Construction of the benzindenoazepine skeleton via cyclopentannulation of Fischer aminocarbene complexes: Total synthesis of bulgaramine

Article abstract of DOI:10.1021/jo050631h

A total synthesis of bulgaramine has been accomplished with a longest linear sequence of eight steps and an overall yield of 23% from commercially available 3,4-dimethoxyphenethyl alcohol. An intramolecular cyclopentannulation reaction of a Fischer aminoc

Full text of DOI:10.1021/jo050631h

Construction of the Benzindenoazepine Skeleton via
Cyclopentannulation of Fischer Aminocarbene Complexes: Total
Synthesis of Bulgaramine
Matthew W. Giese† and William H. Moser*
Department of Chemistry, Indiana University-Purdue University Indianapolis,
Indianapolis, Indiana 46202
wmoser@mmm.com
Received March 31, 2005
A total synthesis of bulgaramine has been accomplished with a longest linear sequence of eight
steps and an overall yield of 23% from commercially available 3,4-dimethoxyphenethyl alcohol. An
intramolecular cyclopentannulation reaction of a Fischer aminocarbene complex provided the key
step and occurred under significantly milder conditions and in higher yields than those of other
reported examples of this reaction type. The reaction solvent was a critical factor in the
cyclopentannulation reaction, with measurable amounts of the desired product observed only when
THF was utilized. The product yield could be further enhanced by the addition of two-electron
donor ligands, demonstrating the first example of this effect on the thermal reaction of aminocarbene
complexes with alkynes.
Introduction
Plants of the genus Fumaria have long been utilized
in several Asian and eastern European countries as folk
medicines for their antipyretic, analgesic, and diuretic
properties.1 A wealth of chemical constituents that may
be responsible for these properties have been extracted
from these plants, including a series of compounds based
on the benzindenoazepine ring system.2 Among the
known benzindenoazepines, bulgaramine (Figure 1) has
received attention because of its apparent central position
in the biogenic pathway of Fumaria-derived alkaloids.3
FIGURE 1. Bulgaramine.
It has been suggested that spirobenzylisoquinolines are
the biological precursors to bulgaramine, which in turn
is converted to other benzindenoazepines and/or related
alkaloids of a higher oxidation state. Evidence for this
hypothesis has been provided by semisynthetic transfor-
mations patterned after these proposed biological path-
ways.4 In addition to these biogenic studies, the valuable
medicinal properties and interesting architectural fea-
tures of the benzindenoazepines have motivated the
development of new synthetic methodology geared toward
their synthesis.5
* Corresponding author. Present address: 3M Pharmaceuticals, 3M
Center, Building 0270-04-S-02, St. Paul, MN 55144.
† Present address: Eli Lilly and Company, Lilly Corporate Center,
Indianapolis, IN 46285.
(1) (a) Blasko, G. In The Alkaloids; Brossi, A., Ed.; Academic Press
Inc.: New York, 1990; pp 157-224. (b) Lee, T. B. Illustrated Flora of
Korea; Hanygmoonsa: Seoul, 1989; p 385.
(2) (a) Blasko, G.; Hussain, S. F.; Freyer, A. J.; Shamma, M.
Tetrahedron Lett. 1981, 22, 3127. (b) Murugesan, N.; Blasko, G.;
Minard, R. D.; Shamma, M. Tetrahedron Lett. 1981, 22, 3131. (c)
Blasko, G.; Murugesan, N.; Hussain, S. F.; Minard, R. D.; Shamma,
M.; Sener, B.; Tanker, M. Tetrahedron Lett. 1981, 22, 3135. (d) Blasko,
G.; Murugesan, N.; Freyer, A. J.; Gula, D. J.; Sener, B.; Shamma, M.
Tetrahedron Lett. 1981, 22, 3139. (e) Blasko, G.; Murugesan, N.; Freyer,
A. J.; Minard, R. D.; Shamma, M. Tetrahedron Lett. 1981, 22, 3143.
(3) Yakimov, G. I.; Mollov, N. M.; Leet, J. E.; Guinaudeau, H.;
Freyer, A. J.; Shamma, M. J. Nat. Prod. 1984, 47, 1048.
(4) For recent examples, see: (a) Kim, I. J.; Park, Y. J.; Kim, J. I.;
Lee, K. T.; Kim, S. K. Arch. Pharm. Res. 1997, 20, 476. (b) Iwasa, K.;
Kamigauchi, M.; Takao, N. J. Nat. Prod. 1988, 51, 1232. (c) Hanaoka,
M.; Inoue, M.; Kobayashi, N.; Yasuda, S. Chem. Pharm. Bull. 1987,
35, 980.
10.1021/jo050631h CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/01/2005
6222
J. Org. Chem. 2005, 70, 6222-6229

Total Synthesis of Bulgaramine
In contrast to the wealth of available information
regarding Fischer alkoxycarbene complexes, thermal
reactions of the corresponding amino-substituted carbene
complexes with alkynes have been explored to a much
lesser degree. Yamashita originally established that the
pentannulation reaction pathway predominates with
aryl-substituted aminocarbene complexes, such that es-
sentially no products are observed in which a carbon
monoxide ligand has been incorporated.9 Subsequent
investigations have verified this motif, although excep-
tions have been reported.8a,10 The electronic difference
between alkoxy- and amino-substituted carbene com-
plexes is thought to provide the key reason for their
divergence in reactivity. The greater electron-donating
capacity of an amino vs an alkoxy substituent results in
increased electron density at the metal center and,
consequently, an increase in the strength of the bonds
to the carbon monoxide ligands. This serves to disfavor
carbon monoxide insertions during annulation processes,
thereby resulting in the observed prevalence of pentan-
nulation-derived products.
