Azidomercurations of alkenes: Mercury-promoted Schmidt reactions

Article abstract of DOI:10.1021/jo001181q

Azides bearing a suitably disposed alkene, when treated with either mercuric perchlorate or mercuric trifluoromethanesulfonate, produce bicyclic iminium ions. This new version of the Schmidt reaction proceeds by capture of the mercuronium ion intermediate by the azide to produce an aminodiazonium ion, which suffers a 1,2-shift to give an iminium ion (e.g., 10 → 16 → 17 → 18). Reduction of the iminium ion may then be carried out to produce an amine. Compared to earlier work on the protic acid-promoted intramolecular Schmidt reaction of azido-alkenes, the mercury-promoted Schmidt reaction has several advantages. First, the acid-promoted Schmidt reaction of azido-alkenes requires that the intermediate carbocations be tertiary, allylic, benzylic, or propargylic. The mercury-promoted method has no such limitation; thus even 1,2-disubstituted alkenes may be used. Second, the mercury-promoted method is milder, allowing the presence of acid-sensitive functionality. The protic version, typically employing trifluoromethanesulfonic acid, is limited in its functional group tolerance. Third, whereas carbocation rearrangement is often observed prior to cyclization/rearrangement in the acid-promoted Schmidt reaction, the mercury-promoted method avoids this problem. Fourth, the presence of the mercurio group during the rearrangment may alter the regioselectivity of the 1,2-migration. Finally, the mercury-bearing iminium ions that are the result of the Schmidt reaction were found to be sensitive to protodemercuration, precluding their use in other transformations.

Full text of DOI:10.1021/jo001181q

8326
J . Org. Chem. 2000, 65, 8326-8332
Azid om er cu r a tion s of Alk en es: Mer cu r y-P r om oted Sch m id t
Rea ction s
William H. Pearson,* Daniel A. Hutta,† and Wen-kui Fang‡
Department of Chemistry, University of Michigan, Ann Arbor, Michigan, USA 48109-1055
wpearson@umich.edu
Received August 4, 2000
Azides bearing a suitably disposed alkene, when treated with either mercuric perchlorate or mercuric
trifluoromethanesulfonate, produce bicyclic iminium ions. This new version of the Schmidt reaction
proceeds by capture of the mercuronium ion intermediate by the azide to produce an aminodiazo-
nium ion, which suffers a 1,2-shift to give an iminium ion (e.g., 10 f 16 f 17 f 18). Reduction of
the iminium ion may then be carried out to produce an amine. Compared to earlier work on the
protic acid-promoted intramolecular Schmidt reaction of azido-alkenes, the mercury-promoted
Schmidt reaction has several advantages. First, the acid-promoted Schmidt reaction of azido-alkenes
requires that the intermediate carbocations be tertiary, allylic, benzylic, or propargylic. The mercury-
promoted method has no such limitation; thus even 1,2-disubstituted alkenes may be used. Second,
the mercury-promoted method is milder, allowing the presence of acid-sensitive functionality. The
protic version, typically employing trifluoromethanesulfonic acid, is limited in its functional group
tolerance. Third, whereas carbocation rearrangement is often observed prior to cyclization/
rearrangement in the acid-promoted Schmidt reaction, the mercury-promoted method avoids this
problem. Fourth, the presence of the mercurio group during the rearrangment may alter the
regioselectivity of the 1,2-migration. Finally, the mercury-bearing iminium ions that are the result
of the Schmidt reaction were found to be sensitive to protodemercuration, precluding their use in
other transformations.
Sch em e 1. Sch m id t Rea ction P a th w a ys
In tr od u ction
We have previously shown that the Schmidt reaction
of hydrazoic acid (HN3) with carbocations can be extended
to the use of aliphatic azides (RN3) in both the inter- and
intramolecular modes (Scheme 1, 2 f 3 f 4, intramo-
lecular mode shown).1,2 Thus, formation of a carbocation
2 from an alcohol 1 or alkene 6 followed by capture by
an aliphatic azide generates a secondary aminodiazonium
ion 3. Bond migration to the electron-deficient nitrogen
atom results in the formation of an iminium ion 4, which
may be reduced to the amine 5 with sodium borohydride.
Other 1,2-migrations of 3 are also possible. Our previous
work was limited to benzylic, allylic, or propargylic
carbocations; less stabilized cations either failed to
produce Schmidt products or suffered rearrangement to
a more stable cation prior to the Schmidt reaction.
Furthermore, acid sensitive functionality was not toler-
ated under the protic or Lewis acidic reaction conditions.
We now wish to report the mercury-promoted Schmidt
reaction of aliphatic azides with alkenes [Scheme 1, 6 f
7 f 8 f 9 (f 5)], a process that avoids some of the
limitations of the carbocation method. These are the first
examples of azidomercuration reactions involving ali-
phatic azides.3,4
† Current address: Provid Research, 10 Knightsbridge Rd., Piscat-
away, New J ersey, 08854.
‡ Current address: Allergan, Inc., Mail Code RD-3D, 2525 Dupont
Drive, Irvine, California, 92612.
(1) (a) Pearson, W. H.; Schkeryantz, J . M. Tetrahedron Lett. 1992,
33, 5291-5294. (b) Pearson, W. H.; Walavalkar, R.; Schkeryantz, J .
M.; Fang, W.-k.; Blickensdorf, J . D. J . Am. Chem. Soc. 1993, 115,
10183-10194. (c) Pearson, W. H.; Fang, W.-k.; Kampf, J . W. J . Org.
Chem. 1994, 59, 2682-2684. (d) Pearson, W. H.; Fang, W.-k. J . Org.
Chem. 1995, 60, 4960-4961. (e) Pearson, W. H. J . Heterocycl. Chem.
1996, 33, 1489-1496. (f) Pearson, W. H.; Fang, W.-k. Isr. J . Chem.
1997, 37, 39-46.
(2) Aube´ has shown that aliphatic azides may participate in Schmidt
reactions with ketones and related electrophiles. See: (a) Aube´, J .;
Milligan, G. L. J . Am. Chem. Soc. 1991, 113, 8965-8966. (b) Aube´, J .;
Milligan, G. L.; Mossman, C. J . J . Org. Chem. 1992, 57, 1635-1637.
(c) Aube´, J .; Rafferty, P. S.; Milligan, G. L. Heterocycles 1993, 35, 1141-
1147. (d) Milligan, G. L.; Mossman, C. J .; Aube´, J . J . Am. Chem. Soc.
