Gold catalysis: AuCl-induced polymerization of styrene and n-butylvinylether

Article abstract of DOI:10.1071/CH13562

Polymerization of styrene, 4-methoxystyrene, and n-butylvinylether was achieved using simple AuCl as catalyst and AgPF₆ as cocatalyst. High molecular weights and low polydispersity indices were obtained. Evidence for a cationic mechanism was obtained by reactions with nucleophiles. The mechanistic study also indicates a living polymerization with the gold(I)-alkene complex as the resting species. CSIRO 2014.

Full text of DOI:10.1071/CH13562

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2014, 67, 500–506
Full Paper
http://dx.doi.org/10.1071/CH13562
Gold Catalysis: AuCl-induced Polymerization of Styrene
and n-Butylvinylether
A B C F
, , ,
C
A
A. Stephen K. Hashmi,
Sascha Schafer, Verena Goker,
¨ ¨
D
D
E
Claus D. Eisenbach, Klaus Dirnberger, Zhirong Zhao-Karger,
E
and Patrick Crewdson
A
Organisch-Chemisches Institut, Ruprecht-Karls-Universitat Heidelberg,
¨
Im Neuenheimer Feld 270, 69120 Heidelberg, Germany.
Chemistry Department, Faculty of Science, King Abdulaziz University,
Jeddah 21589, Saudi Arabia.
B
C
Institut fur Organische Chemie, Universitat Stuttgart, Pfaffenwaldring 55,
¨
¨
70569 Stuttgart, Germany.
D
E
Institut fur Polymerchemie, Universitat Stuttgart, Pfaffenwaldring 55,
¨
¨
70569 Stuttgart, Germany.
Catalysis Research Laboratory (CaRLa) Heidelberg, Universitat Heidelberg
¨
& BASF Im Neuenheimer Feld 584, 69120 Heidelberg, Germany.
Corresponding author. Email: hashmi@hashmi.de
F
Polymerization of styrene, 4-methoxystyrene, and n-butylvinylether was achieved using simple AuCl as catalyst and
AgPF6 as cocatalyst. High molecular weights and low polydispersity indices were obtained. Evidence for a cationic
mechanism was obtained by reactions with nucleophiles. The mechanistic study also indicates a living polymerization
with the gold(I)–alkene complex as the resting species.
Manuscript received: 16 October 2013.
Manuscript accepted: 29 January 2014.
Published online: 24 February 2014.
possibility of b-H eliminations at alkylgold(I) species[6]
suggested that, at least for gold(I), a mechanism based on
carbocations rather than a mechanism based on migratory
insertion should be involved in the polymerization.
So far, neither gold(I) catalysts nor simple gold(III) halides
and their combination with a silver(I) salt as halide scavenger
have been tested. Pioneering work of He et al.[7] on the activa-
tion of alkenes by gold(III) catalysts focussed on the addition of
heteronucleophiles rather than the addition of another alkene
molecule in the sense of a polymerization. Cationic gold(I) can
be considered to be a ‘super proton’ by the isolobal analogy;[8]
thus, testing gold(I) in homogeneous gold-catalyzed polymeri-
zation seemed quite promising.
Introduction
Homogeneous and heterogeneous polymerization reactions
were carried out using a variety of main group Lewis acids like
Al, Si, Ge, Ga, In, Bi, Sn, and some transition metals like Ti, Zr,
Hf, Fe, Co, and Zn.[1] Further, ate complexes of Nb, Mo, and W
were successfully tested in the cationic polymerization of iso-
butyl vinyl ethers.[2] In the few last years, homogeneous gold-
catalyzed reactions have attracted significant attention. Either
simple gold salts or gold complexes in oxidation state I and III
with a vast variety of different ligands have been employed for a
wide range of different reactions.[3]
Although oxidative polymerization reactions using some
gold species in stoichiometric and sub-stoichiometric amounts,
mostly in combination with small heterocycles, are known and
have been investigated,[4a,b] so far only one publication dealing
with gold-catalyzed polymerizations has appeared in the litera-
ture, where Nolan et al. reported that the carbene gold(III)
complex IPrAuBr3 (IPr ¼ 1,3-bis(diisopropylphenyl)imidazol-
2-ylidene) in combination with NaBAr4 (Ar ¼ 3,5-bis
(trifluoromethyl)phenyl) as a halide scavenger is catalytically
active in the polymerization of styrene in the solvents toluene or
dichloromethane (DCM).[4c] No mechanistic research on the
gold-catalyzed olefin polymerization was carried out and thus
the mechanistic picture remained unclear. Recent investigation
into possible insertion reactions into C–Au bonds[5] and into the
We now report our results obtained in the field of gold-
catalyzed olefin polymerization.
