Highly selective supramolecular catalyzed allylic alcohol isomerization

Article abstract of DOI:10.1021/ja068688o

A supramolecular tetrahedral assembly Na₁₂[Ga₄L₆] (L = 1,5-bis-catecholamide naphthalene) has been found to selectively encapsulate monocationic rhodium complexes of the appropriate size and shape. Encapsulation within the chiral environment of the host directly affects the symmetry of the rhodium guest and can be well characterized by NMR spectroscopy. The rhodium complexes were found to be catalytically active for the isomerization of allylic alcohols. Investigations into the catalytic activity of the encapsulated rhodium guests have shown that the constrained cavity of the host exerts a strong influence on the reactivity at the metal center. The supramolecular host prevents substrates of the wrong size and shape from entering the host cavity and reacting with the encapsulated metal center, while substrates of the correct dimensions are allowed ready access. These results suggest that the metal center remains in the active site of the host while reactants and products freely and rapidly access the host cavity. Copyright

Full text of DOI:10.1021/ja068688o

Published on Web 02/16/2007
Highly Selective Supramolecular Catalyzed Allylic Alcohol Isomerization
Dennis H. Leung, Robert G. Bergman,* and Kenneth N. Raymond*
Department of Chemistry, UniVersity of California, Berkeley, California 94720-1460
Received December 4, 2006; E-mail: rbergman@berkeley.edu; raymond@socrates.berkeley.edu
Enzymatic catalysis often relies upon the highly constrained
environment of substrates within the active site cavity to govern
selectivity and reactivity. Recent examples in organometallic
chemistry mimic this approach by using host-guest interactions.1
A supramolecular M4L6 tetrahedral assembly developed by Ray-
mond and co-workers2 provides a host that mediates a variety of
organic and organometallic transformations.3 We have previously
demonstrated the efficient encapsulation of Ir(III) guests by this
M4L6 assembly that are capable of stoichiometric C-H bond
Figure 1. (Left) Schematic of M4L6 assembly with one of the edge-
activation with high size and shape selectivity.3a,c We envisioned
spanning ligands displayed containing a cartoon guest. (Right) Crystal
structure of Fe4L6 assembly with a cartoon guest.
that a similar catalytic encapsulated metal complex system could
also be achieved. Herein we report a remarkable and selective
supramolecular encapsulation of discrete organometallic catalysts
that display catalytic activity controlled by the size and shape of
the host cavity.
12-
The M4L6
host is formed from the self-assembly of four
octahedral metal centers (M ) GaIII) that comprise the vertices and
six bis-catecholamide naphthalene ligands (L4-) that span the edges
of a tetrahedron (Figure 1). The host has been found to encapsulate
monocationic guests ranging from ammonium cations to cationic
organometallic species.2,3
1
Figure 2. (a) H NMR spectrum of Na11[1 ⊂ Ga4L6] displaying upfield
but equivalent PMe3 peaks. (b) 1H NMR spectrum of Na11[2 ⊂ Ga4L6]
displaying upfield but nonequivalent dmpe methyl and methylene peaks.
Due to the large number of known monocationic rhodium
catalysts, we decided to pursue the encapsulation of these species
as potentially reactive guests. Small bis-phosphine rhodium com-
plexes of the general formula [(P-P)Rh(diene)][BF4] (P-P ) 2
PR3 or 1,2-bis(dimethylphosphino)ethane (dmpe), diene ) 1,5-
cyclooctadiene (COD) or norbornadiene (NBD)) were prepared,
and their suitability as guests was investigated (Table 1). Upon
addition of the rhodium complexes to an aqueous solution of the
Na12[Ga4L6] host assembly, encapsulation was observed, yielding
discrete host-guest assemblies of the general form Na11[LnRh ⊂
Ga4L6]. There is a sharp cutoff in the size allowed for the
encapsulated guest. Rhodium complexes that are too sterically
demanding such as [(PEt3)2Rh(COD)]+ were not encapsulated.
The equilibrium constants of encapsulation for these guests were
determined by 1H NMR spectroscopy. The binding affinities are 1
to 2 orders of magnitude lower than those of the iridium complexes
studied previously.3c We attribute this to the poorer shape comple-
mentarity of the square planar geometry compared to the more
compact half-sandwich structure with the host cavity. This shape
selectivity may explain the difference in binding affinity between
different rhodium complexes: the COD-substituted complexes bind
more favorably than the NBD-substituted ones, while the similarly
shaped PMe3 and dmpe complexes have comparable binding
affinities.
