Supporting information for publication
Supporting information for Photochemical production of singlet oxygen from particulate organic matter Elena Appiani† and Kristopher McNeill†* †
Department of Environmental Systems Science, ETH Zurich, Universitätstrasse 16,
8092 Zurich. *Author to whom correspondence should be addressed
TOC art Pages Synthesis of TPMA
S2
1
S8
H NMR spectrum of TPMA
Particle characterization
S9
Screening calculation
S10
Probe partitioning assessed by D2O
S11
Change of the light intensity in the presence of different sensitizer
S13
measured by PNA-Py actinometer FFA consumption in heterogeneous solution
S15
Bibliography
S17
Appiani, et al.
Supporting information
Page S1
Synthesis of 3-((1r,3r,5R,7S)-adamantan-2ylidene(methoxy)methyl)phenoxy)(tert-butyl)dimethylsilane (TPMA) O N
O
N H
+
Si Cl
OH
3x45 W max T 125°C
O
Si
1
O OH S O
O O
O O
O 70 °C
OTBS
OTBS
1
2 O
O O O
O P
O P O O
TiCl4 CH2Cl2
O
OTBS
OTBS
2
3
O
O
O
P O O
O LDA THF OTBS
OTBS 3
4
Scheme S1. Synthesis overview
Appiani, et al.
Supporting information
Page S2
Synthesis of 3-(tert-butyldimethyl)siloxybenzaldehyde (1) O N
O
N H
+
Si Cl
OH
3x45 W max T 125°C
Si
O
1
Scheme S2. TBS protection of 3-hydroxybenzaldehyde.
The procedure was adapted from Bastos et al.1 A CEM Peptide Hydrolysis Discover Microwave was used as a source of radiation. Hydroxylbenzaldehyde (6.12 g, 50 mmol, 1 equiv), tert-butyldimethylsilyl chloride (9.05 g, 60 mmol, 1.2 equiv) and imidazole (10.21 g, 150 mmol, 3 equiv) were placed in a round bottom flask and exposed to three microwave radiation cycles. The power was set to 45 W and the maximum temperature to 125 ºC. The reaction mixture was allowed to cool between cycles and the reaction was monitored by TLC (1:10 EtOAc:n-hexane). After three cycles, the reaction mixture was allowed to stand overnight. The reaction mixture was washed with water (1 × 50 mL) and the product extracted with EtOAc (3 × 50 mL). The organic phase was dried with MgSO4 and then the solvent removed by rotary evaporation. The crude product following extraction contained ca. 10% impurity, as determined by 1H NMR spectroscopy. The product was purified by flash chromatography (SiO2, 1:10 EtOAc:n-hexane). The purified product was isolated in 87 % yield
1
H-NMR (400 MHz, CDCl3) δ: 0.12 (s, 6H MeSi) , 0.90 (s, 9H t-Bu) ,
7.11 (ddd, J=1.18 Hz, 2.46 Hz, 8.1 Hz, 1H, Ar), 7.33 (dd, J=2.21 Hz, 1.67 Hz, 1H, Ar), 7.40 (t, J=7.7 Hz, 1H, Ar), 7.47 (dt, J=1.3 Hz, 7.6 Hz; 1H, Ar), 9.95 (s, 1H,
Appiani, et al.
Supporting information
Page S3
HCO).
13
C-NMR (100 MHz, CDCl3) δ: -4.28, 18.35, 25.76, 120.03, 123.70, 126.70,
130.23, 138.09, 156.56, 192.26. (tert-butyl(3-(dimethoxymethyl)phenoxy)dimethylsilane) (2) O OH S O
O O
O O
O 70 °C
OTBS 1
OTBS 2
Scheme S3. Synthesis of (tert-butyl(3-(dimethoxymethyl)phenoxy)dimethylsilane)
The synthesis of 2 was adapted from Roeschlaub et al.2 Compound 1 (3 g, 12.7 mmol, 1 equiv) was dissolved in 2,2-dimethoxypropane (3.2 mL, 25.8 mmol, 2 equiv) and few crystals of dry p-toluensulfonic acid were added. The reaction mixture was gently heated to 70 ºC while the acetone formed was distilled off the mixture and heating continued until no more acetone was formed. The reaction was cooled to room temperature, quenched with aq. NaHCO3 (50 mL) and extracted with CH2Cl2 (3 × 50 mL). The organic phase was washed with water (1 × 50 mL) and dried over MgSO4. The solvent was removed by rotary evaporation. The product was purified by vacuum distillation. The product (2) (99% purity by NMR) was isolated in 52% yield. 1HNMR (400 MHz, CDCl3) δ 0.19 (6H, s, MeSi), 0.98 (9H, s, t-Bu) 3.32 (6 H, s, MeO), 3.49 (1 H, s, ArCH), 6.80 (1 H, ddd, J 0.8 Hz, 2.4 Hz, 8 Hz, ArH), 6.93(1 H, t, J 1.96 Hz, ArH), 7.04 (1 H, d, J 7.68 Hz ArH), 7.22 (1 H, t, J 7.9 Hz, ArH). 13C-NMR (100 MHz, CDCl3) δ: -4.29, 18.33, 25.82, 52.77, 103.2, 116.70, 120.22, 123.68, 129.29, 139.76, 155.77.
Appiani, et al.
