Standard UKWIR Technical Report Format - Defra, UKrandd.defra.gov.uk/Document.aspx?Document=WT03014_2073... · Web viewEstrogenicity determination in sewage treatment plants and surface

ENDOCRINE DISRUPTERS IN SEWAGE INFLUENTS - ANALYTICAL METHOD

DEVELOPMENT

Defra Report Ref. No. RSE-9

Programme Area & Reference Toxicology: Endocrine Disrupters – Wastewaters and Sludge TX/04

Report Title Endocrine Disrupters in Sewage Influents - Analytical Method Development

Project Management Gordon Wheale, UKWIR

Collaborator Department for Environment, Food and Rural Affairs

Contractor WRc-NSF Limited

Author of Report James H. A.

Kanda, R

Sutton, A.

Report Type Final

Period Covered 2003/2004

UKWIR Report Ref. No. 04/TX/04/9

UK Water Industry Research Limited provides a framework for a common research programme to undertake projects, which are considered to be fundamental to water operators on ‘one voice’ issues. Its contributors are the water and sewerage companies and the water supply companies of England and Wales, Scottish Water and Northern Ireland’s Water Service.

All statements contained in this document are made without responsibility on the part of UK Water Industry Research Limited, and its collaborators and are not to be relied upon as statements or representations of facts; UK Water Industry Research Limited and its collaborators on this project do not make or give, nor has any person authority on its behalf to make or give, any representation or warranty whatever in relation to the contents of this document or any associated software.


Published by UK Water Industry Research Limited1 Queen Anne’s Gate, London SW1H 9BT

First published 2004

ISBN 1 84057 339 2

UK Water Industry Research Limited 2004

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise, without the prior written consent of UK Water Industry Research Limited and its collaborators.

Printed by Webree.com Ltd.

UK WATER INDUSTRY RESEARCH LIMITED

ENDOCRINE DISRUPTERS IN SEWAGE INFLUENTS - ANALYTICAL METHOD DEVELOPMENT

Executive Summary

An analytical method has been developed to determine the natural oestrogens, oestrone and 17--oestradiol and the synthetic oestrogen ethinyl oestradiol in sewage influent.

Samples were extracted after filtration to separate the liquid and solid phase. The liquid phase was extracted using solid phase extraction (SPE) disks and the solid phase was extracted with methanol using accelerated solvent extraction (ASE). The extracts were purified using SPE cartridges. Analysis to determine free steroid oestrogens was undertaken using liquid chromatography mass spectrometry (LCMS) operated in the negative ion electrospray mode.

During method development a number of extraction disks, elution solvents and clean-up sorbents were investigated. Method development for the liquid phase involved the evaluation of a number of SPE disks. Octadecylsilyl (C18) disks were found to provide acceptable recovery of the free steroids. The solids were extracted with methanol using accelerated solvent extraction (ASE). The extracts were purified using C18 and aminopropyl SPE cartridges.

SPE extraction using disks provides a number of advantages over conventional extraction techniques for large volume sample preparation as they provide a large surface area for sorbent/sample contact, and therefore faster flow rates and higher throughput can be obtained for complex environmental samples such as raw or treated sewage.

Accelerated solvent extraction offers a number of advantages over other solvent extraction techniques, including lower solvent consumption, reduced extraction times, higher throughput through automation and ease of use. Furthermore, when used to extract sewage the resulting extract does not require filtration prior to analysis.

Statistical performance testing of the method shows that the procedure can be used for the routine monitoring of free steroids in sewage influent. The statistically derived limits of detection of the method to determine 17--oestradiol, ethinyl oestradiol and oestrone in sewage influent were 2.6, 11.4 and 5.2 ng l -1 respectively in the liquid phase and 65, 32 and 98 ng g-1 respectively in the solid phase.


For further information please contact UK Water Industry Research Limited, 1 Queen Anne’s Gate, London SW1H 9BT quoting the report reference number


Contents Page Number

1 Introduction 1

1.1 Project Objectives 2

2 Analytical Method Development 3

2.1 Strategy for method development 32.2 Sample collection and preparation 42.3 Development of sample extraction technique (Liquid Phase) 52.4 Development of sample extraction technique (Solid Phase) 9

3 Analytical Method validation 9

3.1 Method validation (Liquid Phase) 93.2 Method validation (Solid Phase) 14

4 Conclusion 18

5 References 18

APPENDIX A STANDARD OPERATING PROCEDURE FOR THE DETERMINATION OF FREE STEROID OESTROGENS IN SEWAGE INFLUENT 21

APPENDIX B CHROMATOGRAPHS 33



1 Introduction

The presence of trace quantities of endocrine disrupting chemicals (EDCs) in the aquatic environment has become an issue of considerable concern throughout the world. EDCs have the ability to alter or disrupt the body's hormone or endocrine system. There is evidence that aquatic organisms downstream of some wastewater treatment works (WwTW) show endocrine disruption as a result of exposure to substances in the effluent. Evidence for the oestrogenic activity of such effluents comes from the UK (Purdom et al.,1994)1Germany (Stumpf et al.,1996, Ternes et al.,1999) 2,3 Italy (Baronti et al, 2000) 4 and the USA ( Snyder et al.,1999)5.

The list of suspected endocrine disrupters includes some natural and synthetic steroid oestrogens. Research has indicated that the natural oestrogens, oestrone and 17--oestradiol and the synthetic oestrogen ethinyl oestradiol are responsible for most of the oestrogenic activity found in sewage effluents in the UK (Desbrow et al.,1998) 6.

The majority of work carried out on monitoring EDC concentrations in sewage treatment works has focused on the liquid phase of sewage effluents. Due to the lack of analytical methodology, there has been very little work carried out on levels of steroids in the aqueous phase of sewage influent or in the solid phase of sewage or in sewage sludge. A number of workers have therefore attempted to calculate the concentrations that are likely to be present in sewage influent. These estimates have been based on population equivalents, human excretion and water flow rates through wastewater treatment works.

Johnson et al. (1999)7 have estimated the concentration of the natural steroid oestrogen 17-oestradiol in sewage influent to be in the range 13.6-24 ng l -1 for six UK STWs. In a further study Johnson et al. (2000)8 estimated inputs of the natural steroids and the synthetic steroid 17-ethinyl oestradiol into activated sludge STWs based on population equivalents, contraceptive use and water flow rates. Predicted values were in the range 12-102, 5-44 and 1.1 to 5.1 ng l-1 for oestrone, 17--oestradiol and 17-ethinyl oestradiol respectively. These estimates were then tested against measured concentrations for five Italian and three Dutch activated sludge STWs. Predictions were particularly sensitive to assumptions regarding the number of women taking the oral contraceptive and were not very accurate.

