Deciphering Photoluminescence Dynamics and Reactivity of the Luminescent Metal-Metal-Bonded Excited State of a Binuclear Gold(I) Phosphine Complex Containing Open Coordination Sites

Article abstract of DOI:10.1002/chem.201503045

Luminescent metal complexes having open coordination sites hold promise in the design of sensory materials and photocatalysts. As a prototype example, [Au₂(dcpm)₂)]²⁺ (dcpm = bis(dicyclohexylphosphanyl) is known for its intriguing environmental sensitive photoluminescence. By integrating a range of complementary ultrafast time-resolved spectroscopy to interrogate the excited state dynamics, this study uncovers that the events occurring in extremely rapid timescales and which are modulated strongly by environmental conditions play a pivotal role in the luminescence behavior and photochemical outcomes. Formed independent of the phase and solvent property within 0.15 ps, the metal-metal bonded ³5dσ6pσ state is highly reactive possessing strong propensity toward increasing coordination number at Au^I center, and with 510 ps lifetime in dichloromethane is able to mediate light induced C-X bond cleavage.

Full text of DOI:10.1002/chem.201503045

DOI: 10.1002/chem.201503045
Communication
&
Photochemistry
Deciphering Photoluminescence Dynamics and Reactivity of the
Luminescent Metal–Metal-Bonded Excited State of a Binuclear
Gold(I) Phosphine Complex Containing Open Coordination Sites
[
a, d]
[c]
[a]
[c]
Chensheng Ma,
Chi-Ming Che*
Chris Tsz-Leung Chan, Wai-Pong To, Wai-Ming Kwok,* and
[
a, b]
all of which vary profoundly with factors such as nature of
I
Abstract: Luminescent metal complexes having open co-
ordination sites hold promise in the design of sensory ma-
terials and photocatalysts. As a prototype example,
counter ion and solvent. The open coordination site of Au in
2
+
II
[Au
(diphosphine)
]
2
is reminiscent of that of Pt in
2
4
À
3
[Pt
(P
O
2
H
5
)
]
4
allowing the 5ds*6ps excited state to perform
2
2
2
+
[
Au (dcpm) )]
(dcpm = bis(dicyclohexylphosphanyl) is
CÀX activation and emissive exciplex formation at the inner co-
ordination sphere. An outcome of these excited-state prop-
erties leads to micro-environmental dependence of lumines-
2
2
[
9,10]
known for its intriguing environmental sensitive photolu-
minescence. By integrating a range of complementary ul-
trafast time-resolved spectroscopy to interrogate the ex-
cited state dynamics, this study uncovers that the events
occurring in extremely rapid timescales and which are
modulated strongly by environmental conditions play
a pivotal role in the luminescence behavior and photo-
chemical outcomes. Formed independent of the phase
and solvent property within ~0.15 ps, the metal–metal
I
I
I
cence on Au–Au interacting binuclear and polynuclear Au
complexes, which has been used in the design of lumines-
cence sensors and new materials for optoelectronic
[
4,6,11–14]
2+
switches.
More recently, [Au
(diphosphine)
2
]
2
has been
reported as a new novel photocatalyst for light-induced organ-
[
15,16]
ic transformation reactions.
An archetype example of [Au
2
+
(diphosphine)
(dcpm=bis(dicyclohexylphosphanyl)methane,
]
2
is
2
3
2+
bonded 5ds*6ps state is highly reactive possessing
[Au (dcpm) ]
2 2
I
strong propensity toward increasing coordination number
1, Figure 1), a bi-Au complex bearing the UV-silent phosphine
ligand. Irrespective of solvent properties (e.g. coordinating
I
[7]
at Au center, and with ~510 ps lifetime in dichlorome-
thane is able to mediate light induced C–X bond cleav-
solvent CH
3
CN vs. less coordinating CH Cl ) and phase condi-
2 2
age.
I
Luminescent Au complexes have garnered considerable inter-
est for their fascinating photophysical properties and photo-
[
1–3]
10
10
I
chemistry.