Results and Discussion
During a close investigation of the intermolecular gold-catalyzed
hydrocarboxylation of styrene (1), under reaction conditions
similar to the literature conditions[9] for the reaction of benzoic
acid (2) and phenyl acetylene, using the new catalyst 3, we
detected that the products 4 and 5 were formed in excellent yield
(Scheme 1). After heating for 15h, the NMR spectra indicated an
almost complete consumption of styrene. From additional GC
and gas chromatography–mass spectrometry (GC-MS) data, the
Journal compilation Ó CSIRO 2014
www.publish.csiro.au/journals/ajc

Polymers
501
CD3
D
D
D
D
O
1 mol-% 3
ϩ
OH
ϩ
[D8]toluene
85ЊC, 15h
1
2
4
5
4:5 ϭ 1.3:1
98% overall
Scheme 1. Hydroarylation and dimerization of styrene by catalyst 3.
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
P
Au
Cl
P
Au
Cl
2 AgNTf2
CH2Cl2
2 AuCl(SMe2)
P
P
P
P
Ϫ2 SMe2
Au
Au
6
NTf2 NTf2
Ϫ2 AgCl
7
3
Scheme 2. Preparation of the gold(I) complexes 7 and 3.
C7
C4
C17
C5
C8
P1
O1
C6
C9
Au2
C12
C15
C14
C11
N1
P2
O2
C1
P1
S1
C3
C10
P2
S2
Au1
O4
C2
Au1
C16
O3
C13
F2
O5
O6
N2
F3
O8
Au2
S3
Cl1
O7
Cl2
Fig. 2. ORTEP plot of [(dtbpm)Au2(NTf2)2] (3) (dtbpm, di-tert-butylpho-
sphino)methane) in the solid state. For clarity, all hydrogen atoms are
˚
omitted. Selected bond distances (A): Au(1)–Au(2) 3.587(9), Au(1)–N(2)
Fig. 1. ORTEP plot of [(AuCl)2dtbpm] (7) (dtbpm, di-tert-butylpho-
sphino)methane) in the solid state. For clarity, all hydrogen atoms are
˚
omitted. Selected bond distances (A): Au(1)–Au(2) 3.2101(4), Au(1)–P(1)
2.140(7), Au(1)–P(2) 2.246(3), Au(2)–N(1) 2.100(8), Au(2)–P(1) 2.236(2).
Selected bond angles (8): N(2)–Au(1)–P(2) 169.5(2), N(1)–Au(2)–P(1)
171.9(2), P(1)–C(9)–P(2) 123.5(5).
2.2447(17), Au(1)–Cl(1) 2.2956(18), Au(2)–P(2) 2.2439(17), Au(2)–Cl(2)
2.2836(17). Selected bond angles (8): P(1)–Au(1)–Cl(1) 174.94(8), P(1)–
Au(1)–Au(2) 85.51(4), Cl(1)–Au(1)–Au(2) 99.53(6), P(2)–Au(2)–Cl(2)
175.75(7), P(2)–Au(2)–Au(1) 86.19(5), Cl(2)–Au(2)–Au(1) 97.93(5),
P(1)–C(9)–P(2) 120.1(3).
Complex 7 was formed in 51 % yield, the dinuclear gold(I)
bis-NTf2 complex 3 in quantitative yield. Crystallization from
DCM/hexane delivered single crystals suitable for an X-raysingle-
crystal structure analysis of both complexes (Figs 1 and 2).[13]
The crystal structure of 7 exhibits an aurophilic AuꢁꢁꢁAu inter-
products were identified as 4, which results from intermolecular
hydroarylation by the solvent, and 5, the known[10] product of an
electrophilic dimerization of styrene. Owing to the ‘open-flask’
conditions typical for gold-catalyzed reactions and the presence
of the carboxylic acid, in 4 the deuterium atom that was substi-
tuted in the tolyl ring was not incorporated into the methyl group
attached to the benzyl position of 4.
Catalyst 3 was prepared from the known ligand 6[11] and
AuCl(SMe2), and subsequent exchange of Clꢀ by NTf2ꢀ in 7 in
analogy to the procedure of Gagosz (Scheme 2).[12]
action[8d,14] with a length of 3.21 A. Owing to the interaction
˚
between two Au atoms, the structure of 7 shows a deviation of
the P–Au–Cl angles from linearity and the two P–Au–Cl
fragments are oriented almost perpendicular to each other.
˚
Compared with 7, the AuꢁꢁꢁAu distance in complex 3 at 3.59 A
is clearly beyond an attractive aurophilic interaction and the
P(1)–C–P(2) angle is enlarged by up to 38 as well. This is
presumably due to the repulsion of the sterically demanding
˚
NTf2 groups; the Au–N distances are 2.10 and 2.14 A.

