Addition of Bifunctional Organoboron Reagents to Strained Alkenes. Carbon-Carbon Bond Formation with Rh(I) Catalysis in Aqueous Media

Article abstract of DOI:10.1021/jo049874k

Arylboronate esters bearing a pendant Michael-type acceptor olefin undergo transmetalation with a rhodium-based catalytic complex to generate a functionalized organorhodium intermediate that can cyclize onto strained olefins in good to excellent yields. The catalytic system involves the use of an electron-rich, sterically bulky ligand to stabilize the organorhodium intermediate and reduce the incidence of protodeboronation in aqueous media.

Full text of DOI:10.1021/jo049874k

Ad d ition of Bifu n ction a l Or ga n obor on Rea gen ts to Str a in ed
Alk en es. Ca r bon -Ca r bon Bon d F or m a tion w ith Rh (I) Ca ta lysis in
Aqu eou s Med ia
Mark Lautens* and J ohn Mancuso†
Davenport Chemical Laboratories, Department of Chemistry, University of Toronto, 80 St. George Street,
Toronto, Ontario M5S 3H6, Canada
mlautens@chem.utoronto.ca
Received J anuary 22, 2004
Arylboronate esters bearing a pendant Michael-type acceptor olefin undergo transmetalation with
a rhodium-based catalytic complex to generate a functionalized organorhodium intermediate that
can cyclize onto strained olefins in good to excellent yields. The catalytic system involves the use
of an electron-rich, sterically bulky ligand to stabilize the organorhodium intermediate and reduce
the incidence of protodeboronation in aqueous media.
In tr od u ction
to the seminal studies by Miyaura10 and Hayashi,11 who
demonstrated conjugate addition of arylboronic acids to
enones and aldehydes. Further studies by both groups
have shown that the coupling process can be rendered
asymmetric with the electrophilic component replaced
with activated olefins.12 An important aspect of these
reactions was that water was necessary as cosolvent (or
additive) to promote the coupling process through the
generation of a catalytically active hydroxorhodium(I)
intermediate from Rh-X (X ) acac, Cl, etc.).13 In addi-
tion, reactivity differences between Rh and Pd have been
Transition metal-catalyzed cross-coupling reactions
have proven vital in the assembly and construction of
various compounds for pharmaceutical usage, the syn-
thesis of natural products, and the generation of novel
supramolecular constructs.1 Protocols in frequent use
involve the Stille,2 Suzuki,3 and Heck4 couplings which
established palladium catalysis5 as a robust method for
carbon-carbon σ-bond formation. A major area of re-
search focuses on the solubilization of transition metal
complexes in water through the use of water-soluble
phosphines to carry out catalysis in aqueous media.6,7
As an alternative to palladium catalysis, rhodium-
catalyzed cross-couplings8,9 are gaining in popularity due
(8) Reviews on Rh-catalyzed couplings: (a) Fagnou, K.; Lautens, M.
Chem. Rev. 2003, 103, 169. (b) Hayashi, T.; Yamasaki, K. Chem. Rev.
2003, 103, 2829.
(9) Some recent applications of rhodium catalysis for the formation
of C-C bonds: (a) Dhanalakshmi, K.; Vaultier; M. Tetrahedron 2003,
59, 9907. (b) Lautens, M.; Yoshida, M. J . Org. Chem. 2003, 68, 762.
(c) Zou, G.; Wang, Z.; Zhu, J .; Tang, J . Chem. Commun. 2003, 2438.
(d) Oi, S.; Moro, M.; Ito, H.; Honma, Y.; Miyano, S.; Inoue, Y.
Tetrahedron 2002, 58, 91. (e) Pucheault, M.; Darses, S.; Geneˆt, J .-P.
Tetrahedron Lett. 2002, 43, 6155. (f) Pucheault, M.; Darses, S.; Geneˆt,
J .-P. Eur. J . Org. Chem. 2002, 3552. (g) Amengual, R.; Michelet, V.;
Geneˆt, J .-P. Synlett 2002, 11, 1791. (h) Kuriyama, M.; Nagai, K.;
Yamada, K.; Miwa, Y.; Taga, T.; Tomioka, K. J . Am. Chem. Soc. 2002,
124, 8932. (i) Koike, T.; Du., X.; Mori, A.; Osakada, K. Synlett 2002, 2,
301. (j) Boiteau, J .-G.; Imbos, R.; Minnaard, A. J .; Feringa, B. L. Org.
Lett. 2003, 5, 681. (k) Chapman, C. J .; Frost, C. G. Adv. Synth. Catal.
2003, 345. (l) Venkatraman, S.; Meng, Y.; Li, C.-J . Tetrahedron Lett.
2001, 42, 4459. (m) Murakami, M.; Igawa, H. Chem. Commun. 2002,
390.
† Present address: Department of Chemistry, University of Pitts-
burgh, Pittsburgh, PA 15260.
(1) For general references on cross-coupling reactions, see: (a)
Miyaura, N., Ed. Cross-coupling reactions:
a
practical guide;
Springer: New York, 2002. (b) Murahashi, S.-i., Davies, S. G., Eds.
Transition Metal Catalyzed Reactions; Blackwell Science: Oxford, UK,
1999. (c) Diederich, F.; Stang, P. J . Metal-Catalyzed Cross-Coupling
Reactions; Wiley-VCH: New York, 1998. (d) Ojima, I., Ed. Catalytic
Asymmetric Synthesis; Wiley-VCH: New York, 2000.
