S   1  

Environmental Photochemistry of Fenamate NSAIDs and their Radical 1  Intermediates 2  

3  

Caroline A. Davis1,2, Paul R. Erickson1, Kristopher McNeill1*, Elisabeth M.L. 4  

Janssen2* 5  

6  

1Institute of Biogeochemistry and Pollutant Dynamics, Department of Environmental 7  Systems Science, Swiss Federal Institute of Technology (ETH Zurich), 8006 Zurich, 8  Switzerland 9  

2 Department of Environmental Chemistry, Swiss Federal Institute of Aquatic Science 10  and Technology (Eawag), 8600 Dubendorf, Switzerland 11  

12   13   14  

Electronic Supplementary Information 15   16   17  

18  The electronic supplementary information contains 21 pages numbered S1-S21, 20 19  figures, 4 tables, and details on methods for determining, logDow values, 20  determination of the bimoleular reaction rate constant with 1O2, Fraction of D2O in 21  solutions for KSIE experiments, calculation of steady-state concentrations of singlet 22  oxygen and hydroxyl radicals, calculation of light screening factor, water column 23  model, and calculation of 3PN* quenching caffeic acid (Text S1-S7). 24   25  

Electronic Supplementary Material (ESI) for Environmental Science: Processes & Impacts.This journal is © The Royal Society of Chemistry 2017

S   2  

Table S1. LogDow for Fenamates - Predicted with ACD/Labs software 26  (www.chemspider.com). 27  

Fenamate Drug LogDow (ionized, pH 7.4) Diclofenac

Flufenamic acid Meclofenamic acid

Mefenamic acid Tolfenamic acid

1.69 2.29 2.95 2.04 2.47

28   29  Text S1. Methods for determining the bimolecular rate constant with singlet 30  oxygen. 31  

Rose Bengal and visible light (> 455 nm) 32  This is a steady-state experiment in open borosilicate test tubes containing 40 µM 33  FFA, 3 µM Rose Bengal as a 1O2 source and 5 µM of the test compound. Samples 34  were irradiated with a Xenon lamp using a 455 nm longpass filter to inhibit any direct 35  photolysis of the test compounds. Samples were taken at 0, 2.5, 5, 7.5, 10, 12.5, and 36  15 min for FFA quantification. Aliquots were taken at 0, 5, 10, 15, 20, 25, 30, and 37  40 min for the fenamate drugs. Dark controls of the samples kept in amber vials 38  during the irradiation were also included. UV-vis absorbance measurements of 39  samples at time 0 and 40 min were compared to determine the photobleaching of 40  Rose Bengal during the experiment. 41   42  1O2 phosphorescence (laser spectroscopy) 43  Singlet oxygen phosphorescence was recorded at (1270 ± 5 nm) using a near-IR 44  photo-multiplier tube detector. This configuration was built in our lab and modeled 45  after the singlet oxygen phosphorescence detector described by Bilski et al1 and 46  Jiménez-Banzo et al2. Quartz cuvettes were filled with sensitizer in a solvent 47  composition of 50:50, H2O:ACN. For these experiments 100 µM perinaphthenone 48  and 5 µM Rose Bengal were used as sensitizers. A pump beam of 360 nm and 49  550 nm were used to excite PN and RB, respectively and their corresponding power 50  was 80 and 70 mW. Increasing concentrations of fenamates were present in solution, 51  100 – 500 µM for flufenamic, mefenamic and tolfenamic acid, and 500 µM – 52  3000 µM for diclofenac and meclofenamic acid. The various fenamate concentrations 53  also contained different amount of solvent (acetonitrile), which can also affect the 54  lifetime of 1O2, so controls were done with the same solvent composition, which did 55  not contain fenamates to correct for this difference. The decay portion of the 1O2 56  signal was fitted and represented the lifetime of the 1O2. The decay rate (kobs, s-1) 57  corresponds to 1/lifetime. The decay rate for each amount of fenamates added were 58  plotted against each other in a Stern-Volmer plot. The data was collected using a 59  software called TimeHarp. Data analysis was performed using Origin 9.1 60   61  Non-photochemical generation of 1O2: H2O2 decomposition with molydate catalyst 62  Non-photochemical generation of 1O2 was done using a method adapted from one 63  used by Boreen et al 3. This experiment was carried out in amber sample vials that 64  were covered to prevent photolysis. 50 µL of 1 mM H2O2 was added to 4.95 mL of a 65  carbonate (pH 10.0, 10 mM) solution containing 10 µM fenamate, 40 µM FFA, and 66  1 mM MoO4

