Mechanistic Implications of the Stereochemistry of the Cation Radical Diels-Alder Cycloaddition of 4-(cis-2-Deuteriovinyl)anisole to 1,3-Cyclopentadiene

Full text of DOI:10.1021/jo005530s

6276
J . Org. Chem. 2000, 65, 6276-6277
Sch em e 1
Mech a n istic Im p lica tion s of th e
Ster eoch em istr y of th e Ca tion Ra d ica l
Diels-Ald er Cycloa d d ition of
4-(cis-2-Deu ter iovin yl)a n isole to
1,3-Cyclop en ta d ien e
Daxin Gao and Nathan L. Bauld*
Department of Chemistry and Biochemistry,
The University of Texas at Austin, Austin, Texas 78712
Sch em e 2
bauld@mail.utexas.edu
Received May 30, 2000
The cation radical Diels-Alder cycloaddition of trans-
anethole to 1,3-cyclopentadiene, catalyzed by tris(4-
bromophenyl)aminium hexachloroantimonate, yields both
the trans-endo and trans-exo products, but no traces of
cis adducts are found (<0.1%).1 In sharp contrast, the
addition of cis-anethole to cyclopentadiene is non-
stereospecific, forming all four Diels-Alder cycloadducts
in substantial amounts. These contrasting results are
especially dramatic because of the very high degree of
suprafacial stereoselectivity established in the cycload-
dition of trans-anethole. Since cis-anethole obviously adds
via a stepwise process, it is natural to assume that trans-
anethole does likewise. However, if this is the case it is
necessary to postulate that the cyclization of the distonic
cation radical intermediate is at least 1000 times as rapid
as bond rotation. Presumably, bond rotation in the trans
isomer could be appreciably slowed by the steric hin-
drance encountered by the anisyl substituent as it rotates
into the less stable cisoid conformations, but the existence
of a cyclization step that is so much faster than bond
rotation would be at least somewhat surprising and
would constitute an important precedent in cation radical
cycloaddition chemistry. An alternate, but admittedly less
likely, possibility is that while the cis isomer reacts in
stepwise fasion, the trans isomer reacts via a concerted
mechanism; i.e., there is a mechanistic change between
these two geometric isomers.
and NOESY spectra of both pure isomers were examined,
and it was established that the endo and exo protons at
C6 were well resolved from each other and from other
absorptions for both isomers. The deuterated substrate
was then prepared from 4-ethynylanisole as shown in
Scheme 1.
The results of the reaction of the deuterated substrate
with cyclopentadiene are shown in Scheme 2. The reac-
tion is significantly nonstereospecific in the case of both
adduct diastereoisomers. Although the main cycloaddi-
tion stereochemistry involved in the formation of the endo
isomer is the expected suprafacial addition, the exo
isomer is formed predominantly by an antarafacial path.
This latter observation, although initially somewhat
surprising, is consistent with the proposal that the exo
adduct is predominantly formed via an initial approach
from the endo face, followed by rotation to give the exo
product. Precedent for a preferred endo approach mode
has previously been established.3
These results clearly establish that the addition of the
vinylanisole cation radical to cyclopentadiene, like that
of the cis-anethole cation radical, is a stepwise process,
and they thus strongly suggest that the addition of trans-
anethole is also stepwise, even though both the endo and
exo adducts in the latter case are formed with complete
(>99.9%) suprafacial stereoselectivity. The list of non-
stereospecific and thus obviously stepwise cation radical
Diels-Alder cycoadditions thus continues to grow longer.3,4
These results further emphasize that, especially in the
case of cation radical Diels-Alder cycloadditions, even
very high reaction stereoselectivity may not necessarily
be a strong indicator of reaction concertedness. However,
the uncoupling of concertedness and high reaction ste-
reospecificity (highly stereoselective addition of both cis
and trans dienophile geometric isomers), although also
theoretically possible, still has not been observed and
appears unlikely. It is therefore appropriate to note that
Resu lts a n d Discu ssion
To distinguish between the two previously mentioned
mechanistic possibilities, and specifically to remove the
effect of the methyl substituent upon the rate of bond
rotation, the stereochemistry of the cation radical Diels-
Alder cycloaddition of 4-(cis-2-deuteriovinyl)anisole (1.+)
to 1,3-cyclopentadiene was studied. Since the reaction of
4-vinylanisole with cyclopentadiene had not previously
been reported, we first carried out the reaction of this
substrate with an excess (10-fold) of 1,3-cyclopentadiene
in the presence of tris(4-bromophenyl)aminium hexachlo-
roantimonate (2.+) in dichloromethane solution at 0 °C.2
The reaction afforded a mixture of endo and exo Diels-
Alder adducts (3; endo/exo )3:1), which could be sepa-
1
rated by column chromatography. The H NMR, COSY,
* To whom correspondence should be addressed. Phone: (512) 471-
3017. Fax: (512) 471-8696.
(1) Bauld, N. L.; Gao, D. J . Chem. Soc., Perkin Trans. 2 2000, 931-
934.
