Organometallic enantiomeric scaffolding: General access to 2-substituted oxa- and azabicyclo[3.2.1]octenes via a Bronsted acid catalyzed [5 + 2] cycloaddition reaction

Article abstract of DOI:10.1021/ja800664v

6-Substituted TpMo(CO)₂(η-2,3,4-pyranyl)- and TpMo(CO) ₂(η-2,3,4-pyridinyl) scaffolds (Tp = hydridotrispyrazolylborato) function as reaction partners in an efficient regio- and stereocontrolled synthesis of functionalized oxa- and az

Full text of DOI:10.1021/ja800664v

Published on Web 05/15/2008
Organometallic Enantiomeric Scaffolding: General Access to
2-Substituted Oxa- and Azabicyclo[3.2.1]octenes via a
Brønsted Acid Catalyzed [5 + 2] Cycloaddition Reaction
Ethel C. Garnier and Lanny S. Liebeskind*
Emory UniVersity, Department of Chemistry, 1515 Dickey DriVe, Atlanta, Georgia 30322
Received January 31, 2008; E-mail: chemLL1@emory.edu
Abstract: 6-Substituted TpMo(CO)2(η-2,3,4-pyranyl)- and TpMo(CO)2(η-2,3,4-pyridinyl) scaffolds (Tp )
hydridotrispyrazolylborato) function as reaction partners in an efficient regio- and stereocontrolled synthesis
of functionalized oxa- and azabicyclo[3.2.1]octenes through a novel Brønsted acid catalyzed [5 + 2]
cycloaddition reaction. Excellent exo-selectivities are obtained, and the reaction gives products with complete
retention of enantiomeric purity when carried out with chiral, nonracemic scaffolds. The substituent at C-6
of the η3-coordinated heterocyclic scaffold not only influences [5 + 2] reactivity but also plays a critical role
in the demetalation step directing the reaction to only one of two possible products.
Introduction
For greatest synthetic versatility it is important to develop
practical routes to pyranyl and pyridinyl organometallic scaffolds
bearing additional substitution patterns about the heterocycle
ring. This paper describes the high yield and high enantiopurity
construction of the air-stable 2-oxo-pyranyl and -pyridinyl
scaffolds 3 and 4 and their progeny, the 6-substituted η3-pyranyl-
and η3-pyridinylmolybdenum complexes 7 and 8 (Figure 2).
The latter two systems participate in versatile enantiocontrolled
[5 + 2] cycloadditions taking place through an unusual Brønsted
acid assisted process allowing access, after a completely
regioselectiVe demetalation, to functionalized 2-substituted oxa-
and azabicyclo[3.2.1]octenes of high enantiopurity.
The oxa- and azabicyclo[3.2.1]octene skeletons are key
structural units found in a large and diverse array of
biologically interesting and medicinally important natural and
synthetic products.16–21 Structurally complex molecules pos-
sessing oxa- and azabicyclo[3.2.1]octane skeletons have been
rapidly and efficiently generated by [5 + 2] cycloadditions
of 3-oxidopyrylium22,23 and 3-oxidopyridinium dipoles.24
Although synthetically very powerful, these methods can be
Air-stable metal π-complexes such as the 3-oxopyranyl and
3-oxopyridinylmolybdenum complexes 1 and 2 shown in Figure
1 (Tp ) hydridotrispyrazolylborato) can function as versatile
organometallic enantiomeric scaffolds.1 Short, scalable, and
practicable methods of synthesis of high enantiopurity scaffolds
TpMo(CO)2(3-oxopyranyl) 1 and TpMo(CO)2(3-oxopyridinyl)
2 have been developed,2 and from these simple scaffolds a
diverse array of biologically relevant heterocyclic core structures
can be obtained in an enantiocontrolled fashion (Figure 1).3–15
The synthetic elaboration of the 3-oxopyranyl and -pyridinyl
complexes 1 and 2 often proceeds Via the 5-substituted
η3-pyranyl and -pyridinyl scaffolds 5 and 6 shown in Figure 2.
(1) Organometallic enantiomeric scaffolds: (1) the metal/ligand auxiliary
attached to the unsaturated ligand influences novel reactions, controls
non-traditional selectivities, and provides a dominant regio- and
stereocontrol element on the scaffold that allows the predictable
introduction of new stereocenters at multiple sites around the unsatur-
ated ligand, (2) a single metal/ligand auxiliary controls the introduction
of multiple stereocenters over multiple steps, and (3) the organometallic
nature of the scaffolding enables the use of stereodivergent tactics for
the introduction of ring substituents.
(2) Coombs, T. C.; Lee IV, M. D.; Wong, H.; Armstrong, M.; Cheng, B.;
Chen, W.; Moretto, A. F.; Liebeskind, L. S. J. Org. Chem. 2008, 73,
882–888.
(16) Pollini, G. P.; Benetti, S.; Risi, C. D.; Zanirato, V. Chem. ReV. 2006,
106, 2434–2454.
(17) Lounasmaa, M.; Tamminen, T. In Alkaloids; Academic Press: New
York, 1993; pp 1-114..
(3) Shu, C.; Liebeskind, L. S. J. Am. Chem. Soc. 2003, 125, 2878–2879.
(4) Shu, C.; Alcudia, A.; Yin, J.; Liebeskind, L. S. J. Am. Chem. Soc.
2001, 123, 12477–12487.
(18) Nair, V.; Treesa, P. M.; Rath, N. P.; Kunwar, A. C.; KiranKumar,
K. S.; RaviSankar, A.; Vairamani, M.; Prabhakar, S. Tetrahedron Lett.
2002, 58, 7221–7231.
(5) Yin, J.; Llorente, I.; Villanueva, L. A.; Liebeskind, L. S. J. Am. Chem.
Soc. 2000, 122, 10458–10459.
(19) Singh, S. B.; Jayasuriya, H.; Ondeyka, J. G.; Herath, K. B.; Zhang,
C.; Zink, D. L.; Tsou, N. N.; Ball, R. G.; Basilio, A.; Genilloud, O.;
Diez, M. T.; Vicente, F.; Pelaez, F.; Young, K.; Wang, J. J. Am. Chem.
Soc. 2006, 128, 11916–11920.
(6) Moretto, A. F.; Liebeskind, L. S. J. Org. Chem. 2000, 65, 7445–7455.
