Diastereomeric control of enantioselectivity: Evidence for metal cluster catalysis

Article abstract of DOI:10.1039/c4cc02319f

Enantioselective hydrogenation of tiglic acid effected by diastereomers of the general formula [(μ-H)₂Ru₃(μ₃-S)(CO) ₇(μ-P-P*)] (P-P* = chiral Walphos diphosphine ligand) strongly supports catalysis by intact Ru<

Full text of DOI:10.1039/c4cc02319f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Diastereomeric control of enantioselectivity:
evidence for metal cluster catalysis†
Cite this: Chem. Commun., 2014,
50, 7705
Ahmed F. Abdel-Magied,a Amrendra K. Singh,*a Matti Haukka,b
Michael G. Richmondc and Ebbe Nordlander*a
Received 28th March 2014,
Accepted 21st May 2014
DOI: 10.1039/c4cc02319f
www.rsc.org/chemcomm
Enantioselective hydrogenation of tiglic acid effected by diastereomers asymmetric induction in a catalytic system is observed, and if the
of the general formula [(l-H)2Ru3(l3-S)(CO)7(l-P–P*)] (P–P* = chiral enantioselectivity of the catalytic reaction is reversed by using
Walphos diphosphine ligand) strongly supports catalysis by intact another diastereomer of the cluster, where the chirality of the
Ru3 clusters. A catalytic mechanism involving Ru3 clusters has been ligand remains the same but the chirality of the cluster framework
established by DFT calculations.
has been changed, then this would constitute prima facie evidence
for cluster-promoted asymmetric induction. Here we wish to present
A number of homogeneous catalytic systems that are based on such a catalytic system that is based on two diastereomeric cluster
dimetallic complexes and/or clusters have been developed. complexes, where an unprecedented reversal of the enantioselectivity
In several of these systems, the polynuclear complexes have of a catalytic asymmetric reaction occurs upon reversal of the
been implicated as the active catalysts, or direct precursors chirality of the cluster framework.
to polynuclear catalysts.1 It is however difficult to clearly
In previous studies, we have obtained excellent conversion and
identify clusters as homogeneous catalysts.2 Duckett, Dyson good enantioselectivity in asymmetric hydrogenation of a-unsatu-
and coworkers3 have demonstrated that the clusters rated carboxylic acids, effected by catalytic systems based on
[H2Ru3(CO)10(L)2] (L = phosphine) function as homogeneous [(m-H)4Ru4(CO)10(L)] clusters (L = bulky chiral ferrocene-based diphos-
catalysts for the hydrogenation of alkynes (diphenylacetylene), phine).6 In continuation of this research, we have focused our
although a competing catalytic pathway involving fragmenta- attention on triruthenium hydrido clusters containing triply bridging
tion to form the mononuclear species [Ru(H)2(CO)2(L)(alkyne)] chalcogenide ligands that may function as ‘‘clamps’’ to maintain an
as active catalysts is also observed. Norton4 has defined an intact cluster framework. The trinuclear clusters [(m-H)2Ru3(m3-E)-
irrefutable criterion that identifies a cluster species as an active (CO)9ꢀ2x(dppm)x] (E = O, x = 2; E = S, x = 1,2) have been shown to
catalyst, viz. the observation of asymmetric induction in a be efficient precursors for catalytic olefin hydrogenation reactions.7,8
reaction that is catalysed by a cluster that is chiral by virtue of Reaction of [(m-H)2Ru3(m3-S)(CO)9] with the Walphos diphosphine
the cluster framework only (i.e. excluding chiral ligands). This R,R-1 in the presence of Me3NO results in the formation of clusters
concept has been probed in the silylation of acetophenone using 2 and 3 with the common empirical formula [(m-H)2Ru3(m3-S)(CO)7-
chiral tetrahedrane clusters, but cluster racemization was found (m-1)] (Scheme 1). The bridging of adjacent metals by the hetero-
to occur faster than productive catalysis.5 The generation of a bidentate ligand leads to an intrinsically chiral metallic framework
chiral cluster framework through the coordination of a suitable unless there is complete planarity,9 and the coordination of the
chiral ligand allows the above criterion to be modified such that if enantiomerically pure diphosphine ligand 1 thus yields a mixture of
diastereomers 2 and 3 due to different connectivities of the hetero-
bidentate ligand (cf. Scheme 1). Clusters 2 and 3 could be isolated
a Inorganic Chemistry Research Group, Chemical Physics, Center for Chemistry and
and were characterized by comparing their IR and 1H/31P NMR
Chemical Engineering, Lund University, Box 124, Lund, SE-221 00, Sweden.
spectral data with those of [(m-H)2Ru3(m3-S)(CO)7(dppm)]7 (cf. ESI†).
