uv/nitrilotriacetic acid process as a novel strategy for efficient … · 2018. 2. 5. · 1 1...

Subscriber access provided by Caltech Library

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

Article

UV/nitrilotriacetic acid process as a novel strategy for efficientphotoreductive degradation of perfluorooctane sulfonate

Zhuyu Sun, Chaojie Zhang, Lu Xing, Qi Zhou, Wenbo Dong, and Michael R HoffmannEnviron. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b05912 • Publication Date (Web): 03 Feb 2018

Downloaded from http://pubs.acs.org on February 5, 2018

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are postedonline prior to technical editing, formatting for publication and author proofing. The American ChemicalSociety provides “Just Accepted” as a service to the research community to expedite the disseminationof scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear infull in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fullypeer reviewed, but should not be considered the official version of record. They are citable by theDigital 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. Aftera manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Website and published as an ASAP article. Note that technical editing may introduce minor changesto the manuscript text and/or graphics which could affect content, and all legal disclaimers andethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors orconsequences arising from the use of information contained in these “Just Accepted” manuscripts.

1

UV/nitrilotriacetic acid process as a novel strategy for 1

efficient photoreductive degradation of perfluorooctane 2

sulfonate 3

4

Zhuyu Sun,†‡

Chaojie Zhang,*†‡

Lu Xing,†‡

Qi Zhou,†‡

Wenbo Dong,&

5

Michael R. Hoffmann§ 6

†State Key Laboratory of Pollution Control and Resources Reuse, College 7

of Environmental Science and Engineering, Tongji University, Shanghai 8

200092, China 9

‡ Shanghai Institute of Pollution Control and Ecological Security, 10

Shanghai 200092, China 11

&Shanghai Key Laboratory of Atmospheric Particle Pollution and 12

Prevention, Department of Environmental Science & Engineering, Fudan 13

University, Shanghai 200433, China 14

§Linde-Robinson Laboratories, California Institute of Technology, 15

Pasadena, California 91125, United States 16

*Corresponding author. Tel: +86 21 65981831; fax: +86 21 65983869; 17

E-mail address: [email protected] 18

19

Page 1 of 41

ACS Paragon Plus Environment

Environmental Science & Technology

2

Abstract: Perfluorooctane sulfonate (PFOS) is a toxic, bioaccumulative and highly 20

persistent anthropogenic chemical. Hydrated electrons (eaq–) are potent nucleophiles 21

that can effectively decompose PFOS. In previous studies, eaq– are mainly produced 22

by photoionization of aqueous anions or aromatic compounds. In this study, we 23

proposed a new photolytic strategy to generate eaq– and in turn decompose PFOS, 24

which utilizes nitrilotriacetic acid (NTA) as a photosensitizer to induce water 25

photodissociation and photoionization, and subsequently as a scavenger of hydroxyl 26

radical (·OH) to minimize the geminate recombination between ·OH and eaq–. The net 27

effect is to increase the amount of eaq– available for PFOS degradation. The UV/NTA 28

process achieved a high PFOS degradation ratio of 85.4% and a defluorination ratio of 29

46.8% within 10 h. A pseudo first-order rate constant (k) of 0.27 h-1

was obtained. The 30

laser flash photolysis study indicates that eaq– is the dominant reactive species 31

responsible for PFOS decomposition. The generation of eaq– is greatly enhanced and 32

its half-life is significantly prolonged in the presence of NTA. The electron spin 33

resonance (ESR) measurement verified the photodissociation of water by detecting 34

·OH. The model compound study indicates that the acetate and amine groups are the 35

primary reactive sites. 36

Page 2 of 41



3

1. Introduction 37

Perfluorooctane sulfonate (PFOS, C8F17SO3−) is a useful anthropogenic chemical 38

that has been extensively utilized in industrial and commercial applications for 39

decades.1,2

However, due to its high toxicity, environmental persistence, 40

bioaccumulation and global distribution, PFOS is considered hazardous to 41

environmental and human health.3,4

Moreover, because of the high thermal and 42

chemical stability of C–F bond (~116 kcal/mol, almost the strongest in nature2), PFOS 43

is extremely resistant to traditional chemical and biological degradations. 44

Recently, hydrated electrons (eaq–) mediated photoreductive approaches have 45

garnered special attention for PFOS decomposition owing to their extraordinarily high 46

efficiency and relatively mild conditions. However, in order to produce sufficient 47

amount of eaq– to defluorinate PFOS, chemicals such as iodide

5, sulfite

6, chloride

7 and 48

indole derivatives8 are essential. Lyu et al. developed a catalyst-free PFOS 49

photodecomposition method, but it relied on using strong alkaline conditions (pH = 50

11.8) and high temperatures (100 °C).9 Furthermore, due to the rapid reaction between 51

eaq– and oxygen (Eq. 1, rate constant = 1.9×10

10 M

-1·s

-1)10

, the reduction efficiency of 52

eaq– is significantly affected by dissolved oxygen (O2). Thereby, previous eaq

–53

-mediated photoreductive approaches normally require strict anoxic conditions.9, 11

In 54

view of the drawbacks of current photoreductive technologies, we were motivated to 55

propose a new strategy for PFOS removal, which can be conducted under more 56

environmentally relevant conditions and with minimization of undesirable 57

byproducts. 58

Page 3 of 41



4

In addition to electron photodetachment from anions and aromatic compounds12,

59

13, UV photolysis of liquid water generates eaq

– as well (Eq. 2).

14, 15 However, direct 60

photoionization of liquid water under 254 nm irradiation is difficult due to a 61

negligible absorption coefficient at λ > 200 nm.16

In addition, the quantum yield of 62

eaq– from water photoionization is low because the radical species undergo significant 63

geminate recombination. In pure water photolysis, energy deposition takes place in 64

well-separated local volumes called spurs, where the primary products including eaq–, 65

hydroxyl radicals (·OH), hydrogen atoms (·H) and H3O+ are formed in close vicinity 66

with high initial local concentrations.17

However, as the spurs expand through 67

diffusion, a large proportion of the primary products undergo very efficient back 68

reactions (Eq. 3-5). Thus, only a small fraction of eaq– actually escape into the bulk 69

solution to initiate reductive chemical processes.10

Therefore, there is scarcely any net 70

generation of eaq– in pure water photolysis upon low energy excitation. Based on these 71

facts, we now propose a new strategy to enhance the generation of eaq–, i.e., by 72

minimizing the geminate recombination of eaq– with oxidative species after inducing 73

water photoionization. 74

��– + O� → O�

∙ (1)

H�O��

– +∙ OH +∙ H + H�O� (2)

��– +∙ OH → OH (3)

��– +∙ H → H� + OH

(4)

��– + H�O

� →∙ H + H�O (5)

Aminopolycarboxylic acids (APCAs) are compounds that contain several 75

Page 4 of 41



5

carboxylate groups bound to one or more nitrogen atoms. Due to their excellent 76

chelating properties18

, interests in previous studies were mainly focused in their 77

chelation with metal ions.19-21

Furthermore, APCAs contain several donor atoms 78

whose free electron pairs are easily attacked by ·OH. Leitner et al.22

reported that the 79

rate constants for the reactions of APCAs with ·OH can exceed 1010

M-1

·s-1

. These 80

near diffusion controlled values are 2 to 3 orders of magnitude higher than reactions 81

of APCAs with eaq–. Therefore, APCAs are considered to be excellent candidates to 82

scavenge ·OH, and thus increase the apparent quantum yields of eaq– and in turn 83

promote the photoreductive degradation of PFOS. 84

In this study, we chose to use nitrilotriacetic acid (NTA) as a representative 85

