Phosphorinanes as ligands for palladium-catalyzed cross-coupling chemistry

Article abstract of DOI:10.1021/ol052579h

(Chemical Equation Presented) Phosphorinanes are presented as a class of phosphine ligand suitable for organopalladium cross-coupling chemistry. Prepared via a direct double Michael addition of a monoalkyl- or arylphosphine to phorone followed by a Wolf-K

Full text of DOI:10.1021/ol052579h

ORGANIC
LETTERS
2006
Vol. 8, No. 1
103-105
Phosphorinanes as Ligands for
Palladium-Catalyzed Cross-Coupling
Chemistry
Tim Brenstrum,† Julie Clattenburg,† James Britten,† Serguei Zavorine, Jeff Dyck,‡
Alan J. Robertson,‡ James McNulty,*,† and Alfredo Capretta*,†
Department of Chemistry, McMaster UniVersity, Hamilton, Ontario, Canada L8S 4M1,
and Cytec Canada, Inc., P.O. Box 240, Niagara Falls, Ontario, Canada L2E 6T4
capretta@mcmaster.ca
Received October 24, 2005
ABSTRACT
Phosphorinanes are presented as a class of phosphine ligand suitable for organopalladium cross-coupling chemistry. Prepared via a direct
double Michael addition of a monoalkyl- or arylphosphine to phorone followed by a Wolf Kishner reduction, phosphorinanes are relatively
−
inexpensive to manufacture and allow modification of one of the alkyl moieties permitting steric and electronic fine-tuning of the ligands.
Library screening and applications of these ligands in the Suzuki, Sonogashira, ketone arylation, and aryl amination reactions are presented.
Palladium-catalyzed cross-coupling reactions are now among
the most prominent synthetic methods for mild, chemo- and
stereoselective carbon-carbon and carbon-heteroatom bond-
forming processes.1 The versatility and scope of this family
of cross-coupling reactions has expanded over the past few
years to now include readily available but less reactive aryl
chlorides and the once problematic â-hydride-containing
alkyl halides. The direct coupling of aliphatic â-hydride
containing electrophiles with the standard field of organo-
metallics (including those based on B, Mg, Li, Sn, Al, Zn,
and Zr) has moved from the “remarkable”2 to now standard
protocol.3,4 The success of this chemistry hinges upon the
choice of ligand employed. In particular, the popular tri-
tert-butylphosphane, tricyclohexylphosphane,3 and the di-
tert-butylalkylphosphane (tBu)2PR5-7 family have been used
successfully in addition to the generally useful biaryl ligands
that have emerged.8
In view of these developments, we were attracted to some
chemistry described by Cyanamid chemists over four decades
ago9 that appeared highly applicable as a general entry to a
new class of hindered aliphatic phosphane ligands. The
central scaffold of interest is the phosphorinane skeleton (3)
readily accessed through the direct double Michael addition
of a monoalkyl- or arylphosphine to phorone followed by a
Scheme 1
† McMaster University.
‡ Cytec Canada, Inc.
(1) Metal-Catalyzed Cross Coupling Reactions; Diederich, F., Stang, P.,
Eds.; Wiley-VCH: New York, 1998.
(2) Terao, J.; Watanabe, H.; Ikumi, A.; Kuniyasu, H.; Kambe, N. J. Am.
Chem. Soc. 2002, 124, 4222.
(3) Netherton, M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am. Chem.
Soc. 2001, 123, 10099.
Wolf-Kishner reduction of the central ketone (Scheme 1,
2). There has been no interest in their development as ligands
(4) Powell, A.; Fu, G. C. J. Am. Chem. Soc. 2004, 126, 7788.
(5) Netherton, M. R.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 3910.
(6) Menzel, K.; Fu, G. C. J. Am. Chem. Soc. 2003, 125, 3718.
(7) Lee, J. Y.; Fu, G. C. J. Am. Chem. Soc. 2003, 125, 5616.
(8) Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2004, 43, 1871.
(9) Welcher, R. P.; Johnson, G. A.; Wystrach, V. P. J. Am. Chem. Soc.
1960, 82, 4437. Welcher, R. P.; Day, N. E. J. Org. Chem. 1962, 27, 1824.
10.1021/ol052579h CCC: $33.50
© 2006 American Chemical Society
Published on Web 12/13/2005

