Isomerization of terminal epoxides by a [Pd-H] catalyst: A combined experimental and theoretical mechanistic study

Article abstract of DOI:10.1021/ja400325w

An unusual palladium hydride complex has been shown to be a competent catalyst in the isomerization of a variety of terminal and internal epoxides. The reaction displayed broad scope and synthetic utility. Experimental and theoretical evidence are provided for an unprecedented hydride mechanism characterized by two distinct enantio-determining steps. These results hold promise for the development of an enantioselective variant of the reaction.

Full text of DOI:10.1021/ja400325w

Subscriber access provided by Univ. of Tennessee Libraries
Article
Isomerization of Terminal Epoxides by a [Pd–H] Catalyst: a
Combined Experimental and Theoretical Mechanistic Study.
Devendra J. Vyas, Evgeny Larionov, Celine Besnard, Laure Guénée, and Clément Mazet
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja400325w • Publication Date (Web): 25 Mar 2013
Downloaded from http://pubs.acs.org on April 3, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Isomerization of Terminal Epoxides by a [Pd–H] Catalyst: a
Combined Experimental and Theoretical Mechanistic Study.
Devendra J. Vyas,† Evgeny Larionov,† Céline Besnard,‡ Laure Guénée,‡ Clément Mazet*,†
†Department of Organic Chemistry University of Geneva, Quai Ernest Ansermet 30, 1211 Geneva 4, Switzerland
‡Laboratoire de Cristallographie University of Geneva Quai Ernest Ansermet 24, 1211 Geneva 4, Switzerland
Supporting Information
yields. In the case of 2,2-disubstituted epoxides, evidence is
provided for an unprecedented stereodivergent mechanism
characterized by two distinct enantio-determining steps
which highlights the potential of this method in asymmetric
catalysis.
ABSTRACT: An unusual palladium hydride complex has
been shown to be a competent catalyst in the isomerization
of a variety of terminal and internal epoxides. The reaction
displayed broad scope and synthetic utility. Experimental
and theoretical evidences are provided for an unprecedented
hydride mechanism characterized by two distinct enantio-
determining steps. These results hold promise for the devel-
opment of an enantioselective variant of the reaction.
Mechanistic studies on the isomerization of trisubstituted
epoxides are scarce and the rearrangement shown on Scheme
1 has been generally admitted rather than truly demonstrat-
ed. Alternatives have been proposed in rare occasions.9-11
Terminal epoxides tend to undergo additional rearrange-
ments and, aside from the desired aldehydes, allylic alcohols,
alkenes or ketones are often formed in substantial amounts
(Scheme 1, Eq. 2).3d
The isomerization of epoxides into the corresponding car-
bonyl derivatives is a reaction that bears substantial synthet-
ic potential.1,2 Starting from readily accessible substrates, it
provides a straightforward access to either ketones or alde-
hydes – arguably two of the most valuable and prevalent
functions in synthesis. Surprisingly, this seemingly simple
isomerization still belongs to the repertoire of promising
chemical reactions for which the potentialities have yet to be
fully revealed. Indeed, although many Lewis acids have been
employed to perform this transformation, the number of
catalytic rather than stoichiometric versions remains lim-
ited.3 More importantly, the outcome of the reaction strongly
depends on the substitution pattern of the starting epoxide
and the often modest regioselectivity of the transformation
usually delivers complex mixtures of products. From a mech-
anistic standpoint, the lack of predictability is usually at-
tributed to the competition between hydrogen migration and
alkyl migration upon opening of the epoxide after activation
by the Lewis acid (Scheme 1, Eq. 1). Examples of stereospecif-
ic isomerizations of epoxides are very rare.4 Although the
kinetic resolution of racemic 2,3-disubstituted and trisubsti-
tuted epoxides has been documented, the main focus was to
access either enantioenriched epoxides or enantioenriched
allylic alcohols.5 A variant of this transformation using ter-
minal epoxides yet remains to be developed. Importantly,
such a redox economical6 process would provide a direct
route to chiral aldehydes with a tertiary −stereocenter.
