scientific project can be an isolated effort; let alone a project that attempts to 
study environmental issues.  This work is no exception, and it would have never been 
completed without the help of innumerable individuals.  First, I would like to thank my 
two advisors, Alba Torrents at UMD and Clifford Rice at the USDA, for their guidance 
and friendship.  Alba convinced me to come to Maryland in the first place and always 
encouraged me to take the next step. Cliff introduced me to the alkylphenol ethoxylates
and his door has always been open to discuss any topic, from experimental results to 
Mexican politics.  I also learnt from him how to repair many lab fixtures or, at least, not 
to be afraid to try.
I also wish to thank the rest of my PhD committee, Joel Baker, Oliver Hao and 
Alice Mignerey, for taking the time to read and comment on my dissertation.  Joel Baker 
also gave us the opportunity to take part in some of his team?s sampling trips to the 
Baltimore Harbor area.
Many people from different institutions contributed to the alkylphenol project in 
varied ways. None of the work in the wastewater treatment plants and the Back River 
would have been possible without the cooperation of the plants? personnel.  They 
provided us with samples and a substantial portion of the ancillary information.  Larry 
Barber and the people at EPA Region V were instrumental in obtaining the samples from 
the Chicago area. Our work with Steve Corsi started when he hired our group to analyze 
iv
some of his airport runoff samples and it has morphed into a very prolific and enjoyable 
collaboration.
I wish to acknowledge the financial support from a Conacyt-Fulbright/Garcia-
Robles scholarship provided by the Mexican and American governments.  Partial funding 
for this project also came from the Maryland Water Resources Research Center.  I am 
also indebted to the people who donated analytical standards for this project: P. Lee 
Ferguson provided the [
13
C
6
]-labeled compounds; Larry Barber the Marlophen 810; 
Carter Naylor the NP0EC, OP0EC, and Surfonic N-10; US EPA Region V the NP2EO 
and NP1EC; and Schenectady International the NP standard.  Barbara Losey provided us 
with Surfonic N-95 and facilitated its characterization by Huntsman.
Special thanks to my friends at the Environmental Quality Lab and the University 
of Maryland.  Cristina Nochetto, now at FDA, not only taught me about QA/QC issues, 
but she is also an excellent and generous cook; I have fond memories of many nights at 
the Nochetto?s table with amazing food and better conversation.  For the last few years, 
Krystyna Bialek and I have shared an office and many conversations on the often 
neglected Russian writers and Polish princesses turned into Mexican journalists.  My 
work would have been immensely more difficult and not nearly as much fun without 
Isabelle Schmitz-Afonso?s help with the LC/MS.  Her husband, Carlos, helped us 
elucidate the structure of the APEs? fragment ions.  The long and tedious extractions 
were made much lighter by the help and company of some of our interns, such as 
Fr?d?rick Adam, Anika Lynch, and Natasha Andrade.  Natasha?s drive and enthusiasm 
are very contagious and helped me get through the last part of the PhD.  Special thanks to
Anubha Goel, fellow PhD student and great friend; we shared many of our frustrations 
v
and achievements and she was incredibly supportive in difficult times.  Two of the best 
friends I made in Maryland left several years ago, but have always managed to stay in 
touch: Roya Saffary, who is a real doctor and has always been  a source of inspiration, and 
Daniel Ch?vez who, in my mind, embodies the best of Mexico.
I want to thank my family, especially my parents, Georgina and Eduardo, who 
have always loved and supported me, even when things did not go as they would have 
wanted, and to my grandparents, Georgina e Hip?lito, who have always been role models 
to me.  I am also grateful to my brother Carlos, my sister-in-law Angeles, and their young 
family, and to the Carrillos? Betty, Enrique, Ethel and Luis?for all their love and 
support. My cousins, Elisa and Georgina, who are like sisters to me, have always been a 
great source of advice.  I greatly appreciate Georgina and Pedro?s hospitality in New 
York.
I would not be where I am without Ver?nica?s constant presence during the last 
15 years.  Not only have we shared the best and the worst times of our lives, but also an 
undying love and selfless friendship. I have always admired her courage, her laughter, 
and her passion for life.  Gracias, Vero, h?roe de los QAs, especialmente de aqu?l a quien 
salvaste la vida y no pudo ser lo que quisimos.
Finally, special thanks to my partner Matt, who saw me through the last stage of 
the PhD, putting up with the resulting grumpiness and the coast-to-coast commute; but, 
more than anything, porque juntos aprendimos que dos pueden ser en verdad uno.
vi
TABLE OF CONTENTS
LIST OF TABLES .........................................................................................................viii
LIST OF FIGURES .......................................................................................................... x
LIST OF FREQUENT ABBREVIATIONS ................................................................xiii
CHAPTER 1 ? INTRODUCTION.................................................................................. 1
1.2 Production and uses .................................................................................................. 3
1.3 Toxicity and endocrine disruption effects.................................................................5 
1.4 Concentrations in surface waters and effluents ...................................................... 11
1.5 Fate and distribution in the environment ................................................................ 13
1.6 Regulatory status..................................................................................................... 16
CHAPTER 2 ? ANALYTICAL METHODS.  SHORT-CHAIN ETHOXYLATES 17
2.1 Abstract................................................................................................................... 17
2.2 Introduction............................................................................................................. 18
2.3 Experimental section............................................................................................... 20
2.4 Results and discussion ............................................................................................ 28
2.5 Conclusions............................................................................................................. 43
CHAPTER 3 ? ANALYTICAL METHODS.  LONG-CHAIN ETHOXYLATES 
AND CARBOXYLATES ............................................................................................... 44
3.1 Abstract................................................................................................................... 44
3.2 Introduction............................................................................................................. 45
3.3 Experimental........................................................................................................... 47
vii
3.4 Results and discussion ............................................................................................ 61
3.5 Conclusions............................................................................................................. 77
CHAPTER 4 ? APEO FATE IN WASTEWATER TREATMENT PLANTS.......... 79
4.1 Abstract................................................................................................................... 79
4.2 Introduction............................................................................................................. 80
4.3 Experimental Section.............................................................................................. 82
4.4 Results and discussion ............................................................................................ 86
CHAPTER 5 ? APEO FATE IN BACK RIVER ....................................................... 109
5.1 Abstract................................................................................................................. 109
5.2 Introduction........................................................................................................... 110
5.3 Experimental Section............................................................................................ 113
5.4 Results and Discussion ......................................................................................... 117
CHAPTER 6 ? OTHER APPLICATIONS: NP MIGRATION FROM PLASTIC TO 
BOTTLED WATER ..................................................................................................... 144
6.1 Abstract................................................................................................................. 144
6.2 Introduction........................................................................................................... 145
6.3 Materials and methods .......................................................................................... 147
6.4 Results and discussion .......................................................................................... 151
CHAPTER 7 ? CONCLUDING REMARKS AND RESEARCH NEEDS.............. 160
Research needs............................................................................................................ 162
APPENDIX 1 ? BACK RIVER MODEL ................................................................... 164
APPENDIX 2 ? ANCILLARY AND RAW DATA.................................................... 170
REFERENCES.............................................................................................................. 189
viii
LIST OF TABLES
Table 1.1 Examples of the different CAS numbers and nomenclature used in the literature 
for APE compounds..................................................................................................... 4
Table 1.2 Acute and chronic toxicity values for selected APEs in fish and invertebrates 
compared to reported concentrations in fresh water and WWTP effluents in North 
America, Europe and Asia......................................................................................... 12
Table 2.1 Parent and fragment ions used for quantitation of NP, OP and their respective 
APEOs, and MS parameters used to produce them................................................... 25
Table 2.2 Possible structures for fragments of [NP5EO+NH
4
]
+
....................................... 38
Table 2.3 Concentrations of the APs and APEOs in two sampling sites (1 and 5) in Back 
River, MD.................................................................................................................. 40
Table 2.4 Water and sediment method performance parameters...................................... 42
Table 3.1 Ions and MS/MS parameters used for the quantitation of NPEOs, n=6-16, 
NP0EC, NP1EC and OP0EC. Parent ions correspond to [M+NH
4
]
+
 in ESI(+), and to 
[M-H]
?
 in ESI(-).  Fragment ions for NP6-7EO correspond to [C
6
H
5
-(OCH
2
CH
2
)
x
-
OH + H]
+
 with x = 6-7, to [M+H]
+
 for NP8-16EO, C
9
H
19
-C
6
H
5
-O
?
 for NP0-1EC and 
C
8
H
17
-C
6
H
5
-O
?
 for OP0EC........................................................................................ 55
Table 3.2 Calibration standards for the NP6-16EOs. ....................................................... 59
Table 3.3 Performance parameters for water and sediment methods. LOD: instrumental 
limit of detection; MDL: method detection limit; LOQ: limits of quantitation. ....... 64
Table 3.4 Fraction of the NP0-16EOs in water sorbed to filters. ..................................... 68
Table 4.1 Wastewater treatment plant characteristics and sampling events..................... 84
Table 4.2 Total average APEO and APEC concentrations (dissolved + particulate) in 
WWTP influents and effluents.  Values are arithmetic means of 5 samples, except for 
the summer influent, where n = 4; values in parentheses correspond to standard 
deviations................................................................................................................... 87
Table 4.3 APEO removal from wastewater in WWTPs. % removal = 100 -
([effluent]*100/[influent])). Concentrations units were ?M...................................... 93
Table 4.4 NP0-16EO and OP0-5EO concentrations in sludge from the WWTPs.......... 105
Table 5.1 Sampling sites location and water quality parameters for the second group of 
samples.  MR: Middle River (reference site); BR: Back River.  Data provided by 
John Martin (Back River WWTP) except for the suspended solids........................ 116
Table 5.2 Dissolved AP and APEO concentrations in Back River and Back River 
tributaries, Herring Run and Moores Run, in the 2001 sampling events.  Site 
locations: BR-A: 39?18.01?N, 76?29.07?W; BR-B: 39?17.37?N, 76?28.29?W; BR-C: 
39?16.31?N, 76?26.41?W; BR-D: 39?15.34?N, 76?26.44?W; BR-E: 39?14.49?N, 
76?26.10?W; BR-F: 39?14.36?N, 76?23.59?W........................................................ 118
Table 5.3 NP0-3EO, OP0-2EO, and NP0-1EC concentrations found in the Back River in 
this study compared to other sites in the United States. .......................................... 124
ix
Table 6.1 Concentrations of NP and OP found in spring water bottled in three different 
plastic types (HDPE: high-density polyethylene; PET: polyethylene terephthalate; 
PVC: polyvinyl chloride)......................................................................................... 153
Table A1 Ancillary data (suspended solids concentration, TSS; temperature, T; pH; and 
dissolved organic carbon, DOC) for the samples from the three wastewater treatment 
plants and the Chicago sewers described in Chapter 4............................................ 170
Table A2 Dissolved concentrations of NP0- 16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the 2004 sampling events in Back River WWTP. ................................................... 172
Table A3 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2004 sampling 
events in Back River WWTP................................................................................... 173
Table A4 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the 2004 sampling events in Blue Plains WWTP.................................................... 174
Table A5 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2004 sampling 
events in Blue Plains WWTP. ................................................................................. 175
Table A6 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the 2005 sampling events in Back River WWTP. ................................................... 176
Table A7 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005 sampling 
events in Back River WWTP................................................................................... 177
Table A8 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the 2005 sampling events in Blue Plains WWTP.................................................... 178
Table A9 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005 sampling 
events in Blue Plains WWTP. ................................................................................. 179
Table A10 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the 2005 sampling events in Calumet WWTP......................................................... 180
Table A11 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005 sampling 
events in Calumet WWTP. ...................................................................................... 181
Table A12 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
the Chicago sewage samples. .................................................................................. 182
Table A13 Particulate concentrations of NP0-16EO, and OP0-5EO in the Chicago 
sewage samples........................................................................................................ 183
Table A14 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in 
Back River. .............................................................................................................. 184
Table A15 Particulate concentrations of NP0-16EO, and OP0-5EO in Back River. ..... 185
Table A16 NP0-15EO and OP0-5EO concentrations in sediments from the Baltimore 
Harbor area (Fig. 3.7) .............................................................................................. 186
Table A17 NP0-16EO and OP0-5EO concentrations in sediments from Back River (Fig. 
5.3)........................................................................................................................... 187
Table A18 Mass balance of NPEO and NPEC in three WWTPs (See also Fig. 4.7). IN: 
total mass in raw influent (except for Calumet in August, where it corresponds to the 
primary effluent), OUT (s): total mass in waste primary and secondary sludge (Blue 
Plains includes tertiary sludge); OUT (l): total mass in final effluent; OUT (total) = 
OUT (s) + OUT (l); DEGRADED: estimated from IN-OUT(total)........................ 188
x
LIST OF FIGURES
Figure 1.1 Chemical structures and nomenclature for octyl- and nonylphenol and their 
respective ethoxylates and carboxylates...................................................................... 2
Figure 1.2 Relationship between acute toxicity (LC
50
) and EO chain length in bluegill 
sunfish, Lepomis macrochirus (adapted from Servos 1999, and Talmage 1994). ...... 7
Figure 2.1 MS/MS spectrum of NP?s [M-H]
?
 ion, m/z = 219.  The peak with m/z 93 can 
be attributed to the phenolate ion. Fragments with m/z = 119, 133, 147, 161, 175, 
189 and 203 correspond to fragments with the general formula 
?
O-C
6
H
4
-(CH
2
)
x
-
CH=CH
2
, where x = 0 to 6. ....................................................................................... 31
Figure 2.2 MS/MS spectrum of OP?s [M-H]
?
 ion, m/z = 205. The peak at around m/z 133 
is actually two ions, m/z 133 and 134, which are attributed to 
?
O-C
6
H
4
-C(CH
3
)=CH
2
and 
?
?O=C
6
H
4
=C(CH
3
)-CH
3
 respectively. ................................................................ 33
Figure 2.3 (a) MS/MS spectrum of OP1EO?s [M+NH
4
]
+
 ion, m/z = 268; (b) MS/MS 
spectrum of NP1EO?s [M+NH
4
]
+
 ion, m/z = 282...................................................... 35
Figure 2.4 MS/MS spectrum of NP5EO?s [M+NH
4
]
+
 ion, m/z = 458. Base peak 
corresponds to the ion [C
6
H
5
-(OCH
2
CH
2
)
5
-OH + H]
+
. Proposed structures for the 
fragments labeled as a, b, c, d, and e can be found in Table 2.2. .............................. 37
Figure 3.1 Map of the Baltimore Harbor region showing approximate locations for the 
sediment sampling sites and the Back River wastewater treatment plant.  BC: Bear 
Creek; BR: Back River; GF: Gwynn?s Falls; IH: Inner Harbor; WR: White Rock. . 51
Figure 3.2 Extracted MS-MS chromatograms of the NPEOs in the dissolved fraction of 
WWTP effluent extract (A and B correspond to tertiary effluent, whereas C to 
primary). (A) NP6-11EO, concentrations were 237, 179, 203, 127, 104, and 92 ng/L 
respectively; (B) NP12-16EO, 80, 53, 45, 5, and 15 ng/L; (C) NP0-1EC, OP0EC, 
1050, 1650 and 97 ng/L. ............................................................................................ 56
Figure 3.3 Recoveries of NP6-16EO from sediment and water. ...................................... 62
Figure 3.4 Nonylphenol ethoxylate concentrations in a Mid-Atlantic region WWTP in 
(A) the aqueous phase, and (B) particulate matter.  The numbers above the bars 
indicate the total concentration (?g/L in the aqueous phase and ?g/g in the 
particulate) of NP0-16EOs in each treatment stage................................................... 71
Figure 3.5 Octylphenol ethoxylate concentrations in a Mid-Atlantic region WWTP.  The 
numbers above the bars indicate the total concentration (?g/L) of OP0-5EOs in each 
treatment stage. .......................................................................................................... 73
Figure 3.6 Alkylphenol carboxylate concentrations in a Mid-Atlantic region WWTP 
(values obtained using external standard quantitation).  The numbers above the bars 
indicate the total concentration (?g/L) of NP0EC, NP1EC, and OP0EC in each 
treatment stage. .......................................................................................................... 75
Figure 3.7 Nonylphenol ethoxylate concentrations in sediments from the Baltimore 
Harbor area.  The different sites are identified in Figure 3.1.  The numbers above the 
bars indicate the total concentration (?g/g) of NP0-14EOs in each site.................... 76
xi
Figure 4.1 Total NPEO concentrations (dissolved + particulate) in raw wastewater (A) 
and WWTPs? final effluents (B) for all sampling events in winter and summer and 
for the three WWTPs, Back River (BR), Blue Plains (BP), and Calumet (Cal). ...... 88
Figure 4.2 Total NP0-16EO concentrations (dissolved + particulate) in final effluents 
from three WWTPs as a function of temperature; r = -0.868.................................... 90
Figure 4.3 Logged ratios of degradation compounds? concentrations (NP0-3EO + NP0-
1EC) to parents products? concentrations (NP4-16EO) for (A) Blue Plains, (B) Back 
River and (C) Calumet in the different treatment stages.  Influent refers to raw 
wastewater; prim eff, sec eff, tert eff, and fin eff refer to the primary, secondary, 
tertiary and final effluents respectively. .................................................................... 94
Figure 4.4 Changes in total NP0-16EO, NP0-3EO, NP4-16EO, and NP0-1EC 
concentrations in the different treatment stages in Blue Plains WWTP (February 
2005) .  Influent refers to raw wastewater; prim eff, sec eff, tert eff, and fin eff refer 
to the primary, secondary, tertiary and final effluents respectively. ......................... 97
Figure 4.5 Changes in total NP0-3EC concentrations in the different treatment stages in 
(A) Blue Plains and (B) Back River WWTP (February 2005).................................. 99
Figure 4.6 NP0-16EO (squares) and NP0-16EO + NP0-1EC (diamonds) removal from 
three WWTPs as a function of temperature............................................................. 103
Figure 4.7 NPE mass balance in three WWTPs. Influent: total mass in raw influent 
(except for Calumet in August, where it corresponds to the primary effluent), Sludge: 
total mass in waste primary and secondary sludge (Blue Plains includes tertiary 
sludge); Effluent: total mass in final effluent; Degraded: estimated from Influent-
(Sludge+Effluent). ................................................................................................... 106
Figure 4.8 Total NP0-16EO concentrations (dissolved + particulate) in (A) grab sewage 
samples from commercial, industrial and residential areas in Chicago, IL, and (B) 
24-h composite sewage samples from different residential areas in Chicago. ........ 108
Figure 5.1 Main biological transformation pathways for the APEOs: (1) non-oxidative, 
and (2) oxidative; x = 8 for OPEOs, and x = 9 for NPEOs.  Adapted from 
Montgomery-Brown and Reinhard 2003................................................................. 111
Figure 5.2 Back River, its main tributaries, and the location of the WWTP.  Numbers 1 to 
4 indicate sampling locations BR-1 to BR-4 described in Table 1, and MR is the 
background sampling site in Middle River.............................................................. 114
Figure 5.3 NP0-16EO concentrations in sediments from the Back River in (A) January 
and (B) July 2001. Site locations are in Table 5.2................................................... 119
Figure 5.4 Total NPE concentrations (dissolved + particulate; ?g/L) in the Back River 
WWTP effluent (EFF), Back (BR1-4) and Middle River (MR) water samples.  NP, 
NP1-3EO, and NP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of NP0-1EC are
compared to the total NP0-16EO for September and October 2004 in (B) and for 
March 2005 in (D). .................................................................................................. 120
Figure 5.5 Total NPE concentrations (dissolved + particulate; ?M) in the Back River 
WWTP effluent (EFF), Back (BR1-4) and Middle River (MR) water samples.  NP, 
NP1-3EO, and NP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of NP0-1EC are 
compared to the total NP0-16EO for September and October 2004 in (B) and for 
March 2005 in (D). .................................................................................................. 121
xii
Figure 5.6 NP0-16EO and OP0-5EO concentrations (dissolved + particulate; ?M) in the 
Back River WWTP effluent (EFF), Back (BR1-4) and Middle River (MR) water 
samples in September 2004 (A: NPEOs; B: OPEOs) and March 2005 (C: NPEOs; D: 
OPEOs).................................................................................................................... 123
Figure 5.7 Total OPE concentrations (dissolved + particulate; ?g/L) in the Back River 
WWTP effluent (EFF), Back (BR1-4) and Middle River (MR) water samples.  OP, 
OP1-3EO, and OP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of OP0-1EC are 
compared to the total OP0-16EO for September and October 2004 in (B) and for 
March 2005 in (D). .................................................................................................. 126
Figure 5.8 K
d
 values, L/g, as a function of ethoxylate-chain length for NP0- 16EO in sites 
BR-2 and BR-3, March 2005................................................................................... 128
Figure 5.9. Model versus measured results for sites BR-1, 1400 m, BR-2, 2500 m, BR-3, 
3800 m, and BR-4, 7600 m, in March 2005 for NP0-1EC (A) assuming k = 0; (B) k = 
0.015 d
-1
; and for NP0-16EO (C) assuming k = 0; (D) k  = 0.02 d
-1
. ....................... 133
Figure 5.10 Model versus measured results for sites BR-1, 1400 m, BR-2, 2500 m, BR-3, 
3800 m, and BR-4, 7600 m, in October 2005 for (A) NP0-1EC, and (B) NP0-16EO.
................................................................................................................................. 136
Figure 5.11 NP0-16EO concentrations in Back River particulate matter samples from 
October 2005. .......................................................................................................... 138
Figure 6.1 Migration of NP from two types of plastic bottles (HDPE and PVC) to water 
over time. Error bars represent standard deviation of 3 determinations.................. 156
Figure 6.2 Migration of NP from HDPE bottles to milk surrogate (10% ethanol solution) 
over time at two different temperatures (20 and 40?C). Error bars represent standard 
deviation of 3 determinations. ................................................................................. 158
Figure A1.  MatLab code to calculate the steady state concentrations of a chemical in the 
Back River. .............................................................................................................. 167
Figure A2.  Schematic representation of the Back River used to estimate the estuary?s 
depth at distance x from the head of the estuary...................................................... 169
xiii
LIST OF FREQUENT ABBREVIATIONS
APs alkylphenols
APEs alkylphenolic compounds (includes APs, APEOs, and APECs)
APECs alkylphenoxyethoxyacetic acids
APEOs alkylphenol ethoxylates
ASE accelerated solvent extraction
CAS Chemical Abstracts Service
DCM dichloromethane
DFW Dallas-Fort Worth International Airport
DI deionized carbon-free water
ED endocrine disruption
EO ethylene oxide, ethoxylate
ESI electrospray ionization
FDA Food and Drug Administration
GC/MS gas chromatography/mass spectrometry
HDPE high density polyethylene
HPLC high performance liquid chromatography
HPLC-F HPLC with fluorescence detection
HRT hydraulic retention time
IDMS isotope dilution mass spectrometry
K
ow
octanol-water partition coefficient
LC
50
lethal concentration for 50% of the population
LC/MS liquid chromatography/mass spectrometry
LC/MS/MS liquid chromatography/tandem mass spectrometry
LOD instrumental limit of detection
LOEC lowest observed effect concentration
LOQ limit of quantitation
MDL method detection limit
MQL method quantitation limit
NOEC no observable effect concentration
NP 4-nonylphenol
NPECs nonylphenoxyethoxyacetic acids
NPEOs nonylphenol ethoxylates
OP 4-tert-octylphenol
OPECs octylphenoxyethoxyacetic acids
OPEOs octylphenol ethoxylates
PET poly(ethylene terephthalate)
PVC poly(vinyl chloride)
QA/QC quality assurance/quality control
RSD relative standard deviation
SD standard deviation
xiv
SPE solid phase extraction
SRT sludge retention time
TNPP tris(nonylphenyl)phosphite
USEPA United States Environmental Protection Agency
WWTPs wastewater treatment plants
1
CHAPTER 1 ? INTRODUCTION
1.1 Nomenclature and chemical structures
Alkylphenol ethoxylates (APEOs) are known under different generic names, such 
as -(p-alkylphenyl)--hydroxypoly(oxyethylene), alkylphenoxy [poly(ethyleneoxy)] 
ethanol, alkylphenol polyglycol ethers, and polyoxyethylene alkylphenols (Porter 1994, 
Talmage 1994).  The APEOs are synthesized from alkylphenols (APs), i.e. phenols 
substituted with an alkyl chain of variable length. Substitution is normally directed to the 
para position because 4-alkylphenols are more readily polyethoxylated than the ortho-
isomers (Weinheimer and Varineau 1998). The most commonly used alkylphenol is 
nonylphenol (NP) (Fig. 1.1) followed by 4-tert-octylphenol (OP), whose alkyl chains 
have 9 and 8 carbons respectively.
In reality, NP is not a single compound, but rather a very complex mixture of 
isomers, because the nonene used to synthesize NP is composed by a highly branched 
family of isomers called propylene trimer (Wheeler et al. 1997).  In contrast, OP occurs 
as a single compound because it is synthesized from diisobutylene, a relatively pure 
material (Weinheimer and Varineau 1998).
APEOs are composed of an alkylphenol combined by an ether linkage to one or 
more ethylene oxide (EO) units.  Nonylphenol ethoxylates (NPEOs) are the most 
important APEOs (Fig. 1.1), followed by octyl- and dodecylphenol ethoxylates.  The 
number of EO units found in NPEOs ranges from 1 to 100, with 9-10 being the most 
2
H
2x+1
C
x
OH
Alkylphenols (APs)
nonylphenol (NP), x = 9 
octylphenol (OP), x = 8
H
2x+1
C
x
O
O H
n
Alkylphenol ethoxylates (APEOs), n = 1 - 16
nonylphenol ethoxylates (NPEOs), x = 9 
octylphenol ethoxylates (OPEOs), x = 8
H
2x+1
C
x
O
m
O
O
OH
Alkylphenoxyethoxyacetic acids (APECs), m = 0 - 15
nonylphenoxyethoxyacetic acids (NPECs), x = 9 
octylphenoxyethoxyacetic acids (OPECs), x = 8
Figure 1.1 Chemical structures and nomenclature for octyl- and nonylphenol and 
their respective ethoxylates and carboxylates.
3
commonly used in cleaning products (Talmage 1994).  Octylphenol ethoxylates (OPEOs) 
have a similar structure, but the alkyl chain has eight carbons instead of nine (Fig. 1.1).
The alkylphenoxyethoxyacetic acids (APECs) are APEO metabolites produced 
during their aerobic biotransformation, especially in wastewater treatment plants 
(WWTPs) (Ahel et al. 1994a).  Their structures are also depicted in Fig. 1.1.
Because commercially-available APEOs are mixtures of different homologues, 
NP is an isomeric mixture, and some of the APEO derivatives are biotransformation 
products, the nomenclature and classification of the alkylphenolic compounds (APEs) is 
not standardized and it is often confusing.  A good example of this is the different 
Chemical Abstracts Service (CAS) numbers assigned to these compounds, as shown in 
Table 1.1.  The product defined as CAS 84852-15-3 was agreed to be the most 
representative of the commercial NP formulations (Environment Canada 2001), and it is 
composed mainly by 4-NP, but it also contains smaller amounts of 2-NP and 2,4-
dinonylphenol (European Commission 2002).
1.2 Production and uses
NPEO production in the United States was almost 300,000 tons in 1994 
(Ferguson et al. 2001), which accounts for approximately 80% of total APEO use (APE 
Research Council 1999). NP production increased 2% per year from 1995 to 2000 and it 
was expected to keep growing at the same rate through 2004; demand also increased, and 
was calculated to be 110,000 tons in 2000 (?Nonylphenol? 2001). Only a small 
4
Table 1.1 Examples of the different CAS numbers and nomenclature used in the 
literature for APE compounds.
Compound CAS No. Ref. Name used/Comments
NP 104-40-5 1 4-nonylphenol
136-83-4 1 2-nonylphenol
25154-52-3 1,2,3 originally designated all NPs; currently n-NP only (ref. 3)
84852-15-3 1,2 representative of the commercially available mixture
90481-04-2 2 branched nonylphenol
1300-16-9 4 nonylphenol, mixed isomers
NP1EO 27986-36-3 5 2-(nonylphenoxyl)ethanol
104-35-8 6 4-nonylphenol ethoxylate; Ethanol, 2-(4-nonylphenoxy)-
NP2EO 20427-84-3 4,6 4-nonylphenol diethoxylate; Ethanol, 2-[2-(4-
nonylphenoxy)ethoxy]-
NP3EO 51437-95-7 6 4-nonylphenol triethoxylate; Ethanol, 2-[2-[2-(4-
nonylphenoxy)ethoxy]ethoxy]-
NP4EO 7311-27-5 3,6 4-nonylphenol tetraethoxylate; Ethanol, 2-[2-[2-[2-(4-
nonylphenoxy)ethoxy]ethoxy]ethoxy]-
9016-45-9 3 NP4EO
NPEO 9016-45-9 7 poly (oxy-1,2-ethanediyl), 
-(nonylphenyl)--hydroxy-
26027-38-3 7 poly (oxy-1,2-ethanediyl), 
-(4-nonylphenyl)--hydroxy-
37205-87-1 7 poly (oxy-1,2-ethanediyl), 
-(isononylphenyl)--hydroxy-
68412-54-4 7 poly (oxy-1,2-ethanediyl), 
-(nonylphenyl)--hydroxy-, 
branched
127087-87-0 7 poly (oxy-1,2-ethanediyl), 
-(4-nonylphenyl)--hydroxy-, 
branched
OP 62744-41-6 4 (1,1,3,3-tetramethylbutyl)phenol (mixed isomers)
27193-28-8 4 phenol, (1,1,3,3-tetramethylbutyl)-
1806-26-4 5 4-octylphenol
OPEO 9036-19-5 4 poly(oxy-1,2-ethanediyl), 
-[(1,1,3,3-
tetramethylbutyl)phenyl]--hydroxy-
NP0EC 3115-49-9 6 4-nonylphenoxy acetic acid; Acetic acid, (4-nonylphenoxy)-
NP1EC 106807-78-7 6 4-nonylphenoxy ethoxy acetic acid; Acetic acid, [2 -(4-
nonylphenoxy)ethoxy]-
NP2EC 108149-59-3 6 4-nonylphenoxy diethoxy acetic acid; Acetic acid, [2-[2-(4-
nonylphenoxy)ethoxy]ethoxy]-
NP3EC 184007-22-5 6 4-nonylphenoxy triethoxy acetic acid; Acetic acid, [2-[2-[2-
(4-nonylphenoxy)ethoxy]ethoxy]ethoxy]-
1 U.S. EPA 2003 5 ITC 2002
2 European Commission 2002 6 ITC 2002b
3 Environment Canada 2001 7 APERC 2004
4 CAS 2005
5
proportion of the NP is applied directly (US EPA 2003); most of it  is used for surfactant 
(80%) and phosphite antioxidant production (10%) (?Nonylphenol? 2001).
According to Talmage (1994), APEO uses in the United States are mainly 
industrial (55%) and institutional (30%); household and personal care uses account for 
only 15% of the total. The most important industrial uses are found in the production of 
plastics and elastomers, textile processing and manufacturing, agricultural chemicals 
formulation and the pulp and paper industries, where they have a wide array of functions, 
e.g. as emulsifiers, wetting agents, dispersants or cleaning agents. Institutional uses are 
mainly for cleaning purposes, such as commercial laundry products and hard surface 
cleaners. More than 80% of the household and personal care uses are in the form of 
heavy-duty powders, liquid laundry detergents and hard surface cleaners.
1.3 Toxicity and endocrine disruption effects
Nonionic surfactants are considered to be highly toxic to aquatic organisms 
(Nimrod and Benson 1996). Servos (1999) reviewed published data on the aquatic 
toxicity of APs and APEOs, and concluded that OP and NP are ?relatively toxic and have 
similar acute toxicity in fish, invertebrates and algae?. 96-h LC
50
 values ranged from 17 
to 3000 ?g/L in 22 different species of fish, although most values were between 100 and
300 ?g/L (Servos 1999).
The toxicity of APEOs and APs to aquatic organisms is inversely proportional to 
the hydrophobicity of the compound, i.e. toxicity increases with decreasing number of 
6
EO units and increasing alkyl chain length (Weinheimer and Varineau 1998).  This trend 
is illustrated in Fig. 2.1, where 96-h LC
50
 values in bluegill sunfish are presented as a 
function of the ethoxylate chain length for NPEOs and OPEOs.
Few studies have addressed the toxicity of the APECs, but their results suggest 
that the carboxylated metabolites are less toxic than the respective ethoxylates (Servos 
1999), probably due to their higher water solubility (Nimrod and Benson 1996).
Acute toxicity to mammals is considered to be low.  The FDA approved the use of 
NP9EO (nonoxynol-9) as a spermicide in over-the-counter contraceptive products in 
1980 following a safety study in humans.  Although early human epidemiological studies 
linked spermicide use with congenital malformations, later studies that evaluated 
statistical designs and biases suggested that the risks do not exceed those observed with 
other contraception methods (Nimrod and Benson 1996, Talmage 1994).  Due to its 
ubiquity, however, there is a high potential for human exposure to NP, as evidenced by 
its presence in urine.  Calafat and collaborators (2005) detected NP in 51% of 394 
samples of human urine from a reference population at a median concentration of <0.1 
?g/L.  Unfortunately, this study did not account for metabolites and used n-NP, which is 
barely present in commercial mixtures, as analytical standard.  These two facts combined 
prevent the study from reflecting accurately human exposure to branched NP isomers;
additionally, any possible toxicological effects of continuous long-term exposure to NP 
are unknown.
Interest in the toxicological properties of NP increas ed further after Soto and 
collaborators (Soto et al. 1991) discovered that it was able to induce both cell 
proliferation and the expression of progesterone receptor in human estrogen-sensitive 
7
1
10
100
1000
10000
100000
0246810 12
number of EO units
9
6
-
h
 