(2) For reviews on the Stille coupling, see: (a) Farina, V.; Krishna-
murthy, V.; Scott, W. J . The Stille Reaction; J ohn Wiley & Sons: New
York, 1998. (b) Stille, J . K. Angew. Chem., Int. Ed. Engl. 1986, 25,
508. (c) Mitchell, T. N. Synthesis 1992, 803.
(3) For reviews on the Suzuki coupling, see: (a) Miyaura, N.; Suzuki,
A. Chem. Rev. 1995, 95, 5, 2457. (b) Miyaura, N. Organoboron
Compounds in Topics in Current Chemistry, 2002, Vol. 219, Springer-
Verlag: Heidelberg, p 11. (c) Kotha, S.; Lahiri, K.; Kashinath, D.
Tetrahedron 2002, 58, 9633.
(10) (a) Sakai, M.; Hayashi, H.; Miyaura, N. Organometallics 1997,
16, 4229. (b) Ueda, M.; Miyaura, N. J . Org. Chem. 2000, 65, 4450.
(11) Takaya, Y.; Ogasawara, M.; Hayashi, T.; Sakai, M.; Miyaura,
N. J . Am. Chem. Soc. 1998, 120, 5579.
(4) For reviews on the Heck reaction see: (a) Dounay, A. B.;
Overman, L. E. Chem. Rev. 2003, 103, 2945. (b) Beletskaya, I. P.;
Cheprakov, A. V. Chem. Rev. 2000, 100, 3009.
(5) Tsuji, J . Palladium Reagents and Catalysis: Innovations in
Organic Synthesis; Wiley & Sons: New York, 1995.
(6) For reviews on transition metal catalysis with water-soluble
ligands, see: (a) Pinault, N.; Bruce, D. W. Coord. Chem. Rev. 2003,
241, 1. (b) Lindstro¨m, U. M. Chem. Rev. 2002, 102, 2751. (c) Sinou, D.
Adv. Synth. Catal. 2002, 344, 221.
(7) For a review on organometallic chemistry in water, see: Her-
rmann, W. A. In Aqueous-Phase Organometallic Chemistry; Cornils,
B., Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 1998;
pp 35-45.
(12) Asymmetric addition to enones: (a) Takaya, Y.; Ogasawara,
M.; Hayashi, T.; Sakai, M.; Miyaura, N. J . Am. Chem. Soc. 1998, 120,
5579. (b) Takaya, Y.; Ogasawara, M.; Hayashi, T. Tetrahedron Lett.
1999, 40, 6957. (c) Yamamoto, Y.; Fujita, M.; Miyaura, N. Synlett 2002,
5, 767. Asymmetric addition to acrylates: (d) Takaya, Y.; Senda, T.;
Kurushima, H.; Ogasawara, M.; Hayashi, T. Tetrahedron: Asymmetry
1999, 10, 4047. (e) Sakuma, S.; Miyaura, N. J . Org. Chem. 2001, 68,
8944. Asymmetric addition to vinylphosphonates: (f) Hayashi, T.;
Senda, T.; Takaya, Y.; Ogasawara, M. J . Am. Chem. Soc. 1999, 121,
11591. Asymmetric addition to vinyl nitro compounds: (g) Hayashi,
T.; Senda, T.; Ogaswara, M. J . Am. Chem. Soc. 2000, 122, 10716.
(13) Hayashi, T.; Takahashi, M.; Takaya, Y.; Ogasawara, M. J . Am.
Chem. Soc. 2002, 124, 5052.
10.1021/jo049874k CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/21/2004
3478
J . Org. Chem. 2004, 69, 3478-3487

Catalytic Bifunctional Organoboronate Additions
SCHEME 1
observed. For example, the tolerance of rhodium for aryl-
halide bonds has been demonstrated, which positions this
methodology as complementary to Pd-based methods.
Previous reports have indicated that the C-Br bond in
aryl bromides is not readily prone to oxidative insertion
by rhodium and aryl C-I insertion is promoted only at
high temperatures (>100 °C).14 Furthermore, in boron-
based couplings, the waste product of the catalytic cycle
is usually boric acid, a relatively nontoxic species as
compared to metal halide salts generated by other
coupling processes.
We demonstrated recently that a cross-coupling be-
tween arylboronic acids and styrene-type olefins or
2-vinyl(azaheteroaromatics) could occur in neat water
using SDS (sodium dodecyl sulfate) as a surfactant with
a Rh(I) catalyst.15 The hydrophilic ligand, TPPDS (di-
potassium bis(p-sulfonatophenyl)phenylphosphine),16 was
used to solubilize the catalyst in the aqueous phase.
Styrene-type olefins were found to generate unsaturated
products, presumably arising from a Heck-type process.
Vinyl(azaheteroaromatics) followed a conjugate addi-
tion-protonation pathway, resulting in a net hydroary-
lation process. Protodemetalation of the boronic acid
partner could be attenuated through the use of SDS, but
still required it be used in excess (2.5 equiv) with respect
to alkene.
As the study pointed to reactivity differences between
rhodium and palladium in intermolecular couplings, it
was of interest to determine how intramolecular pro-
cesses would fare. The majority of methodologies reported
with rhodium catalysis focus on intermolecular processes
with the metal mediating the formation of only one
carbon-carbon bond. In contrast, several studies have
demonstrated the flexibility of palladium chemistry,
through the creation of tandem or “cascade” processes,17
promoting the formation of several C-C and C-X bonds
in a “one-pot” fashion.
reliant on the generation of organorhodium intermediates
which cannot undergo undesirable side reactions such as
â-hydride elimination or protodemetalation by water.