and reagents until now.10 The present paper describes the
synthesis of a small library of phosphorinanes and demon-
strates their utility in palladium-catalyzed cross-coupling
chemistry.
of the particular reaction class. Our initial work focused on
the Suzuki coupling of p-bromoanisole with phenylboronic
acid, and while all the ligands successfully facilitated
coupling, ligand 3d revealed itself to be marginally superior
(entry 1, Table 1). Optimizing the conditions for this ligand
Phosphorinanes (3a-f) were obtained via the two-step
procedure outlined in Scheme 1. Phorone was treated with
the appropriate phosphine (1a-f) to provide ketones of the
general type 2. Reduction of the carbonyl (in order to avoid
any potential retro-Michael reaction) was achieved via
treatment with hydrazine under basic conditions. The phos-
phorinane products obtained were viscous oils or low melting
solids and were converted into the more easily handled, air-
stable, tetrafluoroborate salts.11 An X-ray structure for
cyclopentyl-substituted phosphorinane (3c) was obtained and
is illustrated in Figure 1.12 Overall, the procedure is high
Table 1. Selected Examples of Palladium Cross-Couplings
Facilitated by Phosphorinane Ligands
Figure 1. X-ray structure of 1-cyclopentyl-2,2,6,6-tetrameth-
-
ylphosphorinane (BF4 removed for clarity).
yielding and versatile allowing for the production of a family
of ligands.
a Set A: K3PO4‚H2O (2.4 equiv), Pd2(dba)3‚CHCl3 (2%), ligand 3d (4%),
THF. Set B: NaOtBu (1.5 equiv); Pd2(dba)3‚CHCl3 (1%), ligand 3d (2%).
i
With phosphorinanes 3a-f in hand, advantage was taken
of our previously established approach13 wherein the entire
ligand library is screened against a particular palladium-
catalyzed coupling reaction. This parallel screening quickly
establishes the superior ligand to be used for the optimization
Set C: Pr2NH (1.5 equiv); Pd(PhCN)2Cl2 (3%), ligand 3a (6%), CuI (2%).
Set D: KOtBu (1.5 equiv); Pd2(dba)3‚CHCl3 (1%), ligand 3d (2%). Set E:
NaOtBu (1.5 equiv); Pd2(dba)3‚CHCl3 (1%), ligand 3d (2%). b Isolated yields
averaged between two runs.
allowed for a series of Suzuki couplings to be carried out,
and selected examples are shown in Table 1 (entries 1-3).
Activated aryl chlorides such as 4-chloroacetophenone (entry
2) are readily coupled at room temperature, while as
expected, systems such as p-chloroanisole (entry 3) require
heating. In all cases, reaction times could be diminished at
the expense of increased temperature while maintaining
excellent yields.
Attention was turned to the coupling of enolates with aryl
halides. The R-arylation of propiophenone with 4-iodotoluene
was the system chosen to compare the activity of the ligand
library. Using previously optimized conditions,14 with sodium
tert-butoxide base, Pd2dba3‚CHCl3 as the palladium source,
and toluene as solvent, the norbornylphosphorinane 3d
clearly revealed itself again as the ligand of choice, showing
(10) McNulty, J.; Zhou, Y. Tetrahedron Lett. 2004, 45, 407.
(11) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295.
(12) Crystal size, 0.35 × 0.20 × 0.08 mm3; crystal system, monoclinic;
space group, P21; unit cell dimensions, a ) 7.98430(10) Å, b ) 13.9102-
(2) Å, c ) 15.4291(2) Å, R ) 90°, â ) 91.0590(10)°, γ ) 90°; volume,
1713.31(4) Å3; empirical formula, C14H28BF4P; formula weight, 314.14; Z
4; F calcd, 1.218 mg/m3; 2θmax, 136.32°; radiation, Cu KR; wavelength,
1.54178 Å; scan mode, ω scans; T, 293(2) K; reflections collected, 9024;
independent reflections, 4927 [Rint ) 0.0249]; Lorentzian correction applied,
polarization correction applied, absorption correction: SADABS (semiem-
pirical from equivalents, max and min transmission, 1.00 and 0.84);
absorption coefficient, 1.679 mm-1; direct methods solution (SHELXS);
refinement method, full-matrix least-squares on F2; refinement data [I >
-3σI]/restraints/parameters, 4927/61/432; H-atoms riding on C-atoms in
calculated positions, H-atoms on P-atoms found from difference map and
-
refined isotropically, BF4 disorders modeled; small disorder of C3′-C4′
orientation not refined; final R indices [I > σI], R1 ) 0.0571, wR2 )
0.1632; R indices (all data), R1 ) 0.0591, wR2 ) 0.1679; absolute structure
parameter, 0.02(3); extinction coefficient, 0.0032(7); largest diff peak and
hole, 0.296 and -0.270 e‚Å-3; structure deposited with CCDC.
(13) Brenstrum, T.; Gerristma, D. A.; Adjabeng, G. M.; Frampton, C.
S.; Britten, J.; Robertson, A. J.; McNulty, J.; Capretta, A. J. Org. Chem.
2004, 69, 7635.
(14) Adjabeng, G. M.; Brenstrum, T.; Frampton, C. S.; Robertson, A.
J.; Hillhouse, J.; McNulty, J.; Capretta, A. J. Org. Chem. 2004, 69, 5082.
104
Org. Lett., Vol. 8, No. 1, 2006

