Mechanistic insights into the palladium-catalysed asymmetric phosphination of cyclohexenyl triflate

Article abstract of DOI:10.1002/ejic.201000987

Preliminary results on the mechanism of the asymmetricphosphination involving an achiral alkenyl triflate (cyclohexenyl triflate) and a secondary phosphane-borane, (methyl)phenylphosphane-borane (1) are reported. A catalytic cycle is proposed based on the

Full text of DOI:10.1002/ejic.201000987

FULL PAPER
DOI: 10.1002/ejic.201000987
Mechanistic Insights into the Palladium-Catalysed Asymmetric Phosphination
of Cyclohexenyl Triflate
Delphine Julienne,[a] Olivier Delacroix,[a] Jean-François Lohier,[a,b]
Jana Sopkova de Oliveira-Santos,[a,b] and Annie-Claude Gaumont*[a]
Keywords: Organophosphorus chemistry / Homogeneous catalysis / Phosphane ligands / Palladium / C–P coupling / Cross-
coupling
Preliminary results on the mechanism of the asymmetric
phosphination involving an achiral alkenyl triflate (cyclohex-
enyl triflate) and a secondary phosphane–borane, (methyl)-
phenylphosphane–borane (1) are reported. A catalytic cycle
is proposed based on the variable-temperature 31P NMR
characterization of the individual steps (oxidative addition,
transmetallation and reductive elimination). Each likely in-
termediate involved in this asymmetric C–P coupling reac-
tion has been identified and characterized. Hence, cationic
oxidative addition complex 3 has been readily synthesized
and engaged in the stoichiometric reaction with highly en-
antio-enriched secondary phosphane–borane (SP)-1 enabling
to isolate the corresponding diastereomerically pure trans-
metallation adduct (S,S,RP)-5. Its structure has been con-
firmed by X-ray crystallography. Its decomposition by re-
ductive elimination affords the highly enantio-enriched cou-
pling product (SP)-2. This study also provides information on
the origin of the enantioselectivity.
develop more efficient reactions. In this context, we report
herein preliminary results on the mechanism of the palla-
dium catalysed asymmetric phosphination of an alkenyl
triflate using a secondary phosphane–borane as phosphin-
ating agent.
Introduction
Alkenylphosphanes are key compounds in organophos-
phorus chemistry.[1] They proved to be useful in coordina-
tion chemistry, polymer chemistry, organic synthesis for the
preparation of polyphosphanes and more recently in cataly-
sis as ligands in cross-coupling reactions.[2] However, chiral
derivatives and particularly P-chirogenic ones are scarce.[1]
Possible reasons are the lack of general synthetic routes and
tedious preparations using a stoichiometric amount of a
chiral catalyst (resolution processes or asymmetric synthe-
Results and Discussion
After reporting a general and efficient access to achiral
ses).[1] A promising new approach for their synthesis is alkenylphosphanes involving the palladium-catalysed C–P
asymmetric catalysis. Although a few examples have been cross-coupling reaction between secondary phosphane–bo-
recently reported in the literature through the use of cata- ranes and alkenyl triflates,[5] we focused our work on the
lytic hydrophosphination of alkynes[3] and C–P cross-cou- asymmetric version of this C–P cross-coupling reaction and
pling reaction of an alkenyl triflate,[4] the enantioselectivity recently reported that an enantio-enriched P-chirogenic alk-
reached and/or the scope of the reactions are too modest enylphosphane could be formed by asymmetric C–P cross-
to envision preparing efficiently P-chirogenic tertiary phos- coupling reaction (Scheme 1).[4] Promising enantiomeric ra-
phanes through these methodologies. Information on the tios, up to 78:22, could be obtained in the study involving
mechanism and the factors, which affect the enantio- achiral cyclohexenyl triflate and the racemic (methyl)phen-
selectivity of these reactions, would certainly be helpful to ylphosphane–borane
1
when using (S,S)-Me-DU-
PHOSPdCl2 as precatalyst (Scheme 1).
[a] Laboratoire de Chimie Moléculaire et Thioorganique (LCMT),
UMR CNRS 6507, INC3M, FR 3038, ENSICAEN &
Université de Caen,
14050 Caen, France
Fax: +33-2-31452877
E-mail: annie-claude.gaumont@ensicaen.fr
[b] X-ray Center, Centre d’Etudes et de Recherche sur le Médica-
ment de Normandie (CERMN), UPRES EA 4258 - FR CNRS
3038 INC3M,
Boulevard Becquerel, 14032 Caen Cedex, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000987.
Scheme 1. Asymmetric C–P cross-coupling between cyclohexenyl
triflate and phosphane–borane 1.
Eur. J. Inorg. Chem. 2011, 2489–2498
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2489

A.-C. Gaumont et al.
FULL PAPER
To get insights into the origin of the enantioselection, we
embarked in a study aiming at characterizing likely interme-
diates involved in the different steps of the putative catalytic
cycle by using 31P NMR spectroscopy. To the best of our
knowledge, there is no mechanistic information available on
the coupling between a triflate derivative and a phosphane–
borane in the literature. A plausible mechanism based on
previous investigations on the C–P cross-coupling reaction
between secondary phosphane–borane and aryl iodide,[6] is
proposed in Scheme 2.
Scheme 3. Synthesis of complex 3. a) Pd2(dba)3·CHCl3, TMEDA,
toluene, 50 °C, 3 h, 77%; b) (S,S)-Me-DUPHOS, THF, 40 °C, 2 h,
84%; c) AgOTf, CD3CN, –20 °C.
CH2Cl2.[14] The ORTEP diagram is shown in Figure 1. Se-
lected bond lengths and angles are given in Table 1. Com-
plex 7 adopts a near square-planar geometry at palladium
with the cyclohexenyl group oriented orthogonal to the
square plane of the molecule.
Scheme 2. Plausible catalytic cycle for the palladium-catalysed
vinylation of phosphane–borane.
The putative first step is the oxidative addition of the
alkenyl triflate to the low valent Pd0 complex leading to
cationic complex 3.[7] Then, transmetallation with boron
phosphide 4 resulting from the deprotonation of secondary
phosphane–borane 1 occurs, leading to transmetallation
adduct 5 and/or 6. Lastly, reductive elimination enables the
C–P bond formation and regenerates the Pd catalyst.
Synthesis of the Precursors: Secondary Phosphane–Borane
1 and Putative Cationic Oxidative Complex 3
Figure 1. ORTEP diagram of Pd(TMEDA)(cyclohexenyl)(I) 7. Hy-
Racemic secondary phosphane–borane 1 was prepared
according to a literature procedure.[8] Enantiopure (SP)-1,
readily accessible by following the procedure reported by
Livinghouse,[9] was efficiently prepared with an enantio-
meric ratio of 98:2 (See Supporting Information). Alkenyl
cationic oxidative palladium complexes with a labile ligand
and a cis chelating diphosphane being depicted as unstable
species,[7] compound 3 could not be obtained by direct oxi-
dative addition to [(S,S)-Me-Duphos]Pd0 complex. Hence,
the putative catalyst 3 was prepared following the route de-
scribed in Scheme 3, through a synthesis inspired by the
chemistry developed by Brown on arylic cationic oxidative
complexes for the study of reactive intermediates in the
Heck reaction.[10]
drogen atoms are omitted for clarity.
