Experimental Evaluation of Turbidity Impact on the Fluence Rate

Oct 25, 2017 - Lyles School of Civil Engineering, Purdue University, West Lafayette, Indiana 47907, United States. ‡ Key Laboratory of Drinking Wate...
0 downloads 12 Views 751KB Size
Subscriber access provided by UNSW Library

Article

Experimental evaluation of turbidity impact on the fluence rate distribution in a UV reactor using a micro-fluorescent silica detector Mengkai Li, Wentao Li, Dong Wen, Zhimin Qiang, and Ernest R. Blatchley Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b02730 • Publication Date (Web): 25 Oct 2017 Downloaded from http://pubs.acs.org on October 26, 2017

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 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

1

Experimental evaluation of turbidity impact on the fluence rate

2

distribution in a UV reactor using a micro-fluorescent silica

3

detector

4 5

Mengkai Li †,‡, Wentao Li ‡, Dong Wen ‡, Zhimin Qiang ‡,*, Ernest R. Blatchley III †,§,*

6 7



8

United States

9



Lyles School of Civil Engineering, Purdue University, West Lafayette, Indiana 47907,

Key Laboratory of Drinking Water Science and Technology, Research Center for

10

Eco-Environmental Sciences, University of Chinese Academy of Sciences, Chinese

11

Academy of Sciences, 18 Shuang-qing Road, Beijing 100085, China

12

§

13

Lafayette, Indiana 47907, United States

Division of Environmental & Ecological Engineering, Purdue University, West

14 15

*Corresponding authors.

16

*

17

Blatchley III)

18

*Phone: +86 10 62849632; fax: +86 10 62923541; e-mail: [email protected] (Z.

19

Qiang)

Phone: 1-765-494-0316; fax: 1-765-494-0395; e-mail: [email protected] (E.R.

20

1

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 2 of 30 by Li et al.

21

ABSTRACT: Turbidity is a common parameter used to assess particle concentration

22

in water using visible light.

23

scattering, refraction, and reflection) in influencing the optical properties of aqueous

24

suspensions complicates examinations of their effects on ultraviolet (UV)

25

photoreactor performance.

26

in a photoreactor containing various particle suspensions (SiO2, MgO, and TiO2) were

27

measured using a micro-fluorescent silica detector (MFSD).

28

particles, as well as transmittance and scattering properties of the suspensions were

29

characterized at UV, visible, and infrared (IR) wavelengths.

30

measurements indicated that the optical properties of all three particle types were

31

similar at visible and IR wavelengths, but obvious differences were evident in the UV

32

range.

33

suspensions, the weighted average FR (WAFR) increased relative to deionized water.

34

These increases were attributed to low particle photon absorption and strong

35

scattering.

36

TiO2 suspensions because of their high particle photon absorption and low scattering

37

potential.

38

transmittance at UV wavelengths can be used to quantify the effects of turbidity on

39

UV FR distributions.

However, the fact that particles play multiple roles (e.g.,

To address this issue, UV fluence rate (FR) distributions

Reflectance of solid

The results of these

The FR results indicated that for turbidity associated with SiO2 and MgO

In contrast, the WAFR values decreased with increasing turbidity for

The findings also indicate that measurements of scattering and

40 41

Key words: micro-fluorescent silica detector (MFSD); turbidity; scattering; reflection;

42

fluence rate.

2

ACS Paragon Plus Environment

Page 3 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

43

by Li et al.

■ INTRODUCTION

44

In recent years, ultraviolet (UV) technology has been used increasingly for

45

disinfection, photolysis, and advanced oxidation processes in water and wastewater

46

treatment.1, 2

47

delivery, in particular the fluence distribution.

48

applied to calculate fluence delivery, generally based on integrated application of

49

sub-models to simulate the optical field (i.e., fluence rate (FR) distribution) and fluid

50

mechanics, usually by application of computational fluid dynamics software.3, 4

51

important component of accurate simulations of fluence delivery is an accurate

52

description of the FR distribution.5–7

The performance of UV photoreactors is governed by fluence (dose) Numerical models have been widely

An

53

Several mathematical models have been developed for simulation of FR

54

distributions in UV reactors, including the multiple point source summation (MPSS),8

55

line source integration (LSI),9 and extense source with volumetric emission (ESVE)

56

models.10 Meanwhile, hardware and analytical methods to measure (local) FR have

57

been developed to validate optical field models.11,12

58

effective for quantification of the effects of dissolved compounds that may absorb UV

59

radiation.13, 14 The fundamental optical behavior of these dissolved compounds (i.e.,

60

absorbance) can be accurately measured using a spectrophotometer based on the

61

Lambert-Beer Law.

62

In general, FR models are

When UV radiation is imposed on particle suspensions, photon absorption and

63

scattering (or reflection) by suspended particles will occur.

64

behavior is difficult to simulate using existing models, and has not been examined in a

3

ACS Paragon Plus Environment

This complex optical

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 4 of 30 by Li et al.

65

systematic manner, largely because of the dispersed nature of the suspended particles.

