A visible-light photocatalytic thiolation of aryl, heteroaryl and vinyl iodides

Article abstract of DOI:10.1039/c8ob00238j

The general catalytic synthesis of aryl and vinyl thioethers from readily available halides remains a challenge. Herein we report a unified method for the thiolation of aryl and vinyl iodides with dialkyl disulfides using visible light photoredox catalysis. A range of thioether products bearing diverse functional groups can be accessed in high yield and with excellent chemoselectivity. We demonstrate the versatility of this method through the expedient synthesis of a family of thioether-rich natural products. A detailed investigation of the photocatalytic mechanism is presented from both steady-state and time-resolved luminescent quenching as well as transient absorption spectroscopy experiments.

Full text of DOI:10.1039/c8ob00238j

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
A visible-light photocatalytic thiolation of aryl,
heteroaryl and vinyl iodides†
Cite this: DOI: 10.1039/c8ob00238j
d
M. L. Czyz, a G. K. Weragoda, a R. Monaghan,b,c T. U. Connell,
M. Brzozowski, c A. D. Scully, a J. Burton,a,e D. W. Lupton e and
a,c
A. Polyzos
*
The general catalytic synthesis of aryl and vinyl thioethers from readily available halides remains a chal-
lenge. Herein we report a unified method for the thiolation of aryl and vinyl iodides with dialkyl disulfides
using visible light photoredox catalysis. A range of thioether products bearing diverse functional groups
can be accessed in high yield and with excellent chemoselectivity. We demonstrate the versatility of this
method through the expedient synthesis of a family of thioether-rich natural products. A detailed investi-
gation of the photocatalytic mechanism is presented from both steady-state and time-resolved lumines-
cent quenching as well as transient absorption spectroscopy experiments.
Received 3rd December 2017,
Accepted 31st January 2018
DOI: 10.1039/c8ob00238j
rsc.li/obc
Introduction
The development of catalytic methods to generate C(sp2)–S
bonds remains an important endeavor within the chemical
sciences, due to the ubiquity of the aryl thioether motif in
pharmaceuticals1 (Fig. 1, compounds 1–4) and organic
materials.2 Transition metal catalyzed cross-coupling reactions
dominate approaches to contemporary C(sp2)–S bond
construction.3–5 These reactions typically employ a thiolate anion
coupling partner, generated by treatment of thiols with a strong
base, which often cause the rapid onset of irreversible catalyst de-
activation. This undesired pathway mandates expensive ligands
and high temperatures to circumvent catalyst poisoning, which
in combination with strong bases, narrows functional group
compatibility and generality of such approaches.4,6
The continuing demand for general methods of C(sp2)–S
bond formation that operate under benign reaction conditions
has led to the development of radical-based methodologies.
Radical-mediated C(sp2)–S and C(sp)–S bond formation has
been accomplished recently under visible – light photoredox
conditions.7–9,10,11 These methods proceed via synergistic
photoredox/nickel catalytic cycles, in which in situ generated
Fig. 1 Proposed photoredox-catalyzed thiolation of aryl, heteroaryl
and vinyl iodides; sulfur-containing drugs.
aCSIRO Manufacturing, Research Way, Clayton, VIC 3168, Australia.
E-mail: anastasios.polyzos@unimeb.edu.au
bSchool of Chemistry, The University of Melbourne, Melbourne, Victoria 3010,
Australia
thiyl radicals are intercepted by a nickel complex,9 or following
formation of aryl radicals they react with neutral sulfur nucleo-
philes.7,8,11 In the latter approach, seminal work by Jacobi von
Wangelin and co-workers established the formation of
thioethers from disulfides and aryldiazonium salts using the
cDepartment of Chemistry, Durham University, South Road, Durham, UK
dRMIT University, Melbourne, Victoria 3000, Australia
eSchool of Chemistry, Monash University, Clayton 3800, Victoria, Australia
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8ob00238j
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Optimisation of reaction conditionsa
organic photocatalyst eosin Y.7 Fu and co-workers developed
an elegant method for the arylation of thiophenols using the
photocatalyst fac-Ir(ppy)3.8
A complementary, general method for the formation of aryl
and vinyl alkyl thioethers from a range of C(sp2)–I bonds is yet
to be developed. We anticipated this could be achieved using
disulfides in conjunction with aryl, heteroaryl and vinyl
iodides. Disulfides present an attractive alternative to thiols as
thiolating reagents because they are widely available, stable to
air and relatively non-volatile. Their use is also less explored in
the context of both transition metal-catalyzed5 and radical-
based approaches.7 A photocatalytic method for alkylthiolation
of C(sp2)–I bonds would expand approaches available to syn-
thetic chemists for the synthesis of this important structural
motif.
