United States Office of Water
Environmental Protection Agency (4303) September 2010
Stability of Pharmaceuticals,
Personal Care Products, Steroids,
and Hormones in Aqueous
Samples, POTW Effluents, and
Biosolids
September 20 10
-------
U.S. Environmental Protection Agency
Office of Water (4303T)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
EPA-820-R-10-008
-------
TABLE OF CONTENTS
ACKNOWLEDGMENT AND DISCLAIMER ii
EXECUTIVE SUMMARY iii
INTRODUCTION 1
EXPERIMENTAL 2
Analytes and Analytical Methods 2
Holding Time and Preservation Conditions 3
Preparation of Samples for the Holding Time Study 4
Spiking Solution 4
Aqueous Samples 6
Chlorinated Samples 6
Biosolid Samples 6
Effluent Samples 6
Sample Containers 7
Procedure for Biosolids Sample Preparation 7
Storage and Analysis of Samples 8
RESULTS 8
Analytical Results 8
Statistical Analysis 9
Statistical Tests 9
Treatment ofNon-Detects 11
Results of Statistical Analysis 12
Container Types 12
Chlorination and Dechlorinating Agents 13
Temperature and Acid Preservation Effects on Effluent Samples 15
Combination of Best Aqueous and Effluent Treatments 17
Effect of Time and Temperature on Storage of Biosolids 19
Extract Stability 22
DISCUSSION 25
CONCLUSIONS 26
REFERENCES 27
-------
ACKNOWLEDGMENT AND DISCLAIMER
This report was prepared by the Engineering and Analysis Division (BAD) of the U.S.
Environmental Protection Agency and has been peer reviewed. BAD is responsible for the
content of this report.
The laboratory work was conducted by AXYS Analytical Services, Ltd.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
Questions concerning this report may be addressed to:
The Clean Water Act Methods Team
U.S. EPA, Engineering & Analytical Support Branch
Engineering and Analysis Division
Office of Science and Technology, Office of Water
1200 Pennsylvania Avenue NW (4303T)
Washington, DC 20460
E-mail: OSTCWAMETHODS@epa.gov
-------
EXECUTIVE SUMMARY
EPA Methods 1694 and 1698 cover pharmaceuticals and personal care products (PPCPs),
steroids and hormones in sewage influent, effluent and sludge. Although researchers have
published several holding time and preservation studies, the studies have been limited in the
number and variety of matrices, chemicals, and preservation techniques tested. Because these
studies were also conducted with a range of different methods, comparing data between studies
is difficult. In this work, EPA describes a study conducted to revise the holding times and
preservation conditions for EPA Methods 1694 and 1698. This study tested a broad number and
variety of chemicals, matrices and preservation techniques under conditions expected in samples
collected in support of Clean Water Act programs.
EPA has tested 89 chemicals that included PPCPs and hormones in reagent water, sewage
treatment plant effluent and biosolids samples. EPA evaluated the effects of chlorine,
dechlorinating agents, container types and temperature on the target chemicals, which included
macrolide antibiotics, quinoline antibiotics, beta-lactam antibiotics, sulfonamide antibiotics,
tetracycline antibiotics, synthetic hormones and other PPCPs. Samples were tested at 0, 7, 14
and 28 days to identify the most promising holding times and preservation conditions. EPA
selected conditions from this initial work to test a second set of samples at 0, 4, 7 and 14 days.
The choice of the most suitable sample bottle varies between individual PPCPs. Higher doses of
chlorine (2 ppm) significantly decreased the concentration of or completely destroyed the
chemicals. The combination of chlorine and dechlorinating agents (sodium thiosulfate or
ascorbic acid) affected many PPCPs including: antibiotics and various over the counter and
prescription medications. However the combination of chlorine and sodium thiosulfate resulted
in more frequent destruction of all classes of antibiotics studied. In the absence of chlorine, the
addition of sodium thiosulfate resulted in nondetects for clinafloxacin, sarafloxacin, fluoxetine,
diphenhydramine. In contrast, the use of ascorbic acid in the absence of chlorine, did not
significantly affect these compounds. The combined effects of lower doses of chlorine (0.5 ppm)
and ascorbic acid in effluent samples resulted in nondetects for tetracyclines, chlorotetracyclines
and tetracycline degradates as well as several other PPCPs including: 17 alpha dihydroequilin,
desogesterol, carbadox, sulfadimethoxine, sulfathiazole, cefotaxime and linomycin. In contrast,
effluent samples not containing chlorine or ascorbic acid had less frequent nondetects.
EPA has used the results of this study to revise the holding times and preservation conditions in
EPA Methods 1694 and 1698 as follows:
• Bottles - amber high density polyethylene or glass containers.
• Dechlorinating Reagent - ascorbic acid.
• Shipping and Storage Conditions - in the dark at <6°C (optional freezing for biosolids).
• Holding Times - extract samples within seven days; analyze extracts as soon as possible,
not to exceed 30 days (10 days in the case of tetracyclines).
These holding times are precautionary, as they protect the most sensitive compounds. They are
not universal holding times for PPCPs or steroids and hormones. It is also important to point out
that these holding times apply only to wastewater samples and not to drinking water samples.
Due to the diverse nature of these chemicals, more holding time studies may be warranted when
new chemicals are targeted.
iii
-------
INTRODUCTION
Contaminants of emerging concern (CECs) originate from many different sources. The chemical
classes that comprise CECs include but are not limited to: flame-retardants, fluorinated alkyl
phenol surfactants, phthalates bis-phenol A, steroids, hormones, pharmaceuticals, and various
personal care products (PPCPs). Over the last ten years, researchers have identified many CECs
in POTW influents, effluents, biosolids and other environmental samples.
There are many published reviews1 covering general topic knowledge and analytical methods as
well as many reports of the occurrence of PPCPs in various types of samples.2 Some of the
earliest reports detailed the occurrence of a variety of PPCPs in surface waters, and in publicly
owned treatment works (POTWs).3'4'5 These include pollutants, such as 17 alpha-ethynyl
estradiol (EE2) which is designed to affect the human endocrine system.
To support occurrence surveys of CECs, researchers have been developing new, and refining
existing analytical methods. These environmental monitoring tools need to work in a wide
variety of sample types, and measure a wide variety of chemicals. An important part of method
refinement is establishing sample collection, preservation, storage and holding time protocols.
Improving our understanding of the stability of CECs in environmental samples is important for
several reasons: researchers are not collecting and handling CEC samples in a uniform way;
few studies have been conducted in sewage samples; and the majority of PPCPs are bioactive
and hence likely susceptible to breakdown by bacteria or other transformation reactions. Some
biologically active compounds, such as illicit substances, may be actively degraded and/or
metabolized by the bacteria found in sewage matrices.6 Other CECs, such as synthetic and
naturally occurring estrogens, can undergo a variety of transformations in the environment and
exist in a number of conjugated forms. For example, 17 alpha-ethynyl estradiol (EE2), 17 beta-
estradiol, estrone and testosterone are susceptible to biodegradation and transformation by
various types of bacteria.7 Humans can metabolize synthetic and naturally occurring estrogens
to form estrogen sulfates and glucuronides which are excreted in urine8 and potentially
hydrolyzed back to their unconjugated forms in wastewaters. 9'10
Although researchers have published several holding time and preservation studies, the studies
have been limited in the number and variety of matrices, chemicals, and preservation techniques
tested.n'12 There are several studies in surface waters; fewer in sewage matrices. In general,
these studies have covered a small subset of CECs compared to the larger set monitored in
published occurrence studies. These studies have examined a relatively narrow range of sample
handling and preservation conditions. For example, the role of containers in sample preservation
has not been reported. Our study tested a broader number and variety of PPCPs, steroids,
hormones, matrices and preservation techniques, and under conditions expected in samples
collected in support of Clean Water Act (CWA) programs.
This report describes the experimental design, results, and findings from a comprehensive
investigation of sample collection and handling procedures for 89 chemicals in reagent water,
POTW effluent and sewage samples. The conclusions from this study will be used to revise
sampling, storage and holding times in EPA Methods 1694 and 1698.
-------
EXPERIMENTAL
The study consisted of four phases, all of which used the same analytical conditions (Table 1)
and the same target list of analytes (Table 2). In Phase 1, storage containers were tested. In
Phase 2, several combinations of preservation techniques were tested to identify the most
promising holding times and preservation conditions in aqueous samples. In Phase 3, several
temperatures were tested for the preservation of biosolids samples. In the first three phases,
samples were tested at 0, 7, 17 and 28 days. In Phase 4, EPA selected "best combination"
conditions from Phase 1 to test a second set of samples at 0, 4, 7 and 14 days. The study design
and the distribution of replicate samples are described in Table 3.
Analytes and Analytical Methods
There is no uniform or consensus list of PPCPs, steroids or hormones to include in occurrence
surveys of POTW or environmental samples. As our understanding of the occurrence,
persistence and toxicity of these compounds increases, the target lists will be refined both by
elimination of some compounds that are not commonly found, and by addition of others.
In 2007, EPA published an LC/MS/MS method (Method 1694), which was used in this study to
test 72 PPCPs.13a This method was modified to include 17 of the steroids and hormones in EPA
Method 1698. Together these methods defined the target list of 89 chemicals, comprised of six
subsets of analytical run conditions (Tables 1 and 2.)
Table 1. Summary of Method/Analyte Categories
Category
Method 1694 List 1
Method 1694 List 2
Method 1694 List 3
Method 1694 List 4
Hormones List 1
Hormones List 2
No. of Compounds
48
14
6
4
10
7
Extraction
Acidic
Acidic
Acidic
Basic
Acidic
Acidic
LC/MS/MS Mode
ESI +
ESI +
ESI-
ESI +
ESI +
ESI-
Solids were extracted by sonication in buffered acetonitrile and exchanged into water. Solid
extracts and water samples were extracted by solid phase extraction and analyzed by LC/MS/MS
according to the scheme in Table 1. Three to four bottles of water, effluent, or biosolid were
prepared for each sample. Therefore, the multiple extractions required for analysis of each
sample were performed on individual sample aliquots. The target compounds are listed in Table
2 according to a compound classification system that is used in the discussion of results.
