Investigations of the scope and mechanism of the tandem hydroesterification/lactonization reaction

Article abstract of DOI:10.1021/ol048378f

(Chemical Equation Presented) Heating allylic and homoallylic alcohols and 2-pyridylmethyl formate in the presence of Ru₃(CO)₁₂ initiates a tandem sequence of hydroesterification and lactonization. Mechanistic studies suggest that regioselectivity and overall reaction efficiency are governed by the relative rates of reductive elimination and β-hydride elimination for the alkylruthenium intermediates.

Full text of DOI:10.1021/ol048378f

ORGANIC
LETTERS
2004
Vol. 6, No. 23
4207-4210
Investigations of the Scope and
Mechanism of the Tandem
Hydroesterification/Lactonization
Reaction
Lijun Wang and Paul E. Floreancig*
Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
florean@pitt.edu
Received August 14, 2004
ABSTRACT
Heating allylic and homoallylic alcohols and 2-pyridylmethyl formate in the presence of Ru3(CO)12 initiates a tandem sequence of hydroesterification
and lactonization. Mechanistic studies suggest that regioselectivity and overall reaction efficiency are governed by the relative rates of reductive
elimination and
â-hydride elimination for the alkylruthenium intermediates.
Processes that couple alkene hydrometalation and reductive
elimination provide an attractive approach for homologating
and functionalizing olefins.1 Forming the requisite metal
hydrides through carbon-hydrogen bond activation2 is a
particularly desirable aspect of this process with respect to
operational simplicity and reagent accessibility. We became
interested in this reaction class through our application of
Chang’s ruthenium-catalyzed olefin hydroesterification reac-
tion3 in the context of synthetic efforts4 toward the naturally
occurring HIV integrase inhibitor integramycin.5 In this
process we converted an enantiomerically pure homoallylic
alcohol into a δ-lactone (Figure 1) in a single operation by
heating with Ru3(CO)12 in 2-pyridylmethyl formate. This
tandem hydroesterification/lactonization sequence is signifi-
cant in that it is a rare example of an intermolecular process
of this type proceeding with the alkene component as the
limiting reagent. An unexpected outcome in this study was
the isolation of a significant amount of the γ-lactone that
forms from the branched hydroesterification product. This
result contrasts with Chang’s demonstration that allylic
substitution results in excellent selectivity for linear hydro-
esterification products and indicates that the free hydroxyl
(1) For recent examples, see: (a) Willis, M. C.; McNally, S. J.; Beswick,
P. J. Angew. Chem., Int. Ed. 2004, 43, 340. (b) Jun, C.-H.; Moon, C. W.;
Lee, D.-Y. Chem. Eur. J. 2002, 8, 2423. (c) Wiedemann, S. H.; Bergman,
R. G.; Ellman, J. E. Org. Lett. 2004, 6, 1685. (d) DeBoef, B.; Pastine, S.
J.; Sames, D. J. Am. Chem. Soc. 2004, 126, 6556.
(2) (a) Ritleng, V.; Sirlin, C.; Pfeffer, M. Chem. ReV. 2002, 102, 1731.
(b) Labinger, J. A.; Bercaw, J. E. Nature 2002, 147, 507. (c) Dyker, G.
Angew. Chem., Int. Ed. 1999, 38, 1699.
(3) Ko, S.; Na, Y.; Chang, S. J. Am. Chem. Soc. 2002, 124, 750.
(4) Wang, L.; Floreancig, P. E. Org. Lett. 2004, 6, 569.
(5) Singh, S. B.; Zink, D. L.; Heimbach, B.; Genilloud, O.; Teran, A.;
Silverman, K. C.; Lingham, R. B.; Felock, P.; Hazuda, D. C. Org. Lett.
2002, 4, 1123.
Figure 1. Lactone formation through hydroesterification.
10.1021/ol048378f CCC: $27.50
© 2004 American Chemical Society
Published on Web 10/14/2004