Table 1. Selected bond lengths [Å] and bond angles [°] for Pd(TME-
DA)(cyclohexenyl)(I) 7.
Bond
Length
Bond
Angle
Pd1–C7
Pd1–N1
Pd1–N2
Pd1–I1
1.9795(16)
2.1322(15)
2.2249(16)
2.545(5)
N1–Pd1–N2
N2–Pd1–I1
C7–Pd1–I1
C7–Pd1–N1
C7–Pd1–N2
N1–Pd1–I1
83.17(7)
95.08(10)
89.20(9)
92.66(7)
174.21(6)
177.54(13)
Displacement of the TMEDA ligand by (S,S)-Me-
DUPHOS afforded iodide complex 8 in 84% yield. This
The first step is the preparation of the new iodide com- new complex was fully characterized by NMR spec-
plex
7 following an adapted procedure inspired by troscopy, elemental analysis and HRMS. It proved to be
Glueck[11] and Drago’s[12] works on the preparation of aryl- less stable than its precursor 7 both in solution and in solid
iodide oxidative addition complexes. Pd2(dba)3·CHCl3 was state. Although it is possible to handle this complex in air,
first treated with TMEDA and iodocyclohexene,[13] to give, it is necessary to store it at –20 °C to avoid decomposition.
after work-up, the air stable complex Pd(TMEDA)(cyclo- The low stability of 8 was also demonstrated during
hexenyl)(I) 7 in 77% yield. Complex 7 was fully charac- crystallization attempts, which all led to the known by-
[11a]
terized by NMR spectroscopy, elemental analysis and product [(S,S)-Me-DUPHOS]PdI2
whatever the tech-
HRMS. Its structure was also confirmed by X-ray crystal- nique used (evaporation, diffusion at low temperature). The
lography after crystallisation from a mixture of Et2O and assignment of the cyclohexenyl and DUPHOS ligand pro-
2490
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2489–2498

Pd-Catalysed Asymmetric Phosphination
1
tons was established via a H COSY experiment and long- temperature and by HRMS (see Supporting Information).
1
range 13C, H correlations. The coordinated cyclohexenyl The 31P NMR spectrum shows two peaks at δ = 68.6 and
1
fragment exhibited a signal about 5.5 ppm in the H NMR 65.4 ppm (in CD3CN) with a characteristic cis coupling
spectrum and a resonance in 13C NMR about 154 ppm with constant Jpp value about 29 Hz (see Supporting Infor-
2
a characteristic trans coupling constant JC-Ptrans value of mation).
ca. 130 Hz for the σ-bounded alkene (See Supporting Infor-
mation). This high frequency chemical shift is quite similar
to those observed in the literature for aryl complex deriva-
Study of the Catalytic Cycle by vt-NMR Spectroscopy
1
tives.[11] Insight into the structure was then obtained by H
Using (SP)-1: Transmetallation Step
NOESY in order to determine how the cyclohexenyl ligand
and the auxiliary Me-DUPHOS interact in solution. A sec-
Having in hand phosphane 1 and catalyst precursor 3,
1
tion of the H NOESY spectrum of 8 is shown in Figure 2 the catalytic cycle was studied by vt-NMR spectroscopy
and clearly reveals fairly strong NOEs between the methyl using first the highly enantio-enriched (SP)-1 phosphane–
(Me12) and the methine (H8) of the DUPHOS phospholane borane. Treatment of 3 at –60 °C in a mixture of solvents,
ring and the vinylic proton of the cyclohexenyl group (H1) {[D8]THF/[D7]DMF (10 equiv. to Pd complex)},[15] with
(Figure 3). Correlation was also observed between the (SP)-1 (98:2 e.r.) in stoichiometric amount, in the presence
methyl (Me18) and the methine (H11 and H14) of the of Me3SiOK as base,[16] led to the clean formation of a new
DUPHOS and the ortho protons (H21 and H24) of the aro- complex. Low-temperature 31P NMR spectroscopy (see
matic ring of DUPHOS, as shown in Figure 3.
Table 2 and the Supporting Information) supports the for-
mation of the transmetallation complex 5 (Scheme 2). No
trace of a silanolate palladium complex resulting from the
reaction between silanolate and 3 could be detected.[17] This
observation indicates that deprotonation of 1 is fast[18] and
that boron phosphide 4 directly reacts with 3 to give 5. It
is worth noting that since the phosphorus in 5 is tetrahe-
dral, the inversion process at phosphorus should be ar-
rested. Therefore, only one diastereomer of the transmetal-
lation adduct 5 having a controlled stereochemistry at phos-
phorus is expected if the coordination of the phosphide 4
to the metal centre occurs with high stereocontrol.[19] In
fact, we observed in the 31P NMR spectrum two isomers of
the transmetallation adduct 5 in a 1:1.5 ratio.[20] The extra
isomer 5min appeared to be a conformational isomer, which
arose from slow rotation about the Pd–C bond on the
NMR time scale. Similar behaviour was reported in the re-
lated Pt(DIOP)(o-An)(I)[21] and [Pd(DUPHOS)(o-An)-
{PPhMe(BH3)}] complexes.[11]
Figure 2. Section of the 1H NOESY spectrum of 8 (CDCl3,
400 MHz).
Table 2. 31P{H} NMR spectroscopic data for 5 at –30 °C.[a]
Complex
δ(P1)
δ(P2)
δ(P3)
JP1P2 JP1P3
JP2P3
5maj[b]
61.6
59.2
63.3
59.8
–19.7
–15.1
26.0 294.2
25.1 300.1
24.4
29.8
5min[c]
[a] Solvent: [D8]THF/[D7]DMF with the signal locked to [D8]THF.
[b] Major conformational isomer. [c] Minor conformational isomer.
Figure 3. Representation of 8 showing intra- and inter-ligand
NOEs.
In the 31P NMR spectrum, each isomer of the boron
phosphide complex 5 shows a characteristic broad doublet
The putative oxidative addition product 3 was then ob- due to the [P3(Me)(Ph)(BH3)] phosphido group and two
tained after treatment of alkenyliodide complex 8 with sil- doublets of doublets due to the non equivalent P1 and P2
ver triflate in CD3CN at –20 °C followed by low tempera- phosphorus atoms of DUPHOS ligand with small (ca.
ture filtration of the precipitated silver salts. Due to its low 25 Hz) and large (ca. 295 Hz) coupling constants, which are
stability in solution (decomposition about 20 °C), 3 was consistent with JPP cis and JPP trans values, respectively (see
characterized by multinuclear NMR spectroscopy at low Table 2 and Figure 4).[6d]
Eur. J. Inorg. Chem. 2011, 2489–2498
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2491

A.-C. Gaumont et al.
FULL PAPER
Figure 6. 31P{H} spectrum of complex 5 in [D8]THF/[D7]DMF
with the signal locked to [D8]THF at 0 °C.