Ternes et al. (1999)9 reported concentrations of oestrone and 17-oestradiol in the liquid phase of sewage influent in Brazil and Germany; average levels were 21, 40 and 4.2ng l-1

for 17-oestradiol, oestrone and 17-ethinyl oestradiol respectively, in the Brazilian raw sewage. Baronti et al. (2000)Error: Reference source not found monitored steroid oestrogen concentrations in influents of six Italian activated sludge WwTW. Average concentrations were as follows: 52 22 ng l-1, 12 5 and 3 2.6 ng l-1 of oestrone, 17-oestradiol and 17-ethinyl oestradiol respectively. Solé et al. (2000)10 detected oestriol in two sewage influents at ca260 ng l-1, whereas the concentrations of 17-oestradiol and oestrone were below the limit of detection of 50 ng l -1. Gentili et al. (2002)11 have reported an analytical method to determine free and conjugated steroids in the liquid phase of sewage influent using graphitized carbon black cartridges. A 100 ml sample of sewage influent was filtered and extracted using Carbograph 4 SPE cartridges and eluted with 80:20 (v/v) dichloromethane-methanol (10 ml).


The methods utilised in the literature, noted above, for the determination of free steroid oestrogens have not been performance tested. The lack of performance testing and the absence of data on AQC samples or spiked samples makes it is difficult to assess the reliability of data produced. The accepted approach in the UK for the derivation of the accuracy requirements of an analytical system (when monitoring to a particular water quality standard) is described in WRc Report NS30 (Cheeseman et al. 1989)12. NS30 requires a series of tests to be carried out to assess the precision of analysis across the analytical range, over a period of two weeks or longer. The test programme also includes an evaluation of bias (as spiking recovery) and the effect of the sample matrix on the standard deviation of analysis. This approach is now widely accepted as best practice in environmental analysis in Europe and forms the basis of the validation recommended by the Environment Agency. It is consistent with the specifications made by UN/ECE ISO. 13

In order to determine the fate of steroid oestrogens in WwTW, it is essential that reliable analytical methodology is available to determine free steroid oestrogens in crude sewage. The aim of this research was to develop methods for the analysis of free steroid oestrogens in sewage influent including the liquid and solid fractions.

1.1 Project Objectives

The specific objectives of this work programme were:

To develop an analytical method to determine the natural oestrogens, oestrone and 17--oestradiol and the synthetic oestrogen ethinyl oestradiol in sewage influent.

To performance test the methodology to NS30 and to ISO /TR 13530

To validate and report the analytical method as a Standard Operating Procedure (SOP) in a format to be agreed with the steering committee.

To produce a report of the research carried out.

The scope of the method development was:

To select appropriate solid phase extraction sorbents (silica and non-silica) for the extraction of free steroid oestrogens from the liquid phase

To select optimal solvents for elution of analytes from the SPE disks

To develop extraction methodology to determine free steroids in the solid phase

To develop clean-up procedures to remove interferences from sample extracts

To optimise LCMS conditions for analyte resolution


2 Analytical Method Development

2.1 Strategy for method development

Analytical methodology to determine trace organics in environmental samples has a number of components (isolation/extraction, concentration, clean-up, separation and detection/quantification) and each of these aspects needs to be investigated during method development.

Sewage influent is a complex matrix containing high levels of solids and organic matter. The aim of this project was to develop a method, which would measure the concentration of free steroids in both the aqueous and solid phase of the sample. The target for the limit of detection was set at 10 ng l-1 of each individual steroid oestrogen in the total sample.

It was also considered necessary to utilise techniques that could be rapidly deployed and would provide robust and precise methods, and which could be operated using automated laboratory equipment.

2.1.1 Extraction

Extraction of trace organics from water samples containing solids is often carried out by separating the liquid and solid phases and extracting both phases separately using appropriate techniques.

A commonly used technique for the isolation, concentration and purification of trace organics from aqueous samples is solid phase extraction (SPE). Traditionally, SPE sorbents have been packed into disposable cartridges or columns packed with a sorbent. However there are limitations using this approach for large volume samples, such as channelling at high pressures and column blockages due to the presence of solids. Empore (3M) extraction disks provide an alternative SPE format which uses small (8 m) chromatographic particles enmeshed in a network of polytetrafluroethylene (PTFE) fibrils forming a thin particle-loaded membrane. The use of a disk allows higher flow rates due to improved mass transfer kinetics as a result of the smaller particle size sorbent. A method for the determination of free steroid oestrogens in environmental waters using SPE disks has been reported (Kelly, 2000)14. This work demonstrated that C18 solid phase extraction disks could be used for the extraction of free steroids from water samples. However, this method has only been tested with reverse osmosis water although the authors used it to determine steroids in river water and sewage effluent.

Free steroids in the liquid phase of sewage effluent are routinely analysed by WRc-NSF using solid phase extraction (SPE) cartridges. However, as problems with blockages are significantly reduced when an SPE disk is used in place of SPE cartridges, for this work only SPE disks were investigated.

A number of techniques are also available for the extraction of organics from solid matrices. These include Soxhlet extraction and accelerated solvent extraction (ASE). A method for the determination of free steroids in sewage sludge utilising ASE and detection using liquid chromatography mass spectrometry (UKWIR)15 has recently been


developed. It was decided that this method would be used to extract solids from sewage influent and there would be no further investigation of alternative methods (assuming that this protocol was suitable for the solid phase of sewage influent).

2.1.2 Internal standards

The reason for using internal standards is to improve the precision of quantitative analysis. Generally, an internal standard can be classified as one of three types:

A. a stable isotope labelled analogue of the compound(s) to be determined;

B. a homologous compound (which is assumed to behave similarly to the compound of interest);

C. an unrelated compound which behaves similarly (in terms of extraction, clean-up and chromatographic behaviour) to the compound to be determined.

Isotopically labelled internal standards are the preferred choice for analyses which utilise mass spectrometry for detection.

2.1.3 Clean-up

Clean-up of environmental extracts is required to ensure potential analytical interferences or compounds that may degrade chromatographic columns (leading to loss of instrument performance) are removed. In addition a cleaner extract can result in lower detection limits (removal of background results in a higher signal to noise ratio) or (in the case of LCMS) removal of compounds which suppress ionisation again resulting in an improved LOD.

A number of clean-up techniques are available, which can either be carried out using automated (e.g. HPLC, GPC or SEC clean-up) or manual methods (e.g. column chromatography, liquid-liquid partition or SPE).

2.1.4 Analysis

Free steroid oestrogens are polar compounds which can be analysed directly by LCMS or LCMSMS and by GCMS or GCMSMS. Analysis using GCMSMS can be carried out directly but is more usually undertaken after derivatisation of the steroids. However, derivatisation is time consuming and complex to carry out. It was therefore decided that LCMS would be used for this work.

2.2 Sample collection and preparation

Sewage influent was collected from a wastewater treatment works (WwTW) from Southern England. This WwTW serves a population of 14,800 (average flow 3018 l/day) and the input into the works is primarily domestic.

