Asymmetric synthesis and metal complexes of a C 3-symmetric P-stereogenic triphosphine, (R)-MeSi(CH2PMe(t -Bu))3 (MT-Siliphos)

Article abstract of DOI:10.1021/om200930k

Asymmetric deprotonation of PMe₂(t-Bu)(BH₃) with s-BuLi/(-)-sparteine, followed by treatment with MeSiCl₃, gave a 2.6:1 mixture of the C₃- and C₁-symmetric triphosphine-boranes MeSi(CH₂PMe(t-Bu)(BH₃))₃ (3). Recrystallization gave highly diastereomerically and enantiomerically enriched C₃-3. After deprotection with morpholine, the triphosphine (R)-MeSi(CH₂PMe(t-Bu))₃ (4; (Me, t-Bu)-Siliphos or MT-Siliphos) was used to prepare the metal complexes [Rh(MT-Siliphos) (norbornadiene)][OTf] (5), Ru(MT-Siliphos)(DMSO)Cl₂ (6), [Ru(MT-Siliphos)(NCMe)₃][OTf]₂ (7), M(MT-Siliphos) (PPh₃) (M = Pd (8), Pt (9)), Cu(MT-Siliphos)(Cl) (10), and polymeric [Cu₂Cl₂(μ-(κ²,κ¹)-MT- Siliphos)]_∞ (11). The crystal structures of C₃-3 and of complexes 5, 7, and 11 were determined, and dynamic processes in 6 and 8 were characterized by variable-temperature NMR spectroscopy.

Full text of DOI:10.1021/om200930k

Communication
pubs.acs.org/Organometallics
Asymmetric Synthesis and Metal Complexes of a C3-Symmetric
P-Stereogenic Triphosphine, (R)-MeSi(CH2PMe(t-Bu))3 (MT-Siliphos)
Matthew F. Cain,† David S. Glueck,*,† James A. Golen,‡ and Arnold L. Rheingold‡
†Department of Chemistry, Dartmouth College, 6128 Burke Laboratory, Hanover, New Hampshire 03755, United States
‡Department of Chemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093, United States
S
* Supporting Information
ABSTRACT: Asymmetric deprotonation of PMe2(t-Bu)(BH3) with s-BuLi/
(−)-sparteine, followed by treatment with MeSiCl3, gave a 2.6:1 mixture of the
C3- and C1-symmetric triphosphine−boranes MeSi(CH2PMe(t-Bu)(BH3))3 (3).
Recrystallization gave highly diastereomerically and enantiomerically enriched C3-3.
After deprotection with morpholine, the triphosphine (R)-MeSi(CH2PMe(t-Bu))3
(4; (Me, t-Bu)-Siliphos or MT-Siliphos) was used to prepare the metal complexes
[Rh(MT-Siliphos)(norbornadiene)][OTf] (5), Ru(MT-Siliphos)(DMSO)Cl2 (6),
[Ru(MT-Siliphos)(NCMe)3][OTf]2 (7), M(MT-Siliphos)(PPh3) (M = Pd (8), Pt
(9)), Cu(MT-Siliphos)(Cl) (10), and polymeric [Cu2Cl2(μ-(κ2,κ1)-MT-Siliphos)]∞
(11). The crystal structures of C3-3 and of complexes 5, 7, and 11 were determined,
and dynamic processes in 6 and 8 were characterized by variable-temperature NMR
spectroscopy.
C2-symmetric diphosphines are valuable ligands for metal-
catalyzed asymmetric reactions,1 in which they reduce the
number of possible intermediates/transition states (Chart 1).
