UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The application of semipermeable membrane devices to the detection and monitoring of 4-nonylphenol in… Suffredine, Lori Jean 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0098.pdf [ 6.27MB ]
Metadata
JSON: 831-1.0050165.json
JSON-LD: 831-1.0050165-ld.json
RDF/XML (Pretty): 831-1.0050165-rdf.xml
RDF/JSON: 831-1.0050165-rdf.json
Turtle: 831-1.0050165-turtle.txt
N-Triples: 831-1.0050165-rdf-ntriples.txt
Original Record: 831-1.0050165-source.json
Full Text
831-1.0050165-fulltext.txt
Citation
831-1.0050165.ris

Full Text

THE APPLICATION OF SEMIPERMEABLE MEMBRANE DEVICES TO THE DETECTION AND MONITORING OF 4-NONYLPHENOL IN AQUATIC SYSTEMS by LORI JEAN SUFFREDINE B . S c , The University of Victoria, 1977 Post Baccalaureate, Simon Fraser University, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Civil Engineering) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA September 1998 ® Lori Jean Suffredine, 1998  In  presenting  degree freely  at  this  the  available  copying  of  department publication  thesis  in  partial  fulfilment  of  University  of  British  Columbia,  I agree  for  this or of  reference  thesis by  this  for  his thesis  and  study.  scholarly  or for  her  I further  purposes  gain  shall  permission.  Department of  ^.'^'. (  The University of British Vancouver Canada  Date if-  DE-6 (2/88)  A*->V  £->  )\  Columbia  *»--  f  i * * '  —•*-*--•—=-,  requirements that  agree  may  representatives.  financial  the  be  It not  that  the  Library  permission  granted  is  by  understood be  for  allowed  an  advanced  shall make for  the that  without  it  extensive  head  of  my  copying  or  my  written  ABSTRACT  The surfactant 4-nonylphenol (4-NP) is a component of many industrial and domestic cleaning products, and has been demonstrated to elicit estrogenic effects in organisms.  4-NP  enters waterways through surface runoff and as a microbial breakdown product in secondary sewage treatment plant (STP) effluents.  This research project examined the feasibility of  applying a passive concentrator called the semipermeable membrane device (SPMD) to the collection and concentration of 4-NP from water.  A derivatization method was developed to  improve peak separation by GC-FID and provide better integration of peaks. Laboratory tests with SPMDs indicated they could be used successfully to sequester waterborne 4-NP.  An established  acetylation method for chlorinated phenolics was then applied to the 4-NP present in SPMD dialysates.  The results indicated  that only a small portion of the 4-NP was undergoing reaction. A procedure specific to 4-NP in STP effluents and sludges was attempted next, but the results confirmed that only partial derivatization was occurring.  An experiment designed to track  the progress of 4-NP through the acetylation reaction determined that the majority of the target analyte was remaining in the hexane-based SPMD dialysate, where it was unavailable for reaction with acetic anhydride in the water-  ii  based potassium carbonate solution. Subsequent attempts at derivatization continued to yield only partial acetylation of the 4-NP.  Solvent exchange  procedures, aimed at transferring the target analyte from hexane to a water-miscible solvent, produced no improvement. A method which incorporated a methanol solvent exchange with substantially reduced solvent volumes appeared promising with laboratory samples, but failed to yield satisfactory results when applied to SPMD dialysates.  Finally, successful  acetylation of 4-NP in hexane was achieved using a method designed to acetylate analytes in an organic solvent. A replicate test conducted using the new acetylation method provided unexpected insights into the reasons for the difficulties encountered with achieving complete derivatization.  The replicate test results suggested that the  length of time taken to perform the acetylation procedure was critical to derivatization success.  Subsequent experiments  confirmed that although the initial reaction favored formation of the acetylated product, if left in contact with base, an equilibrium was established between acetylated and nonacetylated 4-NP. Preliminary testing with secondary sewage treatment plant effluent confirmed that the conventional SPMD system successfully sequestered waterborne 4-NP.  Resulting  dialysates had estimated NP concentrations of 4 to 5 mg/L.  iii  When performed under controlled conditions, the derivatization procedure developed in this research project quantitatively acetylated NP present in hexane-based SPMD dialysates.  Final  cleanup of acetylated dialysates using 5% deactivated silica gel isolated NP into the second eluting fraction, but failed to separate it from a significant number of background interferences.  Further field testing and additional cleanup  methods are needed to complete the techniques developed in this project and confirm their application under actual field conditions.  iv  TABLE OF CONTENTS  ABSTRACT  ii  LIST OF TABLES  xii  LIST OF FIGURES  XV  1.0  INTRODUCTION  1  1.1  Motivation for Research  1  1.2  Thesis Organization  2  1.3  Background Information and Literature Review 1.3.1  4-Nonylphenol 1.3.1.1  4  Production and Use of 4-Nonylphenol  1.3.1.2  4  Chemical and Physical Characteristics  1.3.1.3  1.3.2  6  Environmental Distribution and Fate  8  1.3.1.4  Biological Effects  14  1.3.1.5  Analytical Methods  18  Semipermeable Membrane Devices 1.3.2.1  Description and Use  1.3.2.2  Estimating Analyte Water Concentrations With SPMDs  1.3.2.3  . . . . 21 21  . . . . 26  Advantages and Limitations of the SPMD System  1.4  . .4  Research Objectives  29 37  v  2.0  MATERIALS AND EXPERIMENTAL METHODS  38  2.1  38  2.2  2.3  Materials 2.1.1  SPMD Components  38  2.1.2  Target Chemical  38  2.1.3  Solvents and Standard Solutions . . . . 38  2.1.4  Other Materials  39  Experimental Methods  42  2.2.1  SPMD Preparation  42  2.2.2  Preparation of Glassware  43  2.2.3  Preparation of Test Solutions  44  2.2.4  Cleanup and Dialysis of SPMDs  44  2.2.5  Analysis of SPMD Extracts  45  2.2.6  Silica Gel Cleanup Procedure  45  Analytical Techniques  46  2.3.1  Gas Chromatography  46  2.3.2  Calibration Standards  47  2.3.3  Method Detection Limit and Reliable Detection Limit  3.0  50  METHOD DEVELOPMENT  51  3.1  Preliminary SPMD Testing  51  3.1.1  Summary of Procedure  51  3.1.2  Results of Preliminary Testing  3.1.3  Discussion of Preliminary Testing Results  . . . . 52  56  vi  3.2  Application of the Voss Procedure for Chlorinated Phenolics  57  3.2.1  Selection of the Voss Procedure . . . . 57  3.2.2  The Voss-Adapted Acetylation Procedure Using an SPMD Extract . . . .59 3.2.2.1  Summary of Method  3.2.2.2  Results Using an SPMD  3.2.2.3 3.2.3  Extract  60  Discussion of Results  61  Applying the Voss Procedure to Laboratory-Prepared Samples 3.2.3.1  Summary of Method  3.2.3.2  Results Using Laboratory-  3.2.3.3 3.3  59  62 62  Prepared Samples  63  Discussion of Results  67  Acetylations Using the Lee and Peart Procedure  68  3.3.1  Selection of the Procedure  68  3.3.2  Summary of Method  68  3.3.3  Results of the Lee and Peart Derivatizations  3.3.4  70  Tracking NP Through the Derivatization Process  72  3.3.4.1  Description of Method  72  3.3.4.2  Results  75  vii  3.3.4.3 3.4  Discussion of Results  76  Solvent Exchange  76  3.4.1  Introduction  76  3.4.2  Summary of Method  76  3.4.3  Results Using Solvent Exchange  3.4.4  Discussion of Results  3.4.5  Examining the Influence of Reaction  . . . . 77 78  Time and Extraction Time  79  3.4.5.1  Summary of Method  79  3.4.5.2  Results of the Reaction Time and Extraction Time Study  3.4.5.3 3.5  Discussion of Results  . . . .81  . . . . . . 84  Application of a Model Compound - Octylphenol . . 85 3.5.1  Introduction  3.5.2  The Lee and Peart Procedure Using Octylphenol 3.5.2.1  Summary of Method  3.5.2.2  Results of the Lee and Peart  3.5.2.3 3.6  85  86 86  Derivatization  86  Discussion of Results  88  The Lee-Adapted Acetylation Procedure  88  3.6.1  Summary of Method  88  3.6.2  Results  90  3.6.3  Discussion of Results  91  viii  3.7  3.8  3.9  Methanol Solvent Exchange  91  3.7.1  Summary of Method  91  3.7.2  Results  92  3.7.3  Discussion of Results  93  Applying Methanol Solvent Exchange to Nonylphenol  93  3.8.1  Summary of Method  93  3.8.2  Results  94  3.8.3  Discussion  95  Acetylation of SPMD Dialysates using Methanol Solvent Exchange and the  3.10  Lee-Adapted Procedure  95  3.9.1  Summary of Method  95  3.9.2  Results  95  3.9.3  Discussion of Results  99  The Wilson Procedure 3.10.1  Introduction  3.10.2  Applying the Wilson Procedure to Laboratory-Prepared Standards . . . .  99 99  100  3.10.2.1  Summary of Method  100  3.10.2.2  Results  101  3.10.2.3  Discussion of Results  102  ix  3.10.3  Applying the Wilson Procedure to SPMD Dialysates  3.10.3.1  Summary of Method  102  3.10.3.2  Results  102  3.10.3.3  Discussion  103  3.10.4  The Replicate Test  3.10.4.1  Summary of Method  3.10.4.2  Results of the Replicate  3.10.4.3 3.10.5  104 104  Testing  104  Discussion of Results  105  The Final Wash Study  106  3.10.5.1  Summary of Method  106  3.10.5.2  Results  108  3.10.5.3  Discussion of Results  109  3.10.6  The Shaking Time Study  110  3.10.6.1  Summary of Method  110  3.10.6.2  Results  Ill  3.10.6.3  Discussion  112  3.10.7  3.11  102  Determination of Percent Recovery . .  113  3.10.7.1  Summary of Method  113  3.10.7.2  Results  114  3.10.7.3  Discussion  116  Conclusions on Method Development  x  116  3.12  4.0  5.0  Preliminary Field Testing  118  3.12.1  Introduction  118  3.12.2  Summary of Method  119  3.12.3  Results  121  3.12.4  Discussion  127  CONCLUSIONS AND RECOMMENDATIONS  130  4.1  Conclusions  130  4.2  Recommendations  131  LITERATURE CITED  132  APPENDICES Appendix A  141 Certificate of Analysis for 4-Nonylphenol  Appendix B  GC-MS Analysis of Underivatized 4-Nonylphenol  Appendix C  143  GC-MS Analysis of Underivatized and Derivatized Octylphenol  Appendix D  141  146  GC-MS Analysis of Derivatized 4-Nonylphenol  xi  148  LIST OF TABLES  Table 1.  Structure and selected physical characteristics of 4-nonylphenol  Table 2.  96 h LC50 values for NP for selected aquatic species  Table 3.  15  Volume of stock added for underivatized standards  Table 4.  49  Volume of stock added for derivatized standards  Table 5,  49  Treatment summary for preliminary SPMD testing  Table 6.  52  Summary of results for preliminary SPMD testing  Table 7.  7  54  Summary of results for acetylation of SPMD extracts using the Voss procedure . . . 60  Table 8,  Summary of results for the Voss acetylation method using a laboratory sample  Table 9.  64  Summary of results for the NP tracking experiment  Table 10.  .  Results of NP acetylation after solvent exchange with THF  Table 11  75  78  Results of the reaction time and extraction time study  xii  83  Table 12,  Results of OP acetylations using the Lee-adapted procedure  Table 13.  Results of the Lee-adapted acetylations using methanol solvent exchange  Table 14,  90  92  Results of the NP acetylations using methanol solvent exchange and the Lee-adapted procedure  Table 15.  94  Results of acetylations on SPMD dialysates using methanol solvent exchange and the Lee-adapted procedure  Table 16.  96  Results of NP acetylations on laboratory-prepared samples using the Wilson procedure  Table 17.  Results of acetylations on SPMD dialysates using the Wilson procedure . .  Table 18.  105  Final washes used for the Final Wash Study  Table 20.  103  Results of replicate testing using the Wilson procedure  Table 19.  101  107  Estimated percent acetylation for five final washing regimes  109  Table 21.  Results of the Shaking Time Study . . . .  112  Table 22.  Percent recoveries for Lab Replicates 1 to 5  114  xiii  Table 23.  Example calculation for determining the percent recovery of NP  Table 24.  115  Field test results for underivatized SPMD dialysates analyzed using GC-MS in SIM mode  xiv  122  LIST OF FIGURES  Figure 1.  Formation of APs from biodegradation of APEs  5  Figure 2.  Photograph of SPMD  25  Figure 3.  Aluminum frame deployment panel for SPMD  Figure 4.  25  Gas chromatogram of 10 mg/L NP standard analyzed 28 May 1996  Figure 5.  Gas chromatogram of hexane after acetylation analyzed 30 April 1996  Figure 6.  66  Gas chromatogram of 100 mg/L NP standard  Figure 8.  65  Gas chromatogram of 2 mg/L NP standard analyzed 30 April 1996  Figure 7.  55  71  Gas chromatogram for Lee and Peart acetylation with a hexane-based sample . . . 71  Figure 9.  Gas chromatogram for Lee and Peart acetylation with a water-based sample  Figure 10.  . . . 72  Collection of hexane aliquots at key stages of the Lee and Peart acetylation procedure  Figure 11,  74  Variations in reaction time and time of hexane addition for the Reaction Time and Extraction study  xv  80  Figure 12.  Gas chromatogram of 50 mg/L OP standard  Figure 13.  Gas chromatogram of acetylated OP sample . . 87  Figure 14.  Gas chromatogram of non-acetylated 100 mg/L NP standard  Figure 15.  .98  Gas chromatogram of acetylated SPMD dialysate  Figure 18,  97  Gas chromatogram of acetylated SPMD Control dialysate  Figure 17.  97  Gas chromatogram of acetylated 40 mg/L NP standard  Figure 16,  . . 87  98  Chromatogram of derivatized 10 mg/L NP standard obtained using GC-MS in SIM mode  Figure 19,  124  Chromatogram of derivatized SPMD dialysate obtained using GC-MS in SIM mode  Figure 20,  124  Chromatogram of silica gel cleanup procedure Fraction 1 obtained using GC-MS in SIM mode  Figure 21  125  Chromatogram of silica gel cleanup procedure Fraction 2 obtained using GC-MS in SIM mode  Figure 22,  125  Chromatogram of silica gel cleanup procedure Fraction 3 obtained using GC-MS in SIM mode  xv i  126  Figure 23.  Chromatogram of silica gel cleanup procedure Fraction 2 obtained using GC-FID  126  xvii  1.0  INTRODUCTION  1.1  Motivation for Research In recent years concern has increased over the presence  of nonionic detergents in the aquatic environment.  These  chemicals and their degradation by-products have been linked to a variety of hormone-disrupting effects in aquatic species with speculative effects in humans. The impact of long-term exposure to hormone-disruptors may have far-reaching consequences related to reproductive success and, ultimately, species survival. One such chemical, 4-nonylphenol, is attracting attention in the research community because of its widespread use in a variety of products and industrial processes. Nonylphenol has been detected in sewage treatment plant effluents, industrial wastewaters, receiving waters and sediments.  Current methods  for measuring nonylphenol present in aquatic systems include gas chromatography-mass spectrometry (GC-MS), high performance liquid chromatography (HPLC), and gas chromatography using a flame ionization detector (GC-FID). An environmental monitoring tool developed by Huckins et al.  has been demonstrated to successfully sequester a variety  of environmental contaminants, including organochlorines, polycyclic aromatic hydrocarbons (PAHs), and pesticides such as mirex and fenvalerate (Huckins et al., 1992; Prest et al.,  1995).  1990; Lebo et  al.,  This passive sampler, called a  1  semipermeable membrane device or SPMD, collects and concentrates low-level contaminants present in air, water and sediment.  When contaminant levels within the SPMD are  sufficiently high, the chemicals are then extracted and quantitatively analyzed. SPMDs present many significant advantages as a monitoring tool, and the chemical characteristics of 4-nonylphenol suggest it should sequester well within an SPMD.  The focus of  this research project was to explore the application of conventional SPMD technology to the detection and monitoring of 4-nonylphenol in aquatic systems.  1.2  Thesis Organization Chapter 1 describes the motivation and objectives of the  research, and gives background information on 4-nonylphenol and SPMDs. Chapter 2 outlines the materials used in the project and the experimental methods used with SPMD testing. Analytical techniques and instrumentation are also described in Chapter 2.  Chapter 3 details the steps taken to the final  achievement of a successful derivatization procedure for 4nonylphenol present in SPMD dialysates.  Chapter 4 provides  conclusions and offers suggestions for further research. The appendices contain supplementary experimental data. Appendix A gives the Certificate of Analysis detailing the isomeric content and purity of the 4-nonylphenol used in this  2  study.  Appendix B contains the gas chromatographic/mass  spectrometric (GC-MS) analysis for underivatized 4-nonylphenol.  Appendix C details the GC-MS results for  octylphenol, and Appendix D provides the GC-MS results for derivatized 4-nonylphenol.  3  1.3  Background Information and Literature Review  1.3.1  4-Nonylphenol  1.3.1.1  Production and Use of 4-Nonylphenol  The occurrence of 4-nonylphenol in the aquatic environment results primarily from microbial degradation of its derivatives, nonylphenol polyethoxylates (NPEs).  NPEs  form part of a major group of nonionic surface-active agents collectively referred to as the alkylphenol polyethoxylates (APEs).  When subjected to microbial biodegradation such as  occurs in secondary sewage treatment, APEs cleave along parts of their long hydrophilic chain to form a series of stable short-chain metabolic products (Figure l)(Giger et al., Giger et al.,  1987; Blackburn and Waldock, 1995).  1984;  The  nonylphenols (NPs), short-chain nonylphenol polyethoxylates, and nonylphenoxy carboxylic acid metabolites formed by biotransformation of NPEs exhibit lipophilicity and have greater toxicity than the original NPE compound (Granmo et al.,  1989; Naylor et al.,  1992).  This toxicity of NP and the  short-chain NPEs has focussed attention on the production and use of NPEs worldwide. In terms of production, the APEs, including the NPEs, form the world's third largest group of surfactants (Blackburn and Waldock, 1995).  In 1990, 450 million pounds (200,000 Mg)  of APEs were produced in the U.S. (Naylor et al.,  4  1992) and  16,000 to 19,000 tonnes (15,000 - 17,000 Mg) were produced in the U.K. (Blackburn and Waldock, 1995).  Canadian demand for  NPEs in that year was 4.1 kton (3700 Mg)(Lee and Peart, 1995). NPEs are widely used as industrial and domestic cleaning products (Naylor et al., 1992).  Though concern about possible  biological effects in recent years has led to a decline in household use, these surfactants continue to have important industrial applications particularly in the textile industry and the pulp and paper sector (Lee and Peart, 1995). AP„E0 JJ. AP„.iEO  jfY°{o^ CH2v t3]^ 2v CHf OH R  Progressive loss of ethoxylate Groups until APEO  ^ £Jf  ^CH °*l <* Alkylphenol Monoethoxylate  jfV 0 "  AP  Alkylphenol Polyethoxylate (n = m+l)  Mk l heno1  yP  R: C9H19 Nonylphenol R: C8Hi7 Octylphenol  Figure 1.  Formation of APs from biodegradation of APEs (from Blackburn and Waldock (1995))  By far the greatest use for NP is in the production of NPEs (Stephanou and Giger, 1982).  However the chemical is  also an important adjuvant in several pesticides and forms a major component (50.5%) of Matacil, the carbamate insecticide  5  used extensively in Canada to control the spruce budworm, Choristoneura  fumiferana  (Clem.) (Sundaram, 1995).  In  pesticide formulations, NP enhances spreading and cuticular penetration of the pesticide's active agent (Sundaram, 1995). For spray-mixes, the low evaporation rate of NP improves targetability and droplet deposition, increasing application efficiency (Sundaram, 1995).  In industry, NP finds  application as an emulsifier and plasticizer in a variety of paints and plastics, including polyvinyl chloride (PVC) and some types of polystyrene (Soto et al., 1991).  NP and NPEs  are used as pitch dispersants in the paper industry, and reduce the deposition of wood resins onto the surfaces of process eguipment (Sithole and Allen, 1989). Concerns regarding the environmental impacts of long-term exposure to NP and NPEs have resulted in voluntary use reductions in a number of European countries (Blackburn and Waldock, 1995).  Switzerland and Germany have imposed bans on  the use of NPEs in cleaning products, and the U.K. has introduced measures to voluntarily restrict their domestic use (Blackburn and Waldock, 1995).  To date, no comparable  measures have been documented elsewhere in the world. 1.3.1.2  Chemical and Physical Characteristics  4-NP is manufactured by alkylation of phenol with nonene (Lee and Peart, 1995).  The resulting technical product is a  complex mixture of isomers, predominantly para-substituted,  6  formed by differential branching of the nonyl side chain (Lee and Peart, 1995).  The structure and selected physical  characteristics of 4-NP are provided in Table l.  Table 1.  Structure and selected physical characteristics of 4-nonylphenol  M o l e c u l a r F o r m u l a : C9H19C6H40H Formula Weight ( d a ) :  220.36  P h y s i c a l Form ( 2 5 ° C ) :  liquid  B o i l i n g P o i n t ( ° C ) : 293 - 297 Density (g/mL):  0.950  (from Merck (1983); Aldrich (1995); Fluka (1995))  The chemical is described as a pale yellow and viscous liquid with a mildly phenolic odor (Merck, 1983).  4-NP dissolves  readily in organic solvents but is practically insoluble in water and dilute sodium hydroxide (Merck, 1983).  With a log  kow of 4.12, 4-NP is hydrophobic (Blackburn and Waldock, 1995) and considered to have only low to moderate biodegradability (Swisher, 1987).  7  1.3.1.3  Environmental Distribution and Fate  It is estimated that 80% of APEs produced are NPEs, and that 60% of APEs end up in the aquatic environment as NP or octylphenol (OP) (Lee and Lee, 1996).  