Samples were collected in 5 litre silanized glass bottles. Bottles were silanized after they had been cleaned using a proprietary cleaning agent (e.g. Decon 90) using a solution of dimethyldichlorosilane in dichloromethane. Excess silanising agent was allowed to drain


off, the bottles rinsed with dichloromethane and the solvent allowed to evaporate prior to use. Samples were stored in a fridge below 6 C prior to extraction. Samples were filtered to separate the solid and liquid phase.

2.3 Development of sample extraction technique (Liquid Phase)

2.3.1 SPE Disk evaluation

A number of Empore extraction disks (3M) were investigated for the extraction of free steroids from water samples:

Empore Disk with C18 sorbent

Empore Disk with a C8 sorbent and

Empore Disk with a styrenedivinylbenzene (SDB-XC) sorbent

Kelly (2000)Error: Reference source not found showed that Empore C18 disks can be used to extract free steroids from water samples. The aim in this present work was to confirm that the conditions used were satisfactory and to compare the extraction efficiency of other sorbents and to optimise the elution and “wash” solvents.

A number of solvents and solvent mixtures were selected for evaluation for use as either wash or elution solvent:

Non-polar solvents including hexane, chloroform, cyclohexane

Methanol and acetone

Aqueous methanol mixtures i.e. 85%, 80%, 60%, 50%, 45%, 40%, 30% and 20% v/v methanol in water

In order to achieve shorter extraction times 100 ml aliquots of groundwater were fortified with the same quantity of each analyte (100 ng l-1). The disks were conditioned with 10 ml of the “elution” solvent, followed by 20 ml methanol and 20 ml of blank water. The spiked water sample was loaded onto the disk at a flow rate of approximately 20ml min-1. After extraction the disk was left under vacuum for 10 minutes and eluted slowly with the “elution solvent” (2 x 20 ml) into a silanized glass collection tube. The extracts were combined and concentrated to a final volume of 200 l using a TurboVap LV ASE concentrator (Zymark Corporation) prior to analysis using LCMS.


Figure 2.1 Showing recovery of free steroid oestrogens using a number of Empore solid phase extraction disks and elution solvents.


6

The results obtained from this study showed that non-polar solvents provided poor recoveries of the analytes. All three disks (C8, C18 and SDVB) provided good recoveries of free steroids using methanol as the elution solvent. Aqueous methanol mixtures used for elution also provided high recovery of free steroids from aqueous samples with 100-85% aqueous methanol solutions providing >90% recovery. The work also showed (see Figure 2.1) that the optimal “wash” solvent for the cartridges was 30% methanol:water, with higher methanol concentrations leading to breakthrough of analytes.

The best recoveries were achieved with C18 disks, therefore these disks were selected for method testing. It was decided that 30% methanol:water would be used as the “wash” solvent and 100% methanol used for elution solvent.

2.3.2 Amino propyl clean-up

Aminopropyl cartridges were investigated for the removal of interferences from extracts. Isolute 500 mg/3 ml aminopropyl SPE cartridges were evaluated. In order to determine the most appropriate elution solvent, standard solutions of free steroids in hexane were applied to the aminopropyl column after it had been primed with hexane (5 ml). The loaded solvent was collected in a 10 ml silanized test-tube. Concentration and analysis of the contents of the collection vessel showed that about 20% of the analytes were present in the eluent.

The SPE cartridges were then eluted with different mixtures of ethyl acetate in hexane and acetone in ethyl acetate to recovery the free steroids from the SPE cartridges. The recoveries of analytes are given in Figure 2.2 which shows that 73-87% of the analyte can be recovered using 50% ethyl acetate in acetone, and when combined with the steroids present in the loading solvent, near complete recoveries of the analytes can be achieved.

Figure 2.2 Recovery of free steroid oestrogens from an aminopropyl clean-up using various elution solvents

2.3.3 Application of the method to sewage influent

After the establishment of the extraction and clean-up procedures using groundwater as described above, the methodology was applied to the determination of free steroids in sewage influent. The methodology is described in detail in Appendix A.


Briefly, sewage influent (6 litres) was filtered to remove solids and the liquid phase was split into six 1 litre aliquots. Three of the six aliquots were spiked with free steroids at 10 ng l-1

and the remaining three were unspiked. All aliquots were spiked with the labelled internal standards and were extracted after conditioning the SPE disks. The disks were eluted with methanol and the extracts were concentrated to 200 l using a TurboVap LV ASE (Zymark Corporation), prior to analysis using LCMS.

At this stage (prior to clean-up) the results obtained show an elevated background in the LCMS chromatograms (Appendix B, Figure B1) and all extracts were also highly coloured, indicating co-extraction of other materials.

Interferences in the methanol extracts were removed using aminopropyl cartridges. The methanol extract was made up to 2 ml in hexane. The aminopropyl cartridge was conditioned with hexane (2 ml) and the extract added to the cartridge. After adding the extract to the cartridge the tube was rinsed with a further 1 ml of hexane which was also added to the cartridge. The cartridge was then washed with 30 % ethyl acetate in hexane (2 x 2 ml) and eluted with 50% ethyl acetate in acetone (2 x 2ml). The extracts were then concentrated to dryness and made up to 100 l in methanol prior to analysis using LCMS.

After clean-up the colour in the extract was removed (retained by the aminopropyl cartridge) and the extract appeared clear. The resulting LCMS chromatogram showed a lower background signal and peaks corresponding to the spiked internal standards and free steroids were visible (Appendix B, Figure B2).

The concentrations of free steroids in sewage influent and spiked sewage influent using C18 disks and aminopropyl clean-up is given in Table 2.1 below. The results obtained showed that the method development was successful and this method was therefore performance tested.

Table 2.1 Concentration of free steroids in the liquid phase of spiked (10 ng/l) and unspiked sewage influent

Sample Concentration ng l-1

Ethinyl oestradiol Oestradiol Oestrone

Sewage influent unspiked (1) <10 37 128



Sewage influent spiked (1) 10 50 145



2.4 Development of sample extraction technique (Solid Phase)

Sewage influent was collected (Section 2.2) and filtered. The solid phase was freeze dried and extracted using accelerated solvent extraction as described below. The methodology described is based on a published method for the analysis of steroids in sewage sludge (UKWIR 2003).15


2.4.1 ASE extraction and C18 Clean-up

The filtered solids were freeze dried and an aliquot of the dried sample (0.5 g) was spiked with free steroids. The sample was ground and mixed with an equal proportion of Hydromatrix and placed into an ASE stainless steel extraction cell and extracted using methanol.

ASE extracts in methanol (40 ml) were diluted with an equal volume of blank reagent water. Isolute C18 non-endcapped cartridges (500 mg, 6ml) were conditioned with methanol followed by 50% aqueous methanol. The sample was passed through the cartridge at less than 10 ml min-1. The cartridge was then dried using nitrogen and eluted with 2 x 5ml methanol. The methanol extract was diluted with water (90 ml) to produce a 10% aqueous methanol solution. An Isolute C18 cartridge (500 mg; non-endcapped) was conditioned with methanol followed by 10% aqueous methanol. The sample was applied to the cartridge, the cartridge dried and eluted with 85% aqueous methanol. The extract was concentrated to 100 l using nitrogen blow-down and analysed using LCMS.

