Chloroform Formation from Swimming Pool Disinfection:

A Significant Source of Atmospheric Chloroform in Phoenix?

by

Christy J. Rose

A Thesis Presented in Partial Fulfillment

of the Requirements for the Degree

Master of Science

Approved November 2014 by the

Graduate Supervisory Committee:

Pierre Herckes, Chair

Matthew Fraser

Mark Hayes

Paul Westerhoff

ARIZONA STATE UNIVERSITY

December 2014

i

ABSTRACT

Chloroform (CHCl3) is an important atmospheric pollutant by its direct health

effects as well as by its contribution to photochemical smog formation. Chloroform

outgassing from swimming pools is not typically considered a source of atmospheric

CHCl3 because swimming pools are scarce compared to other sources. However, large

urban areas in hot climates such as Phoenix, AZ contain a substantial amount of

swimming pools, potentially resulting in significant atmospheric fluxes. In this study,

CHCl3 formation potential (FP) from disinfection of swimming pools in Phoenix was

investigated through laboratory experiments and annual CHCl3 emission fluxes from

swimming pools were estimated based on the experimental data.

Swimming pool water (collected in June 2014 in Phoenix) and model

contaminants (Pharmaceuticals and Personal Care Products (PPCPs), Endocrine

Disrupting Compounds (EDCs), artificial sweeteners, and artificial human waste

products) were chlorinated in controlled laboratory experiments. The CHCl3 production

during chlorination was determined using Gas Chromatography-Mass Spectrometry (GC-

MS) following solid-phase microextraction (SPME). Upon chlorination, all swimming

pool water samples and contaminants produced measureable amounts of chloroform.

Chlorination of swimming pool water produced 0.005-0.134 mol CHCl3/mol C and

0.004-0.062 mol CHCl3/mol Cl2 consumed. Chlorination of model contaminants

produced 0.004-0.323 mol CHCl3/mol C and 0.001-0.247 mol CHCl3/mol Cl2 consumed.

These numbers are comparable and indicate that the model contaminants react similarly

to swimming pool water during chlorination. The CHCl3 flux from swimming pools in

Phoenix was estimated at approximately 3.9-4.3 Gg/yr and was found to be largely

ii

dependent on water temperature and wind speed while air temperature had little effect.

This preliminary estimate is orders of magnitude larger than previous estimates of

anthropogenic emissions in Phoenix suggesting that swimming pools might be a

significant source of atmospheric CHCl3 locally.

iii

For my Family

iv

ACKNOWLEDGMENTS

This material is based upon work supported by the National Science Foundation under

grant number BCS-1026865, Central Arizona Phoenix Long-Term Ecological Research

(CAP LTER). I would like to thank my advisor, Dr. Pierre Herckes and my committee

members, Dr. Matthew Fraser, Dr. Mark Hayes, and Dr. Paul Westerhoff for their

guidance in completing this work. I am indebted to the Herckes group, especially Aurelie

Marcotte, Denise Napolitano, Joana Sipe, and Jinwei Zhang, for their assistance in

sample collection, instrument troubleshooting, and endless encouragement. I also thank

Natalia Fischer, David Hanigan, Heather Stancl and other members of the Westerhoff

group for their help in analyzing water quality parameters.

v

TABLE OF CONTENTS

Page

LIST OF TABLES ........................................................................................................ vii

LIST OF FIGURES ...................................................................................................... viii

CHAPTER

1 INTRODUCTION................. ............................................................................... 1

1.1 Project Motivation and Overview .................................................... 1

1.2 Water Disinfection and DBPs ......................................................... 2

1.3 Organic Contaminants in Swimming Pools ..................................... 3

1.4 Swimming Pool Disinfection and Water Chemistry ......................... 4

1.5 Health Risks of Exposure to DBPs in Swimming Pools ................... 6

1.6 Atmospheric CHCl3: Sources and Sinks .......................................... 8

1.7 Ambient CHCl3 in Phoenix ............................................................10

1.8 Investigations of CHCl3 Formation ................................................11

2 METHODOLOGY...............................................................................................13

2.1 Reagents .......................................................................................13

2.2 Sample Collection and Characterization .........................................13

2.3 Chlorination of Swimming Pool Water...........................................14

2.4 Chlorination of Swimming Pool Contaminants ...............................15

2.5 Extraction and GC-MS Analysis ....................................................16

3 RESULTS AND DISCUSSION ..........................................................................18

3.1 Swimming Pool Water Parameters .................................................18

3.2 Confirmation of Reaction Duration and Methods............................20

vi

CHAPTER Page

3.3 CHCl3 Formation from Chlorination of Swimming Pool Water

Samples ......................................................................................22

3.4 CHCl3 Formation from Chlorination of Swimming Pool

Contaminants ..............................................................................24

3.5 CHCl3 Flux ...................................................................................26

4 SUMMARY AND OUTLOOK............................................................................33

REFERENCES....... ........................................................................................................36

vii

LIST OF TABLES

Table Page

3.1: Water Quality Parameters of Collected Swimming Pool Samples aWang et al.,

2013 bWesterhoff et al., 2005.

cYoon et al., 2006.

dLee and Westerhoff, 2007.

ePehlivanoglu-Mantas and Sedlak, 2008.

fCDC.gov

gSalter and Langhus, 2007.

hAiking et al., 1994.

iWeaver et al., 2009

jWeng and Blatchley, 2011. ............. 18

3.2: Statistics of Swimming Pools in Phoenix (Forrest and Williams, 2010) ............ 29

viii

LIST OF FIGURES

Figure Page

1.1: 2012 Average Annual Ambient CHCl3 Concentrations (EPA Air Toxics Data,

2012) ............................................................................................................. 10

2.1: Schematic of Reaction Vial with Cl2 and CHCl3 Sampling Set-Up .................... 15

2.2: Temperature Program for GC-MS Analysis of CHCl3. ...................................... 17

3.1: Kinetics of CHCl3 Formation From Chlorination of Resorcinol (●) and

Oxybenzone (●) in This Study. Chlorination of Resorcinol by Chaidou et al.

(1999) (—) and Oxybenzone by Duirk et al. (2013) (—) Show Comparable

Completion Timing ........................................................................................ 21

3.2: CHCl3 amd Cl2 Concentrations During Chlorination of Oxybenzone ................. 22

3.3: Molar Yield of CHCl3 per mol C from Chlorination of Collected Swimming Pool

Water. Error Bars Indicate the Standard Deviation of Triplicate Experiments.

Shaded Region indicates 0.01 mol CHCl3/mol C as Measured by Lee et al.

(2007). ........................................................................................................... 23

3.4: Molar Yield of CHCl3 per mol Cl2 Consumed from Chlorination of Collected

Swimming Pool Water. Error Bars indicate the Standard Deviation of Triplicate

Experiments. Shaded Region Indicates 0.039 mol CHCl3/mol Cl2 as Measured

by Weng and Blatchley (2011). ..................................................................... 23

3.5: Molar Yield of CHCl3 per mol C in Precursor from Chlorination of Suspected

Swimming Pool Contaminants. Error Bars Indicate the Standard Deviation of

Triplicate Experiments. *Error Bars Indicate Standard Deviation of Two

ix

Figure Page

Measurements. Shaded Region Indicates Maximum of 0.134 mol CHCl3/mol C

Found from Chlorination of Swimming Pools Samples in This Study. ............. 24

3.6: Molar Yield of CHCl3 per mol Cl2 Consumed from Chlorination of Suspected

Swimming Pool Contaminants. Error Bars Indicate the Standard Deviation of

Triplicate Experiments. *Error Bars Indicate Standard Deviation of Two

Measurements. Shaded Region Indicates Maximum of 0.061 mol CHCl3/mol Cl2

Found from Chlorination of Swimming Pools Samples in This Study. ............. 25

3.7: Daily, Monthly, and Annual CHCl3 Flux from Swimming Pools in Phoenix

Calculated with Wind Speed and Air Temperature Data for 2013 While Holding

Water Temperature Constant. ......................................................................... 29

3.8: Daily, Monthly, and Annual CHCl3 Flux from Swimming Pools in Phoenix

Calculated with Average Annual Wind Speed, Varying Air Temperature and

Water Temperature for 2013. .......................................................................... 30

3.9: Swimming Pool Emissions of CHCl3 in Phoenix compared to estimated natural

and anthropogenic emission estimates.. ........................................................... 31

1

CHAPTER 1

INTRODUCTION

1.1 Project Motivation and Overview

Chloroform (CHCl3) is a well documented disinfection by-product (DBP) of

water chlorination (Richardson et al., 2010). Within the scope of atmospheric pollution,

CHCl3 is of importance due to its direct health effects as well as its contribution to

photochemical smog formation. Natural sources of CHCl3 such as vegetation, soil and

ocean processes account for 90% of global CHCl3 emissions (Khalil and Rasmussen,

1999); however, most human exposure occurs in urban settings. Locally, major sources

of CHCl3 in the urban atmosphere are industrial operations such as pharmaceutical and

refrigerant manufacturing as well as landfill processing. Chloroform outgassing from

swimming pools is not typically considered a source of atmospheric CHCl3 because

swimming pools are scarce compared to other sources. However, urban areas in hot

climates such as Phoenix generally contain a substantial number of swimming pools per

capita. Arizona ranks fourth in the nation for total number of swimming pools and second

in swimming pools per capita with 4.8 swimming pools per 100 people (P.K. Data,

2013). The temperate regions of Arizona limit swimming pools to the central desert

region whereas swimming pools in other states are generally distributed throughout the

state. Additionally, states with more swimming pools than Arizona benefit from coastal

breezes that help clear the air of pollutants. It is possible that the numerous chloroform-

emitting pools within a valley such as Phoenix are as a sum responsible for a substantial

portion of the atmospheric chloroform in urban air. This study investigated CHCl3

formation potential (FP) from chlorination of swimming pool water collected in Phoenix,

2

AZ and suspected swimming pool contaminants to determine maximum CHCl3

production. FP was used to estimate CHCl3 flux from swimming pools in Phoenix to

determine total annual emissions.

