Triplet-State Dissolved Organic Matter Quantum Yields and Lifetimes

Oct 16, 2017 - *Phone: +41-(0)44-6324755; fax: +41-(0)44-6321438; e-mail: [email protected]. ACS AuthorChoice - This is an open access article ...
1 downloads 12 Views 937KB Size
Subscriber access provided by Eastern Michigan University | Bruce T. Halle Library

Article

Triplet state dissolved organic matter quantum yields and lifetimes from direct observation of aromatic amine oxidation Markus Schmitt, Paul R. Erickson, and Kristopher McNeill Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03402 • Publication Date (Web): 16 Oct 2017 Downloaded from http://pubs.acs.org on October 16, 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 28

Environmental Science & Technology

1

Triplet state dissolved organic matter quantum yields and lifetimes from direct

2

observation of aromatic amine oxidation

3

Markus Schmitt, Paul R. Erickson, and Kristopher McNeill *

4 5

Institute of Biogeochemistry and Pollutant Dynamics (IBP), Department of Environmental Systems Science, ETH Zurich, 8092 Zurich, Switzerland

6 7

*Corresponding author

8

Kristopher McNeill

9

Tel. +41 (0)44 6324755; Fax. +41 (0)44 6321438

10

Email: [email protected]

11 12

TOC Art:

13 14

1 ACS Paragon Plus Environment

Environmental Science & Technology

15 16

Page 2 of 28

Abstract Excited triplet state chromophoric dissolved organic matter (3CDOM*) is a short-lived

17

mixture of excited state species that plays important roles in aquatic photochemical processes.

18

Unlike the study of triplet states of well-defined molecules, which are amenable to transient

19

absorbance spectroscopy, the study of 3CDOM* is hampered by it being a complex mixture and

20

its low average intersystem crossing quantum yield (ΦISC). This study is an alternative approach

21

to investigating 3CDOM* using transient absorption laser spectroscopy. The radical cation of

22

N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD), formed through oxidation by 3CDOM*,

23

was directly observable by transient absorption spectroscopy and was used to probe basic

24

photophysical properties of 3CDOM*. Quenching and control experiments verified that TMPD•+

25

was formed from 3CDOM* under anoxic conditions. Model triplet sensitizers with a wide range

26

of excited triplet state reduction potentials and CDOM oxidized TMPD at near diffusion-

27

controlled rates. This gives support to the idea that a large cross-section of 3CDOM* moieties

28

are able to oxidize TMPD and that the complex mixture of 3CDOM* can be simplified to a

29

single signal. Using the TMPD•+ transient, the natural triplet lifetime and ΦISC for different DOM

30

isolates and natural waters were quantified; values ranged from 12 to 26 µs and 4.1-7.8%,

31

respectively.

32

INTRODUCTION

33

Chromophoric dissolved organic matter (CDOM) is a complex mixture of organic molecules

34

originating from the breakdown of plant and microbially derived biomass coming from both

35

aquatic and terrestrial sources.1 In most natural aquatic systems, CDOM is the dominant light

36

absorber.2 Upon light absorption, CDOM is first promoted to its excited singlet state, and then a

37

subset of these species undergoes intersystem crossing (ISC) to the triplet state (Figure 1).3 In 2 ACS Paragon Plus Environment

Page 3 of 28

Environmental Science & Technology

38

the excited triplet state, a molecule is a better oxidant (and reductant) than in the ground state, as

39

the excited state reduction potential (E0* in V) for one-electron charge transfer is a sum of the

40

ground state potential (E0 in V) and the triplet energy ET of the molecule (eq 1; converting ET to

41

a potential using Faraday’s constant F).3

42

 ∗ =   +

43

3



(1)



CDOM* reacts with O2 to give singlet oxygen (1O2), another important photochemically

44

produced reactive intermediate (PPRI).4 3CDOM* and 1O2, along with other PRRI, play a central

45

role in aquatic systems for both natural element cycling processes and the transformation of

46

environmental contaminants.1-9

47

3

CDOM* is more difficult to investigate than most of the other PPRI, as it is not a well-

48

defined species. Rather it is a complex mixture of mostly unknown molecules, which vary in

49

their intersystem crossing quantum yield (ΦISC), triplet lifetimes and energies, and excited state

50

reduction potential.3, 10 Additionally, current estimates for ΦISC for 3CDOM* are quite low,

51

ranging from 0.4 to 11%.11-17

52

Most of our current knowledge about 3CDOM* comes from steady-state experiments with

53

different chemical probes. Phenols bearing electron-donating substituents, such as 2,4,6-

54

trimethylphenol (TMP), are frequently used to study oxidation reactions by 3CDOM*.18 One

55

limitation of TMP as a probe molecule is that it is more readily oxidized by triplets with a higher

56

reduction potential and may miss the less reactive pool of 3CDOM*.

57 58

1

O2 formation14, 16 and diene (1,3-pentadiene16 and sorbic acid12) isomerization have been

used as probe reactions for triplet energy transfer from 3CDOM*. The dienes are only able to 3 ACS Paragon Plus Environment

Environmental Science & Technology

Page 4 of 28

59

quench molecules with a triplet energy higher than about 250 kJ mol-1.3 O2 is most likely able to

60

quench all triplets in CDOM by energy transfer due to its low triplet-singlet energy gap, but not

61

all quenching events form 1O2.3 Therefore, various probes likely respond to different

62

subpopulations of the entire 3CDOM* mixture, and thus yield unique, but overlapping

63

information about different pools of triplets.19

64

The information that can be gained by such steady-state experiments is limited due to the fact

65

that the bimolecular reaction rate constant between the probe and 3CDOM* as well as the

66

steady-state concentration of 3CDOM* cannot be directly measured. So far, these values can

67

only be estimated with kinetic modeling, comparison with model sensitizers, and

68

comparison/competition between different probes.11-14, 16, 17, 20

69

Because triplets have lifetimes typically on the microsecond timescale, their direct

70

measurement requires special analytical techniques. Transient absorption laser spectroscopy is

71

commonly used to study short-lived intermediates like triplets and radicals, as direct detection of

72

these transients by changes in their UV-vis absorbance is possible. Therefore, time-resolved

73

measurements have the potential to provide more quantitative information about 3CDOM*. Past

74

studies have reported the observation of several transients after laser excitation of CDOM.21-24

75

However, these transients are mostly related to radical species, such as phenoxyl radical,24 and

76

hydrated electrons. Hydrated electrons are overproduced under laser irradiation compared to

77

natural sunlit conditions due to a biphotonic process.25 Some transients were partly quenched by

78

O2, a powerful triplet quencher, but were unaffected by less sensitive triplet quenchers.22, 23

79

Hence, these signals are most likely a composite of signals, some of which may be triplet state

80

molecules.22 It remains unclear how representative these signals are of the triplet CDOM pool,

81

and thus far, not much quantitative knowledge has been gained from these experiments. 4 ACS Paragon Plus Environment

Page 5 of 28

Environmental Science & Technology

82

A universal probe for triplets is needed to study CDOM reactivity and properties in more

83

detail. In this study, we report a new approach to investigate the properties of 3CDOM* with

84

transient absorption spectroscopy using the probe molecule N,N,N′,N′-tetramethyl-p-

85

phenylenediamine (TMPD). Due to its low oxidation potential, E0 (TMPD+•/TMPD) = 0.25

86

VSHE,26, 27 TMPD is oxidized to its radical cation (TMPD•+) at near diffusion-controlled rates by

87

triplet sensitizers27-31 with an efficiency close to unity.31 TMPD radical cation has a distinctive

88

absorption spectrum, featuring a distinct double hump absorption with maxima around 560 and

89

610 nm and a high absorption coefficient (12000 M-1 cm-1 at 612 nm).32 Additionally the radical

90

cation is very stable under most conditions.33 The combination of high reactivity, ease of

91

identification, and relative stability make TMPD an attractive probe for 3CDOM* in transient

92

absorption experiments. By following the formation of TMPD•+ in the presence of photoexcited

93

CDOM, we are able to gain knowledge about its precursor, 3CDOM* (Figure 1).