A family of dinuclear d –d Au complexes con-
2
+
taining bridging phosphine ligands, [Au (diphosphine) ] , is of
2
2
particular interest from the perspective of Au–Au bonding in-
[
1,2,4–6]
[1,2,7,8]
teractions
and their intriguing photoluminescence,
[
a] Prof. Dr. C. Ma, Dr. W.-P. To, Prof. Dr. C.-M. Che
State Key Laboratory of Synthetic Chemistry
Institute of Molecular Functional Materials
Department of Chemistry, The University of Hong Kong
Pokfulam Road, Hong Kong (P. R. China)
E-mail: cmche@hku.hk
[
b] Prof. Dr. C.-M. Che
HKU Shenzhen Institute of Research and Innovation
Shenzhen 518053 (P. R. China)
[
c] C. T.-L. Chan, Dr. W.-M. Kwok
Department of Applied Biology and Chemical Technology
The Hong Kong Polytechnic University
Hung Hom, Kowloon, Hong Kong (P. R. China)
E-mail: wm.kwok@polyu.edu.hk
[
d] Prof. Dr. C. Ma
School of Chemistry and Environmental Engineering
Shenzhen University, Shenzhen, 518060 (P. R. China)
[7a]
Figure 1. Top: ORTEP drawing of 1. Bottom: Steady-state absorption, nor-
malized emission (lex =280 nm), and excitation spectra (a) of 1·(ClO
4
)
2
in
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201503045.
degassed CH
temperature.
3
CN (c), CH Cl (c), and in the solid-state (c) at room
2 2
Chem. Eur. J. 2015, 21, 13888 – 13893
13888
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
3
tions (solution vs. solid phase), 1·(ClO ) features characteristic
tion in CH Cl . The 5ds*6ps is highly reactive and events oc-
2 2
4
2
low energy absorption due to the strongly spin and optical al-
lowed metal centred (MC) Au–Au 5ds*!6ps transition (e
curring at very early times play a crucial role in accounting for
the luminescence and photochemical reactivity of
~
max
4
3
À1
À1
2+
2
.6–2.7”10 dm mol cm ). This transition promotes elec-
[Au (diphosphine) ] complexes.
2
2
2
trons from the anti-bonding 5dz to the bonding combination
Fs-TRF measurement realized based on a time-gating tech-
nique (details in the Supporting Information) was done to
probe fluorescence as a function of time after excitation (at
280 nm) into the 5ds*6ps state. Figure 2 shows the TRF spec-
of 6s/6p orbital (with the Au–Au axis defined as the z axis)
z
leading to enhanced metal–metal interactions and an increase
in the Au–Au formal bond order from 0 in the ground state
1
(
S ) to 1 in the 5ds*6ps excited
0
[
17,18]
state.
Complex 1·(ClO4)2 in
the solid state features an in-
tense near-UV emission with
lmax ~370 nm (F = ~0.37). In
e
CH Cl , the complex is much less
2
2
luminescent (F ~0.01) showing
e
still the ~370 nm emission but
with an added weak shoulder at
~
420 nm. In contrast, 1·(ClO ) in
4 2
CH CN exhibits emission (F = ~
3
e
0
.05) dominated by a distinct
broad profile (lmax ~480 nm)
along with a very weak band at
4 2 2 2 3
Figure 2. Fs-TRF spectra recorded with 280 nm excitation of 1·(ClO ) in (a) CH Cl and (b) CH CN. Insets show ex-
perimental (*) and fitted (c) kinetic decay profiles at denoted fluorescence wavelengths. Arrows indicate tem-
poral evolution of the TRF.
~
~
315 nm and the emission at
370 nm has a reduced intensity
compared to that in CH Cl and
2
2
in the solid state. The ~480 nm
emission is typical of those re-
I
[1c,d,7,8]
ported for analogous bi/multi-nuclear Au complexes.
depicted also in Figure 1, the luminescence in CH CN, CH Cl
2
As
tra recorded with 1·(ClO ) in CH Cl and CH CN; the result in
4 2 2 2 3
the solid state is given in Figure S1 in the Supporting Informa-
3
2
and the solid phase features a similar excitation spectrum that
resembles the corresponding low energy absorption. As
a result, these seemingly very different emissions must arise
due to altered dynamic evolutions of the common 5ds*!6ps
excitation. Early studies ascribed the broad visible region emis-
tion. For all the three studied cases, the TRF, which arises from
1
photogenerated 5ds*6ps, display a common profile with l
max
~315 nm and deplete promptly with a ~0.13–0.18 ps time con-
stant (insets in Figure 1 and Figure S1, Supporting Informa-
tion).