Figure 4. 31P{H} spectrum of complex 5 in [D8]THF/[D7]DMF
with the signal locked to [D8]THF at –30 °C.
Table 3. 31P{H} NMR spectroscopic data for 5 at 0 °C.[a]
On warming (–10 °C), one of the isomer disappeared: a
section of 31P{H} spectra of complex 5 at various tempera-
tures (from –50 °C to 20 °C) is reported in Figure 5.
Complex
δ(P1)
δ(P2)
δ(P3)
JP1P2 JP1P3
26.0 294.0
JP2P3
5
61.6
63.4
–19.7
24.5
[a] Solvent: [D8]THF/[D7]DMF with the signal locked to [D8]THF.
Single crystals of 5 were obtained by slow diffusion of
pentane into a concentrated CH2Cl2 solution at –20 °C.[22]
ORTEP diagram is shown in Figure 7; selected bond
lengths and angles are given in Table 4. Additional infor-
mation appears in the Supporting Information. Complex 5
shows a great distortion from planarity as judged by the
distances of ligand atoms from the P1P2PdCP3 plan (up to
0.20 Å for P3). This result is consistent with a great steric
congestion around the palladium atom caused by the
[P3(Me)(Ph)(BH3)] phosphido–borane group and is in
agreement with the behavior of complex 5 in solution. Ab-
solute configuration (R) of the phosphido center P3 in 5
was established by X-ray crystallography, suggesting that
the transmetallation step proceeds with retention of config-
uration (Figure 7).[23]
Figure 5. Section of 31P{H} spectra corresponding to phosphorus
P1 and P2 of DUPHOS ligand in complex 5 at various temperature;
experiments were done in [D8]THF/[D7]DMF with the signal
locked to [D8]THF.
As illustrated in Figure 6, the 31P NMR spectrum at 0 °C
indicates the presence of only one species, which is consis-
tent with the boron phosphide complex 5 (see Table 3 for
31P NMR spectroscopic data). This observation points out
that treatment of the chiral cationic complex 3 with the
highly enantio-enriched (SP)-1 (98:2 e.r.) gives 5 with high
dr (98:2) at low temperature (–60 °C).
Figure 7. ORTEP diagram of [Pd{(S,S)-Me-DUPHOS}(cyclo-
hexenyl)–{P(Me)(Ph)(BH3)}] (5). Hydrogen atoms are omitted for
clarity.
2492
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2489–2498

Pd-Catalysed Asymmetric Phosphination
Table 4. Selected bond lengths [Å] and bond angles [°] for
[Pd{(S,S)-Me-DUPHOS}(cyclohexenyl){P(Me)(Ph)(BH3)}] 5.
Bond
Length
Bond
Angle
Pd1–P1
Pd1–P2
Pd1–P3
Pd1–C19
P3–C26
P3–C25
P3–B1
2.3310(18)
2.2849(19)
2.3531(18)
2.070(6)
1.831(6)
1.839(7)
P1–Pd1–P3
P2–Pd1–P1
C19–Pd1–P2
C19–Pd1–P3
C19–Pd1–P1
P2–Pd1–P3
99.28(6)
85.80(7)
90.1(2)
86.3(2)
171.21(19)
167.79(6)
1.947(8)
Study of the Catalytic Cycle by vt-NMR Spectroscopy
Using (SP)-1: Reductive Elimination Step
Decomposition of 5 on warming to 20 °C through re-
ductive elimination led as expected to tertiary phosphane–
borane 2 along with Pd[(S,S)-Me-Duphos]2, which was
characterized by comparison with an authentic sample.[11a]
Surprisingly, a third product was detected in the reaction
mixture at 20 °C (Figure 8). The 31P{H}NMR spectrum of
this unknown compound shows the presence of two dou-
blets of doublets, indicating that the two phosphorus atoms
P1 and P2 of the [(Me-DUPHOS)Pd] fragment are non
equivalent, along with a broad signal, which is consistent
with the phosphorus atom P3 of the tertiary phosphane–
borane group [MePh(cyclohexenyl)P(BH3)]. These observa-
tions are in agreement with a palladium complex, having
the 3 phosphorus atoms in its coordination sphere. How-
ever, no large Jpp coupling constant (about 300 Hz) between
the phosphorus atoms of the phosphane–borane group and
one of the DUPHOS group could be observed contrarily to
what was observed in 5 (JP1P3 = 294 Hz, see Table 3). In-
Figure 8. 31P{H} spectrum of the reaction mixture after reductive
elimination in [D8]THF/[D7]DMF with the signal locked to [D8]-
THF at 20 °C.
Table 5. 31P{H} NMR spectroscopic data for 9 at 20 °C.[a]
Complex δ(P1)
55.2
δ(P2)
δ(P3)
J12
J13
J23
9
53.9
16.0
56.7
8.9
26.5
[a] Solvent: [D8]THF/[D7]DMF with the signal locked to [D8]THF.
was monitored by 31P{H} NMR spectroscopy (Figure 9).
As expected, increase of the temperature led to the evol-
ution of complex 9 into the coupling product 2 and
Pd[(S,S)-Me-Duphos]2. This reaction proved to be irrevers-
ible since the decrease of the temperature from 50 to 20 °C
did not modify the phosphorus NMR spectrum, indicating
that the palladium complex initially involved in the com-
plexation has evolved (Figure 9).
stead, rather small coupling constants (JP1P3 = 8.9, JP2P3
=
26.5 Hz) were measured (Table 5). From this set of observa-
tions, we can determine that phosphorus P3 in [MePh(cy-
clohexenyl)P(BH3)] is not directly bound to the metal centre
in the palladium complex. We propose structure 9 for this
new complex, which would arise from the coordination of
the C=C bond of the cyclohexenyl group of the coupling
product 2 to the unstable [(Me-DUPHOS)Pd] complex re-
leased in the reaction mixture during the reductive elimi-
nation step. A second lower-intensity signal (set of doublet
of doublets) close to that of the major compound was also
observed. This signal might be due to an isomer of 9, in
which palladium binds the other enantioface of the alkene.
It is well known that coordination of an olefin fragment
to a transition metal complex lowers the stretching fre-
quency of olefin C=C bond.[24] Thus, to get more infor-
mation, we performed the infrared spectrum of a reaction
mixture sample containing complex 9 and coupling product
2. The C=C absorption frequency in 2 is observed at
1649 cm–1 while that of 9 is at 1633 cm–1 (see Supporting
Information). This low-frequency shift of the C=C band is
in agreement with the coordination of the cyclohexenyl
C=C bond on the Pd0.
To confirm the structure of 9, a sample of the reaction
mixture containing compound 2 and complex 9, was heated
Figure 9. 31P{H} spectrum of the reaction mixture resulting from
reductive elimination, at different temperatures between 20 °C and
into the NMR probe from 20 °C to 50 °C and evolution 50 °C, in [D8]THF/[D7]DMF with the signal locked to [D8]THF.
Eur. J. Inorg. Chem. 2011, 2489–2498
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2493

A.-C. Gaumont et al.