Supporting information
Page S4
Dimethyl ((3-(tert-butyldimethylsiloxy)phenyl)(methoxy)methyl)phosphonate (3)
O
O O O
O P
OTBS
O P O O
TiCl4 O
CH2Cl2 OTBS
2
3
Scheme S4. Synthesis of dimethyl ((3-(tert-butyldimethylsilyloxy)phenyl)(methoxy)methyl)phosphonate
The synthesis was adapted from Roeschlaub et al.2 Compound 2 (2.4 g, 8.09 mmol, 1 equiv) and trimethylphosphite (1.4 mL, 12.35 mmol, 1.5 equiv) were dissolved in CH2Cl2 (15 mL, 32 equiv) and cooled to -84 ºC under dry N2 atmosphere. TiCl4 (1.4 mL, 12.8 mmol, 1.5 equiv) were added dropwise by addition funnel over 30 min. The reaction mixture was stirred for 30 min at -84 ºC and then allowed to warm to room temperature and further stirred for 45 min. The reaction mixture was quenched with MeOH:H2O 2:1. The organic phase was extracted with CH2Cl2 (3 × 25 mL), washed with aq. NaHCO3 (1 × 25 mL) and brine (1 × 50 mL), and then dried over Na2SO4. The solvent was removed by rotary evaporation. The product was purified by flash chromatography (EtOAc: petroleum ether 3:1) and isolated in 92% yield. 1H-NMR (400 MHz, CDCl3) δ 0.20 (6H, s, MeSi), 0.98 (9H, s, t-Bu) 3.38 (3 H, s, MeO), 3.72 (6 H, dd, J 7, 10Hz, MeOP), 4.60 (1 H, d, J 15 Hz, CHP), 6.82 (1 H, dddd, J 1, 2.6, 8.1 Hz,ArH), 6.95 (1 H, dd, J 2.2, 4.1,ArH), 7.02 (1 H, d, J 7.7, ArH), 7.24 (1 H, t, J 7.9 Hz, ArH).
13
C-NMR (100 MHz, CDCl3) δ -4.31, 18.37, 25.82, 58.82, 81.02,
119.78, 120.60, 121.28, 129.78, 135.71, 156.02.
Appiani, et al.
Supporting information
Page S5
TPMA (4) O
O
P O O
O
O LDA THF OTBS
OTBS 3
4
Scheme S5. Synthesis of TPMA Diisopropylamine (1.62 mL, 11.28 mmol, 1.6 equiv) was dissolved in THF (9 mL) under dry N2 atmosphere, and cooled to -84 ºC. n-Butyllithium (7.05 mL, 11.28 M in hexanes, 1.6 equiv) was added dropwise and the mixture was stirred for 30 min to allow complete conversion to LDA. Compound 3 (2.54 g, 7.05 mmol, 1 equiv) dissolved in THF (10 mL) was added at -84 ºC dropwise by syringe over 15 min. The mixture was stirred for 1 h. Adamantanone (1.05 g, 7.05 mmol, 1 equiv) was dissolved in THF (3.5 mL) and added dropwise to the reaction mixture by syringe over 30 min. The reaction mixture was allowed to warm to room temperature and stirred for 6 h. The reaction was quenched with pH 7 phosphate buffer 0.2 M (25 mL), and the organic phase was rapidly extracted with CH2Cl2 (3 × 50mL). The extract was washed with aq. NaHCO3 (2 × 50 mL) and then with cold brine (2 × 50 mL). The organic phase was dried over MgSO4, and the solvent was removed by rotary evaporation. The product was purified by flash chromatography (19:1 petroleum ether: EtOAc) The purified product was recovered in 65% yield. 1H-NMR (400 MHz, CDCl3) δ 0.20 (6H, s, MeSi), 0.98 (9H, s, t-Bu), 1.78–1.97 (12 H, m, adamantanyl-H), 2.63 (1 H, br s, adamantanyl-H), 3.24 (1 H, br s, adamantanyl-H), 3.29 (3 H, s, MeO), 6.78 (2 H, tdd, J 0.9 Hz, 2.4 Hz, 8.05 Hz, ArH), 6.91 (1 H, dt, J 1.2 Hz, 7.6 Hz, ArH), 7.20 (1 H, t, J 7.8 Hz, ArH), 13C-NMR (100 MHz, CDCl3) -4.45, 18.23, 25.69, 28.35,
Appiani, et al.
Supporting information
Page S6
30.16, 32.25, 37.47, 38.83, 39.19, 57.64, 119.33, 121.09, 122.53, 128.89, 131.24, 136.79, 155.32.
tert-butyldimethyl(3-((1r,3r,5r,7r)-spiro[adamantane-2,3'-[1,2]dioxetan]-4'yl)phenoxy)silane H3CO
1O
2
O O
OCH3
OTBS TBSO
Scheme S6. Dioxetane synthesis TPMA (150 mg, 0.4 mmol) was dissolved in a solution of methylene blue in CDCl3 (10 µM, 2 mL) and placed in an NMR tube. The solution was cooled to 0 °C in an ice bath and irradiated with a xenon lamp with a 455 nm cutoff filter. The solution was bubbled with synthetic air and the product formation was followed by 1H NMR. After 5 hours the 1H NMR spectrum showed complete consumption of the starting material and formation of two products. The reaction mixture was filtered through a plug of silica to remove methylene blue. The silica was washed with CH2Cl2. The combined organic filtrates were purified by flash column (2:1 CH2Cl2:n-hexane). The pure product was recovered in 38% yield. 1H-NMR (400 MHz, CDCl3) NMR δ 0.19 (6H, s, MeSi), 0.99 (9H, s, t-Bu), 1.45–1.94 (12 H, m, adamantanyl-H), 2.24 (1 H, br s, adamantanyl-H), 3.02 (1 H, br s, adamantanyl-H), 3.24 (3 H, s, MeO), 6.88 (2 H, dd J 9.23 Hz, 2.35 Hz, ArH), 7.29 (2 H, t, J 7.8 Hz, ArH). 13C-NMR (100 MHz, CDCl3)
Appiani, et al.