3
3
[1,2]
sion to the MC triplet state of 5ds*6ps or 5dd*6ps;
more
The very rapid decay of TRF is indicative of the presence
of a highly efficient non-radiative process to deactivate
recent work from both experimental and theoretical perspec-
3
1
tives showed that the 5ds*6ps is in fact responsible for the
the 5ds*6ps. To interrogate the nature of this process, a
~
370 nm emission, whereas the emission at visible wave-
parallel fs-TA measurement was conducted and the spectra
obtained in CH Cl2 and CH CN are compared in Figure 3;
lengths is given by a triplet exciplex that is formed from sub-
2
3
3
strate binding of the 5ds*6ps with solvent or/and the counter
the corresponding TA time profiles are given in Figure S2 in
the Supporting Information. Temporal evolutions of TA in the
two solvents are very similar at early times (< ~1 ps) but they
differ distinctively at time intervals beyond ~1 ps after excita-
tion.
3
[7]
anion ( SB hereafter).
To develop [Au (diphosphine) ] as a new photocatalyst for
2
+
2
2
light-induced inner-sphere CÀX activation, the issues to be ad-
dressed/clarified include the dynamic pathway(s) of the
3
5
ds*6ps state and whether this excited state could be
In both CH Cl and CH CN, the TA at initial time (0.2 ps) fea-
2
2
3
formed with high efficiency and has sufficient lifetime to allow
for substrate binding to occur at the open coordination site of
ture broad absorption with lmax at ~355 and 480 nm; the spec-
trum evolves with a ~0.14–0.16 ps time constant (Figures S2a
and S2c, Supporting Information) into a varied profile with lmax
~365 nm accompanied with an isosbestic point at ~413 nm
(Figure 3a,c). Upon formation, the ~365 nm TA persists in tens
of ps (Figure 3b and Figure S2b, Supporting Information) and
decays with a time constant of ~510 ps (Figure S3, Supporting
Information) in CH Cl . Unlike this example, the ~365 nm TA in
I
Au in solution. To address these issues, a comparative time-re-
solved (TR) study performed with a conjunction of fs TR fluo-
rescence (fs-TRF), transient absorption (fs-TA), ps- and ns-TR
emission (TRE) on 1·(ClO ) in the solid state and in solution of
4
2
CH CN and CH Cl was undertaken. By monitoring directly tem-
3
2
2
poral evolution of excited-state spectra over a time span from
2
2
tens of fs to tens of ms after the 5ds*!6ps excitation, this
CH CN transforms with a ~4.5 ps time constant (Figure S2d,
Supporting Information) into yet another profile with lmax ~
422 nm with a concomitant pair of isosbestic points at ~388
and 527 nm (Figure 3d). The ~422 nm TA is long-lived with an
3
1
3
work unveils an ultrafast 5ds*6ps! 5ds*6ps intersystem
3
crossing (ISC) and a remarkably rapid rate of the 5ds*6ps to
coordinate with solvent in CH CN or to perform a redox reac-
3
Chem. Eur. J. 2015, 21, 13888 – 13893
www.chemeurj.org
13889
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
sion, the TRE were produced by
employing an adequately pro-
longed gating time (~3 ps in
CH Cl ; ~2 ns in CH CN and the
2
2
3
solid state) to allow capturing
emission from a longer-lived and
relatively weakly emissive state
(
details in the Supporting Infor-
mation).
The TRE in CH Cl (Figure 4a)
2
2
and in the solid state (Figure S4,
Supporting Information) feature
similar spectra that are nearly
identical to the ~370 nm steady-
3
[7]
state emission from 5ds*6ps;
the emission lifetime was mea-
sured to be ~4.3 ms in the solid
state and ~510 ps in CH Cl . The
2
2
latter is equivalent to the decay
time of ~365 nm TA in the same
solvent,
a common origin, that is, the
365 nm TA is also from the
which
indicates
Figure 3. Fs-TA spectra recorded for 1·(ClO
4 2 2 2 3
) in (a,b) CH Cl and (c,d) CH CN at early (a,c) and late (b,d) times after
2
80 nm excitation. Arrows show directions of spectral evolution.
~
3
5
ds*6ps (T ), corresponding to
1
absorption signature (T !T ) of
1
n
3
intensity kept unchanged over a timescale up to several ns
after the excitation.
this state. The much shorter lifetime of the 5ds*6ps in CH Cl
2 2
corroborates with the markedly reduced Fe in this solvent
The ~0.14–0.16 ps time constant of the first-phase TA evolu-
tion coincides with the fluorescence decay time observed in
the fs-TRF. The initial ~355/480 nm TA is, therefore, corre-
compared to that in the solid state, revealing the presence of
3
an effective quenching process to the 5ds*6ps (T ) in CH Cl .
