Tellurorhodamine photocatalyzed aerobic oxidation of organo-silanes and phosphines by visible-light

Article abstract of DOI:10.1039/c9dt00487d

Tellurorhodamine, 9-mesityl-3,6-bis(dimethylamino)telluroxanthylium hexafluorophosphate (1), photocatalytically oxidizes aromatic and aliphatic silanes and triphenyl phosphine under mild aerobic conditions. Under irradiation with visible light, 1 can react with self-sensitized ¹O₂ to generate the active telluroxide oxidant (2). Silanes are oxidized to silanols and triphenyl phosphine is oxidized to triphenyl phoshine oxide either using 2, or 1 with aerobic irradiation. Kinetic experiments coupled with a computational study elucidate possible mechanisms of oxidation for both silane and phosphine substrates. First-order rates were observed in the oxidation of triphenyl phosphine and methyldiphenyl silane, indicating a substitution like mechanism for substrate binding to the oxidized tellurium(iv). Additionally, these reactions exhibited a rate-dependence on water. Oxidations were typically run in 50:50 water/methanol, however, the absence of water decreased the rates of silane oxidation to a greater degree than triphenyl phosphine oxidation. Parallel results were observed in solvent kinetic isotope experiments using D₂O in the solvent mixture. The rates of oxidation were slowed to a greater degree in silane oxidation by 2 (k_H/k_D = 17.30) than for phosphine (k_H/k_D = 6.20). Various silanes and triphenyl phosphine were photocatalytically oxidized with 1 (5%) under irradiation with warm white LEDs using atmospheric oxygen as the terminal oxidant.

Full text of DOI:10.1039/c9dt00487d

Dalton
Transactions
View Article Online
View Journal
PAPER
Tellurorhodamine photocatalyzed aerobic
oxidation of organo-silanes and phosphines by
Cite this: DOI: 10.1039/c9dt00487d
visible-light†
Irving D. Rettig, Jackson Van, Jacob B. Brauer, Wentai Luo and
Theresa M. McCormick
*
Tellurorhodamine, 9-mesityl-3,6-bis(dimethylamino)telluroxanthylium hexafluorophosphate (1), photo-
catalytically oxidizes aromatic and aliphatic silanes and triphenyl phosphine under mild aerobic con-
1
ditions. Under irradiation with visible light, 1 can react with self-sensitized O2 to generate the active tell-
uroxide oxidant (2). Silanes are oxidized to silanols and triphenyl phosphine is oxidized to triphenyl phoshine
oxide either using 2, or 1 with aerobic irradiation. Kinetic experiments coupled with a computational study
elucidate possible mechanisms of oxidation for both silane and phosphine substrates. First-order rates were
observed in the oxidation of triphenyl phosphine and methyldiphenyl silane, indicating a substitution like
mechanism for substrate binding to the oxidized tellurium(IV). Additionally, these reactions exhibited a rate-
dependence on water. Oxidations were typically run in 50 : 50 water/methanol, however, the absence of
water decreased the rates of silane oxidation to a greater degree than triphenyl phosphine oxidation.
Parallel results were observed in solvent kinetic isotope experiments using D2O in the solvent mixture. The
rates of oxidation were slowed to a greater degree in silane oxidation by 2 (kH/kD = 17.30) than for phos-
phine (kH/kD = 6.20). Various silanes and triphenyl phosphine were photocatalytically oxidized with 1 (5%)
under irradiation with warm white LEDs using atmospheric oxygen as the terminal oxidant.
Received 1st February 2019,
Accepted 3rd April 2019
DOI: 10.1039/c9dt00487d
rsc.li/dalton
dine-type ligands.5,6 Though extensive research in ligand
modifications on these chromophores have allowed for
1 Introduction
Over the past decade, there has been a tremendous push to improved 1O2 generation, the oxidation reactions are equally
1
develop catalysts for green synthetic methods within the field limited by the low selectivity of transient O2 and its radical-
of organic chemistry.1,2 Catalysts that have been employed in type chemistry.7–10 Organic photosensitizers like Rose Bengal,
oxidation reactions represent one such avenue of research, Erythrosin blue, and Methylene blue have been developed to
however, the utility of many catalytic oxidation reactions is directly address the economic and environmental setbacks of
limited due to the generation of stoichiometric amounts of co- metal-containing photosensitizers, however, these dyes are
oxidant waste.3 Using molecular oxygen as the stoichiometric reactive towards 1O2, and, in many cases, have decreased stabi-
oxidant in catalytic oxidation reactions avoids this issue due to lity in solution because of 1O2 induced degradation.11
its high relative abundance, low cost and negligible environ-
mental toxicity.