-------
Table 2. Analyte Categories Used During This Study
Macrolide Antibiotics -Acidic Extraction ESI+
Compound Compound Classification
Azithromycin Macrolide Antibiotic
Clarithromycin Macrolide antibiotic
Erythromycin-H2O Macrolide antibiotic
Roxithromycin Macrolide antibiotic
Tylosin Macrolide antibiotic
Virginiamycin Macrolide antibiotic
Quinoline Type Antibiotics -Acidic Extraction ESI+
Compound Compound Classification
Ciprofloxacin Quinoline antibiotic
Clinafloxacin Quinoline antibiotic
Lomefloxacin Quinoline antibiotic
Norfloxacin Quinoline antibiotic
Ofloxacin Quinoline antibiotic
Enrofloxacin Quinolone antibiotic
Flumequine Quinolone antibiotic
Oxolinic Acid Quinolone antibiotic
Carbadox Quinoxaline Antibiotic
Sarafloxacin Fluoroquinolone antibiotic
beta-Lactam & Misc Antibiotics - Acidic Extraction ESI+
Compound Compound Classification
Trimethoprim Pyrimidine antibiotic
Lincomycin Lincosamide antibiotic
Cefotaxime Cephalosporin antibiotic
Cloxacillin U-Lactam antibiotics
Oxacillin U-lactam antibiotics
Penicillin G U-Lactam antibiotics
Penicillin V U-Lactam antibiotics
Sulfonamide Antibiotics -Acidic Extraction ESI+
Compound Compound Classification
Sulfachloropyridazine Sulfonamide antibiotic
Sulfadiazine Sulfonamide antibiotic
Sulfadimethoxine Sulfonamide antibiotic
Sulfamerazine Sulfonamide antibiotic
Sulfamethazine Sulfonamide antibiotic
Sulfamethizole Sulfonamide antibiotic
Sulfamethoxazole Sulfonamide antibiotic
Sulfanilamide Sulfonamide antibiotic
Sulfathiazole Sulfonamide antibiotic
Misc PPCPs List 1 -Acidic Extraction ESI-
Compound Compound Classification
Gemfibrozil Antilipemic
Ibuprofen Analgesic
Naproxen Non-steroidal anti-inflammatory drug
Triclocarban Antimicrobial, disinfectant
Triclosan Antimicrobial, disinfectant
Warfarin Anticoagulant
Misc PPCPs List 2 - Basic Extraction ESI+
Compound Compound Classification
Albuterol Antiasthmatic
Cimetidine H2-receptor antagonist
Metformin Anti-diabetic drug
Ranitidine Anti-acid reflux
Misc PPCPs List 3 - Acidic Extraction ESI+
Compound
1 ,7-Dimethylxanthine
Acetaminophen
Caffeine
Carbamazepine
Ormetoprim
Codeine
Cotinine
Dehydronifedipine
Digoxigenin
Digoxin
Diltiazem
Diphenhydramine
Fluoxetine
Miconazole
Norg estimate
Thiabendazole
Tetracyclines, Chlorotetracyclines &
Compound
4-Epianhydrochlortetracycline (EACTC)
4-Epianhydrotetracycline (EATC)
4-Epichlortetracycline (ECTC)
4-Epioxytetracycline (EOTC)
4-Epitetracycline (ETC)
Anhydrochlortetracycline (ACTC)
Anhydrotetracycline (ATC)
Chlortetracycline (CTC)
Demeclocycline
Doxycycline
Isochlortetracycline (ICTC)
Minocycline
Oxytetracyclin (OTC)
Tetracycline (TC)
Compound Classification
Antispasmodic, caffeine metabolite
Antipyretic, Analgesic
Stimulant
Anticonvulsant
Diaminopyrimidine
Opiate
Nicotine metabolite
Nifedipine metabolite
Immunohistochemical Marker Steroid
Cardiac glycoside
Antihypertensive
Antihistamine
SSRI Antidepressant
Antifungal agent
Hormonal contraceptives
Fungicide and parasiticide
Degradates - Acidic Extraction ESI+
Compound Classification
Chlorotetracycline degradate
Chlorotetracycline degradate
Chlorotetracycline degradate
Oxytetracycline degradate
Tetracycline degradate
Chlorotetracycline degradate
Chlorotetracycline degradate
Tetracycline antibiotic
Tetracycline antibiotic
Tetracycline antibiotic
Chlorotetracycline degradate
Tetracycline antibiotic
Tetracycline antibiotic
Tetracycline antibiotic
Steroids & Hormones -Acidic Extraction ESI+ and ESI-
Compound
Allyl Trenbolone
Androstenedione
Androsterone
Desogestrel
Mestranol
Norethindrone
Norgestrel
Progesterone
Testosterone
Estriol
17 alpha-Dihydroequilin
17 alpha-Estradiol
17 alpha-Ethynyl-Estradiol (EE2)
17 beta-Estradiol
Equilenin
Equilin
Estrone
Compound Classification
Sex Hormone
Anabolic agent
Hormone metabolite
Ovulation inhibitor
Ovulation inhibitor
Ovulation inhibitor
Ovulation inhibitor
Sex Hormone
Sex Hormone
Sex Hormone
Sterol
Sex Hormone
Ovulation inhibitor
Sex Hormone
Hormone replacement
Hormone replacement
Sex Hormone
Holding Time and Preservation Conditions
This study was designed to investigate the stability of PPCPs, steroids and hormones in reagent
water and POTW effluent and biosolid matrices. The parameters investigated included sample
containers (glass, silanized glass and high density polyethylene [HDPE] bottles), storage
conditions (-20 °C or 4 °C), the presence of residual chlorine, addition of dechlorinating agent
(sodium thiosulfate or ascorbic acid) and time (from Day 0 to up to 28 days after preparation).
Statistical relevance of the results was evaluated by the use of replicate samples for each set of
conditions. All samples were spiked with known amounts of the target analytes at the beginning
of the study to provide concentrations of all target compounds that could be reliably quantified
by the analytical methods.
-------
Table 3. Holding Time Study Treatments, Conditions, and Parameters
Treatment and Condition
Sample
Bottle
Temp
Treatment
Number of Replicates
DayO
Day 2
Day 4
Day?
Day 14
Day 28
Phase I: Container Type
Reagent Water
Reagent Water
Reagent Water (Control)
HOPE
Unsilanized
Glass
Silanized Glass
4°C
4°C
4°C
-
-
3
3
3
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
Phase II: Preservation Techniques (Dechlorination)
Reagent Water
Reagent Water
Reagent Water
Silanized Glass
Silanized Glass
Silanized Glass
4°C
4°C
4°C
2 mg/L C12
2 mg/L C12
+ 80 mg/L
Na2S2O3
2 mg/L C12
+ 50 mg/L
ascorbic acid
2
2
2
0
0
0
0
0
0
2
2
2
2
2
2
2
0
0
Phase II: Preservation Techniques (Dechlorination follow up samples)
Reagent Water
Reagent Water
HOPE
HOPE
4°C
4°C
80 mg/L
Na2S2O3
50 mg/L
ascorbic acid
1
1
1
1
0
0
0
0
0
0
0
0
Phase II: Preservation Techniques (Temperature)
Unspiked Effluent
Spiked Effluent
Spiked Effluent
Spiked Effluent
Silanized Glass
Silanized Glass
Silanized Glass
Silanized Glass
4°C
4°C
-20 °C
4°C
-
-
-
pH=2.0
3
3
3
2
0
0
0
0
0
0
0
0
0
o
J
o
J
2
0
o
J
o
J
2
0
o
J
o
J
0
Phase III: Biosolids
Unspiked Biosolid
Spiked Biosolid
Spiked Biosolid
Polypropylene
Polypropylene
Polypropylene
4C
4C
-20 °C
-
-
-
3
3
3
0
0
0
0
0
0
0
3
3
0
3
3
0
3
3
Phase IV: POTW Effluent Best Techniques
Spiked Effluent
HOPE
4°C
0.5 mg/L C12
+ 50 mg/L
ascorbic acid
4
0
4
4
4
0
Phase 1 samples were analyzed at 0, 7, 14, and 28 days; and Phase 2 samples were analyzed at 0,
4, 7, and 14 days.
Preparation of Samples for the Holding Time Study
Spiking Solution
All samples were spiked at the mid range of the calibration curve in the corresponding method
used (see Table 4). Each sample was spiked with all 89 chemicals. This produced samples
containing the same amount of target analytes as the on-going precision and recovery sample
(OPR) used for analytical method quality control. The spiking solution was a concentrated
methanol solution so that only 80 jiL was used for each 1-L sample.
-------
Table 4. Spiking Concentration for PPCPs, Steroids, and Hormones
Compound
Mestranol
Estrone
1 7
-------
Aqueous Samples
Samples were not preserved with sodium thiosulfate as specified in EPA Method 1694 unless
sodium thiosulfate was specifically being tested in that sample. Samples were dechlorinated with
50 mg/L of ascorbic acid instead of the 80 mg/L of sodium thiosulfate in Method 1694, and
stored in the dark at 4 °c.llc'14 Samples preserved with sulfuric acid were treated immediately
after collection by adjusting to pH 2 with concentrated sulfuric acid.
Chlorinated Samples
Chlorinated effluent and reagent water (MilliQ water) samples were prepared by adding sodium
hypochlorite solution to reagent water or effluent samples immediately after they had been
spiked with target compounds. Chlorinated samples were chlorinated at either 0.5 or 2 mg/L
which was verified using HACK procedure 8167 for Total Chlorine. Note that for all chlorinated
samples, each sample in this study was prepared (i.e., spiked and chlorinated) in its own bottle,
not in a bulk container and then aliquoted into 1-L sample bottles. Because of this, adding the
quenching agents to the bottles prior to sample collection was not possible.