L
C
5
0
,
 
u
g
/
L
NP OP
Figure 1.2 Relationship between acute toxicity (LC
50
) and EO chain length in 
bluegill sunfish, Lepomis macrochirus (adapted from Servos 1999, and Talmage 
1994).
8
breast tumor cells, suggesting its role as an environmental xenoestrogen.  Since then, a 
number of studies have shown that APs and APEOs cause estrogenic responses in 
different in vivo and in vitro tests (Servos 1999).  Most studies revealed that the 
estrogenic potencies of APs and APEOs are only a fraction of that of 17-estradiol, e.g. 
relative potencies of NP and OP to induce in vitro vitellogenin synthesis were reported as 
9x10
-6
 and 3.7x10
-5
 respectively (Jobling and Sumpter 1993).  The same group also 
showed that OP had a higher capacity than NP to elicit in vivo vitellogenin synthesis in 
rainbow trout (Jobling et al. 1996), and suggested that NP2EO and NP0EC had similar 
potencies as NP.  More recent studies, however, seem to indicate that NP0EC is not as 
potent as suggested; Dussault et al. (2005) reported that NP0EC potency to elicit 
vitellogenin production in rainbow trout is only 3% that of NP, and Balch and Metcalfe 
(2006) calculated no observable effect concentrations (NOEC) for NP, NP1EO and 
NP0EC of 3, 35, and 2010 ?g/L respectively in Japanese medaka, using alteration to sex 
ratios and development of gonadal intersex as endpoints.
In order to evaluate the potential of the APEOs to cause endocrine disruption 
(ED) effects in real environment settings, such as WWTP effluent-impacted waters, 
where most of the estrogenic activity in WWTP effluents has been traced to steroid 
estrogens (Johnson and Sumpter 2001), additional factors have to be considered; e.g. 
duration of exposure?which might be long if not life-spanning in effluent-dominated 
streams, and possible additive effects with other estrogenic compounds.  Indeed, long-
term exposure (1 yr) tests by Ackermann et al. (2002) showed that exposure to NP in 
concentrations as low as 1 ug/L resulted in increased vitellogenin expression in rainbow 
trout, although no gonadal intersex or changes in sex ratios were observed.  As for 
9
combined effects with other substances, the possibility of additive effects was suggested 
by the work of Thorpe et al. (2001) with binary mixtures of NP and estradiol.  In a more 
extensive study, Brian et al. (2005) confirmed the potential of a mixture of NP, OP, 
estradiol, ethynylestradiol, and bisphenol A to act additively even at individual 
concentrations below those needed to elicit vitellogenin induction in fathead minnows 
(Pimephales promelas) by the individual chemicals.  The presence of antiestrogenic 
compounds might also affect the ED potential of the APEOs.  In fact, Hu et al. (2002) 
reported that chlorinated derivatives of NP resulting from the reaction of NP with 
hypochlorite posses this kind of properties, at least in vitro.
The exact mechanism (or mechanisms) of action by which APs and APEOs cause 
ED is unknown.  Typically, direct interaction of the xenoestrogen with the estrogen 
receptor occurs, but other interactions along the reproductive endocrine axis can also lead 
to ED (Nimrod and Benson 1996).  Some studies have presented evidence that NP and 
OP bind to the estrogen receptor, although with less affinity than estradiol (Nimrod and 
Benson 1996, Odum et al. 2001).  Other studies suggest that the effects of NP on fathead 
minnows seem to result not from a direct interaction with the receptor but rather from 
changes in the endogenous concentrations of 17-estradiol through an indirect activation 
mechanism (Giesy et al. 2000). This mechanism could be the aromatase-mediated 
transformation of androgens to estrogens, as suggested by NP?s dose-dependent 
enhancement of the expression of one of the two aromatase genes in zebrafish (Danio 
rerio) (Kazeto et al. 2004).  Additionally, other reports have suggested that toxic effects 
other than ED might lead to reproductive failure.  Schwaiger et al. (2000) proposed that 
10
chronic exposure to NP at environmentally-relevant concentrations might cause severe 
anemia in common carp (Cyprinus carpio) and result in reproductive impairment.
Establishing a threshold value for the onset of ED in aquatic populations is a 
difficult task because of the number of species involved and the different biological 
endpoints available.  Early studies reported the threshold concentration of NP for 
vitellogenin induction as 20-50 ?g/L, but the value for the 91-d lowest-observed-effect 
concentration (LOEC) for NP in juvenile rainbow trout was determined to be 10 ?g/L 
(Nimrod and Benson 1996). The same concentration (10 ?g/L) was reported by Jobling 
and collaborators (1996) as the threshold for vitellogenin induction in rainbow trout.
Staples et al. (2004) reviewed the chronic toxicity data for NP, NPEO, and NPEC to 
propose thresholds that would protect aquatic ecosystems based in current knowledge of 
their adverse effects. They calculated chronic values, the geometric mean of NOEC and 
LOEC, for all the compounds and biological endpoints available in different species, 
which ranged from 900 to 14100 ?g/L for NPEOs with more than 9 EO units; 3200 to 
12000 ?g/L for NP0EC, 11 to 500 ?g/L for NP1-2EO, and 2.3 to 3500 ?g/L for NP.
These values agree with previous observations discussed above that toxicity increases 
with decreasing EO chain length.  As for sensitivity differences between organisms, algae 
seemed to be relatively insensitive to NP and NPEO exposure; whereas fish and 
invertebrates showed similar ranges at least for NP (few data are available for NPEO 
effects on invertebrates).  They proposed the use of a species sensitivities distribution 
(SSD) approach, which calculates a concentration protective of most single species based 
on NOEC and LOEC values, to set the allowed concentration thresholds.  Using this 
approach for NP, 10th and 5th percentile chronic effect values of 5.7 and 3.9 ?g/L 
11
respectively were calculated, which are comparable to the U.S. Environmental Protection 
Agency?s (USEPA) proposed water quality criteria of 5.9 ?g/L for chronic exposure in 
freshwater (U. S. EPA 2003).
1.4 Concentrations in surface waters and effluents
Although interest in the APEOs is not recent (e.g. Giger et al. 1984), Soto?s 
group?s discovery (1991) of the possible role of NP as a xenoestrogen prompted 
numerous researchers to measure the concentrations of the APEs in different 
environmental compartments to assess the risk of these compounds to aquatic biota.
These values have been compiled in several reports, such as Bennie 1999, Ying et al. 
2002, and Montgomery-Brown and Reinhard 2003. Most of the available data are for NP 
or NPEOs, whereas information on the APECs, OP, and the OPEOs is relatively scarce.
With few exceptions, such as a site in Canada (Bennie 1999) with NP 
concentrations of 2600 ?g/L, NP and OP concentrations in North American fresh waters
were below the reported acute toxicity (LC
50
) values or close to the lower extreme, as 
exemplified in Table 1.2.  This observation was supported by a national reconnaissance 
of organic pollutants concentrations in American waters (Kolpin et al. 2002), in which 
median values for NP, NP1EO, and NP2EO were 0.8, 1, and 1 ?g/L respectively. In 
contrast, wastewater effluents, either from WWTPs or industrial facilities, can contain 
12
Table 1.2 Acute and chronic toxicity values for selected APEs in fish and 
invertebrates compared to reported concentrations in fresh water and WWTP 
effluents in North America, Europe and Asia.
concentration, ?g/L
Ref NP NP1-2EO NPnEO (n>9) NP0EC OP
acute value
a
fish 1 17 - 3000 3000
b
300 - 110000 290 ? 450
c
invertebrates 1 20 - 3000 150 ? 1000
b
2900 ? 44200
b
9400-17000 90 ? 270
b
chronic value 2 2.3 - 3500 11 - 500 900 - 14000 3200-12000 3
d
surface waters
US/Can 3,4 <0.01 - 2 ND - 10 <1.6 - 15 ND-2 ND - 3
Europe 3 ND - 644 ND - 32 <1 - 7.1 <1-45
Asia 4 ND - 10 0.04 - 0.81 ND - 0.18
WWTP effluent
US/Can 3,4 <0.02 - 40 0.8 - 5.5 3.9 - 102 1.9-703 <0.005 - 200
Europe 3,4 <0.2 - 340 30 - 77 980 - 1067 ND-224 ND - 1
Asia 4 0.08 - 1.2 0.21 - 3.0 0.02 - 0.48
a
96-h LC
50
b
48-h LC
50
c
24-h LC
50
d
Threshold for vitellogenin induction in rainbow trout (Jobling et al 1996)
References
1 Servos 1999
2 Staples et al. 2004; values include all species tested
3 Bennie 1999
4 Ying et al. 2002
13
much higher concentrations of the APEs; e.g. 6,300 ?g/L of NP in a shipyard oil/water 
separator (Hale et al. 2000), and 200 ?g/L of OP in a WWTP effluent from Philadelphia 
(Sheldon and Hites 1979), and might represent an acute toxicity risk in places where 
there is no further treatment or where low dilution occurs when the effluents are 
discharged directly into water bodies.
In Europe, however, higher concentrations of the APs and APEOs were routinely 
found?although this situation might change due to restriction in APEO use (see section 
1.6 below), and they are usually explained by a greater removal efficiency of the 
American WWTPs (Renner 1997), and dilution effects in American rivers, that are larger 
than those in Europe (Field and Reed 1996).
NP0EC and long-chain NPEO concentrations in surface waters also tended to be 
lower than the threshold for chronic effects (Table 1.2).  Most sites also showed 
concentrations of NP, NP1-2EO, and OP below the thresholds, although some sites were 
clearly above these values (Table 1.2).  This was especially true in Europe, where high 
levels of the NPEs have been found in both Switzerland and Spain.  The median values 
from the American national reconnaissance mentioned above were also below these
chronic effect thresholds.
1.5 Fate and distribution in the environment
It is widely accepted that long-chain APEOs are the least persistent of the group 
because the polyethoxylate chain is readily and rapidly lost to biotransformation under 
14
aerobic conditions (DiCorcia et al. 1998).  Originally, this transformation was assumed to 
occur via ether cleavage, resulting in the formation of AP1-2EOs (Ahel et al. 1994a), 
which could in turn be oxidized to form the respective APECs.  However, more recent 
studies have suggested the immediate oxidation of the long-chain APEOs to long-chain 
APECs, followed by shortening of the EO chain and the simultaneous carboxylation of 
the alkyl group resulting in compounds with carboxylated EO and alkyl chains of diverse 
lengths (Jonkers et al. 2001).  It has not been established with certainty whether the first 
pathway actually occurs or it was mistakenly defined as a consequence of the late 
discovery of the long-chain APECs.  It could also be the case that each pathway occurs in
different environments, e.g. during WWTP treatment versus surface waters.  The 
biotransformation, and ultimate biodegradation, mechanisms of the resulting compound 
from both pathways above is poorly understood (Montgomery-Brown and Reinhard 
2003), but the carboxylated compounds are generally regarded as more persistent (Ahel 
et al. 1994b, Jonkers et al. 2001). 
Besides biotransformation, another important removal mechanism for the APEOs, 
and especially for the APs, is partition to solids, and it has been reported to account for a 
high percentage of the NP removal in WWTPs (Ahel et al. 1994a). The APEOs are 
relatively hydrophobic; log K
ow
 values for NP, NP1EO, NP2EO, NP3EO and OP were 
determined experimentally to be 4.48, 4.17, 4.21, 4.20 and 4.12 respectively, and the 
values for NP4-10EO were estimated to be between 4.3 and 4.1 (Ahel and Giger 1993).
Although these log K
ow
 values suggest the sorption of the APEs to organic matter , 
interaction with mineral surfaces can also occur; therefore, sorption to sediments is the 
15
result of the combination of at least two different types of interaction, hydrophobic with 
organic matter, and hydrophilic with the mineral fraction (John et al. 2000).
NP is also suspected to be lost from the water by volatilization. This was first 
suggested by Ahel and collaborators (Ahel et al. 1994a) and latter confirmed by field 
observations of Dachs et al. (1999), who estimated the Henry?s Law constants for the 
different NP isomers between 3 and 4 x 10
-5
 atm m
3
 mol
-1
, high enough to make 
volatilization a significant removal process in all waters (Lyman et al. 1990).  This group 
(Van Ry et al. 2000) also estimated that 40% of NP removal from the water column of 
the lower Hudson River Estuary was due to volatilization .
Photolysis has also been suggested as an elimination pathway for the APEs, and it 
has been shown to occur to OP9EO (Tanaka et al. 1991); products identified included 
polyethylene glycols of all possible chain lengths, glycolic aldehydes, glycolic ethers, and 
OP1-8EO.  Ahel and collaborators (Ahel et al. 1994c) estimated a first-order rate constant 
of sunlight photolysis (k
p
) for NP of 0.09 h
-1
, corresponding to a half-life of 10-15 hours 
under continuous summer sunlight (0.700 kW/m
2
) in the surface layer of natural water. 
They also reported that the photolysis rate was approximately 1.5 times slower at depths 
of 20 to 25 cm than at the surface, and that photochemical degradation of NPEOs 
occurred at significantly slower rates, whereas OP had a similar behavior to that of NP (k
p
= 0.10 h
-1
).
16
1.6 Regulatory status
The toxicological properties of the APEs and their potential as endocrine 
disrupters have prompted government bodies around the world to introduce regulations
restricting their use or release into the environment.  In the United States, the EPA 
published draft water quality criteria for NP of 27.9 and 5.9 ?g/L for acute and chronic
exposure in freshwater, and 6.7 and 1.4 ?g/L for acute and chronic exposure in saltwater 
(U. S. EPA 2003).  Canada also issued guidelines for freshwater, 1.0 ?g/L of NP toxic 
equivalents or TEQs (a TEQ is defined as the concentrations of all NPEs present 
expressed as the NP concentration having an equivalent toxicological potency), and sea 
water, 0.7 ?g/L of NP TEQs (Environment Canada 2004).  Additionally, the Canadian 
government introduced regulation requiring a 50% reduction in NPE use by 2007, and 
95% by 2010 (both based on 1998 values) for soap and cleaning product manufacturers 
and importers, textile and paper processors, and anyone acquiring 2000 kg or more of NP 
and NPEs (Department of the Environment 2004).  The European Union took a more 
conservative approach and banned the use of NPEs in concentrations equal or higher than 
0.1% by mass in products ranging from industrial and domestic cleaning products, to 
personal care products and pesticide formulations, and in industrial processes such as 
textile, leather and paper manufacturing (?Directive? 2003).
17
CHAPTER 2 ? ANALYTICAL METHODS.  SHORT-CHAIN ETHOXYLATES
This chapter has been published as:
Loyo-Rosales, J. E., Schmitz-Afonso, I., Rice, C. P., and Torrents, A. (2003), ?Analysis 
of octyl- and nonylphenol and their ethoxylates in water and sediments by liquid 
chromatography/tandem mass spectrometry.? Anal. Chem. 75(18), 4811-4817.
Reproduced with permission.  Copyright 2003, American Chemical Society.
2.1 Abstract
The ubiquitous presence of alkylphenol ethoxylates in the environment as well as 
concern for endocrine disruption effects in biota caused by their degradation products 
(such as octyl- and nonylphenol) have raised interest in the environmental fate of these 
compounds.  As part of an effort to model their behavior in a sub-estuary of the 
Chesapeake Bay, a quantitative method for the analysis of octyl- and nonylphenol, and 
their ethoxylates (1 to 5) in water and sediment was developed.  Extraction procedures 
are based on solid-phase extraction techniques.  Identification and quantitation of the 
analytes is done by liquid chromatography coupled to tandem mass spectrometry.  
Instrument detection limits for the compounds ranged from 0.1 to 9 pg injected on 
column, which allowed method detection limits of 0.04 to 3 ng/L in water and 0.2 to 13 
ng/g-dry weight in sediment.  The method was used to analyze water and sediment from 
the Back River, MD, where concentrations for the individual compounds ranged from < 8 
18
to 200 ng/L in water, and < 9 to 6700 ng/g-dry wt. in sediment.  Additionally, structural 
information obtained in the mass spectrometer is presented that supports previous 
observations that nonylphenol and its ethoxylates are composed mainly by isomers with a 
tertiary alpha carbon.
2.2 Introduction
Alkylphenol
*
 ethoxylates (APEOs) have been widely used as part of industrial 
processes and as detergents in both industrial and household applications for more than 
30 years (APERC 1999).  These compounds were the major nonionic surfactants in use 
during the mid-1970s, when low biodegradability concerns led the surfactant industry to 
replace them with alcohol ethoxylates in household products (Vivian 1986).  More bans 
and restrictions followed in Europe in the mid-1980s, when it was found that the products 
of APEO degradation (such as 4-nonylphenol and 4-tert-octylphenol, NP & OP 
respectively) were more toxic than the parent compounds (Renner 1997).  In the last 
decade, the discovery of the estrogen-like activity of NP raised concern over their role in 
endocrine disruption effects observed in aquatic biota.  NP enters the aquatic 
environment mainly as a by-product of nonylphenol ethoxylates (NPEOs) degradation in 
wastewater treatment plants (WWTP).
*
 Although the term ?alkyl? includes all possible classes of alkyl groups occurring in alkylphenol 
ethoxylates and derivatives (e.g. heptyl-, nonyl-, dodecylphenol ethoxylates, etc.), it will refer exclusively 
to the octyl- and nonyl- groups in this work.
19
These toxicity concerns and the possible environmental persistence of the APEOs 
and their metabolic products prompted the development of analytical methodologies to 
determine the presence and concentration of these compounds in different environmental 
matrices (Lee 1999).  Extraction methods for these compounds range from liquid-liquid 
extraction with organic solvents to supercritical fluid extraction and were reviewed 
recently (Lee 1999).  The analysis of the APEOs was typically done by high-performance 
liquid chromatography with fluorescence detection (HPLC-F) (Datta et al. 2002, 
Marcomini et al. 2000) or gas chromatography/mass spectrometry (GC/MS) (Bennett and 
Metcalfe 2000).  But as the advantages of liquid chromatography/mass spectrometry 
(LC/MS) were recognized, several groups developed various LC/MS methods to analyze 
APEOs and their derivatives in different environmental matrices (Ferguson et al. 2001, 
Ferguson et al. 2000, Takino et al. 2000, Di Corcia et al. 2000, Shang et al. 1999a, 
Scullion et al. 1996, Crescenzi et al. 1995).  Although LC/MS methods have resulted in 
lower detection limits for the APEOs, tandem mass spectrometry (MS/MS) techniques 
reportedly improve the identification of complex mixtures such as these compounds
(Petrovic and Barcel? 2001).
The main objectives of this work were to develop an LC/MS/MS method to 
quantify NP, OP, and their AP1-5EOs in water and sediment, and to study the behavior of 
the APs (OP and NP) and APEOs in the MS/MS system to obtain structural information 
on the different NP isomers, specifically the predominance of isomers with tertiary alpha 
carbons in the alkyl chain (Wheeler et al. 1997).  In contrast to previous work that relied 
on mixtures of the APEOs, purified standards of these compounds were used to develop 
and validate the extraction and quantitation methods.  The water extraction method that is 
20
presented allowed the concentration of up to four liters of sample, which is an important 
factor in achieving low detection limits for these compounds, necessary to detect them at 
the concentrations present in the environment.  A similar LC/MS/MS method to analyze 
surface water was recently published by Houde et al. (2002).  Although this method 
allowed the detection of the carboxylated derivatives of the APEOs with the shortest 
chains, it did not include the APs, which are considered the most toxic metabolites of the 
APEOs (Servos 1999).
2.3 Experimental section
Standards and reagents.  NP was obtained from Schenectady International, 
Schenectady, NY (purity 
 95%, CAS 84852-15-3).  OP came from Aldrich, Milwaukee, 
WI (97% purity, CAS 140-66-9).  NP2EO was obtained as an R&D product from 
Aldrich. NP1EO, NP3EO, NP4EO, NP5EO and the OP1-5EOs were purified in the 
laboratory by flash chromatography on silica gel from commercial mixtures as described 
elsewhere (Datta et al. 2002); NP1EO was isolated from Surfonic N-10 (Huntsman 
Chemicals, Austin, TX); NP3EO, NP4EO and NP5EO were isolated from POE(4) 
nonylphenol (Chem Service, West Chester, PA).  The OP1-5EOs were purified from 
POE(3) and POE(5) tert-octylphenol, which were also from Chem Service.  Purity of the 
standards was assessed by high performance liquid chromatography with fluorescence 
detection (HPLC-F); identity of the compounds was confirmed by liquid chromatography 
coupled to tandem mass spectrometry (LC/MS/MS).  Purity of the standards was 
21
calculated to be above 99% in all cases, except for the octylphenol monoethoxylate, 
OP1EO (94%).  A mixture of 
13
C
6
-NP (Cambridge Isotope Laboratories Inc., Andover, 
MA) and 
13
C
6
-NP(1.6)EO synthesized by P. L. Ferguson (Ferguson et al. 2001) was used 
as internal standard for quantitation.  Organic solvents such as dichloromethane (DCM), 
hexane, and methanol were high purity, pesticide grade from Burdick & Jackson 
(Honeywell International Inc., Muskegon, MI).  Deionized (18.2 megohm-cm), carbon-
free water (DI water) was obtained in the laboratory using a NANOpure water 
purification system (Barnstead International, Dubuque, IA).  Anhydrous Na
2
SO
4
, 
granular powder, was purchased from Mallinckrodt Baker Inc. (Paris, KY); and the 
ammonium acetate was acquired from Aldrich (Milwaukee, WI), 99.99+% purity.
Sample collection. Surface water was taken with an 1100 gph (gallons per hour) 
automatic bilge pump (Rule Industries, Gloucester, MA) through Tygon tubing into 1-
gallon amber glass bottles (previously rinsed with acetone and baked at 400?C for 4 
hours).  The pump was operated for at least 2 minutes before filling the first bottle at each 
site.  The bottles were then rinsed 3 times with water from the pump, filled and stored in 
coolers with ice.  After returning to the laboratory, water samples were immediately 
vacuum-filtered through a 1 ?m Multigrade GMF 150 graded density glass microfibre 
filter in series with a 0.7 ?m GF/F glass microfibre filter (both from Whatman Inc., 
Clifton, NJ).  Both filters had been previously baked at 400?C for 4 hours. Filtered water 
was stored in pre-baked glass bottles at 4?C until extraction, which occurred no more 
than 24 hours after filtration. Sediments were collected with a ponar dredge.  Four to six 
grabs of sediment were taken at each site and the top 2 cm of each were removed with a 
stainless steel spoon into a stainless steel container.  The sediments were then 
22
homogenized, poured into 250 mL jars and stored on ice.  Upon returning to the 
laboratory, sediments were stored at -20?C until extraction.
Extraction.  For the extraction of the APEOs from water, we investigated the 
performance of a hyper cross-linked hydroxylated polystyrene-divinylbenzene copolymer 
(Isolute ENV+, 500 mg, 6 mL cartridges from International Sorbent Technology Ltd., 
Hengoed, UK) against octadecylsilica (Supelclean LC-18, 500 mg, 3 mL cartridges from 
Supelco, Bellefonte, PA).  Both solid phases showed similar recoveries, but ENV+ was 
preferred because extraction is faster and larger volumes of water can be extracted, which 
is especially necessary for the quantitation of OP and the octylphenol ethoxylates 
(OPEOs).  The extraction procedure is as follows: the ENV+ cartridges were pre-rinsed 
sequentially with 18 mL DCM, 12 mL acetone and 12 mL organic-free deionized water 
in a vacuum manifold; then, approximately 4 L of sample were passed through each 
cartridge.  The cartridges were then dried with air and eluted sequentially with 12 mL 
DCM, 12 mL methanol and 12 mL acetone.  The collected solvents were evaporated 
under a gentle nitrogen flow and exchanged for methanol to a final volume of 0.5 mL, 
after which 0.5 mL of water was added.  The extracts were then filtered through an 
Acrodisc LC 13-mm syringe filter containing a 0.2 ?m PVDF membrane (Pall Gelman 
Laboratory, Ann Arbor, MI).  Both syringe and filter were then rinsed with 0.5 mL of a 
50:50 v/v methanol:water mixture that was added to the extract to attain a final volume of 
1.5 mL.
Extraction of the analytes from sediment was performed by accelerated solvent 
extraction (ASE), using modifications of the ASE extraction and SPE cleanup methods 
developed by Shang et al (1999a and 1999b).  Before extraction, sediment samples were 
23
thawed, homogenized and then suction dried on a b?chner funnel to remove most of the 
water.  A fraction of each sample was dried in a desiccator under vacuum until constant 
weight was achieved.  Approximately 1 g of dry sediment was ground in a mortar with 50 
g of previously baked Na
2
SO
4
.  This mixture was then packed into 33-mL stainless steel 
cells and extracted in an accelerated solvent extractor (ASE 200 from Dionex Corp., 
Sunnyvale, CA) using a 50:50 v/v acetone:hexane mixture as extraction solvent.  The 
ASE was programmed to operate at 1500 psi and 100?C during 3 cycles of heating (5 
min), static extraction (10 min) and purging (200 sec).  The extracts were exchanged to 
hexane under a gentle nitrogen flow and then cleaned up using the multi-layer modified 
SPE cartridge described by Shang et al. (1999a).  Briefly a 10-mL, 500-mg, BondElut 
aminopropylsilica cartridge (Varian Associates Inc., Harbor City, CA) was stacked with 1 
g of HCl-activated copper and 3 g of previously baked Na
2
SO
4
.  The stacked cartridge 
was conditioned with 30 mL DCM and 15 mL hexane.  The samples were loaded 
dropwise, rinsed with 2 x 5 mL washes of hexane and eluted with 2 x 6 mL acetone.  
Eluates from the cartridges were exchanged to 0.5 mL methanol and then filtered as 
described for the water extracts.
LC/MS/MS analysis.  After cleanup, both water and sediment extracts were 
analyzed by LC/MS/MS.  Chromatographic separation was based on a method by 
Ferguson et al. (2001), which relied on a mixed-mode (size- exclusion and reversed-phase 
adsorption) column. The instrument was a Waters 2690 XE separations module (Waters 
Corp., Milford, MA) with an MSpak GF-310 4D column, 4.6 x 150 mm (Shodex, Shoko 
Co., Tokyo, Japan) at 60?C; injection volume was 10 ?L.  A mobile phase gradient was 
necessary to separate the compounds: solvent A was 10 mM ammonium acetate in a 
24
50:50 v/v methanol:water solution; solvent B was pure methanol.  Initial conditions were 
100% A; a 20 min gradient was started immediately after injection until 90% B was 
reached; these conditions were maintained for 8 minutes and then the percentage of B 
was gradually increased over 2 min to 100%.  Finally, the column was stabilized for 20 
min with 100% A; total run time was 60 min.  Flow rate was set at 0.2 mL/min and all of 
the eluent was allowed into the MS.  Atmospheric pressure ionization-tandem mass 
spectrometry analysis was performed on a benchtop triple quadrupole mass spectrometer 
(Quattro LC from Micromass Ltd., Manchester, UK) with an electrospray ionization 
source.  The source parameters were as follows: capillary voltage was set at 3.5 kV in the 
electrospray positive (ES+) and ?2.9 kV in the electrospray negative mode (ES-); 
extractor voltage was set at 3 and 2 V respectively; RF lens at 0.1 V in both modes; 
source and desolvation temperatures were 140 and 400?C.  Liquid nitrogen was used to 
supply the nebuliser and desolvation gas (flow rates were approximately 80 and 600 L/hr 
respectively).  Argon was used as collision-induced decomposition (CID) gas to fragment 
the parent ions; typical pressure was 2.6 x 10
-3
 mbar.  Both high- and low-mass resolution 
were set at 12.0 for both quadrupoles.  Acquisition was done in the multiple-reaction 
monitoring mode (MRM) in electrospray positive (ES+) for the first 25 minutes of the 
run and then switched to electrospray negative (ES-) for 10 min.  MRM was chosen 
because it allows high sensitivity and selectivity by setting both quadrupoles to transmit 
selected ions only, the first quadrupole selects the parent ions that are allowed into the 
collision cell, and the second one transmits only specific fragment ions.  The parent and 
fragment ions used for compound identification and quantitation are listed in Table 2.1
25
Table 2.1 Parent and fragment ions used for quantitation of NP, OP and their 
respective APEOs, and MS parameters used to produce them.
compound
parent ion, 
Da
fragment ion, Da
retention time, 
min
cone, V collision, eV ion mode
NP 219 133 26.4 -40 30 ES -
NP1EO 282 127 24.1 15 10 ES +
NP2EO 326 183 22.7 20 12 ES +
NP3EO 370 227 21.4 20 14 ES +
NP4EO 414 271 20.3 20 15 ES +
NP5EO 458 315 19.3 20 18 ES +
OP 205 133 25.9 -45 20 ES -
OP1EO 268 113 23.5 15 10 ES +
OP2EO 312 183 21.9 20 12 ES +
OP3EO 356 227 20.6 20 13 ES +
OP4EO 400 271 19.5 20 15 ES +
OP5EO 444 315 18.3 25 18 ES +
13
C-NP 225 139 26.4 -40 30 ES -
13
C-NP1EO 288 127 24.2 15 9 ES +
13
C-NP2EO 332 189 22.6 20 10 ES +
13
C-NP3EO 376 233 21.4 20 12 ES +
13
C-NP4EO 420 277 20.3 20 15 ES +
26
along with the optimum cone voltages and collision energies used.  Optimization was 
performed by infusion of the standards from a syringe pump (10 ?L/min) mixed with the 
LC effluent (100% A; 200 ?L/min), setting high- and low-mass resolution at 15.0 for 
both quadrupoles.  Detector was a photomultiplier set at 650 V.  Analyte concentrations 
were calculated by the internal standard method using 
13
C
6
-NP0-4EOs as internal 
standards.  Calibration curves were prepared in the mobile phase solvents and the analyte 
concentrations ranged from 20 to 700 ng/mL for OP, NP, OP1EO and NP1EO, and from 
6 to 200 ng/mL for the rest of the APEOs (2 to 5).  Peak integration and quantitation were 
performed automatically using the MassLynx 3.5 software (Micromass Ltd, Manchester, 
UK).
Analytes stability.  The stability of the APs and AP1-5EOs was evaluated in 
different matrices.  To test the stability of these compounds in the calibration curve 
solvent (50:50 v/v methanol:water), 10 mL of the 4
th
 calibration point were prepared and 
stored at ?20?C; analyte concentrations were: NP, 205; NP1EO, 214; NP2EO, 64; 
NP3EO, 64; NP4EO, 63; NP5EO, 67; OP, 210; OP1EO, 208; OP2EO, 65; OP3EO, 67; 
OP4EO, 69; and OP5EO, 64 ng/mL.  This solution was analyzed by LC/MS/MS on days 
1, 8 (1 week), 15 (2 weeks), 29 (4 weeks), 43 (6 weeks), and 85 (12 weeks).  The stability 
of these compounds adsorbed onto the ENV+ cartridges after extraction from water was 
tested as follows: seven pre-baked 4-L amber glass bottles were spiked with the analytes, 
filled to the top with DI water and extracted as detailed above.  The dry cartridges were 
stored at ?20?C until analysis on days 1, 5, 8, 29, 43 and 85, when they were extracted 
and analyzed as described before.  Stability of the analytes was also evaluated in DI water 
and compared to stability in water from the Little Paint Branch stream that runs through 
27
the USDA campus.  To carry this out, two sets of pre-baked 4-L amber glass bottles were 
prepared, one was filled with DI water and the other with water from the stream; both 
sets were then spiked with the analytes and stored at 4?C until analysis on days 1, 3, 5, 8, 
15 and 29.  The water was extracted as described above, except that these samples were 
not filtered.
Quality assurance and quality control (QA/QC).  A number of performance 
parameters were evaluated for each analyte in both water and sediment.  Instrument 
detection limits (LOD) were calculated as the amount of analyte giving a peak with a 
signal-to-noise ratio of 3.  Method detection limits (MDL) were calculated from the 
LODs, normalizing to sample amount and the final volume of extract analyzed.  Method 
quantitation limits (MQL) were set at least five times above the MDL values.  
Repeatability of the injections was evaluated by injecting a standard solution 11 
consecutive times from the same vial.  Variations in response ranged from 1 to 8% (RSD) 
for all analytes, whereas the variation in their retention times was  0.1% (RSD).
With every batch of 10 to 12 water or sediment samples, procedural blanks, a 
matrix spike and at least one replicate were included.  Calibration curves including six 
points were analyzed with each set of extracts.  These curves were injected at least twice, 
before and after every 5 to 8 samples.  In order to be considered acceptable, r
2
 values 
from the linear regression were expected to be greater than 0.995 and the residual values 
for each calibration point below 10%.  Any matrix effects on analyte quantitation were 
accounted for by the use of adequate internal standards, as described by Ferguson et al.
(2000).
28
2.4 Results and discussion
Analytes stability.  The concentrations of the APs and AP1-5EOs in solvent 
(50:50 v/v methanol:water) remained constant over the 12-week period, as suggested by 
the low RSD values (4 to 13%) obtained for all compounds, which were close to the RSD 
for injection repeatability, 1 to 8% (discussed in the experimental section).
The concentrations of most of the compounds stayed constant in the ENV+ 
cartridges (RSD = 4 to 19%), except for the AP4-5EOs, whose concentrations decreased 
noticeably after 6 weeks of storage (from 33 to 85%).  It is not clear whether this
decrease was due to degradation of these compounds or to a stronger binding to the solid-
phase resulting in lower recoveries.  But if degradation was indeed occurring, it was not 
by the loss of ethoxylate units from the chain, because no accumulation of the lower 
ethoxymers was observed.  Petrovic and Barcel? (2000) conducted a similar stability 
study using octadecylsilica SPE cartridges.  They used a mixture of NPEOs with an 
average ethoxylate length of 6, and separated them by reversed-phase LC, from which 
they elute as a single peak.  They reported complete recovery of the NPEOs from the 
cartridges even after 90 days of storage at ?20?C, but the analytical method did not allow 
them to monitor changes in the concentration of the individual ethoxylates.  In a separate 
experiment, we spiked DI water with the APEOs, extracted the water with the ENV+ 
procedure and stored the cartridge for one year at ?20?C.  Recoveries for the APEOs 
where n  4 ranged from 63 to 120%, with NP and OP showing the highest recoveries 
and the AP4EOs the lowest; in contrast, only around 40% of the AP5EOs was recovered.  
29
These observations are in agreement with the results from the stability study described 
above.
We expected the APs and APEOs to be stable in pure water because their 
molecular structure makes them resistant to abiotic degradation reactions such as 
hydrolysis.  The results from the stability experiment in DI water showed that the 
concentrations remained constant in most cases (RSD = 4 to 14%, 27% for NP), although 
a declining trend was observed in the AP5EOs.  In contrast to DI water, most compounds 
were rapidly degraded in stream water over the 29 days of the experiment, except for NP 
and OP, whose concentrations remained constant (RSDs were 10 and 13% respectively).  
We also observed increasing concentrations of NP4EO and NP5EO in the last two weeks 
of the study; this accumulation was probably the result of the degradation of NPEOs with 
EO chain length > 5 that were naturally present in the samples.  Petrovic and Barcel?
(2000) evaluated the stability of the NPEOs (average EO chain length = 6) in ground 
water and wastewater using different preservatives.  Their results are in good agreement 
with ours.  They observed good stability in ground water, even without the use of
additives, whereas they report important losses of the analytes (up to 80%, depending on 
the preservative used) in the wastewater samples.  They concluded that acidification to 
pH 3 was the most effective preservation method for the NPEOs in ground water; but 
neither acidification, nor the use of formaldehyde prevented the losses of NPEOs in 
wastewater. In our case, lowering the pH resulted in very low recoveries.  We adjusted 
the pH of two DI water samples to 4, spiked them with the APEOs, and then extracted 
according to our procedure. Recoveries ranged from 0 to 19%.  We believe this is due to 
the nature of the interactions between these compounds and the hydroxylated 
30
polystyrene-divinylbenzene copolymer solid phase.  If the retention of the APEOs in this 
solid phase is mainly due to polar interactions such as hydrogen bonds, the addition of 
free protons to the water would result in the protonation of the oxygen atoms in both the 
hydroxyl groups in the polymer and the ether bonds of the APEOs, thus hindering the 
formation of hydrogen bonds between them.  This would not occur when using 
octadecylsilica as the solid phase because the retention mechanism in this case is non-
polar and would not be affected by acidification of the water.
LC/MS/MS analysis.  Chromatographic separation of NP and the NPEOs was 
achieved as expected from Ferguson et al. (2001).  Similarly, OP and the OPEOs were 
also resolved using this column and chromatographic conditions.  When analyzed 
simultaneously, OP and the OPEOs eluted earlier than NP and the respective NPEOs; 
suggesting that the reversed-phase separation mechanism predominates over size-
exclusion when comparing a particular OPEO to the respective NPEO, because of the 
latter being a larger molecule, it should elute earlier than the former were the separation 
based exclusively on size.  Besides, coelution of the OPEOs with the NPEOs was 
observed; as the length of the ethoxylate chain increased, OPEOs tended to elute closer to 
the NPEO with an additional ethoxylate unit, i.e. OP3EO with NP4EO and OP5EO with 
NP6EO.  In these cases, the compounds were resolved spectrometrically.  Retention 
times were repeatable and reproducible; this is especially crucial to properly quantify 
NP1EO and OP, which elute immediately before and after the shift from ES(+) to ES(-).
NP and OP produce [M-H]
?
 quasi-molecular ions in the ES(-) mode (Ferguson et 
al. 2000).  Fragments of these ions were used for quantitation (see Table 2.1).
31
Figure 2.1 MS/MS spectrum of NP?s [M-H]
?
 ion, m/z = 219.  The peak with m/z 93 
can be attributed to the phenolate ion. Fragments with m/z = 119, 133, 147, 161, 175, 
189 and 203 correspond to fragments with the general formula 
?
O-C
6
H
4
-(CH
2
)
x
-
CH=CH
2
, where x = 0 to 6.
[M-H]
-
0.0E+00
4.0E+04
8.0E+04
1.2E+05
1.6E+05
2.0E+05
70 110 150 190 230
m/z
i
n
t
e
n
s
i
t
y
93
133
119
219
203
189175161
147
32
Fragmentation of NP?s quasi-molecular ion (m/z = 219) results in a series of peaks 
separated by 14 Da (Figure 2.1), which we attribute to the consecutive loss of CH
2
groups, resulting in a series of ions with a general structure equivalent to 
?
O-C
6
H
4
-
(CH
2
)
x
-CH=CH
2
, where x = 0 to 6, with x = 1 being the most abundant fragment (m/z = 
133).  We refer to these structures as general because NP is composed of different 
isomers with branched alkyl chains and not linear as the proposed structure would 
suggest.  In contrast, OP?s [M-H]
?
 ion (m/z = 205) forms only one fragment (m/z = 133) 
in lower abundance than the parent ion (Figure 2.2).  Pedersen and Lindholst (1999)
observed a similar fragmentation pattern in a single quadrupole MS, using high inlet 
voltages to enhance collision-induced dissociation before allowing the ions into the mass 
analyzer.  They attributed the 133 fragment to the loss of a C
5
H
12
 group from the quasi-
molecular ion.  Examining this fragment at a higher resolution we observed that the peak 
is actually composed by two ions, m/z 133 and 134, which could correspond to the 
structures 
?
O-C
6
H
4
-C(CH
3
)=CH
2
 and the radical 
?
?O=C
6
H
4
=C(CH
3
)-CH
3
 respectively.  
Using high resolution results in lower sensitivity for all the compounds; therefore, 
quantitation was performed using m/z 133 at a lower resolution in order to merge the two 
ions and achieve greater ion abundance.  The difference between the fragmentation 
patterns of NP and OP is apparently due to the presence of several different NP isomers 
that results in the formation of different fragments.  The fact that the most abundant 
fragment for NP?s quasi-molecular ion is equivalent to OP?s, suggests that fragmentation 
occurs preferentially at a tertiary alpha carbon.  This observation fits well with the results 
of a GC/MS study by Wheeler et al. (1997), who partially characterized the different 
33
Figure 2.2 MS/MS spectrum of OP?s [M-H]
?
 ion, m/z = 205. The peak at around m/z 
133 is actually two ions, m/z 133 and 134, which are attributed to 
?
O-C
6
H
4
-
C(CH
3
)=CH
2
 and 
?
?O=C
6
H
4
=C(CH
3
)-CH
3
 respectively.
[M-H]
-
0.0E+00
5.0E+04
1.0E+05
1.5E+05
2.0E+05
2.5E+05
3.0E+05
70 110 150 190 230
m/z
i
nt
ens
i
t
y
205
133
34
isomeric groups occurring in commercial NP, and concluded that in more than 80% of 
the isomers the alpha carbon was tertiary.
Previous studies of the APEOs using LC/MS favor the use of sodium salts for 
quantitative analysis because of the abundant formation of stable [M+Na]
+
 adduct ions 
due to the affinity of the APEOs to this cation (Ferguson et al. 2000, Shang et al. 1999).  
The stability of these ions is such that they undergo minimal fragmentation in the 
collision cell.  In contrast, [M+NH
4
]
+
 adducts are more labile, which makes the use of 
ammonium salts more suitable for MS/MS analysis.  Stability of the ammonium adducts 
increases with increasing ethoxy chain length, as it does for the sodium adducts.  After 
maximizing the abundance of the [M+NH
4
]
+
 parent ions by optimizing the cone voltage 
for each one of them, the most abundant fragments obtained in daughter-ion spectra of 
each individual parent ion were chosen for quantitation in ES(+).  The collision energy 
for each compound was optimized to maximize the abundance of the selected fragments 
(Table 2.1).  Fragmentation is very reproducible; in a study published recently, Houde et 
al. (2002) measured the NP1-17EOs by LC/MS/MS using ammonium adducts and 
reported the same daughter ions selected in this study for NP1EO to NP5EO (Table 2.1).
Fragmentation of the [M+NH
4
]
+
 ions of the APEOs in the collision cell splits the 
ions at the alkyl chain ? aromatic ring bond.  The spectra of the AP1EOs are shown in 
Figure 2.3.  For OP1EO (Fig. 2.3a), the base peak can be assigned to the carbocation 
CH
3
C(CH
3
)
2
CH
2
C(CH
3
)
2
+
, whereas the rest of the OP1EO molecule, [C
6
H
5
-OCH
2
CH
2
-
OH + H]
+
, appears as the fragment with m/z = 139.  The peaks at m/z = 57 and 71 may 
correspond to fragments of the alkyl chain, C(CH
3
)
3
+
 and C(CH
3
)
3
CH
2
+
 respectively; the 
35
Figure 2.3 (a) MS/MS spectrum of OP1EO?s [M+NH
4
]
+
 ion, m/z = 268; (b) MS/MS 
spectrum of NP1EO?s [M+NH
4
]
+
 ion, m/z = 282.
[M+H]
+
[M+NH
4
]
+
0.0E+00
5.0E+04
1.0E+05
1.5E+05
2.0E+05
2.5E+05
3.0E+05
50 100 150 200 250 300
m/z
i
nt
ens
i
t
y
a
57
71
113
139
251
268
113
57
71
139
O
O
H
H
[M+NH
4
]
+
0.0E+00
5.0E+04
1.0E+05
1.5E+05
2.0E+05
2.5E+05
50 100 150 200 250 300
m/z
i
nt
ens
i
t
y
57
71
85
127
139
265
282
b
[M+H]
+
36
former ion being produced more abundantly than the latter because of its greater stability.  
The relative size of the [M+H]
+
 and [M+NH
4
]
+
 peaks to the respective base peaks 
suggests that NP1EO (Fig. 2.3b) is more susceptible to fragmentation than OP1EO, as 
previously observed with the APs.  NP1EO?s fragmentation pattern is very similar to 
OP1EO?s, but it also reflects the complex isomeric composition of NP1EO.  The base 
peak (m/z = 127) might also correspond to a tertiary carbocation, C
6
H
13
C(CH
3
)
2
+
, further 
supporting the observation that most of the NP isomers have a tertiary alpha carbon as 
noted above.  Fragments m/z = 57 and 71 also appear in the NP1EO spectrum, but in this 
case the latter is more abundant, possibly because it would correspond to C
2
H
5
C(CH
3
)
2
+
, 
which is more stable than the secondary carbocation assigned to the same peak in 
OP1EO?s spectrum.  The fragment at m/z = 85 would correspond to yet another tertiary 
carbocation, C
3
H
7
C(CH
3
)
2
+
, not present in OP derivatives.
In the case of the AP2-5EOs the most abundant ion corresponds to the phenol-
ethoxylate moieties, which can be assigned the structure [C
6
H
5
-(OCH
2
CH
2
)
x
-OH + H]
+
, 
where x = 2 to 5.  These ions are the same for the NPEOs and their respective OPEO s, 
but the use of the MRM acquisition mode prevented any misidentification of the two 
groups.  Peaks assigned to carbocations in NP1EO and OP1EO spectra were also 
observed, but in less abundance.  As the length of the alkyl chain increases different 
fragmentation patterns occur besides those discussed earlier.  Fragments of the ethoxylate 
chain can be observed, as well as from the phenol-ethoxylate ion shown above.  In Figure 
2.4, the daughter spectrum for [NP5EO+NH
4
]
+
 is presented as an example of this 
fragmentation, and possible structures for the fragments are shown in Table 2.2.  In a 
study of NPEOs by atmospheric pressure chemical ionization (APCI) LC/MS, Castillo 
37
Figure 2.4 MS/MS spectrum of NP5EO?s [M+NH
4
]
+
 ion, m/z = 458. Base peak 
corresponds to the ion [C
6
H
5
-(OCH
2
CH
2
)
5
-OH + H]
+
. Proposed structures for the 
fragments labeled as a, b, c, d, and e can be found in Table 2.2.
0.0E+00
1.0E+06
2.0E+06
3.0E+06
4.0E+06
5.0E+06
0 100 200 300 400 500
m/z
i
nt
ens
i
t
y
315
441
458
aa
a
abb
b
b
b
c
c
c
c
c
d
d
e
e
e
[M+H]
+
[M+NH
4
]
+
38
Table 2.2 Possible structures for fragments of [NP5EO+NH
4
]
+
.
Group m/z Proposed structure
a
57, 71, 
85, 127
H
2x+1
C
x
C
CH
3
CH
3
x = 1, 2, 3, 
6
b
45, 89, 
133, 177, 
221
O
OH
+ H
+
x
x = 0 to 4
c
121, 165, 
209, 253, 
297
+ H
x
+
O
O x = 0 to 4
d
139*, 
183*, 
227, 271, 
315
+ H
x
+
O
OH x = 1 to 5
e
247, 291, 
335, 
379*, 
423*
+ H
x
O
OH
19
C
9
+
x = 0 to 4
* Not visible in figure 5.
39
et al. (1999) observed similar fragmentation patterns in the APCI source and proposed 
comparable structures to those in Table 2.2, except for groups? c and e, for which they 
suggested cleavage of both the alkyl and ethoxylate chains in the same molecule.  The 
structures we propose are simpler (i.e. only one of the chains is fragmented) and, 
therefore, more likely to be produced under the relatively mild fragmentation conditions 
in the collision cell.  In the case of the NPEOs with EO chain length 
 6 fragmentation 
still occurs but it is less extensive.  For n = 6 and 7, the most abundant daughter ion is 
[M+H]
+
, and for compounds with n = 8 to 17 all fragmentation peaks decrease 
substantially, even [M+H]
+
, and the parent ion [M+NH
4
]
+
 becomes the base peak in all 
spectra, suggesting that stability of the ammonium adducts increases with ethoxylate 
chain length.
Method application and QA/QC.  The methods for water and sediment were 
tested with samples obtained in July 2001 from Back River, a tidal sub-estuary of the 
Chesapeake Bay located in Baltimore County, MD.  Results for two sites (1 and 5) are 
shown in Table 2.3.  The levels of APs and APEOs found in Back River were within the 
range of typical values found in fresh waters in North America (Bennie 1999).  NP and 
the NPEOs occurred in higher concentrations than OP and the OPEOs; this has been 
observed before in most studied sites and it is attributed to the usage patterns of these 
compounds (Ferguson et al. 2000).  The single most abundant compound present in water 
and sediment was NP, which agrees with previous findings in different places (Ying et al. 
2002).  Higher concentrations were observed in site 5 because it is close to the outfall of a 
wastewater treatment plant.  A more comprehensive set of data for the Back River and its 
interpretation will be published elsewhere.
40
Table 2.3 Concentrations of the APs and APEOs in two sampling sites (1 and 5) in 
Back River, MD.
water sediment
MQL, conc. (RSD, %), ng/L MQL, conc. (RSD, %), ng/g-dry wt
Compound ng/L Site 1
a
Site 5
b
ng/g-dry wt Site 1
a
Site 5
a
NP 10 140 (15) 200 (3) 40 410 (0.3) 6700 (4)
NP1EO 9 BQL 67 (5) 40 240 (3) 1800 (8)
NP2EO 2 12 (11) 57 (4) 9 100 (2) 910 (0.3)
NP3EO 2 9 (13) 59 (4) 9 37 (3) 340 (5)
NP4EO 5 12 (5) 52 (2) 9 26 (0) 160 (7)
NP5EO 14 14 (4) 21(0.4) 9 24 (0) 120 (2)
OP 9 BQL BQL 40 BQL 410 (7)
OP1EO 9 BQL BQL 40 BQL 110 (9)
OP2EO 8 BQL BQL 9 BQL 55 (2)
OP3EO 11 BQL BQL 9 BQL 33 (6)
OP4EO 14 BQL BQL 9 BQL 30 (0)
OP5EO 12 BQL BQL 9 BQL 31 (3)
a
n = 2; 
b
n = 4; BQL: below MQL
41
The instrument detection limits, LODs, calculated for these compounds (Table 
2.4) are comparable to those obtained using LC/MS (Ferguson 2000), but the use of an 
MS/MS system is advantageous in that it provides better selectivity.  MDLs in water are 
also similar to those reported by Ferguson et al. (2000), but are several orders of 
magnitude lower than the MDLs reported by Houde et al. (2002), who used a similar 
LC/MS/MS method.  The higher sensitivity of our approach is most likely due to the high 
concentration factor (2700) allowed by the SPE procedure utilized here.  MDLs in 
sediment are up to 2 orders of magnitude lower than those reported by Ferguson et al.
(2001).  Recoveries from water and sediment were above 75% for most analytes (Table 
4); we attributed the lower values observed for NP3-5EO and OP5EO in water to a long 
storage period of the cartridges.  Previous extractions of DI water spikes - in which 
cartridges were not stored - yielded recoveries of 83?14, 91?12, 91?11, and 91?8% for 
NP3EO, NP4EO, NP5EO and OP5EO respectively (average of 5 values ? SD).  
Reproducibility of the data was very good for all analytes as suggested by the low RSD 
values obtained for the samples (Table 2.3).  MQLs were at least 5 times the respective 
MDL, but usually higher because we found traces of NP, the NP1-5EOs and the OP2-
5EOs in solvent blanks, and all MQLs were set at least three times the amount of the 
interference.  The source of the contaminants was traced to the LC column; however, 
their presence did not affect quantitation because the contamination was present in all 
samples, including calibration curves, and was reproducible (RSD between 6 and 16%, n 
= 10).  Sediment and water procedural blanks were below MQL for all compounds.
42
Table 2.4 Water and sediment method performance parameters.
Water Sediment
LOD, MDL, Recovery, MDL, Recovery* (SD),
Compound pg ng/L % ng/g-dry wt %
NP 6 0.2 95 0.9 95 (4)
NP1EO 5 0.2 96 0.8 78 (1)
NP2EO 0.7 0.03 81 0.1 110 (4)
NP3EO 0.2 0.008 66 0.03 84 (2)
NP4EO 0.2 0.008 63 0.03 87 (2)
NP5EO 0.2 0.008 38 0.03 99 (4)
OP 9 0.3 110 1.4 67 (2)
OP1EO 5 0.2 96 0.8 92 (1)
OP2EO 0.5 0.02 89 0.08 99 (2)
OP3EO 0.2 0.008 89 0.03 83 (3)
OP4EO 0.1 0.004 77 0.02 92 (3)
OP5EO 0.1 0.004 36 0.02 91 (3)
*n = 2
43
2.5 Conclusions
The methodologies presented allow the simultaneous extraction, identification and 
quantitation of NP, OP, and their respective APEOs in water and sediment.  High 
concentration factors in the case of water extraction and the use of MS/MS detection 
resulted in low detection limits and accurate identification of the analytes in 
environmental samples.  The analysis of NP and the NPEOs by tandem mass 
spectrometry and comparison to the respective OPs permitted insights into the isomeric 
composition of the NPs; although further analysis (such as NMR) is necessary to confirm 
the precise structures of the isomers.  These methodologies are presently being used in 
our laboratory to conduct distribution and fate studies of the APEOs, and the LC/MS/MS 
method has been successfully used to quantify these compounds in fish tissue and eggs, 
which confirm the wide range of applications of the method (Schmitz-Afonso et al. 
2003).
44
CHAPTER 3 ? ANALYTICAL METHODS.  LONG-CHAIN ETHOXYLATES
AND CARBOXYLATES
This chapter to be submitted to the Journal of Chromatography A as ?Determination of 
octyl- and nonylphenol ethoxylates and carboxylates in environmental samples by isotope 
dilution liquid chromatography/tandem mass spectrometry.? by Loyo-Rosales, J. E., 
Rice, C. P., and Torrents, A.
3.1 Abstract
This work presents an LC-MS-MS-based method for the quantitation of 
nonylphenol ethoxylates (NP0-16EO) and octylphenol ethoxylates (OP0-5EOs) in water, 
sediment, and suspended particulate matter, and three of their carboxylated derivatives in 
water.  The alkylphenol ethoxylates (APEOs) were analyzed using isotope dilution mass 
spectrometry with [
13
C
6
]-labeled analogues, whereas the carboxylated derivatives were 
determined by external standard quantitation followed by confirmation using standard 
additions.  The method was used to study the behavior of the APEOs in a wastewater 
treatment plant (WWTP).  Our results indicate that the total dissolved NP0-16EO
concentration was reduced by approximately 99% from influent (390 ?g/L) to final 
effluent (4 ?g/L), and that the total OP0-5EO concentration decreased 94% from 3.1 to 
0.2 ?g/L.  In contrast, the carboxylated derivatives were formed during the process with
45
NP0-1EC concentrations increasing from 1.4 to 24 ?g/L.  It was also observed that short-
chain APEOs were present in higher proportions in particulate matter, presumably due to 
their greater affinity for solids compared to the long-chain homologues.  The method was
also used to analyze a surface-water sample from a WWTP-impacted estuary, where only 
NP and NP0-1EC were detected in concentrations of 0.49 and 4.8 ?g/L respectively; in 
other words, 90% of the mass of NPEO-related compounds detected was in the form of 
carboxylated derivatives.  Application of this method to sediment analysis showed 
nonylphenol to be the single most abundant compound in these sediments.  Differences in 
homologue distributions suggest the presence of treated effluent at some of the sampling
sites and non-treated sources in the rest of them.
3.2 Introduction
The alkylphenol ethoxylates (APEOs) are nonionic surfactants widely used in 
diverse cleaning and industrial applications (Talmage 1994).  The main APEOs produced 
in the United States are the nonylphenol ethoxylates (NPEOs) and the octylphenol 
ethoxylates (OPEOs), accounting for approximately 80 and 20% of the total production 
(APERC 1999).  In terms of volume, the production of NPEOs in the US was close to 
300 million metric tons in 1994 (Ferguson et al. 2001).  Many of the uses for APEOs 
imply their disposal in wastewater, which can then be sent to wastewater treatment plants 
(WWTP) or released into natural bodies of water; in both instances, the APEOs are 
subjected to biotransformation.  It is widely accepted that this transformation results in 
46
the formation of several products, including the mono- and diethoxylates, and the 
carboxylated derivatives of the APEOs (APECs) in aerobic environments, and the parent 
alkylphenols (APs), nonylphenol (NP) or octylphenol (OP), when degradation occurs 
under anaerobic conditions (Montgomery-Brown and Reinhard 2003).  Moreover, studies 
on the environmental occurrence of the APEOs transformation products suggest that they 
are widespread aquatic pollutants (Bennie 1999, Kolpin et al. 2002), especially in waters 
impacted by WWTP effluents.  The ubiquitous presence of these compounds, together 
with evidence of the toxic and estrogenic properties of some of their transformation 
products (Servos 1999), has raised interest in their environmental fate, requiring the 
development of appropriate analytical methodologies.
Traditionally, APEOs and their transformation products were analyzed by GC-MS 
and HPLC with fluorescence detection (HPLC-F), e.g. Ahel et al. 1987, Datta et al. 2002.  
Due to intrinsic limitations of these methods?the low sensitivity of GC-MS to the long-
chain APEOs and the relatively low selectivity of fluorescence detection?and the 
increased availability of LC-MS instrumentation capable of overcoming these limitations, 
several methods based on LC-MS detection have been developed, e.g. Ferguson et al. 
2001, Scullion et al. 1996, Shang et al. 1999a, Di Corcia et al. 2000, Ferguson et al. 2000, 
Takino et al. 2000.  One of the limitations of the analysis of environmental samples by 
LC-MS is the suppression or enhancement of analyte signal by matrix components 
(Ferguson et al. 2000).  In order to account for these effects, isotope dilution mass 
spectrometry (IDMS) techniques can be used.  IDMS is based in the addition of an 
isotope-enriched analogue of the analyte to the sample before processing; the MS 
response of the analyte is then normalized to the response of the analogue to determine 
47
the original analyte concentration using a calibration curve.  This approach not only 
compensates for variations due to matrix effects, but also for losses of the analyte during 
sample extraction (Prichard et al. 1996).  Although IDMS is considered a definitive 
method, it has the inherent disadvantage of the high cost and low availability of 
appropriate isotopic analogues (Sargent et al. 2002).  This is particularly true for the 
APEOs and their transformation products, labeled or not, because only a few of them are 
available as pure analytical standards. 
In a previous report, we presented a quantitative LC-MS-MS method to analyze 
NP, OP, and the short-chain AP1-5EOs in water and sediment (Loyo-Rosales et al. 
2003).  The work presented here is a modification of these methods to employ an isotope 
dilution mass spectrometry (IDMS) quantitation approach using [
13
C
6
]-labeled analogues 
and to include the analysis of the long-chain NP6-16EOs in water and sediment, the OP0-
5EOs and NP0-16EOs in suspended particulate matter, and the following APECs in 
water: nonylphenoxyacetic acid (NP0EC), nonylphenoxyethoxyacetic acid (NP1EC), and 
octylphenoxyacetic acid (OP0EC).  Due to the lack of [
13
C
6
]-labeled carboxylates, the 
APECs were determined using external standard quantitation followed by confirmation 
with a standard additions approach.  Examples of the application of these methods to 
wastewater, freshwater, sediment, and particulate matter samples are also presented.
3.3 Experimental
Standards and reagents.  Individual compounds were used as standards for the 
short-chain APEOs and the carboxylates, and they were either acquired from commercial 
48
sources ? NP (Schenectady International, Schenectady, NY, USA; purity  95%), OP 
(Aldrich, Milwaukee, WI, USA; 97%), NP2EO (R&D product from Aldrich), NP0EC 
(R&D product, Huntsman Chemicals, Austin, TX, USA), NP1EC (Aldrich; 90%), and 
OP0EC (R&D product, Huntsman) ? or purified in the laboratory from commercially-
available mixtures by flash chromatography as described in previous publications (Datta 
et al. 2002, Loyo-Rosales et al. 2003) ? NP1EO from Surfonic N-10 (Huntsman); NP3-
5EOs from POE(4) nonylphenol (Chem Service, West Chester, PA, USA); OP1-5EOs 
from POE(3) and POE(5) tert-octylphenol (Chem Service).  Purity of the isolated 
compounds was > 99% with the exception of OP1EO (94%).  Due to the lack of pure 
standards for the individual NP6-16EOs when this work was performed, previously 
characterized mixtures were used as standards.  Originally, we attempted to use 
Marlophen 810 (Chemische Werke, H?ls, Germany), characterized by Ahel and 
collaborators (Ahel et al. 2000), but the presence of unreported OPEOs in the mixture 
resulted in an overestimation of the results by a factor of 1.7 to 2.8, depending on the 
compound.  Therefore, we used Surfonic N-95 (Schenectady International; > 97%), 
whose characterization was provided by Huntsman.  A mixture of [
13
C
6
]NP (Cambridge 
Isotope Laboratories, Andover, MA, USA), [
13
C
6
]NP(1.6)EO and [
13
C
6
]NP(9.5)EO?
both synthesized by Ferguson et al. (2001)?was used for isotope dilution quantitation.  
All standards were stored at ?20 ?C.  Organic-free, high purity methanol, acetone, and 
dichloromethane (DCM) were acquired from Burdick & Jackson (Honeywell 
International, Muskegon, MI, USA).  Carbon-free deionized water (DI water) was 
obtained in the laboratory using a NANOpure water purification system (Barnstead 
International, Dubuque, IA, USA).  Other reagents included ACS Certified Plus 
49
hydrochloric acid (Fisher Chemicals, Fair Lawn, NJ, USA), anhydrous sodium sulfate in 
granular powder form (Mallinckrodt Baker, Paris, KY, USA) and 99.99+% purity 
ammonium acetate (Aldrich).  Previous to its use, all glassware was baked at 400?C for 4 
h in an industrial oven (Grieve, Round Lake, IL, USA) to avoid contamination with the 
analytes of interest, especially NP.
Water and sediment collection. Wastewater samples from a WWTP in the Mid-
Atlantic region of the United States, and a surface water sample from the Back River, 
MD were analyzed in order to test this method.  All of them consisted of ~ 4-L grab 
samples; they were collected in July 2004 in previously-baked 1-gal amber glass bottles.  
The sampling sites in the WWTP were selected to represent different stages in the 
wastewater treatment, and they included raw wastewater, primary, secondary, tertiary and 
final effluents.  The sample from Back River was provided by the MD Department of 
Natural Resources; it was taken at a location in the river east of Stansbury Point (NAD83 
coordinates: 39.2834?, -76.4497? [Maryland DNR 2004]) and downstream from the Back 
River WWTP effluent.  Upon return to the laboratory, approximately 1 to 4 L of each 
sample was filtered using a glass vacuum filter holder (Millipore, Billerica, MA, USA) 
with two pre-weighed glass fiber filters in series: GF/A, 1.6 ?m particle retention, and 
GF/F, 0.7 ?m (Whatman, Clifton, NJ, USA).  Both filters and the filtration unit were 
previously baked 4 h at 400?C.  Filtered samples were stored overnight at 4?C in pre-
baked amber glass bottles.  Extraction occurred less than 24 h after filtration.  The filters 
were dried overnight in the extraction hood, stored under vacuum in a desiccator with 
silica gel, and finally weighed to calculate the amount of particulate matter present in the 
samples.  
50
Sediment samples were collected from seven different sites in the Baltimore 
Harbor area in October 1999 (Figure 3.1).  Exact locations were: Back River 1 (BR1): 
39.2730?, -76.4419?; Back River 2 (BR2): 39.2450?, -76.4311?; Bear Creek 1 (BC1): 
39.2585?, -76.4777?; Bear Creek 2 (BC2): 39.2272?, -76.4999?; Inner Harbor (IH): 
39.2783?, -76.5931?; Gwynn?s Falls (GF): 39.2615?, -76.6219?; White Rock (WR): 
39.1730?, -76.4863?.  The procedure for sediment collection was reported previously 
(Loyo-Rosales et al. 2003).  Briefly, a ponar dredge was used to collect four to six grabs 
of sediment at each sampling site.  The top 2 cm of each grab were removed with a 
stainless steel spoon and homogenized in a stainless steel container.  The sediments were 
then poured into 250-mL pre-baked glass jars and stored in ice until arrival in the 
laboratory, where they were stored at -20?C until extraction.
Sample extraction. The extraction of APs and APEOs from water was 
performed as reported earlier for the short-chain APEOs (Loyo-Rosales et al. 2003), 
except that the [
13
C
6
]-labeled compounds were added before the extraction rather than 
immediately before LC-MS-MS analysis.  Briefly, solid-phase extraction cartridges with 
a hyper-cross-linked hydroxylated poly(styrene-divinylbenzene) copolymer (SDB, 
Isolute ENV+, 500 mg, 6 mL, International Sorbent Technology, Hengoed, U.K.) were 
rinsed with 10 mL of DCM, 12 mL of acetone, and 12 mL of DI water using a vacuum 
51
Figure 3.1 Map of the Baltimore Harbor region showing approximate locations for 
the sediment sampling sites and the Back River wastewater treatment plant.  BC: 
Bear Creek; BR: Back River; GF: Gwynn?s Falls; IH: Inner Harbor; WR: White 
Rock.
BACK RIVER
BC2
BC1
BR2
BR1
WR
IH
GF
PATAPSCO RIVER
INNER HARBOR
CHESAPEAKE 
BAY
BEAR CREEK
BACK RIVER WWTP
52
manifold.  The water samples were spiked with 20 ?L of the [
13
C
6
]-standard mix and 
passed through the cartridges using vacuum.  Once empty, the bottles that contained the 
water samples were rinsed with 10 mL of methanol.  The cartridges were then dried and 
eluted with 10 mL of DCM, 10 mL of methanol, and 12 mL of acetone.  The eluates for 
each sample and the methanol used to rinse the bottle were pooled together and reduced 
to 0.5 mL of methanol in a nitrogen evaporator (Organomation Associates, South Berlin, 
MA, USA), after which 0.5 mL of DI water was added.  This mixture was then filtered 
through an Acrodisc LC 13-mm syringe filter with a 0.2-?m PVDF membrane (Pall 
Gelman Laboratory, Ann Arbor, MI, USA) using a 1-mL glass syringe (Hamilton, Reno, 
NV, USA).  Both syringe and filter were rinsed with 0.5 mL of a 1+1 methanol/water 
mixture that was added to the extract; final volume was adjusted to 1.5 mL with clean 
methanol/water mixture when necessary.
The APECs were extracted from water by liquid-liquid extraction using DCM as 
follows: a 500 mL aliquot of each filtered water sample was poured into a 1-L separation 
funnel and its pH adjusted to ~2 with HCl.  50 mL of DCM were added to the acidified 
sample, and vigorously shaken for ~2 min.  After allowing both phases to separate, the 
DCM fraction was collected and the extraction was repeated two more times for a total of 
150 mL of DCM.  The solvent in the extract was exchanged to methanol in a rotary 
evaporator and reduced to 0.5 mL in a nitrogen evaporator, after which 0.5 mL of DI 
water was added.  The extract was then filtered and its volume adjusted as described 
above for the ethoxylates.
The APs and APEOs in the particulate matter retained by the filters were 
extracted as follows: the dry filters were spiked with 20 ?L of the [
13
C
6
]-standard mix and 
53
then extracted with ~150 mL of methanol in a Soxhlet apparatus with an Allihn 
condenser for 8 hours.  The methanol extracts were reduced to 0.5 mL by sequential 
rotary and nitrogen evaporation.  The extracts were then filtered and their volumes 
adjusted as described for the ethoxylates in the section on water extraction above.
APs and APEOs were extracted from sediments according to a previously 
published procedure (Loyo-Rosales et al. 2003); as with the water, the addition of the 
[
13
C
6
]-labeled compounds occurred prior to extraction.  Briefly, the sediments were 
thawed, homogenized and air-dried.  Approximately 1 g of dry sediment was mixed with 
sodium sulfate, spiked with 20 ?L of the [
13
C
6
]-standard mix, and extracted by 
accelerated solvent extraction in an ASE 200 (Dionex, Sunnyvale, CA, USA) using 
hexane and acetone as extraction solvents.  These extracts were exchanged to hexane and 
cleaned using a solid-phase extraction procedure with amino-propyl cartridges.  After 
clean-up, the extracts were exchanged to 0.5 mL of methanol and then processed as the 
water extracts.
Sorption to filters.  In order to evaluate the sorption of APs and APEOs to filters, 
six 1-L amber-glass bottles were filled with DI water; four of them (?spikes?) were 
spiked with 50 ?L of the [
13
C
6
]-standard mix, and then all six bottles were filtered as 
described in the section on water collection.  Water from the other two bottles 
(?references?) was spiked with 50 ?L of the [
13
C
6
]-standard mix after filtration.  All six 
water samples were then extracted as the ethoxylates above.  Separately, the filters were 
allowed to dry and extracted as described for the particulate matter, except that in this 
case, the filters from the two ?reference? samples were spiked with 50 ?L of the [
13
C
6
]-
54
standard mix, whereas no additional [
13
C
6
]-standard was added to the filters from the 
?spike? samples.
LC-MS-MS analysis.  The chromatographic and mass spectrometry conditions 
for AP and APEOs analysis were published in detail in a previous work (Loyo-Rosales et 
al. 2003).  Briefly, separation was performed at 60?C in a Waters 2690 XE separations 
module (Waters, Milford, MA, USA) with a 150 x 4.6 mm MSpak GF-310 4D column 
(Shodex, Shoko, Tokyo, Japan).  The mobile phase was 10 mM ammonium acetate in 
50:50 v/v methanol/water and was gradually changed to 100% methanol.  The tandem 
mass spectrometry analysis was performed using a Quattro LC triple quadrupole mass 
spectrometer (Micromass, Manchester, UK) with an electrospray ionization source.  The 
different compounds were analyzed using multiple-reaction monitoring (MRM); APEOs 
in electrospray positive mode (ESI+) and NP and OP in electrospray negative (ESI?).  
The MS conditions, as well as the parent and fragment ions used to identify and 
quantitate the NP6-16EOs and the [
13
C
6
]NP5-15EOs are presented in Table 3.1.  These
parameters for NP, OP, [
13
C
6
]NP, NP1-5EO, OP1-5EO, and [
13
C
6
]NP1-4EO are available 
in  Loyo-Rosales et al. 2003.  Example chromatograms for the NP6-16EO from a tertiary 
effluent extract are shown in Figure 3.2.  Concentrations were calculated by the internal 
standard method using the respective [
13
C
6
]-labeled compounds, e.g. [
13
C
6
]NP6EO was 
used as internal standard for NP6EO.  Eight calibration points were prepared in a 50:50 
methanol/water mixture; the concentrations are listed in Table 3.2.  Peak integration and 
quantitation were performed automatically using MassLynx 4.0 (Micromass).
55
Table 3.1 Ions and MS/MS parameters used for the quantitation of NPEOs, n=6-16, 
NP0EC, NP1EC and OP0EC. Parent ions correspond to [M+NH
4
]
+
 in ESI(+), and to 
[M-H]
?
 in ESI(-).  Fragment ions for NP6-7EO correspond to [C
6
H
5
-(OCH
2
CH
2
)
x
-
OH + H]
+
 with x = 6-7, to [M+H]
+
 for NP8-16EO, C
9
H
19
-C
6
H
5
-O
?
 for NP0-1EC and 
C
8
H
17
-C
6
H
5
-O
?
 for OP0EC.
compound
parent ion,
m/z
fragment ion,
m/z
retention time,
min
cone,
V
collision,
eV
ion mode
NP6EO 502 359 18.2 20 18 ESI + 
NP7EO 546 403 17.4 25 18 ESI +
NP8EO 590 573 16.7 30 18 ESI +
NP9EO 634 617 16.0 30 20 ESI +
NP10EO 678 661 15.3 35 20 ESI +
NP11EO 722 705 14.7 35 20 ESI +
NP12EO 766 749 14.1 35 22 ESI +
NP13EO 810 794 13.4 40 22 ESI +
NP14EO 854 838 12.8 40 24 ESI +
NP15EO 898 882 12.1 45 24 ESI +
NP16EO 942 926 11.4 45 26 ESI +
[
13
C
6
]NP5EO 464 321 19.3 20 18 ESI +
[
13
C
6
]NP6EO 508 365 18.2 20 18 ESI +
[
13
C
6
]NP7EO 552 409 17.4 25 18 ESI +
[
13
C
6
]NP8EO 596 579 16.7 30 18 ESI +
[
13
C
6
]NP9EO 640 623 16.0 30 20 ESI +
[
13
C
6
]NP10EO 684 667 15.3 35 20 ESI +
[
13
C
6
]NP11EO 728 711 14.7 35 20 ESI +
[
13
C
6
]NP12EO 772 755 14.1 35 22 ESI +
[
13
C
6
]NP13EO 816 800 13.4 40 22 ESI +
[
13
C
6
]NP14EO 860 844 12.8 40 24 ESI +
[
13
C
6
]NP15EO 904 888 12.1 45 24 ESI +
NP0EC 277 219 10.2 30 20 ESI -
NP1EC 321 219 10.7 20 15 ESI -
OP0EC 263 205 9.2 30 20 ESI -
56
Figure 3.2 Extracted MS-MS chromatograms of the NPEOs in the dissolved fraction 
of WWTP effluent extract (A and B correspond to tertiary effluent, whereas C to 
primary). (A) NP6-11EO, concentrations were 237, 179, 203, 127, 104, and 92 ng/L 
respectively; (B) NP12-16EO, 80, 53, 45, 5, and 15 ng/L; (C) NP0-1EC, OP0EC, 
1050, 1650 and 97 ng/L.
2.50 5.00 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00
Time3
100
%
6
100
%
2
100
%
6
100
%
8
100
%
10
100
%
APE050520a45 1: MRM of 32 Channels ES+ 
722.3 > 705.3
1.97e4
Area
14.79
17940
6.79
10089
11.88
2584
APE050520a45 1: MRM of 32 Channels ES+ 
678.3 > 661.3
2.51e4
Area
15.39
23927
6.91
13306
12.49
5157
APE050520a45 1: MRM of 32 Channels ES+ 
634.3 > 617.3
3.57e4
Area
16.12
34676
7.03
14253 13.70
5204
APE050520a45 1: MRM of 32 Channels ES+ 
590.3 > 573.5
8.67e4
Area
16.85
87318
14.54
22815
7.16
22465
APE050520a45 1: MRM of 32 Channels ES+ 
546.3 > 403.1
3.07e4
Area
17.57
32462
15.51
5659
6.43
3664
APE050520a45 1: MRM of 32 Channels ES+ 
502.3 > 359.1
6.23e4
Area
18.66
77622
16.48
4081
NP11EO
NP10EO
NP9EO
NP8EO
NP7EO
NP6EO
A
57
Figure 3.2 (cont.)
2.50 5.00 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00
Time13
100
%
18
100
%
26
100
%
32
100
%
47
100
%
APE050520a45 1: MRM of 32 Channels ES+ 
942.6 > 925.8
4.17e3
Area
11.40
25806.31
3551
APE050520a45 1: MRM of 32 Channels ES+ 
898.3 > 881.6
6.17e3
Area
12.12
4756
6.43
3467
APE050520a45 1: MRM of 32 Channels ES+ 
854.3 > 837.6
7.49e3
Area
12.73
5534
6.55
6151
APE050520a45 1: MRM of 32 Channels ES+ 
810.3 > 793.7
1.10e4
Area
13.45
9364
6.67
4467
APE050520a45 1: MRM of 32 Channels ES+ 
766.3 > 749.5
1.51e4
Area
14.18
13646
6.19
10061
NP12EO
NP13EO
NP14EO
NP15EO
NP16EO
B
58
Figure 3.2 (cont.)
2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
Time21
100
%
23
100
%
56
100
%
APE072804a12 MRM of 6 Channels ES- 
263.3 > 205
3.79e3
Area
10.44
927
APE072804a12 MRM of 6 Channels ES- 
321.4 > 219
9.21e3
Area
12.41
6733
APE072804a12 MRM of 6 Channels ES- 
277.5 > 219
9.93e3
Area
11.69
4923
OP0EC
NP1EC
NP0EC
C
59
Table 3.2 Calibration standards for the NP6-16EOs.
ng/mL
nEO CC-1 CC-2 CC-3 CC-4 CC-5 CC-6 CC-7 CC-8 
6 9 19 38 76 150 300 470 940
7 11 23 46 91 180 370 570 1100
8 12 25 50 100 200 400 620 1200
9 12 25 49 99 200 390 620 1200
10 11 22 44 88 180 350 550 1100
11 9 18 36 72 140 290 450 900
12 7 14 27 54 110 220 340 680
13 5 9 19 38 76 150 240 470
14 3 6 12 24 49 97 150 300
15 2 4 7 14 29 58 90 180
16 1 2 4 8 17 34 53 110
60
The carboxylates were analyzed with the same instruments and chromatographic 
column as for the ethoxylates, and using almost identical parameters.  Chromatographic 
separation was also performed at 60 ?C, but in isocratic mode.  The mobile phase was 10 
mM ammonium acetate in 50:50 v/v methanol/water and it was run at a 0.2 mL/min flow 
rate.  Total run time was 20 min.  The conditions for the MS were essentially the same as 
for the ethoxylates, except that the entire run was done in ESI?.  The cone voltages, 
collision energies, parent and fragment ions for the APECs are presented in Table 3.1.  
Example chromatograms for and NP0-1EC and OP0EC from a primary effluent extract 
are also shown in Figure 3.2.  Concentrations were calculated by the external standard 
method; eight calibration points were prepared in a 50:50 methanol/water mixture, and 
concentrations for each of the three compounds, NP0EC, NP1EC and OP0EC, ranged 
from 6.6 to 660 ng/mL.  Peak integration and quantitation were performed automatically 
using MassLynx 4.0 (Micromass).  In order to assess possible matrix effects on 
quantitation, concentrations obtained by the external standard approach were confirmed 
using the following standard additions procedure: samples were reanalyzed three times, 
the first was the extract per se, the second was the extract after being spiked with 20 ?L 
of a mixture containing the three APECs in amounts such that concentrations in the 
sample (as calculated using the external standard method) were doubled, and the third 
time after adding an additional 20 ?L of the same APECs mixture.  The original APEC
concentrations in the extract were then obtained using the equation: C = intercept/slope, 
where C is the original APEC concentration; intercept and slope refer to the regression 
parameters obtained by plotting area versus added APEC concentration.
61
Quality assurance and quality control.  Instrumental limits of detection (LODs) 
were calculated as the amount of the compound analyzed in the LC-MS-MS producing a 
signal-to-noise ratio of 3.  Theoretical method detection limits (MDLs) were calculated 
normalizing the LOD values with original sample amounts and final volume of the 
extracts.  The limits of quantitation (LOQs) were set at least five times the value of the 
MDLs.  Every batch of up to 11 samples included at least one procedural blank.  During 
the analysis by LC-MS-MS, the calibration curves were injected every 8-12 samples; r
2
values from linear regression analysis were expected to be at least 0.995 and the residual 
values for each calibration point below 10%.
3.4 Results and discussion
Long-chain APEOs in water and sediment.  These methods were originally 
developed for the short-chain APEOs, yielding good recoveries for these compounds 
(Loyo-Rosales et al. 2003).  In contrast, the long-chain APEOs were not recovered as 
efficiently; in fact, the larger the number of EO units, the lower the recovery (Fig. 3.3): 
values were 71, 66, 64, 59, 52, 44, 43, 36, 30, 26, and 21% for NP6EO to NP16EO in 
water and 88, 86, 83, 77, 67, 50, 33, 19, and 11% for NP6EO to NP14EO in sediment 
respectively (NP15EO and NP16EO were not recovered from sediment in this case).  The 
reasons behind these losses were not investigated thoroughly, but they are probably due 
to an increase in the affinity of the compounds for materials such as glass and the 
polymers in the SPE cartridges or the sediment particles.  A previous publication reported
62
Figure 3.3 Recoveries of NP6-16EO from sediment and water.
0
10
20
30
40
50
60
70
80
90
100
678910 11 12 13 14 15 16
EO units
r
e
c
o
v
e
r
y
,
 