Table 1. Hydroesterification of Functionalized Olefinsa
Figure 2. Promotion and suppression of branched product forma-
tion.
group plays a role in determining the regiochemical outcome
of the reaction. In this Letter we report our studies directed
toward determining the scope of this process and understand-
ing the function of the hydroxyl group through constructing
a structure-reactivity relationship and through the use of
deuterated pyridylmethyl formate as a mechanistic probe.
Our initial proposal for branched product formation
invoked hydroxyl coordination in the hydrometalation step
(Figure 2) to lower the barrier of formation for the secondary
alkylruthenium species 5. This mechanism presupposes that
product selectivity is determined by hydrometalation regio-
chemistry and is consistent with our observation that
branched product formation is eliminated by suppressing
hydroxyl group coordination through silyl ether formation.
Silyl-protected substrates can also engage in a one-pot
hydroesterification/lactonization reaction by adding HOAc
upon completion of ester formation.
Promoting heteroatom complexation requires an open
coordination site on ruthenium. Although the exact nature
of the catalyst in this process has yet to be determined, clearly
extensive ligand loss from the coordinatively saturated
Ru3(CO)12 must precede C-H insertion and olefin binding.
Inspired by Schreiber’s studies6 on the Pauson-Khand
reaction, we added N-methylmorpholine N-oxide (NMO) to
the reaction mixture to open coordination sites by sequester-
ing and oxidizing transiently dissociated CO. Adding 5-15
mol % NMO resulted in a rate enhancement at 135 °C and
allowed us to lower the reaction temperature from 135 to
95 °C, a temperature at which no reaction was observed
under the original protocol. In addition to providing less
thermally demanding reaction conditions, this change permit-
a General procedure: substrate, Ru3(CO)12 (5 mol %), and NMO (5-15
mol %) were stirred at 100-135 °C in pyridylmethyl formate for 1.5-12
h. HOAc was added to complete lactonization when necessary. b R )
n-hexyl. c Isolated as a 1:1 mixture of diastereomers. d Yield at 89%
conversion.
ted us to conduct subsequent mechanistic studies at two
temperatures and to enhance reaction rates for sterically
hindered substrates at 135 °C.
To gauge the scope of this reaction with functionalized
olefins as limiting reagents and to assess the potential of
hydroxyl coordination to alter the regiochemical outcome
of the hydroesterification reaction, we prepared several
substrates and exposed them to hydroesterification conditions
(Table 1). Homoallylic alcohol 8, in which the absence of
branching at the allylic carbon was expected to be less
detrimental to the formation of the branched product relative
to previous substrates, indeed provided γ-lactone 9 (48%)
as a diastereomeric mixture and δ-lactone 10 (25%). This
result, while not immediately useful in a synthetic sense,
(6) Shambayati, S.; Crowe, W. E.; Schreiber, S. L. Tetrahedron Lett.
1990, 31, 5289.
4208
Org. Lett., Vol. 6, No. 23, 2004

demonstrates that coordination has the capacity to reverse
the regiochemical outcome of hydroesterification reactions.
As expected, hydroesterification of silyl ether 11 provided
a mixture of linear ester 12 and branched ester 13 in an
approximately 3:1 ratio. The major product in this reaction,
however, was olefin isomer 14. Allylic alcohol 15 provided
only a moderate yield of γ-lactone 16, with the major side
product resulting from starting material isomerization to form
ketone 17. Isomerization can be suppressed by silylation of
the hydroxyl group. Ether 18 underwent hydroesterification
to yield ester 19 in good yield, ultimately providing an
excellent method for γ-lactone formation from allylic alco-
hols. Internal olefins proved to be much less reactive than
terminal olefins (data not shown), but 1,1-disubstituted alkene
20 reacted efficiently, albeit slowly, to form lactone 21 as a
1:1 mixture of diastereomers. Hindered neopentyl and tertiary
alcohols proved to be very suitable substrates. Disubstitution
at the allylic position (entry 6) actually resulted in quantita-
tive hydroesterification, though the linear to branched ratio
was only approximately 2:1. Notably, menthone derivatives
25 and 27, which were poor substrates for a recently reported7
metathesis-based lactonization strategy, were converted to
spirocyclic lactones 26 and 298 in satisfactory yields.
Hindered substrates reacted most efficiently in sealed tubes,
most likely because the substrates are somewhat volatile and
can be lost upon prolonged heating.
The formation of ketone 17 and varying amounts of olefin
isomers of the starting materials suggest that â-hydride
elimination can be competitive with reductive elimination,
thereby challenging our original hypothesis of product
selectivity being set in the hydrometalation step. To obtain
a more precise understanding of the reaction mechanism, we
initiated a study in which deuterated pyridylmethyl formate,
easily prepared from commercially available DCO2D, served
as a deuterioesterification reagent.9 In these experiments we
define R-deuteration as deuterium incorporation on the
carbon bearing the ester group and â-deuteration as deute-
rium incorporation on the adjacent carbon. Exclusive â-
deuteration would be expected if product regiochemistry were
solely dictated by the initial hydrometalation step, whereas
a mixture of R- and â-deuteration would be expected if
hydrometalation were reversible. The site of deuterium
incorporation is readily monitored by deuterium NMR. We
subjected 32-35 to deuterioesterification conditions. The
results of this study are shown in Table 2.
Table 2. Deuterium Incorporation Studiesa
a General procedure: substrate, Ru3(CO)12 (5 mol %), and NMO (5-15
mol %) were stirred at 100-135 °C in deuterated pyridylmethyl formate
for 1.5-12 h. b R ) p-methoxyphenoxyethyl. c Reaction conducted at 100
°C.
These results strongly indicate that hydrometalation is
reversible and that regiochemical preferences in this step
cannot be the sole determinant in the partitioning between
linear and branched products for this series of compounds.
A striking observation from this work is that little or no
selectivity is observed in the site of deuterium incorporation
in most of the reaction products. Also noteworthy is that
stopping the hydroesterification of 34 prior to complete
conversion resulted in the isolation of starting material in
which deuterium was incorporated at both vinylic positions.
Figure 3. Mechanistic pathways for hydroesterification.
(7) Cossy, J.; Baraggia, F.; BouzBouz, S. Org. Lett. 2003, 5, 459.
(8) Although ester 28 cyclized only reluctantly under thermal conditions,
it can be converted to lactone 29 quantitatively by treatment with NaH in
THF. See Supporting Information for details.
(9) For related examples of using deuteration as a mechanistic probe in
hydrometalation reactions, see: (a) Casey, C. P.; Martins, S. C.; Fagan, M.
A. J. Am. Chem. Soc. 2004, 126, 5585. (b) Bosnich, B. Acc. Chem. Res.
1998, 31, 667. (c) Evans, D. A.; Fu, G. C.; Anderson, B. A. J. Am. Chem.
Soc. 1992, 114, 6679.
Figure 3 shows a revised mechanism that is consistent with
the deuteration patterns shown in Table 2. Hydrometalation
proceeds with no selectivity, even with substrates that are
branched at the allylic position. â-Hydride elimination is, in
most cases, rapid relative to reductive elimination. Internal
olefin formation by â-hydride elimination accounts for
Org. Lett., Vol. 6, No. 23, 2004
4209