E-mail: Ebbe.Nordlander@chemphys.lu.se; Fax: +46 46 222 4439
The 31P NMR data confirmed the difference in the bridging mode of
b
¨
¨
¨
¨
Department of Chemistry, University of Jyvaskyla, Box 35, FI-40014 Jyvaskyla,
Finland
the diphosphine ligand in 2 and 3. For 2, the P1 signal (cf. Scheme 1)
appears as a triplet, indicating coupling to both hydrides, while the
signal for P2 is a doublet. In contrast, the signal for P1 in 3 appears as
a doublet while that of P2 is a doublet of doublets. The identities of
the two diastereomers were further confirmed by the determination
of their crystal structures (Fig. 1 and 2).
c Department of Chemistry, University of North Texas, Denton, TX 76203, USA
† Electronic supplementary information (ESI) available: Experimental details,
crystallographic details, FTIR and NMR spectra. Complete ref. 14, computational
details, and atomic coordinates of optimized ground-state and transition-state
structures. CCDC 993899 (2) and 993900 (3). For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c4cc02319f
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7705--7708 | 7705

View Article Online
Communication
ChemComm
Scheme 1 The synthesis of clusters 2 and 3.
Table 1 Summary of catalytic asymmetric hydrogenation reactions
Ligand
Catalyst
Conversiona (%)
ee (%)
Configurationb
R,R-1
R,R-1
S,S-1
S,S-1
2
3
4
5
49
79
55
64
23
56
24
52
R
S
S
R
a
The amount of substrate consumed in the catalytic experiments,
assessed by 1H NMR spectroscopy. Favoured enantiomer.
b
significantly higher enantioselectivities in other cluster-based
systems.6b The reversal in enantioselectivity in experiments where
2 and 3 were used as catalysts indicates that it is indeed the
clusters that are the active catalysts, or closely related triruthenium
species.11 Further, a mechanism involving cluster decomposition
during catalysis and regeneration of intact cluster at the end of
the catalytic cycle would be accompanied by formation of both
diastereomers,12 even if only one form of the cluster was used
initially. However, no trace of the other diastereomer was observed
at the end of catalytic cycles employing 2 or 3.
Fig. 1 Molecular structure of [(m-H)2Ru3(m3-S)(CO)7(m-1)] 2 with thermal
ellipsoids drawn at the 50% probability level. C–H hydrogen atoms have
been omitted for clarity. Selected bond distances [Å] and angles [1]: Ru1–Ru
2 2.8951(7), Ru1–Ru3 2.9023(6), Ru2–Ru3 2.7536(6), Ru1–P1 2.359(2),
Ru2–P2 2.333(1), Ru1–S1 2.368(2), Ru2–S1 2.375(2), Ru3–S1 2.346(2),
Ru1–H1 1.75(5), Ru1–H2 1.69(5), Ru2–H2 1.76(4), Ru3–H1 1.76(5).
In previous studies, we have shown that strong chiral induction
by chiral phosphines can be achieved in cluster-based hydrogena-
tions6b and, as a consequence, that reversal of enantioselectivity
can be effected by reversal of phosphine chirality.13 In order to
further investigate the reversal in enantioselectivity observed for
the sulfide-capped clusters 2 and 3, the analogous diastereomeric
pair based on S,S-1, viz. 4 (with diphosphine connectivity corres-
ponding to 2) and 5 (connectivity corresponding to 3) were
prepared, and characterized by comparing spectroscopic data with
those of 2 and 3 (cf. ESI†). Again, a reversal of enantioselectivity
with reversal in diphosphine connectivity was observed when 4
and 5 were used as hydrogenation catalysts (Table 1). Furthermore,
the chiral induction by the diphosphine ligand is confirmed, as
the enantioselectivity is reversed with reversal in diphosphine
chirality (2 vs. 4, and 3 vs. 5).
Fig. 2 Molecular structure of [(m-H)2Ru3(m3-S)(CO)7(m-1)] 3 with thermal
ellipsoids drawn at the 50% probability level. C–H hydrogen atoms have
been omitted for clarity. Selected bond distances [Å] and angles [1]: Ru1–Ru2
2.915(2), Ru1–Ru3 2.894(2), Ru2–Ru3 2.746(2), Ru1–P2 2.335(4), Ru2–P1
2.348(5), Ru1–S1 2.361(5), Ru2–S1 2.358(5), Ru3–S1 2.351(5). Ru1–H1 1.53,
Ru1–H2 1.57, Ru2–H2 1.90, Ru3–H1 1.67 (riding on Ru atoms).