66

In fact, absolute scattering does not induce photon energy consumption; scattered

67

photons remain available for disinfection or photolysis, while absorbed photons are no

68

longer available for reactions.15, 16 Hence the Lambert-Beer law is insufficient for

69

FR distribution simulation in the presence of particle suspensions, and measurements

70

based on standard spectrophotometers do not allow independent characterization of

71

the optical behavior of suspensions. Therefore, in the absence of numerical methods

72

to account for the optical effects of suspended particles, experimental measurements

73

of the FR field could be used to provide direct evidence of the effects of particles.

74

Turbidity measurements are commonly used to characterize the particle

75

concentration in water.17,18

76

generally based on visible radiation, this commonly applied optical parameter has also

77

been used to inform water quality assessments that are applied for performance

78

analysis of UV photoreactors.

79

measured using an electronic turbidity meter, usually expressed in nephelometric

80

turbidity units (NTUs).

81

approved by the United States Environmental Protection Agency (USEPA) and the

82

International Organization for Standardization (ISO), including EPA Method 180.1,

83

ISO7027, Great Lakes Instrument Method 2 (GLI 2), Hach Method 10133, etc.

84

These methods are based on nephelometry, wherein a visible light beam and a

85

photodetector are positioned to allow measurement of visible light scattered

86

perpendicular to the incident beam.

Despite the fact that turbidity measurements are

At present, turbidity in water treatment is usually

Multiple turbidity measurement methods have been

4

ACS Paragon Plus Environment

Page 5 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

87

by Li et al.

To date, most studies of the effects of turbidity on UV FR fields have involved

88

empirical investigations based on bench-scale reactors.

Passantino reported that clay

89

turbidity up to 12 NTU does not have an obvious effect on MS2 inactivation.20

90

Amoah found that UV inactivation of C. parvum and Giardia cysts was not impacted

91

by 10 NTU of natural turbidity.21

92

suspended particles have different inhibition impacts on the performance of MS2

93

virus inactivation by UV.22

94

sludge floc particles can protect MS2 coliphage and bacteriophage T4 from UV

95

inactivation.23

96

spherical actinometry and natural particles.24

97

could yield increases in local FR, depending on the nature of the particles.

However, Malley observed that different

Templeton demonstrated that humic acid and activated

Mamane et al. examined the effects of turbidity on the FR field using Their work demonstrated that particles

98

A micro fluorescent silica detector (MFSD) was developed for in-situ

99

measurement of FR distributions in UV photoreactors in our previous research.12, 25

100

The key features of this detector include 360-degree response, high stability, fast

101

response, water-resistance, and small volume; these features have allowed it to be

102

used for the examination of a wide range of optical characteristics of UV

103

photoreactors, including the effect of inner-wall reflection,25 polychromatic lamp

104

reactor behavior,14 and multi-lamp reactor behavior.26

105

hypothesized that this detector would be suitable for the examination of the turbidity

106

effect in UV photoreactors.

107

impact of turbidity on the FR distribution in a UV reactor.

Therefore, it was

This work aimed to clarify, by use of the MFSD, the

5

ACS Paragon Plus Environment

Environmental Science & Technology

Page 6 of 30

Submission to Environ. Sci. Technol.

by Li et al.

108

A second hypothesis was that measurements of reflectance and scattering at the

109

characteristic wavelength of the UV source would yield better indications of the

110

effects of suspended particles on the FR field than traditional measurements of

111

turbidity.

112

namely SiO2, MgO, and TiO2) with a range of reflection coefficients were examined.

113

The optical properties of the solid particles and suspension were characterized

114

independently, and then UV FR distributions in various particle suspensions at various

115

turbidities were measured.

116

of suspended particles on the FR field of a UV photoreactor.

117

promise for practical examination and monitoring of the particle effect on UV

118

photoreactors used in water treatment.

119

■ EXPERIMENTAL SECTION

With this in mind, three different substituted oxide powders (SOPs,

This study provides experimental evidence of the effects The results suggest

120

Chemicals and analytical methods. Three SOPs, namely TiO2 (R902, DuPont

121

Co., particle diameter (PD) ca. 400 nm, USA), SiO2 (Cabot M-5, PD ca. 300 nm,

122

USA), and MgO (Sinopharm Chemical Reagent Co., PD ca. 1000 nm, China), were

123

used to prepare particle suspensions with various turbidities.

124

coefficients of the solid SOPs were measured at wavelengths of 254, 600, and 800 nm

125

using a spectrophotometer (U-3900, Hitachi, Japan) equipped with an integrating

126

sphere attachment; reported reflectance values were the sum of specular and diffuse

127

measurements.

128

individually for the reflection coefficient measurement.27

129

reflection coefficients of the MgO powder than that of the initial standard plate of this

The reflection

Powders of all three SOPs were compressed into tablet mold

6

ACS Paragon Plus Environment

Because of higher

Page 7 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

130

spectrophotometer (i.e., aluminum-evaporated plane) in the wavelength range of 240–

131

280 nm, MgO powder was used as the standard plate in this study.

132

reported reflection coefficients are relative to those of MgO.

Therefore,

133

Aqueous suspensions of SOPs were prepared with deionized (DI) water.

A

134

HACH 2100N turbidity meter was employed to measure the turbidity of aqueous

135

particle suspensions.

136

the range of 0−1000 NTU.