1995, 117, 10449-10459. (e) Forsee, J . E.; Aube´, J . J . Org. Chem. 1999,
64, 4381-4385.
(3) While azidomercurations have not been carried out using
aliphatic azides, hydrazoic acid and its salts react with alkenes in the
presence of mercuric ion to produce â-azidomercurials. See: (a)
Heathcock, C. H. Angew. Chem., Int. Ed. Engl. 1969, 8, 134-135. (b)
Larock, R. C. Solvomercuration/ Demercuration Reaction in Organic
Synthesis; Springer-Verlag: New York, 1986; pp 443-527. (c) Galle,
J . E.; Hassner, A. J . Am. Chem. Soc. 1972, 94, 3930-3933. (d) Sokolova,
T. N.; Grishin, Y. K.; Timofeev, I. V.; Kartashov, V. R. Russ. Chem. B.
1994, 43, 1044-1047. (e) Marchand, A. P.; Sorokin, V. D.; Rajagopal,
S. D.; Bott, S. G. Synth. Commun. 1994, 24, 3141-3147. (f) Kartashov,
V. R.; Sokolova, T. N.; Pavinskii, A. Y.; Timofeev, I. V.; Radbil’, A. B.
Russ. Chem. B. 1995, 44, 2375-2381.
10.1021/jo001181q CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/28/2000

Mercury-Promoted Schmidt Reactions
J . Org. Chem., Vol. 65, No. 24, 2000 8327
Sch em e 2. P r oton - vs Mer cu r y-P r om oted
Sch m id t Rea ction s
A proposal for the improved regioselectivity of the
mercury-promoted Schmidt reaction is shown in Scheme
3. The proton-initiated reaction begins with protonation
of the alkene, which is followed by rearrangement to the
tertiary carbocation 13, cyclization to the aminodiazo-
nium ion 14, and a nonregioselective migration of bond
a or bond b to produce two regioisomeric iminium ions
15. The mercury-promoted process begins with the
formation of the mercuronium ion 16 or its equivalent,6
which is opened by the azide to produce the aminodia-
zonium ion 17 without rearrangement. Migration of bond
c produces the iminium ion 18, which affords only 11
upon reduction. Other notable differences between these
two routes are the positions of the iminium ion functional
groups in 15 and 18 and the presence of the organomer-
cury group in 18.
Sch em e 3. P r op osed Mech a n ism s for P r oton - a n d
Mer cu r y-P r om oted Sch m id t Rea ction s
A variety of azidoalkenes were synthesized (see Sup-
porting Information) and examined in the mercury-
promoted Schmidt reaction (Table 1). Treatment of the
azidoalkenes 19-28 with 2 equiv of Hg(ClO4)2‚3H2O or
Hg(OTf)2 led to gas evolution and the formation of the
amines 11 and 29-38 after reduction of the intermediate
mercury-bearing iminium ion with NaBH4. For compari-
son, attempted acid-promoted Schmidt reactions of 21,
22, and 25-27 with TfOH led to decomposition. Entry 1
illustrates the use of a regioisomeric alkene 19 to arrive
at the same rearrangement product 11 (cf. Scheme 2, 11
from 10). Entries 1 and 2 provide additional examples of
cyclization without rearrangement; 6-exo ring closure of
the azidomercuronium ion is observed rather than five-
membered ring formation via a rearranged cation, the
observed process using the acidic method. Entries 3-10
show that 1,2-disubstituted alkenes may be used in the
mercury-promoted Schmidt reaction, which is generally
difficult in the acid-promoted method. Especially signifi-
cant are entries 4-10, reactions that would require a
nonstabilized secondary carbocation in the acid-promoted
Schmidt reaction. Indeed, treatment of such compounds
with acid did not produce Schmidt products, whereas the
mercury(II) method was successful. In entry 3, it is
unclear whether the mercury-promoted Schmidt reaction
proceeds by a 5-exo closure at the secondary nonbenzylic
position of the intermediate mercuronium ion followed
by migration of the benzylic organomercury group or by
a 6-endo closure at the benzylic position followed by ring
contraction. Entries 4-6 illustrate functional group toler-
ance that is not possible in the acid-promoted Schmidt
reaction. Entries 4-10 involve hydride migration in the
intermediate aminodiazonium ions (i.e., R2 ) H in
Scheme 1). As expected, 5-exo (vs 6-endo) and 6-exo (vs
7-endo) closure of the intermediate azidomercuronium
ions is apparently involved in entries 4-7. The observa-
tion of 5-endo rather than 4-exo closure in entry 10 is
also expected on the basis of ring strain issues. The
apparent 6-endo closure of the azidomercuronium ions
derived from 26 and 27 (entries 8 and 9) is unusual, since
5-exo closure would normally prevail, as was found in
entry 5. Related aminomercurations3b give 5-exo rather
than 6-endo products, whether under kinetic or thermo-
dynamic conditions.7 Perhaps the azidomercuration
method inherently favors 6-endo closure over 5-exo
(entries 8 and 9) and entry 5 is an exception, e.g., the
Resu lts a n d Discu ssion
The proton-initiated intramolecular Schmidt reaction
of the azidoalkene 10 had been previously examined in
our laboratories, producing a mixture of two regioisomeric
products, 11 and 12 (Scheme 2), the result of carbocation
rearrangement prior to cyclization (Scheme 3, top
pathway).1b The yield of 11/12 was typically modest (ca.
40%) as a result of isolation problems, although yields
as high as 79% could be obtained in larger scale runs.
Attempts were made to avoid carbocation rearrangement
and thus the lack of regioselectivity by treating 10 with
a variety of nonprotic electrophiles. While PhSeCl, Ph-
SeBr, PdCl2, Pd(OAc)2, I2, N-iodosuccinimide, Br2, Hg-
(OAc)2, Hg(O2CCF3)2, HgCl2, HgI2, Hg(NO3)2, and HgO
were unsuccessful in promoting Schmidt reactions, gen-
erally leaving the starting material untouched, exposure
of 10 to a stoichiometric amount of Hg(ClO4)2‚3H2O or
Hg(OTf)2 followed by reduction with sodium borohydride
afforded 11 in 84% and 87% yields, respectively, as
judged by GC against an internal decane standard,
calibrated for relative response.5 A single stereo- and
regioisomer was observed. The isolated yield of 11 was
38% using mercuric perchlorate trihydrate, again reflect-
ing isolation difficulties. The choice of solvent (THF, CH2-
Cl2, or benzene) did not significantly influence the yields.
