Photoreduction of Thioether Gold(III) Complexes: Mechanistic Insight and Homogeneous Catalysis

Article abstract of DOI:10.1002/chem.201804322

Complexes formed between AuCl₃ and thioether ligands underwent a photoinduced reductive elimination under homogeneous conditions in dichloromethane and toluene solutions to afford the corresponding Au^I complexes. All the gold(III) complexes were rapidly reduced to the gold(I) chloride complexes under 365 nm irradiation or ambient light while being thermally stable below 55 °C. The mechanism of photoreduction through Cl₂ elimination is discussed based on a kinetic study and the chemical trapping of chlorine species: Cl₂, radical Cl^., and possibly Cl⁺. The catalytic activities of the gold(III) chloride complexes and the corresponding gold(I) complexes obtained by in situ reduction were evaluated in the cyclization of N-propargylic amides to oxazoles. The merits of such photoreducible complexes in homogeneous gold catalysis are illustrated by a cascade reaction catalyzed by thioether gold complexes that affords a 4H-quinolizin-4-one in high yields.

Full text of DOI:10.1002/chem.201804322

A Journal of
Accepted Article
Title: Photoreduction of Thioether Gold(III) Complexes: Mechanistic
Insight and Homogeneous Catalysis
Authors: Zhen Cao, Dario M. Bassani, and Brigitte Bibal
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201804322
Link to VoR: http://dx.doi.org/10.1002/chem.201804322
Supported by

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
Photoreduction of Thioether Gold(III) Complexes: Mechanistic
Insight and Homogeneous Catalysis
Zhen Cao,[a] Dario M. Bassani,[a] and Brigitte Bibal*[a]
Abstract: Complexes formed between AuCl3 and thioether ligands
undergo a photo-induced reductive elimination under homogeneous
conditions in dichloromethane and toluene solutions to afford the
corresponding Au(I) complexes. All gold(III) complexes were rapidly
reduced to the gold(I) chloride complexes under a 365-nm irradiation
or ambient light while being thermally stable below 55°C. The
mechanism of photoreduction through Cl2 elimination is discussed
based on a kinetic study and the chemical traping of chlorine
species: Cl2, radical Cl• and possibly Cl+. The catalytic activity of
gold(III) chloride complexes and the corresponding gold(I) ones
obtained by in situ reduction was evaluated in the cyclization of
N-propargylic amides to oxazoles. The merits of such photoreducible
complexes in homogeneous gold catalysis are illustrated by a
cascade reaction catalyzed by thioether gold complexes affording a
4H-quinolizin-4-one in high yields.
Recently, interest in the reduction of gold(III) complexes has
increased following the report of two new catalytic pathways: (i)
the known reductive elimination of organogold(III) complexes
elegantly exploited in advanced transformations[7] and (ii) the
redox Au(I)/Au(III) catalytic cycle implemented in the presence of
a
sacrificial oxidant[8]. The latter redox cycle was also
successfully achieved in several radical reactions with aryl
diazonium, diaryliodonium salts or bromoalcanes using gold
catalysis in the presence of photosensitizers[9] or by simply using
light to directly activate the gold species[10]
.
Concerning the reductive elimination from AuCl3 complexes, the
process seems to depend on the nature of the ligands (Scheme
1). Mono- and di- phosphine ligands enabled
a
slow
photoreduction of AuCl3 into AuCl in the presence of alkenes, as
a chemical trap.[11] NHC ligands were reported to favor rapid
photoreduction of AuBr3 into AuBr in methanol under UV light
(280 nm).[12] It was proposed that these photoreductions proceed
through a ligand-to-metal charge transfer (LMCT) state that
results in the direct elimination of X2.
Introduction
Since 1986, homogeneous gold(I) catalysis has played
a
considerable role in the development of synthetic
transformations and the construction of complex molecular
architectures.[1],[2] In contrast, the use of gold(III) catalysts is less
common, despite the description of some remarkable examples
using AuCl3.[3] Three main limitations still exist for the
exploitation of gold(III) complexes in catalysis: (i) the preparation
of gold(III) complexes from AuCl3 is facile when using pyridine
ligands[4] but less convenient with phosphine and N-heterocyclic
carbene (NHC) ligands requiring the formation of the
corresponding Au(I) complexes followed by an oxidation step, (ii)
the formation of cationic Au(III) species by chloride exchange is
not controlled,[5] and (iii) the reduction of AuCl3 species into the
corresponding AuCl complexes is scarcely documented in
organic solvents despite their importance in catalytic processes
(vide infra). Currently, the rationalization of the properties of new
gold catalysts distinguishes between their oxidation state (AuIII or
AuI) and the nature of their substituents on the gold center, with
the notion of precatalysts (neutral gold chloride) and active
species (cationic gold).[6]
Scheme 1. Photoreduction of gold(III) halide complexes depending on ligand:
(A) Mono- or di- phosphine, (B) N-heterocyclic carbenes, and (C)
dialkylthioether.
[a]
Z. Cao, Dr D. M. Bassani, Dr B. Bibal
Université de Bordeaux
Another plausible mechanism for X2 elimination during gold(III)
reduction was proposed earlier by Kochi[7a–b] for dialkyl- and
trialkyl- gold(III) phosphine complexes and later, for NHC[13]
complexes of AuX2R (R: Me, Ar). It consists in the initial loss of
Institut des Sciences Moléculaires, UMR CNRS 5255
351 cours de la Libération, 33405 Talence, France
E-mail: brigitte.bibal@u-bordeaux.fr
Supporting information for this article is given via a link at the end of
the document.
an halide substituent from the gold atom to generate
a
three-coordinate intermediate that is more prone to undergo a
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
Figure 1. Dialkylthioether ligands L1–L4 designed for AuCl3 and AuCl
coordination in dichloromethane or toluene: L1–L2 absorb visible light
absorption whereas L3–L4 are transparent to 365–600 nm irradiation.
reductive elimination than the former square-planar saturated
gold(III) center. Of note, the catalytic properties of these gold(I)
complex obtained by photoreduction were not reported.