These major challenges in aerobic oxidation photocatalysis
can be addressed by using the photosensitized O2 to generate
1
One method of utilizing molecular oxygen in aerobic cata- longer-lived reactive intermediates that will selectively oxidize
lytic oxidation reactions is through the generation of photosen- desired substrates. In an effort to address this, Ando and co-
sitized singlet oxygen (1O2) from ground state triplet oxygen workers developed a series of biaryl organotellurium com-
(3O2). Typically, inorganic metal-containing chromophores pounds that could react with 1O2 generated by an external
with strong spin–orbit coupling are used to photosensitize organic photosensitizer to form a considerably more selective
1
1O2.4 The most common of these complexes involve a heavy oxidant than O2 alone.12 1O2 oxidizes the Te(II) compound to
metal center, like iridium, ruthenium or osmium, with bipyri- Te(IV), which can then reduce upon interaction with a substrate
to regenerate the initial Te(II) compound. Using this method,
Ando reported successful photocatalytic aerobic oxidation of
Department of Chemistry, Portland State University, Portland, Oregon, 97201, USA.
phosphite esters, silanes and thiols.13–15 However, in each
E-mail: t.m.mccormick@pdx.edu
case, an external photosensitizer was needed to initially gene-
rate O2. Detty and coworkers have developed tellurium-based
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
1
c9dt00487d
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
using previously published methods19 A solution of 1 in
aqueous methanol (50 mmol, 5.4 × 10−4 M) was irradiated
with warm white light LEDs while stirring for approximately
8 hours. The absorbance maxima of 1 and 2, (600 nm and
665 nm respectively) in methanol were used to monitor the for-
mation of 2 in situ.24 After the UV-Vis spectra of 2 persisted (no
further oxidation), the solution was evaporated to dryness and
stored in the dark as a teal residue for use in preparing solu-
tion for both UV-Vis absorption and 1H/13C NMR spectroscopic
experiments.
Scheme 1 Oxidation of 1 to 2 with light and 3O2, generating 1O2 which
directly oxidizes 1. Two equivalents of thiol can be reductively elimi-
nated from 2 to return 1.
UV-Vis spectroscopy
UV-Vis absorption spectroscopy was used to elucidate kinetic
and mechanistic details for the reduction of 2 to 1 by phos-
phines and silanes. Rates for oxidation of triphenyl phosphine
and various silanes were determined using pseudo-first order
reaction conditions. In each experiment, a solution of 2,
ranging from 2.4 × 10−6 M to 5.4 × 10−6 M, was prepared in
50 : 50 methanol/water (unless specified otherwise) and spiked
with excess (20–100 equivalents, experiment specific) of reduc-
tant and sealed. Spectra were acquired between 400–800 nm
every 30 seconds for 5 to 15 min and were baseline corrected
in each case. Reactions that took longer than 15 minutes to
complete were stirred continuously between scans, which were
then acquired every 5 minutes. Reaction rates were calculated
from the natural log of the concentration of 2 over time. The
concentration of 2 was determined by the Beer’s law equation
using the absorbance at 665 nm and molar extinction coeffi-
cient for 2 used was 9.5 × 104 M−1 cm−1, as reported by
Detty et al.24
1
organic dyes that react with O2, in a similar vein to the diaryl
telluride compounds but also generate the required 1O2.16–18
We have previously published that a tellurium-based rhod-
amine analog 1 can form telluroxide 2 upon irradiation with
visible light, and then reduce back to 1 by oxidizing various
thiols to their corresponding disulfides, shown in Scheme 1.19
This net oxidative transformation was characterized as
having a substitution-like mechanism via a dithio-tellurane
intermediate. A transfer of the oxygen atom to the substrate
was not involved in the thiol oxidation mechanism.
Phosphine and silane oxidation are two industrially rele-
vant chemical reactions that involve oxygen atom transfer that
can be accessed via aerobic oxidation catalysis. In particular
for silane oxidation, noble metal catalysts and nano-materials
are typically used for reactions of this type.20–23 Oxidized diaryl
tellurium-containing compounds have been observed to
reduce in the presence of both silanes and phosphite
esters,13,15 However, experimentally supported mechanisms
for the oxygen transfer by telluroxides have yet to be explored.
Additionally, oxidation of phosphines and silanes by 2 has not
been reported, in which the only reactants required to oxidize
these substrates is light, oxygen and water. We propose two
different mechanisms involving oxygen transfer from 2 to
afford the experimentally observed phosphine oxides and sila-
nols, supported by kinetic experiments and computations.
1H/13C NMR
1H/13C NMR experiments were conducted using a 600 mHz
Bruker NMR spectrometer. Phosphines and silanes were oxi-
dized catalytically using 5 mol% of 1 in methanol-d4. Samples
were prepared under oxygen atmosphere with 1 (1.0 × 10−3 M),
20 eq. of reductant (phosphine or silane) (3.0 × 10−2 M), 5%
water, and cyclohexane as an internal standard in a 5 mm
Wilmad Lab-Glass low-pressure/vacuum NMR tube. To ensure
sufficient gas exchange, the tubes were stirred via inversion
using a custom-made rotating stage mounted inside of a
white-light LED chamber.
2 Experimental
Mass spectrometry
All reagents and solvents were purchased from Sigma Aldrich
and Tokyo Chemical Industry (TCI). Experiments requiring
anhydrous conditions were performed using Schlenk line tech-
niques or in a nitrogen-containing glovebox. The synthesis of 1
is detailed in the ESI.† 1H/13C NMR spectra (Fig. S1–2†) and
Uv-Vis absorption spectra (Fig. S3†) of 1 are consistent with
previously published literature values.24
Mass spectrometry identified the oxidized products of photo-
catalytic oxidation of both silanes and phosphines with 1.