DMDS
equiv.
Yield 7b
[%]
Yield 8b
[%]
Entry
Solvent
Additive (equiv.)
1
2
3
4
5
6
7
8
DMSO
DMSO
DMF
—
4
4
4
4
4
4
8
2
4
4
4
4
ND
55
30
50
33
40
55
45
ND
45
70
50
67
52
45
52
ND
ND
ND
ND
iPr2EtN (4)
iPr2EtN (4)
iPr2EtN (4)
iPr2EtN (4)
iPr2EtN (2)
iPr2EtN (4)
iPr2EtN (4)
9 (4)
CH3CN
CH3OH
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
9
ND
ND
60
10
11
12
10 (4)
DABCO (4)
DABCO (4) +
DMSO (2)
Results and discussion
Reaction design and optimization
Full (91)c
A mechanistic hypothesis was devised on the basis of known
reduction potentials for aryl iodides (Scheme 1). In the pro-
posed cycle, initiation occurs following irradiation of fac-Ir
(ppy)3 with blue light (λmax = 447 nm) to generate the strongly
reducing triplet excited state [Ir(ppy)3]*,3 (E° [*IrIII/IrIV] = −1.76 V
vs. SCE). This species undergoes a single electron transfer a General conditions: 5 (0.3 mmol), Ir(ppy)3 (2 mol%), solvent (0.1 M
in substrate), 24 h, 10 W blue LED. b Yields determined by 1H NMR
(SET) with iodide I, generating a carbon-centered radical II.
spectroscopic analysis using 2,5-dimethylfuran as an internal stan-
dard. c Isolated yield.
Radical intermediates are susceptible to nucleophilic attack by
disulfides III to generate a trivalent sulfur-based radical
adduct IV.7,12 Previous studies have shown that the adduct IV
is sufficiently reducing to promote electron transfer to the
eosin Y radical cation ([EY•+/EY] = +0.78 V vs. SCE).7 Given the was examined as a model reaction (Table 1, entry 1). The start-
similar reduction potential of [Ir(ppy)3]+ (E° [IrIV/IrIII] = +0.76 V ing material was recovered quantitatively, suggesting that cata-
vs. SCE) we anticipated oxidation of IV to V by [Ir(ppy)3]+ to lyst turnover might be inefficient under the initial redox-
close the catalytic cycle.
neutral conditions. Addition of four equivalents of the reduc-
Studies commenced by examining the feasibility of the tive quencher iPr2EtN resulted in full consumption of the start-
sulfur radical (IV) acting as a single electron donor to the oxi- ing material and 55% yield of the desired product 7. A compet-
dized [Ir(ppy)3]+. The arylation of dimethyl disulfide (DMDS) 6 ing reductive hydrodehalogenation reaction pathway13–15 led to
with methyl 4-iodobenzoate 5 in dimethyl sulfoxide (DMSO) formation of methyl benzoate 8 in 45% yield (Table 1, entry 2).
A solvent screen was performed and acetonitrile found to give
almost identical conversion to the desired product (relative to
DMSO), thus acetonitrile was chosen for further optimisation
due to ease of handling and removal (Table 1, entry 4).
N′,N-Dimethylformamide (DMF) or methanol either promoted
reductive hydrodehalogenation or resulted in incomplete con-
version of the starting material 5 (Table 1, entries 3 and 5). A
i
reduction of Pr2EtN equivalents led to incomplete conversion
of 5 (Table 1, entry 6), while a doubling of dimethyl disulfide
concentration did not increase the yield (Table 1, entry 7).
Alternative reductive quenchers were screened to minimize
reductive hydrodehalogenation. Triarylamines, N,N-bis(4-
methoxyphenyl)aniline 9 (E° = +0.63 V vs. SCE) and 4-methoxy-
N,N-diphenylaniline 10 (E° = +0.76 V vs. SCE) have been
reported to diminish undesired H-atom abstraction pathway.14
i
In this system, substituting Pr2EtN with triarylamines 9 or 10
Scheme 1 A minimalistic model for the proposed Ir(ppy)3-catalyzed
thiolation reaction.
yielded no product despite the matched redox couple of the
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 2 Reaction scope: aryl iodidesa,b
quenchers with Ir(ppy)3 (E° [IrIV/IrIII] = +0.76 V vs. SCE),
(Table 1, entries 9 and 10). The bicyclic base 1,4-diazabicyclo[2.2.2]
octane, DABCO (E° = +0.69 V vs. SCE), has been reported to be
an effective reductive quencher for Ir(III) photocatalyst,
Ir(ppy)2(dtbbpy)PF6.16 It was anticipated that the use of DABCO
would minimize reductive hydrodehalogenation since
α-hydrogen atom transfer from DABCO leads to the formation
of an unfavorable strained bridgehead imine adduct.17
i
Substitution of Pr2EtN with four equivalents of DABCO led to
formation of the desired product in 60% yield with no trace of
the reduced material (Table 1, entry 11). It was also observed
that the reaction mixture was not completely soluble in aceto-
nitrile at ambient temperature. The addition of two equivalents
of DMSO resulted in full conversion of the starting material
with 91% isolated yield of thioether 7 (Table 1, entry 12).