Supporting information
Page S7
0.99
7.0
7.0
0.99 1.92
7.2
6.5
6.0
ppm
5.5
6.897 6.804 6.798 6.794 6.780 6.777 6.774 6.771 6.760 6.757 6.754 6.751
Title: TPMA PROTON in CDCl3
5.0
4.5
4.0
3.5
3.0
2.5
1.9
1.00
3.289 3.241 2.95 1.02
2.0
ppm
1.5
2.625 1.974 1.956 1.950 1.917 1.849 1.822 1.818 1.779 12.39
Supporting information 1.0
0.984 9.24
Appiani, et al. 0.5
ppm
0.196 5.97
F2 - Processing parameters SI 65536 SF 400.1300103 MHz WDW EM SSB 0 LB 0.30 Hz GB 0 PC 1.00
======== CHANNEL f1 ======== NUC1 1H P1 7.00 usec PLW1 11.56900024 W SFO1 400.1324710 MHz
F2 - Acquisition Parameters Date_ 20140107 Time 15.38 INSTRUM spect PROBHD 5 mm PABBI 1H/ PULPROG zg30 TD 65536 SOLVENT CDCl3 NS 16 DS 2 SWH 8223.685 Hz FIDRES 0.125483 Hz AQ 3.9846387 sec RG 203 DW 60.800 usec DE 6.50 usec TE 300.0 K D1 1.00000000 sec
Current Data Parameters NAME 140107 EXPNO 1 PROCNO 1
1
H NMR characterization of TPMA
Page S8
Particle characterization The organic carbon content was measured by TOC analyzer and is summarized in Table S1. Table S1. Carbon content in the different particle samples Baldeggersee Sediment 0.89% 0.49% 1.62% 0.6% 0.3% 8.06 % 1.5% 0.85% 9.68% c Inorganic carbon and PLL derived carbon, Total carbon POM-0.5
OCa content IC/PLLb TCc a Organic carbon, b
POM-1.0
The synthetic particles were further analyzed by zeta potential, to assess that the surface charge was changing from negative (SiO2) to positive (PLL coated SiO2) back to negative (AHA coated PLL-coated SiO2).
Baldeggersee sediment preparation details A sediment sample from Baldeggersee was collected, washed with nanopure water three times and filtered through a 0.5 µm filter. The sediment retained by the filter was dried under vacuum and further used for the photolysis experiment.
Appiani, et al.
Supporting information
Page S9
Screening correction calculation Rate constants obtained from experiments with DOM solutions and DOM+POM samples were corrected for light screening. The absorption spectra were measured in 1 cm quartz cuvettes using a Cary 100 spectrophotometer (Varian) for solutions containing DOM (10 mgC/L), POM (10 mgC/L) and DOM+POM. The relative irradiance, Eλ, of the UV light was recorded with a spectrometer (OceanOptics Inc.). The screening factor, S, is described in eq S1, where α is the optical density at each wavelength, λ (nm), and z (cm) is the optical path length. !!!"!! !
𝑆! = !.!"! !
! !
( S1)
The relative light intensity experienced by OM was estimated as follows, 𝐼=
! 𝑆! × 𝐸!
(S2)
where E is the relative irradiance at each wavelength. Light screening was calculated λ
over the wavelength range in which chromophoric DOM absorbs light (345-410 nm). The observed degradation rates of the probes were corrected for light screening by dividing the measured value by the estimated I value. Table S1 summarizes the relative light intensity experienced for each sample. The POM-containing samples had a light attenuation factor far larger than 50% (and thus were not considered quantitatively useful as they led to correction factors that were larger than the original value) Table S1. Hypothesized attenuation by screening obtained from UV-vis measurements. Sample Blank DOM POM 0.5 POM 1 POM+DOM 0.5 POM+DOM 1 Appiani, et al.
E 0.98 0.83 0.17 0.16 0.16 0.15
Attenuation 2% 17% 84% 84% 85% 85% Supporting information
Page S10
Probe partitioning assessed by D2O According to the KOC of TPMA and FFA, we expect TPMA to partition completely in the DOM and FFA to be evenly distributed in the solution. To verify this model we performed photolysis experiments in D2O and in H2O (with 1% d6-ethanol and ethanol, respectively). Enhancement in the [1O2]aq in the D2O experiment was expected since relaxation by the solvent is the major quenching mechanism for 1O2 in aqueous solution. On the contrary, no enhancement was expected for [1O2]OM, since the main loss pathway is due to diffusion that is substantially unchanged from D2O to H2O. If TPMA is completely partitioned in DOM we expect unchanged [1O2]ss measured by TPMA in D2O and H2O. Table S2 reports the [1O2]ss observed by TPMA and FFA in the two solvents and reflects our expectation concerning the partitions of the probes. Notice that the expected 1O2 lifetime enhancement for pure D2O vs pure H2O is 13-fold,3 but the enhancement is sensitive to H2O traces (dropping in half in the presence of 6% H2O),3 and in the experiment design there are several factors that can affect the 1O2 lifetime such as the presence of a co-solvent in both the experiments, the presence of water associated with the particles which, for experimental reasons, cannot be dried. In addition the error associated with the FFA measurements in the presence of particles is large due to the uncertainties introduced by the heterogeneity of the system that makes the quantification of the enhancement challenging.
Appiani, et al.
Supporting information
Page S11
Table S2. Solvent isotope effect for TPMA and FFA with different sensitizers [1O2]ss (pM)
TPMA
D2 O H2O D2O/H2O
POM 0.5
POM 1
27 ± 3.1 25 ± 3.5 1.1
25 ± 4.8 33 ± 3.3 0.8
D2 O 0.18 ± 0.01 0.15±0.01 FFA H2O 0.02 ± 0.01 0.04 ± 0.02 D2O/H2O 7.6 3.7 1 The [ O2]ss measured by TPMA and FFA are reported in pM concentrations. The ratio between the [1O2]ss detected in the experiment performed in D2O and the [1O2]ss detected in the experiment performed in H2O is also reported.
Appiani, et al.
Supporting information
Page S12
Change of the light intensity in the presence of different sensitizers measured by PNA-Py actinometer The p-nitroanisole / pyridine (PNA / Py) actinometer was developed by Dulin and Mill4 and has been widely employed in the environmental photochemistry because it is not sensitive to wavelength or temperature. We can use this actinometer to estimate the enhancement in irradiance in the presence of particles. O O N
hv NO 2N
NO 2
Scheme S7. p-nitroanisole-pyridine actinometer reaction
The reaction is the photochemical substitution of pyridine for nitrite. Since it is a bimolecular reaction, the quantum yield (Φ, eq S3)4 is dependent on the concentration of pyridine. Φ = 0.44 Py + 2.8 × 10!!