Strategies to access this family of stereo-labile compounds in
high enantiopurity are quite limited.7,8
Scheme 1. Lewis Acid Mediated or Catalyzed Rear-
rangement of Epoxides
Kulawiec and co-workers have reported that a catalytic
combination of Pd(OAc)2 and PPh3 efficiently isomerized a
few 2,3-disubstituted epoxides in tBuOH at elevated temper-
atures.11c Isomerization of 2-methyl-2-(2-naphthyl)-oxirane 1a
into 2-(naphthyl)-propanal 2a was the sole example involving
a terminal epoxide; a reaction for which a SN2-type mecha-
nism was proposed.11c,d It was also shown that 2a enolizes
under the protic reaction conditions employed.11b Despite
this significant shortcoming, we considered this study as a
potential entry for the development of an asymmetric isom-
erization of terminal epoxides. Although we were able to
reproduce the result reported by Kulawiec and co-workers,
more representative epoxides such as 1b were unreactive
Herein, we report on the preparation of a well-defined
Pd(II)-hydride which catalyzes the isomerization of a variety
of epoxides with perfect regioselectivity and in practical

under the same conditions (Table 1). Much to our delight,
the use of chelating ligand (dppp 1,3-bis-
plex [Pd2(dippp)3].13 Subsequent treatment with 1.35 equiva-
lents (per Pd) of trifluoromethanesulfonic acid in diethyl
ether at room temperature provided quantitatively the corre-
sponding dinuclear palladium hydride 3. The solution struc-
ture of 3 was established by spectroscopic and spectrometric
analyses and was found to be consistent with the result of an
X-ray diffraction study (Figure 1).14 In the solid state, each Pd
atom is chelated by one dippp ligand in a boat-like confor-
mation (P–Pd–P = 93.60° and 99.37°) and both are bridged by
the third dippp. The hydrides have been located cis to the
monodentate P-donor. Each Pd atom lies in a slightly dis-
torted square planar geometry and, while one [(dippp)PdH]
unit is nearly coplanar with the linear propane chain of the
bridging dippp ligand, the other is almost perpendicular.
a
=
(diphenylphosphino)-propane) restored the catalytic activity
and 2b was obtained quantitatively (entry 1–3). A rapid sur-
vey of the most common organic solvents revealed a dichot-
omous situation since only tBuOH proved viable whereas all
other options left the substrate unreacted (See Supporting
Information). Questioning whether the initial step of the
catalytic cycle proposed by Kulawiec and co-workers would
really consist in reduction of Pd(II) to Pd(0), we evaluated
various commercially available Pd(0) sources (entry 4–9).
Neither [Pd(PPh3)4], [Pd(dba)2] (dba = dibenzylideneace-
tone), [Pd(PtBu3)2], nor a mixture of [Pd(dba)2] and dppp,
proved competent in the isomerization of 1b. Under other-
wise identical conditions, a combination of [Pd(PtBu3)2] and
dppp delivered 2b in 38% conversion (entry 8). The more
sterically demanding and electron-rich dippp ligand (dippp =
1,3-bis(di-isopropylphosphino)propane) afforded 2b in 98%
conversion (entry 9).
We next investigated the aptitude of complex 3 to perform
the isomerization of 1b into 2b. Among the different solvents
surveyed (toluene, benzene, DMF, dioxane, acetone), THF
turned out to be the best option. With as low as 2 mol% in
[Pd], complete conversion was reached in THF at 85°C after
only 4 h and aldehyde 2b was isolated in 77% yield.
Table 1. Optimization Study
Scheme 2. Substrate Scope (Terminal Epoxides)
entrya epoxide
[Pd]
ligand
PPh3c
PPh3c
dppp
-
conv. (%)b
98d
1
1a
1b
1b
1b
1b
1b
1b
1b
1b
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(PPh3)4
Pd(dba)2
Pd(PtBu3)2
Pd(dba)2
Pd(PtBu3)2
Pd(PtBu3)2
2
3
4
5
6
7
8
-
98
-
-
-
-
-
dppp
dppp
dippp
-
38
9
98
a
b
c
Average of two experiments. Determined by GC. 15 mol
%. d Repeated according to ref. 11c.
Figure 1. Synthesis and X-ray analysis for complex 3.
Suspecting a [Pd–H] species might be responsible for the
catalytic activity in the isomerization of 1b to 2b, we at-
tempted to isolate a well-defined complex following a proto-
col similar to that developed by Bunel and co-workers.12 Re-
action of [Pd(PtBu3)2] with 1.8 equivalents of dippp in tolu-
ene at 70°C, afforded the known dinuclear palladium com-
[Pd–H]-catalyzed isomerization of epoxides using 3 (0.25
mmol scale, yields of isolated product). Conditions A: 1 mol%
in [Pd], 85°C, 4 h; Conditions B: 4 mol% in [Pd], 100°C, 24 h;
Conditions C: 6 mol% in [Pd], 140°C, 24 h. a 2 mol% in [Pd]. b
120°C, 48 h. c 10 mol% in [Pd] d4 mol% in [Pd], 24 h.