2-. 375 µL aliquots were taken at 0, 10, 30, 60, 120, 180, and 360 min 67  and added to 125 µL of sodium azide (507 mM) to quench the reaction. Samples 68  

S   3  

were analyzed with HPLC. Controls were done without MoO42- and without H2O2 to 69  

check whether fenamates were reactive with the catalyst or hydrogen peroxide. 70   71   72  

73  

Figure S1. Comparison of krxn(1O2,fen) determined via different experimental 74  methods. Black squares (¢), and green circles (l), represent 1O2 75  phosphorescence using perinaphthenone (PN) and Rose Bengal (RB) as a 1O2 76  source, respectively. Blue triangles (p) used chemical generation of 1O2 with 77  H2O2 and MoO4

2-. Pink triangles (q), show photochemical generation of 1O2 with 78  RB and light > 455 nm. Red diamonds (®) represent kinetic solvent isotope 79  effect (KSIE) experiments using PN as a 1O2 source. 80  

* = no kinetic solvent isotope effect observed for diclofenac, so no rate constant 81  could be determined. 82   83   84  

85  

86  

*

S   4  

Text S2. Composition of D2O in KSIE experiments 87  

The fraction of D2O in solution can be calculated using the following equation: 88  𝑘"#$% =  𝜒𝐻*𝑂 ∙ 𝑘-*. + 𝜒𝐷*𝑂 ∙ 𝑘1*. 89  

90  with ksolv being the observed reaction rate constant of the probe molecule furfuryl 91  alcohol (FFA) in the solution tested, kH2O being the observed reaction rate constant of 92  FFA in 100% H2O, and kD2O being the reaction rate constant of FFA in 100% D2O, as 93  well as the unknown mole fractions of in H2O and D2O of the tested solution. 94   95  𝜒𝐻*𝑂 =  

23456  7289:2;9:7  289:

; 𝜒𝐷*𝑂 = 1 − 𝜒𝐻*𝑂 96   97  kH2O = 2.54 x 105 s-1; kD2O = 1.79 x 104 s-1, (based on lifetimes of 1O2)4 98  The calculated fractions of H2O and D2O in the experiments for all test compounds 99  are listed in 100  Table S2. 101   102   103  

Table S2. Fraction of D2O in solutions for KSIE experiments. 104  

Fenamate Mole Fraction H2O Mole Fraction D2O Diclofenac Flufenamic acid Meclofenamic acid Mefenamic acid Tolfenamic acid

9.8% 11.3% 18.0% 13.8% 21.6%

90.2% 88.7% 82.0% 86.1% 78.4%

105   106   107  

Text S3. Calculation of steady-state concentration of singlet oxygen, [1O2]ss 108   109  

[ 𝑂? *]"" =  𝑘#A" 𝐹𝐹𝐴𝑘DEF,HHI

110  

kobs = observed degradation rate constant for FFA 111   112  

ln 𝑘DEF,HHI =  −1.59   ± 0.06 ×10S

273.16 + 𝑇[ 𝐶]∘ + (23.82   ±  0.21) 113  

114  T = temperature in degrees Celsius 115  see Appiani et al.5 116   117  

S   5  

Text S4. Calculation of steady-state concentration of hydroxyl radical, [•OH]ss 118   119  

[ 𝑂𝐻• ]""  =  d[ℎ𝑇𝑃𝐴])

𝑑𝑡𝑘DEF,bcI ∙ 𝑇𝑃𝐴 ∙ 𝑌 120  

121  

d[hTPA]/dt = rate of change for hTPA (in M s-1) 122  

[TPA] = initial concentration of TPA, 123  

krxn,TPA = 4.4 x 109 M-1 s-1, Y = 35% (production yield)6 124  

125  

126  Stern-Volmer Plots for Calculating Reaction Rate Constant with triplet 127  

perinaphthenone - krxn(fen,3PN) 128  

129  Figure S2. Stern-Volmer Plot for determination of bimolecular reaction rate 130  constant for reaction between triplet perinaphthenone and diclofenac. 131  

k sen

s

slope 4.24 ± 0.87 x 108 M-1s-1

intercept 1.59 ± 0.04 x 106 M-1s-1

Residual sum of squares 1.123

Adjusted R2 0.849

Diclofenac (M)

