Seasonal Variations in Lead Release to Potable ... - ACS Publications

Subscriber access provided by UNIV OF GEORGIA

Article

Seasonal Variations in Lead Release to Potable Water Sheldon Masters, Gregory J Welter, and Marc A. Edwards Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b05060 • Publication Date (Web): 14 Apr 2016 Downloaded from http://pubs.acs.org on April 19, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1

2

Environmental Science & Technology

Seasonal Variations in Lead Release to Potable Water

3

Sheldon Masters1*, Gregory J. Welter2 and Marc Edwards1

5

Virginia Tech, Blacksburg, VA 24061; 2O’Brien and Gere Engineers, Inc, Bowie, MD;

7

*CORRESPONDING AUTHOR. Via Department of Civil and Environmental

9

(330) 347-7825; Fax: (540) 231-7916

4

6 8

10

11

AUTHOR ADDRESS. 1Via Department of Civil and Environmental Engineering,

Engineering, Virginia Tech, Blacksburg, VA 24061. Email: [email protected]; Phone:

KEYWORDS. Solubility, lead, copper, summer, winter.

ACS Paragon Plus Environment

1


12

Abstract

14

drinking water systems and lead release to potable water was examined. Temperature

16

chloropyromorphite, lead orthophosphate and lead oxide solids; however, in the

18

vs. 1277 ppb) at 20 ˚C compared to 4 ˚C due to accelerated reductive dissolution. The

20

vs. 92 ppb). In full scale pipe rigs using harvested lead service lines in Washington DC

22

particulate lead increased 2-6 times in the summer versus winter. In 4 of the 8 homes

13

The influence of temperature on the solubility of representative lead solids present in

15

had surprisingly little effect on the dissolution of cerrusite, hydrocerussite,

17

Page 2 of 33

presence of natural organic matter, lead oxide dissolution was 36 times greater (36 ppb

19

solubility of plumbonacrite was three times higher at 20 ˚C compared to 4 ˚C (260 ppb

21

and Providence RI, dissolved lead release increased by as much as 2-3 times and

23

sampled in Providence, RI, dissolved lead levels were three times higher during the

25

greater in the winter. These studies demonstrate a need to better understand how lead

27

sometimes vary markedly even within the same distribution system.

24

summer compared to the winter while 5 homes had copper levels that were 2.5-15 times

26

service line scales vary, since patterns of release and temperature dependency

28

29


2

Page 3 of 33


30

INTRODUCTION

32

materials such as pipes, faucets, and fittings.1,2 Human exposure to lead from drinking

34

children and adverse health outcomes.3–7 Water utilities have historically controlled the

31

Potable water can be contaminated with lead due to the corrosion of lead bearing

33

water in modern plumbing systems has been associated with increased blood lead in

35

release of lead from lead bearing pipe materials by maintaining water chemistry

37

coat plumbing materials.1,8

39

lead carbonates, and lead oxides.9–12 Pb(IV) oxides can form in distribution systems with

36

38

conditions (i.e. pH and alkalinity) that reduce the solubility of lead corrosion solids that

Representative lead solids coating plumbing systems include lead phosphates,

40

a free chlorine residual and they have extremely low solubility.1,13 Very low solubility

42

orthophosphate as a corrosion inhibitor,14 and if adequate levels of phosphate or free

41

lead phosphate scales are also formed in distribution systems that employ

43

chlorine are not present higher solubility lead carbonates such as cerussite (PbCO3),

45

form.11,15

47

include water quality (i.e., pH, dissolved oxygen, alkalinity, buffer capacity, phosphate

49

observed variability in metals release in real systems using existing solubility

44

46 48

50

51

hydrocerussite (Pb3(OH)2(CO3)2), and plumbonacrite (Pb5O(CO3)3(OH)2) tend to

The parameters that are often considered when assessing the corrosion of lead

and polyphosphate level). However, these factors alone cannot fully explain the

models.16,17 Seasonal variations in temperature are another potentially important factor that can influence the concentration of lead and copper in potable water. Even though


3


52

the dissolution rates of lead oxide,18–22 lead carbonate,15,23 and lead phosphate24–27 have

54

examined the important role of temperature on the kinetics or equilibrium solubility of

53

been studied under a range of water chemistries, to our knowledge few studies have

55

lead in well-controlled laboratory studies.