These intermediate species should have sufficient longev-
ity and reactivity to proceed in further carborhodative
steps, ultimately building fused ring systems. The pro-
posed concept is shown in Scheme 1, whereby the use of
a bifunctional donor/acceptor I, containing group M (a
metal/metalloid moiety) and group A (an acceptor moiety;
i.e., O, NR, CH-EWG, etc.).
Group M would allow for a transmetalation reaction
to occur in the presence of a rhodium catalyst to generate
an organorhodium intermediate II. The organorhodium
intermediate can coordinate another olefin and undergo
migratory insertion, resulting in the carborhodated spe-
cies III. Internal coordination of the rhodium center to
the acceptor group A may be possible and cyclization can
occur to give IV. At this stage, protodemetalation by a
protic donor would give V or â-hydride elimination of
Rh-H from IV would generate VI, the unsaturated
product.
The key to the success of this endeavor is avoiding
protodemetalation at an intermediate stage in the pro-
cess. Indeed, one of the drawbacks of working in aqueous
organorhodium chemistry is the propensity for reaction
with water to occur faster than carborhodation of the
olefin. This is often compensated for by the addition of
several equivalents (2.5-10) of arylboron substrate thus
dramatically reducing the practical utility18 of the reac-
tion. From a financial perspective, the cost of organoboron
reagents, in particular functionalized or heterocyclic
species, remains high. To render the proposed cascade
reaction synthetically attractive, it was necessary to
devise a method whereby only 1 equiv of the organoboron
substrate can be used, allowing yields to be based on this
coupling partner.
In this article, we present a comprehensive study on
the generation of functionalized indanes from bifunc-
tional arylboronate esters and norbornyl olefins, which
had appeared previously as a communication.19 In addi-
tion, we also describe our efforts to generate products
from thiophene-based boronate ester substrates.
In theory, rhodium catalysis could be used to effect a
similar cascade sequence. The success of this process is
(14) Our previous study demonstrated that halide substituents on
the boronic acid coupling partner survive the cross-coupling conditions;
see ref 15 and for other examples indicating the resistance of Rh
insertion into aryl halide bonds at low temperature; see: (a) Ishiyama,
T.; Hartwig, J . J . Am. Chem. Soc. 2000, 122, 12043. (b) Grigg, R.;
Sridharan, V.; Stevenson, P.; Sukirthalingam, S.; Worakun, T. Tetra-
hedron 1990, 46, 4003.
Resu lts a n d Discu ssion
A communication by Miura disclosed the “merry-go-
round” alkylation of phenyboronic acid by norbornene in
(15) Lautens, M.; Roy, A.; Fukuoka, K.; Fagnou, K.; Martin-Matute,
B. J . Am. Chem. Soc. 2001, 123, 5358.
(16) This ligand is commercially available from Strem Chemicals.
(17) Selected reviews on Pd cascade reactions: (a) Heumann, A.;
Re´glier, M. Tetrahedron 1996, 52, 9289. (b) Grigg, R.; Sridharan, V.
J . Organomet. Chem. 1999, 576, 65.
(18) (a) Trost, B. M. Science 1991, 254, 1471. (b) Trost, B. M. Angew.
Chem., Int. Ed. Engl. 1995, 24, 259.
(19) Lautens, M.; Mancuso, J . Org. Lett. 2002, 4, 2105.
J . Org. Chem, Vol. 69, No. 10, 2004 3479

Lautens and Mancuso
SCHEME 2
TABLE 1. Op tim iza tion of th e Bifu n ction a l Ad d ition w ith a Rh (I) Ca ta lyst a n d Wa ter -Solu ble Liga n d s
entrya
catalyst/ligand
conditions
conv, %b
(2:3)c
69:31
1
2
3
[Rh(cod)Cl]2
80 °C, 2 h
80 °C, 2 h
80 °C, 2 h
80 °C, 2 h
80 °C, 2 h
rt, 14 h
13
[Rh(cod)Cl]2/TPPDSd
>99
>99
>99 (95)
>99
>99
NR
27:73
e
[Rh(cod)Cl]2/t-Bu-amphos-Cl
[Rh(cod)Cl]2/t-Bu-amphos-Cl
[Rh(cod)Cl]2/t-Bu-amphos-Cl
[Rh(cod)Cl]2/t-Bu-amphos-Cl
[Rh(cod)Cl]2/Cy-amphos-Cl
[Rh(cod)OH]2/t-Bu-amphos-Cl
-
4f
5f,g
6f
7f
8f
>99:<1
96:4
>99:<1
rt, 14 h
80 °C, 2 h
>99
>99: <1
a
b
Reactions run in 0.2 M degassed water under nitrogen with 0.6 equiv of SDS and 1 equiv of Na2CO3. Determined by 1H NMR.
Isolated yield in parentheses. c Determined by H NMR. 8.0 mol % of ligand used. e Variable results. f 1:1 water:toluene (degassed) was
used as the solvent. 1 equiv of norbornene was used.
