Possibly the simplest approach to the synthesis of bicyclics is
the direct use of cycloalkenes as Diels-Alder dienophiles, but
a long-standing limitation on this methodology is the recalcitrant
dienophilicity of mono-activated cycloalkenes. The low reactivity
of cyclohexenones is notorious, though many of the problems have
been solved by employing Lewis acid catalysis.
Cycloalkenes of general structure 1 with the activating
group exocyclic to

the ring appear somewhat less reactive and there is a history
of these dienophiles either failing to react in Diels-Alder reactions
or affording low yields after prolonged heating.,
Cyclopentenes are more reactive than cyclohexenes and the best
success has been obtained either with formyl as the activating
group3, or with highly reactive dienes such
as Danishefsky's diene.-
Trivalent boron atoms are powerful activating
substituents for dienophiles, and impart a number of unusual and
useful properties to the Diels-Alder reactions of vinyl- and acetylenic
boranes., A special
feature of these reactions is the potential to control their properties
by variation of the boryl substituent. In this paper we make
use of the dibromoboryl group, apparently the most powerful of
neutral activating substituents, to carry out reactions of cyclopentenyl
and cyclohexenyl dienophiles at room temperature. These reactions
provide a simple, efficient and highly regioselective synthesis
of bridgehead bicyclic alcohols.
We had recently found that 2-alkenyl- and cis-1-alkenyldibromoboranes are highly reactive and selective dienophiles, in contrast to the low reactivity of some alkenyldialkylboranes. Our initial efforts here were directed at determining if the powerful activating effect of the dibromoboryl group would allow the reaction of the more substituted cycloalkenylborane dienophiles.
The cycloalkenylboranes 4-7 were generated in situ from
the boron-tin exchange reaction of cyclopentenyl- and cyclohexenyltributylstannanes
2 and 316 with BBr3, BCl3, or 9-bromo-9-BBN (Br-9-BBN).
A slight excess of cycloalkenylstannane was generally used to
remove adventitious protic acid which might be present in the
stock haloborane solutions; this helped to provide consistent
yields in the subsequent Diels-Alder reactions. The completion
of boron-tin exchange reactions were judged from
NMR spectra of the reaction mixtures taken immediately
at room temperature, based of the disappearance in the 1H
NMR of the vinylic peak of the cycloalkenylstannanes 2 and
3 and appearance of vinylic peaks at d
6.72, 7.21, 7.32, and 7.55 for 4-7, respectively.
No attempt was made to isolate 4-7.

In the exploratory studies, no reaction of 4 with isoprene could be observed by NMR at 55 °C. However, the reactivity of cycloalkenyldihaloboranes was striking. In contrast to the high temperatures required with the other activating groups,6,7 both 5 and 6 reacted with isoprene at 25 °C. The reaction of isoprene with the dichloroborane 5 appeared Å90% complete after 20 hours, and the reaction with the dibromoborane 6 proceeded to completion in less than three hours! For comparison, most Diels-Alder reactions of AlCl3-complexed cycloalkenones still require prolonged heating at 40 - 70 °C,1 and the reaction of 1-nitrocyclohexene with Danishefsky's diene involved heating for 13 hours in refluxing xylene.7
Our studies were focused on the reactions of the dibromoboranes 6 and 7 because of their high reactivities. The Diels-Alder reactions of 6 and 7 with several dienes were characterized by an oxidative workup of the intermediate borane adducts to afford the corresponding alcohols. Our initial attempts at oxidative workups afforded discouragingly large amounts of products apparently derived from protodeboronation of the presumed bicyclic borane intermediates. However, addition of 7 equivalents of triethylamine before proceeding with a standard H2O2/NaOH workup eliminated the protodeboronation side reaction and afforded alcohols in good to excellent yields. Our results are summarized in Table 1.