APCA to explore its impact on the photoreductive degradation of PFOS. NTA is the 86

first chemical synthetic APCA23

and is more biodegradable than other APCAs24

. In 87

order to unravel the underlying reaction mechanisms, laser flash photolysis, ESR 88

measurement and a model compound study were carried out. Furthermore, the effects 89

of NTA concentration, pH and air were investigated. To our best knowledge, this is 90

the first study that utilizes APCAs as ·OH scavengers to promote the photoreductive 91

degradation of refractory organic pollutants. The findings in this study can provide a 92

new strategy for PFOS remediation and a novel application field for APCAs. 93

Page 5 of 41



6

2. Materials and methods 94

2.1. Chemicals 95

Perfluorooctanesulfonic acid (PFOS, ~40% in water), HPLC grade methanol 96

(≥99.9%), 5,5-Dimethyl-1-pyrroline N-oxide (DMPO, ≥98.0%) and model 97

compounds including trisodium citrate (≥99.0%), ethylenediamine-N,N’-disuccinic 98

acid trisodium salt solution (EDDS, ~35% in water) and methylglycine diacetic acid 99

(MGDA, ≥99.0%) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, Mo, 100

USA). Nitrilotriacetic acid (NTA, ≥98.5%), ammonium hydroxide solution (25%), 101

ammonium chloride (≥99.8%), ethylenediaminetetraacetic acid (EDTA, ≥99.5%), 102

iminodiacetic acid (IDA, ≥98.0%), glycine (≥99.5%), oxamic acid (≥98.0%) and 103

sodium oxalate (≥99.8%) were obtained from Sinopharm Chemical Reagent Co. 104

(Shanghai, China). HPLC grade ammonium acetate (97.0%) was purchased from 105

TEDIA (Fairfield, OH, USA). Sodium perfluoro-1-[1,2,3,4-13

C4]octanesulfonate 106

(MPFOS, ≥99%, 13

C4) acquired from Wellington Laboratories Inc. (Guelph, ON, 107

Canada) was used as the internal standard for the quantification of PFOS. Milli-Q 108

water and deionized water were used throughout the whole experiment. 109

2.2. Reductive defluorination 110

The photoreductive degradation of PFOS was conducted under anoxic conditions 111

in a stainless steel cylindrical reactor (Fig. S1, SI). The outer and inner diameters of 112

the reactor were 100 mm and 60 mm, respectively. A low-pressure mercury lamp (14 113

W, Heraeus, Germany) with quartz tube protection was placed in the center of the 114

reactor, emitting 254 nm UV light. 720 mL solution was added to the reactor. Before 115

Page 6 of 41



7

the reaction, the mixture was bubbled with highly purified nitrogen for 20 mins to 116

remove oxygen. The solution pH was adjusted with ammonium hydroxide-ammonium 117

chloride (NH3·H2O-NH4Cl) buffer. The reaction temperature was held constant at 30

118

oC by the circulating cooling system. After various time intervals, samples of the 119

liquid were analyzed after filtration through 0.22 µm nylon filter (ANPEL Laboratory 120

Technologies, Shanghai, China). 121

2.3. Analytical methods 122

The concentrations of PFOS and possible aqueous-phase intermediate analytes 123

were determined by high-performance liquid chromatography/tandem mass 124

spectrometry (HPLC−MS/MS, TSQTM

Quantum AccessTM

, Thermo Finnigan, San 125

Jose, CA, USA). Ion Chromatography (Dionex, ICS-3000, Thermo Fisher Scientific, 126

USA) equipped with a conductivity detector was used for the analysis of fluoride, 127

nitrate and short chain organic acids. NTA and its degradation products including IDA, 128

oxamate and oxalate were detected by UPLC−MS/MS (ACQUITY UPLC&SCIEX 129

SelexION Triple Quad 5500 System, Waters, MA, USA). More detailed information 130

is available in the SI. 131

2.4. Laser flash photolysis experiment and electron spin resonance (ESR) 132

measurement 133

Nanosecond laser flash photolysis for the detection of eaq– was performed using a 134

Quanta Ray LAB-150-10 Nd:YAG laser at an excitation wavelength of 266 nm. The 135

components of the laser flash photolysis apparatus were as described by Ouyang et 136

al..25

Detailed information is available in the SI. 137

Page 7 of 41



8

The ESR spectra were obtained on a Bruker EMXplus-10/12 ESR spectrometer 138

at room temperature. The instrumental parameters for ESR analysis were as follows: 139

microwave frequency, 9.852 GHz; microwave power, 20 mW; modulation amplitude 140

1 G; modulation frequency, 100 kHz; center field, 3500 G; sweep width, 100 G. The 141

simulations of ESR spectra were obtained with the use of Spinfit in Xenon software. 142

Details are provided in the SI. 143

144

Page 8 of 41



9

3. Results and Discussion 145

3.1. NTA-assisted photoreductive degradation and defluorination of PFOS 146

147

Figure 1. The time profiles of PFOS degradation (a) and defluorination (b) under different 148

Page 9 of 41



10

conditions. The treatment experiment: PFOS (0.01 mM), NTA (2 mM), UV irradiation, N2 149

saturated, pH (10.0); the 1st control experiment (direct photolysis): PFOS (0.01 mM), UV 150

irradiation, N2 saturated; the 2nd

control experiment (UV/buffer): PFOS (0.01 mM), UV irradiation, 151

N2 saturated, pH (10.0); the 3rd

control experiment (without UV irradiation): PFOS (0.01 mM), 152

NTA (2mM), N2 saturated, pH (10.0); the 4th

control experiment (UV/sulfite): PFOS (0.01 mM), 153

SO32-

(2mM), UV irradiation, N2 saturated, pH (10.0). Error bars represent standard deviations of 154

triplicate assays. 155

The photochemical decomposition of PFOS was conducted under 254 nm light 156

irradiation in the presence of NTA under anoxic condition. In order to determine the 157

effect of each parameter, four control experiments were conducted. The results of 158

experiments under different conditions are shown in Fig. 1. Since PFOS has a weak 159

absorption at 254 nm and the quantum yield of eaq– by pure water photolysis is 160

negligible26

, the direct photolysis of PFOS was poor. Only 12.9% of the initial PFOS 161

was decomposed after 10 h of irradiation, with a low defluorination ratio of 3.7%. 162

Adjusting solution pH to 10.0 with NH3·H2O-NH4Cl buffer somewhat enhanced the 163

decomposition of PFOS, with the 10-h PFOS degradation and defluorination ratios 164

increased to 20.9% and 9.9%, respectively. Compared with the UV/buffer process, 165

addition of 2 mM NTA significantly accelerated the degradation and defluorination of 166

PFOS. After 10 h of irradiation, the degradation and defluorination ratios of PFOS in 167

the presence of NTA were 85.4% and 46.8%, respectively. In particular, PFOS 168

degraded rapidly during the initial 0.5 h. The 0.5-h PFOS photodegradation and 169

defluorination ratios by UV/NTA process achieved 45.1% and 18.6%, respectively, 170

Page 10 of 41



11

which are 5.2-fold and 8.5-fold, respectively higher than by UV/buffer process. The 171

results of the 3rd

control experiment indicate that PFOS does not degrade without UV 172

irradiation, even in the presence of NTA. Therefore, both UV and NTA are necessary 173

for the efficient degradation and defluorination of PFOS. 174

Gu et al. (2016) demonstrated that UV/sulfite system exhibited a high efficiency 175

in decomposing PFOS.6 In order to better evaluate the efficiency of UV/NTA process 176

for PFOS degradation, the newly developed approach was compared with UV/sulfite 177

process (the 4th

control). The 10-h degradation and defluorination ratios of PFOS by 178

UV/sulfite process were 60.1% and 29.6%, respectively, which were 0.7-fold and 179