137

halogen-filled tungsten filament lamp (emission spectrum spanned the wavelength

138

range 400−1800 nm) to meet the reporting requirements of EPA Method 180.1. This

139

method is based on the principles of nephelometry, and the photodetector, whose

140

response spectral peak was reported to be between 400−600 nm, received light

141

scattered perpendicular to the incident light path.

142

meter did not allow measurements based on radiation in the UV portion of the

143

electromagnetic spectrum, a fluorescence spectrophotometer (Cary Eclipse, Varian,

144

Australia) with a 1 cm path-length cell was utilized to investigate the scattering

145

behavior of each SOP suspension at specific wavelengths in the UVC (254 nm),

146

visible (600 nm), and IR (800 nm) ranges.

147

with the turbidity meter, the excitation and emission beams are perpendicular to each

148

other.

149

set to the same value.

150

represent the scattering property of the test suspension at each of the chosen

151

wavelengths.

Precision of 0.01 NTU was reported by the manufacturer for The turbidity meter was equipped with a stable

Additionally, because the turbidity

In the fluorescence spectrophotometer, as

For these measurements, the excitation and emission wavelengths were both Hence, the receiving “fluorescence” signal was used to

Note that in common spectrofluorometric applications, the emission

7

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 8 of 30 by Li et al.

152

wavelength is set to a longer value than the excitation wavelength, hence the excited

153

fluorescent signal of the suspension cannot interfere the scattering property test.

154

addition, absorbance at each of the three characteristic wavelengths (i.e., 254 nm, 600

155

nm, and 800 nm) of each SOP suspension was measured with a spectrometer

156

(UV2600, Shimadzu, Japan) in a 1 cm path-length cell.

In

157

FR Distribution Measurement platform. Figure 1 represents a schematic

158

diagram of the experimental platform for measuring the FR distributions in a

159

single-lamp UV photoreactor.

160

outside diameter), whose fluorescence was received by a BPW 34 B silicon PIN

161

photodiode, and amplified and displayed by a multimeter, was used as the FR

162

measurement detector in this study.

163

diameter = 190 mm and inner length = 1000 mm) wrapped in a black cotton cloth was

164

used; UV reflectance from the air/quartz/air interfaces of this reactor had previously

165

been estimated at approximately 8%.24 A single low pressure mercury Hg lamp (GL

166

Type, Xiashi Wanhua Co., China) with a quartz sleeve (radius = 11.5 mm) was housed

167

in the center of the cylindrical quartz photoreactor.

168

length = 297 mm; total output power 16 W; UVC efficiency at 254 nm = 26%.

169

A cylindrical MFSD (1.0 mm length × 0.3 mm

A cylindrical quartz-wall UV reactor (inner

The lamp parameters were: arc

The MFSD was fixed onto a two-dimensional guideway with its vertical axis

170

oriented parallel to the lamp axis (Figure 1).

This configuration allowed the detector

171

to move either parallel (displacement precision of 1 mm) or perpendicular

172

(displacement precision of 0.01 mm) to the lamp axis to measure the axial or radial

173

FR distribution, respectively.

The calibration process of the MFSD at 254 nm has

8

ACS Paragon Plus Environment

Page 9 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

174

been described previously.12

175

inserted into the gap between the lamp and the sleeve to monitor lamp output

176

fluctuations.

177

0.5ºC.

178

A second MFSD, used as a reference detector, was

Water temperature was controlled by a recirculator with precision of

FR Distribution Measurement process.

The FR distributions in the reactor

179

were measured for each of several particle suspensions.

180

suspensions of SiO2 and TiO2, the SiO2 powder was first added to reach the turbidity

181

requirement (e.g., 5.0 NTU for the SiO250%-TiO250% case with 10.0 NTU) and then

182

the TiO2 powder was added to reach the final turbidity requirement (e.g., 10.0 NTU

183

for the SiO250%-TiO250% case with 10.0 NTU).

184

Note that for mixed

For each suspension, the test MFSD was placed parallel to the UV lamp at radial

185

distances of 5.00 and 60.00 mm away from the quartz sleeve.

For each fixed radial

186

distance, the detector was moved parallel to the lamp axis to measure the axial FR

187

distribution in the UV reactor.

188

lamp axis at the central cross-section of the lamp to measure the radial FR distribution.

189

The detector was maintained at each test point for one minute to allow a data

190

acquisition switch unit (34971A, Agilent Co., USA) to collect 30 readings from the

191

test MFSD.

192

The final result for each test location was divided by the corresponding reading of the

193

reference detector to correct any potential error induced by the lamp output

194

fluctuation.

195

■ RESULTS AND DISCUSSION

Then the detector was moved perpendicular to the

For each location, the last ten readings were recorded and averaged.

9

ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 30

Submission to Environ. Sci. Technol.

196

by Li et al.

Reflection coefficients of various solid particles.

Table 1 lists the reflection

197

coefficients in the UVC region (i.e., 254 nm) and visible/IR region (i.e., 600 and 800

198

nm) of the three solid SOPs (i.e., MgO, TiO2 and SiO2) that were compressed into

199

plates.

200

reflection coefficients for this material at all wavelengths were reported as 1.000.