(7) Rubio, A.; Liebeskind, L. S. J. Am. Chem. Soc. 1993, 115, 891–901.
(8) Arraya´s, R. G.; Liebeskind, L. S. J. Am. Chem. Soc. 2003, 125, 9026–
9027.
(20) Han, L.; Huang, X.; Sattler, I.; Moellmann, U.; Fu, H.; Lin, W.;
Grabley, S. Planta Med. 2005, 71, 160–164.
(9) Arraya´s, R. G.; Liebeskind, L. S. J. Am. Chem. Soc. 2001, 123, 6185–
6186.
(21) Shen, Y.-H.; Li, S.-H.; Li, R.-T.; Han, Q.-B.; Zhao, Q.-S.; Liang, L.;
Sun, H.-D.; Lu, Y.; Cao, P.; Zheng, Q.-T. Org. Lett. 2004, 6, 1593–
1595.
(10) Arraya´s, R. G.; Yin, J.; Liebeskind, L. S. J. Am. Chem. Soc. 2007,
129, 1816–1825.
(11) Zhang, Y.; Liebeskind, L. S. J. Am. Chem. Soc. 2006, 128, 465–472.
(12) Zhang, Y.; Liebeskind, L. S. J. Am. Chem. Soc. 2005, 127, 11258–11259.
(13) Malinakova, H. C.; Liebeskind, L. S. Org. Lett. 2000, 2, 3909–3911.
(14) Malinakova, H. C.; Liebeskind, L. S. Org. Lett. 2000, 2, 4083–4086.
(15) Yin, J.; Liebeskind, L. S. J. Am. Chem. Soc. 1999, 121, 5811–5812.
(22) Wender, P. A.; Love, J. A. In AdVances in Cycloaddition; Harmata,
M., Ed.; JAI Press: Stamford, CT, 1999; pp 1-45..
(23) Chiu, P.; Lautens, M. In Topic in Current Chemistry; Metz, P., Ed.;
Springer-Verlag: Berlin, 1997; Vol. 190; pp 1-85..
(24) Katritzky, A. R.; Dennis, N. Chem. ReV. 1989, 89, 827–861.
9
10.1021/ja800664v CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 7449–7458 7449

A R T I C L E S
Garnier and Liebeskind
Figure 1. Organometallic enantiomeric scaffolding.
Scheme 1. Comparative Approaches to [5 + 2] Cycloadditions
Figure 2. Core organometallic enantiomeric scaffolds.
problematic. Issues include the limited development of
enantioselective versions,25–28 the lack of control over facial
selectivity29,30 except in the case of internal directing
groups,31–37 and the regiochemistry of the cycloaddition.38
Organometallic scaffolds of the η3-pyranyl and η3-pyridinyl
families such as 5 through 8 shown in Figure 2 offer the
opportunity for topologically similar but mechanistically
distinct [5 + 2] transformations11,13,15 as depicted compara-
tively in Scheme 1. While both of these [5 + 2] approaches
provide controlled access to complex molecular structures,
the latter organometallic protocol affords unique potential
modularity because it is not dependent on the placement of
dipole-generating functionality within the cycloaddition
precursor.
The 6-substituted η3-pyranyl- and η3-pyridinylmolybdenum
scaffolds 7 and 8 prepared in this study serve as chiral,
nonracemic substrates in [5 + 2] cycloadditions with a range
of electron-deficient alkenes, delivering focused libraries of
2-substituted oxa- and azabicyclo[3.2.1]octene ring systems in
high enantiopurity after demetalation (Scheme 2). This reaction
allows the installation of four stereocenters, including one
quaternary carbon, in one step, with total control of the regio-
and the stereochemistry of the obtained products. Additionally,
a well-established demetalation procedure39 offers varied op-
portunities for further regio- and stereoselective functionalization
of the [5 + 2] cycloadducts.
Results and Discussion
(25) Wender, P. A.; Mascaren˜as, J. L. J. Org. Chem. 1991, 56, 6267–6269.
(26) Krishna, U. M.; Trivedi, G. K. Tetrahedron Lett. 2004, 45, 257–259.
(27) Baldwin, J. E.; Mayweg, A. V. W.; Pritchard, G. J.; Adlington, R. M.
Tetrahedron Lett. 2003, 44, 4543–4545.
Synthesis of Lactonyl and Lactamyl Scaffolds 3 and 4.
Racemic lactonyl scaffold (()-3 is readily available from
5-bromo-5,6-dihydro-2H-pyran-2-one40 following an established
procedure.41 The previously disclosed route to the enantiomeri-
(28) Wender, P. A.; Lee, H. Y.; Wilhelm, R. S.; Williams, P. D. J. Am.
Chem. Soc. 1989, 111, 8954–8957.
(29) Furneaux, R. H.; Mason, J. M.; Miller, I. J. J. Chem. Soc., Perkin
Trans. 1 1984, 1923–1928.
(39) Alcudia, A.; Arraya´s, R. G.; Liebeskind, L. S. J. Org. Chem. 2002,
67, 5773–5778.
(30) Lee, H.-Y.; Sohn, J.-H.; Kim, H. Y. Tetrahedron Lett. 2001, 42, 1695–
1698.
(40) Nakagawa, M.; Saegusa, J.; Tonozuka, M.; Obi, M.; Kiuchi, M.; Hino,
T.; Ban, Y. Org. Synth. 1977, 56, 49–52
(31) Roethle, P. A.; Hernandez, P. T.; Trauner, D. Org. Lett. 2006, 8, 5901–
5904.
(32) Krishna, U. M.; Srikanth, G. S. C.; Trivedi, G. K.; Deodhar, K. D.
Synlett 2003, 2383–2385.
(33) Williams, D. R.; Benbow, J. W.; McNutt, J. G.; Allen, E. E. J. Org.
Chem. 1995, 60, 833–843.
(34) Wender, P. A.; Rice, K. D.; Schnute, M. E. J. Am. Chem. Soc. 1997,
119, 7897–7898.
(35) Krishna, U. M.; Deodhar, K. D.; Trivedi, G. K. Tetrahedron 2004,
60, 4829–4836.
(36) Wender, P. A.; Bi, F. C.; Buschmann, N.; Gosselin, N.; Kan, C.; Kee,
J.-M.; Ohmura, H. Org. Lett. 2006, 8, 5373–5376.
(37) Prakash, K. R. C.; Trzcinska, M.; Johnson, K. M.; Kozikowski, A. P.