%
sediment water
63
good recoveries for the long-chain APEOs from water using graphitized non-porous 
carbon (GCB) SPE cartridges (Houde et al. 2002); for the present work, however, it was 
decided to continue the use of the SDB cartridges because our previous experience with
the short-chain APEOs (Loyo-Rosales et al. 2003) showed that SDB allowed a rapid 
extraction of up to 4 L of water (GCB cartridges tend to show high resistance to water 
flow according to Rodr?guez et al. 2000) and provided reasonably cleaning of interfering 
compounds (SDB was originally selected to extract the short-chain APEOs partially 
because it was hypothesized that its structure would allow a more selective extraction of 
the APEOs than GCB, which is a more general adsorbent).  More importantly, NP was 
adequately recovered from SDB cartridges (Loyo-Rosales et al. 2003), but not from GCB 
(Houde et al. 2002); this constituted a serious disadvantage for the GCB method because 
of the high relevance of NP for the assessment of endocrine disruption potential.  In order 
to offset the low recoveries of the long-chain APEOs from SDB cartridges, an isotope 
dilution approach was adopted, where the [
13
C
6
]-labeled compounds were added before 
the extraction instead of before the LC-MS-MS analysis, thus correcting for recovery 
losses in every sample.
As it was the case for the short-chain APEOs (Loyo-Rosales et al. 2003), the 
mixed-mode column allowed the chromatographic separation of the long-chain APEOs 
(Figure 3.2).  Although peak resolution is not essential for the determination of different 
analytes when using tandem MS (the peaks can be resolved spectrometrically), 
chromatographic separation is advantageous because it decreases competition between 
coeluting compounds, which might result in ion suppression and lower sensitivity (Houde 
64
Table 3.3 Performance parameters for water and sediment methods. LOD: 
instrumental limit of detection; MDL: method detection limit; LOQ: limits of 
quantitation.
water sediment
LOD, MDL, LOQ*, ng/L MDL, LOQ,
compound pg ng/L 4 L 1 L 0.5 L ng/g dry wt ng/g dry wt
NP6EO 1.6 0.06 4 14 28 0.2 14
NP7EO 2.8 0.10 4 17 34 0.4 17
NP8EO 1.0 0.04 5 19 37 0.1 19
NP9EO 1.7 0.06 5 18 37 0.3 18
NP10EO 1.7 0.06 4 17 33 0.3 17
NP11EO 2.3 0.09 3 13 27 0.3 13
NP12EO 2.2 0.08 3 10 20 0.3 10
NP13EO 4.7 0.18 2 7 14 0.7 7
NP14EO 5.9 0.22 2 5 9 0.9 5
NP15EO 8.0 0.30 2 3 5 1.2 6
NP16EO 8.2 0.31 2 2 3 1.2 6
NP0EC 17 0.63 --- --- 20 --- ---
NP1EC 22 0.81 --- --- 20 --- ---
OP0EC 7.3 0.27 --- --- 20 --- ---
* LOQ as a function of sample volume
65
et al. 2002).  Also similarly to the short-chain APEOs (Loyo-Rosales et al. 2003), LOD 
values for the long-chain APEOs were in the order of picograms (Table 3.3).  LODs tend 
to increase with the number of EO units due to the increasing resistance of the 
compounds to fragment.  APEOs are detected in this MS/MS method by forming 
[M+NH
4
]
+
 adducts and analyzing one of their fragments, usually the most abundant in 
order to achieve maximum sensitivity.  However, the stability of these adducts increases 
as the length of the ethoxylate chain increases, resulting in less fragmentation and lower 
sensitivity.  In the case of the short-chain APEOs and NP6-7EO, fragments are produced 
with relative abundance and these were used for quantitation; for NP6-7EO, the 
fragments corresponded to the phenol-ethoxylate [C
6
H
5
-(OCH
2
CH
2
)
x
-OH + H]
+
, where x 
= 6-7 (Loyo-Rosales et al. 2003).  Whereas for the NPEOs with EO chain length > 7, 
fragments are not as abundant as their [M+H]
+
 ions, which were therefore used to 
quantitate these compounds.  This phenomenon also explains the low LOD values for 
NP9-12EO compared to the values for NP6-7EO.  The MDLs for the long-chain APEOs 
in water reported here are 1 to 2 orders of magnitude lower than those reported by Houde 
et al. (2002), mainly because in our approach a larger amount of sample is extracted?up 
to 4 L versus 100 mL.  In contrast, MDL values for the same compounds in sediment are 
comparable to those reported by Ferguson et al. (2001), but our approach offers the added 
selectivity of MS-MS.
The use of a mixture as analytical standard, even if it is well characterized, 
presents several disadvantages.  One of them is illustrated in this case with the method 
performance parameters.  The LOD for NP16EO (Table 3.3) is approximately 8 times 
higher than the value for NP8EO, which implies that the instrument is more sensitive to 
66
the latter.  However, LOQ values are up to 7 times higher for NP8EO than for NP16EO, 
suggesting that the method is indeed less sensitive to NP8EO.  This apparently 
contradictory observation arises because LOQs are based in the lowest point of the 
calibration curve, whose concentration is determined by the least abundant compound in 
the mixture?NP16EO.  Because NP8EO is approximately five times more abundant than 
NP16EO in the mixture, its concentration in the first calibration standard is relatively 
high, resulting in a high LOQ.
It was reported earlier (Loyo-Rosales et al. 2003) that the LC column used for this 
work is a source of some of the analytes, especially NP.  This contamination, combined 
with the use of IDMS quantitation, resulted in the overestimation of the NP contents of 
samples with low concentrations or none of this compound present, such as blanks.  This 
situation arises when a NP-free blank spiked with [
13
C
6
]NP is injected, the instrument 
will still detect NP because of the LC column contribution; and, because of extraction 
losses, the amount of [
13
C
6
]NP in the same sample will be lower than the amount in the 
calibration standards.  This results in an artificially-high response ratio, which is 
interpreted by the quantitation software as a relatively high NP concentration.  In order to 
avoid this, if the area counts for NP in a particular sample were less than twice the 
average of the area counts for NP from at least five blank injections, NP was considered 
to be absent from that sample.
In addition to the NPEOs, the long-chain OPEOs were also monitored in the LC-
MS-MS analysis, but they were not quantified due to the lack of appropriate analytical 
standards.  Because the chemical structure of the OPEOs is very similar to the 
corresponding NPEOs, they could be expected to show similar responses in the MS 
67
system.  Therefore, it was attempted to use NPEO calibration curves to quantitate the 
homologous OPEOs.  The results obtained, however, do not support the initial 
hypothesis.  After analyzing a long-chain OPEO mixture and plotting ion intensities
versus ethoxylate chain length, the expected normal distribution centered around OP9EO 
was observed.  But when the NPEO calibration curves were used to estimate OPEO
concentrations, this pattern was lost completely.
Ethoxylates in particulate and sorption to filters.  For the extraction of the 
ethoxylates from the filters using Soxhlet, three different solvents were investigated: 
acetonitrile, hexane:acetone 50:50, and methanol.  Although the ethoxylates were better 
extracted with the hexane:acetone mixture, methanol was chosen because it improved NP 
and OP recovery, which was very poor with the mixture.  In contrast to the sediments, 
where recoveries for the long-chain APEOs were low, the recovery of the NP0-16EOs 
from spiked filters (no matrix present) ranged from 73 to 100% (RSD from 9 to 19%, for 
n=4), and it improved with increasing number of EO units.  However, when particulate 
was present in the filters, recovery values increased by a factor of ~2 due to matrix-
induced ionization enhancement; an effect that was previously documented by Ferguson 
et al. (2001) in sediment extracts.  As in their case, isotope dilution was used here to 
correct for these effects.
The experiment described as sorption to filters in the experimental section was 
conducted to evaluate the sorption of NPEOs by filters in the absence of particulate.  The 
results for this experiment are presented in Table 3.4, and they suggest that a relatively 
small fraction of the NPEOs originally present in the water indeed remains sorbed to a 
clean filter.  The sorbed amount increased with molecular weight, from 0% 
68
Table 3.4 Fraction of the NP0-16EOs in water sorbed to filters.
compound
total 
spiked
recovered in 
water
recovered in 
filter
total 
recovered
NP 1.00 0.86 0.00 0.86
NP1EO 1.00 0.92 0.00 0.92
NP2EO 1.00 0.89 0.03 0.92
NP3EO 1.00 0.89 0.04 0.93
NP4EO 1.00 0.85 0.06 0.91
NP5EO 1.00 0.84 0.07 0.91
NP6EO 1.00 0.73 0.07 0.79
NP7EO 1.00 0.69 0.08 0.77
NP8EO 1.00 0.55 0.08 0.63
NP9EO 1.00 0.47 0.08 0.55
NP10EO 1.00 0.44 0.08 0.52
NP11EO 1.00 0.42 0.08 0.49
NP12EO 1.00 0.41 0.08 0.49
NP13EO 1.00 0.45 0.08 0.53
NP14EO 1.00 0.45 0.09 0.54
NP15EO 1.00 0.51 0.09 0.59
NP16EO 1.00 0.60 0.08 0.68
69
for NP and NP1EO to 9% for the long-chain APEOs.  Even though the fraction of the 
analytes retained by the filters is relatively small, this could result in an overestimation of 
the analytes? concentrations in particulate matter; especially in the case of water samples 
with relatively low concentrations of suspended solids, as discussed in the method 
application section.
Carboxylates in water.  Although it would be ideal to obtain all the compounds 
of interest in a single extraction, the SPE method used for the ethoxylates did not perform 
well with the carboxylates, presumably because acidification of the water prevents the 
ENV+ solid-phase from retaining the analytes, as discussed previously (Loyo-Rosales et 
al. 2003).  Therefore, liquid-liquid extraction with DCM was used.  Recoveries from 
spiked DI water were 93, 93 and 94% for NP0EC, NP1EC and OP0EC respectively (RSD 
= 16, 19, 15%; for n = 4).  LODs and LOQ for these compounds are presented in Table 
3.3.  These compounds tend to have higher LODs than the ethoxylates, suggesting that
the LC-MS-MS is less sensitive to them.  The MDLs for NP0EC and NP1EC are 
approximately one order of magnitude lower than those reported by Houde et al. (2002); 
the main cause for this is most probably the larger amount of sample analyzed in our 
approach.  The ions, fragments, and MS conditions used for the carboxylates 
identification are listed in Table 3.1.  In this case, the parent ions correspond to the [M-
H]
?
 quasi-molecular ions, and the fragments used for MS/MS detection correspond to the 
octyl- or nonyl-phenolate, which were abundantly produced.  Houde et al. (2002)
reported the same fragmentation pattern for NP0EC and NP1EC.  Comparison of the 
concentrations obtained using external standard quantitation to those using standard 
additions indicated little or no matrix effects on APEC quantitation in any of the samples, 
70
either raw wastewater, the different treatment effluents, or in the Back River sample.  For 
NP0EC and NP1EC concentrations obtained by both methods were virtually identical, 
whereas for OP0EC, concentrations calculated with external standard quantitation were 
consistently around 20% higher than the concentrations obtained by standard additions.  
These results suggest a slight matrix enhancement effect for OP0EC.
Method application.  The methods described above were used to analyze 
wastewater and surface water from the WWTP and the Back River, MD, respectively.  
Results for the plant are presented in Figures 3.4 to 3.6.  Figure 3.4A shows the 
concentrations of dissolved NP0-16EOs along the different treatment stages.  Total 
dissolved NP0-16EO concentration was reduced by approximately 99%, from 390 ?g/L 
in the raw wastewater to 4 ?g/L in the final effluent.  Moreover, the relative composition 
of the homologue mixture was enriched in the short-chain APEOs as the treatment 
progressed.  Such phenomena are in agreement with previous observations that 
degradation of the APEOs proceeds by a shortening of the ethoxylate chain (Ahel et al. 
1994a), which results in the formation of the short-chain APEOs.  As a consequence, 
removal of NP, NP1EO and NP2EO at 85% was lower than total NPEO removal.
Due to their affinity for organic matter, the APEOs tend to accumulate in 
suspended solids.  The shorter the ethoxylate chain, the more hydrophobic the compound 
(Ahel and Giger 1993).  Therefore, the short-chain APEOs have a greater affinity for 
particulate matter, as Figure 3.4B exemplifies.  The concentration profile of the 
homologues in the solid phase is similar to the profile of the dissolved compounds, 
except that the short-chain APEOs were present in higher proportions.  In fact, in raw 
wastewater and the primary effluent, more than 60% of the NP, NP1EO and NP2EO 
71
0
5
10
15
20
25
30
35
40
45
50
Influent Prim effluent Sec effluent Tert effluent Final effluent
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14 NP15 NP16
A
390
4
19
130
220
0
100
200
300
400
500
600
700
800
900
Influent Prim effluent Sec effluent Tert effluent Final effluent
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
g
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14 NP15 NP16
B
3100
540
330
2500
2400
Figure 3.4 Nonylphenol ethoxylate concentrations in a Mid-Atlantic region WWTP 
in (A) the aqueous phase, and (B) particulate matter.  The numbers above the bars 
indicate the total concentration (?g/L in the aqueous phase and ?g/g in the 
particulate) of NP0-16EOs in each treatment stage.
72
occurred in the particulate phase.  This situation, combined to a solids removal during the 
process of more than 99%, was reflected in a 93% removal of the three compounds when 
considering both the dissolved and solid phases, in contrast with 85% when considering 
only the dissolved phase.
As mentioned in the last section, the concentration of the APEOs in particulate 
might be overestimated in cases where the amount of solids in water is relatively low.  In 
the case of these samples, the presence of some of the individual NPEOs in the extracts 
from tertiary and final effluent solids is entirely due to direct sorption to the filters.  The 
total concentrations, however, are not as severely affected.  The corrected values for total 
concentrations of NPEOs in solids from raw wastewater, primary, secondary, tertiary, 
and final effluents are 2500, 2000, 2100, 240, and 430 ?g/g respectively.  These values 
represent 74-86% of the non-corrected numbers.
The OPEOs behavior was similar to the NPEOs.  Concentrations of the dissolved
OP0-5EOs are presented in Figure 3.5.  The individual OP homologues were present in 
the raw wastewater in concentrations that were 10 to 30 times lower than the respective 
NPEOs, reflecting volume differences in the use of these surfactants, which is dominated 
by the NPEOs (Ferguson et al. 2000).  The total concentration of OPEOs almost doubled 
in the secondary effluent with respect to the raw wastewater, presumably as a product of 
higher ethoxylate degradation.  As mentioned above, OPEOs with EO chain lengths >5 
were not quantified due to the lack of appropriate standards; they were monitored during 
the LC-MS-MS analysis of these samples, however, and they disappeared gradually as 
the wastewater treatment progressed much like the NPEOs (data not shown). 
73
0
0.5
1
1.5
2
2.5
Influent Prim effluent Sec effluent Tert effluent Final effluent
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
OP OP1 OP2 OP3 OP4 OP5
0.2
0.7
5.8
3.6
3.1
Figure 3.5 Octylphenol ethoxylate concentrations in a Mid-Atlantic region WWTP.  
The numbers above the bars indicate the total concentration (?g/L) of OP0-5EOs in 
each treatment stage.
74
In contrast to the APEOs, the concentration of the APECs increased along the treatment 
as can be seen in Figure 3.6.  In the final effluent, the amount of APECs present was 
approximately six times higher than the amount of the total ethoxylates, representing 
85% of the compounds measured.  Additionally, other carboxylated derivatives might be 
present that may increase the fraction of metabolites in the final effluent.  Examples of 
these derivatives are long-chain APECs, and APEOs and APECs where the alkyl chain is 
also carboxylated, all of which have been previously reported in WWTP effluent and 
river water (Ding and Tzing 1998, Di Corcia et al. 1998).
NP was found in the sample from Back River in a concentration of 0.49 ?g/L, 
whereas OP and the APEOs were not detected.  In contrast, APECs concentrations were 
1.2, 3.6, and 0.056 ?g/L for NP0EC, NP1EC, and OP0EC respectively.  Therefore, 90% 
of the mass of APEO-related compounds detected was in the form of transformation 
products, illustrating the importance of this type of compounds in the study of the 
environmental fate of organic chemicals.  It also suggests that APECs might be better 
indicators for the presence of WWTP effluent in water than the APEOs.
Results for NPEOs in sediments from the Baltimore Harbor area are shown in 
Figure 3.7.  As observed previously (Loyo-Rosales et al. 2003), NP tends to be the most 
abundant compound due at least in part to its higher affinity to solids.  The sites with the 
highest concentrations of NPEOs were in Bear Creek and Back River, presumably 
because of the presence of WWTP effluent in both bodies of water.  These were followed 
by the sites in the Inner Harbor and Gwynn?s Falls; whereas White Rock showed the 
lowest amounts of NPEOs, as expected from its location further away from wastewater 
effluent or other sources for these compounds.  It is worthwhile noticing that the samples 
75
0
5
10
15
20
25
30
Influent Prim effluent Sec effluent Tert effluent Final effluent
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP0EC NP1EC OP0EC
1.5
24
43
25
2.8
Figure 3.6 Alkylphenol carboxylate concentrations in a Mid-Atlantic region WWTP 
(values obtained using external standard quantitation).  The numbers above the 
bars indicate the total concentration (?g/L) of NP0EC, NP1EC, and OP0EC in each 
treatment stage.
76
0
1
2
3
4
5
6
7
8
WR GF BC1 BC2 BR1 BR2 IH
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
g
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14
1.8 4.6
2.9
6.4
116.1
8.4
Figure 3.7 Nonylphenol ethoxylate concentrations in sediments from the Baltimore 
Harbor area.  The different sites are identified in Figure 3.1.  The numbers above 
the bars indicate the total concentration (?g/g) of NP0-14EOs in each site.
77
from Bear Creek and Back River contained a higher proportion of the short-chain APEOs 
than the rest of the samples; and that the concentrations of the long-chain APEOs in 
samples from the Inner Harbor and Gwynn?s Falls seemed to increase with EO number, 
instead of leveling off and decrease, resembling the concentration profile of the most 
commonly used APEO mixtures.  These observations are consistent with the presence of 
a treated effluent in Bear Creek and Back River, and suggest a non-treated source of 
NPEOs in the rest of the sites.
OP and the OPEOs were also detected in these sediment samples, albeit in lower 
concentrations?6 to 17 times less concentrated than their respective NPEOs.  Their 
relative abundance paralleled that of the NPEOs, with OP being the most abundant 
compound, and Bear Creek and Back River the sites with the highest concentrations.
3.5 Conclusions
The methods presented in this article allow for the simultaneous extraction and 
quantitation of NP0-16EOs and OP0-5EOs present at environmentally-relevant 
concentrations in the dissolved and particulate fractions of WWTP effluents and surface 
water, as well as in sediments.  The use of IDMS for these determinations compensates 
for possible matrix effects and for analyte losses during sample extraction and 
processing.  Unfortunately, isotopic-labeled analogues for the APECs are not available, 
precluding the use of IDMS for the quantitation of these compounds in water.  Together 
with the lack of individual long-chain APEOs, this lack of appropriate standards 
78
constitutes one of the main difficulties in conducting studies on the environmental fate of 
compounds that are subject to biotransformation.  This work also illustrates the 
importance of measuring transformation products as well as the parent compounds in this 
kind of studies, not only because the former may be present in larger amounts, but also 
because their relative concentrations might be used as a tool for effluent source 
identification.  A more detailed discussion on the fate of the APEOs in WWTP and 
receiving waters will be presented in a future report.
79
CHAPTER 4 ? APEO FATE IN WASTEWATER TREATMENT PLANTS
This chapter to be submitted to Environmental Science and Technology as ?The fate of 
octyl- and nonylphenol ethoxylates and some of their carboxylated derivatives in three 
American wastewater treatment plants? by Loyo-Rosales, J. E., Rice, C. P., and Torrents, 
A.
4.1 Abstract
The fate of a group of nonylphenol and octylphenol ethoxylates and several of 
their carboxylated derivatives was studied in three American wastewater treatment plants 
(WWTPs), two of which included advanced treatment.  Compounds analyzed 
quantitatively included nonylphenol, octylphenol, nonlyphenol ethoxylates with 
ethoxylate chain lengths of 1 to 16, octylphenol ethoxylates 1 to 5, nonyl- and 
octylphenoxyacetic acids, and nonylphenoxyethoxyacetic acid; whereas the 
nonylphenoxydi- and triethoxyacetic acids were measured qualitatively.  In spite of being 
located in three different metropolitan areas, results showed that influent wastewater to 
the WWTPs had similar concentrations of the alkylphenolic compounds; whereas effluent 
concentrations were only similar when samples from the same season (summer or winter) 
were compared.  Ethoxylate and carboxylate concentrations in winter samples were on 
average seven and four times higher than in summer, with carboxylate accumulation 
80
showing more variability than APEO depletion.  Sorption to particulate was 
approximately 1.6 times higher for nonylphenolic compounds than for their octylphenolic 
counterparts, in agreement with their difference in K
ow
 values.  Both effluent 
concentrations and APEO removal rates?the latter averaging 99% in summer and 94% 
in winter for the NPEOs?were strongly correlated to water temperature, and no 
correlation was found with suspended solids or organic carbon removal.  Additionally, a 
small survey of urban sewers suggested that household products could still constitute an 
important source of the APEOs reaching WWTPs.
4.2 Introduction
The study of organic pollutants in wastewater treatment plants (WWTPs) and 
receiving waters is relevant to water quality scientists because WWTPs funnel residues 
from large areas and the pollutants are subject to a series of biotransformations that might 
not be the same or might not take place to the same extent in the environment.  In the 
particular case of the alkylphenol ethoxylates (APEOs), a widely used family of 
surfactants composed mainly of nonylphenol and octylphenol ethoxylates (NPEOs and 
OPEOs, respectively), they undergo a rapid transformation in WWTPs into short-chain 
APEOs, the parent alkylphenols (APs, octylphenol, OP, and nonylphenol, NP), and 
carboxylated derivatives, which include the alkylphenoxyethoxy carboxylates (APECs), 
and the carboxyalkylphenoxyethoxy carboxylates (Johnson and Sumpter 2001).  Several 
of the transformation products are of toxicological concern due to their estrogenic 
81
properties (Servos 1999), and have been related to endocrine-disruption effects in biota 
exposed to WWTP effluents (e.g. Lye et al. 1999, Sol? et al. 2000).
Most studies on the fate of the APEOs in WWTPs focus exclusively on influent 
and effluent concentrations of the more toxicologically relevant transformation 
products?NP, the low molecular weight ethoxymers (nonylphenol monoethoxylate, 
NP1EO, and nonylphenol diethoxylate, NP2EO), and the carboxylated derivatives 
nonylphenoxyacetic acid (NP0EC), and nonylphenoxyethoxyacetic acid (NP1EC)?
while ignoring the long-chain parent compounds.  It is the latter, however, that reach the 
WWTPs in the influent wastewater, and the analysis of as many of the compounds as 
possible is necessary to conduct environmental fate studies.  The most comprehensive 
study the fate of the APEOs in WWTPs  was performed by Ahel et al. (1994a) in Swiss 
WWTPs with activated-sludge secondary treatment.  In the past decade, however, it has
been argued that American WWTPs are more effective than their European counterparts 
in removing APEOs (Renner 1997), but studies conducted in North America?
specifically in the United States?have not been as comprehensive as those in Europe.  In 
addition, there is a tendency in the United States to add tertiary treatment to WWTPs for 
nutrient removal, especially in sensitive areas such as the Chesapeake Bay.  Few studies 
have addressed APEO degradation in plants with this level of treatment, or the effects of 
temperature on removal efficiency, and those that have been carried out include only a 
few of the APEOs; e.g. Bennie et al. 1998. 
This work describes the fate of a comprehensive set of the APEOs in three 
American WWTPs, two of which with advanced treatment. Compounds analyzed 
quantitatively included the short- (NP1-5EO) and long-chain (NP6-16EO) NPEOs, the 
82
short-chain OPEOs (OP1-5EO), NP, OP, and their carboxylated derivatives NP0EC, 
NP1EC, and OP0EC, whereas the behavior of NP2EC and NP3EC was assessed
qualitatively.  Seasonal changes in the concentrations and removal efficiency of these 
compounds are also discussed, and mass balances to evaluate degradation of the APEOs 
in the plants are presented.  Finally, a short assessment of the APEO sources to one of the 
WWTPs is included.
4.3 Experimental Section
Standards and reagents.  Standards for NP (Schenectady International, 
Schenectady, NY, USA; purity 95%), OP (Aldrich, Milwaukee, WI, USA; 97%), 
NP2EO (R&D product from Aldrich), NP0EC (R&D product, Huntsman Chemicals, 
Austin, TX, USA), NP1EC (Aldrich; 90%), and OP0EC (R&D product, Huntsman) were 
used as provided.  NP1EO, NP3EO, NP4EO, NP5EO, and the OP1-5EOs were purified 
in the laboratory from commercial mixtures as described previously (Datta et al 2002, 
Loyo-Rosales et al 2003).  A commercial mixture (Surfonic N-95, Schenectady 
International; >97%; characterized by Huntsman) was used as standard for the NP6-
16EOs.  
13
C
6
-substituted standards for IDMS were synthesized by Ferguson et al (2001), 
except for 
13
C
6
-NP (Cambridge Isotope Laboratories Inc., Andover, MA, USA).  Solvents 
were high purity, pesticide grade from Burdick & Jackson (Honeywell International Inc., 
Muskegon, MI, USA).  Carbon-free deionized water (DI water) was obtained from a 
NANOpure system (Barnstead International, Dubuque, IA, USA).
83
Site descriptions and sampling procedures.  Grab wastewater and sludge 
samples were obtained in 2004 and 2005 from three large American WWTPs.  Table 4.1
lists the main characteristics of the plants and sampling events.  For the purposes of this 
study, a sample was labeled as a summer sample if, when collected, its temperature was 
above 20?C, and as a winter sample when its temperature was below 15?C.  All the 
samples were collected in previously-baked amber glass containers, and transported in ice 
to the laboratory.  Sludge and solids were frozen at -20?C upon arrival, whereas the 
wastewater samples were filtered as described in Chapter 2; the filters were dried and 
weighed to calculate the amount of suspended solids in the samples, and extracted to 
evaluate their APEO contribution.  Additionally, two sampling events were conducted in 
sewers from the Chicago, IL, area to gain further insight into the origin of the APEOs.  
The first sampling event took place in March 2005 and included 5 grab raw sewage 
samples, two originating mainly from residential zones, one from a commercial area, one 
from an industrial area, and one was the influent to Calumet WWTP.  The second 
sampling event was done in August 2005 and consisted of 5 24-h composite raw sewage 
samples collected in residential areas of Chicago, IL. 
Extraction and analysis.  NP0-16EOs, OP0-5EOs, NP0-1EC, and OP0EC were 
analyzed qualitatively, and NP2-3EC quantitatively, in the dissolved fraction of the 
wastewater samples.  Particulate matter and sludge were analyzed for the APEOs 
mentioned above, but not the APECs.  Extraction methods for the different matrices and 
analytes were published elsewhere (Loyo-Rosales et al. 2003 and Chapter 2), and 
consisted of solid-phase extraction using hyper-cross-linked hydroxylated poly(styrene-
divinylbenzene) copolymer cartridges (Isolute ENV+, 500 mg, 
84
Table 4.1 Wastewater treatment plant characteristics and sampling events.
Name Back River Blue Plains Calumet
Location Baltimore, MD Washington, DC Chicago, IL
Capacity, mgd 180 370 350
Population served, mi 1.3 2.0 1.0
Treatment (after primary) Modified Ludzack-
Ettinger
Activated sludge + 
separate stage
nitrification/denitrification
Activated sludge
Final chlorination yes yes no
Time in biological 
treatment, hrs
a
5-10 ~6 6-14
Sludge retention time, 
days
7-13 Secondary sludge: 1-1.5
Nitrification sludge: 14-20
7-12
Samples obtained
water Raw influent Raw influent Primary influent
Primary effluent Primary effluent Primary effluent
Secondary effluent Secondary effluent Aeration tank effluent
Final effluent Tertiary effluent Final effluent
Final effluent
sludge Primary Primary Primary
Secondary Secondary Prim + secondary
Tertiary
Sampling events Sep 2004 (24.5) July 2004 (25.2) March 2005 (9.3)
b
(average water T, ?C) Oct 2004 (20.5) August 2004 (25.7) August 2005 (23.5)
b
Feb 2005 (13.8) Feb 2005 (14.4)
Mar 2005 (14.0)
b
Mar 2005 (14.0)
b
a
 this time is the hydraulic residence time in activated sludge reactors in Back River and Calumet; but in 
Blue Plains is the total HRT in the secondary, nitrification and denitrification reactors.
b
collected sludge samples
85
6 mL, International Sorbent Technology Ltd., Hengoed, U.K.) for the APEOs in 
wastewater; liquid-liquid extraction with dichloromethane for the APECs; Soxhlet 
extraction with methanol for the APEOs in particulate; and accelerated solvent extraction 
with hexane and acetone, followed by clean-up with a solid-phase extraction procedure 
with amino-propyl silica cartridges (Varian Associates Inc., Harbor City, CA, USA) for 
the sludge samples.  Solvents in all extracts were evaporated and exchanged to 1.5 mL 
methanol/water 50:50 v/v.  These extracts were analyzed by LC-MS-MS using an isotope 
dilution mass spectrometry (IDMS) approach described in detail in Chapter 2 and Loyo-
Rosales et al. 2003.  Briefly, all compounds were separated in a Waters 2690 XE 
separations module (Waters Corp., Milford, MA, USA) with a 4.6 x 150 mm MSpak GF-
310 4D column (Shodex, Shoko Co., Tokyo, Japan) at 60?C.  APs and APEOs were 
analyzed in a single run using 10 mM ammonium acetate in 50:50 v/v methanol/water 
that was gradually changed to 100% methanol; whereas the APECs were separated with 
the same mobile phase in a separate isocratic run.  Tandem mass spectrometry multiple-
reaction monitoring (MRM) analysis was performed on a Quattro LC triple quadrupole 
mass spectrometer (Micromass Ltd., Manchester, UK) in electrospray positive mode 
(ESI+) for the APEOs, and in electrospray negative (ESI?) for the APs and APECs.  
Except for the APECs, which were done by external standard, quantitation was done 
using the 
13
C
6
-substituted compounds as internal standards.  NP2EC and NP3EC were 
monitored qualitatively along the rest of the APECs in the same LC-MS-MS runs using 
the instrumental conditions described in Chapter 2.  Ion transitions monitored were 365.4 
> 219.0 m/z for NP2EC, and 409.4 > 219.0 for NP3EC, whereas cone voltages were 35 
and 40 V respectively, and collision energy 25 eV for both compounds.
86
4.4 Results and discussion
APEO concentrations in WWTPs? influents and effluents.  The three WWTPs 
studied for this work serve three different metropolitan areas in the United States, in 
which the APEOs may be used for varying purposes and in diverse amounts.  However, 
the fact that the plants treat large amounts of wastewater and serve large populations 
(Table 4.1) might have as a consequence wastewater with similar APEO composition, 
regardless of the specific location.  In the present study, total NPEO concentrations 
(dissolved + particulate) in raw wastewater from the different plants averaged 630 
(SD=140) ?g/L in summer and 730 (150) ?g/L in winter (Table 4.2 and Figure 4.1A).  In
other words, the variation in NPEO composition between the different plants was 
relatively low at 23 and 21% (as RSD) in summer and winter respectively, and the total 
NPEO concentration increased only 15% from summer to winter.  However, short-chain
ethoxylate concentrations tended to be higher in summer?almost 50% on a molar basis,
whereas long-chain ethoxylates were higher in winter (40%).  These differences could be 
explained by the transformation of the long-chain ethoxylates into the short-chain
ethoxymers in raw wastewater while in transit to the WWTP, which would occur more 
readily in the summer because of higher temperatures.
In spite of the differences in treatment, total NPEO concentrations in the final 
effluents of the three plants were very similar when comparing samples from the same 
season (Table 4.2 and Figure 4.1B).  However, NPEO concentrations were higher in 
87
Table 4.2 Total average APEO and APEC concentrations (dissolved + particulate) 
in WWTP influents and effluents.  Values are arithmetic means of 5 samples, except 
for the summer influent, where n = 4; values in parentheses correspond to standard 
deviations.
NP0-16EO, ?g/L OP0-5EO, ?g/L APEC*, ?g/L
influent effluent influent effluent influent effluent
summer 630 (140) 3.5 (1.2) 6.4 (1.1) 0.21 (0.22) 1.5 (1.2) 17 (8.9)
winter 730 (150) 27 (7.8) 6.5 (1.2) 1.6 (0.46) 1.6 (0.7) 82 (29)
* APEC includes NP0-1EC and OP0EC.
88
Figure 4.1 Total NPEO concentrations (dissolved + particulate) in raw wastewater 
(A) and WWTPs? final effluents (B) for all sampling events in winter and summer 
and for the three WWTPs, Back River (BR), Blue Plains (BP), and Calumet (Cal).
0
20
40
60
80
100
120
N
P
N
P
1
N
P
2
N
P
3
N
P
4
N
P
5
N
P
6
N
P
7
N
P
8
N
P
9
N
P
1
0
N
P
1
1
N
P
1
2
N
P
1
3
N
P
1
4
N
P
1
5
N
P
1
6
compound
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
BR summer BR summer BP summer BP summer BR winter
BR winter BP winter BP winter Cal winter
A
0
2
4
6
8
10
12
14
N
P
N
P
1
N
P
2
N
P
3
N
P
4
N
P
5
N
P
6
N
P
7
N
P
8
N
P
9
N
P
1
0
N
P
1
1
N
P
1
2
N
P
1
3
N
P
1
4
N
P
1
5
N
P
1
6
compound
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
BR summer BR summer BP summer BP summer Cal summer
BR winter BR winter BP winter BP winter Cal winter
B
89
winter, 27 (7.8) ?g/L, than in summer, 3.5 (1.2) ?g/L, by more than 7 times.  
Furthermore, there was a strong negative correlation between total final NPEO 
concentration and temperature, r = -0.868 (Figure 4.2).  Seasonal differences in 
concentrations were especially pronounced for the long-chain ethoxylates, which were on 
average more than 20 times higher in winter as compared to the short-chain ethoxylates, 
which were only 7 times higher.  This difference might be due to the degradation rate of 
the short-chain EOs being more affected by the temperature changes in these 
observations (correlation NP0-5EO vs temperature r = -0.874) than the de-ethoxylation 
rate of the long-chain NPEOs (correlation NP6-16EO vs temperature r = -0.659), which 
appears to occur rapidly even at low temperatures. It is important to note that there is 
also some degree of correlation between the total NPEO concentration in the effluent to 
the total concentration in the influent (r = 0.543), but it is not as strong as the correlation 
between effluent concentration and temperature.  Furthermore, this correlation decreases 
substantially, r = 0.255, if concentrations are expressed on a molar basis by eliminating 
the effect of the mass loss through long-chain APEO de-ethoxylation; whereas the 
correlation between final concentration and temperature is barely modified, r = -0.861.
The OPEOs behave similarly to the NPEOs: total OP0-5EO concentrations in the 
influents barely change from winter to summer, whereas concentrations in the final 
effluents are more than 7 times higher in winter (see Table 4.2).  There is also a strong 
negative correlation between temperature and final effluent concentration, r = -0.880.
Compared to the NP0-5EOs?, total OP0-5EO concentrations in the influents are less 
variable and an average of 35 times lower in summer and 26 in winter; although in the 
final effluents, these ratios decrease to 16 and 15 respectively.  This difference suggests 
90
0
5
10
15
20
25
30
35
40
45
5 10 15 20 25 30
temperature, ?C
N
P
0
-
1
6
E
O
 