Using only 1 mol% in [Pd], aldehydes 2a and 2c were ob-
tained in good yield (54% and 67% respectively). Two other
sets of reaction conditions (B–C) were applied to a variety of
epoxides (Scheme 2). Substrates with primary (2d,e), sec-
ondary (2f,g) and tertiary (2h) alkyl substituents R2 were all
isomerized in good yield with consistently perfect regioselec-
tivity. A wide range of functional groups potentially sensitive
to Pd catalysis is also well-tolerated.15 In particular, syntheti-
cally useful halogenated products 2j-n were all isolated in
practical yields (52–76%). Similarly, substrates 1o-p with
para-methoxy and benzyloxy substituents were isomerized
into 2o-p in 78% and 70% yield respectively.16 Substrates
possessing two alkyl groups also underwent efficient isomer-
ization (2q in 64% yield and 2r in 61% yield). Epoxides bear-
ing ester (1s), cyano (1t), as well as silylated alkyne (1u) also
afforded the corresponding aldehydes. Although the yields
were sometimes modest, products that would result from
reduction by the Pd−H catalyst were not observed. Remarka-
bly, isomerization of epoxide 1v, derived from the pregne-
nolone scaffold, delivered aldehyde 2v in 51% as a 2:1 mixture
of C-20 diastereoisomers. Interestingly, the silyloxy substitu-
ent was well-tolerated and no isomerization of the internal
double bond was observed, underscoring further the func-
tional group tolerance of the [Pd−H] catalyst.
tored by GC. After 12 h, 2,3-disusbtituted epoxide 1w was
fully isomerized and phenylpropane-2-one 2w was isolated in
82% yield as the sole product of the reaction. Isomerization
of cis-1x and trans-1x gave in each case ketone 2x exclusively
(86%). Cyclic epoxides such as 1y and 1z also underwent
smooth isomerization into 2y and 2z. Of note, the presence
of the carbonyl functionality in 1z was well-tolerated.
Scheme 3. Substrate Scope (2,3- and Tri-substituted
Epoxides)
Figure 2. Mechanistic experiments. A: Isomerization of an
enantioenriched epoxide by 3. B and C: Racemization exper-
iments with and without 3. D: Labeling experiments.
In order to gain insight into the exact role of the dinuclear
palladium hydride precursor in the isomerization of terminal
epoxides, we performed a series of complementary experi-
ments. The use of 2,6-di-tert-butyl-4-methylpyridine does
not inhibit the reaction and rules out the decomposition of 3
into triflic acid (See Supporting Information). When the typ-
ical reaction using 1b was carried out in the presence of 10
mol% of TEMPO no product formation was observed (See
Supporting Information).17 Isomerization of (R)-1b (ee =
94%), delivered 2b quantitatively but in racemic form (Figure
2, A). Taken individually, this result suggests this isomeriza-
tion is a stereoconvergent process. Whereas heating a THF
solution of (R)-2b (ee = 88%) for 4 h did not lead to an ero-
sion of enantiopurity (Figure 2, B), partial racemization was
noticed in the presence of complex 3 (2 mol% in [Pd]) (Fig-
ure 2, C). These results indicate that racemization involves
direct participation of the active form of the palladium hydride
catalyst. Labeling experiments were also conducted using 1b-
d10 and various amounts of 3 (Figure 2, D). When 12 mol% in
[Pd–H] were employed, H-incorporation was observable only
at the benzylic and aldehydic positions (9% and 2% respec-
tively). At higher loadings in palladium (29 and 51 mol%),
incorporation was seen at the same positions. In all cases,
incorporation was not complete and signals of unreacted Pd–
H which integrated for the remainder of ‘hydrogen content’
were observed by 1H NMR at  = –6.9 ppm (1%, 10% and 29%
respectively).18 Collectively these data indicate that a [Pd–H]
species is involved in the isomerization of 2,2-disubstituted
epoxides and, as will be shown hereafter, entry into the cata-
lytic cycle likely proceeds via reversible dissociation of the
bridging dippp ligand in 3. H-incorporation in the aldehydic
position is consistent with involvement of the [Pd–H] in a
reversible last step (vide infra).