The results obtained showed poor recovery of the analytes. Inspection of the original sample and the freeze dried solids showed that they contained a large proportion of fibrous materials. Further work was required to remove these interferences from the final extract.

2.4.2 Amino-propyl Clean-up

Due to the problems encountered with the clean-up methodology used above, it was decided that the aminopropyl cartridge method used for clean-up of the liquid phase would be investigated for the solid phase. Solids were extracted using ASE as above and the methanol extracts concentrated to 200 l prior to clean-up using the aminopropyl method previously described.

The samples cleaned-up using this procedure showed lower levels of solids in the final extract than when using the C18 clean-up, and a number of samples were extracted to determine the reproducibility of the method. The results obtained showed poor reproducibility and this was attributed to interferences or ion suppression during LCMS analysis. Therefore it was decided that the extracts required further clean-up prior to analysis. The two clean-up methods (C18 and aminopropyl) were applied in series. The resulting extracts produced data that was consistent with the expected spiked concentrations.

3 Analytical Method validation

3.1 Method validation (Liquid Phase)

Sewage influent was collected from a sewage treatment works (post primary screen) in a 25 litre stainless steel container that was thoroughly rinsed with blank water followed by the sample, prior to collection. The collected sample was transported to the laboratory where it was immediately filtered and the filtered sample placed in a cold room at 6 C.

The filtered samples (1 litre) were spiked with free steroids as follows:

1. “Blank” - sewage influent unspiked


2. “High Spike” - sewage influent spiked with 70 ng l-1 (oestradiol) and 100 ng l-1 (ethinyl oestradiol)

3. “Low Spike” – sewage influent diluted ten times with blank borehole water and spiked with 7 ng l-1 (oestradiol) and 10 ng l-1 (ethinyl oestradiol)

The intention of the spikes was to obtain a water sample containing 100 ng l-1 of each free steroid in the high spike sample and 10 ng l -1 in each low spike sample. Each analytical batch consisted of two “high spikes”, two “low spikes” and a blank sample. Eleven batches of samples were extracted over the course of five days. Sample extraction, concentration and analysis were carried out using the procedure described in Appendix A.

The results obtained from the performance testing are given in Appendix A.

Based on the performance data of the “low spike” (nominal 10 ng l -1) the limit of detection for the three free steroids 17--oestradiol, ethinyl oestradiol and oestrone in sewage influent were 2.6, 11.4 and 5.2 ng l-1 respectively.


Table 3.2 Performance Data for oestradiol in sewage influent (liquid phase)

Batch Concentrationng l-1

Batch Mean Result

Estimate of within-batch

Standard deviation


variance

1 3.48 3.49 0.01 0.003.50

2 3.33 3.22 0.16 0.023.11

3 3.18 3.07 0.16 0.032.95

4 3.45 3.10 0.50 0.252.74

5 2.56 2.61 0.07 0.012.66

6 2.81 3.18 0.52 0.273.55

7 3.46 4.16 0.99 0.984.86

8 2.90 2.90 0.00 0.002.90

9 2.57 1.88 0.98 0.951.19

10 3.42 3.59 0.24 0.063.76

11 5.05 4.61 0.62 0.394.17

Mean 3.25M1 1.17M0 0.27

F Observed 4.35

Sw 0.52Sb 0.67St 1.14Rel SD(St) 24.7%Est Deg F 14

Limit of Detection 2.6 ng l-1

Table 3.3 Performance data for ethinyl oestradiol in sewage influent (liquid phase)


Batch Concentrationng l-1

Batch Mean Result


Standard deviation


variance

1 7.35 7.41 0.08 0.017.47

2 2.56 5.83 4.62 21.399.10

3 5.66 6.52 1.22 1.487.38

4 8.91 6.92 2.81 7.924.93

5 5.01 7.31 3.25 10.539.60

6 6.67 6.11 0.80 0.645.54

7 4.47 3.37 1.56 2.422.27

8 2.26 2.26 0.00 0.002.26

9 5.31 3.67 2.32 5.382.03

10 2.78 4.00 1.72 2.955.21

11 6.68 5.54 1.62 2.624.39

Mean 5.36M1 15.32M0 5.03

F Observed 3.04



Table 3.4 Performance data for oestrone in sewage influent (liquid phase)

Batch Concentration Batch Mean Estimate of Estimate of


ng l-1 Result within-batch Standard deviation

within-batch variance

1 4.19 4.28 0.13 0.024.37

2 3.76 4.01 0.35 0.134.26

3 4.41 4.49 0.11 0.014.57

4 4.10 4.44 0.47 0.224.77

5 3.84 3.81 0.04 0.003.78

6 4.10 4.42 0.45 0.204.74

7 4.91 5.48 0.81 0.656.05

8 8.25 10.45 3.11 9.6812.65

9 6.87 6.43 0.62 0.395.99

10 4.80 4.77 0.05 0.004.73

11 7.83 7.48 0.50 0.257.12

Mean 5.46M1 8.57M0 1.05

F Observed 8.16




3.2 Method validation (Solid Phase)

Further sewage influent samples were collected and filtered to collect significant quantities of solids in order to carry out performance testing of the methodology to determine free steroids.

The solid phase was freeze dried and the dried samples (0.5 g) were extracted after spiking as follows.

One portion was unspiked to determine levels of natural steroids in the sample and labelled as the blank. The other four samples were spiked with a mixed solution of the three free steroids: oestrone, 17--oestradiol and ethinyl oestradiol. Two samples were spiked at a “low” concentration and two samples were spiked at a “high” concentration. The low spikes were spiked at 50 ng g-1 and the high spikes were spiked at 200 ng g-1.

The spiked samples were mixed using a metal spatula and allowed to stand overnight at room temperature to allow equilibrium with the matrix to take place.

The spiked and dried samples were ground and mixed with an equal proportion of Hydromatrix. The samples were placed into an ASE stainless steel extraction cell and spiked with a mixture of labelled internal standards. The void at the top of the extraction cell was filled using more Hydromatrix. ASE extraction and clean-up was carried out as described in Appendix A.

The results obtained from the performance testing are given in Appendix A.

Based on the performance data of the “low spike” (nominal 10 ng l -1) the limit of detection for the three free steroids 17--oestradiol, ethinyl oestradiol and oestrone in sewage influent were in the 65, 32 and 98 ng g-1 respectively.


