Primary and secondary aminophosphines as novel P-stereogenic building blocks for ligand synthesis

Article abstract of DOI:10.1002/anie.201004041

Set the N free! The reactivity of the amino group of P-stereogenic aminophosphines allows the further elaboration of the aminophosphine unit whilst preserving the original chirality of the phosphorus atom (see picture; Rh green). P-stereogenic aminodiphosphine ligands can easily be prepared in optically pure forms, feature distinct structural and electronic characteristics, and can be used in asymmetric hydrogenation reactions. Copyright

Full text of DOI:10.1002/anie.201004041

Communications
DOI: 10.1002/anie.201004041
P Ligands
Primary and Secondary Aminophosphines as Novel P-Stereogenic
Building Blocks for Ligand Synthesis**
Marc Revꢀs, Catalina Ferrer, Thierry Leꢁn, Sean Doran, Pablo Etayo, Anton Vidal-Ferran,
Antoni Riera,* and Xavier Verdaguer*
Chiral phosphine ligands are central to asymmetric metal
catalysis.[1] The effect of the majority of these ligands arises
from the chirality of their backbones; however, P-stereogenic
(P*) ligands have garnered renewed interest.[2] After the
decisive work of Knowles and co-workers with PAMP and
DIPAMP ligands, several efficient syntheses of all-carbon P*
compounds have been reported.[3,4] In contrast, P* com-
pounds that contain heteroatoms directly linked to the
Scheme 1. Strategy explored to synthesize P*-aminophosphine building
blocks.
phosphorus center are scarce, and have found little applica-
tion in catalysis. This class of substances includes secondary
phosphine oxides, which exist in equilibrium with their
trivalent phosphinite form.[5] P* aminophosphines, which are
the corresponding nitrogen analogues, are even more rare,[6]
as free primary aminophosphines tend to dimerize with the
evolution of ammonia.[7] However, Kolodiazhnyi et al. have
reported that borane aminophosphines of type I are stable
and that they can be obtained in diastereomerically pure form
using 2-phenylethylamine as a chiral amine (Scheme 1).[8]
Nonetheless, type I compounds do not have any reported
applications in asymmetric catalysis, nor has their hydro-
genolysis been described. We envisioned that reductive
cleavage of the arylethyl fragment should provide borane-
protected primary aminophosphines of type II, which would
be amenable to further transformations and become useful P*
building blocks in catalysis. Herein, we report the synthesis of
enantiopure P-chiral primary and secondary aminophos-
phines (II) and diphosphinoamines (III).
We began by investigating the hydrogenolysis of the
known compound 1a, which contains tert-butyl-
(phenyl)phosphinamine moiety (Scheme 2), under various
a
Scheme 2. Synthesis of primary and secondary aminophosphines 3
and 4. Reagents and conditions: a) Li/NH3, tBuOH, À788C, THF;
b) KMnO4, Al2O3, 08C, acetone; c) nBuLi, MeI, THF.
[*] M. Revꢀs, Dr. C. Ferrer, T. Leꢁn, S. Doran, Prof. A. Riera,
Prof. X. Verdaguer
Unitat de Recerca en Sꢂntesi Asimꢃtrica (URSA)-PCB
Institute for Research in Biomedicine (IRB) Barcelona and
Departament de Quꢂmica Orgꢄnica, Universitat de Barcelona
c/Baldiri Reixac 10, 08028 Barcelona (Spain)
Fax: (+34)403-7095
E-mail: antoni.riera@irbbarcelona.org
xavier.verdaguer@irbbarcelona.org
Homepage: http://www.ursapcb.org
reaction conditions; however, all initial attempts failed. We
were pleased to find that reaction of 1a with lithium in
ammonia afforded the desired cleavage of the benzylic
fragment. Although the phenyl group attached to phosphorus
underwent concomitant Birch-type reduction to form the
phosphinodiene 2 (Scheme 2), oxidation of 2 with KMnO4 on
alumina resulted in 3; to the best of our knowledge, this is the
first report of this compound.[9,10] We were able to circumvent
the undesired Birch-type reduction by preparing the naphthyl
derivative 1b. Reductive cleavage of the naphthylethyl frag-
ment in 1b occurs before the undesired Birch reduction of the
phosphine phenyl group. This behavior, which may be
attributed to the difference between the reduction potentials
of the naphthalene and benzene rings,[11] enabled the efficient
syntheses of the primary and secondary aminophosphines 3
and 4, respectively, from 1b (Scheme 2). Notably, no isomer-
ization of the phosphorus center was detected under these
conditions. Thus, the use of diastereomerically pure 1a or 1b
Dr. P. Etayo, Dr. A. Vidal-Ferran
Institute of Chemical Research of Catalonia (ICIQ)
Adva. dels Paꢅsos Catalans 16, 43007 Tarragona (Spain)
Dr. A. Vidal-Ferran
Instituciꢁ Catalana de Recerca i Estudis Avanꢆants (ICREA)
Passeig Lluꢂs Companys 23, 08018 Barcelona (Spain)
[**] We thank MICINN (CTQ2008-00763/BQU and CTQ2008-00950/
BQU), the Generalitat de Catalunya (2009SGR 00901), IRB
Barcelona, and Enantia for financial support. M.R. thanks the MEC
for a fellowship. T.L. thanks the Generalitat de Catalunya for a
Fellowship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201004041.