1
complete conversion (by H NMR) after just 6 h at room
employed (such as tBu3P, for example). The phosphorinanes,
however, are much easier to generate, relatively inexpensive
to manufacture, and allow for modification of one of the
alkyl moieties permitting steric and electronic fine-tuning
of the ligands. Furthermore, the intermediacy of the phos-
phorinones 2 (Scheme 1) allows entry to P-protected 4-hy-
droxyphosphorinanes10 through hydride reduction, providing
a handle by which this family of ligands could be im-
mobilized onto a solid support.15,16 In conclusion, the ease
of preparation of this family of ligands, structural variation
possible, efficacy in assisting [Pd]-mediated transformations,
and potential applicability to solid supported chemistry makes
this new family of ligands highly attractive. Further develop-
ments and applications along these lines are in progress in
our laboratories.
temperature. Using the optimized catalyst system containing
ligand 3d, aryl bromides such as bromobenzene (entry 4)
and sterically demanding 2,4,5-trimethylbromobenzene (entry
5) were readily coupled with propiophenone at room tem-
perature in high yields. Slightly higher temperatures were
required to facilitate coupling of the aryl chlorides (entries
6 and 7) in a reasonable time period.
Screening revealed that the isobutylphosphorinane (3a)
was the most effective ligand for the Sonogashira reaction.
In addition, optimization revealed that Pd(PhCN)2Cl2 was
the superior palladium source for these couplings. Quantita-
tive yields for activated iodides and bromides (entries 8 and
9) were obtained at room temperature, while deactivated
systems such as p-bromoanisole (entry 10) gave excellent
yields of aryl alkynes at room temperature as well.
Finally, aminations of aryl halides were shown to be best
promoted by ligand 3d. Under optimized conditions, aryl
bromides such as p-bromoanisole could be coupled readily
to morpholine (entry 11) and dipropylamine (entry 12) in
high yields at room temperature. Similarly, aryl chlorides
could be coupled to amines in high yields (entries 13 and
14) although NaOtBu was shown to be the optimum base
and higher temperatures were required in these cases.
Overall, the results above demonstrate that the phospho-
rinane family of trialkylphosphine ligands is clearly com-
parable in efficacy to other popular ligands currently
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada and Cytec Canada
for their financial support.
Supporting Information Available: Crystallographic
information, experimental procedures, and compound char-
acterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL052579H
(15) Leadbeater, N. E.; Marco, M. Chem. ReV. 2002, 102, 3217.
(16) Benaglia, M.; Puglisi, A.; Cozzi, F. Chem. ReV. 2003, 103, 3401.
Org. Lett., Vol. 8, No. 1, 2006
105