Such high estimates for  transfer of NP to ecosystems have resulted in a number of studies aimed at determining regional concentrations in water and sediment. In Europe, Stephanou and Giger (1982) measured NP, nonylphenol monoethoxylate (NP1E0) and nonylphenol diethoxylate (NP2EO) in effluents from secondary sewage treatment plants (STPs) near Zurich, Switzerland. Concentrations of NP in the effluents ranged from 10 - 35 M9/L and varied with the origin of the untreated wastewaters (Stephanou and Giger, 1982).  Influents with high levels of  nonionic detergents yielded correspondingly high concentrations of NP in their secondary treatment effluents (Stephanou and Giger, 1982).  Also in Switzerland, studies on  the river Glatt, which receives a series of treated effluents along its length, found water concentrations from < 0.3 to 45 /xg/L total extractable NP (TENP), with most values < 10 /ig/L TENP (Ahel et al. , 1993; Ahel et al. , 1994). In Croatia, researchers measured levels of NP in untreated municipal wastewaters and the nearby Krka River estuary.  Outfall levels ranged from < 0.5 to 419 /xg/L, while  those in the estuary were much lower at < 20 to 1200 ng/L  8  (Kvestak and Ahel, 1994).  Since anaerobic or methanogenic  conditions favor the formation of NP, the very high concentration of 419 /ig/L measured in one wastewater sample was attributed to the presence of anaerobic "pockets" within the sewer system (Kvestak and Ahel, 1994).  The ratio of  parent to metabolite NP compounds varied significantly throughout the estuary, suggesting that biotransformation reactions were important in determining distribution and fate (Kvestak and Ahel, 1994).  However, distance from the outfall  was still the primary factor affecting distribution of NP and NP derivatives in the estuary (Kvestak and Ahel, 1994).  Wind  conditions, and their influence on direction and velocity of surface currents and saline and freshwater mixing, also played a significant role (Legovic et al.,  1989).  The Croatian study raised several points of interest. The researchers determined that the total annual input of NP and short-chain metabolites of NPEs (NPIEO and NP2E0) into the estuary was only 3 - 4 % of the input to the system (Kvestak and Ahel, 1994).  This differed significantly from results  obtained in areas where wastewaters were treated prior to disposal into receiving waters.  In this circumstance, NP,  NPIEO and NP2E0 formed over 30% of the total receiving water input (Giger et al.,  1987; Ahel et al.,  1994).  These results  emphasize the importance of biotransformation processes in the concentration of NP in aquatic ecosystems.  9  As well, the Croatian study revealed a vertical profile within the estuary that included concentration maxima for NP, NPIEO and NP2E0 at phase boundaries (Kvestak and Ahel, 1994). NP metabolites concentrated at the boundary between air and freshwater or freshwater and seawater appeared stable and were not efficiently eliminated by tidal flushing (Kvestak and Ahel, 1994).  The researchers concluded that the residence  time for NP in the estuary exceeded that of the estimated residence time for freshwater or seawater (Kvestak and Ahel, 1994).  Under these circumstances, the rate of formation of  NP, NPIEO and NP2E0 was significantly faster than their rates of elimination from the estuary (Kvestak and Ahel, 1994). In the U.K., Blackburn and Waldock (1995) surveyed NP concentrations in rivers and estuaries throughout England and Wales.  Samples were collected at five or six sites along six  rivers selected to provide a wide range of potential APE inputs and concentrations.  Six estuaries and one harbour were  sampled, along with final effluents from twelve STPs and three outfalls.  Over 80% of the samples collected contained NP  concentrations of < 0.1 nq/l.,  with concentrations in rivers  and sewage effluents ranging from < 0.2 to 12 A»g/L (Blackburn and Waldock, 1995).  One exception was the river Aire, where  river water concentrations were up to 180 M9/L total NP and sewage effluents discharging into the river were as high as 330 /ig/L total NP (Blackburn and Waldock, 1995).  10  Results from the study indicated that high concentrations of NP were derived from industrial wastewaters containing heavy surfactant loads, particularly those associated with textile operations (Blackburn and Waldock, 1995).  This was  the case for the river Aire, where water concentrations of NP at some sampling locations exceeded the Daphnia  chronic No  Observed Effect Concentration (NOEC) of 24 /ig/L (Comber et al. , 1993) and approached acutely toxic levels of 300 /ig/L for shrimp and 190 /i<?/L for salmon (McLeese et al.,  1981).  Concentrations of NP and NPE in U.S. rivers were examined in a 1992 study sponsored by the Chemical Manufacturers Association.  Water column and bottom sediment samples were  collected from thirty rivers selected at random from the Environmental Protection Agency's (Naylor et al.,  1992).  "U.S. river reach" database  Sampling sites were located at points  on the river receiving municipal or industrial wastewater inputs (Naylor et al.,  1992).  The study's findings indicated  that for 60 - 75% of the samples, NP and NPE concentrations were < 0.1 /ig/L (Naylor et al.,  1992).  Concentrations of NP  in the river water samples ranged from < 0.11 to 0.64 /ig/L, averaging 0.12 /ig/L (Naylor et al.,  1992).  Sediment NP  concentrations were higher, ranging from < 2.9 to 2960 /ig/kg and averaging 162 /ig/kg (Naylor et al.,  1992).  The  researchers estimated interstitial water concentrations to be similar to concentrations in the water column and concluded  11  that the tendency of NP to partition into sediments resulted in minimal exposure risk to aquatic life (Naylor et al., 1992). Currently, there is little information regarding the occurrence and distribution of NP in Canadian waters.  Lee and  Peart (1995) developed an analytical method for 4-NP using effluent samples from five STPs in Toronto and Mississauga. Primary effluent and pre- and post-chlorination final effluents at each site were analyzed for the presence of NP. Primary effluent concentrations ranged from 2.8 to 30  fig/L,  while those in the post-chlorination final effluent were 0.8 to 15 /ig/L (Lee and Peart, 1995).  NP was detected in all  samples, with concentrations two to five times higher in primary effluent than in final effluent (Lee and Peart, 1995). The concentration of NP in the final effluent was not affected by chlorination, since pre- and post-chlorination concentrations were found to be comparable (Lee and Peart, 1995). The researchers also examined sludge samples from two Toronto STPs, and seven sediments from nearby receiving waters.  NP concentrations in the sludge samples were very  high, 137 and 470 /ig/g (Lee and Peart, 1995).  Concentrations  of NP in the sediments ranged from 0.29 to 6.73 /ig/g (Lee and Peart, 1995).  A tenth sample collected at an STP outfall was  a mixture of sludge and sediment, and contained an NP  12  concentration of 41.1 fJ.g/g (Lee and Peart, 1995). The findings of Lee and Peart emphasize the importance of sediments and sludge as sinks for NP in the aquatic environment.  With its low water solubility and corresponding  lipophilicity, NP released into a receiving water will partition into the sediment bed at a rate determined primarily by water discharge patterns and the total organic content (TOC) of the sediment (Naylor et al., 1992).  STP sludges,  with high TOCs, serve as reservoirs for NP either present in influent wastewaters or produced as a result of microbial action on NPEs. In Switzerland, concerns over the safety of drinking water supplies led to an examination of the occurrence of NP in groundwater reservoirs. Where river concentrations were 1.8 to 25 /xg/L, groundwater levels were significantly lower and ranged from < 0.1 to 1 M9/L (Ahel et al.,  1996).  The  findings suggested efficient removal of NP occurred during infiltration, particularly in the first 2.5 m of the aquifer (Ahel et al.,  1996).  However, significant seasonal  variability was observed in the elimination of NP from the groundwater, with lowest efficiencies during the winter months (Ahel et al.,  1996).  This observed decrease in efficiency was  attributed to the conditions of lower temperature that predominate in winter, and was considered to provide additional proof linking the elimination of NP from ecosystems  13  to biotic action (Ahel et al.,  1996).  Information on residence times for NP within aquatic ecosystems is scant. Ahel et al.  (1996) determined an average  residence time for NP within an aquifer was > 10 days.  A  combination of biological influences and physicochemical factors, including temperature, sediment organic content and dilution and dispersion forces (Ahel et al.,  1996; Blackburn  and Waldock, 1995), may interact to ultimately determine the length of time the chemical remains in an ecosystem. 1.3.1.4  Biological Effects  With an oral LDso (rat) of 1620 mg/kg (Registry of Toxic Effects, 1990 cited in Encyclopedia of Chemical Technology, 1992), 4-NP can be considered moderately toxic (Klaassen and Doull, 1980) to mammalian systems. The acute effects of NP are more pronounced in aquatic organisms, where 96 h LC^ values range from 130 - 300 /ig/L (Table 2) (McLeese et  al.,  1981; Armstrong and Kingsbury, 1979 cited in McLeese et  al.,  1981). The primary concern with NP is the effect of exposure to low concentrations over long periods of time.  Increasing  evidence suggests that chronic exposure to APs, and particularly NP, can result in disruptions to the endocrine systems of animals.  Specifically, NP has been demonstrated to  elicit an estrogenic response in mammals and fish (Soto et al.,  1991; Jobling and Sumpter, 1993; Lech et al.,  14  1996; Lee  Table 2.  96 h LC50 values for NP for selected aquatic species  Species  shrimp  salmon  fingerling  fingerling  rainbow  brook  trout  trout  LC50 (/ig/L)  130 - 190  300  230  145  (adapted from Stephanou and Giger (1982))  and Lee, 1996).  Natural estrogens in the body are important  regulators of reproductive development and influence the activity of the neuroendocrine and skeletal systems (Lee and Lee, 1996).  Estrogenic compounds are those which mimic the  effect of estrogen, and their presence in female and male organisms can significantly impact normal hormonal activity and regulation.  The ability of NP to disrupt normal hormonal  function within the body raises serious concerns about its impact on reproductive success in individuals and, ultimately, species survival.  Estrogenic and other endocrine-disrupting  chemicals have been associated with abnormal thyroid function in birds (Moccia et al.,  1986) and fish (Moccia et al.,  1981),  decreased fertility and hatching success in birds (Shugart, 1980; Kubiak et al.,  1989), fish (Mac et al.,  15  1988;  Leatherland, 1992 cited in Colburn et al.,  1993) and mammals  (Reijinders, 1986), feminization (Fry and Toone, 1981; Beland, 1989 cited in Colburn et al.,  1993; Munkittrick et al.,  1991)  and masculinization effects (Fry and Toone, 1981; Ellis and Pattisina, 1990; Davis and Bortone, 1992 cited in Colburn et al. , 1993), and alterations in immune system function (Erdman, 1988 cited in Colburn et al.,  1993; Martineua et al.,  1988).  In humans, exposure to estrogenic compounds in the environment may account for observed increases in breast and prostatic cancers (Hoel et al.,  1992), as well as a 400% increase in  ectopic pregnancies in the U.S. between 1970 and 1987 (Nederlof et al.,  1990) and a 50% decrease in sperm counts  worldwide since the 1940's (Sharpe and Skakkebaek, 1993). In 1991, Dr. Ana Soto and her co-workers at Tufts University Health Sciences Schools in Boston, MA published a paper in the journal "Environmental Health Perspectives" describing the isolation and identification of NP leaching from plastic centrifuge tubes (Soto et al.,  1991).  This work  linked NP with increased cell activity in estrogen-sensitive breast tumor cells (Soto et al.,  1991).  NP has been found to  stimulate uterine growth in immature female rats (Lee and Lee, 1996) and to displace the powerful natural estrogen estradiol at estrogen receptor sites (Mueller and Kim, 1978). and Sumpter (1993) used an in vitro  Jobling  bioassay to test the  effect of APs including NP on the production of vitellogenin,  16  a substance secreted by the liver cells of female fish in response to estrogen (Chen, 1983).  Vitellogenin is associated  with yolk production in eggs, and its synthesis is estrogendependent (Jobling and Sumpter, 1993).  NP was found to  stimulate vitellogenin production, further evidence of its estrogenicity (Jobling and Sumpter, 1993).  Lech et al.  (1996)  continued the work with fish liver vitellogenin, and demonstrated that the estrogenic effects of NP occur at concentrations well below levels of acute toxicity (Lech et al.,  1996).  The estrogenic threshold concentration of NP  based on the production of vitellogenin was determined as 14 jug/L, considerably lower than the 72 h LCso value for 50 200 g rainbow trout of 193 //g/L (Lech et al.,  1996).  The estrogenic action of NP is apparently associated with positioning of the alkyl side chain para to the phenolic group (Jobling and Sumpter, 1993), as with 4-NP. group is moved to either the ortho  When the alkyl  or meta position,  estrogenic activity is lost (Jobling and Sumpter, 1993). Colburn et al.  (1993) postulated that NP is an estrogen  agonist, and acts through the estrogen receptor.  This theory  is substantiated by evidence that its action can be blocked by estrogen antagonists (Lee and Lee, 1996). Because of its lipophilicity, NP is taken up by fatty tissues and may bioconcentrate and bioaccumulate.  In a  bioaccumulation study using rainbow trout, Lewis and Lech  17  (1996) found NP concentrated in the liver, fat, kidney and bile and to a lesser extent in the dorsal muscle, heart and gill (Lewis and Lech, 1996).  The apparent half-life for NP in  rainbow trout tissues was approximately 20 hours in fat and dorsal muscle and approximately 6 hours in the liver (Lewis and Lech, 1996).  Since fish utilize fat as an energy source  rather than carbohydrates (Babin and Vernier, 1989), NP present in fatty tissue may be released into the bloodstream when the fat is utilized (Jobling and Sumpter, 1993), thereby re-exposing the animal to the chemical.  There is little  information regarding the need for metabolism in order for NP to express its estrogenicity.  Studies conducted in  vitro,  however, suggest that prior metabolism is not reguired (Lee and Lee, 1996).  NP is eventually metabolized, apparently  through oxidation and glucuronic acid conjugation, and excreted in the bile (Lewis and Lech, 1996). 1.3.1.5  Analytical Methods  Analysis of 4-NP is complicated by its occurrence at relatively low concentrations in complex samples containing a wide variety of interfering substances (Stephanou and Giger, 1982).  For this reason, it is necessary to extract and  isolate the target compound prior to analysis.  Exhaustive  steam distillation (Naylor et al., 1992), solvent-extraction (Stephanou and Giger, 1982; Giger et al.,  1984; Lee and Peart,  1995) or a combination of these (Ahel et al., 1996) are  18  commonly used to extract NP from water and effluent samples. Steam distillation may also be applied to extract NP from sludge and soils after first suspending the solids in water (Ahel and Giger, 1985), and Lee and Peart (1995) used supercritical fluid extraction (SFE) to achieve isolation of NP from sludge samples. The Wickbold procedure of gas stripping and precipitation commonly applied to larger APEs is not suitable for NP (Stephanou and Giger, 1982). Once isolated, 4-NP may be analyzed directly or derivatized prior to analysis.  4-NP consists of a mixture of  isomers and even the cleanest formulations contain a number of low-level impurities (Sundaram, 1995).  This makes  chromatographic resolution difficult and may lead to overestimation of the amount of compound present (Sundaram, 1995). Derivatization provides greater analytical stability, as well as increased refinement and preconcentration of NP in the sample (Lee and Peart, 1995).  Earlier attempts to form  readily detectable fluorobujtyrl (Coburn et al.,  1976;  Lamparski and Nestrick, 1978) and dinitrophenyl (Cohen et  al.,  1969) derivatives of NP were unsuccessful, due to steric interference in the chemical procedure by the large nonyl group (Sundaram et al.,  1980).  Lee and Peart (1995) applied  an acetylation procedure for phenols to NP and developed an in situ  derivatization and extraction method suitable for  effluents, waters and sludges. The procedure reacts NP with  19  acetic anhydride in the presence of a base, yielding a less polar acetyl derivative.  When analyzed using GC-MS operating  in selected ion monitoring (SIM) mode, the NP acetate allows better chromatographic separation of the 4-NP isomer peaks and a detection limit that is lower by 4-fold than the nonacetylated compound (Lee and Peart, 1995).  Blackburn and  Waldock (1995) also used GC-MS, but in multiple ion detection (MID) mode, to measure NP and other APs in water samples, after first concentrating the target compounds onto C18 solid phase extraction (SPE) columns. HPLC methods are less complex than those for GC-MS and allow for simultaneous analysis of several APs in a sample (Lee and Peart, 1995). Ahel et al.,  Both normal (Kvestak and Ahel, 1994;  1996) and reverse (Sundaram, 1995) phase HPLC are  used to measure NP, with fluorescence detection at 278 - 300 ran in the UV range (Naylor et al., 1994; Sundaram, 1995; Ahel et al.,  1992; Kvestak and Ahel, 1996).  HPLC methods have  general application to a wide variety of APs, although Sundaram et al.  have developed procedures specific for NP in  pesticide formulations (Sundaram, 1995) and forestry matrices (Sundaram et al.,  1980).  Methods for HPLC require less sample  manipulation than those for GC-MS and do not involve extensive derivatization.  For this reason, they may include significant  amounts of interfering substances when sewage and water samples are used (Stephanou and Giger, 1982).  20  Sample analysis  using HPLC is simple and rapid (Sundaram, 1995), but also less selective than GC-MS (Lee and Peart, 1995). A third method, GC with flame ionization detector (FID), has been widely applied to APEs (Favretto et al.,  1978) and is  considered well suited to the analysis of NP (Stephanou and Giger, 1982).  Although less sensitive than HPLC (Sundaram,  1995), GC provides a readily available and less expensive alternative.  Stephanou and Giger (1982) applied GC-FID to the  quantitative determination of NP, NPIEO and NP2EO in secondary STP effluents and found the technique to have good specificity.  McLeese et al.  (1981) used GC-FID to measure NP  concentrations in water samples. Method detection limits (MDL) are in the range of 10 /xg/L (Stephanou and Giger, 1982), considerably higher than the 10 - 20 ng/L for HPLC (Kvestak and Ahel, 1994; Ahel et al.,  1996) and 30 - 200 ng/L for GC-MS  (Blackburn and Waldock, 1995; Lee and Peart, 1995).  However,  concentration of the sample prior to analysis can be used to raise analyte levels above the MDL (McLeese et al.,  1981;  Stephanou and Giger, 1982). 1.3.2  Semipermeable Membrane Devices  1.3.2.1  Description and Use  Xenobiotics present in water and sediment may enter the bodies of aquatic organisms through active uptake processes such as eating foods that contain them, or passively by diffusion (Spacie and Hamelink, 1985).  21  During passive uptake,  the chemical travels down the concentration gradient into the organism by crossing semipermeable membranes in the body, such as those in the gills, the lining of the mouth, or the gastrointestinal tract (Spacie and Hamelink, 1985).  When the  rate of uptake of the chemical exceeds its rate of elimination from the body, bioaccumulation occurs. Some chemicals accumulate in fatty tissues where they remain resident, and their continued uptake leads to concentration.  Since the  1960's, bioaccumulated and bioconcentrated chemicals have been associated with individual organism mortality and reduced reproductive success. As well, these chemicals have demonstrated an ability to transfer through food chains, negatively affecting species at successively higher trophic levels (Spacie and Hamelink, 1985). In 1990, Huckins et al. situ  published a description of an in  sampler intended to mimic the action of bioconcentration  in organisms (Huckins et al.,  1990).  The passive  concentrator, called a semipermeable membrane device or SPMD, consisted of a thin layer of fish lipid enclosed within nonporous, low-density polyethylene (LDPE) layflat tubing (Figure 2). SPMDs were designed to sequester dissolved organic contaminants in water, providing information on the nature of bioavailable chemicals present and their concentrations. Chemicals in the surrounding water enter the SPMD by  22  travelling down the concentration gradient across the membrane layer of LDPE tubing.  Although described as nonporous, the  LDPE actually contains transient channels of < 10 Angstroms in al.,  diameter (Hwang and Kammermeyer, 1975 cited in Huckins et 1996).  Materials such as silicone, polypropylene, polyvinyl  chloride, and plasma-treated microporous tubing have also found application as SPMD membranes (Huckins et al.,  1996).  The sequestration phase of the SPMD is most commonly triolein (1,2,3-tri[cis-9-octadecenoyl]glycerol), a neutral triglyceride known to constitute a major portion of the neutral lipids in most fish (Henderson and Tocher, 1987). High molecular weight (> 600 Daltons) lipid surrogates such as some silicone fluids, or a mixture of lipids characteristic to a particular organism of interest may also be used for sequestration (Huckins et al.,  1996).  Triolein, however, is  considered the compound of choice for SPMDs.  It is readily  available in contaminant-free synthetic form and has a relatively low melting point (Huckins et al.,  1996), keeping  it in a liquid state at most ambient water temperatures.  