94 95 96 97 98 99 100 101

Figure 1. Schematic overview of TMPD•+ as a probe for 3CDOM*: 3CDOM* has no distinguishable transient absorption signal, which is indicated with the crossed-out arrow between 3CDOM* and its kinetic trace. Instead, observation of TMPD•+ kinetics indirectly gives the natural lifetime τ0, the bimolecular rate constant with TMPD (ket) and intersystem crossing quantum yield (ΦISC) for 3CDOM*. This is possible as kobs for TMPD•+ formation is equal to the unknown observed rate constant for triplet decay in the presence of TMPD.

The first aim of this study was to characterize TMPD as a probe for excited triplet state electron transfer with a set of model sensitizers covering a wide range of photophysical and

5 ACS Paragon Plus Environment

Environmental Science & Technology

102

chemical properties (Supporting Information, Table S1). This set reflected the possible range of

103

reactivity of different molecules present in CDOM. By varying the reduction potential of the

104

sensitizers, the aim was to assess how general the formation of TMPD•+ is for triplet sensitizers.

105

Using the model sensitizers, we sought to validate that the TMPD•+ signal can be used to

Page 6 of 28

106

determine a sensitizer’s natural triplet lifetime τ0 (the inverse of the rate constants for the O2-

107

independent deactivation pathways),3 its rate constant with TMPD, and its intersystem crossing

108

quantum yield (1S  1T), ΦISC. Once validated with the model sensitizers, these methods could

109

then be applied to determine these values for CDOM. Second, in DOM experiments, we

110

evaluated whether or not TMPD•+ formation was due primarily to electron transfer reaction with

111

3

112

Finally, the versatility of TMPD•+ as a triplet probe was tested by determining ΦISC and natural

113

lifetimes of 3CDOM* for a set of DOM isolates and natural waters of different origin.

114

Materials and Methods

115

CDOM*, and if it could reliably be used as a qualitative and quantitative probe for 3CDOM*.

Chemicals and Preparation of Solutions. The following reagents were used as received.

116

From Sigma-Aldrich: 2-acetonaphthone (2-AN ≥ 99%), perinaphthenone (PN, ≥ 97%), 3’-

117

methoxyacetophenone (3-MAP, ≥ 97%), riboflavin (RF, 98%), 5,7-dimethoxycoumarin (≥

118

98%), coumarin (≥ 99%), Zinc 5,10,15,20-tetra(4-pyridyl)-21H,23H-porphine

119

tetrakis(methochloride) (ZnTMPyP ≥ 80%), sodium 1-pyrenesulfonate (PSA ≥ 97%), 4-

120

methoxyphenol (≥ 99%), sodium nitrite (≥ 99%), and N,N,N',N'-tetramethyl-p-

121

phenylenediamine (TMPD ≥ 98%); from TCI: 1,4 Naphthoquinone (NQ, ≥98%). Zinc

122

tetrakis(sulfonatopheny1)porphyrin (ZnTPPS) was prepared from the free porphyrin (85%, TCI)

123

and ZnCl2 (98%, TCI). Details about the synthesis are provided in Section S1. 6 ACS Paragon Plus Environment

Page 7 of 28

124

Environmental Science & Technology

All solutions were prepared with nanopure water (resistivity >18 MΩ cm, Barnstead

125

Nanopure System) without any cosolvent. All experiments were carried out at pH 7 in 10 mM

126

phosphate buffer. Due to the rapid oxidation of TMPD under oxic conditions, TMPD stock

127

solutions of 5 mM were prepared in argon-purged anoxic water and stored inside a nitrogen

128

atmosphere glovebox.

129

Natural organic matter isolates were obtained from the International Humic Substances

130

Society (IHSS) (St. Paul, MN) and included Leonardite humic acid (LHA), Pony Lake fulvic

131

acid (PLFA) and Suwannee River natural organic matter (SRNOM). DOM solutions were

132

prepared by dissolving the organic matter in 10 mM phosphate buffer, increasing the pH to 10.0

133

by 1M NaOH addition and sonicating the solutions intermittently until the pH remained stable.

134

The DOM solutions were then adjusted to pH 7 with 1 M HCl.

135

Natural water samples from Lake Bradford (Tallahassee, Florida, USA) and the Great Dismal

136

Swamp (Virginia, USA) were filtered through sterile 0.2 µm filters, and stored at 4 °C protected

137

from light. These sample locations were chosen due to their high light absorption by DOM,

138

which was required by the transient absorption method developed in this work. Before the

139

experiments were conducted, the pH was adjusted to 7 by dilution into a phosphate buffer (final

140

concentration 10 mM).

141

Transient absorption spectroscopy experiments. Time-resolved measurements were

142

obtained via pump−probe transient absorption spectroscopy. The setup of the system has been

143

previously described in detail.34 Briefly, pump laser pulses were generated by a Solstice

144

amplified Ti:Sapphire ultrafast laser (Newport Spectra-Physics, Darmstadt, Germany). Here, 795

145

nm pulses (3.2 W, 80 fs pulse with 1 kHz repetition rate) were directed into a Topas optical 7 ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 28

146

parametric amplifier for wavelength conversion (Light Conversion, Vilnius, Lithuania). The

147

output was tuned to generate an excitation wavelength of 346 nm unless otherwise mentioned.

148

Pulses were then steered into an EOS transient absorption spectrometer (Ultrafast Systems,

149

Sarasota, Florida). The pulse energy was adjusted to 3.5 µJ except for the power dependence

150

measurements. At this pump power the system is in the linear range between pump power and

151

initial rate of TMPD•+ formation which is required to obtain reliable quantum yields (Figure S1).

152

The overlap between pump and probe pulses was optimized by observation of the TMPD•+

153

signal and was adjusted before each individual experiment.

154

The inherent time resolution of the instrument is in the nanosecond range; however, the ∆t

155

time window can be optimized for different time scales. For this study, time windows of up to

156

400 µs were used to fit the growth and the decay of the long-lived TMPD•+. The initial time

157

resolution was set to 10 ns to allow the fitting of the initial rate of TMPD•+ formation and the

158

decay of model triplet sensitizer.

159

An excitation wavelength of 346 nm was chosen to avoid absorption by and thus direct

160

photoionization of TMPD.35 At this wavelength, CDOM and most model sensitizers had

161

adequate absorption. 3-MAP, coumarin and 4-methoxyphenol were excited at 328 nm, 337 nm

162

and 300 nm, respectively, to ensure substantial sensitizer absorption. For compounds with high

163

ΦISC, absorbance (A) was between 0.1 and 0.3 to form a range of triplet concentrations. For

164

DOM and coumarins, with expected low ΦISC values, the sample concentration was adjusted to

165

A of 0.4 to have sufficient TMPD•+ formation signal but avoid saturation effects. Natural water

166

samples were diluted with phosphate buffer and also matched by absorbance. In the case of Lake

167

Bradford water, A of the undiluted sample was already lower (0.26) and therefore it was used

168

without any dilution. For quantum yield calculations absorption was corrected for light screening 8 ACS Paragon Plus Environment

Page 9 of 28

Environmental Science & Technology

169

and difference in A. The ground state absorption was measured with a Cary 100 Bio UV-Vis

170

spectrometer (Varian).