FULL PAPER
Hydrolysis of the reaction mixture containing tertiary Table 6. 31P{H} NMR spectroscopic data for diastereomers 5 and
6 at –30 °C.[a]
phosphane–borane 2 and Pd[(S,S)-Me-Duphos]2, followed
by extraction and purification by flash column chromatog-
raphy afforded pure coupling product 2. HPLC analysis
showed that enantiomer e2 (e2: tR 9.34 min, enantiomer e1:
tR 10.95 min, see Supporting Information) was formed with
a 98:2 er. This result indicates that under the conditions
used (transmetallation at low temperature and reductive eli-
mination at room temp.), loss of P-stereochemistry is not
observed.[25] This information is significant since it suggests
that interconversion of phosphide enantiomers 4 does not
occur before Pd–P bond formation or occurs more slowly
than transmetallation at –60 °C (Scheme 2). This observa-
Complex
δ(P1)
δ(P2) δ(P3)
JP1P2
JP1P3
JP2P3
5maj[b]
5min[c]
6maj[b]
6min[c]
60.0
57.6
59.4
58.7
61.6
58.1
60.8
61.5
–21.5
26.0
25.1
25.5
24.7
294.4
300.3
280.1
289.0
24.3
[d]
29.7
[e]
–24.8
–22.5
30.5
tion also highlights the high electrophilic reactivity of 3 [a] Solvent: [D8]THF/[D7]DMF with the signal locked to [D8]THF.
[b] Major conformational isomer. [c] Minor conformational isomer.
even at low temperature. The group of Glueck having pre-
[d] Broad doublet is overlapped by the signal of 1, which is in ex-
viously demonstrated in the synthesis of the Pamp ligand
cess. [e] Not observed.
that P–C reductive elimination from PdII derivatives pro-
ceeded through retention of configuration,[11] (SP) configu-
As anticipated, at –10 °C, signals of the minor isomers
ration was assigned to the major enantiomer of phosphane–
almost disappeared leading to diastereomers (S,S,RP)-5 and
borane 2 obtained in the coupling. The correlation with
(S,S,SP)-6 in an approximate 1:1.4 ratio (within experimen-
HPLC, thus enabled us to assign (SP) configuration to en-
antiomer e2 and (RP) to enantiomer e1.
tal error due to the inaccuracy of 31P NMR integration).
The 31P{H} NMR spectrum and data are reported in Fig-
ure 11 and Table 7, respectively.
Study of the Catalytic Cycle by vt-NMR Spectroscopy
Using (+/–)-1
The mechanistic investigation was then performed with
a 3:1 ratio of racemic phosphane–borane 1 and complex 3
in order to parallel the conditions used in catalytic reactions
(Scheme 2). Other conditions were similar to those em-
ployed in the mechanistic investigation using enantio-en-
riched (SP)-1. Upon reaction between rac-1 and complex 3,
four new sets of signals were observed by 31P vt-NMR at
temperatures between –60 °C and –10 °C. 31P{H} NMR
spectrum and data are reported in Figure 10 and Table 6,
respectively. The observed 31P NMR signals were attributed
to the two expected diastereomers of the transmetallation
complex 5maj and 6maj and to their conformational iso-
mers 5min and 6min (Table 6).
Figure 11. 31P{H} spectrum of complex 5 and 6 in [D8]THF/[D7]-
DMF with the signal locked to [D8]THF at 0 °C.
Table 7. 31P{H} NMR spectroscopic data for diastereomers 5 and
6 at 0 °C.[a]
Complex δ(P1)
δ(P2)
δ(P3)
JP1P2
JP1P3
JP2P3
5
6
59.9
59.5
61.7
61.0
–21.8
–24.5
26.0
25.8
294.3
281.8
24.4
[b]
[a] Solvent: [D8]THF/[D7]DMF (10 equiv. to Pd complex) with the
signal locked to [D8]THF. [b] Not observed.
Since no interconversion of anion 4 was observed at low
temperature using enantiopure (S)-1, the fact that dia-
stereomers (S,S,RP)-5 and (S,S,SP)-6 are formed in unequal
Figure 10. 31P{H} spectrum of complex 5 and 6 in [D8]THF/[D7]-
DMF with the signal locked to [D8]THF at –30 °C.
2494
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2489–2498

Pd-Catalysed Asymmetric Phosphination
vice (activated alumina column containing a copper catalyst and
molecular sieves). Pentane was distilled from calcium hydride be-
fore use.
amounts suggests that one enantiomer, here (RP)-4, reacts
more quickly with chiral 3 than the other one, i.e. (SP)-4,
(kR Ͼ kS), Scheme 4.
The solvents used for the preparation of complexes were distilled
(or purified by an Innovative technology Pure Solv. device) and
degassed. CDCl3, CD2Cl2, CD3CN, [D8]THF and [D7]DMF were
dried with molecular sieves (4 Å) before use.
Column chromatography was performed on Merck silica gel Si 60
(40–63 μm). Solvents were used as purchased. Thin-layer
chromatography (TLC) was performed on silica gel 60 F-254 plates
(0.1 mm) with iodine or UV detection.
1H, 13C, 19F and 31P NMR spectra were obtained on Bruker DPX
250 or AC 400 spectrometers, 11B NMR spectra were recorded on
1
a Bruker AC 400 spectrometer. H and 13C NMR chemical shifts
are reported relative to Me4Si used as an internal standard. 31P,
19F and 11B NMR chemical shifts are reported relative to H3PO4
(85%), CFCl3 and BF3·Et2O used as external references, respec-
tively. Coupling constants are reported in Hertz (Hz). The follow-
Scheme 4. Faster reaction of (RP)-4 vs. (SP)-4.
Warming the solution to 20 °C did not show any signifi-
cant modification of the diastereomeric ratio. Approximate ing symbols were used to describe signal multiplicities: s stands
for singlet, d for doublet, t for triplet, q for quadruplet, quint for
quintuplet, sept for septuplet, oct for octuplet, m for multiplet, and
br. for broad.
half-lives of 5 and 6 under these conditions are 1.25 h and
1 h, respectively. At 20 °C, diastereomers 5 and 6 smoothly
undergo reductive elimination at rather similar rates leading
to the formation of coupling product 2 in a 68:32 enantio- Elemental analyses were obtained from a ThermoQuest NA 2500
CHNS-O instrument.
meric ratio. As expected HPLC analysis shows that enanti-
omer e1, i.e. (RP)-2, is the major one. These observations
are consistent: i) with a reductive elimination, which pro-
ceeds with retention of configuration, enantiomer (RP)-2
arising from the major diastereomer (S,S,SP)-6,[23] and ii)
with a faster reaction of (RP)-4 with 3 compared to (SP)-4.
Mass spectroscopy was performed on a QTOF Micro WATERS.
Infrared spectroscopy was performed on a Perkin–Elmer Spectrum
One ATR-FTIR spectrometer.
High-Performance Liquid Chromatography (HPLC) separations
were achieved using the following components: For flow up to
0.5 mL/min, the components are: a Waters 600 pump, a Waters 996
photodiode array detector (190–250 nm) and Millenium software.
For flow below 0.5 mL/min, the components are: a Waters Alliance
2695 pump, a Waters 2996 photodiode array detector (190–250 nm)
and Empower Software.