8. In this reaction we postulate that branched hydrometalation
is favored over linear hydrometalation, possibly resulting
from hydroxyl coordination, thereby increasing the concen-
tration of the secondary alkylruthenium species and promot-
ing branched product formation.
We have demonstrated that hydroesterification reactions
of functionalized olefins can proceed efficiently even when
the olefin is used as the limiting reagent. These processes
proceed most effectively when branching is present at the
allylic position to suppress olefin isomerization and can be
conducted on 100 mmol scale.11 Homoallylic hydroxyl
groups promote the formation of branched esters to a modest
extent. Mechanistic studies with deuterated pyridylmethyl
formate show that, whereas the hydroxyl group might very
well influence the regiochemistry of the hydrometalation step,
the partitioning of products between linear and branched
isomers is dependent upon the relative rates of reductive
elimination between primary and secondary alkylruthenium
species. The hydroxyl group serves to slow â-hydride
elimination of the secondary alkylruthenium species to an
extent that allows reductive elimination to be a viable
process.
Application of hydrometalation/reductive elimination pro-
cesses to complex molecule synthesis will ultimately require
an understanding of the effects of remote functionality on
regiochemistry and efficiency. Elucidating the consequences
of heteroatom coordination through mechanistic studies of
the type described herein will prove to be beneficial in the
design of procedural variants that expand the scope of this
useful reaction class.
Figure 4. Role of coordination in hydroesterification.
starting material isomerization. As discussed above, these
products undergo hydroesterification reactions only very
slowly under our conditions. Allylic branching suppresses
starting material isomerization, leading to enhanced ef-
ficiency for ester formation for all substrates except allylic
alcohols. The selectivity for linear products thus arises from
reductive elimination being more efficient for primary
alkylruthenium species than for secondary species, either
through having an inherently lower energy of activation or
through â-hydride elimination being significantly faster for
secondary alkylruthenium compounds. Branching provides
further drive for linear product formation. This can be
attributed to steric hindrance disfavoring the branched isomer
as the alkylruthenium intermediates equlibrate prior to
â-hydride elimination.
The conspicuous diminution of R-deuteration in the
branched products of entries 2 and 4 provides compelling
evidence for the role of heteroatom coordination (Figure 4).
These data indicate that the primary alkylruthenium species
undergoes reductive elimination in nearly exclusive prefer-
ence to â-hydride elimination. This result is most economi-
cally ascribed to the hydroxyl group occupying the coordi-
nation site that is required for â-hydride elimination. By
analogy, coordination must slow â-hydride elimination to a
sufficient extent in branched intermediates for reductive
elimination to become a competitive pathway.10 Although
this analysis explains improved efficiency for branched
product formation, it does not explain its preference from
Acknowledgment. This work was supported by generous
funding from the National Science Foundation.
Supporting Information Available: Full experimental
procedures and characterization data for all new compounds
and deuterium NMR spectra. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL048378F
(10) For a discussion on the roles of coordination in reductive elimination
and â-hydride elimination, see: Collman, J. P.; Hegedus, L. S.; Norton, J.
R.; Finke, R. G. Principles and Applications of Organotransition Metal
Chemistry; University Science Books: Mill Valley, CA, 1987.
(11) Seiders, J. R., II. Unpublished results.
4210
Org. Lett., Vol. 6, No. 23, 2004