502
A. S. K. Hashmi et al.
Table 1. Styrene polymerization under different reaction conditions and with different catalysts
Tol, Toluene; DCM, dichloromethane; hmpy, 2-hydroxymethylpyridine; Mw, weight-average molecular weight; Mn, number-average molecular weight; PDI,
polydispersity index; TON, turnover number (moles of substrate that one mole of catalyst can convert before becoming inactive); TOF, turnover frequency
(turnover per time unit); quant., quantitative; RT, room temperature
Entry Solvent Time [h] Temperature Catalyst
Isolated yield
Mw
Mn
PDI TON TOF Consistency
1
Tol
14
3
558C
RT
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
4 mol-% AuCl/AgPF6
1 mol-% AuCl/AgPF6
2 mol-% Ph3PAuNTf2
2 mol-% Ph3PAuNTf2
2 mol-% Ph3PAuCl/AgPF6
2 mol-% Ph3PAuCl/AgPF6
2 mol-% Ph3PAuCl/AgTFA
2 mol-% (2-hmpy)AuIIICl2
Au colloidal
quant.
quant.
quant.
quant.
90 %
48 %
–
960
16000
470 2.04
5700 2.81
50
50
50
25
90
24
–
3.6 Oil
2
DCM
DCM
DCM
DCM
Tol
16.7 Solid
16.7 Solid
8.3 Solid
4
3
ꢀ788C
ꢀ788C
ꢀ788C
558C
RT
210000 77700 2.70
160000 59600 2.68
81000 40500 2.00
3
3
5
3
30
Solid
6
22
72
3
630
360 1.75
1.1 Oil
7
Tol
–
–
–
–
–
–
–
–
8
DCM
Tol
ꢀ788C
408C
408C
558C
458C
458C
–
–
–
–
9
14
14
22
14
14
5 %
–
2300
1740 1.32
2.5
0.2 Solid
10
11
12
13
Tol
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Tol
–
Tol
–
Tol
2 mol-% AgPF6
–
Dimer 5 indicated an electrophilic C–C-bond forming step
comparable with the AuCl-catalyzed reactions (compare entry 6
and entry 1). Even heating and long reaction times could not
induce full conversion. In DCM at ꢀ788C, no reaction was
observed for 2 mol-% Ph3PAuCl/AgPF6 (entry 8). Colloidal
gold and (2-hmpy)AuIIICl[215] (2-hmpy ¼ 2-hydroxymethyl-
pyridine) were not catalytically active (entry 12 and entry 11
respectively). Using triflate as counter-anion of the silver salt led
to no conversion (entry 10); with this counter-ion, the catalyst
seems to be too unstable. The counter-anion may be too nucleo-
philic, causing the polymerization to stop at once. The important
control experiment with 2 mol-% AgPF6 and no gold complex
showed no conversion (entry 13).
between two alkenes; thus, we considered the gold(I) catalysts
to be efficient catalysts for olefin polymerization, too.
Subsequent screening of gold(I) complexes in homogeneous
catalytic polymerization were carried out with styrene as
monomer. In principle, styrene can be polymerized by
a free-radical, an anionic, and a cationic mechanism.[16]
Therefore, styrene monomer seemed to be ideally suited for
an initial screening.
In a typical reaction, the gold(I) complex was first sus-
pended in the solvent. Then AgPF6 was added and after 2 min of
stirring, the styrene monomer was added in one portion. After
completion of the polymerization reaction, the reaction mix-
ture was poured into methanol, and the product precipitated.
Data from the experiments employing styrene monomer are
compiled in Table 1.
Vinyl ether derivatives have been successfully employed for
kinetic studies of the mechanism of cationic polymerization.[1b]
Thus, we carried out kinetic studies of the polymerization
mechanism with n-butylvinylether (Table 2).
When using simple gold(I) chloride and AgPF6 in toluene as
solvent at 558C, a quantitative conversion of 1 was observed
within 14 h, but the weight-average and number-average mole-
cular weights, Mw and Mn respectively, were quite low (entry 1).
Replacing the solvent toluene with DCM resulted in an
,16-fold increase of Mw along with full conversion at room
temperature in only 3 h (entry 2) instead of requiring 14 h in
toluene at 558C. The reaction was exothermic, which might
cause a problem for the formation of longer polymer chains.
Therefore, the reaction mixture was cooled to ꢀ788C before
adding the monomer (entry 3). The isolated polymer showed an
Mw 13 times higher than the one for the same reaction carried out
at room temperature. Using 4 mol-% of catalyst (entry 4), Mw
and Mn remained quite high at 160000 and 59600 respectively,
but were a little lower than with 2 mol-% of catalyst (entry 3).
Switching to 1 mol-% of catalyst (entry 5) resulted in the lowest
Mw and Mn under these reaction conditions, and only ,90 %
conversion was reached. All observed polymers showed
a monomodal distribution as determined by gel permeation
chromatography (GPC).
For this new monomer, we first tested the influence of
different catalyst loadings under the reaction conditions previ-
ously used in the polymerization of styrene. For 1 and 2 mol-%
of catalyst, we observed a quantitative conversion. With 2 mol-%
catalyst, Mw and Mn were higher than with 1 mol-% catalyst.
With 4 mol-% catalyst, only 80 % conversion was observed;
also, the molecular weight of the resulting polymer was found to
be lower compared with that using 2 mol-% of catalyst (Fig. 3),
thus 2 mol-% catalyst loading is the preferred catalyst loading
with regard to conversion and molecular weight in the case of
butylvinyl ether. The dependency of monomer conversion and
of the molecular weight on the catalyst loading was found to be
similar for both n-butylvinylether and styrene.