171

Data collection times of up to 30 min were required to improve signal to noise by averaging

172

multiple runs. The longest acquisition times were necessary when using DOM as the sensitizer

173

or other weak sensitizers. Under anoxic conditions, TMPD•+ rapidly converts back to TMPD

174

following electron transfer, but because conversion is not 100%, the TMPD concentration

175

decreases over time. To minimize this effect and sensitizer bleaching during long experiments, a

176

flow-through configuration was employed with a 100 to 200 mL reservoir. Solutions were

177

continuously cycled through a 10-mm quartz flow-through cuvette. To minimize dark reaction

178

between TMPD and DOM, the reservoir was cooled with an ice bath to a temperature of 10

179

(±1) °C. All solutions were purged with N2O (unless otherwise mentioned) before and during

180

measurements to quench hydrated electron as well to increase the triplet lifetime by the removal

181

of oxygen. Under these conditions, the TMPD•+ signal did not overlap with the hydrated electron

182

signal, and TMPD was a more competitive quencher for triplets than in the presence of O2.

183

TMPD stock solution was spiked into DOM or sensitizer solutions to reach TMPD

184

concentrations of 10 to 90 µM (in the case of NQ, up to 300 µM).

185

Data evaluation. The resulting data from the transient absorption experiments consist of a

186

three-dimensional array of wavelength, ∆A and ∆t. Data analysis was performed with the

187

Surface Xplorer software package (Ultrafast Systems, Sarasota, Florida) to correct for scattered

188

light and Origin 9.1 (OriginLab, Northampton, MA) to fit respective kinetic traces. To obtain

189

kinetics for TMPD•+ formation and triplet decay, the observed ∆A was corrected for the overlap

190

of this signal with the triplet signal of the sensitizer (Section S3).

9 ACS Paragon Plus Environment

Environmental Science & Technology

191 192

Page 10 of 28

Results and discussion Formation of TMPD•+ through electron transfer between TMPD and excited triplet

193

state sensitizers. In the presence of TMPD, a signal appeared in the transient absorption spectra

194

with maxima around 560 and 610 nm (Figure 2A, Figure S2-S9) after excitation of all model

195

sensitizers examined (Table S1) under anoxic conditions (N2O-purged). This was consistent with

196

the expected signal for TMPD•+.28-30, 32, 33, 35 With increasing TMPD concentrations, formation of

197

TMPD•+ occurred with a higher initial rate and produced a signal of greater amplitude Figure

198

2B). With the model sensitizers, the growth of TMPD•+ reached a maximum in the range of 10-

199

25 µs and then decreased.

200 201 202 203 204 205 206 207 208 209 210 211

Figure 2. A: 3D transient absorption spectra of 3-MAP under N2O-purged conditions (excitation wavelength 328 nm) in the presence of 60 µM TMPD. The formation of TMPD•+ (maxima 560, 610 nm) from 33-MAP (400-450 nm region) is visible. The colors indicate transient absorption (∆A) intensity. B: Kinetic trace of the TMPD•+ transient (averaged ∆A of 555-565 nm as well as 605-615 nm and corrected for overlapping triplet signal) in the presence of 10 (black), 30 (red) and 60 µM TMPD (blue) formed from 33-MAP. Solid line shows the nonlinear growth and decay fit (eq S6). Inset: First 2 µs with linear fits, which were used to determine the initial rates of TMPD•+ formation. C: Kinetic trace of the TMPD•+ transient (averaged ∆A of 555-565 nm as well as 605-615 nm and corrected for overlapping signals, excitation wavelength 346 nm) from SRNOM in the presence of 60 µM TMPD under N2O-purged (red) as well as air-purged conditions (blue), with NaNO2 as a sensitizer in the presence of 60 µM TMPD (magenta) and 60 µM TMPD in absence of any sensitizer only in the phosphate buffer (black). Solid lines are monoexponential or biexponential (blue) growth fits.

212

As expected from the reaction kinetic theory (see below), the TMPD•+ formation was found to

213

match the decay of the reacting triplet (Figure 2A, Figure S2-S9). It is known that aromatic

214

amines react by an electron transfer mechanism with triplet sensitizers in aqueous solution36 and 10 ACS Paragon Plus Environment

Page 11 of 28

Environmental Science & Technology

215

due to the high triplet energy of aromatic amines,37 energy transfer is unfavorable. Additionally,

216

an electron transfer mechanism (Figure 3) was supported by the observation of the radical anion

217

of some of the sensitizers, such as 1-pyrenesulfonate, with the same formation rate constant as

218

TMPD•+ (Figure S9). Other potential sources of TMPD•+ were investigated through control and

219

quenching experiments and are discussed in detail in the section on CDOM, where this issue is

220

critical due to overall weak signal and the presence of other PPRI.

221

Obtaining kinetic information about excited triplet state sensitizers using TMPD•+. The

222

relationship between TMPD•+ formation and the decay of the model triplet sensitizers was

223

investigated to calculate 3CDOM* properties directly from the TMPD•+ transient (Figures 1and

224

3). The observed rate constant (kobs) for the decay of a reactant (in our case the triplet state) is the

225

same as for the formation of the product (eq 2 with ket as the bimolecular rate constant of the

226

sensitizer with TMPD and kd =τ0-1 as the radiationless decay of the triplet). This holds true even

227

if there exist multiple competing decay processes (for example, kd, or chemical reactions) for the

228

excited intermediate. The relative importance of these processes is reflected in the

229

preexponential term which includes such a ratio (eqs 3,4).38 The efficiency of TMPD•+ formation,

230

η, was omitted from the pre-exponential term (eq 4) as it was found to be close to 1 (see below).

231 232 233 234

Figure 3: Scheme illustrating the kinetic pathways involved in formation of TMPD•+ and its decay by back electron transfer. Non-electron transfer quenching pathways are not shown as the efficiency (η) of TMPD•+ formation from reaction of 3sens* and TMPD was found to be close to 1.

11 ACS Paragon Plus Environment

Environmental Science & Technology

 

  =  +  ∗  =

235



236

TMPD•+ , = A.1 − e2345+467!"#$8 9

237

A = 34

238

Page 12 of 28

!"#$•&

(2) (3)

467 !"#$ : =sens>init 5 +467 !"#$8

(4)

kobs values for triplet decays were fit mostly with a monoexponential decay function (Section

239

S3). In all cases, quencher concentrations were in large enough excess to assume pseudo-first-

240

order behavior. kobs for TMPD•+ growth has to be corrected for the subsequent decay of TMPD•+.

241

In the case of DOM experiments, the back reaction was much less pronounced (see below). A

242

decay process in the time window of 400 µs may seem surprising as it is known that TMPD•+ is

243

essentially a stable radical species.33 This decay process with the rate constant kq is believed to

244

be due to the back reaction between the reduced sensitizer, sens•–, and TMPD•+ by electron

245

transfer or proton transfer from the methyl group and back hydrogen abstraction (Figure 3), as

246

known for the reaction of triplet sensitizer and amines.39 Considering this mechanism, the

247

following differential equation has to be solved:

248

A!"#$•&  A,

= B, : =sens>init TMPD − C TMPD•+ sens •2 

(5)

249

As the formation is a pseudo-first-order reaction and the decay is a second-order reaction, one

250

can use the solution of Chien (1948)40 for a uni-bimolecular reaction with two starting materials.

251

The equation is a Riccati equation with a non-trivial solution that includes the incomplete Γ-

252

function (see Section S4 for the exact solution). This equation was used to obtain kobs for

253

TMPD•+ growth, but to obtain [3sens]init the initial rate approach was used (see below).

