Isolation of the reactive intermediate in palladium-catalysed coupling of secondary phosphine-boranes with aryl halides

Article abstract of DOI:10.1039/a807830k

Ph₂PH·BH₃ is converted into its K salt with KOSiMe₃ in THF, and the anion reacts with (dppp)Pd(Ph)I to produce the complexed secondary phosphine-borane which decomposes to PPh₃·BH₃ at -10°C; the C

Full text of DOI:10.1039/a807830k

Isolation of the reactive intermediate in palladium-catalysed coupling of
secondary phosphine–boranes with aryl halides
Annie-Claude Gaumont,a Michael B. Hursthouse,b Simon J. Colesb and John M. Browna
a Dyson Perrins Laboratory, South Parks Road, Oxford, UK OX1 3QY. E-mail: bjm@ermine.ox.ac.uk
b Department of Chemistry, University of Wales, Cardiff, UK CF1 3TB
Received (in Cambridge, UK) 8th October 1998, Accepted 5th November 1998
R1
I
Ph2PH·BH3 is converted into its K salt with KOSiMe3 in
THF, and the anion reacts with (dppp)Pd(Ph)I to produce
the complexed secondary phosphine–borane which decom-
poses to PPh3·BH3 at 210 °C; the C6F5 analogue has been
crystallographically characterised.
R2
R2
Ph
KOSiMe3
–
K+
P2Pd
P
H
P
BH3
BH3
Ph
I–
R1
I
R1
The ability of palladium complexes to effect catalysed s-bond
couplings has been extended significantly in recent years with
some emphasis on the formation of C–heteroatom bonds.1
Formation of C–P bonds has been achieved by the Pd-catalysed
reaction of secondary phosphine oxides R2P(H)O with aryl
triflates;2 if the phosphine rather than the phosphine oxide is
desired then either Ni or Pd catalysis with R2PH as the
nucleophile may be effective.3 Less success has been achieved
with secondary phosphine boranes, although the coupling
reaction with ArI has been demonstrated.4
P2Pd
P2Pd
Ph
R2
P
H3B
R2
R1
P
BH3
R1, R2 = Aryl
Ph
Scheme 2
the solution above 230 °C, but intermediates were too ill-
defined to characterise. Hence our attention shifted to the
corresponding dppp complexes, known5,8 to react more slowly
than their dppf analogues in the reductive elimination step as a
consequence of the smaller chelate bite angle.
Following Imamoto’s work,4 satisfactory conditions were
established for the catalytic synthesis of triarylphosphine–
boranes, e.g. Ar3P·BH3 1a–d from ArI, but less cleanly from
aryl triflates (Scheme 1). An important observation was that the
Addition of KOSiMe3 to an equimolar mixture of complex 3a
and Ph2PH·BH3 in THF at 270 °C led to an immediate and
quantitative change to a single new species whose distinctive
31P NMR spectrum is shown in Fig. 1, revealing the trans-
relationship of (P1–Pd–P3–B). This is consistent with the
structure 4a, whose spectrum persisted up to 0 °C. At that
temperature it decomposed (half-life ca. 30 min.) giving rise
only to Ph3P·BH3 and an unidentified Pd product; when reaction
was carried out in the presence of an excess of PhI, species 3a
was regenerated. The same type of transformation was shown to
occur with other secondary phosphine–boranes and other PdArI
complexes, as indicated. With more electron-withdrawing aryl
groups in Ar–Pd, the elimination step was slower.
R2
R2
i
H
P
+
I
P
BH3
BH3
R1
R1
Ph
Ph
R1
H
R2
1a
b
c
Ph
Ph
Ph
p-MeO
o-F
d
H
o-An
Scheme 1 Reagents and conditions: i, dppfPdCl2 (5 mol%), K2CO3 (2
equiv.), MeCN, (for 1a) room temp., 24 h, 92%; (for 1b) room temp., 18 h,
89%; (for 1c) 50 °C, 48 h, 74%; (for 1d) 30 °C, 24 h, 90%.