Biosolid Samples
A bulk biosolid sample was collected from a POTW process employing screening, followed by
treatment with bacteria in a settling pond and then discharge of the aqueous material as final
effluent. To ensure adequate time for homogenization, subsampling and moisture determination
of biosolid samples, these procedures were carried out the day before the study began. Samples
were collected in two 500-mL jars using a large plastic scoop. The contents of the two jars were
emptied into a large, solvent-rinsed stainless steel bowl and, using forceps, the large pieces of
non-sediment material (vegetable skin, paint chips, etc.) were removed and the sample was well
mixed using a disposable teaspoon. The resulting homogenous mixture was portioned out into
ten 125-mL jars for ease of handling. Four separate subsamples were taken from this mixture for
percent solid determination. The sample was determined to be 27.7% solids; therefore 0.5 g dry
weight was equivalent to 1.8 g wet.
Effluent Samples
Effluent samples were obtained from the same POTW as for the biosolids samples. Effluent was
collected using a peristaltic pump from a final discharge pipe containing flowing effluent. Ten
five gallon pails were filled over the course of less than ten minutes. They were not combined
prior to filling the sample bottles. The effluent was collected and immediately taken to the lab for
dispensing into individual sample bottles, treatment and storage. The effluent was analyzed
unspiked as part of this project and tested for total chlorine before use and found to be chlorine-
free. pH or ammonia concentration was not characterized in POTW effluent. Therefore, an
estimate the proportions of free and combined chlorine, or chloramines present in the chlorinated
effluent samples cannot be know. However, by measuring "total chlorine" via Hach Procedure
8167, all of these forms were included in the measurement.
-------
Sample Containers
All sample bottles were silanized amber glass except those indicated as unsilanized or HDPE.
The HDPE bottles had polypropylene lids. The frozen effluents were stored in 2.5-L bottles to
prevent breakage. All other containers for aqueous samples were 1 L. Silanized glass bottles (1-
L and 2.5-L size) were purchased from ESS Environmental Sampling Supply (Oakland, CA).
Procedure for Aqueous Sample Preparation
For each "sample," three individual 1-L bottles were prepared: one each for acid and base
extraction for PPCPs and one for steroids and hormones extraction. Pre-labeled sample bottles
were filled and spiked in random order. For reagent water controls, sample bottles were filled
with 900 mL of reagent water and spiked with target compounds and mixed by shaking. Then
100 mL of reagent water was added to fill the bottle and again mixed. For chlorinated water
samples bottles were filled with 100 mL of reagent water, spiked with target compounds and
mixed by shaking. Then 900 mL of chlorinated water was added and the sample was mixed
again. For sodium thiosulfate samples, 100 mL of reagent water was added and the bottles were
spiked with target compounds and mixed by shaking. 800 mL of chlorinated water was added
and samples were mixed by shaking. 80 mg of sodium thiosulfate was added to the bottles and
samples were again mixed by shaking. Finally, chlorinated water was added to fill the bottle to 1
L and samples were mixed. For ascorbic acid samples, bottles were filled with 100 mL of
reagent water, samples were spiked with target compounds and mixed, then 800 mL of
chlorinated water was added to each bottle, mixed, then 50 mg ascorbic acid was added and
samples were mixed one final time. Bottles were filled to 1 L with reagent water. For reagent
water samples, the bottles were filled with 900 mL of reagent water, spiked with all target
compounds and mixed. Reagent water was added to fill the bottle to 1 L. For effluent samples,
bottles were filled with 900 mL of POTW effluent, spiked with target compounds and mixed.
Effluent was added to fill the bottle to 1 L. For frozen effluent samples, 2.5-L bottles were filled
with 1 L of effluent, spiked with target compounds and mixed. For effluent samples preserved
with sulfuric acid, bottles were filled with 900 mL of effluent, spiked with target compounds and
mixed. The samples were adjusted to pH 2.0 with 1:1 F^SC^ and then mixed. The bottles were
filled with effluent up to 1 L. The pH was tested and adjusted as necessary and mixed. For
aqueous samples with sodium thiosulfate and ascorbic acid, but without chlorine, samples were
prepared as described for other samples containing sodium thiosulfate or ascorbic acid, but were
stored in 1 L HDPE bottles with polypropylene lids. For best evaluated techniques, each HDPE
bottle was filled with 950 mL of effluent, and an aliquot of target analyte spiking solution was
added. The bottle was capped and shaken to mix. Dilute sodium hypochlorite solution (1.9 mL
of a 0.15% NaOCl) was added to give a total chlorine of 0.5 mg/L. The bottle was capped and
shaken to mix. Exactly one minute after addition of the chlorine solution 50 mg of ascorbic acid
was added. The bottle was capped and again shaken to mix. Each sample was then topped up to 1
L with effluent.
Procedure for Biosolids Sample Preparation
For each "sample," three individual containers were prepared: one each for acid and base
extraction of PPCPs and one the extraction of steroids and hormones. Between 1.8 and 1.9 g of
wet material was subsampled into each of 111 50-mL polypropylene tubes that would eventually
be used as extraction containers. This provided the 0.5 g of dry biosolids specified in EPA
-------
Method 1694 for each aliquot. The sample tubes were filled and stored overnight at 4 °C. On
Day 0, 96 of the replicate samples were spiked in random order with the analytes of interest. The
other 15 tubes were "controls" and were not spiked. Then 3.0 mL of reagent water was added to
each sample aliquot (all 111) and the biosolids, water, and spiking solution (80 jiL) were mixed
on a vortex mixer. The resulting replicate samples contained approximately 10% solids, and
were ready for storage or extraction and subsequent analysis on each day of the study.
Storage and Analysis of Samples
Samples were prepared on Day 0 of each study. Analysis of Day 0 samples commenced within
three hours of preparation. All other samples were immediately placed into the designated
storage condition until analyzed. Samples that were stored at 4 °C were kept in the refrigerator
until the morning of their analysis. Samples that were stored at -20 °C were removed from the
freezer the evening before analysis and allowed to thaw overnight in the refrigerator.
RESULTS
Analytical Results
Each sample in Table 3 was analyzed for the 89 target compounds and the results were
statistically analyzed (Tables 5-11). The individual compounds were organized into groups or
classifications to simplify the data analysis. These classifications are listed in Table 2 and
described below. A comprehensive statistical analysis of the results was conducted to determine
which set of conditions caused changes in the target compound concentrations.
Each analysis batch consisted of up to 15 test samples plus a procedural blank and an on-going
precision and recovery (OPR) sample, as specified in EPA Method 1694. Instrument quality
control included initial instrument calibrations consisting of a minimum of five concentrations of
all target compounds and on-going calibration verifications as described in Method 1694. All
data was subjected to two levels of validation. Specifically, a primary validator reviewed data
focusing on instrument performance, batch QC, correctness of calculations and further work
(dilutions, extra cleanup, repeat analysis, etc.). A secondary validator spot checked the primary
validator's work and focused on overall data set quality, flagging of results, reasonableness,
completeness of documentation, accuracy and completeness of the electronic data deliverable
(EDD) and final report. A final check was performed by the laboratory project manager prior to
reporting the data. The percent recoveries for target analytes from 20 randomly selected OPR
samples from each study phase are shown in Appendix 1.
Replicate samples were prepared and analyzed (see Table 3) for each sample treatment to
provide data that would be amenable to statistical analysis. Replicate results are therefore an
indication of overall precision, including both analytical and sampling variance. Of the over
10,000 results collected for aqueous samples, less than 1% were associated with analyte
recoveries being outside the specified range for OPR samples; 2.5% were associated with labeled
internal standard recoveries being outside the method specified range; and less than 1.2 % were
associated with blank contamination. Of the over 2,700 results collected for biosolid samples,
none was associated with analyte recoveries being outside the specified range for OPR samples;
1.8 % were associated with labeled internal standard recoveries being outside the method
specified range; and less than 1.0 % was associated with blank contamination.
-------
Statistical Analysis
Classification of Compounds for Statistical Analysis
Given that this work utilized screening analytical methods and evaluated a total of 89 analytes
with multiple replicate analyses for the majority of tests being conducted, we realized that to
accurately and efficiently examine the large amount of data produced, and to identify statistically
significant changes in concentration, a robust statistical approach would be needed. Some sort of
system for classifying these compounds into distinct groupings was required to facilitate
statistical analysis.
One limitation in performing a statistical analysis for any group of chemicals is that in some
cases, each subset of compounds represents a different class or classes of chemicals that may
behave very differently under similar conditions, due to differences in their physicochemical
properties. Specifically for this work, various chemicals may be prone to different reaction
pathways in environmental samples than others. For this reason, the chemicals in this study were
examined both by statistical analysis of different subsets (see Table 2) and, when appropriate, on
a compound-by-compound basis. Appendix 2 provides additional data on the degradation of
assorted compounds in this study which may not fit well within subgroups identified for the
purposes of statistical analysis.
PPCPs were divided into groups (Table 2) for the purposes of statistical analysis. Different
antibiotic classes, namely, macrolides, quinolines, sulfonamides, and beta lactam antibiotics, all
refer to classes of antibiotics that were grouped individually to make statistical analysis more
meaningful. Even though most of these chemicals may be amenable to extraction by similar SPE
conditions and analysis by similar instrumental conditions (in this case acidic extraction followed
by ESI+ LC/MS/MS analysis), they may react very different under environmental conditions.
For similar reasons, tetracyclines, chlorotetracyclines and degradates, all of which were extracted
under acidic conditions followed by ESI+ LC/MS/MS analysis, were designated as one group of
chemicals for the purposes of statistical analysis because of chemical and structural similarities.
Several steroids and hormones also were grouped together for the purposes of statistical analysis.