[Pd–H]-catalyzed isomerization of epoxides using 3 (0.25
mmol, yields of isolated product). Reaction performed on a
1.0 mmol scale.
a
Although our initial focus was the isomerization of termi-
nal epoxides, we also explored the ability of complex 3 to
isomerize internal epoxides (Scheme 3). Reactions were per-
formed in THF using 2 mol% in [Pd−H] at 85°C and moni-

In the perspective of developing an asymmetric version of
this [Pd–H]-catalyzed isomerization of 2,2-disubstituted
epoxides, we have optimized an in situ protocol that avoids
systematic isolation of well-defined [Pd–H] complexes when
precious chiral ligands are employed (Table 2).
48% ee, S = 4.1; 2f: 13% ee, S = 1.1). This indicates a kinetic
resolution (i.e. a stereodivergent process) rather than a dy-
namic kinetic resolution (i.e. a stereoconvergent process) has
taken place. Of important note, the discrepancy observed
between the selectivity factors calculated for (S)-1f and (S)-2f
is in line with the partial racemization observed in Experi-
ments B and C (Figure 2).
Table 2. Optimization of Reaction Conditions for In
Situ Catalysis
en- [Pd]
Ligand
(6 mol%)
Additive
(10 mol%)
conv.
(%)a
try
(4 mol %)
1
3
-
-
99b
2
[Pd2(dippp)3]
[Pd2(dippp)3]
[Pd2(dippp)3]
[Pd(PtBu3)2]
[Pd(PtBu3)2]
-
-
-
-
-
3
-
tBuOH
4
-
4-CN-C6H4OH 95
5
dippp
dippp
dippp
-
-
-
-
-
-
Figure 4. Proposed catalytic cycle.
6
7
tBuOH
On the basis of the experiments detailed on Figure 2 and 3,
a mechanistic proposal that accounts for the isomerization of
terminal epoxides using 3 is depicted on Figure 4. Initial
decoordination of the monodentate dippp ligand allows
binding of the epoxide and entry of monomeric [Pd–H] units
5 into the catalytic cycle. Opening of the epoxide generates
the most stable carbocationic intermediate (6), hydride
transfer (7) and a subsequent -hydride elimination of a
methylenic proton delivers the aldehyde and concomitantly
regenerates the [Pd–H] intermediate. This manifold is con-
sistent with the results of the labeling experiments (Figure 2,
C). The result of Experiment A supports formation of a sp2
hybridized carbocationic intermediate. We propose that the
formation of 6 is both the rate-determining step of the reac-
tion and the selectivity-determining step that accounts for
enantio-enrichment of the epoxide. The enantioselection
could occur either by selective binding of one of the two en-
antiomeric epoxides to the palladium center or by selective
opening of one of the two diastereomeric intermediates of
type 5. Remarkably, the subsequent hydride transfer (6 → 7)
is likely to be the enantio-determining step during product
formation. To the best of our knowledge, the observation of
distinct enantio-determining steps in a kinetic resolution has
been rarely documented.19 On the sole basis of our experi-
mental results, we initially hypothesized that reversible for-
mation of the carbocationic intermediate (2 → 8 → 7 → 6)
could explain partial racemization of the aldehyde.
4-CN-C6H4OH
4-CN-C6H4OH
8
9
10
[Pd(PtBu3)2]
[Pd(PtBu3)2]
[Pd(PtBu3)2]
dippp
dippp
4-CN-C6H4OH 98
4-CN-C6H4OH 98b
a Determined by GC. b Reaction performed in THF.
This approach was again inspired by the work of Bunel and
co-workers who showed that formation of [Pd−H] is directly
linked to the pKa of the protic additive employed.12 Although
it could be used as a solvent, tBuOH (pKa = 19) did not prove
viable as an additive. In contrast, 4-cyanophenol (pKa = 4.5)
revealed as the protic additive of choice. When [Pd2(dippp)3]
is used, 4-cyanophenol is the only necessary additive. In the
case of [Pd(PtBu3)2], both added dippp and 4-cyanophenol
were required. Notably, THF and toluene were equally com-
petent solvents for the reactions performed in situ.