As  well, the equilibrium octanol-water partition coefficient (log kow) and bioconcentration factor (BCF) values for many hydrophobic organic compounds correlate well with the equilibrium triolein-water partition coefficients (log ktw) , allowing estimation of the uptake of a target chemical by the  23  triolein in an SPMD from the published log kow value for the chemical (Huckins et al.,  1996).  Two U.S. patents (#5,098,573 and #5,395,426) licensed to Environmental Sampling Technologies (EST), St. Joseph, MO. cover the assembly of SPMDs and the dialytic recovery of analytes from them (Huckins et al.,  1996).  The SPMD is  deployed in the field for a specified period of time, commonly 28 or 60 days, then retrieved for processing.  In the  laboratory, external debris is removed by washing, and the cleaned SPMD is submerged in an organic solvent, usually hexane, for 48 hours (Huckins et al.,  1996).  Chemicals  seguestered in the triolein during the field exposure period dialyze into the hexane, which is then concentrated and analyzed (Huckins et al.,  1996).  Sample clean-up procedures,  such as gel permeation chromatography (GPC), may be applied to the hexane dialysate prior to analysis (Huckins et al.,  1996).  During field exposures, SPMDs may be suspended in the water column, or placed on or directly beneath the surface of the sediment bed.  To prevent entanglement and to maximize  surface area exposure, SPMDs are usually anchored by attachment to aluminum frame panels (Figure 3)(Ellis et 1995) or enclosed within galvanized conduits (Huckins et 1996).  al., al.,  These deployment structures permit adeguate water  exchange while protecting the SPMD from the physical effects of water turbulence or interference from animals and humans  24  Figure 2.  Photograph of SPMD  Lipid-Filled Layfl*t LDPE Tubing Aluminum Frame Copper Screen  Machine Screws  Figure 3.  Aluminum frame deployment panel for SPMD (from Ellis et al.  25  (1995))  (Huckins et al.,  1996).  Copper mesh screens are often used to  cover SPMD field deployment structures.  In addition to  providing additional physical protection, the copper is toxic to aquatic organisms and acts to reduce biofouling of the SPMD's surface (Huckins et al., 1.3.2.2  1996).  Estimating Analyte Water Concentrations With SPMDs  Mathematical models developed by Huckins et al. et al.,  1993; Huckins et al.,  (Huckins  1996) allow estimation of the  water concentration of an analyte from levels measured in the triolein of an exposed SPMD.  The calculated result provides a  time-averaged measure of analyte water concentrations during the exposure period of the SPMD (Prest et al.,  1995).  At the start of the exposure period, the concentration of analyte in the water is much greater than that in the triolein.  Moreover, the ratio of analyte concentration in the  lipid to that in water is much less than its equilibrium partition coefficient (Ellis et al.,  1995).  An analyte which  is not readily water-soluble will preferentially move from the water into the triolein at a rate described by linear firstorder kinetics (Huckins et al.,  1993).  The relationship  between the analyte's water concentration and the concentration measured in the triolein is then described by:  CL = C w .K m w k 0 A t / V L  26  (1)  where, CL  = the concentration of analyte in triolein  Cw  = the concentration of analyte in water  K^  = the membrane/water partition coefficient  k0  = the overall mass transfer coefficient of the analyte into the triolein  A  = the membrane surface area  t  = time  VL  = the volume of the lipid (from Huckins et al.  (1996))  The relationship between Cw and CL can be established by repeated laboratory determinations of C^, when Cw is known (Huckins et al.,  1996).  For a system at equilibrium, the model simplifies to:  CL = CMKLW  (2)  where, KLW  = the equilibrium lipid/water partition coefficient (from Huckins et al.  (1996))  The KLW is approximated by the equilibrium octanol/water partition coefficient, and the C„ can then be estimated using the measured CL value for the analyte and its KOM (Huckins et al.,  1996).  27  Substantial absorption of the analyte from the water may occur within the membrane layer of the SPMD itself (Huckins et al.,  1996).  The membrane's contribution to the analyte  concentration in the final hexane dialysate may be as high as 50%, and should be accounted for by:  CL = Ko/(^  + KML + MJ  (3)  where, A^, = the total mass of the analyte in the dialysate ML  = the mass of the triolein  KHL  = the membrane/triolein partition coefficient  MM  = the mass of the membrane (from Huckins et al.  (1996))  The passage of chemicals across the LDPE membrane forms al.,  the rate-limiting step of the uptake process (Huckins et 1993).  Uptake is independent of water turbulence and is  controlled primarily by the available membrane surface area (Huckins et al.,  1996).  The high membrane-surface-area-to-  lipid-volume ratio of the SPMD maximizes the opportunity for uptake, allowing sampling rates of 1 - 10 L per day for most organic contaminants (Huckins et al.,  1994).  The physico-  chemical nature of the contaminant, including its KOM value and molecular size and shape, also play a role in determining the rate at which it is sequestered by the SPMD (Huckins et  28  al.,  1996). Uptake rates for SPMDs are influenced by the presence of biofouling on the membrane surface and by changes in the temperature of the exposure system (Huckins et al.,  1996).  Biofouling reduces the rate of uptake by increasing membrane resistance to mass transfer, but generally does not stop uptake completely (Ellis et al.,  1995).  Fluctuations in  temperature influence the chemical diffusion rate and may alter the free volume characteristics of the membrane (Huckins et al.,  1996).  The effects of biofouling and variable  temperature on the sampling rate may be corrected for by the use of a permeability reference compound (PRC), an isotopically-labelled compound with properties similar to those of the contaminant of interest (Ellis et al.,  1995).  A  known volume of the PRC is spiked into the triolein at the beginning of the exposure period.  The dissipation rate of the  PRC from the triolein then provides an indirect measure of membrane permeability and uptake rates (Ellis et al., 1.3.2.3  1995).  Advantages and Limitations of the SPMD System  Evidence amassed on the severe impact of pesticides such as DDT on organism survival and reproduction has demonstrated that even at trace levels some organic contaminants may have significant biological and ecosystem effects.  Detecting these  low-level pollutants and tracking their passage through aquatic systems is often difficult, because of the transient  29  nature of the discharges and the large number of non-point sources (Ellis et al.,  1995).  Methods such as  ultrafiltration, which extract trace contaminants from water, require that large volumes of water be passed through the sampling system and sample over only a short time period (eg. 8 h ) , so that non-continuous or episodic discharges may be missed (Ellis et al.,  1995).  Bioavailability studies which examine the levels of contaminants within the bodies of aquatic organisms also face a number of obstacles.  Different animal species have varying  degrees of mobility, and an animal which is motile may travel outside the boundaries of the sampling area.  Although  confining the test organisms, as within a cage or trap, eliminates this difficulty, observed differences in contaminant levels between caged and free-swimming animals suggest the data are not strictly transferrable.  Ellis et  (1995) found that contaminant residue levels for six organochlorine pesticides were significantly lower in feral than in caged fish, possibly as a result of enhanced metabolism associated with activity.  Metabolism and  biotransformation within organisms may complicate identification and tracking of the parent compound.  If  metabolites are excreted, equilibrium for the target parent chemical may never be attained and only low concentrations will be detected within the body of the organism (Ellis  30  al.  et al., 1995).  Bioavailability studies which rely on  concentration of the target chemical within the lipid tissues of the organism face the added dilemma of inter- and intraspecies variability.  The total body lipid content of an  animal varies between species, between individuals of the same species, and even within a particular animal at different life stages, influenced by factors such as age and gender (Schneider, 1982; Schmitt et al.,  1993).  Finally,  bioavailability studies often require large numbers of organisms (Ellis et al.,  1995) in order to ensure a suitably  large data set, a necessity that can be expensive and ecologically wasteful. The application of passive samplers like the SPMD to detection and monitoring of trace contaminants in aquatic systems addresses many of these issues.  SPMDs have been used  successfully with a wide variety of organic contaminants, including a number of halogenated compounds (Erhardt-Zabik et al.,  1990), polycyclic aromatic hydrocarbons (PAHs), alkanes,  phthalates, and organochlorines (Ellis et al.,  1995).  Because  they mimic organic contaminant uptake by the lipid tissues of animals (Huckins et al.,  1996), SPMDs have potential  application to any contaminant known or suspected to accumulate in organisms. Using SPMDs to monitor for trace contaminants has a number of advantages over other sampling procedures. The  31  commercially-produced SPMD is relatively inexpensive to obtain and use (Prest et al.,  1995), and is manufactured essentially  contaminant-free (Huckins et al.,  1996).  Its uniform design  permits development of standardized protocols, improving comparability between data collected from different sites (Prest et al.,  1995).  SPMDs can be deployed simultaneously at  a number of locations, permitting insights into the residence time, transport dynamics, and spatial distribution of a target contaminant (Prest et al.,  1995).  Of key importance is the ability of the SPMD to concentrate contaminants which are below readily detectable limits to levels that can be accurately measured.  Since only  truly dissolved forms of chemicals are seguestered by SPMDs (Huckins et al.,  1990; Huckins et al.,  1993), concentrations  measured in the SPMD dialysate are directly applicable to published water guality criteria (Prest et al.,  1995).  The  long exposure period of an SPMD increases the probability of capturing and recording episodic discharge events (Huckins et al.,  1996). Although many of these advantages are common to all  passive sampling devices (PSDs), the SPMD developed by Huckins et al.  (1990) has several unigue features.  Early passive  samplers such as those of Sodergren (1987), Hassett et  al.  (1989), and Johnson (1991) consisted of a nonpolar solvent such as hexane enclosed within a polar or nonpolar membrane.  32  These solvent-filled models experienced difficulties with leakage of the sequestering solvent (Zabik et al., Litten et al.,  1992;  1993), while PSDs with polar membranes  precluded sequestration of nonpolar contaminants such as DDE and PCBs (Paasivirta et al.,  1991).  The lipid-filled SPMD  provided a stable, high molecular weight sequestration phase, eliminating loss through the membrane and permitting very high concentration factors (Prest et al.,  1995).  Since the SPMD  membrane controls the rate of uptake, uptake kinetics for a particular chemical are more reproducible and modelling is simplified (Huckins et al.,  1993).  The development of  reliable and consistent mathematical models increases the accuracy of ambient water concentration estimates for dissolved contaminants (Ellis et al.,  1995).  Both in structure and function, SPMD membranes resemble those of organisms (Ellis et al.,  1995), and contaminant  uptake by SPMDs is considered to mimic bioconcentration processes (Ellis et al., al.,  1996).  1995; Prest et al.,  1995; Huckins et  Comparative studies indicate that nonporous  polymers such as LDPE have a molecular permeability size limitation that is very similar to that of biomembranes (Opperhuizen et al.,  1985).  As well, the process of  unmediated diffusion is similar to that in biomembranes (Lieb and Stein, 1969) and requires that the diffusing molecule be nonionic in nature and fully dissolved (Huckins et al.,  33  1996).  Ionic particles and those associated with particulates or dissolved organic carbon are unable to cross the nonpolar, nonporous membrane (Huckins et al.,  1996).  However, the SPMD is a thermodynamically-passive system (Ellis et al.,  1995), while organisms participate in a number  of active absorption and accumulation processes.  Dissolved  and particle-bound organic contaminants may enter animals through direct ingestion (active uptake) or by ingestion of foods that contain them (Ellis et al.,  1995).  For  contaminants which accumulate up food chains to successively higher trophic levels, dietary ingestion is a major route of uptake.  Metabolism, biotransformation, and active excretion  of compounds are also processes unique to organisms that SPMDs cannot duplicate.  For a compound which is biotransformed, the  SPMD may more realistically portray ambient water concentrations (Ellis et al.,  1995).  In addition, the  distribution of chemicals sequestered may differ somewhat between an animal and an SPMD.  In a comparative study of 25  organochlorine pesticides, Prest et al.  (1995) found that  although the total mass of compounds accumulated by SPMDs and bivalves was comparable, their distributions differed.  SPMDs  had more penta- and tetra-chlorinated organochlorines, while bivalves sequestered primarily hexa- congeners (Prest et 1995).  al.,  For this reason, the most accurate representation of  the bioavailability of a contaminant utilizes information  34  obtained from both sources (Prest et al., al.,  1995; Huckins et  1996). The inanimate nature of SPMDs allows their placement in  locations where conditions such as high temperature or turbidity preclude the use of living organisms (Prest et 1995).  al.,  Their greater lipid content - 15 - 20% lipid as  compared with 0.5 - 10% for aquatic organisms - increases analyte capacity and facilitates the determination of ambient water concentrations (Huckins et al.,  1996).  Recently,  attention has turned to using SPMDs to extract and concentrate ultra-trace level compounds from water for subsequent testing of the dialysate using bioassay techniques such as MutatoxR, MicrotoxR, or fish EROD induction (Prest et al., et al.,  1996).  1995; Huckins  Screening tests using SPMD dialysates are used  to isolate samples producing a toxic response, which then undergo further testing with more expensive analytical methods such as GC-MS (Huckins et al.,  1996).  SPMDs will sequester air-borne contaminants such as PCBs very effectively (Petty et al.,  1993), and care must be taken  during transport, deployment and retrieval to avoid accidental contamination by non-target compounds (Ellis et al.,  1995).  SPMDs should be transported to and from testing sites in gastight metal cans, and field blanks or travel controls employed to account for accidental exposure to extraneous chemicals (Ellis et al.,  1995).  Biofouling of the external surface of  35  the LDPE membrane is common during field exposures. This microbial and microalgal growth, termed Aufwuch's colonies, reduces the sampling rate of the SPMD and impedes contaminant uptake (Ellis et al.,  1995).  Periodic application of a  biocide to the surface of the SPMD may reduce the amount of Aufwuch's growth, although Prest et al. (1995) reported that they found this practice labor-intensive and largely ineffective.  Ellis et al.  (1995) employed copper-plated  screens in their field deployment structures, the copper acting to inhibit microbial growth.  In cases of heavy  Aufwuch's colonization, particulates may become imbedded in the matrix of the colony (Ellis et al.,  1995).  These  particulates are not considered to contribute to the chemical residues sequestered by the SPMD, however, since the polar, proteinaceous nature of the Aufwuch's growth restricts their transport to the membrane (Ellis et al.,  1995).  SPMDs have found application in the sampling and concentration of trace and ultra-trace level contaminants in aquatic systems. By applying mathematical models, and with only limited information on exposure conditions (Prest et  al.,  1995), average ambient water concentrations for low-level contaminants can be calculated.  SPMDs also provide insights  into bioaccumulation processes in aquatic organisms, supplementing the information obtained using biomonitoring techniques.  36  1.4  Research Objectives The objective of this thesis research project was to  assess the suitability of SPMDs for sequestration of waterborne 4-NP.  It was evident during initial instrumental  analysis that a derivatization procedure for 4-NP was required to ensure consistent, stable chromatographic results. Subsequent research focussed on development of a suitable acetylation method for 4-NP present in SPMD dialysates.  37  2.0  MATERIALS AND EXPERIMENTAL METHODS  2.1  Materials  2.1.1  SPMD Components Low-density layflat polyethylene tubing (27 mm wide and  60 /im wall thickness) was obtained from Cope Plastics Inc., St. Louis, MO.  Triolein (1,2,3-Tri-[(cis)-9-  octadecenoyl]glycerol, 99%) was purchased from Sigma Chemical Co., St. Louis, MO.  To maximize compound integrity throughout  testing, the contents of the original 5 g snap-top ampoule of triolein were distributed into 4 clean 14 mL amber vials. All vials were stored in the dark at -15°C.  Immediately prior to  SPMD preparation, a single vial was removed from cold storage and allowed to equilibrate to room temperature.  The contents  of each vial were used in succession. 2.1.2  Target Chemical 4-NP was purchased from FLUKA Chemical Corp.,  Switzerland.  A manufacturer's Certificate of Analysis  (Appendix A) confirmed 90.0% purity by HPLC assay and approximately 85% para-isomer content.  The chemical was  stored in the dark at room temperature. 2.1.3  Solvents and Standard Solutions All solvents were HPLC or pesticide grade and were  obtained from Fisher Scientific, Nepean, Ont.  Solutions of  underivatized standards were prepared in hexane. standards were prepared in methanol.  38  Derivatized  Stock solutions for SPMD  testing were prepared in methanol. All standards and stocks were stored in the dark at 5°C. 2.1.4  Other Materials One liter wide mouth Mason canning jars were obtained  from local retail outlets. Amber glass bottles (500 mL) were from VWR Scientific Ltd., London, Ont. Aluminum foil was obtained from local retail outlets. An Eppendorf Series 2000 adjustable volume pipettor (100 - 1000 /iL capacity) was purchased from Brinkmann Industries (Canada) Ltd., Mississauga, Ont.  Two Nichiryo  Model 5000 adjustable volume pipettors (40 - 200 /iL and 200 1000 juL capacity) were also used for volumes less than 1000 /iL. The heat sealer used to prepare SPMDs was an Impulse Sealer Model CD-300 (320 watts), from Dea Lun Co. Ltd., Taiwan. A 46 cm x 100 cm glass plate was used as a work surface in SPMD preparation. A polypropylene roller was used to distribute the triolein within the LDPE tubing during SPMD preparation. Glass weights used to keep SPMDs submerged were prepared from 5 mm diameter glass rods. Dialysis tubing closures (40 mm width) were obtained from Spectrum Medical Industries Inc., Houston, TX.  39  1000 mL PyrexR and 250 mL Kimax separatory funnels were used to perform acetylations.  40 mL Kimax clear glass vials  were used to perform the Lee-adapted and Wilson procedures. Acetic anhydride (Certified A.C.S.) was obtained from Fisher Scientific, Fair Lawn, N.J.  Subsamples were double-  distilled and washed twice with hexane prior to use in testing.  Cleaned acetic anhydride was stored in the dark at  5°C. Anhydrous potassium carbonate (K2C03, 99.9%) was purchased from BDH Inc., Toronto, ON.  The 5M solution of K2C03 used as  stock in the acetylation procedures was prepared by measuring 18.0 g into 25 mL of Alpha-QR water.  The solution was washed  prior to use with two 30 mL aliquots of hexane to remove organic contaminants.  A 0.1M K2C03 solution was prepared from  the 5M stock by dilution with Alpha-QR water.  The 0.1M K2C03  solution was sometimes prepared directly by adding 6.91 g of K2C03 to 500 mL of Alpha-QE water.  The solution was washed  twice with hexane prior to use. Alpha-QK water was provided by an Alpha-QR water purification system from Millipore Corporation, Bedford, MA. Rotoevaporation was accomplished using a 115 volt Flash Evaporator from Buchler Instruments, Fort Lee, NJ. Tetrahydrofuran (99.96%) was purchased from BDH Inc., Toronto, ON. and stored in the dark at room temperature. The model compound, octylphenol (4-[tert-octyl]phenol,  40  97%), was purchased from Aldrich Chemical Co., Inc., Milwaukee, WI. The dry chemical was stored in the dark at room temperature. Silica gel (GC grade 950, 60 - 200 mesh) was obtained from Fisher Scientific Company, Fair Lawn, NJ.  The silica gel  was activated by heating 12 to 13 hours at 200°C, then stored in a vacuum dessicator prior to use.  The 5% deactivated  silica gel used for sample cleanup was prepared by measuring 4.75 g of activated silica gel into a clean amber vial, then adding 0.25 mL of Alpha-QR water.  The mixture was handshaken  for about a minute, then placed on an automatic shaker for 15 minutes.  At the end of the shaking period, the vial contents  were placed in the vacuum dessicator for use the following day. Anhydrous sodium sulfate was obtained from Fisher Scientific Company, Fair Lawn, NJ.  The sodium sulfate was  heated 12 to 13 hours at 500°C, cooled, then stored in a clean amber bottle. The glass columns used in the silica gel cleanup procedure were from Supelco, Inc., Bellefonte, PA.  