We previously described
a 9,10-diphenylanthracene (DPA)
derivative L1 (Fig. 1) with two thioether groups as coordinating
sites that allowed facile preparation of a AuCl3 complex, through
liquid-liquid extraction.[14] The corresponding gold complex is
The corresponding gold(III) chloride complexes underwent
photo- and thermal reductions to generate Au(I) chloride
complexes. The mechanism of these reductions was
investigated to define the impact of the ligand, i.e. its nature and
its number of coordination sites. Interestingly, the observation of
chlorinated side-products arising from the trapping of different
halogen species provides indirect support for the heterolytic
cleavage of the Au–Cl bond. Finally, the catalytic properties of
these AuCl3 and in situ generated AuCl complexes were
evaluated under homogeneous conditions in both a model
cyclization reaction and in new one-pot cascade reaction for the
construction of a fused polyaromatic compound.
directly
soluble
in
hydrophobic
solvents
such
as
dichloromethane and toluene, thereby avoiding the two-step
procedure typically used to prepare AuCl3 complexes with
ligands other than pyridine.[4] Importantly, the liquid-liquid
extraction procedure guarantees that the L1.2AuCl3 complex is
confined to the organic phase and that it is not exposed to
additional reduction processes as observed in other reported
protocoles. For example, the preparation of (tht)AuCl or
(Me2S)AuCl is based on the in situ redox reaction between the
neat organic ligand (tht or Me2S) and an aqueous solution of
HAuCl4.[15]
The photoreduction of L1.2AuCl3 complex into L1.2AuCl in
dichloromethane or toluene at 365 nm is fast and can lead to the
formation of gold nanocrystals in the presence of water.
However, in anhydrous media, this interesting example of a
photoreducible thioether complex of AuCl3 remains under-
exploited with respect to its catalytic properties at either
oxidation state (+III) or (+I) and the mechanism of the
photoreduction still requires additional clarifications. We
envisioned that not only L1, but other dialkylthioether
compounds would be useful ligands for AuCl3 with the ensuing
complexes reduced in situ into Au(I) complexes in a controlled
fashion. Both gold species can thus contribute to homogeneous
Results and Discussion
Complexes between thioethers L1–L4 and AuCl3
New ligands L2–L4 were prepared through straigthforward
routes in high yields (see supporting information, SI) and the
synthesis of L1 was achieved according to our previous
procedure.[14] The complexes between thioethers Ln (n = 1–4)
and AuCl3 were prepared by liquid-liquid extraction in the dark
by dissolving the desired ligand in the organic phase and
shaking with an aqueous solution of KAuCl4. This procedure
takes advantage of the solubility of the ligands and gold
complexes in low polarity solvents such as dichloromethane or
toluene while free Au(III) is confined to the aqueous phase.
Another benefit of this protocol is that the solution of gold
complex can be directly prepared in common solvents for the
further use in organic synthesis.
gold catalysis. Herein, we report
a series of lipophilic
dialkylthioether ligands for AuCl3 coordination, appended with
(L1 – L2) or without (L3 – L4) a chromophore in order to
elucidate the exact role of the photosensitizing anthracene
moiety in L1. Ligands L3 – L4 were chosen as reference
thioether ligands that do not absorb light between 365 and 600
nm.
Figure 2. UV-Visible spectra of AuCl3 complexes (c = 60 µM, toluene) and the
gold(III) contribution at 330 nm.[9]
For each complex, the stoichiometry in solution was determined
by UV–Visible spectroscopy by monitoring the absorbance at
330 nm due mainly to the contribution of gold(III) (Figure 2).
Despite a small absorption by ligands L1 and L2, the spectra
unambiguously indicated a (Ligand: AuCl3) stoichiometry of 1 : 2
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
for L1 and L3, whereas a 1 : 1 value was found for L2 and L4.
Each dialkylthioether group therefore contributes to the
coordination of one AuCl3 species.
photoreduction versus 365-nm irradiation (77–80% conv. vs.
93-95% respectively, see Table 3, entry 2). Accurate analysis of
cyclization products further indicated that gold(I) complexes
obtained through daylight exposure lead to side-products in
higher proportion than those obtained using catalysts prepared
by 365-nm irradiation (Table S3).
All gold(III) complexes were characterized by NMR (CDCl3) and
mass spectroscopy. Interestingly, their 1H NMR spectra featured
a series of distinct signals for the aliphatic protons (CH2) near
the thioether group due to metal coordination of the sulfur atom
(Figure 3). Concerning the mass spectra, complex L1•2AuCl3
and complex L2•AuCl3 were clearly identified by FD technique,
along with reduced complexes containing AuCl, obtained
through an in situ reductive elimination of Cl2. For the complexes
without a DPA chromophore i.e. L3•2AuCl3 and L4•AuCl3, their
mass spectra showed several reduced species such as
[L3.AuCl2]+ and [L4.2Au]+, indicating a lower stability under
different analysis conditions (Figures S3–4). The occurrence of
reduced species and unusual ionic complexes reflects the lability
of these gold complexes under mass spectrometry operating
conditions.
Finally, the strong fluorescence emission of 9,10-
diphenylanthracene is quenched in L1•2AuCl3 and L2•AuCl3
complexes due to an intramolecular energy transfer from the
polyaromatic ligand to the gold(III) atom, as previously
evidenced.[14] The classical emission of anthracene is recovered
after photoreduction to the corresponding L1•2AuCl and
L2•AuCl complexes, as Au(I) species do not absorb at 260-600
nm (Figures S18, S20).
Figure 3. 1H NMR spectra (CDCl3) of L1•2AuCl3 before (up) and after (down)
photoreduction towards the corresponding L1•2AuCl complex, showing shifted
thioether protons due to the change in oxidation state of gold chloride.
Photoreduction towards AuCl complexes
Thermal reduction
Ln•xAuCl gold(I) complexes were obtained by irradiating the
corresponding solutions of Ln•xAuCl3 (4.8 µmol, 6.0 mM) in
dichloromethane (0.8 mL) for 1h using a 365-nm UV lamp (6W).