Samples were prepared with 1 (1.0 × 10−3 M), 20 eq. of reduc-
tant (phosphine or silane) (3.0 × 10−2 M) and 5% water in
methanol with an oxygen atmosphere. Triphenyl phosphine
and aromatic and aliphatic silanes were oxidized for the same
period of time as in the NMR experiments (18–24 hours).
Samples were diluted to 100 ng μl−1 before analysis. Agilent
(Santa Clara, California, USA) 7693 autosampler, Agilent 7890A
Oxidation of phosphines or silanes using 2
A bulk solution of 2 was prepared to allow us to study the oxi- GC and Agilent 5975C MS was used for the analysis. The GC
dation of different substrates by 2, which is the proposed cata- column was Restek (Bellefonte, PA) Rxi-5Sil MS of 30 m length,
lyst turnover step. Photo-oxidation of 1 to 2 was performed 0.25 mm id and 0.25 μm film thickness. One microlitre of the
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
sample solution was injected on the GC injector at 225 °C with
a 20 : 1 split. The GC temperature program for all analyses was:
80 °C hold for 2 min; 15 °C min−1 to 300 °C; hold at 300 °C for
1 min; then 15 °C min−1 to 320 °C. The MS operated in elec-
tron impact ionization mode. The MS source temperature was
225 °C and the scan range was 34 to 400 amu. The identifi-
cations were made by matching the mass spectra of character-
ized compounds to their NIST library mass spectra or the
mass spectra generated from pure compounds purchased from
Sigma Aldrich and the GC retention times of the compounds.
Scheme 2 Stoicheometric reduction of 2 by excess triphenyl phos-
phine. Various concentrations of triphenyl phosphine were reacted with
2 (2.4 × 10−6 M) and monitored by UV-Vis spectroscopy.
Rates were determined for a range of concentrations of tri-
phenyl phosphine (2.4 × 10−5 M, 4.8 × 10−5 M, 7.2 × 10−5 M,
9.6 × 10−5 M) with 2 (2.4 × 10−6 M) in 50 : 50 water/methanol.
The UV-Vis traces of 2 reducing to 1 with 30 eq. of triphenyl
phosphine (7.2 × 10−5 M) are shown in Fig. 1a. Full conversion
of 2 to 1, indicated by the disappearance of the peak at
665 nm and persistence of the peak at 600 nm, occurred
within 15 minutes. An isosbestic point is observed at 618 nm
indicating direct conversion from 2 to 1. The natural log of the
concentration of 2, when plotted against time, gave a linear fit
(Fig. 1b). Plotting these rates against the concentrations of tri-
phenyl phosphine gave a pseudo-first-order rate constant of
16.67 M−1 s−1 (Fig. 1 insert). The observed linear relationship
between the concentration of triphenyl phosphine and time
would suggest a first order reaction.
Computational details
Calculations were completed with Gaussian0925 suit of pro-
grams, and results were visualized using GaussView05.26 The
structures were optimized to a minima using the B3LYP27–29
level of theory with a split basis set (6-31+G(d)30–32 for all light
atoms, LanL2DZ33–35 for Te). Gibbs free energies were obtained
from thermally corrected energy values from the frequency
calculations on minimized structures. Hirshfeld charge
population analyses were calculated from optimized
geometries.36,37
3 Results and discussion
Stoichiometric reduction of 2 to 1
We determined the stoichiometry of this reaction by adding
half or one stoichiometric equivalent of triphenyl phosphine
to 2 (5.4 × 10−6 M) in methanol and stirred in the dark for two
hours. The half equivalent sample had nearly equal amounts
of 2 and 1 present. One equivalent of triphenyl phosphine was
sufficient to fully reduced 2 to 1 (Fig. S4†). We later used this
stoichiometry in conjunction with computational modeling to
propose a reaction mechanism for oxygen transfer from 2 to
phosphine to afford the corresponding phosphine oxide.
The kinetics of silane to silanol oxidation by 2 was also
explored experimentally. Initially, triphenyl silane was selected
as an appropriate analog to triphenyl phosphine to determine
reaction kinetics, however, the reaction of 2 with 100 equiva-
lents of triphenyl silane took 11 hours to fully reduce
(Fig. S5†). Instead, diphenyl methyl silane was oxidized by 2
within 5 min and was used to determine rates of silane oxi-
dation. Scheme 3 shows the reaction conditions for stoichio-
metric methyl diphenyl silane oxidation by 2.
Information about the reduction step of the catalytic cycle was
obtained from kinetic experiments on reduction of 2 to 1 with
varying amounts of excess substrate. Compound 2 was pre-
pared by irradiating a solution of 1 under atmospheric con-
ditions and the solvent was removed. We monitored the reac-
tion between 2 and both phosphines and silanes using UV-Vis
spectroscopy to determine the concentration of 1 and 2 over
time using previously published photophysical data (Table 1).
The overall rates of both phosphine and silane oxidation are
slow enough that standard UV-Vis spectroscopic methods can
be used to monitor the reaction. This is ∼103 times slower
than our previously reported rates of oxidation of aromatic
thiols by the reduction of 2 to 1.19
To determine the rate constants for the oxidation reactions,
the concentration of 2 was determined by monitoring the
absorption at 665 nm over time, after the addition of various
concentrations of the reductants. Scheme 2 shows the reaction
conditions for stoichiometric oxidation of triphenyl phosphine
oxidation by 2.