254

Thereby, the decay of the triplet can be directly compared with TMPD•+ formation and the

255

Stern-Volmer plots from both signals were similar for the model sensitizer (Figure 4A). The

256

slope for the different model sensitizers with a  ∗ between 0.25 and 2.4 V were in the same

257

range and resulted in near diffusion-controlled ket values at 10 °C between 1.4 and 3.7 x 109 M-1

258

s-1 (Figure 4B and Table S1). There was no clear trend for ket and sensitizer reduction potential,

259

indicating that TMPD is a good choice to capture essentially all 3CDOM* moieties. In contrast

260

to the slope, the intercept of the Stern-Volmer plot was quite different depending on model 12 ACS Paragon Plus Environment

Page 13 of 28

Environmental Science & Technology

261

sensitizer, as it reflected the natural lifetime τ0 (kd-1) of the triplets (Figure 4B and Table S1).

262

This suggests that the TMPD•+ transient can be used to investigate the lifetime of triplets over a

263

wide range. The difference between the two methods used to obtain ket and kd (triplet decay and

264

TMPD•+ formation) was generally within the error of the fits (Table S1). The larger deviation

265

can be partially explained by errors due to corrections of overlapping signals (Section S3).

266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284

Figure 4. A: The observed rate constant (kobs) for triplet riboflavin (RF) decay as well as TMPD•+ formation versus TMPD concentration. B and C: kobs for TMPD•+ formation for different model sensitizers (B) and DOM solutions (C). The intercept is the inverse of the natural lifetime τ0 of the sensitizer D: Initial rate of TMPD•+ formation from 5,7-dimethoxycoumarin (black) and PLFA (red) to determine initially formed triplets (3sensinit). E: 3sensinit from model sensitizer (normalized to the absorbance coefficient α346 at excitation wavelength 346 nm) related to intersystem crossing quantum yield (ΦISC, literature vales Table S1). Numbers match Table S1 and are ordered from the highest to the lowest triplet energy. 1: 3-methoxyacetophenone; 2: coumarin; 3: 5,7-dimethoxycoumarin; 4: 2acetonaphthone; 5: 1,4-naphthoquinone (not shown because of the short triplet lifetime, 3sensinit could not determine with the initial rate approach); 6: riboflavin; 7: sodium 1-pyrenesulfonate; 8: perinaphthenone; 9: zinc 5,10,15,20tetrakis(N-methylpyridinium-4-yl)porphyrin; 10: zinc tetrakis(sulfonatophenyl)porphyrin. Blue points from porphyrin were not included in the linear fit (red line). Error bars shows the uncertainty of each fit. The point with ΦISC= 0 and without a number is from NaNO2 experiments to obtain background value without TMPD•+ formation due to triplet oxidation. NaNO2 acts in this case as a light screener as TMPD•+ formation from •OH radicals only took place at later time points and did not interfere notably with the initial rate during the first 2 µs (Section S7). F: Intensity dependency of hydrated electron (red) and TMPD•+(black) formation from SRNOM under Ar purged conditions shown as ∆A between 600-650 nm (immediately after laser excitation) respective 555-565, 605-615 nm (40-50 µs after laser excitation). Hydrated electron data were fitted with an empirical exponential function and TMPD•+ with a linear fit (excluding the last two points due to saturation effects).

13 ACS Paragon Plus Environment

Environmental Science & Technology

Page 14 of 28

285

Relationship between ΦISC and TMPD•+ formation. To calculate ΦISC for 3CDOM, a

286

connection between TMPD•+ formation and ΦISC for model sensitizers with known ΦISC must

287

first be made. Initial triplet concentration ([3sens]init) scales linearly with ΦISC, according to eq 6

288

289

: =sens>init = 2.303 HI JI ΦISC,λ

(6)

where Iλ is the intensity of the laser irradiation (mol photon cm-2 pulse-1), αλ is the decadic

290

absorbance coefficient (cm-1) at the excitation wavelength λ, and the value 2.303 (ln 10) is the

291

correction from the decadic to Naperian logarithm scale. This equation assumes that the

292

formation of 3sens is much faster than its decay, which is reasonable given that singlet lifetimes

293

are on the order of ps to ns whereas triplet lifetimes are on the order of µs. Analysis of the initial

294

rates of TMPD•+formation (Figure 2B, eq 7, details about the derivation can be found in Section

295

S5) was found to be the best approach to calculate [3sens]init from the TMPD•+ signal.

296

!"#$•&  

Rate

=  TMPD : =sens>init

(7)

297

Here ket is the electron transfer rate constant for the oxidation of TMPD by 3sens, and was

298

found above to be between 1.4 and 3.7 x 109 M-1 s-1 (10 °C) for the model sensitizers. As will be

299

discussed below, 2.3 x 109 M-1 s-1 (10 °C) is the value we use for CDOM. The initial rate

300

method of determining [3sens]init was more reliable than attempting to fit the full formation and

301

decay of TMPD•+, especially for DOM, because at later time points other PPRI may begin

302

producing TMPD•+ (see below). The slope of the regression of the initial rate over the TMPD

303

concentration (Figure 4D) divided by ket was used to calculate [3sens]init for each model sensitizer.

304

For NQ, it was not calculated since this approach did not work due to the very short 150 ns

305

lifetime of the triplet (Figure S6). 14 ACS Paragon Plus Environment

Page 15 of 28

Environmental Science & Technology

Values of [3sens]init/α346 determined from TMPD•+ initial rates of formation correlated linearly

306 307

with ΦISC (eq 6, R2 = 0.89, Figure 4E), with two porphyrin sensitizer outliers. Thereby,

308

differences in TMPD•+ formation can mostly be explained by ΦISC. Only for porphyrin

309

sensitizers was TMPD•+ formation clearly less efficient, possibly due to their very low triplet

310

energies (Section S5). The contribution of molecules with similar properties as porphyrins in

311

3

312

CDOM* can most likely be neglected. The efficiency for TMPD•+ formation, η(TMPD•+), from 3RF was 0.94, which is the fraction

313

of TMPD•+ formed per triplet quenching event (Section S5). Unfortunately, triplet-triplet

314

absorption coefficient in water were not available for most sensitizers and therefore η could not

315

be calculated. Nevertheless, it can be assumed that values of η were all quite high, as there were

316

no major differences between different sensitizers (expect the porphyrin-based sensitizers) in

317

their ability to form TMPD•+. In this study, we assume η = 1 for all sensitizers. This is in

318

agreement with previous studies that showed a high efficiency of radical cation formation for

319

different aromatic amines, especially with N-methyl substituents,41 and specifically TMPD.31

320

These results overall demonstrated that the initial rate of TMPD•+can be used to calculate ΦISC of

321

an unknown chromophore.

322

Formation of TMPD•+ from photoexcited DOM. Formation of the typical double hump

323

absorbance of TMPD•+ was observed under N2O-purged conditions for all DOM solutions

324

employed in the transient absorption spectra after adding 30-90 µM TMPD (shown for SRNOM

325

in Figure 2C). While other attempts have been made,42 this is to our knowledge the first study in

326

which the formation of a radical cation from a reaction with photoexcited DOM has been

327

directly observed. A number of control and quenching experiments were performed to access the

328

contribution of processes other than oxidation by 3CDOM* to the formation of TMPD•+. 15 ACS Paragon Plus Environment

Environmental Science & Technology

Page 16 of 28

329

TMPD•+ formation in the absence of a sensitizer or DOM was insignificant, indicating that direct

330

photoionization of TMPD was not important (Figure 2C). Due to light screening, this possibility

331

is even lower in the presence of DOM. The observed formation of TMPD•+ takes place on the µs

332

timescale (Figure 2C). Excited singlet state electron transfer would be on the ns time scale.

333

Thereby singlet state electron transfer is likely not the reason for TMPD•+ formation from

334

photoexcited DOM.