The assignment of the regiochemistry of 10-19 was complicated by the coupling of the vinylic protons to both CH2's of the cyclohexenyl ring. In addition, the relatively small vicinal coupling constants between the vinylic protons and the adjacent CH2 could not be clearly distinguished from the allylic coupling of the vinylic protons with the other CH2. Instead, the CH2's of the cyclohexenyl rings were distinguished by phase-sensitive NOESY experiments. The regiochemistry was finalized by DQCOSY or selective homonuclear decoupling experiments to determine which CH2 was coupled to the bridgehead proton. Figure 1 shows the key coupling and NOE observations used in the regiochemical assignments of 10-13, 16, and 18. In the cases of 13 and 18 no definitive NOE was observed between the vinylic protons and the adjacent CH2, but the regiochemistry could still be assigned based on an NOE between the methyl or t-butyl groups and the other CH2 group of the cyclohexene ring. The regiochemistry of 14 and 15 were assigned by analogy with 12 and 13. These analogies were supported by the observation of very similar chemical shifts and coupling patterns for the cyclohexenyl ring CH2's for 12 and 14 and for 13 and 15 respectively. The stereochemistry of 20 was assigned on the basis of NOEs between the methine proton of the cyclohexenyl ring and three protons on the cyclopentyl ring. There was also observed an NOE between the -OH and the bridgehead proton, indicative of a cis ring fusion. In all other cases a cis ring fusion was assumed. The stereochemistry of 21 was assigned on the basis of an 8.2 Hz coupling between the bridgehead proton and the methine proton on the cyclohexenyl ring.

Aside from the high reactivity of 6 and 7, the regioselectivity and endo-stereoselectivity exhibited in these reactions was striking. With the exception of the 2-t-butyl-1,3-butadiene reactions, only a single regioisomeric or stereoisomeric product alcohol could be detected in the crude product mixtures for each reaction. For comparison, uncatalyzed Diels-Alder reactions of carbonyl-activated dienophiles are normally neither very regioselective with isoprene nor very stereoselective with trans-piperylene, while Lewis acid catalyzed reactions afford fairly high (though usually not complete) selectivity. Regular cycloalkenyl dienophiles appear to follow the same trend, though product epimerization has often been a problem under Lewis acid catalyzed conditions.1
In the reactions of 6 and 7 with both 1- and 2-substituted dienes the major or exclusive product is derived from a 1,3-orientation of diene substituent and dibromoboryl group. This electronic preference for the "meta" product is an unique property of some but not all boron-activated dienophiles. Along with 6 and 7, simple 2-alkenyl- and cis-1-alkenyldihaloboranes and alkynylboranes favor the meta product, while vinyldialkylboranes, vinyldihaloboranes, and trans-1-alkenyldihaloboranes favor a normal regioselectivity.10,12
In keeping with the proposed explanation for the unusual regiochemistry with alkynylboranes (based on ab initio calculations),10 the unusual regiochemistry with 6 and 7 could be the result of a [4 + 3] transition state that looks something like 22, in which there is advanced bonding of B to C1 relative to the bonding of C5 to C4. This would result in a partial positive charge buildup on C2 and C4 which would be stabilized by R or R', favoring the observed regiochemistry. Alternatively, the unusual regiochemistry could be explained by a stepwise reaction involving the zwitterionic intermediate 23. However, this would not explain the high endo-stereoselectivity observed in these reactions.

The normally poor dienophilicity of cycloalkenes is overcome through
the use of a dibromoboryl activating group. Cycloalkenyldibromoboranes
are uniquely reactive with simple dienes at 25 °C and exhibit
exceptional selectivity among Diels-Alder reactions. Combination
of the high reactivity, regioselectivity, and endo-stereoselectivity
of these reactions with the potential versatility of the product
boranes should provide easy access to a variety of bicyclics not
otherwise available from Diels-Alder reactions.
All reactions were carried out in dry glassware under a nitrogen atmosphere using solvents dried by standard techniques. 1H NMR spectra of reaction mixtures were taken in sealed glass capillaries within NMR tubes and were referenced approximately to the peaks of the solvent hexanes or CH2Cl2. Cyclopentenyltributylstannane (2) and cyclohexenyltributylstannane (3) were prepared by literature methods.