Chart 2. C3-Symmetric Chiral Triphosphines with C (1) and
P (2) Stereocenters
Chart 1. Homotopic Coordination sites in C2- and C3-
Symmetric Metal Complexes
in Rh-catalyzed acetalization.6 Its chemistry has also not been
pursued, likely because its synthesis required resolution of the
intermediate PMe(Ph)(t-Bu)(BH3) by preparative chromatog-
raphy on a chiral support.5
No analogues of C3-symmetric P-stereogenic Siliphos are
known.7 To enable investigation of the potential of such ligands
in asymmetric catalysis, we report here a new approach to
asymmetric synthesis of a related triphosphine, (R)-MeSi-
(CH2PMe(t-Bu))3 (4; (Me, t-Bu)-Siliphos or MT-Siliphos),
and describe its coordination chemistry in several late-metal
complexes.
For example, in a square-planar complex, coordination sites A
and B are identical (homotopic). Similarly, C3-symmetric
triphosphines were proposed to be useful in octahedral
complexes, where positions A−C are homotopic.2
Asymmetric deprotonation8 of PMe2(t-Bu)(BH3) with
s-BuLi/(−)-sparteine in ether at −78 °C,9 followed by treatment
with MeSiCl3, gave a 2.6:1 mixture of the air-stable diastereo-
meric C3- and C1-symmetric triphosphine−boranes MeSi-
(CH2PMe(t-Bu)(BH3))3 (3, Scheme 1).
Testing this hypothesis has been limited by the lack of
synthetic routes to C3-symmetric chiral triphosphines.3 Very
few of these ligands are known, and their use in coordination
chemistry or catalysis has been little explored.4 The most
successful application, Rh-catalyzed asymmetric hydrogena-
tion using tris(phospholane) 1 (Chart 2), was slow at room
temperature; Burk proposed that dissociation of one arm
of the triphosphine or of a chelating olefin substrate was
required to generate a vacant coordination site.2 Perhaps because
of this observation, such ligands have not been investigated
since the initial publications more than 20 years ago. Similarly,
P-stereogenic Siliphos (2; Chart 2)5 was used with limited success
Cooling hot methanol or isopropyl alcohol solutions of 3
gave large white crystals; repeated recrystallization resulted
in increased enrichment of C3-3. Scaling up the synthesis
gave over 2 g of highly diastereomerically and enantiomerically
Received: October 4, 2011
Published: January 13, 2012
© 2012 American Chemical Society
775
dx.doi.org/10.1021/om200930k | Organometallics 2012, 31, 775−778

Organometallics
Communication
Scheme 1. Synthesis and Deprotection of Triphosphine−
Borane 3
product ratios are dr = 3.5, er(C3) = 1521, and er(C1) = 11.5.
Instead, we observed a lower dr value (2.6). None of the minor
enantiomer C3′ was observed (as expected), but neither was C1′.
This indicates nonstatistical selectivity (substrate control) in the
reaction of the chiral nucleophile with the chiral electrophiles
MeSi((CH2PMe(t-Bu)(BH3))nCl3−n (n = 1, 2).10,11
Treatment of [Rh(NBD)Cl]2 with AgOTf and MT-Siliphos
gave the cation [Rh(MT-Siliphos)(NBD)][OTf] (5; NBD =
norbornadiene, Scheme 2).14 The structure of 5 (Figure 2),
Scheme 2. Synthesis of Rhodium and Ruthenium MT-
Siliphos Complexes
enriched C3-3 from a single preparation.10 This was only 15%
yield, but the short, scalable synthesis (two one-pot procedures
from commercially available t-BuPCl2),9 ease of handling
crystalline C3-3, and possible recovery of (−)-sparteine10
offer practical advantages in comparison to other syntheses of
P-stereogenic phosphines.1 The crystal structure of C3-3 (Figure 1)
Figure 1. ORTEP diagram of (R)-C3-MeSi(CH2PMe(t-Bu)(BH3))3
(C3-3).