57

plumbing systems with higher temperature, and it is commonly stated that this is due

59

For instance, the Environmental Protection Agency warns consumers to avoid

61

because these conditions are suspected to have highest potential for lead release from

63

lead levels in water and higher seasonal temperatures.3,30,33,41–45 Also, observations of a

56 58 60

Page 4 of 33

Nonetheless, field research has often associated measured higher levels of lead in

to higher solubility and/or dissolution rates for lead solids at higher temperature.8,28–38

consuming hot water and recommends utility lead monitoring in the summer months,

62

plumbing.39,40 Anecdotally, a few field studies have confirmed links between higher

64

higher incidence of lead poisoning during the summer months, once exclusively

66

levels of lead documented in potable water in warmer summer months.38 Thus, the

68

lead to higher levels of lead in water, although exceptions have been hypothesized

70

release in hard water containing orthophosphate had little temperature dependency

72

reductive dissolution of PbO2 oxides by NOM can be an important pathway of lead

65

attributed to increased exposure to soils and lead paint46–49 might also be due to higher

67

general expectation based on the conventional wisdom is that higher temperatures will

69

where the opposite trend is expected.1,32,33 For example, field data indicate that lead

71

and apparently a small decrease in lead levels with increasing temperatures.33 Finally,


4

Page 5 of 33

73


release to potable water in some systems, and the role of temperature in the rate of this

74

reaction has not been previously assessed.18

76

temperature dependence of lead corrosion by-product release and solubility, using data

78

examined as part of this overall evaluation, which has important implications for

80

Lead and Copper Rule (EPA LCR) and human lead exposure.

82

MATERIALS AND METHODS

84

by-product release were assessed using bench scale solubility tests, pipe loop studies,

86

Lead Mineral Solubility Experiments

88

D.C., Providence, RI and other distribution systems were examined in laboratory

90

2000, the dominant lead scale in the Washington D.C. distribution system was identified

92

disinfectant byproduct formation but also produced a significant increase in lead.3 Since

94

release, 7,23,52 and characterization of lead scales from a Washington Aqueduct pipe loop

75

The goal of this research is to address these knowledge gaps, by examining the

77

from bench, pilot and full-scale studies. Both soluble and total lead release were

79

monitoring worst case lead levels under a revised Environmental Protection Agency

81 83

85 87

The effects of seasonal variations in temperature on lead solubility and corrosion

and distribution system sampling.

Dissolution of six lead solids that have been identified on pipes in Washington,

89

experiments. Prior to a change in disinfectant from chlorine to chloramine in November

91

as plattnerite (PbO2).12,21,50,51 The change in disinfectant reduced the potential for

93

2005, orthophosphate has been added to the finished water in order to reduce lead


5


Page 6 of 33

95

study found that the importance of pyromorphite in the pipe scale increased.12,51,52 Even

97

the dominant crystalline phase in some pipe scales from the Providence, RI distribution,

99

Hydrocerussite was also found to be a dominant pipe scale at some sample sites within

101

Lead Solids. Cerussite and hydrocerussite were purchased from Alfa Aesar, and

96

though lead(II) hydroxycarbonate plumbonacrite is rare in nature, it was found to be

98

perhaps due to the unusually high pH and low alkalinity in this system.52,53

100

the same Providence, RI distribution system.53

102

other representative solids including lead oxide (PbO2), lead orthophosphate

104

formed in the laboratory using established methods. Specifically, lead oxide was

106

oxidizing a 1000 mg/L as Pb solution of lead chloride via addition of excess

108

Na2HPO4 to 0.1 M lead acetate at 80 ˚C and maintaining a low heat on a hot plate for 3

110

thoroughly mixing 0.25 M PbCl2 solution with 0.15 M phosphoric acid (H3PO4) resulting

103

(Pb3(PO4)2), lead phosphate chloropyromorphite (Pb5(PO4)3Cl) and plumbonacrite were

105

produced using the method outlined by Triantafyllidou et al. (2007) by completely

107

hypochlorous acid.54 Lead orthophosphate solids were formed by slowly adding 0.1 M

109

hours.24,55 Lead phosphate chloropyromorphite (Pb5(PO4)3Cl) was synthesized by

111

in a white precipitate.25,26 Lead carbonate plumbonacrite was synthesized using the

113

200 mL solution of K2CO3 (13 g) and KOH (2 g) and stirring the suspension for 48 hours

112

method described by Taylor and Lopata (1984) by suspending hydrocerussite (5 g) in a

114

at 100 ˚C. The filtered product contained plumbonacrite (> 80%) and hydrocerrusite.56

116

both filtered (0.45 µm pore size mixed cellulose filter, Whatman) and unfiltered lead in

115

Collection of Synthesized Lead Solids. After measuring the concentration of


6

Page 7 of 33


117

each synthesized sample (i.e., lead oxide, lead phosphates, and plumbonacrite), the

119

was calculated.