1
2
2
On the other hand, the TRE in CH CN shows spectra resem-
3
1
3
sponding to absorption (S !S ) of the 5ds*6ps (S ) while the
bling the ~480 nm steady-state emission due to SB, the decay
1
n
1
~
365 nm (Figure 3a,b) absorption is assigned to an immediate
of which occurs on a timescale consistent with the ~422 nm
species in fs-TA. The ~422 nm TA can thus be ascribed to the
1
product produced by non-radiative decay of the 5ds*6ps. Of
3
note, the ~365 nm (in both CH Cl and CH CN) and ~422 nm
absorption spectrum of the SB.
2
2
3
(
in CH CN only) species in TA shows no dynamic counterparts
The decay dynamics of TRE (in CH CN) from tens of ns to
3
3
in the fs-TRF (Figure 2). Therefore, these species, which are
formed in high yield (as suggested by their rapid formation
rates), must be very weakly emissive featuring a reduced (by
several ms was, however, found to be complex with the time
profile deviating strongly from singlet exponential and the
emission spectrum displaying a subtle but clear redshift from
lmax ~480 nm at 10 ns to lmax ~510 nm at 5 ms after excitation.
A proper fit to the kinetic decay requires a multi-exponential
function described by at least three components of ~52 ns,
1
0-fold or more) radiative rate constant (k ) when compared to
r
1
that of the ds*ps (which displays spin-allowed fluorescence
8
À1
emission with k = ~2.6”10 s according to the Strickler–Berg
r
equation).
To resolve the identities of
~
365 and ~422 nm TA (Fig-
ure 3a,b) and their relevance to
the steady-state emission in
Figure 1, Figure 4 shows TRE
spectra and the related decay
profiles acquired with 1·(ClO ) in
4
2
CH Cl2 and CH CN, the data in
2
3
the solid state is depicted in Fig-
ure S4 in the Supporting Infor-
mation. Unlike fs-TRF (Figure 2),
which was recorded by using an
ultra-short
gating
duration
(
~100 fs) for the detection of
Figure 4. TRE spectra recorded with 1·(ClO
4 2 2 2 3
) at various times after 280 nm excitation in (a) CH Cl and (b) CH CN.
strongly optically allowed emis- Insets show the experimental (*) and fitted (c) decay profile from the corresponding TRE.
Chem. Eur. J. 2015, 21, 13888 – 13893
www.chemeurj.org
13890
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
ISC in ultrafast (tens of fs to several ps) timescale has been
documented for a number of 1st, 2nd and 3rd row metal com-
Table 1. Spectral and dynamic parameters obtained for excited states of
1
·(ClO
4
)
2
in the solid state, CH
2
Cl
2
and CH
3
CN.
II [20]
II [21]
I [22]
II[23]
plexes containing metals such as Fe , Ru , Re, Pt
or
[a]
[b]
[c]
l
abs [nm]
l
em [nm] Solid state t CH
2
Cl
2
CH
3
CN
I [24]
(
mononuclear) Au. The work here is, however, the first direct
1
10
10
I
I
ds*ps 355, 480 315
0.18 ps
4.3 ms
0.15 ps
510 ps
0.13 ps
4.5 ps
52 ns, 910 ns, 3.2 ms
study on ISC for the type of d –d bi-Au (Au for short) com-
2
3
1
3
ds*ps 365
SB 422
370
480–510
plexes. The ~150 fs ISC ( 5ds*6ps! 5ds*6ps) of 1 appears
not surprisingly in regard to the large spin-orbital coupling
(SOC) constant (x) of gold and the heavy atom effect in pro-
moting the ISC. The result is, however, significant when noting
the relatively slow ISC reported for the counterpart states in
3
[
a] Absorption wavelength maximum from fs-TA. [b] Emission wavelength
maximum from fs-TRF and ns-TRE. [c] Decay time constant.
8
8
2+
I
d –d analogues such as [Rh L ]
(denoted Rh , L=2,5-di-
2
2
4
[
25]
methyl-2,5-diisocyanohexane, t = ~820 ps) and the classic
Pt (P O H ) ] (denoted Pt , t = ~15-29 ps). Indeed, there
ISC
4À
II
2
[26]
[
2
2
5
2 4
ISC
is increasing evidence that the ISC rate of metal complex is
[
24–28]
not governed solely by SOC strength of the metal.