Reaction scope
With optimised conditions in hand, we investigated reaction
scope with respect to substituted aryl iodides. In general, the
arylation of DMDS (6) afforded the corresponding methyl-
thioether products in good to excellent yields (50–91%,
Table 2), with para substituted electron-withdrawing substitu-
ents giving higher yields than electron rich substrates.
Pleasingly, the highly hindered 2,4,6-trimethyliodobenzene
also reacted to provide a good yield of thioether 22. The reac-
tion was compatible with ester, ketone, ether, nitrile and
amide functionalities (products 7, 11–16). Notably, the
method achieves excellent chemoselectivity with exclusive thio-
lation of the iodide in the presence of bromo, chloro and
fluoro groups (products 17–21). It is important to note that the
reaction is compatible with nitrogen, sulfur and oxygen con-
taining heterocycles, highlighting the applicability of this
method to the synthesis of medicinally important structures
(products 25 to 30).
To further assess the versatility of this reaction, we exam-
ined vinyl iodides as potential radical precursors. Reaction
between E-1-(2-iodovinyl)benzene and dimethyl disulfide
under the optimised conditions gave a mixture of the E- and
Z-alkene thioethers 31 in a combined 21% yield. The majority
of the starting material isomerised to the unreactive Z isomer,
even when performed whilst the reaction was exposed to
ambient atmosphere.18 Increasing the temperature to 60 °C
minimised isomerisation of the vinyl iodide substrate, and
increased the yield of the desired thiolated product to 45%
yield and 1.5 : 1 Z/E ratio (Table 3). A mixture of disubstituted
products 32 was also isolated in 43% yield. Despite moderate
yields, this reaction represents one of relatively few viable
a General conditions: Aryl iodide (0.3 mmol), dimethyl disulfide
(4 equiv.), DABCO (4 equiv.), DMSO (2 equiv.), fac-Ir(ppy)3 (2 mol%),
acetonitrile (0.1 M in substrate), 24 h, 10 W blue LEDs. b Isolated yield.
methods for the synthesis of alkyl vinyl sulfides using easily methyl group resulting in isolated excellent yields (products
accessible vinyl halides in a single step.19 Furthermore, for 36–39).
products 31, 33–35, the thermodynamically less stable Z-vinyl
We next examined a diverse set of disulfide nucleophiles
sulfide was accessed from the readily available E-vinyl iodide, under optimized reaction conditions (Table 4). Simple dialkyl
highlighting applicability of this reaction to the synthesis of disulfides were found to be effective coupling partners,
these valuable intermediates in organic synthesis. The reaction although yields decreased as chain length increased, likely due
was more efficient for electron rich vinyl iodides (products 33 to decreased solubility of these disulfides in acetonitrile (pro-
and 34). Formation of the major disubstituted by-product was ducts 40–42). Both ester and Boc-protected amino functional-
minimised when the benzylic position was substituted with a ities on the alkyl chains were tolerated, affording the desired
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 3 Reaction scope: vinyl iodidesa,b,c
Table 4 Reaction scope: disulfidesa,b
a General conditions: Vinyl iodide (0.3 mmol), dimethyl disulfide
(4 equiv.), DABCO (4 equiv.), fac-Ir(ppy)3 (2 mol%), acetonitrile (0.1 M
in substrate), 60 °C, 24 h; 10 W blue LEDs. b Isolated yield. c Z/E ratios
are based on the 1H NMR spectra of the crude reaction mixtures.
a General conditions: Methyl 4-iodobenzoate (0.3 mmol), disulfide
(4 equiv.), DABCO (4 equiv.), DMSO (2 equiv.), fac-Ir(ppy)3 (2 mol%),
acetonitrile (0.1 M in substrate), 24 h, 10 W blue LEDs. b Isolated yield.
products 43 and 44 in acceptable yields. We found the reaction
was not compatible with diphenyl disulfide and sterically con-
gested disulfides including di-tert-butyl disulfide and di-tert-
hexyl disulfide.
possess reduction potentials more negative than −2.0 V vs.