0.63-fold, respectively lower than by UV/NTA process. PFOS degradation by the 180

UV/NTA process follows pseudo first-order kinetics (Fig. S4, SI), with an apparent 181

reaction rate constant (k) of 0.27 h-1

, and a half-life (t1/2) of 2.6 h. The k value for the 182

UV/NTA process was higher than other photochemical approaches including UV/KI 183

process5, UV/sulfite process, UV/Fe

3+ process

27, UV/alkaline 2-propanol process

26 184

and UV/K2S2O8 process28

(Table S1, SI). Therefore, UV/NTA process exhibited a 185

high efficiency in the degradation and defluorination of PFOS. 186

As PFOS decomposed, short-chain-length perfluorinated intermediates including 187

PFBS, PFBA, PFHxS, PFHxA, PFHpA and PFOA were detected (Fig. S5, SI). Based 188

on the concentrations of PFOS, intermediates and F−, the mass balance of F was 189

calculated (Fig. S6, SI). The loss of F recovery during the reaction implies the 190

formation of partially fluorinated intermediates. Therefore, based on the product 191

distribution and F balance, three possible degradation pathways of PFOS are likely to 192

Page 11 of 41



12

occur in UV/NTA process: a) PFOS undergoes direct defluorination to form 193

polyfluorinated intermediates; b) PFOS firstly undergoes desulfonation to form PFOA, 194

and then degrades to shorter-chain-length perfluoroalkyl carboxylic acids (PFCAs); c) 195

C-C bond in PFOS molecules cleaves, in which case short-chain-length 196

perfluoroalkane sulfonates (PFSAs), such as PFHxS and PFBS, were produced. The 197

proposed PFOS degradation pathways in UV/NTA process are similar with other 198

photochemical processes5, 9, 26

, and accord well with the mechanisms suggested by 199

molecular orbitals and thermodynamic analyses6. 200

As mentioned above, most photochemical technologies for PFOS degradation are 201

faced with challenges of secondary pollution due to by-product formation. In order to 202

evaluate the potential post-treatment impact of the UV/NTA process, the degradation 203

products of NTA were quantified. As is shown in Fig. S7, SI, NTA degraded rapidly 204

during the first 1 h of PFOS degradation. 97.7% of the initial NTA decomposed within 205

1 h, along with the concomitant formation of three intermediates including IDA, 206

oxamate and oxalate. The maximum concentrations of IDA, oxamate and oxalate 207

were 0.75 mM, 0.26 mM and 0.68 mM, respectively, observed at 0.5 h, 2 h and 6 h, 208

respectively. IDA and oxamate underwent nearly complete decomposition after 10 h, 209

with their concentrations falling below 0.03 mM. Oxalate was completely 210

decomposed after 14 h (data not shown). Ammonium (NH4+) and nitrate (NO3

−) were 211

the major N-containing end products. Based on the product distribution and the NTA 212

photooxidation mechanism in literatures29, 30

, a possible NTA decomposition pathway 213

is illustrated in Fig. S8, SI. The end products of NTA degradation are NH4+, NO3

− and 214

Page 12 of 41



13

carbon dioxide (CO2), which are innocuous and can be easily removed by biological 215

transformation. Furthermore, NTA is reported to have relatively low environmental 216

risks to sewage treatment or aquatic life.18, 31

It is readily biodegradable by natural 217

microbial processes, and the biodegradation is complete without accumulation of 218

unwanted intermediates.24

In contrast, the secondary pollutions of other approaches 219

appear to be severer. For example, the iodide by-products include residual iodide, 220

iodine, polyiodide and iodate, which may cause detrimental effects to aquatic 221

organisms32

and human health33

. The various iodide species are also potential sources 222

for iodinated disinfection byproducts (iodo-DBPs). Therefore, UV/NTA process is 223

expected to provide a relatively green alternative for efficient photoreductive 224

degradation of PFOS. 225

As is shown in Fig. 1, the PFOS decomposition rate attenuated after 1 h. This is 226

primarily due to the nearly complete degradation of NTA within 1 h. The subsequent 227

PFOS degradation and defluorination after 1 h is attributed to the effect of NTA 228

degradation products such as IDA. The effects of NTA degradation products are 229

discussed in section 3.3. 230

Page 13 of 41



14

3.2. Mechanism of NTA-assisted UV photoreductive degradation of PFOS 231

Photodegradation of organic compounds can take place through direct and/or 232

indirect photolysis. Since direct PFOS photolysis under UV irradiation is very slow, 233

PFOS degradation in UV/NTA process is believed to occur through indirect pathway. 234

In order to ascertain the dominant reactive species that is responsible for the 235

degradation of PFOS, experiments with nitrous oxide (N2O) were conducted. As is 236

shown in Fig. S9 in the SI, PFOS degradation and defluorination were substantially 237

suppressed in the presence of N2O. N2O is a well-known scavenger of eaq–, which 238

quenches eaq– rapidly to form ·OH (Eq. 6, rate constant = 9.1×10

9 M

-1·s

-1)

10, whereas 239

·OH has a poor reactivity towards perfluorinated acids that it cannot decompose PFOS 240

effectively.34, 35

Therefore, these results imply the crucial role of eaq– in the PFOS 241

degradation in UV/NTA process. Besides eaq–, N2O can also react with ·H.

10 However, 242

under alkaline conditions, the yield of ·H is much lower than that of eaq–, and the rate 243

constant for the reaction of N2O and ·H at alkaline pH is 2.1×106 M

-1·s

-1, which is 244

over 3 orders of magnitude lower than that of N2O and eaq–. 10

Therefore, the effect of 245

·H can be excluded. 246

��– +N�O → OH

+∙ OH + N� (6)

247

Page 14 of 41



15

248

Figure 2. Transient absorption spectra following the laser flash photolysis of 10 mM NTA 249

solution at pH 10.0. Insertion in Figure 2 is the decay of eaq– detected at 630 nm in the presence of 250

PFOS with different concentrations. The transient absorption curves are fitted. 251

The N2O scavenging experiment provides an indirect proof for the potential role 252

of eaq–. However, a more direct proof is needed to reveal the formation and decay of 253

eaq– in UV/NTA process. Therefore, a laser flash photolysis study was conducted. The 254

absorption spectrum of intermediates was obtained at 10 nm intervals between 260 255

nm and 700 nm upon the photolysis of 10 mM NTA in N2 saturated water (Fig. 2). 256

The broad optical absorption band with a peak at around 630 nm is attributed to eaq– 257

since it is identical to the eaq– absorption spectrum acquired both theoretically

36 and 258

experimentally37

. The absorption peak of eaq– decayed continuously with monitoring 259

time, indicating eaq– gradually reacts back with other primary species. The decays of 260

Page 15 of 41



16

eaq– in the presence of PFOS with different concentrations are shown in the inserted 261

figure in Fig. 2. In the absence of NTA (i.e., the laser photolysis of pure water), eaq– 262

was not produced. By adding 10 mM NTA, the absorbance at 630 nm increased 263

significantly, confirming the enhanced production of eaq– in the presence of NTA. The 264

addition of PFOS accelerated the decay of eaq–, and its decay rate increased at higher 265

PFOS concentration level. These results further verify the role of eaq– in PFOS 266

degradation. 267

Thus, the generation of eaq– in UV/NTA process has been clearly confirmed by 268

the laser flash photolysis study. However, compared with common eaq– source 269

chemicals like sulfite38

, ferrocyanide39

, iodide40

, and aromatic compounds like 270

pyrenetetrasulfonate41

, the transient yield of eaq– by UV/NTA process is much lower. 271

The relatively low transient yield of eaq– cannot explain the high efficiency of 272