3 PN

*de

cay

rate

con

stan

t(s-1

)

S   6  

132  Figure S3. Stern-Volmer Plot for determination of bimolecular reaction rate 133  constant for reaction between triplet perinaphthenone and flufenamic acid. 134  

135  Figure S4. Stern-Volmer Plot for determination of bimolecular reaction rate 136  constant for reaction between triplet perinaphthenone and meclofenamic acid. 137  

138  

k rxn

slope 5.68 ± 0.50 x 108 M-1s-1

intercept 1.66 ± 0.07 x 106 M-1s-1

Residual sum of squares

0.062

Adjusted R2 0.969

Flufenamic acid (M)

3 PN

*de

cay

rate

con

stan

t(s-1

)

k rxn

slope 11.93 ± 1.84 x 108 M-1s-1

intercept 1.53 ± 0.06 x 106 M-1s-1

Residual sum of squares

0.642

Adjusted R2 0.910

Meclofenamic acid (M)

3 PN

*de

cay

rate

con

stan

t(s-1

)

S   7  

139  Figure S5. Stern-Volmer Plot for determination of bimolecular reaction rate 140  constant for reaction between triplet perinaphthenone and tolfenamic acid. 141  

 142  

143  Figure S6. Stern-Volmer Plot for determination of bimolecular reaction rate 144  constant for reaction between triplet perinaphthenone and caffeic acid. 145  

k rxn

slope 8.93 ± 1.26 x 108 M-1s-1

intercept 1.78 ± 0.05 x 106 M-1s-1

Residual sum of squares

1.413

Adjusted R2 0.924

Tolfenamic acid (M)

3 PN

*de

cay

rate

con

stan

t(s-1

)

Caffeic acid (M)

3 PN

*de

cay

rate

con

stan

t(s-1

)

k rxn

slope 3.53 ± 1.26 x 109 M-1s-1

intercept 0.50 ± 0.08 x 106 M-1s-1

Residual sum of squares

0.395

Adjusted R2 0.96

S   8  

Stern-Volmer Plots for Calculating Reaction Rate Constant with Antioxidant - 146  krxn(fen+.,ascorbic acid) 147  

148  

149  Figure S7. Stern-Volmer Plot for determination of bimolecular reaction rate 150  constant for reaction between flufenamic acid radical cation (Flu+•) and ascorbic 151  acid. 152  

 153  

154  Figure S8. Stern-Volmer Plot for determination of bimolecular reaction rate 155  constant for reaction between flufenamic acid radical cation (Mec+•) and 156  ascorbic acid. 157  

Ascorbic acid (M)

Flu+⦁

deca

yra

te c

onst

ant(

s-1)

k rxn

slope 6.93 ± 1.31 x 107 M-1s-1

intercept 7.54 ± 0.70 x 103 M-1s-1

Residual sum of squares

0.370

R2 0.900

Ascorbic acid (M)

Mec

+⦁de

cay

rate

con

stan

t(s-1

)

k rxn

slope 3.27 ± 0.61 x 107 M-1s-1

intercept 2.43 ± 0.33 x 103 M-1s-1

Residual sum of squares

0.213

R2 0.903

S   9  

158  

Figure S9. Stern-Volmer Plot for determination of bimolecular reaction rate 159  constant for reaction between flufenamic acid radical cation (Tol+•) and ascorbic 160  acid. 161  

 162      163  

Ascorbic acid (M)

Tol+⦁

deca

yra

te c

onst

ant(

s-1)

k rxn

slope 2.06 ± 0.12 x 108 M-1s-1

intercept 7.01 ± 0.40 x 103 M-1s-1

Residual sum of squares

0.076

R2 0.993

S   10  

 164  

165  Figure S10. Pseudo-first order degradation plots of the fenamate drugs in the 166  solar simulator controlled at pH 7.5 for mefenamic acid (1, blue), tolfenamic 167  acid (2, green), meclofenamic acid (3, red), flufenamic acid (4, purple), and 168  diclofenac (5, black). The table inset describes their half-lives under 169  experimental conditions. 170  

171  

-7.0

-6.0

-5.0

-4.0

-3.0

-2.0

-1.0

0.0

0 2 4 6 8

ln(C

t/C0)

exposure time (hours)