5,6-Dimethyl-2,3,3a,4,7,7a-hexahydro-(1H)-inden-3a-ol (8): 1H NMR (CDCl3) d 2.26-2.10 (m, 2 H), 2.04 (d, J = 17 Hz, 1 H) 1.96-1.84 (m, 2 H) 1.79-1.54 (m, 11 H), 1.45 (s, 1 H), 1.27 (m, 1 H); 13C NMR (CDCl3) d 124.3, 123.1, 79.9, 44.6, 42.3, 38.4, 35.1, 30.5, 20.6, 19.1, 18.7; HRMS (EI) calcd for C11H18O 166.1358, found 166.1355.
Anal. Calcd for C11H18O: C, 79.47; H, 10.91. Found: C, 79.17; H, 10.74.
2,3-Dimethyl-1,4,4a,5,6,7,8,8a-octahydro-4a-naphthalenol (9): 1H NMR (CDCl3) d 2.48-2.32 (m, 2 H), 1.74-1.53 (m, 13H) 1.50 (m, 1 H), 1.46-1.08 (m, 4 H); 13C NMR (CDCl3) d 123.2, 121.2, 71.4, 41.4, 40.3, 38.0, 35.4, 30.2, 25.0, 24.0, 19.0, 18.8; HRMS (EI) calcd for C12H20O 180.1514, found 180.1521.
Anal. Calcd for C12H20O: C, 79.94; H, 11.18. Found: C, 79.99; H, 10.94.
2,3,3a,4,7,7a-Hexahydro-5-phenyl-(1H)-inden-3a-ol (10): 1H NMR (CDCl3) d 7.40-7.36 (m, 2 H), 7.34-1.28 (m, 2 H), 7.22 (tt, J = 7.3, 1.5 Hz, 1 H), 6.09 (m, 1 H) 2.70 (dm, J = 17.6 Hz, 1 H) 2.57 (dm, J = 17.6 Hz, 1 H), 2.48 (dm J = 17.6 Hz, 1 H), 2.06-1.94 (m, 3 H), 1.83-1.68 (m, 4 H), 1.54 (s, 1 H), 1.38 (m, 1 H); 13C NMR (CDCl3) d 141.6, 134.6, 128.3, 126.8, 125.1, 123.0, 80.0, 43.3, 38.2, 37.6, 30.4, 29.0, 20.3; HRMS (EI) calcd for C15H18O 214.1358, found 214.1356.
Anal. Calcd for C15H18O: C, 84.07; H, 8.47. Found: C, 83.86; H, 8.21.
1,4,4a,5,6,7,8,8a-Octahydro-3-phenyl-4a-naphthalenol (11): 1H NMR (CDCl3) d 7.42-7.37 (m, 2 H), 7.35-7.29 (m, 2H) 7.23 (tt, J = 7.2, 1.6 Hz, 1 H), 6.1 (m, 1 H), 2.82 (dm, J = 17.6, 1.8 Hz, 1 H), 2.69 (d of hextets, J = 19.0, 3.6 Hz, 1 H), 2.27 (d, J = 17.2 Hz, 1 H), 2.05 (dm, J = 17.2 Hz, 1 H), 1..87-1.58 (m, 6 H), 1.55-1.43 (m, 2 H), 1.41-1.21 (m, 2 H); 13C NMR (CDCl3) d 141.6, 132.3, 128.2, 126.8, 125.0, 122.1, 71.4, 40.1, 38.4, 35.6, 30.0, 29.4, 25.0, 24.0; HRMS (EI) calcd for C16H20O 228.1514, found 228.1521.
Anal. Calcd for C16H20O: C, 84.16; H, 8.83. Found: C, 83.89; H, 8.79.
2,3,3a,4,7,7a-Hexahydro-5-methyl-(1H)-inden-3a-ol (12): 1H NMR (CDCl3) d 5.31 (m, 1 H), 2.25-2.14 (m, 2 H), 2.02 (d, J = 17 Hz, 1 H), 1.94-1.56 (m, 10 H), 1.52 (s, 1 H), 1.31 (m, 1 H); 13C NMR (CDCl3) d 131.2, 119.0, 79.5, 42.7, 40.0, 37.7, 29.6, 28.1, 23.5, 19.8; HRMS (EI) calcd for C10H16O 152.1201, found 152.1186.