9452
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 9452 –9455

Angewandte
Chemie
provided optically pure P* aminophosphines (as determined
by HPLC on a chiral stationary phase). Most gratifyingly,
compounds 3 and 4 were stable crystalline solids.
Having optimized the hydrogenolysis, we then sought to
develop a practical and useful methodology for the diaste-
reoselective synthesis of aminophosphines. Substitution reac-
tions of racemic chlorophosphines with chiral alcohols and
amines are known to occur under dynamic kinetic resolution
at the phosphorus center to produce an uneven mixture of the
diastereomers 1 and 1’.[12] Thus, we tested several commer-
cially available chiral amines with (Æ )-tBuPhPCl and
(Æ)-tBuMePCl (Table 1). Among the amines tested, use of
Scheme 3. Synthesis of primary and secondary aminophosphines 5
and 6. Reagents and conditions: a) Li/NH3, tBuOH, À788C, THF;
b) NaH, MeI, RT, DMF.
Table 1: Diastereoselective synthesis of bulky aminophosphine boranes.
result in novel structures that preserve the original P chirality.
A straightforward application of these P* aminophosphines is
the preparation of chiral aminodiphosphine (PNP) ligands.
Although many PNP ligands have been reported, they have
met little success in asymmetric catalysis,[14] as most chiral
PNP ligands reported to date derive from chiral amines, and
therefore, the chiral center remains far away from the metal
center. Thus, the aminophosphines 3–6 are ideally suited for
the synthesis of PNP* and P*NP* ligands. As a proof of
principle, we synthesized a series of diphosphine ligands from
the N-methylaminophosphines 4 and 6 (Scheme 4). Forma-
Entry
R1
R2, R3
Yield [%][a]
d.r.[b]
Product
1
2
3
4
5
6
7
8
9
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Ph, Me
1-naph, Me
p-tol, Me
p-MeOC6H4, Me
p-ClC6H4, Me
Ph, CONH2
Ph, Me
1-naph, Me
Ph, CONH2
87
91
88
90
88
95
91
90
98
5:1
1a,1a’
1b,1b’
1c,1c’
1d,1d’
1e,1e’
1 f,1 f’
1g,1g’
1h,1h’
1i,1i’
6:1[c]
4:1[c]
4:1[c]
4.5:1[c]
1:1
1:2[d]
1:3[d]
1.5:1[d]
[a] Yields correspond to product mixtures purified by flash chromatog-
1
raphy. [b] Diastereomeric ratios were determined by H NMR spectros-
copy. [c] Absolute configuration at the phosphorus center was deter-
mined by comparison with data in Ref. [8]. [d] Absolute configuration at
the phosphorus center was determined from the X-ray crystal structure of
the resulting Rh complex 14.
1-naphtylethylamine resulted in the highest selectivities for
both chlorophosphines (Table 1, entries 2 and 8). Phenyl-
glycinamide was also found to be a highly efficient resolving
agent. Although we encountered low diastereoselectivities,
we were able to readily separate both sets of diastereomers
(1 f/1 f’ and 1i/1i’) by chromatography, and, more importantly,
these products were highly crystalline, which facilitated their
purification. Thus, based on reactivity and separability, we
chose 1-naphtylethylamine for the preparation of 3 and 4
(Scheme 2). Alternatively, we found phenylglycinamide to be
the most effective resolving agent for the syntheses of 5 and 6
(Scheme 3). Diastereomerically pure 1i and 1i’ (> 99% de)
were obtained by flash chromatography (SiO2, CH2Cl2/
EtOAc 1:1). Multigram quantities of 1i were obtained by
crystallization in EtOH.[13] Reductive cleavage of 1i provided
ready access to optically pure aminophosphine 5 (as deter-
mined by GC on a chiral support; Scheme 3). Alternatively,
exhaustive methylation of 1i with NaH and MeI followed by
treatment with Li/NH3 afforded the N-methyl secondary
aminophosphine 6.