335

DOM photolysis is known to generate OH•,43 which can also oxidize TMPD to form

336

TMPD•+.32 Reported OH• quantum yield values (ΦOH•) for DOM are less than 0.01%,44 strongly

337

suggesting TMPD•+ formation from this source is unimportant. However, due to purging with

338

N2O, OH• can be formed from eaq- and N2O (eq 8).45

339

eaq- + N2O + H2O  N2 + OH• + OH-

(8)

340

To investigate the influence of N2O on the TMPD•+ formation in the presence of photoexcited

341

DOM, experiments were conducted under argon-purged conditions and no significant difference

342

between these two conditions was seen for both signal intensity and kinetics after corrections of

343

interference from hydrated electron signals (Section S6 and Figure S13). To further verify that

344

OH• was not important for the observed TMPD•+ formation, NaNO2 was used to generate OH• in

345

the same concentration range as 3CDOM* to study the reaction between TMPD and OH•. The

346

kinetics for TMPD•+ formation from reaction with OH• were completely different than from

347

DOM (Figure 2C, Figure S14) and slower, with a rate constant 3.3 x 108 M-1 s-1. This slow

348

formation is in agreement with previous studies, which proposed an OH-adduct intermediate32 or

349

a carbon-centered radical (R2NCH2•) intermediate46 in the reaction of OH• with TMPD.

16 ACS Paragon Plus Environment

Page 17 of 28

Environmental Science & Technology

350

Consequently, TMPD•+ formation from OH• did not affect our approach of using initial rates to

351

determine ΦISC for 3CDOM* (Section S7). Hydrated electrons (e-aq) were formed with a quantum yield of 2.9% (determined for SRNOM;

352 353

Section S8). Biphotonic process, due to high light intensity in laser experiments, can lead to

354

unreasonably high e-aq concentrations compared to natural sulight.25 The higher quantum yield

355

found here compared to previous determinations25, 47, 48 was ascribed to the very short pulse

356

width of our laser system (100 fs), which enhances non-linear effects. Since each e-aq that is

357

generated is accompanied by a “hole”, a potential photooxidant, these species must be

358

considered as potential (artefactual) precursors of TMPD•+. Power dependency experiments were

359

able to rule out this possibility. Specifically, it was found that the e-aq and TMPD•+ formation

360

had different dependencies on laser intensity (Figure 4F). TMPD•+ formation increased linearly

361

with laser intensity until saturation effects started to occur. This behavior is typical for

362

monophotonic processes.49 Whereas e-aq (and accompanying hole) production increased

363

nonlinear, which is typical for biphotonic processes.49 This deviation in behavior clearly

364

indicates that the formation of TMPD•+ is not dominated by the biphotonic process that forms e-

365

aq.

366

reacts with the hole radicals, could be that the reactivity of such radicals is lower than triplet

367

oxidants. If the hole formed in the biphotonic ionization is a relatively stable radical (e.g.,

368

phenoxy radical), kobs for TMPD•+ formation might be lower because of the longer intrinsic

369

lifetime and lower bimolecular rate constant than for triplet sensitizers. This was supported by

370

experiments in which 4-methoxyphenoxyl radical was generated by direct ionization of 4-

371

methoxyphenol and allowed to react with TMPD. The observed kobs value for phenoxy radical-

372

mediated oxidation of TMPD was significantly lower than that observed from triplet-mediated

Potential reasons for the absence of this relationship, even if it is quite likely that TMPD

17 ACS Paragon Plus Environment

Environmental Science & Technology

Page 18 of 28

373

oxidation of TMPD by 2-AN, a triplet sensitizer, and also lower than DOM-sensitized oxidation

374

of TMPD (Figure S14). This indicates that while phenoxy radicals can oxidize TMPD, they may

375

react slower than triplet sensitizers. Another reason for the lack of an apparent effect from such

376

holes is that such radicals may react with TMPD in a different manner, analogous to how OH•

377

reacts with TMPD initially through addition32 or H-atom abstraction.46

378

Additional support for the idea that radicals are slower oxidants of TMPD•+ than triplet

379

oxidants came from experiments with the model sensitizer, pyrene sulfonic acid (PSA). Both

380

triplet PSA (3PSA) and PSA radical cation (PSA•+) were observed upon excitation of PSA

381

(Figure S9). PSA•+ and 3PSA were quenched with rate constants of 8 x 108 and 2.0 x 109 M-1 s-1,

382

respectively. In the same experiment, TMPD•+ was formed with a rate constant of 2.5 x 109 M-1 s-

383

1

384

[3sens]init/α346 value based on TMPD•+ formation was not overestimated in the case of PSA

385

(Figure 4E) as one would expect if the radical contributes significantly to TMPD•+ formation.

, having a greater contribution from the triplet decay than the radical decay. Furthermore the

The formation processes of some PPRI, like 1O2, H2O2 and organic peroxyl radicals, require

386 387

O2,18 while the formation of other species, such as phenoxyl radicals, are O2-independent.24

388

3

389

is enhanced in the absence of O2, as O2 is an efficient 3CDOM* quencher.16 O2 quenching

390

experiments showed that TMPD•+ formation strongly decreased with increasing O2 concentration

391

(Figure 2C, Figure S16) and the quenching rate constant (quenching of 3CDOM* by O2) was

392

roughly estimated to be in the order of 109 M-1 s-1 for SRNOM (Section S9). This is in agreement

393

with previous determinations for the rate constants between 3CDOM* and O216 and thereby was a

394

strong indication that the observed formation of TMPD•+ under anoxic conditions was

395

dominated by 3CDOM*. One could argue that the decrease in TMPD•+ signal in the presence of

CDOM* is the most important photochemically produced pool of oxidants whose concentration

18 ACS Paragon Plus Environment

Page 19 of 28

Environmental Science & Technology

396

O2 is not due to quenching of 3CDOM*, but rather quenching of TMPD•+ or CDOM•+ by

397

superoxide (O2•–) formed from either e–aq or CDOM•– reacting with O2. We have considered this

398

possibility and believe it can be discounted on kinetic grounds. Even assuming diffusion-

399

controlled reaction between O2•– and TMPD•+/CDOM•+, we would expect a maximum of 3%

400

reduction in the signal intensity over 15 µs. Instead, we observe a 66% reduction in the signal

401

over this time frame, which is more consistent with O2 quenching of the precursor 3CDOM*.

402

Under air-purged conditions a second slower component became visible (Figure 2C), which

403

contributed around 50% of the overall TMPD•+ formation under these conditions (Section S9).

404

These transients could be successfully fit assuming a second TMPD•+-forming process and gave

405

a bimolecular rate constant for this second oxidant with TMPD of 5 x 108 M-1 s-1 and a lifetime

406

of 50 µs (Section S9). The above discussed radicals from DOM ionization are candidates for this

407

second component, as other photooxidants like 1O2 and OH• have different kinetic properties

408

(Figure 2C, Section S7, S9).

409

Time-resolved methods, in contrast to steady-state probe methods, make it possible to

410

separate oxidation reactions taking place on different time scales. Therefore, to reduce the

411

influence of the radicals only the initial rate (first 2 µs) was used to calculate ΦISC for 3CDOM*.

412

Kinetic modelling showed only a small influence of radicals (Section S10). Furthermore, only

413

the first 30 to 50 µs (depending on TMPD concentration) were used to fit kobs to reduce the

414

contribution of radicals to the TMPD•+ growth and thereby avoid influence on the determination

415

of triplet lifetime (Section S10).