Conclusions
In summary, we have determined the individual steps and
characterized the main intermediates in a potential mecha-
nism for the [(S,S)-Me-DUPHOS]PdCl2-catalysed coupling
reaction involving a racemic secondary phosphane–borane
and an alkenyl triflate. This mechanistic study also provides
information on the origin of enantioselectivity. Both Pd–P
(transmetallation) and P–C bond formation (reductive eli-
mination) proceeds with retention of configuration; and the
enantioselectivity is proposed to result from a kinetic reso-
lution when transmetallation and reductive elimination are
performed under mild conditions (–60 °C for transmetalla-
tion and room temp. for reductive elimination). Influence
of higher temperature and extension to more dissymmetric
and bulky secondary phosphane–boranes in order to favour
the anion interconversion and improve the transmetallation
selectivity are currently in progress.
Enantiomeric ratio of compound 2 was measured by HPLC using
two different conditions, determined using racemic 2:[5]
Conditions A: Daicel Chiralpack OJ-H (250ϫ4.6 mm, LϫID)
column (n-heptane/2-propanol 90:10, 1 mLmin–1, tR1 = 9.34 min:
enantiomer e2, tR2 = 10.95 min: enantiomer e1) at 20 °C.
Conditions B: Daicel Chiralpack AD-H (250ϫ4.6 mm, LϫID)
column [n-heptane/(MeOH/EtOH, 50:50) 98:2, 0.2 mLmin–1, tR1
37.08 min, tR2 = 39.48 min] at 12 °C
=
Given enantiomeric ratio of compound 2 are average of at least
two experiments.
Xray structures were recorded using a Bruker Kappa Apex II CCD
diffractometer.
Synthesis of (Sp)-(Methyl)phenylphosphane–borane (1): The en-
antio-enriched (Sp)-(methyl)phenylphosphane–borane 1 (98:2 e.r.)
was prepared by asymmetric lithiation/trappingϪreductive elimi-
nation as reported by Livinghouse.[9] The enantiomeric ratio of
compound 1 was measured by HPLC using conditions described in
the literature:[25a] Daicel Chiralpack OJ-H (250ϫ4.6 mm, LϫID)
column (n-heptane/2-propanol 90:10, 1 mLmin–1, tR1 = 14.72 min,
tR2 = 15.54 min) at 20 °C.
Experimental Section
General: All reactions were carried out under nitrogen atmosphere.
All glassware was flame-dried before use. (Methyl)phenylphos-
phane–borane[8] and 1-iodocyclohexene[13] were prepared accord-
ing to literature procedures. (S,S)-Me-DUPHOS, commercially
available, was used as purchased.
Synthesis of Pd(TMEDA)(cyclohexenyl)(I) (7): In a Schlenk tube,
flushed under nitrogen, TMEDA (102 μL, 0.68 mmol, 2.6 equiv.)
and 1-iodocyclohexene (130 mg, 0.62 mmol, 2.4 equiv.) are added
to a dark suspension of Pd2(dba)3·CHCl3 (269.5 mg, 0.26 mmol) in
Toluene, diethyl ether, tetrahydrofuran, acetonitrile and dichloro-
methane were purified by an Innovative technology Pure Solv. de-
Eur. J. Inorg. Chem. 2011, 2489–2498
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2495

A.-C. Gaumont et al.
C16), 36.7 (s, C9), 36.6 (d, JCP = 19.2 Hz, overlapping with other
FULL PAPER
1
toluene (3 mL). The reaction mixture is heated to 50 °C for 3 h in
darkness. During the reaction, the mixture turns green and then
yellow. After cooling to room temperature, the reaction mixture is
filtered and an orange solid is obtained. This solid is washed with
diethyl ether (6ϫ2 mL) to remove dba, then dissolved in dichloro-
methane, and filtered (through a 0.2 μm Nylon filter) to remove Pd
metal particles. The solution is concentrated under reduced pres-
sure affording a yellow-orange solid (172 mg, 77%).
signals, C17), 36.4 (d, JCP = 2.8 Hz, C15), 35.9 (d, JCP = 4.7 Hz,
2
2
4
4
C10), 33.7 (br., C8), 29.5 (dd, JCP = 12.2, JCP = 2.7 Hz, C2), 26.5
(d, JCP = 6.5 Hz, C5), 25.6 (s, C4), 24.0 (s, C3), 17.4 (d, JCP
=
3
2
12.4 Hz, C13), 15.9 (d, JCP = 8.7 Hz, C12), 14.7 (d, JCP = 1.7 Hz,
2
2
C18), 14.4 (d, JCP = 1.0 Hz, C7) ppm. C24H37IP2Pd (620.82): C
2
46.43, H 6.01; found C 46.21, H 6.28. HRMS calcd. for C24H37IN-
aP2Pd [M + Na]+: 643.0348, found 643.0328.
Synthesis of the Cationic Complex [Pd{(S,S)-MeDUPHOS}-
(cyclohexenyl)]+ (TfO)– (3): In a Schlenk tube, flushed under nitro-
gen, a cooled solution (–20 °C) of [Pd{(S,S)-Me-DUPHOS}-
(cyclohexenyl)](I) (20 mg, 0.032 mmol) in CD3CN (0.5 mL) is
added to AgOTf (8.3 mg, 0.032 mmol) at –20 °C in darkness. After
5 min, the precipitate is removed by filtration (through a 0.2 μm
Nylon filter) leading to a clear and pale-yellow solution which is
immediately used for NMR analysis.
1H NMR (400.1 MHz, CD2Cl2): δ = 5.02–4.96 (m, 1 H, H1), 3.01–
2.87 (m, 1 H, H9a), 2.71–2.44 (m, 2 H, H9b and H10a), 2.65 (s, 3 H,
H7), 2.56 (s, 6 H, H8 and H11), 2.51 (s, 3 H, H12), 2.42–2.29 (m, 2
H, H5a and H10b), 2.22–2.09 (m, 3 H, H2 and H5b), 1.69–1.43 (m,
4 H, H3 and H4) ppm. 13C NMR (100.6 MHz, CD2Cl2): δ = 142.8
(s, C6), 123.3 (s, C1), 62.3 (s, C9), 58.1 (s, C10), 51.5 (s, C8), 50.7 (s,
C11), 48.1 (s, C12), 47.6 (s, C7), 38.8 (s, C5), 28.5 (s, C2), 26.5 (s,
C4), 24.4 (s, C3) ppm. C12H25IN2Pd·0.11CH2Cl2 (438.75): C 32.85,
H 5.74, N 6.30; found C 32.87, H 6.08, N 6.52. HRMS calcd. for
C12H25IN2Pd·CH3CN [M
+
H]+: 472.0441, found 472.0457
(CH3CN is the solvent used for HRMS analysis).