Akin to phosphine oxidation experiments, various concen-
trations of diphenyl methyl silane (5.4 × 10−5 M, 1.1 × 10−4 M,
1.6 × 10−4 M, 2.2 × 10−4 M, 2.7 × 10−4 M) were reacted with 2
(5.4 × 10−6 M) to determine substrate dependence on reaction
Table 1 Photophysical information for 1 and 220
kinetics. Full reduction of
5–15 minutes, dependent on silane concentration. Plots of the
concentration of versus time gave exponential decays
2 to 1 occurred between
Molar extinction
Absorbance
coefficient
Singlet oxygen
quantum yield (ϕ)
2
Compound maximum (nm) (M−1 cm−1
)
(Fig. 2a). Taking the natural log of the concentration of 2 gave
linear traces (Fig. 2b), corresponding to pseudo-first-order
reaction rates. These were plotted against the concentration of
methyl diphenyl silane to give a rate constant of 31.18 M−1 s−1
(Fig. 2 insert). Attempts to stoichiometrically reduce 2 with
1
2
600
665
86 000
95 000
0.75 0.03
n/aa
a Triplet quantum yields are not available for 2 due to fluorescent
emission.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 (a) Example UV-Vis absorbance spectra of reduction of 2 to 1 with 30 eq. triphenyl phosphine, (b) natural log of the concentration of 2
versus time at increasing concentrations of triphenyl phosphine and the corresponding rate constant plot as the insert in (b).
intermediate 2-P, as discussed below, has partial negative and
positive charges located on the tellurium-bound oxygen and
phosphine atoms respectively. In this case, because water is
more polar than methanol, water solvation could stabilize
intermediate 2-P. This increased stabilization due to water sol-
vation could explain the increase in reaction kinetics in
aqueous versus neat methanol.
Scheme 3 Stoicheometric reduction of 2 with excess methyl diphenyl
silane. Various concentrations of silane were reacted with 2 (5.4 × 10−6
Under the same conditions, the rate of reaction between
methyl diphenyl silane (2.2 × 10−5 M) and 2 (5.4 × 10−6 M) was
2.61 0.28 × 10−5 s−1, compared to 1.04 0. 017 × 10−3 s−1 in
the wet reaction. Dividing kwet/kdry gives a ratio of 39.85, con-
siderably higher than kwet/kdry of triphenyl phosphine. The
large difference in the rates between wet and dry conditions
shows the significance of water solvation effects in silane oxi-
dation by 2. The computationally proposed intermediate 2-Si,
as discussed below, has both a hydroxyl group that can partici-
pate in hydrogen bonding and a partially negative silane group
attached to the tellurium core that can be solvated by water.
Stabilization from these multiple solvent interactions could
account for the slowed kinetics in anhydrous solvent.
Additionally, because kwet/kdry of both oxidation reactions
differ from one another, it can be said that water dependence
varies with substrate and that the role of water is different in
silane oxidation than in phosphine oxidation.
M) and monitored by UV-Vis spectroscopy.
one equivalent of silane resulted in only partial return of 1 due
to the slow reaction kinetics at these low concentrations.
Effect of water on reaction rates
We had previously determined that water plays a possible role
in the substitution-like mechanism of thiol oxidation.19 The
rates of phosphine and silane oxidation under anhydrous con-
ditions were determined to elucidate the role of water in each
reaction.
The ratio of the reaction rates under wet (kwet) and dry (kdry
)
conditions can be compared to determine the role of water in
directly or indirectly solvating the reaction. The observed
pseudo-first-order rates were slower under dry conditions in
each case. Under anhydrous conditions, the pseudo-first order
rate of triphenyl phosphine (2.2 × 10−5 M) reacting with 2
(5.4 × 10−6 M) was measured at 6.06 1.2 × 10−4 s−1 (kdry). In
Solvent kinetic isotope effect
the aqueous solvent mixture, the pseudo-first-order rate was To further explore the role of water solvation in these reac-
1.62 0.4 × 10−3 s−1 (kwet), which when compared gives a ratio tions, substrate oxidation by 2 was carried out in deuterium
of 2.67. Computational details show that the phosphine-bound oxide (D2O) and methanol to measure the solvent kinetic
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Fig. 2 (a) Pseudo-first-order rate plots of [2] at various concentrations of methyl diphenyl silane with respect to time, (b) linear traces obtained by
taking the natural log of [2] versus time and the corresponding rate constant plot in the insert.
1
isotope effect. When D2O was substituted for H2O, the standard to products in each H NMR spectrum. The scope of
reduction of 2 to 1 by triphenyl phosphine (2.2 × 10−5 M) the reactants successfully oxidized to their corresponding
under pseudo-first order conditions was much slower, with a phosphine oxide and silanol analogs are compiled in Table 2.
rate of 2.61 0.5 × 10−4 s−1 versus 1.62 0.4 × 10−3 s−1 at the
Triphenyl phosphine was catalytically oxidized to triphenyl
same concentration in 50 : 50 H2O/MeOH. The kinetic isotope phosphine oxide in methanol-d4 within 18 hours by 1 with
effect in this case is 6.20, suggesting that water plays a direct quantitative yield, shown in Scheme 4.
role in solvating the reaction of phosphine oxidation by 2 via
To determine if 1 was responsible for the catalytic oxidation
hydrogen bonding.34,35
of triphenyl phosphine or 1O2 directly, a 1O2 photosensitizer
When reacting methyl diphenyl silane with 2 in 50 : 50 D2O/ (Rose Bengal) (1.0 × 10−3 M) was combined with triphenyl
MeOH, the pseudo-first-order rate was observed to be 6.01
0.14 × 10−5 −1. In H2O/MeOH this was 1.04 0.017 × 10−3 s−1
phosphine (3.0 × 10−2 M) and irradiated under the same reac-
tion conditions and analyzed by GCMS. After 18 hours, a mass
s
.