The photoreduction was monitored by UV–Visible spectroscopy
(Figure 4A). The absorbance at 330 nm decreases due to the
transformation of the more highly absorbing gold(III) chloride into
the weakly absorbing gold(I) chloride. This process is fast (about
30 min) and does not require any additive such as alkenes
which were previously employed to achieve the photoreduction
of gold complexes with phosphines[11]. A pseudo-first-order
kinetics was observed over the first 7–10 minutes of the process
(Table 1), which is consistent with the photoreductive elimination
mechanism proposed in the literature (see vide infra).[11],[12]
Following photoreduction, all Ln•xAuCl complexes (n = 1 – 4, x
= 1 – 2) were characterized by proton NMR which revealed
upfield shifts for protons next to the sulfur atom, when compared
to the corresponding AuCl3 complexes (Figure 3). Mass
spectroscopy indicated that the expected reduced gold(I)
complexes were formed (Figures S11–14). As observed for
gold(III) complexes, gold(I) complexes with thioether ligands
appeared to be labile under mass spectrometry operating
conditions.
In the dark, the L1•2AuCl3 complex is thermally stable up to
55°C (See Figure S21). To further investigate the impact of
temperature, a solution of L1•2AuCl3 in toluene was monitored
by UV–Visible spectroscopy at 80 °C in the dark over several
hours (Figure 4B).
The decrease in gold(III) absorption at 330 nm was similar to
that observed during photoreduction (Figure 4A), but much
slower (12 hours vs. 30 min). Therefore, it can be concluded that
background thermal reduction is negligible under irradiation
conditions.
The 1H NMR spectra of crude product(s) after the thermal
treatment presented the signature of the expected gold(I)
complexes and several signals that could be attributed to
degradation of the aliphatic chains belonging to the ligands (Fig.
S23–24). Notably, the catalytic properties of L1•2AuCl and
L2•AuCl complexes obtained by thermal reduction were far
lower than those of the same complexes prepared by
photoreduction (40–67% conv. versus 93–95%, see Table 3,
entry 2). For the sake of preparation and effectiveness, the
following studies were focused on gold(I) complexes obtained by
photoreduction.
Remarkably, all solution of Ln•xAuCl3 complexes are stable in
the dark but sensitive to ambient visible light. We found that
solutions of AuCl3 complex exposed to daylight inside the
laboratory were prone to be photoreduced to Au(I) at rates
similar to samples exposed to the 365-nm irradiation (Table 1).
However, a closer inspection revealed that the catalytic activity
was somewhat lower for gold(I) complexes obtained by daylight
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
A)
suggested that reductive elimination in the case of binuclear
gold complexes might be facilitated through the formation of
short-lived halide-bridged intermediates that weaken the Au–Cl
bond.
Table 1. Pseudo-first-order rate constants for photoreduction of gold(III)
complexes under 365 nm irradiation and daylight.[a]
Gold(III) Complexes
L1•2AuCl3
k365 (min-1)[b]
0.0802
kdaylight (min-1)[c]
0.0733
B)
L2•AuCl3
0.0571
0.0539
L3•2AuCl3
0.0836
0.079
L4•AuCl3
0.0644
0.0673
[a] All rate constants were determined for solutions of gold(III) complex (60
µM) in toluene at room temperature. [b] Irradiation under a xenon-mercury
lamp at 365 nm. [c] Indoor daylight in a sunny day.
To explore the elimination of dichlorine, the photoreduction of
each Ln•xAuCl3 complex (ca. 20–30 mM in CH2Cl2) was
conducted in the presence of cyclohexene (380 mM) as a
chemical trap (Table 2). The crude mixture was analyzed by
GC–MS, which revealed the presence of chlorocyclohexane and
1,2-dichlorocyclohexane, as well as 2-chlorocyclohexanol (see
Figures S25–S28). This result suggests that Cl2 and Cl• were
released following the photoexcitation of the gold(III) complex
(Figure 6). In the presence of alkene, the excitation of the LMCT
band was shown to induce the ejection of Cl• to give an Au(II)
intermediate that was further reduced.[16] On the contrary, the in
Figure 4. Reductions of L1•2AuCl3 in toluene monitored by UV-Visible
spectroscopy: (A) Photoreduction (c = 60 µM) using a TLC lamp at 365 nm;
(B) Thermal reduction (c = 70 µM) at 80°C.
Mechanistic aspect of AuCl3 photoreduction:
Previously, we showed that an intramolecular energy transfer
occurs within the L1•2AuCl3 complex upon excitation of the DPA
chromophore at 365 – 400 nm. We proposed that this energy
transfer might accelerate the rapid and clean reduction into
L1•2AuCl.[14] To better understand the photoreduction
mechanism of gold(III) thioether complexes, we further
investigated the role of the diphenylanthracene chromophore,
taking into account the plausible reductive elimination
pathway.[7a–b],[11–13]
The kinetics of reduction were investigated under irradiation at
365 nm or ambient daylight by monitoring the absorption of the
gold(III) species. For each Ln•xAuCl3 (n = 1–4, x = 1–2) complex,
the decrease in absorbance was fitted to a pseudo-first-order
reaction over the first ten minutes of irradiation (Table 1), with
similar rates observed upon irradiation at 365 nm (k365 = ca.
0.06–0.08 min–1) and under ambient daylight (kdaylight = ca. 0.05–
0.08 min–1). The nature of the dialkylthioether ligand, i.e. the
presence of diphenylanthracene, aryl or alkyl, substituents, does
not impact the rate of the photoinduced process. This is in
agreement with our previous finding[14] of near-quantitative
energy transfer from the excited DPA to the gold(III) center since,
under these experimental conditions, all incident light is
absorbed either by the DPA or by the gold(III).
situ dichlorine formation is characteristic of
a non–radical
reductive elimination mechanism in agreement with Kochi's
proposed mechanism or a direct cis-elimination process.[11],[13b]
Therefore, both radical and non–radical photoreduction
pathways appear to be operating during the reduction of gold(III)
chloride thioether complexes. Surprisingly, the GC-MS analysis
of the crude photoreducted gold(III) complexes indicated that 2-
chlorocyclohexanol was also formed. This compound may be
generated from cyclohexene and electrophilic Cl+ in the
presence of residual water. To the best of our knowledge, this is
the first time that the cleavage of AuIII–Cl bond into –AuIII and Cl+,
was indirectly observed through a chemical trap.