Although this observed preference for pentannulation
reactions would appear to provide the inviting op-
portunity for the construction of five-membered rings, a
problematic corollary is that the increased stability of
aminocarbene complexes relative to their alkoxy-substi-
tuted counterparts typically necessitates forcing condi-
tions for reactions with alkynes to proceed.9,11 This often
leads to lower product yields and the formation of
additional byproducts. In efforts to increase the produc-
tivity of annulation reactions with aminocarbene com-
plexes, several researchers have attempted to tabulate
additional factors that can affect product yield and
distribution, including solvent, concentration, and the
presence of ligand additives.8b,c,12 It is clear, however, that
the behavior of Fischer aminocarbene complexes remains
less well understood than their alkoxy-substituted coun-
terparts.
FIGURE 2. Proposed route to benzindenoazepine alkaloids.
Research in our laboratories has focused on thermal
reactions of Fischer chromium carbene complexes with
silyl-substituted alkynes in an inter- or intramolecular
fashion. Under certain experimental conditions, we have
observed novel and synthetically valuable annulation
reactions that can be achieved with high efficiency.6
During the course of our studies, we became interested
in the benzindenoazepine ring system, noting that a
Fischer aminocarbene complex could, in theory, provide
a useful precursor for its assembly. Specifically, success-
ful thermal intramolecular cyclopentannulation of a
Fischer aminocarbene complex possessing an appropri-
ately functionalized nitrogen-tethered alkyne would allow
the benzindenoazepine B and C rings to be constructed
in a single step (Figure 2).
Thermal reactions of aryl- and alkenyl-substituted
Fischer carbene complexes with alkynes have been
vigorously studied for several decades, and the number
of different types of products that derive from these
reactions is exceedingly large.7 Trends in reactivity have
emerged, however, and are based largely on the substitu-
tion pattern of the Fischer carbene complex. The most
carefully studied examples entail thermal reactions of
alkoxy-substituted carbene complexes with alkynes, which
undergo benzannulation and/or pentannulation trans-
formations as the major product-forming pathways. The
benzannulation process, in which fragments from the
carbene complex, alkyne, and a carbon monoxide ligand
join together to form a new phenolic product, is commonly
referred to as the Do¨tz reaction. The pentannulation
process, in which a carbon monoxide is not incorporated
into the products, is often a competitive reaction pathway
to the Do¨tz reaction and, in fact, is hypothesized to derive
from a common intermediate. Nonetheless, researchers
have established a number of experimental parameters,
including solvent, temperature, and ligand additives, by
which either the benzannulation or pentannulation path-
way can be favored.8
The differences in reactivity between alkoxy- and
aminocarbenes have also resulted in a disparity in their
utilization in synthetic endeavors. On one hand, the
efficient and predictable nature of the Do¨tz benzannu-
lation, coupled with the comparatively mild reaction
conditions required, has motivated numerous natural
product syntheses that use this transformation in either
inter- or intramolecular fashion as a key step.13 In
contrast, the recalcitrant behavior of aminocarbene com-
plexes in annulation reactions appears to have impeded
their application in the total synthesis arena.14 We,
therefore, became highly motivated to pursue a total
(8) (a) Barluenga, J.; Lopez, L. A.; Martinez, S.; Tomas, M. Tetra-
hedron 2000, 56, 4967. (b) Wulff, W. D.; Bax, B. M.; Brandvold, T. A.;
Chan, K. S.; Gilbert, A. M.; Hsung, R. P.; Mitchell, J.; Clardy, J.
Organometallics 1994, 13, 102. (c) Wulff, W. D.; Gilbert, A. M.; Hsung,
R. P.; Rahm, A. J. Org. Chem. 1995, 60, 4566.
(9) Yamashita, A. Tetrahedron Lett. 1986, 27, 5915.
(5) Fidalgo, J.; Castedo, L.; Dominguez, D. Tetrahedron Lett. 1993,
34, 7317.
(10) (a) Leese, T.; Do¨tz, K. H. Chem. Ber. 1996, 129, 623. (b) Rahm,
A.; Wulff, W. D. Tetrahedron Lett. 1995, 36, 8753. (c) Chelain, E.;
Parlier, A.; Audouin, M.; Rudler, H.; Daran, J.-C.; Vaissermann, J. J.
Am. Chem. Soc. 1993, 115, 10568.
(6) Moser, W. H.; Sun, L.; Huffman, J. Org. Lett. 2001, 3, 3389.
(7) For recent reviews on synthetic applications of Fischer carbene
complexes, see: (a) de Meijere, A.; Duetsch, M.; Schirmer, H. Angew.
Chem., Int. Ed. Engl. 2000, 39, 3964. (b) Do¨tz, K. H.; Tomuschat, P.
Chem. Soc. Rev. 1999, 28, 187. (c) Harvey, D. F.; Sigano, D. M. Chem.
Rev. 1996, 96, 271. (d) Wulff, W. D. In Comprehensive Organometallic
Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.;
Pergamon: New York, 1995; Vol. 12, p 469. (e) Herndon, J. W. Coord.