The  columns were 160 mm in length, with a 6 mm stem ID and a 30 mL capacity reservoir. PyrexR 15 mL graduated centrifuge tubes were used to perform the nitrogen blowdown procedure.  41  2.2  Experimental Methods  2.2.1  SPMD Preparation Procedures for preparation, deployment, cleanup, and  dialysis of SPMDs were based on those described in Rohr (1994).  A 48 h pre-extraction period was used to remove  additives and contaminants from the polyethylene tubing prior to its use in SPMDs.  Forty-six inch (46") segments of tubing  were cut from the roll, and each segment was rinsed inside by rolling approximately 25 mL of hexane back and forth along its length.  The rinsed tubing was placed in a clean Mason jar,  eight pieces of tubing per jar, and 1 L of hexane was added. The jar was covered with hexane-rinsed aluminum foil, capped tightly, and the contents left at room temperature. After 48 h, the tubing segments were removed using clean forceps and polyethylene gloves.  Each piece was pressed  gently to remove hexane, then draped on hexane-rinsed aluminum foil.  The tubing segments were left to air dry several  minutes, then folded gently and placed in clean amber glass bottles.  The pre-extracted tubing segments were stored in the  dark at 5°C until required. To reduce contamination during SPMD preparation, polyethylene gloves were worn and handling of the tubing was kept to a minimum.  To prepare an SPMD, a piece of pre-  extracted tubing was placed on hexane-washed aluminum foil atop a clean glass plate. All folds or creases were removed,  42  then 1.0 mL (0.91 g) of triolein was slowly drawn up and pipetted into the tubing approximately 10 cm from one end. The pipettor was allowed to drain into the tubing for about 30 seconds after ejection to ensure complete transfer of the viscous triolein, then the tubing was double heat-sealed directly above the triolein.  Care was taken to ensure no  triolein contacted the heat seal as this would compromise the seal's effectiveness.  The tubing was laid flat on the glass  plate and a polypropylene roller was used to distribute the triolein in a thin, even layer along the length of the tubing. Air bubbles were removed during distribution of the triolein, as these would interfere with diffusion.  The open end of the  tubing was double heat-sealed, and one to two glass clips were threaded on to aid in submersion of the SPMD.  Finally, one  end of the tubing was rotated 180 degrees to form a Mobius strip formation and the two ends were double heat-sealed together.  SPMDs were stored at 5°C in the dark in sealed  amber glass bottles until reguired. 2.2.2  Preparation of Glassware Glassware used in testing was washed with phosphate-free  cleanser and rinsed thoroughly with Alpha-QB water, then fired for 1 h in a muffle furnace at 425°C to remove organic contaminants.  Immediately prior to use, the glassware was  rinsed with hexane then allowed to dry.  43  2.2.3  Preparation of Test Solutions Test solutions were prepared from concentrated (1 g/L)  methanol-based stocks. A measured volume of concentrated stock was added to tap water in a volumetric flask, and the volume brought to 1 L.  The solution was mixed well by  inverting the flask several times and 800 mL distributed into a clean Mason jar. Control jars consisted of tap water only. After distribution of the test solution, SPMDs were carefully added to each jar using hexane-rinsed forceps and polyethylene gloves for handling.  Jars were covered with two layers of  hexane-rinsed foil, then capped tightly and placed in the dark at room temperature. 2.2.4  Cleanup and Dialysis of SPMDs SPMD cleanup procedures were designed to simulate  conditions after field exposure, and consisted of an organic solvent rinse followed by a tap water wash.  With field  testing, these steps would assist in the removal of extraneous organic material from the exterior of the SPMD that could interfere with or complicate analysis of the dialysate.  At  the end of the exposure period, the SPMD was removed from the test solution and transferred to a clean amber glass bottle. Approximately 80 mL of hexane was added and the bottle was rolled gently to rinse the exterior of the tubing.  After  rolling for about 30 seconds to ensure contact of the solvent with all of the tubing surface, the SPMD was removed using  44  hexane-rinsed forceps and polyethylene gloves, and placed over hexane-rinsed aluminum foil to air dry.  The SPMD was then  rinsed under a gently running stream of cool tap water and allowed to air dry for 3 - 5 minutes.  Finally, acetone was  applied to the exterior of the tubing using a Pasteur pipette, to remove any last traces of water.  The clean, dry SPMD was  transferred to a clean 250 mL erlenmeyer flask and a dialysis tubing closure attached at the position of the heat seal.  The  closure could then be rested on the mouth of the erlenmeyer, allowing the tubing to hang suspended in the flask.  210 mL of  hexane was added and three layers of hexane-rinsed foil used to cap the flask.  The SPMD was placed in the dark at room  temperature to dialyze for 48 h. After 48 h, the tubing was removed from the erlenmeyer flask and discarded.  The flask containing the dialysate was  covered with two layers of hexane-rinsed foil and stored in the dark at -15°C to await analysis. 2.2.5  Analysis of SPMD Extracts Dialysates were first concentrated by rotoevaporation  before being analyzed using GC-FID.  A water bath of 40 -  45°C was used to concentrate the dialysate to 2 mL. The concentrated sample was then run on the GC. 2.2.6  Silica Gel Cleanup Procedure To perform the silica gel cleanup procedure, a small plug  of glass wool was first placed in the bottom of a glass  45  column, then the column was rinsed and filled to the base of the reservoir with hexane.  5% deactivated silica gel was  added slowly to the column to a height of 5 cm.  The column  was drained to just above the silica bed and a 5 mm layer of anhydrous sodium sulfate was added.  The prepared column was  then washed with 5 mL of hexane. The sample of organic extract collected after derivatization was concentrated to approximately 3 mL by rotoevaporation in a 40 - 45°C water bath, then further concentrated to 1 mL under a gentle stream of nitrogen.  One  mL of iso-octane was used as a keeper during concentration of the sample.  The concentrated sample was added to the silica  gel bed, and sequentially eluted using 10 mL of hexane, 10 mL of 50% methylene chloride in hexane, and 10 mL of 1% methanol in methylene chloride.  The first fraction was discarded.  Fractions 2 and 3 were collected and concentrated to 2 mL using the procedure described above before being analyzed.  2.3 2.3.1  Analytical Techniques Gas Chromatography Initially, gas chromatography was conducted using a  Hewlett Packard 5890 Series II gas chromatograph with flame ionization detector.  The fused silica DB-5 column was 30.0 m  long with 0.32 mm ID and film thickness 0.25 /im. Helium was used as carrier gas, with a linear velocity of 20 cm/s at  46  280°C.  Make-up gas consisted of helium at 20 mL/min, hydrogen  at 30 mL/min and air at 400 mL/min.  The oven temperature was  raised from 50°C to 270°C at 20°/min to 145°C, 2°/min to 175°C and 20°/miri to 270°C. Commencing in August 1996, gas chromatography was conducted using a Hewlett Packard Model 6890.  To achieve  sharper chromatographic resolution and to shorten retention times, the carrier gas was changed from helium to hydrogen. The average linear velocity for the hydrogen carrier was 66 cm/s.  Make-up gas flow for the FID was hydrogen at 30 mL/min  and air at 400 mL/min.  The oven temperature was raised from  50°C to 275°C, at 15"/rain to 140°C, 2°/min to 155°C and 30°/min to 275°C. Selected samples were analyzed by combination gas chromatography-mass spectrometry (GC-MS) using a low resolution instrument (HP 6890) operated in the selected ion monitoring (SIM) mode.  The fused silica DB-5 GC  column was 30.0 m long with 0.25 mm ID and film thickness 0.25 /im. Helium was used as carrier gas, with a linear velocity of 37 cm/s. The oven temperature was raised from 50°C to 250°C at 15°/min  to  150°C and 5°/min to 250°C. The  electron voltage during operation was 1920 eV. 2.3.2  Calibration Standards Underivatized 4-NP calibration standards were prepared in  50 mL volumes of hexane using a hexane-based 1 g/L 4-NP stock.  47  The volume of stock added for each concentration of standard is provided in Table 3. Derivatized standards were prepared using 25 mL volumes of 0.1M K2C03, 3 mL of acetic anhydride and a methanol-based 1 g/L 4-NP stock.  Table 4 provides the volume of stock added  for each standard.  The reactants were mixed by hand shaking  or using an automatic shaker, and the 4-NP acetate was then extracted using a 5 mL volume of hexane. Once the Wilson procedure was adopted, the procedure for preparing derivatized standards was adjusted to make it comparable with that of sample derivatization.  A 5 mL volume of 0.1M K2C03 was placed  in a clear glass vial and a known volume of methanol-based 1 g/L NP stock added.  Two mL of acetic anhydride and 3 mL  of hexane were added and the vial was hand-shaken for 2 minutes venting often.  The organic layer was collected and  the aqueous fraction extracted once more using a 2 mL aliquot of hexane.  The combined organic extract was washed with 3 mL  of 0.05M K2C03, then collected for analysis.  48  Table 3.  Volume of stock added for underivatized standards  Concentration of Underivatized  (mL)  Standard (mg/L) 100  5.0  50  2.5  25  1.25  12.5  Table 4.  Volume of stock added  0.625  Volume of stock added for derivatized standards  Concentration of Derivatized Standard (mg/L)  Volume of stock added (ML)  160  800  80  400  40  200  20  100  49  2.3.3  Method Detection Limit and Reliable Detection Limit A series of experiments was conducted to establish  detection limits for testing.  By employing progressively  lower chemical concentrations, the Method Detection Limit (MDL) and Reliable Detection Limit (RDL) for the final GC injectate were determined to be 0.1 mg/L and 1.0 mg/L, respectively.  50  3.0  METHOD DEVELOPMENT  3.1  Preliminary SPMD Testing  3.1.1  Summary of Procedure Two static nonrenewal tests and one static renewal test  were performed.  All tests consisted of one control and three  replicates of the test 4-NP concentration.  4-NP  concentrations of 10 and 100 iiq/'L and exposure periods of 1 to 3 weeks were used for the static nonrenewal tests.  For these  tests, jars were left undisturbed for the duration of the exposure period. The static renewal test consisted of a 25 /xg/L test concentration with seven test solution renewals over a 3 week exposure period.  This resulted in a total exposure  concentration of 200 ixg/h NP.  On test solution renewal days,  fresh control and NP solutions were prepared.  Starting with  the control jar, each jar was opened and tilted carefully to allow the water to drain out.  SPMDs were held in place during  draining using hexane-rinsed forceps. Water was removed to less than 100 mL, then new solution was added to bring the volume to 800 mL.  The jars were covered with fresh hexane-  rinsed foil, covered securely, and returned to the dark exposure chamber.  Test solutions in the jars were changed one  at a time and in succession.  A summary of treatment regimes  for the preliminary SPMD testing is provided in Table 5.  For  this preliminary work, SPMD dialysates were analyzed after  51  straight rotoevaporation.  Table 5.  Treatment summary for preliminary SPMD testing  Test Initiation Date  Total Water  Exposure  Renewal  Concentration  Duration  or  (Mg/L)  (wks)  Nonrenewal  5 March 1996  10  1  Nonrenewal  6 March 1996  10  2  Nonrenewal  9 April 1996  100  3  Nonrenewal  29 April 1996  200  3  Renewal  3.1.2  Results of Preliminary Testing A chromatogram for the SPMD processing of a 10 mg/L NP  standard is shown in Figure 4, and the results of the preliminary testing are summarized in Table 6.  Standard  solutions were used to establish the retention time range for NP peaks. The presence of NP in a test dialysate was then determined by a visual comparison of the chromatogram for the dialysate with that of a concurrently run standard.  Peaks on  the test chromatogram with retention times corresponding to those on the NP standard were considered a match and included in the determination of total peak area.  52  Chromatograms for  Control SPMD extracts were used to evaluate background levels of NP and contaminating non-target compounds.  Peaks on test  chromatograms which corresponded with those for background contaminants in the Control were not included in the determination of total peak area for NP. The results indicated that NP was present in all treatment dialysates and absent in the three Controls. No attempt was made to estimate NP concentrations in the test dialysates analyzed May 22 and May 28. For these dialysates, the total NP peak areas greatly exceeded those of the concurrently run standards, and fell outside the limits of the established standard curve.  For the dialysates analyzed March  28, concentration estimates for NP were determined by comparing the total NP peak area in the dialysate with that in the concurrently run 10 mg/L NP standard.  Based on linearity  of response, NP concentrations in the March 28 test dialysates were estimated at 2 - 4 mg/L.  From this, SPMD uptake  efficiencies were estimated by comparing the amount of NP in the initial test solution with that in the final dialysate. The initial amount of NP added to the test solution was 10 /ig/L in a total volume of 800 mL, or 8 /jg of NP.  Final  dialysate amounts were 2 or 4 mg/L in a rotoevaporate volume of 2 mL, or 4 - 8 jug of NP.  The efficiency of uptake for the  SPMD was then the ratio of initial to final NP amount, expressed as a percentage.  For the March 28 test dialysates,  53  SPMD uptake efficiencies were estimated at 50 - 100%.  Table 6.  Summary of results for preliminary SPMD testing  Analysis Date  Sample I.D.  Total NP Peak Area  28 March 1996  10 mg/L NP standard Control  22 May 1996  0  10 /ig/L, 1 wk. exp.  5870  10 /ig/L, 2 wk. exp.  2355  2 mg/L NP standard  11,884  Control  0 3  wk.exp.  160,707  10 mg/L NP standard  61,082  100 jug/L/ 28 May 1996  15,213  Control  0  100 /xg/L, 3 wk.exp.  190,667  200 Mg/ L / 3 wk.exp.  337,230  54  MfiV 28, 1.996  Figure 4.  18:51:48  Gas chromatogram of 10 mg/L NP standard analyzed 28 May 1996  55  3.1.3.  Discussion of Preliminary Testing Results  Results of the preliminary SPMD testing confirmed that the conventional SPMD system could be used to successfully sequester waterborne 4-NP.  Over the concentration range  tested, 10 to 200 /ig/L NP, the amount of 4-NP present in the SPMD dialysate, as represented by the peak area, increased with increasing water exposure concentration.  This result was  consistent with the model of first-order uptake kinetics associated with an SPMD system not in equilibrium. In the comparatively clean setting of laboratory-prepared samples, the NP peaks were easily distinguishable.  Field  samples, however, could be expected to contain many more background contaminants.  Background chemicals with GC peaks  which overlapped those of NP would complicate the analysis, possibly resulting in overestimation of the amount of NP present.  In addition, the analyte exhibited chromatographic  instability, as evidenced by marked variability in total peak area values for replicates of the same sample.  The 10 mg/L NP  standard in Figure 4, for example, had total peak area values for three successively-run replicates of 54,717, 61,082, and 55,819.  This instability was attributed to adsorption of the  NP at the GC injection port.  The tendency of NP and NPEs to  adsorb onto inner surfaces of instrumentation was examined by Sithole et al.  (1990) and Dr. Coreen Hamilton of The Axys  Group, Sidney, B.C. (personal communication, July 1996).  56  Adsorption of NP at the injection port could affect measurement in two ways.  NP adsorbed at the GC injection port  would be lost from analysis and the resulting measurement based on peak area would underestimate the amount of analyte present in the sample. A portion of the adsorbed NP could then slough away from the port during analysis of a subsequent sample, and the measurement for that sample would be higher than the actual amount present. Derivatization of the NP prior to GC analysis should eliminate the adsorption problem and stabilize the chromatography.  As well, derivatization would provide a  preliminary cleanup of the sample by excluding some of the background contaminants not participating in the derivatization reaction.  A well-established acetylation  procedure was selected as a possible derivatization method for 4-NP present in the hexane-based SPMD dialysates.  3.2  Application of the Voss Procedure for Chlorinated Phenolics  3.2.1  Selection of the Voss Procedure An acetylation procedure developed by Voss et al.  (1981)  was selected as a potential derivatization method for 4-NP present in SPMD dialysates. The Voss procedure is well established and used routinely for the determination of chlorinated phenolics in pulp mill effluents and receiving  57  waters.  The procedure involves addition of a strong potassium  carbonate buffer to the effluent or receiving water sample, followed by acetylation with acetic anhydride.  The acetylated  phenol is extracted using hexane and analyzed by gas chromatography. Success of the procedure relies on conversion of the phenolic compound to its potassium salt, accomplished by addition of the strong potassium carbonate buffer.  The  primary concern in adapting the Voss procedure for use with SPMD dialysates is the immiscibility of the carbonate solution in the hexane-based dialysate.  This effectively isolates the  target NP present in the hexane from the water-based potassium carbonate buffer solution.  Some conversion of the NP to its  potassium salt would be expected to occur at the interface between the hexane and water fractions, as NP in the hexane comes into physical contact with potassium carbonate in the water.  The more water-soluble NP salt would move into the  aqueous layer, vacating a position at the interface for another NP molecule and setting up a dynamic equilibrium between the aqueous and organic fractions. Agitation of the mixture by shaking would increase the surface area for reaction and the frequency of contact, accelerating the process.  These considerations led to the decision to amend  the original Voss procedure to include two or three initial reactions with potassium carbonate solution, aimed at  58  transferring the NP from the hexane-based dialysate to the aqueous layer.  Once there, acetylation would proceed in the  manner described by Voss. 3.2.2  The Voss-Adapted Acetylation Procedure Using an SPMD Extract  3.2.2.1  Summary of Method  To perform the acetylation, a 200 mL sample of SPMD dialysate was placed in a separatory funnel and 4 mL of 5M potassium carbonate solution added.  The mixture was shaken  vigorously for 30 seconds, then allowed to stand for 2 minutes.  The aqueous layer was collected and the dialysate  reacted twice more with 4 mL aliquots of 5M potassium carbonate.  The organic layer was given a final wash using 25  mL of cold tap water, and the resulting aqueous fraction was added to those previously collected. The total aqueous fraction was transferred to a clean separatory funnel and 3 mL of acetic anhydride added.  The  mixture was shaken vigorously, venting often, then allowed to stand for 2 minutes. A second 3 mL aliquot of acetic anhydride was added and the procedure repeated.  The resulting  mixture was then extracted three times with a total of 30 mL of hexane.  The collected organic fraction was rotoevaporated  to 2 mL in a 40 - 45°C water bath, and analyzed using GC-FID. The procedure was repeated with a second SPMD extract. This time, the NP was back-extracted from hexane into the  59  aqueous potassium carbonate solution but no acetylation was performed.  The collected aqueous fraction was acidified to  pH 2 using a 5M solution of sulphuric acid, then extracted three times with a total of 50 mL of hexane. rotoevaporated to 2 mL and analyzed.  The hexane was  Results from this second  sample would provide a measure of the degree to which NP in the hexane-based dialysate was successfully transferring to the aqueous potassium carbonate. 3.2.2.2  Results Using an SPMD Extract  The results for the Voss acetylation testing are summarized in Table 7.  Table 7.  Summary of results for acetylation of SPMD extracts using the Voss procedure  Sample I.D.  Total NP Peak Area  10 mg/L NP standard  59,033  Control  0  acetylated SPMD extract  253?  back-extracted SPMD extract  339?  60  The retention time range for NP in the 10 mg/L NP standard was 19.6 to 22.1 minutes.  For both the acetylated  and back-extracted samples, one peak occurred in this range at just over 19.6 minutes.  The peak was small in both  chromatograms, and may indicate the presence of a small amount of NP or a background contaminant. 3.2.2.3  Discussion of Results  The results for the Voss testing with SPMD extracts were inconclusive.  No distinctive pattern of peaks characteristic  of NP was evident in the chromatograms for the acetylated or back-extracted dialysates.  The failure to detect measurable  quantities of NP may indicate that the necessary transfer of NP from the hexane-based dialysate to the water-based carbonate solution was not successful. Alternatively, the transfer and acetylation may have been wholly or partially successful, but NP was not identifiable due to the large amount of background contamination.  Losses during sample  manipulation may also have reduced the amount of NP present to a level below the minimum detection limit for the instrument. The GC operating program may have contributed to the overall inability to obtain conclusive results. The program used included the initiation of a temperature ramping procedure at approximately 23 minutes, a feature designed to clean the column in preparation for the next sample.  