From Table 1, it is evident that in these catalytic systems the
The catalytic activities of 2 and 3 were examined in asymmetric stereochemistry, as well as the bridging mode of the coordi-
hydrogenation of tiglic acid.6,10 Spectroscopic measurements (IR, nated chiral ligand, strongly affects the outcome of hydrogena-
1H/31P NMR, cf. ESI†) and mass spectrometry showed no signs of tion in terms of both conversion and enantioselectivity. Use of
changes in the clusters after catalytic runs. The catalysis results are cluster 3 yields an impressive 56% ee for (S)-2-methylbutyric
summarized in Table 1. The enantioselectivities are low in com- acid. The mechanism responsible for the observed catalysis and
parison to mononuclear catalysts, but relatively high by the details concerning the enantioselectivity were computationally
standards of cluster-based systems. This reflects an inherent investigated by electronic structure calculations.14,15 Scheme 2
weakness in the chiral induction effected by clusters – the sub- shows the computed catalytic cycle for the hydrogenation of
strate is likely to bind at a metal site with minimum steric tiglic acid to (S)-2-methylbutyric acid employing cluster 3
hindrance (vide infra) and the chiral induction effected by bulky (species A), where catalysis is initiated by the site-selective loss
chiral ligands is thus diminished. However, we have observed of CO from the Ru(CO)3 center, as shown in Fig. 3.16
7706 | Chem. Commun., 2014, 50, 7705--7708
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
energy than the si-face route. The energetics computed for the
reaction profiles depicted in Fig. 4 are in concert with the
observed preference for the (S) enantiomer when cluster 3 is
employed as the catalyst precursor. The mechanistic steps
computed involve a fixed stereochemistry at the cluster, whose
chirality directly influences the asymmetric induction observed
in the hydrogenation product. The observed reversal in enantio-
selectivity with the different cluster diastereomers containing
either R,R-1 or S,S-1 strongly support the involvement of intact
Ru3 clusters as the active hydrogenation catalysts.
In conclusion, we have prepared four diastereomers of the
cluster [(m-H)2Ru3(m3-S)(CO)7(m-1)], containing chiral cluster
frameworks and chiral diphosphine ligands. All diastereomers
display different catalytic behaviour in the enantioselective
hydrogenation of tiglic acid. The enantioselectivity is not only
reversed with reversal in ligand chirality, but also with reversal in
cluster chirality. The latter observed reversal in enantioselectivity
with the different cluster diastereomers containing either R,R-1
or S,S-1 strongly supports the involvement of intact Ru3 clusters
as the active hydrogenation catalysts. While the (transient)
formation of mononuclear chiral catalysts may be envisaged,11
such reactivity can neither explain the observed enantioselectivi-
ties for the diastereomeric pairs 2 and 3, and 4 and 5, nor the
isolation of pure cluster diastereomers after catalysis. Applica-
tion of Occam’s razor implicates cluster catalysis.
Scheme 2 Catalytic cycle for the hydrogenation of tiglic acid to (S)-2-
methylbutyric acid by cluster 3.
Fig. 3 Site-selective loss of CO from the Ru(CO)3 center in cluster 3 (see text).
AFA thanks the EU Erasmus Mundus program for a pre-
doctoral fellowship. AKS thanks the Carl Trygger Foundation
for a postdoctoral fellowship. MGR thanks the Robert A. Welch
Foundation (grant B-1093) for financial support; NSF support of
the computational facilities at the University of North Texas
through grant CHE-0741936 is acknowledged. We thank Dr
David Hrovat and Prof. Xinzheng Yang for many helpful
ONIOM-based discussions.
Cluster B serves as the entry point into the catalytic cycle, and
the free energy profile associated with this reaction is depicted in
Fig. 4. Si-face coordination of tiglic acid to B affords the alkene-
substituted cluster D. Regiospecific insertion of the alkene into
the proximal bridging hydride occurs via transition structure
TSDE, whose energy is 29.6 kcal molꢀ1 uphill relative to B and C.
The alternative alkene insertion route, which involves hydride
transfer to the ester-substituted alkene carbon and the creation
of the (S) stereogenic center, lies 6.6 kcal molꢀ1 above TSDE. The
resulting agostic alkyl cluster E reacts with H2 to furnish the
transient trihydride cluster F. Reductive elimination in F gives
(S)-2-methylbutyric acid (G) and regenerates cluster B, complet-
Notes and references
1 (a) P. J. Dyson, Coord. Chem. Rev., 2004, 248, 2443; (b) R. D. Adams
and F. A. Cotton, Catalysis by Di- and Polynuclear Metal Cluster
Complexes, Wiley-VCH, Weinheim, 1998; (c) R. D. Adams,
T. S. Barnard, Z. Li, W. Wu and J. H. Yamamoto, J. Am. Chem.
Soc., 1994, 116, 9103; (d) R. D. Adams and T. S. Barnard, Organome-
tallics, 1998, 17, 2567.
ing the catalytic cycle with a net release of 13.2 kcal molꢀ1
.