There were a wide variety of miscellaneous PPCPs analyzed in this work which did not fit well
into groups with chemical similarities. For this reason, they were grouped into four groups,
based on the Table 1 conditions of analysis. Specifically miscellaneous (Misc.) PPCP list 1
(Table 1) refers to chemicals extracted under acidic conditions and analyzed by ESI-
LC/MS/MS, Misc. PPCPs list 2 refers to chemicals extracted under basic conditions and
analyzed by ESI+ LC/MS/MS and Misc. PPCP list 3 refers to compounds that were extracted
under acidic conditions and analyzed by ESI+ LC/MS/MS.
Statistical Tests
Statistical analyses were performed on the analytical results for each compound to determine
which set of conditions resulted in a significant change in the concentration of target analytes.
Identification of statistical differences for each compound was performed using a multi-step
processes that involved a one-way Analysis of Variance (ANOVA) and a Dunnett's test. The
results from the holding time study were stratified by analyte and treatment. For each
-------
stratification, an ANOVA model was fit to assess the effect of holding time on the analyte
recoveries, in a two-step process.
In the first step, an overall F-test in the ANOVA model was used to determine whether there was
an overall difference between holding times (i.e., whether recovery at one holding time was
significantly different from recovery for at least one other holding time). The ANOVA asks the
question: "Does the mean recovery at any one holding time differ significantly from the mean
recovery at any other holding time?" Phrased differently, the question is: "Based on the observed
variability between replicate analyses, can we rule out the possibility that there are no differences
in analyte recovery between any of the different holding times?" If the ANOVA determines that
the answer is "No, there is not a difference," then we can stop evaluating the results for that
analyte and treatment.
If the ANOVA gives a "yes" answer, we concluded that at least one holding time had a mean
recovery significantly different from at least one other holding time. Therefore, the second step
involved pairwise comparisons between Day 0 and each of the other holding times, using
Dunnett's test. Dunnett's test was used to specifically compare results from one reference time
(Day 0) to all other holding times, without directly comparing the individual non-zero times to
each other.
Because fewer comparisons are run using Dunnett's test than for other pairwise comparison
procedures, Dunnett's test yields more statistical power for the given set of analyses. In addition,
because the Dunnett's test compared all of the results back to the Day 0 result, it ensured that
small changes over time were not overlooked (e.g., a gradual downward trend might not be
apparent when comparing results from adjacent days).
Based on the ANOVA and the Dunnett's test, we determined whether or not there was a
statistically significant difference in the mean result for each analyte on each day of the study
(e.g., 7, 14, and 28), compared to the results for the analyte on Day 0. The differences for each
analyte were classified as significant (Y) or not significant (N).
Magnitude of Statistical Differences: If a statistical difference was observed using the tests
described above, the magnitude of the difference for each analyte, in terms of percent difference
in mass compared to the mass on Day 0 was also calculated.
Using a 20% difference (positive or negative) in mass as an indication of a change that could
affect a decision about the data in a real-world sample, we counted the number of analytes for
which there was a statistically significant change (i.e., Y) and the magnitude of the change was
greater than 20%. We tabulated the number of analytes in each family that had significant
differences greater than 20% and those that did not. If there was a statistically significant change
with a magnitude less than 20%, we did not count that change in this evaluation of holding time
(e.g., a change of 5% in mass, while perhaps statistically significant, is not sufficient reason to
establish a holding time).
The use of the 20% threshold provided additional protection against a high frequency of false
positive decisions being made due to the large number of matrices, test conditions and analytes
for which holding time comparisons were performed. Considering the variability observed in the
analysis of biosolids and some other sewage matrices, we believe that this threshold provides
10
-------
added protection against counting variations in recovery as significant concentration increases or
decreases.
While other approaches to holding times determinations are available, the ANOVA along with
the use of a Dunnett's test offered a number of benefits. For instance, in ASTM D4841,15 the
study data quality objective is to be able to detect a 15% difference between holding times.
Ultimately, under ASTM D4841, holding time study data is evaluated using a separate t-test for
each holding time. The ANOVA used in conjunction with a Dunnett's test is more statistically
powerful than using a separate t test for each compound.
We performed our statistical analysis at a 95% confidence level rather than the 99% confidence
level used in ASTM D4841, to increase the statistical power (the ability to identify a holding
time difference). Even using a Dunnett's test, the statistical power would have been rather low if
we had run the test at the 99% confidence level given the practical resource limitations in the
study (the number matrices which needed to be covered and the number of replicate analysis
which could be reasonably performed). A result of using this approach is that it introduces a 5%
risk of concluding there is a holding time effect when there is not (rather than 1% as it would be
if the tests were run at the 99% confidence level).
In cases where the variability between replicate analyses in this study is low, utilizing the 20%
cutoff mentioned above decreases the 5% risk of concluding there is an effect when there is not.
Ultimately, the choice to identify both statistically significant increases and decreases was based
on previous work10, and on the observation that, for some of these analytes, a temporal effect, or
change in concentration might not result only in a decrease in concentration. For example, some
of the analytes in this study are breakdown products of other analytes being assessed. A
significant decrease for one analyte could result in a significant increase for another analyte. In
other cases, increases in concentration might be the result of conjugated forms of analytes
breaking down into unconjugated forms.
A tally was made of the numbers of analytes that "survived" on each day of the study (i.e., the
mean did not differ by more than 20% from the mean on Day 0). Each total was divided by the
total number of analytes in that chemical group, to arrive at the "percent survival" for each
treatment/family/day combination. We also calculated a survival rate for all 89 analytes in each
treatment.
Treatment of Non-Detects
Some target analytes did not survive under some conditions (e.g. chlorination) and therefore
were not detected even at Day 0 of the study. For the analytes that were not detected (ND) on
Day 0 and never observed on later days, we subtracted those analytes from the total in
calculating the survival rate. For example, if two analytes were ND on Day 0 and one other
analyte had a statistically significant difference greater than 20% on Day 7, all three analytes
were treated as not surviving on Day 7. If there were only four analytes in that family to start
with, then the survival rate was only 25% (1/4) on Day 7. On Day 14, analytes which were
initially ND, were also treated as not surviving.
11
-------
Results of Statistical Analysis
Container Types
Although silanized, unsilanized and HDPE containers were evaluated in this study, silanized
glass containers were chosen for control samples, POTW effluent and biosolids samples
primarily because many of the studies reported in the literature used this type of bottle.
Results of the statistical analysis of the effect of sample container on target compound stability
are presented in Table 5. A comparison of the percentage of analytes at Day 7 without a
statistically significant change greater than 20% for reagent water samples stored in silanized
bottles with those stored in plain glass or HDPE suggests that for some compounds there may be
little benefit when using silanized glass bottles. The potential benefit of the silanized glass
bottles exists primarily for quinoline antibiotics. Data indicated that for days 14 the "advantage"
of the silanized glass bottles is less evident for quinoline antibiotics and by Day 28, increases in
concentration are seen for some compounds, possibly due to a desorption effect. For quinoline
antibiotics, HDPE and unsilanized glass containers had more statistically significant changes on
days 7, 14 and 28. Sulfonamide antibiotics showed the least statistically significant changes in
concentration when stored in silanized glass and HDPE. Analyte-specific results for these
compounds indicate that it is difficult to determine which of the evaluated containers is the most
beneficial overall and container selection should be project specific and based on the compounds
of greatest interest.
Table 5. Percentage of Analytes at Day 7 without a Statistically Significant Change
Greater than 20%, by Bottle Type
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Steroids/Hormone s
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total Number
of Analytes
14
7
6
9
10
17
16
6
4
Bottle Type
Silanized Glass
57.1
71.4
50
100
100
94.1
75
83.3
50
Plain Glass
92.9
71.4
28.6
88.9
50
100
100
100
100
HDPE
85.7
57.1
57.1
100
50
100
87.5
83.3
100
Samples stored in unsilanized containers had fewer statistically significant changes in
concentration for tetracyclines, however, this category of compounds showed changes after 7
days regardless of container. In addition, 4-epianhydrochlorotetracycline, 4-epianhydro-
tetracycline and anhydrochlorotetracycline either decreased significantly or were not recovered
past Day 0 when stored in silanized glass containers. Decreases for these compounds in HDPE
containers were less dramatic than in silanized containers. Ultimately, it may be the stability of
these compounds that plays a role in these decreases rather than the container selected.
Macrolide antibiotics had statistically significant changes when stored in HDPE. One compound,
azithromycin, was not recovered when stored in HDPE. A number of p-lactam antibiotics had
statistically significant increases in concentration on Days 7, 14 and 28 when being stored in
12
-------
HDPE but had statistically significant decreases in unsilanized and silanized glass containers.
For sulfonamide antibiotics, both HDPE and silanized containers yielded results with fewer
statistically significant decreases. Steroids and hormones showed few statistically significant
changes in concentration at Day 7 in all containers tested.
The benefits obtained from using silanized glass containers are evident primarily for quinoline
and sulfonamide antibiotics. In contrast, unsilanized glass containers appear to yield fewer
statistically significant changes for a number of miscellaneous PPCPs, steroids, and hormones.
Although the cost of using silanized containers is not prohibitive, there is added complexity in
sampling procedures using different types of glass containers for different analytes. Although
our results suggest that the limited benefit observed in this study from silanization may not
warrant the use of silanized glass bottles in all situations, we recommend that this be evaluated,
when appropriate, on a case-by-case basis for the specific compounds being tested.
It is possible that adsorption to container surfaces would be reduced in more complex sewage
matrices relative to the reagent water samples studied in the container component of this work.
This may mean that choice of container type may be more critical for some types of samples than
others. We note that our data may not be relevant to analysis in drinking water samples where the
concentrations of PPCPs are expected to be considerably lower than those being spiked into
samples in this study. In those cases, slight concentration decreases from adsorption could mean
the difference between presence and absence.
Chlorination and Dechlorinating Agents
The effect of various preservatives and dechlorinating agents has been investigated for specific
chemicals. Glassmeyer and Shoemaker (See Reference 12) reported degradation of PPCPs in
chlorinated samples Vanderford, et al. (See References 11 and 12), tested sodium thiosulfate,
ammonium chloride and ascorbic acid to dechlorinated surface water samples. They reported
that sodium thiosulfate degraded trimethoprim, erythromycin, fluoxetine, atrazine, diazepam,
progesterone and diclofenac, while ascorbic acid only partially degraded erythromycin.