We then tried to gain some mechanistic insight. For kinetic
studies, 2 mol-% of AuCl/AgPF6 were chosen. The reaction was
quenched after different reaction times by pouring the reaction
mixture into methanol. The resulting slurry was separated from
the solvent, dissolved in DCM, and precipitated once more by
adding methanol to separate the catalyst from the polymer.
In a short induction period of at least 2 min, the active catalyst
is formed and polymerization starts (Fig. 4). Smaller oligomers
are likely to be formed during this period. Because of their low
molecular weight, they could not be precipitated by adding
methanol. Evaporation of the solvent afforded no product. After
3 min, we observed 32 % monomer conversion to poly(n-butyl-
vinylether), and 90 % conversion after 4 min. In 4.5 min, nearly
Having established these conditions for the polymerization
of styrene with AuCl/AgPF6 (ꢀ788C in DCM with 2 mol-%
AuCl/AgPF6 from entry 3), we then tried to improve the catalyst
system, but none of the tested catalysts led to better results. The
Gagosz salt Ph3PAuNTf[212] or the corresponding precursor
Ph3PAuCl and AgPF6 as chloride scavenger produced an Mw

Polymers
503
all of the monomer was consumed and converted into polymeric
material, i.e. the reaction is fast (Fig. 3). In the cationic
polymerization of vinyl ethers using transition metal complexes,
the fastest polymerization reaction is described for FeCl3.[17]
Depending on the solvent used, the authors reported that the
reaction was finished within 1 s to 15 h with full conversion. The
reaction was carried out at 08C. If one considers the drastically
lower reaction temperature (ꢀ788C) that we applied for the
polymerization reaction, our results are comparable with the
results obtained with FeCl3. Nolan and coworkers reported that
their gold–carbene catalyst system needed 15 min at room
temperature for full conversion of styrene.
Interestingly, Mw and Mn kept growing while the polydisper-
sity index (PDI) decreased (from 2.4 to 1.4) for longer reaction
times, even after all the starting monomer was converted into
polymeric material (Figs 5 and 6). This observation gives rise to
the deduction that a living or at least a controlled cationic
polymerization is taking place. Obviously, the reaction system
stays active although all the monomer is consumed. The mono-
modal molecular weight distribution did not change. This
indicates that polymer chains join to produce highly polymeric
material.
Similar behaviour is observed in controlled radical poly-
merization reactions[18] and controlled group-transfer poly-
merizations.[19] Also in these types of reactions, the PDI
decreases as the reaction progresses. Finding this behaviour,
which is characteristic of controlled polymerization reactions,
led to the assumption that in the case of polymerization with
gold catalysts as initiators, the mechanism also proceeds in a
living or controlled manner (probably as the gold(I)–olefin
complex at the terminus of the growing chain, see species G in
Scheme 3).
A mechanistic picture is drawn in Scheme 3. A cationic
gold(I) complex reacts with a monomer to form the complex A.
The addition of n monomers results in the complex B containing
a polymeric chain. From this intermediate, one can surmise
several pathways. Reaction with a nucleophile forms product E.
Protodesauration of B leads to compound C, which could
react with an external nucleophile to form the final product D.
In the absence of nucleophiles, B could lose a proton to form
complex F, and the latter suffers protodesauration to form the
olefinic product G, which in the non-coordinating solvent will
almost certainly exist as the gold(I)–alkene complex. The reac-
tion product G carrying the coordinated olefin as the end group
thus could readily enter the previously discussed reaction
Mn v. mol-% catalyst
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0
1
2
3
4
5
mol-%
Fig. 3. Observed Mn (number-average molecular weight) at different
catalyst concentrations (n-butylvinylether, ꢀ788C).
As no methanol was added to the reaction mixture, the
solution stayed colourless to slightly orange. Shortly after the
reaction mixture was poured into methanol, a black or pinkish
slurry was observed, which is due to colloidal gold formed on
the surface of the polymer. This indicates that the catalyst stayed
active and was not reduced to Au0 as long as no nucleophile was
added to the reaction mixture.
In the previous publication on gold-initiated polymeriza-
tion,[4c] the authors suggested that simple protons could also be
the active species for the initiation of polymerization, which is
provided by the reaction of the gold catalyst with water present
in the mixture. We observed that the reaction is stopped by
adding a protic solvent like methanol. If protons were the active
initiator, the reaction should not be completely suppressed in the
presence of methanol.
Conversion of monomer
100%
80%
60%
40%
20%
0%
0
1
2
3
4
5
6
Time [min]
Fig. 4. Conversion of monomeric n-butylvinylether (2 mol-% catalyst,
ꢀ788C).