201

the visible/IR light region, all three SOPs displayed similar and high reflectance,

202

while at 254 nm the reflection coefficient of TiO2 was substantially lower than other

203

two SOPs.

204

behavior in the visible/IR region; however, the low UV reflection coefficient of TiO2

205

may be expected to yield a different UV FR distribution as compared to the other two

206

SOP suspensions.

207

study were pigment grade, which often contain other metal oxides (e.g., SiO2, Al2O2,

208

ZrO2) as surface modifications.

209

on the optical behavior of TiO2 particles.

210

TiO2 particles may be different than those of other TiO2 particles.

211 212

Because MgO was used as the reference standard for reflection in this study, In

These results indicate that the all three SOPs may have similar optical

It should be noted that the TiO2 particles selected for use in this

These surface impurities can have profound effects As such, the optical characteristics of these

Table 1 Scattering properties of various suspensions.

Measurements of UVT values

213

at 600 nm as a function of the turbidity for various SOP suspensions are shown in

214

Figure 2a. Because of their similar reflection coefficients at 600 nm, all three SOPs

215

displayed similar relationships between UVT600 and turbidity.

216

suspensions did not display the same trends for UVT254 vs. NTU (Figure 2b) and

217

scattering intensity (SI) at 254 nm, measured by the fluorescent spectrometer and

10

ACS Paragon Plus Environment

However, the particle

Page 11 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

218

defined by the strength of the signal scattered at an angle perpendicular to the incident

219

beam (Figure 2c).

220

suspension; therefore, as turbidity increases it is reasonable to expect a decrease in

221

UVT254 and a corresponding increase in SI.

222

suspensions, the SiO2 suspension demonstrated the fastest decrease of UVT254 and the

223

fastest increase of SI, while the TiO2 suspension demonstrated opposite trends.

224

These results suggest that these two parameters are complementary.

225

measurements in the spectrometer and fluorescent spectrometer, incident UV photons

226

will be scattered by an SOP suspension in the test cell, which would decrease the rate

227

at which photons are received by the sensor oriented in the incident direction (i.e.,

228

UVT254) and increase the rate at which photons are received by the sensor oriented

229

perpendicular to the incident direction (i.e., SI).

230

could be used to evaluate the scattering property of the suspension, and in this study

231

the SiO2 and TiO2 suspensions displayed the strongest and weakest scattering ability,

232

respectively.

A high particle concentration translates to high turbidity of a

However, among the three SOP

For the

Therefore, these two parameters

Figure 2

233 234

Note that the results of UVT254 and SI254 for the suspensions did not absolutely

235

agree with the relative reflection coefficient measurements reported at 254 nm in

236

Table 1.

237

behavior of particles will depend on several factors, including particle size, particle

238

shape, particle concentration, and chemical composition.

239

data in presented in Table 1 only express the effects of chemical composition.

We hypothesize that this is attributable to the fact that the scattering

11

ACS Paragon Plus Environment

The reflection coefficient By

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 12 of 30 by Li et al.

240

contrast, the scattering property test (i.e., Figures 2b and 2c) is an integral

241

characterization for the suspension, providing a measure of optical (scattering)

242

behavior that accounts for all of the factors listed above.

243

Radial FR distribution in various particle suspensions.

Figure 3 illustrates

244

the measured radial FR distributions in the central cross-section of the UV reactor

245

filled with DI water, as well as the SiO2, TiO2, and MgO suspensions with various

246

turbidities. When the test MFSD was moved from 10 to 70 mm from the sleeve

247

surface, the FR decreased for all conditions.

248

3a), the local measurements of FR for all values of turbidity were higher than the

249

corresponding measurements in DI water.

250

also increased.

251

or redirect UV photons in various directions, thereby allowing photons to travel along

252

a longer optical path than they would in the absence of these particles. In theory, this

253

provides more opportunities for photons to interact with photochemical or

254

photophysical targets in the system, such as a suspended microorganisms, absorbing

255

dissolved chemicals, or an MFSD detector.

256

reflective particles could promote photoreactor efficiency.

257

reported previously with an aluminum clay suspension.23

258

For the SiO2 suspension (i.e., Figure

As turbidity increased, local FR values

These results imply that SiO2 particles in water will strongly reflect

Therefore, the presence of suspended, Similar results have been

Figure 3

259

Similar results were also observed with the MgO suspensions shown in Figure 3b.

260

The FR distribution in the MgO suspension increased with increasing turbidity and

261

was consistently higher than with DI water.

However, for the TiO2 suspensions, a

12

ACS Paragon Plus Environment

Page 13 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

262

consistent increase of FR, relative to DI water, was evident only for the 2.0 NTU

263

suspension.

264

locations more distant from the lamp.

265

the FR distribution dropped.

266

absorption at 254 nm (i.e., low refection coefficient) and low scattering ability of the

267

TiO2 particles, as compared to the other two SOPs.

268

reactor wall, photons had more opportunities to be absorbed by the TiO2 particles and

269

were less likely to be scattered than with the other two particle suspensions.

270

FR enhancement was more evident in the near-lamp region than for As turbidity increased from 2.0 to 20.0 NTU,

This was consistent with the relatively strong

As the MFSD approached the

Axial FR distribution in various particle suspensions.