121

the filter with the captured lead solids (~0.15 g total Pb) or pure lead solids (cerussite

123

and mixing at 100 rpm using an orbital shaker. Control experiments indicated that the

125

lead oxide reactors had an initial free chlorine residual of 2.2 mg/L while the lead

127

phosphate levels were measured using a DR2700 spectrophotometer (HACH, Loveland,

118

120

volume of solution that should be filtered to capture a total of 0.15 g lead on each filter

Dissolution of Lead Solids. A dissolution experiment was conducted by adding

122

and hydrocerussite) into a 1 L jar containing 800 mL of synthesized tap water (Table S1)

124

presence of the filter had no detectable effect on the dissolution of the lead solid. The

126

phosphate reactors had an orthophosphate residual of 1 mg/L as P. Chlorine and

128

CO).

130

fulvic acid natural organic matter (NOM) isolated by XAD from Silver Lake, WA was

132

with NaOH.57,58 Total organic carbon (TOC) was measured by persulfate-ultraviolet

129

The dissolution of lead oxide in synthetic water with 12 mg C/L of purified

131

used to create an NOM stock solution (~600 mg/L as C) and the pH was adjusted to 7.0

133

detection using a Sievers Model 5300C with an autosampler according to Standard

134

Method 5310 C.

136

0.1 µm pore size filter at 6 and 24 hours after wasting the first 3 drops of solution. The

138

mLs, providing negligible opportunity for CO2 transfer during the test. Each condition

135

137

Sample aliquots of 10 mL were collected from each reactor and filtered through a

system was closed except during sampling and headspace was on the order of a few


7


Page 8 of 33

139

was tested in triplicate at 4 ˚C and 20 ˚C and the pH did not change significantly (< 0.2

141

were dissolved in 5% nitric acid and heat treated at 50 ˚C for at least 24 hours in order to

143

analyzed using inductively coupled plasma mass spectrometry (ICP-MS) using

145

Pipe loop Studies

140

pH units) over the 24 hours of testing. The remaining lead solids in each batch reactor

142

confirm that ~0.15 g of lead was added to each condition. Metal concentrations were

144

Standard Method 3125B.59

146

147

Washington Aqueduct Pipe Loop Study Washington Aqueduct Treatment Process. The Washington Aqueduct draws water

148

from the Potomac River which is then treated at two plants, Dalecarlia and McMillan.

150

with free chlorine followed by secondary disinfection with chloramines. The plants use

152

using lime and since August 2004, orthophosphate has been added as a corrosion

149

Both plants perform pre-sedimentation, coagulation/flocculation, primary disinfection

151

an alum coagulant and polyaluminum chloride as a filtration aid. The pH is controlled

153

inhibitor.60

155

pipe loop study at the Dalecarlia Water Treatment Plant using harvested LSLs to

157

in the Washington DC area. Data from the control pipe loop, which was exposed to

159

seasonal variations in temperature on total and dissolved lead release.61 Standard

154

Pipe Loop Study. Beginning in January 2005, the Washington Aqueduct initiated a

156

simulate distribution system conditions and to better understand lead corrosion issues

158

finished effluent from the Aqueduct, was analyzed herein to determine the effects of

160

methods were used for all data measurements. The racks consisted of a single pass flow


8

Page 9 of 33


161

regime, with a 16-hour period of flow followed by an 8-hour stagnation period.60 Each

163

diameter pipe, yielding a total volume of 1.1 liters per pipe loop. Water temperature

165

aliquot samples were also collected to operationally determine the fraction of soluble

167

the sampling period.