If the  peaks for the acetylated NP eluted at a time greater than 23  61  minutes, they would have been compressed by the ramping program and not readily discernible. The next step was to repeat the Voss acetylation using a clean standard solution, where the NP concentration was high enough to be readily measurable and interference from background peaks was minimized. 3.2.3  Applying the Voss Procedure to Laboratory-Prepared Samples  3.2.3.1  Summary of Method  A test solution of 10 mg/L NP was prepared by adding 0.5 mL of 1 g/L NP stock to 50 mL of hexane. The sample was transferred to a separatory funnel and 25 mL of 0.1M potassium carbonate solution was added.  The mixture was hand-shaken  vigorously for 30 seconds and allowed to separate for 2 minutes, then the procedure was repeated with a second 25 mL aliquot of 0.1M potassium carbonate.  The molarity of the  potassium carbonate buffer was altered from the 5M solution used with the SPMD dialysates to a 0.1M solution, while the corresponding volume of carbonate added was adjusted from 4 mL to 50 mL.  This was done to more closely align the  derivatization method to that described by Voss et al.  (1981).  After settling, the aqueous and organic layers were collected separately.  The aqueous layer was returned to the  separatory funnel and 1 mL of acetic anhydride was added.  The  mixture was hand-shaken for 30 seconds, venting often, then  62  allowed to separate for 2 minutes.  The resulting acetylated  solution was extracted with 5 mL of hexane and analyzed. The hexane remaining after reaction of the original NP solution with potassium carbonate was also collected for analysis.  Any NP present in this fraction represented that  which had failed to make the critical transfer into carbonate for subsequent acetylation. A second 50 mL sample consisting only of 0.1M potassium carbonate was placed in a separatory funnel and processed through all of the steps of the acetylation procedure. The resulting final hexane extract was analyzed to provide a measure of background peaks originating with the potassium carbonate and acetic anhydride solutions. 3.2.3.2  Results Using Laboratory-Prepared Samples  The results for the Voss testing using laboratoryprepared samples are presented in Table 8.  Based on the  standard solution, the retention time range for underivatized NP was 18.5 to 20.9 minutes. NP was absent from the background sample, where no contaminant peaks appeared in the NP retention time range.  Substantial NP was detected in the  hexane sample collected after the potassium carbonate wash, indicating that transfer of the NP from hexane to the aqueous potassium carbonate solution was incomplete.  63  Table 8.  Summary of results for the Voss acetylation method using a laboratory sample  Sample I.D.  Total NP Peak Area  2 mg/L NP standard  70,338  background  0  residual hexane after K2C03  32,501  wash hexane extract after  5092  acetylation in K2C03  A small cluster of peaks with a retention time range of 21.8 to 24.5 minutes appeared in the chromatogram for the hexane extract after acetylation (Figure 5).  No such cluster  appeared in the underivatized standard (Figure 6), and so these peaks were considered to indicate the presence of acetylated NP.  The longer retention times were consistent  with the supposition that the larger acetate derivative would require more time to travel through the GC column.  64  a*pp°  9 ..19? 1.1 ..B48  Figure 5  Gas chroitiatogram of hexane after acetylation analyzed 30 April 1996  65  r~  15.489  Ts4fijg?3,  Figure 6.  Gas chromatogram of 2 mg/L NP standard analyzed 30 April 1996  66  3.2.3.3  Discussion of Results  Derivatization of NP in hexane using the Voss method was only partially successful. Although some portion of the NP transferred to the potassium carbonate solution and was acetylated, the larger fraction, as indicated by total peak area on the chromatograms, remained in the hexane. Use of a clean laboratory-prepared sample allowed clarification of issues raised during the testing with SPMD dialysates.  It was evident that the transfer of NP from  hexane to carbonate was critical to successful derivatization. NP that did transfer appeared to acetylate completely, as evidenced by an absence of peaks in the underivatized region on the chromatogram for the hexane extract after acetylation. The use of a higher test concentration for the laboratory sample yielded larger, more distinctive NP peaks and allowed easier interpretation of the chromatograms.  This was  complemented by a marked reduction in the number of background contaminant peaks from that seen with the SPMD traces. Finally, the GC operating program was adjusted to delay onset of the temperature ramping, and the results from the acetylated sample confirmed that the derivatized NP eluted well before the onset of ramping.  However, a method which  promoted transfer of the NP from hexane to carbonate was still reguired.  67  3.3  Acetylations Using the Lee and Peart Procedure  3.3.1  Selection of the Procedure An acetylation method published in 1995 by Drs. Lee and  Peart of the National Water Research Institute in Burlington, Ontario, provided a procedure specific for 4-NP present in STP effluents and sludges. Although the Lee and Peart procedure contained similarities to that of Voss, reaction times were increased substantially from the single shaking in a separatory funnel prescribed by Voss, to a total mixing time of 1 hour.  As well, three separate acetylations were  performed, compared with the single acetylation of the Voss procedure.  These changes afforded greater opportunity for  more complete derivatization of the NP. However, as in the Voss method, the sample in Lee and Peart's procedure was water-based.  Due to concerns about the  effect of a hexane-based dialysate on derivatization efficiency of the NP, the procedure was assessed using two separate samples - one water-based and the second in hexane. 3.3.2  Summary of Method To perform the acetylations, a 250 mL sample of water-  based NP solution was prepared by adding 0.2 mL of 1 g/L NP in methanol to Alpha-QR water.  The sample was transferred to a  400 mL beaker and 1 g of potassium carbonate was added.  The  sample was mixed on a magnetic stir plate for 30 seconds to ensure complete dissolution of the potassium carbonate, then  68  1.0 mL of acetic anhydride and 30 mL of hexane were added. The sample was covered and left to mix vigorously on the stir plate for 30 minutes. After mixing, the beaker contents were transferred to a 1 L separatory funnel and left to settle 10 minutes.  The aqueous layer was collected and returned to the  400 mL beaker. Acetylation was performed twice more using 100 /iL volumes of acetic anhydride, 30 mL volumes of hexane and 15 min mixing times. The organic fraction was composited and washed with 20 mL of 1% potassium carbonate solution to remove co-extracted acetic acid, then transferred to a roundbottomed flask and rotoevaporated to 2 mL at 40 - 45°C. A subsample of the concentrated extract was collected for analysis. A second sample, based in hexane, was prepared by adding 0.2 mL of 1 g/L NP dissolved in hexane to 250 mL of hexane. Potassium carbonate was added by first dissolving 1 g of the dry chemical in 5 mL of Alpha-QR water.  The NP and potassium  carbonate solutions were added together to a 400 mL beaker and mixed.  After 30 seconds, 1.2 mL of acetic anhydride was  added, the beaker was covered and the contents were mixed vigorously for 2 hours. The beaker contents were then transferred to a 1 L separatory funnel, washed with 20 mL of 1% potassium carbonate solution and allowed to settle 10 minutes.  The organic fraction was collected, rotoevaporated  to 2 mL, and analyzed.  69  3.3.3  Results of the Lee and Peart Derivatizations Chromatograms for the Lee and Peart derivatizations are  provided in Figures 8 and 9.  A chromatogram for an  underivatized 100 mg/L NP standard is provided in Figure 7. These chromatograms were the first done on the new gas chromatograph Model 6890, and were used for gualitative analysis only.  NP peaks in the underivatized region were  integrated over a retention time range of 16.6 to 18.6 minutes, and appear in the data section of the chromatogram printout as a single peak area at 17.6 minutes. A visual review of the underivatized NP standard showed distinctive peaks in the 16.6 to 18.6 minute retention time range (Figure 7).  A similar but smaller cluster appeared on  the chromatogram for the acetylated hexane-based sample (Figure 8).  On the chromatogram for the water-based sample  (Figure 9), peaks in the retention time range for underivatized NP were much reduced in height and area, and a second cluster occurred from 19.6 to 22.0 minutes. The results suggested that partial conversion to the acetate had occurred with the water-based sample and essentially no derivatization had occurred in the hexane-based sample. The procedure was repeated with a hexane-based sample, this time collecting subsamples at key stages of the process. The goal was to pinpoint steps in the procedure that were preventing successful acetylation of the NP.  70  20505:D)"  jo  10  Figure 7,  20  15  Gas chromatogram of 100 mg/L NP standard  O) ID  (D CD CO  O (D  5£F 10  Figure 8.  15  1, iqflffifYfy 20  Gas chromatogram for Lee and Peart acetylation with a hexane-based sample  71  CO  m CO CO  <5 £  co oco  CD  -^J- CM  ^ ••" „_  04  0  Figure 9.  '  '  1^  °>dco °- rocMcsr r; C M ^  i sT %*M\g.  co  1  _^ oo S o  O  '  '  1*5  n  '  '  '  <H  '  cstt> ostein rEsj  2<D  '  '~  Gas chromatogram for Lee and Peart acetylation with a water-based sample  3.3.4  Tracking 4-NP Through the Derivatization Process  3.3.4.1  Description of Method  The fate of 4-NP through the acetylation procedure was monitored by analysing hexane subsamples collected at various stages of the derivatization.  A sample of NP in hexane was  prepared by adding 0.3 mL of 1 g/L NP in hexane to 50 mL of hexane.  A 30 mL aliquot of this sample, labelled "Hexane NP",  was set aside for use in the derivatization.  The balance was  analyzed directly on the GC. A 250 mL volume of Alpha-QR water, representing the "effluent" sample in the Lee and Peart procedure, was measured into a 400 mL beaker.  The beaker was placed on a magnetic  72  stir plate and 1.0 g of potassium carbonate was added.  After  stirring 30 seconds to dissolve the carbonate, 1.0 mL of acetic anhydride and the 30 mL NP sample "Hexane NP" were added to the beaker.  The beaker contents were mixed for 30  minutes on the magnetic stirrer, then transferred to a 1 L separatory funnel. The aqueous and organic fractions were collected separately, and the aqueous fraction was returned to the 400 mL beaker.  The organic fraction was set aside in a  separate beaker labelled "Hexane 1". A further 100 juL of acetic anhydride and 30 mL of hexane were added to the aqueous sample in the 400 mL beaker and the mixture was stirred for 15 minutes before being transferred back to the separatory funnel. The collected organic fraction was set aside as "Hexane 2" and the aqueous fraction was returned to the beaker for a third acetylation with 100 juL of acetic anhydride and 30 mL of hexane. After mixing 15 minutes and separating again in the separatory funnel, the aqueous fraction was discarded and the organic fraction was collected as "Hexane 3". With the final acetylation completed, each hexane fraction was washed with 10 mL of 1% potassium carbonate, then rotoevaporated to 2 mL and analyzed.  To provide a measure of  contamination levels, a background sample was prepared as described in Section 3.2.3.1.  Figure 10 provides a schematic  representation of the collection of hexane aliquots.  73  NP in hexane = "Hexane NP"  + water + K2C03 + acetic anhydride extraction 1  aqueous fraction  and  "Hexane 1"  + acetic anhydride + hexane extraction 2  aqueous fraction  and  "Hexane 2"  + acetic anhydride + hexane extraction 3  aqueous fraction  and  "Hexane 3"  (discarded)  Figure 10.  Collection of hexane aliquots at key stages of the Lee and Peart acetylation procedure  74  3.3.4.2  Results  The results for the NP tracking experiment are summarized in Table 9.  Table 9.  Summary of results for the NP tracking experiment  Sample I.D.  Total NP Peak Area  100 mg/L NP standard  1930  background  3  Hexane NP  1029  Hexane 1  1446  Hexane 2  88  Hexane 3  12  The largest amount of NP occurred in Hexane 1, the fraction collected after the first acetylation.  Only small  amounts of NP, as indicated by total peak area, were present in Hexanes 2 and 3.  Significantly, only underivatized NP was  present.  75  3.3.4.3  Discussion of Results  The results of the NP tracking experiment clearly indicated that NP was not making the necessary transfer from hexane to carbonate, and so was remaining unacetylated.  It  was evident that the hexane base of the SPMD dialysate was presenting a significant barrier to successful acetylation. Transfer of the NP to a water-miscible solvent was needed if the analyte was going to contact the acetic anhydride in the carbonate solution and undergo derivatization.  3.4 3.4.1  Solvent Exchange Introduction Solvent exchange was proposed as a means of eliminating  the hexane blockage of the derivatization process.  It was  theorized that the NP in hexane was effectively isolated from acetic anhydride present in the water-based carbonate solution, due to the immiscibility of the two solvents. Transferring the NP into a water-miscible solvent would provide opportunity for contact and acetylation with acetic anhydride.  Successful solvent exchange required a solvent  intermediary that was miscible in both hexane and water. Based on its nearly universal miscibility, tetrahydrofuran (THF) was selected for trial. 3.4.2  Summary of Method A standard solution was prepared by adding 160 /iL of  76  1 g/L NP to 200 mL of hexane.  The solution was transferred to  a round-bottomed flask and rotoevaporated in a 40 - 45°C water bath to approximately 5 mL.  A 30 mL volume of tetrahydrofuran  (THF) was added to the round-bottomed flask and the contents again rotoevaporated to 5 mL.  The solvent exchange was  performed twice more using 30 mL aliguots of THF. After the final rotoevaporation, the 5 mL NP solution was transferred to a volumetric flask and the volume made up to 250 mL using Alpha-QR water.  Derivatization then proceeded as described in  Section 3.3.2.  Two background samples were prepared, in order  to provide a separate measure of contaminant levels from the acetylation step and from THF. 3.4.3  Results Using Solvent Exchange Table 10 summarizes the results obtained.  Based on the  100 mg/L NP standard, the retention time range for underivatized NP extended from 19.2 to 21.4 minutes.  Peaks in  this range were present on the chromatogram for the derivatized sample, indicating that some of the NP present in the sample did not undergo acetylation.  However, a small  cluster of peaks extending from 23.6 to 28.5 minutes confirmed that derivatization was at least partially successful. A significant amount of background contamination appeared on the chromatogram for the derivatized NP sample, and this contamination was traced to the THF.  The majority of the THF  contaminant peaks, however, eluted at retention times of less  77  than 15 minutes, so that little overlap occurred with those of NP.  Table 10.  Results of NP acetylation after solvent exchange with THF  Total NP  Sample I.D.  Peak Area  100 mg/L NP standard  1878  procedural background  0  THF background  0  acetylated NP sample  3.4.4  697  Discussion of Results  ' The acetylation procedure continued to be only partially successful, despite the transfer of NP to water-miscible THF. The reasons for incomplete derivatization were unclear.  Two  variables seemed inherent in the procedure - the length of the reaction time between NP and acetic anhydride, and the time of addition of hexane.  Lee and Peart added their organic solvent  78  at the same time as the acetic anhydride, effectively combining the acetylation and extraction steps. The experiments to date, though, had established the strong affinity of NP for hexane.  Perhaps the presence of hexane  during acetylation extracted a portion of the NP from the aqueous carbonate solution before it was derivatized.  An  experiment was devised to test the effect of the two variables - reaction time and extraction time with hexane. The goal was to optimize the acetylation conditions for NP. 3.4.5  Examining the Influence of Reaction Time and Extraction Time  3.4.5.1  Summary of Method  Shaking time and hexane addition were varied during the first acetylation in an experiment designed to examine the effects of reaction and extraction times on successful NP acetylation.  Four samples were prepared following procedures  described in Section 3.4.2.  After solvent exchange with THF,  each sample was reconstituted to 250 mL using Alpha-QR water, and 2.0 g of potassium carbonate was added.  The samples were  hand-shaken to dissolve the potassium carbonate, then 1.0 mL of acetic anhydride was added to each. The samples were separated according to the length of the shaking period and time of hexane addition.  Two of the four  samples received a 30 mL aliquot of hexane, and all samples were placed on an automatic shaker.  79  Two samples, one with  hexane and one without, were shaken for 30 minutes. Ten minutes prior to the end of the shaking period, the sample without hexane received a 30 mL aliquot of hexane and the shaking continued. The other two samples were shaken for 6 hours. Thirty minutes prior to the end of the shaking period, the sample without hexane received a 30 mL aliquot of hexane.  Figure 11  provides a schematic representation of experimental conditions for the first acetylation.  Sample 1  Sample 2  Sample 3  Sample 4  + K2C03 + acetic anhydride /  \  30 min shaking  6 hour shaking  /  /  hexane in at start  Figure 11.  \  hexane added at 20 min  hexane in at start  \  hexane added at 5 h 30 min  Variations in reaction time and time of hexane addition for the Reaction Time and Extraction study  The effect of reaction time between NP and acetic anhydride on the proportion of successfully derivatized NP was  80  measured by altering the duration of the shaking period.  The  addition of hexane at the beginning or the end of the shaking period was intended to assess the effect of the presence of hexane during the acetylation. At the end of the first acetylation, all four samples were transferred to separatory funnels and the aqueous layers collected.  Second and third acetylations proceeded as  previously described in Section 3.3.2.  A background sample  was also prepared, following the procedure outlined in Section •J m J  • Ci •  3.4.5.2  Results of the Reaction Time and Extraction Time Study  The results are summarized in Table 11. Based on a concurrently run NP standard, the R.T. range for underivatized NP was 19.2 to 21.4 minutes. A cluster of peaks with R.T. range of 22.2 to 24.8 minutes was evident on the chromatograms for Samples 1 to 4, and was considered to indicate the presence of acetylated NP. Using these two R.T. ranges, the proportions of acetylated and nonacetylated NP were estimated for each sample by calculating their peak area values and comparing these with the total peak area for NP. The highest estimated percent conversion (55%) occurred with the sample receiving hexane at the start of a 6 hour shaking period.  Derivatization efficiencies for the other  three samples were remarkably similar, and ranged from 21 to  81  30%. Total peak area values for the samples ranged from 813 to 1206.  The wide range in values for total peak area was  unexpected, as all samples were prepared the same way. Further, both of the extended shaking time samples produced lower total peak area values than their 30 minute counterparts.  This suggested that procedural losses may have  been incurred during the longer shaking period, although the nature of these is unclear.  Despite the discrepancy in total  peak area, the proportion of derivatized NP as indicated by the estimated percent derivatization remained comparable between Sample 4 (6 h shaking, hexane later) and the two 30 minute samples.  For this reason, the results were considered  to indicate a trend toward higher conversion efficiencies when a longer shaking time was combined with hexane added at the start of the acetylation.  82  Table 11.  Results of the reaction time and extraction time study  Sample I.D.  Peak Area in R.T. Range  Total NP  Estimated  Peak Area  Percent Acetylated  19.2-21.4  22.2-24.8  minutes  minutes  Sample 1  839  239  1078  22  Sample 2  840  366  1206  30  Sample 3  368  445  813  55  Sample 4  734  199  933  21  Sample 1 = 30 min shaking period; hexane added at start Sample 2 = 30 min shaking period; hexane added at 20 min Sample 3 = 6 h shaking period; hexane added at start Sample 4 = 6 h shaking period; hexane added at 5 h 30 min  83  3.4.5.3  Discussion of Results  The results raised several questions regarding the relative importance of reaction vs extraction time. Clearly, the optimal condition was a long reaction time with hexane present throughout the acetylation.  This testing regime  yielded an estimated 55% conversion of NP to the acetate, an efficiency at least 25% higher than for the other three samples.  Trends were less clear for the balance of the  testing regimes. The second highest rate of conversion was obtained with a shorter reaction time and hexane added towards the end of the acetylation period.  Addition of hexane at the  start of the 30 minute reaction period reduced derivatization by 8%.  