Chem. Rev. 2000, 206, 237.
(11) (a) Yamashita, A.; Toy, A.; Watt, W.; Muchmore, C. R. Tetra-
hedron Lett. 1988, 29, 3403. (b) Grotjahn, D. B.; Do¨tz, K. H. Synlett
1991, 381.
(12) (a) Chan, K. S.; Peterson, G. A.; Brandvold, T. A.; Faron, K. L.;
Challener, C. A.; Hyldahl, C.; Wulff, W. D. J. Organometallic Chem.
1987, 334, 9. (b) Bos, M. E.; Wulff, W. D.; Miller, R. A.; Chamberlin,
S.; Brandvold, T. A. J. Am. Chem. Soc. 1991, 113, 9293.
J. Org. Chem, Vol. 70, No. 16, 2005 6223

Giese and Moser
SCHEME 1. Synthesis of Amine 5
tetrafluoroborate to provide alkoxycarbene complex 7 in
moderate to good yields. Subsequent treatment of 7 with
amine 5 generated aminocarbene complex 8, which was
readily methylated under sodium hydride/iodomethane
conditions to afford aminocarbene complex 9. Notably,
aryl(amino)carbene complexes 8 and 9a were obtained
as approximately 4:1 mixtures of rotamers because of
hindered rotation about the nitrogen-carbene carbon
bond. Partial separation of the rotamers via flash chro-
matography could be achieved to characterize the indi-
vidual isomers. However, attempts at complete separa-
tion were not successful, and the originally obtained
mixture was thus utilized in subsequent steps including
cyclopentannulation reactions. This was not anticipated
to pose a problem, as Wulff had previously demonstrated
that the independent thermolysis of rotational isomers
of alkenyl(amino)carbene complexes with nitrogen-
tethered alkynes resulted in identical product distribu-
tions and yields.10b
We also planned to study the effect of the alkyne
substituent on the course of the reaction and thus
modified the carbene complex 9a at this juncture. Ami-
nocarbene complex 9a was subjected to desilylation with
tetrabutylammonium fluoride (TBAF) to yield aminocar-
bene complex 9b possessing a terminal alkyne function-
ality. Aminocarbene complex 9b was then silylated to
afford aminocarbene complexes 9c and 9d bearing tri-
methylsilane (TMS) and triethylsilane (TES) groups,
respectively. Aminocarbene complexes 9b-d were also
isolated as 4:1 mixtures of rotamers, and these mixtures
were also directly utilized in subsequent steps.
synthesis of the alkaloid bulgaramine in which an
intramolecular cyclopentannulation reaction of a Fischer
aminocarbene complex provided the key assembly of the
benzindenoazepine ring system. Success in this venture
would not only provide valuable information regarding
the cyclopentannulation reaction itself, but would also
represent the first natural product synthesis derived from
Fischer aminocarbene complexes.
Results and Discussion
A variety of factors including temperature, solvent,
concentration, and additives are known to affect the
course of thermal reactions between Fischer carbene
complexes and alkynes.7,8,12 With aminocarbene com-
plexes 9a-d in hand, the task was to determine optimal
reaction conditions for a cyclopentannulation reaction
that would lead to the indenobenzazepine skeleton.
The first parameter explored was the solvent (Table
1, entries 1-11). In annulation reactions involving
alkoxycarbene complexes and alkynes, the effect of
solvent on product distribution has been thoroughly
studied. Previous observations are that polar and/or
coordinating solvents such as DMF and acetonitrile favor
cyclopentannulations, whereas less polar solvents such
as benzene or toluene lead to benzannulation as the
predominant pathway. Although one would assume that
polar solvents would have a similar beneficial effect on
pentannulation reactions of aminocarbene complexes,
data to verify this hypothesis are scarce. The most
commonly utilized solvent for pentannulation reactions
with aminocarbene complexes is DMF. As alluded to
previously, this is often a necessity because a solvent with
higher refluxing temperature is typically required to
induce decarbonylation of the robust aminocarbene com-
plexes, which serves as the initial step of the pentannu-
lation reaction.9,11 We were, therefore, surprised to find
that the desired pentannulation transformation of ami-
nocarbene complex 9a proceeded in refluxing THF at a
reaction temperature that would be anticipated to be too
low for the decarbonylation step. In contrast, thermolysis
of 9a in DMF at 120 °C resulted in no reaction and
raising the temperature to 152 °C caused decomposition
with no observable product formation. Similarly, ther-
The synthesis of appropriately substituted Fischer
aminocarbene complexes began with the preparation of
requisite amine 5, as summarized in Scheme 1. Regio-
selective ortho iodination of 3,5-dimethoxyphenethyl
alcohol (1) was achieved with AgO2CCF3/I2 to produce
aryl iodide 2,15 which was subsequently utilized in a
Sonogashira coupling with (triisoproylsilyl)acetylene to
provide aryl alkyne 3 in good overall yield. The alcohol
functionality was then converted to an amino group.
Specifically, phthalimide derivative 4 was prepared from
alcohol 3 under Mitsunobu conditions and then treated
with excess methylamine to afford the desired amine 5.16
Construction of the desired Fischer aminocarbene
complexes, depicted in Scheme 2, began with aryl bro-
mide 6, which was prepared according to literature
methods.17 Metal-halogen exchange of 6 was followed
by the addition of the resultant aryllithium to chromium
hexacarbonyl and methylation with trimethyloxonium
(13) (a) Do¨tz, K. H.; Popall, M. Angew. Chem., Int. Ed. Engl. 1987,
26, 1158. (b) Boger, D. L.; Huter, O.; Mbiya, K.; Zhang, M. J. Am. Chem.