Table 2. Polymerization of n-butylvinylether at different reaction times and with different catalyst loadings
BVE, n-butylvinylether; Mw, weight average molecular weight; Mn, number average molecular weight; PDI, polydispersity index; TON, turnover number
(moles of substrate that one mole of catalyst can convert before becoming inactive); TOF, turnover frequency (turnover per time unit); quant., quantitative;
DCM, dichloromethane. All experiments were performed with n-butylvinylether in DCM at ꢀ788C
Entry
Time [h]
Temperature
Catalyst
Isolated yield
Mw
Mn
PDI
TON
TOF
Consistency
1
2
3
4
5
6
7
8
3
ꢀ788C
ꢀ788C
ꢀ788C
ꢀ788C
ꢀ788C
ꢀ788C
ꢀ788C
ꢀ788C
4 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
1 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
2 mol-% AuCl/AgPF6
80 %
99 %
quant.
32 %
90 %
97 %
quant.
quant.
52000
63000
48000
11000
20000
23000
26000
31000
37000
44000
5800
1.41
1.43
8.28
2.44
2.08
2.32
2.36
1.94
20
49.5
100
7.68
45
6.7
16.5
33.3
153.6
675
Oil
Oil
Oil
Oil
Oil
Oil
Oil
Oil
3
3
0.05
0.067
0.075
0.083
0.092
4500
9600
9900
48.5
50
646
11000
16000
602
50
543

504
A. S. K. Hashmi et al.
pathways to form highly polymeric material of higher molecular
weight with increasing reaction time.
For assignment of the terminal group of the polymer, the
reaction was quenched after 5-min reaction time by adding
trifluoroethanol (CF3CH2OH). Subsequent measurement of
19F NMR spectra of the resulting polymeric material showed a
triplet signal shifted to lower field compared with the 19F signal
of CF3CH2OH (see Supplementary Material). This shows that
the polymerization can be stopped by adding a nucleophile,
which proves the assumption of a cationic mechanism. This
result is not only interesting from a mechanistic but also from a
synthetic point of view. The terminal groups of the resulting
polymer are formed by the addition of a nucleophile. Quenching
with different nucleophiles at different reaction times provides
access to a large variety of different polymers exhibiting differ-
ent molecular weight distributions and terminal groups.
Mn versus conversion
100%
80%
60%
40%
20%
0%
0
2000 4000 6000 8000 10000 12000 14000 16000 18000
In the absence of a nucleophile, Mw and Mn still increase even
after all monomers have been converted into polymeric mate-
rial. This result is explained by, first, the abstraction of a proton
(from B to form F), and second, the participation of the
unsaturated end group in further polyreactions, resulting in
high-molecular-weight polymers. If the assumption is correct
that further participation of low-molecular-weight polymers
(such as G after protodesauration of F) leads to the formation
of longer chains, one would expect to find a decreasing PDI over
time. This is exactly what was observed during our experiments
(PDI decreased from 2.4 at 3 min to 1.4 after 3 h).
If the assumption of a cationic polymerization mechanism is
correct, with an electron-rich monomer the reaction rate must
increase, and the reaction must proceed quite readily even with
lower catalyst loadings. Thus 4-methoxystyrene was employed.
Polymerization was carried out under the same reaction condi-
tions (room temperature and at ꢀ788C) employing 2, 0.25 or
0.24 mol-% of the AuCl/AgPF6-catalyst (Table 3). The reaction
with 0.24 mol-% catalyst was finished within 4 min, and a
polymer of high molecular weight was obtained. The molecular
weight of all the polymeric materials obtained was 500000.
When the polymerization was carried out at room tempera-
ture (0.24 mol-% catalyst), a high-molecular-weight polymeric
material of monomodal distribution was obtained. In contrast
Mn
Fig. 5. Increase of Mn (number-average molecular weight) with conver-
sion (n-butylvinylether, 2 mol-% catalyst, ꢀ788C).
Mn
PDI
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
3.0
2.5
2.0
1.5
1.0
0.5
0
32
90
97
100
Conversion [%]
Fig. 6. Mn (number-average molecular weight), PDI (polydispersity
index) v. monomer conversion (n-butylvinylether, 2 mol-% catalyst,
ꢀ788C).
ϩ
AuL
ϩHϩ
H
High-molecular-
weight polymer
LAu
R
R
R
R
n
n
R
G
F
ϩ
ϩ
[LAu
]
ϪHϩ
Nu
NuH
LAu
ϩ
n
R
LAu
LAu
R
ϪHϩ
R
R
R
R
A
n
n
B
E
R ϭ
Ϫ[LAuϩ]
Ϫ[LAuϩ]
ϩHϩ
ϩHϩ
ϩ
Nu
NuH
H
H
MeO
R
R
ϪHϩ
R
R
n
n
OnBu
C
D
Scheme 3. Mechanistic pathways for cationic polymerization with gold.