The lowest conversion rate, 21%, was obtained with 6  hours of reaction time and a later addition of hexane. The impact of the observed decrease in total peak area values on the estimated percent derivatization of NP was undetermined. The most pressing issue was that of incomplete derivatization.  The highest estimated conversion was still  only 55%, and this required 6 hours of reaction time. The reasons for this partial derivatization were unclear. A method was required that allowed a more concentrated focus on each step of the derivatization procedure.  84  3.5  Application of a Model Compound - Octylphenol  3.5.1  Introduction The complex chromatography of NP presented difficulties  in determining precisely the extent to which derivatization had occurred.  At least twelve peaks appeared on the  chromatogram for 4-NP.  Derivatization shifted these peaks on  the chromatogram to longer retention time values, so that no overlap occurred with the underivatized compounds. However, in the case of partial acetylation, employing a chemical with less complex chromatography would permit more readily discernible results. Once complete derivatization was accomplished with the surrogate compound, the method could be applied to the more complex NP. The primary requirement for a suitable surrogate was comparability in chemical behaviour to that of the target analyte.  On the basis of its structural and physical  characteristics, 4-tert-octylphenol was selected to model the NP response in the acetylation process. Octylphenol (OP) is similar structurally to NP but carries one less carbon on its alkyl chain.  The chemicals are markedly different, though, in  their degree of complexity.  Where the formulation for 4-NP  contains a complex mixture of isomers, OP is a single compound yielding one chromatographic peak.  The less complicated  chromatogram would allow easier monitoring of progress towards complete acetylation.  85  3.5.2  The Lee and Peart Procedure Using Octylphenol  3.5.2.1  Summary of Method  A 200 mL sample of OP in hexane was prepared by adding 80 /iL of 1 g/L OP stock to hexane.  The sample was  rotoevaporated and solvent exchanged into THF using the procedure outlined in Section 3.4.2.  The resulting sample was  reconstituted with Alpha-QR water and acetylated as described in Section 3.3.2. A shaking time of 4 hours was used for the first acetylation.  All other aspects of the procedure were  performed as previously.  A 50 mg/L OP standard, prepared by  adding 2.5 mL of 1 g/L OP stock into a final volume of 50 mL of hexane, was run concurrently in order to establish the R.T. for underivatized OP.  80 juL volumes of two internal  standards, 4-sec-butylphenol and 1-phenylnonane (nonylbenzene), were added to the OP standard. 3.5.2.2  Results of the Lee and Peart Derivatizations  Chromatograms for the Lee and Peart testing using OP are provided in Figures 12 and 13. The retention time for underivatized OP was 11.7 minutes (Figure 12).  The  proliferation of background peaks on the chromatogram for the derivatized OP sample were most probably from THF, and made interpretation of the chromatogram results difficult (Figure 13).  A peak occurred at 11.7 minutes, indicating the presence  of underivatized OP. Another large peak occurred at 13.8 minutes.  Although no procedural or THF background samples  86  <L HL)1 A, (LOKI\H2000007.D) m  pA  I  c i c i c i  i--  c  ui  *~  i  »  i  90~~J  0  80-  70-  ' • •  I  (J-  {  [  —I  N  i  t  c  i  n  |  60-  >-  1 i  50-  I  J  0 o w  &  _  I  m  CM  CM CO •*  i  CM  d  30-  20-  u  \f  40-  C f. a  CO CM  ?  CO  i  SB CD  ... .11 _  o  i  <b  •  C  '  '  Figure 12  '  •  5  '•  '  ib  •  '  •  •  1'5  2'0  •  '  •  Gas chromatogram of 50 mg/L OP standard  FID1 A, (LORI\H2000009.DT pA  90 80 70 60 50 40 30 20  10  Figure 13.  15  20  Gas chromatogram of acetylated OP sample  87  '  i  were run with this testing, previously run backgrounds failed to disclose a peak in the vicinity of 13.8 minutes. This suggested this new peak was that of acetylated OP. 3.5.2.3  Discussion of Results  The test results confirmed that only partial derivatization of the target compound was occurring.  Although  the reasons for this were unclear, the difference between the hexane solution of the SPMD dialysate and the water-based effluents for which the Lee and Peart procedure was designed remained the most likely reason for the lack of success. Solvent exchanging into THF did not circumvent this problem, and the very messy chromatography of THF complicated the analysis.  Since successful derivatization had not been  achieved, there were no derivatized standards. This made analysis of acetylated samples difficult, as peaks in the proposed R.T. region indicative of acetylation could not be compared with those in a standard solution.  These  considerations led to development of a new method, based on the principles of the Lee and Peart procedure but with a much reduced volume of solvent.  3.6  The Lee-Adapted Acetylation Procedure  3.6.1  Summary of Method A 50 mL solution of OP in hexane was prepared by adding  600 fiL of 1 g/L OP stock.  The solution was rotoevaporated and  88  solvent exchanged as described in Section 3.4.2, using AlphaQR water in place of THF.  Following solvent exchange, the OP  solution was transferred to a graduated cylinder and the volume made up to 20 mL using Alpha-QR water. Acetylation was performed using a 40 mL clear glass screw-top vial.  The 20 mL OP sample was transferred to the  glass vial and 0.2 g of K2C03 was added, followed by 1.0 mL of acetic anhydride.  The vial was hand-shaken gently for about 1  minute, venting often, then placed on an automatic shaker for 15 minutes. After shaking, 10 mL of hexane was added and the vial was returned to the shaker for a further 10 minutes. The vial contents were then allowed to settle, and the organic layer was collected for analysis. In order to compare the results obtained with hexane to those of OP in a water-miscible solvent, a second sample was prepared using OP in methanol. Ten mL of 0.1M K2C03 was added to a 40 mL clear glass vial, followed by 100 /LtL of 1 g/L OP in methanol.  The contents were mixed by gentle hand-shaking,  then 0.5 mL of acetic anhydride was added.  The vial was hand-  shaken vigorously for about one minute, venting often, then placed on an automatic shaker for 15 minutes. At the end of the shaking period, 5 mL of hexane was added.  The sample was  returned to the shaker for a further 5 minutes, then removed and allowed to separate. After separation, the organic layer was collected for analysis.  89  3.6.2  Results The results for the Lee-adapted acetylations are  summarized in Table 12. Significant conversion to the acetate occurred with both samples, as evidenced by large peak area values at the 13.8 minute retention time for the acetylated compound.  For both samples, estimated percent acetylated  values were greater than 90%. A small amount of underivatized analyte was also present in both samples at a retention time of 11.7 minutes.  Table 12.  Results of OP acetylations using the Leeadapted procedure  Sample I.D.  Peak Area  Total OP  Estimated  in R.T. Range  Peak Area  Percent Acetylated  OP in  11.7  13.8  minutes  minutes  39  602  641  94  18  365  384  95  water OP in methanol  90  3.6.3  Discussion of Results The results from the Lee-adapted acetylation procedure  indicated that reducing the volume of solvent in the sample markedly improved the efficiency of derivatization.  With both  test samples, a large proportion of the OP had acetylated. Although the results with the water solvent exchange were good, the greater solubility of OP in methanol could improve the efficiency of transfer from hexane. The derivatization procedure was repeated, this time employing a solvent exchange with methanol.  3.7  Methanol Solvent Exchange  3.7.1  Summary of Method Two samples of OP in hexane were prepared by adding 80 /iL  volumes of 1 g/L OP to 200 mL of hexane.  Each sample was  rotoevaporated in a 40 - 42°C water bath to a volume of less than 2 mL, then solvent exchanged three times using 20 mL volumes of methanol. After the final solvent exchange, the samples were rotoevaporated to less than 2 mL and acetylated using the Lee-adapted procedure described in Section 3.6.  For  one sample, the acetylation was performed in the roundbottomed flask used for rotoevaporation and sample mixing was accomplished using a magnetic stir plate.  The second sample  was transferred to a 40 mL clear glass vial after solvent exchange into methanol, and acetylation was performed in the  91  vial. The solvent exchange and derivatization procedure was repeated a third time to confirm the results obtained.  For  the third sample, acetylation was performed in the roundbottomed flask as described above. 3.7.2  Results Table 13 summarizes the results obtained.  Table 13.  Results of the Lee-adapted acetylations using methanol solvent exchange  Sample  Peak Area  I.D.  in R.T. Range  Total OP  Estimated  Peak Area  Percent Acetylated  OP in  11.8  14.0  minutes  minutes  5  264  269  98  16  300  316  95  5  247  252  98  vial OP in flask 1 OP in flask 2  92  Both the flask and vial methods yielded high conversion efficiencies, with estimated percent derivatizations ranging from 95 to 98%. It is important to note that the results were strictly qualitative and no attempt was made at this point to incorporate a quantitative analysis of procedural losses. Still, the results suggested that solvent exchanging the target analyte out of hexane and into methanol prior to derivatization increased acetylation efficiency dramatically. 3.7.3  Discussion of Results Based on the results obtained with OP, the methanol  solvent exchange procedure and Lee-adapted derivatization method were applied next to the true target analyte, NP.  3.8  Applying Methanol Solvent Exchange to Nonylphenol  3.8.1  Summary of Method Two samples of NP in hexane were prepared by adding  160 liL of 1 g/L NP to 200 mL of hexane. The samples were rotoevaporated and solvent exchanged as described in Section 3.7.1, then acetylated in situ in the round-bottomed flask using the Lee-adapted derivatization procedure.  At the end of  derivatization, the contents of the round-bottomed flasks were transferred to 40 mL clear glass vials to allow easier removal of the hexane layer.  The flasks were each rinsed with 1.0 mL  of hexane, which was then added to the glass vial.  A third  200 mL sample was prepared using 320 juL of 1 g/L NP stock,  93  and solvent exchange and derivatization were performed as above. 3.8.2  Results The results are summarized in Table 14. Based on a  100 mg/L NP standard, the R.T. range for underivatized NP was 12.9 to 14.5 minutes. A cluster of peaks with R.T. range 15.1 to 16.8 minutes was present only on the chromatograms for the derivatized samples, and was considered to indicate the presence of acetylated NP. An estimated value for the percent acetylation was determined using the procedure described in Section 3.4.5.2.  Table 14.  Results of the NP acetylations using methanol solvent exchange and the Lee-adapted procedure  Peak Area  Sample I.D.  in R.T. Range  Total NP  Estimated  Peak Area  Percent Acetylated  12.9 - 14.5  15.1 - 16.8  minutes  minutes  NP 1  112  753  865  87  NP 2  45  619  664  93  NP 3  126  989  1115  89  94  The estimated percent conversion values ranged from 87 to 93%.  These values were slightly lower than the 95 to 98%  estimated acetylation obtained using OP. Again, no measure of procedural losses was undertaken. 3.8.3  Discussion Results obtained with NP correlated well with those of  OP.  For a hexane-based sample, performing a solvent exchange  into methanol prior to acetylation, and reducing the volume of solvent present during the acetylation, dramatically increased derivatization efficiency.  3.9  Acetylation of SPMD Dialysates using Methanol Solvent Exchange and the Lee-Adapted Procedure  3.9.1  Summary of Method SPMD dialysates were obtained using procedures described  in Section 2.2. A three-week exposure period and 200 /xg/L NP test water concentrations were applied to the SPMDs. Dialysates were rotoevaporated to 2 mL in a 40 - 45°C water bath, then solvent exchanged and acetylated following the procedure outlined in Section 3.8.1.  Acetylation was  performed in the round-bottomed flask used for rotoevaporation. 3.9.2  Results Chromatograms for the acetylated Control and a test SPMD  dialysate are provided in Figures 16 and 17. Chromatograms  95  for non-acetylated and acetylated standards are given in Figures 14 and 15. Table 15 summarizes the results obtained.  Table 15.  Results of acetylations on SPMD dialysates using methanol solvent exchange and the Leeadapted procedure  Sample I,D.  Peak Area in R.T. Range  Total NP  Estimated  Peak Area  Percent Acetylated  12.8 - 14.5  15.1 - 16.8  minutes  minutes  Control  2  -  2  -  Test A  273  164  438  37  Test B  204  121  325  37  The R.T. range for underivatized NP was 12.8 to 14.5 minutes (Figure 14), and derivatized NP was 15.1 to 16.8 minutes (Figure 15).  One peak at 12.8 minutes occurred in the  Control dialysate (Figure 16), and may indicate contamination  96  HL>1 A, (LOKIVNPIUUUUy.U) pA 80-  70-  60-  50  40  30  20  10  10  Figure 14.  15  20  Gas chromatogram of non-acetylated 100 mg/L NP standard  ~TT01 A, (LORI\NP100008T3T pA 35-  30  25  20  15-  10  ,_:  CM  CM  ,1  ^j^Jt*A+J\fo*. 10  Figure 15,  15  15"  Gas chromatogram of acetylated 40 mg/L NP standard  97  Figure 16.  Gas chromatogram of acetylated SPMD Control dialysate  HL)1 A, (LOKKNMOaUOS.Dr PA  80-  60-  40 -| I  <o CD  20  _i_ 0-,  10  Figure 17.  15  20  25  Gas chromatogram of acetylated SPMD dialysate  98  by NP or a background chemical with similar R.T.  Both  acetylated and non-acetylated NP were present in the test dialysate (Figure 17).  Interestingly, the estimated percent  conversion was the same for both test replicates.  The results  indicated that only about 37% of the target analyte underwent derivatization. 3.9.3  Discussion of Results Despite promising results using laboratory-prepared  standards, the derivatization of NP in SPMD dialysates remained incomplete. A derivatization method designed for use with organic solutions was next applied.  3.10  The Wilson Acetylation Procedure  3.10.1  Introduction  A derivatization method described by Wilson (1996) paralleled many aspects of the Lee-adapted procedure, but possessed several key advantages. Like the Lee-adapted procedure, the Wilson procedure used small volumes of sample and reactants.  The Wilson procedure, however, was designed to  acetylate analytes in an organic solvent.  This eliminated the  need for solvent exchange, removing the difficulty of transferring NP out of hexane and reducing procedural losses. As well, reaction times with the Wilson procedure were shorter than those used previously, reducing the overall time required to perform the derivatization.  The method was rapid and  99  efficient, incorporating the simultaneous acetylation and extraction seen previously in the Lee and Peart procedure. 3.10.2  Applying the Wilson Procedure to LaboratoryPrepared Standards  3.10.2.1  Summary of Method  A 100 mg/L NP test solution was prepared by adding 0.2 mL of 1 g/L NP stock into hexane to a final volume of 2 mL. The sample was transferred to a 40 mL clear glass vial and 1 mL of acetone was added, so that the proportion of acetone to hexane was 1:2.  Five mL of 0.1M potassium carbonate and 400 jxL of  acetic anhydride were added to the vial, and the contents were hand-shaken, venting often.  Shaking continued until bubble  formation in the solution slowed, an indication that acetylation was complete (Wilson, 1996).  The organic layer  was collected, and the remaining aqueous fraction was extracted once more using 1.5 mL of 1:2 acetone:hexane.  The  combined organic extract was washed with 3 mL of 0.0.5M potassium carbonate, then collected for analysis. After first test results appeared positive, the procedure was repeated using successively greater volumes of acetic anhydride.  This was done to ascertain whether a larger amount  of acetic anhydride than that specified in the original method was required because of the high concentration of NP present. For acetic anhydride volumes of 2.0 and 2.5 mL, the volume of solvent used was also adjusted upward.  100  Each extraction was  then performed using 3.0 mL of 1:2 acetone:hexane. 3.10.2.2  Results  The results of testing with the Wilson procedure are provided in Table 16.  Table 16.  Results of NP acetylations on laboratoryprepared samples using the Wilson procedure  Volume of Acetic  Peak Area in R.T. Range  Total NP  Estimated  Peak Area  Percent  Anhydride  Acetylated  (mL) 12.8-14.5  15.1-16.8  minutes  minutes  0.4  919  684  1603  43  0.8  269  601  870  69  1.0  238  1374  1612  85  1.5  264  1188  1452  82  2.0  17  637  689  92  32  672  704  95  46  581  627  93  2.0 repeat 2.5  101  Estimated percent conversion values for NP increased with increasing volumes of acetic anhydride, up to a maximum value of 95% at 2.0 mL of acetic anhydride. 3.10.2.3  Discussion of Results  The qualitative results obtained using the Wilson procedure and laboratory-prepared standards indicated high conversion efficiencies for the target analyte. The procedure was next applied to NP present in SPMD dialysates. 3.10.3 3.10.3.1  Applying the Wilson Procedure to SPMD Dialysates Summary of Method  SPMD dialysates were obtained from SPMDs exposed for three weeks to a water concentration of 200 nq/L NP. The dialysates were rotoevaporated to 2 mL and acetylated using the procedures outlined in Section 3.10.2.1. 3.10.3.2  Results  Table 17 summarizes the results obtained.  Estimated  percent conversion values ranged from 87 to 98%. As with the laboratory-prepared standards, the highest conversion occurred with 2.0 mL of acetic anhydride.  This volume was incorporated  into the procedure and applied to all subsequent acetylations.  102  Table 17.  Results of acetylations on SPMD dialysates using the Wilson procedure  Volume of Acetic  Peak Area  Total NP  Estimated  in R.T. Range  Peak Area  Percent  Anhydride  Acetylated  (ittL)  12.8-14.5  15.1-16.8  minutes  minutes  2.0  14  242  256  95  2.5  37  252  289  87  6  268  274  98  2.0 repeat  3.10.3.3  Discussion  Results of dialysate acetylations using the Wilson procedure confirmed the method provided reliable and consistent conversion of NP to the acetate. However, assessments of successful acetylation were still qualitative. A replicate test was undertaken next, aimed at providing an initial measure of the amount of variability present in the derivatization results.  103  3.10.4  The Replicate Test  3.10.4.1  Summary of Method  A 100 mg/L NP standard was prepared by adding 5.0 mL of 1 g/L NP stock into hexane to a final volume of 50 mL.  Five  2 mL aliquots (Lab Samples 1 to 5) were distributed into individual 40 mL clear glass vials and acetylated simultaneously using the procedure described in Section 3.10.2.1.  After preliminary results from Lab Samples 1 to 3  indicated that only partial acetylation had occurred, five more 2 mL aliquots were taken from the 50 mL 100 mg/L NP standard, and processed individually using the same acetylation procedure. 3.10.4.2  Results of the Replicate Testing  Table 18 summarizes the results obtained.  Results for  Lab Samples 4, 5, and 10 are not included, as Lab Samples 4 and 5 were not analyzed and Lab Sample 10 was used in development of the silica gel cleanup method. Based on concurrently run standards, the R.T. ranges for underivatized and derivatized NP were 12.8 to 14.5 minutes and 15.1 to 16.8 minutes, respectively.  Estimated percent  acetylated values for the replicate samples 1, 2, and 3 ranged from 46 to 53%, with a mean of 50% (s.d. 3.8). Lab Samples 6 to 9, processed individually, exhibited much higher conversion values ranging from 87 to 97% with a mean of 94% (s.d. 4.6).  104  Table 18.  Results of replicate testing using the Wilson procedure  Sample I.D.  Peak Area in R.T. Range  Total NP  Estimated  Peak Area  Percent Acetylated  12.8-14.5  15.1-16.8  minutes  minutes  1  418  468  886  53  2  533  459  992  46  3  447  476  923  52  6  109  733  842  87  7  34  895  929  95  8  31  912  943  97  9  33  856  889  96  3.10.4.3  Discussion of Results  The observed differences in conversion success between samples processed individually and those done as part of a  105  group were unexpected.  The treatment of the replicate samples  differed only from individually processed samples in the length of time taken to perform the acetylation. Individually-processed samples were carried through the steps in the procedure without time delays, while replicate samples were left for several minutes while awaiting the next step in the procedure.  Results from the replicate testing suggested  that the amount of time taken to perform the acetylation procedure played a critical role in derivatization success. It was theorized that the acetylated compound might be unstable or reactive.  Left in contact with the potassium  carbonate base solution, the reaction, instead of favoring formation of the acetylated product, might result in a stable equilibrium between acetylated and non-acetylated NP.  In this  case, the resulting GC chromatogram would indicate the presence of both compounds. An experiment was devised to test the stability of the acetylated NP product after final washing of the organic fraction. 3.10.5  The Final Wash Study  3.10.5.1  Summary of Method  A 100 mg/L NP standard was prepared as described in Section 3.10.4.1, and five 2 mL aliquots were measured into individual 40 mL clear glass vials.  Each aliquot was  acetylated using the procedure described in Section 3.10.2.1. The aliquots were processed individually, and differed only in  106  their treatment of the final combined organic extract. Aliquot 1 was treated with an aqueous base wash of 3 mL 0.05M potassium carbonate as described in the Wilson procedure to remove co-extracted acetic acid.  