Ammonium chloride did not affect compound recoveries.
The chlorine concentrations used in this study (2 mg/L or 0.5 mg/L) were selected to reflect a
range of concentrations that might be found in NPDES permits, drinking water, and POTW
effluents. The effects of the chlorine on the PPCPs studied in this work were dramatic. This is
shown in Table 6, which compares the numbers of study analytes that were spiked into each
sample, but not detected on Day 0, the same day that the spiked samples were prepared. The
majority of antibiotics, PPCPs, steroids and hormones were not detected in chlorinated samples
that had not been dechlorinated. This is consistent with literature reports that chlorine destroys a
number of PPCPs (See Reference 12).
Table 6. Numbers of Analytes Not Detected on Day 0 in Samples Chlorinated to 2 mg/L
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Total Number of
Analytes
14
7
6
9
10
Chlorinated
Water
14
7
3
9
10
With Ascorbic
Acid
12
3
0
0
0
With Sodium
Thiosulfate
14
7
6
9
10
13
-------
Table 6. Numbers of Analytes Not Detected on Day 0 in Samples Chlorinated to 2 mg/L
Analytical Family
Steroids/Hormones
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total Number of
Analytes
17
16
6
4
Chlorinated
Water
13
13
4
2
With Ascorbic
Acid
1
0
1
2
With Sodium
Thiosulfate
1
12
0
2
This study examined the effect of two different reagents that are often used to remove residual
chlorine from samples: ascorbic acid and sodium thiosulfate.16 In general, for antibiotics, the use
of sodium thiosulfate (80 mg/L) was least effective in the protection of analytes in the presence
of 2 mg/L of chlorine (Table 6). In contrast, there were many fewer non-detects in the
chlorinated samples treated with ascorbic acid (50 mg/L). Both dechlorinating agents protected
the majority of steroids and hormones in this study, however 17a-dihydroequilin was not
recovered using either dechlorinating agent. In the miscellaneous PPCP category, sodium
thiosulfate appeared to protect warfarin, while ascorbic acid did not, however, in general,
samples dechlorinated with sodium thiosulfate showed more non-detects. With the exception of a
few PPCPs, steroids, and hormones (data not shown), the majority of compounds were found to
be better preserved when samples were dechlorinated with ascorbic acid. Dechlorination with
sodium thiosulfate in the presence of 2 ppm chlorine had low or no recovery of a number of P-
lactam, tetracycline and quinoline antibiotics. It is evident that some of the compounds in this
study are simply too labile to withstand the effects of chlorine or dechlorination agents.
Specifically, 17 a-dihydroequilin, cimetidine, ranitidine and many of the tetracycline,
chlorotetracyclines and their degradates were not detected in chlorinated or dechlorinated
samples.
The direct effect of each dechlorinating agent was evaluated in reagent water samples without
any added chlorine. Four reagent water samples were prepared and spiked with all of the
analytes. Ascorbic acid was added to two of the samples, and sodium thiosulfate to the other two
samples. One sample from each treatment was analyzed on Day 0 and the remaining two
samples were analyzed on Day 2. All of the analytes were present on Day 0 in the samples
treated with ascorbic acid. However, clinafloxacin, diphenhydramine, fluoxetine, and
sarafloxacin, were not detected on Day 0 or Day 2 in the samples treated with sodium thiosulfate.
That only one sample of each treatment was analyzed on each day makes statistical comparisons
impossible, however this data may be used to better assess the direct effects of these
dechlorinating agents on PPCPs.
For almost all tetracycline, chlorotetracycline, and degradates, there were notable decreases by
Day 2 in the presence of both dechlorinating agents. Decreases in concentration for sulfonamide
antibiotics, steroids, and hormones in these samples appeared to be minimal in the presence of
either sodium thiosulfate or ascorbic acid.
This data, in combination with that from the earlier tests which evaluated chlorine and
dechlorinating agents, suggests that ascorbic acid is a more suitable dechlorinating agent for
PPCPs than sodium thiosulfate and that sodium thiosulfate destroys some of the PPCPs
evaluated in this study.
14
-------
Temperature and Acid Preservation Effects on Effluent Samples
Snyder et al. (See Reference 11) reported rapid decreases in the concentrations of
androstenedione, testosterone, progesterone, trimethoprim, acetaminophen, and fluoxetine in
unpreserved surface waters (See Reference 11). Snyder hypothesized that this was due to
microbial degradation and subsequently tested some biocide techniques. The addition of
formaldehyde (1%, v/v) to freshly collected samples stabilized these compounds, while other
compounds, including acetaminophen, meprobamate, dilantin, TCEP, and isopromide, showed
signs of degradation by the formaldehyde. Sulfuric acid preservation prevented degradation of
the compounds affected by formaldehyde and did not adversely affect the recoveries of the other
compounds.
While there are a variety of antimicrobial agents such as CuSC>4 and diazolidinyl urea17 that have
been used to prevent microbial degradation in drinking water samples, their use has seldom been
extended to sewage samples. On the other hand, EPA has used acid to prevent the bacterial
degradation of some aromatic compounds in wastewater samples. The most notable of these are
benzene, toluene, and ethyl benzene which are susceptible to rapid biological degradation under
certain environmental conditions.18 Strong mineral acids such as HC1 or H2SO4 can be used to
reduce the pH of a sample to less than 2, thereby causing many bacterial cells to lyse and
minimize the biological activity present in the sample. Acid preservation in conjunction with
reduced temperatures is not uncommon for preservation of samples.
Sewage treatment plant effluent samples were stored at either 4 °C (cold) or -20 °C (frozen). The
goal of both temperature treatments was to reduce the biological activity of the samples, as well
as reduce the rates of any chemical reactions in the effluent that might affect the analytes of
interest. Acidification of the sample is a treatment that also might reduce biological activity in
the samples.
The results of the statistical analysis of the storage temperature and acidification data are
summarized in Table 7. For macrolide antibiotics, sulfonamide antibiotics and steroids and
hormones, freezing the effluent samples did not increase survival on Day 7. For quinoline and P-
lactam antibiotics, little to no differences were observed between samples which were frozen and
those which were stored at 4 °C for a number of compounds (data not shown).
Table 7. Percentage of Analytes in Effluents at Day 7 without a Statistically
Significant Change Greater than 20%, by Storage Condition
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Steroids/Hormones
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total Number of
Analytes
14
7
6
9
10
17
16
6
4
Storage Condition
Cold
21.4
71.4
83.3
88.9
60
88.2
62.5
83.3
100
Cold and pH < 2
7.1
28.6
66.6
0
70
82.4
87.5
83.3
75
Frozen
64.3
57.1
50
66.7
90
47.1
75
100
100
15
-------
For some miscellaneous PPCPs, tetracycline antibiotics, and quinolone antibiotics which were
initially protected by freezing, concentrations for some these compounds in frozen samples
dropped off dramatically in the later days of the study. These results suggest that freezing may
not provide substantial benefits to effluent samples which are expected to be much less
biologically active than biosolids. And samples collected in glass and then frozen can break if
not filled and stored correctly.
Increased concentrations of certain analytes, such as desogesterol, digoxin, equilenin,
virginiamycin, roxithromycin, and sulfanilamide, were observed in some samples after storage
(See Figure 1). It is possible that the increases in concentration over the duration of the study
may reflect the presence of conjugated forms of some of these analytes or interactions between
the spiked analytes and the sample matrix that vary with time.
Figure 1. Macrolide Antibiotics in Frozen POTW Effluent
POTW Effluent -20°C
o
0)
u
<§
n
N
jr
Macrolide Antibiotics
The combination of acidification and cold storage in this study achieved only mixed success
relative to cold storage or freezing alone. Our studies revealed that samples preserved with
sulfuric acid had marked decreases in recoveries for p-lactam, sulfonamide, tetracycline
antibiotics. When comparing samples preserved with acid to those simply stored at 4 °C for a
series of other PPCPs (Table 7), preservation with acid provided few benefits and actually
destroyed a number of PPCPs on Day 0, including norgestimate and ranitidine.
For steroids and hormones, treatment with acid actually produced increases in the concentrations
of a number of compounds. Deconjugation of estrogen conjugates, matrix enhancement, or other
reactions in POTW samples may have contributed to this increase, however further analysis
would be required to prove this hypothesis.
16
-------
Combination of Best Aqueous and Effluent Treatments
The goal of Phase IV of this study was to test the "best combination" of storage and preservation
conditions that had been observed in Phase 1 and 2. These "best combination" conditions were:
chlorinated effluent samples, held in HDPE bottles, dechlorinated with ascorbic acid, and stored
at 4 °C, with no pH adjustment. HDPE containers were selected because earlier results did not
reveal that one type of container was significantly better than another for all of the PPCPs and
hormones in this study, and because these bottles had been used in previous EPA studies of
sewage samples. Ascorbic acid was selected because it proved to be the best dechlorinating
agent tested. Storage at 4 °C was selected as opposed to freezing because freezing provided
minimal benefit and presented the additional problem of breakage. Four replicate samples were
analyzed on each of Day 0, Day 4, Day 7, and Day 14 (16 samples total). Analyses on Day 4
were included to see if any changes over the first 7 days would be apparent prior to Day 7.
Effluent samples used for this portion of the study had no detectable levels of chlorine, so they
needed to be chlorinated. The spiking levels of chlorine were decreased from 2 mg/L to 0.5
mg/L to better reflect what might be more commonly found in typical POTW samples. In
addition, given the dramatic losses of analytes in the chlorinated reagent water samples earlier in
the study, this reduction in the concentration could offer more useful information. The use of
four replicate samples on each day added statistical power to the study, compared to the three
replicates used for earlier effluent samples.