1.2 Water Disinfection and DBPs

Disinfectants are necessary to inactivate biological contaminants in water such as

algae, viruses and bacteria. The most common chemical disinfection methods include

treatment with chlorine and chloramines. Chlorine is a very effective disinfectant and is

therefore used in heavily contaminated waste waters. If wastewater is intended for return

to consumers as non-potable irrigation water, as is increasingly common in light of water

conservation efforts, further purification such as filtration or adsorption with activated

carbon is required (Redman et al., 2001). Chloramination is preferred in drinking water

disinfection because it tends to produce less odor and taste-causing byproducts

(Boorman, 1999). Chlorination and bromination are the most common disinfectants used

in swimming pools. Disinfectants work by oxidizing the cell membranes of biological

contaminants. Because of these strong oxidative properties, organic contaminants are also

oxidized resulting in DBPs; however these DBPs are not generally removed from the

water.

DBPs were first recognized in 1974 by Rook et al. and public interest soon

followed. Despite a strong interest in understanding and extensive investigation of DBP

production, much is still unknown. The major contribution to this gap in knowledge is the

large variability in precursors resulting in a seemingly infinite number of DBP species

and therefore a tedious process in identifying and characterizing these DBPs. Several

3

classes of disinfection byproducts have been reported. Chlorination of drinking water has

been found to produce trihalomethanes (THMs), haloacetic acids (HAAs),

haloacetonitriles (HANs) and haloketones while ozonation has been found to produce

aldehydes and brominated compounds (Boorman, 1999). Over 500 DBPs have been

detected in chlorinated waters, but 50% of organic halide mass remains uncharacterized.

Other studies have identified several hundred DBPs in swimming pools alone (Kim et al.,

2002; Zweiner et al., 2007; Richardson et al., 2010).

1.3 Organic Contaminants in Swimming Pools

Organic contaminants commonly found in swimming pools are similar to those

found in waste water and surface water. These include human waste products such as

sweat and urine, Pharmaceuticals and Personal Care Products (PPCPs) such as lotions

and perfumes, and Endocrine Disrupting Compounds (EDCs) such as synthetic hormones

from oral contraceptives.

PPCPs and EDCs are of particular interest because of their ability to affect

ecosystem health and their potential to mimic or block hormones. Medicinal metabolites

such as 1,7-dimethylxanthine (caffeine metabolite) and 17α-ethinylestradiol (oral

contraceptive metabolite) are excreted and flushed down the toilet which must then be

treated at waste water treatment plants. Items such as sunscreen, lotions and perfumes are

released into the freshwater supply during swimming or bathing. Although several studies

have shown that these contaminants pose no immediate human health risks, long-term

effects are not known (Christensen, 1998; Schulman et al., 2002; Webb et al., 2003).

Swimming pools are different from surface water in that the water is continuously cycled

4

and rarely replenished, causing a build-up of contaminants. DEET, an insect repellant,

and caffeine were found in swimming pools at concentrations of 100~2100 ng/L and

50~500 ng/L, respectively (Weng et al., 2014), compared to 2~20 ng/L in various

Arizona ground and surface waters (Chiu and Westerhoff, 2010).

Ongoing efforts to reduce occurrences of contaminants and DBPs in swimming

pools focus on preventing introduction of precursors into waters. Public pools often

require swimmers to shower before entering the pool to remove body care products such

as lotions and perfumes as well as body oils and sweat. Additionally, the Center for

Disease Control (CDC) recommends showering with soap before engaging in water

activities to reduce introduction of recreational water illness (RWI)-causing pathogens

that can survive several days even in well-maintained waters, such as Cryptosporidium,

Giardia, Shigella, norovirus, and E. coli (Castor and Beach, 2004).

DBP removal from swimming pools generally occurs only through outgassing of

volatile compounds at the surface. Swimming pool water is not well-mixed in unused

pools; therefore conditions vary in vertical profiles due to temperature gradients and UV

penetration in addition to out-gassing at the surface (Lian et al., 2014). Turbulence from

splashing can bring more DBPs to the water’s surface causing an increase in outgassing

of DBPs and therefore increased exposure rates in swimmers (Aggazzotti et al., 1995).

1.4 Swimming Pool Disinfection and Water Chemistry

Chlorination is the most common type of swimming pool disinfection because it

is the cheapest and most effective method. When chlorine gas is used, Cl2 hydrolyzes in

5

water to form hypochlorous acid (HOCl) (Equation 1.1) which then dissociates to

hypochlorite (OCl-) (Equation 1.2).

Cl2 (g) + H2O (l) ↔ HOCl (aq) + HCl (aq) pKeq=-4.60 Equation 1.1

HOCl ↔ OCl- (aq) + H

+ (aq) pKa=7.54 Equation 1.2

Hypochlorous acid is more often added directly to pools as aqueous sodium hypochlorite

(NaOCl), commonly referred to as liquid chlorine or bleach, or solid calcium

hypochlorite (Ca(OCl)2) via tablets. Speciation of hypochlorous acid is largely pH

dependent. Below pH ~5, hypochlorous acid exists primarily as hypochlorite. Between

pH 5 and pH 8, HOCl and OCl- exist in similar concentrations and above pH 8,

hypochlorous acid exists primarily as the protonated form. Hypochlorite, referred to as

active chlorine, is the reactive species responsible for disinfection whereas HOCl,

referred to as available chlorine, is known to have no bactericidal properties but rather

serves as a reservoir to replace OCl- as it is consumed (Black et al., 1970). Rate of

formation of DBPs has been found to be largely pH dependent due to this speciation. The

recommended range for pH in swimming pools is 7.2-7.8. Within this range, chlorine is

most effective as a disinfectant without irritating eyes and nasal linings (CDC.gov).

Chlorine is measured as chlorine residual, which is expressed in two forms: free

and total chlorine. Free chlorine is a measure of the disinfection power of the chlorine in

water and includes both OCl- and HOCl. Because chlorine exists in several forms,

representation is standardized as mg/L Cl2 and refers to all available forms of reactive

chlorine. Hypochlorous acid, OCl- and Cl2 are considered available forms while chloride

ion (Cl-) is not reactive and therefore is not considered available chlorine. Combined

chlorine is a measure of the concentration of chlorine present in non-reactive compounds

6

such as DBPs. Total chlorine refers to the sum of free and combined chlorine in the

sample.

Spectrophotometric measurement of free and total chlorine using N,N-diethyl-p-

phenylenediamine (DPD) as an indicator is well-outlined in EPA Method 334.0

(Wendelken et al., 2009) and is similarly used in waste water, drinking water, and

swimming pool water testing (Moore et al., 1984). Free chlorine reacts with DPD to form

a magenta solution. In the case of total chlorine, potassium iodide is added to liberate any

combined chlorine to form free chlorine that then can then react with DPD.

Several other chemicals are also added to swimming pool water to maintain

adequate water chemistry. Bicarbonate buffer is added to maintain pH during heavy use.

Cyanuric acid is added as a stabilizer to protect chlorine from UV degradation. Algaecide

is added to treat or prevent algae growth. Copper-containing algaecides are most effective

but can cause staining in pools and are only used in cases of heavy algae growth. Alkyl

dimethyl benzyl ammonium chlorine, commonly called quat algaecide (referring to the

quaternary ammonium functional group) is cheaper and does not stain pools and therefore

is the most commonly used algaecide (Rowhani and Lagalante, 2007).

1.5 Health Risks of Exposure to DBPs in Swimming Pools

Although swimming is often applauded as a low-impact fitness activity and fun

pastime, exposure to DBPs in swimming pool water can pose various health risks. The

most common complaint from swimmers is of irritation to mucus membranes in the eyes

and nose. This irritation is caused by exposure to chloramines, which are responsible for

the distinctive “chlorine” smell of pools (Hery et al., 1995). Prolonged breathing of

7

indoor swimming pool air can also cause irritation in lungs and is blamed for adolescent

and occupational asthma symptoms (Font-Ribera et al., 2009; Jacobs et al., 2007;

Thickett et al., 2002). Chloramines are formed by chlorination of nitrogen-containing

compounds which are often introduced to the pool as biological waste products such as

sweat and urine.