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
Figure 4.2 Total NP0-16EO concentrations (dissolved + particulate) in final effluents 
from three WWTPs as a function of temperature; r = -0.868.
91
that the NPEOs are preferentially eliminated during wastewater treatment in comparison 
to the OPEOs.  The elimination mechanism in question is most likely abiotic, rather than 
microbial degradation, because the difference in the ratios above can be explained by the 
higher affinity of the NPEOs for the WWTP solids as a consequence of their higher log 
K
ow
 values (see discussion in the particulate section), resulting in a higher proportion of 
them being eliminated with the WWTP solids.
Carboxylates also show a similar but less pronounced behavior.  There is no 
change from winter to summer in total average APEC concentrations in the influents 
(Table 4.2).  In contrast, total APEC concentrations in the effluents are 4 times higher in 
winter than in summer, and there is a negative correlation between temperature and final 
concentration, r = -0.745.  It is interesting to note, however, that in spite of the increase in 
short- versus long-chain EO concentrations in raw wastewater in the summer?
presumably due to higher degradation rates as discussed above, and assuming that 
APEOs are transformed to APECs (Montgomery-Brown et al. 2003), APEC 
concentrations did not increase accordingly.  This suggests that either the transformation 
rate of APEOs to APECs is affected to a different extent by temperature than the 
degradation rate of APECs (this one increasing more in summer), or that APEO 
carboxylation does not occur as readily in sewers as it does in WWTPs possibly due to 
lower oxygen availability.
Compared to previous reports, influent concentrations of the APEOs presented 
here are within the ranges observed in different parts of the world (Ahel et al. 1994a, 
Bennie 1999); effluent concentrations, also within observed ranges, tend to be in the 
lower end (Ahel et al. 1994a, Bennie 1999, Ying et al. 2002, Montgomery-Brown and 
92
Reinhard 2003).  This is not surprising because the plants studied here have advanced 
treatment, whereas most of the plants reported in the literature are limited to secondary or 
even primary treatment.  In contrast, the APECs tend to be in lower concentrations in the 
influents reported here (Ahel et al. 1994a, Bennie 1999), but the concentrations in the 
final effluents are in the same order of magnitude as those observed elsewhere (Ahel et 
al. 1994a, Bennie 1999, Montgomery-Brown and Reinhard 2003).
APEO behavior in WWTPs.  In general terms, the behavior of the APEOs was 
very similar in the three WWTPs studied.  There was a drastic decrease in total APEO 
concentrations (up to >99%, Table 4.3), and a shift in the ethoxylate distribution towards 
the low-ethoxylates and APs as observed previously (Chapter 2).  In contrast, total APEC 
concentrations increased anywhere from 8 to 120 times from influent to effluent.  
Accumulation of the APECs results from the balance between formation from the APEOs 
and degradation of the APECs themselves.  Although APEC accumulation showed a 
larger variation than APEO degradation between WWTPs, there were still statistically 
significant differences (t-test,  = 0.05) between summer and winter.  APEC 
concentrations increased by an average of 66 (SD = 40) times in winter compared to 17 
(SD = 8) times in summer.  These observations suggest that degradation rates for the 
APECs are lower than the de-ethoxylation rates for the APEOs (because there is APEC 
accumulation), and more affected by temperature than the APEO de-ethoxylation rates 
(there is further APEC accumulation in winter than in summer).
Observing more closely the behavior of the NPEOs at the different stages of 
wastewater treatment, it is clear that most of the transformation occurred in the biological 
93
Table 4.3 APEO removal from wastewater in WWTPs. % removal = 100 -
([effluent]*100/[influent])). Concentrations units were ?M.
total NP0-16EO removal, %
WWTP summer winter
Back River 99.3 99.0 92.9 92.6
Blue Plains 98.7 99.4 94.8 95.4
Calumet 99.1 93.1
mean 99.1 93.7
SD 0.28 1.2
total NP0-5EO removal, %
WWTP summer winter
Back River 98.9 97.9 84.6 77.7
Blue Plains 96.4 99.1 87.7 85.1
Calumet 98.8 75.5
mean 98.2 82.1
SD 1.1 5.2
total OP0-5EO removal, %
WWTP summer winter
Back River 92.1 97.0 66.1 64.9
Blue Plains 94.1 99.9 81.5 84.0
Calumet 99.7 61.1
mean 96.6 71.5
SD 3.4 10.5
94
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Influent Prim eff Sec eff Tert eff Final eff
l
o
g
(
N
P
0
-
3
E
O
+
N
P
0
-
1
E
C
/
N
P
4
-
1
6
E
O
)
summer summer winter winter
Figure 4.3 Logged ratios of degradation compounds? concentrations (NP0-3EO + 
NP0-1EC) to parents products? concentrations (NP4-16EO) for (A) Blue Plains, (B) 
Back River and (C) Calumet in the different treatment stages.  Influent refers to 
raw wastewater; prim eff, sec eff, tert eff, and fin eff refer to the primary, 
secondary, tertiary and final effluents respectively.
A
95
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
Influent Prim eff Sec eff Final eff
l
o
g
(
N
P
0
-
3
E
O
+
N
P
0
-
1
E
C
/
N
P
4
-
1
6
E
O
)
summer summer winter winter
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
Influent Prim eff Sec eff Final eff
l
o
g
(
N
P
0
-
3
E
O
+
N
P
0
-
1
E
C
/
N
P
4
-
1
6
E
O
)
winter summer
Figure 4.3 (cont.)
C
B
96
treatment stage (secondary treatment), as can be observed in Figure 4.3, and as it has 
been extensively reported in the literature; e.g. Montgomery-Brown and Reinhard 2003.
Figure 4.3 shows the logged ratios of the concentrations of ?degradation? (NP0-
3EO + NP0-1EC, ?M) to ?parent? compounds (NP4-16EO, ?M) to  for all the sampling 
events.  In most cases, it was only after secondary treatment when the degradation 
products concentration surpassed the parent compounds?; i.e. the value of the logged ratio 
was > 0.  In a few cases?all in the summer?the long-chain EOs degradation occurred 
so early that the degradation products? concentrations were higher than the parents? even 
in the influent.  It is also interesting to note that, regardless of differences in treatment, 
carboxylation seems to have occurred earlier than APEO degradation, supporting the 
oxidative pathway.  This is best observed in Blue Plains, where most of the APECs were 
formed in the secondary treatment, whereas most of the APEOs degradation occurred in 
the nitrification/denitrification stage (see Figure 4.4).  The figure shows that 
concentrations of the NP4-16EO decreased in the secondary treatment, but at the same 
time the NP0-3EOs increased by almost 50%.  NP0-1EC increased 55 times in this stage, 
presumably due to the intense aeration.  Finally, degradation of NP0-3EO and even the 
NPECs occurred in the tertiary treatment.  Although nitrifying conditions have been 
correlated to improved APEO degradation (Ahel et al. 1994a), sludge retention times 
(SRTs) could also be an important factor (Environment Canada 2001), especially in Blue 
Plains, where the SRT in the nitrification step is longer than the SRT in the secondary 
treatment (Table 4.1).  Interestingly, in Back River APEC accumulation in the primary 
treatment could be as important as in the secondary; APEO degradation occurred mainly 
in the secondary treatment.  Note that secondary treatment in Back River is a modified 
97
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Influent Prim eff Sec eff Tert eff Final eff
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
NP0-16EO NP0-3EO NP4-16EO NP0-1EC
Figure 4.4 Changes in total NP0-16EO, NP0-3EO, NP4-16EO, and NP0-1EC 
concentrations in the different treatment stages in Blue Plains WWTP (February 
2005) .  Influent refers to raw wastewater; prim eff, sec eff, tert eff, and fin eff refer 
to the primary, secondary, tertiary and final effluents respectively.
98
Ludzack-Ettinger, which includes an anoxic step, as opposed to the purely aerobic 
activated sludge process in Blue Plains.
Two different degradation pathways have been proposed for the long-chain 
APEOs (Montgomery-Brown and Reinhard 2003); the original model involves their 
transformation to the short-chain APEOs via a stepwise ether hydrolysis, which are then 
converted to the APs under anaerobic conditions or oxidized to the homologous short-
chain carboxylates under aerobic conditions.  In the alternative pathway?limited to 
aerobic environments, terminal alcohol oxidation of the long-chain APEOs occurs first, 
followed by the shortening of the EO chain.  Because all three WWTPs in this study 
include aerobic treatment, long-chain APEO degradation might occur preferably through 
this pathway.  In order to assess this hypothesis, NP2EC and NP3EC were monitored in 
all samples in spite of the lack of analytical standards.  In most instances in Blue Plains 
WWTP, higher concentrations of NP2EC and NP3EC were observed in the tertiary 
effluent than in the secondary effluent, even when NP0EC and NP1EC concentrations 
decreased considerably, as exemplified in Figure 4.5A.  In Back River, the behavior of 
NP2EC and NP3EC tended to mirror that of NP0EC and NP1EC, as shown in Figure 
4.5B.  In both sites, however, the concentrations of NP2EC and NP3EC kept increasing 
even when NP3EO and NP4EO decreased.  All this suggests that either both NP2EC and 
NP3EC were being formed from long-chain APECs that were not monitored, or that they 
were being formed from NP3EO and NP4EO.  In any case, these observations support the 
alternative degradation pathway described in Montomery-Brown and Reinhard (2003); 
although they do not disprove that both pathways could take place during treatment, 
either simultaneously or in different stages, as proposed by Di Corcia et al. (1994).  
99
Figure 4.5 Changes in total NP0-3EC concentrations in the different treatment 
stages in (A) Blue Plains and (B) Back River WWTP (February 2005).
0.0E+00
1.0E+05
2.0E+05
3.0E+05
4.0E+05
5.0E+05
6.0E+05
Influent Prim eff Sec eff Tert eff Final eff
a
r
e
a
,
 
a
r
e
a
 
c
o
u
n
t
s
NP0EC NP1EC NP2EC NP3EC
A
0.0E+00
1.0E+05
2.0E+05
3.0E+05
4.0E+05
5.0E+05
6.0E+05
7.0E+05
Influent Prim eff Sec eff Final eff
a
r
e
a
,
 
a
r
e
a
 
c
o
u
n
t
s
NP0EC NP1EC NP2EC NP3EC
B
100
Particulate.  The fraction of the APEOs associated to particulate varied widely 
and there was no apparent correlation between the values observed and temperature.  In
most cases, and as it would be expected from its log K
ow
(4.48, Ahel and Giger 1993), NP 
was the compound that partitioned the most to the particulate.  In the influents, 43 to 88% 
of the NP was bound to the particulate; this fraction increased slightly to 50-94% in the 
primary effluents.  This can be attributed to the longer time NP had to partition to the 
particulate and the production of additional NP from NPEOs degradation.  The 
percentage of the long-chain ethoxylates associated to the particulate was also very 
variable, but tended to decrease with molecular weight and could be as low as 10%, 
although in some cases NP14EO and higher showed an increasing trend.  As the 
treatment progressed, the percentage of APEOs associated to the particulate decreased, 
partly because of the lower contact time and the lower concentrations of both solids and 
APEOs.  Therefore, the percentages in the final effluents range from 2 to 20% for NP.  
Another difference in the final effluents is that NP1EO tends to be present in a higher 
percentage, 10-42% in this case, reflecting the higher concentrations of NP1EO vs NP in 
the dissolved phase.  The OPEOs showed a similar behavior to the NPEOs, but the latter 
tend to be in higher proportions in the particulate.  On average, the fraction of a given 
short-chain NPEO in particulate was 1.6 (SD = 0.8, n = 156) times larger than the 
respective OPEO.  This value is close to that (1.85) predicted from the difference in their 
log K
ow
 values, which are an average of 0.268 units lower for the short-chain OPEOs 
(Ying et al. 2002).
APEO removal versus WWTPs efficiency.  As a consequence of similar 
influent and effluent concentrations, APEO removal is equally comparable across the 
101
three studied WWTPs in the same season, and it is significantly different in summer than 
in winter.  Table 4.3 shows the removal percentage for NPEOs and OPEOs calculated 
using molar concentrations because mass-based concentrations yield artificially high 
removal rates due to the loss of mass from the long-chain APEOs? de-ethoxylation, which 
cannot be considered an actual removal, especially when the short-chain homologues are 
more relevant in toxicological terms.  These removal rates are significantly higher than 
those reported by Ahel et al. (1994a) for 11 Swiss WWTPs with activated-sludge 
treatment, 59% (SD = 18).  The relatively low standard deviation values in Table 4.3 (0.3
to 15% expressed as RSD) attest to the low variability of APEO removal efficiency 
across the three WWTPs.  The mean values in Table 4.3 were compared using t-tests at a 
0.05 level of significance, and all three groups of compounds showed statistically 
significant higher removal in summer than in winter.  It is interesting to note that average 
removal for the OP0-5EOs tends to be lower than removal for the NP0-5EOs.  Although 
this difference is not statistically significant in the summer, it is significant in winter 
when biodegradation rates are lower, and might be partially due to the NPEOs having 
higher affinity than the OPEOs for the solids, although proper assessment of this 
phenomenon would require the quantitation of the long-chain OPEOs, which was not 
performed for this work.
If overall removal of APEO-related compounds is calculated using APEO and 
APEC concentrations to account for APEC formation, the removal percentages behave 
differently according to the season.  In summer, due to the relatively low accumulation of 
APECs, removal percentages decrease only slightly from the values in Table 4.3 to an
average of 95% (SD = 2.8%) for the NP0-16EOs, 90 (7.0) for the NP0-5EOs, and 92 
102
(6.4) for the OP0-5EOs. In winter, however, APEC accumulation was higher and resulted 
in a larger drop in the removal rates to 75 (8.7) for the NP0-16EOs, 25 (31) for the NP0-
5EOs, 0.5 (53) for the OP0-5EOs.  Note that the high standard deviation values for the 
short-chain APEOs removal results partially from the higher variability in the APEC 
production, but also from negative removal rates, implying a net production of 
alkylphenolic compounds in the plants, but are only a consequence of excluding the long-
chain APEOs from the calculations.
It has been suggested that APEO removal from WWTPs is correlated to WWTP 
performance (e.g. Fensterheim 2001), specifically to organic carbon removal.  The data 
obtained for this study do not support this suggestion; no correlation was found between 
APEO removal and total suspended solids, total organic carbon or dissolved organic 
carbon removal.  In fact, these three parameters remained relatively unchanged from 
summer to winter, which should be expected because the WWTPs control them.  As 
expected from the discussion on concentrations in influents and effluents, APEO removal 
is strongly correlated with temperature (Figure 4.6), with r = 0.932 for NP0-16EO 
removal, 0.937 for NP0-5EO, and 0.890 for OP0-5EO.  Including the NPEC 
concentrations to NP0-16EO removal decreases the strength of the correlation (Figure 
4.6), r = 0.775.  This seems to be related to the higher variability in APEC accumulation 
between WWTPs mentioned above, but a larger amount of data would be necessary to 
show that there is indeed a statistically significant difference.  However, if this were the 
case, previous statements (Fensterheim 2001) that APECs are highly biodegradable and, 
hence, completely removable from properly-operating WWTPs would not necessarily 
apply to every type of wastewater treatment.
103
60
65
70
75
80
85
90
95
100
5.0 10.0 15.0 20.0 25.0 30.0
temperature, ?C
r
e
m
o
v
a
l
,
 