Soc. 1995, 117, 11839. (c) Bao, J.; Wulff, W. D.; Dragisich, V.;
Wenglowsky, S.; Ball, R. G. J. Am. Chem. Soc. 1994, 116, 7616. (d)
Do¨tz, K. H.; Pruskil, I.; Muhlemeier, J. Chem. Ber. 1982, 115, 1278.
(e) Semmelhack, M. F.; Bozell, J. J.; Keller, T.; Sato, T.; Spiess, E.;
Wulff, W. D.; Zask, A. Tetrahedron 1985, 41, 5803. (f) Wulff, W. D.;
Xu, Y.-C. J. Am. Chem. Soc. 1988, 110, 2312. (g) Parker, K. A.; Coburn,
C. A. J. Org. Chem. 1991, 56, 1666. (h) King, J.; Quayle, P.; Malone,
J. F. Tetrahedron Lett. 1990, 31, 5221. (i) White, J. D.; Smits, H. Org.
Lett. 2005, 7, 235. (j) Pulley, S. R.; Czako, B. Tetrahedron Lett. 2004,
45, 5511.
(14) For studies towards substituted lycoranes, see: Bouaucheau,
C.; Parlier, A.; Rudler, H. J. Org. Chem. 1997, 62, 7247.
(15) Mulholland, G. K.; Zheng, Q. Synth. Commun. 2001, 31, 3059.
(16) Sen, S. E.; Roach, S. L. Synthesis 1995, 756.
(17) Klix, R. C.; Cain, M. H.; Bhatia, A. V. Tetrahedron Lett. 1995,
36, 6413.
6224 J. Org. Chem., Vol. 70, No. 16, 2005

Total Synthesis of Bulgaramine
SCHEME 2. Synthesis of Aminocarbene Complexes
molysis in other polar coordinating solvents (acetonitrile
or pyridine), polar noncoordinating solvents (chloroform),
or nonpolar solvents (benzene or toluene) at their reflux
temperatures resulted in decomposition of the starting
aminocarbene complex with formation of little or none
of the desired pentannulation product. When considering
the successful pentannulation reaction, it is possible that
the alkyne or oxygen substituents on the carbene complex
may serve as intramolecular two-electron donors to the
metal center, thereby decreasing the activation barrier
to initial decarbonylation. However, this hypothesis still
does not appear to explain the failure of the reaction with
solvents other than THF.
Although the desired cyclopentannulation product was
the major component observed in thermal reactions with
THF, the conversion was relatively poor and multiple
byproducts were also being formed.18 In studies with
alkoxycarbene complexes, it has previously been demon-
strated that reducing the concentration of the alkyne
substrate can favor cyclopentannulation pathways rela-
tive to benzannulations because of a decrease of the
allochemical effect.19 More specifically, it has been pro-
posed that the presence of alkynes as available ligands
can facilitate CO insertions by serving as 4π-electron
donors to the metal center, thereby avoiding higher
energy electron-deficient organometallic intermediates
that would normally ensue from the CO insertion step.
Assuming that one or more of the desired byproducts
could derive from pathways involving CO insertions, the
thermolysis of 9a in THF was repeated at a more dilute
concentration (0.008 M vs 0.022 M). This alteration did
result in a modest increase in the yield of 10 (Table 1,
entry 12). However, in the absence of further experimen-
tation, we can only surmise that this increase may be
due to a reduction of the allochemical effect or a mini-
mization of intermolecular cyclizations.
To further probe the effect of electron donors on the
efficiency of the reaction, a series of experiments were
attempted in which ligating reagents were added to the
reaction solution. It has been reported by both De-
Meijere19 and Wulff8c that the presence of ligating
reagents increases the selectivity for cyclopentannulation
pathways in reactions with Fischer alkoxycarbene com-
plexes. Indeed, we were pleased to find that the addition
of 1 equiv of either tri-n-butylphosphine, triphenylphos-
phine, or pyridine resulted in a significant increase in
the yield of 10, along with an essentially complete
suppression of byproduct formation (Table 1, entries 13-
17). This result is consistent with the ability of a two-
electron donor to stabilize reaction intermediates leading
to chromacycle formation and reductive elimination,
which are the hypothesized key steps leading to the
cyclopentannulation product. Although conversion was
much slower in the presence of added pyridine, the
desired product appeared to be completely stable to the
reaction conditions. The reaction time of all subsequent
experiments was therefore extended to ensure complete
conversion. Because the yield of 10 was essentially the
same in the presence of all ligands that were attempted,
tri-n-butylphosphine was employed in all future experi-
ments due to the ease of workup.
(18) Isolation of reaction products was extremely difficult because
of oxidative degradation during flash chromatography on silica gel.
Initially, the product yields and/or ratios were estimated by 1H NMR
analysis of the crude reaction mixtures. Fortuitously, it was discovered
that treatment of THF solutions of the crude reaction mixtures with
methanolic HCl allowed the desired product 10 to be cleanly isolated
as its HCl salt.