2
31P NMR (161.9 MHz, CD3CN, at –20 °C): δ = 68.6 (d, JPP
=
Synthesis of [Pd{(S,S)-Me-DUPHOS}(cyclohexenyl)](I) (8): In a
Schlenk tube, flushed under nitrogen, a solution of Pd(TMEDA)-
(cyclohexenyl)(I) (140 mg, 0.32 mmol) and (S,S)-Me-DUPHOS
(99.6 mg, 0.32 mmol) in THF (3.6 mL) is stirred in darkness at
40 °C for 2 h. After cooling to room temperature, the reaction mix-
ture is partially concentrated under reduced pressure and the
orange product is precipitated by addition of pentane at 0 °C, fil-
tered and washed with pentane (5ϫ3 mL). The orange solid ob-
tained is dissolved in dichloromethane, and filtered (through
0.2 μm Nylon filter). The solution is concentrated under reduced
pressure giving a pale orange solid (169 mg, 84%).
28.8 Hz), 65.4 (d, JPP = 28.8 Hz) ppm. 1H NMR (400.1 MHz,
CD3CN, at –20 °C): δ = 8.02–7.63 (m, 4 H, H21, H22, H23 and H24),
5.51–5.42 (m, 1 H, H1), 3.16–3.00 (m, 1 H, H17), 2.89–2.69 (m, 3
2
H, H8, H11 and H14), 2.58–2.19 (m, 6 H, H4a H9a, H10a, H15 and
,
H16a), 2.15–2.06 (m, 2 H, H2), 2.02–1.94 (m, 1 H, H4b), 1.92–1.75
(m, 3 H, H9b, H10b and H16b), 1.73–1.60 (m, 4 H, H3 and H5), 1.37
(dd, JHP = 18.0, JHH = 7.2 Hz, 3 H, H13), 1.32 (dd, JHP = 18.0,
3
3
3
3JHH = 6.8 Hz, 3 H, H12), 0.84 (dd, JHP = 14.8, JHH = 7.2 Hz, 3
3
3
H, H7), 0.83 (dd, JHP = 15.6, JHH = 7.2 Hz, 3 H, H18) ppm. 13C
3
3
2
NMR (100.6 MHz, CD3CN, at –20 °C): δ = 153.9 (d, JCP
=
112.7 Hz, C6), 143.1 (br., C19 or C20), 137.9 (br., C19 or C20), 134.2
(d, JCP = 13.0 Hz, C21), 133.7 (d, JCP = 16.2 Hz, C24), 132.4 (d,
2
2
3JCP = 6.5 Hz, C22 or C23), 132.2 (br., C22 or C23), 128.3 (br., C1),
1
121.3 (q, JCF = 320.7 Hz, overlapping with solvent signal, CF3),
42.9 (br., C11), 40.4 (d, JCP = 19.4 Hz, C14), 37.0 (s, C16), 36.6 (d,
1
2JCP = 4.0 Hz, C9), 36.4 (d, 1JCP = 20.8 Hz, C17), 36.3 (s, C15), 35.9
(d, JCP = 5.3 Hz, C10), 35.1 (br., C8), 29.0 (dd, JCP = 11.3, JCP
2
4
4
= 3.6 Hz, C2), 25.9 (d, 3JCP = 5.7 Hz, C5), 25.7 (s, C4), 23.8 (s, C3),
17.6 (d, JCP = 12.7 Hz, C13), 16.1 (d, JCP = 6.2 Hz, C12), 14.1 (d,
2JCP = 2.5 Hz, C18), 14.0 (s, C7) ppm. 19F NMR (376.5 MHz, [D7]-
DMF, at –50 °C): δ = –79.4 (s) ppm. HRMS calcd. for C24H37P2Pd
[M]+: 493.1405, found 493.1384. ESI MS (CF3O3S): m/z 149 [M]–
2
2
2
31P NMR (161.9 MHz, CDCl3): δ = 65.9 (d, JPP = 31.8 Hz), 60.4
(d, 2JPP = 31.8 Hz) ppm. 1H NMR (400.1 MHz, CDCl3): δ = 7.77–
7.64 (m, 2 H, H21 and H24), 7.64–7.56 (m, 2 H, H22 and H23), 5.45
NMR Spectroscopy Monitoring of the Catalytic Cycle
4
(dm, JHP = 12.0 Hz, 1 H, H1), 3.62–3.46 (m, 1 H, H17), 3.30–3.10
(m, 1 H, H8), 2.80–2.60 (m, 1 H, H11), 2.56–2.40 (m, 2 H, H4a and
H14), 2.38–2.26 (m, 3 H, H9a, H10a and H16a), 2.26–2.04 (m, 5 H,
From (Sp)-(Methyl)phenylphosphane–Borane (1). Stoichiometric
Conditions: In a Schlenk tube, flushed under nitrogen, a cooled
H2, H4b and H15), 1.92–1.72 (m, 3 H, H9b, H10b and H16b), 1.72– solution (–60 °C) of the oxidative addition complex 3 (30 mg,
3
3
1.62 (m, 4 H, H3 and H5), 1.50 (dd, JHP = 18.4, JHH = 6.8 Hz, 3
0.048 mmol) in a [D8]THF (0.2 mL)/[D7]DMF (38 μL) solvent mix-
ture is added to AgOTf (12.4 mg, 0.048 mmol) at –60 °C in dark-
H, H13), 1.40 (dd, 3JHP = 18.8, JHH = 6.8 Hz, 3 H, H12), 0.93 (dd,
3
3JHP = 15.6, JHH = 7.6 Hz, 3 H, H7), 0.87 (dd, JHP = 14.0, JHH ness. The reaction mixture is stirred for 5 min giving a yellow pre-
3
3
3
= 7.2 Hz, 3 H, H18) ppm. 13C NMR (100.6 MHz, CDCl3): δ = cipitate. The solution is filtered (through a 0.2 μm Nylon filter),
153.7 (d, JCP = 130.4 Hz, C6), 144.7 (dd, JCP = 45.7, JCP
=
affording a clear and pale-yellow solution, which was transferred
into a precooled NMR tube (–60 °C) fitted with a rubber septum.
In another Schlenk, flushed under nitrogen, (Sp)-(methyl)phenyl-
phosphane–borane 1 (6.7 μL, 0.048 mmol, 98:2 e.r.) is added to a
solution of Me3SiOK (7.45 mg, 0.058 mmol, 1.2 equiv.) in [D8]THF
2
1
2
33.1 Hz, C19 or C20), 143.0 (dd, JCP = 31.9, JCP = 23.7 Hz, C19
1
2
or C20), 133.2 (d, JCP = 13.9 Hz, C21), 132.4 (d, JCP = 15.4 Hz,
C24), 130.8 (s, C22 or C23), 130.7 (s, C22 or C23), 127.4 (br., C1),
43.5 (br., C11), 42.2 (d, 1JCP = 18.7 Hz, C14), 37.7 (d, 2JCP = 2.7 Hz,
2
2
2496
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2489–2498

Pd-Catalysed Asymmetric Phosphination
(0.2 mL) at –60 °C. The solution is stirred for 1 min at –60 °C and
then transferred rapidly into the precooled NMR tube (–60 °C)
containing the solution of the cationic oxidative addition complex.