Regarding the stoichiometry of this mercury-promoted
Schmidt reaction, at least 1 equiv of the mercury salt is
required. In principle, demercuration of an intermediate
such as 9 (see Scheme 1) by a nucleophilic counterion
could render the reaction catalytic in mercury, but we
were unable to promote such a process.
(4) For a review on the related area of alkene cyclofunctionalization,
i.e., cyclizations involving electrophile-induced additions of heteronu-
cleophiles to alkenes, see: Harding, K. E.; Tiner, T. H. In Compre-
hensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon
Press: Oxford, 1991; Vol. 4; pp 363-421.
(6) Varhan, H. B.; Bach, R. D. J . Org. Chem. 1992, 57, 4948-4954.
(7) (a) Tokuda, M.; Yamada, Y.; Takagi, T.; Suginome, H.; Furusaki,
A. Tetrahedron Lett. 1985, 26, 6085. (b) Tokuda, M.; Yamada, Y.;
Suginome, H. Chem. Lett. 1988, 1289.
(5) The results of these studies are presented in the Supporting
Information in tabular form.

8328 J . Org. Chem., Vol. 65, No. 24, 2000
Pearson et al.
Sch em e 4. 5-en d o-tr ig Cycliza tion s:
Ta ble 1. Mer cu r y-P r om oted Sch m id t Rea ction s
Regioselectivity in P r oton - vs Mer cu r y-P r om oted
Sch m id t Rea ction s
Sch em e 5. 5-exo-tr ig Cycliza tion s:
Regioselectivity in P r oton - vs Mer cu r y-P r om oted
Sch m id t Rea ction s
this point. Treatment of 398 with trifluoromethane-
sulfonic acid followed by reduction afforded the two
benzo-fused pyrrolizidines 40 and 41 in a 1.5:1 ratio,
presumably resulting from a 5-endo cyclization to the
aminodiazonium ion 42 followed by a 1,2-aryl (rather
than alkyl) migration.8 Using mercuric trifluoromethane-
sulfonate, 40 and 41 were obtained with the same
efficiency and regioselectivity. Apparently, the mercury
substituent in the aminodiazonium ion 43 exerted no
effect on the migratory preference of the aryl and alkyl
groups, a scenario that is reasonable given that the
mercury substituent is not directly bonded to the migrat-
ing center in the rearrangement step. In contrast, the
cyclization of the azido-alkene 44,8 proceeding through
an initial 5-exo cyclization, shows a clear influence of the
mercury substituent. The acid-promoted Schmidt cycliza-
tion of 44,8 proceeding through the aminodiazonium ion
47, gave the two benzo-fused indolizidines 45 and 46 in
roughly equal amounts, demonstrating that aryl and
alkyl migrations occurred to approximately the same
extent. The mercury-promoted Schmidt cyclization of 44,
however, produced only the regioisomer 45. If the mercury-
bearing aminodiazonium ion 48 is involved, this experi-
ment indicates that the mercurio group inhibits the
migration of the alkyl group, allowing aryl migration to
dominate. Examination of the mercury-promoted cycliza-
tions in Scheme 2 and Table 1 reveals a similar observa-
tion: with the exception of entry 3 in Table 1, all of these
cyclizations proceed by migration of a group that does
not bear a mercurio substituent. It is difficult to rational-
ize entry 3, since it is not known whether the cyclization
proceeds in a 5-exo or 6-endo fashion, although it is likely
a
Conditons: 2 equiv of Hg(ClO4)2‚3H2O in THF; NaBH4, 15%
NaOH unless otherwise noted. In cases where new chiral centers
are formed, only the isomer shown was detected by 1H NMR.
Isolated, chromatographed yields unless otherwise noted. c Yield
b
determined by GC against a decane internal standard, calibrated
d
for relative response. Conditions: 2 equiv of Hg(OTf)2 in THF;
NaBH4, MeOH. e After storage at -5 °C for 2 weeks. f Conditons:
2 equiv of Hg(OTf)2 in THF; NaBH4, MeOH; NaOMe, MeOH,
reflux.
inductive effect of the allylic oxygen on the regioselec-
tivity of the ring-opening of the mercuronium ion derived
from 23 might favor 6-endo rather than 5-exo ring
closure. Clearly, the 6-endo closures apparently involved
in entries 8 and 9 are unexpected and unprecedented in
the realm of cyclofunctionalization chemistry. While no
explanation is available at this time, this result is a
useful one, providing a novel route to piperidines. Finally,
the use of the secondary azide 27 (entry 9) is also
significant. The cis-stereoselectivity observed is likely a
result of axial hydride delivery to the best chairlike
conformation of the intermediate iminium ion.
When comparing the proton- and mercury-promoted
Schmidt reactions, the regioselectivity of the 1,2-shift
may be influenced by the presence of the mercury
substituent (compare 3 and 8 in Scheme 1 above).
Intramolecular Schmidt reactions of the azido-alkenes 39
and 44 (Schemes4 and 5, respectively) serve to illustrate
(8) Pearson, W. H.; Fang, W.-k. J . Org. Chem. 2000, 65, 7158-7174.

Mercury-Promoted Schmidt Reactions
J . Org. Chem., Vol. 65, No. 24, 2000 8329
Sch em e 6. P r op osed Ra d ica l Ch em istr y of th e
Sch em e 7. P r obin g th e Na tu r e of th e
Or ga n om er cu r y In ter m ed ia te
r-Mer cu r io Im in iu m Ion s
that a mercurio-bearing group migrates in either case,
i.e., a mercuriobenzyl group or a mercurioalkyl group
must migrate in the 5-exo- or 6-endo-derived aminodia-
zonium ions, respectively.
While the mercury-promoted Schmidt reaction is a
useful complement to the proton-initiated version, this
process would be more powerful if the organomercury
intermediate (e.g., 3 in Scheme 1 and 18 in Scheme 3)
could be used for a subsequent carbon-carbon formation.