Table 2. Outcome of chlorinated species during photoreduction of gold(III)
complexes in toluene under 365-nm irradiation (UV lamp).[a]
Conditions
GC-MS Analysis of products[c]
Ln•xAuCl3 +
chlorocyclohexene, 1,2-dichlorocyclohexene, 2-
chlorocyclohexanol
cyclohexene[b]
Ln•xAuCl3
benzaldehyde, benzyl chloride, o- and p- chlorotoluene
Comparing complexes with one or two gold atom(s), the
photoreduction of L1•2AuCl3 and L3•2AuCl3 were found to be
faster than those of L2•AuCl3 and L4•AuCl3 complexes, as
observed for phosphine-based complexes[11]. In the latter, it was
[a] All Ln•xAuCl3 complexes (20-30 µM, toluene; n = 1–4, x = 1–2)
revealed similar GC-MS analysis after 4h of irradiation at room
temperature. [b] cyclohexene concentration was 300 µM. [c] for detailed
analysis, see S.I.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
In the absence of cyclohexene, the analysis of the crude
irradiated gold(III) chloride complexes revealed the presence of
benzaldehyde and species resulting from the chlorination of
toluene, i.e. benzyl chloride and ortho- and para- chlorotoluene
(Table 2). These products could be obtained from reaction of
toluene: oxidation by Cl2 (PhCHO), radical substitution by radical
Cl• (PhCH2Cl) and electrophilic substitution by Cl+ (Cl-C6H4-CH3).
The formation of the chloronium ion is unexpected, although
supported by the formation of both 2-chlorocyclohexanol and
o/p-chlorotoluene. It may arise from an ionic pathway that is
reminiscent of the mechanism initially proposed by Kochi (Figure
6). This ionic pathway would imply the formation of a chloronium
cation and a tricoordinated gold(III) [Ln•xAuCl2]– species. The
reactive Cl+ reacts with toluene to form o- and p- chlorotoluene.
If the chloronium is not captured by toluene, then [Ln•xAuCl2]–
can undergo a reductive elimination of Cl2 followed by a
recombination with Cl+ to lead to the expected Ln•xAuCl
complex.
the product of 5-exo-dig cyclization i.e alkylidene oxazoline
(compounds 5b – 7b) was observed. Our study was focused on
three different propargylic amides appended with aliphatic (5),
phenyl (6) and electron-rich thiophene (7) substituents (Table 3).
The catalytic activities of Ln•xAuCl3 (termed gold(III) chloride),
the corresponding Ln•xAuCl (gold(I) chloride) obtained by in situ
photoreduction at 365 nm and the cationic Ln•xAuNTf2
complexes (obtained from the latter Ln•xAuCl in the presence of
AgNTf2) were compared to PPh3AuNTf2, an efficient gold(I)
phosphine complex for homogeneous catalysis.
Amides 5–7 were converted to the aromatic 2,5-disubstituted
oxazoles 5a–7a using either the gold(III) chloride catalysts (85–
99% conv.), or the gold(I) chloride catalysts (80–99% conv.) in a
similar fashion as AuCl3. To the best of our knowledge, this is
the first report that complexes of AuCl can also catalyze this
transformation. Slightly lower conversions for amide 5 might be
result from steric hindrance between the catalyst and the
reactant.
The use of toluene as solvent is thus beneficial for the protection
of the catalyst during its photoreduction as it captures any
reactive halogenated species that are generated. Indeed, when
dichloromethane was employed as a solvent for photoreductions,
some degradation of dialkylthioether ligands was noticed,
presumedly due to their reactivity towards Cl• and/or Cl+.
Gold(III) or Gold(I)
R
R
H
O
Cationic gold(I)
(2 mol%)
O
Chloride(s)
R
N
N
N
O
(2 mol%)
5b/6b/7b
R = tBu
R = phenyl
R = thiophene
5
5a/6a/7a
6
7
Table 3. Catalytic activity of gold(III) chlorides and gold(I) catalysts obtained
by photoreduction.[a],[b]
Conversion,[c]
Substrates
Entry
Catalyst
Product
1
2
Gold(III) chloride
Gold(I) chloride
85–90 %, 5a
80–92 %, 5a
77–80 %, 5a
40–67 %, 5a
Gold(I) chloride[d] (daylight)
Gold(I) chloride[e]
(thermal reduction)
[f]
3
4
Gold(I) chloride/AgNTf2
85–93 %, 5b
95 %, 5b
[g]
PPh3AuNTf2
Figure 6. Plausible pathways for the photoreduction of complexes between
dialkylthioether and AuCl3 in solution: a radical-based mecanism induced by
excitation of the LMCT band (release of Cl•), a direct reductive elimination
through a concerted mechanism (release of Cl2) and two ionic pathways
through transient tricoordinated Au(III) ions (negatively or positively charged)
that undergo a reductive elimination. Each reactive chlorinated species (Cl•,
Cl2, Cl+) can be trapped by cyclohexene or toluene (solvent) molecules.
5
6
Gold(III) chloride
Gold(I) chloride
96–99 %, 6a
96–99 %, 6a
91–96 %, 6b
94 %, 6b
[f]
7
Gold(I) chloride/AgNTf2
8[
PPh3AuNTf2
[g]
Gold catalysis
9
Gold(III) chloride
Gold(I) chloride
95–99 %, 7a
90–99 %, 7a
79–95 %, 7b
96 %, 7b
The catalytic properties of thioether-based complexes of AuCl3
and AuCl (obtained by in situ photoreduction) were exploited in
two model reactions.
In a first example, the cyclization of N-propargylic amides
developed by Hashmi was chosen as a model reaction.[17] Two
different isomers were prepared depending on the nature of the
catalyst. An oxazole (compounds 5a – 7a) was isolated in the
presence of AuCl3, due to a a 5-exo-dig cyclization followed by
an isomerization of the allyl group towards the aromatic
compound. Meanwhile, in the presence of cationic Au(I) catalyst,
10
11
12
[f]
Gold(I) chloride/AgNTf2
[g]
PPh3AuNTf2
[a] Unless specified, all the reactions were carried out in the dark an NMR tube
(CDCl3) without stirring, using an amide substrate (0.11 mmol), Ln.xAuCl3 or
Ln.xAuCl (formed by 365 nm irradiation of the precursor using a TLC lamp) as
a catalyst (n = 1–4, x = 1–2; 2 mol% loading), at room temperature in the dark.
Reactions were monitored by 1H NMR for 24-30h until full conversion was
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
reached. [b] For detailed experiments depending on the ligand nature, see
supporting information, Table S2. [c] Determined by 1H NMR. [d] Obtained
through photoirradiation by ambient daylight. [e] Obtained by thermal
reduction. [f] Prepared by mixing gold(I) chloride complex and AgNTf2 for 1h.