Bioorg. Med. Chem. Lett. 2000, 10, 1443–1446.
(38) Ohmori, N. J. Chem. Soc., Perkin Trans. 1 2002, 755–767.
(41) Ward, Y. D.; Villanueva, L. A.; Allred, G. D.; Liebeskind, L. S. J. Am.
Chem. Soc. 1996, 118, 897–898.
9
7450 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008

Access to 2-Substituted Oxa-/Azabicyclo[3.2.1]octenes
A R T I C L E S
Scheme 2. 6-Substituted η3-Pyranyl and η3-Pyridinyl Scaffolds in [5 + 2] Cycloadditions
Scheme 3. Synthesis of Chiral, Nonracemic Lactonyl Scaffold (+)-3a
a (a) HClO4, Ac2O, P, Br2, 23 °C; (b) Zn, CuSO4, NaOAc, EtOAc, reflux; (c) 10 mol % InCl3, IBX, CH3CN/H2O 9:1, 23 °C; (d) (toluene)Mo(CO)3, then
KTp, CH2Cl2, 23 °C.
Scheme 4. Synthesis of Racemic Lactamyl Scaffold (()-4a
butylbenzenesulfinimidoyl chloride (PhS(Cl)dNstBu).50 Bro-
mination of 15 with NBS in refluxing carbon tetrachloride in
the presence of benzoyl peroxide followed by metalation of the
corresponding allylic bromide 16 afforded the lactamyl scaffold
(()-4 in 15% overall yield from 13 (Scheme 4).
The separate antipodes (+)-4 and (-)-451 are readily acces-
sible from racemic (()-4 by (1) removal of the N-Boc protective
group in slightly acidic media giving racemic (()-17, (2)
reprotection with a (S)-1-phenylbutyl-1H-imidazole-1-carboxy-
late auxiliary52 providing the diastereomers (+)-18 and (-)-
18′ (Scheme 5), and (3) classical resolution. Both diastereoi-
somers (+)-18 and (-)-18′ are isolable in good yield and high
enantiopurity (99.9% ee) by chromatography. If desired, these
compounds can be easily converted to the corresponding N-Boc
derivatives (+)-4 and (-)-4 by hydrogenation of the chiral
auxiliary and reprotection of the resulting secondary amine.53
a (a) n-BuLi, THF, -78 °C then Boc2O, THF, -78 °C; (b) LDA, THF,
-78 °C then PhS(Cl)dNstBu, THF, -78 °C; (c) NBS, benzoyl peroxide,
CCl4, reflux; (d) (DMF)3Mo(CO)3 then KTp, CH2Cl2, 23 °C.
cally pure version of 341 relied upon a known synthesis of allylic
acetate (S)-12.42 The preparation of (S)-12 has now been
significantly enhanced (Scheme 3). D-Arabinose 9 was first
transformed into 3,4-di-O-acetyl-D-arabinal 11 by a peracety-
lation/bromination/elimination sequence.43 An allylic rearrange-
ment/oxidation of compound 11 induced by indium chloride
and IBX44 provided the enantiopure allylic acetate (S)-1245 in
75% yield. Treatment of compound (S)-12 with 1.0 equiv of
Synthesis of 6-Substituted η3-Pyranyl- and η3-Pyridinylmo-
lybdenum Scaffolds. By mimicking a nucleophilic functionaliza-
tion-dehydration protocol used to convert glycones into
46
(toluene)Mo(CO)3 in dichloromethane at room temperature
over 4 days, followed by the addition of potassium hydridot-
ris(pyrazolyl)borate (K+TP-),47 provided complex (+)-3 in 62%
yield with excellent enantiomeric excess (97% ee) Via a retention
of the configuration pathway.41
The synthesis of racemic lactamyl scaffold (()-4 has also
been dramatically improved compared to the previously de-
scribed 11-step process.48 Racemic (()-4 is now available by
a straightforward four-step sequence. Commercially available
2-piperidone 13 is N-Boc protected in quantitative yield giving
14 (Scheme 4). Dehydrogenation of compound 14 to the
corresponding R,ꢀ-unsaturated carbonyl compound 15 was
achieved under mild conditions49 in 76% yield using N-tert-
(49) Matsuo, J.-i.; Aizawa, Y. Tetrahedron Lett. 2005, 46, 407–410.
(50) Commercially available from Tokyo Kasei Kogyo Co., Ltd (TCI). For
a reliable procedure, see: Matsuo, J.-i.; Iida, D.; Tatani, K.; Mu-
kaiyama, T. Bull. Chem. Soc. Jpn. 2002, 75, 223–234.
(51) The absolute configuration of scaffolds (+)-4 and (-)-4 was deter-
mined through the synthesis of R-(-)-coniine starting from scaffold
(-)-4.
(42) Lichtenthaler, R. W.; Ro¨nninger, S.; Jarglis, P. Liebigs Ann. Chem.
1989, 1153–1161.
(43) Areces, P.; Carrasco, E.; Mancha, A.; Plumet, J. Synthesis 2006, 946–
948.
(44) Yadav, J. S.; Subba Reddy, B. V.; Suresh Reddy, C. Tetrahedron Lett.
2004, 45, 4585–4585.
(45) The other antipode (R)-6 would be available starting from L-arabinose
following the same synthetic sequence.
(52) This reagent was obtained from condensation of (S)-1-phenylbutanol
with CDI in the presence of catalytic DMAP in dichloromethane. For
more details, see the Supporting Information.
(53) Introduction of the chiral auxiliary directly onto compound 13 was
problematic, especially at the purification stage. The sequence indicated
in the text proved more flexible and allowed isolation of both racemic
and chiral, nonracemic scaffolds.
(46) Nesmeyanov, A. N.; Krivykh, V. V.; Kaganovich, V. S.; Rybinskaya,
M. I. J. Organomet. Chem. 1975, 102, 185–193.
(47) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 3170–3171.
(48) Arraya´s, R. G.; Alcudia, A.; Liebeskind, L. S. Org. Lett. 2001, 3,
3381–3383.
9
J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008 7451

A R T I C L E S
Garnier and Liebeskind
Scheme 5. Synthesis of Chiral, Nonracemic Scaffolds (+)-4 and (-)-451a
a (a) TFA, CH2Cl2, 0 °C; (b) n-BuLi, THF, -78 °C then (S)-1-phenylbutyl-1H-imidazole-1-carboxylate, THF, -78 °C; (c) H2, 10 mol % Pd/C, MeOH,
23 °C; (d) n-BuLi, THF, -78 °C then Boc2O, THF, -78 °C.