Table 1. Encapsulation Affinities for Rhodium Complexes in D2O
as Determined by 1H NMR Spectroscopy
1
Guest
Encapsulation Guest Affinity (M-
)
(1) (PMe3)2Rh(COD)+
(2) (dmpe)Rh(COD)+
(3) (PMe3)2Rh(NBD)+
(4) (dmpe)Rh(NBD)+
(5) (PEt3)2Rh(COD)+
5.2 × 102
5.7 × 102
1.2 × 102
1.2 × 102
not encapsulated
1
(COD)]+ (2). In the H NMR spectrum of [1 ⊂ Ga4L6], the PMe3
ligands remain equivalent due to the retention of C2 symmetry
1
(Figure 2a). In the H NMR spectrum of [2 ⊂ Ga4L6], the dmpe
methyl groups are split into two resonances; while there is a lack
of mirror symmetry relating the methyl groups, C2 symmetry still
relates two of the four methyl groups (Figure 2b). This loss of
symmetry is consistent with free rotation of the encapsulated guest
within the chiral host cavity; restricted rotation would result in a
further loss of symmetry.
Cationic rhodium complexes catalyze a wide variety of chemical
transformations4 including the isomerization of allylic alcohols.5
This reaction serves as an ideal model reaction for the host-guest
assembly since it involves a single substrate reacting with the
encapsulated catalyst. In addition, many allylic alcohols are water
soluble, which should ensure homogeneous reactivity.
The NMR resonances for the guest are shifted upfield by 2-3
ppm due to shielding by the naphthalene walls of the host, which
is diagnostic of encapsulation. Additionally, the chirality of the host
cavity breaks the symmetry of the guest. Since the tetrahedral host
has T symmetry, no mirror planes exist within the cavity; however,
C2 axes remain intact. This is exemplified by the NMR spectra of
the encapsulated guests [(PMe3)2Rh(COD)]+ (1) and [(dmpe)Rh-
While rhodium complexes bearing large phosphine groups have
been used in these isomerizations,5 little work has been done on
complexes bearing small phosphine groups such as PMe3. Therefore,
the scope of the reaction of unencapsulated 1 with allylic alcohols
was first examined (Table 2). In order to generate the active catalyst
+
(PMe3)2Rh(OD2)2 (6),6 1 atm of H2 was added to hydrogenate
9
2746
J. AM. CHEM. SOC. 2007, 129, 2746-2747
10.1021/ja068688o CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Isomerization of Allylic Substrates by 6
the substrate to enter the host cavity through the channels in the
ligand walls, preventing it from interacting with the active catalyst.
This is an example of the high shape selectivity of the supramo-
lecular catalyst: the isosteric linear allyl methyl ether is isomerized
quantitatively (Table 3, entry 7).
When both allyl alcohol and crotyl alcohol were added to [6 ⊂
Ga4L6], selective isomerization of allyl alcohol to propionaldehyde
was observed. This suggests that, while the supramolecular host
prevents crotyl alcohol from entering the cavity and deactivating
the catalyst, allyl alcohol has ready access to the encapsulated
catalyst. In combination with the slow rate of catalyst dissociation,
these selectivity results indicate that the highly specific substrate
selectivities are enforced by encapsulation within the well-defined
cavity of the host.
In conclusion, a supramolecular Ga4L6 host has been shown to
encapsulate a variety of cationic rhodium catalysts. This is a
remarkable example of the supramolecular encapsulation of discrete
organometallic catalysts, in this case resulting in supramolecular
control over the catalytic isomerization of allylic alcohols to give
highly specific size and shape substrate selectivities. In contrast to
the behavior of the free rhodium catalyst, in situ selectivity between
allyl alcohol and crotyl alcohol resulted in no inhibition by crotyl
alcohol and complete isomerization of allyl alcohol. These results
demonstrate that a synthetic supramolecular outer-sphere environ-
ment can direct control of catalytic reactivity at a metal centers
much like enzymes do in nature.
a Determined by 1H NMR spectroscopy. b1:1 E:Z enol ether was obtained.
Table 3. Isomerization of Allylic Substrates by [6 ⊂ Ga4L6]
Acknowledgment. We thank Dr. Dorothea Fiedler for helpful
discussion and Drs. Ulla Anderson and Georg Seeber for mass
spectrometry assistance. This work was supported by the Director,
Office of Energy Research, Office of Basic Energy Sciences
Division, of the U.S. Department of Energy under contract DE-
AC02-05CH11231.
a Determined by 1H NMR spectroscopy. b 1:1 E:Z enol ether was
obtained.
the COD ligand. Upon addition of the allylic alcohol, complete
conversion to the corresponding isomerized product was observed
within 0.5 h at 25 °C (Table 1, entry 1). Isomerization of allylic
ethers to the corresponding enol ethers was also observed (Table
1, entries 7 and 8). The reaction does not tolerate terminal
substitution; no isomerization of crotyl alcohol was observed (Table
1, entry 4). Significantly, when both allyl alcohol and crotyl alcohol
were added to the active catalyst, no isomerization of either substrate
was observed, indicating that crotyl alcohol inhibits the catalyst.