[g] reaction performed in a 10 mL vial with stirring.
by the gold(III) chloride catalyst whereas the disubstituted alkyne
reacts in the presence of cationic gold(I). Step 1 is a 6-exo-dig
cyclization followed by isomerization leading to the new β–
carbolinone 9. A similar electrophilic activation of monoalkynes
by gold(III) was described by A. Padwa using AuCl3.[18] Step 2 is
a novel 6-endo-dig cyclization towards 4H-quinolizin-4-one 10,
As expected, dihydrooxazoles 5b–7b were obtained by
employing cationic Ln•xAuNTf2 complexes (79–96% conv.) or
commercially available PPh3AuNTf2 (94–96% conv.)[17c] through
a 5-exo-dig cyclization without further isomerization. The model
cyclization of N-propargylic amides can thus be readily
controlled by employing Ln•xAuCl3 and Ln•xAuCl gold chloride
catalysts or cationic gold(I) catalyst, Ln•xAuNTf2.
inspired by the synthesis of
a
related regioisomer[19]
.
Straightforward routes to 4H-quinolizin-4-ones are rare and
mainly based on Pd or Rh catalysis.[20] The development of new
synthetic pathways is thus of interest as these π-extended N-
heteroarenes exhibited biological activities with potential
applications against Alzheimer's disease, Type 2 diabetes and
as antibiotics.[21] Herein, we propose a cascade strategy towards
model substrate 10 that can be attained using gold catalysts at
different oxidation states: either Au(III) or Au(I) that is obtained
from photoreduction of the corresponding Au(III) complex.
In
a
second example of the catalytic properties of
dialkylthioether gold complexes, a more sophisticated one-pot
reaction was developed to transform amide 8 with two alkyne
groups into indolino-4H-benzoquinolizin-4-one 10. In principle,
the two sequential chemioselective cyclizations are catalyzed by
different gold species: the monosubstituted alkyne is activated
Table 4. One-pot cascade cyclization of a propargylindole 8 towards fused polyheteroaromatic compound 10 catalyzed by gold(III) or gold(I) catalysts.
O
N
Step 1
Step 2
N
N
N
6-endo-dig
6-exo-dig
N
N
O
O
8
9
10
Step 2
Irradiation (365 nm)[c]
Step 1
Entry[a]
Catalyst
Yield, Product[d]
Time
16 h
Conversion (9)[b]
Additive
–
[e]
1
2
AuCl3
–
–
–
trace
L1•2AuCl3
L2•AuCl3
L3•2AuCl3
L4•AuCl3
L1•2AuCl[f]
L2•AuCl[f]
AgOTf
5 h (48 h)
12 h
99 % (99 %)
99 %
99 %
99 %
99 %
99 %
99%
–
93 %, 9
90 %, 9
95 %, 9
92 %, 9
80 %, 10
71 %, 10
n.d.[g], 9
76 %, 10
72 %, 10
83 %, 10
72 %, 10
80 %, 10
73 %, 10
4
–
–
5
12 h
–
–
6
12 h
–
–
7
16 h
–
AgOTf
AgOTf
–
8
16 h
–
9
16h
–
10
11
12
13
14
15
L1•2AuCl3
L2•AuCl3
L1•2AuCl3
L2•AuCl3
L3•2AuCl3
L4•AuCl3
16 h
99 %
99 %
99 %
99 %
99 %
99 %
–
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
16 h
–
12 h
Yes
Yes
Yes
12 h
12 h
12 h
Yes
[a] Unless specified, all the reactions were performed in dry dichloromethane at ambient temperature and in the dark, in the presence of a catalyst (2 mol%),
amide 8 (0.1 mmol) and when necessary, AgOTf (6 mol%) was added after the completion of step 1 (except entry 9). [b] monitored by TLC. [c] TLC lamp (6W)
irradiation for 2h. [d] Isolated yield. [e] Compound 9 was not detected by TLC. [f] Freshly prepared through irradiation of the corresponding gold(III) complex at
365 nm . [g] n.d.: not determined.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
Indole-based compound 8 was synthesized in four steps with a
37 % overall yield (see SI). This model substrate possesses
both an N-propargyl amide and an N-diphenylacetylene group
to successively allow the formation of a pyridone (step 1) and a
quinolizin-4-one ring (step 2).
cyclization of disubstituted alkynes, with a similar performance
to that of cationic Au(I) catalysts.
Conclusion
Initially, the reaction was tested using AuCl3 as catalyst (Table
4, entry 1). Several products were observed by TLC but the
expected cyclization product was not detected. The smooth
transformation of amide 8 to 9 (step 1) was seen when
thioether gold(III) chlorides Ln•xAuCl3 (n = 1–4, Table 4, entries
2–6, 90–95% yield) or the corresponding gold(I) catalyst
Ln•xAuCl obtained by photoreduction at 365 nm (n = 1–2,
entries 7–8, 99% conversion) were used as catalysts.
Notably, indole 9 can be obtained with 99% conversion within
5h in the presence of L1•2AuCl3, and this product was isolated
in 93 % yield (Table 4, entry 2). As noticed for the catalyzed
cyclization of propargylic amides, the synthesis of indole-
pyridone 9 is catalyzed by thioether complexes with gold(III)
chloride or gold(I) chloride. To catalyze the second cyclization
and obtain the desired quinolizin-4-one 10 (step 2), the higher
activity of a cationic gold species is required. The classical
cationic Au(I) species obtained from Ln•xAuCl complexes in the
presence of silver salts were explored first (Table 4, entries 7–
8).[22] The sequential cyclization product 10 is obtained in
excellent yield (71–80 %, entries 7–8) by tuning the Au(I)
catalysts over the two sequential steps: Au(I) chloride complex
was used in the first step (99% conv.) whereas, in the second
step, the corresponding cationic Au(I) complex was in situ
generated by the addition of AgOTf in the reaction medium.
Along step 2, the role of silver triflate was shown to be
restricted to chloride exchange, as no catalytic activity was
detected when using the silver salt alone (entry 9).