The results for the best combination analyses are compared to those for unchlorinated Phase 2
effluent samples in Table 8 to illustrate comparison of chlorinated POTW effluent treated with
ascorbic acid and unchlorinated POTW effluent not treated with ascorbic acid. The third and
fifth columns in Table 8 illustrate the differences in the number of non-detects at the start of each
experiment (Day 0) and the fourth and sixth columns illustrate the percentage of analytes in each
family that survived to Day 7 (e.g., those without a statistically significant difference greater than
20%).
Table 8. Comparison of Results for the Best Combination Treatment of Effluents and
Simple Cold Storage
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Steroids/Hormones
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total
Number of
Analytes
14
7
6
9
10
17
16
6
4
Best Combination (n=4)
# NDs Day 0
9
2
0
2
1
0
0
0
0
% Survival on
Day 7
21.4
28.6
83.3
66.7
30
64.7
81.3
83.3
75
Cold Storage only (n=3)
# NDs Day 0
0
0
0
0
0
0
0
0
0
% Survival
on Day 7
21.4
71.4
83.3
88.9
60
88.2
62.5
83.3
100
Note: % Survival indicates those analytes without a statistically significant difference greater than 20%
Even under these "best" conditions several compounds were not detected at Day 0. Specifically,
nine tetracycline compounds (4-epaanhydrochlorotetracycline, 4-epianydrotetracycline, 4-
17
-------
epichlorotetracycline, 4-epioxytetracycline, anhydrochlorotetracycline, anhydrotetracycline,
chlorotetracycline, isochlorotetracycline and oxytetracycline), two p-lactam compounds
(cefotaxime and lincomycin), two sulfonamide (sulfadimethoxine and sulfathiazole) and one
quinolone antibiotic (carbadox) were ND in these samples (data shown in appendix 2). As this
extent of loss was not previously observed in POTW effluents stored cold without treatment
(Table 8), we concluded that the selected treatment combination affected the survival of a
number of analytes on Day 0. The results from earlier chlorinated reagent water analyses show
that all of the tetracycline, chlorotetracycline and degradates, and a number of antibiotics were
lost on Day 0 in those samples, supporting the conclusion that it may be more likely that the
chlorine, rather than the ascorbic acid, destroyed most of these compounds.
The number of non-detects on Day 0 and the percentage of analytes that survived on Day 7 for
both treatments may illustrate the combined effects of chlorine and ascorbic acid. With the
exception of macrolide antibiotics, fewer antibiotics, steroids and hormones survived in the
presence of chlorine and ascorbic acid in these samples. The survival rates for tetracyclines and
several miscellaneous PPCPs were not affected, although the analytes that survived in best
combination samples and simple cold storage samples differed to some degree. Only for the
Misc. PPCPs in List 3, was survival higher in the best combination treatments.
Table 9. Comparison of Results for the Best Combination Treatment of Effluents on
Days 4, 7, and 14
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Steroids/Hormones
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total Number of
Analytes
14
7
6
9
10
17
16
6
4
# NDs on Day 0
9
2
0
2
1
0
0
0
0
% of Analytes Surviving
Day 4
35.7
57.1
83.3
66.7
50
88.2
75
100
50
Day 7
21.4
28.6
83.3
66.7
30
64.7
81.3
83.3
75
Day 14
21.4
42.9
83.3
77.8
70
64.7
75
83.3
100
Comprehensive data for the best combination of samples are shown in Table 9. These results
illustrate many of the same time trends observed in Phase 1 of this study. Some of the
statistically significant differences in these results reflect increases in analyte concentrations over
time. For example, the survival rates for quinolines, sulfonamides and PPCPs lists 2 and 3
(cimetidine, diphenhydramine, ranitidine and warfarin) on Day 7 are lower than the survival on
later Days. Figure 2 shows the results for Quinoline and several other antibiotics. These results
suggest that the background compounds present in the effluent may be undergoing changes as
the samples age.
18
-------
Figure 2. Quinoline and Several Other Antibiotics in POTW Effluent with 0.5
mg/L Ch, Preserved with Ascorbic Acid and Stored at 4 °C in HDPE
c
o
I
0)
u
c
o
o
•o
0)
N
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
^
i
\\
A
I
i
3
tf
il
/ y y x y y y./ y
Quinoline & Other Antibiotics
The results for steroids and hormones reflect decreases in the concentrations for 17 alpha-
dihydroequilin, androsterone and desogesterol. Mestranol had a statistically significant loss on
Day 7 that was not apparent on Day 14. In fact, the mean result for this analyte on Day 14 was
within 0.5% of the Day 0 mean result. Examination of the Day 7 data for mestranol revealed that
there is no evidence of an outlier among the four replicate results that would explain the
magnitude of the loss of this analyte.
Effect of Time and Temperature on Storage ofBiosolids
The increased biological activity of biosolids may account for changes in the concentration of
PPCPs in biosolids. It is generally not practical to add chemical preservatives to solid samples at
the time of collection. Therefore, most methods for solid matrices rely solely on reduced storage
temperatures to minimize biological activity and preserve the analytes of interest. Two
temperatures were investigated in this holding time study: half of the samples were stored at 4
°C, and half at -20 °C. The results are summarized in Table 10, in terms of the percentage of
analytes surviving on each day without a statistically significant difference greater than 20%.
There were several analytes that were not detected in either set of biosolids samples on Day 0,
despite being spiked into all of the sample aliquots. Both cefotaxime and desogestrel were non-
detects on Day 0, and on all subsequent days in the biosolids samples stored at 4 °C and those
stored -20 °C. Although sarafloxacin was not detected on Day 0 in all four of the frozen
biosolids samples, it was detected on all subsequent days in the frozen samples, at levels that
represent 60 to 75% of the initial spike level. Examination of the analysis QC data associated
with the Day 0 analyses did not identify any analytical problems for sarafloxacin, however
matrix enhancement and/or suppression cannot be ruled out as contributing factors.
19
-------
Table 10. Percentage of Analytes in Biosolids Surviving without a Statistically
Significant Change Greater than 20%, by Storage Temperature
Analytical Family
Tetracyclines
beta-Lactam + misc antibiotics
Macrolide antibiotics
Sulfonamide antibiotics
Quinolone type antibiotics
Steroids/Hormones
Misc PPCPs List 3
Misc PPCPs List 1
Misc PPCPs List 2
Total
Number of
Analytes
14
7
6
9
10
17
16
6
4
% Survival
Day 7
Cold
64.3
42.9
83.3
22.2
90
29.4
68.8
83.3
75
Frozen
64.3
42.9
66.6
88.9
40
52.9
68.8
83.3
75
Day 14
Cold
78.6
0
83.3
33.3
70
41.2
56.3
100
75
Frozen
78.6
42.9
100
88.9
60
53.0
68.8
100
75
Day 28
Cold
50
20
66.6
22.2
80
47.1
50
66.7
75
Frozen
50
42.9
100
55.6
90
70.6
68.8
100
75
Comparing the data for Day 7 in Table 10, freezing biosolids samples does not protect more of
the analytes than simple cold storage, except for sulfonamide antibiotics and perhaps some
steroids and hormones. When examining data past Day 7, the benefit for freezing seems more
apparent for beta-lactam, macrolide antibiotics and steroids and hormones. However, as seen in
results for effluent samples, the results for the later days in the study show statistically significant
increases in concentration, rather than decreases, for some analytes when compared to Day 7
results. For example, all five of the statistically significant differences for tetracyclines on Day 7
were increases greater than 20% (See Figures 3a and 3b). The only statistically significant
decreases in tetracycline, chlorotetracycline, and several of their degradates appeared on Day 28.
The data for tetracyclines in cold biosolids samples suggests that concentrations peak at Day 14,
but in frozen samples, the results continue to increase to Day 28. The lag between the peaks in
the cold and frozen samples may reflect a temperature dependence of any reactions occurring in
the stored sample.
20
-------
Figures 3a and 3b. Tetracycline and Tetracycline Degradates in Biosolids Samples at
4 °C and -20 °C
.0
-------
concentration in biosolids samples which were not frozen. Decreases in concentration for these
compounds were less substantial in biosolids samples that were frozen. Cefotaxime was not
recovered in frozen biosolids samples. For quinoline, sulfonamide and macrolide antibiotics, a
combination of statistical analysis (Table 10) and examination of data for individual compounds
(data not shown) indicates there may only be moderate benefits to freezing.
For the steroids and hormones examined, there were marked increases in concentrations for 17
alpha-estradiol, equilenin, equilin and estrone on Day 7 in both the cold and frozen samples. For
a few analytes, the increases continued through the later days in the study even in samples stored
at -20 °C. Table 11 provides the mean results for these analytes in the frozen biosolids samples
over time.
Table 11. Mean Results for Several Hormones in the Frozen Biosolids Samples over Time
Analyte
17 alpha-dihydroequilin
17 alpha-Estradiol
17 beta-Estradiol
Equilenin
Equilin
Estrone
Mean Day 0
69.0
32.5
76.5
16.6
149.0
64.7
Mean Day 7
80.9
203.3
100.3
71.6
208.7
389.7
Sig Dif ?
N
Y
N
Y
N
Y
Mean Day 14
86.3
248.3
102.4
46.4
424.7
453.7
Sig Dif?
N
Y
N
Y
Y
Y
Mean Day 28
81.5
80.1
71.2
21.4
124.1
162.0
Sig Dif?
N
N
N
N
N
N
There are different trends for some of the steroids and hormones over time. For example, for
both the frozen and cold biosolids samples, the results for estrone and equilin appear to peak at
Day 14, and then decrease dramatically by Day 28. The increases to Day 14 are statistically
significant for both of these analytes in the frozen and cold samples.
Extract Stability
Extracts of effluent and OPR samples stored at 4 °C were re-analyzed at various time intervals
after initial analysis to determine their stability over time. A minimal number of replicates were
included in this portion of the study and therefore rigorous statistical analysis was not possible.