Scheme 4. P-stereogenic aminodiphosphine ligands derived from
(+)-4 and (+)-6. Cy=cyclohexyl; Tol=tolyl.
tion of the anion with nBuLi and reaction with R2PCl
provided the C1-symmetric ligands 7–10 with identical optical
purity as the starting phosphinoamine. The enantiomeric
excess for ligands 7 and 8 was determined to be greater than
99% by HPLC on a chiral stationary phase, thus ruling out
racemization at this point. Alternatively, reaction of 6 with
racemic tBuMePCl provided a mixture of the meso compound
and the C2-symmetric ligand 11 in 41 and 32% yield
respectively.[15] Ligand 11 is a nitrogen analogue of the
ligand miniPHOS reported by Imamoto and co-workers.[16]
The presence of an alkyl group on the nitrogen atoms
generally stabilizes the resulting PNP ligands. However, the
free primary aminophosphine is required for reaction with the
bulky tBu2PCl. Thus, reaction of optically pure (+)-5 with
tBu2PCl provided the bisphosphinamine 12 (Scheme 5).
À
The aminophosphines 3–6 are attractive building blocks
for the construction of chiral ligands. The reactivity of the
amino group should permit further functionalization that can
Compound 12 was found to exist exclusively as its P H
tautomer, which prevented the oxidation of phosphorus.[17]
Removal of the borane moiety with HBF4 in MeOH at 608C
Angew. Chem. Int. Ed. 2010, 49, 9452 –9455
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9453

Communications
Table 2: Asymmetric hydrogenation reactions using either phosphonium
salt 13 or rhodium complex 14.
Entry
Substrate
Cat.[a] (mol%)
H2 [bar]
ee[b] [%]
1
2
A (0.3)[c]
B (1.0)
3
3
99
99
3
4
A (0.3)
B (3.0)
3
99
99
10
Scheme 5. Synthesis of the aminodiphosphine rhodium complex 14.
Reagents and conditions: a) NaH, tBu2PCl, 608C, THF. b) HBF4, 658C,
MeOH. c) [Rh(cod)2]BF4, Na2CO3, THF. cod=1,5-cyclooctadiene.
[a] Catalyst A: complex 14. Catalyst B: Rh complex formed in situ by
mixing iPr2EtN or Et3N, 13 and [Rh(cod)2]BF4, in a 2.4/1.2/1 ratio. All
hydrogenation reactions resulted in greater than 99% conversion.
[b] Enantiomeric excess was determined by GC or HPLC on a chiral
stationary phase. [c] Reaction time less than 9 h.
provided the air-stable phosphonium salt 13 in 93% yield.[18]
Compound 13 is a nitrogen-containing analogue of the
trichickenfootPHOS (TCFP) ligand reported by Hoge et
al.; this compound is one of the most efficient ligands ever
developed for asymmetric hydrogenation reactions.[19] Finally,
reaction with [Rh(cod)2]BF4 in the presence of sodium
carbonate provided the corresponding cationic complex 14.
The X-ray structure of 14 is shown in Figure 1.[20] The solid-
iently generated in situ by mixing [Rh(cod)2]BF4, 13, and an
organic base. High levels of selectivity were also achieved by
following this procedure (Table 2, entry 2). Finally, reduction
of N-(3,4-dihydronaphthalen-2-yl)acetamide provided the
corresponding amine in 99% ee (Table 2, entry 4). This
pharmaceutically relevant intermediate is known to be a
difficult substrate for asymmetric hydrogenation reaction.[21]
To the best of our knowledge, our result represents the
highest enantioselectivity ever reported for this compound.[22]
In summary, we have reported the first synthesis of
optically pure, borane-protected primary and secondary
aminophosphines. These compounds were found to be
valuable P-stereogenic building blocks in catalysis, as dem-
onstrated with the synthesis of new, chiral asymmetric
aminodiphosphine ligands. These building blocks are easily
prepared in optically pure form, and, owing to their distinct
structural and electronic characteristics, are ideal for exploi-
tation in catalysis.
Received: July 2, 2010
Revised: September 2, 2010
Published online: October 26, 2010
Keywords: asymmetric catalysis · cleavage reactions · P ligands ·
.
phosphorus · rhodium
Figure 1. Crystal structure of complex 14 (ORTEP drawing showing
thermal ellipsoids at 50% probability). The BF4 counterion is omitted
for clarity.
À
[1] Phosphorus Ligands in Asymmetric Catalysis, Vol. I–III (Ed.: A.