UV/NTA process in PFOS degradation and defluorination. For instance, the 10-h 273

degradation and defluorination ratios of PFOS by UV/NTA process were 1.42-fold 274

and 1.58-fold, respectively higher than by UV/sulfite process. Therefore, we speculate 275

that besides direct photoejection of eaq– from NTA, there may be another dominant 276

mechanism for the efficient generation of eaq–. 277

Grossweiner et al. once proposed an alternative mechanism for the generation 278

of eaq– based on their observations, that is the photoionizaton of water itself sensitized 279

by the light-absorbing substances.12

In this case, a hydroxyl radical is produced by 280

dissociation of the photoionized water.12

According to the UV-Vis spectra (Fig. S10, 281

SI), direct photoionization of water under 254 nm light irradiation can be excluded 282

Page 16 of 41



17

since water is transparent in the near UV spectral domain (λ > 200 nm)16

. However, 283

NTA has an absorbance at 254 nm, which means NTA is able to absorb photons and 284

possibly sensitize water ionization. Furthermore, APCAs can form hydration 285

complexes with several water molecules through hydrogen bonding. For instance, 286

eight water molecules are required to fully hydrate the first hydration shell of 287

deprotonated glycine.42

The solvation process may make it easier for the interactions 288

between NTA and water. Therefore based on foregoing facts, we speculate that water 289

photoionization sensitized by NTA is another major source of eaq– in UV/NTA 290

process. 291

In addition, an important reason for the low quantum yield of eaq– from pure 292

water photolysis is the rapid recombination of eaq– with ·OH, ·H and H3O

+ (Eq. 3-5), 293

especially the reaction with ·OH, which accounts for more than 82% ± 3% of the 294

recombination of eaq–.43

The geminate recombination between eaq– and ·OH greatly 295

decreases the survival probability of eaq–. However in UV/NTA process, NTA is 296

highly reactive towards ·OH,30, 44

with a high reaction rate constant of 4.2×109 M

-1·s

-1 297

at pH of 10.0.22, 45

The strong scavenging capacity of NTA towards ·OH can 298

effectively protect eaq– from being quenched by ·OH, which increases the steady-state 299

concentration of eaq– and in turn facilitates the decomposition of PFOS. The eaq

–300

protection mechanism can be demonstrated by the long lifetime of eaq– in UV/NTA 301

process (Fig. 2). For instance, the half-life of eaq– in UV/NTA process is about 11.6 µs, 302

whereas it is only 1 ns in UV/sulfite process38

and lower than 2 ns in UV/KI process40

. 303

Therefore, the presence of NTA can dramatically prolong the survival time of eaq– by 304

Page 17 of 41



18

scavenging ·OH, thus promoting the degradation of PFOS. This explains why 305

UV/NTA process has a low transient quantum yield of eaq–, whereas has a high 306

efficiency in PFOS degradation. 307

Overall, the generation of eaq– during the UV/NTA process appears to involve 308

three steps. In aqueous solution, NTA is fully hydrated with both primary and 309

secondary hydration spheres. The fully hydrated NTA induces water photodissociation 310

and photoionization as a photosensitizer with the generation of eaq– and ·OH. Finally, 311

the NTA core scavenges ·OH to decrease the geminate recombination between ·OH 312

and eaq–, thus increasing the amount of eaq

– available for PFOS degradation. 313

314

Figure 3. (A) DMPO spin-trapping ESR spectra recorded in the UV/NTA and UV/buffer 315

(NH3·H2O-NH4Cl) processes. (B) Simulated data of various radicals trapped by DMPO in 316

UV/NTA process after 60 s irradiation: (c) total ESR signal; (e) simulation of DMPO-·OH; (f) 317

simulation of DMPO-·CH2COOH; (g) simulation of NO·. 318

In order to verify our speculation, DMPO was used as an ESR spin-trap to 319

identify possible short-lived radicals produced in the UV/NTA process. As shown in 320

Fig. 3, after 30 s irradiation, several peaks were clearly observed in the ESR spectra, 321

Page 18 of 41



19

indicating formation of radicals and their involvement in the reaction. As the 322

irradiation time increased to 60 s, the ESR signal intensity also increased. No 323

well-defined ESR signal was observed in the UV/buffer process, indicating that the 324

radicals were not produced in the absence of NTA. In order to further confirm the 325

radical species, the ESR data of UV/NTA-60s reaction were simulated. The ESR 326

signal of 1:2:2:1 quartets is thus assigned to DMPO-·OH (curve e). The spectrum of 327

curve f is speculated to be the signal of DMPO-·CH2COOH. The three-line spectrum 328

(curve g) indicates the detection of free nitroxide (NO·).46, 47

The occurrence of ·OH 329

suggests that water indeed undergoes photodissociation in the UV/NTA process. NO· 330

and ·CH2COOH are presumably the reaction products of NTA with ·OH, which is 331

consistent with the reactive sites in NTA molecule (discussed in detail in section 3.3) 332

and also consistent with the known steady-state products of NTA degradation (Fig. S7, 333

SI). The ESR data of UV/NTA-30s has a similar fitting result with UV/NTA-60s 334

(shown in Fig. S11, SI). Therefore, the ESR measurement detected the occurrence of 335

·OH, ·CH2COOH and NO·, further confirming the mechanism that NTA induces water 336

photodissociation and photoionization, and scavenges ·OH to increase the quantum 337

yield of eaq–.338

Page 19 of 41



20

3.3. Model compound study 339

340

Page 20 of 41



21

Figure 4. The time profiles of PFOS degradation and defluorination in the presence of model 341

compounds (NTA, IDA, EDTA, EDDA, citrate, glycine, oxamate, EDDS and MGDA): PFOS 342

(0.01 mM), model compounds (2 mM), UV irradiation, N2 saturated, pH (10.0). Error bars 343

represent standard deviations of triplicate assays. 344

In order to further validate the proposed mechanism and to test the effects of 345

other chelating agents, IDA, EDTA, EDDA, citric acid, glycine, oxamic acid, oxalic 346

acid, EDDS and MGDA were chosen as model compounds to investigate the 347

structure-activity relationship of APCAs in terms of impacting the photodegradation 348

of PFOS. The structural formulas of the model compounds are shown in Table S2, SI. 349

The results of the model compound study are shown in Fig. 4. Except oxamate and 350

oxalate, all of the tested model compounds accelerated the PFOS degradation and 351

defluorination compared with the control experiment, although their enhancement 352

effects varied a lot. For example, NTA was clearly better than IDA in accelerating 353

PFOS degradation. The 10-h PFOS degradation and defluorination ratios in presence 354

of NTA were 85.4% and 46.8%, respectively, while in the case of IDA they were only 355

54.8% and 25.6%, respectively. NTA and IDA are APCAs with only one nitrogen 356

atom. However, the nitrogen atom in NTA molecule is attached one more acetate 357

group (-CH2COOH) than IDA. Similarly, in the comparison of EDDA to EDTA, the 358

PFOS degradation and defluorination ratios were found to be higher in the presence of 359

EDTA (10-h values of 78.5% and 43.7%, respectively for EDTA and 51.1% and 360

24.2%, respectively for EDDA). This is because each nitrogen atom in EDTA 361

molecule is connected to one more acetate group than EDDA. Therefore, the acetate 362

Page 21 of 41



22

group density appears to be a factor that determines the efficiency of the specific 363