Polymers
505
Table 3. Reaction of 4-methoxystyrene with AuCl/AgPF6
Mw, weight average molecular weight; Mn, number average molecular weight; PDI, polydispersity index; TON, turnover number
(moles of substrate that one mole of catalyst can convert before becoming inactive); TOF, turnover frequency (turnover per time unit);
quant., quantitative; RT, room temperature; n.d., not determined
Entry Solvent Time [h] Temperature Catalyst
Isolated yield Mw
Mn
PDI TON TOF Consistency
1
2
DCM
DCM
0.06
3
RT
0.24 mol-% AuCl/AgPF6 quant.
0.25 mol-% AuCl/AgPF6 91 %
500000 112000 n.d. 417 6944 Solid
ꢀ788C
500000 112000 n.d. 364 121 Solid or oil
4670
500000 112000 n.d.
1800 710 2.53
2270 2.06
3
DCM
3
ꢀ788C
2 mol-% AuCl/AgPF6
quant.
50
17 Solid
to this, the polymerization at ꢀ788C resulted in a bimodal
distribution with a low-molecular-weight fraction and a high-
molecular-weight fraction.
used. Chemical shifts are referenced to residual solvent protons.
Mass spectra were recorded on a Finnigan TSQ700 triple
quadrupole mass spectrometer. IR spectra were measured on a
Bruker Equinox 55 FT-IR (Fourier-transform) spectrometer.
GPC analysis was recorded on Viscotek TPA 302 and Waters
system with Waters 515 HPLC-pump, Waters 410 IR detector
and Waters 996 photo diode array (PDA). Column material was
crosslinked polystyrene (Styragel HR) with columns of separa-
tion ranges 500–30000, 5000–600000, 2ꢂ 105–4 ꢂ 107 g molꢀ1
and 0–1000, 100–5000, 500–20000, 500–30000 g molꢀ1 and
columns from Polymer Standards Service (500mm, 106, 105 and
There are two important points to consider: (i) the catalyst
system is very efficient and even lower catalyst loadings should
be possible, and (ii) the reaction is very sensitive to nucleophilic
impurities. The gold catalyst and the silver cocatalyst are very
hygroscopic but were handled under a normal atmosphere for
these results: with 2 mol-% catalyst, more water was added to
the reaction mixture. Either way, the high Mw and Mn achieved
in these experiments indicate a cationic mechanism for the
polymerization.
˚
1000A). The sample was filtered through PTFE (450-mm pore
size), THF was used for elution (flow rate 1 mLminꢀ1) at 258C
and calibration was done with polystyrene.
Conclusion
We established the use of a simple and efficient gold catalyst
for the polymerization of styrene, n-butylvinylether, and
4-methoxystyrene under different reaction conditions allowing
full monomer conversion and resulting in high-molecular-
weight polymers in a fast and controlled polymerization reac-
tion. Quenching the reaction at different times allowed control
of the average molecular weight and the terminal groups. The
experiments indicate a ‘living polymerization’ that may be
based on a gold(I)–alkene complex as the resting species.
No complicated ligands on the gold catalyst were needed for
the reaction. We found evidence for a cationic mechanism by
quenching the reaction with a nucleophile and by kinetic studies.
Mw and Mn for poly(styrene) were much higher than those under
the standard conditions recently reported for polymerization
with gold as an initiator. Only three superior examples have been
reported, but there more complicated gold(I)–N-heterocyclic
carbine (NHC) complexes were used.[4c]
Synthesis of 7
Bis(di-tert-butylphosphino)methane
(dtbpm)
(106 mg,
350 mmol) in DCM (10 mL) was added dropwise to AuCl(SMe2)
(206 mg, 700 mmol) in DCM (15 mL). Complex 7 (146 mg,
190 mmol, 54 %) was obtained as colourless crystals. dH
(200.1 MHz, CD2Cl2, 298 K) 1.52 (‘d’, J(H,P) 15.2, 36H, C
2
(CH3)3), 2.25 (t, J(H,P) 10.4, 2H, PCH2P). dC (50.3 MHz,
CD2Cl2, 298 K) 30.4 (t, J 2.21, CH3), 38.3 (t, J(C,P) 13.5,
PCH2P), 39.5 (s, C(CH3)3). dP (81.02 MHz, CD2Cl2, 298 K) 79.7
(s, PCH2P). nmax (KBr)/cmꢀ1 (s), 3001 (m), 2957 (s), 2901 (s),
2869 (w), 1630 (s), 1470 (s), 1394 (w), 1373 (m), 1175 (m), 810
(w), 769 (w), 705 (m), 693 (w), 598 (s), 523 (s), 496 (m), 468
(m), 448 (m), 425 (m), 418 (w). m/z (high-resolution MS elec-
trospray ionization (HRMS ESI)) 733.14628; calc. for MþClꢀ
[C17H38Au2ClP2] 733.14683.
3
Synthesis of 3
Complex 7 (385 mg, 500 mmol) was dissolved in DCM (10 mL),
and AgNTf2 (388 mg, 1.00 mmol) was added as a solid, resulting
in the instantaneous formation of a silver chloride precipitate.