Bꢁrner), Wiley-WCH, Weinheim, 2008.
[2] For a review of P-stereogenic compounds, see: A. Grabulosa, J.
Granell, G. Muller, Coord. Chem. Rev. 2007, 251, 25.
state structure confirmed that the aminodiphosphine ligand
worked as a C1-symmetric P–P ligand. The P-N-P angle in 14
is 103.048, which is wider than the P-C-P angle in TCFP.
[3] a) W. S. Knowles, M. J. Sabacky, B. D. Vineyard, D. J. Weinkauff,
J. Am. Chem. Soc. 1975, 97, 2567; b) B. D. Vineyard, W. S.
Knowles, M. J. Sabacky, G. L. Bachman, D. J. Weinkauff, J. Am.
Chem. Soc. 1977, 99, 5946; PAMP: o-anisylmethylphenylphos-
phine, DIPAMP: 1,2-bis(o-anisylphenylphosphino)ethane.
[4] a) S. Jugꢂ, M. Stephan, J. A. Laffitte, J. P. Genet, Tetrahedron
Lett. 1990, 31, 6357; b) T. Imamoto, K. Sugita, K. Yoshida, J. Am.
Chem. Soc. 2005, 127, 11934; c) T. Imamoto, J. Watanabe, Y.
Wada, H. Masuda, H. Yamada, H. Tsuruta, S. Matsukawa, K.
Yamaguchi, J. Am. Chem. Soc. 1998, 120, 1635; d) B. Wolfe, T.
Livinghouse, J. Am. Chem. Soc. 1998, 120, 5116; e) E. Bergin,
C. T. OꢃConnor, S. B. Robinson, E. M. McGarrigle, C. P. OꢃMah-
À
However, the P N bonds (1.701 and 1.688 ꢀ) in 14 are
shorter than those in TCFP, thus resulting in a bite angle of
70.08, which is slightly smaller than that of TCFP (72.68).[19a]
We undertook a preliminary study of the performance of
14 in asymmetric hydrogenation reactions in order to check
whether aminodiphosphines could be efficient ligands in
catalysis (Table 2). (Z)-a-Acetamidocinnamate and (E)-b-
acetamidobutanoate were reduced with complete selectivity
using 0.3 mol% of 14 as catalyst (Table 2, entries 1 and 3).
Alternatively, the active ligand–metal species can be conven-
9454 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 9452 –9455

Angewandte
Chemie
ony, D. G. Gilheany, J. Am. Chem. Soc. 2007, 129, 9566; f) W.
Tang, X. Zhang, Angew. Chem. 2002, 114, 1682; Angew. Chem.
Int. Ed. 2002, 41, 1612; g) C. Scriban, D. S. Glueck, J. Am. Chem.
Soc. 2006, 128, 2788; h) F. Maienza, F. Spindler, M. Thommen, B.
Pugin, C. Malan, A. Mezzetti, J. Org. Chem. 2002, 67, 5239; i) R.
den Heeten, B. H. G. Swennenhuis, P. W. N. M. van Leeuwen,
J. G. de Vries, P. C. J. Kamer, Angew. Chem. 2008, 120, 6704;
Angew. Chem. Int. Ed. 2008, 47, 6602; j) J. M. Brown, J. C. P.
Laing, J. Organomet. Chem. 1997, 529, 435.
[13] A single crystallization process of the initial 1.5:1 mixture of 1i/
1i’ afforded a crystalline solid with 98:2 d.r. An additional
crystallization from EtOH afforded 1i/1i’ with > 99:1 d.r.
(determined by HPLC on a chiral stationary phase).
[14] a) T. S. Venkatakrishnan, S. K. Mandal, R. Kannan, S. S. Krish-
namurthy, M. Nethaji, J. Organomet. Chem. 2007, 692, 1875;
b) D. Konya, F. Robert, Y. Gimbert, A. E. Greene, Tetrahedron
Lett. 2004, 45, 6975.
[15] Reaction of 6 with a twofold excess of racemic tBuMePCl
resulted in similar ratios between meso and C2 isomers, and no
kinetic resolution was observed.
[5] a) D. Moraleda, D. Gatineau, D. Martin, L. Giordano, G. Buono,
ˇ
Chem. Commun. 2008, 3031; b) M. Stankevic, K. M. Pietrusie-
wicz, J. Org. Chem. 2007, 72, 816; c) J. Bigeault, L. Giordano, G.
Buono, Angew. Chem. 2005, 117, 4831; Angew. Chem. Int. Ed.
2005, 44, 4753; d) X. Jiang, A. J. Minnaard, B. Hessen, B. L.