Coordination of the re face of the prochiral substrate proceeds
through a series of identical steps, all of which lie higher in
2 (a) Y. Lin and R. G. Finke, Inorg. Chem., 1994, 33, 4891;
(b) C. M. Hagen, L. Vieille-Petit, G. Laurenczy, G. Su¨ss-Fink and
R. G. Finke, Organometallics, 2005, 24, 1819.
3 (a) D. Blazina, S. B. Duckett, P. J. Dyson and J. A. B. Lohman, Angew.
Chem., Int. Ed., 2001, 40, 3874; (b) D. Blazina, S. B. Duckett,
P. J. Dyson and J. A. B. Lohman, Chem. – Eur. J., 2003, 9, 1045.
4 J. R. Norton, in Fundamental Research in Homogeneous Catalysis, ed.
M. Tsutsui, Plenum Press, New York, 1977.
5 C. U. Pittman, Jr., M. G. Richmond, M. Absi-Halabi, H. Beurich,
F. Richter and H. Vahrenkamp, Angew. Chem., Int. Ed. Engl., 1982,
21, 786.
6 (a) V. Moberg, M. Haukka, I. O. Koshevoy, R. Ortiz and
E. Nordlander, Organometallics, 2007, 26, 4090; (b) V. Moberg,
¨
R. Duqueesne, S. Contaldi, O. Rohrs, J. Nachtigall, L. Damoense,
A. T. Hutton, M. Green, M. Monari, D. Santelia, M. Haukka and
E. Nordlander, Chem. – Eur. J., 2012, 18, 12458.
7 S. J. Ahmed, M. I. Hyder, S. Kabir, M. A. Miah, A. J. Deeming and
E. Nordlander, J. Organomet. Chem., 2006, 691, 309.
8 C. Bergounhou, P. Fompeyrine, G. Commenges and J. J. Bonnet,
J. Mol. Catal., 1988, 48, 285.
Fig. 4 Free energy profiles for the hydrogenation of tiglic acid catalysed
by cluster 3. Black and red profiles are for the si- and re-face alkene
coordination routes, respectively.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7705--7708 | 7707

View Article Online
Communication
ChemComm
9 S. P. Tunik, I. O. Koshevoy, A. J. Poe, D. H. Farrar, E. Nordlander, 13 P. Homanen, R. Persson, M. Haukka, T. A. Pakkanen and
M. Haukka and T. Pakkanen, J. Chem. Soc., Dalton Trans., 2003, 2457. E. Nordlander, Organometallics, 2000, 19, 5568.
10 E. Tyrell, M. W. H. Tsang, G. A. Skinner and J. Fawcett, Tetrahedron, 14 M. J. Frisch et al., Gaussian 09, revision B.01, Gaussian, Inc., Wallingford,
1996, 52, 9841.
11 Formation of
CT, 2010.
a
Noyori-type catalyst of the general formula 15 The DFT calculations were performed using a two-layered ONIOM
[Ru(1)(O2CR)2] is in principle possible but cannot explain a reversal of
enantioselectivity. Formation of a [Ru(1)2X2] (X = arbitrary monodentate
anion) complex is even less likely, but could in principle form L and
D-isomers; however, a racemic mixture would then expected to be
formed, and the reversal of enantioselectivity would not be effected.
12 This is especially true with cluster 3, whose DFT-computed DG1 lies
2.0 kcal molꢀ1 above that of cluster 2. Any fragmentation of 3 during
(B3LYP/genecp:PM6) approach, where the Walphos ligand, except
for the phosphorus and iron atoms, was confined to the lower
layer. The Ru and Fe atoms were treated using SDD effective core
potential (ecp) basis sets, and all of the other high level atoms
were described by a 6-31G(d0) basis set. Standard-state corrections
have been applied to all species to convert concentrations from
1 atm to 1 M.
catalysis, followed by cluster reassembly, would thermodynamically 16 The unsaturated cluster B and free CO lie 23.0 kcal molꢀ1 (DG)
afford 2.
above cluster 3.
7708 | Chem. Commun., 2014, 50, 7705--7708
This journal is ©The Royal Society of Chemistry 2014