Table 3.5 Performance data for oestradiol in sewage influent (solid phase)

Batch Concentrationng g-1

Batch Mean Result


Standard deviation


variance

1 42.71 37.97 6.71 45.0333.22

2 21.30 15.14 8.72 76.018.97

3 33.87 41.08 10.20 103.9748.29

4 84.55 71.12 19.00 361.0057.68

5 43.68 42.61 1.51 2.2941.54

6 21.27 21.41 0.20 0.0421.55

7 8.40 24.56 22.85 522.2940.72

8 43.76 26.18 24.86 618.118.60

9 9.80 8.30 2.12 4.506.80

10 12.60 15.60 4.24 18.0018.60

11 35.20 37.75 3.61 13.0140.30

Mean 31.06M1 2136M0 160.4

F Observed 13.32


Limit of Detection 64.5 ng g-1

Table 3.6 Performance data for ethinyl oestradiol in sewage influent (solid phase)



Batch Mean Result


Standard deviation


variance

1 55.34 59.73 6.21 38.5464.12

2 23.07 18.41 6.59 43.4313.75

3 56.18 57.66 2.09 4.3559.13

4 17.81 27.76 14.06 197.8137.70

5 6.39 6.42 0.04 0.006.45

6 19.49 24.20 6.65 44.2728.90

7 32.48 32.51 0.04 0.0032.54

8 34.94 28.58 9.00 81.0322.21

9 32.50 34.10 2.26 5.1235.70

10 10.80 11.65 1.20 1.4512.50

11 32.20 33.85 2.33 5.4535.50

Mean 30.44M1 726.31M0 38.31

F Observed 18.96



Table 3.7 Performance data for oestrone in sewage influent (solid phase)


Batch Mean Result




Standard deviation

variance

1 17.99 49.28 44.25 195880.57

2 103.84 96.57 10.28 10589.30

3 31.22 44.91 19.36 37458.60

4 10.81 35.56 35.00 122560.31

5 54.14 47.89 8.84 78.1341.64

6 64.69 57.07 10.78 116.2849.44

7 42.00 48.57 9.28 86.2055.13

8 43.71 41.86 2.62 6.8840.00

9 42.00 43.65 2.33 5.4545.30

10 42.90 49.25 8.98 80.6555.60

11 45.50 44.05 2.05 4.2142.60

Mean 50.79M1 5655M0 367.42

F Observed 15.39



4 Conclusion

An analytical method has been developed to determine the natural oestrogens, oestrone and 17--oestradiol and the synthetic oestrogen ethinyl oestradiol in sewage influent.


During method development a number of extraction disks and elution solvents and clean-up sorbents were investigated. Octadecylsilyl (C18) disks were found to provide acceptable recoveries of the free steroids from the liquid phase. The solids were extracted with methanol using accelerated solvent extraction (ASE). Interferences were removed from the extracts using C18 and aminopropyl SPE cartridges.

SPE extraction using disks provides a number of advantages for large volume sample preparation over other extraction techniques such as liquid-liquid extraction or SPE extraction using packed cartridges. SPE disks provide a large surface area for sorbent/sample contact and faster flow rates and higher throughput can be obtained from complex environmental samples such as raw or treated sewage.

Accelerated solvent extraction offers a number of advantages over other solvent extraction techniques, including lower solvent consumption, reduced extraction times, higher throughput through automation and ease of use. Furthermore, when used to extract sewage the resulting extract does not require filtration prior to analysis.

Statistical performance testing of the method shows that the procedure can be used for the routine monitoring of free steroids in sewage influent. The statistically derived limits of detection of the method to determine oestrone, 17--oestradiol and ethinyl oestradiol in sewage influent were in the range 2.6-11.4 ng l-1 in the liquid phase and 31-100 ng g-1 in the solid phase.

The methods presented can be applied to the determination of free steroid oestrogens in raw sewage influent, which will assist in the optimisation and assessment of the removal efficiencies of steroid oestrogens from waste water treatment works.

5 References

1. Purdom, C. E.; Hardiman, P. A. and Bye, V. J (1994) Estrogenic Effects of Effluents from Sewage Treatment Works. Chem. Ecol., 8, 275-285.

2. Stumpf, M.; Ternes, T. A.; Haberer, K.; Baumann, (1996) Determination of Natural and synthetic Estrogens in Sewage Plants and River Water, W. Vom. Wasser., 87, 251-261.

3. Ternes, T. A.; Stumpf, M.; Müller, J.; Haberer, K.; Wilken, R. D.; Servos, M. (1999) Behavior and occurrence of estrogens in municipal sewage treatment plants –– I. Investigations in Germany, Canada and Brazil, Sci. Total. Environ., 225, 81-90.

4. Baronti, C.; Curini R.; D’Ascenzo, G.; Di Corca, A.; Gentili, A.; Samperi, R. (2000) Monitoring Natural and Synthetic Estrogens at Activated Sludge Sewage Treatment Plants and in a Receiving River Water Environ. Sci. Technol., 34, 5059-5066.

5. Snyder, S. A.; Keith, T. L.; Verbrugge, D. A.; Snyder, E. M.; Gross, T. S.; Kannan, K.; Giesy, J. P. (1999) Analytical Methods for Detection of Selected Estrogenic Compounds in Aqueous Mixtures Environ. Sci. Technol., 33, 2814-2820

6. Desbrow, C., Routledge, E.J., Brighty, G.C., Sumpter, J.P. and Waldock, M. (1998) Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environmental Science and Technology, 32, 1549-1558


7. Johnson, A.C, Williams, R.J. and Ulahannan, T (1999) Comment on ‘Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environmental Science and Technology, 33(2), 369-370.

8. Johnson, A.C., Belfroid, A. and DiCorcia, A. (2000) Estimating steroid oestrogen inputs into activated sludge treatment works and observations on their removal from the effluent. Sci. Total. Environ., 256, 163-173.

9. Ternes, T.A., Stumpf, M., Mueller, J., Haberer, K., Wilken, R.-D. and Servos, M. (1999) Behaviour and occurrence of estrogens in municipal sewage treatment plants - I. Investigations in Germany, Canada and Brazil. Sci. Total. Environ., 225, 81-90.

10. Solé, M., López De Alda, M.J., Castillo, M., Porte, C., Ladegaard-Pedersen, K., and Barceló, D (2000). Estrogenicity determination in sewage treatment plants and surface waters from the Catalonian area (NE Spain). Environmental Science and Technology, 34: 5076-5083.

11. Gentili, A. Perret, D. Marchese, S. Mastropasqua, R. Curini, R. Di Corcia, A. (2002) Analysis of Free Estrogens and their Conjugates in Sewage and River Waters by Solid-Phase Extraction then Liquid Chromatography-Electrospray-Tandem Mass Spectrometry. Chromatographia, 56, 25-32

12. Cheeseman, R.V., Gardner, M.J. & Wilson, A.L. (1989) A manual on analytical quality control for the water industry WRc Rept NS30, ISBN 0902156853

13. Timmerman, J, Gardner, M.J and Ravenscroft J.E UN/ECE Task Force on Monitoring and Assessment, Volume 4 - Quality Assurance. ISBN 9036945860.(1996). ) ISO /TR 13530 (1998) Water Quality - Guide to Analytical Quality Control for Water Analysis

14. Kelly, C (2000) Analysis of steroids in environmental water samples using solid-phase extraction and ion-trap gas chromatography-mass spectrometry and gas chromatography-tandem mass spectrometry

15. UKWIR (2003) Endocrine Disrupters in Sewage Sludge: Analytical Method Development, UKWIR Report 03/TX/04/7 (ISBN 1 84057 307 4)



APPENDIX A STANDARD OPERATING PROCEDURE FOR THE DETERMINATION OF FREE STEROID OESTROGENS IN SEWAGE INFLUENT

1 SCOPE

This document outlines the procedure for the determination of oestrone, 17-oestradiol and 17-ethinyl oestradiol in sewage influent using solid phase extraction disks for extraction of the liquid phase, accelerated solvent extraction (ASE) for extraction of the solid phase and liquid chromatography mass spectrometry (LCMS) for detection and quantification.