Different washing regimes  were applied to the remaining four aliquots in order to assess their effect on the stability of acetylated NP in the organic extract.  Table 19 describes the final aqueous wash used with  each aliquot.  Table 19.  Aliquot  Final washes used for the Final Wash Study  #  N a t u r e of t h e F i n a l  Wash  1  3 mL 0 . 0 5 M K 2 C0 3  2  3 mL 0 . 0 5 M K 2 C0 3 + 2 X 3 mL A l p h a - Q R water  3  3 mL 0 . 0 5 N NaOH  4  3 mL 0 . 0 5 N NaOH + 2 x 3 mL A l p h a - Q R water  5  2 x 3 mL A l p h a - Q R  107  water  A washing period of two minutes was used.  After shaking,  the organic extract was left in contact with the base washing solution and subsamples were collected from the organic layer at timed intervals of 1, 5, 15, 30 and 60 minutes. A sixth subsample was collected the following morning. 3.10.5.2  Results  Table 20 summarizes the results obtained.  The estimated  percent acetylated value was determined for each subsample of the organic fraction by comparing NP peak areas in the derivatized R.T. range with the total NP peak area. For all final washing alternatives, the amount of acetylated NP in the organic fraction dropped sharply after 15 minutes in contact with the agueous base washing solution. Estimated percent acetylated values were reduced by approximately 15 to 20% in the time period between 15 and 30 minutes, and only a small amount of acetylated NP was detected in overnight samples. Final washes with potassium carbonate, sodium hydroxide and straight Alpha-QR water yielded very similar results for estimated percent acetylation.  Those washing methods which  incorporated several washing steps resulted in lower values for percent acetylation than their single wash counterparts.  108  Estimated percent acetylation for five final  Table 20,  washing regimes  Final  Time After Shaking (min.)  Wash  1  5  15  30  60  o/n  K2C03  97  96  95  80  34  2  K2C03 +  63  77  71  53  4  2  NaOH  97  97  95  83  52  17  NaOH +  87  86  78  53  32  2  96  95  90  74  31  insuff.  2 X H20  2 x H20 2 x H20  sample o/n = leflt overnighIt  3.10.5.3  Discussion of Results  Results from the Final Wash Study indicated that time was an important component in determining the success of acetylation.  The longer the organic fraction was left in  contact with the aqueous base solution, the lower was the yield of derivatized NP.  This applied to all base solutions  109  used.  As well, final washing methods that included a base  wash followed by two water washes took longer to perform than the single washing with base solution, and resulted in correspondingly lower values for estimated percent conversion. It is important to note that interpretation of the Final Wash results was strictly qualitative.  Each testing procedure  used only a single replicate, and no attempt was made to incorporate a quantitative measure of the acetylated NP present or procedural losses. However, consistent trends were evident that indicated that the length of time the aqueous and organic fractions were left in contact was an important factor in determining the final amount of acetylated NP.  Contact  between the organic and aqueous fractions also occurs during the acetylation and extraction period.  The influence of time  during this part of the procedure was next examined by varying the length of the shaking time. 3.10.6  The Shaking Time Study  3.10.6.1  Summary of Method  Three 100 mg/L NP solutions were prepared by adding 0.2 mL of 1 g/L NP stock to hexane to a final volume of 2 mL. The samples were derivatized individually using the procedures described in Section 3.10.2.1, but varying the amount of time taken to perform the acetylation and extraction.  Shaking  times of one minute were used for each acetylation/extraction with Sample 1, two minutes for Sample 2, and three minutes for  110  Sample 3.  The total mixing time for Sample 1 was therefore s  two minutes, Sample 2 was four minutes, and Sample 3 was six minutes.  The organic fractions were washed with 0.05M  potassium carbonate, then concentrated to 2 mL and analyzed. A background sample was also prepared and analyzed. 3.10.6.2  Results  Table 21 summarizes the results obtained.  Using an  acetylated 40 mg/L NP standard, the R.T. range for derivatized NP was confirmed as 15.1 to 16.8 minutes. No contaminant peaks were present in this R.T. range on the chromatogram for the background sample. The estimated percent acetylation was high for all samples, ranging from 97 to 99%. Sample 1, with a total shaking time of two minutes, had the highest estimated percent conversion value of 99%. Conversion rates to the NP acetate declined slightly as the total shaking time increased, with the total mixing time of six minutes used for Sample 3 yielding the lowest conversion rate. The reverse trend was observed with total NP peak area, where longer shaking times yielded higher values. The significance of this was undetermined, but may indicate that total recoveries of NP were improved with longer mixing times.  Ill  Table 21.  Results of the Shaking Time Study  Total  Peak Area  Mixing  in R.T. Range  Total NP  Estimated  Peak Area  Percent  Time  Acetylated  (min.) 12.8-14.5  15.1-16.8  minutes  minutes  2  8  693  701  9?  4  16  851  867  98  6  27  870  898  97  3.10.6.3  Discussion  The results of the Timed Shaking Study were a further indication of the critical role played by reaction time in achieving successful derivatization.  All shaking times used  yielded estimated conversion values that were greater than 95%, but higher conversion efficiencies occurred with shorter reaction times. The observed increase in total NP peak area with increasing reaction time, however, suggested analyte recoveries could be adversely affected if mixing times were  112  too short. A four minute total mixing time was selected to provide a compromise between reaction and recovery efficiencies, and percent recovery work was undertaken. 3.10,7  Determination of Percent Recovery  3.10.7.1  Summary of Method  The ratio between the estimated amount of target analyte measured by an analytical method to the initial amount known to be present in the sample, describes the recovery efficiency for the procedure.  The method used to determine percent  recovery was based on procedures described by Dr. Hing-Biu Lee of the National Water Research Institute, Environment Canada, Burlington, Ontario (personal communication, January 1998). A 100 mg/L NP standard solution was prepared by adding 5.0 mL of 1 g/L NP stock into hexane to a final volume of 50 mL.  Five 2 mL aliquots (Lab Replicates 1 to 5) were  distributed into individual 40 mL clear glass vials, and each aliquot was acetylated individually using the procedure outlined in Section 3.10.2.1.  A series of three calibration  standards was also prepared, according to the procedure described in Section 2.3.2, and used to establish response linearity over the concentration range tested.  As described  in Dr. Lee's procedure, the recovery for the calibration standards was considered to be 100%. The amount of NP in each lab replicate after acetylation was then determined by comparing the NP peak area for the lab replicate with  113  that of the calibration standards. 3.10.7.2  Results  Based on the calibration standards, the R.T. range for derivatized NP was 15.1 to 16.8 minutes. The percent recovery for each lab replicate is given in Table 22. Table 23 provides an example calculation for determining the percent recovery of NP.  Table 22.  Percent recoveries for Lab Replicates 1 to 5  Sample I.D.  Total PA for  Percent Recovery  derivatized NP  of NP  311  100  1  391  93  2  424  102  3  430  102  4  442  105  5  376  90  25 mg/L standard  114  Table 23.  Example calculation for determining the percent recovery of NP  Total PA for derivatized NP in 25 mg/L standard =  311  Total PA for derivatized NP in Lab Replicate 1  391  =  Since response is linear, NP concentration in 1 =  (391)(25 mg/L)/311  =  31 mg/L  Initial amount of NP in Lab Replicate 1 = 2 mL  x  100 mg/L  =  0.200 mg NP  Final amount of NP in Lab Replicate 1 6 mL  x  31 mg/L  =  Percent recovery of NP 0.186 mg/0.200 mg  x  =  0.186 mg NP = 100  =  93  Recovery of NP after acetylation was 90% or greater for all five lab replicates, and ranged from 90% to 105%. Mean percent recovery was 98% with a standard deviation of 6.  115  3.10.7.3  Discussion  Percent recoveries using the Wilson procedure were high for all of the test samples. Procedural simplicity and the minimal sample handling required by the method most likely contributed significantly to the observed high recovery efficiencies.  However, results from the Replicate Test and  Timed Shaking Study established that recovery efficiencies for derivatized NP decrease substantially when the overall time to perform the acetylation exceeds about 15 minutes. When measuring derivatized NP, then, the Wilson method can be considered to have high recovery efficiency provided it is performed without time delays.  3.11  Conclusions on Method Development The difficulties encountered in achieving successful  acetylation of NP shifted the research focus toward development of a derivatization method.  The experimental  evidence indicated that an acetylation method which relied upon the action of acetic anhydride and potassium carbonate could not be applied directly to NP in hexane.  The NP,  however, proved unexpectedly resistant to transfer out of hexane.  Methods which provided high derivatization  efficiencies incorporated a solvent that possessed some miscibility in both hexane and the water-based potassium carbonate.  The Lee-adapted procedure used methanol as this  116  intermediate solvent, and the Wilson procedure used acetone. The third solvent may act as a "bridge" between the two immiscible solvents, facilitating reaction.  The failure of  the Lee-adapted procedure to produce good results when applied to SPMD dialysates may relate more to the reactant volumes used, particularly acetic anhydride, than to any weakness in the method itself. Another unexpected complication was the apparent tendency for the acetylation reaction to produce a final equilibrium between the acetylated and non-acetylated compounds. The results of the Final Wash Study suggested that reaction times in excess of about fifteen minutes were sufficiently long to permit establishment of such an equilibrium.  However,  applying Le Chatelier's principle to the reaction, the excess of the reactants NP and acetic anhydride should have produced a shift to the right in the equilibrium, favoring formation of the acetylated product.  It may be that the presence of a  base, K2C03, or an acid, acetic acid, produces a counteracting shift to the left if the acetylated NP ester formed is very acid- or base-labile (Morrison and Boyd, 1973).  The answer  may be found in a thorough investigation into the kinetics of the reaction, but such an investigation is beyond the scope of this research project.  117  3.12  Preliminary Field Testing  3.12.1  Introduction  Experiments performed under controlled laboratory conditions indicated that the conventional SPMD system could be used successfully to sequester waterborne 4-NP.  Further,  NP present in the resulting dialysates could be acetylated and collected in the second eluting fraction of a silica gel cleanup procedure.  The final test was application of these  methods under field conditions. Limited by time, a preliminary field test was devised.  The objective for this  field study was to evaluate the methods developed in this research project with actual field samples. The results obtained would pinpoint weaknesses in the procedures and define areas for future work. For the field study, samples of final effluent were collected from the UBC Pilot Plant.  The UBC Pilot Plant is a  2500 L capacity biological nutrient removal (BNR) waste treatment plant located on university lands, where research is conducted on municipal wastewater and sludge treatment. Domestic sewage from campus buildings serves as the feed source for the facility.  The samples used for the field  testing program were collected after secondary treatment, and brought to the laboratory for testing.  In this way, SPMDs  were exposed to field samples in a controlled laboratory setting.  118  3.12.2  Summary of Method  Samples of secondary STP effluent from the UBC Pilot Plant were collected in 4 L amber glass bottles and returned to the laboratory for testing.  The effluent samples were  stored in the dark at 4°C for up to 10 days, until ready for use in testing. SPMDs were prepared using the procedures described in Section 2.2.1.  The test consisted of one tap water Control  and seven replicates of undiluted effluent.  A three week  exposure period was used, with test solution renewals performed three times each week.  Test solution renewals were  made as described in Section 3.1.1.  Test jars were held at  room temperature in the dark. At the end of the exposure period, SPMDs were processed according to methods described in Section 2.2.4. The resulting dialysates were covered securely with hexane-rinsed aluminum foil and placed in the dark at -15°C to await processing. To establish initial NP concentration ranges, one test dialysate was concentrated by rotoevaporation in a 40 - 45°C water bath to a volume of approximately 10 mL. The dialysate was further concentrated to 2 mL under a gentle stream of nitrogen, using 1 mL of iso-octane as a keeper. A 1 mL subsample of the concentrated SPMD dialysate was collected and analyzed using GC-MS in selected ion monitoring (SIM) mode.  Ion peaks selected for analysis were at m/z 107,  119  121, 135, 149, and 220. A second test dialysate was concentrated to 2 mL as described above and then derivatized.  To perform the  derivatization, the 2 mL concentrate was transferred to a 40 mL clear glass vial and 3 mL of 1:2 acetone:hexane was added.  The vial was shaken gently to mix the contents, then  5 mL of 0.1M K2C03 and 2 mL of acetic anhydride were added. The vial was hand-shaken for 2 minutes, venting often, then the organic layer was collected using a Pasteur pipette and removed to a 14 mL amber vial.  A second 2-minute extraction  with 3 mL of 1:2 acetone:hexane was performed on the remaining aqueous fraction.  The resulting organic fraction  was collected and removed into the amber vial, and the pooled organic fractions washed for l minute with 3 mL of 0.05M K2C03.  The washed organic fraction was removed into a 15 mL  graduated centrifuge tube. The total volume of organic fraction collected was 5 mL.  The sample was then concentrated  to 2 mL under a gentle stream of nitrogen, using 1 mL of isooctane as a keeper.  A 1 mL subsample of the final concentrate  was collected for analysis using GC-MS in SIM mode.  Ion peaks  selected for the acetylated SPMD sample were at m/z 177, 205, 220, and 262. A second set of samples was processed once results from the above two samples had provided preliminary measures for NP.  A test dialysate and the Control dialysate were  120  concentrated to 2 mL by rotoevaporation and nitrogen blowdown, and analyzed directly.  The derivatization procedure described  above was performed on two more test dialysates.  The first of  these was analyzed after derivatization, while the second was further concentrated with nitrogen to approximately 1 mL and passed through a column containing 5 cm of 5% deactivated silica gel covered by a 5 mm layer of sodium sulfate. The sample was sequentially eluted using 10 mL each of hexane, 50% methylene chloride in hexane, and 1% methanol in methylene chloride.  All three fractions were collected and concentrated  to 2 mL for analysis.  Samples processed in the second set of  testing were analyzed using GC-MS in SIM mode and GC-FID. 3.12.3  Results  The laboratory testing results for underivatized SPMD dialysates analyzed using GC-MS in SIM mode are provided in Table 24. NP was present in test dialysates, and absent from the Control.  Estimates for test dialysate NP concentrations  were obtained by comparison with peak areas for concurrently run standards, and ranged from 4 to 5 mg/L NP. The results for testing with derivatized samples are provided in Figures 18 and 19. The R.T. range of derivatized NP for samples analyzed with GC-MS in SIM mode was 14.6 to 15.8 minutes. The chromatograms for a derivatized 10 mg/L NP standard (Figure 18) and a derivatized SPMD dialysate  121  Table 24.  Field test results for underivatized SPMD dialysates analyzed using GC-MS in SIM mode  Analysis Date  Sample 1.0.  Total NP Peak Area  13 July 1998  21 July 1998  5 mg/L standard  13,967,111  Test dialysate  15,016,362  10 mg/L standard  38,512,631  Control dialysate  0  Test dialysate  15,900,544  (Figure 19) showed good correlation of the NP peaks. This confirmed that the NP present in an SPMD dialysate was successfully acetylated using the Wilson procedure. Chromatograms for the derivatized SPMD extract following silica gel cleanup are provided in Figures 20 to 22. NP was absent in Fraction 1 (Figure 20) and Fraction 3 (Figure 22) and present in Fraction 2 (Figure 21), indicating successful isolation of the target analyte into Fraction 2. A comparison of the NP peaks in Fraction 2 (Figure 21) with those of the  122  acetylated SPMD dialysate (Figure 19) confirmed that the acetylation was successful and complete in the silica gel sample. Analysis of the samples with GC-MS in SIM mode greatly improved the ease and accuracy of data interpretation when compared with analysis using GC-FID.  The chromatogram of the  Fraction 2 silica gel sample obtained with GC-MS-SIM (Figure 21) showed clearly defined NP peaks. The same sample analyzed using GC-FID (Figure 23) contained a profusion of background peaks that made the definitive identification of NP difficult. The many background peaks indicated, as well, that although the silica gel procedure was successful at collecting NP in the second eluting fraction, it did not isolate it from the many contaminants present in the SPMD dialysate.  123  Abundance ; 100000  TIC: 10PPMD.D  13.9354 i*48ftt  80000  6000040000 20000 0  Fime->  I M M | I I I I |  A i-, y^r-iA^  1111  I/xMvUv^. I i ' "  'i r n  i | i i i i | i i i i | i i i ,.  6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.  Figure 18.  Chromatogram of derivatized 10 mg/L NP standard obtained using GC-MS in SIM mode  Time  6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.1  Figure 19.  Chromatogram of derivatized SPMD dialysate obtained using GC-MS in SIM mode  124  TIC:SPMDF1S.D  Abundance  2074  15000  10000  5000  J  _-_A ~~A-X  i_  JJML~  I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I. I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I  Time->  6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23  Figure 20.  Chromatogram of silica gel cleanup procedure Fraction 1 obtained using GC-MS in SIM mode  TIC: SPMDF2S.D  Abundance  13.22  60000  40000 16.65 20000  j  oJllL—^uJL JUUIA  I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' " ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' l ' ' ' ' [  Time >  6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.  Figure 21.  Chromatogram of silica gel cleanup procedure Fraction 2 obtained using GC-MS in SIM mode  125  TIC: SPMDF3.D  Abundance 40000  30000-  20000-  10000  Ji Time~->  JL  6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 23.'  Figure 22.  Chromatogram of silica gel cleanup procedure Fraction 3 obtained using GC-MS in SIM mode "Hn"lTA7'(LpRI\NP1CiCHU5:i3)_ pA  5$ ; j r - * 25  20-  15  10-  S^  h- 'ten  OT-  co oxo t ' - o J  ^r'VW*S*+flffi'  10  Figure 23.  12  16  Chromatogram of silica gel cleanup procedure Fraction 2 obtained using GC-FID  126  3.12.4  Discussion  The field test results provided evidence that SPMDs could be used successfully to sequester NP from secondary sewage treatment plant effluents. Although effluent concentrations of NP were not measured, the presence of NP in the test dialysates indicated that the SPMD system was collecting and concentrating the chemical from the wastewater. 4-5  The  mg/L NP concentration estimated for the field sample  dialysates was comparable with the 2 - 4 mg/L NP dialysate concentration estimated for the preliminary SPMDs analyzed March 28, 1996 (Section 3.1.2).  These SPMDs were exposed to a  water concentration of 10 /ig/L NP, with no test solution renewal.  The field sample SPMDs received nine test solution  renewals over the 3 week exposure period, for a total of ten exposures to fresh test solution.  Assuming a total water  exposure concentration in the range of 10 ng/L NP, provides an estimated effluent concentration of one-tenth of 10 (Mg/L, or 1 Mg/L NP. However, the values for NP dialysate concentrations are very broad estimates only, based on the assumption of linearity in response for standards and samples. SPMDs exposed under actual field conditions could be expected to yield higher dialysate NP concentrations, as the exposure period is typically 28 to 60 days rather than the 21 days used for this study.  In addition, SPMDs in the field receive  127  continuous, flow-through exposures as opposed to the static renewal system used here. The success of the derivatization procedure when applied to dialysates from effluent-exposed SPMDs indicated that the method is suitable for field study.  The silica gel cleanup  procedure was only partially successful.  NP was isolated in  the second eluting fraction as intended, but the presence of many background peaks indicated there were a large number of chemicals present in the effluent with polarities similar to that of NP. Adjusting the polarities of the eluting solvents may improve separation efficiency.  As well, employing another  cleanup method such as gel permeation chromatography (GPC) alone or in combination with silica gel may prove useful. Data interpretation was easiest using GC-MS in SIM mode. Here, there was the least opportunity for incorrect interpretation when matching retention times in test dialysate chromatograms with those of standard solutions. When GC-FID was applied to chromatograms having substantial background material, the characteristic pattern of NP peaks was obscured and assessment for NP made difficult.  