2-substituted glycals,54,55 the lactonyl and lactamylmolybdenum
scaffolds 3 and 4 function as precursors to the synthetically
versatile 6-substituted η3-pyranyl- and η3-pyridinylmolybdenum
scaffolds 7 and 8. Both 3 and 4 are susceptible to nucleophilic
1,2-functionalization with Grignard reagents. By careful control
of the reaction temperature and time, and by using a slight
modification of Martin’s conditions for the in situ dehydration
of lactols (DMAP/TFAA),55 the undesired ring opening of the
hemiketal alkoxide intermediate can be prevented and a high-
yield dehydration ensues (see Table 1).56 The key reaction
factors are maintenance of the reaction temperature for the
Grignard reagent addition below -40 °C and initiation of the
in situ dehydration with DMAP/TFAA at -78 °C. The nucleo-
philic addition-in situ dehydration is efficient in all cases
studied, although the dehydration pathway diverges to two
different products as depicted in Table 1.
Both (()-3 and (()-4 and their high enantiopurity versions
(+)-3, (+)-4, and (-)-4 have been studied in this chemistry.
These scaffolds react with sp2 Grignard reagents (aromatic and
alkenyl) giving, after in situ dehydration, access to 6-substituted
η3-pyranylmolybdenum and η3-pyridinylmolybdenum com-
plexes 7 and 8 in good to excellent yields (Table 1, entries 1-8)
and with excellent retention of enantiopurity. However, when
these same scaffolds are treated with sp3 Grignard reagents, the
elimination proceeds exocyclic to the ring to generate 2-alky-
lidene-η-3,4,5-pyranyl- and -pyridinylmolybdenum complexes
19 and 20, respectively, as shown in Table 1 (entries 9-14). In
the specific case of the pyranylmolybdenum scaffold 3, two of
the anticipated 2-alkylidene-η-3,4,5-pyranylmolybdenum com-
plexes, 19a and 19c, are apparently very electron-rich substrates
and undergo an in situ Friedel-Crafts-like reaction with TFAA
to generate the corresponding trifluoromethylketones 19b and
19d, respectively (Table 1, entries 9 and 10). This side reaction
could not be eliminated by reducing the amount of trifluoroacetic
anhydride or by reversing the order of addition of reactants.
The Z-stereochemistry depicted for 19b and 19d was confirmed
by an X-ray diffraction study on compound 19d (see the
Supporting Information for details). In comparison to the
electron-rich and reactive alkylidene complexes generated from
lactonylmolybdenum scaffold 3, the addition of sp3 Grignard
reagents to the lactamyl scaffold 4 led to stable 2-alkylidene-
η-3,4,5-pyridinylmolybdenum complexes 20a,b (Table 1, entries
12, 13) in excellent yields. No trace of a Friedel-Crafts-like
side-reaction with TFAA was observed. Interestingly, the
2-alkylidene-η-3,4,5-dihydropyridinylmolybdenum compounds
20a and 20b were easily isomerized to their η-2,3,4-pyridinyl-
molybdenum isomers 8e and 8d, respectively. The latter product
can also be produced directly by the addition of (E)-
CH3CHdCHMgCl to the lactamyl scaffold 4 (Table 1, entry
4). The isomerization was accomplished either in refluxing
toluene over two days in 75% yield (Scheme 6) or more rapidly
and more efficiently under microwave irradiation within 30 min
(quartz reactor, 150 W, 110 °C) to give compounds 8d,e in 91
and 87% yield, respectively. Importantly, all substrates 7-8 and
19-20 are easily handled, air-stable, orange/yellow solids that
can be stored in the refrigerator for months.
η-2,3,4-Pyridinylmolybdenum complexes can therefore be
generated bearing C-sp2 and C-sp3 substituents at the 6-position,
either directly from lactamylmolybdenum complex 4 when sp2
Grignard reagents are the reactants or indirectly Via the
2-alkylidene-η-3,4,5-dihydropyridinylmolybdenum complexes
after thermal isomerization in the case of sp3 Grignard reagents.
Scheme 6. Thermal Isomerization
Brønsted Acid Promoted [5 + 2] Cycloaddition Reactions
with Compounds 7 and 8. In earlier studies13–15 it was shown
that 5-substituted-η-2,3,4-pyranyl- and -pyridinylmolybdenum
complexes were highly effective substrates in Lewis acid
catalyzed [5 + 2] cycloaddition reactions with electron-deficient
alkenes and alkynes. An exploration of the reactivity of
6-phenyl-η-2,3,4-pyranylmolybdenum complex 7a with the
electron-deficient alkenes methyl vinyl ketone (MVK) and ethyl
vinyl ketone (EVK) revealed unanticipated subtleties of the
reaction system. Of the various Lewis acids initially studied
(EtAlCl2, Et2AlCl, Sc(OTf)3, TiCl4, BF3 ·Et2O), EtAlCl2 was
able to initiate a rapid and highly efficient [5 + 2] cycloaddition
between 7a and MVK giving the expected cycloadduct 21a in
(54) Boyd, V. A.; Drake, B. E.; Sulikowski, G. A. J. Org. Chem. 1993,
58, 3191–3193.
(55) Li, H.; Procko, K.; Martin, S. F. Tetrahedron Lett. 2006, 47, 3485–
3488.
(56) Addition of Grignard reagents at-40 °C was followed by trapping of
the hemiketal at-78 °C for 1 h. The reaction mixture was then allowed
to warm over 2 h.
(57) The regiochemistry and the exo stereochemistry of cycloadduct 21a
was established unequivocally by NMR: the regiochemistry of the
reaction was determined by COSY experiment, which showed a vicinal
relationship of H6 to H7. Moreover small coupling constants were
observed for cycloadduct 21a between H7 and the two vicinal hydrogen
pairs (0 and 1 Hz), allowing us to assume an exo stereochemistry.
The exo relationships for the other examples described later in this
1
publication were readily determined by analogy from their H NMR
spectra.
9
7452 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008

Access to 2-Substituted Oxa-/Azabicyclo[3.2.1]octenes
A R T I C L E S
Table 1. Synthesis of TpMo(CO)2(6-Substituted-η3-pyranyl) and -Pyridinyl Complexes
a Yield of pure, isolated compound. b Only the Z isomer was observed. c An X-ray diffraction study conducted on compound 19d confirmed the Z
stereochemistry.
