Singlet Oxygen Phosphorescence as a Probe for Triplet-State

Subscriber access provided by UNIV OF DURHAM

Environmental Processes

Singlet Oxygen Phosphorescence as a Probe for Triplet-State Dissolved Organic Matter Reactivity Paul R. Erickson, Kyle J. Moor, Jeffrey J. Werner, Douglas E. Latch, William A. Arnold, and Kristopher McNeill Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b02379 • Publication Date (Web): 19 Jul 2018 Downloaded from http://pubs.acs.org on July 20, 2018

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 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 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.

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 31

Environmental Science & Technology

1

Singlet Oxygen Phosphorescence as a Probe for Triplet-State Dissolved

2

Organic Matter Reactivity

3 ◊,†

4

Paul R. Erickson,

5

Arnold,ϒ and Kristopher McNeill*

Kyle J. Moor,

◊,†

Jeffrey J. Werner,§ Douglas E. Latch,‡ William A.

,†

6 7 8

†

Institute of Biogeochemistry and Pollutant Dynamics (IBP), Department of Environmental

Systems Science, ETH Zurich, 8092 Zurich, Switzerland

9 10

§

Chemistry Department, SUNY-Cortland, Cortland, New York 13045, United States

11 12

‡

Department of Chemistry, Seattle University, Seattle, Washington 98122, United States

13 14

ϒ

15

Minneapolis, Minnesota 55455, United States

Department of Civil, Environmental, and Geo- Engineering, University of Minnesota,

16 17

◊

Authors contributed equally to this work

18 19

Abstract

20 21

Triplet-state chromophoric dissolved organic matter (3CDOM*) plays an important

22

role in aquatic photochemistry, yet much remains unknown about reactivity of these

23

intermediates. To better understand the kinetic behavior and reactivity of 3CDOM*, we have

ACS Paragon Plus Environment


24

developed an indirect observation method based on monitoring time-resolved singlet oxygen

25

(1O2) phosphorescence kinetics. The underpinning principle of our approach relies on the

26

fact that O2 quenches almost all triplets with near diffusion-limited rate constants, resulting in

27

the formation of 1O2, which is kinetically linked to the precursors. A kinetic model relating

28

1

29

substances and in whole natural water samples (hereafter referred to as 3CDOM*) was

30

developed and used to determine rate constants governing 3CDOM* natural lifetimes and

31

quenching by oxygen and 2,4,6-trimethylphenol (TMP), a common triplet probe molecule.

32

3

O2 phosphorescence kinetics to triplet excited states produced from isolated humic

CDOM* was found to exhibit smaller O2 and TMP quenching rate constants, ~9 ×108 and

33

~8 ×108 M-1 s-1 respectively, compared to model sensitizers, such as aromatic ketones.

34

Findings from this report shed light on the fundamental photochemical properties of CDOM

35

in organic matter isolates and whole waters and will help refine photochemical models to

36

more accurately predict pollutant fate in the environment.

37 38

Introduction

39 40

An important driver of indirect photochemistry in natural waters is chromophoric

41

dissolved organic matter (CDOM), which is the light-absorbing fraction of the complex

42

mixture of biologically-derived organic molecules present in all aquatic systems. Light

43

absorption by CDOM initiates the formation of several reactive intermediates, including

44

hydroxyl radical (•OH), hydrogen peroxide (H2O2), singlet oxygen (1O2), and others,

45

collectively known as photochemically-produced reactive intermediates (PPRIs).1 Many of

46

these PPRIs are generated from triplet-state CDOM (3CDOM*), which are CDOM molecules

47

in their electronically excited triplet state.

48


Page 2 of 31

Page 3 of 31


3

49

CDOM* is both a producer of PPRIs, and an important oxidant itself, playing a role

50

in the transformation of aquatic contaminants, biomolecules, and the cycling of carbon and

51

other elements.2 As opposed to other PPRIs that exist as single, distinct species, 3CDOM* is

52

an ill-defined mixture of a myriad of discrete molecules, whose individual chemical

53

properties combine to form the macroscale properties of 3CDOM*.3 Thus, the properties of

54

3

55

progress is being made on understanding the reactivity of 3CDOM*, there is much room for

56

improvement in the techniques utilized to study 3CDOM*. One common method used to

57

assess 3CDOM* reactivity is to follow the loss of 2,4,6-trimethylphenol (TMP), which is

58

known to be oxidized by 3CDOM*.4 This method is generally effective, but only serves as a