416

Kinetics of TMPD•+ formation from 3CDOM*. Upon DOM excitation, TMPD•+ formed

417

faster and in higher concentration as the TMPD concentration was increased (Figure S10). The 19 ACS Paragon Plus Environment

Environmental Science & Technology

Page 20 of 28

418

decay of the TMPD•+ signal was not very pronounced, but increased slightly (depending on

419

DOM source) with increasing TMPD concentration (Figure S10). In general, TMPD•+ decay was

420

slower when oxidized by 3CDOM* than by model sensitizers (Figure S2-S9), for which we

421

believe recombination is the main decay pathway. The difference can be explained with

422

scavenging of the reduced CDOM, CDOM•–, by electron-accepting moieties such as quinones in

423

the DOM pool.50 This would reduce the likelihood of back reaction between CDOM•– and

424

TMPD•+. Alternatively, other PPRI, for instance the radicals from DOM ionization, could form

425

TMPD•+ at later time points, and these processes may mask the decay of TMPD•+ formed from

426

3

427

a function fitting the growth part (Section 4, eq 3) was used to analyze the DOM data.

428

CDOM*. The exact reason remains unclear and could be a combination of both. Therefore, only

Stern–Volmer plots showed a linear relationship between kobs on TMPD concentration

429

(Figure 4C) and ket was near diffusion-controlled ranging between 1.2 and 2.4 x109 M-1 s-1 for the

430

different DOM isolates and natural waters (Table 1, Figure 4C). This demonstrates that the probe

431

was properly chosen, as we sought a probe that reacts at diffusion-controlled rates with

432

3

CDOM* and yields a single signal with one rate constant for the whole complex mixture of

433

3

CDOM*.

434 435 436

Table 1. Rate constant with TMPD (ket), natural triplet lifetime (τ0) and intersystem crossing quantum yield (ΦISC) at 10 °C for waters and DOM isolates of different origin. Errors indicate standard deviation of the fit for at least 3 replicates. DOM source

ket (109 M-1 s-1)

τ0 (µs)

ΦISC (%)

PLFA

2.4 (±0.2)

26 (±8)

6.9 (±1.1)

SRNOM

2.3 (±0.2)

18 (±4)

4.1 (±0.5)

LHA

1.2 (±0.2)

13 (±2)

4.1(±0.6)

Lake Bradford

1.2 (±0.2)

12 (±1)

7.8 (±1)

Dismal Swamp

1.8 (±0.2)

12 (±2)

5.6 (±0.7)

20 ACS Paragon Plus Environment

Page 21 of 28

Environmental Science & Technology

437 438

We have higher confidence in the larger rate constants obtained with PLFA and SRNOM due

439

to the fact that the slightly lower rate constants for the other samples are most likely a result of

440

some influence of radicals (Section S10). For this reason, a value of ket of 2.3 x109 M-1 s-1 was

441

used for all DOM solutions to calculate ΦISC, as we believe this is most likely the value that best

442

represents 3CDOM* without interferences. It must be noted that while we believe the

443

differences are most likely due to interference of long-lived signals in some samples, we cannot

444

rule out that there are some intrinsic rate constant differences between the DOM samples.

445

The TMPD•+ transient signal was then used to obtain 3CDOM* lifetimes from Stern-Volmer

446

plots (Figure 4C), ranging from 12 to 26 µs for a series of DOM isolates and natural waters

447

(Table 1). We view these lifetimes as average value for the triplets weighted by their abundance,

448

which is in turn is proportional to each triplet’s ground state precursor abundance, precursor

449

absorptivity, and its ΦISC. Our experiments do not give any indication of the distribution of

450

lifetimes within a DOM sample. Two other estimates of 3CDOM* lifetimes have been

451

previously reported. One used formation of 1O2 from energy transfer to O2 as a probe reaction14

452

for 3CDOM* and the other used decay of TMP13 by 3CDOM*-mediated oxidation. Because of

453

the differences in the probe reaction types, these methods may represent different pools of

454

triplets. The TMP method yielded an estimate of 2 to 15 µs and the 1O2 method gave two

455

components with the first near 20 µs and the second up to 80 µs. Our results fall between these

456

two prior estimates. It must be noted that our experiments were performed at 10 °C and the

457

temperature dependency of 3CDOM* lifetimes is unknown.

21 ACS Paragon Plus Environment

Environmental Science & Technology

Page 22 of 28

458

Transients formed directly from photoexcited CDOM. Transient signals formed directly

459

from CDOM were observed with a maximum around 450 nm (Figure S18), in agreement with

460

previous reports, which assigned these signals to transients with radical or triplet character.21-24

461

To further investigate the contribution of triplets to these transient signals, the kinetics were

462

compared to TMPD•+ formation. TMPD was a poor quencher of the 450 nm transient. Either no

463

quenching or quenching to a small extent was observed (Figure S18). As TMPD is expected to

464

react with nearly all triplets by electron transfer, the lack of quenching is taken as evidence that

465

the 450 nm transient is not primarily due to 3CDOM*. The small quenching in some cases might

466

indicate some triplet contribution, but could also be attributed to radicals that also react with

467

TMPD.51 At the present time, it is unclear what connection the 450 nm transient has to 3CDOM*

468

and whether any quantitative or qualitative information about 3CDOM* can be gained from it.

469

ΦISC for 3CDOM*. The initial concentration of 3CDOM*, [3sens]init, was calculated in the

470

same way as for the model sensitizers with the initial rate of TMPD•+ formation (Figure 4D) and

471

ket for 3CDOM*. To calculate ΦISC for 3CDOM*, we exploited the relationship between [3sens]init

472

and ΦISC (eq 6). Rearranging equation 6 gives equation 9

473

R

ΦISC,λ = S.== T ∙ U

: Wsens>

init

XU

(9)

474

which shows the linear relationship between ΦISC and [3sens]init/α. Using the model sensitizers

475

with known ΦISC gave a reasonable linear relationship (Figure 4E), with a value for I346 of 3.4 x

476

10-9 mol photons cm-2 pulse-1 and an intercept of 1.7 x 10-10 mol cm-2 pulse-1, which is close to

477

the expected intercept of 0. This value of I346 agrees with the independent determination of the

478

intensity of the laser irradiation from laser power and laser beam diameter measurements. Using

22 ACS Paragon Plus Environment

Page 23 of 28

Environmental Science & Technology

479

the linear relationship from Figure 4E, ΦISC values were calculated for 3CDOM* and were found

480

to vary between 4.1 and 7.8% depending on DOM source (Table 1).

481

Previous ΦISC estimates range between 0.4 and 11% 11-17 based on 1O2, TMP, and sorbic acid

482

probe experiments, placing our values in the same range but on the higher end. These values

483

were not all directly comparable to our results because they used different wavelengths or DOM

484

of different origins. Most comparable were the results of Sharpless14 and Grebel and coworkers12

485

in terms of wavelength (peak at 370 respectively 350 nm) and DOM isolate studied (always

486

SRNOM) whereas the temperature was different. The 1O2 quantum yield was 1.4%14 and thus 34%

487

of our ΦISC, which is a reasonable number for the fraction of triplet quenching events by O2

488

which result in 1O2 formation. The quantum yield based on sorbic acid was 1.2%12 which is 29%

489

of our value, and is in good agreement with earlier studies showing around one third of all

490

triplets in CDOM have a triplet energy below that of a 1,3-pentadiene (250 kJ mol-1).16 This

491

further supports that nearly all triplets in CDOM were able to oxidize TMPD to form TMPD•+.