(S3)
We followed the actinometer reaction under the different POM and DOM sensitizer conditions. In addition we monitored the reaction in the presence of uncoated SiO2 beads of both sizes. The measured rates are reported in Table S3. The quantum yield function is given in eq S4. 𝑘!"!!" = 2.303 × 𝐸! ×𝜀! Φ
(S4)
𝑘!"!!" is the observed rate, 𝜀! is the molar extinction coefficient (6.3 ×103) and 𝐸! is the irradiance. Using eq S4, and the known value of Φ from eq S2, the irradiance is estimated.
Appiani, et al.
Supporting information
Page S13
Table S3. Irradiance of the solutions in the presence of different sensitizers Solution DOM POM-1 POM-0.5 Blank SiO2-1 SiO2-0.5
kPN-Py (× 10-4M-1 s-1) 0.73 1.00 1.18 1.34 1.46 1.65
Eλ (× 10-6 W m-2) 1.21 1.64 1.94 2.20 2.40 2.71
Eλ, n/Eλ, blank 0.55 0.75 0.88 1.00 1.09 1.23
The results reported in Table S3 suggest that the screening is partially compensated by the scattering of the particles. In the solution containing POM-0.5 and in the one containing POM-1 the OM content is the same resulting in a larger number of particles in solution in the case of POM-0.5 (~1.09 1013 units/L) compared to POM-1 (~3.25 1012 units/L). This can esplain the larger enhancement due to scattering effect in the case of POM-0.5 compared to POM-1 particles, both coated and uncoated.
Appiani, et al.
Supporting information
Page S14
FFA consumption in heterogeneous solution In order to exclude the geometric effect explanation for the difference in the ! !"# [ ! 𝑂! ]!"# !" and [ 𝑂! ]!" detected by FFA, we estimated the difference of FFA inside
and outside DOM under the experimental condition. To do so, we need to compare the moles of FFA consumed per second in the OM unit with the moles of FFA per second supplied to the OM region trough diffusion. We calculate the gradient between inside the DOM and outside the DOM as a limiting case, since we know the [1O2]OM from the TPMA measurement. The first step is to estimate the consumption rate of FFA inside one DOM aggregate (
!!!!",!" !"
) (eq S5) !!!!",!" !"
!
= 𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! π 𝑟 !
(S5)
Where krxn is the FFA and 1O2 bimolecular reaction rate constant (estimated to be 8.3 107 M-1 s-1)5, [FFA] is the concentration of FFA in solution (10-4 M) ,[1O2]OM is the concentration of 1O2 in the organic matter (estimated by TPMA to be roughly 3 10-11 M, Table 1) and r is the radius of the OM aggregate (that, as a semplification, we assume spherical) and is estimated 1 nm.6
Equation 6 represents the moles of FFA supplied to the OM trought the OM surface area and is equal to the flux (derived through the Fick’s law of diffusion) times the surface area (eq S6). 𝐽 × 𝑆𝐴 = D ×
[!!"]! ![!!"]! !
×4π 𝑟 !
(S6)
Where J is the diffusion flux in units of mmol s-1 cm-2, SA is the surface area of the OM aggregate (cm2 with radius r), D is the diffusivity (estimated, according to the Stoke-Einstein equation,7 to be 4.7 x 10-6 cm2 s -1 for FFA) and l is the distance
Appiani, et al.
Supporting information
Page S15
troughout the flux (cm). The maximum diffusion length along which we can estimate the flux is through the entire radius of the OM particle, where we set equal l to r. Combining eq S5 and S6 we obtain eq S7 !
𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! π 𝑟 ! = D ×
[!!"]! ![!!"]! !
×4π 𝑟 !
(S7)
And upon rearrangiament we obtain eq S8 !
𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! 𝑟 ! = D ×([𝐹𝐹𝐴]! − [𝐹𝐹𝐴]! )
(S8)
Solving eq 8 for ([FFA]0-[FFA]r) we obtain a difference between outside and inside OM of 1.8 10-16M. We can conclude that under the experimental condition we do not have a significant gradient from the inside to the outside of the OM region. If we set ([FFA]0-[FFA]r) to a significant change such as 10 % (10-5 M), using the same relationship, we can propose the scenarios where such an eventuality can occur. One possibility is to have 1011 more concentrated 1O2 in OM. This would mean a [1O2] of 1 M, which is physically impossible. A second possibility would be to have a much larger OM aggregate. In order to obtain a 10-5 M difference between the interior and exterior concentrations of 1O2, we need a radius of OM of 100 µm.
A more intuitive explanation of this problem is given by calculating the distance FFA can diffuse in the time it takes to consume 10% of it through the reaction with 1O2. The calculated time for 3 ×10-11 M 1O2 is 120 s, which corresponds to a distance of 230 µm. For distances significantly shorter than 230 µm, diffusion is much faster than loss of FFA.
Appiani, et al.
Supporting information
Page S16
References 1. Bastos, E. L.; Ciscato, L. F. M. L.; Baader, W. J., Microwave‐Assisted Protection of Phenols as tert‐Butyldimethylsilyl (TBDMS) Ethers Under Solvent‐Free Conditions. Synthetic Commun. 2005, 35, 1501-1509. 2. Roeschlaub, C. A.; Sammes, P. G., Use of the Wadsworth-Emmons reaction for preparing hindered vinyl ethers and related 1,2-dioxetanes. J. Chem. Soc. Perk. Trans. 1 2000, 2243-2248. 3. Rodgers, M. A. J.; Snowden, P. T., Lifetime of oxygen (O2(1∆g)) in liquid water as determined by time-resolved infrared luminescence measurements. J. Am. Chem. Soc. 1982, 104, 5541-5543. 4. Dulin, D.; Mill, T., Development and evaluation of sunlight actinometers. Environ. Sci. Technol. 1982, 16, 815-820. 5. Latch, D. E.; Stender, B. L.; Packer, J. L.; Arnold, W. A.; McNeill, K., Photochemical fate of pharmaceuticals in the environment: cimetidine and ranitidine. Environ. Sci. Technol. 2003, 37, 3342-3350. 6. Latch, D. E.; McNeill, K., Microheterogeneity of Singlet Oxygen Distributions in Irradiated Humic Acid Solutions. Science 2006, 311, 1743-1747. 7. Cussler, E. L., Diffusion. Cambridge University Press: 2009.
Appiani, et al.