The
II
slow ISC rate of Pt 2 was ascribed to symmetry forbidden (in
1,3
D ) for the SOC between 5ds*6ps, a large singlet-triplet
4
h
À1
II
2
energy splitting (DE= ~5000 cm1 in Pt ), and a lack of media-
ting triplet state between the 5ds*6ps. The tens of ps ISC
,3
II
rate of Pt 2 was proposed in a recent study to occur due to
symmetry lowering induced by structure distortion in the ex-
[
26]
1,3
I
cited state from the S0. The 5ds*6ps of Au have an elec-
2
À1
Scheme 1. Dynamics and deactivation diagram proposed for 1·(ClO
4
)
2
in the
tronic nature and energy gap (DE= ~4700 cm according to
the ~315/370 nm emission for ds*ps/ ds*ps, Figures 2 and 4
and Table 1) similar to that of the 5ds*6ps of Pt . One major
solid state (middle), CH
3
CN (left) and CH
2
Cl
2
(right).
1
3
1,3
II
2
1,3
I
2
II
~
910 ns and 3.2 ms time constant with a fraction contribution
difference in the 5ds*6ps between Au and Pt 2 is that, as
I
of ~79.6, ~19.9 and ~0.5%, respectively.
a result of the linear 2-coordination in Au versus the planar 4-
II
I
Taking together the key spectral features and the direct dy-
namic interplays between them (Table 1), the following excit-
ed-state cascade for the environmental dependent emission of
coordination in Pt , the Au features symmetry (D ) lower
2 2h
II
than that of Pt . We conceive that, apart from the somewhat
more favourable SOC strength (x= ~5000/4000 cmÀ1 for Au/
2
1
5
·(ClO4)2 is constructed (Scheme 1). Upon excitation, the
Pt), the low symmetry could be among the main factors for
1
2
II
ds*6ps (S ) which features lmax ~315 nm emission, depletes
the ultrafast ISC (by >10 -fold faster than t of Pt ) exhibited
1
ISC
2
3
I
by ISC with a ~0.15 ps time constant to the 5ds*6ps (T ). This
occurs as a common process invariant with the phase condi-
by the Au complex 1.
1
2
3
3
The very rapid rate (~4.3 ps) of 5ds*6ps! SB conversion
3
3
tion and solvent properties. The 5ds*6ps (T ) after formation,
testifies that the 5ds*6ps state has an unprecedentedly
1
decays with lifetime of ~510 ps in CH Cl and ~4.3 ms in the
strong proclivity towards increasing coordination number at
2
2
I
solid phase producing the ~370 nm emission. In CH CN, the
the Au centres. This is associated with the open coordination
3
3
I
5
ds*6ps (T ) transforms with a ~4.5 ps time constant through
framework of the Au centres in 1. That the binding occurs in
1
3
substrate binding to yield SB, which gives the ~480 nm emis-
such a rapid timescale is remarkable. It is worth noting that
sion exhibiting multi-exponential decay with 52 ns/910 ns/
the binding to CH CN, albeit very fast, is slow compared to the
3
[
29]
3
.2 ms time constants. The ~0.15 and 4.3 ps rates of ISC (tISC)
reorientation solvation dynamics (~0.24 ps) of CH CN. This
3
and substrate binding are very rapid, exceeding greatly those
of the radiative decay or other likely deactivation process. This
implies close to unitary quantum efficiency for both the pro-
cesses, when considering their non-radiative nature is responsi-
ble for the minute ~315 nm fluorescence in the steady-state
implies the presence of a certain free-energy barrier along the
3
3
path from precursor 5ds*6ps to the SB product. Most likely,
this is due to an involvement of substantial entropy of activa-
tion that is caused by the steric hindrance imposed by the
bulky phosphine groups and to reposition solvent molecules
to the required orientation for the development of the binding
emission in all three examined conditions and the minimal
3
~
370 nm 5ds*6ps (T ) phosphorescence in CH CN. Moreover,
through the coordinating end (cyano group) of CH CN to the
1
3
3
3
I
that the SB is involved only in CH CN but not in CH Cl pres-
Au centre. Moreover, the multi-exponential decay and the con-
3
2
2
3
3
ents compelling evidence that the SB is brought about by
current redshift of the SB emission are evidence indicating the
3
À
3
5
ds*6ps binding to CH CN solvent, rather than ClO counter
complex composition of the SB, which we ascribe to stem
3
4
3
anion. This assignment is corroborated by the ~4.5 ps binding
rate, which is much faster than could be expected for binding
with the counter anion. The latter is a process limited by the
dynamics of diffusion, which occurs typically in the timescale
from several to tens of ns under the conditions of solvent and
sample concentration (~10À4 m) used in this study.[19]
from the SB being made up of an ensemble of species that
I
have the Au centres coordinated to a varied number of CH CN
3
molecules. These species, depending on the coordination
number, might be energetically stabilized to a varied degree
3
(with respect to the 5ds*6ps) and have a different tendency
towards de-solvation into solvent molecules and the S0 of
Chem. Eur. J. 2015, 21, 13888 – 13893
www.chemeurj.org
13891
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
3
1
(non-radiative decay of the SB). The one stabilized to a great-
photochemical CÀX and CÀH bond cleavage reactions of
4
À [9b]
er degree (therefore emission at longer wavelength) is less
prone to de-solvation and thus has a longer lifetime. This ex-
[Pt (P O H ) ] .