Half-life1. MEF 8.3 days2. TOL 5.1 h3. MEC 47 min4. FLU 37 min5. DIC 24 min

1

2

3

4

5

S   11  

172  Figure S11. Absorbance spectra for diclofenac and fenamates from 200 – 400 nm 173  (left axis). The photo fluence rate for the solar simulator output is plotted on the 174  right axis. Overlap of molar absorptivity peaks with solar simulator output peak 175  represents light that can be absorbed by the molecules. 176  

177  

178  

S   12  

Text S3. Calculation of Light Screening Correction Factor. 179  

Using Figure S11 above, decide the range of wavelengths where the light source and 180  the test compound overlap. This range will be used to determine the screening 181  correction factor based on method from Leifer et al.7. 182   183  First, calculate the “S-factor”. 184  

𝑆 =1 − 107fg

2.303 ∙ 𝑎𝑧 185  𝑎 = optical density at λ, dependent on the absorption of the sensitizer; DOM. 186  𝑧 = pathlength of light through test-tube 187   188  Next calculate the intensity of light absorbed by the system, 𝐼kl. 189  

𝐼kl =𝑆 ∙ 𝐼ml𝐼ml

190  

𝐼ml  = incident light intensity of the light source, measured using a radiometer 191   192  Then, the amount of light that is transmitted (T) is calculated, 193   194  

𝑇 =𝐼kl𝐼ml

195  

And finally from the transmission, it is possible to calculate the correction factor. 196   197  

𝐶𝐹 =1𝑇 198  

CF = correction factor 199   200   201  

Table S3. Light Screening Correction Factor for diclofenac and fenamates in 202  10 mgc L-1 PLFA to account for light attenuation from DOM. 203  

Fenamate Drug Light Screening Correction Factor Diclofenac

Flufenamic acid Meclofenamic acid

Mefenamic acid Tolfenamic acid

1.15 1.09 1.10 1.08 1.09

204   205  

206  

207  

208  

S   13  

Table S4. Summary of contribution to indirect photodegradation from various 209  photochemically produced reactive intermediates (PPRIs) under simulated 210  sunlight with 10 mgC L-1 PLFA. 211  

Compound

Effect of DOM Dkobs (%)

1O2 (%)

•OH (%)

3CDOM* (%)

Mefenamic acid +95 24.0 1 71

Tolfenamic acid +9 9.9 >1 0

Meclofenamic acid -14 4.0 >1 n.a.a

Flufenamic acid -2 1.4 >1 n.a.a

Diclofenac +19 0.0 >1 19 aNot analyzed because of overall net quenching effect of DOM 212   213   214   215   216  

S   14  

Text S4. Water Column Model (Figure 2b) 217  

To estimate the change of direct and indirect photochemical degradation in a water 218  column, the absorption spectrum of the organic matter solution and the test 219  compounds was recorded and the light intensity of solar irradiation were used. 220  

First, the wavelength dependent change of light intensity as a function of water 221  depth 𝐼l,g (mE cm-2 s-1), was calculated as 222   223  

𝐼l,g = 𝐼l,m ∙  107g∙f 224   225  

with 𝐼l,m  being incident light intensity at the water surface, a (cm) being the optical 226  density at wavelength, λ, dependent on the absorption of the sensitizer; DOM, and z 227  (cm) being the water pathlength of light representing the water depth. 228  

229  Second, the depth and wavelength dependent rate of light absorbance, ka,z 230  

(mE cm-3 s-1), by the sensitizer DOM and the test compound were estimated as 231   232  

𝑘f,g = 2.303   ∙ 𝑎   ∙ 𝐼l,g. 233   234  

At each depth, the rates were summed across the wavelength spectrum where the 235  DOM and test compounds absorb light within the solar spectrum, being 290-500 nm 236  for DOM and 290-400 nm for diclofenac, mefenamic acid, tolfenamic acid, 237  meclofenamic acid, and flufenamic acid. 238  

The change in ka,z as a function of depth for DOM and test compounds is directly 239  proportional to the relative decrease of indirect and direct photodegradation, 240  respectively. 241  

242  Figure S12. Light intensity of wavelengths 200-700 nm up to 1 m in the water 243  column. Purple = 1 cm, blue = 10 cm, green = 20 cm, yellow = 30 cm, orange = 50 244  cm and red = 1 m depth. 245  