APCA in promoting PFOS photodegradation. 364

Citric acid is an important organic tricarboxylic acid. According to the results 365

shown in Fig. 4, citrate also promotes PFOS degradation and defluorination as 366

compared with the control experiment in its absence. However, the enhancement 367

effect of citrate is lower than NTA. The 10-h PFOS degradation ratios in the presence 368

of NTA and citrate were 85.4% and 50.7%, respectively, with the corresponding 10-h 369

defluorination ratios of 46.8% and 24.1%, respectively. Citric acid has a similar 370

chemical structure with NTA, both containing acetate groups, whereas citric acid has 371

no amine group. Therefore, the greater acceleration in the presence of NTA than 372

citrate can be probably attributed to the electron-rich center of amine group, whose 373

lone pair on the nitrogen atom favors the electrophilic attack of ·OH. 374

Glycine and oxamic acid both contain an amine group and a carboxyl group, but 375

their relative impacts on PFOS degradation are totally different. Glycine clearly 376

promoted PFOS degradation compared with the control experiment, whereas oxamate 377

has a significant inhibitory effect. The 10-h PFOS degradation and defluorination 378

ratios in the presence of glycine were 61.5% and 29.4%, respectively, whereas they 379

were only 1.8% and 0.4%, respectively in the presence of oxamate. These results are 380

mainly due to the different moieties present in glycine and oxamic acid that connect 381

amine group and carboxyl group, i.e., methylene group (CH2) for glycine and 382

carbonyl group (C=O) for oxamic acid. Therefore, it is presumably the methylene 383

group in acetate group that offers a reactive site for ·OH attack. The significant 384

Page 22 of 41



23

inhibition of oxamate is resulted from the electron-withdrawing property of 385

C=O-COOH group, which inductively withdraws electron density from the 386

neighboring N atom, thus reducing its reactivity with ·OH. Oxamic acid is reported to 387

be highly persistent during ·OH oxidation process.48, 49

In contrast, oxamic acid has a 388

high reactivity towards eaq–. The rate constant for the reaction of oxamate with eaq

– is 389

reported to be 5.7×109 M

-1·s

-1 (pH = 9.2). The second-order rate constant for the 390

reactions between oxamate and eaq– is more than 3 orders of magnitude higher than 391

the corresponding rate constant for eaq– with glycine (1.7×10

6 M

-1·s

-1, pH = 11.8). 392

Therefore, oxamic acid competes with PFOS for eaq–, and thus inhibits PFOS 393

degradation. Similar to oxamic acid, the presence of oxalate significantly inhibited 394

PFOS degradation and defluorination (Fig. S12, SI). This result coincides with the 395

low reaction rate constant for oxalate with ·OH (7.7×106 M

-1·s

-1, pH = 6.0)

10 and 396

further indicates that carboxyl group alone was unable to accelerate PFOS 397

degradation. In other words, the methylene group in acetate moieties is essential for 398

the effective attack by ·OH via H-atom abstraction. 399

In comparing EDTA to EDDS, we find that EDTA is more efficient with respect 400

to acceleration of PFOS degradation than EDDS. Although EDTA and EDDS both 401

have four acetate groups, two acetate groups in EDDS molecule are attached to the 402

α-C instead of the N atom. Therefore, EDTA is a tertiary amine while EDDS is a 403

secondary amine. The hydrogen bonding to the N atom in EDDS decreases the 404

availability of N-electrons and hinders the N-electrons transfer. Furthermore, the 405

presence of hydrogen on carbon next to the amine (α-hydrogen) was reported to be a 406

Page 23 of 41



24

key factor for electron-transfer interactions.50

Therefore, an extra α-hydrogen in 407

EDTA may result in a higher electron-transfer capability. Compared with NTA, the 408

enhancement is more favored for MGDA. This is because MGDA has an additional 409

CH3 group on the carbon α of amine, which increases the electron density of N due to 410

the electron-donating inductive effect, and adds a reactive site for ·OH attack. 411

Therefore, the electronic distribution of N atom, especially the availability of the lone 412

pair on N atom also makes a difference for the efficiencies of APCAs. 413

In summary, the acetate group and amine group in APCAs play important roles 414

in accelerating the photodegradation of PFOS. This conclusion is consistent with the 415

reactive sites for the reactions of APCAs with ·OH as pointed out in literatures.22, 29, 45,

416

50-53 Therefore, the results of our model compound study confirm that the scavenging 417

effect on ·OH by APCAs is the primary mechanism for the enhanced 418

photodegradation of PFOS. 419

The effects of NTA degradation products including IDA, glycine, oxamate and 420

oxalate were summarized in Fig. S12, SI. Their efficiencies for the degradation and 421

defluorination of PFOS followed the order of NTA > glycine ≈ IDA > control > 422

oxamate ≈ oxalate. Compared with the control experiment, NTA, glycine and IDA 423

obviously accelerated the photodegradation of PFOS, whereas oxamate and oxalate 424

significantly inhibited. Since NTA degradation products have either lower efficiencies 425

than NTA or inhibiting effect, the PFOS degradation and defluorination rates 426

decreased gradually as NTA degraded, especially after 1 h of reaction. 427

Page 24 of 41



25

3.4. Effects of pH and air 428

429

Figure 5. The effect of pH on the degradation (a) and defluorination (b) of PFOS: PFOS (0.01 430

mM), NTA (2 mM), UV irradiation, N2 saturated. Error bars represent standard deviations of 431

Page 25 of 41



26


As is shown in Fig. 5, the degradation and defluorination of PFOS by UV/NTA 433

process is strongly dependent on pH. The PFOS degradation ratios under the pH 434

conditions of 6.0, 7.0, 8.0, 9.0, 10.0 and 11.0 after 10 h were 17.4%, 29.9%, 47.7%, 435

56.6%, 85.4% and 99.5%, respectively. The 10-h defluorination ratios of PFOS under 436

the pH conditions of 6.0, 7.0, 8.0, 9.0, 10.0 and 11.0 were 8.5%, 8.0%, 15.1%, 34.9%, 437

46.8% and 72.3%, respectively. After 2 h irradiation, the degradation and 438

defluorination ratios of PFOS at the pH value of 11.0 had reached 82.2% and 48.8%, 439

respectively, whereas they were only 2.1% and 0.1%, respectively at the pH value of 440

6.0. The strong dependence of PFOS degradation on pH is mainly attributed to the 441

following two reasons. First, at low pH values, eaq– is quickly quenched by H

+ with a 442

high reaction rate constant of 2.3×1010

M−1

·s−1

(Eq. 5). 10

The excessive consumption 443

of eaq– by H

+ inhibits the degradation of PFOS and results in a low defluorination 444

efficiency under acidic condition. Second, pH affects the reactivity of NTA towards 445

·OH. The pKa values for successive deprotonation of NTA are 0.8, 1.9, 2.48 and 446

9.65.45, 54

The principal forms of NTA at pH 2.0 are HN+(CH2COOH)2(CH2COO

−) 447

and HN+(CH2COOH)(CH2COO

−)2; its major form at pH 6.0 is HN

+(CH2COO

−)3; 448

while at pH 10.0, NTA is near fully deprotonated in the form of :N+(CH2COO

−)3. As 449

concluded in the model compound study, the lone pair on the nitrogen atom is one of 450

the primary reactive sites for the electrophilic attack of ·OH on APCAs molecules. At 451

lower pH values, the addition of a single proton to :N(CH2COO−)3 or to 452

HN+(CH2COO

−)3 can decrease the rate constant for ·OH reaction by about one order 453

Page 26 of 41



27

of magnitude.45

Therefore, the deprotonated form of NTA is the more reactive species 454

towards ·OH than its protonated or partially protonated counterparts. Leitner et al. 455

drew a similar conclusion that the reactivity of ·OH with NTA is related to the amount 456

of deprotonated nitrogen.22

The rate constant for the reaction of ·OH with NTA was 457

reported to be 6.1×107

M−1

·s−1

, 5.5×108

M−1

·s−1

and 4.2×109

M−1

·s−1

, respectively, at 458

the pH values of 2.0, 6.0 and 10.0.45

Furthermore, since the pKa4 value of NTA is 459

9.65,45

the PFOS degradation rate increased significantly as pH increased from 9.0 to 460