Most PPCPs (with the exception of tetracyclines, Figures 5a and 5b) and hormones in this work
appeared to be stable in extracts stored for at least one month (data not shown). While there were
some compounds that were stable beyond one month, this limited data set suggests that for most
compounds (with the exception of tetracyclines) precautionary extract holding time of up to 30
days may be sufficient to preclude extensive degradation of some compounds in extracts. It is
likely that for other compounds (e.g. tetracyclines, an extract holding time of less than 30 days
may be required.
22
-------
Figures 4a and 4b. 4a. PPCPs extract stored in methanol/formic acid buffer solution and
analyzed at Day 0 and day 29. Samples were reagent water samples.
4b. Steroids and Hormones extract stored in methanol/formic acid
buffer solution and analyzed at Day 0 and Day 38. Samples were
reagent water samples. Concentration is normalized to Day 0.
Reagent Water Sample Extract at 0 and 29 days
Reagent Water Sam pie Extract at 0 and 38 days
O
C
O
O
T3
-------
Figure 5a and 5b. 5a. Tetracycline, chlorotetracyclines and degradates data from two
separate extracts stored in methanol/formic acid buffer solution and
analyzed at 0 and 24 days. Samples were reagent water treated with
ascorbic acid in the presence of 0.5 mg/L chlorine.
5b. Data for miscellaneous PPCPs for extracts stored in methanol/
formic acid buffer solution and analyzed at 0 and 130 days. Samples
were reagent water samples. Concentration is normalized to Day 0.
.§
"ro
-------
DISCUSSION
It is not surprising that there is no clear-cut answer to the question "What should the holding
times and preservation conditions for PPCPs, steroids and hormones be?" For instance, for some
samples and analytes, different containers may be appropriate. The results of this study can be
used to recommend some sample preservation and treatment conditions that are more useful than
others, but ultimately, for some parameters tested, there is no one best choice for all chemicals.
For example, HDPE containers may be adequate for some classes of analytes but silanized glass
containers may be more appropriate for others. Containers may need to be selected based on the
analytes being studied.
Likewise, chlorinated samples treated with ascorbic acid exhibited fewer nondetects than those
treated with sodium thiosulfate. The data for various samples which did not contain chlorine
indicate that some classes of analytes will survive for 7 days without a statistically significant
change greater than 20%. We observed that some analytes exhibit statistically significant
changes in concentration in as few as seven days, even in the absence of chlorine (Table 7).
Thus, for aqueous samples not containing chlorine, a maximum holding time of 7 days until time
of analysis may be advisable.
The addition of sulfuric acid as a preservative reduced the recovery of a number of compounds
while freezing was only marginally successful at preventing the degradation of analytes and may
lead to problems with sample breakage. Thus, based on this study, we recommend storing
sewage treatment plant effluent samples at 4 °C and without sulfuric acid. Even in best
combination samples the concentrations of tetracyclines, chlorotetracyclines and tetracycline
degradates as well as several other PPCPs and hormones including: 17 alpha dihydroequilin,
desogesterol, carbadox, sulfadimethoxine, sulfathiazole, cefotaxime and linomycin were shown
to decrease quickly.
For biosolids samples, we observed that the concentrations of some analytes increase
significantly in samples held up to 14 days, possibly as a result of the release of conjugate forms
of some of the analytes from the biosolids. Cold storage, at 4 °C appears to be sufficient to
preserve many of the analytes in biosolids for this period, however, for antibiotics such as beta
lactams, freezing samples at -20 °C may be more effectively prevent degradation. For these
reasons, a holding time of 7 days cold or frozen for biosolids samples is likely reasonable.
There are several limitations to conclusions to draw from the data in this study. Because the
methods used in this work cover a wide range of matrices and analytes, the results include a
subset of non-quantitative results, specifically for those compounds not measured by isotope
dilution. It is not likely that our findings may be extended to matrices such as drinking water,
where small differences in concentration can mean the difference between presence and absence.
We relied on statistical analysis, and in some cases, presence/absence to make our holding time
determinations. Studies using more precise analytical methods and focusing on fewer target
compounds may reveal more subtle differences than those observed in this study. Ultimately,
holding times arising from this study are protective of analytes which degraded more quickly
than others and are more suited to Methods 1694 and 1698 than to analytical methods targeting a
subset of the chemicals in EPA Methods 1694 and 1698.
25
-------
CONCLUSIONS
Most CEC analytical methods cover a large number of chemicals from different classes of
chemicals. Thus, uniform holding times and preservation conditions may not apply across all of
these classes. More relevant data might be obtained when using holding times and storage
conditions specific to the class of compounds and samples being tested with an analytical method
designed specifically for the target CECs. The results of this 2008 study may help researchers
design stability studies to identify holding times and preservation conditions appropriate to their
samples and target analytes.
The results of this study have been used to revise the holding times and preservation conditions
for PPCPs in EPA Method 1694 and steroids and hormones in Method 1698. Previous holding
time and preservation conditions in these methods specified the use of amber glass containers
(optionally Method 1698 allowed the use of amber plastic), sodium thiosulfate for
dechlorination, storage at less than 4 °C for samples (with freezing as an option), storage of
sample extracts in the dark at less than -20 °C, start of sample extraction within 7 days of
collection (within 48 hours is strongly encouraged), analysis of sample extracts within 40 days of
extraction, and pH adjustment to 5.0 - 9.0 for samples not extracted within 48 hours of
collection.
Based on the results of this study, these conditions have been revised as follows:
• Bottles - amber high density polyethylene or glass containers.
• Dechlorinating Reagent - ascorbic acid.
• Shipping and Storage Conditions - in the dark at <6 °C (optional freezing for biosolids).
• Holding Times - extract samples within seven days; analyze extracts as soon as possible,
not to exceed 30 days (10 in the case of tetracyclines).
It is essential to point out that the above holding times are not universal, but are protective due to
the large number of compounds found in Method 1694.
26
-------
REFERENCES
1. a) Susan D. Richardson. Environmental Mass Spectrometry: Emerging Contaminants and
Current Issues. Anal. Chem. 2002, 74,2719-2742.
b) Susan D. Richardson. Water Analysis: Emerging Contaminants and Current Issues. Anal.
Chem.2003, 75,2831-2857.
c) Susan D. Richardson. Environmental Mass Spectrometry: Emerging Contaminants and
Current Issues. Anal. Chem.2004, 76,3337-3364.
d) Susan D. Richardson, and Thomas A. Ternes. Water Analysis: Emerging Contaminants and
Current Issues. Anal. Chem.2005, 77,3807-3838.
e) Susan D. Richardson. Environmental Mass Spectrometry: Emerging Contaminants and
Current Issues. Anal. Chem.2006, 78,4021.
f) Susan D. Richardson. Water Analysis: Emerging Contaminants and Current Issues. Anal.
Chem.2007, 79, 4295-4324.
g) Brenton Nicholson. Organic Chemical Issues in Wastewater Quality, A Review of Current
Analytical Methods. Australian Water Quality Centre. CRC for Water Quality and Treatment.
September 2006. Available at
http://www.waterquality.crc.org.au/publications/Chemical issues in wastewater quality.pdf
h) Christian G. Daughton and Thomas A. Ternes. Pharmaceuticals and Personal Care Products in
the Environment: Agents of Subtle Change? Environmental Health Perspectives Volume 107,
Supplement 6, December 1999.
i) TL Jones-Lepp, Rick Stevens. Pharmaceuticals and Personal Care Products in
Biosolids/Sewage Sludge - The Interface between Analytical Chemistry and Regulation.
Available at www.epa.gov/esd/bios/pdf/ABC-biosolids.pdf
2. a) Michele E. Lindsey, Michael Meyer, and E. M. Thurman. Analysis of Trace Levels of
Sulfonamide and Tetracycline Antimicrobials in Groundwater and Surface Water Using Solid-
Phase Extraction and Liquid Chromatography/Mass Spectrometry. Anal. Chem. 2001, 73, 4640-
4646.
b) Dana W. Kolpin, Edward T. Furlong, Michael T. Meyer, E. Michael Thurman, Steven D.
Barber and Herbert T. Buxton. Pharmaceuticals, Hormones, and Other Organic Wastewater
Contaminants in U.S. Streams, 1999-2000: A National Reconnaissance. Environ. Sci.
Technol.2002, 36,1202-1211.
3. a) Stumm-Zollinger, E. and G. M. Fair. "Biodegradation of steroid hormones." Journal of the
Water Pollution Control Federation. 1965,37: 1506-1510.
b) Tabak, H. H. and R. L. Bunch. Steroid hormones as water pollutants. I. Metabolism of natural
and synthetic ovulation-inhibiting hormones by microorganisms of activated sludge and primary
settled sewage." Dev. Ind. Microbiol. 1970. 11: 367-376.
c) A.W. Garrison, J.D. Pope, F.R. Allen, 'GC/MS Analysis of Organic Compounds in Domestic
Wastewaters', in Identification and Analysis of Organic Pollutants in Water, ed. C.H. Keith, Ann
Arbor Science Publishers, Ann Arbor, 517-556, 1976.
d) C. Hignite, D.L. Azarnoff, 'Drugs and Drug Metabolites As Environmental Contaminants:
Chlorophenoxyisobutyrate and Salicylic Acid in Sewage Water Effluent', Life Sci. 1977. 20,
337-341.
27
-------
4. Britt E. Erickson, Analysing the Ignored Environmental Contaminants. Environ. Sci. Technol.,
2002, 36, 140A-145A. Article at: pubs.acs.org/subscribe/journals/esthag-
w/2002/mar/science/data/Erickson.pdf.
5. a) Kolpin, D. W.; Furlong, E. T.; Meyer, M. T.; Thurman, E. M.; Zaugg, S. D.; Barber, L. B.;
Buxton, H. T. Environ. Sci. Technol. 2002, 36, 1202-1211.
b) Kinney, C. A.; Furlong, E. T.; Zaugg, S. D.; Burkhardt, M. R.; Werner, S. L.; Cahill, J. D. and
Jorgensen, G. R. Environ. Sci. Technol.; (Article); 2006; 40(23); 7207-7215.