To further substantiate our mechanistic model, we per-
formed a series of DFT calculations using dtbpp as a model
ligand (dtbpp = 1,3-bis(di-tert- butylphosphino)propane). We
first investigated the thermodynamics of the dissociation of
dinuclear palladium complex [(dtbpp)3(PdH)2]2+ 3b into
mononuclear species (Figure 5). Dissociation of
[(dtbpp)3(PdH)2]2+ 3b into two monomeric palladium hy-
dride complexes [(dtbpp)(PdH)]+ 3b-1 and [(dtbpp)2(PdH)]+
3b-2 was found to be slightly exergonic (-2.2 kcal/mol). Fur-
ther dissociation of one dtbpp ligand in 3b- 2 to generate the
coordinatively unsaturated palladium complex 3b-1 is for-
mally spontaneous (-11.4 kcal/mol).
Figure 3. Kinetic Resolution of 1f.
We next attempted to develop an enantioselective variant
of the isomerization of terminal epoxides using substrate 1f
and the in situ protocol ([Pd(PtBu3)2], 4-cyanophenol and
chiral ligand). A mixture of toluene and THF was employed
to ensure complete solubility of all reaction components.
Ligand 4 was initially selected for its steric and electronic
resemblance with dippp (Figure 3). A conversion of 18% was
reached after 4 h at 60°C. Recovered 1f was obtained in 13%
ee (S = 4.4) and aldehyde 2f in 10% ee (S = 1.3). Prolonged
heating (16 h, C = 51%) led to further racemization of 2f (1f:

Figure 6. Computed reaction profile using dtbpp (1,3-bis(di-tert-butylphosphino)propane) as a model ligand. The most relevant
intermediates and transition states are shown. DFT method: B3LYP/[LANL2DZ on Pd, 6-31G(d) on H, C, O, P], PCM in THF at
the same level. G values are given in kcal/mol.
a G value of 15.4 kcal/mol indicating such a mechanism is
less likely to occur.
Next, starting form mononuclear complex 3b-1 a reasona-
ble reaction profile has been calculated (Figure 6).20,21 Com-
plex 5b-1 lies 7.8 kcal/mol higher in energy than the coordi-
natively unsaturated complex (1b + 3b-1). Ring opening of
the epoxide followed by hydride transfer in overall activation
free energy of 26.4 kcal/mol leads to the pallada-alkoxy in-
termediate 7b and constitutes the rate-determining step of
the transformation. This value is in good agreement with the
experimental reaction conditions (T = 85°C for isomerization
of 1b). The sequence passes through the carbocationic inter-
mediate 6b which is consistent with the collective results of
experiments A–D (Figure 2) and the kinetic resolution de-
scribed on Figure 3. Aldehyde 2b is liberated after a -
hydride abstraction (facilitated by an agostic interaction in
7b) and subsequent decoordination in a slightly exergonic
process.
Unexpectedly, we found that epimerization of the alde-
hyde occurs by a formal 1,2-addition of the Pd–H into the
carbonyl (9b) followed by -hydride elimination of the ben-
Figure 5. Generation of mononuclear [Pd−H] units from
zylic proton to generate the enol-bound intermediate 10b
dimer 3b. DFT method: B3LYP/[LANL2DZ on Pd, 6-31G(d)
rather than by the reverse reaction from (2b + 3b-1) to 6b.
on H, C, O, P], PCM in THF at the same level.
This route is more favorable by 23.4 kcal/mol (47.7 kcal/mol
In contrast, coordination of one explicit molecule of THF or
one molecule of substrate 1b to 3b-1 were identified as en-
dergonic processes, with the latter being slightly disfavored
(3.6 vs. 7.8 kcal/mol respectively). These results are in line
with the incomplete dissociation of 3 in the labeling experi-
ments at high catalyst concentration (Figure 2, D). An hypo-
thetical associative mechanism involving [(dtbpp)2(PdH)]+
3b-2 and substrate 1b leading to intermediate 5b-2 displayed
vs 24.3 kcal/mol). This observation prompted us to consider
alternative tautomerization pathways that could account for
post-reaction racemization of the enantioenriched aldehyde
(Figure 7). The O-bound enol isomer 10b’ was found to be
almost iso-energetic to the C-bound enol 10b, and to further
evolve into dissociated enol 11b and the coordinatively un-
saturated palladium complex 3b-1 (G = -15.2 kcal/mol).

Supporting Information
Experimental procedures, spectral, crystallographic for com-
plex 3, computational data, and full reference 20. This mate-
rial is available free of charge via the internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
clement.mazet@unige.ch
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported by the University of Geneva and
Roche. We thank the SMS platform of the University of Ge-
neva for mass spectrometric measurements and Prof. Tomasz
A. Wesolowski (University of Geneva) for offering us access
to his computational facilities. Mr. Houhua Li is acknowl-
edged for technical assistance in the preparation of sub-
strates 1v and 1z. This article is dedicated to Prof. Lutz H.