59

qualitative probe because the bimolecular reaction rate constant between 3CDOM* and TMP,

60

kTMP, is not known and has only been estimated and may vary as a function of dissolved

61

organic matter (DOM) structure and properties. If kTMP were known and consistent across

62

DOM samples, TMP could be used as a quantitative probe to determine 3CDOM* steady-

63

state concentrations ([3CDOM*]SS) and intersystem crossing quantum yields (Φ ).5,

64

Typically, triplet reactivity is assessed using transient absorption (TA) spectroscopy, wherein

65

the triplet excited state intermediates are directly observed, allowing the bimolecular reaction

66

rate constant to be determined. For many organic molecules this is possible, but the diverse

67

mixture of triplets within 3CDOM* leads to unresolvable TA signals, resulting from not only

68

triplets, but also photo-produced radicals and hydrated electrons.7

CDOM*, at best, are only generalized with most available measurement techniques. While

6

69 70

As 3CDOM* is the direct precursor to 1O2, it is possible to indirectly obtain kinetic

71

information about 3CDOM* through direct observation of 1O2, which is formed via energy

72

transfer from 3CDOM* to O2, but with varying yield.8 1O2 is an excellent indirect probe

73

because O2 exhibits near diffusion-limited quenching rate constants with most triplets (~1-3



74

×109 M-1 s-1 in water)9 and possesses a low singlet-triplet energy gap (94 kJ mol-1),8 thereby

75

allowing the capture of most triplets in 3CDOM* when using high O2 concentrations. 1O2 is

76

short lived in aqueous solution (lifetime = 3.6 µs) and weakly phosphorescent, emitting in the

77

near-infrared (NIR, 1268 nm).10 This signal is observable with modern NIR

78

photomultipliers,11 which allows for the time-resolved determination of 1O2 formed in

79

aqueous solution.

80

making it an effective tool to access the reactivity of 3CDOM*.

1

O2 is formed almost exclusively by this pathway in natural waters,

81 82

In this work, we describe a method which uses time-resolved 1O2 phosphorescence to

83

study the reactivity of 3CDOM*. Laser excitation of CDOM samples, which were either

84

aqueous solutions of humic substance isolates or whole water samples, yielded clear 1O2

85

phosphorescence growth and decay traces, from which 3CDOM* kinetic information was

86

extracted.

87

governing 3CDOM* quenching by oxygen and TMP as well as the natural triplet lifetimes for

88

a number of DOM isolates and natural water samples.

89

comparing traditional TA and our 1O2-based method to determine bimolecular rate constants

90

for the quenching of triplet-excited riboflavin (3riboflavin*) by O2, tryptophan, or TMP.

A kinetic model is proposed and used to determine average rate constants

The method was validated by

91 92

Materials and Methods

93 94

Chemicals, organic materials, and natural waters

95 96

Riboflavin (≥ 98%), tryptophan (≥ 98%), perinaphthenone (PN; 97%), 2,4,6-

97

trimethylphenol (TMP; 99%), sodium phosphate dibasic (≥ 99%), potassium phosphate

98

monobasic (≥ 99%), were purchased from Sigma-Aldrich. Methanol (HPLC grade) was


Page 4 of 31

Page 5 of 31


99

purchased from Merck. All reagents were used as received. Water (18 MΩ· cm) was obtained

100

from a Barnstead Nanopure Diamond system. The DOM isolates Suwannee River fulvic acid

101

#2S101F (SRFA), Suwannee River humic acid #2S101H (SRHA), Suwannee River natural

102

organic matter #2R101N (SRNOM), Pony Lake fulvic acid #1R109F (PLFA), and upper

103

Mississippi River natural organic matter #1R110N (MRNOM) were purchased from the

104

International Humic Substances Society (St. Paul, MN). Whole water samples were obtained

105

from the Great Dismal Swamp (Virginia, USA) along the Jericho ditch on September 8, 2016

106

by Vivian Lin and from Lake Bradford (Tallahassee, FL) on December 28, 2015 by Paul R.

107

Erickson. Whole waters were filtered (Whatman Polycap TC 75, pore size 0.2 µm)

108

immediately after collection, stored in acid-washed brown plastic bottles at approximately 3

109

ºC, and were allowed to warm to room temperature before use.

110 111

Time-resolved 1O2 phosphorescence

112 113

Time-resolved 1O2 phosphorescence experiments were performed using an in-house

114

constructed system based on a previously published design.12 Further details about the

115

experimental configuration are provided in the supporting information (SI).