10.0. Therefore, alkaline conditions are most favorable for the reaction of ·OH with 461

NTA, thus making more eaq– available for the degradation of PFOS. 462

Page 27 of 41



28

463

Figure 6. The effect of air on the degradation (a) and defluorination (b) of PFOS: PFOS (0.01 464

mM), NTA (2 mM), UV irradiation, pH (10.0). Error bars represent standard deviations of 465


Page 28 of 41



29

Besides pH, air is another important parameter that affects the photoreductive 467

degradation efficiency of PFOS. In general, the eaq–-mediated degradation of PFOS 468

requires strict anoxic conditions to prevent eaq– from being quenched by molecular 469

oxygen (Eq. 1).9, 11

For instance, Park et al. reported that PFOS was degraded rapidly 470

by UV/Iodide process in the presence of Ar with a rate constant of 6.5×10-3

min-1

, 471

whereas it was not degraded in the presence of air.5 Therefore in this study, PFOS 472

degradation was conducted under both N2 and air saturated conditions to evaluate the 473

effect of air. Under N2 saturation, the solution was pre-bubbled with N2 for 20 mins 474

and kept bubbling during the whole reaction period. Under air saturation, the solution 475

was not pre-bubbled and the reactor was kept open with the solution exposed to air 476

during the whole reaction period. The results are shown in Fig. 6. The degradation 477

ratios of PFOS under N2 and air saturated conditions after 10 h irradiation were 85.4% 478

and 75.2%, respectively. The 10-h defluorination ratios of PFOS under N2 and air 479

saturation were 46.8% and 35.8%, respectively. The PFOS degradation and 480

defluorination efficiencies under N2 saturation were just slightly higher than under air 481

condition. No significantly negative impact of air was observed. This is because the 482

excited NTA under UV irradiation (NTA*) can react with O2.30

Meanwhile, ·OH can 483

abstract the alpha hydrogen from the methylene group in NTA molecule.45

The 484

hydrogen abstracted radical intermediate can also react with O2.45

As a reductant, 485

NTA can scavenge a variety of oxidizing species produced in the presence of O2 such 486

as superoxide radicals (O2·−) and HO2 radicals (HO2·), which are also active 487

quenchers of eaq–.55

The sequestration of these oxidants by NTA protects eaq–, and thus 488

Page 29 of 41



30

enhances the degradation of PFOS. Therefore, the photodegradation of PFOS by 489

UV/NTA process is less sensitive to air than previously reported photoreductive 490

approaches like UV/KI11

and UV/sulfite56

processes. This result bodes well for future 491

applications of in situ or ex situ PFOS photo-remediation since strict anoxic 492

conditions are often difficult to achieve in practical engineering applications. 493

The effect of NTA concentration was also evaluated in this study. A higher NTA 494

concentration over the range of 0 to 4.0 mM resulted in higher PFOS degradation and 495

defluorination ratios (Fig. S13, SI). A detailed discussion is available in the SI. 496

4. Environmental Implications 497

This study presents a new strategy for efficient photoreductive degradation and 498

defluorination of PFOS. Compared with previous photochemical approaches, the 499

newly developed UV/NTA process has three remarkable advantages. First, PFOS 500

undergoes rapid photodegradation in the presence of NTA, with a high defluorination 501

rate. Second, no significant detrimental impact of air was observed, which means 502

UV/NTA process has a certain tolerance to oxygen. Third, the end products of NTA 503

degradation are CO2, NH4+ and NO3

−, which are easily biodegradable with minimized 504

secondary pollution risks. Therefore, UV/NTA may provide a novel, efficient and 505

relatively green technology for future in situ or ex situ PFOS remediation. 506

In previous studies, the most common way for the generation of eaq– to 507

decompose PFOS was by adding eaq– source chemicals such as iodide, sulfite and 508

indole derivatives. Meanwhile, attentions were more often paid on the improvement 509

of the transient generation of eaq–, such as increasing the reaction temperature or 510

Page 30 of 41



31

applying high photon flux. Actually, the apparent quantum yield of eaq– or its ultimate 511

efficiency is not just determined by its transient quantum yield. It is also greatly 512

influenced by the twinborn species such as ·OH and oxidative byproducts. For 513

instance, the unsatisfactory activity of UV/iodide process towards PFOS degradation 514

is likely a result of triiodide scavenging of eaq–.57

In this study, we proposed a new 515

way to generate eaq–, which utilizes NTA to sensitize water photoionization, and 516

subsequently to scavenge ·OH, in order to minimize the geminate recombination 517

between ·OH and eaq–. The net effect is to increase the steady-state level of eaq

– that is 518

available for PFOS degradation. The laser flash photolysis results indicate that 519

although UV/NTA process has a low transient yield of eaq–, it has a long half-life of 520

eaq–, which is considered to be the main reason for the high efficiency of UV/NTA 521

process. 522

APCAs is a class of compounds acting as chelating agents. Interests in previous 523

studies were mainly focused in their chelating effect with metal ions. Other APCAs’ 524

properties receive little attention. For example, they are excellent electron donors and 525

they have a high reactivity with ·OH. In this study, we revealed that the 526

photoreductive degradation of PFOS can be enhanced by various APCAs such as 527

NTA, IDA, EDTA, EDDA, EDDS and MGDA. The acetate and amine groups are the 528

primary reactive sites in APCAs molecules. Therefore, APCAs-mediated 529

photoreductive process may be a novel application field for APCAs, which is also 530

expected to be a promising replacement for the current photoredutive technologies for 531

PFOS decomposition. 532

Page 31 of 41



32

Associated content 533

Supporting Information 534

Schematic diagram of the photochemical reactor; Technical data of the low-pressure 535

mercury lamp; additional details on analytical methods; kinetics for PFOS 536

degradation; abbreviation and chemical structure of model compounds. Figures 537

showing time profiles of PFOS degradation products; mass balance of F; NTA and 538

NTA degradation products; NTA degradation pathway; effect of N2O; UV-Vis 539

absorption spectra; ESR spectra; effect of NTA degradation products and effect of 540

NTA concentration. Discussion on the effect of NTA concentration. This information 541

is available free of charge via the Internet at http://pubs.acs.org. 542

Acknowledgements 543

The authors greatly thank Dr. Jiahui Yang from Bruker (Beijing) Scientific 544

Technology Co., Ltd for her kind assistance on the ESR result analysis. The authors 545

also gratefully acknowledge Prof. Side Yao and Dr. Huijie Shi for their valuable 546

comments on the discussion. This study has been supported by the National Natural 547

Science Foundation of China (Project No. 21677109), and the State Key Laboratory 548

of Pollution Control and Resource Reuse Foundation (No. PCRRT16001). 549

550

References 551

1. Renner, R. Growing concern over perfluorinated chemicals. Environ. Sci. 552

Technol. 2001, 35 (7), 154A-160A. 553

2. Giesy, J. P.; Kannan, K. Peer reviewed: perfluorochemical surfactants in the 554

Page 32 of 41



33

environment. Environ. Sci. Technol. 2002, 36 (7), 146A-152A. 555

3. Lau, C.; Butenhoff, J. L.; Rogers, J. M. The developmental toxicity of 556

perfluoroalkyl acids and their derivatives. Toxicology and Applied Pharmacology 557