In particular, upon hydride reduction, â-hetero organo-
mercury compounds are known to produce â-hetero
radicals that can be trapped with electron-poor alkenes.9
Attempts were therefore made to capture the presumed
radical intermediate from the borohydride-promoted de-
mercuration reaction of the intermediate 18 encountered
in the mercury-promoted Schmidt reaction of 10 (see
Scheme 6) in an attempt to form the chain-extended
compounds 49. Treatment of 10 with Hg(OTf)2 or Hg-
(ClO4)2‚3H2O in THF or dichloromethane followed by
reduction with NaBH4 or NaHB(OMe)3 in the presence
of acrylonitrile or ethyl acrylate failed to induce carbon-
carbon bond formation. In related work, 1-(1-pyrrolidi-
nyl)cyclohexene 50 was treated with Hg(OTf)2 or Hg-
(OAc)2 followed by the same hydrides and traps, also
failing to produce the coupling products 52. Studies on
the interaction of enamines with mercury(II) salts have
been reported, indicating that R-mercurio iminium ions
such as 51 are involved and may be reduced to amines
with sodium borohydride.10-12
borodeuteride, producing only the monodeuterated prod-
uct 11a , the result of deuteride reduction at the iminium
ion carbon (i.e., R1 ) D). No evidence for a second
deuterium at R2, the position presumably occupied by
mercury, was found, i.e., no 11b (R1 ) R2 ) D) was
formed. When the reduction was carried out with sodium
borodeuteride in D2O (or CD3OD), two deuteria were
incorporated, producing 11b (R1 ) R2 ) D), implying that
deuterio-demercuration had occurred prior to iminium
ion reduction. Further evidence for this pathway was
obtained when sodium borohydride in D2O (or CD3OD)
was used, leading to 11c (R1 ) H, R2 ) D). Again,
deuterio-demercuration had occurred prior to iminium
ion reduction. Similar results were obtained when the
enamine 50 was mercurated and subjected to these three
sets of reductive conditions, producing 53a -c. Appar-
ently, 10 and 50 are converted to the R-mercurio-
enamines 54, which undergo proto- or deuteriodemercu-
ration in the presence of a protic or deuteric solvent,
respectively, to produce the iminum ions 55 and thus the
amines 11 or 53 after hydride or deuteride reduction.
Examples of protodemercuration of organomercury com-
pounds have been reported.16-18 A consequence of these
studies is that R-mercurio iminium ions, while probably
present, are too fragile to use in subsequent chemistry.
The inability to capture the presumed radical inter-
mediate involved in hydride reductions of the R-mercurio
iminium ions led us to seek evidence for the presence of
the mercury(II) group in the Schmidt reaction product.
Specifically, reduction with sodium borodeuteride should
replace the mercury with deuterium (Scheme 7, 54 f 55-
d 1 f 11-d 2 or 53-d 2), a well-known process in organo-
mercury chemistry.13-15 The mercury-promoted Schmidt
reaction of 10 was thus followed by reduction with sodium
Con clu sion
In conclusion, alkenes undergo azidomercuration with
aliphatic azides and mercury(II) salts, resulting in mer-
cury-promoted Schmidt reactions. This mild method
complements the proton-initiated Schmidt method, which
proceeds under strongly acidic conditions. Evidence for
R-mercurio iminium ion intermediates was obtained,
which are readily proto- or deuterio-demercurated by
(9) Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds; Pergamon Press: Oxford, 1986; pp 44-49.
(10) Brodersen, K.; Opitz, G.; Breitinger, D.; Menzel, D. Chem. Ber.
1964, 97, 1155-1162.
(11) Brodersen, K.; Opitz, G.; Breitinger, D. Chem. Ber. 1964, 97,
2046-2049.
(12) Bach, R. D.; Mitra, D. K. J . Chem. Soc., Chem. Commun. 1971,
1433-1434.
(13) Bordwell, F. G.; Douglass, M. L. J . Am. Chem. Soc. 1965, 88,
993-999.
(16) Iwayanagi, T.; Matsuo, M.; Saito, Y. J . Organomet. Chem. 1977,
135, 1-11.
(14) Pasto, D. J .; Gontarz, J . A. J . Am. Chem. Soc. 1968, 91, 719-
721.
(17) Beletskaya, I. P.; Artamkina, G. A.; Reutov, O. A. Dokl. Akad.
Nauk SSSR 1980, 251, 358-361.
(15) J ensen, F. R.; Miller, J . J .; Cristol, S. J .; Beckley, R. S. J . Org.
Chem. 1972, 37, 4341-4349.
(18) Giffard, M.; Cousseau, J .; Gouin, L.; Crahe, M.-R. Tetrahedron
1986, 42, 2243-2252.

8330 J . Org. Chem., Vol. 65, No. 24, 2000
Pearson et al.
Meth od E. A solution of 1-methyl-6-(3-azidopropyl)cyclohex-
1-ene 19 (0.20 g, 1.1 mmol) in THF (2 mL) was added in a
dropwise fashion to a solution of mercuric perchlorate trihy-
drate (0.51 g, 1.1 mmol) in THF (4 mL) at room temperature.
Gas evolution was observed. After 15 min, a solution of sodium
borohydride (0.26 g, 6.7 mmol) in 15% aqueous sodium
hydroxide (2 mL) was added. After 2 h, the mixture was diluted
with ether (15 mL) and decanted away from the bead of
mercury. The organic phase was washed twice with brine,
dried (MgSO4), and concentrated. Chromatography (20% ethyl
acetate/hexanes to 5% MeOH/ethyl acetate gradient) gave
0.052 g (31%) of the title compound, identical to the material
obtained using Method A.
solvent, producing mercury-free iminium ions that may
be reduced to amines.