2 PRINCIPLE

Samples are collected in silanized glass bottles and the sample is filtered to separate the liquid and solid phase. Prior to extraction, the solids are freeze dried, spiked with isotopically labelled internal standards and extracted using ASE. The liquid phase is extracted using C18 SPE disks after spiking the sample with labelled internal standards. The extract is cleaned-up using solid phase extraction (SPE) cartridges. The extracts are analysed using LCMS operated in negative ion electrospray mode using selected ion monitoring.

3 TERMS, DEFINITIONS AND SYMBOLS

For the purposes of this standard operating procedure, the following terms, definitions and symbols apply:

3.1 Analyte

The three free steroid oestrogens determined using this method: oestrone, oestradiol and ethinyl oestradiol.

3.2 Calibration standard

A solution prepared from a secondary standard and/or stock solutions and used to calibrate the response of the instrument with respect to the analyte concentration.

3.3 Calibration verification standard or continuing calibration check

A mid point calibration standard that is used to verify calibration.

3.4 Internal Standard Spiking

The addition of d4-labelled steroid oestrogens. The recovery of these standards is used to correct values of native analytes of interest.

3.5 Statistical Performance Characteristics

Quantification for measured values of the possible deviations resulting from the random part of the measuring process; e.g. repeatability or instability [ISO 6879:1995]


3.6 Method Blank

A blank water sample that has been treated exactly as a sample through the complete analytical procedure including extraction, clean-up, identification and quantification including all the relevant reagents and materials.

3.7 ASE

Accelerated Solvent Extraction.

3.8 LCMS

Liquid Chromatography Mass Spectrometry.

3.9 MDL

Method detection limit.

4 PERFORMANCE CHARACTERISTICS

See Annex A.

5 HAZARDS

Hazard assessments should be carried out for all of the chemicals and procedures used and should be consulted prior to carrying out any work with the chemicals involved.

All of the steroids determined are oestrogenically active, and it must be assumed that the internal standards are also active. Appropriate precautions should be taken when handling the pure compounds and standard solutions of these compounds.

Several of the reagents used are potentially hazardous. Methanol and acetonitrile are toxic and flammable.

6 APPARATUS

6.1 General

Pasteur pipette

Analytical balance (4 place)

Volumetric flasks (range between 5 ml and 50 ml)

Range of glass syringes (between 10 µl and 1 ml)

6.2 Freeze drying apparatus

6.3 Accelerated solvent extraction

Dionex ASE 200 Accelerated Solvent extractor with 33 ml stainless cells.

Dionex vials for collection of extracts (60 ml)


Dionex ASE cell cellulose filter (P/N 049458)

6.4 Solid Phase extraction apparatus

Vacuum manifold to which several solid phase extraction cartridges can be attached via 2 litre conical flasks. The flow rate through each individual cartridge is controlled by adjusting the vacuum applied to each one.

3M Empore Vacuum Manifold Station - 6 station for 47 mm disks - 3M Part #206-47

6.5 Extract Concentration Equipment

Test Tubes (10ml)

TurboVap concentrator with thermostatically controlled water bath.

6.6 LCMS

Liquid Chromatograph (LC) conditions:

LC: Agilent 1100 system with autosampler, binary pump, degasser and column heater.

Column: Phenomenex LUNA 5u C18(2), 150 x 4.60 mm (P/N 00F-4252-E0)

Flow: 0.5 ml/min

LC Eluents: Solvent A: Water containing 0.1% ammonia (3 ml of 30% ammonium hydroxide per litre)

Solvent B: Methanol

LC Gradient:

Time 0 min 10 min 18 min 24 min 28 min

%A 90% 50% 10% 10% 90%

Injection volume: 20 l

Mass Spectrometer (MS): Micromass Quattro LC

Source: Electrospray (negative ion)

Source Temp: 150 C

Desolvation Temp: 400 C

Capillary voltage: 3.1 kV


Cone Voltage: 45 V

Nebuliser: 75 litres/hour

Drying: 535 litres/hour

Acquisition mode: Selected ion monitoring (SIM)

The system performance should be optimised by injecting a mixed tuning solution containing the three free steroids in methanol. The mixed standard (500 l of 10 ng l-1) is introduced using a direct injection loop.

Ions monitored:

269.30 (oestrone)

273.30 (d4-oestrone)

271.30 (17-oestradiol)

275.30 (d4-17-oestradiol)

295.25 (17-ethinyl oestradiol)

299.25 (d4-17-ethinyl oestradiol)

The retention times of the free steroids are about 19-21 minutes and the data system is programmed to monitor over the time period 15-30 minutes following the injection of the extract.

7 REAGENTS

All reagents must be of sufficient purity that they do not give rise to significant interfering peaks in the analysis. Purity must be checked for each batch of materials by the running of procedural blanks with each batch of samples analysed.

7.1 Standards and internal standards.

Oestrone, 17-oestradiol and 17-ethinyl oestradiol are available as crystalline solids from various sources. The internal standards used are 2,4,16,16-d4-oestrone, 2,4,16,16-d4-17-oestradiol and 2,4,16,16-d4-17-ethinyl oestradiol. These are available from CDN isotopes (UK agents Univar plc, K&K-Greeff Ltd, Suffolk House, George Street, Croydon CR9 3QL).

7.2 Silanising solution

A solution (10% v/v) of dimethyldichlorosilane in dichloromethane is used to deactivate the surface of all glassware which comes into contact with the samples and extracts.

Silanising solution should preferably be made up as required, but excess solution can be stored for short periods (up to 1 week).


7.3 Solvents

Methanol - Rathburns HPLC grade or equivalent

Acetonitrile – Rathburn silylation grade

Deionised water

7.4 Hydromatrix

7.5 Solid phase cartridges

500 mg/6 ml BondElute C18 non-end-capped solid phase extraction cartridges,

500 mg/6 ml Isolute aminopropyl solid phase extraction cartridges

7.6 Solid phase disks

3M Empore C18 disks 47 mm 3M Part#2215

8 STANDARD SOLUTIONS

8.1 Internal Standard Stock Solutions, 1mg/ml

Dissolve 10mg of each compound into a 10ml volumetric flask containing acetone and then make up to the mark. The solution codes are given below.