2004, 198 (2), 231-241. 558

4. Lau, C.; Anitole, K.; Hodes, C.; Lai, D.; Pfahles-Hutchens, A.; Seed, J. 559

Perfluoroalkyl acids: A review of monitoring and toxicological findings. 560

Toxicological Sciences 2007, 99 (2), 366-394. 561

5. Park, H.; Vecitis, C. D.; Cheng, J.; Choi, W.; Mader, B. T.; Hoffmann, M. R. 562

Reductive Defluorination of Aqueous Perfluorinated Alkyl Surfactants: Effects 563

of Ionic Headgroup and Chain Length. J. Phys. Chem. A 2009, 113 (4), 690-696. 564

6. Gu, Y. R.; Dong, W. Y.; Luo, C.; Liu, T. Z. Efficient Reductive Decomposition of 565

Perfluorooctanesulfonate in a High Photon Flux UV/Sulfite System. Environ. Sci. 566

Technol. 2016, 50 (19), 10554-10561. 567

7. Guo, R.; Zhang, C. J.; Zhang, G.; Zhou, Q. Degradation of Perfluorooctanoic 568

Acid by UV/Chloride Process. Chem. J. Chin. Univ.-Chin. 2016, 37 (8), 569

1499-1508. 570

8. Tian, H. T.; Gao, J.; Li, H.; Boyd, S. A.; Gu, C. Complete Defluorination of 571

Perfluorinated Compounds by Hydrated Electrons Generated from 572

3-Indole-acetic-acid in Organomodified Montmorillonite. Scientific Reports 573

2016, 6, 9. 574

9. Lyu, X. J.; Li, W. W.; Lam, P. K. S.; Yu, H. Q. Insights into perfluorooctane 575

sulfonate photodegradation in a catalyst-free aqueous solution. Scientific Reports 576

Page 33 of 41



34

2015, 5, 1-6. 577

10. Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. Critical review of 578

rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl 579

radicals (⋅OH/⋅O−) in aqueous solution. Journal of physical and chemical 580

reference data 1988, 17 (2), 513-886. 581

11. Qu, Y.; Zhang, C. J.; Li, F.; Chen, J.; Zhou, Q. Photo-reductive defluorination of 582

perfluorooctanoic acid in water. Water Research 2010, 44 (9), 2939-2947. 583

12. Grossweiner, L. I.; Swenson, G. W.; Zwicker, E. F. Photochemical Generation of 584

the Hydrated Electron. Science (New York, N.Y.) 1963, 141 (3583), 805-6. 585

13. Sauer, M. C.; Crowell, R. A.; Shkrob, I. A. Electron photodetachment from 586

aqueous anions. 1. Quantum yields for generation of hydrated electron by 193 587

and 248 nm laser photoexcitation of miscellaneous inorganic anions. J. Phys. 588

Chem. A 2004, 108 (25), 5490-5502. 589

14. Elles, C. G.; Jailaubekov, A. E.; Crowell, R. A.; Bradforth, S. E. 590

Excitation-energy dependence of the mechanism for two-photon ionization of 591

liquid H2O and D2O from 8.3 to 12.4 eV. J. Chem. Phys. 2006, 125 (4), 12. 592

15. Elles, C. G.; Shkrob, I. A.; Crowell, R. A.; Bradforth, S. E. Excited state 593

dynamics of liquid water: Insight from the dissociation reaction following 594

two-photon excitation. J. Chem. Phys. 2007, 126 (16), 8. 595

16. Nikogosyan, D. N.; Angelov, D. A. Formation of free-radicals in water under 596

high-power laser UV irradiation. Chem. Phys. Lett. 1981, 77 (1), 208-210. 597

17. Gobert, F.; Pommeret, S.; Vigneron, G.; Buguet, S.; Haidar, R.; Mialocq, J. C.; 598

Page 34 of 41



35

Lampre, I.; Mostafavi, M. Nanosecond kinetics of hydrated electrons upon water 599

photolysis by high intensity femtosecond UV pulses. Res. Chem. Intermed. 2001, 600

27 (7-8), 901-910. 601

18. Schmidt, C. K.; Brauch, H. J. Impact of aminopolycarboxylates on aquatic 602

organisms and eutrophication: Overview of available data. Environ. Toxicol. 603

2004, 19 (6), 620-637. 604

19. Repo, E.; Warchol, J. K.; Bhatnagar, A.; Mudhoo, A.; Sillanpaa, M. 605

Aminopolycarboxylic acid functionalized adsorbents for heavy metals removal 606

from water. Water Research 2013, 47 (14), 4812-4832. 607

20. Saifullah; Meers, E.; Qadir, M.; de Caritat, P.; Tack, F. M. G.; Du Laing, G.; Zia, 608

M. H. EDTA-assisted Pb phytoextraction. Chemosphere 2009, 74 (10), 609

1279-1291. 610

21. De Luca, A.; Dantas, R. F.; Esplugas, S. Study of Fe(III)-NTA chelates stability 611

for applicability in photo-Fenton at neutral pH. Appl. Catal. B-Environ. 2015, 612

179, 372-379. 613

22. Leitner, N. K. V.; Guilbault, I.; Legube, B. Reactivity of OH center dot and 614

e(aq)(-) from electron beam irradiation of aqueous solutions of EDTA and 615

aminopolycarboxylic acids. Radiat. Phys. Chem. 2003, 67 (1), 41-49. 616

23. Heintz, W. Ueber dem Ammoniaktypus angehörige organische Säuren. Justus 617

Liebigs Annalen der Chemie 1862, 122 (3), 257-294. 618

24. Bucheli-Witschel, M.; Egli, T. Environmental fate and microbial degradation of 619

aminopolycarboxylic acids. Fems Microbiol. Rev. 2001, 25 (1), 69-106. 620

Page 35 of 41



36

25. Ouyang, B.; Dong, W. B.; Hou, H. Q. A laser flash photolysis study of nitrous 621

acid in the aqueous phase. Chem. Phys. Lett. 2005, 402 (4-6), 306-311. 622

26. Yamamoto, T.; Noma, Y.; Sakai, S. I.; Shibata, Y. Photodegradation of 623

perfluorooctane sulfonate by UV irradiation in water and alkaline 2-propanol. 624

Environ. Sci. Technol. 2007, 41 (16), 5660-5665. 625

27. Jin, L.; Zhang, P. Y.; Shao, T.; Zhao, S. L. Ferric ion mediated 626

photodecomposition of aqueous perfluorooctane sulfonate (PFOS) under UV 627

irradiation and its mechanism. J. Hazard. Mater. 2014, 271, 9-15. 628

28. Yang, S. W.; Cheng, J. H.; Sun, J.; Hu, Y. Y.; Liang, X. Y. Defluorination of 629

Aqueous Perfluorooctanesulfonate by Activated Persulfate Oxidation. PLoS One 630

2013, 8 (10), 10. 631

29. Chen, D.; Martell, A. E.; McManus, D. Studies on the mechanism of chelate 632

degradation in iron-based, liquid redox H2S removal processes. Can. J. 633

Chem.-Rev. Can. Chim. 1995, 73 (2), 264-274. 634

30. Sorensen, M.; Frimmel, F. H. Photodegradation of EDTA and NTA in the 635

UV/H2O2 process. Z.Naturforsch.(B) 1995, 50 (12), 1845-1853. 636

31. Anderson, R. L.; Bishop, W. E.; Campbell, R. L. A review of the environmental 637

and mammalian toxicology of the nitrilotriacetic acid. Crc Critical Reviews in 638