%
NPEO + NPEC NPEO
Figure 4.6 NP0-16EO (squares) and NP0-16EO + NP0-1EC (diamonds) removal 
from three WWTPs as a function of temperature.
104
APEO removal can also be affected by other factors, such as hydraulic residence 
time (HRT) and SRT (Environment Canada 2001).  In this study, the three WWTPs 
sampled operate with similar SRTs and HRTs (Table 4.1).  Although Blue Plains
activated sludge treatment has a short SRT compared to the other two WWTPs, the SRT 
for its nitrification/denitrification step is significantly longer.  This might explain that the 
overall APEO removal efficiency is not affected, because nitrifying conditions have been 
correlated with high APEO degradation rates (Ahel et al. 1994a).  Removal rates from 
Blue Plains? secondary treatment, which can be as low as 18% for NP0-16EO in winter, 
compared to Calumet?s (Table 4.3) higher rates achieved with the same type of treatment, 
substantiate the influence of HRT and SRT on APEO removal.
APEO concentrations in sludge and total mass balances.  Due to the affinity of 
the short-chain APEOs to solids, a complete assessment of APEO removal, and 
especially degradation, in WWTPs cannot be performed without considering the APEOs 
leaving the plant with the sludge.  Primary, secondary and tertiary sludge (in Blue Plains) 
were collected during the March 2005 sampling events and analyzed for the APEOs.  
Results are presented in Table 4.4.  The concentrations in sludge were used with the 
concentrations in water to construct mass balances (Figure 4.7) and compare them to 
Ahel and collaborators? (1994a) study of 2 Swiss WWTPs, where at least 60 to 65% of 
the NPEOs incoming to these plants are discharged into the environment, mostly in the 
form of degradation products.  Our results varied widely; Back River performed in a
similar way as the European plants, releasing almost 60% of the incoming mass of 
APEOs in the liquid effluent and the primary and secondary sludge.  Calumet released a 
smaller amount, slightly more than 40% in the winter, and around 30% in the summer.
105
Table 4.4 NP0-16EO and OP0-5EO concentrations in sludge from the WWTPs.
Concentration, umole/g
Back River WWTP Blue Plains WWTP Calumet WWTP
March 2005 March 2005 March 2005
Comp Primary Secondary Primary Secondary Nitr Primary Secondary
NP 0.3991 0.3264 0.2255 1.3727 0.1791 0.3841 0.1900
NP1 0.2864 0.4470 0.3902 3.6780 0.1943 0.2610 0.2519
NP2 0.2451 0.2568 0.2347 0.7013 0.0513 0.2058 0.4156
NP3 0.1935 0.0497 0.0565 0.1287 0.0141 0.1136 0.0756
NP4 0.2030 0.0125 0.0634 0.0323 0.0023 0.0730 0.0126
NP5 0.1705 0.0062 0.0464 0.0138 0.0012 0.0559 0.0063
NP6 0.2025 0.0041 0.0585 0.0064 0.0011 0.0628 0.0044
NP7 0.1572 0.0026 0.0447 0.0042 0.0009 0.0422 0.0028
NP8 0.1316 0.0021 0.0465 0.0033 0.0007 0.0311 0.0017
NP9 0.1000 0.0016 0.0388 0.0024 0.0006 0.0232 0.0012
NP10 0.0730 0.0013 0.0295 0.0018 0.0004 0.0168 0.0009
NP11 0.0494 0.0009 0.0207 0.0014 0.0003 0.0119 0.0006
NP12 0.0317 0.0007 0.0151 0.0009 0.0002 0.0084 0.0004
NP13 0.0199 0.0005 0.0096 0.0007 0.0002 0.0056 0.0003
NP14 0.0119 0.0003 0.0058 0.0004 0.0001 0.0036 0.0002
NP15 0.0063 0.0002 0.0035 0.0003 0.0001 0.0022 0.0001
NP16 0.0033 0.0001 0.0020 0.0002 0.0001 0.0010 0.0001
OP 0.0110 0.0231 0.0048 0.0568 0.0063 0.0226 0.0077
OP1 ND 0.0234 ND 0.0904 ND 0.0148 0.0065
OP2 0.0098 0.0092 0.0049 BQL 0.0009 0.0075 0.0228
OP3 0.0061 0.0030 0.0031 0.0012 ND 0.0045 0.0030
OP4 0.0046 0.0004 ND ND ND 0.0019 0.0004
OP5 0.0033 0.0002 ND ND ND 0.0011 0.0002
106
Figure 4.7 NPE mass balance in three WWTPs. Influent: total mass in raw influent 
(except for Calumet in August, where it corresponds to the primary effluent), 
Sludge: total mass in waste primary and secondary sludge (Blue Plains includes 
tertiary sludge); Effluent: total mass in final effluent; Degraded: estimated from 
Influent-(Sludge+Effluent).
Influent
557 mol/d
100 %
Back River (March 2005)
Biological
(MLE)
Effluent
216 mol/d
39 %
Degraded
235 mol/d
42 %
Sludge
106 mol/d
19 %
Primary Filt/Chlor
Influent
1765 mol/d
100 %
Blue Plains (March 2005)
Biological
(AS/Denit)
Effluent
423 mol/d
24 %
Degraded
215 mol/d
12 %
Sludge
1128 mol/d
64 %
Primary Filt/Chlor
Influent
1152 mol/d
100 %
Calumet (March 2005)
Biological
(MLE)
Effluent
228 mol/d
20 %
Degraded
677 mol/d
59 %
Sludge
247 mol/d
21 %
Primary Filt/Chlor
107
Whereas in the case of Blue Plains, most of the APEO elimination occurred in the solids, 
resulting in almost 90% of the APEOs being released from the WWTP.  This might be 
due to the shorter SRT causing a higher production and elimination of biosolids.  
Presumably, the short-chain APEOs formed during secondary treatment would attach to 
the solids and would not be available for further microbial degradation.  A more accurate 
evaluation of these phenomena would require the analysis of a more comprehensive set 
of the APEO biotransformation products, such as long-chain APECs and dicarboxylates.
APEO sources.  Due to biodegradability concerns in the mid-1970s, the 
surfactant industry substituted APEOs in household products with alcohol ethoxylates 
(Vivian 1986), and according to Talmage (1994), the household uses of the APEOs 
account for only 15% of the total.  However, a survey of sewage from different areas in 
Chicago suggests that residential areas might still be important sources of the NPEOs in 
WWTP influent (Figure 4.8).  NPEO concentrations and individual oligomer distributions 
in residential samples were comparable to the industrial and commercial sources, as well 
as to the WWTP influent (Figure 4.8A), and in one case (Residential 1) NPEO 
concentrations were significantly higher than in the other sites.  A more detailed survey 
of sewers in different residential areas of Chicago supported the earlier observations 
(Figure 4.8B).  The OPEOs were also detected and followed similar patterns, but OPEO 
concentrations did not vary in accordance to the NPEOs, suggesting different origins for 
these two families of compounds, even within the residential sewers.
108
Figure 4.8 Total NP0-16EO concentrations (dissolved + particulate) in (A) grab 
sewage samples from commercial, industrial and residential areas in Chicago, IL, 
and (B) 24-h composite sewage samples from different residential areas in Chicago.
0
50
100
150
200
250
300
350
WWTP Influent Residential 1 Residential 2 Commercial Industrial
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP NP1EO NP2EO NP3EO NP4EO NP5EO NP6EO NP7EO NP8EO
NP9EO NP10EO NP11EO NP12EO NP13EO NP14EO NP15EO NP16EO
0
1
2
3
4
5
6
Residential 2 Commercial
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
A
0
20
40
60
80
100
120
140
160
180
200
Chicago 2 Chicago 3 Oak Lawn 1 Oak Lawn 2 Bridgeview
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP NP1EO NP2EO NP3EO NP4EO NP5EO NP6EO NP7EO NP8EO
NP9EO NP10EO NP11EO NP12EO NP13EO NP14EO NP15EO NP16EO
B
109
CHAPTER 5 ? APEO FATE IN BACK RIVER
This chapter to be submitted to Environmental Science and Technology.
5.1 Abstract
The concentrations of nonylphenol (NP), octylphenol (OP), their ethoxylates 
(NP0-16EO and OP0-5EO respectively) and some of their carboxylated derivatives
(NP0-1EC and OP0EC quantitatively; NP2-3EC and OP1EC qualitatively) were 
evaluated in water samples from the Back River, MD, a sub-estuary of the Chesapeake 
Bay that receives effluent from Back River wastewater treatment plant.  The most 
abundant of the alkylphenolic compounds (APEs) were the carboxylates (APECs, > 95% 
on mass basis), followed by NP in September and October, and NP1-2EO in March.  NP 
concentrations found in this study, 0.087 ? 0.69 ?g/L, were below acute toxicity 
thresholds, and generally below recently proposed water quality criteria by the US EPA; 
although in March, concentrations can be close to half of the chronic exposure limit (1.4 
?g/L) for saltwater.  Total NPE concentrations in the estuary seemed to vary in 
accordance to the concentrations in the WWTP effluent, especially in the case of the 
APECs.  However, a closer analysis of the data suggested that in the fall sampling events, 
when rain occurred, the ethoxylates present in the particulate matter originated in the 
river?s tributaries rather than the WWTP.
110
5.2 Introduction
The ubiquitous presence in diverse environmental compartments of alkylphenol 
ethoxylates (APEOs), a widely-applied family of nonionic surfactants, has been 
extensively documented, e.g. Montgomery-Brown and Reinhard 2003, Kolpin et al. 2002.  
The most abundantly used are the nonylphenol ethoxylates (NPEOs), followed by their 
octylphenolic homologues (OPEOs) (Talmage 1994), and the difference in usage volume 
is reflected in the concentrations measured in the environment (e.g. Ferguson et al.
2001b).  Numerous reports also exist on the biological transformations of these 
compounds in the environment (e.g. Montgomery-Brown and Reinhard 2003), although 
most studies focus on the transformation reactions occurring in wastewater treatment 
plants (WWTPs) because they are considered the dominant environmental sources of the 
APEOs and, especially, of their transformation products, which are of interest due to their 
toxicological properties, specifically their potential for endocrine disruption (ED) 
(Nimrod and Benson 1996).  The main biological transformation pathways of the APEOs 
are depicted in Figure 5.1; the non-oxidative pathway was recognized first (Ahel et al. 
1994a) and seems to occur mainly in WWTPs, whereas the oxidative was proposed more 
recently (Jonkers et al. 2001) and appears to be the dominant biotransformation pathway 
in natural waters.  In either case, the resulting transformation products, specifically 
nonylphenol and octylpenol (NP and OP respectively), are suspected to be more toxic 
than the parent compounds and with the highest ED-inducing potential.  The mono- and 
111
H
2x+1
C
x
O
O H
n
H
2x+1
C
x
O
O H
2
H
2x+1
C
x
O
O H
H
2x+1
C
x
OH
H
2x+1
C
x
O
n-1
O
O
OH
H
2x+1
C
x
O
O
O
OH
H
2x+1
C
x
O
O
OH
H
2x+1
C
x
O
O H
n-1
?
12
 
Figure 5.1 Main biological transformation pathways for the APEOs: (1) non-
oxidative, and (2) oxidative; x = 8 for OPEOs, and x = 9 for NPEOs.  Adapted from 
Montgomery -Brown and Reinhard 2003.
112
diethoxylates (AP1EO and AP2EO) show similar toxicological properties, albeit to a 
lesser degree; whereas evidence for the ED properties of the oxidation products, the 
alkylphenoxy carboxylates (APECs), is variable (see discussion in the section on 
toxicological significance below), but they are generally considered to be more soluble 
and persistent than the APEOs (Jonkers et al. 2001).  These concerns prompted the 
European Union to heavily restrict the use of NPEs in concentrations equal or higher than 
0.1% by mass in products ranging from industrial and domestic cleaning products, to 
personal care products and pesticide formulations, and in industrial processes such as 
textile, leather and paper manufacturing (?Directive? 2003).  In the United States, the 
EPA published draft water quality criteria for NP of 27.9 and 5.9 ?g/L for acute and 
chronic exposure in freshwater, and 6.7 and 1.4 ?g/L for acute and chronic exposure in 
saltwater (U. S. EPA 2003).
In spite of the abundance of occurrence data, few studies have dealt with the fate 
of the entire range of APEOs (Ahel et al. 1994b, Ferguson et al. 2001b, Jonkers et al. 
2005), especially in the United States.  In this study, a comprehensive set of APEOs and 
transformation products (collectively referred as APEs) were analyzed in water from a 
sub-estuary of the Chesapeake Bay that receives WWTP effluent from Baltimore, MD, as 
well as urban runoff, in an attempt to explain their origin and fate, and seasonal 
differences in their concentrations.
113
5.3 Experimental Section
Standards and Reagents.  Standards and reagents used for the analysis of the 
samples presented here were described in detail in previous reports, Rice et al. 2003, 
Loyo-Rosales et al. 2003 and Chapter 2.  Analytical standards were acquired from 
commercial sources or purified in the laboratory by flash chromatography, whereas the 
[
13
C
6
]-labeled internal standards were obtained from Ferguson et al. 2001.  All the 
reagents were of the highest purity available and the organic solvents were high purity 
and pesticide grade, mainly acquired from Burdick and Jackson (Honeywell 
International, Muskegon, MI, USA).  Deionized carbon-free water was prepared in the 
laboratory with a NANOpure water purification system (Barnstead International, 
Dubuque, IA, USA).
Site Descriptions and Sampling Procedures.  The Back River is a relatively 
small sub-estuary of the Chesapeake Bay situated in Baltimore County, Maryland.  Its 
watershed is approximately 158 km
2
 and consists predominantly of urban areas (MDE 
2005).  Several small streams discharge into the Back River, but an important source of 
water to the river is the Back River WWTP, situated in the upper reaches of the estuary 
(Fig. 5.2); no other municipal or industrial point sources exist in the Back River (MDE 
2005).
Two groups of samples from the Back River were collected and analyzed in 2001 
and 2004-2005.  The first group consisted of three sampling trips in the R/V Orion in 
January, February and July 2001; surface water samples were collected in the three 
events, whereas sediments were collected in January and July only.  Sampling procedures 
114
Figure 5.2 Back River, its main tributaries, and the location of the WWTP.  
Numbers 1 to 4 indicate sampling locations BR-1 to BR-4 described in Table 1, and 
MR is the background sampling site in Middle River.
0 12
kilometers
N
Chesapeake Bay
Patapsco River
Back River
Middle River
Stemmers Run
Moores Run
Herring Run
2
1
4
3
MR
WWTP
115
for these events were described previously (Loyo-Rosales et al. 2003).  Additionally, 
surface water samples were collected from two tributaries of the Back River, Herring Run 
and Moores Run (Fig. 5.2), to assess non-point source contributions to the river.  The 
samples in the second group were collected by John Martin and other personnel from 
Back River WWTP in September 2004, October 2004, and March 2005.  They included a 
background sample from the Middle River (Fig. 5.2), and surface water samples from 4 
sites in the Back River (Table 5.1).  Grab samples of the plant?s final effluent were 
collected the day after river sampling occurred.  In all cases, water samples were 
collected in previously baked 4-L amber glass bottles; they were kept on ice during 
transport to the laboratory and in refrigeration (4?C) until they were filtered and extracted 
within 48 hrs of collection.  Sediments were collected in previously baked 250 mL wide-
mouth glass jars, transported in ice, and frozen at -20?C until analysis.  No chemical 
preservatives were added to any of the samples.
Extraction and Analysis. Because the first group of samples was collected 
during the analytical method development stage, water samples were analyzed with 
different methods and for different sets of compounds.  Water samples collected in 
January and February 2001 were analyzed by GC/MS as described in Rice et al. (2003) 
for NP0-3EO and OP0-3EO exclusively, whereas water samples from July 2001 were 
analyzed by LC/MS/MS for NP0-5EO and OP0-5EO using the methods described in 
Loyo-Rosales et al. 2003.  All water samples from the second group, i.e. those collected 
in 2004 and 2005, were analyzed by LC/MS/MS using the methods described in Chapter 
2 for NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in the dissolved fraction, and for the 
same analytes, except for the APECs, in suspended particulate matter.  Additionally, 
116
Table 5.1 Sampling sites location and water quality parameters for the second group 
of samples.  MR: Middle River (reference site); BR: Back River.  Data provided by 
John Martin (Back River WWTP) except for the suspended solids.
MR BR-1 BR-2 BR-3 BR-4 
 Location
latitude 39? 18.73' 39? 18.10' 39? 17.70' 39? 17.30' 39? 15.45'
longitude 76? 26.36' 76? 29.30' 76? 28.40' 76? 27.90' 76? 26.60' 
Sep 15, 2004
temperature, ?C 24.2 23.3 23.6 23.8 23.8
pH 7.7 7.9 8.2 8.3 7.9
dissolved oxygen, mg/L 7.1 7.6 9.3 9.0 7.3
salinity, ppt 1.50 0.47 0.60 0.62 1.16
total suspended solids, mg/L 10.6 36.6 31.5 32.8 25.4
Oct 20, 2004
temperature, ?C 14.8 13.6 14.5 14.7 13.9
pH 7.6 7.3 7.8 7.9 7.9
dissolved oxygen, mg/L 8.6 8.4 9.6 9.0 9.5
salinity, ppt 1.02 0.31 0.40 0.44 0.89
total suspended solids, mg/L 8.4 37.8 21.5 26.4 23.0
March 16, 2005
temperature, ?C 5.8 5.7 7.5 6.1 5.0
pH 7.7 7.4 7.5 7.7 7.5
dissolved oxygen, mg/L 12.4 15.0 13.6 14.3 13.8
salinity, ppt 2.26 1.44 1.29 1.57 3.27
total suspended solids, mg/L 17.0 25.0 21.7 24.1 21.9
117
NP2-3EC and OP1EC were monitored qualitatively in the dissolved fraction (Chapter 2).  
All the sediment samples were analyzed for the NP0-16EOs and OP0-5EOs using the 
methods described in Chapter 2.
For the second group of samples, APEC concentrations were recalculated using a 
standard additions approach (Chapter 2) because of the lack of appropriate internal 
standards and concern that matrix interferences might result in erroneous results.  Results 
indicated that the external standard method tended to overestimate APEC concentrations 
by an average of 13%, suggesting a slight matrix-induced ionization enhancement.  
However, in the case of OP0EC, which showed the highest overestimation?an average 
of 32% in September and October, the cause was most likely that the values were close to 
the detection limit.  As a consequence, the concentration values obtained by standard 
additions were used for this work. 
5.4 Results and Discussion
APEs concentrations in Back River.  Results from the set of water samples 
collected in 2001 from the Back River and the tributaries, Herring and Moores Run, are 
summarized in Table 5.2; whereas Figure 5.3 shows the results for the sediments 
collected in January and July 2001.  Concentrations of all the NPEs found in Back River 
and the Back River WWTP effluent from the second sample set are presented in Figures 
5.4 and 5.5.  As in the effluent, most of the NPEs found in the river were NPECs, 84-95% 
on mass basis (86-93% molar basis), except for BR-1 in October 2004 (65 and 66% in 
Table 5.2 Dissolved AP and APEO concentrations in Back River and Back River tributaries, Herring Run and Moores Run, in 
the 2001 sampling events. Site locations: BR-A: 39?18.01?N, 76?29.07?W; BR-B: 39?17.37?N, 76?28.29?W; BR-C: 39?16.31?N, 
76?26.41?W; BR-D: 39?15.34?N, 76?26.44?W; BR-E: 39?14.49?N, 76?26.10?W; BR-F: 39?14.36?N, 76?23.59?W.
concentration, ?g/L
Jan 2001 Feb 2001 Jul 2001
NP NPnEO* OP OPnEO* NP NPnEO* OP OPnEO* NP NPnEO* OP OPnEO*
BR-A 0.35 2.3 0.022 0.064 0.22 1.5 0.014 0.044 0.21 0.28 0.022 0.022
BR-B 0.37 3.7 0.024 0.14 0.52 3.4 0.030 0.010 0.20 0.26 0.021 0.021
BR-C 0.24 2.2 0.018 0.061 0.17 1.1 0.011 0.016 0.089 0.14 0.018 0.025
BR-D 0.13 2.8 0.012 0.036 0.10 0.55 0.010 0.010 0.12 0.10 0.014 0.015
BR-E 0.079 0.46 0.0056 0.010 0.11 1.1 0.011 0.030 0.12 0.071 0.012 0.020
BR-F 0.048 0.18 0.0049 0.0071 0.018 0.052 0.0013 0.0026 0.14 0.067 0.059 0.020
upstream 0.42 2.9 0.0039 0.099 0.054 0.12 0.0048 0.0077 0.16 3.7 0.024 0.36Herring 
Run
downstream 0.047 0.31 0.0094 0.037 0.041 0.21 0.0049 0.020 0.021 0.088 0.0059 0.017
upstream 0.070 0.39 0.0060 0.020 0.063 0.57 0.0049 0.031 0.018 2.8 0.0035 0.16Moores
Run
downstream 0.031 0.20 0.0044 0.019 0.030 0.054 0.0029 0.0017 0.065 0.27 0.027 0.073
* n = 1-3 in January and February; n = 1-5 in July
119
Figure 5.3 NP0-16EO concentrations in sediments from the Back River in (A) 
January and (B) July 2001. Site locations are in Table 5.2.
0.0
0.5
1.0
1.5
2.0
2.5
BR-A BR-C
c
onc
en
tr
ati
on, ug/g
A
0
1
2
3
4
5
6
7
8
9
BR-A BR-B BR-C BR-D BR-E BR-F
c
onc
entr
ati
on, ug/g
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14 NP15 NP16
B
Figure 5.4 Total NPE concentrations (dissolved + particulate; ?g/L) in the Back River WWTP effluent (EFF), Back (BR1-4) 
and Middle River (MR) water samples.  NP, NP1-3EO, and NP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of NP0-1EC are compared to the total NP0-16EO for September 
and October 2004 in (B) and for March 2005 in (D).
0
5
10
15
20
25
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP1-3EO NP NP4-16EO
C
0
5
10
15
20
25
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
B
A
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
A
0
20
40
60
80
100
120
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
NP0-1EC NP0-16EO
D
Figure 5.5 Total NPE concentrations (dissolved + particulate; ?M) in the Back River WWTP effluent (EFF), Back (BR1-4) 
and Middle River (MR) water samples.  NP, NP1-3EO, and NP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of NP0-1EC are compared to the total NP0-16EO for September 
and October 2004 in (B) and for March 2005 in (D).
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
A
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
B
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
NP1-3EO NP NP4-16EO
C
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
NP0-1EC NP0-16EO
D
122
mass and molar basis respectively) (Figs. 5.4B, 5.4D, 5.5B, and 5.5D).  NP was the 
second most abundant compound in the estuary in September (Fig. 5.6A) and October 
2004, whereas in March 2005 NP was found in lower concentrations than NP1EO and 
NP2EO (Fig. 5.6C).  Table 5.3 shows that NP concentrations found in the Back River 
were below the median value found in a nationwide USGS reconnaissance study that 
targeted streams susceptible to wastewater contamination (Kolpin et al. 2002), and 
similar to the levels found in Jamaica Bay, which is also a WWTP-effluent impacted 
estuary (Ferguson et al. 2001b). NP concentrations in the Back River were also below 
those measured by Barber et al. (2000) in the Des Plaines River, IL (Table 5.3).  They 
also analyzed water from the Illinois and Minnesota Rivers, but they did not detect NP in 
either site.  NP1EO and NP2EO in September and October 2004 were significantly lower 
than the median value reported in the national reconnaissance, but the concentrations 
increased in March 2005 and were 2 to 3 times higher than the median average in some 
instances (Table 5.3); similar results were observed in 2001.  As it was for NP, NP1-3EO 
concentrations in the Back River during the summer months were comparable to the 
levels found in 1998-1999 in Jamaica Bay (Table 5.3), but tended to be higher in Back 
River in January, February and March.  NP1EO and NP2EO concentrations found in the 
Des Plaines River were in the same range as in Back River, while NP3EO was not 
detected in the former.  NPEC levels were in a similar range as the concentrations found 
by Field and Reed (1996) and by Barber et al. (2000) in several American rivers (Table 
5.3), and NP1EC was more abundant than NP0EC in the three studies.
In contrast to the Back River, NPE concentrations in the Middle River were 
relatively low (Figs. 5.4 and 5.5), and the proportion of NPECs was significantly lower, 0 
Figure 5.6 NP0-16EO and OP0-5EO concentrations (dissolved + particulate; ?M) in the Back River WWTP effluent (EFF), 
Back (BR1-4) and Middle River (MR) water samples in September 2004 (A: NPEOs; B: OPEOs) and March 2005 (C: NPEOs; 
D: OPEOs).
0
0.001
0.002
0.003
0.004
0.005
0.006
EFF BR-1 BR-2 BR-3 BR-4 MR
September
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
A
0
0.0001
0.0002
0.0003
0.0004
0.0005
0.0006
0.0007
0.0008
0.0009
0.001
EFF BR-1 BR-2 BR-3 BR-4 MR
September
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
B
0
0.005
0.01
0.015
0.02
0.025
0.03
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14 NP15 NP16
C
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
M
OP OP1 OP2 OP3 OP4 OP5
D
Table 5.3 NP0-3EO, OP0-2EO, and NP0-1EC concentrations found in the Back River in this study compared to other sites in 
the United States.
concentration, ?g/L
NP NP1EO NP2EO NP3EO OP OP1EO OP2EO NP0EC NP1EC
US (median values)
a
0.8 1 1 --- --- 0.2 0.1 --- ---
Jamaica Bay
b
0.077-0.42 0.061-0.22 0.039-0.40 0.026-0.25 0.0016-0.007 0.0023-0.026 0.0017-0.016 --- ---
US rivers 
c
ND-1.7 ND-1.6 ND-0.50 ND ND ND ND 0.10-19 0.60-31
US rivers
d
--- --- --- --- --- --- --- ND-2.0 ND-12
Back River (this study)
Jan 2001 
e
0.048-0.37 0.084-1.0 0.094-1.4 0.45-1.9 0.0049-0.024 0.0071-0.061 ND-0.012 --- ---
Feb 2001 
e
0.018-0.52 0.041-1.3 0.012-1.4 ND-0.63 0.0013-0.030 0.0026-0.067 ND-0.0063 --- ---
July 2001 
e
0.089-0.21 0.020-0.076 0.011-0.060 0.009-0.066 0.012-0.059 ND-0.0030 0.0026-0.004 --- ---
Sept 2004 
f
0.087-0.21 0.026-0.081 0.016-0.036 0.013-0.037 ND ND ND 0.56-1.4 0.99-4.1
Oct 2004 
f
0.17-0.23 0.038-0.16 0.013-0.099 0.015-0.055 0.012-0.019 ND ND 1.2-1.8 3.2-6.0
March 2005 
f
0.38-0.69 0.84-2.7 0.30-1.9 0.16-0.67 0.13-0.22 0.12-0.30 0.011-0.12 9.4-22 11-26
a
From Kolpin et al. 2002
b
From Ferguson et al. 2001b
c
From Barber et al. 2000.  NPEOs detected in Des Plaines River only; NPECs detected in all the rivers sampled (Des Plaines, Illinois and Minnesota)
d
From Field and Reed 1996
e
Dissolved concentrations
f
Total (dissolved + particulate) concentrations
125
to 48% on mass basis or 0-65% on molar basis.  The latter difference suggests different 
origins for the NPEs in the two estuaries; whereas Back River?s main NPE source is 
expected to be the WWTP, the Middle River is not exposed to treated wastewater, and 
any effluent or urban runoff contributing NPEs to the site is more likely to contain a 
partially or non-degraded mixture.  This is also supported by the fact that NP4-16EO, 
thecompounds most often used in consumer applications, were the most abundant NPEs 
in Middle River (Figs. 5.4A and 5.4C), as opposed to the rest of the NPEs that are 
considered degradation products.
OPE concentrations in Back River and the WWTP effluent were approximately 
one order of magnitude lower than the NPEs (Fig. 5.7), and they showed a similar 
behavior to that of the NPEs described above; OP0EC was the most abundant species in 
similar proportions to OP0-5EOs as NP0EC was to NP0-5EO, i.e. approximately 80%.  
OPEs were only detected in March in the Middle River; they were present in lower 
concentrations than in the Back River and exclusively in two forms, OP (80%) and 
OP0EC (20%) (Fig. 5.7C and inset in Fig. 5.7D).  OP and OP1EO concentrations in the 
Back River, when detected, tended to be higher than those observed in Jamaica Bay, 
whereas OP2EO was only higher in March 2005.  The concentrations of the two OPEOs 
in March 2005 were close to the median values from the USGS reconnaissance study 
(Table 5.3), but lower at all other times.  Barber?s group did not detect any OPEs in their 
river samples.
Although they were not quantified, NP2EC, NP3EC and OP1EC were monitored 
in the Back River and WWTP effluent samples and they were detected in all of them. 
This implies that the proportion of APECs vs total APEs are in reality higher than the 
Figure 5.7 Total OPE concentrations (dissolved + particulate; ?g/L) in the Back River WWTP effluent (EFF), Back (BR1-4) 
and Middle River (MR) water samples.  OP, OP1-3EO, and OP4-16EO concentrations for September and October 2004 are 
shown in (A) and for March 2005 in (C).  The concentrations of OP0-1EC are compared to the total OP0-16EO for September 
and October 2004 in (B) and for March 2005 in (D).
0
0.1
0.2
0.3
0.4
0.5
0.6
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
A
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
OP1-3EO OP OP4-5EO
C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
EFF BR-1 BR-2 BR-3 BR-4 MR EFF BR-1 BR-2 BR-3 BR-4 MR
September October
c
o
n
c
e
n
t
r
a
t
i
o
n
s
,
 
u
g
/
L
B
0
1
2
3
4
5
6
7
8
EFF BR-1 BR-2 BR-3 BR-4 MR
March
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
OP0EC OP0-5EO
D
0
0.02
0.04
0.06
0.08
MR
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
L
127
values reported above.  In the case of the NPECs, and assuming that the LC/MS/MS 
response of NP2-3EC is similar to those of the NP0-1EC, the former, although present in 
lower concentrations than the latter, would represent at least 40% of the total APEC 
concentration in September and October, and 15% in March.  Barber et al. (2000) 
reported the presence of NP2-3EC in the three rivers they studied, also in lower 
concentrations than NP0-1EC.  In contrast, Field and Reed (1996) did not detect NP2EC 
or NP3EC in their river samples.
APEs partitioning to suspended solids. APEs are relatively hydrophobic 
compounds as attested by their log K
ow
 values, which were measured by Ahel and Giger 
(1993) at 4.5 to 4.2 for NP0-3EO, and estimated to be between 4.3 and 4.1 for NP4-10EO 
by the same group.  Due to their hydrophobic nature, APEs have the tendency to partition 
to the organic matter in suspended solids and are therefore removed from the water 
column by deposition, as observed by Ferguson et al. (2001b).  Although APE sorption to 
solids has been correlated to the amount of organic matter present (e.g. Ferguson 2001b, 
John et al. 2000), some experimental evidence suggests that sorption to the mineral 
fraction may play an important role too.  Brownawell et al. (1997) studied the sorption of 
three alcohol ethoxylates?A
13
EO
3
, A
13
EO
6
, and A
13
EO
9
?to sediment.  These 
compounds are also nonionic surfactants, and they have almost identical chemical 
structures to the APEOs; they only differ in that the APEOs include a benzene ring.  
Brownawell?s group found that the extent of sorption increased with the number of 
ethoxylate (EO) units, and they could not find a strong correlation between the affinity of 
the AEs for the sediment and its organic carbon content probably due to the adsorption to 
mineral surfaces in the sediment (Brownawell et al. 1997).  Further evidence of the 
128
Figure 5.8 K
d
 values, L/g, as a function of ethoxylate-chain length for NP0-16EO in 
sites BR-2 and BR-3, March 2005.
y = 2.4616x + 11.683
R2 = 0.9374
y = 1.9303x + 18.527
R2 = 0.8251
0
10
20
30
40
50
60
N
P
N
P
1
E
O
N
P
2
E
O
N
P
3
E
O
N
P
4
E
O
N
P
5
E
O
N
P
6
E
O
N
P
7
E
O
N
P
8
E
O
N
P
9
E
O
N
P
1
0
E
O
N
P
1
1
E
O
N
P
1
2
E
O
N
P
1
3
E
O
N
P
1
4
E
O
N
P
1
5
E
O
N
P
1
6
E
O
compound
K
d
,
 