Exp er im en ta l Section 19,20
(2R*,7R*)-2-Meth yl-1-a za bicyclo[5.3.0]d eca n e (11). Me-
th od A. A solution of 2-(3-azidopropyl)methylenecyclohexane
1020 (0.20 g, 1.1 mmol) in THF (2 mL) was added in a dropwise
fashion to a solution of mercuric perchlorate trihydrate (0.51
g, 1.1 mmol) in THF (4 mL) at room temperature. Gas
evolution was observed. After 15 min, a solution of sodium
borohydride (0.26 g, 6.7 mmol) in 15% aqueous sodium
hydroxide (2 mL) was added. After 2 h, the mixture was diluted
with ether and decanted away from the bead of mercury. The
organic phase was washed twice with brine, dried (MgSO4),
and concentrated. Chromatography (20% ethyl acetate/hexanes
to 5% MeOH/ethyl acetate gradient) gave 0.065 g (38%) of the
title compound as a single diastereomer and regioisomer as
judged by proton and carbon NMR spectroscopy: Rf ) 0.15
(30% ethyl acetate/hexanes); 1H NMR (CDCl3, 300 MHz) δ 3.28
(ddd, J ) 9.5 Hz, 7.4 Hz, 2.9 Hz, 1H), 2.61 (q, J ) 7.7 Hz, 1H),
2.16-2.41 (m, 2H), 1.37-2.06 (m, 12H), 1.12, (d, J ) 6.4 Hz,
3H); 13C NMR (CDCl3, 75 MHz) δ 62.50(+), 60.75(+), 55.77(-
), 36.99(-), 35.75(-), 33.68(-), 26.74(-), 23.68(-), 22.82(+),
22.75(-); IR (neat) 1461(m) cm-1; MS (EI, 70 eV) m/z (rel int)
153 (18.0), 152 (14.8), 138 (100); HRMS (CI with NH3) calcd
for C10H19N 153.1517, found 153.1516. The proton NMR
resonances of this compound matched one set of resonances
found in an inseparable mixture of 11 and 12 as previously
reported.1b
(2R*,8R*)-2-Meth yl-1-a za bicyclo[6.3.0]u n d eca n e (29). A
solution of 2-(3-azidopropyl)methylenecycloheptane 2020 (0.20
g, 1.0 mmol) in THF (3 mL) was added in a dropwise fashion
to a solution of mercuric perchlorate trihydrate (0.94 g, 2.0
mmol) in THF (5 mL) at room temperature. Gas evolution was
observed. After 15 min, a solution of sodium borohydride (0.19
g, 5.1 mmol) in 15% aqueous sodium hydroxide (2 mL) was
added. After 2 h, the mixture was diluted with ether and
decanted away from the bead of mercury. The organic phase
was washed twice with brine, dried (MgSO4), and concentrated.
Chromatography (1:1 ethyl acetate/hexanes to 20% acetone/
ethyl acetate gradient) gave 0.05 g (31%) of a single stereo-
isomer of the title compound, as tetermined by proton NMR
spectroscopy: Rf ) 0.09 (30% ethyl acetate/hexanes); 1H NMR
(CDCl3, 300 MHz) δ 3.25 (t, J ) 6.5 Hz, 1H), 2.96-3.07 (m,
1H), 2.58-2.69 (m, 1H), 2.22-2.34 (m, 2H), 1.24-2.06 (m,
13H), 1.26 (d, J ) 6.4 Hz, 3H);); 13C NMR (CDCl3, 75 MHz,
J MOD) δ 67.5 (+), 64.1 (+), 58.6 (-), 33.5 (-), 33.4 (-), 29.6
(-), 24.2 (-), 23.0 (-), 22.6 (-), 21.7 (-), 17.1 (+); IR (neat)
2785 (s) cm-1; MS (CI with CH4) m/z (rel int) 182 (M + CH3,
Meth od B. A solution of 2-(3-azidopropyl)methylenecyclo-
hexane 1020 (0.100 g, 0.560 mmol) in THF (2 mL) was added
in a dropwise fashion to a solution of mercuric perchlorate
trihydrate (0.254 g, 0.560 mmol) and decane (internal stan-
dard, 0.080 g, 0.560 mmol) in THF (3 mL) at room tempera-
ture. Gas evolution was observed. After 15 min, an aliquot was
removed from the reaction mixture and quenched with a
solution of sodium borohydride (0.05 g, 1.31 mmol) in 1.0 mL
of 15% aqueous sodium hydroxide. Ether (3 mL) was added
14.1), 168 (M + H, 100); HRMS (CI with CH4) calcd for C11H21
NH 168.1752, found 168.1749.
-
N-(4-Meth oxyben zyl)p yr r olid in e (30). This compound
has been previously synthesized by the coupling of an aryl-
stannane with an iminium ion.21 Our mercury-promoted
Schmidt method was carried out as follows. A solution of (Z)-
1-azido-5-(p-methoxyphenyl)pent-4-ene 2120 (0.17 g, 0.81 mmol)
in THF (2 mL) was added in a dropwise fashion to a solution
of mercuric perchlorate trihydrate (0.73 g, 1.6 mmol) in THF
(5 mL) at room temperature. Gas evolution was observed. After
15 min, a solution of sodium borohydride (0.290 g, 7.50 mmol)
in 15% aqueous sodium hydroxide (2 mL) was added. After 2
h, the mixture was diluted with ether and decanted away from
the bead of mercury. The organic phase was washed twice with
brine, dried (MgSO4), and concentrated. Chromatography (1:1
ethyl acetate/hexanes to 20% acetone/ethyl acetate gradient)
gave 0.66 g (43%) of the title compound: Rf ) 0.21 (30% ethyl
acetate/hexanes); proton and carbon NMR data for this
compound matched the literature values.21
(2R,3R,4R,5R)-2-[3-(Ca r boeth oxy)p r op yl]-3,4,5-tr i(ben -
zyloxy)p ip er id in e (31) a n d (1R,2R,3R,9a R)-1,2,3-Tr i(ben -
zyloxy)quinolizidin-6-one (32).Asolutionofmethyl(2R,3S,4R,5E)-
9-azido-6,7,8-tri(benzyloxy)non-5-enoate 2222 (0.431 g, 0.950
mmol) in THF (2 mL) was added in a dropwise fashion to a
solution of mercuric perchlorate trihydrate (0.200 g, 0.368
mmol) in THF (2 mL) at room temperature. Gas evolution was
observed. After 15 min, a solution of sodium borohydride (0.072
g, 1.90 mmol) in 15% aqueous sodium hydroxide (2 mL) was
added. After 2 h, the mixture was diluted with ether and
decanted away from the bead of mercury. The organic phase
was washed twice with brine, dried (MgSO4), and concentrated.
Chromatography (1:1 ethyl acetate/hexanes to 25% acetone/
ethyl acetate gradient) gave 0.078 g (41%) of a single stereo-
isomer of the title compound 31 as a colorless oil: Rf ) 0.66
(50% ethyl acetate/acetone); 1H NMR (CDCl3, 300 MHz) δ
7.23-7.44 (m, 15H), 4.98 (d, J ) 3.8 Hz, 1H), 4.69 (s, 2H),
4.59 (q, J ) 1.4 Hz, 2H), 4.07 (q, J ) 2.9 Hz, 2H),3.56 (br. s,
1H), 3.45 (d, J ) 2.4 Hz, 2H), 3.12 (dd, J ) 5.0 Hz, J ) 1.3 Hz,
and 1 µL of the organic phase was injected into a GCMS.19
A
product peak having a retention time of 5.56 min was observed.
An 84% yield was calculated by comparing the ratio of decane
to product, corrected for relative response using a predeter-
mined linear response curve.