SCE (e.g. 4-iodoanisole, E° = −2.31 V vs. SCE).22 Control experi-
ments (Table S1,† entries 4–5) demonstrated that both the
photocatalyst and light are required for the formation of the
product. The absence of an observable colour change upon
mixing with the exclusion of Ir(ppy)3 and only trace conversion
to product (Table S1,† entry 4) excludes visible-light-promoted
C–S cross coupling via intermolecular charge transfer.23
Furthermore, the strong dependence of the concentration of a
sacrificial reductive quencher on yield (Table S1,† entries 1–3)
implies that single electron transfer from DABCO to the oxi-
dized photocatalyst is likely required to complete the catalytic
cycle and regenerate the ground state Ir(ppy)3. This require-
ment for a sacrificial reductant was not observed in photo-
catalytic thiolation of aryl diazonium salts.7 Consequently,
spectroscopic studies were necessary to further probe the reac-
tion mechanism. Photoluminescent quenching experiments
identified that the aryl iodide (methyl 4-iodobenzoate 5 in
these experiments) exclusively quenched the Ir(ppy)3 photo-
catalyst phosphorescence (Fig. S2, ESI†). The quenching of
phosphorescence from the triplet excited state of Ir(ppy)3,
[Ir(ppy)3]*,3, by 5, as measured by time-resolved luminescence
spectroscopy, displays the linear relationship expected from a
Application to the synthesis of simple floral semiochemicals
To illustrate potential application to the synthesis of bioactive
and pharmaceutically relevant products, we applied the newly
developed methodology to the preparation of (methylthio)
phenols 48–51,
a family of floral semiochemicals from
Caladenia crebra (Australian terrestrial spider orchids)
(Scheme 2).20
Semiochemicals 48–51 were synthesised in excellent yields
in one to three steps through photoredox-catalyzed thiolation
and deprotection sequences. Our method provides an expedi-
ent route to these materials from commercially available aryl
iodides, under mild conditions.
Mechanistic insights
The photoredox catalytic cycle. Studies were undertaken to
examine the reaction mechanism. Consistent with previous
reports,13,21 we propose that the reaction proceeds via electron
transfer from the excited state of Ir(ppy)3 (E° [*IrIII/IrIV] = −1.76
V vs. SCE) to aryl, heteroaryl and vinyl iodides, some of which
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Mechanistic elucidation experiments. (A) Stern–Volmer plot of [Ir
(ppy)3]*,3 (0.01 mM) phosphorescence in the presence of methyl
4-iodobenzoate. Transient absorption difference spectra of Ir(ppy)3
(0.1 mM) in the absence (B) and presence (C) of methyl 4-iodobenzoate
(50 mM) recorded at various delay times after excitation. Solutions were
prepared using acetonitrile and were sparged with argon (6 min) before
measurement; λex = 355 nm; gate times were 0.89 μs (B) and 0.44 μs (C).
Scheme 2 (A) The structures of four (methylthio)phenol semiochem-
icals 48–51 identified from the spider orchid Caladenia crebra (B) total
synthesis of semiochemicals 48–51; reagents and conditions:
(a) dimethyl disulfide (4 equiv.), DABCO (4 equiv.), DMSO (2 equiv.), fac-
Ir(ppy)3 (2 mol%), acetonitrile (0.1 M in substrate), 24 h, 10 W blue LEDs;
(b) A solution of BBr3 in CH2Cl2 (1 M) (10 equiv.), −78 °C to R.T., 2 h;
(c) NaBH4 (1.5 equiv.), MeOH, 14 h.
Stern–Volmer analysis, but the corresponding steady-state exist in the +4 oxidation state, with the transferred electronic
quenching displays upward curvature that was better described charge residing on the cyclometallating ligands. The absorp-
using a quadratic function (Fig. 2A). This non-linearity implies tion spectrum of [Ir(ppy)3]*,3 in acetonitrile solution (Fig. 2B)
contributions from excited-state and ground-state processes to is similar to that reported by Ichimura et al.,26 and displays a
the phosphorescence quenching. The phosphorescence life- number of notable features, including several absorption
time of [Ir(ppy)3]*,3, τP, measured in the present work is bands (λmax ≈ 320, 370, 470 and 700 nm) which are assigned
1.79 μs which is in excellent agreement with previously to ligand-centred transitions associated with the ppy•− radical
reported values.24 The rate constant for the bimolecular phos- anion, based on the close correlation with the reported absorp-
phorescence quenching reaction, kq, was calculated to be 1.8 × tion spectrum of the analogous 2,2-bipyridine (bpy) radical
106 M−1 s−1 by using the value of τP, together with the gradient anion.27 The bands at λmin ≈ 400 nm and 520 nm are ascribed
of the Stern–Volmer plot of the time-resolved phosphorescence to transient bleaching of ground-state Ir(ppy)3 and phosphor-
quenching data. The magnitude of kq is around 4 orders of escence from the triplet excited state respectively. The decay of
magnitude slower than a diffusion-limited process which the transient absorption signals is consistent with the value of
implies a relatively inefficient SET reaction between [Ir(ppy)3]*,3 τP mentioned above, with complete relaxation to the ground
and the aryl iodide 5.