The axial FR

271

distributions in the far-lamp region of the test UV reactor (60 mm from the sleeve

272

surface) for each SOP suspension are illustrated in Figure 4.

273

the MFSD was moved from the lamp center to the lamp end.

274

region, the FR values of all three SOPs were similar for conditions of turbidity of 2.0

275

NTU; at this relatively low value of turbidity, the effects of suspended particles on the

276

FR field were limited.

277

the SiO2 and MgO suspensions were higher than those of the TiO2 suspension at the

278

same turbidity.

279

scattering behavior of the particles (Figures 2b and 2c).

280

the near-lamp region (5 mm from the sleeve surface) for each particle suspension are

281

shown in Figure S1.

282

were similar in all turbidity cases.

283

The FR decreased as In the central lamp

For turbidities of 10.0 NTU and 20.0 NTU, the FR values for

These results were also consistent with the previously described The axial FR distributions in

Because of the short path-length, FR values of all three SOPs

Figure 4

13

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

284

Mixed particle suspensions.

Page 14 of 30 by Li et al.

The particles in practical waters come principally

285

from soil (erosion).18, 19 The composition of natural particles is variable and depends

286

largely on local geology and land use practices.

287

comprise mixtures of silica and metal oxides, as well as organic materials.

288

for the practical applications of UV photoreactors, it is relevant to examine the

289

behavior of mixed particle suspensions involving mixtures of highly-reflective (e.g.,

290

SiO2) and minimally-reflective (TiO2) particles.

291

suspensions of TiO2 and SiO2 particles were prepared.

292

mixtures of SiO2 and TiO2 with turbidity of 10.0 NTU are shown in Figure 5.

293

Turbidity contribution ratios of SiO2 to TiO2 were 7:3, 5:5, and 3:7. Because of the

294

stronger scattering property of SiO2 as compared to that of TiO2, the FR distribution

295

of the suspension with SiO270%–TiO230% was higher than that of suspension with

296

SiO230%–TiO270%, and the FR distribution of the suspension with SiO250%–TiO250%

297

was intermediate to the other two suspensions. Similar results were observed in the

298

suspensions with turbidity of 20.0 NTU, as shown in Figure S2.

299 300

Natural particle suspensions likely Hence,

To address this issue, mixed Radial FR distributions for

Figure 5 Weighted average FR and minimum FR.

Tables 2 and S1 illustrate the

301

weighted average FR (WAFR) values and the FR values at 70 mm distance to the

302

sleeve surface in the central cross-section (regarded as the minimum FR) for DI water,

303

as well as for the SiO2, TiO2, and MgO suspensions with various turbidity values.

304

The WAFR values for the SiO2 suspensions with turbidity values of 2.0, 10.0, and

305

20.0 NTU were 1.05, 1.10, and 1.34 times higher than those of DI water, respectively.

14

ACS Paragon Plus Environment

Page 15 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

306

Similarly, the minimum FR values for the SiO2 suspensions with turbidity values of

307

2.0, 10.0 and 20.0 NTU were 1.02, 1.12, and 1.29 times higher than those in DI water,

308

respectively.

309

end of the dose distribution), which limits the performance of UV disinfection.28

The minimum FR will be related to the minimum dose (or low-dose

310

The results reported herein, which were developed through the use of the MFSD,

311

demonstrate that the presence of reflective suspended particles (i.e., particles that

312

scatter UV photons) can actually enhance the FR field and the overall performance of

313

a photoreactor.

314

This seems contrary to the conventional view of UV photoreactors in the sense that

315

turbidity is commonly regarded as an inhibiting factor for performance of these

316

systems.

317

the reactor.23

A previous study also demonstrated this enhancement effect.23

In fact, the reflection by the particles could extend the photon pathlength in

318

The low photon energy loss during the absorption/reflection process on SiO2

319

particles (high UV reflection coefficient) could reduce interactions of UV photons

320

with the reactor wall where they would be absorbed or inefficiently reflected, thereby

321

enhancing WAFR in the reactor.

322

can shield microbes from UV exposure, thereby counteracting the benefits described

323

above.

324

of UV disinfection systems can be quite complex.20–23, 29

325

Note that for UV disinfection, suspended particles

Therefore, the effects of suspended particles (turbidity) on the performance

Table 2

326

The WAFR values and minimum FR values for the MgO suspensions increased

327

with increasing turbidity, and all WAFR values and minimum FR values for the MgO

15

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 16 of 30 by Li et al.

328

suspensions were higher than those for DI water.

For each turbidity condition (i.e.,

329

2.0, 10.0, or 20.0 NTU), because the MgO particles were less effective at scattering

330

UV254 radiation than SiO2 particles (Figure 2c), its WAFR value and minimum FR

331

values were lower than the corresponding values for the SiO2 suspension.

332

Calculated values of WAFR for the TiO2 suspensions dropped from 6.278 to

333

3.979 mW cm-2 as turbidity increased from 2.0–10.0 NTU. Meanwhile, the minimum

334

FR value decreased from 4.01 to 1.95 mW cm-2 in the turbidity range of 2.0–10.0

335

NTU.

336

that of DI water.

337

highest transmittance (Figure 2b) and the lowest scattering at 254 nm (Figure 2c).