162

loop included two or three separate sections of LSL, for a total 13 feet of ¾ inch-

164

measurements were taken immediately after the samples were collected.60 Filtered

166

lead (< 0.45 µm). Table S2 summarizes the water quality in the control pipe loop during

168

169

Providence, RI, Pipe Loop Study Providence, RI Treatment Process. Providence, RI receives water from the Scituate

170

Reservoir which is treated at the Philip J. Holton Water Purification Plant. The

172

coagulation, lime addition for pH/alkalinity adjustment supporting coagulation,

174

has relatively low pH and low alkalinity which are altered during the treatment

176

increased to 9.9-10.7 in the final effluent. The daily raw water alkalinity range from 5.0-

171

treatment process consists of preliminary aeration, ferric sulfate addition for

173

sedimentation, chlorination and lime addition for final pH adjustment. The raw water

175

process. For example, in 2013 the daily raw water pH ranged from 6.0-6.8 but was

177

7.1 mg/L compared to 13.6-22.0 mg/L in the final effluent.62

179

order to determine the efficacy of orthophosphate as a lead corrosion inhibitor at pH

181

effect of temperature on lead release. The control loop was filled with finished water

178

Pipe Loop Study. In March 2014, Providence Water initiated a pipe loop study in

180

~10.3. Data from the control and experiment pipe loop was analyzed to monitor the

182

from the Providence treatment plant while the experimental condition was augmented


9


Page 10 of 33

183

with 3 mg/L orthophosphate as PO4. Two pipe loops, each consisting of four 24-inch

185

rack included sampling ports, rotameters for measuring flow rate, isolation valves used

187

remain full during stagnant periods between pumping flows and throttling valves for

189

provide a target flow rate of 0.75 gpm through each lead pipe with five 4-minute flush

191

measurements, soluble lead samples were collected by filtering an aliquot through a

193

Providence, RI Distribution Sampling

195

February 2014) as part of an assessment of water treatment process modifications (Table

197

seasonal variations in temperature on lead and copper release at individual homes.

199

spigot. The inside sampling was done by homeowners while outside sampling from

201

spigot were done on different days so that both samples could be collected after a 6-

203

records were used to estimate the volume of water contained in the plumbing between

184

lead pipes harvested from the distribution system, were operated Monday-Friday. Each

186

during sampling, upstream and downstream venting structures to allow the pipes to

188

flow adjustment (Figure S1 and Figure S2). The 8 throttling valves were adjusted to

190

periods at (2:00 am, 5:00 pm, 7:00 pm, 9:00 pm, and 11:00 pm). In addition to total lead

192

0.45 µm pore size filter. Temperature was measured in samples from the source tanks.63

194

Monthly samples were collected from 8 homes in Providence, RI (January 2013-

196

1). Data from this monitoring program were analyzed to characterize the effects of

198

Sampling was conducted from two points: an inside kitchen faucet and an outside

200

spigots was done by trained water quality technicians. Sampling the faucet and the

202

hour minimum stagnation time. On-site premise plumbing surveys along with utility

204

the sampling tap and lead service line (LSL). A 500 mL first draw sample was collected,


10

Page 11 of 33


205

followed by wastage of a site-specific volume, and then followed by 1 L LSL sample.

207

characterize the water in the distribution main. Since sample flow rate can significantly

209

of ~2 gpm for the outside sampling. During the initial inspection of the premises, the

211

sample at the target flow rate. The pH and temperature of first draw spigot and LSL

213

taken by utility staff, and all samples were analyzed for total and dissolved lead and

215

total that passes through a 0.45 µm pore size filter. Field sampling results focus

217

range found in kitchen faucets and temperature measurements were only taken during

219

during the sampling period.

221

Statistical Analyses

223

difference in lead release or copper release between different seasons (i.e., summer vs.

206

After three minutes of additionally flushing, a final 1 L sample was collected to

208

affect lead release,64–67 an effort was made to consistently collect samples at a flow rate

210

flow rates at outside spigots were measured and the spigots marked to reproducibly

212

samples were measured in the field immediately after sampling for the spigot samples

214

copper.62 Dissolved lead and copper were operationally defined as the fraction of the

216

exclusively on spigot samples since lead levels from the spigot samples were within the

218

outside sampling. Table S3 summarizes the finished treatment plant water quality

220 222

A non-parametric two-tailed Wilcoxon test and t-test were used to compare the

224

winter). In this study, winter months are defined as December to March and summer

226

log-transformed normalized data with a Tukey’s HSD post-hoc analysis was used to

225

months are defined as June to September. An analysis of variance (ANOVA) test on the


11


227

compare the difference in soluble lead release in the batch reactors (α = 0.05).

229

between metals release and temperature for non-parametric data while Pearson

228

Spearman’s rank correlation coefficient (ρ) was used to measure the relationship

230

correlation coefficient was used for parametric data.