9
J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008 7453

A R T I C L E S
Garnier and Liebeskind
Scheme 7. Preliminary Observations in the [5 + 2] Cycloaddition with MVK vs EVK
Table 2. Brønsted Lewis Acid Promoted [5 + 2] Cycloadditions
an excess of the proton source effectively shuts down the
chemistry (Table 2, entry 2, footnote c). Pairing a nonprotic
coadditive with the EtAlCl2 was also ineffective (Table 2, entry
6). Therefore, the 6-phenyl-η-2,3,4-pyranylmolybdenum scaffold
requires the dual catalyst system EtAlCl2/AcOH to engage in
productive [5 + 2] cycloadditions with electron-deficient
alkenes. In actual practice stoichiometric rather than catalytic
quantities of ethylaluminum dichloride were used to increase
the reaction rate and avoid possible racemization of the scaffolds
during long reaction times.11
Collectively, these data are consistent with a strong Brønsted
acid activation of the electron-deficient alkene leading to the
formation of the cationic η4-diene adduct 22 (Scheme 8) which
is stabilized by the TpMo(CO)2 moiety. The [5 + 2] cycload-
dition is completed by internal quenching of the cationic
complex with the pendant enol delivering the exo isomer
exclusively. Although a potential pathway for endo product
formation is available,13–15 its formation was not observed.
Intermediate 22 may undergo a nonreversible ring closure
exclusively to the exo adduct, or, more likely, the exo isomer
may be the thermodynamic cycloadduct formed under equili-
brating reaction conditions.
The scope of the Brønsted acid promoted [5 + 2] cycload-
dition of molybdenum scaffolds 7a and 8a was explored with
a variety of functionalized alkenes. The results depicted in Table
3 highlight the range of electron-deficient alkenes that can
participate in this transformation (alkyl or aryl substituents at
the R- or ꢀ-carbons are tolerated). In each case single cycload-
ducts 21a-i, 23a-i were obtained in good to excellent yields
(69-98%) simply by stirring the molybdenum complexes in
dichloromethane at 0 °C in the presence of the appropriate
electron-deficient alkene that was pretreated with EtAlCl2/
AcOH.
R,ꢀ-Unsaturated ketones (Table 3, entries 1-6), N-methyl-
maleimide (Table 3, entries 7-8), and an R,ꢀ-aldehyde (Table
3, entries 11-12) gave the expected cycloadducts in good yields
from either scaffold 7a or 8a. The reaction also proceeded with
less activated substrates like ethyl acrylate (Table 3, entries 13
and 14), acrylonitrile (Table 3, entries 15 and 16), and phenyl
vinyl sulfone (Table 3, entries 17 and 18) by using longer
reaction times (3-6 h) and excess quantities of the electron-
deficient alkene (5 equiv). The ester and nitrile functional groups
of these latter products could be used as anchors for further
functionalization in potential biological applications.61,62 Most
importantly, the cycloadducts can be prepared in high enantio-
meric excess (96 to >99%). Racemization, a problem recognized
during earlier studies of [5 + 2] reactions of 5-substituted
pyranyl and pyridinylmolybdenum scaffolds,13–15 was insig-
entry
Lewis acid (1 equiv)
cocatalyst (0.7 equiv)
yield (%)a
1
2
3
4
5
6
EtAlCl2
EtAlCl2
---
EtAlCl2
EtAlCl2
EtAlCl2
---
AcOH
AcOH
HCl
0b
84c
0b
58d
79
MeOH
Ph3PdNdPPh3+Cl-
0b
a Yield of pure, isolated compound. b Starting material was entirely
recovered. c If the ratio between EtAlCl2/AcOH is <1, starting material
is entirely recovered. d Decomposition was observed with HCl as
cocatalyst.
91% yield within 30 min. Only the exo isomer of compound
21a was formed (Scheme 7).57 It was therefore totally unex-
pected that EVK proved to be completely unreactive toward
7a in the presence of EtAlCl2; the starting material was
recovered untouched (Scheme 7 and Table 2, entry 1).
The cause of this puzzling yet reproducible observation was
eventually traced to the presence of different stabilizers in
commercially available MVK and EVK (Table 2). MVK is sold
stabilized with 0.1% AcOH, while commercially available EVK
is provided with lower levels of 2,4-di-tert-butyl-4-hydroxy-
toluene (BHT) as a stabilizer. The key role played by traces of
AcOH in this chemistry was confirmed: addition of a catalytic
quantity of AcOH to a mixture of 6-phenyl-η-2,3,4-pyranyl-
molybdenum complex 7a, EVK, and EtAlCl2 initiated a rapid
reaction affording the [5 + 2] cycloadduct 21b in 84% yield
after 30 min (Table 2, entry 2). Control experiments suggested
that the [5 + 2] cycloaddition is catalyzed by an in situ generated
“super Brønsted acid” that forms upon treatment of the Lewis
acidic EtAlCl2 with a deficiency of AcOH.58,59 Neither EtAlCl2
nor acetic acid alone was effective in inducing a reaction, while
a cocatalyst generated from a 1.0:0.7 mixture of EtAlCl2 and
AcOH initiated a rapid and efficient reaction (Table 2, entries
1 and 3). An effective catalyst system was also generated by
replacing the AcOH with other protic acids such as HCl or
MeOH (Table 2, entries 4 and 5).60 The critical nature of the
ratio between ethylaluminum dichloride and acetic acid further
supports the role of a super Brønsted acid in this reaction. The
reaction proceeds when the EtAlCl2/AcOH ratio is >1.0, while
(58) Garnett, J. L.; Long, M. A.; Vining, R. F. W.; Mole, T. Tetrahedron
Lett. 1973, 4075–4078.
(61) Huscroft, I. T.; Carlson, E. J.; Chicchi, G. G.; Kurtz, M. M.; London,
C.; Raubo, P.; Wheeldom, A.; Kulagowski, J. J. Bioorg. Med. Chem.
Lett. 2006, 16, 2008–2012.
(59) Magagnini, P. L.; Cesca, S.; Giusti, P.; Priola, P.; Di Maina, M.
Makromol. Chem. 1977, 178, 2235–2248.