Toxicology 1985, 15 (1), 1-102. 639

32. Laverock, M. J.; Stephenson, M.; Macdonald, C. R. Toxicity of iodine, iodide, 640

and iodate to daphnia-magna and rainbow-trout (oncorhynchus-mykiss). 641

Archives of Environmental Contamination and Toxicology 1995, 29 (3), 344-350. 642

Page 36 of 41



37

33. Burgi, H.; Schaffner, T.; Seiler, J. P. The toxicology of iodate: A review of the 643

literature. Thyroid 2001, 11 (5), 449-456. 644

34. Hori, H.; Hayakawa, E.; Einaga, H.; Kutsuna, S.; Koike, K.; Ibusuki, T.; 645

Kiatagawa, H.; Arakawa, R. Decomposition of environmentally persistent 646

perfluorooctanoic acid in water by photochemical approaches. Environ. Sci. 647

Technol. 2004, 38 (22), 6118-6124. 648

35. Zhang, Z.; Chen, J. J.; Lyu, X. J.; Yin, H.; Sheng, G. P. Complete mineralization 649

of perfluorooctanoic acid (PFOA) by gamma-irradiation in aqueous solution. 650

Scientific Reports 2014, 4, 1-6. 651

36. Abramczyk, H.; Kroh, J. Absorption spectrum of the solvated electron. 2. 652

Numerical calculations of the profiles of the electron in water and methanol at 653

300 K. The Journal of Physical Chemistry 1991, 95 (16), 6155-6159. 654

37. Hart, E. J.; Boag, J. Absorption spectrum of the hydrated electron in water and in 655

aqueous solutions. Journal of the American Chemical Society 1962, 84 (21), 656

4090-4095. 657

38. Lian, R.; Oulianov, D. A.; Crowell, R. A.; Shkrob, I. A.; Chen, X.; Bradforth, S. 658

E. Electron photodetachment from aqueous anions. 3. Dynamics of geminate 659

pairs derived from photoexcitation of mono-vs polyatomic anions. The Journal 660

of Physical Chemistry A 2006, 110 (29), 9071-9078. 661

39. Huang, L.; Dong, W. B.; Hou, H. Q. Investigation of the reactivity of hydrated 662

electron toward perfluorinated carboxylates by laser flash photolysis. Chem. 663

Phys. Lett. 2007, 436 (1-3), 124-128. 664

Page 37 of 41



38

40. Qu, Y.; Zhang, C. J.; Chen, P.; Zhou, Q.; Zhang, W. X. Effect of initial solution 665

pH on photo-induced reductive decomposition of perfluorooctanoic acid. 666

Chemosphere 2014, 107, 218-223. 667

41. Yuan, H. X.; Pan, H. X.; Wu, Y. L.; Zhao, J. F.; Dong, W. B. Laser Flash 668

Photolysis Mechanism of Pyrenetetrasulfonate in Aqueous Solution. Acta 669

Phys.-Chim. Sin. 2012, 28 (4), 957-962. 670

42. Yao, Y. H.; Chen, D.; Zhang, S.; Li, Y. L.; Tu, P. H.; Liu, B.; Dong, M. D. 671

Building the First Hydration Shell of Deprotonated Glycine by the MCMM and 672

ab Initio Methods. J. Phys. Chem. B 2011, 115 (19), 6213-6221. 673

43. Thomsen, C. L.; Madsen, D.; Keiding, S. R.; Thogersen, J.; Christiansen, O. 674

Two-photon dissociation and ionization of liquid water studied by femtosecond 675

transient absorption spectroscopy. J. Chem. Phys. 1999, 110 (7), 3453-3462. 676

44. Sillanpaa, M. E. T.; Kurniawan, T. A.; Lo, W. H. Degradation of chelating agents 677

in aqueous solution using advanced oxidation process (AOP). Chemosphere 2011, 678

83 (11), 1443-1460. 679

45. Sahul, K.; Sharma, B. K. Gamma-radiolysis of nitrilotriacetic aicd (NTA) in 680

aqueous solutions. J. Radioanal. Nucl. Chem.-Artic. 1987, 109 (2), 321-327. 681

46. Rehorek, D.; Janzen, E. G. On the formation of arseno aminoxyls (nitroxides) by 682

spin trapping of arseno radicals. Polyhedron 1984, 3 (5), 631-634. 683

47. Buettner, G. R. Spin trapping - electron-spin resonance parameters of spin 684

adducts. Free Radic. Biol. Med. 1987, 3 (4), 259-303. 685

48. Ganiyu, S. O.; Oturan, N.; Raffy, S.; Esposito, G.; van Hullebusch, E. D.; Cretin, 686

Page 38 of 41



39

M.; Oturan, M. A. Use of Sub-stoichiometric Titanium Oxide as a Ceramic 687

Electrode in Anodic Oxidation and Electro-Fenton Degradation of the 688

Beta-blocker Propranolol: Degradation Kinetics and Mineralization Pathway. 689

Electrochim. Acta 2017, 242, 344-354. 690

49. Antonin, V. S.; Garcia-Segura, S.; Santos, M. C.; Brillas, E. Degradation of 691

Evans Blue diazo dye by electrochemical processes based on Fenton's reaction 692

chemistry. J. Electroanal. Chem. 2015, 747, 1-11. 693

50. Chen, Y.; Hu, C.; Hu, X. X.; Qu, J. H. Indirect Photodegradation of Amine Drugs 694

in Aqueous Solution under Simulated Sunlight. Environ. Sci. Technol. 2009, 43 695

(8), 2760-2765. 696

51. Furlong, D. N.; Wells, D.; Sasse, W. H. F. Photooxidation at platinum colloidal 697

TiO2 aqueous-solution interfaces 1. Ethylenediaminetetraacetic acid and 698

related-compounds. Aust. J. Chem. 1986, 39 (5), 757-769. 699

52. Anipsitakis, G. P.; Dionysiou, D. D. Radical generation by the interaction of 700

transition metals with common oxidants. Environ. Sci. Technol. 2004, 38 (13), 701

3705-3712. 702

53. Morooka, S.; Ikemizu, K.; Kamano, H.; Kato, Y. Ozonation rate of water-soluble 703

chelates and related-compounds. J. Chem. Eng. Jpn. 1986, 19 (4), 294-299. 704

54. Martell, A. E.; Smith, R. M. Critical Stability Constants Vol. Amino Acids. 705

Plenum Press: New York, 1974. 706

55. Larson, R. A.; Stabler, P. P. Sensitized photooxidation of nitrilotriacetic and 707

iminodiacetic acids. Journal of Environmental Science & Health Part A 1978, 13 708

Page 39 of 41



40

(8), 545-552. 709

56. Song, Z.; Tang, H.; Wang, N.; Zhu, L. Reductive defluorination of 710

perfluorooctanoic acid by hydrated electrons in a sulfite-mediated UV 711

photochemical system. J. Hazard. Mater. 2013, 262 (22), 332-338. 712

57. Park, H.; Vecitis, C. D.; Cheng, J.; Dalleska, N. F.; Mader, B. T.; Hoffmann, M. 713

R. Reductive degradation of perfluoroalkyl compounds with aquated electrons 714

generated from iodide photolysis at 254 nm. Photochemical & Photobiological 715

Sciences 2011, 10 (12), 1945-1953. 716

Page 40 of 41



water photoionization

O

Geminate

recombination

hv

H2O

eaq−

NTA

scavenges

∙OH

CO2, NH4+, NO3

−

∙OH

NTA sensitized

H C N O F S

CC

CC

C

C

N

O

O

O

OO

HH

H

H

HH

O

Degradation & DefluorinationPFOS

Page 41 of 41



uv/nitrilotriacetic acid process as a novel strategy for efficient … · 2018. 2. 5. · 1 1...

Documents