232

Providence, RI pipe loop were examined using linear regressions to determine if the

234

of comparison from one condition to another and to quantify the magnitude of the

231

Page 12 of 33

Simple linear regression model for Providence, RI pipe loop study. Data from the

233

concentration tended to increase or decrease with temperature. To have a common basis

235

effect, the increase in lead release with temperature (e.g., ppb/˚C) is calculated as the

237

degrees Celsius.

239

the Washington Aqueduct pipe loop study were examined using multiple-linear

236 238

slope of the regression line from a graph of lead concentration versus temperature in

Multiple linear-regression model for Washington Aqueduct pipe loop study. Data from

240

regression to determine the functional relationship between lead release (dependent

242

phosphate concentration). The lead data were normalized (Figure S3 and Figure S4)

241

variable) and several independent variables (i.e., temperature, pH, total chlorine and

243

using a log10 transformation and the log-linear model equation is given by:

244

where Y is the lead concentration (dependent variable), α is the intercept, x is the

246

and β1 is the estimated slope of a regression of Y on x1, if all other x variable are kept

245 247

𝑙𝑙𝑙10 𝑌 = 𝛼 + 𝛽1 𝑥1 + 𝛽2 𝑥2 …

predictor variables (i.e., temperature, pH, total chlorine and phosphate concentration)

constant. In other words, for each 1-unit increase in x multiplies the expected value of Y


12

Page 13 of 33


β

248

by 10 . In the Providence pipe loop study disinfectant, pH and phosphate

250

multifactor analysis could not be performed.

252

RESULTS

254

representative lead solids in laboratory studies, the trends in lead release to water from

256

and regulatory implications of the results are then discussed.

249

251

253

measurements were not done on the same samples for lead, therefore, a similar

After reporting the effects of temperature on the dissolution and solubility of

255

lead pipe loop studies and distribution system field sampling are presented. Practical

257

Lead Mineral Solubility Experiments

259

temperature on solubility of lead solids. The few stability constants that exist suggest

261

decrease with decreasing temperature; however, the temperature effect between 0 ˚C

263

resolve this issue, the effect of temperature on the dissolution of representative lead

265

collected at 6 hours or 24 hours (Figure S5).

267

lead carbonate solids (i.e., cerussite, hydrocerussite, and plumbonacrite) had higher

258

Available solubility models lack necessary enthalpy data to predict the effect of

260

that the solubility of lead carbonate commonly found in distribution systems may

262

and 25 ˚C does not appear to be large in comparison to other factors.68,69 To practically

264

solids was directly determined herein by experiment at 4 ˚C and 20 ˚C with samples

266

Consistent with expectations based on solubility and dissolution kinetics, the

268

solubility compared to PbO2 or Pb3(PO4)2 regardless of temperature (Figure


13


269

1).1,11,13,15,70,71 This verifies the basis for reducing lead in water by creating conditions

271

dosing of phosphate.72,73

273

free chorine is removed during stagnation, the PbO2 can reductively dissolve in the

275

processes rather than oxidation).18,74 After 6 hours in the solubility experiment, lead

270

272

Page 14 of 33

that support formation of these low solubility solids, including use of free chlorine or

On the other hand, if PbO2 is present on pipes after years of chlorine dosing, or if

274

presence of NOM (i.e., the dissolution of the PbO2 scale is mediated by reductive

276

oxide in the presence of 12 mg/L of NOM resulted in 36 times (36 ppb vs. 1277 ppb)

278

0.0001) (Figure 1 and Table S4).18,19,22 Thus, this work demonstrates that temperature is

280

whereas temperature itself had little impact on its own.

282

almost three times more soluble lead was released at 20 ˚C compared to 4 ˚C (260 ppb

284

commonly found on lead plumbing surfaces (i.e., cerussite, hydrocerussite, lead

286

between 4 ˚C and 20 ˚C (p > 0.05).