Chloroform as a DBP is of particular interest because of its known carcinogenic

properties. A positive correlation has been found between bladder cancer mortality rates

and CHCl3 concentrations in drinking water (Cantor et al., 1978; Hogan et al., 1979).

High doses of CHCl3 inhalation have also been found to induce liver cancer in mice

(Larson et al., 1996). Cancer risk from dermal exposure to CHCl3 while swimming was

found to be substantially higher than gastrointestinal or dermal exposure to CHCl3 in tap

water (Panyakapo et al., 2008).

Inhalation and dermal exposure to CHCl3-containing waters such as when

swimming leads to higher blood levels than does oral exposure such as drinking

chlorinated water (Aggazzotti et al., 1993; Weisel and Jo, 1996; Kuo et al., 1998). In

adults, CHCl3 in exhaled air was found to be 6.8 times higher after 40 minutes of indoor

swimming (Kogevinas et al., 2010). Elevated CHCl3 concentrations in blood plasma and

exhaled air were detected up to 10 hours after exposure (Aggazzotti et al., 1990). These

symptoms are thought to be exacerbated in indoor swimming pools compared to outdoor

pools due to a build-up of volatile chemicals in indoor pool air despite required

ventilation (Aiking et al., 1994). In outdoor swimming pools, these volatile chemicals are

readily dissipated from the surface of the water by wind, resulting in decreased exposure.

Chloroform concentrations in blood and urine were found to be lower when swimming in

8

outdoor pools compared to indoor pools (Aiking et al., 1994). However, this dissipation

can potentially result in elevated atmospheric concentrations.

1.6 Atmospheric CHCl3: Sources and Sinks

Atmospheric CHCl3 is important as a main transporter of chlorine through the

troposphere, however global budgets show large missing sources (Rhew et al., 2008a).

Total CHCl3 flux through the environment is estimated to be 660±220 Gg∙yr-1

of which

approximately 90% is from natural sources (McCulloch, 2003). The largest natural

sources of CHCl3 are from biological processing in ocean water and soil. Ocean algae

and seaweed have been reported to produce CHCl3 in measureable concentrations

(McCulloch, 2003; McConnell and Fenical, 1997; Nightingale et al., 1995). Emissions

from soil were found to much smaller in dry soils such as in deserts than moist soils such

as in rain forests or summer tundra (Rhew et al., 2008b).

Several anthropogenic sources of CHCl3 are known. Anaerobic decomposition

such as in landfills and feedlot waste has been reported to produce significant amounts of

CHCl3 (Aucott et al., 1999). Pharmaceutical manufacturing and disposal are also known

to utilize large amounts of CHCl3 as solvent. However, paper mills are by far the largest

source of anthropogenic CHCl3 emissions as CHCl3 is a known by-product of chemical

decomposition of wood pulp and bleaching of paper with chlorine. In 1990, paper

manufacturing and pulp processing plants were responsible for release of 9.97 Gg CHCl3

to the environment in the United States or 90% of all reported CHCl3 releases (Aucott et

al., 1999). Human exposure to CHCl3 primarily occurs in urban settings from

anthropogenic sources. Potable water is a main source of exposure through drinking and

9

showering in water containing CHCl3 (Aucott et al., 1999; Kuo et al., 1998; Richardson

et al., 2007). Drinking water and waste water treatment plants generated an estimate 2

Gg/yr (Aucott et al., 1999). Use of bleach and other chlorine-containing cleaners also

introduces exposure potential. CHCl3 concentrations were found to exceed 1 mg/L in

washwater of residential washing machines during laundering with bleach (Shepherd et

al., 1996).

Industrial emissions of several toxic chemicals are monitored by the

Environmental Protection Agency (EPA) via the Toxic Release Inventory (TRI) Program.

Under this program, facilities are required to report annual disposal management or

release to the environment of certain toxic chemicals if the facility is in a specified

industry such as manufacturing or mining, employs 10 or more full- time employees, and

manufactures or process more than 25,000 lbs of a TRI-listed chemical or otherwise uses

more than 10,000 lbs of a TRI-listed chemical each year (EPA TRI Program, 2014). This

data can be used to provide a rough estimate for anthropogenic emissions but is limited in

that it only contains larger point sources.

Chloroform readily hydrolyzes in water; however the high volatility of CHCl3

leads to transport into the gas phase at a faster rate than hydrolysis occurs (McCulloch,

2003). Therefore, water is not considered to be an acceptable CHCl3 sink. Recent studies

report possible sinks in soils and plants through unexplained uptakes (Liu, 2006; Wang et

al., 2007). Atmospheric removal of CHCl3 is the main removal pathway from the

environment and occurs through oxidation by hydroxyl radical (OH•) (McCulloch, 2003)

10

1.7 Ambient CHCl3 in Phoenix

Figure 1.1 shows annual average ambient CHCl3 concentrations for several cities

in the United States from 2012. Seattle, WA and Cedar Rapids, IA show the highest

concentrations of CHCl3 followed by Tulsa, OK and Phoenix, AZ. Based upon TRI data,

CHCl3 sources can be identified in Seattle, Cedar Rapids and Tulsa but not in Phoenix.

The large paper mill industry in Washington surrounding Seattle leads to high emissions

of CHCl3. A large number of pharmaceutical companies in Cedar Rapids report use of

CHCl3. In Tulsa, CHCl3 several oil refineries report production of CHCl3. However, no

CHCl3 use in Phoenix has been reported to TRI since 1999. Therefore, a disparity exists

between ambient CHCl3 concentrations in Phoenix and recognized emission sources. It is

important to note that emission of CHCl3 does not always correlate directly to ambient

concentrations. For example, presence of oil refineries throughout Texas is not consistent

with high ambient CHCl3 concentrations, possibly due to meteorological factors such as

Figure 1.1: 2012 Average Annual Ambient CHCl3 Concentrations

(EPA Air Toxics Data, 2012)

11

coastal breeze clearing the air. However, anthropogenic influences are a good indicator

for presence of CHCl3 and other pollutants. Urban areas typically see air toxic

concentrations well above the one in one million cancer risk level. A 2005 study found

that average annual CHCl3 ambient concentrations in Phoenix were exceptionally high

(>75th percentile) despite a lack of reported sources (Hafner and O’Brien, 2006). The

Motorola, Inc. Superfund Site in Phoenix is an area with known chemical contamination

and CHCl3 emission from this site was investigated (North Indian Bend Wash, 2005).

Modeling estimates report 3.94 kg/yr CHCl3 is released from the superfund site and is

expected to decrease as cleanup for the site continues. Additionally, likely local

emissions sources including semiconductor manufacturing, have abandoned the use of

CHCl3 as a solvent in recent years.

Although previous studies provide estimates on anthropogenic CHCl3 emissions,

natural sources are not well characterized in Phoenix. Studies of natural CHCl3 emissions

in the desert regions of southern California can provide reasonable comparisons to

Phoenix desert emissions. Chloroform emission in dry dessert shrubland was found to be

6.6 nmol∙m-2∙day

-1 (Rhew et al., 2008a). This estimate is possibly high for Phoenix

because desert plants are sparser in Arizona than in California.

1.8 Investigations of CHCl3 Formation

Although a mechanism for formation of CHCl3 is still not known, other aspects of

CHCl3 formation have been well-studied. Chloroform rate constants from chlorination of

resorcinol change non-linearly with pH, and occur as a minimum at pH ~6 with a

maximum at pH~8 (Gallard and von Gunten, 2002a). Additionally, CHCl3 formation in

12

swimming pools has been shown to be closely related to bather load due to introduction

of precursors. One study showed diurnal variations in CHCl3 during daily heavy

swimming pool use (Weng and Blatchley, 2011). Chloroform formation is often express

as formation potential (FP) which describes the maximum formation of a product without

regard to rate of formation. This value is determined to be the concentration of product

after all reactions have reached equilibrium.

13

CHAPTER 2

METHODOLOGY

2.1 Reagents

All glassware was washed and solvent rinsed with isopropanol (Optima grade,

Fisher Scientific; Fairlawn, NJ) and baked at 450 °C for 12 hours and stored in aluminum

foil to keep them clean until used. Chloroform (99.8% purity) was purchased from

Mallinckrodt Chemicals (St. Louis, MO) and deuterated chloroform (99.96% purity) was

purchased from Cambridge Isotope Labs (Tewksbury, MA). Sodium hypochlorite (10-

12% active chlorine) and all other reagents (≥98% purity) were purchased from Sigma

Aldrich (St. Louis, MO). All water used was 18.2 MΩ∙cm from a Millipore water

filtration system (EMD Millipore; Billerica, MA). Stock buffer solution was made using

monohydrogen sodium phosphate (HNa2PO4) and dihydrogen sodium phosphate

(H2NaPO4) and pH was adjusted to 7.2 using 10 N sulfuric acid (H2SO4) and 3 M sodium

hydroxide (NaOH).