The lower efficiency could be due to the
2
2
5
2 4
3
shorter lifetime of the 5ds*6ps excited state of 1 in solution.
3
plains the result that the longer wavelength SB emission fea-
In summary, we report the first fs time-resolved spectroscop-
I
tures slower decay.
ic characterization for excited states of a prototype Au2 com-
The 510 ps dynamic quenching of ~370 nm emission in
CH Cl provides compelling evidence for strong reactivity of
plex 1, and the first direct comparison of excited state dynam-
ics in inert conditions (the solid state) versus in solvents
(CH CN and CH Cl ) to unravel early time events that are in-
2
2
3
2+
the 5ds*6ps state. It was noted that [Au (dppm) ] complex
2
2
3
2
2
(
dppm=bis(diphenylphosphino)methane) is photochemically
volved in accounting for the intriguing environmentally depen-
dent photoluminescence and the varied photochemical behav-
reactive towards alkyl halide (RÀX) leading to the RÀX bond
I
3
cleavage and formation of [Au (dppm)X ] upon light excita-
iour of Au 2 complexes. The 5ds*6ps state, which is formed,
2
2
[
1b,c,d]
tion.
On the basis of this, we tentatively attribute the
regardless of the environmental conditions, with a ~0.15 ps
1
quenching of ~370 nm emission in CH Cl to arising due to an
inner-sphere reaction between the 5ds*6ps and surrounding
time constant through ISC from the initially excited 5ds*6ps
2
2
3
state, is viable and highly reactive, being at the same time
a powerful reagent for photochemical reactions and with as-
tonishingly strong propensity toward increasing the coordina-
CH Cl molecule(s). Associated with this, we noticed observa-
2
2
3
tion in the fs-TA measurement of the 5ds*6ps spectrum
evolving into an alternative profile (Figure S3a, Supporting In-
formation) which is attributable to a transient species formed
intimately by the quenching reaction and being involved in
the dynamic processes that yield the final photoproducts. The
entire reaction mechanism, which is complex and presumably
I
tion number at the Au centres. These findings provide a bench-
mark example for understanding the photophysics and photo-
I
chemistry of di/multi-nuclear Au complexes.
proceeding through multiple steps over extensive timescales, Acknowledgements
was, however, not assessed owing to the limited detection
time window (< ~6 ns) of the current fs-TA measurement. Of
note, the 510 ps quenching rate is rather rapid, implying that
the related reaction, either by a mechanism of atom or elec-
tron transfer, could be highly effective. We noted that steady-
We thank the National Natural Science Foundation of China
(No. 21473114), the National Key Basic Research Program of
China (2013CB834802), the Area of Excellence Program (AoE/P-
03/08), the Research Grants Council of Hong Kong (PolyU/
5009/11P and PolyU/5009/13P), and the Hong Kong Polytech-
nic University (G-YBAU and G-YM51) for the financial support.
2
+
state photolysis of [Au (dcpm) ]
in CD Cl2 gave
2
2
2
[
[
Au (dcpm)Cl ] and similar photochemistry was observed with
2
2
2
+ [1b]
31
Au (dppm) ]
.
The P NMR spectrum of the solution (Fig-
2
2
ure S5, Supporting Information) showed that, after irradiation
Keywords: femtosecond time-resolved fluorescence · gold ·
photocleavage · substrate binding · transient absorption
with light (l>260 nm, 300 W Xenon arc lamp) for 45 min, the
31
P signal of [Au (dcpm)Cl ] at 46.16 ppm emerged as the
2
2
major new species; both the free dcpm ligand (d=
À10.3 ppm) and its oxide (d=48.5 ppm) were not observed
[
1] a) C.-M. Che, S.-W. Lai, Coord. Chem. Rev. 2005, 249, 1296–1309; b) D. Li,
C.-M. Che, H.-L. Kwong, V. W.-W. Yam, J. Chem. Soc. Dalton Trans. 1992,
3325–3329; c) C.-M. Che, H.-L. Kwong, V. W.-W. Yam, K.-C. Cho, J. Chem.