1
d
g
SCHEME 3
the presence of a rhodium catalyst.20 Through carborho-
dation of norbornene and a series of Rh migrations, it
was observed that norbornyl groups were introduced at
multiple positions around the phenyl ring. Furthermore,
this experiment demonstrated the resistance of nor-
bornylrhodium intermediates toward â-hydride elimina-
tion and showed that norbornene could function as a
potential “trapping” agent for organorhodium intermedi-
ates.
reactions in aqueous media,21 the amphos ligands were
Test substrate 1 was synthesized from the readily
available 2-formylphenylboronic acid, which was esteri-
a viable alternative as they could be readily prepared.22
Recently, Shaughnessy and co-workers have reported the
use of tert-butyl-amphos chloride as a soluble ligand in
a water/acetonitrile-based Suzuki coupling system giving
excellent yields.23 Grubbs and co-workers have also
demonstrated the use of Cy-amphos chloride for the
solubilization of Ru metathesis catalysts in water and
methanol.24
As seen in Table 1, entry 2, the use of the TPPDS
ligand allowed for complete consumption of the starting
material, with protodemetalation as the predominant
reaction pathway. The use of the tert-butyl-amphos
ligand gave excellent product conversion but the ratio of
products 2 and 3 was variable. Problems with irrepro-
ducibility were believed to result from poor solubilization
of the substrate or intermediates in the organic phase
fied as its pinacol derivative to facilitate product recovery
and purification following the Wittig coupling (Scheme
2). Compound 1 was isolated as a mixture of E/Z isomers,
with the E isomer being predominant.
Substrate 1 was designed to incorporate a pendant
acrylate group, a moiety that is known to be highly
reactive as an electrophile with organorhodium nucleo-
philes. It was tested initially with a catalytic amount of
[Rh(cod)Cl]2 in water with SDS and norbornene under
the same conditions which were successful for the
styrene/boronic acid coupling.15 Consumption of 1 was
slow and although the desired product 2 was detected, it
was in very low yield as protodemetalation to generate
3 was a problematic side reaction (Table 1, entry 1).
The desired indane 2 was isolated as a single diaste-
reomer and a ROESY 2D-NMR experiment was used to
assign the stereochemical configuration as the exo adduct.
To improve the conversion and reduce the incidence of
deboronated compound 3, a variety of water-soluble
ligands were tested (Scheme 3).
(21) Some recent articles concerning the use of TPPDS in catalysis:
(a) Peng, Q. R.; Yang, Y.; Wang, C. J .; Liao, X. L.; Yuan, Y. Z. Catal.
Lett. 2003, 88, 219. (b) Chen, H.; Li, Y. Z.; Chen, J . R.; Cheng, P. M.;
Li, X. J . Catal. Today 2002, 74, 131. (c) Li, Y. Z.; Chen, H.; Chen, J .
R.; Cheng, P. M.; Hu, J . Y.; Li, X. J . Chin. J . Chem. 2001, 19, 58.(d)
Thorpe, T.; Brown, S. M.; Crosby, J .; Fitzjohn, S.; Muxworthy, J . P.;
Williams, J . M. J . Tetrahedron Lett. 2000, 41, 4503.
(22) Smith, R. T.; Baird, M. C. Inorg. Chim. Acta 1982, 62, 135.
(23) Shaughnessy, K. H.; Booth, R. S. Org. Lett. 2001, 3, 2757.
(24) (a) Lynn, D. M.; Mohr, B.; Grubbs, R. H.; Henling, L. M.; Day,
M. W. J . Am. Chem. Soc. 2000, 122, 6601. (b) Mohr, B.; Lynn, D. M.;
Grubbs, R. H. Organometallics 1996, 15, 4317.
In addition to the TPPDS ligand, which has been
proven to be ideal for several transition metal-mediated
(20) Oguma, K.; Miura, M.; Satoh, T.; Nomura, M. J . Am. Chem.
Soc. 2000, 122, 10464.
3480 J . Org. Chem., Vol. 69, No. 10, 2004

Catalytic Bifunctional Organoboronate Additions
TABLE 2. Scr een in g of Heck Accep tor Olefin
TABLE 3. Scr een in g of Diels-Ald er Ad d u cts a n d
Nor bor n a d ien e Der iva tives a s Cou p lin g P a r tn er s
a
b
a
b
Isolated by column chromatography. Slow addition of bor-
Ratio of boronate ester to alkene is 1:2. Isolated by column
onate ester to reaction mixture over 12 h at rt. c Protodeboronation
of the substrate only.
chromatography. Diastereomeric excess is >20:1 for all entries.
provided by the surfactant. It was gratifying to discover
that this problem could be alleviated through the use of
an organic cosolvent. Entries 4-8 employ the use of a
1:1 toluene-to-water system containing SDS as a surfac-
tant. This system resulted in the formation of an opaque
yellow emulsion and allowed complete conversion of
substrate to product with very little protodemetalation.
Entry 5 indicated that excess olefin was unnecessary as
only 1 equiv of norbornene gave no significant decrease
in yield. Entry 6 demonstrated that conversion at room
temperature was feasible, albeit with longer reaction
time. However, the use of cyclohexyl-amphos chloride did
not result in any conversion at room temperature (entry
7). Entry 8 supported our hypothesis that a hydroxo-
rhodium(I) intermediate was the active catalytic species,
eochemical configuration of the products was assigned
based on NMR analysis and comparison to compound 2.