(60) 2,4-Di-tert-butyl-4-hydroxytoluene (BHT), the stabilizer used in
commercially available ethyl vinyl ketone, could also generate a “super
Brønsted acid” but is assumed to be too hindered to be effective.
(62) Thompson, C. G.; Carlson, E.; Chicchi, G. G.; Kulagowski, J. J.; Kurtz,
M. M.; Swain, C. J.; Tsao, K.-L. C.; Wheeldom, A. Bioorg. Med.
Chem. Lett. 2006, 16, 811–814.
9
7454 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008

Access to 2-Substituted Oxa-/Azabicyclo[3.2.1]octenes
A R T I C L E S
Scheme 8. Suggested Mechanism of the [5 + 2] Cycloaddition
Table 3. [5 + 2] Cycloadditions of 7a and 8a with Electron-Deficient Various Alkenes
entry
Z
alkene
yield (%)c
product
R1
R2
R3
R4
% ee
1
2
O
NBoc
CH2dCHCOMe
CH2dCHCOMe
98
(+)-21a
(+)-23aa
(-)-23ab
(()-21b
(()-23b
(()-21c
(+)-23ca
(()-21d
(()-23d
(()-21e
(()-23e
(()-21f
H
H
H
H
H
H
H
H
H
Ph
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
COMe
COMe
COMe
COEt
H
97
>99
>99
--
--
--
>99
--
--
--
--
--
--
96
99
74 [9]d
73 [9]d
84
H
H
H
H
H
H
H
H
H
H
Me
Me
H
H
H
H
H
H
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
O
NBoc
O
NBoc
O
NBoc
O
NBoc
O
NBoc
O
NBoc
O
NBoc
O
CH2dCHCOEt
CH2dCHCOEt
73 [7]d
81
COEt
2-cyclohexenone
2-cyclohexenone
N-methylmaleimide
N-methylmaleimide
(E)-PhCHdCHCOMe
(E)-PhCHdCHCOMe
CH2dC(Me)CHOe
CH2dC(Me)CHOe
CH2dCHCO2Etf
CH2dCHCO2Etf
CH2dCHCNf
s(CH2)3COs
s(CH2)3COs
sCON(Me)COs
sCON(Me)COs
75 [8]d
95
72 [10]d
73 [8]d
70 [15]d
88
H
H
H
H
H
H
H
H
H
H
COMe
COMe
CHO
CHO
CO2Et
CO2Et
CN
CN
SO2Ph
SO2Ph
72 [13]d
79 [9]d
78 [7]d
77 [9]d
77 [7]d
71 [20]d
69 [23]d
(()-23f
(+)-21 g
(+)-23 ga
(()-21 h
(+)-23 ha
(()-21i
CH2dCHCNf
99
--
99
CH2dCHSO2Phf
CH2dCHSO2Phf
NBoc
(+)-23ia
a From (+)-8a. b From (-)-8a. c Yield of pure, isolated compound. d Recovered starting material [%]. e NOE experiments confirmed the relative
position of R3 and R4: NOE effects were observed between the methyl group and H5 in (()-21f. The structure of (()-23f was assumed by analogy.
f Longer time of reaction (3-6 h) or larger amounts of alkene and catalysts were required. For more details, see the Supporting Information.
Table 4. Generalization of the [5 + 2] Cycloadditions to Different
6-Substituted Molybdenum Complexes in the Presence of MVK
nificant using these 6-substituted scaffolds (1-2%) and only
noted during cycloadditions of 6-substituted scaffolds with some
polar alkenes that required longer reaction times (Table 3, entry
13).
To further explore the generality of this organometallic
enantiomeric scaffold-based cycloaddition, other 6-complexes
were studied as substrates for [5 + 2] cycloaddition reactions
using MVK (Table 4). The process appears general for 6-aryl
and 6-alkyl scaffolds which reacted to give the expected products
in good yields with complete retention of the enantiomeric
excess. Unfortunately, the 6-vinyl-(η3-pyranyl- and η3-pyridi-
nylmolybdenum) scaffolds 7c and 8c led to complex reaction
mixtures (Table 4, entries 2 and 4) and were not further studied.
Demetalation and Access to 2-Substituted Oxa- and
Aza[3.2.1]octene Derivatives. The synthetic potential of this [5
+ 2] cycloaddition was realized through an efficient oxidative
demetalation using PDC/SiO2.39 This demetalation procedure
regiospecifically introduced an oxo functional group at the more
hindered terminus of the allyl moiety providing the correspond-
ing oxa- and azabicyclo[3.2.1]octenes 26-35 in good yields
(61-71%)(Table 5). Complete retention of enantiomeric purity
was observed when this demetalation protocol was applied to
chiral, nonracemic cycloadducts, highlighting the potential of
this methodology for enantiocontrolled synthesis of oxa- and
azabicyclo[3.2.1]octane derivatives.
entry compound
Z
R
product
yield (%)a
% ee
1
2
(+)-7b
(()-7c
O
O
s(m-MeO)C6H4 (+)-24a 93
97
--
sCHdCH2
(()-24b complex
mixture
(()-24c 81
3
4
5
(()-7ec
(-)-8b
(()-8c
O
sEt
--
>99
--
NBoc s(m-MeO)C6H4 (-)-25a 88 [5]b
NBoc sCHdCH2
(()-25b complex
mixture
6
(()-8e
NBoc sMe
(()-25c 92
--
a Yield of pure, isolated compound. b Recovered SM (%).
c Compound 7e is derived from compound 7c by hydrogenation of the
vinyl side chain. For more details, see Supporting Information.
The structure assigned to the product of oxidative demeta-
1
lation was determined by studying the H NMR spectrum of
(+)-26 (Figure 3). For cycloadduct 26 the enone protons H4
and H5 are readily apparent and assigned on the basis of
chemical shifts and their coupling to each other and to the allylic
9
J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008 7455

A R T I C L E S
Garnier and Liebeskind
Table 5. Demetalation Protocol with PDC/SiO2
a Yield of pure, isolated compound.
9
7456 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008

Access to 2-Substituted Oxa-/Azabicyclo[3.2.1]octenes
A R T I C L E S
Figure 3. NOE Interactions in Cycloadduct (+)-26
Scheme 9. Mechanistic Proposal for the Regiospecific Oxidative Demetalation
hydrogen H6. Consistent with earlier studies of exosubstituted
oxa- and azabicyclo[3.2.1]octenes the two methine hydrogens,
H6 and H7, do not couple to each other.15 NOE experiments
showed correlations between H6 and H7, but no correlation was
observed between the phenyl ring hydrogens and H5, the enone
ꢀ-H. These data are fully consistent with formation of regioi-
somer (+)-26 and not 26′ in the oxidative demetalation. The
structures of the other oxidative demetalation products were
assigned by analogy.