(3) Bauld, N. L.; Yang, J .; Gao, D. J . Am. Chem. Soc. 2000, 207-
210.
(2) Walter, R. I. J . Am. Chem. Soc. 1955, 77, 5999-6005.
(4) Bauld, N. L.; Yang, J . Tetrahedron Lett. 1999, 8519-8522.
10.1021/jo005530s CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/24/2000

Notes
J . Org. Chem., Vol. 65, No. 19, 2000 6277
DA adducts (3:1) was obtained (313 mg, 70%). Endo adduct
(3a ): 1H NMR (500 MHz, CDCl3) δ 1.21 (1H, m, H6 endo), 1.42
(1H, m, H7 anti), 1.50 (1H, m, H7 syn), 2.18 (1H, m, H6 exo),
2.91 (1H, s, H1), 3.02 (1H, s, H4), 3.32 (1H, m, H5), 3.78 (3H, s,
OCH3), 5.75 (1H, m, H3), 6.22 (1H, m, H2), 6.76 (2H, d, J )
the cation radical Diels-Alder cycloadditions of the three
geometric isomers of 2,4-hexadiene to cyclohexadiene are
highly stereospecific5 as are those of the cis,trans isomers
of 1,2-diaryloxyethenes to 1,3-cyclopentadiene.6
8.63 Hz), 7.03 (2H, d, J ) 8.23 Hz); HRMS calcd for C14H17
O
201.127940, found 201.127755. Exo adduct (3b): 1H NMR (500
MHz, CDCl3) δ 1.40 (1H, m, H7 anti), 1.57 (1H, m, H7 syn), 1.61
(1H, m, H6 endo), 1.70 (1H, m, H6 exo), 2.65 (1H, dd, J ) 8.63,
4.62 Hz, H5), 2.82 (1H, s, H4), 2.97 (1H, s, H1), 3.79 (3H, s,
OCH3), 6.14 (1H, dd, J ) 5.62, 2.81 Hz, H2), 6.23 (1H, dd, J )
5.62, 3.01 Hz, H3), 6.82 (2H, d, J ) 8.83 Hz), 7.18 (2H, dq, J )
8.23, 0.6 Hz); HRMS calcd for C14H17O 201.127940, found
201.127745.
Exp er im en ta l Section
Ch em ica ls a n d Solven ts. All chemicals used as starting
materials were purchased from the Aldrich Company and used
as received unless otherwise specified. The dichloromethane
solvent was dried by refluxing it over calcium hydride. The
catalyst, tris(4-bromophenyl)aminium hexachloroantimonate,
was synthesized according to the literature procedure.2
Syn th esis of 1-(2′-Deu ter o)eth yn yl-4-m eth oxyben zen e.