492

Implications. The formation of the TMPD•+ transient from photoexcited DOM was directly

493

observed and the oxidation due to 3CDOM* could be temporally separated from that by other

494

photooxidants. To this end, the whole complex mixture of 3CDOM* was simplified to a single

495

signal and triplet natural lifetime could be directly determined from Stern-Volmer plots. This is

496

an advantage over already established steady-state probes for 3CDOM* like TMP, sorbic acid

497

and 1O2, for which modeling or estimation (based on model sensitizers) of rate constants is

498

needed. Due to the low redox potential of TMPD, a large cross-section of 3CDOM* moieties are

499

represented by this method, as shown by near diffusion-controlled rates of TMPD•+ formation

500

with a high yield for model sensitizers with a wide range of reduction potentials. Moreover, from

23 ACS Paragon Plus Environment

Environmental Science & Technology

Page 24 of 28

501

initial rates of TMPD•+ formation, ΦISC for CDOM can be calculated by comparing it with model

502

sensitizers of known ΦISC. As a result, two important properties of 3CDOM*, natural lifetime and ΦISC were obtained for

503 504

DOM from different origins. These properties are important in determining 3CDOM* steady-

505

state concentrations, along with the rate of light absorption (Rabs), the O2 quenching rate

506

constant, and the concentration of O2 (eq.10).3 : =CDOM*>

= 4

507

YZ[\ ]^_` O Oa +45

(10)

2

Thus, this study contributes to improving models that predict the role of 3CDOM* for the

508 509

environmental fate of certain compounds.52 Furthermore TMPD•+, as a universal probe for

510

3

511

method may also be useful for determining quenching rate constants for 3CDOM* and

512

environmentally relevant molecules through competition kinetics. Such rate constants would be

513

an upper limit estimation for the reaction rate constant between 3CDOM* and the molecule of

514

interest, as physical and chemical quenching cannot be differentiated with this method.

515

Supporting Information Available

516

CDOM*, can potentially be used in the future to investigate different pools of triplets. This

This material is available free of charge on the ACS Publications website http://pubs.acs.org.

517

Supporting figures, transient absorption spectra, table with photophysical properties and reaction

518

rate constants for model sensitizers, description of data treatment, derivation of equations and

519

calculations, kinetic modeling and additional experiments.

520

Acknowledgements

24 ACS Paragon Plus Environment

Page 25 of 28

Environmental Science & Technology

521

We gratefully thank Prof. Douglas Latch (Seattle Univ.) for the helpful discussion of the

522

results, Prof. Kenneth Mopper (Old Dominion University) for providing a sample of Dismal

523

Swamp water and Dr. Vivian Lin for the helpful comments on the manuscript. This work was

524

financially supported by a grant from the Swiss National Science Foundation (Grant number

525

200021-156198).

526

References

527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558

1. McKay, G.; Rosario-Ortiz, F. L., Photochemical reactivity of organic matter and its size fractions. In Surface Water Photochemistry, Calza, P., Vione, D., Eds.; Royal Society of Chemistry, 2016; pp 77-95. 2. Zafiriou, O. C.; Joussot-Dubien, J.; Zepp, R. G.; Zika, R. G., Photochemistry of natural waters. Environ. Sci. Technol. 1984, 18, (12), 358A-371A. 3. McNeill, K.; Canonica, S., Triplet state dissolved organic matter in aquatic photochemistry: reaction mechanisms, substrate scope, and photophysical properties. Environ. Sci. Processes Impacts 2016, 18, (11), 1381-1399. 4. Latch, D. E. The Role of Singlet Oxygen in Surface Water Photochemistry, In Surface Water Photochemistry, Calza, P., Vione, D., Eds.; Royal Society of Chemistry, 2016; pp 77-96. 5. Chu, C.; Erickson, P. R.; Lundeen, R. A.; Stamatelatos, D.; Alaimo, P. J.; Latch, D. E.; McNeill, K., Photochemical and Nonphotochemical Transformations of Cysteine with Dissolved Organic Matter. Environ. Sci. Technol. 2016, 50, (12), 6363-6373. 6. Richard, C.; Canonica, S., Aquatic phototransformation of organic contaminants induced by coloured dissolved natural organic matter. In Environmental Photochemistry Part II, Springer: 2005; pp 299-323. 7. Janssen, E. M.; Marron, E.; McNeill, K., Aquatic photochemical kinetics of benzotriazole and structurally related compounds. Environ. Sci. Processes Impacts 2015, 17, (5), 939-946. 8. Ren, D.; Huang, B.; Bi, T.; Xiong, D.; Pan, X., Effects of pH and dissolved oxygen on the photodegradation of 17α-ethynylestradiol in dissolved humic acid solution. Environ. Sci. Processes Impacts 2016, 18, (1), 78-86. 9. Davis, C. A.; Erickson, P. R.; McNeill, K.; Janssen, E. M., Environmental photochemistry of fenamate NSAIDs and their radical intermediates. Environ. Sci. Processes Impacts 2017, 19, (5), 656-665 10. Arnold, W. A.; Oueis, Y.; O'Connor, M.; Rinaman, J. E.; Taggart, M. G.; McCarthy, R. E.; Foster, K. A.; Latch, D. E., QSARs for phenols and phenolates: oxidation potential as a predictor of reaction rate constants with photochemically produced oxidants. Environ. Sci. Processes Impacts 2017, 19, (3), 324-338. 11. Canonica, S.; Jans, U.; Stemmler, K.; Hoigne, J., Transformation kinetics of phenols in water: photosensitization by dissolved natural organic material and aromatic ketones. Environ. Sci. Technol. 1995, 29, (7), 1822-1831.

25 ACS Paragon Plus Environment

Environmental Science & Technology

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603

Page 26 of 28

12. Grebel, J. E.; Pignatello, J. J.; Mitch, W. A., Sorbic acid as a quantitative probe for the formation, scavenging and steady-state concentrations of the triplet-excited state of organic compounds. Water Research 2011, 45, (19), 6535-6544. 13. Golanoski, K. S.; Fang, S.; Del Vecchio, R.; Blough, N. V., Investigating the mechanism of phenol photooxidation by humic substances. Environ. Sci. Technol. 2012, 46, (7), 3912-3920. 14. Sharpless, C. M., Lifetimes of triplet dissolved natural organic matter (DOM) and the effect of NaBH4 reduction on singlet oxygen quantum yields: Implications for DOM photophysics. Environ. Sci. Technol. 2012, 46, (8), 4466-4473. 15. Peterson, B. M.; McNally, A. M.; Cory, R. M.; Thoemke, J. D.; Cotner, J. B.; McNeill, K., Spatial and temporal distribution of singlet oxygen in Lake Superior. Environ. Sci. Technol. 2012, 46, (13), 7222-7229. 16. Zepp, R. G.; Schlotzhauer, P. F.; Sink, R. M., Photosensitized transformations involving electronic energy transfer in natural waters: role of humic substances. Environ. Sci. Technol. 1985, 19, (1), 74-81. 17. Maizel, A. C.; Remucal, C. K., Molecular Composition and Photochemical Reactivity of Size-Fractionated Dissolved Organic Matter. Environ. Sci. Technol. 2017, 51, (4), 2113-2123. 18. Rosario-Ortiz, F. L.; Canonica, S., Probe Compounds to Assess the Photochemical Activity of Dissolved Organic Matter. Environ. Sci. Technol. 2016, 50, (23), 12532-12547. 19. Maizel, A. C.; Remucal, C. K., The effect of probe choice and solution conditions on the apparent photoreactivity of dissolved organic matter. Environ. Sci. Processes Impacts 2017, 19, 1040-1050. 20. Halladja, S.; Ter Halle, A.; Aguer, J.-P.; Boulkamh, A.; Richard, C., Inhibition of humic substances mediated photooxygenation of furfuryl alcohol by 2, 4, 6-trimethylphenol. Evidence for reactivity of the phenol with humic triplet excited states. Environ. Sci. Technol. 2007, 41, (17), 6066-6073. 21. Power, J. F.; Sharma, D. K.; Langford, C. H.; Bonneau, R.; Joussot-Dubien, J., Laser flash photolytic studies of a well-characterized soil humic substance. In Photochemistry of environmental aquatic systems. ACS symposium series, 1987. 22. Fischer, A. M.; Winterle, J. S.; Mill, T., Primary Photochemical Processing in Photolysis Mediated by Humic Substances. In Photochemistry of environmental aquatic systems. ACS symposium series, 1987. 23. Sul’timova, N.; Levin, P.; Chaikovskaya, O.; Sokolova, I., Laser photolysis study of the triplet states of fulvic acids in aqueous solutions. High Energy Chemistry 2008, 42, (6), 464-468. 24. Martin, M. V.; Mignone, R. A.; Rosso, J. A.; Gara, P. D.; Diez, R. P.; Borsarelli, C. D.; Martire, D., Transient spectroscopic characterization and theoretical modeling of fulvic acid radicals formed by UV-A radiation. Journal of Photochemistry and Photobiology A-Chemistry 2017, 332, (1), 571-579. 25. Wang, W.; Zafiriou, O. C.; Chan, I.-Y.; Zepp, R. G.; Blough, N. V., Production of hydrated electrons from photoionization of dissolved organic matter in natural waters. Environ. Sci. Technol. 2007, 41, (5), 1601-1607. 26. Nickel, U.; Jaenicke, W., 1‐und 2‐Elektronenschritte bei der Oxidation substituierter Paraphenylendiamine mit verschiedenen Oxidationsmitteln. II. Reaktionen mit Hexacyanoferrat. Berichte der Bunsengesellschaft für physikalische Chemie 1982, 86, (8), 695-701. 27. Bock, C.; Connor, J.; Gutierrez, A.; Meyer, T. J.; Whitten, D.; Sullivan, B.; Nagle, J., Estimation of excited-state redox potentials by electron-transfer quenching. Application of