Gade (University of Heidelberg) on the occasion of his 50th
birthday.
REFERENCES
(1) For reviews, see: (a) Jørgensen, K. A.; Schiøtt, B. Chem. Rev.
1990, 90, 1483–1506. (b) Silva Jr. L. F. Tetrahedron 2002, 58, 9137–9161.
(2) For applications in synthesis, see: (a) Bando, T.; Shishido, K.
Chem. Commun. 1996, 1357–1358. (b) Matsushita, M.; Maeda, H.;
Kodama, M. Tetrahedron Lett. 1998, 39, 3749–3752. (c) Kinura, T.;
Yamamoto, N.; Suzuki, Y.; Kawano, K.; Norimine, Y.; Ito, K.; Nagato,
S.;. Iimura, Y.; Yonaga, M. J. Org. Chem. 2002, 67, 6228–6231.
(3) For seminal contributions, see: (a) House, H. O. J. Am. Chem.
Soc. 1954, 77, 3070–3075. (b) Naqvi, S. M.; Horowitz, J. P.; Filler, R. J.
Am. Chem. Soc. 1957, 79, 6283–6286. (c) Meinwald, J.; Labana, S. S.;
Chadha, M. S. J. Am. Chem. Soc. 1962, 85, 582–585. For recent
examples, see: (d) Miyashita, A.; Shimada, T.; Sugawara, A.; Nohira,
H. Chem. Lett. 1986, 1323–1326. (e) Ranu, B. C.; Jana, U. J. Org. Chem.
1998, 63, 8212–8216. (f) Anderson, A. M.; Blazek, J. M.; Garg, P.;
Payne, B. J.; Mohan, R. S. Tetrahedron Lett. 2000, 41, 1527–1530. (g)
Martinez, F.; Del Campo, C.; Llama, E. F. J. Chem. Soc. Perkin Trans. 1
2000, 1749–1751. (h) Kamaré, I.; Tommasino, M. L.; Lemaire, M. Te-
trahedron Lett. 2003, 44, 7687–7689. (i) Procopio, A.; Dalpozzo, R.;
De Nino, A.; Nardi, M.; Sindona, G.; Tagarelli, A. Synlett 2004, 2633–
2635. (j) Robinson, M. W. C.; Pillinger, K. S.; Graham, A. E. Tetrahe-
dron Lett. 2006, 47, 5919–5921. (k) Ertürk, E.; Göllü, M.; Demir, A. S.
Tetrahedron 2010, 66, 2373–2377. (l) Robinson, M. W. C.; Pillinger, K.
S.; Mabbett, L.; Timms, D. A.; Graham, A. E. Tetrahedron 2010, 66,
8377–8382.
(4) (a) Maruoka, K.; Ooi, T.; Yamamoto, H. J. Am. Chem. Soc.
1989, 111, 6431–6432. (b) Suda, K.; Kikkawa, T.; Nakajima, S.-i.;
Takanami, T. J. Am. Chem. Soc. 2004, 126, 9554–9555. (c) Suda, K.;
Nakajima, S.-i.; Satoh, Y.; Takanami, T. Chem. Commun. 2009, 1255–
1257. (d) Hrdina, R.; Müller, C. E.; Wende, R. C.; Lippert, K. M.;
Benassi, M.; Spengler, B.; Schreiner, P. R. J. Am. Chem. Soc. 2011, 133,
7624–7627. For a related study on enol ester epoxides, see: (e) Feng,
X.; Shu, L.; Shi, Y. J. Org. Chem. 2002, 67, 2831–2836.
(5) (a) Naruse, Y.; Esaki, T.; Yamamoto, Y. Tetrahedron 1988, 44,
4747–4756. (b) Gayet, A.; Bertilsson, S.; Andersson, P. G. Org. Lett.
2002, 4, 3777–3779. (c) Sodergren, M. J.; Bertilsson, S. K. ;
Andersson, P. G. J. Am. Chem. Soc. 2000, 122, 6610-6618. For reviews
on meso epoxides, see: (a) Hodgson, D. M.; Gibbs, A. R.; Lee, G. P.
Tetrahedron 1996, 52, 14361−14384. (b) Asami, M. J. Synth. Org. Chem.