277

more lead at 20 ˚C compared to 4 ˚C due to a higher reductive dissolution rate (p =

279

an important dimension in control of reductive dissolution in the presence of NOM,

281

Plumbonacrite also demonstrated a strong temperature dependency, in that

283

vs. 92 ppb) (p < 0.0001). Somewhat surprisingly, the other common lead solids

285

phosphates, and lead oxide) did not have a significant temperature dependency

287

Pipe Loop and Field Studies

289

mg/L as CaCO3 while the Providence finished water had a pH of ~10.4 and alkalinity of

288

290

In the Washington, D.C. study, the average pH was 7.8 and the alkalinity was 83

16 mg/L as CaCO3 (Table S2 and Table S3). Even though the two study sites had widely


14

Page 15 of 33


291

different water chemistries, higher temperatures would be expected to increase soluble

293

and plumbonacrite in the Providence system. The field work also provided an

292 294

lead in these systems, given the reported importance of PbO2/NOM in the D.C. system

opportunity to track the potentially important role of temperature in particulate lead

295

release from pipes.

297

soluble lead release in the Washington Aqueduct pipe loop study varied with seasonal

299

before the peak temperature (Figure 2). Therefore, the absolute temperature as well as

301

main breaks where a drop in temperature resulted in an increase in main breaks due to

303

correlation coefficient of the normalized lead data, there was a significant relationship

305

dissolved lead, p < 0.001). Average particulate lead release during the summer months

307

average dissolved lead was 3 times higher in the summer compared to the winter (10.8

309

months with higher temperature,8,28–38 but the higher particulate lead release was an

296

Washington Aqueduct Pipe Loop Study. Seasonal variation in particulate and

298

changes in temperature. However, it appears that lead levels tend to peak slightly

300

the rise in temperature affects lead release. A similar phenomenon is observed with

302

a drop in temperature as well as the absolute temperature.75 Based on the Pearson

304

between lead release and temperature (R = 0.73 for particulate lead and 0.70 for

306

was 6 times higher compared to the winter (48.4 ppb vs. 7.6 ppb, p < 0.05). Similarly,

308

ppb vs. 3.6 ppb, p < 0.05). Thus, this work confirmed that soluble lead did increase in

310

even greater concern.

312

chlorine (R = -0.4 for both particulate and dissolved lead, p < 0.001) as expected, given

311

There was also a weak negative relationship between lead release and total


15


Page 16 of 33

313

that chloramine decayed more rapidly at higher temperature (R = -0.73 for correlation

315

chloramine levels were at least 2.14 mg/L 99% of the time which would have helped to

317

lead release and pH or phosphate (Figure 3 and Figure S6). A multiple linear regression

319

temperature, phosphate concentration, total chlorine concentration and pH) were

321

R2) of the multiple regression model for particulate lead and dissolved lead was 0.55

323

significant effects of temperature which would be expected given the results of the

325

chloramine concentration (Table S5 and Table S6). This may be due to the fact that

314

between chloramine concentration and temperature). However, it should be noted that

316

stabilized PbO2 during stagnation.76 There was also weak or no relationship between

318

model was used to determine whether other available water quality parameters (i.e.,

320

deterministic in both soluble and total lead release. The coefficient of determination (i.e.,

322

and 0.53, respectively (p < 0.001). Moreover, the regression analyses suggested

324

bench scale tests and previous literature. There was also a significant effect of

326

monochloramine is capable of reducing lead oxide to form Pb(II) and the amount of

328

decomposition.20 Assuming all other variables are kept constant, each 1 ˚C rise in

330

and dissolved lead by a factor of 1.17 (i.e., 17%). Even though seasonal variations in lead

332

does not always result in an increase in lead release as demonstrated in bench scale

327

Pb(II) released tends to be proportional to absolute rate of monochloramine

329

temperature is predicted to increase particulate lead release by a factor of 1.36 (i.e., 36%)

331

release are strongly influenced by temperature, as a general rule, higher temperature

333

tests. In the end the interpretation of these trends are complicated by changes in other

334

factors such as seasonal variations in NOM.


16

Page 17 of 33

335


Providence, RI Pipe Loop Study. Data from the pipe loop with and without

336

phosphate (control) were analyzed to determine the effect of seasonal variation in

338

release was highly correlated with seasonal changes in temperature in the control

340

< 0.0001) (Figure 4). Particulate lead release in the control condition was weakly

342

lead release increased at a rate of 4.5 ppb/˚C in the control loops (p < 0.0001). However,

344

significant (Slope = 17.4 ppb/˚C, p = 0.08).

337

temperature on lead release. Over the course of the 21-month study, dissolved lead

339

conditions with the highest lead levels occurring during the summer months (ρ = 0.80, p

341

correlated with temperature (ρ = 0.31,

Seasonal Variations in Lead Release to Potable ... - ACS Publications

Recommend Documents