The linkage between the stoichiometric and catalytic reac-
tions was strengthened by experiments where the breakdown of
adduct 4a, which led quantitatively to the formation of
Ph3P·BH3 1a, was followed by 31P NMR (Fig. 2). An identical
reaction was carried out with a four-fold excess of both the
reaction took place at ambient temperature or below (0 °C).
Monitoring by 31P NMR demonstrated that after completion of
the reaction, only (dppf)Pd(Ar)I 2 was present when an excess
of ArI was used.5 Initial attempts at defining intermediates in
the catalytic reaction were unsuccessful. When complex 2 (Ar =
Ph) was generated in situ from either the corresponding
(dppf)PdC8H8 complex6 or (dppf)Pd(CH2NCHCO2Me)7 rapid
2
reaction with K+Ph2P·BH3 (optimally generated from the
secondary phosphine–borane and KOSiMe3; quantitative by
NMR spectroscopy) was observed at 270 °C in THF
(Scheme 2). The product Ph3P·BH3 could be characterised in
Ph2
Ph2
P
Ph2
P
P
R1
Ar
I
R
I
Pd
Pd
Fe
Pd
PPhR2•BH3
P
P
P
Ph2
Ph2
Ph2
4a R1 = Ph, R2 = Ph
1 = C6F5, R2 = Ph
3a R = Ph
2
b
R
b
R = C6F5
Ph2
P
Ph2
P
Ph
Me
Me
Me
Pd
Pd
Fe
Fig. 1 The 31P NMR spectrum of complex 4b in THF solution at 250 °C,
taken directly after mixing of precursors at 270 °C. Assignments: d 9.0 (P3,
JP2P3 31 Hz), 5.2 (P1, JP1P3 297, JP1P2 43 Hz), 21.35 (P2). On standing at
or above 0 °C the reductive elimination product grows in at ca. d 1.
N(C6H4CH3)2
Ph
P
H
Me
Fe
P
P
Ph2
P
OC
Ph2
Ph
BH3
CO
5
6
7
Chem. Commun., 1999, 63–64
63

These experiments indicate that the catalytic cycle for
phosphinylation of aryl iodides with secondary phosphine–
boranes is very simple, and the true reactive intermediate is
amenable to characterisation. Unlike the corresponding inter-
mediate 6 observed in catalytic amination,11 it reacts rapidly
below ambient temperature and very likely does not involve
prior chelate dissociation. Studies of the reductive elimination
1
of neutral Pd phosphido complex 7 and analogous (dppe)Pd h -
1
Ar h -(PPh2) complexes have been carried out by Glueck and
co-workers,12 who generated these intermediates by deprotona-
tion of cationic phosphine complexes.
Future work will be directed towards stereochemical control
of the P–Pd addition step with a view to applications in the
asymmetric synthesis of arylphosphines.
We thank CNRS for granting leave of absence to A-C. G.,
EPSRC and NATO for support and Johnson-Matthey for loans
of Pd salts. We greatly appreciate an open exchange of
information with Professor David Glueck (Dartmouth).
Fig. 2 (5) Stoichiometric formation of borane 1a from breakdown of
complex 4a, 0.0532 in THF at 0 °C, reaction followed by loss of 4a in the
31P NMR spectrum, dotted line is smoothed curve. (8) Catalytic turnover of
a mixture of PhI and KPh2P·BH3 (both 0.213 ) in THF at 0 °C in the
presence of complex 3a (0.0532 ), followed by the appearance of borane
1a. The dashed line is the predicted rate of turnover based on the rate-
constant 2.9 3 1024 s21 derived for the stoichiometric process above.