The final parameter explored was the alkyne substitu-
ent. Subjecting aminocarbene complex 9b, containing a
terminal alkyne, to the optimal conditions observed for
9a resulted in a rapid disappearance of the carbene
complex. After 6 h, 1H NMR analysis of the crude reaction
mixture indicated that bulgaramine was being formed
(19) Flynn, B. L.; Schirmer, H.; Duetsch, M.; de Meijere, A. J. Org.
Chem. 2001, 66, 1747.
J. Org. Chem, Vol. 70, No. 16, 2005 6225

Giese and Moser
TABLE 1. Cyclopentannulation of Aminocarbene
difficult, and analytically pure samples could not be
isolated. Although these experiments indicate that car-
bene complex 9a, in which the pendant alkyne bears a
TIPS group, provides the most effective precursor in the
context of the bulgaramine synthesis, they do little to
shed light on the effect of the steric bulk of the silyl group
on the annulation process. That is, although the dimin-
ished yields observed with the TMS- or TES-substituted
alkynes suggest a reduced efficiency for the annulation
process, it is possible that the greater fragility of these
groups relative to the TIPS group leads to decomposition
following the annulation step.
Taken together, these results are consistent with the
following mechanism, as discussed by others in the
context of related results (Figure 3). Reversible decar-
bonylation of A to yield tetracarbonyl complex B occurs
at an uncharacteristically low temperature for aminocar-
bene complexes and appears to be aided by the presence
of THF as a donor solvent. Intramolecular chelation by
oxygen atoms has been postulated in related examples
to facilitate decarbonylation and may also have the same
effect in this case.21 However, the unreactivity of carbene
complexes 9a-d in solvents other than THF suggests
that the solvent must play the key role. Intramolecular
displacement of the solvent by the alkyne provides
intermediate C, which would then insert the alkyne to
yield the 18-electron allylidene chelate complex D. Co-
ordination of the two-electron donor phosphine ligand
dechelates the alkyene moiety in intermediate D to give
the 18-electron vinyl carbene complex E and induces
chromacycle formation and reductive elimination to
afford cyclopentadiene G. At this point, rearomatization
could occur to yield two possible isomers, allylamine H
or enamine I. Although Do¨tz has reported a similar
cyclopentannulation where tautomerization occurs via a
suprafacial [1,5]-H shift to afford an enamine intermedi-
ate,21 only the allylamine is observed on thermolysis of
9a, 9c, or 9d. This appears to be a consequence of
stabilization by the silyl group because thermolysis of
carbene complex 9b, containing a terminal alkyne,
proceeds directly to bulgaramine, which contains the
enamine moiety.
With the optimal conditions for the cyclopentannula-
tion reaction now established, our attention was turned
toward the conversion of the benzindenoazepine skeleton
10a to bulgaramine via desilylation and double bond
migration (Scheme 3). The resiliency of the TIPS group,
which allowed precipitation of 10a as the HCl salt
without desilylation,18 rendered this transformation dif-
ficult. A variety of common desilylating reagents were
attempted, including TBAF, aqueous acids and bases,
and anhydrous organic acids such as methanesulfonic
acid and TFA, all with no success. Desilylation was
finally accomplished by the addition of 10% HBr/HOAc
in dry dichloromethane, which afforded two products as
indicated by 1H NMR analysis of the crude reaction
mixture. These two products were presumed to be alkene
regioisomers because interconversion of similar regio-
isomers has previously been reported.22 Indeed, treat-
ment of this mixture for 5 min in a 10% solution of KOH
Complexes 9a-d
temp time
solvent (°C) (h)
yield
of 10
entry
R
M
additive
1
2
3
4
5
6
7
8
9
TIPS benzene
TIPS benzene
80 24 0.022 none
80 72 0.022 none
no reaction
decompa
tracea
TIPS toluene 111 48 0.022 none
TIPS DMF
TIPS DMF
TIPS ACN
TIPS CDCl3
TIPS CDCl3
120 20 0.022 none
152 24 0.022 none
trace
decomp
decomp
no reaction
decomp
no reaction
decomp
24%
82
62
4
4
0.022 none
0.022 none
62 24 0.022 none
0.022 none
TIPS pyridine 66
8
10 TIPS pyridine 80 18 0.022 none
11 TIPS THF
12 TIPS THF
13 TIPS THF
66
66
66
8
8
8
0.022 none
0.008 none
0.022 PBu3
32%
62%
(1.0 equiv)
14 TIPS THF
15 TIPS THF
16 TIPS THF
17 TIPS THF
66
8
0.022 PPh3
63%
62%
67%b
0%c
(1.0 equiv)
66 20 0.022 pyridine
(1.0 equiv)
66 20 0.008 PBu3
(1.0 equiv)
66 20 0.008 PBu3
(3.0 equiv)
0.008 PBu3
(1.0 equiv)
66 20 0.008 PBu3
(1.0 equiv)
66 20 0.008 PBu3
(1.0 equiv)
18
H
THF
66
6
38%
(enamine)d
56%
19 TMS THF
20 TES THF
(1:1 mix)e
37%e
a Minor amounts of starting carbene obtained. b Average of 6
trials. c No starting carbene or cyclopentannulation product identi-
fied by crude NMR. d Bulgaramine was obtained directly as the
only product. e Contained inseparable impurities.
directly as the only identifiable product. Isolation of this
product from the crude reaction mixture, however, proved
to be difficult because of the propensity for oxidation
under chromatographic conditions. Fortunately, we ob-
served that the use of argon pressure rather than air
during flash chromatographic purification minimized
oxidative degradation, affording a 38% isolated yield of
bulgaramine (Table 1, entry 18). The fact that bulgar-
amine, which contains an enamine moiety, was generated
rather than the allylamine functionality present in
product 10 attests to the ability of the silyl group to
localize the double bond following cyclopentannulation.