showed it contained RP stereocenters, as expected from the
known stereoselectivity of (−)-sparteine-mediated asymmetric
deprotonation.8 Heating C3-3 in morpholine gave MT-Siliphos
(4; Scheme 1) without P-epimerization.5
Attachment of three enantiomerically enriched units A to an
achiral linker, as in Scheme 1, yields a mixture of C3- and C1-
symmetric products.11 Under statistical control, this “Horeau
amplification” gives the product ratio C3:C3′:C1:C1′ = x3:[(1 −
x)3]:[3x2(1 − x)]:[3x(1 − x)2], where x is the mole fraction of
the major enantiomer of A.12 With x = 0.92, as observed in
(−)-sparteine-mediated asymmetric deprotonation of PMe2-
(t-Bu)(BH3) with quenching by SiMe2PhCl,13 the predicted
Figure 2. ORTEP diagram of [Rh(MT-Siliphos)(norbornadiene)]-
[OTf] (5), showing one of the two independent molecules, with the
anion omitted. Selected bond lengths (Å) and angles (deg): Rh(1)−
P(1) = 2.3503(5), Rh(1)−P(2) = 2.4265(6), Rh(1)−P(3) =
2.4391(5), Rh(1)−C(21) = 2.128(2), Rh(1)−C(20) = 2.161(2),
Rh(1)−C(23) = 2.254(2), Rh(1)−C(24) = 2.300(3); P(1)−Rh(1)−
P(2) = 91.305(19), P(1)−Rh(1)−P(3) = 92.272(19), P(2)−Rh(1)−
P(3) = 93.754(19).
with P−Rh−P angles ranging from 91.3 to 93.8°, was similar to
that of [Rh(Siliphos)(NBD)][OTf]; the switch from P−Ph to
P−Me substituents had little structural effect.5
776
dx.doi.org/10.1021/om200930k | Organometallics 2012, 31, 775−778

Organometallics
Communication
Figure 3. ORTEP diagram of one of the four independent cations of [Ru(MT-Siliphos)(NCMe)3][OTf]2 (7), along with a view down the C3 axis
(compare Chart 1). Selected bond lengths (Å) and angles (deg): Ru(1)−P(1) = 2.3476(18), Ru(1)−N(1) = 2.087(6); P(1)−Ru(1)−P(1A) =
92.24(7), N(1)−Ru(1)−N(1A) = 83.9(2), N(1)−Ru(1)−P(1) = 95.90(16), N(1A)−Ru(1)−P(1) = 171.85(16), N(1B)−Ru(1)−P(1) = 87.93(16).
Heating a toluene slurry of Ru(DMSO)4Cl2 with MT-
Siliphos gave Ru(MT-Siliphos)(DMSO)Cl2 (6; Scheme 2).15
At room temperature, its 31P NMR spectrum contained a single
Scheme 3. Synthesis of Four-Coordinate d10 MT-Siliphos
Complexes
1
broad peak (δ 34.7, CD2Cl2), while the H and 13C NMR
spectra showed symmetrical Siliphos signals. At −70 °C, the
Siliphos arms were inequivalent (31P{1H} NMR (CD2Cl2):
three apparent triplets (δ 40.3, 31.3, 28.7, J = 34 Hz)). These
observations are consistent with reversible formation, via ligand
dissociation, of a fluxional five-coordinate intermediate, whose
pseudorotation makes the three ligand arms equivalent on
the NMR time scale at room temperature. Although DMSO
exchange was proposed to explain similar behavior in
Ru(triars)(DMSO)Cl2 (triars = MeC(CH2AsPh2)3),15a room-
temperature 1H and 13C NMR spectra showed that the DMSO
methyl groups in 6 were inequivalent, ruling out this process
and instead suggesting that the fluxionality arises from
reversible dissociation of a Siliphos arm.14b
Complex 6 could not be separated from an impurity which
gave rise to a sharp 31P NMR peak, suspected to be [Ru2(MT-
Siliphos)2(μ-Cl)3][Cl]; a related cation was the sole product
on treatment of Ru(DMSO)4Cl2 with triphos (MeC-
(CH2PPh2)3).15a The reaction of this mixture with 2 equiv of
AgOTf in acetonitrile gave the C3-symmetric dication [Ru(MT-
Siliphos)(NCMe)3][OTf]2 (7; Scheme 2).15 The crystal
structure (Figure 3) showed that 7 was a distorted octahedron,
with a Siliphos bite angle of 92.24(7)° and N−Ru−N angles
(83.9(2)°) less than the idealized 90°. Washing crude 7 to
remove excess acetonitrile, or recrystallization, resulted in forma-
tion of a small amount of [Ru(MT-Siliphos)(NCMe)2(OTf)]-
[OTf] (7′), whose 31P NMR spectrum (CD2Cl2) was similar to
that of 6 (three apparent triplets, δ 46.6, 27.4, 24.1 (J = 32 Hz)).