M
M
M
Notes and references
† Selected data for 4b: dP(CD2Cl2, 202 MHz) 14.3 (P3, J1,3 296), 1.95 (P1,
J
1,2 49.5, J1,3 296) 21.2 (P2, J2,1 49.5, J2,3 34); dB(CD2Cl2, 80 MHz) 233.2
(br s); dF(CD2Cl2, 235 MHz) 2164.3 (m), 2163.4 (m), 2114.8 (m);
dH(CD2Cl2, 500MHz) 1.90 (m, CH2) , 2.35 (m, CH2P) , 2.55 (m, CH2P) ,
6.95, 7.04, 7.2, 7.24, 7.3, 7.4 (Har); dC(CD2Cl2, 125 MHz) 18.1 (CH2), 23.6
(CH2P), 27.3 (CH2P), 126.9 (Ar), 127.4 (Ar), 127.8 (Ar), 129.3 (Ar), 130.2
(Ar), 130.9 (Ar), 132.3 (Ar), 132.9 (Ar), 133.8 (Ar), 138.5 (Ar); m/z (ES,
+30 V) 883.02 ([M 2 1]+, 100%), 717.06 ([M 2 C6F5]+, 70%), 703.03 ([M
2 C6F5 2 BH3]+, 68%).
anion and PhI, and catalytic turnover was followed by the
appearance of 2a. Within experimental error, the rate of
catalytic turnover is defined by the rate of the reductive
elimination step. This provides firm support for the mechanism
suggested in Scheme 2. A correlation between reductive
elimination and catalytic turnover has been observed for the C–
C bond forming step in the breakdown of (dppf)Pd(Ph)Me.5
In the extreme case of 3b, the intermediate complex 4b was
formed cleanly at 270 °C but was stable in solution at ambient
temperature for a significant period of time (half-life ca. 96 h)
and could be isolated by filtration through silica and precipita-
tion. The isolated material gives rise to phosphine–borane 2e on
redissolution and prolonged standing, accompanied by some
decomposition to the phosphine oxide. In characterisation of
complex 4b, we observed an unusual [M 2 1] cation peak in the
electrospray mass spectrum† which will be the subject of
further investigation; such behaviour has previously been
observed in the mass spectra of amine–boranes.9 Recrystallisa-
tion (CH2Cl2, Et2O) gave blocks suitable for X-ray analysis.‡
The structure is shown in Fig. 3 and demonstrates a slightly
distorted square planar arrangement with the Pd–P bond of the
‡ Crystal data for C45H39BF5P3Pd·0.5CH2Cl2 Mr = 927.35, monoclinic,
space group P21/c, a = 11.481(2), b = 27.923(6), c = 14.005(3) Å, b =
107.15(3)°, U = 4290(2) Å3, Z = 4, Dc = 1.436 g cm23, m = 0.660 mm21
,
F(000) 1884, T = 150(2) K, Crystal size 0.18 3 0.14 3 0.10 mm, 6219
independent reflections, 15833 collected. Refinement method: full-matrix
least-squares on F2, Goodness-of-fit on F2
= 0.877, Final R indices
[I > 2s(I)] R1 = 0.0424, wR2 = 0.0951. CCDC 182/1086.
1 R. A. Widenhoefer and S. L. Buchwald, J. Am. Chem. Soc., 1998, 120,
6504; B. C. Hamann and J. F. Hartwig, J. Am. Chem. Soc., 1998, 120,
3694; D. Baranano, G. Mann and J. F. Hartwig, Curr. Org. Chem., 1997,
1, 287; S. Wagaw, R. A. Rennels and S. L. Buchwald, J. Am. Chem.
Soc.,1997,119, 8451 and earlier papers from these groups.
2 L. Kurz, G. Lee, D. Morgans, M. J. Waldyke and T. Ward, Tetrahedron
Lett., 1990, 31, 6321; J. M. Brown and H. Doucet, Tetrahedron:
Asymmetry, 1997, 8, 3775 .