The formation of enone regioisomer 26 from the oxidative
demetalation can be rationalized by a mechanism that begins
with removal of one electron from the TpMo(CO)2(allyl)
complex generating the cationic allylmolybdenum species 36
(Scheme 9). An incoming water nucleophile would likely attack
the cationic allyl 36 anti to the TpMo(CO)2 moiety and at one
of the two allylic termini. Although the η3-allyl terminus
adjacent to the quaternary carbon of the [5 + 2] adduct clearly
presents a more hindered trajectory for attack by an incoming
nucleophile, the immediate product of that nucleophilic addition
of water is an η2-alkene complexed to TpMo(CO)2 (structure
38 in Scheme 9). The stability of the “product” (the η2-alkene
complex) must therefore control the regiochemical outcome of
the oxidative demetalation and lead to the less sterically
encumbered η2-alkene complex 37. This produces the allylic
alcohol bearing its hydroxyl substituent adjacent to the hindered
quaternary carbon; it is oxidized to the observed enone 26 by
excess PDC.
resent an important class of tropane derivatives that have been
studied as CNS penetrant hNK1 receptor antagonists.61,62
Oxabicyclo[3.2.1]octanes have also been used as versatile
building blocks in organic synthesis,23 which has spurred the
development of enantiocontrolled routes to these bicycles.63
Experimental Section
Representative experimental procedures are listed here. Full
experimental details for all reactions and characterization data for
all compounds can be found within the Supporting Information.
(-)-Dicarbonyl[hydridotris(1-pyrazolyl)borato][(η-2,3,4)-1-
(tert-butyloxycarbonyl)-6-phenyl-1,2-dihydropyridin-2-yl]moly-
bdenum, (+)-8a. Under argon, at -78 °C, to a suspension of a
99.9% ee sample of the molybdenum complex (+)-4 (0.100 g, 0.18
mmol) in dry THF (1.8 mL) was added dropwise a 0.3 M solution
of phenyl magnesium bromide (650 µL, 0.20 mmol, 1.1 equiv).
After 5 min, the reaction mixture was allowed to warm up at -40
°C and was stirred for 1 h. After cooling down at -78 °C,
4-dimethylaminopyridine DMAP (0.025 g, 0.20 mmol, 1.15 equiv)
and trifluoroacetic anhydride TFAA (62 µL, 0.44 mmol, 2.5 equiv)
were sequentially added, and the mixture was stirred for 1 h at
-78 °C, then slowly warmed up at room temperature over 1 h,
and stirred for further 1 h. The reaction was hydrolyzed at 0 °C by
addition of ice and extracted twice with ethyl acetate and brine.
The organic layers were combined, dried over magnesium sulfate,
filtrated, and concentrated under vacuum. The crude product was
purified by chromatography (silica gel, hexanes/EtOAc 2:1 v/v) to
20
afford compound (+)-8a {91% yield and 99.7% ee, [R]D +278
(c 0.32, CH2Cl2)} as a yellow powder. TLC (n-hexanes/CH2Cl2
2:1) Rf ) 0.25; mp ) 204-208 °C with decomp.; IR (cm-1): 3127
(w), 3057 (w), 2980 (w), 2934 (w), 1930 (s), 1849 (s), 1702(m);
1H NMR (CDCl3, 300 MHz): δ 8.52 (d, J ) 2.0 Hz, 1H), 8.23 (d,
J ) 2.0 Hz, 1H), 7.71 (d, J ) 2.0 Hz, 1H), 7.64 (dd, J ) 6.0, 2.4
Hz, 1H), 7.62 (d, J ) 2.0 Hz, 1H), 7.61 (d, J ) 2.0 Hz, 1H), 7.48
(d, J ) 2.0 Hz, 1H), 7.28-7.20 (m, 3H), 7.17-7.12 (m, 2H), 6.28
(app t, J ) 2.0 Hz, 1H), 6.21 (app t, J ) 2.0 Hz, 1H), 6.18 (app t,
J ) 2.0 Hz, 1H), 5.90 (d, J ) 6.0 Hz, 1H), 4.96 (td, J ) 6.0, 2.4
Hz, 1H), 2.84 (dd, J ) 6.0, 1.2 Hz, 1H), 1.11 (s, 9H); 13C NMR
(CDCl3, 100 MHz): δ 231.6, 223.8, 157.2, 153.2, 146.1, 144.9,
140.3, 139.9, 136.2, 136.0, 134.6, 131.4, 129.0, 127.7, 127.5, 127.4,
126.8, 118.0, 106.0, 105.6, 93.7, 82.6, 62.2, 53.1, 27.6 (*3); (+)-
8a: HRMS-FAB (m/z): [M]+ calcd for C27H28BMoN7O4, 623.1350;
found, 623.1354.
(+)-Dicarbonyl[hydridotris(1-pyrazolyl)borato][η-(2,3,4)-
(1S,2R,5S,6R)-8-aza-8-(tert-butyloxycarbonyl)-6-ethoxycarbonyl-
1-phenylbicyclo[3.2.1]oct-3-en-2-yl]molybdenum, (+)-23g. Under
argon, at 0 °C, to a solution of compound (+)-8a (0.125 g, 0.20
mmol) in dry dichloromethane (1.6 mL) were sequentially added
ethyl acrylate (120 µL, 1.21 mmol, 6 equiv) followed by a premixed
Conclusions
Efficient and scalable syntheses of the molybdenum π-com-
plex scaffolds (+)-3, racemic (()-4, and chiral, nonracemic
(+)-4 and (-)-4 are reported. Grignard reagents add to these
substrates in excellent yields (72-97%) to provide 6-substituted
TpMo(CO)2(η-2,3,4-pyranyl) and TpMo(CO)2(η-2,3,4-pyridi-
nyl) complexes 7 and 8 with complete retention of enantiomeric
purity. By way of a novel and versatile Brønsted acid mediated
[5 + 2] cycloaddition, scaffolds 7 and 8 generate functionalized
oxa- and azabicyclo[3.2.1]octenes in a regio- and enantiocon-
trolled fashion. The reaction proceeds in moderate to excellent
yields (69-98%), with excellent exo-selectivity, and affords
products 21/23a-i and 24/25a-c with complete retention of
enantiomeric purity. Finally, highly functionalized 6-substituted
oxa- and azabicyclo[3.2.1]octane 26-35 derivatives were gener-
ated in good yields (61-71%) and high enantiopurity (96-99%
ee) by applying a very regioselective oxidative demetalation
protocol. 6-Substituted-1-aryl-8-azabicyclo[3.2.1]octenes rep-
9
J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008 7457

A R T I C L E S
Garnier and Liebeskind
solution of a 1.0 M solution in dichloromethane of ethylaluminum
dichloride (201 µL, 0.20 mmol, 1 equiv) and glacial acetic acid (8
µL, 0.14 mmol, 0.7 equiv). After 4 h at 25 °C, the reaction mixture
was hydrolyzed with ice at 0 °C and extracted with dichloromethane
(2 × 5 mL). The combined organic layers were dried over
magnesium sulfate, filtered and evaporated under reduced pressure.