246  

S   15  

247  Figure S13. Pseudo-first order degradation of (A) diclofenac, (B) flufenamic 248  acid, (C) meclofenamic acid, (D) mefenamic acid, (E) tolfenamic acid in 0.7 µM 249  perinaphthenone. Blue diamonds represent degradation in 90% D2O and red 250  hollow diamonds represent degradation in 100% H2O. Panel F shows the KSIE 251  (ratio of kobs (D2O/H2O)). 252  

 253   254  

255  Figure S14. Competition Plots vs. pseudo-first order degradation of benzoic acid 256  (BZA) for (A) diclofenac, (B) flufenamic acid, (C) meclofenamic acid, (D) 257  tolfenamic acid, and (E) mefenamic acid. Panel F shows the calculated 258  bimolecular reaction rate constants for diclofenac and the fenamates. 259  

260  

y"="$0.0005x"$ 0.3156R²"="0.9047

y"="$0.0004x"$ 0.3265R²"="0.8725

!3.5

!3

!2.5

!2

!1.5

!1

!0.5

0

0 1000 2000 3000 4000

Ln(c

t/c0)

Irradiation"time"(s)

Diclofenac"

diclofenac2+2PN2(90%2D2O)diclofenac2+2PN2(100%2H2O) y"="$0.0002x"+"0.0064

R²"="0.9912

y"="$0.0001x"+"0.008R²"="0.98719

!0.7

!0.6

!0.5

!0.4

!0.3

!0.2

!0.1

0

0.1

0 1000 2000 3000

Ln(c

t/c0)

Irradiation"time"(s)

Flufenamic"acid

flufenamic2+2PN2(90%2D2O)flufenamic2+2PN2(100%2H2O)

y"="$0.0005x"$ 0.0264R²"="0.99134

y"="$0.0003x"$ 0.0126R²"="0.99274

!1.2

!1

!0.8

!0.6

!0.4

!0.2

0

0.2

0 500 1000 1500 2000 2500 3000

Ln(c

t/c0)

Irradiation"time"(s)

Meclofenamic"acid

meclofenamic2acid2+2PN2(90%2D2O)meclofenamic2acid2+2PN2(100%2H2O)

y"="$0.0004x"$ 0.0288R²"="0.9852

y"="$0.0003x"$ 0.0281R²"="0.9787

!0.8!0.7!0.6!0.5!0.4!0.3!0.2!0.1

00.1

0 500 1000 1500 2000

Ln(c

t/c0)

Irradiation"time"(s)

Mefenamic"acid

mefenamic2+2PN2(90%2D2O)mefenamic2+2PN2(100%2H2O)

y"="$0.0004x"$ 0.0044R²"="0.9906

y"="$0.0003x"$ 0.0016R²"="0.9933

!1.2

!1

!0.8

!0.6

!0.4

!0.2

0

0.2

0 500 1000 1500 2000 2500 3000

Ln(c

t/c0)

Irradiation"time"(s)

Tolfenamic"acid

tolfenamic2+2PN2(90%2D2O)tolfenamic2+2PN2(100%2H2O)

KSIE"± errorDiclofenac 1.122± 0.14Flufenamic2acid 1.532± 0.11Meclofenamic2acid 1.532± 0.13Mefenamic2acid 1.322± 0.14Tolfenamic2acid 1.202± 0.09

A

FED

CB

y"="2.7647x"* 0.0149R²"="0.9873

!0.35

!0.3

!0.25

!0.2

!0.15

!0.1

!0.05

0

0.05

!0.15 !0.1 !0.05 0 0.05

Ln(c

t/c0)"Dic

Ln(ct/c0)"BZA

Dic vs"BZA

y"="1.795x"* 0.0336R²"="0.95026

!0.5

!0.4

!0.3

!0.2

!0.1

0

0.1

!0.2 !0.1 0 0.1

Ln(c

t/c0)"Flu

Ln(ct/c0)"BZA

Flu"vs"BZA"