state occurring within approximately 10 μs of excitation. The
Transient absorption spectroscopy was used to examine the transient absorption spectrum differs markedly for Ir(ppy)3 in
nature of the interaction between [Ir(ppy)3]*,3 and the aryl the presence of 5 (Fig. 2C). Consistent with the phosphor-
iodide 5. Given the well-documented metal-to-ligand charge- escence quenching experiments, addition of the aryl iodide
transfer nature of the absorption bands for ground-state results in a dramatic shortening of the lifetime of the triplet
Ir(ppy)3,25 the metal centre in [Ir(ppy)3]*,3 is considered to excited state, coupled with a decrease in phosphorescence
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
intensity. Interestingly, the ground-state bleach observed at To probe this step, the recovery of the transient ground-state
around 400 nm undergoes a slight hypsochromic shift and, bleach signal at 390 nm of an acetonitrile solution comprising
along with the broad absorption band at 700 nm, the signal Ir(ppy)3 and methyl 4-iodobenzoate 5 (0.1 mM and 50 mM,
intensity displays little decay over the first 10 μs after exci- respectively) was measured as a function of DABCO concen-
tation. It was found that neither of these spectral features tration (Fig. 3). Increasing the concentration of DABCO from
decay fully even on a 100 μs timescale (Fig. 3, top panel). The 0 mM to 2.5 mM results in a progressive shortening of the life-
creation of this new longer-lived chemical species arising from time of [Ir(ppy)3]+ which is reflected by the corresponding
the interaction between [Ir(ppy)3]*,3 and the aryl iodide 5 is increase in the recovery rate of the ground-state bleach signal
attributed to the formation of [Ir(ppy)3]+ following a single at 390 nm. This supports the proposed mechanism in which
electron-transfer (SET) event in which [Ir(ppy)3]*,3 is oxidised the initial oxidation of [Ir(ppy)3]*,3 by 5 to produce [Ir(ppy)3]+
by the aryl iodide. The apparent absence of MLCT absorption is followed by an efficient bimolecular SET reduction of
bands in the absorption spectrum of [Ir(ppy)3]+ (Fig. 2C) [Ir(ppy)3]+ by DABCO, resulting in rapid and quantitative re-
implies that the metal remains in the +4 state in this complex, generation of Ir(ppy)3. The lifetime of [Ir(ppy)3]+ (measured
and the loss of the structured absorption bands below 400 nm without quencher as 113 μs) is reduced by approximately 97%
seen in the absorption spectrum of [Ir(ppy)3]*,3 (Fig. 2B) pro- on addition of 2.5 mM DABCO whereas, in contrast, the [Ir
vides evidence that the electron is extracted from a ppy radical (ppy)3]*,3 lifetime is only reduced by 22% in the presence of
anion ligand. The transient absorption spectra of Ir(ppy)3 in 100 mM aryl iodide 5. These results, combined with the rapid
the presence of either DABCO or DMDS (6) are identical to rate of generation of [Ir(ppy)3]*,3 which is unresolvable on the
that of the photocatalyst alone (Fig. S3 and S4 respectively, timescales used in the present work,28 indicates that the initial
ESI†) which is consistent with the absence of phosphorescence SET to form the oxidised [Ir(ppy)3]+ is the rate-limiting step in
quenching, and indicates an absence of interaction between the photoredox catalytic cycle.
[Ir(ppy)3]*,3 and either of these species.
Homolytic cleavage pathway. Once the nature and order of
The next step in the proposed photocatalyst reaction cycle single electron transfer steps in the photocatalytic cycle was
is the reductive turnover of [Ir(ppy)3]+ to Ir(ppy)3 by DABCO. verified, attention was shifted to the mechanism for the for-
mation of aryl sulfides from aryl radicals and disulfides. Given
the critical role of DABCO in the catalytic cycle, we considered
the formation of the proposed trivalent sulfur-based radical
adduct IV (Scheme 1). Nicewicz and co-workers demonstrated
that diaryl disulfides undergo blue light promoted homolytic
cleavage to form thiyl radicals, which are subsequently
reduced to thiolate anions.29
To investigate the homolytic cleavage pathway in our
system, cross-over experiments29 were performed by combining
equimolar solutions of dimethyl disulfide 6 and diethyl di-
sulfide 57 in deuterated acetonitrile (Scheme 3) and irradiating
1
the solution with blue light. Analysis by H NMR spectroscopy
and electrospray ionisation-mass spectrometry (ESI-MS)
revealed no exchange between the two disulfides after
24 hours. The equivalent experiment between diphenyl di-
sulfide 58 and dimethyl disulfide 6 resulted in a statistical
mixture of the three disulfides 6, 58 and 59 after just
Fig. 3 Ground-state bleach recovery profiles (λmeas
= 390 nm) of
Ir(ppy)3 (0.1 mM) and methyl 4-iodobenzoate (50 mM) in acetonitrile at
various DABCO concentrations. Solutions were sparged with argon
(6 min) before measurement; λex = 355 nm.