The diminished yield obtained with 9b also mirrors
previous literature reports in which intramolecular reac-
tions of alkoxycarbene complexes containing terminal
alkynes have met with disappointing results.13e,20
Aminocarbene complexes 9c and 9d were also sub-
jected to the identical reaction conditions, and crude H
NMR analysis of the reaction mixtures indicated forma-
tion of the desired cyclopentannulation products along
with varying amounts of byproducts (Table 1, entries 19
and 20). Isolating the products again proved to be
1
(21) Do¨tz, K. H.; Schafer, T.; Harms, K. Angew. Chem., Int. Ed. Engl.
(20) (a) Wulff, W. D.; McCallum, J. S.; Kunng, F.-A. J. Am. Chem.
Soc. 1988, 110, 7419. (b) Gross, M. F.; Finn, M. G. J. Am. Chem. Soc.
1994, 116, 10921.
1990, 29, 176.
(22) Irie, H.; Tani, S.; Yamane, H. J. Chem. Soc., Perkin Trans. 1
1972, 2986.
6226 J. Org. Chem., Vol. 70, No. 16, 2005

Total Synthesis of Bulgaramine
FIGURE 3. Proposed mechanism for the optimized cyclopentannulation reaction.
SCHEME 3. Completion of Bulgaramine via
Cyclopentannulation Reaction
by the addition of two-electron donor ligands, demon-
strating the first example of this effect on the thermal
reaction of aminocarbene complexes with alkynes. We
anticipate that the methodology reported here will prove
useful in the synthesis of additional indenobenzazepine
alkaloids.
Experimental Section
2-(2-Iodo-4,5-dimethoxyphenyl)ethanol (2). To a solu-
tion of 3,4-dimethoxy phenethyl alcohol (5.02 g, 27.6 mmmol)
and AgO2CCF3 (7.00 g, 31.7 mmol) in CHCl3 at 0 °C was added
I2 (8.04 g, 31.7 mmol) in 150 mL of CHCl3 via cannula. After
the addition, stirring was continued for 15 min. The mixture
was then filtered through Celite, and the filtrate was washed
sequentially with saturated aqueous Na2S2O5, water, and
brine. The organic layer was then dried over MgSO4, filtered,
and concentrated. Purification of the residue via flash chro-
matography (60 g of SiO2, ramp eluent from 2:1 to 1:2 hexane/
ethyl acetate) afforded 2 as a white solid after trituration with
1
Et2O (8.25 g, 97%): IR (neat) 3499 cm-1; H NMR (CDCl3) δ
in a 1:1 H2O/EtOH mixture afforded analytically pure
bulgaramine in 80% yield as the sole product. The
spectroscopic properties of this synthetic material were
identical with those reported for the naturally occurring
compound.3
7.21 (s, 1H), 6.77 (s, 1H), 3.84 (s, 3H), 3.83 (s, 3H), 3.82 (m,
2H), 2.93 (t, J ) 6.6 Hz, 2H), 1.46 (s, 1H); 13C NMR (CDCl3) δ
149.5, 148.4, 133.6, 122.0, 113.2, 88.3, 62.5, 56.2, 56.0, 43.3;
HRMS (ESI) [M + Na]+ m/z calcd for C10H13NaO3I 330.9807,
found 330.9807.
2-{4,5-Dimethoxy-2-[(triisopropylsilanyl)ethynyl]-
phenyl}ethanol (3). A solution of aryl iodide 2 (7.93 g, 25.7
mmol), triisopropylsilyl acetylene (6.35 mL, 28.3 mmol), and
freshly distilled triethylamine (10.76 mL, 77.2 mmol) in 100
mL of acetonitrile was freeze-thaw degassed three times. To
this solution was added CuI (980 mg, 5.20 mmol) and PdCl2-
(PPh3)2 (1.81 g, 2.60 mmol), and the red mixture was stirred
at room temperature overnight. The resulting black solution
was filtered through Celite, and the filtrate was washed
sequentially with saturated aqueous NH4Cl, water, and brine.
The organic layer was then dried over MgSO4, filtered, and
concentrated to an orange solid. The orange solid was recrys-
tallized from boiling hexane and purified via flash chroma-
tography (70 g of SiO2, ramp eluent from 3:2 to 2:3 hexane/
ethyl acetate) to afford 3 as a pale yellow solid (7.04 g, 75%):
Conclusions
In conclusion, a rapid and efficient synthesis of bul-
garamine has been accomplished with a longest linear
sequence of eight steps and an overall yield of 23% from
commercially available 3,4-dimethoxyphenethyl alcohol.