Feringa, A. L. L. Duchateau, J. G. O. Andrien, J. A. F. Boogers,
J. G. de Vries, Org. Lett. 2003, 5, 1503; e) G. Y. Li, Angew. Chem.
2001, 113, 1561; Angew. Chem. Int. Ed. 2001, 40, 1513; f) R. K.
Haynes, R. N. Freeman, C. R. Mitchell, S. C. Vonwiller, J. Org.
Chem. 1994, 59, 2919.
[16] D. Gridnev, Y. Yamanoi, N. Higashi, H. Tsuruta, M. Yasutake,
T. Imamoto, Adv. Synth. Catal. 2001, 343, 118; miniPHOS:
(R,R)-1,1-bis(tert-butylmethylphosphino)methane.
[17] a) A. M. Caminade, E. Ocando, J. P. Majoral, M. Cristante, G.
Bertrand, Inorg. Chem. 1986, 25, 712; b) J. S. Ritch, T. Chivers,
D. J. Eisler, H. M. Tuononen, Chem. Eur. J. 2007, 13, 4643.
[18] In general, aminophosphines are prone to hydrolysis and
alcoholysis, especially under acidic conditions. However, the
presence of bulky tert-butyl groups on the phosphorus atom
makes the aminophosphines less likely to undergo hydrolysis,
see: C. Bauduin, D. Moulin, E. B. Kaloun, C. Darcel, S. Jugꢂ,
J. Org. Chem. 2003, 68, 4293. The fact that ligand 13 resists harsh
acidic deprotection conditions (5 equiv HBF4, MeOH, 658C,
overnight) without decomposition demonstrates its stability
towards hydrolysis in protic solvents.
[6] For the chromatographic resolution of tBuPh(Me2N)P, see: J.
Naubron, L. Giordano, F. Fotiadu, T. Buergi, N. Vanthuyne, C.
Roussel, G. Buono, J. Org. Chem. 2006, 71, 5586.
[7] Dimerization can be avoided by attaching bulky substituents to
the phosphorus atom, see: A. Jana, G. Schwab, H. W. Roesky, D.
Stalke, Inorg. Chem. 2008, 47, 8990.
[8] O. I. Kolodiazhnyi, E. V. Gryshkun, N. V. Andrushko, M.
Freytag, P. G. Jones, R. Schmutzler, Tetrahedron: Asymmetry
2003, 14, 181.
[19] a) G. Hoge, H. Wu, W. S. Kissel, D. A. Pflum, D. J. Greene, J.
Bao, J. Am. Chem. Soc. 2004, 126, 5966; b) I. D. Gridnev, T.
Imamoto, G. Hoge, M. Kouchi, H. Takahashi, J. Am. Chem. Soc.
2008, 130, 2560.
[9] The Birch-type reduction of phenylphosphines has been recently
ˇ
described; see: M. Stankevic, K. M. Pietrusiewicz, Tetrahedron
Lett. 2009, 50, 7093.
[10] C. M. McBride, W. Chrisman, C. E. Harris, B. Singaram,
Tetrahedron Lett. 1999, 40, 45.
[20] CCDC 782690 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[21] J. Meeuwissen, M. Kuil, A. M. van der Burg, A. J. Sandee,
J. N. H. Reek, Chem. Eur. J. 2009, 15, 10272.
[22] The highest enantioselectivity described for this hydrogenation
reaction is 94% ee, see: X. Jiang, L. Lefort, P. E. Goudriaan,
A. H. M. de Vries, P. W. N. M. van Leeuwen, J. G. de Vries,
J. N. H. Reek, Angew. Chem. 2006, 118, 1245; Angew. Chem.
Int. Ed. 2006, 45, 1223.
[11] K. Meerholz, J. Heinze, J. Am. Chem. Soc. 1989, 111, 2325.
[12] This phenomenon is known to occur with either phosphorus(III)
and phosphorus(V) chloride, see: a) K. M. Pietrusiewicz, M.
Zablocka, Chem. Rev. 1994, 94, 1375; b) I. Fernꢄndez, N. Khiar,
A. Roca, A. Benabra, A. Alcudia, J. L. Espartero, F. Alcudia,
Tetrahedron Lett. 1999, 40, 2029; c) T. A. Hamor, W. B. Jennings,
C. J. Lovely, K. A. Reeves, J. Chem. Soc. Perkin Trans. 2 1992,
843; d) J. F. Cavalier, F. Fotiadu, R. Verger, G. Buono, Synlett
1998, 73.
Angew. Chem. Int. Ed. 2010, 49, 9452 –9455
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9455