Supporting information
Page S17
Department of Environmental Systems Science, ETH Zurich, Universitätstrasse 16,
8092 Zurich. *Author to whom correspondence should be addressed
TOC art Pages Synthesis of TPMA
S2
1
S8
H NMR spectrum of TPMA
Particle characterization
S9
Screening calculation
S10
Probe partitioning assessed by D2O
S11
Change of the light intensity in the presence of different sensitizer
S13
measured by PNA-Py actinometer FFA consumption in heterogeneous solution
S15
Bibliography
S17
Appiani, et al.
Supporting information
Page S1
Synthesis of 3-((1r,3r,5R,7S)-adamantan-2ylidene(methoxy)methyl)phenoxy)(tert-butyl)dimethylsilane (TPMA) O N
O
N H
+
Si Cl
OH
3x45 W max T 125°C
O
Si
1
O OH S O
O O
O O
O 70 °C
OTBS
OTBS
1
2 O
O O O
O P
O P O O
TiCl4 CH2Cl2
O
OTBS
OTBS
2
3
O
O
O
P O O
O LDA THF OTBS
OTBS 3
4
Scheme S1. Synthesis overview
Appiani, et al.
Supporting information
Page S2
Synthesis of 3-(tert-butyldimethyl)siloxybenzaldehyde (1) O N
O
N H
+
Si Cl
OH
3x45 W max T 125°C
Si
O
1
Scheme S2. TBS protection of 3-hydroxybenzaldehyde.
The procedure was adapted from Bastos et al.1 A CEM Peptide Hydrolysis Discover Microwave was used as a source of radiation. Hydroxylbenzaldehyde (6.12 g, 50 mmol, 1 equiv), tert-butyldimethylsilyl chloride (9.05 g, 60 mmol, 1.2 equiv) and imidazole (10.21 g, 150 mmol, 3 equiv) were placed in a round bottom flask and exposed to three microwave radiation cycles. The power was set to 45 W and the maximum temperature to 125 ºC. The reaction mixture was allowed to cool between cycles and the reaction was monitored by TLC (1:10 EtOAc:n-hexane). After three cycles, the reaction mixture was allowed to stand overnight. The reaction mixture was washed with water (1 × 50 mL) and the product extracted with EtOAc (3 × 50 mL). The organic phase was dried with MgSO4 and then the solvent removed by rotary evaporation. The crude product following extraction contained ca. 10% impurity, as determined by 1H NMR spectroscopy. The product was purified by flash chromatography (SiO2, 1:10 EtOAc:n-hexane). The purified product was isolated in 87 % yield
1
H-NMR (400 MHz, CDCl3) δ: 0.12 (s, 6H MeSi) , 0.90 (s, 9H t-Bu) ,
7.11 (ddd, J=1.18 Hz, 2.46 Hz, 8.1 Hz, 1H, Ar), 7.33 (dd, J=2.21 Hz, 1.67 Hz, 1H, Ar), 7.40 (t, J=7.7 Hz, 1H, Ar), 7.47 (dt, J=1.3 Hz, 7.6 Hz; 1H, Ar), 9.95 (s, 1H,
Appiani, et al.
Supporting information
Page S3
HCO).
13
C-NMR (100 MHz, CDCl3) δ: -4.28, 18.35, 25.76, 120.03, 123.70, 126.70,
130.23, 138.09, 156.56, 192.26. (tert-butyl(3-(dimethoxymethyl)phenoxy)dimethylsilane) (2) O OH S O
O O
O O
O 70 °C
OTBS 1
OTBS 2
Scheme S3. Synthesis of (tert-butyl(3-(dimethoxymethyl)phenoxy)dimethylsilane)
The synthesis of 2 was adapted from Roeschlaub et al.2 Compound 1 (3 g, 12.7 mmol, 1 equiv) was dissolved in 2,2-dimethoxypropane (3.2 mL, 25.8 mmol, 2 equiv) and few crystals of dry p-toluensulfonic acid were added. The reaction mixture was gently heated to 70 ºC while the acetone formed was distilled off the mixture and heating continued until no more acetone was formed. The reaction was cooled to room temperature, quenched with aq. NaHCO3 (50 mL) and extracted with CH2Cl2 (3 × 50 mL). The organic phase was washed with water (1 × 50 mL) and dried over MgSO4. The solvent was removed by rotary evaporation. The product was purified by vacuum distillation. The product (2) (99% purity by NMR) was isolated in 52% yield. 1HNMR (400 MHz, CDCl3) δ 0.19 (6H, s, MeSi), 0.98 (9H, s, t-Bu) 3.32 (6 H, s, MeO), 3.49 (1 H, s, ArCH), 6.80 (1 H, ddd, J 0.8 Hz, 2.4 Hz, 8 Hz, ArH), 6.93(1 H, t, J 1.96 Hz, ArH), 7.04 (1 H, d, J 7.68 Hz ArH), 7.22 (1 H, t, J 7.9 Hz, ArH). 13C-NMR (100 MHz, CDCl3) δ: -4.29, 18.33, 25.82, 52.77, 103.2, 116.70, 120.22, 123.68, 129.29, 139.76, 155.77.
Appiani, et al.