A kinetic isotope effect of 17.30 was obtained in this case, peak of 277 m/z was observed, which indicates the presence of
which is too large to be described by a single proton exchange. triphenyl phosphine oxide (Fig. S16†). Although one would
Compounding primary and secondary kinetic isotope effects expect competing reaction between 1O2 and 2, 2 gets formed
can occur when more than one proton is involved in exchan- very rapidly, as shown in our previously published work.19 The
ging with and solvating a reaction intermediate.36,37 This large proximity of 1O2 to the tellurium core after formation could
kinetic isotope effect could therefore be the result of proton
exchange from solvent, like in the dehydration step to produce
2-Si shown below, as well as hydrogen-bonding stabilization
from other solvent molecules of this intermediate.
Table 2 Scope of photocatalytic phosphine and silane oxidation using
5% 1 in methanol-d4 irradiated with warm white LEDs and monitored by
1H NMR
Catalytic substrate oxidation by 1
Irradiation
time (h)
Conversion
(%)
Yield
(%)
Phosphine and silanes were oxidized catalytically to their
corresponding phosphine oxides and silanols with 5% 1.
1H/13C NMR spectroscopy was used to monitor the conversion
of starting materials to products (Fig. S6–15†). Yields were
quantified using cyclohexane as an internal standard, compar-
ing the integration ratio of standard to starting material and
Entry
Reactant
1
2
3
4
5
Triphenyl silane
24
18
18
18
18
96
99
≥99
≥99
≥99
85
98
99
95
Methyl diphenyl silane
Dimethyl phenyl silane
Triethyl silane
Triphenyl phosphine
≥99
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
277 m/z was the only peak observed in the spectrum, which
corresponds to triphenyl phosphine oxide (Fig. S18†). Silane
oxidation products were also characterized with GCMS. Masses
of 276, 214, 152, and 132 m/z were obtained, matching the
reference mass peaks of triphenyl silanol, methyldiphenyl
silanol, dimethylphenyl silanol and triethyl silanol respectively
(Fig. S18–21†). A large peak appeared in each sample 14 mass
units above the molecular ion peak. This indicates solvent
exchange wherein –OMe interchanges with the –OH of each
silanol (Fig. S22–25†). Another common byproduct of silane
oxidation is disiloxane caused by unfavorable condensation. A
small mass peak in the triethyl silane oxidation sample corre-
lated to the corresponding disiloxane (Fig. S26†). However disi-
loxane formation was not observed for any other tested silane
even in aqueous conditions. We suggest the low steric bulk of
the ethyl moieties significantly increases ease dimerization of
the silanol product compared to the bulkier substrates.38,39
Scheme 4 Catalytic oxidation of triphenyl phosphine with 5 mol% of 1.
Reactions were run in an oxygen atmosphere and irradiated with light
for 18 hours.
increase its likelihood of reacting with tellurium versus phos-
phine. Furthermore, we have shown that triphenyl phosphine
can react with 2 in the dark to form triphenyl phosphine oxide
1
and 1. Thus, O2 or photocatalytically formed 2 could result in
the observed phosphine oxidation.
Catalytic oxidation of both aromatic and aliphatic silanes to
their corresponding silanols were obtained with 5% 1 in
methanol-d4 within 18–24 hours of irradiation, shown in
Scheme 5. Nearly all silanols were obtained in quantitative
yields, as determined by 1H NMR spectroscopy.
Computationally supported mechanism
DFT methods were used to further elucidate potential mecha-
nisms of phosphine and silane oxidation by 2. Thermal and free
energy corrected to total energy values of the optimized struc-
tures were used to calculate the Gibbs free energy changes (ΔG)
between reactants and products. To reduce overall computational
time, trimethyl phosphine and trimethyl silane were used to
model the mechanisms of oxidation. Calculations included
implicit water solvation to reflect experimental findings.
The disappearance of the silane proton peak was observed
in all cases, as well as an up field shift in the alpha-carbon
protons of the aliphatic and aromatic moieties. Unlike the
other silanes, after 24 hours triphenyl silane had only partially
oxidized, evidenced by the persistence of the silane proton
peak in the 1H NMR spectrum. As seen in the above rate
studies, the large steric bulk of triphenyl silane significantly
slows the reaction with 2, which could account for increased
time of oxidation and the less than quantitive yield. A control
experiment was also conducted to determine if 1O2 could
oxidize silane to silanol with Rose Bengal (Fig. S17†).