116 117

DOM isolate samples were prepared in pH 7 phosphate buffer (10 mM) at

118

concentrations of 66 -112 mg/L DOM and used within two weeks of preparation. Natural

119

waters were analyzed undiluted. Sample absorbance at the excitation wavelength ( , 365

120

nm) ranged from 0.2-0.8 for both the prepared isolates and natural waters; UV-visible

121

absorption spectra are displayed in Figure S1. DOM samples were irradiated with 365 nm

122

pulsed laser light (further details in SI) for 7.2 ×105 pulses; riboflavin samples only required

123

1-2 ×104 pulses to obtain comparable 1O2 phosphorescence intensities. DOM samples were



124

not reused and only fresh, non-irradiated solutions were used in our analysis, despite no

125

significant change of the sample absorbance and 1O2 production after two collection cycles.

126

For all DOM samples, fluorescence interference was removed by subtracting a background

127

spectrum obtained from an argon-purged sample for each individual isolate or natural water.

128 129

Samples were purged with O2 or N2/O2 mixtures, as required for each experiment.

130

For oxygen variation experiments, [O2] was adjusted by mixing O2 and N2 with a gas-mixing

131

flow rotameter. Oxygen concentrations were measured with a commercial fiber optic O2-

132

sensing micro-optrode (PreSens Regensburg, Germany) prior to each 1O2 phosphorescence

133

measurement. TMP quenching experiments were performed with 100% O2 purging. TMP

134

was spiked into DOM solutions from a concentrated methanol stock with additional methanol

135

added to maintain 1% (v/v) methanol across all samples, resulting in concentrations of 100-

136

1000 µM TMP, a range necessary to observe quenching of 1O2 transients.

137 138

Transient absorption measurements

139 140

Pump-probe TA spectroscopy was employed to monitor 3riboflavin* quenching using

141

a previously described experimental design.13 TMP and tryptophan quenching experiments

142

were performed with 20% O2 purging. To determine riboflavin’s intrinsic triplet lifetime, a

143

flow-through cuvette configuration with a 200-mL reservoir under Ar-purging was used as

144

previously described.14 Transient decay lifetimes were determined from fits obtained from

145

Surface Explorer (Ultrafast Systems, Sarasota, FL, USA) and Origin 9.1 (OriginLab,

146

Northhampton, MA USA). In Stern-Volmer quenching experiments, 3riboflavin* decay rate

147

constants (kobs) were determined at 660, 679, and 703 nm and 623, 669, and 710 nm as a


Page 6 of 31

Page 7 of 31


148

function of added TMP or tryptophan, respectively. Calculated rate constants were averaged

149

for 3riboflavin* transients at these wavelengths.

150 151

Results and Discussion

152 153

Kinetic analysis

154 155

The processes considered in the kinetic analysis for CDOM and the nomenclature

156

used for each respective rate constant are listed in Scheme 1. A completely analogous kinetic

157

scheme is valid for model triplet sensitizers.

158

corresponding ki is the second-order rate constant. Upon absorption of a photon (hv), CDOM

159

is excited to singlet state CDOM (1CDOM*), with rate of light absorption Rabs. A portion of

160

1

161

substrate C or O2, with a fraction of O2 quenching events (f∆) forming 1O2, which typically

162

ranges from ~0.1-1 depending on the sensitizer.15 is the pseudo-first-order rate constant

163

for all deactivation processes acting on 3CDOM* other than quenching by substrate C or O2.

164

1

165

whose rate is highly solvent dependent and described by the pseudo-first-order rate

166

constant ∆ . The rate constant

167

deactivation (NIR photon emission), which is much smaller than ∆ and can be ignored in

168

the kinetic expression for 1O2.16 The resulting rate expression for 1O2 is shown in Equation 1:

For the bimolecular reactions listed, the

CDOM* forms 3CDOM* with efficiency Φ .

3

CDOM* can undergo quenching with

O2 similarly undergoes quenching with substrate C or non-radiative deactivation processes,

∆ is the pseudo-first order rate constant for radiative

169 170

[ ]

∆ = ∆ [ ][ ∗ ] − ! [][ " ] − ∆ [ " ]

171


1


The value [3CDOM*]t is the concentration of 3CDOM* at time t, expressed in eq. 2

172 173

Page 8 of 31

relative to the initial concentration, [3CDOM*]0.

174 175

,

,

[ ∗ ] = [ ∗ ]# $ %&'( [])'*+* [])'- .