338

Collectively, these attributes increase the likelihood of UV254 photons being incident

339

on and absorbed by the reactor wall.

340

TiO2 powder at 254 nm (i.e., 0.135) implies that a large fraction of the UV photons

341

were absorbed by the TiO2 particles in suspension.

342

MgO suspensions, an inhibition effect of TiO2 particles was observed.

343

At turbidity values of 10.0 and 20.0 NTU, the WAFR values were lower than At any given value of turbidity, the TiO2 suspensions had the

Moreover, the low reflection coefficient of

Therefore, unlike the SiO2 and

Table 2 also illustrates WAFR calculations for mixed suspensions with various

344

ratios of SiO2 to TiO2.

At a fixed value of turbidity, WAFR increased with SiO2

345

content. Moreover, as turbidity increased, the WAFR values of all three mixed

346

suspensions decreased, and the suspension with the highest TiO2 ratio demonstrated

347

the most rapid decrease.

348

SiO2 and TiO2 at fixed turbidity (5.0 NTU and 10.0 NTU) were greater than based on

349

TiO2 alone, and lower than those of SiO2 alone.

In addition, all WAFR values of the mixed suspensions of

16

ACS Paragon Plus Environment

Page 17 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

350

Discussion.

by Li et al.

This study illustrated the impact of turbidity, a conventional water

351

parameter used for characterizing the concentration of suspended particles in water,

352

on the FR field of a simple UV photoreactor.

353

optical properties of suspended particles can induce the FR distribution in a UV

354

reactor at an identical turbidity value.

355

measurements of turbidity involve visible radiation, whereas the optical properties of

356

particles in the UVC range are perhaps more relevant to the behavior of a UV

357

photoreactor.

358

differences in optical behavior in the visible and UVC ranges.

359

conventional turbidity measurements (based on visible light) are inappropriate for

360

characterization of the effects of suspended particles on the FR field.

361

measurements of reflectance and scattering at the characteristic wavelength of the UV

362

source are more representative of the effects of suspended particles on the FR field.

The results demonstrate that several

This is explained by the fact that conventional

Moreover, differences in particle composition are associated with Therefore,

As alternatives,

363

The measurements reported herein demonstrate that the optical field (FR

364

distribution or WAFR) was principally dependent on the photon scattering ability of

365

the suspension, which corresponded with measurements of photon absorption and

366

reflection by the solid powders in the suspension (reflection coefficient), as well as

367

other factors such as the particle size and shape.29

368

turbidity, particles that absorb poorly in a UVC wavelength range (e.g., SiO2 and

369

MgO) could enhance the FR distribution and performance of a photoreactor.

370

Conversely, for particles that absorb strongly in UVC range (e.g., TiO2 ), their

371

presence in suspension could diminish the FR distribution.

For suspensions with identical

17

ACS Paragon Plus Environment

Meanwhile, the

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 18 of 30 by Li et al.

372

turbidity value could represent the particle concentration in the suspensions, implying

373

that the increasing turbidity could amplify the effect of the particles on the FR

374

distribution in the reactor, including positive effects (e.g., SiO2 and MgO), and

375

negative effects (e.g., TiO2).

376

The effects of suspended particles on the optical behavior of UV radiation were

377

also characterized by measurements of transmittance and scattering intensity

378

developed using a spectrometer and a fluorescence spectrometer, respectively.

379

two tests have the potential to be used as characterization methods for evaluating the

380

effects of turbidity on FR distributions.

381

■ ASSOCIATED CONTENT

382

Supporting Information.

383

distributions in the near-lamp region (5 mm from the sleeve surface) with various

384

particle suspensions with various turbidities (Figure S1), radial FR distributions with

385

mixed particle suspensions with various turbidities of 20.0 NTU (Figure S2), and

386

minimum FR values (Table S1)

387

■ AUTHOR INFORMATION

388

Corresponding Authors

389

*

390

Blatchley III)

391

*Phone: +86 10 62849632; fax: +86 10 62923541; e-mail: [email protected] (Z.

392

Qiang).

These

Two figures and one table are provided. Axial FR

Phone: 1-765-494-0316; fax: 1-765-494-0395; e-mail: [email protected] (E.R.

18

ACS Paragon Plus Environment

Page 19 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

393

Notes

394

The authors declare no competing financial interest.

395

■ ACKNOWLEDGEMENTS

396

The authors gratefully acknowledge the financial support from the National

397

Natural Science Foundation of China (51408592, 21590814, 51525806), Edward M.

398

Curtis Visiting Professorship of Purdue University, and National Geographic Air and

399

Water Conservation Fund (GERC23-15).

400

■ REFERENCES

401

(1)

Rizzo, L.; Della Sala, A.; Fiorentino, A.; Li Puma, G., Disinfection of urban

402

wastewater by solar driven and UV lamp - TiO(2) photocatalysis: Effect on a multi

403

drug resistant Escherichia coli strain. Water Res. 2014, 53, 145–52.

404

(2)

Zhao, Q.; Shang, C.; Zhang, X.; Ding, G.; Yang, X., Formation of

405

halogenated organic byproducts during medium-pressure UV and chlorine coexposure

406

of model compounds, NOM and bromide. Water Res. 2011, 45, (19), 6545–54.