Supporting information
Page S4
Dimethyl ((3-(tert-butyldimethylsiloxy)phenyl)(methoxy)methyl)phosphonate (3)
O
O O O
O P
OTBS
O P O O
TiCl4 O
CH2Cl2 OTBS
2
3
Scheme S4. Synthesis of dimethyl ((3-(tert-butyldimethylsilyloxy)phenyl)(methoxy)methyl)phosphonate
The synthesis was adapted from Roeschlaub et al.2 Compound 2 (2.4 g, 8.09 mmol, 1 equiv) and trimethylphosphite (1.4 mL, 12.35 mmol, 1.5 equiv) were dissolved in CH2Cl2 (15 mL, 32 equiv) and cooled to -84 ºC under dry N2 atmosphere. TiCl4 (1.4 mL, 12.8 mmol, 1.5 equiv) were added dropwise by addition funnel over 30 min. The reaction mixture was stirred for 30 min at -84 ºC and then allowed to warm to room temperature and further stirred for 45 min. The reaction mixture was quenched with MeOH:H2O 2:1. The organic phase was extracted with CH2Cl2 (3 × 25 mL), washed with aq. NaHCO3 (1 × 25 mL) and brine (1 × 50 mL), and then dried over Na2SO4. The solvent was removed by rotary evaporation. The product was purified by flash chromatography (EtOAc: petroleum ether 3:1) and isolated in 92% yield. 1H-NMR (400 MHz, CDCl3) δ 0.20 (6H, s, MeSi), 0.98 (9H, s, t-Bu) 3.38 (3 H, s, MeO), 3.72 (6 H, dd, J 7, 10Hz, MeOP), 4.60 (1 H, d, J 15 Hz, CHP), 6.82 (1 H, dddd, J 1, 2.6, 8.1 Hz,ArH), 6.95 (1 H, dd, J 2.2, 4.1,ArH), 7.02 (1 H, d, J 7.7, ArH), 7.24 (1 H, t, J 7.9 Hz, ArH).
13
C-NMR (100 MHz, CDCl3) δ -4.31, 18.37, 25.82, 58.82, 81.02,
119.78, 120.60, 121.28, 129.78, 135.71, 156.02.
Appiani, et al.
Supporting information
Page S5
TPMA (4) O
O
P O O
O
O LDA THF OTBS
OTBS 3
4
Scheme S5. Synthesis of TPMA Diisopropylamine (1.62 mL, 11.28 mmol, 1.6 equiv) was dissolved in THF (9 mL) under dry N2 atmosphere, and cooled to -84 ºC. n-Butyllithium (7.05 mL, 11.28 M in hexanes, 1.6 equiv) was added dropwise and the mixture was stirred for 30 min to allow complete conversion to LDA. Compound 3 (2.54 g, 7.05 mmol, 1 equiv) dissolved in THF (10 mL) was added at -84 ºC dropwise by syringe over 15 min. The mixture was stirred for 1 h. Adamantanone (1.05 g, 7.05 mmol, 1 equiv) was dissolved in THF (3.5 mL) and added dropwise to the reaction mixture by syringe over 30 min. The reaction mixture was allowed to warm to room temperature and stirred for 6 h. The reaction was quenched with pH 7 phosphate buffer 0.2 M (25 mL), and the organic phase was rapidly extracted with CH2Cl2 (3 × 50mL). The extract was washed with aq. NaHCO3 (2 × 50 mL) and then with cold brine (2 × 50 mL). The organic phase was dried over MgSO4, and the solvent was removed by rotary evaporation. The product was purified by flash chromatography (19:1 petroleum ether: EtOAc) The purified product was recovered in 65% yield. 1H-NMR (400 MHz, CDCl3) δ 0.20 (6H, s, MeSi), 0.98 (9H, s, t-Bu), 1.78–1.97 (12 H, m, adamantanyl-H), 2.63 (1 H, br s, adamantanyl-H), 3.24 (1 H, br s, adamantanyl-H), 3.29 (3 H, s, MeO), 6.78 (2 H, tdd, J 0.9 Hz, 2.4 Hz, 8.05 Hz, ArH), 6.91 (1 H, dt, J 1.2 Hz, 7.6 Hz, ArH), 7.20 (1 H, t, J 7.8 Hz, ArH), 13C-NMR (100 MHz, CDCl3) -4.45, 18.23, 25.69, 28.35,
Appiani, et al.
Supporting information
Page S6
30.16, 32.25, 37.47, 38.83, 39.19, 57.64, 119.33, 121.09, 122.53, 128.89, 131.24, 136.79, 155.32.
tert-butyldimethyl(3-((1r,3r,5r,7r)-spiro[adamantane-2,3'-[1,2]dioxetan]-4'yl)phenoxy)silane H3CO
1O
2
O O
OCH3
OTBS TBSO
Scheme S6. Dioxetane synthesis TPMA (150 mg, 0.4 mmol) was dissolved in a solution of methylene blue in CDCl3 (10 µM, 2 mL) and placed in an NMR tube. The solution was cooled to 0 °C in an ice bath and irradiated with a xenon lamp with a 455 nm cutoff filter. The solution was bubbled with synthetic air and the product formation was followed by 1H NMR. After 5 hours the 1H NMR spectrum showed complete consumption of the starting material and formation of two products. The reaction mixture was filtered through a plug of silica to remove methylene blue. The silica was washed with CH2Cl2. The combined organic filtrates were purified by flash column (2:1 CH2Cl2:n-hexane). The pure product was recovered in 38% yield. 1H-NMR (400 MHz, CDCl3) NMR δ 0.19 (6H, s, MeSi), 0.99 (9H, s, t-Bu), 1.45–1.94 (12 H, m, adamantanyl-H), 2.24 (1 H, br s, adamantanyl-H), 3.02 (1 H, br s, adamantanyl-H), 3.24 (3 H, s, MeO), 6.88 (2 H, dd J 9.23 Hz, 2.35 Hz, ArH), 7.29 (2 H, t, J 7.8 Hz, ArH). 13C-NMR (100 MHz, CDCl3)
Appiani, et al.