Jpn. 1996, 54, 188−199. (c) O'Brien, P. J. Chem. Soc., Perkin Trans. 1
1998, 1439−1457.
Figure 7. Competing tautomerization pathways. DFT meth-
od: B3LYP/[LANL2DZ on Pd, 6-31G(d) on H, C, O, P], PCM in
THF at the same level.G values are given in kcal/mol.
Metal-free tautomerization of 2b to 11b can occur either in a
unimolecular (grey) or a bimolecular (blue) process. The
latter was found to be more favorable and the corresponding
6-membered Zimmerman-Traxler transition state TS2b-11b lied
at 27.8 kcal/mol (G‡298 = 49.1 kcal/mol; 2b → 11b); far above
the [Pd−H]-catalyzed tautomerization route (in red on Fig-
ure 7). These values are in agreement with the results of ex-
periments B and C reported on Figure 2. Finally, tautomeri-
zation via an enolate Lewis acid-assisted pathway was found
to be unrealistic as the corresponding transition state TS8b-
10b’ lied at 47.6 kcal/mol (purple).
In conclusion, we have synthesized and characterized a
well-defined Pd–H dinuclear complex which proved compe-
tent in the selective isomerization of terminal epoxides into
aldehydes. Experimental and theoretical investigations point
to an unprecedented hydride-type mechanism. Preliminary
observations revealed two distinct enantio-determining steps
for the kinetic resolution of racemic epoxides. A pathway
that accounts for partial racemization of the enantio-
enriched aldehyde and that lies off the productive catalytic
cycle was also identified. Importantly, although far from
practicality, these results clearly set a precedent that will
serve as a blueprint for further developments of a more effi-
cient asymmetric isomerization of terminal epoxides. Future
work will be aimed at the identification of highly enantiose-
lective catalysts for this isomerization and for related trans-
formations.
ASSOCIATED CONTENT

(6) Burns, N. Z.; Baran, P. S.; Hoffmann, R. W. Angew. Chem. Int.
Ed. 2009, 48, 2854-2967.
exchange between the palladium hydride catalyst and deuterium
from substrate 1b-d10.
(7) (a) Enders, D.; Wortmann, L.; Peters, R. Acc. Chem. Res. 2000,
33, 157–169. (b) For reviews on asymmetric hydroformylation: (b)
Nozaki, K.; in Comprehensive Asymmetric Catalysis, Vol.1 (Eds:
Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H.), Springer: Berlin, 1999,
Chapter 11; (c) Claver, C.; van Leuween, P. N. M. W. in Rhodium-
catalyzed Hydroformylation (Eds: van Leuween, P. N. M. W. ; Claver,
C.), Kluwer Academic Publishers: Dodrecht, Boston, London, 2000,
Chapter 5; (d) Klosin, J.; Landis, C. R. Acc. Chem. Res. 2007, 40, 1251–
1259; (e) Guai, A.; Godard, C.; Castillón, S.; Claver, C. Tetrahedron:
Asym. 2010, 15, 1135–1146.
(19) (a) Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249–
331. (b) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. Adv. Synth. Catal.
2001, 1, 5–26. (c) Steinreiber, J.; Faber, K.; Griengl, H. Chem. Eur. J.
2008, 14, 8060–8072.
(20) Gaussian 09, Revision A.02, M. J. Frisch et al. Gaussian, Inc.,
Wallingford CT, 2009.
(21) (a) LANL2 ECP: Hay, P. J.; Wadt, W. R. J. Chem. Phys., 1985,
82, 270–283. (b) PCM: Tomasi, J.; Mennucci, B.; Cammi, R. Chem.
Rev., 2005, 105, 2999–3094.
(8) (a) Vignola, N.; List, B. J. Am. Chem. Soc. 2004, 126, 450–451.
(b) Ibrahem, I.; Córdova, A. Angew. Chem. Int. Ed. 2006, 45, 1952–
1956; (c) Beeson, T. D.; Mastraccio, A.; Hong, J. B.; Ashton, K.;
MacMillan, D. W. C. Science 2007, 316, 582–585. (d) Xie, H.; Zu, L.;
Li, H.; Wang, J.; Wang, W. J. Am. Chem. Soc. 2007, 129, 10886-10894.
(e) Rios, R.; Sundén, H.; Vesely, J.; Zhao, G.-L.; Eriksson, L.; Córdova,
A. Adv. Synth. Catal. 2007, 349, 1028-1032. (f) Rios, R.; Vesely, J.;
Sundén, H.; Ibrahem, I.; Zhao, G.-L.; Córdova, A. Tetrahedron Lett.