407 408 409 410 411 412 413 414

(3)

Duran, J. E.; Taghipour, F.; Mohseni, M., Evaluation of model parameters for

simulating TiO(2) coated UV reactors. Water Sci. Technol. 2011, 63, (7), 1366–72. (4)

Bagheri, M.; Mohseni, M., Computational fluid dynamics (CFD) modeling of

VUV/UV photoreactors for water treatment. Chem. Eng. J. 2014, 256, 51–60. (5)

Taghipour, F.; Sozzi, A., Modeling and design of ultraviolet reactors for

disinfection by-product precursor removal. Desalination 2005, 176, (1-3), 71–80. (6)

Sozzi, D. A.; Taghipour, F., UV reactor performance modeling by Eulerian

and Lagrangian methods. Environ. Sci. Technol. 2006, 40, (5), 1609–1615. 19

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

415

(7)

Page 20 of 30 by Li et al.

Jin, S. S.; Linden, K. G.; Ducoste, J.; Liu, D., Impact of lamp shadowing and

416

reflection on the fluence rate distribution in a multiple low-pressure UV lamp array.

417

Water Res. 2005, 39, (12), 2711–2721.

418 419 420

(8)

Jacob, S. M.; Dranoff, J. S., Light intensity profiles in a perfectly mixed

photoreactor. AIChE J. 1970, 16, (3), 359–366. (9)

Blatchley III, E. R., Numerical modelling of UV intensity: Application to

421

collimated-beam reactors and continuous-flow systems. Water Res. 1997, 31, (9),

422

2205–2218.

423

(10)

Irazoqui, H. A.; Cerda, J.; Cassano, A. E., Radiation profiles in an empty

424

annular photoreactor with a source of finite spatial dimensions. AIChE J. 1973, 19, (3),

425

460–467.

426

(11)

Rahn, R. O.; Bolton, J.; Stefan, M. I., The iodide/iodate actinometer in UV

427

disinfection: determination of the fluence rate distribution in UV reactors. Photochem.

428

Photobiol. 2006, 82, (2), 611–615.

429

(12)

Li, M. K.; Qiang, Z.; Li, T. G.; Bolton, J. R.; Liu, C. L., In situ measurement

430

of UV fluence rate distribution by use of a micro fluorescent silica detector. Environ.

431

Sci. Technol. 2011, 45, (7), 3034–3039.

432

(13)

Bolton, J. R., Calculation of ultraviolet fluence rate distributions in an

433

annular reactor: Significance of refraction and reflection. Water Res. 2000, 34, (13),

434

3315–3324.

435 436

(14)

Li, M. K.; Qiang, Z. M.; Wang, C.; Bolton, J. R.; Blatchley III, E. R.,

Experimental assessment of photon fluence rate distributions in a medium-pressure

20

ACS Paragon Plus Environment

Page 21 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

437

UV photoreactor. Environ. Sci. Technol. 2017, 51, (6), 3453–3460. (15)

438 439

Bohren, C. F.; Nevitt, T. J., Absorption by a sphere: a simple approximation.

Appl. Opt. 1983, 22, (6), 774–775. (16)

440 441

by Li et al.

Qualls, R. G.; Johnson, J. D., Bioassay and dose measurement in UV

disinfection. Appl. Environ. Microbiol. 1983, 45, (3), 872–877. (17)

442

American Public Health Association, Standard Methods for the Examination

443

of Water and Wastewater, 14th ed., American Public Health Association, New York,

444

1975.

445

(18)

446

4th ed., McGraw-Hill, New York, 2008. (19)

447 448

Davis, M. L.; Cornwell, D. A., Introduction to Environmental Engineering,

Mihelcic, J.R.; Zimmerman, J.B., Environmental Engineering: Fundamentals,

Sustainability, Design, 2nd ed., John Wiley & Sons, Hoboken, NJ. (20)

449

Passantino, L.; Malley, J.; Knudson, M.; Ward, R.; Kim, J. Effect of low

450

turbidity and alage on UV disinfection performance. J. Am. Water Works Assoc. 2004,

451

96(6), 123–137. (21)

452

Amoah, K.; Craik, S.; Smith, D. W.; Belosevic, M. Inactivation of

453

Cryptosporidium oocysts and Giardia cysts by ultraviolet light in the presence of

454

natural particulate matter. J. Water Supply Res. Technol. Aqua. 2005, 54, 165–178. (22)

455

Malley, R.; Vernacchio, L.; Devincenzo, J.; Ramilo, O.; Dennehy, P. H.;

456

Meissner, H. C.; Gruber, W. C.; Jafri, H. S.; Sanchez, P. J.; Macdonald, K.; Montana,

457

J.

458

(23)

Templeton, M. R.; Andrews, R. C.; Hofmann, R., Inactivation of

21

ACS Paragon Plus Environment

Environmental Science & Technology Submission to Environ. Sci. Technol.

Page 22 of 30 by Li et al.

459

particle-associated viral surrogates by ultraviolet light. Water Res. 2005, 39, (15),

460

3487–3500.