An intramolecular cyclopentannulation reaction of a
Fischer aminocarbene complex provided the key step and
occurred under significantly milder conditions and in
higher yields than other reported examples of this
reaction type. The reaction solvent was a critical factor
in the cyclopentannulation reaction, with measurable
amounts of the desired product observed only when THF
was utilized. The product yield could be further enhanced
IR (neat) 3503, 2143 cm-1
;
1H NMR (CDCl3) δ 6.93 (s, 1H),
J. Org. Chem, Vol. 70, No. 16, 2005 6227

Giese and Moser
6.72 (s, 1H), 3.87 (m, 2H), 3.87 (s, 3H), 3.86 (s, 3H), 3.01 (t, J
) 6.6 Hz, 2H), 1.45 (s, 1H), 1.11 (s, 21H); 13C NMR (CDCl3) δ
149.6, 147.4, 134.5, 115.5, 115.2, 112.9, 105.6, 92.8, 63.1, 56.1,
56.0, 37.9, 18.7, 11.4; HRMS (ESI) [M + Na]+ m/z calcd for
C21H34NaO3Si 385.2175, found 385.2171.
mixture of rotamers) as a fluffy yellow solid (7.48 g, 98%). For
characterization purposes, partial separation of the rotamers
could be achieved via flash chromatography utilizing a solvent
system of 4:1 hexane/ether. Minor rotamer: IR (neat) 3229,
2141, 2054, 1979, 1917 cm-1 1H NMR (CDCl3) δ 8.80 (br s,
;
1H), 6.91 (s, 1H), 6.75 (m, 3H), 6.35 (d, J ) 8.8 Hz, 1H), 5.84
(s, 2H), 4.39 (m, 2H), 3.87 (s, 3H), 3.84 (s, 3H), 3.30 (m, 2H),
1.05 (s, 21H); 13C NMR (CDCl3) δ 275.0, 223.5, 217.1, 150.1,
147.9, 147.5, 138.0, 136.0, 132.4, 121.9, 115.7, 115.3, 115.2,
112.1, 108.0, 105.1, 100.9, 94.1, 56.1, 55.9, 53.7, 33.5, 18.7, 11.3,
10.5; HRMS (EI) [M]+ m/z calcd for C34H39NO9CrSi 685.1799,
found 685.1811. Major rotamer: IR (neat) 3347, 3291, 2141,
2054, 1979, 1917 cm-1; 1H NMR (CDCl3) δ 8.98 (br s, 1H), 6.93
(s, 1H), 6.81 (t, J ) 7.4 Hz, 3H), 6.67 (d, J ) 8.8 Hz, 1H), 6.11
(d, J ) 7.4 Hz, 1H), 5.96 (s, 1H), 5.84 (s, 1H), 3.87 (s, 6H),
3.60 (m, 2H), 2.99 (m, 2H), 1.08 (s, 21H); 13C NMR (CDCl3) δ
277.8, 224.6, 218.6, 151.6, 149.7, 149.0, 138.0, 133.6, 132.4,
123.8, 117.5, 116.7, 115.7, 114.0, 108.9, 106.3, 102.7, 95.6, 57.8,
57.5, 52.7, 36.0, 20.3, 12.9; HRMS (EI) [M]+ m/z calcd for
C34H39NO9CrSi 685.1799, found 685.1832.
1-(2-{4,5-Dimethoxy-2-[(triisopropylsilanyl)ethynyl]-
phenyl}ethyl)-3-propenyl-4-vinylpyrrole-2,5-dione (4). To
a solution of alcohol 3 (6.90 g, 19.0 mmol), phthalimide (2.94
g, 19.9 mmol), and PPh3 (5.24 g, 19.9 mmol) in 75 mL of THF
at 0 °C was added diisopropyl azodicarboxylate (3.91 mL, 19.9
mmol) dropwise via syringe. The resulting mixture was
allowed to warm to room temperature overnight with stirring.
The mixture was then partitioned between saturated aqueous
NaHCO3 and EtOAc, and the aqueous layer was extracted with
additional portions of EtOAc (3 × 30 mL). The combined
organic layers were washed with brine, dried over MgSO4,
filtered, and concentrated to a tan solid. Recrystallization from
methanol provided 4 as a white crystalline solid (8.19 g,
88%): IR (neat) 2145, 1712, 1767 cm-1 1H NMR (CDCl3) δ
;
7.76 (m, 2H), 7.66 (m, 2H), 6.90 (s, 1H), 6.51 (s, 1H), 3.99 (t, J
) 6.6 Hz, 2H), 3.83 (s, 3H), 3.60 (s, 3H), 3.16 (t, J ) 6.6 Hz,
2H), 1.13 (s, 21H); 13C NMR (CDCl3) δ 168.0, 149.3, 147.3,
133.9, 133.8, 132.1, 123.1, 115.4, 115.3, 112.1, 105.2, 93.0, 55.9,
55.7, 38.2, 32.7, 18.7, 11.4; HRMS (ESI) [M]+ m/z calcd for
C29H37NO4Si 491.2492, found 491.2502.