L
/
g
BR-2 BR-3 Linear (BR-2) Linear (BR-3)
129
sorption of APEOs to mineral surfaces was provided by John et al. (2000), who observed 
that sorption to organic-free sediment, kaolinite, or silica increased with the number of 
EO units.  In the present study, observed sorption coefficients, K
d
, for the March 2005 
samples tended to increase with the length of the ethoxylate chain too (Fig. 5.8), an 
opposite trend to what would be expected from the decreasing log K
ow
 values discussed 
above, therefore suggesting that sorption to mineral surfaces might be important in the 
Back River.  This trend did not change even after normalizing the K
d
 values for organic 
carbon content, which was an average of 10 %.  The organic-carbon normalized sorption 
coefficients, log K
oc
 (K
oc
 = K
d
/f
oc
, where f
oc
 is the fraction of organic carbon in the solids), 
for NP0-16EO showed little variation among sites and different compounds, 5.27 to 5.64, 
in agreement with the observations of Ferguson et al. (2001b) in Jamaica Bay for NP0-
3EO, who reported values of 5.39, 5.46, 5.18, and 4.87 for NP, NP1EO, NP2EO, and 
NP3EO respectively compared to 5.34 ? 0.09, 5.45 ? 0.13, 5.40 ? 0.15, and 5.43 ? 0.16 
(n = 4) calculated by our group.
Model of NPE fate in Back River.  In order to understand the behavior of the 
NPEs in the Back River, and to evaluate the relative contributions of different removal 
mechanisms, a steady state model of the compounds in the estuary was used.  The steady 
state assumption was considered reasonable because the residence time of the water in 
the portion of the estuary where the sampling points were located is approximately 39 
days (108 days for the whole estuary), and from a previous study (Chapter 4) it appears 
that effluent concentrations remain relatively constant during a season, i.e. summer or 
winter.  Back River was modeled as an incompletely mixed system based on the 
following basic mass balance equation, which includes the different processes considered 
130
relevant in determining the concentration of a given NPE compound at a certain location 
in the river (the mathematical details can be found in Appendix 1):
0 = external mass load ? advection ? dispersion ? deposition ? transformations Eq. 5.1
The model was developed using the results for the second set of samples, i.e. 
samples taken in 2004-2005, because this set included a larger number of NPE species, 
such as the NPECs.  In order to simplify the model, the different individual NPE species 
were initially grouped into two broad groups: NPEOs, encompassing NP0-17EO; and 
NPECs, including NP0-1EC.  After the model was developed with these two groups, 
individual NPE species, such as NP, were also modeled independently as discussed later 
in this chapter.
Only two external NPE sources were considered, the WWTP effluent and the 
tributaries (Fig. 5.2).  After an initial inspection of the results, the effluent was assumed 
to be the main contributor of NPEs to the Back River, because total NPE concentrations 
in the Back River seemed to vary in accordance to the concentrations in the WWTP 
effluent (Figs. 5.4B and 5.4D).  This assumption held for the NPECs, and for the NPEOs 
in March 2005, but the tributaries seemed to be the main contributors of NPEOs in both 
September and October 2004, when rainfall occurred before and during both sampling 
events (see discussion below).
Advection was assumed to be a function of the net estuary flow, which included 
the WWTP effluent and the average watershed discharge.  Dispersion was modeled using 
the physical characteristics of the river and an average dispersion coefficient calculated 
131
using the salinity data in Table 5.1.  More specific details for the approach used to model 
both transport phenomena are described in Appendix 1.
Removal from the water column by sorption to suspended solids and subsequent 
deposition was assumed to be relevant for the NPEOs only.  The NPECs were expected 
to be less prone to sorption than the NPEOs due to their higher solubility?although no 
solubility data are available for the NPECs, they are most probably present in their ionic 
form in the Back River because of their pK
a
 of approximately 4 and water pH > 7 (Table
5.1).  Furthermore, although information on NPEC sorption is scarce, previous attempts 
to measure them in sediment and sludge suggest that it is not an important removal 
mechanism (see Petrovic et al. 2002, and Petrovic and Barcelo 2001b for sediment, and 
Lee et al. 1997 for sludge).  For this work, NPEO sorption to particulate and deposition 
was assumed to be the predominant removal mechanism for these compounds, especially 
in winter.  Sorption was modeled using weight-averaged values of the observed sorption 
coefficients, K
d
, for the different NPEOs along with average suspended solids 
concentrations in Back River to calculate the fraction of the NPEOs in the particulate, F
p
.  
Because no data were available on the settling velocity, 
s
, for the river, a value of 0.25 
m/d was chosen, which corresponds to the average value for phytoplankton and organic 
solids (Chapra 1997).  More details can be found in Appendix 1.
NPEs might be subject to different transformation reactions, such as 
volatilization, photolysis, and microbial-mediated biotransformation.  For the purposes of 
this model, all transformation reactions were assumed to follow first-order kinetics and 
aggregated into a single rate constant, k, which was used as a fitting parameter also.  As 
for the relevance of the different processes, biotransformation is believed to be the most 
132
important, and it has been widely studied as discussed in the introduction; whereas 
volatilization and photolysis might be confined to the water surface. Furthermore, 
volatilization is expected to be significant only in the case of NP because of the lower 
Henry?s law constants and molecular diffusion coefficients of the NPEOs (Ferguson et al. 
2001b).  This process was reported to be the main removal mechanism for NP in the 
lower Hudson River estuary (Van Ry et al. 2000), but data from the present study, 
however, do not support volatilization as a significant removal mechanism for NP in the 
Back River because the preferential removal of NP would imply an increase in the 
NPEO/NP ratios and in this site these ratios decreased as discussed below.  The same 
would apply for OP; although proper assessment of the importance of volatilization for 
both compounds would require additional data, specifically atmospheric concentrations.  
On the other hand, information on photolysis is scarce (Montgomery-Brown and 
Reinhard 2003), especially on its relevance in natural waters, and it was not directly 
addressed here; if photodegradation is an important removal process in the Back River, it 
would be integrated as part of the lumped reaction rate constant, k, in the model.
NPE model results.  Figure 5.9 shows the model versus measured concentrations 
for the NPECs (Figs. 5.9A and 5.9B) and the NPEOs (Figs. 5.9C and 5.9D) for the March 
2005 data.  In Fig. 5.9A, the NPECs were modeled considering all the factors in Eq. 5.1 
except for deposition and transformations, resulting in an overestimation of the measured 
concentrations.  Assuming biotransformation occurred and assigning a value of 0.015 d
-1
 
to the rate constant k resulted in a closer fit (Fig. 5.9B), with measured to predicted 
concentration ratios of 1.15, 0.92, 0.94, and 0.98 for BR-1, BR-2, BR-3, and BR-4 
respectively.  This value for k implies that approximately 1.5% of the NPECs are lost per 
Figure 5.9. Model versus measured results for sites BR-1, 1400 m, BR-2, 2500 m, BR-3, 3800 m, and BR-4, 7600 m, in March 
2005 for NP0-1EC (A) assuming k = 0; (B) k = 0.015 d
-1
; and for NP0-16EO (C) assuming k = 0; (D) k = 0.02 d
-1
.
134
day to biotransformation, or a t
?
 of 46 days, and it is close to the range calculated by 
Jonkers et al. (2005) in the Scheldt estuary also in winter, 0.016 to 0.020 d
?1
.  Note that 
the loss of NPECs by biotransformation might be compensated by their production from 
the NPEOs or long-chain NPECs (Fig. 5.1), and k in this case would be a net value 
comprising the formation and degradation of these compounds.  Although the data 
suggest that transformation of long to short-chain NPECs is not occurring because NP0-
1EC/NP2-3EC ratios barely changed from the site at the WWTP effluent to BR-4, from 
5.03 to 5.16, or less than 3%, and were very similar to the ratio in the effluent itself, 4.91.  
Biotransformation of the NPECs is believed to proceed through oxidation of the nonyl 
chain, resulting in the formation of NPE derivatives with carboxylated alkyl and 
ethoxylate chains (Jonkers et al. 2001), but the present samples were not analyzed for 
these compounds.
Fig. 5.9C shows that modeling the NPEOs with all the factors in Eq. 5.1 except 
for transformation reactions also underestimates the observed concentrations.  Assuming 
k = 0.02 d
-1
 yields the following measured to predicted ratios: 0.87, 0.94, 0.95, and 1.21
(Fig. 5.9D).  This value for k is close to the value obtained for the carboxylates above, 
and it is lower than the value reported by Jonkers et al. (2005), 0.061 d
-1
 for NPEO >2, 
0.070 for NP1-2EO, and 0.032 for NP.  This difference arises because of their assumption 
that biodegradation would be the main removal process in the Scheldt estuary, and the 
biodegradation rates were fitted neglecting sorption.  In our case, had we assumed that 
sorption did not occur, a very similar k value for the NPEOs, 0.065 d
-1
, would have been 
obtained.  As for the specific NPEO biotransformation pathways occurring in the Back 
River, it appears that de-ethoxylation of the higher NPEOs via the non-oxidative pathway 
135
depicted in Fig. 5.1 is not occurring, as suggested by the changes in the ratio NP1-
3EO/NP4-16EO (total molar-based concentrations), which changed only 5% from the site 
at the WWTP effluent, BR-2, to the downstream site BR-4.  In contrast, the ratio NP0-
1EC/NP1-16EO increased from 4.8 in the effluent to 6.7 in BR-1 and 11.4 in BR-4, 
suggesting the oxidation of NPEOs to NPECs, although the evidence of hydrolytic 
degradation of the carboxylates as expected from the oxidative hydrolytic pathway was 
very weak, because the ratio NP0-1EC/NP2-3EC remained almost constant in the river as 
discussed above.
In regard to NP, a decrease in the value of the NP0-1EC/NP (50 to 38) and NP1-
16EO/NP (7.5 to 3.4) ratios from BR-2 to BR-4 points towards the formation of NP from 
the rest of the compounds.  The model also suggests that NP concentration decreases 
more slowly than expected considering partitioning to solids or biotransformation.  NP 
formation in the estuary would seem unlikely because it is linked to anaerobic 
environments (Ahel et al. 1994a) and dissolved oxygen in the water was measured at 13 
to 15 mg/L, but it has been argued that it could occur in anaerobic microenvironments 
present in the particulate (Jonkers et al. 2001).  Alternative explanations for these 
observations are that the NP could be more resistant to degradation than the NPECs and 
the NPEOs; or that sources other than the WWTP exist for the NP in the estuary.
In contrast to the observations above for March, results for September and 
October were not as readily described by the model.  This is at least in part due to rainfall 
that occurred during and before both sampling episodes.  Whereas in March only 0.07? of 
rain occurred five days prior to the sampling, 0.64? fell the day before the sampling in 
September, and 0.16? on the sampling day.  In October, it had been raining for a week 
Figure 5.10 Model versus measured results for sites BR-1, 1400 m, BR-2, 2500 m, BR-3, 3800 m, and BR-4, 7600 m, in October 
2005 for (A) NP0-1EC, and (B) NP0-16EO.
137
prior to the sampling event, with daily rain amounts of 0.09, 0.21, 0.05, 0.05, 0.0, 0.08, 
0.12, and 0.08? (Values are 24 hr readings from the National Weather Service rain gage 
at the Maryland Science Center in Baltimore, as reported in NCDC 2004a, 2004b, and 
2005).  In order to account for the rain in the model, water flow rates from the tributaries 
were increased.  For the NPECs in October, the average amount of rain in the last three 
days was added to the mean flow from the tributaries, and it was assumed that 
degradation and sorption did not occur; results are presented in Fig. 5.10A.  Using these 
flow rate values, however, failed to explain the behavior of the NPEO concentrations 
(Figure 5.10B), with the model overestimating the observations.  A similar situation 
occurs in September, with the adjusted water flow dictating the behavior of the NPECs, 
but not that of the NPEOs.  Numerically, the overestimation is a consequence of the high
NPEO concentrations in the effluent relative to the concentrations found in the estuary 
(Fig. 5.4A).
An alternative explanation might be that the NPEOs stemming from the WWTP 
are rapidly transformed and that most of the NPEOs sorbed to the particulate, especially 
the long-chain NPEOs, are coming from the tributaries.  Evidence for the NPEOs 
originating in the tributaries comes from the relatively high amounts of long-chain 
NPEOs in particulate (the effluent has extremely low concentrations of these, as 
illustrated in Fig. 5.6A), especially at the site closest to the tributaries discharge, BR-1, 
and their decrease along the estuary (Fig. 5.11).  Also, NPEO concentrations are higher in 
BR-1 than BR-2 (Fig 5.4A), whereas NPEC concentrations are higher in the latter as 
expected if the main source were the WWTP (Fig. 5.4B).  Moreover, suspended solids 
(SS) concentrations are rain-event driven in the Back River (MDE 2003), and in both 
138
Figure 5.11 NP0-16EO concentrations in Back River particulate matter samples 
from October 2005.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
BR-1 BR-2 BR-3 BR-4
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
u
g
/
g
NP NP1 NP2 NP3 NP4 NP5 NP6 NP7 NP8 NP9 NP10
NP11 NP12 NP13 NP14 NP15 NP16
139
sampling events in the fall, in contrast to the data from March, the concentrations of the 
NPEOs were correlated to SS concentrations, r = 0.929 for September, and 0.648 for 
October, as opposed to -0.059 in March.  Although the tributaries were not sampled in the 
2004 events, SS contribution of the Back River watershed was estimated at 3531 tons/yr 
(MDE 2003); this value was used together with the annual average water discharge, 8.53 
x 10
10
 L/yr to obtain an average SS concentration of 41.4 mg/L, which is close to the 
values observed in the Back River site closest to the tributaries mouth (BR-1), 36.6 mg/L 
in September and 37.8 in October.  Analysis of water from the tributaries conducted in 
2001 provides evidence of NPEO presence; although only a few of the NPEs were 
analyzed, and only those in the dissolved phase, results in Table 5.2 suggest that the 
concentrations can be important and in a similar order of magnitude to those found in the 
estuary.  The NPEOs in these waters might originate from raw sewage leaks into the 
tributaries, and this would explain the presence of long-chain NPEOs. 
Evidence for the rapid biotransformation through the non-oxidative pathway (Fig. 
5.1) in the fall is provided by the changes in NPEO distribution in the estuary compared 
to the effluent, where the most abundant homologue is NP2EO, followed by NP1EO and 
NP in both the dissolved and particulate phases, whereas in the estuary the distribution 
order shifts to NP, NP1EO, and NP2EO (Fig. 5.6A).  This did not occur in March, when 
NPEO distribution in the estuary corresponds to the distribution in the effluent (Fig. 
5.6C).  Further evidence for NPEO transformation via the non-oxidative pathway in the 
estuary comes from the changes in the NP1-3EO/NP4-16EO ratio (total molar-based 
concentrations), which increased 21% from BR-2 to BR-4 in September, and more than 
three times in October.  Furthermore, the ratio changed from 12.1 to 1.9 from the final 
140
effluent to site BR-2 in September, and from 19.5 to 7.3 in October, again implying that 
the short-chain ethoxylates are removed faster than their long-chain homologues.
Evidence for the occurrence of the oxidative pathway is contradictory, because the ratio 
NP0-1EC/NP1-16EO increased almost three times in October form BR-2 to BR-4, but 
decreased almost 50% in September.  This difference cannot be explained by differences 
in dissolved oxygen concentrations, which were very similar (Table 5.1), but water 
temperature was approximately 10?C lower in October, which might slow NPEC 
degradation and result in higher ratios.  The ratio NP0-1EC/NP2-3EC from BR-2 to BR-4 
increased 8.6% in September and 21% in October, suggesting that the transformation of 
long- to short-chain NPECs might occur, albeit slowly.
Model limitations. The model proposed, although simple, explained the general 
behavior of the NPEs in the Back River.  The agreement between observed and modeled 
APE concentrations suggests that the main assumptions summarized in Eq. 5.1 and 
subsequent comments are adequate.  However, in order to make this into a truly 
predictive model, there are several limitations that need to be addressed.  Flow rates and 
other Back River parameters used for this model are yearly averages or estimates, 
whereas APE concentrations are ?snapshot? values.  A better characterization of the 
estuarine parameters and the temporal variations of APE concentrations should yield 
more exact results.  The inclusion of other environmental compartments, especially 
sediment, would also be necessary to fully understand the fate of the NPEs.  Although no 
sediments were collected in 2004 and 2005, analysis of samples collected in 2001 showed 
that the NPEOs are present in the Back River (Fig. 5.3), and the possibility of NPE 
exchange between sediment and water column needs to be addressed.  Additionally, a 
141
better characterization of all the NPE transformation products, such as long-chain 
NPECs, is necessary to fully understand the ultimate fate of these compounds.
OPE fate in the Back River.  As it was the case for the NPEs, OPE 
concentrations in the estuary were also proportional to the concentrations in the WWTP 
effluent (Figs. 5.7B and 5.7D), with the exception of the samples from September 2004 
where the concentration in the effluent was disproportionately higher than the 
concentrations in the Back River.  The general behavior of the OPEs was expected to be 
similar to the NPEs, therefore the same model was used.  OP0EC?s behavior in March 
was explained by the model with no sorption and a k = 0.03 d
-1
.  This value is two times 
higher than the value obtained for the NPECs, implying that OPEC is eliminated twice as 
fast as the NPECs measured.  In contrast, OP1-5EO matched the behavior of NP1-5EO 
more closely, with a k of 0.02 d
-1
.  OP, however, showed much higher K
d
 values than the 
OPEOs, or even NP, by almost one order of magnitude.  This value stems from the 
relatively high proportion of OP associated to the particulate, including the Back River 
WWTP effluent.  The average log K
oc
 value derived from our data, 6.13 ? 0.26, is also 
higher than values found by Ferguson et al. (2001) in Jamaica Bay, 5.18 ? 0.28, and by 
Johnson et al. (1998) in three different English rivers, 4.89 ? 0.97.  The reasons for these 
observations are not clear, but they contradict what would be expected from the 
difference in log K
ow
 between NP and OP.  OP in an independent effluent sample from 
February 2005, however, showed a similar behavior to NP as expected.
Data for the fall were limited, because only OP was observed in the estuary in 
October, and none of the OPEs were found in September (Figure 5.7A).  OP0EC was 
detected in both sampling events, however, and its behavior is analogous to that of the 
142
NPECs described above, being described by the hydraulic flow.  OP in October also 
behaves similarly to the NPEOs; although a spike in concentration was observed in BR-4, 
there is not enough information to assume that this was due to OP formation.  Because of 
the similarity of OP?s behavior to the NPEOs, it is likely that it is also entering the 
estuary from the tributaries; samples from the tributaries in 2001 also contained the 
OPEOs in significant concentrations (Table 5.2).  Finally, and also in close agreement to 
the NPEOs, biotransformation seems to be an important process in the fall, because 
relatively large amounts of OPEOs are released from the WWTP effluent but these 
cannot be detected in the estuary.  Also, OPEO distribution shifted from OP2EO being 
the most abundant in the effluent, to OP being the only compound detected in the estuary, 
whereas in March the distribution was similar in both.
Toxicological significance.  Because NP is more abundantly found in the 
environment than OP, and it is considered the most toxic of the NPEs, NP has been 
widely studied and more toxicological information is available for this compound than 
for any of the other APEs.  NP concentrations in the Back River (Table 5.3) were well 
below the range of 96-h LC
50
 values compiled by Servos (1999) for 22 different species 
of fish, 17-3000 ?g/L (although most values were between 100 to 300 ?g/L).  NP 
concentrations were also below the U.S. EPA proposed water quality criteria discussed in 
the introduction, although in March they were close to half of the chronic exposure limit 
for saltwater, 1.4 ?g/L (U. S. EPA 2003).
As for endocrine disruption effects, NP concentrations are also below the threshold 
concentration of NP for in vivo vitellogenin induction in rainbow trout, 10 ?g/L, as 
determined by Jobling and collaborators (1996).  Based on data from this study, the 
143
authors also suggested that the thresholds for NP2EO and NP0EC would be similar, and 
lower, 3 ?g/L, for OP.  If this were the case, OP and NP2EO concentrations in the Back 
River would also be below the threshold values, but NP0EC concentrations would be 
higher in March.  More recent studies, however, seem to indicate that NP0EC is not as 
potent as suggested.  Dussault et al. (2005) showed that NP0EC potency to elicit 
vitellogenin production in rainbow trout is only 3% that of NP, and Balch and Metcalfe 
(2006) calculated no observable effect concentrations (NOEC) for NP, NP1EO and 
NP0EC of 3, 35, and 2010 ?g/L respectively in Japanese medaka, using alteration to sex 
ratios and development of gonadal intersex as endpoints.  On the other hand, biota in sites 
such as Back River, where they are constantly exposed to low concentrations of the 
APEs, might show detrimental effects as suggested by long-term exposure (1 yr) tests by 
Ackermann et al. (2002), which showed that exposure to NP in concentrations as low as 1 
ug/L resulted in increased vitellogenin expression in rainbow trout, although no gonadal 
intersex or changes in sex ratios were observed.
Although most of the estrogenic activity in WWTP effluents has been traced to 
steroid estrogens (Johnson and Sumpter 2001), the possibility of additive effects was 
suggested by the work of Thorpe et al. (2001) with binary mixtures of NP and estradiol.  
In a more extensive study, Brian et al. (2005) confirmed the potential of a mixture of 
nonylphenol, octylphenol, estradiol, ethynylestradiol, and bisphenol A to act additively 
even at individual concentrations below those needed to elicit a response by the 
individual chemicals.  These observations suggest that the APEOs need to be included in 
any risk-assessment study addressing endocrine disruption in aquatic biota.
144
CHAPTER 6 ? OTHER APPLICATIONS: NP MIGRATION FROM PLASTIC TO 
BOTTLED WATER
This chapter has been published as:
Loyo-Rosales, J. E., Rosales-Rivera, G. C., Lynch, A. M., Rice, C. P., and Torrents, A. 
(2004), ?Migration of nonylphenol from plastic containers to water and a milk surrogate.? 
J. Agric. Food  Chem. 52(7), 2016-2020.
Reproduced with permission.  Copyright 2004, American Chemical Society.
6.1 Abstract
Nonylphenol (NP) is used as an antioxidant and plasticizer in some plastic 
products. After the discovery of its endocrine-disrupting potential, concern over human 
exposure to this chemical has increased.  Recently, a group in Germany estimated the 
average daily intake of NP from food (7.5 ?g/day), excluding water.  In the present study, 
NP, octylphenol (OP), and their respective ethoxylates (1 to 5) were measured in spring 
water bottled in three different types of plastic (HDPE, PET, and PVC).  NP was present 
in water from HDPE and PVC containers, at 180 and 300 ng/L respectively, which 
represent 4.8 and 8% of the value calculated by the German group assuming a 
consumption of 2 L of water per day.  OP was found in water from HDPE extracts in 
lower amounts, 12 ng/L, and neither the NP- nor the OP-ethoxylates were detected in any 
of the samples.  Attempts to measure these compounds in tap water were unsuccessful, 
145
probably because reaction with residual chlorine results in the formation of chlorinated 
by-products.  Migration of NP from HDPE containers to a milk surrogate was also 
evaluated; results indicate that the amounts of NP leaching into milk might be similar to 
those in bottled water.
6.2 Introduction
After the serendipitous discovery in 1991 of the estrogenic properties of p-
nonylphenol (NP) by Soto et al. (1991), NP and other related compounds ? such as 
octylphenol (OP) ? and their ethoxylated derivatives (nonyl- and octylphenol ethoxylates, 
NPEOs and OPEOs respectively) ? have been the subject of different studies addressing 
their toxicological properties and their ubiquitous presence in the environment (e.g.
Servos 1999, Bennie 1999).  Due to their uses as industrial and heavy-duty surfactants, 
they are introduced to the environment mainly from wastewater discharges.  Therefore, 
most of the studies have focused on their presence in aquatic media and their impact on 
aquatic biota (Bennie 1999).
However, according to Talmage (1994), one of the main industrial uses of 
alkylphenol ethoxylates is in plastic production; the production of 
tris(nonylphenyl)phosphite (TNPP), an antioxidant used in plastics, demands 
approximately 10% of the total NP used in the US (?Nonylphenol? 2001).  It is also 
acknowledged that some compounds present in plastic packaging have the capacity to 
migrate into foods and several studies have addressed this phenomenon using specific 
146
examples of plastic and chemicals of concern (e.g., Eberhartinger et al. 1990, Gilbert et 
al. 1986, McNeal et al. 2000).  Both the presence of NP and NPEOs in plastics and the 
widespread distribution of these compounds in the environment prompted Guenther et al. 
(Guenther et al. 2002) to estimate the average daily intake of NP from food by German 
consumers.  They analyzed a wide variety of food products for NP content.  Although 
they found NP in all their samples (ranging from 0.1 to 19.4 ?g/kg fresh weight), the 
amount of NP in the foodstuff did not correlate with the packaging material or the fat 
content of the food, suggesting that NP is introduced to food from a variety of sources, 
one of them being plastic wrapping and/or containers.
In order to have a complete assessment of the daily human intake of NP, it is also 
necessary to include the amounts ingested from water.  There has been a significant 
increase in the consumption of bottled water in the last decade; the market, with a value 
of 5.2 billion dollars, is projected to grow as much as 30% each year (Hollingsworth 
2002).  Bottled water is usually sold in plastic containers, normally polyethylene 
terephthalate (PET) and high-density polyethylene (HDPE); polyvinyl chloride (PVC) is 
also used in smaller quantities.  There are few studies (most of them Japanese) addressing 
the migration of NP from plastic bottles to water. Toyo?oka and Oshige (2000) reported 
the presence of NP in mineral water from PET bottles in concentrations that ranged from 
19 to 78 ng/L, but were unable to conclude if the presence of NP in the water was due to 
leaching from the plastic.
The objectives of this work were (1) to evaluate the presence of NP, OP, and their 
respective ethoxylates in bottled water to estimate daily human intake from water; (2) and 
to investigate whether the origin of these compounds is the plastic bottle. 
147
6.3 Materials and methods
Reagents.  NP was obtained from Schenectady International, Schenectady, NY 
(purity 
 95%, CAS 84852-15-3), and OP from Aldrich, Milwaukee, WI (97% purity, 
CAS 140-66-9).  NP2EO was an R&D product from Aldrich. NP1EO, NP3EO, NP4EO, 
NP5EO and the OP1-5EOs were purified in the laboratory by flash chromatography on 
silica gel from commercial mixtures as described elsewhere (Datta et al. 2002): NP1EO 
from Surfonic N-10 (Huntsman Chemicals, Austin, TX); NP3EO, NP4EO, and NP5EO 
from POE(4) nonylphenol (Chem Service, West Chester, PA);  the OP1-5EOs from 
POE(3) and POE(5) tert-octylphenol (Chem Service).  Purity of the standards was 
assessed by high performance liquid chromatography with fluorescence detection 
(HPLC-F) and was above 99% in all cases, except for the octylphenol monoethoxylate, 
OP1EO (94%).  Identity of the compounds was confirmed by liquid chromatography 
coupled to tandem mass spectrometry (LC/MS/MS).  A mixture of n-NP (Lancaster 
Synthesis, Windham, NH) and n-nonylphenol triethoxylate (n-NP3EO) synthesized by P. 
L. Ferguson (Ferguson et al. 2000) was used as an internal standard for quantitation.  
Ethanol was HPLC/spectrophotometric grade, 200 proof (Aldrich); dichloromethane 
(DCM) and methanol were high purity, pesticide grade from Burdick & Jackson 
(Honeywell International Inc., Muskegon, MI).  Deionized (18.2 megohm-cm), carbon-
free water (DI water) was obtained in the laboratory using a NANOpure water 
purification system (Barnstead International, Dubuque, IA).  Anhydrous Na
2
SO
4
, 
granular powder, was purchased from Mallinckrodt Baker Inc. (Paris, KY); both sodium 
148
hypochlorite (available chlorine  4%) and ammonium acetate (99.99+%) were from 
Aldrich.
Bottled water samples.  Spring water jugs (1 gallon) were acquired from three 
different local stores and brought to the laboratory.  Water in HDPE and PET containers 
was from the same bottler, although the water originated from different springs in the US; 
water in PVC containers was sourced from a different brand.  Special attention was taken 
to choose all the bottles for a single experiment from the same lot to minimize variation 
among samples.  The water jugs were stored at room temperature and in the dark until 
analysis, which occurred within 48 hours of acquisition.
Extraction.  Extraction of the analytes from the milk surrogate was performed by 
liquid-liquid extraction with DCM; whereas extraction from water was done by solid-
phase extraction (SPE) as described below.
The procedure for liquid-liquid extraction was as follows:  two 500 mL aliquots 
from each sample were measured and poured into two 1-L separation funnels.  Each 
aliquot was extracted three times with 50 mL DCM.  DCM extracts from both aliquots 
were mixed together and dried by passing the liquid through approximately 50 g of 
Na
2
SO
4
.  DCM was exchanged to methanol in a rotary evaporator and volume reduced to 
approximately 5 mL.  Methanol extracts were transferred to 15-mL graduated tubes and 
the volume was reduced to 0.5 mL under a gentle nitrogen stream, after which 0.5 mL of 
water was added.  The extracts were then filtered through an Acrodisc LC 13-mm syringe 
filter containing a 0.2 ?m PVDF membrane (Pall Gelman Laboratory, Ann Arbor, MI).  
Both syringe and filter were then rinsed with 0.5 mL of a 50:50 v/v methanol:water 
mixture that was added to the extract.  Finally, volume was adjusted to 1.5 mL with the 
149
methanol:water mixture.  Recoveries (average of 3 determinations ? SD) were NP 
78?3%, NP1EO 80?9%, NP2EO 89?11%, NP3EO 106?12%, and NP4EO 110?13%.
The SPE method is described in detail elsewhere (Loyo-Rosales et al. 2003).  
Briefly, ENV+ SPE cartridges (Isolute ENV+, 500 mg, 6 mL; International Sorbent 
Technology Ltd., Hengoed, UK) were pre-rinsed sequentially with 18 mL DCM, 12 mL 
acetone and 12 mL DI water in a vacuum manifold before passing the entire contents of 
the water jugs through them.  The cartridges were then dried for 2 hr by passing air 
through them and eluted sequentially with 12 mL DCM, 12 mL methanol and 12 mL 
acetone.  The collected solvents were evaporated under a gentle nitrogen flow and 
exchanged for methanol to a final volume of 0.5 mL.  Water was added and the solution 
was filtered as described above for liquid-liquid extraction.  Recoveries (average from 2 
determinations ? SD) from spiked bottled water in PET containers using SPE extraction 
were: NP 87?9%, NP1EO 90?8%, NP2EO 90?4%, NP3EO 97?8%, NP4EO 99?20%, 
NP5EO 92?17%, OP 76?4%, OP1EO 94?4%, OP2EO 100?2%, OP3EO 99?7%, OP4EO 
110?15%, and OP5EO 130?16%.
LC/MS/MS analysis.  After extraction, the samples were analyzed by 
LC/MS/MS as described by Loyo-Rosales et al. (2003).  A Waters 2690 XE separations 
module (Waters Corp., Milford, MA) coupled to a benchtop triple quadrupole mass 
spectrometer with an electrospray interface (Quattro LC, Micromass Ltd., Manchester, 
UK) was used for the analysis.  Chromatographic separation was achieved with an 
MSpak GF-310 4D column, 4.6 x 150 mm (Shodex, Shoko Co., Tokyo, Japan) at 60?C; 
injection volume was 10 ?L.  Mobile phase was composed of solvents A (50:50 v/v 10 
mM ammonium acetate in DI water:methanol) and B (100% methanol).  Initial 
150
conditions were 100% A; the amount of B was increased to 90% in 20 min, held for 8 
min, and increased to 100% in 2 min.  The column was then stabilized for 20 min at 
100% A; total run time was 60 min.  Flow rate was set at 0.2 mL/min and all of the eluent 
was allowed into the MS.  Source parameters were: capillary voltage 3.5 kV in 
electrospray positive (ES+) and ?2.9 kV in electrospray negative (ES- ); extractor voltage 
3 and 2 V respectively; RF lens 0.1 V in both modes; source and desolvation 
temperatures 140 and 400?C.  Nitrogen was used as nebuliser and desolvation gas (~80 
and 600 L/hr, respectively).  The photomultiplier was set at 650 V.  Acquisition was done 
in the multiple-reaction monitoring mode (MRM) in ES+ for the first 25 minutes of the 
run and then switched to ES- for 10 min.  For NP the parent ion was 218.9 m/z and its 
fragment 132.8 m/z; cone voltage was set at ? 40 V, and collision energy to 30 eV.  The 
reader is referred to Loyo-Rosales et al. (2003) for the ions, cone voltages and collision 
energies used for the rest of the compounds.  Analyte concentrations were calculated by 
the internal standard method using n-NP and n-NP3EO as ES- and ES+ internal standards 
respectively.  Six-point calibration curves were prepared in 100% A; analyte 
concentrations ranged from 20 to 700 ng/mL for OP, NP, OP1EO and NP1EO, and from 
6 to 200 ng/mL for the rest of the AP2-5EOs.  Peak integration and quantitation were 
performed automatically using MassLynx 3.5 and 4.0 (Micromass Ltd, Manchester, UK).
Migration of NP from plastic bottles to water.  This procedure was based on 
US Food and Drug Administration (FDA) recommendations for the estimation of 
substance migration from packaging materials to food (US FDA 2002).  The original 
water contained in 15 HDPE, 15 PVC and 6 PET jugs was either extracted using the SPE 
method and analyzed as described above or discarded, after which all the jugs were 
151
rinsed three times and then filled to the top with DI water.  The mouth of all the jugs was 
covered with Teflon tape before capping them to prevent contact with the caps.  The 
water from three HDPE, three PVC and two PET jugs were extracted by the SPE method 
and analyzed as described above (time 0). The remaining bottles were stored in a 
controlled-temperature growth chamber (PGR 15; Controlled Environments, Ltd., 
Winnipeg, Canada) at 40?C and analyzed at different intervals: for HDPE and PVC, three 
bottles for each type of plastic were analyzed after 48, 120, 240, and 360 hours; for PET, 
two bottles were analyzed after 120 and 240 hours.
Migration of NP from plastic bottles to milk surrogate.  This procedure was 
also based on FDA guidelines (US FDA 2002).  The original water from 27 HDPE jugs 
was discarded, after which all the jugs were rinsed three times with DI water and then 
filled to the top with a 10% v/v ethanol solution.  The mouth of all the jugs was covered 
with Teflon tape before capping them to prevent contact with the caps.  Three jugs were 
extracted with DCM and analyzed as described above (time 0).  Half of the bottles were 
stored in a controlled-temperature growth chamber (PGR 15; Controlled Environments, 
Ltd., Winnipeg, Canada) at 40?C, and half at 20?C.  Three bottles from each temperature 
were analyzed after 48, 120, 240, and 360 hours.
6.4 Results and discussion
Bottled water analysis.  Spring water bottled in HDPE, PET, and PVC 
containers was extracted using the SPE method above, and the extracts were analyzed for 
152
NP, OP, and their respective ethoxylates (n = 1 to 5).  Results are summarized in Table 
6.1.  NP was found in HDPE and PVC containers, whereas OP was present in all three 
types of container materials, albeit at lower concentrations.  The variation (expressed as 
RSD) in NP and OP concentrations among HDPE samples, 29 and 23% respectively, 
reflects differences among bottles in the lot rather than variability of the analytical 
method.  Variation in recovery experiments of spiked water was consistently lower ?
below 10% for both compounds (data not shown).
From the results above, it follows that the amount of NP ingested from water 
depends on the nature of the container.  Assuming an average consumption of 2 L of 
bottled water per day, a person would ingest an average of 360 ng/day of NP if all water 
intake came from HDPE jugs; this represents 4.8% of the average daily intake value for 
NP calculated by Guenther et al. (2002), 7.5 ?g/day.  This percentage could increase to 
8% if the water was from PVC jugs.  Although there are no data available for daily intake 
of NP from food in the US, a comparison of these values to the contributions of the 24 
different food groups reported by Guenther et al. (2002) in Germany suggests that water 
from HDPE and PVC containers could represent one of the most important individual 
sources of NP, only behind sausages, apples and tomatoes.  In contrast, water from PET 
containers would not significantly increase NP ingestion.
Tap water analysis and the effect of hypochlorite on analyte recovery.
Attempts were made to measure the analytes in laboratory tap water, but the analysis of 
spiked samples revealed that NP and OP were not recovered, while recoveries for the 
153
Table 6.1 Concentrations of NP and OP found in spring water bottled in three 
different plastic types (HDPE: high-density polyethylene; PET: polyethylene 
terephthalate; PVC: polyvinyl chloride).
concentration, ng/L
Sample NP OP
HDPE avg (n = 6) 180 12
HDPE SD 53 2.8
HDPE RSD, % 29 23
PET avg (n = 6) ND BQL
PVC (n = 12) 300 BQL
PVC SD 44 NA
PVC RSD, % 15 NA
ND: not detected
BQL: < 8 ng/L
154
ethoxylates (75 to 110%) were approximately 20% lower than those for bottled spring 
water.  No NPEOs or OPEOs were detected in the tap water samples analyzed.
Because of their phenolic nature, NP and OP undergo chlorine substitution 
reactions in the presence of hypochlorite, resulting in the rapid formation of diverse 
chlorinated by-products (Hu et al. 2002).  This might explain why NP and OP are not 
recovered from spiked tap water, which contains residual amounts of chlorine from the 
disinfection procedure.  To test this hypothesis, sodium hypochlorite (1.2 ppm) was 
added to DI water and spiked with NP, OP and their ethoxylates.  A reference solution 
(DI water spiked with the analytes) without hypochlorite was also prepared as reference.  
All the solutions were stored in the dark for 24 hr to allow sufficient reaction time.  After 
this period, all the samples were extracted using the SPE method described above and the 
extracts analyzed by LC/MS/MS.  As expected, NP and OP were not recovered from 
samples containing NaClO, while there was no significant difference in the recoveries of 
the NPEOs and the OPEOs between the chlorinated and non-chlorinated samples.  It is 
therefore important to consider the reactivity of NP and OP when measuring NP and OP 
concentrations in water treated with hypochlorite, which includes not only tap water, but 
also effluent from wastewater treatment plants, where NP is often measured.  The toxicity 
of the chlorination products is also of importance when conducting risk assessment for 
these compounds, because they have been found to elicit antiestrogenic effects (Hu et al. 
2002).
Migration of NP from plastic bottles to water.  The sole presence of NP in 
bottled spring water is not enough evidence to conclude that this compound migrated 
from the plastic jugs.  NP could have been present in the water itself, or remained as a 
155
residue of washing steps during the bottle manufacturing process.  Therefore, 
experiments based on FDA guidelines for migration testing (US FDA 2002) were 
conducted as described in the experimental section.  Although all samples were analyzed 
for NP, OP and their ethoxylates 1 to 5, only NP was found in amounts above 
quantitation limits (8 ng/L) in water from HDPE and PVC jugs; OP was detected in both 
types of plastic, but always below quantitation limits (8 ng/L) and no increasing trend 
was observed in this experiment.  Neither NP nor OP were found in extracts from water 
stored in PET containers, even after 240 hours; and none of the ethoxylates were detected 
in any of the samples, regardless of the nature of the plastic.
Results for NP migration from PVC and HDPE containers are shown in Figure 
6.1.  The levels of NP increased during the first hours, and tended to stabilize after 120 
hours at around 140 ng/L for PVC and 230 ng/L for HDPE.  Using these values, NP 
ingestion from bottled water would represent 6% of the daily intake calculated by 
Guenther et al. (Guenther et al. 2002) if all water consumption was from HDPE jugs, and 
4% for PVC, in contrast with 5 and 8% calculated above.  Interestingly, the amount of 
NP was higher for HDPE than for PVC in this case, whereas the analysis of the original 
water contained in these jugs showed higher NP values for PVC.  It was also surprising to 
find lower amounts of NP in the PVC bottles stored at 40?C than in the original water, 
which we assumed was stored at room temperature (the actual conditions during storage 
and transportations are unknown).  Both observations might be an indication of exposure 
to higher temperatures during transport or storage of the PVC bottles, which would result 
in an increase of NP migration from the plastic to the water.  Although these variations 
could also be caused by other factors (or a combination of them), such as (i) different 
156
Figure 6.1 Migration of NP from two types of plastic bottles (HDPE and PVC) to 
water over time. Error bars represent standard deviation of 3 determinations.
0
50
100
150
200
250
300
0 50 100 150 200 250 300 350 400
time, hr
N
P
 