3 D. W. Cai, J. F. Payack, D. R. Bender, D. L. Hughes, T. R. Verhoeven
and P. J. Reider, J. Org. Chem., 1994, 59, 7180; D. J. Ager, M. B. East,
A. Eisenstadt and S. A. Laneman, Chem. Commun., 1997, 2359; G.
Martorell, X. Garcias, M. Janura and J. M. Saa, J. Org. Chem., 1998, 63,
3463.
4 T. Imamoto, T. Oshiki, T. Onozawa, M. Matsuo, T. Hikosaka and M.
Yanagawa, Heteroatom Chem., 1992, 3, 799; T. Imamoto, M. Matsuo,
T. Nonomura, K. Kishikawa and M. Yanagawa, Heteroatom Chem.,
1993, 4, 475; T. Imamoto, T. Yoshizawa, K. Hirose, Y. Wada, H.
Masuda, K. Yamaguchi and H. Seki, Heteroatom Chem., 1995, 6, 99.
5 J. M. Brown and P. J. Guiry, Inorg. Chim. Acta, 1994, 220, 249.
6 J. M. Brown and N. A. Cooley, Organometallics, 1990, 9, 343; F.
Schager, K. J. Haack, R. Mynott, A. Rufinska and K. R. Porschke,
Organometallics, 1998, 17, 807.
7 A. Jutand, K. K. Hii, M. Thornton-Pett and J. M. Brown, unpublished
work.
8 M. Kranenburg, P. Kamer and P. Van Leeuwen, Eur. J. Inorg. Chem.
1998, 155; J. E. Marcone and K. C. Moloy, J. Am. Chem. Soc., 1998,
120, 8527.
9 Z. Polivka, V. Kubelka, N. Holubova and M. Ferles, Collect. Czech.
Chem. Commun., 1970, 35, 1131.
10 W. Angerer, W. S. Sheldrick and W. Malisch, Chem. Ber. 1985, 118,
1261; see also: D. Dou, G. L. Wood, E. N. Duesler, R. T. Paine and H.
Noth, Inorg. Chem., 1992, 31, 1695; D. A. Hoic, W. M. Davis and G. C.
Fu, J. Am. Chem. Soc., 1996, 118, 8176.
11 M. S. Driver and J. F. Hartwig, J. Am. Chem. Soc., 1997, 119, 8232.
12 D. K. Wicht, I. V. Kourkine, B. M. Lew, J. M. Nthenge and D. S.
Glueck, J. Am. Chem. Soc., 1997, 119, 5039; M. D. Zhuravel, R. D.
Sweeder and D. S. Glueck, to be published; see D. K. Wicht, S. N.
Paisner, B. M. Lew, D. S. Glueck, G. Yap, L. M. LiableSands, A. L.
Rheingold, C. M. Haar and S. P. Nolan, Organometallics, 1998, 17, 652
for related platinum chemistry.
Fig. 3 The X-ray crystal structure of complex 4b. Selected bond lengths and
angles: Pd(1)–P(1) 2.337(2), Pd(1)–P(2) 2.334(2), Pd(1)–P(3) 2.375(2),
Pd(1)–C(4) 2.053(6), P(3)–B(1) 1.925(6); C(4)–Pd(1)–P(2) 170.4(2), C(4)–
Pd(1)–P(1) 87.6(2), P(2)–Pd(1)–P(1) 90.84(6), C(4)–Pd(1)–P(3) 88.0(2),
P(2)–Pd(1)–P(3) 94.35(6), P(1)–Pd(1)–P(3) 173.09(6) B(1)–P(3)–Pd(1)
119.4(2).
phosphine borane only slightly longer than Pd–P bonds in the
chelate (0.237 vs. 0.233 nm). The only previous example of a
1
simple h -coordinated phosphine–borane structure is that of
complex 5, which is a stable isolable material.10
Communication 8/07830K
64
Chem. Commun., 1999, 63–64