2.2 Sample Collection and Characterization

Swimming pool samples were collected in June 2014 from outdoor pools

throughout the Phoenix metro area. Samples were collected at approximately 30-40 cm

below the water’s surface into 1 L high-density polyethylene (HDPE) bottles that had

been cleaned and solvent-rinsed with isopropanol. Samples were immediately stored

under refrigeration at 4°C. Prior to use, all pool samples were filtered with pre-baked

quartz fiber filters (0.7 μm, Whatman, Piscataway, NJ). Sample pH was measured using

an Accumet (Hudson, MA) AP62 portable pH meter calibrated with pH 4, 7 and 10

14

buffers. UV absorbance at 254 nm was performed with a Shimadzu (Columbia, MD)

PharmaSpec UV-1700 UV-Vis Spectrophotometer in photometric mode. Chlorine

residual was measured spectrophotometrically as free and total chlorine according to

EPA Method 344 (Wendelken et al., 2009) using commercially prepared DPD reagent

pillows (Hach, Loveland, CO). Absorbance of DPD was measured at 514.6 nm using a

handheld Vernier (Beaverton, OR) LabQuest with SpectroVis Plus attachment.

Conductivity was measured using a VWR (Radnor, PA) Model 2052 digital conductivity

meter. Dissolved organic carbon (DOC) and total dissolved nitrogen (TDN) were

measured by Natalia Fischer of the Westerhoff group (ASU) using a Shimadzu Total

Organic Carbon (TOC)-VCSH after acidifying and purging with carbon-free oxygen.

Dissolved organic nitrogen (DON) was calculated as the difference between TDN and

dissolved inorganic nitrogen (TDN) (ammonia, nitrate, and nitrite). Ammonia, nitrate,

and nitrite measurements were performed spectrophotometrically using commercially

prepared Hach reagent sets with absorbance measured in a Hach DR 5000

spectrophotometer. Bromide was measured using a Dionex (Sunnyvale, CA) DX-120 ion

chromatograph (IC) system by Heather Stancl of the Westerhoff group (ASU).

2.3 Chlorination of Swimming Pool Water

Pool samples were mixed with 1 M phosphate buffer at pH 7.2 to achieve a 0.5 M

phosphate buffer solution. Previously measured DOC was used to determine chlorine

demand of pool water and chlorine (NaOCl, 10-15% active chlorine) was added at 100x

excess. Deuterated chloroform (CDCl3) was added as an internal standard, free and total

chlorine were measured, and pH was recorded. The sample solution was transferred to an

15

amber glass reaction vial and capped with a polytetrafluoroethylene (PTFE) septum cap

within 30 seconds of addition of chlorine. Initial CHCl3 concentration was measured. The

septum allowed liquid sample retrieval via syringe and headspace extraction while

maintaining a gastight system. Reactions were stored at constant room temperature (22

°C) for at least 85 hours. Then final concentrations of CHCl3, chlorine residual and pH

were measured. Figure 2.1 shows a schematic of the reaction vial and sampling set-up.

Figure 2.1: Schematic of Reaction Vial with Cl2 and CHCl3 Sampling Set-Up.

2.4 Chlorination of Swimming Pool Contaminants

Chlorination of various precursors (PPCPs, EDCs, artificial sweeteners, and

artificial human waste products) in simulated swimming pool water were investigated to

determine possible contributions to CHCl3 FP. Precursors were chosen because of known

or suspected occurrence in swimming pool water or detection in Arizona surface water

(Chiu and Westerhoff, 2010). Reactions were performed in 0.5 M phosphate buffer at pH

7.2. Precursors were dissolved in dimethylsulfoxide (DMSO) then added at ~1 mg/L.

16

Precursor concentrations in this study were chosen at higher concentrations than those

found in surface water due to instrument sensitivity. Chlorine was added at 100x excess

and CDCl3 was added as an internal standard. A reaction blank using DMSO was also

performed. Free and total chlorine was measured, pH was recorded, and the reaction

solution was placed in an amber glass reaction vial then capped within 30 sections of

chlorine addition. Reactions were stored at 22 °C. Final CHCl3 concentration, chlorine

residual and pH were measured after at least 85 hours.

2.5 Extraction and GC-MS Analysis

Chloroform was extracted by 15 minute headspace extraction via solid phase

microextraction (SPME) using a 100 μm film-coated polydimethylsiloxane (PDMS) fiber

assembly with manual holder (Supelco, Bellefonte, PA). After extraction, an Agilent

(Santa Clara, CA) 6890N Gas Chromatograph (GC) coupled with Agilent MSD 5973

Mass Spectrometer (MS) in electron impact (EI) ionization mode was used to isolate and

determine CHCl3. Separation was performed with a SPB-1 fused silica capillary column

(30 m x 0.25 mm x 0.25 μm, Supelco) with helium as the carrier gas at 1.2 mL/min after

splitless injection at inlet temperature 230 °C. The total analysis run time was 5 minutes,

with constant temperature of 40 °C for 2 min then temperature ramp at a rate of 20

°C/min until 100 °C was achieved. Figure 2.2 shows a graphical representation of the

temperature program. Chloroform elution occurred at approximately 2 minutes with

additional temperature ramp to remove impurities on SPME fiber. Deuterated chloroform

was added to all samples as an internal standard to determine instrument response and

calculate CHCl3 concentration. Mass-to-charge ratios of 83 and 84 give the strongest

17

signals for CHCl3 and CDCl3, respectively. A ratio of peak areas for m/z=83 and 84 was

used to form a calibration curve.

Figure 2.2: Temperature Program for GC-MS Analysis of CHCl3.

20

40

60

80

100

120

0 1 2 3 4 5

tem

pera

ture

(°C

)

time (min)

18

CHAPTER 3

RESULTS AND DISCUSSION

3.1 Swimming Pool Water Parameters

Table 3.1 shows the water quality parameters of collected pool samples. Pools

were categorized into three types: private, community, and public pools. These categories

can be used to describe bather load of the swimming pool to estimate the rate at which

contaminants are introduced into a pool. Private pools have tens of bathers per season,

and might not be used daily. Community pools such as those in apartment complexes or

country clubs may have hundreds of bathers per season, while public pools might see as

many as 1000 bathers per season.

DOC was measured to determine total organic contaminants in the water. All

pools except pool 6 contained DOC concentrations above the expected 2-13 mg/L DOC

for swimming pools (Zweiner et al., 2007; Kanan and Karanfil, 2011; Wang et al., 2013).

This might be attributed to potentially high use of alkyl-containing algaecides during

summer months. Haboobs in Phoenix kick up iron-rich dust which is then deposited

throughout the city, including into outdoor swimming pools. Iron is known to promote

pool # pool type

DOC

(mg/L)

SUV254

(L∙mg-1∙m-1)

DON

(mg/L) pH

free Cl2

(mg/L)

total Cl2

(mg/L)

Conductivity

(mS/cm)

CHCl3

(μM)

1 private 41 0.07 52 7.50 0.03 0.15 2.01 4.04

2 private 49 0.06 62 7.32 8.7 9.2 2.37 0.11

3 community 69 0.10 80 7.47 19.6 25.7 2.19 6.85

4 community 138 0.06 195 7.27 8.3 9.4 3.81 1.47

5 community 92 0.03 118 7.08 3.0 3.4 3.41 0.74

6 public 2.1 0.75 10 7.16 15 15 4.72 0.07

expected: 2-13a 1-6b,c,d 2-14e 7.2-7.8f 1-3g <3h 0.15-0.21h,i,j

Table 3.1: Water Quality Parameters of Collected Swimming Pool Samples. aWang et al.,

2013 bWesterhoff et al., 2005.

cYoon et al., 2006.

dLee and Westerhoff, 2007.

ePehlivanoglu-Mantas and Sedlak, 2008.

fCDC.gov

gSalter and Langhus, 2007.

hAiking et

al., 1994. iWeaver et al., 2009

jWeng and Blatchley, 2011.

19

rapid algal growth (Griffin and Kellogg, 2005), therefore algaecides are added as

preventative maintenance. Addition of alkyl-containing algaecides might result in high

DOC measurements. Pool Time (Lawrenceville, GA) preventative algaecide, advertises

20% w/w alkyl (50% C14, 40% C12 and 10% C10) dimethyl benzyl ammonium chloride as

the active ingredient. Weekly addition of 3 fl. oz. per 10,000 gallons as directed would

result in 4.15 mg/L DOC added per week. For an average swimming season of 27 weeks

(Forrest and Williams, 2010), algaecides would contribute 112 mg/L DOC per swimming

season.

SUV254 is calculated by Equation 3.1 and gives a measure of aromatic content to

help determine degree of natural organic matter. SUV254 was lower than expected which

indicates low aromaticity and high non-aromatic carbon. This is consistent with frequent

alkyl-containing algaecide use.

Equation 3.1

DON was measured to determine biological matter in pool samples. Ammonia

and nitrite were not present in any pool samples while nitrate was present in all samples

in small concentrations (<9mg/L). DON was higher than expected in all pools. This

might be due to the use of alkyl algaecides, which contain quaternary ammonium centers,

or organic stabilizers such as cyanuric acid (C3H3N3O3).

Measurements of pH in all pools were close to or within the recommended range

of 7.2-7.8 and were consistent with pH ranges found in other studies (Wang et al., 2013).