Conditions and concentrations for the control were replicated
with triethyl silane (1.0 × 10−3 M) and photosensitizer (3.0 ×
10−2 M) in 95 : 5 MeOD/H2O. After 24 hours of irradiation, the
1H NMR spectrum of silane was unchanged. This not onlyindi-
The reaction coordinate diagram of trimetyl phosphine oxi-
dation by 2 is shown in Fig. 3. The first step in the proposed
reaction pathway is the binding of phosphine to the tellurium(IV)
core of 2 to give 2-P, which has a ΔG of 11.97 kcal mol−1
.
Hirshfeld charge analysis was used to determine the change in
electron density at the tellurium core between 2 and 2-P to verify
the bonding interaction. A decrease in positive charge density is
observed at the tellurium atom, 0.39 in 2 and 0.34 in 2-P, which
indicates electron donation from the phosphorus atom to the
tellurium. The Te–P bond length of 3.59 Å is within the sum of
van der Waals radii for tellurium and phosphorus, further sup-
porting bond formation in the 2-P intermediate.
1
cates that silane does not react with O2 generated in situ but
that the silane is unreactive towards water under these
conditions.
We have previously reported that 2 can react with water to
form a hydrated intermediate 2a, shown below.19 In this case,
substitution of phosphine at the Te atom in 2a to generate
monophospho-hydroxy tellurane is energetically unfavorable,
with a ΔG of +68.55 kcal mol−1 in vacuum. Including water sol-
vation in the calculation of this intermediate resulted in a
structural minimum where neither the phosphine nor hydroxyl
group were still attached to the tellurium core. Direct substi-
tution of 2, in phosphine oxidation without proceeding
through a hydrated intermediate first is a new mechanistic pre-
diction based on experimental and computational results.
Following the binding of phosphine to form 2-P, reductive
elimination of trimethyl phosphine oxide regenerated 1 with a
ΔG of −69.80 kcal mol−1. The overall ΔGrxn of phosphine oxi-
dation by 2 was calculated to be −57.83 kcal mol−1. This
mechanism supports stoichiometric experimental results that
one equivalent of phosphine is sufficient reduce 2 to 1.
Catalytic phosphine and silanol formation were confirmed
with GC-MS. Samples were prepared using the NMR experi-
mental conditions, but with non-deuterated solvent. Triphenyl
phosphine and aromatic and aliphatic silanes were oxidized
for the same period of time as in the NMR experiments
(18–24 hours). For triphenyl phosphine oxidation, a peak at
Scheme 5 General catalytic oxidation of silanes with 5 mol% of 1.
Reactions were run in an oxygen atmosphere and irradiated with light
for 18–24 hours.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Fig. 3 Reaction coordinate of trimethyl phosphine oxidation by 2.
We also propose a mechanism for silane oxidation, which by 2 by proceeding through hydrated intermediate 2a.19 The
reflects a different pathway to silanol formation. This mecha- reaction coordinate diagram of silane oxidation is shown in
nism more closely resembles the mechanism of thiol oxidation Fig. 4. The initial reaction step is the binding of water by 2 to
Fig. 4 Reaction coordinate of trimethyl silane oxidation. Reductive elimination of silanol from the tellurium core of 2-Si yields the reduced
catalyst 1.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
form 2a which has a ΔG of 9.44 kcal mol−1. Following this is
the substitution of one of the hydroxyl groups for one moiety
of silane to generate 2-Si and water, which results in the silane
bound to the tellurium core in a seesaw conformation.
Without the inclusion of a water solvation model, the calcu-
lated ΔG of silane substitution was 8.93 kcal mol−1; however,
including the water solvation model reduced this ΔG to
0.47 kcal mol−1. Reductive elimination of silanol regenerates 1,
which has a ΔG of −70.88 kcal mol−1. The overall ΔGrxn for the
oxidation of trimethyl silane to trimethyl silanol by 2 is
References
1 P. T. Anastas, M. M. Kirchhoff and T. C. Williamson,
Catalysis as a foundational pillar of green chemistry, Appl.
Catal., A, 2001, 221, 3–13.
2 S. K. Ritter, Green chemistry gets greener, Chem. Eng. News,
2002, 80, 38–42.
3 A. N. Campbell and S. S. Stahl, Overcoming the ‘oxidant
problem’: Strategies to use O2 as the oxidant in organo-
metallic C-H oxidation reactions catalyzed by Pd (and Cu),
Acc. Chem. Res., 2012, 45, 851–863.
−60.97 kcal mol−1
.
4 K. Kawaoka, A. U. Khan and D. R. Kearns, Role of singlet
excited states of molecular oxygen in the quenching of
organic triplet states, J. Chem. Phys., 1967, 46, 1842–1853.
5 J. N. Demas, E. W. Harris and R. P. McBride, Energy
Transfer from Luminescent Transition Metal Complexes to
Oxygen, J. Am. Chem. Soc., 1977, 99, 3547–3551.
6 C. Tanielian, C. Wolff and M. Esch, Singlet oxygen pro-
duction in water: Aggregation and charge-transfer effects,
J. Phys. Chem., 1996, 100, 6555–6560.
7 A. A. Abdel-Shafi, P. D. Beer, R. J. Mortimer and
F. Wilkinson, Photosensitized generation of singlet oxygen
from (substituted bipyridine)ruthenium(II) complexes,
Helv. Chim. Acta, 2001, 84, 2784–2795.