2007, 48, 5835-5839. (g) Hodgson, D. M.; Kaka, N. S. Angew. Chem.
Int. Ed. 2008, 47, 9958-9960. (h) Nicewicz, D. A.; MacMillan, D. W.
C. Science 2008, 322, 77-80. (i) Allen, A.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2011, 133, 4260–4263; (j) Skucas, E.; MacMillan, D.W. C. J.
Am. Chem. Soc. 2012, 134, 9090–9093. (k) Shaikh, R. R.; Mazzanti, A.;
Petrini, M.; Bartoli, G.; Melchiorre, P. Angew. Chem., Int. Ed. 2008,
47, 8707-8710.
(9) (a) Milstein, D.; Buchman, O.; Blum, J. J. Org. Chem. 1977, 42,
2299–2309. (b) Aye, K.-T.; Gelmini, L.; Payne, N. C.; Vittal, J. J.;
Puddephatt, R. J. J. Am. Chem. Soc. 1990, 112, 2464–2465. For
coupling reactions using epoxides as structural element, see: (c)
Hirai, A.; Yu, X.-Q.; Tonooka, T.; Miyashita, M. Chem. Commun.
2003, 2482–2483. (d) Beaver, M. G.; Jamison, T. F. Org. Lett. 2011, 13,
4140–4143. (e) Shaghafi, M. B.; Grote, R. E.; Jarvo, E. R. Org. Lett.
2011, 13, 5188–5191. (f) Nielsen, D. K.; Doyle, A. G. Angew. Chem. Int.
Ed. 2011, 50, 6056–6059.
(10) For recent studies on related aziridines, see: (a) Lin, B. L.;
Clough, C. R.; Hillhouse, G. L. J. Am. Chem. Soc. 2002, 124, 2890–
2891. (b) Wolfe, J. P.; Ney, J. E. Org. Lett. 2003, 5, 4607–4610. (c) Ney,
J. E.; Wolfe, J. P. J. Am. Chem. Soc. 2006, 128, 15415–15422. (d) Huang,
C.-Y; Doyle, A. G. J. Am. Chem. Soc. 2012, 134, 9541–9544.
(11) (a) Kulasegaram, S.; Kulawiec, R. J. J. Org. Chem. 1994, 59,
7195–7196. (b) Kim, J.-H.; Kulawiec, R. J. J. Org. Chem. 1996, 61, 7656–
7657. (c) Kulasegaram, S.; Kulawiec, R. J. J. Org. Chem. 1997, 62,
6547–6561. (d) Kulasegaram, S.; Kulawiec, R. J. Tetrahedron 1998, 54,
1361–1374.
(12) Perez, P. J.; Calabrese, J. C.; Bunel, E. E. Organometallics 2001,
20, 337–345.
(13) For an alternative synthesis, see: Portnoy, M.; Milstein, D.
Organometallics 1993, 12, 1655–1664.
(14) CCDC 905952 for complex 3 contains the supplementary
crystallographic data for this article. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre (CCDC)
at www.ccdc.cam.ac.uk/data_request/cif. The structure in Figures 1
and 6 were generated with iew ega t, . . iew, ersion
1. .561b ni ersit de herbroo e herbroo e, , 2009;
http://www.cylview.org.
(15) Reduction of the Pd(II) catalyst to a Pd(0) intermediate might
generate a competent catalyst for cross-coupling reactions.
(16) For examples of synthetic derivatizations of methoxy-
substituted aryl compounds, see: Li, B.J.; Yu, D.-G.; Sun, C.-L.; Shi,
Z.-J. Chem. Eur. J. 2011, 17, 1728–1759.
(17) TEMPO is a potent [Pd–H] inhibitor through reduction from
Pd(II) to Pd(I), see: (a) A b niz, A. . Espinet, P. όpez-Fernández,
R.; Sen, A. J. Am. Chem. Soc. 2002, 124, 11278–11279. (b) Tan, E. H. P.;
Lloyd-Jones, G. C.; Harvey, J. N.; Lennox, A. J. J.; Mills, B. M. Angew.
Chem. Int. Ed. 2011, 50, 9602–9606.
(18) Inspection of the 13C{1H} NMR spectrum did not indicate H
incorporation at other positions. Formation of the Pd–D complex
analogous to 3 was observed by 31P{1H} NMR and confirmed H/D