y"="4.5945x"+"0.0467R²"="0.98393

!2

!1.5

!1

!0.5

0

0.5

!0.4 !0.2 0 0.2

Ln(c

t/c0)"Mec

Ln(ct/c0)"BZA

Mec vs"BZA

y"="2.1272x"* 0.0611R²"="0.69602

!0.35

!0.3

!0.25

!0.2

!0.15

!0.1

!0.05

0

0.05

!0.15 !0.1 !0.05 0 0.05

Ln(c

t/c0)"Tol

Ln(ct/c0)"BZA

Tol vs"BZA

y"="1.857x"* 0.0079R²"="0.90964

!0.2

!0.15

!0.1

!0.05

0

0.05

!0.1 !0.05 0 0.05

Ln(c

t/c0)"Mef

Ln(ct/c0)"BZA

Mef vs"BZA krxn(⦁OH)"x"1010"M!1 s!1

Diclofenac 1.635± 0.11Flufenamic5acid 1.065± 0.12Meclofenamic5acid 2.715± 0.21Mefenamic5acid 1.105± 0.17Tolfenamic5acid 1.265± 0.41

A

FED

CB

S   16  

Role of Triplet Sensitizing and Antioxidant moieties in DOM 261  

Below are the remaining pseudo-first order degradation plots for flufenamic, 262  

meclofenamic, mefenamic, tolfenamic acid for the steady-state experiment with 263  

0.77 µM perinaphthenone in air, argon sparged and with 10 µM caffeic acid. 264  

265  

266  

267  Figure S15. Pseudo-first order degradation of flufenamic acid (5 µM) in 268  enhanced UVA light in phosphate buffer (pH 7.5) only: black squares (n), in the 269  presence of the triplet sensitizer perinaphthenone (PN, 0.7 µM): red circles (¢), 270  with PN and argon sparged: blue triangles (r), and with PN and the 271  antioxidant caffeic acid (CA, 10 µM): green diamonds (¯), and the inset shows 272  the reaction rate constants, kobs and the log-normalised ratio of kobs, normalized 273  to kobs while sensitized with PN, ln(k/kPN). 274  

Flufenamic+acid

kobs (10-4 s-1) ln(k/kPN)

Buffer 0.26 -2.19PN + CA 0.10 -3.18PN 2.30 0PN (Ar) 8.92 1.35

S   17  

275  Figure S16. Pseudo-first order degradation of meclofenamic acid (5 µM) in 276  enhanced UVA light in phosphate buffer (pH 7.5) only: black squares (n), in the 277  presence of the triplet sensitizer perinaphthenone (PN, 0.7 µM): red circles (¢), 278  with PN and argon sparged: blue triangles (r), and with PN and the 279  antioxidant caffeic acid (CA, 10 µM): green diamonds (¯), and the inset shows 280  the reaction rate constants, kobs and the log-normalised ratio of kobs, normalized 281  to kobs while sensitized with PN, ln(k/kPN). 282  

283  

284  Figure S17. Pseudo-first order degradation of mefenamic acid (5 µM) in 285  enhanced UVA light in phosphate buffer (pH 7.5) only: black squares (n), in the 286  presence of the triplet sensitizer perinaphthenone (PN, 0.7 µM): red circles (¢), 287  with PN and argon sparged: blue triangles (r), and with PN and the 288  antioxidant caffeic acid (CA, 10 µM): green diamonds (¯), and the inset shows 289  the reaction rate constants, kobs and the log-normalised ratio of kobs, normalized 290  to kobs while sensitized with PN, ln(k/kPN). 291  

292  

kobs (10-4 s-1) ln(k/kPN)

Buffer 0.43 -2.39PN + CA 0.30 -2.74PN 4.65 0PN (Ar) 20.3 1.47

Meclofenamic+acid

Mefenamic)acid

kobs (10-4 s-1) ln(k/kPN)

Buffer 0.12 -3.63PN + CA 0.33 -2.62PN 4.50 0PN (Ar) 7.63 0.53

S   18  

293  Figure S18. Pseudo-first order degradation of tolfenamic acid (5 µM) in 294  enhanced UVA light in phosphate buffer (pH 7.5) only: black squares (n), in the 295  presence of the triplet sensitizer perinaphthenone (PN, 0.7 µM): red circles (¢), 296  with PN and argon sparged: blue triangles (r), and with PN and the 297  antioxidant caffeic acid (CA, 10 µM): green diamonds (¯), and the inset shows 298  the reaction rate constants, kobs and the log-normalised ratio of kobs, normalized 299  to kobs while sensitized with PN, ln(k/kPN). 300  

301  

Tolfenamic+acid

kobs (10-4 s-1) ln(k/kPN)