A boronate ester 8 containing an R,â-unsaturated alde-
hyde moiety gave a mixture of products. In this case,
competitive 1,2-addition to the aldehyde by another
equivalent of substrate might occur. To improve the yield
of 9, it was necessary to slowly add the substrate as a
solution in toluene to the reaction mixture by syringe
pump at room temperature (entry 3). An acrylonitrile-
based substrate 10 (entry 4) was found to hinder the rate
of transmetalation at room temperature, presumably by
coordination of multiple cyano groups to the rhodium
center, thus attenuating the metal’s reactivity. The use
of the R,â-unsaturated sulfone 11 (entry 5) allowed for
transmetalation of the substrate, as evidenced from
isolation of protodemetalated material, but did not yield
any desired product.
25
as the addition of pregenerated [Rh(cod)OH]2 showed
no difference from the standard chloride catalyst precur-
sor.
Norbornyl derivatives with increased functionality
were examined (Table 3). The Diels-Alder adduct of
cyclopentadiene and maleic anhdydride (12) resulted in
a strongly exothermic reaction with no conversion to the
desired product (entry 1). It was speculated that rhodium
insertion into the C-O bonds of the anhydride group had
occurred as it has been previously shown by Frost that
anhydrides can be used to acetylate arylboronic acids
with use of a rhodium catalyst.26 The Diels-Alder adduct
of N-methylmaleimide (13) proceeds to give the penta-
cycle 14.
It is also interesting to note that the geometric con-
figuration of the double bond in the arylboronate sub-
strate 1 seemed to have no effect on the diastereoselec-
tivity. Variations in the ratio of E-to-Z isomers were
tolerated and gave the same product diastereomer.
tert-Butyl-amphos was chosen for further studies as it
exhibited higher activity as compared to cyclohexyl-
amphos or TPPDS. Functionality on the arylboronate
ester component was examined further by testing various
pendant Heck-acceptor groups (Table 2).
Entries 1 and 2 illustrate that both enone and acry-
lamide functionalites are compatible. The relative ster-
Norbornadiene-based coupling partners27 were also
investigated and entry 3 demonstrates the addition to
(25) Uson, R.; Oro, L. A.; Cabeza, J . A. Inorg. Synth. 1985, 23, 126.
(26) Frost, C. G.; Wadsworth, K. J . Chem. Commun. 2001, 22, 2316.
J . Org. Chem, Vol. 69, No. 10, 2004 3481

Lautens and Mancuso
TABLE 4. Scr een in g of Oxa bicyclic Alk en es a s
Cou p lin g P a r tn er s
SCHEME 4
norbornadiene. The stereochemistry of compound 16 was
verified by hydrogenation, which yielded compound 2.
Entry 4 indicates that the reaction is compatible with
substituted norbornadiene 17 and coupling is selective
for the unsubstituted double bond.
An example of a postsynthetic modification to yield
functionalized tricyclic systems is shown in Scheme 4.
Ozonolysis of 16 followed by reductive workup with
NaBH4 gave the diquinane 19 in quantitative yield.
Compound 19 bears five stereocenters, three of which are
created in the coupling step.
Since norbornyl- and norbornadienyl-alkenes were
successful coupling partners, coupling partners contain-
ing heteroatoms were examined. A variety of oxabicyclic
alkenes were screened and the results are listed in Table
4.
Entry 1 indicated that oxabenzonorbornadiene 20
inhibited the reaction, a surprising result since previous
studies with oxabicyclic systems and arylboronic acids
with rhodium catalysis showed this substrate was com-
patible with the reaction conditions.28 Oxidative addition
into one of the strained C-O bonds of the oxabicyclic
bridge is a possibility and may account for termination
of the reaction.
Entries 2-4 indicate the generality of the reaction as
both [2.2.1] and [3.2.1] systems add very well. Interest-
ingly, no ring-opened products could be detected. [3.2.1]
systems are known to be resistant to ring opening but
the integrity of the [2.2.1] system, a substrate that is
known to be a highly reactive strained olefin, is preserved
in this case.
a
Isolated by column chromatography. Diastereomeric excess is
>20:1 for all entries.
which are less strained, but are expected to be more
sensitive to substitution patterns as the bridgehead
substituents “shield” the alkene to a greater extent, due
to increased bond angles within the ring. Substrate 32
(entry 4) gives only deboronated material as expected for
a disubstituted system while substrate 33 (entry 5) gives
the desired product 34 as a single regio- and diastereo-
isomer, and its relative configuration was confirmed by
X-ray analysis. Entry 6 shows that the presence of a
single sterically encumbering bridgehead substituent
prevents addition completely, resulting only in debor-
onation. Entry 7 demonstrates that a phenyl group at
the bridgehead is not tolerated, resulting in a complete
shutdown of the reaction. In a similar process seen for
oxabenzonorbornadiene, it is possible that the benzylic
C-O bond is weakened and prone to oxidative insertion
by the metal, resulting in a Rh-substrate complex that
is inactive and sequesters the metal.
An interesting aspect of the reaction was discovered
when substituents were placed at the bridgehead carbons
of the oxabicyclic systems (Table 5).
Entry 1 suggests that double bridgehead substitution
prevents attack of the arylrhodium nucleophile onto the
alkene, presumably due to a steric shielding effect by the
methyl groups. However, placement of only one methyl
group (entry 2) yields a product in near quantitative yield
as one diastereomer and regioisomer. The relative stereo-
chemistry was confirmed by single-crystal X-ray analysis.
Further investigation showed that a larger group such
as -CH2CH2OTHP (entry 3) gives a lower yield of
product but still as only one diastereomer and regioiso-
mer. A similar trend is noted for the [3.2.1] systems
Bicyclic alkenes containing nitrogen atoms were also
briefly examined (Table 6).