Scheme 3 Disulfide cross-over experiments between aliphatic and aro-
matic disulfides to investigate the homolytic cleavage pathway.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
30 minutes. These results suggest that dialkyl disulfides in
this reaction do not undergo blue light promoted homolytic
cleavage observed for diaryl disulfides. Absorption spec-
troscopy confirmed the absence of spectral overlap between
dialkyl disulfides and the emission spectrum of the LED lamp
employed in this study (Fig. S5†). These findings are consistent
with the observed unreactivity of diphenyl disulfide 58
(Table 3, product 47 and Scheme 4A). It is likely that formation
of thiyl radicals from diphenyl disulfide (E° [PhS•/PhS−] =
+0.100 V vs. SCE)30 leads to competitive quenching of the
Ir(ppy)3 excited state.29
The formation of the trivalent sulfur radical intermediate
IV is further evidenced by the disulfide by-product 61 identi-
fied by 1H NMR spectroscopy and ESI-MS (35% 1H NMR yield),
generated from the reaction of methyl 4-iodobenzoate 5 and
dimethyl 2,2′-disulfanediyldiacetate 60 (Scheme 4B). These
Scheme 5 Revised reaction mechanism for photoredox thiolation of
results support the absence of the blue light promoted homo- aryl iodides; Path A shows direct homolysis of trivalent sulfur-based
radical; Path B shows oxidation of the radical by DABCO.
lysis pathway and direct nucleophilic attack of the disulfide on
the aryl radical as a key step in the mechanism.
On the basis of the steady-state luminescent quenching and
oxidant for the sulfur radical cannot be excluded (Scheme 5,
Path B). There is an ongoing investigation in our laboratory to
further elucidate the role of DABCO in this system.
transient absorption spectroscopy experiments, and the com-
bined cross over studies, we propose a plausible mechanism
for the C–S bond forming reaction (Scheme 5). The excited
[Ir(ppy)3]* first donates an electron to an iodide I to generate
[Ir(ppy)3]+ and an aryl radical, followed by a SET from DABCO
to return to the ground state. The aryl radical is quenched with
dialkyl disulfide III. The fate of the resulting sulfur radical
adduct IV is unclear. The observation of unimolecular
decomposition of disulfide radical adducts has been described
in the literature31 and this pathway could be accelerated by
blue light, in a fashion similar to the homolytic cleavage of
diaryl disulfides under analogous conditions29 (Scheme 5,
Path A). On the other hand, DABCO can function as an
electron shuttle16,32 and the possibility of it acting as an
Conclusion
A new thiolation method has been developed using visible
light photoredox catalysis. The reaction occurs under mild
conditions and is compatible with a wide range of aryl, hetero-
aryl and vinyl iodides including those with synthetically useful
functional groups and biologically important scaffolds. The
new methodology was applied to the synthesis of a series of
floral semiochemicals from Caladenia crebra (Australian terres-
trial spider orchids) in excellent yields from commercially
available aryl iodides.
Through the application of steady-state luminescent
quenching and transient absorption spectroscopy experiments,
we explicitly demonstrate that the aryl iodide substrate under-
goes a SET process with the triplet excited state of Ir(ppy)3 to
generate an aryl radical and [Ir(ppy)3]+. The addition of
DABCO reduces the lifetime of [Ir(ppy)3]+, consistent with a
second SET event to regenerate ground state Ir(ppy)3.
Furthermore, cross-over experiments provided evidence that
nucleophilic attack of disulfide on the intermediate aryl
radical is the key step of the mechanism and that thiyl radicals
and thiolate anions are not involved in the process. Studies to
further expand this methodology to a broader range of sub-
strates and late-stage functionalization are currently underway
in our laboratory.
Scheme 4 (A) Reaction using diphenyl disulfide as thiolating reagent. Conflicts of interest
(B) Structures of product (43) and by-product (61) from reaction with
bis-methoxycarbonyl disulfane (60).