Free and total Cl2 were above the recommended values of 1-3 mg/L, respectively,

for all pools except pool 1 and 5. High Cl2 dosing is typical during seasons of high use to

eliminate the need for frequent monitoring.

20

Conductivity was measured to determine total inorganic content in pool samples.

These species include any ions in the water such as Na+, Ca

2+ and Cl

-. Pools 1-3 were

within the expected range of <3 mS/cm. Above this value, water chemistry becomes

difficult to maintain but swimming comfort is not affected.

Chloroform was present in all pools at the time of collection with an average 2.21

μM. Chloroform concentrations in pools 2 and 6 were similar to concentrations found in

other swimming pools (Aiking et al., 1994; Weaver et al., 2009; Weng and Blatchley et

al., 2011). All other pools were higher but still below the recommended 8 μM limit.

Bromide measurements were performed but bromide was not detected. This was

expected as bromide levels are generally very low except in brominated pools and spas.

3.2 Confirmation of Reaction Duration and Methods

Resorcinol and oxybenzone were used as model compounds to determine reaction

time to reach completion of CHCl3 production. This point was determined as the point at

which the rate of CHCl3 formation significantly decreased. Resorcinol displayed

completion within 1 hour (Figure 3.1) and is consistent with findings from previous

studies which report a range of 1 to 10 hours (Gallard and von Gunten, 2002a; Gallard

and von Gunten, 2002b; Blatchley et al., 2003; Chang et al., 2006). Oxybenzone

displayed completion after approximately 85 hours whereas previous studies have shown

that CHCl3 formation reached completion after 24 hours in a pH range of 6-8 and a 25 °C

for oxybenzone (Duirk et al., 2013). This difference might be due to pH and temperature

variations. Rate constants have been shown to decrease at higher pH and lower

21

temperatures. Therefore, CHCl3 formation was allowed to occur for 85 hours before final

measurements were taken to ensure completion of CHCl3 formation had been reached.

Resorcinol displayed a molar yield of 0.251 mol CHCl3/mol resorcinol which is

typical of other molar yields for phenolic compounds with a range of 0.01-0.3 mol

CHCl3/mol phenol (Gallard and von Gunten, 2002a). This similarity with literature data

further confirms validity of these methods.

Figure 3.2 shows an increase in CHCl3 concentration as free chlorine

concentration decreases. Consistent presence of free chlorine after 85 hours confirms that

chlorine was not a limiting factor of CHCl3 FP. Limit of quantification (LOQ) of CHCl3

was determined to be 50 nM and the limit of detection (LOD) was determined to be 20

nM.

Figure 3.1: Kinetics of CHCl3 Formation from Chlorination of Resorcinol (●) and

Oxybenzone (●) in This Study. Chlorination of Resorcinol by Chaidou et al. (1999) (—)

and Oxybenzone by Duirk et al. (2013) (—) Show Comparable Completion Timing.

0

2

4

6

8

10

12

14

16

18

0 20 40 60 80 100 120 140 160 180

CH

Cl 3

(μ

M)

time (hrs)

resorcinol

resorcinol, scaled

(Chaidou et al., 1999)

oxybenzone

oxybenzone, scaled

(Duirk et al., 2013)

22

Figure 3.2: CHCl3 amd Cl2 Concentrations During Chlorination of Oxybenzone.

3.3 CHCl3 Formation from Chlorination of Swimming Pool Water Samples

Figure 3.3 shows the molar yield of CHCl3 per mole C present in the pool water.

Chlorination of pool 6 exhibited chloroform formation similar to other pools despite

containing the smallest concentration DOC, resulting in the largest molar yield at 0.134

mol CHCl3/mol C while Pools 1-5 produced less than 0.03 mol CHCl3/mol C. Molar

yield does not appear to be dependent on any of the measured water quality parameters.

A correlation is expected between DOC and CHCl3 formation as carbon should be the

limiting reactant, however this trend is not seen, suggesting the formation of other DBPs

competes with formation of CHCl3.

Figure 3.4 shows molar yield of CHCl3 per mole Cl2 consumed. Pool 6 again

showed a higher molar yield than other pools with 0.062 mol CHCl3/mol Cl2, indicating

that CHCl3 was formed while consuming less Cl2 than other pools.

110

120

130

140

150

160

0

5

10

15

0 50 100 150 200

free

ch

lori

ne

(mg

Cl 2

/L)

CH

Cl 3

(μ

M)

time (hrs)

23

Figure 3.3: Molar Yield of CHCl3 per mol C from Chlorination of Collected Swimming

Pool Water. Error Bars Indicate the Standard Deviation of Triplicate Experiments.

Shaded Region indicates 0.01 mol CHCl3/mol C as Measured by Lee et al. (2007).

Figure 3.4: Molar Yield of CHCl3 per mol Cl2 Consumed from Chlorination of Collected

Swimming Pool Water. Error Bars indicate the Standard Deviation of Triplicate

Experiments. Shaded Region Indicates 0.039 mol CHCl3/mol Cl2 as Measured by Weng

and Blatchley (2011).

24

3.4 CHCl3 Formation from Chlorination of Swimming Pool Contaminants

Figure 3.5 shows molar yield of CHCl3 per mole C chlorinated for several

suspected swimming pool contaminants. A ratio of 1 would indicate complete conversion

to CHCl3. Chlorination of urea showed the highest molar yield at 0.323 mol CHCl3/mol C

and artificial sweat showed the lowest with 0.004 mol CHCl3/mol C. Molar yields for

these model compounds are comparable to those found from chlorination of swimming

pool water which indicates that these model compounds behaved similarly to swimming

pool water when chlorinated.

Nitrogen-containing compounds are thought to result in smaller molar yields because

the fast formation of chloramines shifts by-product formation from CHCl3. Previous

studies reported a complete lack of CHCl3 upon chlorination of urea with NCl3 being the

main by-product (Blatchley and Cheng, 2010). In this study, a considerable amount of

Figure 3.5: Molar Yield of CHCl3 per mol C in Precursor from Chlorination of Suspected

Swimming Pool Contaminants. Error Bars Indicate the Standard Deviation of Triplicate

Experiments. *Error Bars Indicate Standard Deviation of Two Measurements. Shaded

Region Indicates Maximum of 0.134 mol CHCl3/mol C Found from Chlorination of

Swimming Pools Samples in This Study.

25

CHCl3 was produced from chlorination of urea. In contrast, artificial urine (urea plus

creatinine) and artificial sweat (urea plus lactic acid) showed low molar yields. Other

nitrogen-containing compounds (acetaminophen, carbemazepine, DEET, primidone, and

sulfamethoxazole) do not consistently display decreased CHCl3 yields; therefore presence

or absence of nitrogen is not the main factor in determining formation potential with

these precursors.

Low production of CHCl3 per mole Cl2 consumed (Figure 3.6) indicates high

formation of combined chlorine as other DBPs. In sucralose, the low CHCl3/C molar

ratio with high CHCl3/Cl2 molar ratio indicates that less Cl2 is consumed for consumption

of carbon when compared to other compounds. This is because sucralose is chlorinated

on three carbons, and the presence of chlorine in the compound results in less chlorine

addition needed to form a CHCl3. Overall reactivity of sucralose is expected to be low

because halogenation leads to stability of sucralose, as seen by the inability for sucralose

Figure 3.6: Molar Yield of CHCl3 per mol Cl2 Consumed from Chlorination of Suspected

Swimming Pool Contaminants. Error Bars Indicate the Standard Deviation of Triplicate

Experiments. *Error Bars Indicate Standard Deviation of Two Measurements. Shaded

Region Indicates Maximum of 0.061 mol CHCl3/mol Cl2 Found from Chlorination of

Swimming Pools Samples in This Study.

26

to be digested by the human body, hence use as a zero-calorie sweetener (Roberts et al.,

2000). Previous studies have also found sucralose to be persistent in waste water after

chlorination and UV radiation (Torres et al., 2011). However, triclosan, which also

contains chlorinated carbons, exhibited opposite behavior, with high CHCl3/C ratio and

low CHCl3/Cl2 ratios.

3.5 CHCl3 flux

A simple boundary layer model is used to estimate flux of chloroform from a pool

to the atmosphere across a water-air boundary (Schwarzenbach et al., 2003). In this

model, the flux can be evaluated by considering factors affecting activity on both sides of

the boundary. Flux across an air-water boundary, Fa/w (mol∙cm-2

∙s-1

), is calculated by

Equation 3.2

where va/w is the overall transfer velocity (cm/s) across the air-water boundary, cw is the

concentration (M) of chloroform in the water layer, ca is the concentration (M) of

chloroform in the air-side layer, and Ka/w is the dimensionless Henry’s law constant of

chloroform. The average CHCl3 concentration found from chlorination of swimming pool

water in this study, 1.7μM, was used for cw. This value was similar to concentrations

found in swimming pools at the time of collection. Values for ca were determined from

the partial pressure above the CHCl3-pool water mixture and is dependent on air

temperature. The overall transfer velocity va/w is related to the transfer velocity

contributions from the water-side and air-side boundaries by

Equation 3.3

27

where va is the transfer velocity (cm/s) of chloroform air-side boundary and vw is the

transfer velocity (cm/s) of chloroform on the water-side boundary. The dimensionless

Henry’s law constant for chloroform is 0.1445 at 25 °C (Schwarzenbach et al., 2003).