4 Conclusion
Tellurorhodamine chromophores have a rich oxidation chem-
istry that has, until now, been relatively unexplored. We have
successfully shown that both phosphine and various silanes
can be catalytically oxidized by 1 to the corresponding phos-
phine oxide and silanols. Reaction kinetics of the oxidation of
triphenyl phosphine and methyl diphenyl silane by 2 were
explored. Substrate dependence on reaction kinetics was
observed in both cases, which points to an associative substi-
tution-like mechanism for both phosphine and silane.
The role of water in each oxidation reaction was explored.
Kinetic isotope effect studies showed that hydrogen bonding
from water solvent largely effects the relative rates of oxidation,
with silane affected more so, likely due to explicit proton
exchange with water. Triphenyl phosphine and all silanes were
catalytically oxidized within 18–24 hours with irradiation, con-
firmed by 1H/13C NMR spectroscopy and mass spectrometry.
The mechanisms by which 2 oxidizes phosphine and silane
were also explored. Using the results of kinetic experiments
and computation, different mechanisms were proposed for
phosphine and silane oxidation. In the case of phosphine,
direct addition to 2 affords a tetra-substituted intermediate
before reductive elimination of phosphine oxide and regener-
ation of 1. Substitution of phosphine on 2a was found to be
energetically unfavorable. Silane oxidation proceeded through
a substitution-like mechanism wherein a single hydroxyl group
from 2a is displaced by silane, to form water and a silane
bound intermediate 2-Si. Similar reductive elimination from
the tellurium core affords the corresponding silanol and 1.
8 E. K. Pefkianakis, D. Christodouleas, D. L. Giokas,
K. Papadopoulos and G. C. Vougioukalakis, A family of
RuIIphotosensitizers with high singlet oxygen quantum
yield: Synthesis, characterization, and evaluation, Eur,
J. Inorg. Chem., 2013, 4628–4635.
9 A. A. Abdel-Shafi, M. D. Ward and R. Schmidt, Mechanism
of quenching by oxygen of the excited states of ruthenium
(II) complexes in aqueous media. Solvent isotope effect and
photosensitized generation of singlet oxygen, O2(1Δg), by
[Ru(diimine)(CN)4]2− complex ions, J. Chem. Soc., Dalton
Trans., 2007, 2517–2527.
10 A. A. Abdel-Shafi, P. D. Beer, R. J. Mortimer and
F. Wilkinson, Photosensitized Generation of Singlet Oxygen
from
Vinyl
Linked
Benzo-Crown-Ether-Bipyridyl
Ruthenium(II) Complexes, J. Phys. Chem. A, 2000, 104, 192–
202.
11 F. Wang, S. Min, Y. Han and L. Feng, Visible-light-induced
photocatalytic degradation of methylene blue with polyani-
line-sensitized TiO2 composite photocatalysts, Superlattices
Microstruct., 2010, 48, 170–180.
12 M. Oba, M. Endo, K. Nishiyama, A. Ouchi and W. Ando,
Photosensitized oxygenation of diaryl tellurides to tellurox-
ides and their oxidizing properties, Chem. Commun., 2004,
2, 1672–1673.
Conflicts of interest
There are no conflicts to declare.
13 Y. Okada, M. Oba, A. Arai, K. Tanaka, K. Nishiyama and
W. Ando, Diorganotelluride-catalyzed oxidation of silanes
to silanols under atmospheric oxygen, Inorg. Chem., 2010,
49, 383–385.
Acknowledgements
This work was funded by the American National Science
Foundation, Directorate for Mathematical and Physical 14 M. Oba, K. Tanaka, K. Nishiyama and W. Ando, Aerobic oxi-
Science, Division of Chemistry, award #1800599. Funding was
also provided from the Portland State University Faculty
Development Grant.
dation of thiols to disulfides catalyzed by diaryl tellurides
under photosensitized conditions, J. Org. Chem., 2011, 76,
4173–4177.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
15 M. Oba, Y. Okada, K. Nishiyama and W. Ando, Aerobic
photooxidation of phosphite esters using diorganotelluride
catalysts, Org. Lett., 2009, 11, 1879–1881.
application of a time-dependent density functional theory
polarizable continuum model, J. Chem. Phys., 2006, 124,
094107.
16 M. W. Kryman, G. A. Schamerhorn, J. E. Hill, B. D. Calitree, 26 R. Dennington, T. Keith and J. Millam, GaussView, Version
K. S. Davies, M. K. Linder, T. Y. Ohulchanskyy and
M. R. Detty, Synthesis and Properties of Heavy Chalcogen
5, Semichem Inc., Shawnee Mission, KS Semichem Inc,
2009, DOI: 10.1136/amiajnl-2013-001993.
Analogues of the Texas Reds and Related Rhodamines, 27 A. D. Becke, A new mixing of Hartree-Fock and local-
Organometallics, 2014, 33, 2628–2640.
density-functional theories, J. Chem. Phys., 1993, 98, 1372–
17 M. R. Detty and S. L. Gibson, Tellurapyrylium Dyes as cata-
1377.
lysts for the Conversion of Singlet Oxygen and Water to 28 B. Miehlich, A. Savin, H. Stoll and H. Preuss, Results
Hydrogen Peroxide, J. Am. Chem. Soc., 1990, 112, 4086–
4088.