Anal. Calcd for C10H16O: C, 78.90; H, 10.59. Found: C, 78.54; H, 10.54.
3-Methyl-1,4,4a,5,6,7,8,8a-octahydro-4a-naphthalenol (13): 1H NMR (CDCl3) d 5.33 (m, 1 H), 2.44 (dm, J = 18 Hz, 1 H) 2.34 (d, J = 18 Hz, 1 H), 1.81-1.58 (m, 9 H), 1.58-1.44 (m, 2 H), 1.44-1.30 (m, 2 H), 1.30-1.10 (m, 2 H); 13C NMR (CDCl3) d 129.7, 118.7, 71.3, 40.4, 38.8, 38.2, 30.0, 28.8, 25.0, 24.0, 23.4; HRMS (EI) calcd for C11H18O 166.1358, found 166.1345.
Anal. Calcd for C11H18O: C, 79.47; H, 10.91. Found: C, 79.57; H, 11.22.
2,3,3a,4,7,7a-Hexahydro-5-(4-methyl-3-pentenyl)-(1H)-inden-3a-ol (14): 1H NMR (CDCl3) 5.34 (m, 1 H), 5.06 (t of quintets, J = 6.8, 1.4 Hz, 1 H), 2.28-2.15 (m, 2 H) 2..11-1.55 (m, 18 H), 1.48 (s, 1 H) 1.30 (m, 1 H); 13C NMR (CDCl3) d 135.1, 131.5, 124.1, 119.0, 79.6, 43.1, 38.3, 37.7, 37.5, 29.9, 28.1, 26.3, 25.7, 20.0, 17.7; HRMS (EI) calcd for C15H24O 220.1827, found 220.1813.
3-(4-Methyl-3-pentenyl)-1,4,4a,5,6,7,8,8a-octahydro-4a-naphthalenol (15): 1H NMR (CDCl3) d 5.35 (m, 1 H), 5.06 (t-quintet, J = 6.9, 1.4 Hz, 1 H) , 2,45 (dm, J = 20 Hz, 1 H), 2.36 (d, J = 19, 1 H) 2.12-2.03 (m, 2 H), 2.02-1.91 (m, 2 H), 1.78 (d, J = 19 Hz, 1 H), 1,73-1.45 (m, 13 H), 1.45-1.09 (m, 4 H); 13C NMR (CDCl3) d 133.6, 131.4, 124.4, 118.6, 71.3, 40.6, 38.2, 37.6, 37.2, 30.1, 28.9, 26.5, 25.6, 25.0, 24.0, 17.7; HRMS (EI) calcd for C16H26O 234.1984, found 234.1973.
5-(1,1-Dimethylethyl)-2,3,3a,4,7,7a-hexahydro-(1H)-inden-3a-ol (16) and 6-(1,1-Dimethylethyl)-2,3,3a,4,7,7a-hexahydro-(1H)-inden-3a-ol (17): Isolated as a 63:37 mixture of isomers from which samples of purified 16 could be obtained after a second chromatography: (16) 1H NMR (CDCl3) d 5.44 (m, 1 H), 2.32-2.19 (m, 2 H), 2,13 (d, J = 16 Hz, 1 H), 1.94 (m, 1 H), 1.88-1.56 (m, 6 H), 1.45 (br s, 1 H), 1.25 (m, 1 H) 1.01 (s, 9 H); (16+17) 1H NMR (CDCl3) d 5.44 (m), 5.40 (m), 2.34-2.08 (m, 3 H), 2.00-1.56 (m, 7 H), 1.50 (br s, 1 H), 1.25 (m, 1 H), 1.01 (s, 9 H); (16) 13C NMR (CDCl3) d 144.1, 116.3, 80.8, 44.1, 38.3, 35.3, 35.0, 31.3, 28.7, 21.1; (17) 13C NMR (CDCl3) d 146.3, 115.1, 80.2, 46.0, 39.2, 36.5, 31.1, 29.0, 28.2, 21.5; HRMS (EI, of mixture) calcd for C13H22O 194.1671, found 194.1674.