To 1.32 g (0.01mol) of 1-ethynyl-4-methoxybenzene7,8 dissolved
in 20 mL of dry THF at -78 °C was added 7.54 mL (0.012 mol,
1.6 M in hexane) of n-BuLi. The mixture was stirred for 18 min
and then quenched with 2.5 mL of D2O at -78 °C. The usual
two-phase water/dichloromethane workup yielded 1.33 g (99%)
Rea ction of 4-(cis-2′-Deu ter iovin yl)a n isole w ith Cyclo-
p en ta d ien e. To tris(4-bromophenyl)aminium hexachloroanti-
monate (326.6 mg; 0.4 mmol) in 20 mL of dichloromethane and
5 mL of H2O at 0 °C was added a solution of 300 mg (2.24 mmol)
of 4-(cis-2′-deuteriovinyl)anisole and 1.76 g (26.7 mmol) of 1,3-
cyclopentadiene in 5 mL of dichloromethane in one portion. The
reaction mixture was then quenched with saturated methanolic
potassium carbonate after 1.5 min. After aqueous workup, the
adducts were purified by column chromatography (silica gel;
hexanes/dichloromethane ) 10:1), giving 285 mg (63.8% yield)
of the endo and exo DA adducts. Pure fractions of both adducts
were obtained by this means. Endo adduct: 1H NMR (500 MHz,
CDCl3) δ 1.21 (21% H, m, H6 endo), 1.42 (1H, m, H7 anti), 1.50
(1H, m, H7 syn), 2.18 (79% H, m, H6 exo), 2.91 (1H, s, H1), 3.02
(1H, s, H4), 3.30 (1H, dd, J ) 9.5, 3.5 Hz, H5), 3.78 (3H, s, OCH3),
5.76 (1H, dd, J ) 5.75, 3 Hz, H3), 6.22 (1H, dd, J ) 5.5, 3 Hz,
H2), 6.76 (2H, d, J ) 8.83 Hz), 7.03 (2H, dd, J ) 8.83, 0.6 Hz);
HRMS calcd for C14H15DO 202.135765, found 202.134651. Exo
adduct: 1H NMR (500 MHz, CDCl3) δ 1.40 (1H, m, H7 anti),
1.57 (1H, m, H7 syn), 1.60 (78% H, m, H6 endo), 1.70 (22% H,
m, H6 exo), 2.62 (1H, d, J ) 8.5 Hz, H5), 2.81 (1H, s, H4), 2.92
(1H, s, H1), 3.79 (3H, s, OCH3), 6.12 (1H, dd, J ) 5.5, 3 Hz, H2),
6.22 (1H, dd, J ) 5.5, 3 Hz, H3), 6.82 (2H, d, J ) 8.5 Hz), 7.19
(2H, d, J ) 8.53 Hz); HRMS calcd for C14H15DO 202.135765,
found 202.134655.
1
of the pure product: H NMR (250 MHz, CDCl3) δ 3.75 (s, 3H),
6.82 (2H, dd, J ) 8.82, 2.05 Hz), 7.40 (2H, dt, J ) 8.87, 2.07
Hz); HRMS calcd for C9H7DO 134.071617, found 134.071724.
Syn th esis of cis-2-Deu ter iovin yla n isole (1). To 100 mg
(0.75 mmol) of 1-(2′-deutero)ethynyl-4-methoxybenzene of dis-
solved in 8 mL of CH3OD was added 5 mg of Lindlar catalyst
and 50 mg of quinoline. Hydrogen gas contained in a balloon
was bubbled into the reaction mixture at room temprature for
about 16 min (monitored by GC). Workup gave the pure product
(100 mg; 98.5% yield), which was found by NMR to be at least
98.8% cis-2-deuteriovinylanisole: 1H NMR (250 MHz, CDCl3) δ
3.8 (3H, s), 5.13 (1H, dd, J ) 10.97, 3.25 Hz), 6.66 (1H, m), 6.84
(2H, d, J ) 8.56 Hz), 7.32 (2H, d, J ) 8.74 Hz); HRMS calcd for
C9H9DO 1.36.087267, found 136.087118.
Rea ction of 4-Vin yla n isole w ith 1,3-Cyclop en ta d ien e. To
tris(4-bromophenyl)aminium hexachloroantimonate (274 mg;
0.335 mmol) dissolved in 20 mL of dichloromethane and 5 mL
of water at 0 °C was added a solution of 300 mg (2.24 mmol) of
4-vinylanisole and 1.48 g (22.4 mmol) of 1,3-cyclopentadiene
dissolved in 5 mL of dichloromethane, all in one portion. After
1.5 min, the reaction mixture was quenched with saturated
methanolic potassium carbonate. After aqueous workup and
purification by column chromatography (silica gel; elution with
hexanes/dichloromethane ) 10:1), the mixture of endo and exo
Ack n ow led gm en t. We wish to acknowledge the
support of this research by the National Science Foun-
dation (CHE-9610227) and the Robert A. Welch Foun-
dation (F-149).
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of
obtained compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(5) Bauld, N. L.; Wirth, D. D.; Bellville, N. L. J . Am. Chem. Soc.
1981, 718-720.
(6) Bauld, N. L.; Yang, J . Org. Lett. 1999, 1, 773-774.
(7) Okamoto, Y.; Kundu, S. K. J . Org. Chem. 1970, 35, 4250-4254.
(8) Kunckell, F.; Eras, K. Chem. Ber. 1903, 36, 915-918.
J O005530S