Another difficulty that  was encountered with data analysis resulted from the multiple peaks characteristic of NP. As the concentration of chemical decreased, lower peaks disappeared first below the detection limit.  The resulting decrease in total NP peak area was not  always linear, as would be the case for a chemical which  128  produced a single chromatographic peak.  The effect was most  marked at very low NP concentrations, when only the highest peaks were evident on the chromatogram, and greatly hindered concentration estimates.  Finally, it was evident particularly  on the GC chromatograms that the derivatization procedure used as a preliminary cleanup, did not sufficiently clean the sample.  Analysis using GC-MS in SIM mode, while removing some  portion of the NP from subsequent concentration estimates, provided the most reliable method for determining the presence of the chemical.  129  4.0  CONCLUSIONS AND RECOMMENDATIONS  4.1  Conclusions This project investigated the application of SPMDs to the  sequestration of waterborne 4-NP.  A method was developed for  acetylating NP present in SPMD dialysates, first by examining existing water-based procedures and then by adapting a method designed for use with organic solvents.  Finally, a cleanup  procedure using activated silica gel was applied to acetylated NP samples. The results obtained led to the following conclusions:  (1) SPMDs can be applied successfully to the sequestration of waterborne NP;  (2) NP present in hexane-based SPMD dialysates does not acetylate efficiently when a water-based acetylation method is applied directly;  (3) the NP in SPMD dialysates will not readily or efficiently transfer from hexane to a water-miscible solvent using solvent exchange procedures;  (4)  for an acetylation method which relies on the use of acetic anhydride and potassium carbonate, including an intermediate solvent with at least partial miscibility in  130  both hexane and water greatly improves acetylation efficiency; and  (5) the acetylation reaction between hexane-based NP and acetic anhydride, in the presence of potassium carbonate, appears to favor establishment of a final equilibrium between acetylated and non-acetylated NP, rather than one where the acetylated product is predominant.  4.2  Recommendations Based on the findings of this research project, the  following recommendations are made:  (1) that a field study be undertaken to examine the uptake of 4-NP from water by SPMDs under field conditions and to assess the presence of 4-NP in local waters;  (2) that a quantitative analysis of the kinetics associated with uptake of 4-NP by SPMDs be done that permits estimation of ambient water concentrations; and  (3) that the derivatization method developed in this research project be verified by further application to SPMD dialysates from samples of STP and industrial wastewaters.  131  5.0  LITERATURE CITED  Ahel, M. and W. Giger. 1985. Determination of alkylphenols and alkylphenol mono and diethoxylates in environmental samples by HPLC. Anal. Chem. 57: 1576-1583. Ahel, M., J. McEvoy and W. Giger. 1993. Bioaccumulation of the lipophilic metabolites of nonionic surfactants in freshwater organisms. Environ. Pollut. 79: 243-248. Ahel, M., W. Giger and M. Koch. 1994. Behavior of alkylphenol polyethoxylate surfactants in the aquatic environment. Water Res. 28: 1131-1152. Ahel, M., C. Schaffner and W. Giger. 1996. Behavior of alkylphenol polyethoxylate surfactants in the aquatic environment. III. Occurrence and elimination of their persistent metabolites during infiltration of river water to groundwater. Water Res. 30(1): 37-46. Aldrich. 1996/97 Catalog Handbook of Fine Chemicals. Aldrich Chemical Company, Inc., Milwaukee, WI. Armstrong, J.A. and P.D. Kingsbury. 1979. Interim Progress Report. Forest Pest Management Institute, Canadian Forestry Service. 63 p. Babin, P.G. and J.M. Vernier. 1989. fish. J. Lipid Res. 30: 467-480.  Plasma lipoproteins in  Beland, P. 1989. Annual report 1989. Quebec: St. Lawrence National Institute of Ecotoxicology. Blackburn, M.A. and M.J. Waldock. 1995. Concentrations of alkylphenols in rivers and estuaries in England and Wales. Water Res. 29(7): 1623-1629. Chen, T.T. 1983. Identification and characterisation of estrogen responsive gene products in the liver of the rainbow trout. Can. J. Biochem. Cell. Biol. 61: 802-810. Coburn, J.A., B.D. Ripley and A.S.Y. Chau. 1976. Analysis of pesticide residues by chemical derivatization. II. N-Methylcarbamates in natural water and soils. JAOAC 59(1): 188-196. Cohen, I.e., J. Norcup, J.H.A. Ruzicka and B.B. Wheals. 1969. Trace determination of phenols by gas chromatography as their 2,4-dinitrophenyl esters. J. Chromatogr. 44: 251-255.  132  5.0  LITERATURE CITED  continued  Colburn, T., F.S. vom Saal and A.M. Soto. 1993. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspect. 101: 378-384. Comber, M.H.I., T.D. Williams and K.M. Stewart. 1993. The effects of nonylphenol on Daphnia magna. Wat. Res. 27: 273276. Davis, W.P. and S.A. Bortone. 1992. Effects of Kraft mill effluent on the sexuality of fishes: an environmental early warning? In: Chemically induced alterations in sexual and functional development: the wildlife/human connection. T. Colburn and C. Clement, editors. Princeton Scientific Publishing, Princeton, N.J. Ellis, D.V. and L.A. Pattisina. 1990. Widespread neogastropod imposex: a biological indicator of global TBT contamination. Mar. Pollut. Bull. 21: 248-253. Ellis, G.S., J.N. Huckins, C.E. Rostad, C.J. Schmitt, J.D. Petty and P. MacCarthy. 1995. Evaluation of lipid-containing SPMDs for monitoring organochlorine contaminants in the Upper Mississippi River. Environ. Tox. Chem. 14(11): 1875-1884. Encyclopedia of chemical technology. 1992. Fourth edition. Volume 2. J.I. Kroschwitz and M. Howe-Grant, editors. John Wiley and Sons, New York, NY. Erdman, T.C. 1988. Report to the U.S. Fish and Wildlife Service on common and Forster's tern productivity on Kidney Island confined disposal facility, Green Bay, 1987, with supplemental necropsy and pathology reports. Green Bay: University of Wisconsin. Erhardt-Zabik, S., J.T. Watson and M.J. Zabik. 1990. Selective sensitivity of highly chlorinated species in negative ion mass spectrometry - implications for complex mixture analysis. Biomed. Environ. Mass Spectrom. 19: 101108. Favretto, L., B. Stancher and F. Tunis. 1978. Extraction and determination of polyoxyethylene alkyl ether non-ionic surfactants in water at trace levels. Analyst 103: 955-962. Fluka Chemika-BioChemika Analytika 1995/96. Switzerland.  133  Fluka Chemie AG.  5.0  LITERATURE CITED  continued  Fry, D.M. and C.K. Toone. 1981. DDT-induced feminization of gull embryos. Science 231: 919-924. Giger, W., P.H. Brunner and C. Schaffner. 1984. 4Nonylphenol in sewage sludge: accumulation of toxic metabolites from nonionic surfactants. Science 225: 623-625. Giger, W., M. Ahel, M. Koch, H.U. Laubscher, C. Schaffner and J. Schneider. 1987. Behavior of alkylphenol polyethoxylate surfactants and of nitrilotriacetate in sewage treatment. Wat. Sci. Technol. 19: 449-460. Granmo, A., R. Ekelund, K. Magnusson and M. Berggren. 1989. Lethal and sublethal toxicity of 4-nonylphenol to the common mussel (Mytilus edulis L.). Envir. Pollut. 59: 115-127. Hassett, J.P., M. Force and H. Song. 1989. A passive in-situ sampler for organic compounds in water. Proceedings, 198th National Meeting of the American Chemical Society, Miami Beach, FL., September 10-15. 180: 496-497. Henderson, J.R. and D.R. Tocher. 1987. The lipid composition and biochemistry of freshwater fish. Prog. Lipid Res. 26: 281-347. Hoel, D.G., D.L. Davis, A.B. Miller, E.J. Sondik and A.J. Swerdlow. 1992. Trends in cancer mortality in 15 industrialized countries, 1969-1986. J. Natl. Cancer Inst. 84: 313-320. Huckins, J.N., M.W. Tubergren and G.K. Manuweera. 1990. Semipermeable membrane devices containing model lipid: a new approach to monitoring the bioavailability of lipophilic contaminants and estimating their bipconcentration potential. Chemosphere 20(5): 533-552. Huckins, J.N., G.K. Manuweera, J.D. Petty, D. Mackay and J.A. Lebo. 1993. Lipid-containing semipermeable membrane devices for monitoring organic contaminants in water. Environ. Sci. Technol. 27(12): 2489-2496.  Huckins, J.N., J.D. Petty, C.E. Orazio, J.L. Zajicek, V.L. Gibson, R.C. Clark and K.R. Echols. 1994. 15th Annual Meeting, Society of Environmental Toxicology and Chemistry. Denver, CO. (Abstr.)  134  5.0  LITERATURE CITED  continued  Huckins, J.N., J.D. Petty, J.A. Lebo, C.E. Orazio, H.F. Prest, D.E. Tillitt, G.S. Ellis, B.T. Johnson and G.K. Manuweera. 1996. Semipermeable membrane devices (SPMDs) for the concentration and assessment of bioavailable organic contaminants in aquatic environments. In: Techniques in aquatic toxicology. G.K. Ostrander, editor. Lewis Publishers, Boca Raton, FL. Hwang, S.T. and K. Kammermeyer. 1975. Membranes in separations. Robert E. Krieger Publishing, Malabar, FL. pp.  559  Jobling, S. and J.P. Sumpter. 1993. Detergent components in sewage effluent are weakly oestrogenic to fish: An in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquatic Toxicology 27: 361-372. Johnson, G.D. 1991. Hexane-filled dialysis bags for monitoring organic contaminants in water. Environ. Sci. Technol. 25: 1897-1903. Klaassen, C D . and J. Doull. 1980. Evaluation of safety: Toxicologic evaluation. In: Casarett and Doull's toxicology, the basic science of poisons. Second edition. J. Doull, C D . Klaassen and M.O. Amdur, editors. Macmillan Publishing Co., Inc., New York, NY. Kubiak, T.J., H.J. Harris, L.M. Smith, T.P. Schwartz, D.L. Stalling, J.A. Trick, L. Sileo, D.E. Docherty and T.C. Erdman. 1989. Microcontaminants and reproductive impairment of the Forster's tern on Green Bay, Lake Michigan - 1983. Arch. Environ. Contam. Toxicol. 18: 706-727. Kvestak, R. and M. Ahel. 1994. Occurrence of toxic metabolites from nonionic surfactants in Krka River estuary. Ecotoxicology and Environmental Safety 28: 25-34. Lamparski, L.L. and T.J. Nestrick. 1978. Determination of trace phenols in water by gas chromatographic analysis of heptafluorobutyryl derivatives. J. Chromatogr. 156: 143-151. Leatherland, J. 1992. Endocrine and reproductive function in Great Lakes salmon. In: Chemically induced alterations in sexual and functional development: the wildlife/human connection. Colburn, T. and C Clement, editors. Princeton Scientific Publishing, Princeton, NJ.  135  5.0  LITERATURE CITED  continued  Lebo, J.A., J.L. Zajicek, J.N. Huckins, J.D. Petty and P.H. Peterman. 1992. Use of semipermeable membrane devices for the in situ monitoring of polycyclic aromatic hydrocarbons in aquatic environments. Chemosphere 25: 697-718. Lech, J.J., S.K. Lewis and L. Ren. 1996. In vivo estrogenic activity of nonylphenol in rainbow trout. Fund, and Applied Toxicol. 30: 229-232. Lee, P.C. and W. Lee. 1996. In vivo estrogenic action of nonylphenol in immature rats. Bull. Environ. Contam. Toxicol. 57: 341-348. Lee, H.B. and T.E. Peart. 1995. Determination of 4Nonylphenol in effluent and sludge from sewage treatment plants. Anal. Chem. 67: 1976-1980. Legovic, T., Z. Grzetic and A. Smircic. 1989. Effects of wind on a stratified estuary. Mar. Chem. 32: 153-161. Lewis, S.K. and J.J. Lech. 1996. Uptake, disposition, and persistence of nonylphenol from water in rainbow trout (Oncorhynchus mykiss). Xenobiotica 26(8): 813-819. Lieb, W.R. and W.D. Stein. 1969. Biological membranes behave as non-porous polymeric sheets with respect to the diffusion of non-electrolytes. Nature 224: 240-243. Litten, S., B. Mead and J. Hassett. 1993. Application of passive samplers (PISCES) to locating a source of PCBs on the Black River, New York. Environ. Toxicol. Chem. 12: 639-647. McLeese, D.W., V. Zitko, D.B. Sergeant, L. Burridge and C D . Metcalfe. 1981. Lethality and accumulation of alkylphenols in aquatic fauna. Chemosphere 10(7): 723-730. Mac, M.J., T. Schwartz and C.C. Edsall. 1988. Correlating PCB effects on fish reproduction using dioxin equivalents. Presented at the Ninth Annual SETAC Meeting, Arlington, Virginia. Martineua, D., A. Lagace, P. Beland, R. Higgins, D. Armstrong and L.R. Shugart. 1988. Pathology of stranded beluga whales (Delphinapterus leucas) from the St. Lawrence estuary, Quebec, Canada. J. Comp. Pathol. 98: 287-311.  136  5.0  LITERATURE CITED  continued  The Merck Index. An encyclopedia of chemicals, drugs, and biologicals. Tenth edition. 1983. M. Windholz, editor. Merck and Co., Inc., Rahway, NJ. Moccia, R.D., J.F. Leatherland and R.A. Sonstegard. 1981. Quantitative interlake comparison of thyroid pathology in Great Lakes coho (Oncorhynchus kisutch) and Chinook (Oncorhynchus tschawytscha) salmon. Cancer Res. 41: 22002210. Moccia, R.D., G. Fox and A.J. Britton. 1986. A quantitative assessment of thyroid histopathology of herring gulls (Larus argentatus) from the Great Lakes and a hypothesis on the causal role of environmental contaminants. J. Wild. Dis. 22: 60-70. Morrison, R.T. and R.N. Boyd. 1973. Organic chemistry. Third edition. Allyn and Bacon, Inc., Boston, MA. Mueller, G.C. and U. Kim. 1978. Displacement of estradiol from estrogen receptors by simple alkyl phenols. Endocrinology 102(5): 1429-1435. Munkittrick, K.R., C.B. Port, G.J. Van Der Kraak, I.R. Smith and D.A. Rokosh. 1991. Impact of bleached kraft mill effluent on population characteristics, liver MFO activity, and serum steroids of a Lake Superior white sucker (Catostomus commersoni) population. Can. J. Fish. Aquat. Sci. 48: 1-10. Naylor, C.G., J.P. Mieure, W.J. Adams, J.A. Weeks, F.J. Castaldi, L.D. Ogle and R.R. Romano. 1992. Alkylphenol ethoxylates in the environment. JAOCS 69(7): 695-703. Nederlof, K.P., H.W. Lawson, A.F. Saftlas, H.K. Atrash and E.L. Finch. 1990. Ectopic pregnancy surveillance, United States, 1970-1987. MMWR 39: 9-17. Opperhuizen, A., E.W.v.d. Velde, F.A.P.C. Gobas, D.A.K. Liem and J.M.D.v.d. Steen. 1985. Relationship between bioconcentration in fish and steric factors of hydrophobic chemicals. Chemosphere 14: 1871-1896. Paasivirta, J., R. Paukku, S. Herve, P. Heinonen and A. Sodergren. 1991. Uptake of organochlorines from lake water by hexane-filled dialysis membranes and by mussels. Chemosphere 22(11): 997-1001.  137  5.0  LITERATURE CITED  continued  Petty, J.D., J.N. Huckins and J.L. Zajicek. 1993. Application of semipermeable membrane devices (SPMDs) as passive air samplers. Chemosphere 27: 1609-1624. Prest, H.F., J.N. Huckins, J.D. Petty, S. Herve, J. Paasivirta and P. Heinonen. 1995. A survey of recent results in passive sampling of water and air by SPMDs. Mar. Poll. Bull. 31: 306312. Registry of toxic effects of chemical substances. 1990. National Institute for Occupational Safety and Health, Cincinnati, OH. Reijinders, P.J.H. 1986. Reproductive failure in common seals feeding on fish from polluted coastal waters. Nature 324: 456-457. Rohr, A.C. 1994. Application of semipermeable membrane devices (SPMDs) to the monitoring of kraft mill effluents with emphasis on potential fish-tainting compounds. M.A.Sc. thesis. The University of British Columbia, Vancouver, B.C. Schmitt, C.J., J.N. Huckins, J.D. Petty, C.E. Rostad and G.S. Ellis. 1993. Halogenated organic contaminants in the Upper Mississippi: Test of a passive accumulator. Abstracts, 14th Annual Meeting, SETAC, Houston, TX., November 14-18. Schneider, R. 1982. Polychlorinated biphenyls (PCBs) in cod tissues from the western Baltic: Significance of equilibrium partitioning and lipid composition in the bioaccumulation of lipophilic pollutants in gill-breathing animals. Meeresforschung 29: 69-70. Sharpe, R.M., N.E. Skakkebaek. 1993. Are oestrogens involved in falling sperm count and disorders of the male reproductive tract? Lancet 341: 1392-1395. Shugart, G. 1980. Frequency and distribution of polygony in Great Lakes herring gulls in 1978. Condor 82: 426-429. Sithole, B.B. and L.H. Allen. 1989. Determination of nonionic nonylphenol ethoxylate surfactants in pulp and paper mill process samples by spectrophotometry and liquid chromatography. J. Assoc. Off. Anal. Chem. 72(2): 273-276.  138  5.0  LITERATURE CITED  continued  Sithole, B.B., B. Zvilichovsky, C. Lapointe and L.H. Allen. 1990. Adsorption of aqueous nonylphenol ethoxylate surfactants on metal sample loops: effect on quantitation by liquid chromatography. J. Assoc. Off. Anal. Chem. 73(2): 322324. Sodergren, A. 1987. Solvent-filled dialysis membranes simulate uptake of pollutants by aquatic organisms. Environ. Sci. Technol. 21: 855-863. Soto, A.M., H. Justicia, J.W. Wray and C. Sonnenschein. 1991. p-Nonyl-Phenol: an estrogenic xenobiotic released from "modified" polystyrene. Environ. Health Perspect. 92: 167173. Spacie, A. and J.L. Hamelink. 1985. Bioaccumulation. In: Fundamentals of aquatic toxicology: Methods and applications. G.M. Rand and S.R. Petrocelli, editors. Taylor and Francis, Bristol, PA. Stephanou, E. and W. Giger. 1982. Persistent organic chemicals in sewage effluents. 2. Quantitative determinations of nonylphenols and nonylphenol polyethoxylates by glass capillary gas chromatography. Environ. Sci. Technol. 16(11): 800-805. Sundaram, K.M.S. 1995. Liquid chromatographic method for the determination of nonyl phenol surfactant present in the commercial and spray formulations of Aminocarb (MatacilR) insecticide. J. Liquid Chromatogr. 18(9): 1787-1799. Sundaram, K.M.S., S. Szeto, R. Hindle and D. MacTavish. 1980. Residues of nonylphenol in spruce foliage, forest soil, stream water and sediment after its aerial application. J. Environ. Sci. Health B15(4): 403-419. Swisher, R.D. 1987. Surfactant biodegradation. Dekker, New York, NY.  Marcel  Voss, R.H., J.T. Wearing and A. Wong. 1981. A novel gas chromatographic method for the analysis of chlorinated phenolics in pulp mill effluents. In: Advances in the identification and analysis of organic pollutants. L. H. Keith, editor. Ann Arbor Science Publishers, Ann Arbor, MI.  139  5.0  LITERATURE CITED  continued  Wilson, Ann Elise-Jordan. 1996. Characterization and enumeration of the resin acid-degrading bacterial population of a sequencing batch reactor: an emphasis on the isopimaric acid-degrading bacteria. M.Sc. thesis. University of British Columbia, Vancouver, B.C. Zabik, J.M., L.S. Aston and J.N. Seiber. 1992. Rapid characterization of pesticide residues in contaminated soils by passive sampling devices. Environ. Toxicol. Chem. 11: 765770.  140  APPENDIX A  Certificate of analysis for 4-nonylphenol  141  A**** OC fluka Chamie A G T e L 0*1 753 2511 Industriajtrwie 2S Teles tS52S2 CH-9471 Buchs IMofax 081756 54 49  Fluka A  CERTIFICATE OF ANALYSIS PRODUCT-NO". PRODUCT  PURITY FORMULA HOLECUUR MASS  74430 4-NONYL-PHENOL 4-NONYLPHENOL  TECHN C1SH240  220.36  GEHALT (HPLC) ASSAY (HPLC)  90.0 X  ASPEKT APPEARANCE  COLOURLESS CLEAR VISCOUS LIQUID  OICHTE 020/4 DENSITY 020/4  0.949  BRECHUNGSINDEX N20/0 REFRACTIVE INOEX N20/D  1.512  INFRAROT-SPEKTRUM IR SPECTROSCOPY  CORRESPONDING  CHARGE/LOT  303517/1  13.12.94  Fluka Chemie AG Quality Control  ^ Or. G. van Look  s o Niaka  FluRa warrants t h a t - I t s products conform to the information contained m this and ether rt«ks publication*. Purchaser Must determine the suitability af the praduet for lee particular use. See reverse side ef invoice far additional tares ana conditions of sale.  142  APPENDIX B  GC-MS analysis of underivatized 4-nonylphenol  143  GC-MS Analysis of Nonylphenol  A 1000 mg/L 4-nonylphenol (4-NP) standard was analyzed by combination gas chromatography-mass spectrometry (GC-MS), using an HP 6890 in multiple ion detection (MID) mode. The Total Ion Current (TIC) gave a cluster of peaks between 28 and 33 minutes. A library spectrum of nonylphenol gave large peaks at m/e 107, 121, and 149, as well as a small base peak ion at m/e 220. Three spectra in the TIC of the 4-NP standard matched these criteria.  These were peaks at 30.00, 30.15, and  30.28 minutes. A library search for specific nonylphenol congeners gave two spectra - one for 2-nonylphenol and one for 4-nonylphenol. These spectra showed four large peaks at m/e 135, 107, 121, and 149, as well as a small base peak at m/e 220. Only two peaks met all these criteria. minutes.  These were at 30.02 and 30.16  Structures can be postulated for the fragment ions  m/e 107, 121, 135, and 149 based on the structure of nonylphenol and the cleavage of specific bonds under electron bombardment. From this analysis it would appear that the nonylphenol peaks in this sample came out at between 29 and 31.5 minutes.  144  ~~Joia[  800000  Xovi  Cui'-^ciT  nonpheno TIC 30 88 y  750000  30,15 /  700000  650000  600000 550000  500000  450000  400000  350000 300000  250000  200000  150000  100000  50000  —i  26.0  2b  1  rt  ~i  3t  29  rvcviul o^enol  145  33  X  35"  36.0  APPENDIX C  GC-MS analysis of underivatized and derivatized octylphenol  146  Scan 1103 (12.414 min): OP50.D  rwance  8000  6000  4000  2000  0  'i i i i I ' I  m/z->  'I"[''I  40  'I'"I  i I'"I  60  80  ' i'''i'"i  | 'i'"i' i ' i " ' ! 1 1  100  91  i 'i'"! 1  120  i i i i | i i i 'i |  140  160  180  I, 217 22S239 253263  23129230331*  | i - I i i | i i i i | i i i i | i i. i i | i i i i | i i i i | i  200  220_. 240  260  280  300  320  Octylphenol  Scan 1284 (13.863 min): OPDER.D  dance i ,  85  8000  6000  1 ,?  4000  2000 V,)"! 57 0  i  i  1 i  i  r V  1 -i' i  77 , •'•I  119  * •••'••'.  ! - • • •\  I ' I ' \< 1 1  | I1"  .  1 9  * 1S1  '>o« 191 --VG217  •I ' ' ' ' I ' ' ' '  Octylphenol acetate  147  /?48 j 2331 1 758 27 1 2 ^ 2 9 : 304 3153 1 1 1 1 1 1 1  APPENDIX D  GC-MS analysis of derivatized 4-nonylphenol  148  'Abundance  Ion plot for m/e 262, the molecular ion for 4-NP acetate  149  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0050165/manifest

Comment

Related Items