Anal. (of mixture) Calcd for C13H22O: C, 80.36; H, 11.41. Found: C, 80.45; H, 11.21.
3-(1,1-Dimethylethyl)-1,4,4a,5,6,7,8,8a-octahydro-4a-naphthalenol (18) and 2-(1,1-Dimethylethyl)-1,4,4a,5,6,7,8,8a-octahydro-4a-naphthalenol (19): Isolated as a 73:27 mixture of isomers from which samples of purified 18 could be obtained after a second chromatography: 1H NMR (CDCl3) d 5.41 (m, 1 H), 2.43 (dm, J = 19 Hz, 1 H), 2.35 (dm, J = 18 Hz, 1 H), 1,90-1.78 (m, 2 H), 1.74-1.48 (m, 6 H), 1.47-1.34 (m, 2 H), 1.34-1.08 (m, 2 H), 1.01 (s, 9 H); (18+19) 1H NMR (CDCl3) d 5.41 (m), 5.30 (m) 2.25-2.55 (m), 1.95-0.9 (m) (18) 13C NMR (CDCl3) d 141.3, 115.1, 71.4, 39.9, 37.6, 35.1, 33.0, 30.1, 29.1, 28.9, 24.8, 23.7; (19) 13C NMR (CDCl3) d 143.0, 113.0, 70.4, 41.4, 38.2, 33.9, 29.6, 29.3, 28.9, 27.5, 25.1, 24.1; HRMS (EI, of mixture) calcd for C14H24O 208.1827, found 208.1832.
Anal. (of mixture) Calcd for C14H24O: C, 80.71; H, 11.61. Found: C, 80.16; H, 11.55.
2,3,3a,4,7,7a-Hexahydro-7-methyl-(1H)-inden-3a-ol (20): 1H NMR (CDCl3) d 5.56 (m, 1 H), 5.47 (dq, J = 9.9, 2.2 Hz, 1 H), 2.30 (d-quintet, J = 17.1, 2.4 Hz, 1 H), 2.20 (dq, J = 17.1, 2.7 Hz , 1 H), 2.04 (m, 1 H), 1.84-1.56 (m, 5 H), 1.51-1.40 (m, 2 H), 1.33 (s, 1 H), 1.03 (d, J = 7.0 Hz, 3 H); 13C NMR (CDCl3) d 132.3, 123.3, 80.3, 51.9, 37.3, 35.6, 35.0, 29.2, 20.6, 20.5; HRMS (EI) calcd for C10H16O 152.1201, found 152.1189.
2,3,3a,4,7,7a-Hexahydro-7-phenyl-(1H)-inden-3a-ol (21): 1H NMR (CDCl3) d 7.34-7.27 (m, 2 H), 7.25-7.19 (m, 3 H), 5.78 (dm, J = 10.0 Hz, 1 H), 5.64 (dq, J = 10.0, 2.3 Hz, 1 H), 2.97 (m, 1 H), 2.49-2.34 (m, 2 H), 2.02-1.75 (m, 5 H), 1.70 (m, 1 H), 1.57 (m, 1 H), 1.39 (br s, 1 H); 13C NMR (CDCl3) d 145.2, 130.1, 128.4, 128.3, 126.3, 125.0, 80.1, 52.5, 47.3, 37.5, 35.7, 28.7, 20.2; HRMS (EI) calcd for C15H18O 214.1358, found 214.1346.
Anal. Calcd for C15H18O: C, 84.07; H, 8.47. Found: C, 84.14; H, 8.39.
Acknowledgment. We thank the Institute of General Medical Sciences of the National Institutes of Health for support of this research.
Supporting Information Available: 1H and 13C NMR spectra for 10, 12-15, 20 (12 pages). This material is contained in many libraries on microfiche, immediately follows this article in the microfilm edition of the journal, can be ordered from the ACS, and can be downloaded from the Internet; see any current masthead page for ordering information and Internet access instructions.