An azabenzonorbornadiene substrate 37 gave similar
results as the oxygen analogue and did not yield the
desired product (entry 1) even when changing the N-
group from Boc to Ts or Ph. The Diels-Alder adduct of
di-tert-butyl azodicarboxylate and cyclopentadiene (38)
could be coupled to give the product 39 with the hy-
drazido bond intact (entry 2).29 The Diels-Alder adduct
of N-Boc-pyrrole and DMAD 40 failed, resulting only in
deboronation of the substrate. Substrate 41 (entry 4), a
(27) The ratio of arylboronate substrate to alkene deserves some
comment. In cases where the alkene is volatile and any excess can be
easily removed by application of vacuum, 2 equiv are usually used to
ensure complete reaction conversion. In cases where the alkene is
heavily functionalized, or the coupling partner has a similar Rf to the
product on silica gel, the loading is reduced to 1.05-1.1 equiv. In all
cases, this reduction has no detrimental effect on selectivity or yield
of the reaction.
(29) The N-N bond in cyclic compounds of this type can be cleaved
with Na/NH3, see: Mellor, J . M.; Smith, N. M., J . Chem. Soc., Perkin
Trans. 1 1984, 12, 2927.
(28) Lautens, M.; Dockendorff, C.; Fagnou, K.; Malicki, A. Org. Lett.
2002, 4, 1311.
3482 J . Org. Chem., Vol. 69, No. 10, 2004

Catalytic Bifunctional Organoboronate Additions
TABLE 5. Scr een in g of Br id geh ea d -Su bstitu ted
Oxa bicyclic Alk en es
TABLE 6. Scr een in g of Aza bicyclic Alk en es a s Cou p lin g
P a r tn er s
a
Products isolated by column chromatography. Diastereomeric
excess is >20:1 for all entries.
mine if the present coupling methodology could be used
to synthesize new analogues of tropane alkaloids. Prior
studies in the literature concerning the generation of
tropane analogues have pointed to the utility of scopo-
lamine as a starting scaffold for further elaboration. (-)-
Dehydrohyoscamine (42) could be easily generated from
commercial (-)-scopolamine as substrate for the cross-
coupling reaction (Scheme 5).31
Coupling with the boronate ester 1 under standard
conditions with KF as base gave the desired adduct 43
in good yield and as a mixture of diastereomers in a 1:1.3
ratio. More importantly, neither epimerization of the
R-center nor hydrolysis of the tropic acid side chain was
observed, emphasizing the mild conditions of the coupling
process.
In general, the catalytic cycle of this tandem cyclization
is believed to involve the presence of hydroxorhodium(I)
intermediates25,32 (Scheme 6). The rhodium(cyclooctadi-
ene)chloride dimer is known to hydrolyze at room tem-
perature in a basic aqueous solution to generate the
corresponding hydroxorhodium(I) species.33 Although the
presence of a base such as sodium carbonate is needed
to generate a small quantity of hydroxide ion to produce
the active rhodium catalyst and to render the boron
center tetracoordinate, it was found that fluoride is also
suitable, as has been previously observed for studies on
a
Products isolated by column chromatography. Diastereomeric
excess is >20:1 for all entries. Product isolated as one diastereomer
b
and regioisomer. Relative stereo- and regiochemistry verified by
X-ray single-crystal diffraction. c Relative stereo- and regiochem-
istry assigned based on carbon and proton NMR spectroscopy.
[4.2.2] bicyclic ring system, did not undergo any coupling,
implying that the alkene may not be activated enough
as ring strain is not as predominant as it is in the [3.2.1]
and [2.2.1] ring systems.
Azabicycles based on the [3.2.1] system are known
collectively as the tropane class of alkaloids, and have a
seven-membered ring with a bridging aminomethyl group
as the core structure. Members of this class are of interest
as pharmaceutical agents as they exhibit high and
diverse biological activity.30 It was of interest to deter-
(31) (a) Aberle, N. S.; Ganesan, A.; Lambert, J . N.; Saubern, S.;
Smith, R. Tetrahedron Lett. 2001, 42, 1975. (b) Hayakawa, Y.; Baba,
Y.; Makino, S.; Noyori, R. J . Am. Chem. Soc. 1978, 100, 1786.
(32) (a) Brune, H.-A.; Unsin, J .; Hemmer, R.; Reichardt, M. J .
Organomet. Chem. 1989, 369, 335. (b) Grushkin, V. V.; Kuznetsov, V.
F.; Bensimon, C.; Alper, H. Organometallics 1995, 14, 3927. (c)
Hayashi, T.; Takahashi, M.; Takaya, Y.; Ogasawara, M. J . Am. Chem.
Soc. 2002, 124, 5052.
(30) For a general review on the tropane class of alkaloids, see:
Fodor, G.; Dharanipragada, R. Nat. Prod. Rep. 1994, 11, 443.
(33) J oo´, F.; Kova´cs, J .; Be´nyei, A. C.; Na´dasdi, L.; Laurenczy, G.
Chem. Eur. J . 2001, 7, 193.
J . Org. Chem, Vol. 69, No. 10, 2004 3483

Lautens and Mancuso
SCHEME 5
SCHEME 6
the Suzuki coupling. Two equivalents of fluoride are
closure becomes kinetically favored over ring opening of
the strained system by â-deoxyrhodation.