Meth od C. A solution of 2-(3-azidopropyl)methylenecyclo-
hexane 1020 (0.100 g, 0.560 mmol) in THF (2 mL) was added
in
a dropwise fashion to a solution of mercuric trifluo-
romethanesulfonate (0.278 g, 0.560 mmol) and decane (internal
standard, 0.080 g, 0.560 mmol) in THF (3 mL) at room
temperature. After 15 min, quenching of an aliquot and
analysis by GC as above indicated an 87% yield.
Meth od D. A solution 1-methyl-6-(3-azidopropyl)cyclohex-
1-ene 19 (0.100 g, 0.560 mmol) in THF (2 mL) was added in a
dropwise fashion to a solution of mercuric trifluoromethane-
sulfonate (0.28 g, 0.56 mmol) and decane (internal standard,
0.080 g, 0.560 mmol) in THF (3 mL) at room temperature. Gas
evolution was observed. After 15 min, quenching of an aliquot
and analysis by GC as above indicated an 45% yield.
(19) Proton NMR spectral assignments were based on two-dimen-
sional correlated off-resonance spectroscopy (COSY) experiments,
nuclear Overhauser and exchange spectroscopy (NOESY) experiments,
or comparisons with analogous compounds reported in the literature.
J -Modulated spin-echo Fourier transform (J MOD) 13C NMR experi-
ments are reported as (+) for CH3 and CH or (-) for CH2 and
quarternary carbon and are used as an alternative to off-resonance
decoupling. GC and GC/MS data were obtained on a Hewlett-Packard
6890 series gas chromatograph equipped with a 5973 mass selective
detector. The column was a Hewlett-Packard 5-MS comprising 5%
phenyl methyl siloxane with a film thickness of 0.25 µM, internal
diameter of 0.25 mm, and a length of 30 m. Temperature program:
50 °C for 4 min, then increase to 200 °C at 40 °C/min, then hold 10
min. For exploratory experiments, accurate yields were obtained by
gas chromatography against an internal decane standard, where a
calibration curve was used to correct for relative response factors.
(20) See the Supporting Information for the preparation of the
Schmidt substrates.
(21) Cooper, M. S.; Fairhurst, R. A.; Heaney, H.; Papageogiou, G.;
Wilkins, R. F. Tetrahedron 1989, 45, 1155-1166.
(22) Hembre, E. J . Ph.D. Thesis, University of Michigan, 1996.

Mercury-Promoted Schmidt Reactions
J . Org. Chem., Vol. 65, No. 24, 2000 8331
1H), 2.12-2.53 (m, 6H), 1.53-1.97 (m, 3H), 1.23 (t, J ) 3.1
Hz, 3H). Upon standing at -5 °C for 2 weeks, this material
cyclized quantitatively to give 74 mg of the lactam 32 after
removal of a trace of ethanol in vacuo: 1H NMR (CDCl3, 400
MHz) δ 7.29-7.44 (m, 15H), 5.14 (dd, J ) 14.3 Hz, J ) 3.3
Hz, 1H), 5.04 (d, J ) 11.0 Hz. 1H), 4.81 (d, J ) 12.8, 1H), 4.49-
4.65 (m, 5H), 3.83 (t, J ) 2.6 Hz, 1H), 3.73 (t, 9.5 Hz, 1H),
3.50 (dd, J ) 9.2 Hz, J ) 2.9 Hz, 1H), 3.16-3.22 (m, 1H), 2.39
(t, J ) 6.2 Hz, 1H), 2.32 (d, J ) 14.3 Hz, 1H), 2.10-2.16 (m,
1H), 1.76-1.89 (m, 2H), 1.60-1.67 (m, 1H); 13C NMR (CDCl3,
100 MHz) δ 170.2, 138.3, 138.1, 138.0, 128.4, 128.3, 128.2,
128.0, 127.8, 127.7, 127.6, 83.3, 79.0, 75.7, 71.2, 70.4, 70.1, 59.2,
41.7, 32.7, 25.4, 18.5; IR (neat) 1695 (s) cm-1; MS (CI with
NH4+) m/z (rel int) 472 (M + H, 100); HRMS (CI with NH3)
calcd for C30H34NO4 472.2487, found 472.2473. The configu-
ration of the title lactam was assigned based on its similarity
to the quinolizidine 34 synthesized below, a compound we had
previously made and whose configuration had been assigned
by 2-D NOESY NMR spectroscopy.22
(w), 1453 (m) cm-1; MS (CI with NH4+) m/z (rel int) 458 (M+,
7), 212 (100); HRMS (CI with NH3) calcd for C30H36NO3 (M +
H) 458.2695, found 458.2717. The proton and carbon NMR
data for this compound matched the values we had obtained
previously.22
2-Hep tylp ip er id in e (35). This compound has been made
by Meyers and co-workers by the alkylation of a metalated
formamidine derivative of piperidine.24 The mercury-promoted
Schmidt method was carried out as follows: A solution of (Z)-
1-azidododec-5-ene 2520 (0.26 g, 0.560 mmol) in THF (2 mL)
was added in a dropwise fashion to a solution of mercuric
perchlorate trihydrate (1.13 g, 2.50 mmol) in THF (5 mL) at
room temperature. Gas evolution was observed. After 15 min,
a solution of sodium borohydride (0.290 g, 7.50 mmol) in 15%
aqueous sodium hydroxide (2 mL) was added. After 2 h, the
mixture was diluted with ether and decanted away from the
bead of mercury. The organic phase was washed twice with
brine, dried (MgSO4), and concentrated. Chromatography (1:1
ethyl acetate/hexanes to 20% acetone/ethyl acetate gradient)
gave 0.167 g (73%) of the title compound, Rf ) 0.13 (30% ethyl
acetate/hexanes), which was found to have proton and carbon
NMR spectroscopic data and MS data that matched the
literature values.24
2-Hexylp ip er id in e (36). This compound has been made
by Meyers and co-workers by the alkylation of a metalated
formamidine derivative of piperidine.25 Our mercury-promoted
Schmidt method was carried out as follows. A solution of (E)-
1-azidoundec-4-ene 2620 (0.250 g, 1.28 mmol) in THF (2 mL)
was added in a dropwise fashion to a solution of mercuric
perchlorate trihydrate (1.16 g, 2.56 mmol) in THF (8 mL) at
room temperature. Gas evolution was observed. After 45 min,
a solution of sodium borohydride (0.290 g, 7.68 mmol) in 15%
aqueous sodium hydroxide (2 mL) was added. After 2 h, the
reaction mixture was diluted with ether and decanted away
from the bead of mercury. The organic phase was washed twice
with brine, dried (MgSO4), and concentrated. Chromatography
(1:1 ethyl acetate/hexanes to 20% acetone/ethyl acetate gradi-
ent) gave 0.126 g (58%) of the title compound, Rf ) 0.16 (30%
ethyl acetate/hexanes), which was found to have proton and
carbon NMR spectroscopic data and MS data that matched
the literature values.25
(1S,2R,8aR)-1,2-O-Isopr opyliden e-1,2-dih ydr oxyin doliz-
id in -5-on e (33). A solution of ethyl (4E,6S,7R)-6,7-O-isopro-
pylidine-6,7-dihydroxyoct-4-enoate 2322 (0.100 g, 0.371 mmol)
in THF (3 mL) was added in a dropwise fashion to a solution
of mercuric perchlorate trihydrate (0.338 g, 0.743 mmol) in
THF (2 mL) at room temperature. Gas evolution was observed.