The mixture was stirred for an additional 20 min. Removal of the
silver chloride salt by filtration resulted in a colourless to light-
purple solution. Complex 3 (629 mg, 500 mmol) was obtained in
quantitative yield by removing the solvent under vacuum. dH
Experimental
General Methods
Chemicals (Aldrich, Fluka, Strem) were used without
further purification unless otherwise indicated. (Ph3P)
AuCl was recrystallized from DCM before use. Styrene and
4-methoxystyrene were filtered through Al2O3, stirred over-
night over CaH2 and distilled under a nitrogen atmosphere
before use. n-Butylvinylether was washed twice with water,
stirred overnight over Na and distilled under a nitrogen atmo-
sphere before use. Toluene was dried over Na/benzophenone
before use. Dry DCM under a protecting atmosphere with a
crown cap was bought and used without further purification. All
reactions were carried out using the Schlenk technique under a
nitrogen atmosphere. NMR spectra were recorded on Bruker
AvanceII 250 (250 MHz) and Bruker Avance 200 spectro-
meters. For the 13C NMR spectra, phosphorous decoupling was
3
(200.1 MHz, CD2Cl2, 298 K) 1.52 (‘d’, J(H,P) 16.4, 36H, C
2
(CH3)3), 2.28 (t, J(H,P) 11.2, 2H, PCH2P). dC (50.3 MHz,
CD2Cl2, 298 K) 30.2 (t, J 2.21, CH3), 39.3 (t, J(C,P) 13.8,
PCH2P), 39.6 (s, C(CH3)3, Cquart.), 121.1 (q, J(C–F) 321.4). dP
(81.02 MHz, CD2Cl2, 298 K) 74.2 (s, PCH2P). nmax (KBr)/cmꢀ1
3442 (s), 2959 (w), 1629 (w), 1470 (w), 1399 (s), 1349 (s), 1332
(s), 1201 (s), 1140 (s), 1061 (m), 1022 (w), 796 (w), 768 (w), 742
(w), 706 (w), 692 (m), 649 (m), 600 (w), 575 (w), 515 (w).
m/z (HRMS ESI) 978.09472; calc. for MþNSO2CF3
[C19H38Au2F6NO4P2S2] 978.09527.

506
A. S. K. Hashmi et al.
Procedure 1: Polymerization with (Ph3P)AuNTf2, (2-
Hydroxymethylpyridine)AuIIICl2, Au colloidal, AgPF6
(g) A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180. doi:10.1021/
CR000436X
[4] (a) C. Della Pina, E. Falletta, M. Rossi, Catalysis 2010, 22, 279.
doi:10.1039/9781847559630-00279
The catalyst was dissolved or suspended (10 ꢂ vol. solvent : vol.
substrate) in a Schlenk flask under a nitrogen atmosphere. Sty-
rene was added in one portion. The reaction mixture was heated
to the temperature given in Table 1 and stirred. Then the reaction
mixture was poured into methanol (10 ꢂ vol. methanol : vol.
solvent). The solution was decanted, the residue was dissolved
in DCM, and again poured into methanol. After decantation, the
rest of the solvent was evaporated under reduced pressure to give
the polymer material.
(b) L. Sun, Y. Shi, B. Li, L. Chu, Z. He, J. Liu, Colloids Surf.
A Physicochem. Eng. Asp. 2012, 397, 8. doi:10.1016/J.COLSURFA.
2012.01.008
(c) J. Urbano, A. J. Hormigo, P. de Fre´mont, S. P. Nolan,
M. M. D´ıaz-Requejo, P. J. Pe´rez, Chem. Comm. 2008, 759.
doi:10.1039/B716145J
[5] A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, J. Organomet.
Chem. 2009, 694, 592. doi:10.1016/J.JORGANCHEM.2008.11.054
[6] G. Klatt, R. Xu, M. Pernpointner, L. Molinari, T. Q. Hung,
F. Rominger, A. S. K. Hashmi, H. Ko¨ppel, Chem. Eur. J. 2013, 19,
3954. doi:10.1002/CHEM.201203043
[7] C.-G. Yang, C. He, J. Am. Chem. Soc. 2005, 127, 6966. doi:10.1021/
JA050392F
[8] (a) R. Hoffmann, Angew. Chem. Int. Ed. Engl. 1982, 21, 711.
doi:10.1002/ANIE.198207113
Procedure 2: Polymerization with Ph3PAuCl/AgPF6,
Ph3PAuCl/AgTFA or AuCl/AgPF6
In a Schlenk flask under a nitrogen atmosphere, the gold catalyst
was suspended (10 ꢂ vol. solvent: vol. substrate) and then the
silver salt was added. The mixture was stirred at room temper-
ature for 2 min. If required, the mixture was then cooled to
ꢀ788C. Afterwards, the substrate was added at once. The
reaction mixture was stirred at the temperatures given in Tables
1–3. Then the reaction mixture was poured into methanol
(10 ꢂ methanol : vol. solvent). The solution was decanted, the
residue was dissolved in DCM and again poured into methanol.
After decantation, the rest of the solvent was evaporated under
reduced pressure to give the polymeric material.