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
9 M. S. Oderinde, M. Frenette, D. W. Robbins, B. Aquila and
J. W. Johannes, J. Am. Chem. Soc., 2016, 138, 1760–1763;
M. Jouffroy, C. B. Kelly and G. A. Molander, Org. Lett., 2016,
18, 876–879; B. A. Vara, X. Li, S. Berritt, C. R. Walters,
E. J. Petersson and G. A. Molander, Chem. Sci., 2018, 9,
336–344.
Acknowledgements
MLC and GKW gratefully acknowledge CSIRO ResearchPlus
Post-Doctoral Fellowships program. AP acknowledges the
University of Melbourne and CSIRO for the joint
Establishment Grant.
10 X. Wang, G. D. Cuny and T. Noël, Angew. Chem., Int. Ed.,
2013, 52, 7860–7864; J. Santandrea, C. Minozzi, C. Cruché
and S. K. Collins, Angew. Chem., Int. Ed., 2017, 56, 12255–
12259; T. C. Johnson, B. Elbert, A. J. M. Farley,
T. W. Gorman, C. Genicot, B. Lallemand, P. Pasau, J. Flasz,
J. L. Castro, M. MacCoss, D. J. Dixon, R. S. Paton,
C. J. Schofield, M. D. Smith and M. C. Willis, Chem. Sci.,
2018, 9, 629–633.
11 Y. Li, M. Wang and X. Jiang, ACS Catal., 2017, 7, 7587–
7592; C. Bottecchia, M. Rubens, S. B. Gunnoo, V. Hessel,
A. Madder and T. Noël, Angew. Chem., Int. Ed., 2017, 56,
12702–12707.
Notes and references
1 F. Minghao, T. Bingqing, H. L. Steven and J. Xuefeng, Curr.
Top. Med. Chem., 2016, 16, 1200–1216; E. A. Ilardi,
E. Vitaku and J. T. Njardarson, J. Med. Chem., 2014, 57,
2832–2842.
2 D. A. Boyd, Angew. Chem., Int. Ed., 2016, 55, 15486–15502.
3 T. Kondo and T.-a. Mitsudo, Chem. Rev., 2000, 100, 3205–
3220; C.-F. Lee, Y.-C. Liu and S. S. Badsara, Chem. – Asian J.,
2014, 9, 706–722; M. A. Fernández-Rodríguez, Q. Shen and 12 M. Bonifacic and K. D. Asmus, J. Phys. Chem., 1984, 88,
J. F. Hartwig, Chem. – Eur. J., 2006, 12, 7782–7796; 6286–6290.
M. A. Fernández-Rodríguez and J. F. Hartwig, J. Org. Chem., 13 J. D. Nguyen, E. M. D’Amato, J. M. R. Narayanam and
2009, 74, 1663–1672; P. Saravanan and P. Anbarasan, Org.
Lett., 2014, 16, 848–851; L. Cai, J. Cuevas, Y.-Y. Peng and
V. W. Pike, Tetrahedron Lett., 2006, 47, 4449–4452; N. Park,
C. R. J. Stephenson, Nat. Chem., 2012, 4, 854–859; I. Ghosh,
L. Marzo, A. Das, R. Shaikh and B. König, Acc. Chem. Res.,
2016, 49, 1566–1577.
K. Park, M. Jang and S. Lee, J. Org. Chem., 2011, 76, 4371– 14 L. Furst, B. S. Matsuura, J. M. R. Narayanam, J. W. Tucker
4378; M. Murata and S. L. Buchwald, Tetrahedron, 2004, 60, and C. R. J. Stephenson, Org. Lett., 2010, 12, 3104–3107.
7397–7403; M. Sayah and M. G. Organ, Chem. – Eur. J., 15 I. Ghosh and B. König, Angew. Chem., Int. Ed., 2016, 55,
2011, 17, 11719–11722; F. Y. Kwong and S. L. Buchwald, 7676–7679.
Org. Lett., 2002, 4, 3517–3520; Y. Jiang, Y. Qin, S. Xie, 16 J. L. Jeffrey, F. R. Petronijević and D. W. C. MacMillan,
X. Zhang, J. Dong and D. Ma, Org. Lett., 2009, 11, 5250– J. Am. Chem. Soc., 2015, 137, 8404–8407.
5253; F. Ke, Y. Qu, Z. Jiang, Z. Li, D. Wu and X. Zhou, Org. 17 F. S. Fawcett, Chem. Rev., 1950, 47, 219–274; G. Köbrich,
Lett., 2011, 13, 454–457; L. Rout, T. K. Sen and Angew. Chem., Int. Ed. Engl., 1973, 12, 464–473.