Henry’s law constant changes with temperature and can be estimated by

Equation 3.4

Where temperature is expressed in Kelvin and c is 4600K (Gossett, 1987) and is an

experimentally determined constant. Transfer velocities on the air-side boundary are

determined by comparing molecular diffusivity ratios of different substances, which can

easily be determined by a molar mass relationship of those substances. In this case,

chloroform is compared to the well-known properties of water vapor by

Equation 3.5

where va∙water is the transfer velocity of water vapor (cm/s) and M is molecular weight.

Transfer velocity of water vapor is directly related to wind speed by

Equation 3.6

where u10 is the wind speed (m/s) at a standard height of 10 meters.

Transfer velocity on the water-side boundary is controlled by molecular

diffusivity and kinematic viscosity which are expressed through the Schmidt number, Sc.

Transfer velocity can only be calculated when Sc=600. At other Schmidt numbers,

transfer velocity can be determined through a relationship with CO2 where the transfer

velocity at Sc=600 is known:

α

Equation 3.7

28

where vw∙CO2 is the transfer velocity (cm/s) when ScCO2=600, Scw∙CHCl3 is the Schmidt

number of chloroform, and α=2/3 for u10≤5 m/s and α=1/2 for u10>5 m/s. The transfer

velocity of CO2 at 20°C when Sc=600 is found by

Equation 3.8

where u10 is the wind speed (m/s) at a height of 10 meters. The Schmidt number of

chloroform can be calculated by

Equation 3.9

where νw is the viscosity of water (cm2/s) and Dw∙CHCl3 is the molecular diffusivity (cm

2/s)

of chloroform in water. Molecular diffusivity is calculated by

Equation 3.10

where M is molar mass. The transfer velocity of chloroform on the water-side boundary

can then be evaluated for different temperatures by comparing against the viscosity of

water at different temperatures.

Equation 3.11

After flux is calculated per unit area, this value is extrapolated to determine total

annual CHCl3 for all pools in Phoenix Ftot according to

Equation 3.12

29

where Fa/w is the flux across the air-water surface per unit area, aavg is the average area of

swimming pools in Phoenix and nPHX is the number of swimming pools in Phoenix

(Table 3.2).

Sensitivity tests were performed to investigate impacts of wind speed, air

temperature, and water temperature on CHCl3 flux. To investigate this, flux was

calculated using several combinations of these factors when holding some constant and

varying others. Additionally, these factors were averaged daily, monthly, and annually to

determine if averaging these factors was representative of total annual flux. Figure 3.7

shows the CHCl3 flux calculated from swimming pools in Phoenix using wind speed and

air temperature data from 2013 while keeping water temperature constant at 29.4 °C,

which is the recommended pool water temperature for a multi-purpose pool

Estimated number of pools 286,000

average area of a pool 41.1 m2

Table 3.2: Statistics of Swimming Pools in Phoenix (Forrest and Williams, 2010)

Figure 3.7: Daily, Monthly, and Annual CHCl3 Flux from Swimming Pools in Phoenix

Calculated with Wind Speed and Air Temperature Data for 2013 While Holding Water

Temperature Constant.

50

100

150

200

250

300

350

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan

CH

Cl 3

flu

x (

kg∙d

ay

-1)

daily flux

monthly flux

annual flux

30

(USWFA.com). Chloroform flux shows very little daily variation at 90 kg∙day-1

. Days

with significantly higher CHCl3 fluxes are a result of high daily wind speeds. Averaging

of daily, monthly or annual wind speed and air temperature shows little variation in

CHCl3 fluxes.

Figure 3.8 shows CHCl3 flux from swimming pools when keeping wind speed

constant while varying air and water temperature. No reliable records for annual water

temperature variations were available. For these calculations, water temperature was

estimated by a linear increase/decrease between observed summer (28 °C) and winter (8

°C) water temperatures in an unheated swimming pool. A large increase in CHCl3 flux

can be seen as a result of increasing water temperature. Therefore, CHCl3 flux from

swimming pools in Phoenix is not driven directly by the extreme air temperatures for

which Phoenix is known, but rather indirectly through warming of swimming pool water.

However, solar irradiation might be a stronger force in warming swimming pool water.

Figure 3.8: Daily, Monthly, and Annual CHCl3 Flux from Swimming Pools in Phoenix

Calculated with Average Annual Wind Speed, Varying Air Temperature and Water

Temperature for 2013.

40

50

60

70

80

90

100

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

CH

Cl 3

flu

x (

kg∙d

ay-1

)

daily flux

monthly flux

annual flux

31

Flux calculated by averaging daily and monthly air and water temperature resulted in

values similar to annual CHCl3 fluxes, which indicates that using anually averaged

meterological data provides a good estimate for annual CHCl3 flux.

Summing daily CHCl3 flux produced a total annual emission rate. Depending on

variations in meteorological factors, calculations from this study indicated that swimming

pools in Phoenix might produce an estimated 3.9-4.3 Gg/yr. Figure 3.9 shows this

emission estimate from swimming pools in Phoenix compared to natural and

anthropogenic emission estimates. If Phoenix is assumed to be similar to desert shrubland

in the southern California desert, an area the size of Phoenix (2970 km2) with a flux of

6.6 nmol∙m-2

∙day-1

would generate an estimated 0.854 Mg/yr as natural emissions (Rhew

et al., 2008a). Chloroform emission from swimming pools is four orders of magnitude

larger than this number. In 2011, 0.063 Gg/yr was reported as total anthropogenic

Figure 3.9: Swimming Pool Emissions of CHCl3 in Phoenix compared to estimated

natural and anthropogenic emission estimates.

32

emissions for the Phoenix metropolitan area (Hafner and O’Brien, 2006) which is two

orders of magnitude less than CHCl3 emission from swimming pools found in this study.

This suggests that CHCl3 from swimming pools might be a significant source of

atmospheric CHCl3 in Phoenix.

33

CHAPTER 4

SUMMARY AND OUTLOOK

Unexplained elevated atmospheric CHCl3 concentrations in Phoenix, AZ lead to a

necessity to evaluate unconventional sources as a supply for CHCl3 flux into the

atmosphere. This study evaluates chlorination of swimming pools as a potential source

of atmospheric CHCl3 in Phoenix. Swimming pool samples from Phoenix, AZ were

collected and characterized, and CHCl3 formation upon chlorination in a controlled

laboratory experiment was determined using Gas Chromatography-Mass Spectrometry

(GC-MS) after headspace solid phase microextraction (SPME). Measurements from this

study found CHCl3 present immediately upon collection in all pool samples at an average

concentration of 2.21 μM. Upon chlorination, CHCl3 formation occurred in all pool

samples. These samples gave molar yields of 0.005-0.134 mol CHCl3/mol C and 0.004-

0.062 mol CHCl3/mol Cl2 consumed. However no visible trends relating swimming pool

water parameters with formation potential of CHCl3 could be determined and therefore

predictions could not be made as to factors affecting CHCl3 formation in swimming

pools.

Model contaminants were also chlorinated in controlled laboratory experiments.

These contaminants were chosen because of their known presence in swimming pool

water or their detection in Arizona surface water from anthropogenic contamination.

Chlorination of these compounds gave molar yields of 0.004-0.323 mol CHCl3/mol C and

0.001-0.247 mol CHCl3/mol Cl2 consumed which are similar to molar yields from

34

chlorination of swimming pool water. These results did not display any trends in CHCl3

yield based on elements, functional groups, or molecular size.

Chloroform flux was calculated using Henry’s Law and a simple boundary-layer

model. Sensitivity tests were performed to investigate impact of air temperature, water

temperature, and wind speed on CHCl3 flux. These tests determined that water

temperature had the largest effect on CHCl3 flux. Wind speed showed a minor effect on

the flux and air temperature had no effect on flux values even though Phoenix is known

for its high summer temperatures. Variations in wind speed and water temperature

resulted in potential flux calculations in the range of 3.9-4.3 Gg∙yr-1

. This number is

larger than previous anthropogenic source estimates and is four orders of magnitude

larger than estimated natural emissions in Phoenix. However, more work is required to

confirm initial results.

Measurements of annual variation in swimming pool water temperature would

also significantly refine flux measurements. This would aid in validating flux calculations

using a simple boundary-layer model. Water temperature was assumed as a linear

increase/decrease symmetrically about the calendar year, when annual variation of water

temperature is not known. In future work, these results might be validated through flux

measurements in sample swimming pools throughout Phoenix. Air measurements of

CHCl3 concentrations over a period of time can be taken near swimming pools to

measure CHCl3 concentrations in outdoor swimming environments. Additionally,

swimming pool water should be collected during different times of the year. These

samples were not time-resolved and significant changes in water chemistry between

seasons might affect CHCl3 formation.