26 ACS Paragon Plus Environment

Page 27 of 28

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649

Environmental Science & Technology

electron-transfer theory to excited-state redox processes. J. Am. Chem. Soc. 1979, 101, (17), 4815-4824. 28. Lu, C.; Han, Z.; Liu, G.; Cai, X.; Chen, Y.; Yao, S., Photophysical and photochemical processes of riboflavin (vitamin B2) by means of the transient absorption spectra in aqueous solution. Science in China Series B: Chemistry 2001, 44, (1), 39-48. 29. Porcal, G.; Bertolotti, S.; Previtali, C.; Encinas, M., Electron transfer quenching of singlet and triplet excited states of flavins and lumichrome by aromatic and aliphatic electron donors. Physical Chemistry Chemical Physics 2003, 5, (19), 4123-4128. 30. Arbogast, J. W.; Foote, C. S.; Kao, M., Electron transfer to triplet fullerene C60. J. Am. Chem. Soc. 1992, 114, (6), 2277-2279. 31. Shioyama, H.; Masuhara, H.; Mataga, N., Radical yield in electron transfer quenching of the excited tris (2, 2'-bipyridine) ruthenium (II) complex. Chemical Physics Letters 1982, 88, (2), 161-165. 32. Rao, P.; Hayon, E., Oxidation of aromatic amines and diamines by hydroxyl radicals. Formation and ionization constants of amine cation radicals in water. The Journal of Physical Chemistry 1975, 79, (11), 1063-1066. 33. Michaelis, L.; Schubert, M.; Granick, S., The free radicals of the type of Wurster's salts. J. Am. Chem. Soc. 1939, 61, (8), 1981-1992. 34. Wenk, J.; Eustis, S. N.; McNeill, K.; Canonica, S., Quenching of excited triplet states by dissolved natural organic matter. Environ. Sci. Technol. 2013, 47, (22), 12802-12810. 35. Richards, J.; Thomas, J., Effect of solvent on the flash photolysis and laser photolysis of N, N, N', N'-tetramethyl-paraphenylene diamine in solution. Transactions of the Faraday Society 1970, 66, 621-632. 36. Erickson, P. R.; Walpen, N.; Guerard, J. J.; Eustis, S. N.; Arey, J. S.; McNeill, K., Controlling factors in the rates of oxidation of anilines and phenols by triplet methylene blue in aqueous solution. The Journal of Physical Chemistry A 2015, 119, (13), 3233-3243. 37. Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T., Handbook of photochemistry. CRC press: 2006. 38. Klán, P.; Wirz, J., Photochemistry of organic compounds: From concepts to practice. John Wiley & Sons: 2009; p 99-101. 39. Turro, N. J.; Ramamurthy, V.; Scaiano, J. C., Modern molecular photochemistry of organic molecules. University Science Books: 2010; p 643-646. 40. Chien, J.-Y., Kinetic analysis of irreversible consecutive reactions. J. Am. Chem. Soc. 1948, 70, (6), 2256-2261. 41. Sivakumar, V.; Ponnamma, D.; Hussein, Y. H., Photoinduced electron transfer interaction of anthraquinones with aniline quenchers: Influence of methyl substitution in aniline donors. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 2017, 173, 931938. 42. Canonica, S.; Kohn, T.; Mac, M.; Real, F. J.; Wirz, J.; von Gunten, U., Photosensitizer method to determine rate constants for the reaction of carbonate radical with organic compounds. Environ. Sci. Technol. 2005, 39, (23), 9182-9188. 43. Page, S. E.; Arnold, W. A.; McNeill, K., Assessing the contribution of free hydroxyl radical in organic matter-sensitized photohydroxylation reactions. Environ. Sci. Technol. 2011, 45, (7), 2818-2825. 44. Dong, M. M.; Rosario-Ortiz, F. L., Photochemical formation of hydroxyl radical from effluent organic matter. Environ. Sci. Technol. 2012, 46, (7), 3788-3794. 27 ACS Paragon Plus Environment

Environmental Science & Technology

650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670

Page 28 of 28

45. Janata, E.; Schuler, R. H., Rate constant for scavenging eaq-in nitrous oxide-saturated solutions. The Journal of Physical Chemistry 1982, 86, (11), 2078-2084. 46. Steenken, S.; Vieira, A. J., Water‐Assisted Intramolecular Electron Transfer from the Ring to the Side Chain in N, N, N', N'‐Tetraalkyl‐para‐phenylenediamine Radicals—The Reverse of Side Chain Deprotonation of Radical Cations. Angewandte Chemie International Edition 2001, 40, (3), 571-573. 47. Thomas-Smith, T. E.; Blough, N. V., Photoproduction of hydrated electron from constituents of natural waters. Environ. Sci. Technol. 2001, 35, (13), 2721-2726. 48. Zepp, R. G.; Braun, A. M.; Hoigne, J.; Leenheer, J. A., Photoproduction of hydrated electrons from natural organic solutes in aquatic environments. Environ. Sci. Technol. 1987, 21, (5), 485-490. 49. Lachish, U.; Shafferman, A.; Stein, G., Intensity dependence in laser flash photolysis experiments: hydrated electron formation from ferrocyanide, tyrosine, and tryptophan. The J. Chem. Phys. 1976, 64, (10), 4205-4211. 50. Aeschbacher, M.; Sander, M.; Schwarzenbach, R. P., Novel electrochemical approach to assess the redox properties of humic substances. Environ. Sci. Technol. 2009, 44, (1), 87-93. 51. Neta, P.; Huie, R. E.; Ross, A. B., Rate constants for reactions of peroxyl radicals in fluid solutions. Journal of Physical and Chemical Reference Data 1990, 19, (2), 413-513. 52. Bodrato, M.; Vione, D., APEX (Aqueous Photochemistry of Environmentally occurring Xenobiotics): a free software tool to predict the kinetics of photochemical processes in surface waters. Environ. Sci. Processes Impacts 2014, 16, (4), 732-740.

671

28 ACS Paragon Plus Environment