18 B. Calitree, D. J. Donnelly, J. J. Holt, M. K. Gannon,
obtained with the correlation energy density functionals of
becke and Lee, Yang and Parr, Chem. Phys. Lett., 1989, 157,
200–206.
C. L. Nygren, D. K. Sukumaran, J. Autschbach and 29 C. Lee, W. Yang and R. G. Parr, Development of the Colle-
M. R. Detty, Tellurium Analogues of Rosamine and
Rhodamine Dyes: Synthesis, Structure, 125Te NMR, and
Heteroatom Contributions to Excitation Energies,
Organometallics, 2007, 26, 6248–6257.
Salvetti correlation-energy formula into a functional of the
electron density, Phys. Rev. B: Condens. Matter Mater. Phys.,
1988, 37, 785–789.
30 W. J. Hehre, R. Ditchfield, L. Radom and J. A. Pople,
Molecular orbital theory of the electronic structure of
organic compounds. V. Molecular theory of bond separ-
ation, J. Am. Chem. Soc., 1970, 92, 4796–4801.
19 L. V. Lutkus, H. E. Irving, K. S. Davies, J. E. Hill,
J. E. Lohman, M. W. Eskew, M. R. Detty and
T. M. McCormick, Photocatalytic Aerobic Thiol Oxidation
with a Self-Sensitized Tellurorhodamine Chromophore, 31 R. Ditchfield, W. J. Hehre and J. A. Pople, Self-Consistent
Organometallics, 2017, 36, 2588–2596.
20 K. Mori, M. Tano, T. Mizugaki, K. Ebitani and K. Kaneda,
Efficient heterogeneous oxidation of organosilanes to sila-
Molecular-Orbital Methods. IX. An Extended Gaussian-
Type Basis for Molecular-Orbital Studies of Organic
Molecules, J. Chem. Phys., 1971, 54, 724–728.
nols catalysed by a hydroxyapatite-bound Ru complex in 32 P. C. Hariharan and J. A. Pople, The influence of polariz-
the presence of water and molecular oxygen, New J. Chem.,
2002, 26, 1536–1538.
ation functions on molecular orbital hydrogenation ener-
gies, Theor. Chim. Acta, 1973, 28, 213–222.
21 E. Villemin, E. Gravel, D. V. Jawale, P. Prakash, 33 P. J. Hay and W. R. Wadt, Ab initio effective core potentials
I. N. N. Namboothiri and E. Doris, Polydiacetylene
Nanotubes in Heterogeneous Catalysis: Application to the
Gold-Mediated Oxidation of Silanes, Macromol. Chem.
Phys., 2015, 216, 2398–2403.
for molecular calculations. Potentials for K to Au including
the outermost core orbitals, J. Chem. Phys., 1985, 82, 299–
310.
34 P. J. Hay and W. R. Wadt, Ab initio effective core potentials
for molecular calculations. Potentials for the transition
metal atoms Sc to Hg, J. Chem. Phys., 1985, 82, 270–283.
35 W. R. Wadt and P. J. Hay, Ab initio effective core potentials
for molecular calculations. Potentials for main group
elements Na to Bi, J. Chem. Phys., 1985, 82, 284–298.
22 M. Jeon, J. Han and J. Park, Transformation of Silanes into
Silanols using Water and Recyclable Metal Nanoparticle
Catalysts, ChemCatChem, 2012, 4, 521–524.
23 F. Minisci, M. Francesco, R. Francesco, P. Carlo, G. Chiara,
F. Francesca and P. Gian Franco, A New, Highly Selective,
Free-Radical Aerobic Oxidation of Silanes to Silanols 36 F. L. Hirshfeld, Bonded-atom fragments for describing
Catalysed
Conditions, Synlett, 2002, 1173–1175.
by
N-Hydroxyphthalimide
under
Mild
molecular charge densities, Theor. Chim. Acta, 1977, 44,
129–138.
24 M. W. Kryman, G. A. Schamerhorn, K. Yung, 37 F. De Proft, C. Van Alsenoy, A. Peeters, W. Langenaeker and
B. Sathyamoorthy, D. K. Sukumaran, T. Y. Ohulchanskyy,
J. B. Benedict and M. R. Detty, Organotellurium
Fluorescence Probes for Redox Reactions: 9-Aryl-3,6-diami-
P. Geerlings, Atomic charges, dipole moments, and Fukui
functions using the Hirshfeld partitioning of the electron
density, J. Comput. Chem., 2002, 23, 1198–1209.
notelluroxanthylium Dyes and Their Telluroxides, 38 E. G. Rochow and W. F. Gilliam, Polymeric Methyl Silicon
Organometallics, 2013, 32, 4321–4333. Oxides, J. Am. Chem. Soc., 1941, 63, 798–800.
25 G. Scalmani, M. J. Frisch, B. Mennucci, J. Tomasi, 39 R. O. Sauer, Derivatives of the Methylchlorosilanes. I.
R. Cammi and V. Barone, Geometries and properties of
excited states in the gas phase and in solution: Theory and
Trimethylsilanol and its Simple Ethers, J. Am. Chem. Soc.,
1944, 66, 1707–1710.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.