2

176 177

Combining eq. 1 and 2 and integrating to solve for [1O2] as a function of time yields eq. 3:

178 179

[ " ] =

, [])' , /∆ '( [ ][ 201∗ ]3 %&'( [ ])'*+* - . ∆ [])' ∆ %' , []%' , %' [ ] 4$ '*+* ( *+* -

∆

∆

− $ %&'*+*[])'-. 5

3

180 181

1

O2 growth and decay kinetics are related to the observed 1O2 phosphorescence signal (St) by

182

eq. 4:

183 184

6 = 7 ∆ [ " ]

4

185 186

where 7 is an instrument response factor that accounts for optical efficiencies and detector

187

response. Combining eq. 3 and 4 yields the biexponential equation (eq. 5) that is used

188

throughout this work, which includes scaling parameter A0 (eq. 6).

189 190

[6] = ∆ '

83 '( [ ]

∆ , , *+* [])'- %'*+* []%'- %'( []

,

,

∆

∆

4$ %&'( [])'*+*[])'- . − $ %&'*+* [])'-. 5

5

191 192

9# = 7 ∆ ∆ [ ∗ ]#

6

193 194 195

Similar equations have been used to describe 1O2 phosphorescence kinetics.10,

11

Based on eq. 5, it is evident that the 1O2 phosphorescence signal growth rate constant


Page 9 of 31


196

[] ( [ ] + ! + ) inherently contains information about the triplet excited state,

197

because it is comprised of the rate constants for quenching triplets with O2 ( ) and

198

substrate C ( ! ) and the triplet’s natural non-radiative decay constant ( ). By monitoring

199

1

200

observing the triplet intermediates.17 Such an approach using time-resolved

201

phosphorescence has previously been used to extract reaction and relaxation rate constants of

202

sensitizer triplets.18 Additional kinetic information can be derived from the 1O2 signal decay

203

∆ [] rate constant ( ! + ∆ ), which has been used to determine 1O2 quenching rate constants

204

∆ ( ! ) with various molecules.12, 19, 20

O2 kinetics, it is possible to obtain triplet reactivity and kinetic behavior without directly 1

O2

205 206

Triplet lifetime and quenching with O2 validation

207 208

Figure 1a displays time-resolved 1O2 phosphorescence signals for riboflavin as a The 1O2

209

function of [O2] including raw data and fitted curves as discussed below.

210

phosphorescence signal rises with the faster decaying component, 1O2 or the triplet excited

211

state, and decays with the slower. At high O2 concentrations, the initial growth of the 1O2

212

signal relates to triplet deactivation processes and the decay with 1O2 deactivation processes.

213

Conversely, at low O2 levels, triplet decay is dominated by non-radiative relaxation ( ) with

214

typically lower rate constants than decay of 1O2 ( ∆ ), resulting in the initial 1O2 signal growth

215

being associated with 1O2 deactivation and the decay with triplet deactivation. From the

216

traces, it is evident that the 1O2 growth rate is directly proportional to [O2], with decreasing

217

[O2] causing the maximum signal intensity to shift to longer timescales, reflecting an increase

218

in observed triplet lifetime, the inverse of the relaxation rate. The data were analyzed in the

219

context of a simplified form of eq. 5, where the concentration of quencher C is set to zero,

220

resulting in an equation containing only the parameters [ ], , ∆ , and A0. A global



Page 10 of 31

221

fitting procedure was performed on six data sets obtained at various [O2] to simultaneously

222

solve for A0, , and with inputs of the constant ∆ = 2.76 × 105 s-1 and [O2] values as

223

measured for each kinetic trace;12 further details are provided in the SI. Residuals of the fits

224

are included in Figure 1b. Sensitivity analysis on the kinetic model was performed to assess

225

the effects of the fit variables A0, , and on the overall fit of the data set. In general, the

226

model was constrained and sensitive to all the fit variables. (Figure S2).

227 228

Kinetic analysis yielded a value for the quenching rate constant of 3riboflavin* by O2,

229

, of 1.2 ±0.1 ×109 M-1 s-1, which agrees with past reported values of 1.0 ±0.2 and 1.1 ±0.3

230

×109 M-1 s-1.18,

231

determined rate constants are related to the sensitivity (standard error) of the model fit to the

232

data. For the values determined with 1O2 phosphorescence the error is higher, likely around

233

±10%, based on sensitivity analysis. The natural triplet lifetime (

Singlet Oxygen Phosphorescence as a Probe for Triplet-State

Recommend Documents