Buffer 0.19 -3.00PN + CA 0.48 -2.09PN 3.90 0PN (Ar) 9.65 0.90

S   19  

Text S5. Calculation of 3PN* quenching by caffeic acid for steady-state 302  experiments 303  

The decay (kd) of 3PN* in air sparged solution was calculated by the inverse its 304  

lifetime (τ = 1.7 µs). 305  

𝑘n = 5.8  ×  10o  𝑠7? 306  The amount of triplet decay due to caffeic acid (kCA) can be calculated by 307  

multiplying the bimolecular reaction rate constant of k(3PN*, Figure S6) with the 308  

caffeic acid concentration for the steady-state experiment. 309  

310  

𝑘qI = (3.5  ×  10r𝑀7?𝑠7?)  ×  (10  𝜇𝑀) 311  

= 3.5  ×10u  𝑠7?                                                   312  

313  

𝐴𝑑𝑑𝑖𝑡𝑖𝑜𝑛𝑎𝑙  𝑑𝑒𝑐𝑎𝑦 =𝑘n +  𝑘qI

𝑘n 314  

                                               = 6% 315   316  Therefore, caffeic acid would contribute to 6% decay of 3PN* in solution. 317   318   319   320   321  

322  Figure S19. Blank for triplet Lumichrome (3LC*). Pump beam = 360 nm. 323  Sparged continuously with argon gas. 324  

325  

S   20  

326  

327  

Figure S20. (A) Pseudo-first order degradation plots of the fenamate drugs in 328  the solar simulator in solution with 10 mgc L-1 Suwannee River Fulvic Acid, 329  controlled at pH 7.5, for mefenamic acid (1, blue), tolfenamic acid (2, green), 330  flufenamic acid (4, purple), meclofenamic acid (3, red), and diclofenac (5, black). 331  The table inset shows the half-lives under experimental conditions and the net 332  change in reaction rate (Dkobs) due to the DOM. 333  

334   335   336  

-7.0

-6.0

-5.0

-4.0

-3.0

-2.0

-1.0

0.0

0 4 8 12 16 20 24

Ln(C

t/C0)

exposure time (hours)

1

2

3

4

5

r2 = 0.997

r2 = 0.995

r2 = 0.982

r2 = 0.993

r2 = 0.982

t1/2 ∆kobs(%)

1. MEF 1.1 d + 90

2. TOL 3.7 h + 17

4. FLU 56 min - 27

3. MEC 48 min + 6

5. DIC 13 min + 11

S   21  

References 337   338  1. Bilski, P.; Daub, M. E.; Chignell, C. F., Direct detection of singlet oxygen via its 339  

phosphorescence from cellular and fungal cultures. Redox Cell Biology and 340  Genetics, Part A 2002, 352, 41-52. 341  

342  2. Jimenez-Banzo, A.; Ragas, X.; Kapusta, P.; Nonell, S., Time-resolved methods in 343  

biophysics. 7. Photon counting vs. analog time-resolved singlet oxygen 344  phosphorescence detection. Photochem Photobiol Sci 2008, 7, (9), 1003-10. 345  

346  3. Boreen, A. L.; Arnold, W. A.; McNeill, K., Triplet-sensitized photodegradation of 347  

sulfa drugs containing six-membered heterocyclic groups: Identification of an SO2 348  extrusion photoproduct. Environmental science & technology 2005, 39, (10), 349  3630-3638. 350  

351  4. Rodgers, M. A. J.; Snowden, P. T., Lifetime of O-2(1delta-G) in Liquid Water as 352  

Determined by Time-Resolved Infrared Luminescence Measurements. J Am 353  Chem Soc 1982, 104, (20), 5541-5543. 354  

355  5. Appiani, E.; Ossola, R.; Latch, D. E.; Erickson, P.R.; McNeill, K., Aqueous 356  

singlet oxygen reaction kinetics of furfuryl alcohol: Effect of temperature, pH, and 357  salt content. Submitted to Environmental Science: Processes & Impacts. 358  

359   360  6. Page, S. E.; Arnold, W. A.; McNeill, K., Terephthalate as a probe for 361  

photochemically generated hydroxyl radical. J Environ Monitor 2010, 12, (9), 362  1658-1665. 363  

364  7. Leifer, A.: The kinetics of environmental aquatic photochemistry theory and 365  

practice. American Chemical Society: Washington, 1988; p 85, 304 S. 366   367   368