After 30 min, a solution of sodium borohydride (0.056 g, 1.48
mmol) in MeOH (2 mL) was added. After 18 h, MeOH (3 mL)
and NaOMe (0.004 g, 0.074 mmol) were added and the solution
was heated at reflux for 3 h. The reaction mixture was then
cooled and concentrated. Chromatography (1:1 ethyl acetate/
hexanes to 25% acetone/ethyl acetate gradient) of the residue
gave 0.027 g (34%) of a single isomer of the title compound:
1
Rf ) 0.12 (10% MeOH/CHCl3); H NMR (CDCl3, 400 MHz) δ
4.73 (t, J ) 5.1 Hz, 1H), 4.60 (dd, J ) 6.0 Hz, J ) 4.4 Hz, 1H),
4.2 (d, J ) 13.6 Hz, 1H), 3.4 (m, 1H), 3.09 (dd, J ) 13.6 Hz, J
) 5.1 Hz, 1H), 2.22-2.46 (m, 2H), 1.61-2.07 (m, 4H), 1.41 (s,
3H), 1.32 (s, 3H); 13C NMR (CDCl3, 75 MHz, J MOD) δ 168.8
(-), 81.3 (+), 77.8 (+), 77.4 (+), 76.6 (-), 61.3 (+), 50.3 (-),
31.4 (-), 26.6 (+), 24.9 (+), 22.7 (-), 21.1 (-). The configuration
of the title compound was assigned by comparison to the
known compound (1S,2R,8R,8aR)-1,2-O-isopropylidene-1,2,8-
trihydroxyindolizidin-5-one.23
(2R*,5S*)-2-Eth yl-5-h exylp ip er id in e (37). A solution of
(E)-3-azidotridec-6-ene 2720 (0.350 g, 01.57 mmol) in THF (2
mL) was added in a dropwise fashion to a solution of mercuric
perchlorate trihydrate (1.42 g, 3.14 mmol) in THF (8 mL) at
room temperature. Gas evolution was observed. After 20 min,
a solution of sodium borohydride (0.360 g, 9.42 mmol) in 15%
aqueous sodium hydroxide (3 mL) was added. After 2 h, the
mixture was diluted with ether and decanted away from the
bead of mercury. The organic phase was washed twice with
brine, dried (MgSO4), and concentrated. Chromatography (1:1
ethyl acetate/hexanes to 20% acetone/ethyl acetate gradient)
gave 0.136 g (44%) of the title compound as a single stereo-
isomer as judged by proton and carbon NMR: Rf ) 0.13 (30%
ethyl acetate/hexanes); 1H NMR (CDCl3, 360 MHz) δ 2.43-
2.49 (m, 1H), 2.34-2.43 (m, 1H), 1.73-1.82 (m, 1H), 1.61-
1.72 (m, 2H), 1.27-1.40 (m, 14H), 0.94-1.07 (m, 2H), 0.91 (t,
J ) 7.48 Hz 3 H), 0.88 (t, J ) 6.87 Hz, 3H); 13C NMR (CDCl3,
90 MHz, J MOD) δ 58.76 (+), 57.16 (+), 37.44 (-), 32.69 (-),
32.19 (-), 31.79 (-), 30.11 (-), 29.48 (-), 25.92 (-), 24.81 (-),
22.58 (-), 14.06 (+), 10.43 (+); IR (neat) 3286 (br. w), 2794
(w), 2717 (w) cm-1; MS (CI with CH4) m/z (rel int) 198 (9.6),
112 (100); HRMS (CI with CH4) calcd for C13H27NH 198.2222,
found 198.2214. The configuration of this compound was
assigned as 2,5-cis on the basis of a comparison of the chemical
shifts of carbons C(2) and C(5) in the 13C NMR spectrum (δ
58.8, 57.2, order not known) to the values reported for cis- and
trans-2-methyl-5-undecylpiperidine (δ 52.6, 57.2 and δ 45.9,
50.9, respectively).26
(1R,2R,3R,9a R)-1,2,3-Tr i(ben zyloxy)qu in olizid in e (34).
This compound has been prepared in our laboratories by the
thermal cyclization of (2R,3S,4R,5Z)-1-azido-2,3,4-tri(benzyl-
oxy)-9-chloro-5-nonene and the reduction of the resultant
iminium ion.22 Our mercury-promoted Schmidt method for the
preparation of this compound was carried out as follows. A
solution of (2R,3S,4R,5Z)-1-azido-2,3,4-tri(benzyloxy)-9-chloro-
5-nonene 2422 (0.374 g, 0.721 mmol) in THF (2 mL) was added
in
a dropwise fashion to a solution of mercuric trifluo-
romethanesulfonate (0.720 g, 1.44 mmol) in THF (5 mL) at
room temperature. Gas evolution was observed. After 15 min,
a solution of sodium borohydride (0.164 g, 4.32 mmol) in
methanol (3 mL) was added. After 4 h, sodium methoxide
(0.116 g, 2.16 mmol) was added and the solution was warmed
to reflux for 4 h. After cooling, the contents of the flask were
diluted with ether and decanted away from the bead of
mercury. The solution was washed twice with brine, dried
(MgSO4), and concentrated. Chromatography (100% dichlo-
romethane to 5% methanol in dichloromethane gradient) gave
0.213 g (63%) of a single isomer of the title compound as a
colorless oil: Rf ) 0.45 (10% MeOH/CH2Cl2); 1H NMR (CDCl3,
400 MHz) δ 7.24-7.42 (m, 15H), 4.99 (d, J ) 10.6 Hz, 1H),
4.80 (ABq, J AB ) 13.2 Hz, ∆ν ) 25.6 Hz, 2H), 4.57-4.67 (m,
4H), 3.73-3.78 (m, 1H), 3.66 (t, J ) 9.3 Hz, 1H), 3.39 (dd, J )
9.3 Hz, J ) 3.3 Hz, 1H), 2.89 (dd, J ) 12.5 Hz, J ) 2.9 Hz,
1H), 2.82 (br. d, J ) 11.4 Hz, 1H), 2.16 (br. d, J ) 12.5 Hz,
2H), 1.84-1.95 (m, 2H), 1.67-1.74 (m, 2H), 1.52-1.58 (m, 1H),
1.13-1.38 (m, 1H); 13C NMR (CDCl3, 100 MHz) δ 138.8, 138.6,
128.5, 128.3, 128.2, 128.1, 127.6, 127.5, 126.9, 83.9, 80.4, 75.7,
71.9, 71.6, 71.2, 66.1, 57.6, 55.8, 28.2, 25.1, 23.7; IR (neat) 1496
(24) Meyers, A. I.; Poindexter, G. S.; Brich, Z. J . Org. Chem. 1978,
43, 892-898.