Homogeneous
gold(III)
complexes
with
well-defined
stoichiometry are readily obtained by liquid-liquid extraction
using lipophilic dialkylthioether ligands and the resulting
complexes are stable in the dark below 55°C. Independently of
the ligand nature (anthracene, phenyl, alkyl), the gold(III)
chloride complexes are rapidly photoreduced to the
corresponding gold(I) chloride complexes using a 365-nm
irradiation. The course of this photoreduction was more rapid
(30 min) than thermal reduction at 80°C (12h). The
photoreduction under a conventional TLC lamp (365 nm) also
induces the formation of Au(I) exhibiting a higher catalytic
activity than those obtained by daylight exposure or heating.
Under our conditions, we found no variation in the rate of
photoreduction ascribable to the presence of the
diphenylanthracene chromophore. A possible antenna effect
resulting from efficient intramolecular energy transfer from the
organic chromophore to the gold center would presumably
become more pronounced under low light absorption conditions.
The photoreductive elimination of X2 seems to occur under 365
nm or daylight irradiation as evidenced by the formation of
chlorinated organic by-products. In the case of the
dialkylthioether gold complexes, reductive elimination might be
triggered through a direct excitation on the gold center and/or
through a ligand-to-metal charge transfer (LMCT), possibly
accompanied by the formation of a chloronium ion.
The catalytic properties of gold complexes at different oxidation
states were evaluated in cyclization reactions of alkynes under
homogeneous reaction conditions for single and sequential
double cyclization(s). All Au(III) chloride and in situ generated
Au(I) chloride complexes showed excellent efficiency (high yield,
reasonnable reaction times). The corresponding cationic Au(I)
complexes obtained in the presence of AgOTf were also active.
Interestingly, the catalytic system composed of the Au(III)
chloride catalyst and AgOTf (in the dark) showed a similar
activity to classical cationic Au(I) catalysts. The complexes of
thioethers and AuCl3 are therefore highly versatile complexes
with the possibility to prepare and tailor their catalytic activity by
using light to control the oxidation state and silver salts to
modify the coordination sphere of the gold center.
We then decided to try an alternative catalytic system,
composed of Ln•xAuCl3 complexes in the presence of silver
salts, in the absence or presence of irradiation during step 2.
This complex system is composed of cationic gold(III) species
and possibly traces of cationic gold(I). The catalytic effect of
Au(III) chloride complexes (L1•2AuCl3 and L2•AuCl3) alone for
the first step and in the presence of a silver triflate for the
second step, allowed the synthesis of compound 10 in excellent
yield (72–76 %, entries 10–11). It should be noted that several
Au(III) active species can co-exist and no data was available
concerning their nature under our homogeneous reaction
conditions. We also cannot exclude the occurence of Au(I)
traces (through undesired photoreduction) that might participate
to the final cyclization.
To better investigate the possible effect of in situ reduction of
Au(III) into Au(I), we proceed to the final cyclization using
gold(III) chloride complexes in the presence of AgOTf and
under a 365-nm irradiation (entries 12–15). Compound 10 was
isolated in excellent yields (72–82 %) that are similar to those
determined in the dark. The catalytic system composed of
AgOTf and complexes based on thioether ligands L3 – L4 and
Au(III) chloride are therefore efficient gold species for the
Experimental Section
The synthesis of ligands 1–4 and indole 8 is detailed in the supporting
information. N-propargylic amides 5–7 were prepared according to the
literature.[17c]
Preparation of AuCl3 complexes: in an oven-dried glass centrifuge
tube (50 mL), were charged a solution of NaAuCl4•2H2O (4.0 equiv. for
ligands 1 and 3; 2.0 equiv for ligands 2 and 4) in MilliQ water (10 mL)
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
and a solution of thioether ligand Ln (n= 1-4, 1.0 equiv) dissolved in
toluene (12 mL) was added. After shaking in the dark for 1 to 2 hours
and centrifugating at 3000 rpm for 15 min, the upper golden organic
phase was collected, concentrated under vacuum and dried.
We are grateful to Dr. Didier Bourissou for helpful comments on
mechanism.
Keywords: gold • photoreduction • homogeneous catalysis •
thioether ligand • mechanism
Photoreduction of AuCl3 complexes: a solution of Ln•AuCl3 complex
(n = 1–4, 60 µM) in dry dichloromethane or toluene (20 mL) was
irradiated at 365 nm, using either an optical bench equipped with a
xenon-mercury lamp and a high-intensity monochromator, or a TLC
lamp (6 W). The photoreduction process was monitored by UV-Visible
spectroscopy.
[1]
[2]
a) Y. Ito, M. Sawamura, T. Hayashi, J. Am. Chem. Soc. 1986, 108,
6405-6406; b) Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56, 3729-
3731; c) J. H. Teles, S. Brode, M. Chabanas, Angew. Chem. Int. Ed.
1998, 37, 1415-1418.
For a selection, see: a) A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180-
3211; b) D. J. Gorin, B. D. Sherry, F. D. Toste, Chem. Rev. 2008, 108,
3351-3378; c) A. S. K. Hashmi, Acc. Chem. Res. 2014, 47, 864-876;
d) L. Zhang, Acc. Chem. Res. 2014, 47, 877-888; e) A. Fürstner, Acc.
Chem. Res. 2014, 47, 925-938; f) M. Joost, A. Amgoune, D. Bourissou,
Angew. Chem. Int. Ed. 2015, 54, 115022-15045.
Thermal reduction: a solution of Ln•AuCl3 complex (n = 1–4, 70 µM) in
dry toluene (10 mL) was placed in a sealed glassware equipped with a
cuvette (l = 1 cm) and a round flask (25 mL). The solution was heated at
different temperatures (25–80°C). The experiment was monitored by
UV-Visible spectroscopy.
[3]
a) A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost, Angew. Chem.
Int. Ed. 2000, 39, 2285-2288; b) A. S. K. Hashmi, T. M. Frost, J. W.
Bats, J. Am. Chem. Soc. 2000, 122, 11553-11554; c) F. Mo, J. M. Yan,
D. Qiu, F. Li, Y. Zhang, J. Wang, Angew. Chem. Int. Ed. 2010, 49,
2028-2032; d) D. M. Barber, H. Sanganee, D. J. Dixon, Chem.
Commun. 2011, 47, 4379-4381; e) Y. Qiu, W. Kong, C. Fu, S. Ma, Org.