T. Punniyamurthy, Angew. Chem., Int. Ed., 2007, 46, 5583– 18 K. Singh, S. J. Staig and J. D. Weaver, J. Am. Chem. Soc.,
5586; H.-J. Xu, Y.-Q. Zhao, T. Feng and Y.-S. Feng, J. Org.
Chem., 2012, 77, 2878–2884; C. Uyeda, Y. Tan, G. C. Fu and
2014, 136, 5275–5278; W. Cai, H. Fan, D. Ding, Y. Zhang
and W. Wang, Chem. Commun., 2017, 53, 12918–12921.
J. C. Peters, J. Am. Chem. Soc., 2013, 135, 9548–9552; 19 A. A. Heredia and A. B. Peñéñory, Eur. J. Org. Chem., 2013,
V. Percec, J.-Y. Bae and D. H. Hill, J. Org. Chem., 1995, 60, 991–997.
6895–6903; Y. Zhang, K. C. Ngeow and J. Y. Ying, Org. Lett., 20 B. Bohman, R. D. Phillips, G. R. Flematti, R. A. Barrow and
2007, 9, 3495–3498; X.-B. Xu, J. Liu, J.-J. Zhang, Y.-W. Wang R. Peakall, Angew. Chem., Int. Ed., 2017, 56, 8455–8458.
and Y. Peng, Org. Lett., 2013, 15, 550–553; L. B. Junquera, 21 Y. Cheng, X. Gu and P. Li, Org. Lett., 2013, 15, 2664–
F. E. Fernández, M. C. Puerta and P. Valerga, Eur. J. Inorg. 2667.
Chem., 2017, 2017, 2547–2556; M. J. Iglesias, A. Prieto and 22 T. V. Magdesieva, I. I. Kukhareva, E. N. Shaposhnikova,
M. C. Nicasio, Adv. Synth. Catal., 2010, 352, 1949–1954;
N. Taniguchi, J. Org. Chem., 2004, 69, 6904–6906.
G. A. Artamkina, I. P. Beletskaya and K. P. Butin,
J. Organomet. Chem., 1996, 526, 51–58.
4 M. A. Fernández-Rodríguez, Q. Shen and J. F. Hartwig, 23 B. Liu, C.-H. Lim and G. M. Miyake, J. Am. Chem. Soc.,
J. Am. Chem. Soc., 2006, 128, 2180–2181; S. V. Ley and
A. W. Thomas, Angew. Chem., Int. Ed., 2003, 42, 5400–5449.
5 Y. Liu, H. Wang, J. Zhang, J.-P. Wan and C. Wen, RSC Adv.,
2014, 4, 19472–19475.
6 J. Louie and J. F. Hartwig, J. Am. Chem. Soc., 1995, 117,
11598–11599; A. Correa, M. Carril and C. Bolm, Angew.
Chem., Int. Ed., 2008, 47, 2880–2883.
2017, 139, 13616–13619.
24 S. Takayasu, T. Suzuki and K. Shinozaki, J. Phys. Chem. B,
2013, 117, 9449–9456.
25 M. G. Colombo, T. C. Brunold, T. Riedener, H. U. Guedel,
M. Fortsch and H.-B. Buergi, Inorg. Chem., 1994, 33, 545–
550.
26 K. Ichimura, T. Kobayashi, K. A. King and R. J. Watts,
J. Phys. Chem., 1987, 91, 6104–6106.
7 M. Majek and A. J. von Wangelin, Chem. Commun., 2013,
49, 5507–5509.
27 C. Creutz, Comments Inorg. Chem., 1982, 1, 293–311.
8 M. Jiang, H. Li, H. Yang and H. Fu, Angew. Chem., Int. Ed., 28 K.-C. Tang, K. L. Liu and I. C. Chen, Chem. Phys. Lett.,
2017, 56, 874–879.
2004, 386, 437–441; G. J. Hedley, A. Ruseckas and
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
I. D. W. Samuel, Chem. Phys. Lett., 2008, 450, 292– 31 F. Yin, D. Grosjean, R. C. Flagan and J. H. Seinfeld,
296. J. Atmos. Chem., 1990, 11, 365–399.
29 N. A. Romero and D. A. Nicewicz, J. Am. Chem. Soc., 2014, 32 M. H. Shaw, J. Twilton and D. W. C. MacMillan, J. Org.
136, 17024–17035.
30 A. G. Larsen, A. H. Holm, M. Roberson and K. Daasbjerg,
J. Am. Chem. Soc., 2001, 123, 1723–1729.
Chem., 2016, 81, 6898–6926; K. Xu, Z. Tan, H. Zhang, J. Liu,
S. Zhang and Z. Wang, Chem. Commun., 2017, 53, 10719–
10722.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.