The tube is stirred and placed into the precooled NMR probe and
the reaction is monitored by 31P{H} NMR spectroscopy from
–60 °C to 20 °C. At 20 °C, the transmetallation adduct 5 undergoes
reductive elimination. The reaction is completed after 14 h, giving
the cross-coupling product. Then, the mixture is hydrolyzed with
degassed water (0.5 mL). The aqueous layer is extracted with di-
ethyl ether (3ϫ2 mL). The combined organic layers are dried with
MgSO4 and concentrated under reduced pressure. The crude prod-
uct is then purified by silica-gel column chromatography with pen-
tane/AcOEt (95:5, Rf = 0.30) as eluent affording the pure coupling
product 2[5] as a yellow oil with an enantiomeric ratio of 98:2 in
favour of enantiomer e2 (the enantiomeric ratio was measured by
chiral HPLC using conditions described above, see Supporting In-
formation).
raphy with pentane/AcOEt (95:5, Rf = 0.30) as eluent affording the
pure coupling product 2[5] as a yellow oil with an enantiomeric
ratio of 68:32 in favour of enantiomer e1 (Enantiomeric ratio is
measured by HPLC using conditions described above, see Support-
ing Information).
31P NMR Characterization of the Transmetallation Adducts 5 and
6: 31P NMR {161.9 MHz, [D7]DMF (10 equiv. to Pd complex)/
2
[D8]THF (signal locked), at –30 °C}: δ = 61.6 (dd, JP2P1 = 26.0,
2
2
2JP2P3 = 24.3 Hz, P2, 5maj), 61.5 (dd, JP2P1 = 24.7, JP2P3
=
30.5 Hz, P2, 6 min), 60.8 (apparent t, JP2P1 = 25.5 Hz, P2, 6maj),
2
60.0 (dd, JP1P2 = 26.0, JP1P3 = 294.4 Hz, P1, 5maj), 59.4 (dd,
2
2
2JP1P2 = 25.5, 2JP1P3 = 280.1 Hz, P1, 6maj), 58.7 (dd, 2JP1P2 = 24.7,
2
2
2JP1P3 = 289.0 Hz, P1, 6 min), 58.1 (dd, JP2P1 = 25.1, JP2P3
=
29.7 Hz, P2, 5min), 57.6 (dd, JP1P2 = 25.1, JP1P3 = 300.3 Hz, P1,
2
2
5min), –21.5 (br. d, 2JP3P1 = 294.4 Hz, P3, 5maj), –22.5 (br. d, 2JP3P1
= 289.0 Hz, P3, 6 min), –24.8 (br. d, JP3P1 = 280.1 Hz, P3, 6maj)
2
ppm, P3 of complex 5min is not observed (overlapping by signal of
31P NMR Characterization of the Transmetallation Adduct 5: 31P
secondary phosphane–borane).
31P NMR {161.9 MHz, [D7]DMF (10 equiv. to Pd complex)/[D8]-
NMR {161.9 MHz, [D7]DMF (10 equiv. to Pd complex)/[D8]THF
2
2
2
2
(signal locked), at –30 °C}: δ = 63.3 (dd, JP2P1 = 26.0, JP2P3
=
THF (signal locked), at 0 °C}: δ = 61.7 (dd, JP2P1 = 26.0, JP2P3
2
2
24.4 Hz, P2, 5maj), 61.6 (dd, JP1P2 = 26.0, JP1P3 = 294.2 Hz, P1,
2
= 24.4 Hz, P2, 5), 61.0 (apparent t, JP2P1 = 25.8 Hz, P2, 6), 59.9
2
2
5maj), 59.8 (dd, JP2P1 = 25.1, JP2P3 = 29.8 Hz, P2, 5min), 59.2
(dd, 2JP1P2 = 26.0, 2JP1P3 = 294.3 Hz, P1, 5), 59.5 (dd, 2JP1P2 = 25.8,
(dd, 2JP1P2 = 25.1, 2JP1P3 = 300.1 Hz, P1, 5min), –15.1 (br. d, 2JP3P1
2
2JP1P3 = 281.8 Hz, P1, 6), –21.8 (br. d, JP3P1 = 294.3 Hz, P3, 5),
2
= 300.1 Hz, P3, 5min), –19.7 (br. d, JP3P1 = 294.2 Hz, P3, 5maj)
2
–24.5 (br. d, JP3P1 = 281.8 Hz, P3, 6) ppm.
ppm.
Supporting Information (see also the footnote on the first page of
this article): HPLC of 1 and 2, NMR characterisation of 3, 5, 6, 7
and 8, X-ray structure of 5 and 7, IR spectra of 2 and 9.
31P NMR {161.9 MHz, [D7]DMF (10 equiv. to Pd complex)/[D8]-
2
2
THF (signal locked), at 0 °C}: δ = 63.4 (dd, JP2P1 = 26.0, JP2P3
= 24.5 Hz, P2), 61.6 (dd, 2JP1P2 = 26.0, 2JP1P3 = 294.0 Hz, P1), –19.7
2
(br. d, JP3P1 = 294.0 Hz, P3) ppm.
Acknowledgments
Recrystallisation of Diastereomerically Pure Transmetallation Ad-
duct 5: At –10 °C, only the diastereomer (S,S,RP) is observed by
31P NMR spectroscopy. The reaction is stopped at –10 °C and the
reaction mixture is concentrated under reduced pressure. The re-
sulting residue is dissolved in CH2Cl2 (0.4 mL), filtered (through a
0.2 μm Nylon filter), and transferred into a precooled NMR tube
(–20 °C) followed by addition of pentane. By this way, orange crys-
tals suitable for X-ray analysis are obtained by slow diffusion of
pentane into the CH2Cl2 solution at –20 °C.
This work has been performed within the PUNCHOrga interre-
gional network. We gratefully acknowledge financial support from
the Ministère de la Recherche et des Nouvelles Technologies, Cen-
tre National de la Recherche Scientifique (CNRS), the Région
Basse-Normandie and the European Union, FEDER funding). We
thank Sébastien Thomas from the Laboratoire de Catalyse et Spec-
trochimie de l’ENSICAEN for IR analysis. Margareth Lemarié and
Rémi Legay are acknowledged for HPLC analyses and NMR stud-
ies, respectively. European Cooperation in the Field of Science and
Technology (COST) ACTION CM0802 “PHOSCINET” is also ac-
knowledged for support. Lastly, the authors would like to thank a
referee for helpful comments and suggestions.
From Excess Racemic (Methyl)phenylphosphane–borane (1): In a
Schlenk tube, flushed under nitrogen, a cooled solution (–60 °C) of
the oxidative addition complex 3 (30 mg, 0.048 mmol) in a [D8]-
THF (0.2 mL)/[D7]DMF (38 μL, 0.48 mmol, 10 equiv.) solvent
mixture is added to AgOTf (12.4 mg, 0.048 mmol) at –60 °C in
darkness. The reaction mixture is stirred for 5min giving a yellow
precipitate. The solution is filtered (through a 0.2 μm Nylon filter),
resulting in a clear and pale-yellow solution and then transferred
into a precooled NMR tube (–60 °C) fitted with a rubber septum.