Compound Code

d4-oestrone d4-ESTN-ST-Ax

d4-17-oestradiol d4-ESTRA-ST-Ax

d4-17-ethinyl oestradiol d4-EEST-ST-Ax

These solutions are stable for one year when stored in a freezer at –18 C.

8.2 Internal Standard Intermediate solution, 10ng/ul (d4-STR-INT-Ax)

Add 100µl of each stock solution into a 10ml volumetric flask containing acetone, and then make up to the mark.

This solution is stable for 1 year if stored in a freezer –18 C.

8.3 Steroid Standard Stock Solutions, 1 mg/ml

Dissolve 10 mg of each compound into a 10 ml volumetric flask containing acetone, then make up to the mark. The solution codes are given below.

Compound Code

oestrone ESTN-ST-Ax


17-oestradiol ESTRA-ST-Ax

17-ethinyl oestradiol EEST-ST-Ax

These solutions are stable for one year when stored in a freezer.

8.4 Steroid Standard Intermediate solution, 1 ng/ul (STR-INT-Ax)

Add 50 µl of each stock solution to a 50 ml volumetric flask containing acetone, then make up to the mark.

This solution is stable for 1 year if stored in the freezer.

9 SAMPLING

Sample bottles should be made of glass. Plastic screw tops are suitable provided they are fitted with PTFE or PTFE-faced liners. Alternatively, ground-glass stoppered glass bottles may be used. Prior to use, the bottles should be cleaned with a suitable proprietary cleaning agent (e.g. Decon 90) and following thorough rinsing, with deionised water, and dried. The bottles should be deactivated with 20-30ml of silanising solution.

Samples should be kept at 6 C and extracted as soon as possible after sampling.

10 PROCEDURE

10.1 Glassware pre-treatment

A solution of a dimethyldichlorosilane in dichloromethane is used to deactivate the surface of all of the glassware that comes into contact with the samples and extracts. Sufficient of this solution (0.5 - 30 ml, depending on the surface area of the glassware involved) is poured into all of the sample bottles, ASE vials and conical flasks and swirled over the surface, ensuring complete coverage. Excess silanising solution is then allowed to drain and disposed of. Finally, the glassware is rinsed with dichloromethane.

10.2 Sample collection and preservation

Samples should be collected in suitable glass bottles (1 litre). Since the steroids to be monitored are not stable in for more than a few days, samples should be analysed as soon as possible following collection. If storage is unavoidable, samples should be kept in a refrigerator at 4-6°C.

10.3 Sample pretreatment

All samples should be filtered prior to extraction of 1 litre of the liquid phase and 0.5 g of the solid phase.

The solids should be freeze dried prior to extraction.

Solid samples are extracted using the procedure given in Section 10.4.

To each liquid sample 100µl of the internal standard solution (d4-STR-INT-Ax) is added, giving a concentration of 250 ngl-1.


Liquid samples are extracted using the procedure given in Section 10.5

10.4 ASE Extraction of solids

10.4.1 Addition of dispersing agent

An aliquot of the freeze-dried sample (0.5 g) is mixed with approximately 7-8 g of Hydromatrix in a small beaker using a spatula (do not use sodium sulfate as this will dissolve in methanol using the applied ASE conditions).

10.4.2 Filling the ASE cell

Insert a disposable cellulose filter in the bottom of the 33 ml stainless steel ASE cell using the cellulose filter insertion tool. Transfer the entire contents of the beaker containing the sample and Hydromatrix into the ASE extraction cell. Press down the contents using the insertion tool.

Add 25 µl of the internal standard solution (d4-STR-INT-Ax) giving a concentration of 50 ng g-1 to each cell. Add 25 µl of the matrix spike (STR-INT-Ax) to the control sample.

Fill the void volume in the cells with Hydromatrix and place another cellulose filter at the outlet of the cell. Screw the top cap onto the cell body and hand tighten. Check the end of each cap to verify that the O-rings are in place and in good condition; replace if necessary.

10.4.3 Loading the cell tray

Load the ASE cells into the tray slots. Hang the cells vertically from their top caps. Also, load the collection tray with the appropriate number of 60 ml pre-cleaned and silanized sample vials and rinse tubes in the required positions.

10.4.4 Rinsing/Priming the system

Fill the solvent reservoir with methanol and set it inside the ASE 200 sample compartment. Press the rinse button to prime and rinse the ASE.

10.4.5 Sample extraction

The samples should be extracted using the following conditions:

Oven Temperature: 100C

Pressure: 10 Mpa (1500 psi)

Oven heat-up time: 5 min

Static time: 5 min

Flush volume: 60% of extraction cell volume

Solvent: Methanol


10.4.6 Post-extraction procedure

Allow the cells to cool for at least 15 minutes, discard the contents and clean the cells.

Remove sample extracts from the ASE and transfer the contents to 150 ml pre-cleaned, silanized conical flasks.

10.5 SPE Extraction of liquids and ASE extracts

Pre-treatment of liquids and ASE extracts:

Filter the ASE extract using methanol rinsed silanised glass wool. Add 160 ml blank water to the ASE extract to obtain a 20% aqueous methanol solution.

No pre-treatment of the liquid fraction is required.

The liquid and ASE extracts are now ready for SPE extraction.

10.5.1 C18 SPE cartridge conditioning

An Isolute C18 cartridge (500 mg; non-end-capped) is conditioned by passing 100% methanol (5 ml), then 20% aqueous methanol (5 ml) through it at a flow rate of less than 10 ml min-1. The flow rate of the sample is controlled by application of a vacuum. Ensure that the cartridge does not dry out during this process or prior to passage of a sample through the cartridge.


The sample is passed through the cartridge at a flow rate of less than 10 ml min-1, the flow rate again being controlled by application of a vacuum.

10.5.3 SPE cartridge drying

After all of the sample has passed through the cartridge, it is left under vacuum for a further 5 minutes, then further dried for 10 minutes using clean nitrogen gas.

10.5.4 Elution of steroids

The steroids are eluted from the cartridge with methanol (2 x 2 ml) at a flow rate of 2 ml min-1 and collected in a 10 ml test-tube. Concentrate the contents to a final volume of 1 ml using a TurboVap concentrator and then to 200 l using a nitrogen-blow down apparatus in an autosampler vial.

10.6 SPE Clean-up

Add 1.8 ml hexane to the 200 l methanol extract (mix vigorously or sonicate to dissolve any solids) prior to clean-up

10.6.1 Aminopropyl SPE cartridge conditioning

An Isolute aminopropyl cartridge (500 mg) is conditioned by passing hexane (2 ml) through it at a flow rate of less than 10 ml min-1. The flow rate of the sample is controlled by


application of a vacuum. Ensure that the cartridge does not dry out during this process or prior to passage of a sample through the cartridge.


The sample extract is passed through the cartridge at a flow rate of less than 10 ml min-1, the flow rate again being controlled by application of a vacuum. Collect the solvent in a 10 ml test tube and remove the collection vessel (do not discard the liquid).