Treatment of 7 with LiOTf increased the amount of 7′, while
addition of MeCN to the mixture converted it all to 7.16
Treatment of the palladium and platinum complexes
M(PPh3)4 with MT-Siliphos generated M(MT-Siliphos)(PPh3)
(M = Pd (8), Pt (9); Scheme 3) and 3 equiv of PPh3.17 For
M = Pd, the 31P NMR spectrum of the reaction mixture showed
a peak due to coordinated Siliphos and a broad PPh3 signal;
on cooling to −70 °C, exchange between free and bound
PPh3 became slow on the NMR time scale and the expected
31P NMR doublet/quartet pattern for 8 was observed (JPP = 25 Hz).
PPh3 exchange in the analogous Pt complex 9 was slow on
the NMR time scale at room temperature.18 However, attempts
to recrystallize these mixtures led to decomposition, yielding
MT-Siliphos and other unidentified products.
Similar lability of the tridentate ligand in tetrahedral
complexes was observed with Cu(I) (Scheme 3). The reaction
of MT-Siliphos with CuCl gave adduct 10, whose room-
temperature NMR spectra were consistent with the expected C3
symmetry. However, the broad 31P NMR singlet for 10 became
several broad overlapping peaks at −70 °C, suggesting that
dynamic processes fast on the NMR time scale at room
temperature may account for the high symmetry observed.
Attempted recrystallization of 10 gave, via formal loss of MT-
Siliphos, polymeric 11, in which the triphosphine acts as a
bidentate ligand to a four-coordinate copper, which is
connected by unsymmetrical chloride bridges to a three-
coordinate copper ligated by the “third arm” of another
Siliphos, forming a polymeric chain. Figure 4 shows the
monomeric unit in the crystal structure of 11; the bite angle of
bidentate MT-Siliphos (107.27(3)°) was larger than that seen
for tridentate coordination in 5 and 7.19
777
dx.doi.org/10.1021/om200930k | Organometallics 2012, 31, 775−778

Organometallics
Communication
(4) (a) Baker, M. J.; Pringle, P. G. J. Chem. Soc., Chem. Commun.
1993, 314−316. (b) Castaldi, M. P.; Gibson, S. E.; Rudd, M.; White,
́
A. J. P. Angew. Chem., Int. Ed. 2005, 44, 3432−3435. (c) Brule, E.; Pei,
Y.; Lake, F.; Rahm, F.; Moberg, C. Mendeleev Commun. 2004, 276−
277. (d) Pei, Y.; Brule, E.; Moberg, C. Org. Biomol. Chem. 2006, 4,
́
544−550. (e) Pastor, S. D.; Shum, S. P. Tetrahedron: Asymmetry 1998,
9, 543−546. (f) Shum, S. P.; Pastor, S. D.; DeBellis, A. D.; Odorisio, P.
A.; Burke, L.; Clarke, F. H.; Rihs, G.; Piatek, B.; Rodebaugh, R. K.
Inorg. Chem. 2003, 42, 5097−5106.
(5) Ward, T. R.; Venanzi, L. M.; Albinati, A.; Lianza, F.; Gerfin, T.;
Gramlich, V.; Ramos Tombo, G. M. Helv. Chim. Acta 1991, 74,
983−988.