In another Schlenk, flushed under nitrogen, racemic (methyl)phen-
ylphosphane–borane (+/–)-1 (20 μL, 0.145 mmol, 3 equiv.) is added
to a solution of Me3SiOK (22.3 mg, 0.174 mmol, 3.6 equiv.) in
[D8]THF (0.2 mL) at –60 °C. The solution is stirred for 30 min at
–60 °C and then transferred into the precooled NMR tube (–60 °C)
containing the solution of the cationic oxidative addition complex.
The tube is stirred and placed into the precooled NMR probe and
the reaction is monitored by 31P{H} NMR spectroscopy from
–60 °C to 20 °C. At 20 °C, the transmetallation adducts 5 and 6
undergo reductive elimination. The reaction is completed after
14 h, giving the cross-coupling product. Then, the mixture is hy-
drolyzed with degassed water (0.5 mL). The aqueous layer is ex-
tracted with diethyl ether (3ϫ2 mL). The combined organic layers
are dried with MgSO4 and concentrated under reduced pressure.
The crude product is then purified by silica-gel column chromatog-
[1] For a recent review on the synthesis of alkenylphosphanes, see:
D. Julienne, O. Delacroix, A.-C. Gaumont, Curr. Org. Chem.
2010, 14, 457–482.
[2] For a recent review on the application of alkenylphosphanes,
see: D. Julienne, F. Toulgoat, O. Delacroix, A.-C. Gaumont,
Curr. Org. Chem. 2010, 14, 1195–1222.
[3] B. Join, D. Mimeau, O. Delacroix, A.-C. Gaumont, Chem.
Commun. 2006, 3249–3251.
[4] D. Julienne, O. Delacroix, A. C. Gaumont, C. R. Chim. 2010,
13, 1099–1103.
[5] D. Julienne, J.-F. Lohier, O. Delacroix, A.-C. Gaumont, J. Org.
Chem. 2007, 72, 2247–2250.
[6] a) D. S. Glueck, Chem. Eur. J. 2008, 14, 7108–7117; b) D. S.
Glueck, Synlett 2007, 2627–2634, and references cited therein;
c) S. Pican, A.-C. Gaumont, Chem. Commun. 2005, 2393–2395;
d) A.-C. Gaumont, M. B. Hursthouse, S. J. Coles, J. M. Brown,
Chem. Commun. 1999, 63–64.
[7] For related cationic complexes having an alkenyl group, sta-
bility and kinetics of formation by oxidative addition between
alkenyl triflates and Pd0(PPh3)4 were reported: A. Jutand, S.
Négri, Organometallics 2003, 22, 4229–4237.
Eur. J. Inorg. Chem. 2011, 2489–2498
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2497

A.-C. Gaumont et al.
FULL PAPER
[8]
[9]
[16] Me3SiOK, was selected to ensure the complete deprotonation
of the secondary phosphane–borane at –60 °C, weaker base
such as K3PO4 leading to an incomplete reaction.
[17] This observation parallels the one made by Glueck in the palla-
dium-catalyzed coupling of aryl iodide with borane-free sec-
ondary phosphanes, see: N. F. Blank, J. R. Moncarz, T. J.
Brunker, C. Scriban, B. J. Anderson, O. Amir, D. S. Glueck,
L. N. Zakharov, J. A. Golen, C. D. Incarvito, A. L. Rheingold,
J. Am. Chem. Soc. 2007, 129, 6847–6858.
[18] Reaction between 1 and Me3SiOK is completed in 1 min at
–60 °C in [D8]THF yielding phosphido-borane anion 4.
[19] M. D. Fryzuk, G. R. Giesbrecht, S. J. Retting, Inorg. Chem.
1998, 37, 6928–6934.
[20] This ratio depends on reaction conditions. We could not deter-
mine if this ratio of atropoisomers represents a kinetic or a
thermodynamic preference or depends on other solution com-
ponents.
[21] J. M. Brown, J. J. Pérez-Torrente, N. W. Alcock, Organometal-
lics 1995, 14, 1195–1203.
H. Lebel, S. Morin, V. Paquet, Org. Lett. 2003, 5, 2347–2349.
Enantio-enriched (Sp)-1 (98:2 e.r.) was prepared by (–)-sparte-
ine-mediated asymmetric lithiation/trapping-reductive elimi-
nation as reported by: B. Wolfe, T. Livinghouse, J. Org. Chem.
2001, 66, 1514–1516. Only the (S) enantiomer is available by
this method.
Related alkenyl derivatives being depicted as unstable species
(see ref. 7), compound 3 could not be obtained by direct oxidat-
ive addition to [(S,S)-Me-Duphos]Pd0. It was prepared follow-
ing chemistry developed by Brown et al. for the synthesis of
arylic cationic oxidative complexes involved in the Heck reac-
tion: J. M. Brown, K. K. Hii, Angew. Chem. Int. Ed. Engl. 1996,
35, 657–659.
a) J. R. Moncarz, T. J. Brunker, J. C. Jewett, M. Orchowski,
D. S. Glueck, R. D. Sommer, K.-C. Lam, C. D. Incarvito, T. E.
Concolino, C. Ceccarelli, L. N. Zakharov, A. L. Rheingold, Or-
ganometallics 2003, 22, 3205–3221; b) J. R. Moncarz, T. J.
Brunker, D. S. Glueck, R. D. Sommer, A. L. Rheingold, J. Am.
Chem. Soc. 2003, 125, 1180–1181.
[10]
[11]
[22] Experimental details of the structure determination can be
found in the Supporting Information. CCDC-696196 contains
the supplementary crystallographic data for compound 5.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Center via www.ccdc.cam.ac.uk/data_r-
equest/cif.
[12] D. Drago, P. S. Pregosin, Organometallics 2002, 21, 1208–1215.
[13] For the preparation of iodocyclohexene, see: a) K. Lee, D. F.
Wiemer, Tetrahedron Lett. 1993, 34, 2433–2436; for characteri-
zation, see: b) J. K. Stille, J. H. Simpson, J. Am. Chem. Soc.
1987, 109, 2138–2152; c) A. G. Martinez, R. M. Alvarez, A. G.
Fraile, L. R. Subramanian, H. Hanack, Synthesis 1986, 222–
224.
[23] The apparent inversion is due to a Cahn–Ingold–Prelog pri-
ority change.
[14] Experimental details of the structure determination can be
found in the Supporting Information. CCDC-696197 contains
the supplementary crystallographic data for compound 7.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[15] THF was selected to ensure the complete deprotonation of the
secondary phosphane–borane at low temperature, and DMF
to stabilize the cationic complex 3.
[24] M. J. Grogan, K. Nakamoto, J. Am. Chem. Soc. 1968, 90, 918–
922.
[25] For boron phosphide interconversion see, for example: a) T.
Imamoto, T. Oshiki, T. Onozawa, M. Matsuo, T. Hikosaka, M.
Yanagawa, Heteroat. Chem. 1992, 3, 563–575; b) M. Al-Ma-
sum, G. Kumuraswamy, T. Livinghouse, J. Org. Chem. 2000,
65, 4776–4778.
Received: September 15, 2010
Published Online: January 12, 2011
2498
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2489–2498