6. S. Castiglioni, E. Zuccato, E. Crisci, C. Chiabrando, R. Fanelli, and R. Bagnati. Anal. Chem.
2006, 78, 8421-8429.
7. a) P. M. Bradley, L. B. Barber, F. H. Ghapelle, J.L. Gray, D. W. Kolpin, and P. B. McMahon.
Environ. Sci. Technol. 2009, 43, 1902-1910.
b) J. Skotnicka-Pitak, W. O. Khunjar, N. G. Love, and D. S. Aga Environ. Sci. Technol., 2009,
43 (10), pp 3549-3555.
8. a) H. Zhang and J. Henion. Anal. Chem. 1999, 71, 3955-3964.
b) S. Reddy, C. R. Iden, and B. J. Brownawell. Anal. Chem. 2005, 77, 7032-7038.
9. A. Gentili, D. Ferret, S. Marchese, R. Mastropasqua, R. Curini, and A. Di Corcia.
Chromatographia. 2002, 56, 25-32.
10. a) T. Isobe , H. Shiraishia, M. Yasudab, A. Shinodab, H. Suzuki, M. Morita. Journal of
Chromatography A, 984 (2003) 195-202.
b) A.C. Belfroid, A.V.D. Horst, A.D. Vethaak, A.J. Schafer, G.B.J. Rijs, J.Wegener, W.P.
Cofmo, Sci. Total Environ. 225. (1999) 101.
11. a) B. J. Vanderford, R. A. Pearson, D. J. Rexing, and S. A. Snyder. Anal. Chem. 2003, 75,
6265-6274.
b) R. A. Trenholm, B. J. Vanderford, J. C. Holady, D. J. Rexing, and S. A. Snyder.
Chemosphere. 2006, 65, 1990-1998.
c) P. Westerhoff, Y. Yoon, S. Snyder, and E. Wert. Environ. Sci. Technol. 2005, 39, 6649-6663.
12. S. T. Glassmeyer, and J. A. Shoemaker. Bull. Environ. Contam. Toxicol. 2005, 74: 24 - 31.
13. EPA Method 1694. Pharmaceuticals and Personal Care Products in Water, Soil, Sediment,
and Biosolids by HPLC/MS/MS. EPA, Engineering and Analysis Division.
http ://www. epa.gov/waterscience/methods
14. Z. Ye, H. S. Weinberg andM. T. Meyer. Anal. Chem., 79, 3, 1135 -1144, 2007.
15. ASTM D4841-88. Standard Practice for Estimation of Holding Time for Water Samples
Containing Organic and Inorganic Constituents.
16. a) P. E. Stackelberg, J. Gibbs, E. T. Furlong, M. T. Meyer, S. D. Zaugg, R. L. Lippincott.
Science of the Total Environment 377 (2007) 255-272.
28
-------
16. b) EPA Method 526. S.D. Winslow, B. Prakash, M.M. Domino, B.V. Pepich, and D. J.
Munch. Determination of Selected Semivolatile Organic Compounds in Drinking Water by Solid
Phase Extraction and Capillary Column Gas Chromatography/Mass Spectrometry. National
Exposure Research Laboratory, Office of Research and Development. U.S. EPA. Cincinnati,
Ohio 45268.
c) EPA Method 524.2. A. Alford-Stevens, J. W. Eichelberger, W. L. Budde, R. W. Slater, J. W.
Munch and T. A. Bellar. Measurement of Purgeable Organic Compounds in Water by Capillary
Column Gas Chromatography/Mass Spectrometry. National Exposure Research Laboratory,
Office of Research and Development. U.S. EPA. Cincinnati, Ohio 45268.
17. a) S. D. Winslow, B. Prakash, M. M. Domino, B. V Pepich and D. J. Munch. Environ. Sci.
Technol. 2001, 35, 1851-1858. b) S. D. Winslow, B. V Pepich M. V. Bassett, S. C. Wendelken,
J. L. Sinclair and D. J. Munch. Environ. Sci. Technol. 2001, 35, 4103-4110
18. See 40 CFR, Part 136: a) EPA Method 624. Volatile Organic Compounds by GCMS.
b) EPA Method 1624b. Volatile Organic Compounds by Isotope Dilution GCMS.
c) T. A. Bellar and J. J. Lichtenberg. "Semi-Automated Headspace Analysis of Drinking Waters
and Industrial Waters for Purgeable Volatile Organic Compounds," Measurement of Organic
Pollutants in Water and Wastewater, C.E. Van Hall, editor, American Society for Testing and
Materials, Philadelphia, PA. Special Technical Publication 686, 1978.
29
-------
APPENDIX 1. % Recoveries for Target Analytes from 20 Randomly Selected OPR
Samples by Study Phase
Phase 1 and 2 OPRs
Compound
Gemfibrozil
Cimetidine
d6-Norgestrel
de-Thiabendazole
Diltiazem
Sarafloxacin
Sulfamethizole
Digoxigenin
Norfloxacin
Trimethoprim
Ormetoprim
Ciprofloxacin
17 alpha-Estradiol
Cotinine
Warfarin
Estrone
Virginiamycin
d4-17 alpha-Ethinyl-Estradiol
Sulfamethizole
Albuterol
Diltiazem
d4-17 alpha-Ethinyl-Estradiol
Albuterol
Chlortetracycline (CTC)
Equilenin
4-Epitetracycline (ETC)
Ibuprofen
4-Epioxytetracycline (EOTC)
da-Albuterol
Anhydrochlortetracycline (ACTC)
Testosterone
Penicillin G
Progesterone
1,7-Dimethylxanthine
13C2-Erythromycin-H2O
Caffeine
4-Epichlortetracycline (ECTC)
4-Epianhydrotetracycline (EATC)
dg-Progesterone
Enrofloxacin
% Recovery
95
70.5
74.7
92.6
60.6
138
76.6
97.2
128
95.5
120
119
136
113
113
121
57.1
78.9
81.5
114
66.2
90.3
110
99.2
129
113
104
125
105
81.6
93.5
99.1
127
122
65.4
83.6
80.7
74.8
57.2
110
Phase 3 OPRs
Compound
Sulfamethoxazole
Erythromycin-H2O
Tetracycline (TC)
Triclocarban
Ranitidine
Mestranol
Lomefloxacin
Estrone
Virginiamycin
Digoxin
Diphenhydramine
Diphenhydramine
Lincomycin
Oxacillin
Androsterone
Diltiazem
Oxytetracyclin (OTC)
Oxytetracyclin (OTC)
Penicillin G
Oxytetracyclin (OTC)
% Recovery
94
97.4
94.9
99.2
54.2
102
458
98.3
147
106
103
124
58.8
120
95.7
104
105
85.5
130
88.2
Phase 4 OPRs
Digoxin
Sulfadimethoxine
Demeclocycline
Anhydrotetracycline (ATC)
17 alpha-Ethinyl-Estradiol
Sulfachloropyridazine
Norfloxacin
Estrone
Oxacillin
Cefotaxime
Clarithromycin
Oxytetracyclin (OTC)
Caffeine
Chlortetracycline (CTC)
4-Epianhydrotetracycline (EATC)
Sulfamethoxazole
Trimethoprim
Sarafloxacin
Oxolinic Acid
Doxycycline
80.9
100
99.3
120
120
106
133
111
100
130
120
92.8
75.9
110
98
107
104
130
112
88.2
Note: Phase 1 and 2 are shown together as they were performed simultaneously.
30
-------
APPENDIX 2. Data for Assorted Miscellaneous PPCPs
1
ro
o 0.8-
° OR
N °'6 "
O
z
Biosolids 4°C
!|
h.
i
1
1
b
1
ui
fi
i
L
' 1 'l
1
II
Iif
k
trff no
|L "7
1 D14
D28
Misc PPCPs
Biosolids -20°C
Misc FFCFs
31
-------
FOTWBfluent4°C
g
is
/vyx///X/>>y/y>vy>V
y^y *•- 'V ° WT *w~ «*
1^
Misc PPCPs
£= 1 9 _
"ro
0)
o
0 0.8 - p
-a 4
OJ n o
.bl U.o -
"ro I
1 0.4-
z
0.2 -
|
J
, EL
lT~
T 111]
y
1 1
1
POTWBfluent-20°C
ffT rTr JlTl^rllrT
"ni
I'-MrWL
'
. .
"
i -J I I Wi T j
I ft j
f - It
li '
1
rr no
< J •?
,T [ [ 1
fl a 14
028
Misc PPCPs
32
-------
Biosolids 4°C
os-
^
Misc PPCPs
Biosolids -20° C
DO
• 7
D 14
D28
&
Misc PPCPs
33
-------
POTW Effluent 4° C
Misc PPCPs
POTW Effluent -20°C
Misc PPCPs
34
-------
Data for "best combination samples"
1.4
POTW Effluent, 0.5 mg/L CL2, Ascorbic Acid, 4°C
Misc PPCPs
o
•*=
ro
i=
(D
O
c
o
O
T3
(D
N
"ro
E
POTW Effluent, 0.5 mg/L CL2, Ascorbic Acid, 4°C
Misc PPCPs
35
-------
POTW Effluent, 0.5 mg/L CL2, Ascorbic Acid, 4°C
ro
'E
o
c
o
O
Ij
"ro
B Lactam & Other Antibiotics
POTW Effluent, 0.5 mg/L CL2, Ascorbic Acid, 4°C
o
'•a
0)
o
c
o
O
ro
no
D4
D7
D14
Macrolide Antibiotics
36
-------
POTW Effluent, 0.5 mg/L CL2, Ascorbic Acid, 4°C
o
'ra
'c
-------
FOTW Effluent, 0.5 mg/L CL2, Ascorbc Acid, 4°C
1.4 -
1.2 -
0.8 -
0.6 -
0.4 -
0.2 -
0 -
t
rtr-
DO
04
07
014
Steroids & Hormones
38
------- |