(6) (a) Ward, T. R. Ph.D. thesis, ETH-Zurich, 1991. (b) Ott, J.;
Schmid, B.; Venanzi, L. M.; Wang, G.; Ward, T. R.; Ramos Tombo,
G. M. New J. Chem. 1990, 14, 495−498. (c) Ott, J.; Ramos Tombo,
G. M.; Schmid, B.; Venanzi, L. M.; Wang, G.; Ward, T. R. Tetrahedron
Lett. 1989, 30, 6151−6154.
(7) Related P-stereogenic triphosphines MeC(CH2PAr(R))3 have
been prepared, without enantiocontrol, as mixtures of C1- and C3-
symmetric diastereomers: (a) Walter, O.; Klein, T.; Huttner, G.;
Zsolnai, L. J. Organomet. Chem. 1993, 458, 63−81. (b) Buchner, M.;
Huttner, G.; Winterhalter, U.; Frick, A. Chem. Ber./Recl. 1997, 130,
1379−1392.
(8) (a) Muci, A. R.; Campos, K. R.; Evans, D. A. J. Am. Chem. Soc.
1995, 117, 9075−9076. (b) Imamoto, T.; Watanabe, J.; Wada, Y.;
Masuda, H.; Yamada, H.; Tsuruta, H.; Matsukawa, S.; Yamaguchi, K.
J. Am. Chem. Soc. 1998, 120, 1635−1636.
(9) Crepy, K. V. L; Imamoto, T. Org. Synth. 2005, 82, 22−27.
(10) See the Supporting Information for details of the synthesis and
NMR methods for estimation of the diastereomeric ratio and
enantiomeric ratio (dr >100:1; er >100:1) of C3-3. This is consistent
with the predicted high er of C3-3.11
(11) (a) Glueck, D. S. Catal. Sci. Technol. 2011, 1, 1099−1108.
(b) Dornan, P. K.; Leung, P. L.; Dong, V. M. Tetrahedron 2011, 67,
4378−4384. (c) Vigneron, J. P.; Dhaenens, M.; Horeau, A.
Tetrahedron 1973, 29, 1055−1059.
(12) Grassi, M.; Di Silvestro, G.; Farina, M. Tetrahedron 1985, 41,
177−181.
(13) Gammon, J. J.; Canipa, S. J.; O’Brien, P.; Kelly, B.; Taylor, S.
Chem. Commun. 2008, 3750−3752.
Figure 4. ORTEP diagram of polymeric [Cu2Cl2(μ-(κ2,κ1)-MT-
Siliphos)·CH2Cl2]∞ (11·CH2Cl2), showing only the asymmetric unit,
with the solvent molecule omitted.
In summary, we have reported a straightforward asymmetric
synthesis of a novel C3-symmetric P-stereogenic triphosphine,
which has enabled investigation of the coordination chemistry
of this rare class of ligands. MT-Siliphos smoothly formed
five- and six-coordinate metal complexes but displayed remark-
able lability in four-coordinate pseudotetrahedral d10 com-
plexes, perhaps because of strain resulting from its preferred
bite angle of about 90°. Applications of such ligands in asym-
metric catalysis are now under investigation.
ASSOCIATED CONTENT
■
S
* Supporting Information
Text, tables, figures, and CIF files giving additional experimental
and crystallographic details and NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
(14) The apparent C3 symmetry observed in the NMR spectra of 5
presumably results from pseudorotation, as reported previously for
related five-coordinate triphos complexes: (a) Frazier, J. F.; Merola,
J. S. Polyhedron 1992, 11, 2917−2927. (b) Thaler, E. G.; Folting, K.;
Caulton, K. G. J. Am. Chem. Soc. 1990, 112, 2664−2672. (c) Ott, J.;
Venanzi, L. M.; Ghilardi, C. A.; Midollini, S.; Orlandini, A.