-
believed to promote the formation of an Ar(BFnOH3-n
)
species, which transmetalates more readily.34 Indeed, the
reduced basicity of fluoride has proven beneficial in cases
where very acidic protons are present within the coupling
partners, thus rendering the use of protecting groups
unnecessary.
The 5-exo-trig ring closure is rendered diastereoselec-
tive due to the significant steric interactions between the
-COR group of the pendant acceptor alkene and the
bridge group (apical -CH2-, -O-, or -NR′). Thus, the
conformer in which the R group is oriented toward the
endo face of the norbornyl system is favored and leads
to the (oxa-π-allyl)rhodium species V. This is rapidly
protodemetalated by water, which releases the product
VI and regenerates the hydroxorhodium(I) species.
The catalytic cycle is initiated by transmetalation of
LnRh(OH) I with the arylboronate substrate II to gener-
ate the arylrhodium(I) species III and release B(pin)OH.
Coordination of the alkene followed by selective carbo-
rhodation on the exo face ensues to give IV in which the
rhodium is presumably coordinated to the internal
pendant olefin. This coordination may be crucial for cases
such as the oxa[2.2.1]bicyclic substrates whereby the ring
In cases where there is a substituent at one of the
bridgeheads, the selectivity of the syn-carborhodation
process is dictated by sterics; the aryl group adds at the
most sterically accessible carbon, placing the rhodium
group proximal to the bridgehead substituent (Scheme
7, path B favored over path A). When the bridgehead
(34) Wright, S. W.; Hageman, D. L.; McClure, L. D. J . Org. Chem.
1994, 59, 6095.
3484 J . Org. Chem., Vol. 69, No. 10, 2004

Catalytic Bifunctional Organoboronate Additions
SCHEME 7
SCHEME 8
Thienylboronate ester 44 was tested under a variety
of conditions and the results are reported in Table 7.
group is very large, both orientations are disfavored by
steric factors, thus addition is prevented and the compet-
ing protodemetalation pathway becomes predominant
(C).
Under the standard conditions employed previously
with the phenylboronate ester, no reaction was observed
with substrate 44, using sodium carbonate (entry 1) or
potassium fluoride (entry 2) as base at room temperature.
Using potassium fluoride (entry 3) with heating led to
the formation of two products, the desired tetracyclic
species 46 as a minor component and a hydroarylated
product 47, believed to arise from a C-H insertion
process (vide infra).37 These two products were insepa-
rable by column chromatography. The use of the cyclo-
hexyl-amphos ligand (entry 4) also resulted in a similar
reaction outcome, giving a mixture of 46 and 47. Switch-
ing the base to silver oxide (entry 5) led to deboronation.
The use of sodium hydroxide at room temperature led to
a mixture of starting material and deboronated material
(entry 6). Repeating the reaction at higher temperatures
simply accelerated the deboronation process (entry 7).
An alternate catalytic pathway involving oxidative
addition of Rh(I) into the C-B bond of the phenylboronic
acid could be proposed; however, Hayashi’s mechanistic
studies and NMR analysis on the conjugate addition of
boronic acids to enones have indicated that only rhodium-
(I) intermediates are present within the catalytic cycle.32c
Our previous catalytic system, used for the coupling
of arylboron substrates with vinylpyridines and sty-
renes,15 was reactive enough to couple these weakly
reactive alkenes, but suffered from selectivity issues,
preferring protodemetalation.35 Although yields could be
improved by increasing the boronic acid loading to 2.5
equiv and heating at 80 °C, the high cost of these
substrates limits large-scale applications. In the present
case, the ligand promotes the transmetalative step,
resulting in smooth conversion at room temperature.
Since there is no appreciable difference between loading
of 2 equiv of ligand with respect to the metal over 1 equiv,
it is believed that the active species is a monoligated
metal intermediate stabilized by the high basicity of the
amphos ligand. In addition, the large cone angle of tert-
butyl-amphos Cl (approximately 185°)36 may prevent
strong binding of another equivalent of ligand to the
metal center.
Substitution of the phenyl nucleus of arylboronate
ester 1 with other heterocyclic rings would provide access
to new structural motifs and compounds with useful
properties. Heterocycles are present as core structures
and often act as isosteres for the phenyl group within
many natural products and pharmaceuticals. Thiophene
was a suitable choice for a heterocyclic backbone as its
aromatic character is more similar to benzene than either
pyrrole or furan. The requisite boronate ester substrates
44 and 45 were synthesized from the corresponding
commercially available boronic acids.
Concerning the mechanism for the formation of the
C-H insertion product 47 from the boronate ester 44, it
is possible that a rhodium(I)-rhodium(III) manifold is
being followed. There are two possible divergent path-
ways that can be followed after the carborhodation of the
strained alkene (Scheme 9, intermediate A). One is the
expected attack onto the pendant alkene to generate the
cyclized product 46. The other pathway generates the
hydroarylated (C-H inserted) product 47. For C-H
insertion to occur, the intermediate must be at the
organorhodium(I) stage, as the resulting rhodacycle B,
from C-H insertion at the 4-position of the thienyl ring,
contains Rh in the higher valent state.38 Reductive
elimination then occurs to give C, with the overall result
in the rhodium “jumping” from the oxabicycle to the
thienyl ring.39 Performing the reaction in deuterium oxide
showed >95% incorporation of deuterium at the 4-posi-
tion (47A). It is interesting to note that C-H insertion
occurs only with the thienyl system and was never
observed with the phenylboronate esters.