Lett. 2012, 14, 6198-6201.
β–Carbolinone 9: to a solution of 2-amido-indole 8 (17 mg, 44 µmol) in
dichloromethane (2 mL) was added a solution of L1•2AuCl3 catalyst (2.2
µmol, 5 mol%) in dichloromethane (0.5 mL). The reaction was stirred at
room temperature for 40 h. After concentration under vacuum, the
residue was purified by column chromatography on silica gel (eluent:
petrol ether/AcOEt, 1:1). Compound 9 was isolated as a white solid (93 %
yield). m.p. 163.7 °C; FTIR (NaCl): ν 2923, 2854, 2219, 1663, 1590,
[4]
a) E. C. Constable, T. A. Leese, J. Organomet. Chem. 1989, 363, 419-
424; (b) S. K. Hashmi, C. Lothschütz, M. Ackermann, R. Doepp, S.
Anantharaman, B. Marchetti, H. Bertagnolli, F. Rominger, Chem. Eur.
J. 2010, 16, 8012-8019; (c) R. Kumar, A. Linden, C. Nevado, Angew.
Chem. Int. Ed. 2015, 54, 14287-14290.
1331, 738 cm–1 1H NMR (300 MHz, CDCl3) δ 8.18 (d, J = 8.1 Hz, 1H),
;
7.70-7.72 (m, 1H), 7.42-7.58 (m, 5H), 7.27-7.33 (m, 1H), 7.08-7.22 (m,
5H), 6.93 (d, J = 0.88 Hz, 1H), 4.35 (s, 3H), 2.63 (d, J = 0.88 Hz, 3H);
13C NMR (75 MHz, CDCl3) δ 156.0, 142.7, 141.2, 132.9, 131.5, 129.1,
128.6, 128.5, 128.4, 128.2, 126.8, 126.7, 126.5, 125.0, 122.9, 122.8,
122.3, 122.2, 120.2, 111.9, 110.3, 94.7, 85.5, 31.5, 17.0; HRMS (ESI):
m/z calculated for C27H21N2O [M+H]+: 389.1654; found: 389.1638.
[5]
[6]
M. T. Reetz, K. Sommer, Eur. J. Org. Chem. 2003, 3485-3496.
a) B. Ranieri, I. Escofet, A. M. Echavarren. Org. Biomol. Chem. 2015,
13, 7103-7118; b) J. H.Teles, Angew. Chem. Int. Ed. 2015, 54, 5556-
5558.
[7]
a) S. Komiya, T. A. Albright, R. Hoffmann, J. K. Kochi, J. Am. Chem.
Soc. 1976, 98, 7255-7265; b) S. Komiya, J. K. Kochi, J. Am. Chem.
Soc. 1976, 98, 7599-7607; c) J. Vicente, M. D. Bermudez, J.
Escribano, Organometallics 1991, 33, 6358-6368; d) S. Lavy, J. J.
Miller, M. Pazicky, A.-S. Rodrigues, F. Rominger, C. Jakel, D. Serra, N.
Vinokurov, M. Limbach, Adv. Synth. Catal. 2010, 352, 2993-3000; d)
M. Pazicky, A. Loos, M. J. Ferreira, D. Serra, N. Vinokurov, F.
Rominger, C. Jäkel, A. S. K. Hashmi, M. Limbach, Organometallics
2010, 29, 4448-4458; e) M. Blaya, D. Bautista, J. Gil-Rubio, J. Vicente,
Organometallics 2014, 10, 3380-3384; f) W. J. Wolf, M. S. Winston, F.
D. Toste, Nat. Chem. 2014, 6, 159-164.
General procedure for one-pot cyclizations: to a solution of 2-amido-
indole 8 (17 mg, 44 µmol) in dichloromethane (2 mL) was added a
solution of gold catalyst (2.2 µmol, 5 mol%) in dichloromethane (0.5 mL).
The reaction was stirred at room temperature for 8 h. Then silver
trifluoromethanesulfonate (1.0 mg, 7 mol%) was added and the reaction
was stirred for 4 hours.
Indolino-4H-benzoquinolizin-4-one 10: following the general
procedure with catalyst L3•2AuCl3, the reaction medium was finally
concentrated under vacuum. The residue was purified by column
chromatography on silica gel (eluent: CH2Cl2/MeOH, 50:1). Compound
10 was isolated as yellow sticky oil (80 % yield): FTIR (NaCl): ν 2925,
2854, 1630, 1525, 1334, 1266, 1030, 756, 637 cm–1; 1H NMR (300 MHz,
CD3CN) δ 8.41 (s, 1H), 8.31 (d, J = 7.95 Hz, 1H), 8.06 (d, J = 8.28 Hz,
1H), 7.96 (d, J = 7.62 Hz, 1H), 7.62-7.82 (m, 5H), 7.40-7.53 (m, 4H),
6.96 (s, 1H), 3.91 (s, 3H), 2.90 (s, 3H); 13C NMR (75 MHz, CD3CN) δ
146.2, 144.6, 141.4, 134.8, 133.3, 132.9, 132.7, 2.1, 131.4, 130.3, 129.6,
129.4, 126.8, 126.1, 125.2, 124.2, 123.9, 123.5, 122.8, 122.2, 121.1,
119.9, 119.6, 112.3, 111.3, 34.0, 17.3; HRMS (ESI): m/z calculated for
C27H21N2O [M+H]+: 389.1654; found: 389.1648.
[8]
a) T. de Haro, C. Nevado, J. Am. Chem. Soc. 2010, 132, 1512-1513;
b) W. E. Brenzovich Jr., D. Benitez, A. D. Lackner, H. P. Shunatona, E.
Tkatchouk, W. A. Goddard III, F. D. Toste, Angew. Chem. Int. Ed.
2010, 49, 5519-5522; c) G. Zhang, L. Cui, Y. Wang, L. Zhang, J. Am.
Chem. Soc. 2010, 132, 1474-1475; d) L. T. Ball, G. C. Lloyd-Jones, C.
A. Russell, Science 2012, 337, 1644-1648; e) W. Wang, J. Jasinski, G.
B. Hammond, B. Xu, Angew. Chem. Int. Ed. 2010, 49, 7247-7252;
a) B. Sahoo, M. N. Hopkinson, F. Glorius, J. Am. Chem. Soc. 2013,
135, 5505-5508; b) M. N. Hopkinson, B. Sahoo, F. Glorius, Adv. Synth.