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
n
g
/
L
PVC HDPE
157
types of water: DI for migration experiments versus spring water in the original analysis; 
(ii) storage time before analysis: water in HDPE bottles was analyzed 13 days after being 
bottled, whereas water from PVC bottles had been in storage for 27 days; (iii) spring 
water in the PVC jugs could have contained NP before being bottled. 
The FDA guidelines for migration testing of products stored at room temperature 
claim that testing at 40?C for 10 days reflects migration levels obtained after 6-12 months 
storage at 20?C (US FDA 2002).  Assuming that bottles are always stored at room 
temperature (which is very possibly a conservative assumption, especially during the 
summer months), a better estimation for NP intake from bottled water would require the 
use of new plastic jugs and the spring water being bottled, whose NP content should be 
known.  In any case, our estimates indicate that bottled water could contribute with up to 
8% of the daily NP intake estimated by Guenther et al. (2002).
Migration of NP from plastic bottles to milk surrogate.  HDPE jugs are also 
used to bottle milk.  Because milk contains fat and NP is relatively lipophilic (log K
ow
4.48, Ahel and Giger 1993) it is possible that NP migration from plastics to milk is larger 
than to water.  To test this, we used a 10% ethanol solution as a surrogate for milk, as 
suggested in the FDA guidelines (US FDA 2002).  These guidelines recommend storage 
at 20?C to simulate long-term storage of refrigerated foods.  We also conducted a test at 
40?C to compare with the results for migration to water.  After extraction with DCM, all 
the samples were analyzed for NP, OP and their ethoxylates.  As found for the water 
tests, only NP was found in concentrations above quantitation limits, OP was detected but 
below quantitation limits, and the ethoxylates were not detected.  Results for NP are 
summarized in Figure 6.2.  NP concentration increased only slightly with time in the 
158
Figure 6.2 Migration of NP from HDPE bottles to milk surrogate (10% ethanol 
solution) over time at two different temperatures (20 and 40?C). Error bars 
represent standard deviation of 3 determinations.
0
100
200
300
400
500
600
700
0 50 100 150 200 250 300 350 400
time, hr
N
P
 
c
o
n
c
e
n
t
r
a
t
i
o
n
,
 
n
g
/
L
20C 40C
159
containers stored at 20?C, whereas it increased 3.5 times after 15 days at 40?C.  Although 
this increment might be only due to an increase in NP?s solubility and diffusion rate from 
the plastic matrix to the interface with the solvent, an additional consideration is the 
possible presence of TNPP in the plastic, whose reactions with water and ethanol yield 
NP.  The rates of these two reactions would also increase with temperature and affect the 
final concentration of NP found in the milk surrogate.  As expected, the amount of NP 
found in the milk surrogate after 15 days at 40?C, 580?25 ng/L, was higher than in water, 
230?12 ng/L; presumably due to a higher affinity of NP for the milk surrogate.
Averaging the concentration of NP found in the milk surrogate at 20?C at all 
times, a value of 186?21 ng/L is obtained, which is very close to the concentration found 
in the spring water originally contained in the jugs (180?53 ng/L, see Table 6.1).  It 
would appear from these numbers that milk bottled in HDPE may not contribute 
significantly more NP than water packed in the same containers, probably because milk is 
normally kept at lower temperatures than water and it is stored for shorter periods of 
time.
160
CHAPTER 7 ? CONCLUDING REMARKS AND RESEARCH NEEDS
The ubiquitous presence of the APEs in water bodies, combined with their toxic 
effects and endocrine disruption potential to aquatic biota, has raised considerable interest 
in their environmental fate.  When the present work was initiated, the main obstacles to 
study the fate of these compounds were the limitations of the most commonly used 
analytical methods at the time, namely HPLC-F and GC/MS, and the lack of pure
analytical standards.  Therefore, the purpose of the initial part of this study, described in 
chapters 2 and 3, was the development of appropriate extraction and quantitation methods
for the APEs. As a first step, NP1-5EO and OP1-5EO were isolated to be used as more 
reliable analytical standards, whereas a characterized commercial mixture of NPEOs was 
obtained for use as a standard for the NP6-16EOs, and R&D products for the NP0-1EC 
and OP0EC.  Using these standards, analytical methods based on isotope dilution
LC/MS/MS were developed that allowed the determination of a comprehensive set of the 
APEs?NP, OP, NP1-16EO, OP1-5EO, NP0-1EC and OP0EC quantitatively, and OP6-
16EO, NP2-3EC and OP1EC qualitatively?with low detection limits compared to other 
existing methods, and with the added selectivity inherent to the tandem mass 
spectrometry approach.
The second part of this work addressed the fate of the APEs in wastewater 
treatment plants, specifically the differences in treatment efficiency between American 
and European WWTPs, and the effects of tertiary treatment on removal.  As the results in 
161
chapter 4 show, APE removal rates were similar in the three plants studied in spite of two 
of the plants using tertiary treatment and one only secondary.  Furthermore, mass 
balances performed in the three WWTPs showed that the percentage of NPEs leaving the 
plants was comparable to that found in two Swiss plants with secondary treatment.  APE 
removal rates were strongly correlated to water temperature and had no correlation with 
suspended solids or organic carbon removal.  NPEOs were eliminated in higher rates than 
the OPEOs, apparently as a consequence of greater partition to the solids and removal 
with the sludge.  Additionally, analysis of raw wastewater from the Chicago metropolitan 
area suggested that municipal sources might still be important sources of the APEOs in 
spite of their substitution in some household products.
The fate of the APEs in Back River (Chapter 5) was studied in the third part of the 
study.  The nature and concentrations of the APEs present in river water followed closely 
the composition of the WWTP effluent, especially the APECs?that were the most 
abundant APEs present.  A simple model based on average net estuarine flow and the 
WWTP discharge, and also including dispersion and deposition, closely predicted 
observed NPE concentrations.  The contribution of biotransformation reactions was 
estimated as a fitting parameter of the model; these transformation rates were almost 
identical to those obtained independently in two sites in the Netherlands by a separate 
group.  However, at higher temperatures and during rain events, the concentrations of 
APEOs in the river appear to be controlled by flow from the tributaries, whereas the 
APECs still originate in the WWTP.
In the last part of this work, the presence of APEs bottled water in three different 
types of plastic was determined.  NP was found in PVC and HDPE (although not in PET)
162
in concentrations that might represent a considerable portion of the amounts ingested 
through food. Leaching studies in the bottles provided evidence that the plastic is the 
source of NP in the water.
Research needs
Although a wide array of analytical methods is available for the quantitation of 
APEs in the environment, there is a need for a systematic study of their advantages and 
disadvantages and their performance parameters, such as inter-laboratory precision, 
before any of the methods can be used as a standard quantitation methodology.  This 
would be especially important in the United States if the water quality criteria proposed 
by EPA were to be enforced.  Moreover, there is a lack of analytical standards that 
hinders the study of the APEs.  Not only are NP and derivatives mixtures or isomers due 
to the nature of NP?s synthetic method, but the APEOs are only available as mixtures, 
that may or may not be well characterized, and very few standards exist for the APE 
transformation products, such as APECs.
Although there is ample evidence of the APEs ubiquity in the environment, their 
behavior and ultimate fate are not yet fully understood.  It has been established that, after 
being released in natural waters, the APEs will partition into solids and undergo multiple 
transformation reactions, but their ultimate fate has not been thoroughly elucidated.  This 
is closely related to the lack of understanding of all the possible degradation pathways 
and their kinetics.  It is not even clear if all the possible, or at least most commonly 
occurring, transformation products have been identified, and it is not well known in 
163
which environmental compartments the different degradation pathways would occur; e.g.
some transformation mechanisms might be more prevalent in fresh water than in 
WWTPs.
The identification and study of the most abundant APE transformation products 
must be linked to the risks they may represent to exposed organisms, which tend to be 
aquatic biota.  There are many studies on NP and the NPEOs, but more attention should 
be placed on the APECs, because they tend to be the most abundant transformation 
products released by WWTPs as discussed in Chapter 4.
Finally, although wastewater treatment plants might be the APEs most common 
point of entry to the environment, they are not the only source.  APEs are used in a wide 
array of products, such as aircraft deicing fluids, which are not always discarded into
sewers and/or treated, and might represent significant releases of APEOs into the 
environment at least seasonally.  Besides the deicing applications in airports, the use of 
these surfactants as adjuvants in pesticide formulations comes to mind.
164
APPENDIX 1 ? BACK RIVER MODEL
Back River was modeled as an incompletely mixed system at steady state using a 
control-volume approach (Chapra 1997).  The estuary was divided into n = 1200
segments with constant length, dx = 10 m, and surface area calculated for every segment.  
The mass balance equation for segment i is:
()()
iiiiiiiiiii
ii
ii
ii
iii
cVkccEccE
cc
Q
cc
QW ++

	





 +


	





 +
+=
++
+
+

 11,1,1
1
1,
1
,1
''
22
0 Eq. A1 
where c is the concentration of the chemical in the corresponding segment [mmol/m
3
], W
the mass loading of the chemical to the segment [mmol/d], Q is the flow rate [m
3
/d], E?
the bulk dispersion coefficient [m
3
/d], k a lumped first-order reaction rate constant for the 
chemical [d
-1
], and V the volume of the segment [m
3
].
Dirichlet boundary conditions were used to for the first, 0, and last, n, segments.  
At segment 0, the interface with the tributaries was assumed to be advective only; i.e. no 
dispersive mass transport occurred; whereas the boundary of cell n with the Chesapeake 
Bay was treated as an open boundary, with the concentration in the Bay assumed to be 0.
The system of n equations was solved using MatLab Version 6.5.0.1924 Release 
13 (The MathWorks, Inc., Natick, MA) with the script provided in Fig. A1.
Model parameters
Estuary dimensions.  The total length of the Back River was calculated from 
topographic maps and rounded to 12000 m.  Depths were calculated for each segment of 
the model assuming geometric dimensions for the estuary (Fig. A2), and depths of 1.5 m 
165
at the head of the estuary, 2.7 m at 6000 m, and 7.6 m at the mouth (MDE 2005).  For a 
segment situated at length x, if x  6000, depth, h(x), was calculated as
( )
5.1
6000
5.17.2
)( +

= xxh Eq. 2a
and for x > 6000,
( )
7.2
6000
7.26.7
)( +

= xxh Eq. 2b
Widths were calculated in an analogous manner, assuming that the estuary was 800,
1000, 1500, 1600, 1700, and 1800 m at a distance of 0, 2000, 3000, 6000, 8400, and
12000 m respectively from the estuary head.
Flow rate.  The net estuary flow, 6.5 x 10
5
 m
3
/d, was assumed to be the sum of 
the WWTP discharge, 110 mgd (4.2 x 10
5
 m
3
/d) plus the average watershed discharge, 
2.3 x 10
5
 m
3
/d, which was assumed to occur mainly through Back River?s tributaries 
(MDE 2003).
Mass loadings.  Total APE mass loadings to the Back River were estimated using 
the water flow rates above and the APE concentrations measured in the WWTP effluent 
and the tributaries.
Lumped reaction rate constant.  The first-order reaction rate constant k [d
-1
] in 
this model is a lumped parameter that includes all the processes other than advection and 
dispersion that may affect the concentration of the chemical.  One such process is 
biotransformation, and the specific constant for this process is noted as k
bio
.  When 
partition to solids exists, k is the sum of k
bio
 and k
s
, the latter defined as:
h
k
s
s