(b) K. P. Hall, D. M. P. Mingos, Prog. Inorg. Chem. 1984, 32, 237.
doi:10.1002/9780470166338.CH3
(c) H. Schmidbaur, Chem. Soc. Rev. 1995, 24, 391. doi:10.1039/
CS9952400391
(d) P. Pyykko¨, Angew. Chem. Int. Ed. 2004, 43, 4412 (especially pp.
4423–4424). doi:10.1002/ANIE.200300624
(e) L. G. Kuz’mina, A. A. Bagatur’yants, A. V. Churakova,
J. A. K. Howard, Chem. Commun. 2001, 1394. doi:10.1039/
B102938J
[9] E. Genin, P. Y. Toullec, S. Antoniotti, C. Brancour, J. P. Genet,
V. Michelet, J. Am. Chem. Soc. 2006, 128, 3112. doi:10.1021/
JA056857J
[10] A. Sen, T. W. Lai, R. R. Thomas, J. Organomet. Chem. 1988, 358, 567.
doi:10.1016/0022-328X(88)87104-3
Supplementary Material
The 19F NMR spectrum for assignment of the terminal group of
the polymer is available on the Journal’s website.
[11] (a) P. Hofmann, H. Heiss, G. Mu¨ller, Z. Naturforsch. B 1987, 42, 395.
(b) P. Hofmann, H. Heiss, Ger. Offen. DE 4134772, A1 35 19920507
1992; see also Chemical Abstracts Number 117:171685.
[12] N. Me´zailles, L. Ricard, F. Gagosz, Org. Lett. 2005, 7, 4133.
doi:10.1021/OL0515917
Acknowledgements
Zhirong Zhao-Karger and Patrick Crewdson worked at CaRLa, Heidelberg
University, cofinanced by the University of Heidelberg, the State of Baden-
Wu¨rttemberg, and BASF SE.
[13] CCDC 677699 (3) and 677698 (7) contain the supplementary crystal-
lographic data for this paper. These data can be obtained online free of
charge (or from the Cambridge Crystallographic Data Centre, 12,
Union Road, Cambridge CB2 1EZ, UK; fax: (þ44) 1223 336033; or
deposit@ccdc.cam.ac.uk).
References
[1] (a) J. V. Crivello, S. K. Rajaraman, J. Polym. Sci. A Polym. Chem.
1997, 35, 1593. doi:10.1002/(SICI)1099-0518(199706)35:8,1593::
AID-POLA28.3.0.CO;2-1
[14] P. Pyykko¨, Chem. Rev. 1997, 97, 597. doi:10.1021/CR940396V
[15] A. S. K. Hashmi, J. P. Weyrauch, M. Rudolph, E. Kurpejovic, Angew.
Chem. Int. Ed. 2004, 43, 6545. doi:10.1002/ANIE.200460232
[16] D. B. Priddy, Adv. Polym. Sci. 1994, 111, 67. doi:10.1007/
BFB0024127
[17] A. Kanazawa, Y. Hirabaru, S. Kanaoka, S. Aoshima, J. Polym. Sci.
A Polym. Chem. 2006, 44, 5795. doi:10.1002/POLA.21632
[18] F. Lecolley, C. Waterson, A. J. Carmichael, G. Mantovani,
S. Harrisson, H. Chappell, A. Limer, P. Williams, K. Ohno,
D. M. Haddleton, J. Mater. Chem. 2003, 13, 2689. doi:10.1039/
B303759B
(b) S. Aoshima, T. Yosida, A. Kanazawa, S. Kanaoka, J. Polym. Sci.
A Polym. Chem. 2007, 45, 1801. doi:10.1002/POLA.21995
[2] E. Ihara, N. Yoshida, J. I. Ikeda, T. Itoh, K. Inoue, J. Polym. Sci.
A Polym. Chem. 2006, 44, 2636. doi:10.1002/POLA.21383
[3] (a) G. Dyker, Angew. Chem. Int. Ed. 2000, 39, 4237. doi:10.1002/
1521-3773(20001201)39:23,4237::AID-ANIE4237.3.0.CO;2-A
(b) A. S. K. Hashmi, Gold Bull. 2003, 36, 3. doi:10.1007/BF03214859
(c) A. S. K. Hashmi, Gold Bull. 2004, 37, 51. doi:10.1007/BF03215517
(d) A. Hoffmann-Ro¨der, N. Krause, Org. Biomol. Chem. 2005, 3, 387.
doi:10.1039/B416516K
(e) A. S. K. Hashmi, Angew. Chem. Int. Ed. 2005, 44, 6990.
doi:10.1002/ANIE.200502735
(f) A. S. K. Hashmi, G. Hutchings, Angew. Chem. Int. Ed. 2006, 45,
7896. doi:10.1002/ANIE.200602454
[19] O. W. Webster, W. R. Hertler, D. Y. Sogah, W. B. Farnham,
T. V. Rajan Babu, J. Am. Chem. Soc. 1983, 105, 5706. doi:10.1021/
JA00355A039