Catal. 2014, 356, 2794-2800; c) X.-Z. Shu, M. Zhang, Y. He, H. Frei, F.
D. Toste, J. Am. Chem. Soc. 2014, 136, 5844-5847; d) Y. He, H. Wu, ,
F. D. Toste, Chem. Sci. 2015, 6, 1194-4758.
[9]
[10] a) G. Revol, T. McCallum, M. Morin, F. Gagosz, L. Barriault, Angew.
Chem. Int. Ed. 2013, 52, 13342-13345; b) T. McCallum, L. Barriault,
Chem. Sci. 2016, 7, 4754-4758; c) L. Huang, M. Rudolph, F.
Rominger, A. S. K. Hashmi, Angew. Chem. Int. Ed. 2016, 55, 4808-
4813; d) L. Huang, F. Rominger, M. Rudolph, A. S. K. Hashmi, Chem.
Commun. 2016, 52, 6435-6438; e) S. Witzel, J. Xie, M. Rudolph, A. S.
K. Hashmi, Adv. Synth. Catal. 2017, 359, 1522-1528.
Acknowledgements
Financial support was provided by the China Scholarship
Council (Z.C. Ph.D. fellowship), the University of Bordeaux and
Centre National de la Recherche Scientifique. Claire Mouche
and Patricia Castel are thanked for mass spectrometry analysis.
[11] T. S. Teets, D. G. Nocera, J. Am. Chem. Soc. 2009, 131, 7411-7420.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
[12] C. Hirtenlehner, C. Krims, J. Hölbling, M. List, M. Zabel, M. Fleck, R. J.
F. Berger, W. Schoefberger, U. Monkowius, Dalton Trans. 2011, 40,
9899-9910.
[18] D. B. England, A. Padwa, Org. Lett. 2008, 10, 3631-3634.
[19] K. Nakamura, S. Furumi, M. Takeuchi, T. Shibuya, K. Tanaka, J. Am.
Chem. Soc. 2014, 136, 5555-5558.
[13] a) V. J. Scott, J. A. Labinger, J. E. Bercaw, Organometallics 2010, 29,
4090-4096; b) M. J. Ghidiu, A. J. Pistner, G. P. A. Yap, D. A.
Lutterman, J. Rosenthal, Organometallics 2013, 32, 5026-5029.
[14] C. Mongin, I. Pianet, G. Jonusauskas, D. M. Bassani, B. Bibal, ACS
Catal. 2015, 5, 380-387.
[20] a) H. Yu, G. Zhang, H. Huang, Angew. Chem. Int. Ed. 2015, 54,
10912-10916; b) J. Li, Y. Yang, Z. Wang, B. Feng, J. You, Org. Lett.
2017, 19, 3083-3086.
[21] a) S. D. Kuduk, R. K. Chang, C. N. Di Marco, W. J. Ray, L. Ma, M.
Wittmann, M. A. Seager, K. A. Koeplinger, C. D. Thompson, G. D.
Hartman, M. T. Bilodeau, ACS Med. Chem. Lett. 2010, 1, 263-267; b)
S. D. Kuduk, C. N. Di Marco, 2010, WO 2010042347; c) A. Zolotoy, E.
Hayes, 2006, WO 2006110477; d) J. A. Wiles, A. Hashimoto, J. A.
Thanassi, J. Cheng, C. D. Incarvito, M. Deshpande, M. J. Pucci, B. J.
Bradbury, J. Med. Chem. 2006, 49, 39-42; e) B. J. Bradbury, M. J.
Pucci, Curr. Opin. Pharmacol. 2008, 8, 574-581.
[15] a) C. K. Mirabelli, D. T. Hill, L. F. Faucette, F. L. McCabe, G. R. Girard,
D. B. Bryan, B. M. Sutton, J. O'Leary Bartus, S. T. Crooke, R. K.
Johnson, J. Med. Chem. 1987, 30, 2181-2190; b) R. Uson, A. Laguna,
M. Laguna, D. A. Briggs, H. H. Murray, J. P. Fackler Jr., Inorg. Synth.
1989, 26, 85-91.
[16] A. Vogler, H. Kunkely, Coord. Chem. Rev. 2001, 219-221, 489-507.
[17] a) A. S. K. Hashmi, J. P. Weyrauch, W. Frey, J. W. Bats, Org. Lett.
2004, 6, 4391-4394; b) A. S. K. Hashmi, A. M. Schuster, F. Rominger,
Angew. Chem. Int. Ed. 2009, 48, 8247-8249; c) J. P. Weyrauch, A. S.
K. Hashmi, A. Schuster, T. Hengst, S. Schetter, A. Littmann, M.
Rudolph, M. Hamzic, J. Visus, F. Rominger, W. Frey, J. W. Bats,
Chem. Eur. J. 2010, 16, 956-963; d) A. S. K. Hashmi, A. Loos, S.
Doherty, J. G. Knight, K. J. Robson, F. Rominger, Adv. Synth. Catal.
2011, 353, 749-759 and cited references.
[22] Under UV-Visible irradiation, in the presence or absence of gold(III) or
gold(I) chloride complexes, compound 9 showed an uncontrolled
reactivity as several spots were observed by TLC. No trace of product
10 was detected. Only cationic gold catalysts successfully lead to the
6-endo-dig product. As a consequence, no in situ photoreduction of
gold(III) complexes can be conducted in the presence of 9, to achieve
a in situ Au(III)/Au(I) sequential catalysis.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804322
Chemistry - A European Journal
FULL PAPER
FULL PAPER
Four homogeneous dialkylthioether
gold(III) complexes undergo a rapid
halogen photoreductive elimination
process into the corresponding gold(I)
complexes. The catalytic properties of
these gold chlorides at different
oxidation states were evaluated in the
cyclization of propargylamines and a
one-pot cascade reaction towards a
fused heteropolycyclic compound.
Zhen Cao, Dario M. Bassani, Brigitte
Bibal*
Page No. – Page No.
Photoreduction of Thioether Gold(III)
Complexes: Mechanistic Insight and
Homogeneous Catalysis
This article is protected by copyright. All rights reserved.