Soc. Chem. Commun. 1989, 13, 885–886; d) C.-M. Che, H.-L. Kwong, C.-
K. Poon, V. W.-W. Yam, J. Chem. Soc. Dalton Trans. 1990, 11, 3215–3219.
31
(
the P NMR spectroscopic data of [Au (dcpm)Cl ] was also ob-
2 2
tained by independent preparation of this complex in this
3
study). It is proposed that the 5ds*6ps excited state of 1 ini-
2
+
[2] a) V. W.-W. Yam, E. C.-C. Cheng, Top. Curr. Chem. 2007, 281, 269–309;
b) V. W.-W. Yam, E. C.-C. Cheng, Chem. Soc. Rev. 2008, 37, 1806–1813.
[3] R. F. Ziolo, S. Lipton, Z. Dori, J. Chem. Soc. D 1970, 1124–1125.
[4] M. J. Katz, K. Sakai, D. B. Leznoff, Chem. Soc. Rev. 2008, 37, 1884–1895.
tially abstracts chloride from CH Cl to give [Au (dcpm) Cl]
.
2
2
2
2
2
+
The as-formed [Au (dcpm) Cl] intermediate is unstable, and
2
2
undergoes further complex reactions to give [Au (dcpm)Cl ] as
2
2
[
5] a) P. Pyykkç, F. Mendizabal, Inorg. Chem. 1998, 37, 3018–3025; b) P.
Pyykkç, Y. Zhao, Angew. Chem. Int. Ed. Engl. 1991, 30, 604–605; Angew.
Chem. 1991, 103, 622–623.
the observed photoproduct. The mechanism for conversion of
2
+
[
Au (dcpm) Cl] to give [Au (dcpm)Cl ] is complex and further
2 2 2 2
3
study is necessary. Given the reactivity of the 5ds*6ps excited
state of 1, we performed the photochemical reactions of
[
6] a) Y. Takemura, H. Takenaka, T. Nakajima, T. Tanase, Angew. Chem. Int. Ed.
2009, 48, 2157–2161; Angew. Chem. 2009, 121, 2191–2195; b) M. A.
Rawashdeh-Omary, M. A. Omary, H. H. Patterson, J. P. Fackler, J. Am.
Chem. Soc. 2001, 123, 11237–11247.
1
·(OTf) with secondary alcohols and cyclohexene with an ob-
2
jective to observe light-induced CÀH bond cleavage (details
given in the Supporting Information). The triflate salt of 1 was
used because of better solubility. Upon irradiation of 1 in the
presence of isopropyl alcohol at l>260 nm with a 300 W
Xenon arc lamp, production of hydrogen, acetone and pinacol
was observed with turnovers of 2.5, 6.2 and 2.6 respectively.
Similar photolysis of 1 in the presence of cyclohexanol gave
cyclohexanone with 19 turnovers. For the photochemical reac-
tion with cyclohexene, 15 turnovers of 1,1’-bi(cyclohex-2-ene),
which is the coupling product of two allylic radicals of cyclo-
hexene, have been furnished. These findings are similar to the
[
7] a) W.-F. Fu, K.-C. Chan, V. M. Miskowski, C.-M. Che, Angew. Chem. Int. Ed.
1999, 38, 2783–2785; Angew. Chem. 1999, 111, 2953–2955; b) W.-F. Fu,
K.-C. Chan, K.-K. Cheung, C.-M. Che, Chem. Eur. J. 2001, 7, 4656–4664.
8] G. S. M. Tong, S. C. F. Kui, H.-Y. Chao, N. Zhu, C.-M. Che, Chem. Eur. J.
[
[
2009, 15, 10777–10789.
9] a) D. C. Smith, H. B. Gray, Coord. Chem. Rev. 1990, 100, 169–181; b) D. M.
Roundhill, H. B. Gray, C.-M. Che, Acc. Chem. Res. 1989, 22, 55–61.
[10] A. E. Stiegman, S. F. Rice, H. B. Gray, V. M. Miskowski, Inorg. Chem. 1987,
6, 1112–1116.
11] V. W.-W. Yam, C.-K. Li, C.-L. Chan, Angew. Chem. Int. Ed. 1998, 37, 2857–
859; Angew. Chem. 1998, 110, 3041–3044.
2
[
[
2
12] M. A. Mansour, W. B. Connick, R. J. Lachicotte, H. J. Gysling, R. Eisenberg,
J. Am. Chem. Soc. 1998, 120, 1329–1330.
Chem. Eur. J. 2015, 21, 13888 – 13893
www.chemeurj.org
13892
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[
[
[
13] Y. Ma, C.-M. Che, H.-Y. Chao, X. Zhou, W.-H. Chan, J. Shen, Adv. Mater.