35

An alternate pathway suggesting swimming pool chlorination as a source of

atmospheric CHCl3 is through outgassing of Cl2 and HOCl with CHCl3 formation in the

atmosphere above swimming pools rather than in the water. Therefore, future work

should also investigate the impacts of these pathways. Flux estimates from swimming

pools in a previous study were used to calculate possible flux of Cl2 and HOCl in

Phoenix, suggesting pools in Phoenix might outgas Cl2 and HOCl at rates of 0.1 Gg/yr

and 1 Gg/yr, respectively (Chang et al., 2001). Depending on chlorine demand and

conversion rate, this Cl2 might produce 0.2 Gg/yr CHCl3 in the atmosphere and this

HOCl might produce 1 Gg/yr CHCl3. Investigations to determine competition between

chlorine reactivity and outgassing should be performed to investigate possibility of

alternate CHCl3 formation pathways.

36

REFERENCES

Aggazzotti, G.; Fantuzzi, G.; Tartoni, P.L.; Predieri, G. (1990) Plasma chloroform

concentrations in swimmers using indoor swimming pools. Arch. Environ. Health

45, 175–179.

Aggazzotti, G.; Fantuzzi, G.; Righi, E.; Tartoni, P.; Cassinadri, T.; Predieri, G. (1993)

Chloroform in alveolar air of individuals attending indoor swimming pools. Arch.

Environ. Health 48, 250–254.

Aggazzotti, G.; Fantuzzi, G.; Righi, E.; Predieri, G. (1995) Environmental and biological

monitoring of chloroform in indoor swimming pools. J. Chromatogr., A 710,

181–190.

Aiking, H.; van Acker, M.B.; Scholten, R.J.; Feenstra, J.F.; Valkenburg, H.A. (1994)

Swimming pool chlorination: a health hazard? Toxicol. Lett. 72, 375–380.

Aucott, M.L.; Mcculloch, A.; Graedel, T.E.; Kleiman, G.; Midgley, P.; Li, Y. (1999)

Anthropogenic emissions of trichloromethane (chloroform, CHCl3) and

chlorodifluoromethane (HCFC-22): Reactive Chlorine Emissions Inventory. J.

Geophys. Res. 104, 8405–8415.

Black, A.P.; Keirn, M.A.; Smith, J.J.; Dykes, G.M.; Harlan, W.E. (1970) The

Disinfection of Swimming Pool Water. II. A Field Study of the Disinfection of

Public Swimming Pools. Am. J. Public Heal. Nations Heal. 60, 740–750.

Blatchley, E.R.; Margetas, D.; Duggirala, R. (2003) Copper catalysis in chloroform

formation during water chlorination. Water Res. 37, 4385–4394.

Blatchley, E.R. and Cheng, M. (2010) Reaction mechanism for chlorination of urea.

Environ. Sci. Technol. 44, 8529–34.

Boorman, G.A.; Dellarco, V.; Dunnick, J.K.; Robert E. C.; Hunter, S.; Hauchman, F.;

Gardner, H.; Cox, M.; Sills, R.C. (1999) Drinking Water Disinfection

Byproducts: Review and Approach to Toxicity Evaluation. Environ. Health

Perspect. 107, 207–217.

Cantor, K.P.; Hoover, R.; Mason, T.J.; McCabe, L.J. (1978) Associations of Cancer

Mortality With Trihalomethanes in Drinking Water. J. Natl. Cancer Inst. 61, 979-

985.

Castor, M.L. and Beach, M. (2004) Prevention of recreational water illnesses: Is

chlorination enough to ensure healthy swimming? Infect. Dis. Child. 17.

37

CDC.gov. Health Housing Reference Manual: Residential Swimming Pools and Spas.

http://www.cdc.gov/nceh/publications/books/housing/cha14.htm (accessed Nov 2,

2014).

Centerwall, B.S.; Armstrong, C.W.; Funkhouser, L.S.; Elzay, R.P. (1986) Erosion of

Dental Enamel Among Competitive Swimmers at a Gas-Chlorinated Swimming

Pool. Am. J. Epidemiol. 123, 641-647.

Chaidou, C.I.; Georgakilas, V.I.; Stalikas, C.; Saraçi, M.; Lahaniatis, E.S. (1999)

Formation of Chloroform by Aqueous Chlorination of Organic Compounds.

Chemosphere 39, 587–594.

Chang, E.E.; Chiang, P.C.; Chao, S.H.; Lin, Y.L. (2006) Relationship between chlorine

consumption and chlorination by-products formation for model compounds.

Chemosphere 64, 1196–1203.

Chiu, C. and Westerhoff, P.K. (2010) Trace Organics in Arizona Surface and

Wastewaters, in Contaminants of Emerging Concern in the Environment:

Ecological and Human Health Considerations, pp 81–117.

Christensen, F.M. (1998) Pharmaceuticals in the environment--a human risk? Regul.

Toxicol. Pharmacol. 28, 212–221.

Duirk, S.E.; Bridenstine, D.R.; Leslie, D.C. (2013) Reaction of benzophenone UV filters

in the presence of aqueous chlorine: Kinetics and chloroform formation. Water

Res. 47, 579–587.

EPA Air Toxics Data (2012) http://www.epa.gov/ttnamti1/toxdat.html#data (accessed

Jan. 14, 2014)

EPA TRI Program (2014) http://www2.epa.gov/toxics-release-inventory-tri-program

(accessed Nov 1, 2014).

Font-Ribera, L.; Kogevinas, M.; Zock, J.P.; Nieuwenhuijsen, M.J.; Heederik, D.;

Villanueva, C.M. (2009) Swimming pool attendance and risk of asthma and

allergic symptoms in children. Eur. Respir. J. 34, 1304–1310.

Forrest, N. and Williams, E. (2010) Life cycle environmental implications of residential

swimming pools. Environ. Sci. Technol. 44, 5601–7.

Gallard, H. and von Gunten, U. (2002a) Chlorination of Phenols: Kinetics and Formation

of Chloroform. Environ. Sci. Technol. 36, 884–890.

Gallard, H. and von Gunten, U. (2002b) Chlorination of Natural Organic Matter: Kinetics

of Chlorination and of THM Formation. Water Res. 36, 65–74.

38

Griffin, D. and Kellogg, C. (2004) Dust Storms and Their Impact on Ocean and Human

Health: Dust in Earth's Atmosphere. Ecohealth 1, 284–295.

Hafner, H.R. and O’Brien, T.E. (2006) Analysis of Air Toxics Collected as Part of the

Joint Air Toxics Assessment Project. http://www.epa.gov/ttnamti1/files

/20032004csatam/FinalreportJATAP2005.pdf (accessed Aug 31, 2014).

Heller-Grossman, L.; Manka, J.; Limoni-Relis, B.; Rebhun, M. (1993) Formation and

Distribution of Haloacetic Acids, THM, and TOX in CHlorination of Bromide-

Rich Lake Water. Wat. Res. 27, 1323–1331.

Hery, M.; Hecht, G.; Gerber, J. M.; Gendre, J. C.; Rebuffaudj, J. (1995) Exposure to

Chloramines in the Atmosphere of Indoor Swimming Pools. Ann. Occup. Hyg. 39,

427–439.

Hogan, M.D.; Chi, P.Y.; Hoel, D.G.; Mitchell, T.J. (1979) Association between

chloroform levels in finished drinking water supplies and various site-specific

cancer mortality rates. J. Environ. Pathol. Toxicol. 2, 873-887.

Jacobs, J.H.; Spaan, S.; van Rooy, G.B.G.J.; Meliefste, C.; Zaat, V.A.C.; Rooyackers,

J.M.; Heederik, D. (2007) Exposure to trichloramine and respiratory symptoms in

indoor swimming pool workers. Eur. Respir. J. 29, 690–698.

Judd, S. J. and Bullock, G. (2003) The fate of chlorine and organic materials in

swimming pools. Chemosphere 51, 869–879.

Kanan, A. and Karanfil, T. (2011) Formation of disinfection by-products in indoor

swimming pool water: the contribution from filling water natural organic matter

and swimmer body fluids. Water Res. 45, 926–32.

Khalil, M.A.K. and Rasmussen, R.A. (1999) Atmospheric chloroform. Atmos. Environ.

33, 1151–1158.

Kogevinas, M.; Villanueva, C. M.; Font-Ribera, L.; Liviac, D.; Bustamante, M.;

Espinoza, F.; Nieuwenhuijsen, M.J.; Espinosa, A.; Fernandez, P.; DeMarini,

D.M.; Grimalt, J.O.; Grummt, T.; Marcos, R. (2010) Genotoxic effects in

swimmers exposed to disinfection by-products in indoor swimming pools.

Environ. Health Perspect. 118, 1531–7.

Kuo, H.W.; Chiang, T.; Lo, I.I.; Lai, J.S.; Chan, C.C.; Wang, J.D. (1998) Estimates of

cancer risk from chloroform exposure during showering in Taiwan. Sci. Total

Environ. 218, 1–7.

39

Kim, H.; Shim, J.; Lee, S. (2002) Formation of disinfection by-products in chlorinated

swimming pool water. Chemosphere 46, 123–130.

Larson, J.L.; Templin, M.V.; Wolf, D.C.; Jamison, K.C.; Leininger, J.R.; Mery, S.;

Morgan, K.T.; Wong, B.A.; Connolly, R.B.; Butterworth, B.E. (1996) A 90-Day

Chloroform Inhalation Study in Female and Male B6C3F1 Mice: Implications for

Cancer Risk Assessment. Fundam Appl Toxicol 137, 118–137.