Supporting information
Page S7
0.99
7.0
7.0
0.99 1.92
7.2
6.5
6.0
ppm
5.5
6.897 6.804 6.798 6.794 6.780 6.777 6.774 6.771 6.760 6.757 6.754 6.751
Title: TPMA PROTON in CDCl3
5.0
4.5
4.0
3.5
3.0
2.5
1.9
1.00
3.289 3.241 2.95 1.02
2.0
ppm
1.5
2.625 1.974 1.956 1.950 1.917 1.849 1.822 1.818 1.779 12.39
Supporting information 1.0
0.984 9.24
Appiani, et al. 0.5
ppm
0.196 5.97
F2 - Processing parameters SI 65536 SF 400.1300103 MHz WDW EM SSB 0 LB 0.30 Hz GB 0 PC 1.00
======== CHANNEL f1 ======== NUC1 1H P1 7.00 usec PLW1 11.56900024 W SFO1 400.1324710 MHz
F2 - Acquisition Parameters Date_ 20140107 Time 15.38 INSTRUM spect PROBHD 5 mm PABBI 1H/ PULPROG zg30 TD 65536 SOLVENT CDCl3 NS 16 DS 2 SWH 8223.685 Hz FIDRES 0.125483 Hz AQ 3.9846387 sec RG 203 DW 60.800 usec DE 6.50 usec TE 300.0 K D1 1.00000000 sec
Current Data Parameters NAME 140107 EXPNO 1 PROCNO 1
1
H NMR characterization of TPMA
Page S8
Particle characterization The organic carbon content was measured by TOC analyzer and is summarized in Table S1. Table S1. Carbon content in the different particle samples Baldeggersee Sediment 0.89% 0.49% 1.62% 0.6% 0.3% 8.06 % 1.5% 0.85% 9.68% c Inorganic carbon and PLL derived carbon, Total carbon POM-0.5
OCa content IC/PLLb TCc a Organic carbon, b
POM-1.0
The synthetic particles were further analyzed by zeta potential, to assess that the surface charge was changing from negative (SiO2) to positive (PLL coated SiO2) back to negative (AHA coated PLL-coated SiO2).
Baldeggersee sediment preparation details A sediment sample from Baldeggersee was collected, washed with nanopure water three times and filtered through a 0.5 µm filter. The sediment retained by the filter was dried under vacuum and further used for the photolysis experiment.
Appiani, et al.
Supporting information
Page S9
Screening correction calculation Rate constants obtained from experiments with DOM solutions and DOM+POM samples were corrected for light screening. The absorption spectra were measured in 1 cm quartz cuvettes using a Cary 100 spectrophotometer (Varian) for solutions containing DOM (10 mgC/L), POM (10 mgC/L) and DOM+POM. The relative irradiance, Eλ, of the UV light was recorded with a spectrometer (OceanOptics Inc.). The screening factor, S, is described in eq S1, where α is the optical density at each wavelength, λ (nm), and z (cm) is the optical path length. !!!"!! !
𝑆! = !.!"! !
! !
( S1)
The relative light intensity experienced by OM was estimated as follows, 𝐼=
! 𝑆! × 𝐸!
(S2)
where E is the relative irradiance at each wavelength. Light screening was calculated λ
over the wavelength range in which chromophoric DOM absorbs light (345-410 nm). The observed degradation rates of the probes were corrected for light screening by dividing the measured value by the estimated I value. Table S1 summarizes the relative light intensity experienced for each sample. The POM-containing samples had a light attenuation factor far larger than 50% (and thus were not considered quantitatively useful as they led to correction factors that were larger than the original value) Table S1. Hypothesized attenuation by screening obtained from UV-vis measurements. Sample Blank DOM POM 0.5 POM 1 POM+DOM 0.5 POM+DOM 1 Appiani, et al.
E 0.98 0.83 0.17 0.16 0.16 0.15
Attenuation 2% 17% 84% 84% 85% 85% Supporting information
Page S10
Probe partitioning assessed by D2O According to the KOC of TPMA and FFA, we expect TPMA to partition completely in the DOM and FFA to be evenly distributed in the solution. To verify this model we performed photolysis experiments in D2O and in H2O (with 1% d6-ethanol and ethanol, respectively). Enhancement in the [1O2]aq in the D2O experiment was expected since relaxation by the solvent is the major quenching mechanism for 1O2 in aqueous solution. On the contrary, no enhancement was expected for [1O2]OM, since the main loss pathway is due to diffusion that is substantially unchanged from D2O to H2O. If TPMA is completely partitioned in DOM we expect unchanged [1O2]ss measured by TPMA in D2O and H2O. Table S2 reports the [1O2]ss observed by TPMA and FFA in the two solvents and reflects our expectation concerning the partitions of the probes. Notice that the expected 1O2 lifetime enhancement for pure D2O vs pure H2O is 13-fold,3 but the enhancement is sensitive to H2O traces (dropping in half in the presence of 6% H2O),3 and in the experiment design there are several factors that can affect the 1O2 lifetime such as the presence of a co-solvent in both the experiments, the presence of water associated with the particles which, for experimental reasons, cannot be dried. In addition the error associated with the FFA measurements in the presence of particles is large due to the uncertainties introduced by the heterogeneity of the system that makes the quantification of the enhancement challenging.
Appiani, et al.
Supporting information
Page S11
Table S2. Solvent isotope effect for TPMA and FFA with different sensitizers [1O2]ss (pM)
TPMA
D2 O H2O D2O/H2O
POM 0.5
POM 1
27 ± 3.1 25 ± 3.5 1.1
25 ± 4.8 33 ± 3.3 0.8
D2 O 0.18 ± 0.01 0.15±0.01 FFA H2O 0.02 ± 0.01 0.04 ± 0.02 D2O/H2O 7.6 3.7 1 The [ O2]ss measured by TPMA and FFA are reported in pM concentrations. The ratio between the [1O2]ss detected in the experiment performed in D2O and the [1O2]ss detected in the experiment performed in H2O is also reported.
Appiani, et al.
Supporting information
Page S12
Change of the light intensity in the presence of different sensitizers measured by PNA-Py actinometer The p-nitroanisole / pyridine (PNA / Py) actinometer was developed by Dulin and Mill4 and has been widely employed in the environmental photochemistry because it is not sensitive to wavelength or temperature. We can use this actinometer to estimate the enhancement in irradiance in the presence of particles. O O N
hv NO 2N
NO 2
Scheme S7. p-nitroanisole-pyridine actinometer reaction
The reaction is the photochemical substitution of pyridine for nitrite. Since it is a bimolecular reaction, the quantum yield (Φ, eq S3)4 is dependent on the concentration of pyridine. Φ = 0.44 Py + 2.8 × 10!!
(S3)
We followed the actinometer reaction under the different POM and DOM sensitizer conditions. In addition we monitored the reaction in the presence of uncoated SiO2 beads of both sizes. The measured rates are reported in Table S3. The quantum yield function is given in eq S4. 𝑘!"!!" = 2.303 × 𝐸! ×𝜀! Φ
(S4)
𝑘!"!!" is the observed rate, 𝜀! is the molar extinction coefficient (6.3 ×103) and 𝐸! is the irradiance. Using eq S4, and the known value of Φ from eq S2, the irradiance is estimated.