999, 11, 852–857.
14] R. L. White-Morris, M. M. Olmstead, A. L. Balch, J. Am. Chem. Soc. 2003,
25, 1033–1040.
15] a) G. Revol, T. McCallum, M. Morin, F. Gagosz, L. Barriault, Angew. Chem.
Int. Ed. 2013, 52, 13342–13345; Angew. Chem. 2013, 125, 13584–13587;
b) T. McCallum, E. Slavko, M. Morin, L. Barriault, Eur. J. Org. Chem. 2015,
[23] a) C. E. Whittle, J. A. Weinstein, M. W. George, K. S. Schanze, Inorg. Chem.
2001, 40, 4053–4062; b) G. Ramakrishna, T. Goodson III, J. E. Rogers-
Haley, T. M. Cooper, D. G. McLean, A. Urbas, J. Phys. Chem. C 2009, 113,
1060–1066.
[24] a) W. Lu, W.-M. Kwok, C. Ma, C. T.-L. Chan, M.-X. Zhu, C.-M. Che, J. Am.
Chem. Soc. 2011, 133, 14120–14135; b) C. Ma, C. T.-L. Chan, W.-M. Kwok,
C. M. Che, Chem. Sci. 2012, 3, 1883–1892.
1
1
8
1–85.
[25] a) J. R. Winkler, J. L. Marshall, T. L. Netzel, H. B. Gray, J. Am. Chem. Soc.
1986, 108, 2263–2266; b) V. M. Miskowski, G. L. Nobinger, D. S. Kliger,
G. S. Hammond, N. S. Lewis, K. R. Mann, H. B. Gray, J. Am. Chem. Soc.
1978, 100, 485–488.
[
16] J. Xie, S. Shi, T. Zhang, N. Mehrkens, M. Rudolph, A. S. K. Hashmi, Angew.
Chem. Int. Ed. 2015, 54, 6046–6050; Angew. Chem. 2015, 127, 6144–
6
148.
[
[
[
[
[
17] K. H. Leung, D. L. Phillips, M.-C. Tse, C.-M. Che, V. M. Miskowski, J. Am.
Chem. Soc. 1999, 121, 4799–4803.
18] a) H.-X. Zhang, C.-M. Che, Chem. Eur. J. 2001, 7, 4887–4893; b) Q.-J. Pan,
H.-X. Zhang, J. Phys. Chem. A 2004, 108, 3650–3661.
19] N. J. Turro, Modern Molecular Photochemistry; (University Science Book)
Mill Valley, CA, 1991.
20] W. Gawelda, A. Cannizzo, V.-T. Pham, F. van Mourik, C. Bressler, M. Cher-
gui, J. Am. Chem. Soc. 2007, 129, 8199–8206.
[26] R. M. van der Veen, A. Cannizzo, F. van Mourik, A. Vl cˇ ek Jr., M. Chergui,
J. Am. Chem. Soc. 2011, 133, 305–315.
[27] A. Steffen, M. G. Tay, A. S. Batsanov, J. A. K. Howard, A. Beeby, K. Q.
Vuong, X.-Z. Sun, M. W. George, T. B. Marder, Angew. Chem. Int. Ed.
2010, 49, 2349–2353; Angew. Chem. 2010, 122, 2399–2403.
[28] G. S. M. Tong, P. K. Chow, C.-M. Che, Angew. Chem. Int. Ed. 2010, 49,
9206–9209; Angew. Chem. 2010, 122, 9392–9395.
[29] M. L. Horng, J. A. Gardecki, A. Papazyan, M. Maroncelli, J. Phys. Chem.
1995, 99, 17311–17337.
21] a) N. H. Damrauer, G. Cerullo, A. Yeh, T. R. Boussie, C. V. Shank, J. K.
McCusker, Science 1997, 275, 54–57; b) A. T. Yeh, C. V. Shank, J. K.
McCusker, Science 2000, 289, 935–938.
ˇ
ˇ
Received: August 3, 2015
[
22] A. Cannizzo, A. M. Blanco-Rodríguez, A. E. Nahhas, J. Sebera, S. Zµlis, A.
Vl cˇ ek Jr., M. Chergui, J. Am. Chem. Soc. 2008, 130, 8967–8974.
Published online on September 4, 2015
Chem. Eur. J. 2015, 21, 13888 – 13893
www.chemeurj.org
13893
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim