Rate and mechanism of the oxidative addition of a silylborane to Pt 0 complexes - Mechanism for the Pt-catalyzed silaboration of 1,3-cyclohexadiene

Article abstract of DOI:10.1002/ejic.200800285

The chemical reduction of Pt(acac)₂ by DIBALH in the presence of phosphanes, which is used to generate active Pt⁰ complexes in the Pt-catalyzed silaboration of cyclohexadiene by 2-(dimethylphenylsilyl)-4,4,5,5- tetramethyl-1,3,2-dioxaborolane (1) leading to the 1,4-silaborated product, was mimicked by the electrochemical reduction of Pt(acac)₂ in the presence of 2 equiv. of PR₃ (R = Ph, nBu). The electrochemical reduction generates free acac anions and neutral Pt⁰-(PR ₃)₂ complexes. The kinetics of the oxidative addition of bromobenzene (used first as a model molecule) and silylborane 1 to the Pt ⁰ complexes was investigated and the rate constants determined. Pt⁰(PnBu₃)₂ is much more reactive than Pt ⁰(PPh₃)₂ towards 1. From the electrochemical study, it emerges that the acac anions released in the reduction of Pt(acac)₂ do not coordinate to the Pt⁰(PR ₃)₂ complexes. Consequently, the rate of the oxidative addition of 1 to Pt⁰(PR₃)₂, generated either by the electrochemical reduction or by the chemical reduction by DIBALH, is not affected by the acac anions and a posteriori not by aluminum cations. The oxidative addition and the further step of the catalytic cycle [insertion of the diene into the Pt-B bond of the Si-Pt-B complex generated in the oxidative addition, with formation of the (η³-allyl)Pt-Si complex] were monitored by NMR spectroscopy. Pt⁰ and Pt^II complexes involved in the catalytic cycle were characterized. The oxidative addition is faster when the ligand is PMe₂Ph relative to that obtained with PPh₃, in agreement with the electrochemical data. No reductive elimination within the (η³-allyl)Pt-Si complex is observed when the ligand is PMe₂Ph, whereas reactions in the presence of PPh ₃ proceeded to give the final product. As a consequence, PPh ₃ is a better ligand than PMe₂Ph for the catalytic reaction, as observed experimentally. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200800285

FULL PAPER
DOI: 10.1002/ejic.200800285
Rate and Mechanism of the Oxidative Addition of a Silylborane to Pt0
Complexes – Mechanism for the Pt-Catalyzed Silaboration of
1,3-Cyclohexadiene
Guillaume Durieux,[a] Martin Gerdin,[b] Christina Moberg,*[b] and Anny Jutand*[a]
Keywords: Platinum / Silaboration / Reaction mechanisms / Kinetics
The chemical reduction of Pt(acac)2 by DIBALH in the pres-
ence of phosphanes, which is used to generate active Pt0
complexes in the Pt-catalyzed silaboration of cyclohexadiene
by 2-(dimethylphenylsilyl)-4,4,5,5-tetramethyl-1,3,2-dioxa-
borolane (1) leading to the 1,4-silaborated product, was mim-
icked by the electrochemical reduction of Pt(acac)2 in the
presence of 2 equiv. of PR3 (R = Ph, nBu). The electrochemi-
cal reduction generates free acac anions and neutral Pt0-
(PR3)2 complexes. The kinetics of the oxidative addition of
bromobenzene (used first as a model molecule) and silylbo-
rane 1 to the Pt0 complexes was investigated and the rate
constants determined. Pt0(PnBu3)2 is much more reactive
than Pt0(PPh3)2 towards 1. From the electrochemical study, it
emerges that the acac anions released in the reduction of
chemical reduction by DIBALH, is not affected by the acac
anions and a posteriori not by aluminum cations. The oxida-
tive addition and the further step of the catalytic cycle [inser-
tion of the diene into the Pt–B bond of the Si–Pt–B complex
generated in the oxidative addition, with formation of the
(η3-allyl)Pt–Si complex] were monitored by NMR spec-
troscopy. Pt0 and PtII complexes involved in the catalytic cy-
cle were characterized. The oxidative addition is faster when
the ligand is PMe2Ph relative to that obtained with PPh3, in
agreement with the electrochemical data. No reductive eli-
mination within the (η3-allyl)Pt–Si complex is observed when
the ligand is PMe2Ph, whereas reactions in the presence of
PPh3 proceeded to give the final product. As a consequence,
PPh3 is a better ligand than PMe2Ph for the catalytic reaction,
Pt(acac)2 do not coordinate to the Pt0(PR3)2 complexes. Con- as observed experimentally.
sequently, the rate of the oxidative addition of 1 to Pt0(PR3)2,
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
generated either by the electrochemical reduction or by the
Germany, 2008)
hexadiene.[5] In order to be able to further develop this type
of processes, detailed knowledge about the reaction mecha-
nism is essential.
Introduction
Additions of element–element linkages to unsaturated
substrates catalyzed by the platinum group metals consti-
tutes a synthetically versatile process whereby compounds
containing two equal or different reactive groups are ob-
tained.[1] Among the additions studied, silaborations are
particularly interesting, as they provide adducts containing
both silicon and boron functionalities,[2] which serve as
powerful synthetic intermediates.[3] The silaboration of 1,3-
dienes was studied by Ito, Suginome, and coworkers, who,
using a nickel catalyst, found conditions for selective forma-
tion of 1,4-functionalized Z adducts from linear dienes and
cis-1,4 adducts from cyclic dienes [Equation (1)].[4] We later
found that Pt0 in combination with chiral phosphoramidite
ligands catalyzed enantioselective silaborations of 1,3-cyclo-
(1)
The mechanism of the related silaboration of alkynes
[Equation (2)] has been investigated by Ozawa et al. by me-
ans of NMR spectroscopy.[6]
(2)
[a] Ecole Normale Supérieure, Département de Chimie, CNRS
24, Rue Lhomond, 75231 Paris Cedex 5, France
Fax: +33-1-44322402
The reaction was shown to follow the generally accepted
catalytic cycle proceeding by three main steps: (i) oxidative
addition of the silylborane to a Pt0 complex leading to cis-
silyl–PtII–boryl complexes, (ii) insertion of the acetylene
into the Pt–B bond of the silyl–PtII–boryl complex, and (iii)
reductive elimination. Although the carbopalladation and
reductive elimination steps were investigated in full detail,
including the determination of the rate constants of each
E-mail: Anny.Jutand@ens.fr
[b] KTH School of Chemical Science and Engineering, Organic
Chemistry
10044 Stockholm, Sweden
Fax: +46-8-791-2333
E-mail: kimo@kth.se
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
4236
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 4236–4241

Mechanism for the Pt-Catalyzed Silaboration of 1,3-Cyclohexadiene
step, the mechanism and rate of the oxidative addition of and potential of Pt(acac)2 were not affected, but a new oxi-
silylborane 1 to Pt0 complexes were not investigated due to dation peak was detected on the reverse scan at Ep
=
O1
too fast reactions which could not be followed by NMR +0.04 V vs. SCE (Figure 1b). No oxidation peak was ob-
spectroscopy. In the catalytic reactions of 1, transient Pt0 served when the cyclic voltammetry was first performed
complexes were generated in situ by reduction of Pt(acac)2 towards oxidation potentials. The oxidation peak current at
with DIBALH in the presence of the appropriate phos- O1 doubled in the presence of a second equivalent of PPh3,
phane ligands.[5] Some questions arise. What is the structure whereas R1 remained unaffected. Consequently, Pt(acac)2
of the reactive Pt0 complex? Do the acac anions stabilize did not coordinate to PPh3 at room temperature, because
the Pt0 moiety by forming anionic complexes? Do alumi- its reduction peak did not change in the presence of PPh3 (1
num cations delivered by the chemical reductant DIBALH or 2 equiv.). This was confirmed by 31P NMR spectroscopy
play a role in the oxidative addition? Because electrochemis- performed on a mixture of Pt(acac)2 and 2 equiv. of PPh3
try can easily mimic chemical reduction processes, Pt0 com- in DMF; only the singlet from free PPh3 was observed at δ
plexes can be generated by the electrochemical reduction of = –5.35 ppm.
Pt(acac)2 in the presence of phosphanes, and the oxidative
The oxidation peak at O1 disappeared in the presence of
addition of the electrogenerated Pt0 complexes can be fol- an excess amount of PhBr, attesting that a Pt0 complex,
lowed by electrochemical techniques.[7] We report herein oxidized at O1, was generated in the electrochemical re-
data on the rate and mechanism of the oxidative addition duction of Pt(acac)2 performed in the presence of PPh3.
of bromobenzene (used first as a model molecule) and that This Pt0 complex was stabilized by two PPh3 as in Pt0-
of a silylborane, 2-(dimethylphenylsilyl)-4,4,5,5-tetramethyl- (PPh3)2,[8a] as its oxidation peak current at O1 doubled
1,3,2-dioxaborolane (1), to Pt0 complexes ligated by phos- upon the addition of a second equivalent of PPh3 (Fig-
phane ligands (PPh3 or PnBu3). The oxidative addition and ure 1b, Scheme 1). Further addition of PPh3 (1 equiv.) did
insertion steps of the catalytic cycle were also followed by not modify the oxidation peak potential significantly. How-
1H and 31P NMR spectroscopy starting from Pt0 prepared ever, the peak was slightly broader, suggesting the forma-
in situ by reduction of Pt(acac)2 with DIBALH.
tion of different Pt0 species, as Pt0(PPh3)3 and Pt0-
(PPh3)2 in equilibrium with the ligand.[8] Because the
Pt0(PPh3)2 complex is reported to be unstable at long
times[8a] (it is transiently generated in cyclic voltammetry at
short times, t Ͻ 5.5 s), an exhaustive electrolysis of a solu-
tion of Pt(acac)2 (0.02 mmol) was performed in DMF
(12 mL) in the presence of 3 equiv. of PPh3 at a potential
of –2 V (divided cell equipped with a 4-cm2 gold grid cath-
ode). The electroreduction consumed two electrons per mol.
At the end of the electrolysis, the reduction peak of
Pt(acac)2 was no longer observed when the cyclic voltam-
metry was performed towards reduction potentials (from
–0.4 to –2.5 V). An oxidation peak was observed at Ep =
+0.04 V when the CV was performed towards oxidation po-
tentials first (from –0.4 to +0.3 V), that is, at the same oxi-
dation potential as that of O1 observed by CV when
Pt(acac)2 was reduced in the presence of PPh3 (3 equiv.).
Consequently, Pt(acac)2 was reduced by two electrons in the
electrolysis and its reduction performed in the presence of
3 equiv. of PPh3 provided the same complex as the one ob-
served by cyclic voltammetry when Pt(acac)2 was reduced
in the presence of PPh3 (3 equiv.), that is, Pt0-
(PPh3)3.[8b]
Results and Discussion
Structure and Reactivity of Pt0 Complexes Generated in the
Electrochemical Reduction of Pt(acac)2 in the Presence of
PR3 (R = Ph, nBu)
The electrochemical properties of Pt(acac)2 (2 m) were
first investigated in DMF containing nBu4NBF4 (0.3 ).
Pt(acac)2 exhibited a reduction peak at Ep = –2.05 V vs.
R1
SCE at a gold disk electrode (Figure 1a). Only one tiny oxi-
dation peak was detected on the reverse scan, close to the
oxidation peak of acac anions, establishing that the Pt0
complex generated in the reduction of Pt(acac)2 was not
stable or made aggregates that could not be detected on the
reverse scan.
Figure 1. (a) Reduction of Pt(acac)2 (2 m) in DMF (containing
nBu4NBF4, 0.3 ) at a steady gold disk electrode (d = 1 mm) with
a scan rate v = 0.5 Vs–1, at 27 °C; (b) reduction of Pt(acac)2 (2 m)
in the presence of PPh3
conditions as in (a).
Scheme 1.
(
____) 2 m; (- - -) 4 m, under the same
When cyclic voltammetry (CV) was performed in the
A second oxidation peak was observed on the reverse
presence of 1 equiv. of PPh3, the reduction peak current scan at EpO2 = +0.42 V vs. SCE when Pt(acac)2 was reduced
Eur. J. Inorg. Chem. 2008, 4236–4241
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4237

G. Durieux, M. Gerdin, C. Moberg, A. Jutand
FULL PAPER
in the presence of 2 equiv. of PPh3 (Figure 1b). The oxi-
dation peak O2 was assigned to the oxidation of the acac
anions released in the reduction process by comparison
with the oxidation peak potential of Zn(acac)2, observed
independently under the same conditions. The oxidation
peak current of the acac anion released in the reduction
process was not modified when Pt(acac)2 was reduced in
the presence of 1 or 2 equiv. of PPh3 (see O2 in Figure 1b).
It is then established that the two acac anions of Pt(acac)2
were quantitatively released in the two-electron reduction
of Pt(acac)2. Consequently, the electrogenerated Pt0(PPh3)2
is a neutral complex that is not ligated by the acac anions
(Scheme 1).
Figure 2. Kinetics of the oxidative addition of PhBr (᭹: 0.02 ; ϫ:
0.04 ; o: 0.08 ; ᭢: 0.1 ; +: 0.2 ; ᭡: 0.4 ) to Pt0(PPh3)2 (2 m)
generated by the electrochemical reduction of Pt(acac)2 (2 m) in
the presence of 2 equiv. of PPh3 in DMF at 27 °C. Plot of the molar
fraction x of Pt0(PPh3)2 vs. log([PhBr]0/v). x = i/i0 i: oxidation peak
current of Pt0(PPh3)2 at O1 in the presence of PhBr (in the range
0.02–0.4 ); i0: oxidation peak current of Pt0(PPh3)2 in the absence
of PhBr, at the same scan rate v. Oxidation current intensities were
measured after a fivefold amplification.
Because PhBr is well known to undergo oxidative ad-
dition to M0 complexes,[9] it was selected as a model sub-
strate to test the reactivity of the electrogenerated Pt0-
(PPh3)2. The oxidation peak current at O1 of the electro-
generated Pt0(PPh3)2 (2 m) was lower when the cyclic vol-
tammetry was performed in the presence of PhBr
(10 equiv.) (scan rate, v = 0.5 Vs–1) due to oxidative ad-
dition (rate constant k in Scheme 1). This reaction took
place within the time elapsed between the generation of
Pt0(PPh3)2 at R1 and its oxidation at O1 (δt = 4.8 s at a scan
rate of 0.5 Vs–1). When the scan rate v was increased, the
oxidation peak current of O1 increased, because the time
scale for the oxidative addition was shortened. Because the
oxidation peak current of Pt0(PPh3)2 is proportional to its
concentration, the molar fraction x of Pt0(PPh3)2 was deter-
mined for different scan rates as x = i/i0 [i: oxidation peak
current of Pt0(PPh3)2 at O1 in the presence of PhBr (in the
range 0.02–0.4 ); i0: oxidation peak current of Pt0(PPh3)2
in the absence of PhBr, at the same scan rate v]. The plot
of x against log([PhBr]0/v) shown in Figure 2, attests to a
first-order reaction for PhBr and Pt0(PPh3)2.[7] The kinetic
law for the oxidative addition is: ln(x) = –k[PhBr]0t. When
x = 0.5, ln(0.5) = –k[PhBr]0t1/2 = –k[PhBr]0(∆E)/v1/2 with
∆E = (|Einv| – |EpR1|) + (|Einv| – |EpO1|) = 2.33 V (Einv: inver-
sion potential of the scan). The rate constant k for the oxi-
dative addition was then determined from the value of
[PhBr]0/v1/2 corresponding to the half reaction, as indicated
in Figure 2 (Table 1). The reaction was slower in the pres-
ence of an extra equivalent of PPh3. Pt0(PPh3)2 is thus in
equilibrium with PPh3 and Pt0(PPh3)3, but Pt0(PPh3)2 re-
mains the reactive complex in the oxidative addition.[8]
The behavior of the more electron-donating ligand
PnBu3 was very similar. No detectable PtII phosphane com-
plex was formed by treating Pt(acac)2 with 2 equiv. of
PnBu3. The electrochemical reduction of Pt(acac)2 in the
presence of 2 equiv. of PnBu3 delivered a Pt0 complex
Table 1. Oxidation peak potential Epox of Pt0(PR3)2 complexes gen-
erated in the reduction of Pt(acac)2 in the presence of 2 equiv. of
PR3 and rate constant k of the oxidative addition of PhBr and Si–
B derivative 1[a] as followed by fast cyclic voltammetry, in DMF at
27 °C (Scheme 1, Scheme 2).
Pt0(PR3)2
Ep (V vs. SCE)[b]
k [–1s–1]
Ox
PhBr
Si–B 1
Pt0(PPh3)2
+0.04
–0.48
1.9Ϯ0.2
32Ϯ1
Ͻ0.2
384Ϯ1
Pt0(PnBu3)2
[a] Pt0(PR3)2 was first generated by the electrochemical reduction
of Pt(acac)2 in the presence of 2 equiv. of PR3 and then in the
presence of PhBr or 1. [b] Potentials were determined at a gold disk
electrode (d = 1 mm) with a scan rate of 0.5 Vs–1.
Figure 3. Reduction of Pt(acac)2 (2 m) in DMF (containing
nBu4NBF4, 0.3 ) in the presence of PnBu3 (4 m) at a steady gold
disk electrode (d = 1 mm) with a scan rate v = 0.5 Vs–1, at 27 °C.
above and Figure S1 in Supporting Information). As ex-
pected, the more electron-rich Pt0(PnBu3)2 is more reactive
(by a factor of 17) towards PhBr than Pt0(PPh3)2 (Table 1).
ligated by two PnBu3 and oxidized at Ep
(Figure 3,
O3
Scheme 1). Pt0(PnBu3)2 was more easily oxidized than
Pt0(PPh3)2, as could be expected, because it is more electron
rich (Table 1). A second oxidation peak was also observed
Reactivity of Pt0(PR3)2 (R = Ph, nBu) in the Oxidative
Addition of 2-(Dimethylphenylsilyl)-4,4,5,5-tetramethyl-
1,3,2-dioxaborolane (1) in DMF
at Ep = +0.42 V, characteristic of free acac ions released
O2
in the reduction of Pt(acac)2 (Figure 3, Scheme 1).
Kinetic data on the reactivity of the electrogenerated
Pt0(PnBu3)2 with PhBr were obtained by performing cyclic
The reactivity of 1 towards Pt(PnBu3)2 and Pt(PPh3)2
voltammetry at various scan rates, as done for PPh3 (see was investigated by cyclic voltammetry as done for PhBr.
4238
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 4236–4241

Mechanism for the Pt-Catalyzed Silaboration of 1,3-Cyclohexadiene
1
From the value of their respective rate constants (Scheme 2, phosphane, was also followed by recording the H NMR
Table 1, Figure S2 in Supporting Information), it is clear spectra every 30 min at room temperature. In order to
that the reaction of 1 with Pt(PnBu3)2 is much faster than achieve full conversion of 1 an excess amount of Pt was
that with Pt(PPh3)2 (by a factor of at least 1500, Table 1).
employed (Pt/1, 2:1; C6D6/CD2Cl2, 1:1; 0.1  in Pt). Under
such conditions, the half-life time of the oxidative addition
was estimated to t1/2 = 200 min.
A second complex 3 was slowly formed in reactions be-
tween 1 and Pt0. After 3 d at room temperature, the ratio
1/2/3 was ca. 2:4:1 when a 1:1 ratio of Pt/1 was used. Com-
plex 3 could be isolated and separated from 2 by precipi-
tation from CH2Cl2 with diethyl ether. According to NMR
spectroscopic analysis, it contained one phosphane for each
silyl group, and showed a single 31P NMR signal at δ =
4.20 ppm (JPt,P = 1519 Hz, JP,Si = 68 Hz) and a singlet with
Scheme 2.
As illustrated above, electrochemical techniques can pro-
vide kinetic data, and the rate constants of oxidative ad-
ditions of PhBr and 1 with Pt0 complexes were obtained for
the first time. However, electrochemical techniques cannot
provide any structural information. Consequently, the com-
plexes formed in the oxidative addition and in the following
steps were investigated by NMR spectroscopy.
1
Pt satellites at δ = 0.56 (1JH,Pt = 27 Hz) in the H NMR
spectrum. Comparison with literature NMR spectroscopic
data[10] (CD2Cl2) suggests that the compound is bis[(di-
methylphenyl)phosphane]bis[(dimethylphenyl)silyl]-
platinum(II) (3). In reactions using a Pt/1 ratio of 2:1 merely
ca. 2% of this complex was observed, suggesting that it is
formed from 2 and silylborane 1 rather than by metathesis
of 2. This hypothesis was later contradicted: when 1,3-cy-
clohexadiene was added to complex 2, isolated in pure
form, large amounts of complex 3 were still formed, with-
out any noticeable amount of silylborane being present.[11]
The rate of oxidative addition was not affected by the
presence of 1,3-cyclohexadiene. Addition of an excess
amount of 1,3-cyclohexadiene (10 equiv.) to 2 generated as
above (by using a Pt/1 ratio of 1:1) caused the formation of
a new complex, 4. At room temperature, the process was
very slow, but at 80 °C signals for the new species appeared
within 1 h. In the 31P NMR spectrum a singlet flanked by
131Pt satellites (JP,Pt = 3524 Hz) appeared for the new com-
plex at δ = 3.20 ppm, but no new broad signal characteristic
of phosphorus trans to boron was found. As expected, the
η1 complex obtained by Ozawa by reaction of 2 with phen-
ylacetylene contained two chemically different phosphanes,
observed as two doublets in the 31P NMR spectrum with
NMR Spectroscopic Investigations of the Reactions of 1,3-
Cyclohexadiene with Silylborane 1
Two ligands were investigated: PPh3, which was known
to be efficient for the catalytic reaction depicted in Equa-
tion (1), and PMe2Ph. In order to study the reactivity of
silylborane 1 towards Pt0 generated in a stoichiometric reac-
tion under the conditions used in the catalytic process,[5]
1
(1 equiv.) was first added to a mixture of Pt(acac)2/
DIBALH/PMe2Ph (1:2:2) in C6D6/CD2Cl2 (1:1). This re-
sulted in slow formation of a complex showing two 31P
NMR signals: one sharp doublet centered at δ = 5.0 ppm
with 195Pt satellites (JP,Pt = 1400 Hz, JP,P = 29 Hz) and a
broad signal at δ = 5.7 ppm characteristic of a phosphorus
1
atom trans to a boron atom (JP,Pt = 1404 Hz). In the H
NMR spectrum the signal from the methyl protons of the
silylborane at δ = 0.20 ppm decreased in intensity and a
new singlet with Pt satellites at δ = 0.47 ppm (JH,Pt = 32 Hz)
appeared. The complex could be isolated in pure form from
the reaction mixture (see Experimental Section). Compari-
son of the spectra (in CD2Cl2) with those reported by Oz-
awa for the product obtained by oxidative addition of 1 to
the Pt0 complex generated from Pt0(cod)2 and PMe2Ph,[6]
confirmed that the same product 2 was obtained under our
conditions (Scheme 3). Thus, in accordance to what was
shown in the electrochemical studies, the complex does not
contain the anion (acac) or any Al species.
additional splitting due to Pt.[6] The presence of a single 31
P
resonance from the complex obtained from cyclohexadiene
is consistent with the formation of a η3-allyl complex con-
taining only one phosphane ligand (4; Scheme 3). This as-
sumption was supported by the presence of a three-spin sys-
tem characteristic of a metal-coordinated allyl group, con-
sisting of a doublet at δ = ca. 4.92 ppm (J = 7 Hz), a triplet
at δ = 4.54 ppm (J = 7 Hz), and a broad signal at δ = ca.
3.6 ppm in the 1H NMR spectrum of 4 (see Supporting
Information). As expected, separate signals were observed
for the diastereotopic methyl groups on silicon (δ =
0.23 ppm, JH,Pt = 30 Hz and δ = 0.28 ppm, JH,Pt = 30 Hz).
1
When the formation of 4 was followed by H NMR spec-
troscopy at room temperature by using 1-methoxynaphtha-
lene as an internal standard, it was observed that the
amount of 4 increased at the expense of 2, whereas the
amount of bis(silyl) complex 3 was essentially constant.[12]
No product 5 from reductive elimination was observed even
Scheme 3.
The reaction between 1 and Pt0(PMe2Ph)2, generated by after prolonged heating of the mixture containing 4 or after
reduction of Pt(acac)2 with DIBALH in the presence of the the addition of an excess amount of triphenylphosphane
Eur. J. Inorg. Chem. 2008, 4236–4241
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4239

G. Durieux, M. Gerdin, C. Moberg, A. Jutand
FULL PAPER
(see below). Instead, complex 4 slowly decomposed.[13] In cal reduction of Pt(acac)2 by DIBALH in the presence of
agreement with this observation, no product was isolated PPh3, no oxidative addition of 1 was observed, although in
from a catalytic reaction employing PMe2Ph as the ligand. the presence of 1,3-diene, the reaction proceeded to yield
Several attempts were made to isolate complex 4 by unidentified intermediate complexes and finally the desired
crystallization, but the complex was elusive.[14]
1,4-silaborated product. A more rapid oxidative addition of
In contrast to the situation using PMe2Ph, the product 1 occurred with PMe2Ph as ligand, and the resulting Si–
from silaboration (5) is obtained in reactions with PPh3 as PtII–B complex was shown to react with the diene, giving a
the ligand. In this case, oxidative addition was shown by η3-allyl complex. No silaborated product was observed
cyclic voltammetry to be extremely slow at 27 °C (k Ͻ from reactions using PMe2Ph, due to unfavored reductive
0.2 –1 s–1), and no product was observed by NMR spec- elimination.
troscopy when 1 (one equiv.) was added to a mixture of
PPh3, Pt(acac)2, and DIBALH (2:1:2), even after heating at
80 °C for 18 h. Even a 10-fold excess of 1 did not lead to
Experimental Section
General Remarks: All reactions were prepared inside a nitrogen
filled glove box by using oven-dried glassware or under a nitrogen
atmosphere by using standard Schlenk techniques. Hexane,
CH2Cl2, and Et2O were dried by using a Glass-contour solvent
dispensing system. Benzene was distilled from CaH2 prior to use. 2-
(Dimethylphenylsilyl)-4,4,5,5-tetrametyl-1,3,2-dioxaborolane (1)[15]
was synthesized according to a literature procedure. All other
chemicals were of at least 97% purity and used as received. 1H
NMR spectra were recorded at 500 or 400 MHz, 13C spectra at
any observed product from oxidative addition. When the
stoichiometric reaction was repeated in the presence of 1,3-
1
cyclohexadiene and monitored by H NMR spectroscopy
by using 1-methoxynaphthalene as an internal standard no
products were observed after one day at room temperature.
However, after heating overnight at 50 °C, 55% unreacted
silylborane 1 remained together with unidentified com-
plexes and 18% of the product from silaboration (5). After
the same time at 60 °C, 4 % of 1 and 58% 5 were observed.
At 80 °C, only a trace amount of 1 remained and 95%
product was formed. It is interesting to note that whereas
96% of silylborane 1 was consumed after being heated at
60 °C overnight, only 58% of product 5 was formed, but a
final 95% yield was recorded. This shows that Si–B contain-
ing complexes, leading to product, accumulate before the
final reductive elimination to give 5.
From these results it is clear that the relative rates of
oxidative addition, insertion, and reductive elimination are
highly dependent on the structure of the phosphane ligand.
In accordance with expectations, as well as with previous
observations,[6] oxidative addition proceeds faster with the
more electron-donating phosphane PMe2Ph than with PPh3
as ligand. With the former ligand, the barrier for reductive
elimination was too high to allow the formation of the final
product. With PMe2Ph as ligand, oxidative addition was
found to be faster than migratory insertion, whereas with
PPh3, oxidative addition was not observed in the absence
of 1,3-cyclohexadiene. However, in the presence of the un-
saturated substrate the small amount of complex from oxi-
dative addition reacted further to give complexes that ac-
cumulated before reductive elimination took place to yield
the final product 5, the same as that obtained in the cata-
lytic reaction.
1
125 MHz, and 31P spectra at 202 MHz. H chemical shifts are re-
ported relative to residual solvent peak, and 31P chemical shifts are
reported relative to external 85% H3PO4. Cyclic voltammetry was
performed under an atmosphere of argon at a steady gold disk
electrode (d = 1 mm) with a homemade potentiostat and a wave
form generator EG&G Parc Model 175. The voltammograms were
recorded with a Nicolet 301 oscilloscope.
Procedure for the Oxidative Addition of PhBr to Electrogenerated
Pt0 Ligated by two PR3 as Monitored by Cyclic Voltammetry: Ex-
periments were carried out in a three-electrode thermostatted cell
connected to a Schlenk line. The counter electrode was a platinum
wire of ca. 1-cm2 apparent surface area; the reference saturated
calomel electrode (Radiometer) was separated from the solution by
a bridge filled with DMF (2 mL) containing nBu4NBF4 (0.3 ).
The working electrode was a steady gold disk electrode (d = 1 mm).
To DMF (10 mL) containing nBu4NBF4 (0.3 ) was added Pt-
(acac)2 (7.9 mg, 0.02 mmol, 2 m) followed by PPh3 (10.5 mg,
0.04 mmol). The cyclic voltammetry was performed at different
scan rates (from 0.2 to 1 Vs–1) in the absence of PhBr and then in
the presence of increasing amounts of PhBr [10 (14.5 µL), 20, 50,
100 equiv.] at the same scan rates.
Similar experiments were performed from Pt(acac)2 (7.9 mg,
0.02 mmol, 2 m) and PnBu3 (10 µL, 0.04 mmol). Cyclic voltam-
metry was performed at different scan rates (from 0.1 to 2 Vs–1) in
the absence of PhBr and then in the presence of PhBr (1.45 µL,
0.02 mmol) at the same scan rates.
Procedure for the Oxidative Addition of 1 to Electrogenerated Pt0
as Monitored by Cyclic Voltammetry: Experiments were performed
from Pt(acac)2 (7.9 mg, 0.02 mmol) and PnBu3 (10 µL, 0.04 mmol).
The cyclic voltammetry was performed at different scan rates (from
0.2 to 2 Vs–1) in the absence of 1 and then in the presence of 1
(5.24 mg, 0.02 mmol) at the same scan rates.
General Procedure for the Oxidative Addition of 1 to Pt0 as Moni-
tored by 1H NMR Spectroscopy: Pt(acac)2 (19.65 mg, 0.05 mmol)
was dissolved in C6D6/CD2Cl2 (1:1, 300 µL). PPhMe2 (14.2 µL,
0.10 mmol) was then added by syringe, and the resulting mixture
was cooled to –35 °C over 15 min. DIBALH (1  in cyclohexane,
100 µL, 0.10 mmol) was added by syringe, and the solution was
Conclusions
Silylborane 1 undergoes oxidative addition to neutral
Pt0L2 (L = monophosphanes) generated from Pt(acac)2
either by electrochemical reduction or by chemical re-
duction by DIBALH. No coordination of phosphanes to
Pt(acac)2 was observed, whereas the rate of oxidative ad-
dition to Pt0L2 was influenced by the structure of the phos-
phane ligand. From electrochemical data, oxidative ad-
dition of the silylborane to Pt(PnBu3)2 was found to be con-
siderably much faster than that to Pt(PPh3)2. In the chemi- allowed to stand at room temperature for 2 h. Compound 1
4240 Eur. J. Inorg. Chem. 2008, 4236–4241
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Mechanism for the Pt-Catalyzed Silaboration of 1,3-Cyclohexadiene
(13.1 mg, 0.05 mmol) dissolved in C6D6/CD2Cl2 (1:1, 200 µL) was
then added to the catalyst solution. The resulting mixture was
transferred to an NMR tube equipped with a screw cap, and the
reaction was monitored by 1H NMR spectroscopy. In order to iso-
late complex 2, the crude product was dissolved in Et2O/DCM (1:1)
and hexane was added. Precipitated solids and sticky brown resi-
dues were removed by filtration (in the glove box) and the solvents
removed in vacuo. The solids were washed with Et2O and then with
[1]
a) K. A. Horn, Chem. Rev. 1995, 95, 1317–1350; b) H. K.
Sharma, K. H. Pannell, Chem. Rev. 1995, 95, 1351–1374; c) I.
Beletskaya, C. Moberg, Chem. Rev. 1999, 99, 3435–3461; d) R.
Zimmer, C. U. Dinesh, E. Nandanan, F. A. Khan, Chem. Rev.
2000, 100, 3067–3312; e) M. Suginome, Y. Ito, Chem. Rev.
2000, 100, 3221–3256; f) I. Beletskaya, C. Moberg, Chem. Rev.
2006, 106, 2320–2354.
a) Y. Ito, J. Organomet. Chem. 1999, 576, 300–304; b) M. Sugi-
nome, Y. Ito, J. Organomet. Chem. 2003, 680, 43–50.
a) J. W. J. Kennedy, D. G. Hall, Angew. Chem. Int. Ed. 2003,
42, 4732–4739; b) I. Fleming, A. Barbero, D. Walter, Chem.
Rev. 1997, 97, 2063–2192.
M. Suginome, T. Matsuda, T. Yoshimoto, Y. Ito, Org. Lett.
1999, 1, 1567–1569.
M. Gerdin, C. Moberg, Adv. Synth. Catal. 2005, 347, 749–753.
T. Sagawa, Y. Asano, F. Ozawa, Organometallics 2002, 21,
5879–5886.
For the use of electrochemical techniques to generate and fol-
low the reactivity of Pd0 complexes in oxidative addition to
aryl halides, see: a) C. Amatore, M. Azzabi, A. Jutand, J. Am.
Chem. Soc. 1991, 113, 8375–8384; b) J. Pytkowicz, S. Roland,
P. Mangeney, G. Meyer, A. Jutand, J. Organomet. Chem. 2003,
678, 166–179; c) S. Roland, P. Mangeney, A. Jutand, Synlett
2006, 3088–3094.
a) R. Ugo, F. Cariati, G. La Monica, Chem. Commun. (Lon-
don) 1968, 868–869; b) L. Malatesta, C. Cariello, J. Chem. Soc.
1958, 2323–2328; c) J. P. Birk, J. Halpern, A. L. Pickard, J. Am.
Chem. Soc. 1968, 90, 4491–4492.
[2]
[3]
1
hexane (3ϫ) to yield complex 2 as a light-yellow solid. H NMR
(C6D6/CH2Cl2, 1:1): δ = 0.47 (s, JPt,H = 32 Hz, 6 H), 0.92 (d, JP,H
= 7 Hz, JPt,H = 16 Hz, 6 H), 0.96 (s, 12 H), 1.28 (d, JP,H = 7 Hz,
JPt,P = 21 Hz, 6 H), 6.92–7.09 (m, 9 H), 7.15–7.28 (m, 4 H), 7.58–
7.63 (m, 2 H) ppm. 31P NMR (C6D6/CD2Cl2, 1:1): δ = 5.0 (d, JP,P
= 29 Hz, JPt,P = 1400 Hz), 5.7 (br., JPt,P = 1404) ppm.[6]
[4]
Procedure for the Reaction of 2 with 1,3-Cyclohexadiene: The reac-
tion was performed according to the general procedure, with the
addition of 1,3-cyclohexadiene (50 µL, 0.5 mmol) before the solu-
tion was transferred into the NMR tube. Characterization of com-
[5]
[6]
[7]
1
plex 4: H NMR (400 MHz, C6D6/CD2Cl2, 1:1): δ = 0.23 (s, JH,Pt
= 30 Hz, 3 H), 0.28 (s, JH,Pt = 30 Hz, 3 H), 3.6 (br., 1 H), 4.54 (app
t, J = 7 Hz, 1 H), 4.92 (d, J = 7 Hz, 1 H) ppm. The other 1H NMR
signals were obscured due to overlap with the signals from 1, 2,
and 3. H,H-COSY correlations were observed between protons ap-
pearing at δ = 3.6 ppm and δ = 4.54 ppm and also between those
at δ = 4.54 ppm and δ = 4.92 ppm (Supporting Information). 31P
NMR (C6D6/CD2Cl2, 1:1): δ = 3.20 (s, JP,Pt = 3524 Hz) ppm.
[8]
Preparation of 3: Pt(acac)2 (98.3 mg, 0.25 mmol) was dissolved in
benzene/CH2Cl2 (1:1, 1.5 mL). PPhMe2 (71 µL, 0.50 mmol) was
then added by syringe, and the resulting mixture was cooled to
–35 °C over 15 min. DIBALH (1  in cyclohexane, 500 µL,
0.50 mol) was added by syringe, and the solution was allowed to
stand at room temperature for 2 h. Compound 1 (13.1 mg,
0.05 mmol) dissolved in benzene/CH2Cl2 (1:1, 1 mL) was then
added to the catalyst solution. The resulting solution was heated
in a sealed vial at 50 °C for 48 h, giving a ca. 1:1 mixture of com-
plexes 2 and 3. The solvents were removed over the manifold. The
residue was suspended in CH2Cl2 (0.4 mL) and Et2O (6 mL) was
then gently layered on top, and the mixture was left at room tem-
perature overnight. The solvents were removed by syringe, and the
remaining solid was washed repeatedly with Et2O (6ϫ3 mL) until
the washings were colorless. The resulting solid was dried in vacuo,
yielding 3 (12 mg, 6%) as a light-yellow solid. 1H, 31P, and 13C
NMR spectra (CD2Cl2) were in accordance with previously re-
ported data[10] and showed the compound to be of more than 95%
purity.
[9] R. Ugo, Coord. Chem. Rev. 1968, 319–364.
[10] F. Ozawa, J. Kamite, Organometallics 1998, 17, 5630–5639.
[11] When 1,3-cyclohexadiene was added to complex 3 in C6D6/
DCM (1:1) and heated at 50 °C overnight a 0.84:1:0.48 ratio
of 2/3/4 was obtained, along with considerable amounts
(Ͼ10% of total amount of -SiMe2Ph) of unidentified products.
[12] The reaction was performed in C6D6/CD2Cl2 (1:1) at 0.1  con-
centration [Pt(acac)2] on a 0.70:1.16:0.33 mixture of complexes
1/2/3; values calibrated to internal standard (1-methoxynaph-
thalene). 1,3-Cyclohexadiene (10 equiv.) was added and the for-
1
mation of 4 was monitored by H NMR spectroscopy at 24 h
intervals. After 6 d, a 0.54:0.62:0.35:0.62 mixture of 1/2/3/4 was
obtained.
[13] After 24 h at 80 °C, no trace of complexes 2–4 was present. No
product of silaboration (5) was found.
[14] Crystallization from Et2O, benzene, DCM, or CDCl3 with hex-
ane or pentane was attempted by direct precipitation, diffusion
chamber, or diffusion in the NMR tube. Typically, a brown oil
was obtained, or complex 3 was isolated in pure form. Further
attempts to isolate complex 4 were also made, but met with
little success: reaction conditions were optimized in terms of
stoichiometries, order and time of addition, and temperature,
but no increase in the yield of 4 relative to that obtained under
the standard conditions was observed. Exchange of Pt(acac)2
for Pt(cod)2 or addition of 1,3-cyclohexadiene to complex 2
isolated in pure form did not lead to any increase in amount
or purity of complex 4. Finally, attempts to react 4 with sodium
dimethyl malonate (A. J. Blacker, M. L. Clarke, M. S. Loft,
M. F. Mahon, M. E. Humphries, J. M. J. Williams, Chem. Eur.
J. 2000, 6, 353–360) met with no success – heating at 60 °C
overnight only led to decomposition.
Procedure for Catalytic Reactions Involving PPh3: The reaction was
performed according to the general procedure, except for PPh3
(26.5 mg, 0.10 mmol) replacing PPhMe2 as the ligand and the ad-
dition of 1-methoxynaphthalene (7.25 µL, 0.05 mmol) and 1,3-cy-
clohexadiene (50 µL, 0.5 mmol) before the solution was transferred
into the NMR tube.
Supporting Information (see footnote on the first page of this arti-
cle): Graphs for the determination of rate constants; 31P, 13C, and
1H NMR spectra.
[15] M. Suginome, T. Matsuda, Y. Ito, Organometallics 2000, 19,
Acknowledgments
4647–4649.
Received: March 18, 2008
Published Online: August 19, 2008
This work was supported by the Swedish Research Council, the
Centre National de la Recherche Scientifique (CNRS, UMR 8640),
and the Ecole Normale Supérieure.
Eur. J. Inorg. Chem. 2008, 4236–4241
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4241