Lee, W. and Westerhoff, P (2007) Dissolved Organic Nitrogen as a Precursor for

Chloroform, Dichloroacetonitrile, N-Nitrosodimethylamine, and

Trichloronitromethane. Environ. Sci. Technol. 41, 5485–5490.

Lian, L.; E, Y.; Li, J.; Blatchley, E.R. (2014) Volatile disinfection byproducts resulting

from chlorination of uric acid: implications for swimming pools. Environ. Sci.

Technol. 48, 3210–3217.

Liu, X.F. (2006) Evidence of Biodegredation of Atmospheric Carbon tetrachloride in

Soils: Field and Microcosm Studies. Ph.D. Thesis, Columbia University, New

York, pp. 139.

McConnell, O. and Fenical, W. (1977) Halogen chemistry of the red Alga Asparagopsis.

Phytochemistry 16, 367-374.

McCulloch, A. (2003) Chloroform in the environment: occurrence, sources, sinks and

effects. Chemosphere 50, 1291–308.

Monster, A.C. (2010) Biological Markers of Solvent Exposure. Arch. Environ. Health 43,

90–93.

Moore, H.E.; Garmendla, M.J.; Cooper, W.J. (1984) Kinetics of Monochloramine

Oxidation of N,N-Diethyl-p-phenylenediamine. Environ. Sci. Technol. 18, 348–

353.

Nightingale, P.D.; Malin, G.; Liss, P.S.; Production of chloroform and other low-

molecular-weight halocarbons by some species of macroalgae. Limnology and

Oceanography 40, 680-689.

North Indian Bend Wash. (2005) Draft Preliminary Risk Assessment: Area 12

Groundwater Extraction and Treatment System. http://www.motorola.com/EHS

/environment/remediation/NIBW/reports/37%20Area%2012%20EmissHRA%20

Rpt%20Ver%20012605.pdf (accessed Nov 21, 2014).

Panyakapo, M.; Soontornchai, S.; Paopuree, P. (2008) Cancer risk assessment from

exposure to trihalomethanes in tap water and swimming pool water. J. Environ.

Sci. 20, 372-378.

40

Pehlivanoglu-Mantas, E. and Sedlak, D.L. (2008) Measurement of dissolved organic

nitrogen forms in wastewater effluents: Concentrations, size distribution and

NDMA formation potential. Water Res. 42, 3890-3898.

P.K. Data, Inc. U.S. Swimming Pool and Hot Tub Market, 2013. http://apsp.org/resources

/apsp-research-resources/industry-statistics.aspx (accessed Oct 26, 2014).

Redman, J.A,; Grant, S.B.; Olson, T.M.; Estes, M.K. (2001) Pathogen filtration,

Heterogeneity, and the Potable Reuse of Wastewater. Environ. Sci. Technol. 35,

1798–1805.

Rhew, R.C.; Miller, B.R.; Weiss, R.F. (2008a) Chloroform, carbon tetrachloride and

methyl chloroform fluxes in southern California ecosystems. Atmos. Environ. 42,

7135–7140.

Rhew, R.C.; Teh, Y.A.; Abel, T.; Atwood, A.; Mazéas, O. (2008b) Chloroform emissions

from the Alaskan Arctic tundra. Geophys. Res. Lett. 35, L21811.

Richardson, S.D.; Plewa, M.J.; Wagner, E.D.; Schoeny, R.; Demarini, D.M. (2007)

Occurrence, genotoxicity, and carcinogenicity of regulated and emerging

disinfection by-products in drinking water: a review and roadmap for research.

Mutat. Res. 636, 178–242.

Richardson, S.D.; DeMarini, D.M.; Kogevinas, M.; Fernandez, P.; Marco, E.;

Lourencetti, C.; Ballesté, C.; Heederik, D.; Meliefste, K.; McKague, A.B.;

Marcos, R.; Font-Ribera, L.; Grimalt, J.O.; Villanueva, C.M. (2010) What’s in the

Pool? A Comprehensive Identification of Disinfection By-products and

Assessment of Mutagenicity of Chlorinated and Brominated Swimming Pool

Water. Environ. Health Perspect. 118, 1523–1530.

Roberts, A.; Renwick, A.G.; Sims, J.; Snodin, D.J. (2000) Sucralose metabolism and

pharmacokinetics in man. Food Chem. Toxicol. 38, 31–41.

Rook, J.J. (1974) Formation of Haloforms During Chlorination of Natural Waters. J.

Water Treat. Exam. 23, 234-243.

Rowhani, T. and Lagalante, A.F. (2007) A colorimetric assay for the determination of

polyhexamethylene biguanide in pool and spa water using nickel-nioxime.

Talanta 71, 964–970.

Salter, C. and Langhus, D.L. (2007) Products of Chemistry The Chemistry of Swimming

Pool Maintenance. Chem. Everyone 84, 1124–1128.

41

Schulman, L.J.; Sargent, E.V.; Maumann, B.D.; Faria, E.C.; Dolan, D.G.; Wargo, J.P.

(2002) A human health risk assessment of pharmaceuticals in the aquatic

environment. Hum. Ecol. Risk Assess. 8, 657-680.

Shepherd, J.L.; Corsi, R.L.; Kemp, J. (1996) Chloroform in Indoor Air and Wastewater:

The Role of Residential Washing Machines. J. Air Waste Manage. Assoc. 46,

631–642.

Schwarzenbach, R.P.; Gschwend, P.M.; Imboden, D.M. Environmental Organic

Chemistry; John Wiley & Sons: Hoboken, 2003.

Thickett, K.M.; McCoach, J.S.; Gerber, J.M.; Sadhra, S.; Burge, P.S. (2002)

Occupational asthma caused by chloramines in indoor swimming-pool air. Eur.

Respir. J. 19, 827–832.

Torres, C. I.; Ramakrishna, S.; Chiu, C.A.; Nelson, K.G.; Westerhoff, P.; Krajmalnik-

Brown, R. (2011) Fate of Sucralose During Wastewater Treatment. Environ. Eng.

Sci. 28, 325–331.

USWFA.com. Suggested Swimming Pool Temperatures. http://www.uswfa.com

/suggested_pool_temps.asp (accessed Nov. 1, 2014).

Wang, J.; Qin, P.; Sun, S. (2007) The flux of chloroform and tetrachloromethane along an

elevational gradient of coastal salt marsh, East China. Environ. Pollut 148, 10-20.

Wang, W.; Qian, Y.; Boyd, J.M.; Wu, M.; Hrudey, S.E.; Li, X.F. (2013)

Halobenzoquinones in swimming pool waters and their formation from personal

care products. Environ. Sci. Technol. 47, 3275–82.

Weaver, W.A., Li, J.; Wen, Y.; Johnston, J.; Blatchley, M.R.; Blatchley, E.R. (2009)

Volatile disinfection by-product analysis from chlorinated indoor swimming

pools. Water Res. 43, 3308–18.

Webb, S.; Ternes, T.; Gilbert, M.; Olejniczak, K. (2003) Indirect exposure to

pharmaceuticals via drinking water. Toxicol. Lett. 142, 157-167.

Weisel, C.P. and Jo, W. (1996) Ingestion, Inhalation, and Dermal Exposures to

Chloroform and Trichloroethene from Tap Water. Environ. Health Perspect. 104,

48-51.

Wendelken, S.C.; Losh, D.E.; Fair, P.S. (2009) Method 334.0: Determination of Residual

Chlorine in Drinking Water Using an On-Line Chlorine Analyzer.

http://www.epa.gov/ogwdw/methods/pdfs/methods/met334_0.pdf (accessed Oct.

27, 2013).

42

Weng, S. and Blatchley, E.R. (2011) Disinfection by-product dynamics in a chlorinated,

indoor swimming pool under conditions of heavy use: national swimming

competition. Water Res. 45, 5241–5248.

Weng, S.C.; Sun, P.; Ben, W.; Huang, C.H.; Lee, L.T.; Blatchley, E.R. (2014) The

Presence of Pharmaceuticals and Personal Care Products (PPCPs) in Swimming

Pools. Environ. Sci. Technol. Lett. Just Accep, 141024150330002.

Westerhoff, P.; Yoon, Y.; Snyder, S.; Wert, E. (2005) Fate of Endocrine-Disruptor,

Pharmaceutical, and Personal Care Product Chemicals during Simulated Drinking

Water Treatment Processes. Environ. Sci. Technol. 39, 6649–6663.

Yoon, Y.; Westerhoff, P.; Snyder, S.A.; Wert, E.C. (2006) Nanofiltration and

ultrafiltration of endocrine disrupting compounds, pharmaceuticals and personal

care products. J. Memb. Sci. 270, 88–100.

Zweiner, C.; Richardson, S.D.; DeMarini, D.M.; Grummt, T.; Glauner, T.; Frimmel, F.H.

(2007) Drowning in Disinfection Byproducts? Assessing Swimming Pool Water.

Environ. Sci. Technol. 41, 363–372.