Appiani, et al.
Supporting information
Page S13
Table S3. Irradiance of the solutions in the presence of different sensitizers Solution DOM POM-1 POM-0.5 Blank SiO2-1 SiO2-0.5
kPN-Py (× 10-4M-1 s-1) 0.73 1.00 1.18 1.34 1.46 1.65
Eλ (× 10-6 W m-2) 1.21 1.64 1.94 2.20 2.40 2.71
Eλ, n/Eλ, blank 0.55 0.75 0.88 1.00 1.09 1.23
The results reported in Table S3 suggest that the screening is partially compensated by the scattering of the particles. In the solution containing POM-0.5 and in the one containing POM-1 the OM content is the same resulting in a larger number of particles in solution in the case of POM-0.5 (~1.09 1013 units/L) compared to POM-1 (~3.25 1012 units/L). This can esplain the larger enhancement due to scattering effect in the case of POM-0.5 compared to POM-1 particles, both coated and uncoated.
Appiani, et al.
Supporting information
Page S14
FFA consumption in heterogeneous solution In order to exclude the geometric effect explanation for the difference in the ! !"# [ ! 𝑂! ]!"# !" and [ 𝑂! ]!" detected by FFA, we estimated the difference of FFA inside
and outside DOM under the experimental condition. To do so, we need to compare the moles of FFA consumed per second in the OM unit with the moles of FFA per second supplied to the OM region trough diffusion. We calculate the gradient between inside the DOM and outside the DOM as a limiting case, since we know the [1O2]OM from the TPMA measurement. The first step is to estimate the consumption rate of FFA inside one DOM aggregate (
!!!!",!" !"
) (eq S5) !!!!",!" !"
!
= 𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! π 𝑟 !
(S5)
Where krxn is the FFA and 1O2 bimolecular reaction rate constant (estimated to be 8.3 107 M-1 s-1)5, [FFA] is the concentration of FFA in solution (10-4 M) ,[1O2]OM is the concentration of 1O2 in the organic matter (estimated by TPMA to be roughly 3 10-11 M, Table 1) and r is the radius of the OM aggregate (that, as a semplification, we assume spherical) and is estimated 1 nm.6
Equation 6 represents the moles of FFA supplied to the OM trought the OM surface area and is equal to the flux (derived through the Fick’s law of diffusion) times the surface area (eq S6). 𝐽 × 𝑆𝐴 = D ×
[!!"]! ![!!"]! !
×4π 𝑟 !
(S6)
Where J is the diffusion flux in units of mmol s-1 cm-2, SA is the surface area of the OM aggregate (cm2 with radius r), D is the diffusivity (estimated, according to the Stoke-Einstein equation,7 to be 4.7 x 10-6 cm2 s -1 for FFA) and l is the distance
Appiani, et al.
Supporting information
Page S15
troughout the flux (cm). The maximum diffusion length along which we can estimate the flux is through the entire radius of the OM particle, where we set equal l to r. Combining eq S5 and S6 we obtain eq S7 !
𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! π 𝑟 ! = D ×
[!!"]! ![!!"]! !
×4π 𝑟 !
(S7)
And upon rearrangiament we obtain eq S8 !
𝑘!"# 𝐹𝐹𝐴 [ ! 𝑂! ]!" × ! 𝑟 ! = D ×([𝐹𝐹𝐴]! − [𝐹𝐹𝐴]! )
(S8)
Solving eq 8 for ([FFA]0-[FFA]r) we obtain a difference between outside and inside OM of 1.8 10-16M. We can conclude that under the experimental condition we do not have a significant gradient from the inside to the outside of the OM region. If we set ([FFA]0-[FFA]r) to a significant change such as 10 % (10-5 M), using the same relationship, we can propose the scenarios where such an eventuality can occur. One possibility is to have 1011 more concentrated 1O2 in OM. This would mean a [1O2] of 1 M, which is physically impossible. A second possibility would be to have a much larger OM aggregate. In order to obtain a 10-5 M difference between the interior and exterior concentrations of 1O2, we need a radius of OM of 100 µm.
A more intuitive explanation of this problem is given by calculating the distance FFA can diffuse in the time it takes to consume 10% of it through the reaction with 1O2. The calculated time for 3 ×10-11 M 1O2 is 120 s, which corresponds to a distance of 230 µm. For distances significantly shorter than 230 µm, diffusion is much faster than loss of FFA.
Appiani, et al.
Supporting information
Page S16
References 1. Bastos, E. L.; Ciscato, L. F. M. L.; Baader, W. J., Microwave‐Assisted Protection of Phenols as tert‐Butyldimethylsilyl (TBDMS) Ethers Under Solvent‐Free Conditions. Synthetic Commun. 2005, 35, 1501-1509. 2. Roeschlaub, C. A.; Sammes, P. G., Use of the Wadsworth-Emmons reaction for preparing hindered vinyl ethers and related 1,2-dioxetanes. J. Chem. Soc. Perk. Trans. 1 2000, 2243-2248. 3. Rodgers, M. A. J.; Snowden, P. T., Lifetime of oxygen (O2(1∆g)) in liquid water as determined by time-resolved infrared luminescence measurements. J. Am. Chem. Soc. 1982, 104, 5541-5543. 4. Dulin, D.; Mill, T., Development and evaluation of sunlight actinometers. Environ. Sci. Technol. 1982, 16, 815-820. 5. Latch, D. E.; Stender, B. L.; Packer, J. L.; Arnold, W. A.; McNeill, K., Photochemical fate of pharmaceuticals in the environment: cimetidine and ranitidine. Environ. Sci. Technol. 2003, 37, 3342-3350. 6. Latch, D. E.; McNeill, K., Microheterogeneity of Singlet Oxygen Distributions in Irradiated Humic Acid Solutions. Science 2006, 311, 1743-1747. 7. Cussler, E. L., Diffusion. Cambridge University Press: 2009.
Appiani, et al.
Supporting information
Page S17