(25) Meyers, A. I.; Dickman, D. A.; Bailey, T. R. J . Am Chem. Soc.
1985, 107, 7974-7978.
(23) Pearson, W. H.; Hembre, E. J . J . Org. Chem. 1996, 61, 7217-
7221.

8332 J . Org. Chem., Vol. 65, No. 24, 2000
Pearson et al.
2-Hexylp yr r olid in e (38). This compound has previously
been synthesized by Buchwald and co-workers by a titanocene-
mediated reduction of 2-hexyl-1-pyrroline.27 Our mercury-
promoted Schmidt method was carried out as follows. A
solution of (Z)-1-azidododec-5-ene 2820 (0.269 g, 1.38 mmol)
in THF (2 mL) was added in a dropwise fashion to a solution
of mercuric perchlorate trihydrate (1.25 g, 2.76 mmol) in THF
(10 mL) at room temperature. Gas evolution was observed.
After 20 min, sodium borohydride (0.31 g, 8.3 mmol) in 15%
aqueous sodium hydroxide (3 mL) was carefully added. The
solution instantly became gray in color and after 14 h, the fine
black particles coalesced into a bead of mercury. The mixture
was diluted with ether and decanted away from the bead of
mercury. The organic phase was separated and washed with
brine, then dried (MgSO4) and concentrated. Chromatography
(1:1 ethyl acetate/hexanes to 20% acetone/ethyl acetate gradi-
ent) gave 0.079 g (37%) of the title compound as a pale yellow
oil. The proton and carbon NMR data for this compound
matched the literature values.27
NMR, 13C NMR, and MS data for these compounds matched
the values we had previously observed8 and the literature
values.28-30
(1R*,3aS*)-6-Meth oxy-1-(2-pr open yl)-1,2,3,3a,4,5-h exah y-
d r op yr r olo[1,2-a ]qu in olin e (45a ) a n d (1R*,3a R*)-6-Meth -
oxy-1-(2-p r op en yl)-1,2,3,3a ,4,5-h exa h yd r op yr r olo[1,2-a ]-
qu in olin e (45b).8 Mercuric trifluoromethanesulfonate (300
mg, 0.60 mmol) was added to a solution of 3-(3-azido-5-
hexenyl)-7-methoxy-1H-indene 448 (100 mg, 0.37 mmol) in
THF (8 mL) at room temperature. After 30 min, the reaction
mixture was cooled to 0 °C and sodium borohydride (84 mg,
2.22 mmol) in methanol (3 mL) was added. After the mixture
warmed to room temperature for 14 h, 15% aqueous sodium
hydroxide was added and the mixture was extracted with ether
(3x). The combined organic phases were washed twice with
brine, dried (Na2SO4), and concentrated. Chromatography (1:7
ether/hexanes) gave 42 mg (47%) of an inseparable 2:1 mixture
of 45a and 45b as a clear oil, Rf ) 0.61, as judged by
integration of the hydrogens at δ 3.12-3.04 (m, 0.5H) and 2.96
(dd, J ) 17.3, 5.6 Hz, 1H) in the 1H NMR spectrum. The 1H
NMR, 13C NMR, and MS data for these compounds matched
the values we had previously observed.8
2,3,8,8a-Tetr ah ydr o-1H-3a-azacyclopen ta[a ]in den e (40)
a n d 2,3,5,9b-Tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ole (41).
Mercuric trifluoromethanesulfonate (1.00 g, 2.01 mmol) was
added to a solution of 7-(3-azidopropylidene)bicyclo[4.2.0]octa-
1(6),2,4-triene 398 (280 mg, 1.51 mmol, 2:1 E:Z) in THF (25
mL) at room temperature. After 1 h, the mixture was cooled
to 0 °C and treated with borane dimethyl sulfide complex (4.50
mL of a 2.0 M in THF, 9.00 mmol). After 14 h, 15% aqueous
sodium hydroxide was added and the mixture was extracted
with ether (3x). The combined organic phases were washed
with brine (3x), dried (MgSO4) and concentrated. Chromatog-
raphy (1:7 ether/hexanes) gave 120 mg (50%) of an inseparable
Ack n ow led gm en t. We thank the National Insti-
tutes of Health (GM-35572 and CA-77365) for funding
this research. We also thank Dr. Frank E. Lovering for
carrying out initial experiments with mercury-mediated
Schmidt cyclizations.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails and spectral data for compounds not reported in the
Experimental Section, deuterium labeling studies, copies of
1H and/or 13C NMR spectra for all new compounds without
elemental analysis. This material is available free of charge
via the Internet at http://pubs.acs.org.
1
1.5:1 mixture of 4028,29 and 4130 as judged by 1H NMR. The H
(26) Watanabe, Y.; Iida, H.; Kibayashi, C. J . Org. Chem. 1989, 54,
4088-4097.
(27) Willoughby, C. A.; Buchwld, S. L. J . Am. Chem. Soc. 1994, 116,
8952-8965.
J O001181Q
(28) Danishefsky, S.; Doehner, R. Tetrahedron Lett. 1977, 35, 3029-
3030.
(29) Ziegler, F. E.; J eroncic, L. O. J . Org. Chem. 1991, 56, 3479-
3486.
(30) Meyers, A. I.; Santiago, B. Tetrahedron Lett. 1995, 36, 5877-
5880.