= Eq. 3
166
where 
s
 corresponds to the apparent settling velocity of the particles [m/d], and h is the 
segment depth as defined before.  Note that elimination through settling with the solids 
applies only to the fraction of the chemical sorbed to the particulate that is defined as
mK
mK
F
d
d
p
+
=
1
Eq. 4
where F
p
 is the fraction of the chemical in the particulate, K
d
 is the sorption coefficient 
for the chemical [L/g], and m the suspended solids concentration [g/L].
Dispersion coefficient.  An average dispersion coefficient, E [m
2
d
-1
] for Back 
River was estimated assuming constant physical characteristics for the estuary and using 
the slope obtained from a plot of the salinity data in Table 1 according to equation 5 
(Chapra 1997)
x
E
U
s
s
=
0
ln Eq. 5
where s is the salinity [ppt] at distance x from the river?s boundary with the Bay (x = 0), 
s
0
 the salinity at the boundary, and U the net estuarine velocity, 110 m d
-1
.  The latter was 
estimated from the net estuary flow and the average cross-sectional area of the river, 5.8 
x 10
3
 m
2
.  The estimated value for E, 8.1 x 10
5
 m
2
d
-1
, is approximately one order of 
magnitude lower than the value reported by Chapra (1997) due to the higher value used 
for U in his case.
167
Figure A1.  MatLab code to calculate the steady state concentrations of a chemical 
in the Back River.
% BRcentered_diff.m
% Back River as a mixed flow reactor with Dirichlet and Neumann boundary conditions;
% solved by centered-difference control-volume approach.
% Jorge Loyo
clear
% River characteristics and other inputs
Qt = 2.3e5; % [m^3/d] flow from the tributaries
Qp = 4.2e5; % [m^3/d] flow from the WWTP
Qtot = Qt+Qp; % [m^3/d] total flow (WWTP + tributaries)
Ltot = 12000; % [m] total length of the river
loc = 2500; % [m] location of the WWTP in the estuary
E = 8.1e5; % [m^2/d] calculated in dispersion coefficient.xls file
kbio = 0.00; % [d^-1] biotransformation rate
vs = .25; % [m/d] settling velocity
frpart = 0.38; % Fraction of NPEOs sorbed to particulate
frcarboxy = .16; % Fraction of the total NPs that are subject to sorption, i.e. EOs.
fr = frpart*frcarboxy;
dx = 10; % [m] length of modeling interval
% Loads
ceff = 0.380; % [mmol/m^3] compound concentration in the plant's effluent
ctrib = 0; % [mmol/m^3] compound concentration in the river's tributaries
Wp = ceff*Qp; % [mmol/d] load from the WWTP to the river
Wt = ctrib*Qt; % [mmol/d] load from the tributaries to the river
% Define variables
X = [0:dx:Ltot]; % [m]
n = length(X);
B=zeros(1,n); % width vector
H=zeros(1,n); % depth vector
k=zeros(1,n); % deposition rate vector
% Calculate widths [m] from width.m
M = [800 1000 1500 1600 1700 1800]; % [m] 
P = [0 2000 3000 6000 8400 12000]; % [m] points at which the widths in M were measured
for j=1:5
    if j==1
        i=(P(j)+1):find(X==(P(j+1)));
B(i)=((M(j+1)-M(j))*(X(i)-P(j))/(P(j+1)-P(j)))+M(j);
else
        i=find(X==P(j)):find(X==(P(j+1)));
B(i)=((M(j+1)-M(j))*(X(i)-P(j))/(P(j+1)-P(j)))+M(j);
end
end
168
% Calculate depths [m] from depth.m
for i=1:n
    if X(i)<=6000
      H(i)=(X(i)*1.2/6000)+1.5;
    else H(i)=((X(i)-6000)*4.9/6000)+2.7;
    end    
end
% Calculate bulk diffusion coefficients [m^3/d], deposition rates [day^-1],
% and volumes for each cell
for i=1:n
    Ep(i)= B(i)*H(i)*E/dx/2;
    k(i) = kbio + vs*fr/H(i);
    V(i) = B(i)*H(i)*dx;
end
% Define Q for each cell
for i=1:n
    if X(i)<loc
Q(i)=Qt;
    else
        Q(i)=Qtot;
end
end
% W vector
W = zeros(n,1);
W(1) = Wt; % load in the first cell from tributaries
W(find(X==loc)) = Wp; % load in the cell where the WWTP discharges
% A matrix
A = zeros(n,n);
A(1,1) = Ep(2) + k(1)*V(1);
A(1,2) = Q(2)/2 - Ep(2);
A(n,n) = ((Q(n)-Q(n-1))/2) + Ep(n-1) + Ep(n) + k(n-1) * V(n-1);  
A(n,n-1) = -(Q(n-1)/2 + Ep(n-1));
for i = 2:n-1,   
A(i,i-1)=-(Q(i)/2 + Ep(i));
    A(i,i) = ((Q(i+1)-Q(i))/2) + Ep(i) + Ep(i+1) + k(i) * V(i);
A(i,i+1) = -Ep(i+1) + Q(i+1)/2;
end
Ainv = inv(A);
c = Ainv*W;   
169
Figure A2.  Schematic representation of the Back River used to estimate the 
estuary?s depth at distance x from the head of the estuary.
0 6000 12000
1.5
7.6
2.7
Horizontal scale: 1 cm = 1000 m
Vertical scale: 1 cm = 1 m
x
170
APPENDIX 2 ? ANCILLARY AND RAW DATA
Table A1 Ancillary data (suspended solids concentration, TSS; temperature, T; pH; 
and dissolved organic carbon, DOC) for the samples from the three wastewater 
treatment plants and the Chicago sewers described in Chapter 4.
Sample
TSS, 
mg/L
T, ?C pH
DOC, 
ppm
Back River WWTP
Sep 2004
Influent 148 23.9 8.1 64
Prim eff 111 24.5 8.5 71
Sec eff 7.5 24.7 8.6 24
Fin eff 2.5 24.8 8.2 27
Oct 2004
Influent 172 20.3 7.9 115
Prim eff 199 20.5 8.1 77
Sec eff 4.0 20.5 8.1 26
Fin eff 2.1 20.8 8.4 24
Feb 2005
Influent 166 15.0 7.8 90
Prim eff 101 13.0 7.8 79
Sec eff 39 13.0 7.7 29
Fin eff 14 14.0 7.7 19
Mar 2005
Influent 283 14.0 7.9 73
Prim eff 141 14.0 8.0 76
Sec eff 13 14.0 8.2 29
Fin eff 4.9 14.0 8.3 32
Blue Plains WWTP
July 2004
Influent 90 25.7 8.9 23
Prim eff 88 25.6 8.9 16
Sec eff 18 24.1 9.0 13
Tert eff 3.5 26.0 7.8 7.5
Fin eff 0.81 24.7 7.9 7.3
Aug 2004
Influent 146 25.8 8.8 23
Prim eff 166 25.9 8.7 17
Sec eff 15 25.4 8.5 14
Tert eff 4.4 25.9 7.9 8.5
Fin eff 1.5 25.6 6.9 8.4
171
Feb 2005
Influent 171 14.0 8.0 32
Prim eff 131 14.0 8.0 12
Sec eff 49 14.0 7.9 14
Tert eff 4.4 15.0 8.1 9.1
Fin eff 3.0 15.0 7.7 11
Mar 2005
Influent 236 14.0 7.7 25
Prim eff 62 14.0 7.5 20
Sec eff 54 14.0 7.7 15
Tert eff 3.7 14.0 7.7 12
Fin eff 2.0 14.0 7.5 10
Calumet WWTP
Mar 2005
Influent 75 9.9 7.7 42
Prim eff 51 9.1 7.9 33
Sec eff --- 8.8 7.9 25
Fin eff 14 9.3 7.7 18
Aug 2005
Influent --- 23.0 --- ---
Prim eff 53 23.3 8.2 55
Sec eff --- 24.0 8.0 20
Fin eff 3.9 23.5 8.2 17
Chicago sewers
Industrial 163 21.0 8.6 108
WWTP Influent 90 18.0 8.1 68
Residential 1 159 16.9 8.2 58
Residential 2 78 18.3 8.7 31
Commercial 115 20.4 8.2 129
Chicago 2 68 --- 8.2 39
Chicago 3 22 --- 8.2 16
Oak Lawn 1 517 --- 8.1 78
Oak Lawn 2 145 --- 8.1 86
Bridgeview 130 --- 8.4 68
172
Table A2 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC 
in the 2004 sampling events in Back River WWTP.
Concentration, ng/L
Sep 2004 Oct 2004
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 43451 33104 548 446 19564 35085 460 484
NP1 33283 31620 704 636 17035 28508 905 965
NP2 24993 19466 1695 1567 17268 22485 2064 1763
NP3 24507 16447 435 518 21548 20297 354 388
NP4 27676 20003 155 203 30343 25723 123 125
NP5 27054 23040 41 74 30734 29852 62 56
NP6 39948 38684 ND 34 45204 51104 59 ND
NP7 41786 42221 ND ND 47455 57123 60 ND
NP8 46314 50516 ND ND 56953 67582 ND ND
NP9 50570 52454 ND ND 57300 69470 ND ND
NP10 47268 52305 ND ND 57809 68952 59 20
NP11 43517 48499 ND ND 52081 66429 ND ND
NP12 31995 35599 ND ND 37123 45347 ND ND
NP13 23043 27004 ND ND 27192 36732 ND ND
NP14 5470 8137 ND ND 8092 12074 ND ND
NP15 2385 1486 ND ND 2689 2257 ND ND
NP16 13302 14365 ND ND 6657 14711 9 21
OP 999 780 ND 81 1320 2198 21 36
OP1 468 536 BQL 113 862 1599 34 57
OP2 329 303 30 276 691 1085 108 98
OP3 332 313 BQL 65 724 1267 18 18
OP4 355 238 ND BQL 771 1140 BQL BQL
OP5 326 268 ND ND 908 1328 ND ND
NP0EC 76 899 1381 1481 262 1391 213 3543
NP1EC 204 3541 6110 5816 748 6983 714 14750
OP0EC ND 59 157 159 26 98 79 428
173
Table A3 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2004 
sampling events in Back River WWTP.
Concentration, ng/g
Sep 2004 Oct 2004
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 380083 491544 30317 45500 208955 197151 29697 33966
NP1 269231 332552 23250 33386 ND ND 67318 38655
NP2 156856 187594 29017 39000 48245 140004 77682 69621
NP3 112530 104094 8500 8957 46624 68475 13879 12638
NP4 97348 96940 2817 3243 46109 73759 5379 4000
NP5 62594 83449 1233 1143 40248 63796 3924 1897
NP6 95810 146060 3000 2057 66373 103945 6167 2552
NP7 85974 138414 2500 BQL 69646 88832 6485 1983
NP8 77626 134784 3000 BQL 61510 100484 7076 2328
NP9 70383 132994 2217 BQL 64937 94554 6803 BQL
NP10 59971 117339 2450 BQL 47727 83649 6394 BQL
NP11 40636 84195 2417 BQL 39374 61156 5682 2172
NP12 30910 59293 2333 1129 24616 48971 4909 2138
NP13 19539 40648 BQL 1214 17080 32962 4030 2069
NP14 11449 22881 1450 814 8984 23293 2879 1621
NP15 7733 16577 1400 714 5052 10962 2258 1534
NP16 4808 11912 1300 571 2088 10089 1364 1741
OP 12804 17843 BQL 3914 8248 11507 6288 BQL
OP1 5206 8314 ND ND ND ND ND ND
OP2 2385 3313 blk 1657 1369 5599 4000 2500
OP3 1854 2478 ND ND 1221 3510 BQL ND
OP4 2169 2321 ND ND ND 3368 BQL ND
OP5 1057 1508 ND ND 1046 2421 ND ND
174
Table A4 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC 
in the 2004 sampling events in Blue Plains WWTP.
Concentration, ng/L
Jul 2004 Aug 2004
Comp Influent
Prim 
eff
Sec 
eff
Tert 
eff
Fin 
eff Influent
Prim 
eff
Sec 
eff
Tert 
eff
Fin 
eff
NP 4118 5107 11078 3170 613 4418 3652 5562 167 130
NP1 8015 7386 34369 8149 1382 8800 8205 18298 844 585
NP2 9231 9849 31302 5189 1381 8893 6257 8362 298 241
NP3 12084 7163 10813 1669 614 6646 3642 2711 96 86
NP4 16974 8780 5529 583 191 5744 2711 967 28 26
NP5 23554 11569 5261 333 121 7007 2263 533 12 11
NP6 41138 18365 5227 93 ND 10165 2806 542 ND ND
NP7 40296 21763 4106 51 ND 10027 2367 325 ND ND
NP8 46064 23968 3844 43 ND 10864 2342 204 ND ND
NP9 45744 24032 3382 ND ND 10067 2186 170 ND ND
NP10 41316 23627 2958 ND ND 9082 1953 146 ND ND
NP11 34674 20049 3141 ND ND 7455 1631 141 BQL BQL
NP12 28390 16493 3089 ND ND 5515 1297 135 BQL BQL
NP13 20072 10608 3285 ND ND 3598 912 123 ND ND
NP14 10386 5749 1941 6 4 2280 639 97 2 BQL
NP15 4054 2141 394 1 ND 898 270 49 BQL BQL
NP16 3475 1893 536 5 3 764 229 50 1 ND
OP 397 844 899 101 37 520 408 342 11 BQL
OP1 432 711 1819 302 86 448 360 918 BQL BQL
OP2 309 537 2041 196 81 444 378 409 7 5
OP3 602 521 553 59 26 356 253 125 BQL BQL
OP4 564 393 246 15 7 234 144 28 ND ND
OP5 773 555 286 7 3 244 145 29 ND ND
NP0EC 462 1048 7080 15953 8410 1072 3261 22871 8164 7196
NP1EC 965 1651 17100 26185 15370 2075 6022 19392 18718 18398
OP0EC 49 97 601 919 475 82 226 1662 337 291
175
Table A5 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2004 
sampling events in Blue Plains WWTP.
Concentration, ng/g
Jul 2004 Aug 2004
Comp Influent
Prim 
eff Sec eff
Tert 
eff Fin eff Influent
Prim 
eff Sec eff
Tert 
eff Fin eff
NP 121478 179436 177473 40498 29876 218952 316390 419368 11607 16703
NP1 122911 224439 838240 172422 343810 376266 715913 1881589 84338 289822
NP2 147722 199024 711477 83989 117509 203441 197690 631524 18645 47554
NP3 148725 112844 207493 24646 36958 102744 59630 195304 4174 8414
NP4 147994 101540 83349 5749 11601 66835 30983 67984 1174 1805
NP5 233920 126925 69096 1818 BQL 59815 27248 44528 BQL ND
NP6 335317 197015 69776 BQL ND 125638 49774 68763 ND ND
NP7 324669 203767 53134 BQL BQL 131341 46597 50594 ND ND
NP8 337069 194503 46029 BQL ND 144456 48770 40596 ND ND
NP9 307793 197892 39601 BQL ND 151842 50713 35986 ND ND
NP10 266609 183381 35242 BQL ND 143362 47470 34067 ND ND
NP11 218033 149115 33684 ND ND 131279 43191 36436 ND ND
NP12 149802 111499 32063 ND ND 105530 38015 37889 685 BQL
NP13 98099 78488 28663 ND ND 79863 26196 35366 ND ND
NP14 75043 56996 27405 351 ND 58204 22856 32603 496 BQL
NP15 39629 29890 18221 ND ND 13646 6612 14296 327 457
NP16 25584 24790 15727 ND ND 29556 9154 18948 217 361
OP 3252 5089 ND ND ND 12166 10815 9129 ND ND
OP1 3445 5880 18146 ND ND 10700 21759 41737 ND ND
OP2 2792 4174 15161 ND ND 3594 3351 13042 BQL BQL
OP3 2357 1859 2934 766 BQL 2003 1419 4068 ND ND
OP4 2604 1995 1979 BQL ND 1366 784 992 ND ND
OP5 3565 2438 1078 ND ND 1538 821 1147 ND ND
176
Table A6 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC 
in the 2005 sampling events in Back River WWTP.
Concentration, ng/L
Feb 2005 Mar 2005
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 12700 16900 4020 3240 2580 9740 1560 1570
NP1 12700 12400 13300 12200 2450 5100 7710 6250
NP2 24000 16500 9080 6810 3670 6530 5600 5550
NP3 36700 21900 3410 2710 6030 8540 2040 1800
NP4 45600 31900 1830 1310 9460 13300 995 1010
NP5 52800 45500 1360 1370 15300 20300 748 736
NP6 70600 70800 662 722 24300 32400 231 303
NP7 74600 81500 309 348 29500 38400 161 243
NP8 77800 93400 200 248 35900 44900 164 254
NP9 78700 97400 130 174 40500 48600 154 280
NP10 73400 97800 108 149 38600 47500 171 315
NP11 61800 84900 95.2 145 34300 41900 182 370
NP12 46900 66100 82.8 117 24500 32700 211 401
NP13 34000 47000 59.2 94.8 18100 23300 187 400
NP14 13700 19100 37.6 66.6 5960 8740 168 295
NP15 2330 2190 13.4 8 1080 1350 82.8 93.7
NP16 1440 1900 22.3 14 568 1240 76.8 78
OP 848 1010 347 262 249 631 112 142
OP1 575 700 1040 748 129 313 628 582
OP2 599 771 581 504 235 390 392 481
OP3 986 981 184 152 450 559 120 135
OP4 1240 1200 59.5 57 671 839 25.1 32.2
OP5 1160 1470 27.3 28.7 1010 1070 11.2 14.7
NP0EC 504 3190 43000 57200 228 3020 38200 38400
NP1EC 1150 4940 36400 57500 572 6530 61300 54700
OP0EC 30 356 5240 6090 BQL 461 5860 6210
177
Table A7 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005
sampling events in Back River WWTP.
Concentration, ng/g
Feb 2005 Mar 2005
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 58500 205000 113000 138000 13200 134000 30800 45200
NP1 98700 123000 36200 63100 19700 104000 233000 221000
NP2 136000 137000 213000 264000 40100 91600 109000 180000
NP3 85100 113000 56900 61400 43000 80700 26900 43700
NP4 92600 154000 17700 21900 57400 92900 13000 22900
NP5 80400 155000 7350 12100 64100 106000 7200 13400
NP6 89400 218000 9720 15000 99500 171000 4770 8400
NP7 75400 200000 6270 8200 96900 169000 2660 4240
NP8 74000 191000 4430 6070 92600 167000 3170 5780
NP9 68600 174000 3770 5040 84300 160000 3370 7370
NP10 58900 142000 3380 4700 71900 134000 3470 7810
NP11 47100 108000 2700 3870 48900 93100 3920 9550
NP12 33800 77000 2270 3340 30900 72700 4210 11200
NP13 24300 53700 1700 2530 21400 49100 4150 11200
NP14 12900 26100 1250 1800 12000 24200 3810 9680
NP15 7820 16300 642 973 7380 15800 2600 6880
NP16 4000 6640 507 888 3780 8480 2220 5770
OP 1900 3320 12000 10900 2080 7990 12400 34700
OP1 2500 4670 3060 3930 971 6320 6780 8510
OP2 1950 2700 8240 4870 1330 2650 3000 3850
OP3 1240 2020 2490 2500 1560 2630 1020 1530
OP4 1680 3870 916 898 2280 3450 ND ND
OP5 1170 2760 459 648 2410 3350 BQL ND
178
Table A8 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC 
in the 2005 sampling events in Blue Plains WWTP.
Concentration, ng/L
Feb 2005 Mar 2005
Comp Influent
Prim 
eff
Sec 
eff
Tert 
eff
Fin 
eff Influent
Prim 
eff Sec eff
Tert 
eff
Fin 
eff
NP 15600 7450 9750 1170 1140 10400 4560 13200 10300 1740
NP1 19400 9360 45600 5320 5700 12700 8310 109000 31700 8760
NP2 13200 15100 17200 6330 6820 8930 6180 14900 11100 2150
NP3 13400 8520 8230 3830 4020 9130 9190 5730 7800 1790
NP4 23400 13600 5110 1740 2890 15600 15500 4100 3030 721
NP5 34800 18800 4750 700 1190 28500 22500 4700 1230 278
NP6 51600 29000 2690 237 654 43000 35000 1330 124 28.4
NP7 53600 32500 2240 179 589 49500 39800 608 64.9 ND
NP8 78000 41200 2610 203 733 60500 48700 595 60.8 ND
NP9 78900 42700 2450 127 434 66200 48800 463 52.1 ND
NP10 84100 44800 2330 104 352 67400 47900 370 45.4 ND
NP11 67900 41300 2360 92 335 57500 41100 422 36.5 ND
NP12 50200 32700 1660 79.7 307 44300 31200 291 34.7 ND
NP13 40700 20500 1630 53.4 233 34600 24900 252 29.5 ND
NP14 18900 7610 830 44.5 143 18700 14900 179 21.9 ND
NP15 1340 862 34.8 ND 11.7 2230 2120 38.5 3.68 1.65
NP16 1750 1290 35.7 15.2 26.9 3340 2620 112 9.07 5.85
OP 1250 509 850 86 126 886 526 1140 812 166
OP1 1180 400 1960 277 358 784 603 5680 1460 468
OP2 1140 778 917 497 636 575 541 559 417 118
OP3 731 512 165 152 202 566 602 102 105 35.7
OP4 1100 734 116 72 145 841 913 86 63.9 15.7
OP5 601 814 74.7 10.6 27.6 493 487 28.9 2.92 BQL
NP0EC 379 499 47400 19200 23000 466 1410 80400 60300 48300
NP1EC 898 980 34000 27900 32700 778 1430 26600 46900 27000
OP0EC 34.2 50.4 3680 1650 1900 40.8 153 7390 4460 4440
179
Table A9 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005 
sampling events in Blue Plains WWTP.
Concentration, ng/g
Feb 2005 Mar 2005
Comp Influent
Prim 
eff Sec eff
Tert 
eff Fin eff Influent
Prim 
eff Sec eff
Tert 
eff Fin eff
NP 115000 80300 206000 48600 72000 51300 37300 275000 310000 78300
NP1 113000 113000 1020000 174000 444000 16700 98200 1800000 651000 476000
NP2 96300 280000 245000 122000 373000 25200 78500 296000 224000 124000
NP3 58200 68600 60400 47600 127000 23000 59600 83500 107000 59000
NP4 68300 60100 29600 18800 66500 39900 85500 51400 31900 24700
NP5 76900 48900 23500 7130 21500 51200 120000 36400 11100 12600
NP6 96200 64900 14600 6150 16500 76900 162000 16600 6380 11500
NP7 85300 70200 12500 4380 11100 84800 174000 10000 6900 8290
NP8 97900 70900 14200 6390 15900 83800 174000 10300 7380 10300
NP9 93700 69100 12600 5980 8900 74500 160000 8030 8090 12900
NP10 83300 67100 11600 5040 7270 69000 139000 6760 7230 11000
NP11 65900 52900 11000 4400 6630 48300 109000 5760 5380 8980
NP12 52400 43800 9200 3490 5560 39100 80300 4380 4310 7090
NP13 40200 33300 7620 2780 4860 27200 53700 3630 3110 4990
NP14 22300 22200 5740 2130 3870 16100 32000 2600 1970 3160
NP15 13100 11600 1860 1360 2590 7510 17300 1180 1240 1980
NP16 7150 8300 2870 1060 2450 4240 12200 1360 798 1270
OP 3520 4960 7250 ND ND 2880 3140 15900 9020 7460
OP1 2870 4650 20400 BQL 6160 ND 2810 63600 10700 7500
OP2 2390 8630 3700 1830 7080 1070 2170 5980 1870 BQL
OP3 1390 2620 819 1330 2720 1070 1660 1450 1050 1200
OP4 1710 2270 339 BQL BQL 1430 2160 811 ND ND
OP5 1300 1940 393 BQL ND 1420 2730 443 ND ND
180
Table A10 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and 
OP0EC in the 2005 sampling events in Calumet WWTP.
Concentration, ng/L
Mar 2005 Aug 2005
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 4870 4960 2110 1340 --- 33500 663 368
NP1 4810 4050 1200 3760 --- 33000 485 733
NP2 6870 7380 4790 10600 --- 22600 635 890
NP3 9390 8630 1620 3680 --- 11100 206 280
NP4 19000 19000 1020 1710 --- 11800 151 134
NP5 31700 26900 877 1550 --- 12400 138 99.8
NP6 49800 39400 343 636 --- 19600 112 78.1
NP7 58600 44400 105 200 --- 21800 86.2 60.6
NP8 70600 51400 84.1 154 --- 27800 82.8 64.5
NP9 75100 52000 57.8 131 --- 29400 73.1 58.6
NP10 69400 46000 55.7 115 --- 30000 52.2 55.4
NP11 63300 40800 37.8 98 --- 30700 36.7 43.9
NP12 47400 32100 39.7 78.5 --- 24100 26.5 33.6
NP13 33700 22600 20 52.9 --- 19700 21.8 25.2
NP14 22000 12700 17.1 32 --- 9330 12.6 14.8
NP15 2990 2600 8.01 8.1 --- 627 6.59 8.11
NP16 4860 3090 11.3 11.5 --- 1340 3.9 ND
OP 530 673 113 76.5 --- 2950 BQL ND
OP1 423 456 135 306 --- 2320 ND ND
OP2 474 514 670 1140 --- 1830 25.1 32.4
OP3 534 496 116 188 --- 1530 ND 6.74
OP4 973 942 26.7 36.5 --- 1500 ND ND
OP5 1370 1070 9.09 9.9 --- 2100 ND ND
NP0EC 698 959 12100 9660 --- 658 1900 1480
NP1EC 1860 2670 47200 39500 --- 2340 5290 6040
OP0EC 89.1 85.5 1140 777 --- 116 106 62.7
181
Table A11 Particulate concentrations of NP0-16EO, and OP0-5EO in the 2005 
sampling events in Calumet WWTP.
Concentration, ng/g
Mar 2005 Aug 2005
Comp Influent Prim eff Sec eff Fin eff Influent Prim eff Sec eff Fin eff
NP 116000 147000 21300 21300 --- 786000 9210 9210
NP1 83500 86800 30600 30600 --- 467000 28400 28400
NP2 65100 74600 79600 79600 --- 246000 22200 22200
NP3 60000 68100 22900 22900 --- 88500 6390 6390
NP4 77900 92100 9030 9030 --- 56300 3060 3060
NP5 88800 111000 5200 5200 --- 50100 ND ND
NP6 128000 141000 2890 2890 --- 62700 ND ND
NP7 110000 120000 1870 1870 --- 57000 ND ND
NP8 109000 112000 1150 1150 --- 54500 ND ND
NP9 98500 93800 824 824 --- 44300 ND ND
NP10 72000 75200 461 461 --- 34600 ND ND
NP11 54700 55600 233 233 --- 28100 ND ND
NP12 36100 39200 ND ND --- 19400 ND ND
NP13 24800 25800 blk blk --- 13300 ND ND
NP14 15000 16600 153 153 --- 9130 ND ND
NP15 7850 9190 ND ND --- 4840 ND ND
NP16 4210 5220 88.1 88.1 --- 3000 405 405
OP 4100 5500 ND ND --- 19800 ND ND
OP1 1810 3190 ND ND --- 8000 ND ND
OP2 1530 900 3760 3760 --- 5890 BQL BQL
OP3 986 1130 710 710 --- 2390 ND ND
OP4 1250 1530 205 205 --- 2060 ND ND
OP5 1330 1440 ND ND --- 1940 ND ND
182
Table A12 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and 
OP0EC in the Chicago sewage samples.
Concentration, ng/L
Comp Industr Infl Res 1 Res 2 Comm
Chicago 
2
Chicago 
3
Oak 
Lawn 1
Oak 
Lawn 2 Brdgview
NP 5730 14600 13000 2500 231 4770 9540 33800 6330 20100
NP1 3860 23800 13400 2910 277 3730 8390 22400 4320 24800
NP2 3360 18500 17900 2420 910 3050 8890 13300 5050 16900
NP3 4140 14900 20000 1790 1780 3260 6090 11300 6820 13400
NP4 8880 23400 45800 2060 2400 6540 10400 25500 19500 28300
NP5 18600 34000 84300 3170 2840 11600 15500 46800 35200 45000
NP6 35500 60500 172000 4530 3330 21500 26500 89600 64300 81500
NP7 41500 64800 221000 4890 3010 24700 30600 116000 80700 105000
NP8 47100 69000 242000 4820 2640 33500 39000 145000 101000 127000
NP9 47900 69700 275000 4720 2320 33400 43800 165000 112000 141000
NP10 44100 64300 273000 4010 2040 30100 41500 153000 103000 137000
NP11 37300 55400 248000 3650 1800 27000 33200 137000 94200 107000
NP12 28800 43100 210000 2300 1260 22800 27000 102000 70600 88600
NP13 19800 33400 140000 1690 895 15600 19800 81100 50300 65100
NP14 13500 18900 96300 1050 466 7280 11000 45400 28900 32500
NP15 5620 3300 37500 222 296 722 1110 4820 2840 4200
NP16 2630 3730 14000 161 96 1550 2140 12700 7220 7830
OP 589 878 474 167 152 BQL 565 5250 178 1080
OP1 ND 1170 431 148 ND ND 418 1070 295 805
OP2 109 670 427 97 84 73 350 459 218 417
OP3 157 587 462 168 169 71 280 665 281 435
OP4 386 698 730 461 238 145 393 1240 671 628
OP5 629 798 1210 622 382 191 519 1950 913 796
NP0EC 532 707 649 464 388 370 467 360 375 520
NP1EC 167 1630 1690 1110 558 665 1150 936 1050 1380
OP0EC 37 48 125 62 52 21 27 35 30 28
183
Table A13 Particulate concentrations of NP0-16EO, and OP0-5EO in the Chicago 
sewage samples.
Concentration, ng/L
Comp Industr Infl Res 1 Res 2 Comm
Chicago 
2
Chicago 
3
Oak 
Lawn 1
Oak 
Lawn 2 Brdgview
NP 60100 312000 113000 32300 ND 27000 71800 114000 39100 287000
NP1 BQL BQL 33800 17500 ND 45800 282000 32800 14200 172000
NP2 20800 57300 34300 10100 7470 21100 144000 17600 14500 95300
NP3 20400 55300 42500 8340 7420 14700 82900 14300 18900 58100
NP4 30500 59300 80500 6800 7310 22000 109000 21400 38100 72100
NP5 57700 99500 161000 7950 4360 28500 143000 30000 58500 95500
NP6 105000 112000 255000 8800 4100 42600 197000 52000 89800 139000
NP7 98500 92200 266000 7290 2330 43000 207000 50300 89900 132000
NP8 121000 87400 291000 5970 1620 42300 207000 49100 92700 131000
NP9 121000 75200 285000 4600 1180 35000 181000 44500 77700 107000
NP10 116000 57900 231000 3250 931 30400 153000 32700 53800 77300
NP11 94700 43900 184000 2850 614 22900 118000 26400 43400 58200
NP12 72500 29800 127000 1690 544 15700 84000 17600 30600 39100
NP13 52600 19800 96000 1160 380 9530 45600 11400 18700 25800
NP14 34000 12100 49500 598 198 7240 29100 5590 10200 13800
NP15 21600 6680 23500 485 146 3390 14300 3000 5660 7830
NP16 13900 4220 11300 261 61 2470 9280 1460 2560 4470
OP ND 12800 1710 ND ND BQL 2460 10700 673 4850
OP1 ND ND 1300 ND ND ND ND 879 BQL 2580
OP2 937 1430 625 ND 1050 247 2440 356 333 1200
OP3 1050 1010 566 315 849 ND 854 360 266 654
OP4 2060 1040 1070 382 977 ND 1330 469 447 730
OP5 2820 1220 1240 357 631 ND 1500 436 581 627
Table A14 Dissolved concentrations of NP0-16EO, OP0-5EO, NP0-1EC, and OP0EC in Back River.
Concentration, ng/L
Mar 2005 Sep 2004 Oct 2004
Comp MR BR-1 BR-2 BR-3 BR-4 MR BR-1 BR-2 BR-3 BR-4 MR BR-1 BR-2 BR-3 BR-4
NP ND 338 511 432 237 ND 121 96 71 57 ND 128 128 137 76
NP1 BQL 1210 1960 1190 385 ND 53 24 14 26 ND 122 127 120 38
NP2 ND 559 1250 775 189 ND 23 15 12 12 6 63 75 71 13
NP3 ND 218 445 329 102 ND 26 16 13 10 8 40 32 32 10
NP4 ND 174 316 216 97 ND 27 21 21 21 15 28 19 20 ND
NP5 ND 155 254 163 87 ND 13 6 6 6 14 20 9 9 BQL
NP6 40 99 149 97 56 ND ND ND ND ND 24 30 ND ND ND
NP7 45 77 121 80 43 ND ND ND ND ND ND ND ND ND ND
NP8 55 79 124 78 41 ND ND ND ND ND ND ND ND ND ND
NP9 57 71 124 74 34 ND ND ND ND ND ND ND ND ND ND
NP10 50 66 115 66 29 ND ND ND ND ND ND ND ND ND ND
NP11 44 56 108 60 24 ND ND ND ND ND ND ND ND ND ND
NP12 36 49 105 50 19 ND ND ND ND ND ND ND ND ND ND
NP13 22 37 83 44 14 ND ND ND ND ND ND ND ND ND ND
NP14 14 35 67 35 15 ND ND ND ND ND ND ND ND ND ND
NP15 13 22 38 22 8 ND ND ND ND ND ND ND ND ND ND
NP16 6 15 30 18 10 ND ND ND ND ND ND ND ND ND ND
OP ND 48 51 50 25 ND BQL ND ND ND ND 15 13 12 ND
OP1 ND 128 188 129 34 ND ND ND ND ND ND BQL BQL BQL ND
OP2 ND 29 100 83 11 ND ND ND ND ND ND BQL BQL BQL ND
OP3 ND 10 30 21 4 ND ND ND ND ND ND ND ND ND ND
OP4 ND BQL 16 9 BQL ND ND ND ND ND ND ND ND ND ND
OP5 ND BQL 11 7 BQL ND ND ND ND ND ND ND ND ND ND
NP0EC 127 23600 23500 19300 11000 BQL 1190 1696 1383 657 37 550 2141 2227 1542
NP1EC 207 25400 28400 22100 11600 ND 2319 3825 2710 1004 ND 1133 5948 6270 3467
OP0EC BQL 2340 2800 2220 1100 ND 78 114 92 33 ND 62 171 179 99
Table A15 Particulate concentrations of NP0-16EO, and OP0-5EO in Back River.
Concentration, ng/g
Mar 2005 Sep 2004 Oct 2004
Comp MR BR-1 BR-2 BR-3 BR-4 MR BR-1 BR-2 BR-3 BR-4 MR BR-1 BR-2 BR-3 BR-4
NP ND 6740 8120 9320 6800 2512 2086 3550 2941 1179 2241 2373 3620 3591 4290
NP1 ND 25300 35200 30900 20800 ND 788 ND 441 BQL ND 1154 1361 1438 BQL
NP2 BQL 10700 31200 20700 4900 BQL 361 573 318 152 BQL 791 1102 1072 ND
NP3 ND 5970 10400 8440 2630 BQL 290 382 274 107 ND 415 372 423 196
NP4 BQL 2950 5650 4820 1400 ND 305 415 324 119 ND 265 185 226 110
NP5 ND 2510 5100 3820 1340 ND 116 122 82 ND ND 257 BQL 85 BQL
NP6 ND 2450 4410 3070 1330 ND 182 171 BQL BQL BQL 548 277 203 ND
NP7 BQL 1690 3590 2580 982 ND 170 BQL BQL BQL BQL 607 279 172 BQL
NP8 ND 2050 3870 2670 815 ND 151 152 BQL BQL BQL 697 292 203 ND
NP9 ND 2160 4440 2880 623 ND BQL BQL BQL BQL BQL 683 BQL BQL BQL
NP10 ND 1970 4500 2680 646 BQL BQL BQL BQL BQL BQL 673 224 150 ND
NP11 ND 1720 4620 2740 683 ND 114 116 BQL BQL BQL 542 218 136 ND
NP12 ND 1560 4640 2580 578 ND 118 120 BQL BQL 290 420 211 129 ND
NP13 ND 1280 4060 2190 514 ND 115 96 ND 87 273 320 147 112 ND
NP14 ND 987 3170 1700 365 ND 75 74 56 60 162 207 64 BQL ND
NP15 ND 567 2080 1120 253 170 65 46 ND 39 162 130 40 46 96
NP16 ND 381 1550 731 217 ND 52 31 ND 32 153 102 55 37 ND
OP 3200 5880 7730 3930 4880 ND ND ND ND ND BQL ND ND ND 855
OP1 ND 3910 4980 4020 4190 ND ND ND ND ND ND ND ND ND ND
OP2 ND 312 1000 779 BQL ND ND ND ND ND ND ND ND ND ND
OP3 ND 354 546 480 189 ND ND ND ND ND ND ND ND ND ND
OP4 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND
OP5 ND ND BQL BQL ND ND ND ND ND ND ND ND ND ND ND
186
Table A16 NP0-15EO and OP0-5EO concentrations in sediments from the 
Baltimore Harbor area (Fig. 3.7)
Concentration, ug/g
WR GF BC1 BC2 BR1 BR2 IH
NP 0.3352 1.1044 3.767 7.0755 3.0932 1.2523 1.2731
NP1 0.1353 0.5081 0.4625 1.7358 0.6002 0.7083 0.7568
NP2 0.0583 0.4149 0.1742 0.5798 0.1912 0.2522 0.3211
NP3 0.0424 0.1499 0.103 0.1364 0.1156 0.0911 0.139
NP4 0.0382 0.1612 0.0879 0.0901 0.1123 0.0401 0.0727
NP5 0.0536 0.2589 0.0986 0.0915 0.1753 0.0346 0.0763
NP6 0.0862 0.4581 0.1615 0.1207 0.2915 0.0519 0.1141
NP7 0.0979 0.5755 0.1739 0.1313 0.3484 0.0637 0.1255
NP8 0.12 0.5863 0.1852 0.1231 0.3682 0.06 0.139
NP9 0.1449 0.6382 0.1911 0.1315 0.3538 0.0751 0.1644
NP10 0.1502 0.697 0.1968 0.1494 0.29 0.0757 0.2017
NP11 0.1211 0.6098 0.1392 0.1133 0.2045 0.0544 0.2186
NP12 0.1731 0.773 0.128 0.1189 0.1618 0.0633 0.332
NP13 0.2495 1.4708 0.1172 0.146 0.1354 0.0862 0.6283
NP14 0.0079 0.0079 0.1175 0.0079 0.0079 0.0079 0.0079
NP15 ND 0.0076 0.0076 0.0076 ND ND 0.0076
OP ND 0.0972 0.4979 0.8112 0.1605 0.1002 0.1279
OP1 ND ND ND 0.1005 ND ND ND
OP2 0.0125 0.0218 0.031 0.0463 0.0139 0.0147 0.0221
OP3 ND ND 0.0138 0.0151 ND ND BQL
OP4 BQL 0.0115 0.0256 0.0287 BQL BQL 0.0158
OP5 0.0136 0.0127 0.0245 0.0283 0.0111 BQL 0.0167
187
Table A17 NP0-16EO and OP0-5EO concentrations in sediments from Back River 
(Fig. 5.3).
concentration, ug/g
Jan 2001 July 2001
Comp BR-A BR-C BR-A BR-B BR-C BR-D BR-E BR-F 
NP 2.2662 1.7445 2.5200 7.7900 4.0400 1.8200 0.8870 0.7470
NP1 1.1327 0.8923 1.4200 3.8600 1.7600 1.0900 0.5670 0.3310
NP2 0.7052 0.3306 0.6180 1.0400 0.8100 0.3950 0.2540 0.1520
NP3 0.2606 0.1216 0.2870 0.3980 0.2530 0.1060 0.0642 0.0482
NP4 0.1604 0.0621 0.1410 0.2390 0.1500 0.0739 0.0528 0.0489
NP5 0.1143 0.0442 0.0824 0.1560 0.1010 0.0499 0.0382 0.0336
NP6 0.1490 0.0569 0.1170 0.2320 0.1290 0.0741 0.0544 0.0591
NP7 0.1827 0.0678 0.1180 0.2010 0.1230 0.0602 0.0530 0.0497
NP8 0.1897 0.0575 0.1170 0.1890 0.1230 0.0656 0.0528 0.0495
NP9 0.2256 0.0636 0.1160 0.1960 0.1200 0.0602 0.0518 0.0467
NP10 0.2353 0.0622 0.1150 0.1910 0.1220 0.0607 0.0516 0.0459
NP11 0.1954 0.0512 0.0968 0.1660 0.1020 0.0490 0.0452 0.0373
NP12 0.1816 0.0506 0.0866 0.1440 0.0956 0.0423 0.0366 0.0345
NP13 0.1360 0.0415 0.1100 0.1790 0.1170 0.0468 0.0295 0.0229
NP14 0.0079 0.0079 0.1990 0.2780 0.2790 0.0942 0.0312 0.0384
NP15 ND ND 0.0709 0.1600 0.1370 0.0000 0.0206 0.0392
NP16 ND ND 0.0027 0.0030 0.0026 0.0000 0.0000 0.0000
OP ND ND BQL 1.1500 0.1800 ND BQL ND
OP1 ND ND ND ND ND ND ND ND
OP2 0.0244 0.0158 0.0436 0.1320 0.0859 0.0446 0.0284 0.0311
OP3 ND ND 0.0159 0.0654 0.0309 0.0130 BQL ND
OP4 BQL BQL BQL 0.0584 0.0253 ND ND ND
OP5 0.0102 BQL ND 0.0505 0.0205 ND ND ND
188
Table A18 Mass balance of NPEO and NPEC in three WWTPs (See also Fig. 4.7). 
IN: total mass in raw influent (except for Calumet in August, where it corresponds 
to the primary effluent), OUT (s): total mass in waste primary and secondary sludge 
(Blue Plains includes tertiary sludge); OUT (l): total mass in final effluent; OUT 
(total) = OUT (s) + OUT (l); DEGRADED: estimated from IN-OUT(total).
Back River 
(March)
Blue Plains 
(March)
Calumet
(March)
Calumet 
(August)
moles/day % moles/day % moles/day % moles/day %
IN 557 100 1765 100 1152 100 1106 100
OUT (s) 106 19 1128 64 247 21 334 30
OUT (l) 216 39 423 24 228 20 33 3
OUT (total) 322 58 1550 88 476 41 367 33
DEGRADED 235 42 215 12 677 59 739 67
Q (l), mgd 150 350 250 250
Q (s), tons/day 85 313 218 218
189
REFERENCES
Ackermann, G. E., Schwaiger, J., Negele, R. D., and Fent, K. (2002), ?Effects of long-
term nonylphenol exposure on gonadal development and biomarkers of estrogenicity 
in juvenile rainbow trout (Oncorhynchus mykiss).? Aquat. Toxicol. 60(3-4), 203-221.
Ahel, M., Giger, W., Molnar, E., and Ibric, S. (2000), ?Determination of nonylphenol 
polyethoxylates and their lipophilic metabolites in sewage effluents by normal-phase 
high-performance liquid chromatography and fluorescence detection.? Croat. Chem. 
Acta 73(1), 209-227.
Ahel, M., Giger, W., and Koch, M. (1994a), ?Behaviour of alkylphenol polyethoxylate 
surfactants in the aquatic environment ? I. Occurrence and transformation in sewage 
treatment.? Wat. Res. 28(5), 1131-1142.
Ahel, M., Giger, W., and Schaffner, C. (1994b), ?Behaviour of alkylphenol 
polyethoxylate surfactants in the aquatic environment ? II. Occurrence and 
transformation in rivers.? Wat. Res. 28(5), 1143-1152.
Ahel, M., Scully, Jr., F. E., Hoign?, J., and Giger, W. (1994c), ?Photochemical 
degradation of nonylphenol and nonylphenol polyethoxylates in natural waters.? 
Chemosphere 28(7), 1361-1368.
Ahel, M., and Giger, W. (1993), ?Partitioning of alkylphenols and alkylphenol 
polyethoxylates between water and organic solvents.? Chemosphere 26(8), 1471-
1478.
Ahel, M., Conrad, T., and Giger, W. (1987), ?Persistent organic chemicals in sewage 
effluents. 3. Determination of nonylphenoxy carboxylic acids by high-resolution gas 
chromatography/mass spectrometry and high-performance liquid chromatography.? 
Environ. Sci. Technol. 21(7), 697-703.
APE Research Council (2004), ?Nomenclature and CAS numbers for nonylphenol 
ethoxylates.? <http://www.aperc.org/nonylphenol_ethoxylates0204.htm> (February 
13, 2006).
APE Research Council (1999), ?Alkylphenols and alkylphenol ehtoxylates. An overview 
of safety issues.? <http://www.aperc.org/docs/whitepaper1-13-99.htm> (February 
12, 2006).
Balch, G., and Metcalfe, C. (2006), ?Developmental effects in Japanese medaka (Oryzias 
latipes) exposed to nonylphenol ethoxylates and their degradation products?. 
Chemosphere 62(8), 1214-1223.
190
Barber, L. B., Brown, G. K., and Zaugg, S. D. (2000), ?Potential endocrine disrupting 
organic chemicals in treated municipal wastewater and river water,? Analysis of 
Environmental Endocrine Disruptors, L. H. Keith, T. L. Jones-Lepp, and L. L.
Needham, Eds., American Chemical Society, Washington, DC; ACS Symposium 
Series 747, 97-123.
Bennett, E. R., and Metcalfe, C. D. (2000), ?Distribution of degradation products of 
alkylphenol ethoxylates near sewage treatment plants in the lower Great Lakes,
North America.? Environ. Toxicol. Chem. 19(4), 784-792.
Bennie, D. T. (1999), ?Review of the environmental occurrence of alkylphenols and 
alkylphenol ethoxylates.? Water Qual. Res. J. Canada 34(1), 79-122.
Bennie, D. T., Sullivan, C. A., Lee, H.-B., and Maguire, R. J. (1998), ?Alkylphenol 
polyethoxylate metabolites in Canadian sewage treatment plant waste streams.?
Water Qual. Res. J. Canada 33(2), 231-252.
Brian, J. V., Harris, C. A., Scholze, M., Backhaus, T., Booy, P., Lamoree, M., Pojana, G.,
Jonkers, N., Runnalls, T., Bonfa, A., Marcomini, A., and Sumpter, J. P. (2005), 
?Accurate prediction of the response of freshwater fish to a mixture of estrogenic 
chemicals?. Environ. Health Perspect. 113(6), 721-728.
Brownawell, B. J., Chen, H., Zhang, W., and Westall, J. C. (1997), ?Sorption of nonionic 
surfactants on sediment materials.? Environ. Sci. Technol. 31(6), 1735-1741.
Calafat, A. M., Kuklenyik, Z., Reidy, J. A., Caudill, S. P., Ekong, J., and Needham, L. 
(2005), ?Urinary concentrations of bisphenol A and 4-nonylphenol in a human 
reference population.? Environ. Health. Perspect. 113(4), 391-395.
CAS (2005), SciFinder Scholar Database, Chemical Abstracts Service.
Castillo, M ., Ventura, F., and Barcel?, D. (1999), ?Sequential solid phase extraction 
protocol followed by liquid chromatography-atmospheric pressure chemical 
ionization-mass spectrometry for the trace determination of non ionic 
polyethoxylated surfactants in tannery wastewaters.? Waste Management. 1999, 
19(2), 101-110.
Chapra, S. C. (1997), Surface Water-Quality Modeling. McGraw-Hill, New York.
Crescenzi, C., Di Corcia A., Samperi, R., and Marcomini, A. (1995), ?Determination of 
nonionic polyethoxylate surfactants in environmental waters by liuqid-
chromatography electrospray mass-spectrometry.? Anal. Chem. 67(11), 1797-1804.
Dachs, J., Van Ry, D. A., and Eisenreich, S. J. (1999), ?Occurrence of estrogenic 
nonylphenols in the urban and coastal atmosphere of the Lower Hudson River 
Estuary.? Environ. Sci. Technol. 33(15), 2676-2679.
191
Datta, S., Loyo-Rosales, J. E., and Rice, C. P. (2002), ?A simple method for the 
determination of trace levels of alkylphenolic compounds in fish tissue using 
pressurized fluid extraction, solid phase cleanup, and high-performance liquid 
chromatography fluorescence detection.? J. Agric. Food Chem. 50(6), 1350-1354.
Department of the Environment (2004), ?Canadian Environmental Protection Act, 1999. 
Notice requiring the preparation and implementation of pollution prevention plans in 
respect of nonylphenol and its ethoxylates contained in products.? Canada Gazzette, 
138(49), 3564-3579.
Di Corcia, A., Cavallo, R., Crescenzi, C., and Nazzari, M. (2000), ?Occurrence and 
abundance of dicarboxylated metabolites of nonylphenol polyethoxylate surfactants 
in treated sewages.? Environ. Sci. Technol. 34(18), 3914-3919.
Di Corcia, A., Costantino, A., Crescenzi, C., Marinoni, E., and Samperi, R. (1998), 
?Characterization of recalcitrant intermediates from biotransformation of the 
branched alkyl side of nonylphenol ethoxylate surfactants.? Environ. Sci. Technol.
32(16), 2401-2409.
Di Corcia, A., Samperi, R., and Marcomini, A. (1994), ?Monitoring aromatic surfactants 
and their biodegradation intermediates in raw and treated sewages by solid-phase 
extraction and liquid chromatography.? Environ. Sci. Technol. 28(5), 850-858.
Ding, W.-H., and Tzing, S.-H. (1998), ?Analysis of nonylphenol polyethoxylates and 
their degradation products in river water and sewage effluent by gas 
chromatography-ion trap (tandem) mass spectrometry with electron impact and 
chemical ionization.? J. Chromatogr. A. 824(1), 79-90.
?Directive 2003/53/EC of the European Parliament and of the Council of 18 June 2003?
Official Journal L 178, 17/7/2003, 24-27.
Dussault, E. B., Sherry, J. P., Lee, H.-B., Burnison, B. K., Bennie, D. T., and Servos, M. 
R. (2005), ?In vivo estrogenicity of nonylphenol and its ethoxylates in the Canadian 
environment.? Hum. Ecol. Risk Assess. 11(2), 353-364.
Eberhartinger, S., Steiner, I., Wash?ttl, J., and Kroyer, G. (1990), ?Study of the migration 
of acetaldehyde from PET bottles into soft drinks containing carbonic acid?. Z. 
Lebensm. Unters. Forsch. 191(4-5), 286-289.
Environment Canada (2004), ?Canadian water quality guideline. Nonylphenol and its 
ethoxylates? <http://www.ec.gc.ca/ceqg-
rcqe/English/Html/GAAG_NPE_WQC.cfm> (February 17, 2006).
Environment Canada, and Health Canada (2001), Nonylphenol and its ethoxylates. 
Priority substances list assessment report, Canadian Environmental Protection Act.
ISBN 0-662-29248-0.
192
European Commission (2002), European Union risk assessment report. 4-nonylphenol 
(branched) and nonylphenol. Vol. 10, B.G. Hansen, S.J. Munn, J. De Bruijn, S. 
Pakalin, M. Luotamo, F. Berthault, S. Vegro, C.J.A. Heidorn, G. Pellegrini, K. 
Vormann, R. Allanou, S. Scheer, editors. Office for Official Publications of the 
European Communities, Luxembourg.
Fensterheim, R. J. (2001), ?Comment on Estrogenicity determination in sewage treatment 
plants and surface waters from the Catalonian area (NE Spain).? Environ. Sci. 
Technol. 35(20), 4156.
Ferguson, P. L., Iden, C. R., and Brownawell, B. J. (2001), ?Analysis of nonylphenol and 
nonylphenol ethoxylates in environmental samples by mixed-mode high-
performance liquid chromatography-electrospray mass spectrometry.? J. 
Chromatogr. A 938(1-2), 79-91.
Ferguson, P. L., Iden, C. R., and Brownawell, B. J. (2001b), ?Distribution and fate of 
neutral alkylphenol ethoxylate metabolites in a sewage-impacted urban estuary.?
Environ. Sci. Technol. 35(12), 2428-2435.
Ferguson, P. L., Iden, C. R., and Brownawell, B. J. (2000), ?Analysis of alkylphenol 
ethoxylate metabolites in the aquatic environment using liquid chromatography-
electrospray mass spectrometry.? Anal. Chem. 72(18), 4322-4330.
Field, J. A., and Reed, R. L. (1996), ?Nonylphenol polyethoxy carboxylate metabolites of 
nonionic surfactants in U.S. paper mill effluents, municipal sewage treatment plant 
effluents, and river waters.? Environ. Sci. Technol. 30(12), 3544-3550.
Giesy, J. P., Pierens, S. L., Snyder, E. M., Miles-Richardson, S., Kramer, V. J., Snyder, S. 
A., Nichols, K. M., and Villeneuve, D. A. (2000), ?Effects of 4-nonylphenol on 
fecundity and biomarkers of estrogenicity in fathead minnows (Pimephales 
promelas).? Environ. Toxicol. Chem. 19(5), 1368-1377.
Giger, W., Brunner, P. H., and Schaffner, C. (1984), ?4-nonylphenol in sewage sludge: 
accumulation of toxic metabolites from nonionic surfactants.? Science 225(4662), 
623-625.
Gilbert, J., Startin, J. R., and McGuinness, J. D. (1986), ?Compositional analysis of 
commercial PVC bottles and studies of aspects of specific and overall migration into 
foods and simulants.? Food Addit. Contam. 3(2), 133-144.
Guenther, K., Heinke, V., Thiele, B., Kleist, E., Prast, H., and Raecker, T. (2002), 
?Endocrine disrupting nonylphenols are ubiquitous in food.? Environ. Sci. Technol.
36(8), 1676-1680.
Hale, R. C., Smith, C. L., de Fur, P. O., Harvey, E., Bush, E. O., La Guardia, M. J., and 
Vadas, G. G. (2000), ?Nonylphenols in sediments and effluents associated with 
diverse wastewater outfalls.? Environ. Toxicol. Chem. 19(4), 946-952.
193
Hollingsworth, P. (2002), ?Profits pouring from bottled water.? Food Technol. 56(5), 18.
Houde, F., DeBlois, C., and Berryman, D. (2002), ?Liquid chromatographic?tandem 
mass spectrometric determination of nonylphenol polyethoxylates and nonylphenol 
carboxylic acids in surface water.? J. Chromatogr. A. 961(2), 245-256.
Hu, J.-Y., Xie, G.-H., and Aizawa, T. (2002), ?Products of aqueous chlorination of 4-
nonylphenol and their estrogenic activity.? Environ. Toxicol. Chem. 21(10), 2034-
2039.
ITC (2002), ?Fiftieth report of the TSCA Interagency Testing Committee to the 
Administrator? Federal Register (67 FR 49530, July 30, 2002) (FRL-7183-7).
ITC (2002b), ?Forty-ninth report of the TSCA Interagency Testing Committee to the 
Administrator? Federal Register (67 FR 10298, March 6, 2002) (FRL-6820-8).
Jobling, S., Sheahan, D., Osborne, J. A., Matthiessen, P., and Sumpter, J. P. (1996), 
?Inhibition of testicular growth in rainbow trout (Oncorhynchus mykiss) exposed to 
estrogenic alkylphenolic chemicals.? Environ. Toxicol. Chem. 15(2), 194-202.
Jobling, S., and Sumpter, J. P. (1993), ?Detergent components in sewage effluent are 
weakly oestrogenic to fish: An in vitro study using rainbow trout (Oncorhynchus 
mykiss) hepatocytes.? Aquat. Toxicol. 27(3-4), 361-372.
John, D. M., House, W. A., and White, G. F. (2000), ?Environmental fate of nonylphenol 
ethoxylates: Differential adsorption of homologs to components of river sediment.? 
Environ. Toxicol. Chem. 19(2), 293-300.
Johnson, A. C., and Sumpter, J. P. (2001), ?Removal of endocrine-disrupting chemicals 
in activated sludge treatment works?. Environ. Sci. Technol. 35(24), 4697-4703.
Johnson, A. C., White, C., Besien, T. J., and J?rgens, M. D. (1998), ?The sorption 
potential of octylphenol, a xenobiotic oestrogen, to suspended and bed-sediments 
collected from industrial and rural reaches of three English rivers.? Sci. Total 
Environ. 210(1-6), 271-282.
Jonkers, N., Laane, R. W. P. M., de Graaf, C., and de Voogt, P. (2005), ?Fate modeling 
of nonylphenol ethoxylates and their metabolites in the Dutch Scheldt and Rhine 
estuaries: validation with new field data.? Estuarine, Coastal Shelf Sci. 62(1-2), 141-
160.
Jonkers, N., Knepper, T. P., and de Voogt, P. (2001), ?Aerobic biodegradation studies of 
nonylphenol ethoxylates in river water using liquid chromatography-electrospray 
tandem mass spectrometry.? Environ. Sci. Technol. 35(2), 335-340.
Kazeto, Y., Place, A. R., and Trant, J. M. (2004), ?Effects of endocrine disrupting 
chemicals on the expression of CYP19 genes in zebrafish (Danio rerio) juveniles.? 
Aquat. Toxicol. 69(1), 25-34.
194
Kolpin, D. W., Furlong, E. T., Meyer, M. T., Thurman, E. M., Zaugg, S. D., Barber, L. 
B., Buxton, H. T. (2002), ?Pharmaceuticals, hormones, and other organic wastewater 
contaminants in U.S. streams, 1999-2000: A national reconnaissance?. Environ. Sci. 
Technol. 36(6), 1202-1211.
Lee, H.-B. (1999), ?Review of analytical methods for the determination of nonylphenol 
and related compounds in environmental samples.? Water Qual. Res. J. Canada
34(1), 3-35.
Lee, H.-B., Peart, T. E., Bennie, D. T., and Maguire, R. J. (1997), ?Determination of 
nonylphenol polyethoxylates and their carboxylic acid metabolites in sewage 
treatment plant sludge by supercritical carbon dioxide extraction.? J Chromatogr A
785(1-2), 385-394.
Loyo-Rosales, J. E., Schmitz-Afonso, I., Rice, C. P., and Torrents, A. (2003), ?Analysis 
of octyl- and nonylphenol and their ethoxylates in water and sediments by liquid 
chromatography/tandem mass spectrometry.? Anal. Chem. 75(18), 4811-4817.
Lye, C. M., Frid, C. L. J., Gill, M. E., Cooper, D. W., and Jones, D. M. (1999), 
?Estrogenic alkylphenols in fish tissues, sediments, and waters from the U.K. Tyne 
and Tees estuaries.? Environ. Sci. Technol. 33(7), 1009-1014.
Marcomini, A., Pojana, G., Sfriso, A., and Alonso, J.-M. Q. (2000), ?Behavior of anionic 
and nonionic surfactants and their persistent metabolites in the Venice lagoon, Italy.?
Environ. Toxicol. Chem. 19(8), 2000-2007.
Maryland Department of the Environment (2005), Total Maximum Daily Loads of 
Nitrogen and Phosphorus for Back River in Baltimore City and Baltimore County, 
Maryland. Baltimore, MD.
Maryland Department of the Environment (2003), Patapsco/Back River Watershed 
SWMM Model Report. Baltimore, MD.
Maryland Department of Natural Resources (2004), ?Water Quality Monitoring Data.?
<http://mddnr.chesapeakebay.net/bay_cond/bay_cond.cfm?param=bdo&station=WT
41> (November 25, 2004).
McNeal, T. P., Biles, J. E., Begley, T. H., Craun, J. C., Hopper, M. L., and Sack, C. A. 
(2000), ?Determination of suspected endocrine disruptors in food and food 
packaging,? Analysis of Environmental Endocrine Disruptors, L. H. Keith, T. L.
Jones-Lepp, and L. L. Needham, Eds., American Chemical Society, Washington, 
DC; ACS Symposium Series 747, 33-52.
Montgomery-Brown, J., and Reinhard, M. (2003), ?Occurrence and behavior of 
alkylphenol polyethoxylates in the environment.? Environ. Eng. Sci. 20(5), 471-486.
National Climatic Data Center (NCDC) (2005), Climatological Data. Maryland and 
Delaware. 109(3), 8-9.
195
National Climatic Data Center (NCDC) (2004a), Climatological Data. Maryland and 
Delaware. 108(10), 8-9.
National Climatic Data Center (NCDC) (2004b), Climatological Data. Maryland and 
Delaware. 108(9), 8-9.
Nimrod, A. C., and Benson, W. H. (1996), ?Environmental estrogenic effects of 
alkylphenol ethoxylates.? Crit. Rev. Toxicol. 26(3), 335-364.
?Nonylphenol. Chemical Profile? (2001) Chemical Market Reporter. Vol. 260(2). July 9, 
2001 revision. <http://www.aperc.org/nonylphenol.pdf> (February 12, 2006).
Odum, J., Tinwell, H., Van Miller, J., Joiner, R., and Ashby, J. (2001), ?The uterotrophic 
activity of nonylphenol in the rat is not mediated by aromatase enzyme induction.? 
Toxicol. Lett. 118(3), 165-169.
Pedersen, S. N., and Lindholst, C. (1999), ?Quantification of the xenoestrogens 4-tert.-
octylphenol and bisphenol A in water and in fish tissue based on microwave assisted 
extraction, solid-phase extraction and liquid chromatography?mass spectrometry.? J. 
Chromatogr. A. 864(1), 17-24.
Petrovic, M., Sol?, M., L?pez de Alda, M. J., and Barcel?, D. (2002), ?Endocrine 
disruptors in sewage treatment plants, receiving river waters, and sediments: 
Integration of chemical analysis and biological effects on feral carp.? Environ. 
Toxicol. Chem. 21(10), 2146-2156.
Petrovic, M., and Barcel?, D. (2001), ?Analysis of ethoxylated nonionic surfactants and 
their metabolites by liquid chromatography/atmospheric pressure ionization mass 
spectrometry.? J. Mass Spectrom. 36(11), 1173-1185.
Petrovic, M., and Barcel?, D. (2001b), ?Determination of phenolic xenoestrogens in 
environmental samples by liquid chromatography with mass spectrometric 
detection.? J. AOAC Int. 84(4), 1074-1085.
Petrovic, M., and Barcel?, D. (2000), ?The stability of non-ionic surfactants and linear 
alkylbenzene sulfonates in a water matrix and on solid-phase extraction cartridges.? 
Fresenius J. Anal. Chem. 368(7), 676-683.
Porter, M. R. (1994), Handbook of surfactants. 2nd Edition, Blackie Academic & 
Professional, Glasgow.
Prichard, E., MacKay, G.M., and Points, J., editors (1996) Trace analysis: A structured 
approach to obtaining reliable results. Royal Society of Chemistry, Cambridge, p. 
280.
Renner, R. (1997), ?European bans on surfactant trigger transatlantic debate.? Environ. 
Sci. Technol. 31(7), 316A-320A.
196
Rice, C. P., Schimtz-Afonso, I., Loyo-Rosales, J. E., Link, E., Thoma, R., Fay, L.,
Altfater, D., and Camp, M. J. (2003), ?Alkylphenol and alkylphenol-ethoxylates in 
carp, water, and sediment from the Cuyahoga River, Ohio.? Environ. Sci. Technol.
37(17), 3747-3754.
Rodr?guez, I., Llompart, M. P., and Cela, R. (2000), ?Solid-phase extraction of phenols.?
J. Chromatogr. A. 885(1-2), 291-304.
Sargent, M., Harte, R., and Harrington, C., editors (2002) Guidelines for achieving high 
accuracy in isotope dilution mass spectrometry (IDMS). Royal Society of Chemistry, 
Cambridge, p. 3.
Schmitz-Afonso, I., Loyo-Rosales, J. E., Avil?s, M. P., Rattner, B. A., and Rice, C. P. 
(2003), ?Determination of alkylhenol and alkylphenolethoxylates in biota by liquid 
chromatography with detection by tandem mass spectrometry and fluorescence 
spectroscopy.? J. Chromatogr. A. 1010(1), 25-35.
Schwaiger, J., Spieser, O. H., Bauer, C., Ferling, H., Mallow, U., Kalbfus, W., and 
Negele, R. D. (2000), ?Chronic toxicity of nonylphenol and ethinylestradiol: 
haematological and histopathological effects in juvenile Common carp (Cyprinus 
carpio).? Aquat. Toxicol. 51(1), 69-78.
Scullion, S. D., Clench, M. R., Cooke, M., and Ashcroft, A. E. (1996), ?Determination of 
surfactants in surface water by solid-phase extraction, liquid chromatography and 
liquid chromatography-mass spectrometry.? J. Chromatogr. A 733(1-2), 207-216.
Servos, M. R. (1999), ?Review of the aquatic toxicity, estrogenic responses and 
bioaccumulation of alkylphenols and alkylphenol polyethoxylates.? Water Qual. Res. 
J. Canada 34(1), 123-177.
Shang, D. Y., Ikonomou, M. G., and Macdonald, R. W. (1999a), ?Quantitative 
determination of nonylphenol polyethoxylate surfactants in marine sediment using 
normal-phase liquid chromatography?electrospray mass spectrometry.? J. 
Chromatogr. A 849(2), 467-482.
Shang, D. Y., Macdonald, R. W., and Ikonomou, M. G. (1999b), ?Persistence of 
nonylphenol ethoxylate surfactants and their primary degradation products in 
sediments from near a municipal outfall in the strait of Georgia, British Columbia, 
Canada.? Environ. Sci. Technol. 33(9), 1366-1372.
Sheldon, L. S., and Hites, R. A. (1979), ?Sources and movement of organic chemicals in 
the Delaware River.? Environ. Sci. Technol. 13(5), 574-579.
Sol?, M., L?pez de Alda, M. J., Castillo, M., Porte, C., Ladegaard-Pedersen, K., and 
Barcel?, D. (2000), ?Estrogenicity determination in sewage treatment plants and 
surface waters from the Catalonian area (NE Spain).? Environ. Sci. Technol. 34(24), 
5076-5083.
197
Soto, A. M., Justicia, H., Wray, J. W., and Sonnenschein, C. (1991), ?p-Nonyl-phenol: 
An estrogenic xenobiotic released from ?modified? polystyrene.? Environ. Health 
Perspect. 92, 167-173.
Staples, C., Mihaich, E., Carbone, J., Woodburn, K., Klecka, G. (2004), ?A weight of 
evidence analysis of the chronic ecotoxicity of nonylphenol ethoxylates, nonylphenol 
ether carboxylates, and nonyphenol.? Hum. Ecol. Risk Assess. 10(6), 999-1017.
Takino, M., Daishima, S., and Yamaguchi, K. (2000), ?Determination of nonylphenol 
ethoxylate oligomers by liquid chromatography?electrospray mass spectrometry in 
river water and non-ionic surfactants.? J. Chromatogr. A 904(1), 65-72.
Talmage, S. S. (1994), Environmental and human safety of major surfactants. Alcohol 
ethoxylates and alkylphenol ethoxylates. The Soap and Detergent Association and 
Lewis Publishers, Boca Raton.
Tanaka, F. S., Wien, R. G., and Zaylskie, R. G. (1991), ?Photolytic degradation of a 
homogeneous Triton X nonionic surfactant: nonaethoxylated p-(1,1,3,3-
tetramethylbutyl)phenol.? J. Agric. Food Chem. 39(11), 2046-2052.
Thorpe, K. L., Hutchinson, T. H., Hetheridge, M. J., Scholze, M., Sumpter, J. P., and 
Tyler, C. R. (2001), ?Assessing the biological potency of binary mixtures of 
environmental estrogens using vitellogenin induction in juvenile rainbow trout 
(Oncorhynchus mykiss)?. Environ. Sci. Technol. 35(12), 2476-2481.
Toyo?oka, T., and Oshige, Y. (2000), ?Determination of alkylphenols in mineral water 
contained in PET bottles by liquid chromatography with coulometric detection.?
Anal. Sci. 16(10), 1071-1076.
U. S. EPA (2003), Ambient aquatic life water quality criteria for nonylphenol ? draft. 
Washington, DC.
U. S. Food and Drug Administration. Guidance for Industry ? Preparation of food 
contact notifications and food additive petitions for food contact substances: 
Chemistry recommendations. Final guidance, April 2002.
Van Ry, D. A., Dachs, J., Gigliotti, C. L., Brunciak, P. A., Nelson, E. D., and Eisenreich, 
S. J. (2000), ?Atmospheric seasonal trends and environmental fate of alkylphenols in 
the lower Hudson River estuary.? Environ. Sci. Technol. 34(12), 2410-2417.
Vivian, C. M. G. (1986), ?Tracers of sewage-sludge in the marine environment - A 
review.? Sci. Total Envir. 53(1-2), 5-40.
Weinheimer, R. M., and Varineau, P. T. (1998), ?Polyoxyethylene alkylphenols?, 
Nonionic surfactants. Organic chemistry, N. M. van Os, editor, Marcel Dekker, New 
York, 39-85.
198
Wheeler, T. F., Heim, J. R., LaTorre, M. R., and Janes, A. B. (1997), ?Mass spectral 
characterization of p-nonylphenol isomers using high-resolution capillary GC-MS.? 
J. Chromatogr. Sci. 35(1), 19-30.
Ying, G.-G., Williams, B., and Kookana, R. (2002), ?Environmental fate of alkylphenols 
and alkylphenol ethoxylates?a review.? Environ. Intern. 28(3), 215-226.