10.6.3 SPE cartridge wash

Wash the cartridge with 30% ethyl acetate in hexane (2 x 2 ml) and dispose of the washings.

10.6.4 Elution of steroids

The steroids are eluted from the cartridge with 50% ethyl acetate in acetone (2 x 2 ml) at a flow rate of 2 ml min-1 and collected in a 10 ml test-tube. Concentrate the contents to a final volume of 1 ml using a TurboVap concentrator and then to dryness using a nitrogen-blow down apparatus, add dryness and add 100 l methanol and transfer the extract to an autosampler vial. Cap the vial and analyse the contents using LCMS.

10.7 LCMS analysis

Optimise the operating conditions of the LCMS system e.g. according to the manufacturers instructions.

LC: HP 1100 SERIES

Column: Phenomenex LUNA 5 m C18(2),

150 x 4.60 mm (P/N 00F-4252-E0)

Flow: 0.5 ml/min

LC Initial: Solvent A: Water containing 0.1% ammonia

Solvent B: Methanol

LC Gradient:

Time 0 min 10 min 18 min 24 min 28 min

%A 90% 50% 10% 10% 90%

Injection volume: 20 ul

MS: Micromass Quattro LC

Source: Electrospray (negative ion)


Source Temp: 150 C

Desolvation Temp: 400 C

Capillary voltage: 3.1 kV

Cone Voltage: 45 V

Nebuliser gas flow: 75 litres/hour

Drying gas flow: 535 litres/hour

Ions monitored: 269.30 (oestrone)

273.30 (d4-oestrone)

271.30 (17-oestradiol)

275.30 (d4-17-oestradiol)

295.25 (17-ethinyl oestradiol)

299.25 (d4-17-ethinyl oestradiol)

11 RESULTS

Using the mass spectrometer software, the area of each specific peak can be measured. For each determinand the response ratio is then calculated.

Pk Area (D)

[Response] = ---------------------------------

Pk Area (I.S.)

where:

Pk Area (D) is the peak area of the determinand

Pk Area (I.S.) is the peak area of the d4-labelled corresponding internal standard

Using the data system attached to the analytical instrument plot the response ratio against the concentration for the standards. From the plotted calibration curve the slope and intercept are determined by linear regression.

By determining the response ratio in the unknown samples, AQC blanks and controls, described above, this can then be applied to the following equation and the concentration of each determinand calculated.

Concentration = [Response - Intercept ] / [Slope]


12 QUALITY CONTROL

The quality of the analysis is assured through reproducible calibration and testing of the extraction, clean-up and LCMS systems. A series of quality control samples should be analysed with each batch of samples and monitored through control charting and other quality review procedures. It is recommended that with every batch of samples extracted, a blank and a spiked control sample (spiked with oestrone, 17-oestradiol and 17-ethinyl oestradiol ) are produced.

Annex A

PERFORMANCE CHARACTERISTICS

A.1. Type of sample: Sewage influent

A.2. Range of application: Between the “MDL” to 1000 ng l-1

A.3. Linear range: This method is linear between 0 – 1000 ng l-1

A.4. Standard deviation and limit of detection:

Sewage Influent - Determined by analysis of eleven batches of duplicates of spiked samples

Steroid Liquid phase spiked at 100 ng l-1

(n=22)

% RSD LOD (ng g-1)

Oestrone 38.1 5.2

17-oestradiol 24.7 2.6

17-ethinyl oestradiol 44.1 11.4

Sewage Influent - Determined by analysis of eleven batches of duplicates of spiked samples

Steroid Solid phase spiked at 100 ng g-1

(n=22)

% RSD LOD (ng g-1)

Oestrone 41.3 97.6

17-oestradiol 62.7 64.5

17-ethinyl oestradiol 55.2 31.5


A.5 Interferences

None identified but any compounds that produce the same mass fragments and elute at the same retention time as the determinands could result in false positives.


APPENDIX B CHROMATOGRAPHS

Figure B1: LCMS Chromatogram of a 1 litre sample of sewage influent extracted using various disks prior to clean-up

SDVB C8 C189256 00:35:5605-Nov-2003E9SDBS

19.40 19.60 19.80 20.00 20.20 20.40 20.60 20.80 21.00 21.20 21.40 21.60 21.80 22.00Time0

100

%

0

100

%

0

100

%

0

100

%

0

100

%

0

100

%

T7A8724 SIR of 6 Channels ES- 299.258.88e4

21.8721.6121.2720.9020.7620.4620.0419.8819.74 20.2520.99

T7A8724 SIR of 6 Channels ES- 295.251.30e5Area

20.769554


21.6821.2220.8720.4819.9019.65 20.02 20.2720.1621.2921.48 21.91


21.04

20.18 20.4621.22

21.75


19.8811523


21.2444843

9256 01:59:0905-Nov-2003E9C8S

19.40 19.60 19.80 20.00 20.20 20.40 20.60 20.80 21.00 21.20 21.40 21.60 21.80 22.00Time0

100

%

0

100

%

0

100

%

0

100

%

0

100

%

0

100

%


21.9120.08 20.9220.45


20.7822632


21.8921.8221.7321.2919.85 21.1521.08

20.9220.5220.0620.85


21.03

19.90 21.22


19.9030158


21.2658545

9256 21:48:5206-Nov-2003E11C18BS

19.40 19.60 19.80 20.00 20.20 20.40 20.60 20.80 21.00 21.20 21.40 21.60 21.80 22.00Time0

100

%

0

100

%

0

100

%

0

100

%

0

100

%

0

100

%


20.11

20.7520.48 21.4720.96


20.8066664


20.082511


21.248377


19.97141711


21.29264000

Figure B2: A 1 litre sewage influent sample extract before and after aminopropyl clean-up and a corresponding LCMS chromatogram showing free steroids and internal detected in the cleaned up extract.

9256

20:10:3107-Nov-2003

E1325C

19. 19. 19. 20. 20. 20. 20. 20. 21. 21. 21. 21. 21. 22.Tim

0

100%

0

100%

0

100%

0

100%

0

100%

0

100%

T7A8784

SIR of 6ChannelsES-

299.252.79e5Area

20.806183

T7A8784

SIR of 6ChannelsES-

295.251.25e5Area

20.807728

T7A8784

SIR of 6ChannelsES-

275.31.31e5Area

19.90;4802

T7A8784

SIR of 6ChannelsES-

273.39.75e5Area21.

2620075

T7A8784

SIR of 6ChannelsES-

271.31.58e5Area

19.948597

T7A8784

SIR of 6ChannelsES-

269.33.97e5Area

21.2845208


Department for Environment, Food & Rural Affairs

Nobel House, 17 Smith Square, London SW1P 3JR

Documents

Standard UKWIR Technical Report Format - Defra, UKrandd.defra.gov.uk/Document.aspx?Document=WT03014_2073... · Web viewEstrogenicity determination in sewage treatment plants and surface