J. Organomet. Chem. 1985, 291, 89−100. (d) Dahlenburg, L.;
Mirzaei, F. Inorg. Chim. Acta 1985, 97, L1−L3.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Glueck@Dartmouth.Edu.
ACKNOWLEDGMENTS
■
We thank the Department of Education for a GAANN
fellowship for M.F.C. and the National Science Foundation
for support.
(15) For spectroscopic assessment of DMSO coordination mode in
related complexes, see: (a) Rhodes, L. F.; Sorato, C.; Venanzi, L. M.;
Bachechi, F. Inorg. Chem. 1988, 27, 604−610. (b) Iengo, E.; Zangrando, E.;
Baiutti, E.; Munini, F.; Alessio, E. Eur. J. Inorg. Chem. 2005, 1019−1031.
(c) Bratsos, I.; Alessio, E. Inorg. Synth. 2010, 35, 148−152. The DMSO
1H and 13C NMR chemical shifts of 6 (δ 2.61 and 2.57; δ 40.7 and 40.4,
toluene-d8, 21 °C) were consistent with Ru−O coordination. Its IR
REFERENCES
■
(1) (a) Grabulosa, A. P-Stereogenic Ligands in Enantioselective
Catalysis; RSC: Cambridge, U.K., 2011. (b) Phosphorus Ligands in
Asymmetric Catalysis. Synthesis and Applications; Borner, A., Ed.; Wiley-
−1
spectrum (1095, 1010, 881, 759 cm ) was ambiguous, showing peaks in
̈
−1
regions characteristic of Ru−O (∼900 cm ) and Ru−S coordination
VCH: Weinheim, Germany, 2008.
−1
(∼1100 cm ), respectively.
(2) (a) Burk, M. J.; Harlow, R. L. Angew. Chem., Int. Ed. 1990, 29,
1462−1464. (b) Burk, M. J.; Feaster, J. E.; Harlow, R. L. Tetrahedron:
Asymmetry 1991, 2, 569−592. (c) Fillol, J. L.; Kruckenberg, A.; Scherl,
P.; Wadepohl, H.; Gade, L. H. Chem. Eur. J. 2011, 17, 14047−14062.
(3) (a) Moberg, C. Angew. Chem., Int. Ed. 1988, 37, 248−268.
(b) Keyes, M. C.; Tolman, W. B. Adv. Catal. Proc. 1997, 2, 189−219.
(c) Gibson, S. E.; Castaldi, M. P. Chem. Commun. 2006, 3045−3062.
(d) Gibson, S. E.; Castaldi, M. P. Angew. Chem., Int. Ed. 2006, 45,
4718−4720. (e) Gade, L. H.; Bellemin-Laponnaz, S. Chem. Eur. J.
2008, 14, 4142−4152. (f) Gade, L. H.; Marconi, G.; Dro, C.; Ward, B.
D.; Poyatos, M.; Bellemin-Laponnaz, S.; Wadepohl, H.; Sorace, L.;
Poneti, G. Chem. Eur. J. 2007, 13, 3058−3075.
(16) A similar process may account for the observation that
1
[Ru(triphos)(NCMe)3][OTf]3 gave H NMR spectra consistent with
a C3-symmetric octahedral complex only if trace acetonitrile was
present.15a
(17) David, M.-A.; Wicht, D. K.; Glueck, D. S.; Yap, G. P. A; Liable-
Sands, L. M.; Rheingold, A. L. Organometallics 1997, 16, 4768−4770.
(18) Chatt, J.; Mason, R.; Meek, D. W. J. Am. Chem. Soc. 1975, 97,
3826−3827.
(19) Whitesides, G. M.; Casey, C. P.; Krieger, J. K. J. Am. Chem. Soc.
1971, 93, 1379−1389.
778
dx.doi.org/10.1021/om200930k | Organometallics 2012, 31, 775−778