The crude product was purified by flash chromatography (SiO2,
hexanes/EtOAc 2:1 v/v) to afford after evaporation exo-(+)-23g
in 78% yield and 99.6% ee {[R]D20 +284 (c 0.16, CH2Cl2)} as an
orange solid. TLC (n-hexanes/EtOAc 2:1) Rf ) 0.25; mp )
193-194 °C with decomp.; IR (cm-1): 3120 (w), 2951 (w), 1933
(s), 1847 (s), 1732 (m), 1708 (m); 1H NMR (CDCl3, 300 MHz): δ
8.42 (d, J ) 1.8 Hz, 1H), 7.70 (dd, J ) 1.8, 2.1 Hz, 1H), 7.61 (d,
J ) 2.1 Hz, 1H), 7.59 (d, J ) 2.1 Hz, 1H), 7.53 (d, J ) 2.1 Hz,
1H), 7.48 (d, J ) 2.1 Hz, 1H), 7.30-7.18 (m, 5H), 6.23 (app t, J
) 2.4 Hz, 1H), 6.19 (app t, J ) 2.1 Hz, 1H), 6.16 (app t, J ) 2.1
Hz, 1H), 4.46 (d, J ) 5.8 Hz, 1H), 4.39 (d, J ) 6.8 Hz, 1H), 4.32
(app t, J ) 6.6 Hz, 1H), 4.13 (d, J ) 3.0 Hz, 1H), 3.75 (m, 1H),
3.63 (q, J ) 12.0 Hz, 2H), 2.40 (m, 1H), 1.26 (t, J ) 12.0 Hz,
3H), 1.11 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 226.4, 224.3,
173.1, 157.7, 146.3, 145.9, 139.4, 137.7, 136.6, 135.7, 134.0, 128.7
(*2), 128.2 (*2), 125.9, 105.9, 105.7, 105.2, 79.8, 76.1, 65.2, 61.9,
61.6, 50.1, 44.9, 42.4, 32.4, 28.5 (*3), 14.1; HRMS-FAB (m/z):
[M + H]+ calcd for C32H36BMoN7O6, 723.1874; found, 723.1869.
(-)-(1S,2R,5S,6R)-11-aza-11-(tert-butyloxycarbonyl)-6-ethy-
loxycarbonyl-1-phenylbicyclo[3.2.1]oct-3-ene-2-one, (-)-30. To
a solution of a 99.7% ee sample of compound (+)-23g (0.144 g,
0.20 mmol) in dichloromethane (2 mL) was added a mixture of
PDC (0.240 g, 0.64 mmol, 3.2 equiv) and silica gel (0.240 g) in
one portion as a solid. The mixture was stirred at room temperature
overnight. It was then passed through a short pad of Celite and
eluted with dichloromethane/ethyl acetate 2:1. After concentration,
the residue was purified by flash chromatography to afford (-)-30
20
in 68% yield and 99.6% ee {[R]D -81 (c 0.16, CH2Cl2)} as a
white solid. TLC (n-hexanes/EtOAc 1:1) Rf ) 0.27; mp ) 105-106
°C with decomp.; IR (cm-1): 2956 (w), 1731 (s), 1715 (s), 1202
1
(m); H NMR (CDCl3, 300 MHz): δ 7.72 (dd, J ) 9.8, 5.7 Hz,
1H), 7.31-7.22 (m, 5H), 6.77 (d, J ) 9.8 Hz, 1H), 5.26 (d, J )
5.7 Hz, 1H), 4.12 (q, J ) 7.1 Hz, 2H), 2.95 (dd, J ) 3.5, 9.2 Hz,
1H), 2.92-2.83 (m, 1H), 2.17 (dd, J ) 9.2, 14.0 Hz, 1H), 1.95 (s,
9H), 1.30 (t, J ) 7.1 Hz, 3H); 13C NMR (CDCl3, 100 MHz): δ
197.6, 173.1, 157.7, 147.8, 144.2, 128.6 (*2), 128.2 (*2), 128.1,
126.0, 79.8, 76.7, 61.6, 45.9, 37.0, 36.5, 28.5 (*3), 14.1; HRMS-
FAB (m/z): [M]+ calcd for C21H25NO5, 371.1733; found, 371.1745.
Acknowledgment. This work was supported by Grant GM
522238, awarded by the National Institute of General Medical
Sciences, Department of Health and Human Services. We are most
grateful to our colleague Dr. Kenneth Hardcastle for his skilled
and efficient assistance with X-Ray crystallography and to Dr. Pilar
Areces of the Universidad de Extremadura, Badajoz, Spain for
providing unpublished procedures and copies of spectra for
compounds 10 and 11 that are mentioned in ref 43. We thank
colleague Minkoo Ji for carrying out preliminary experiments
confirming the thermal isomerization of exocyclic 2-alkylidene-η-
3,4,5-dihydropyridinylmolybdenum complexes to their endocyclic
η-2,3,4-pyridinylmolybdenum isomers.
Supporting Information Available: Full experimental details
and characterization data for all compounds; copies of proton
and carbon NMR spectra of all new compounds prepared in
this study. Full X-ray crystallography data for compound 19d.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(63) Lo´pez, F.; Castedo, L.; Mascaren˜as, J. L. Org. Lett. 2000, 2, 1005–
1007.
JA800664V
9
7458 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008