Having established the catalytic behavior of 6, the encapsulated
catalyst was generated by addition of 1 atm H2 to an aqueous
solution of [1 ⊂ Ga4L6]. Under these conditions, a new species
was observable by NMR spectroscopy after several hours at room
temperature, which was identified as [6 ⊂ Ga4L6]. However, after
12 h at room temperature, the guest was no longer encapsulated
by the host, indicating that the equilibrium favors the unencapsulated
rhodium complex. This is not surprising, considering that 6 is highly
solvated. This result suggests that 6 is kinetically formed within
the host cavity and slowly dissociates irreversibly into the bulk
solution. Indeed, when 6 was independently prepared and added
to an aqueous solution of Na12[Ga4L6], no encapsulation was
observed. Thus, any reactivity with [6 ⊂ Ga4L6], including substrate
entrance and product release, must be rapid enough to occur prior
to guest dissociation.
Supporting Information Available: Experimental details and
characterization of key intermediates. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Ito, H.; Kusukawa, T.; Fujita, M. Chem. Lett. 2000, 598-599. (b)
Slagt, V. F.; Kramer, P. C. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H.
J Am. Chem. Soc. 2004, 126, 1526-1536. (c) Kuil, M.; Soltner, T.; van
Leeuwen, P. W. N. M.; Reek, J. N. H. J. Am. Chem. Soc. 2006, 128,
11344-11345. (d) Kleij, A. W.; Lutz, M.; Spek, A. L.; van Leeuwen, P.
W. N. M.; Reek, J. N. H. Chem. Commun. 2005, 29, 3661-3663. (e)
Morris, G. A.; Nguyen, S. T.; Hupp, J. T. J. Mol. Catal. A: Chem. 2001,
174, 15-20. (f) Lee, S. J.; Hu, A.; Lin, W. J. Am. Chem. Soc. 2002, 124,
12948-12949. (g) Gianneschi, N. C.; Masar, M. S., III; Mirkin, C. A.
Acc. Chem. Res. 2005, 38, 825-837. (h) For a recent review, see: Kleij,
A. W.; Reek, J. N. H. Chem. Eur. J. 2006, 12, 4218-4227 and references
therein.
(2) (a) Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N. Angew.
Chem., Int. Ed. 1998, 37, 1840-1843. (b) Caulder, D. L.; Bruckner, C.;
Powers, R. E.; Konig, S.; Parac, T. N.; Leary, J. A.; Raymond, K. N. J.
Am. Chem. Soc. 2001, 123, 8923-8938.
(3) (a) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. Acc.
Chem. Res. 2005, 38, 351-360. (b) Fiedler, D.; van Halbeek, H.; Bergman,
R. G.; Raymond, K. N. J. Am. Chem. Soc. 2006, 128, 10240-10252. (c)
Leung, D. H.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2006,
128, 9781-9797.
(4) Evans, P. A., Ed. Modern Rhodium-Catalyzed Organic Reactions; Wiley-
VCH: Weinheim, 2005.
In sharp contrast to the results with unencapsulated 6, selective
isomerization of specific allylic alcohols was observed when
substrates were added to [6 ⊂ Ga4L6] (Table 3). Only alcohols of
the correct size and shape are able to enter the host cavity and
react with the metal catalyst. For example, the small allyl alcohol
is isomerized within 0.5 h at room temperature (Table 3, entry 1),
whereas larger substrates are not isomerized. Substrates with methyl
branching do not react at all, in contrast to the results with 6 (Table
3, entries 2 and 3). We infer that branching inhibits the ability of
(5) (a) Tani, K. Pure Appl. Chem. 1985, 57, 1845-1854. (b) Uma, R.; Cre´visy,
C.; Gre´e, R. Chem. ReV. 2003, 103, 27-52.
(6) Rhodium bis-phosphine solvento species have been implicated in the
isomerization of allylic substrates. See: Bergens, S. H.; Bosnich, B. J.
Am. Chem. Soc. 1991, 113, 958-967. We thank a referee for the
suggestion that in this case a rhodium hydride species may also be the
active catalytic species for this transformation. The NMR resonance for
the hydride may be masked by H-D exchange with the bulk solvent.
Since 6 is highly reactive, we were unable to isolate and fully characterize
it to definitively distinguish these two possibilities.
JA068688O
9
J. AM. CHEM. SOC. VOL. 129, NO. 10, 2007 2747