461

(24)

Mamane, H.; Ducoste, J. J.; Linden, K. G., Effect of particles on ultraviolet

462

light penetration in natural and engineered systems. Appl. Opt. 2006, 45, (8), 1844–

463

1856.

464

(25)

Li, M. K.; Qiang, Z. M.; Bolton, J. R.; Ben, W. W., Impact of reflection on

465

the fluence rate distribution in a UV reactor with various inner walls as measured

466

using a micro-fluorescent silica detector. Water Res. 2012, 46, (11), 3595–3602.

467

(26)

Li, M. K.; Qiang, Z. M.; Bolton, J. R., In situ detailed fluence rate

468

distributions in a UV reactor with multiple low-pressure lamps: Comparison of

469

experimental and model results. Chem. Eng. J. 2013, 214, 55–62.

470

(27)

Cao, Y.; Feng, J.; Wu, P. Preparation of organically dispersible graphene

471

nanosheet powders through a lyophilization method and their poly (lactic acid)

472

composites. Carbon, 2010, 48(13), 3834–3839.

473

(28)

Chiu, K.; Lyn, D. A.; Savoye, P.; Blatchley III, E. R. Integrated UV

474

disinfection model based on particle tracking. J. Environ. Eng.-ASCE, 1999, 125(1),

475

7–16.

476

(29)

477

Mamane, H., Impact of particles on UV disinfection of water and wastewater

effluents: A review. Rev. Chem. Eng., 2008, 24(2–3), 67–157.

478 479

22

ACS Paragon Plus Environment

Page 23 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

Table 1. wavelengths.

by Li et al.

Relative reflection coefficients of various materials at different At each wavelength, the reflection coefficient was normalized

relative to the value recorded for MgO. Wavelength (nm)

MgO

TiO2

SiO2

254

1.000

0.135

0.849

600

1.000

0.929

0.947

800

1.000

0.916

0.954

23

ACS Paragon Plus Environment

Environmental Science & Technology

Page 24 of 30

Submission to Environ. Sci. Technol.

Table 2.

by Li et al.

Weighted average fluence rates for various particle suspensions. Weighted average fluence rate (mW cm-2)

Suspensions 0 NTU

2.0 NTU

10.0 NTU

20.0 NTU

SiO2

6.120

6.386

7.784

TiO2

6.278

4.870

3.979

MgO

5.973

6.244

6.454

5.0 NTU

10.0 NTU

20.0 NTU

SiO230%–TiO270%

5.911

4.890

3.996

SiO250%–TiO250%

N/A

5.111

4.831

SiO270%–TiO230%

5.985

5.314

5.213

Deionized water

5.824

24

ACS Paragon Plus Environment

Page 25 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

Figure 1. Schematic diagram of the experimental platform for measuring the FR distributions in a UV photoreactor.

25

ACS Paragon Plus Environment

Environmental Science & Technology

Page 26 of 30

Submission to Environ. Sci. Technol.

by Li et al.

UVT600

1.1

(a)

1.0 0.9

SiO2

0.8

TiO2

0.7

MgO

UVT254

0.6 1.1

(b)

1.0 0.9 0.8 0.7

SI (A.U.)

0.6

(c)

600 400 200 0 0

10

20

30

40

50

Turbidity (NTU) Figure 2. UVT at 600 nm (a), UVT at 254 nm (b), and scattering intensity (SI) at 254 nm (c) as a function of turbidity.

All tests were repeated three times and all values of

relative standard derivation were lower than 10%.

26

ACS Paragon Plus Environment

Page 27 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

DI water 2.0 NTU 10.0 NTU 20.0 NTU

20 16 12

(a)

Fluence rate (mW cm-2)

8 4 0 20

(b)

16 12 8 4 0 20

(c)

16 12 8 4 0 0

10

20

30

40

50

60

70

80

Distance from the lamp center (mm)

Figure 3. Radial FR distributions with various turbidities in (a) SiO2 suspension, (b) TiO2 suspension, and (c) MgO suspension.

27

ACS Paragon Plus Environment

Environmental Science & Technology

Page 28 of 30

Submission to Environ. Sci. Technol.

by Li et al.

20

(a)

SiO2

16

TiO2

12

MgO

8 -2

Fluence rate (mW cm )

4 0 20

(b)

16 12 8 4 0 20

(c)

16 12 8 4 0 0

3

6

9

12

15

Distance from the lamp center (mm)

Figure 4. Axial FR distributions in the far-lamp region (60 mm from the sleeve surface) in various particle suspensions with various turbidities of (a) 2.0 NTU, (b) 10.0 NTU, and (c) 20.0 NTU.

28

ACS Paragon Plus Environment

Page 29 of 30

Environmental Science & Technology Submission to Environ. Sci. Technol.

by Li et al.

20

-2

Fluence rate (mW cm )

18 SiO2 70%-TiO2 30% SiO2 50%-TiO2 50% SiO2 30%-TiO2 70%

16 14 12 10 8 6 4 2 0

10

20

30

40

50

60

70

80

Distance from the sleeve surface (mm)

Figure 5. Radial FR distributions in mixed particle suspensions with turbidity of 10.0 NTU.

29

ACS Paragon Plus Environment

Environmental Science & Technology

Table of Contents 211x198mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 30 of 30