(35) A control reaction was undertaken to determine the relative
propensity of a strained alkene and
a
Michael acceptor toward
(37) The more reactive [2.2.1]oxabicyclic alkenes were also tested,
as well as norbornene, and did not yield any desired ring-closed
products.
(38) The C-H insertion product as well as the minor cyclized
product was only observed when fluoride was used as base at high
temperatures. Fluoride may promote the oxidative process required
for C-H insertion, by stabilizing the Rh(III) state.
(39) Presumably, installation of a functional group at the 4-position,
such as a methyl group, would prevent this C-H insertion process
and promote formation of the desired cyclized product. This process is
under examination.
carborhodation in the present system. When 1.5 equiv each of nor-
bornene and ethyl acrylate were mixed with 1 equiv of phenylboronic
acid under standard reaction conditions, only ethyl cinnamate was
obtained, indicating that the arylrhodium species generated in this
reaction is highly nucleophilic in character. Increasing the amount of
norbornene to 5 equiv did not change the outcome of the reaction.
(36) Grubbs has tested Cy-amphos Cl and has determined its
Tolman cone angle to be very similar to that of Cy3P (170°). Although
t-Bu-amphos chloride has not been measured, its cone angle is assumed
to be similar to that of t-Bu3P (185°). See ref 24b.
J . Org. Chem, Vol. 69, No. 10, 2004 3485

Lautens and Mancuso
TABLE 7. Scr een in g of Rea ction Con d ition s for Cou p lin g of Com p ou n d 44
entry
base
temp/°C
ligand
46a
47a
deboronated
1
2
3
4
5
6
7
Na2CO3
KF
KF
KF
Ag2O
rt
rt
70
70
rt
t-Bu-amphos Cl
t-Bu-amphos Cl
t-Bu-amphos Cl
Cy-amphos Cl
t-Bu-amphos Cl
t-Bu-amphos Cl
t-Bu-amphos Cl
14
9
60
52
>99
>50
>99
5 M NaOH
5 M NaOH
rt
70
a
Yields estimated by 1H NMR.
SCHEME 9
The thienylboronate ester 44 seems to be very sensitive
to reaction conditions; in particular, the choice of base is
important. The use of stronger bases accelerates the
transmetalation to rhodium, but the resultant thienyl-
rhodium species either reacts too quickly with water or
poorly coordinates the strained alkene coupling partner.
Con clu sion s
We have developed a rhodium-catalyzed tandem cy-
clization involving arylboronic esters and various strained
olefins as coupling partners, which gives access to highly
functionalized polycyclic systems. The reaction was run
in the presence of water, whereby the water is not only
a solvent, but is also required to provide a proton source
to form the product and regenerate the catalyst.
Arylboronic esters bearing a pendant Michael acceptor
react with strained olefins to give highly functionalized
indanes in excellent diastereoselectivity. Both oxa- and
azabicycles were compatible as strained olefin coupling
partners and an example involving the synthesis of a
tropane alkaloid analogue was provided. The steric and
electronic character of the bifunctional partner is impor-
tant and may require modification of the reaction condi-
tions, including investigation of other amphos ligands,
to generate desired products, especially in the case of
heterocyclic systems. This observation is consistent with
cross-coupling reactions such as the Suzuki and Stille
couplings in which a variety of conditions have been
developed for the successful carbon-carbon bond forma-
tion for specific substrate classes. Investigations into the
Thienylboronate ester 45 could not be successfully
coupled to any strained alkene and simply underwent
deboronation under conditions involving a variety of
bases (Na2CO3, KF, CsF, Ba(OH)2, NaOH, and Tl2CO3),
temperatures, and ranges of water content. Difficulties
in encouraging this substrate to add to oxabicyclic
alkenes can be assigned to the fact that the boron group
is proximal to the thienyl sulfur. B-C hydrolytic cleavage
is known to be accelerated by R-heteroatoms (i.e. 2-py-
ridylboronic acid) or the presence of an adjacent Lewis
basic substituent (i.e. o-formylphenyl boronic acid).40,41
(40) Brown, H. C.; Molander, G. A. J . Org. Chem. 1986, 51, 4512.
(41) (a) Kuivilla, H. G.; Nahabedian, K. V. J . Am. Chem. Soc. 1961,
83, 2159. (b) Kuivilla, H. G.; Nahabedian, K. V. J . Am. Chem. Soc.
1961, 83, 2164. (c) Kuivilla, H. G.; Nahabedian, K. V. J . Am. Chem.
Soc. 1961, 83, 2167. (d) Kuivila, H. G.; Reuwer, J . F.; Mangravite, J .
A. J . Am. Chem. Soc. 1964, 86, 2666.
3486 J . Org. Chem., Vol. 69, No. 10, 2004

Catalytic Bifunctional Organoboronate Additions
use of other heteroaromatic boronic esters as well as
couplings involving acyclic vinyl boronic esters are cur-
rently underway.
acquisition of X-ray crystallographic data and Dr. Tz-
vetelina Marquardt for helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: Detailed experi-
mental procedures, as well as NMR spectra for all new
compounds, and ORTEP diagrams and X-ray crystallographic
data for compounds 29 and 34. This material is available free
of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. The Natural Sciences and En-
gineering Research Council of Canada (NSERC), the
Merck Frosst Centre for Therapeutic Research, and the
University of Toronto are gratefully acknowledged for
financial support. We thank Dr. Alan J . Lough for
J O049874K
J . Org. Chem, Vol. 69, No. 10, 2004 3487