Aminocarbene Complex 9a. A solution of aminocarbene
complex 8 (7.48 g, 10.9 mmol) in 110 mL of THF was cooled to
-78 °C, and NaH (60% dispersion in mineral oil, 622 mg, 15.5
mmol) and iodomethane (968 µL, 15.5 mmol) were added. The
resultant mixture was allowed to warm to room temperature
and then quenched with saturated aqueous NH4Cl. The
organic layer was washed with water and brine, dried over
MgSO4, filtered, and concentrated to a yellow oil. Purification
via flash chromatography (120 g of SiO2, 3:1 hexane/ethyl
acetate eluent) provided 9a (inseparable 4:1 mixture of rota-
mers) as a fluffy yellow solid (6.96 g, 91%). For characterization
purposes, partial separation of the rotamers could be achieved
via flash chromatography using 4:1 hexane/ether eluent. Minor
2-{4,5-Dimethoxy-2-[(triisopropylsilanyl)ethynyl]-
phenyl}ethylamine (5). Phthalimide derivative 4 (8.19 g,
16.7 mmol) was dissolved in 330 mL of a 2:1 mixture of EtOH/
THF, and 83 mL of 40% aqueous methylamine was added. The
solution was warmed to 70 °C for 4 h and then diluted with
ethyl acetate and washed sequentially with saturated aqueous
NH4Cl and water. The organic layer was dried over MgSO4,
filtered, and concentrated to a white solid. Purification via
flash chromatography (60 g of Al2O3; ramp eluent from EtOAc
to 3:1 EtOAc/MeOH to MeOH with 10% NH3/MeOH) provided
rotamer: IR (neat) 2142, 2054, 1973, 1917 cm-1 1H NMR
;
(CDCl3) δ 6.90 (m, 3H), 6.66 (d, J ) 6.6 Hz, 1H), 6.28 (d, J )
6.6 Hz, 1H), 5.99 (s, 1H), 5.84 (s, 1H), 4.50 (t, J ) 8.1 Hz, 2H),
3.874 (s, 3H), 3.871 (s, 3H), 3.36 (m, 2H), 3.07 (s, 3H), 1.11 (s,
21H); 13C NMR (75 MHz, CDCl3) δ 269.9, 223.4, 217.2, 150.1,
147.8, 147.7, 135.1, 135.0, 132.3, 122.6, 115.9, 114.9, 113.3,
112.6, 106.6, 105.6, 101.0, 93.5, 63.7, 56.1, 55.9, 43.7, 33.5, 18.7,
11.4; HRMS (EI) [M - 5CO]+ m/z calcd for C30H41NO4CrSi
559.2210, found 559.2184 (because of low volatility and
thermal instability this was the only peak observed). Major
5 as a white solid (4.41 g, 73%): IR (neat) 3368, 2143 cm-1
;
1H NMR (CDCl3) δ 6.91 (s, 1H), 6.69 (s, 1H), 3.86 (s, 3H), 3.85
(s, 3H), 2.95 (m, 2H), 2.89 (m, 2H), 1.62 (br s, 2H), 1.11 (s,
21H); 13C NMR (CDCl3) δ 149.6, 147.2, 135.7, 115.5, 115.4,
112.6, 105.8, 92.5, 56.1, 56.0, 43.0, 38.7, 18.7, 11.4; HRMS
(ESI) [M + H]+ m/z calcd for C21H36NO2Si 362.2515, found
362.2516.
Alkoxycarbene Complex 7. To a solution of aryl bromide
6 (5.43 g, 27.0 mmol) in 50 mL of THF at -78 °C was added
n-BuLi (1.6 M in hexane, 17.7 mL, 28.4 mmol) dropwise via
syringe, and the mixture was stirred at -78 °C for 1.5 h. In a
separate flask, Cr(CO)6 (6.24 g, 28.4 mmol) was suspended in
50 mL of THF, cooled to -78 °C, and the aryllithium was
transferred to this slurry via cannula. The resulting mixture
was warmed to room temperature and allowed to stir for 3 h.
The volume of the solvent was reduced, and the resulting solids
were removed by filtration through Celite. The filtrate was
then concentrated to a thick yellow oil. This residue was taken
up in 30 mL of H2O, Me3OBF4 (4.79 g, 32.4 mmol) was added,
and the mixture was stirred for 15 min to afford a thick red
solution. This solution was extracted with EtOAc (5 × 50 mL),
and the combined organic layers were dried over MgSO4,
filtered, and concentrated to a red oil. Purification by flash
chromatography (100 g of SiO2, 9:1 hexane/ethyl acetate) and
recrystallization from hexane provided 7 as a red crystalline
rotamer: IR (neat) 2144, 2054, 1973, 1917 cm-1 1H NMR
;
(CDCl3) δ 6.83 (m, 3H), 6.65 (d, J ) 7.4 Hz, 1H), 6.18 (s, 1H),
6.01 (s, 1H), 5.81 (s, 1H), 3.91 (s, 3H), 3.83 (s, 3H), 3.74 (s,
3H), 3.63 (m, 2H), 2.99 (m, 2H), 1.11 (s, 21H); 13C NMR (CDCl3)
δ 271.8, 223.6, 217.0, 149.9, 147.8, 147.6, 135.4, 134.1, 132.1,
122.1, 115.4, 114.9, 113.9, 111.6, 106.4, 104.8, 100.9, 93.1, 59.0,
56.1, 55.7, 49.2, 33.3, 18.7, 11.4; HRMS (EI) [M - 5CO]+ m/z
calcd for C30H41NO4CrSi 559.2210, found 559.2195 (because
of low volatility and thermal instability this was the only peak
observed).
Aminocarbene Complex 9b. A solution of aminocarbene
complex 9a (3.50 g, 5.00 mmol) in 100 mL of THF was cooled
to 0 °C, and TBAF (1.0 M in THF, 6.11 mL, 6.11 mmol) was
added dropwise via syringe. The solution was allowed to stir
at 0 °C for 30 min and then was diluted with Et2O and
quenched by the addition of saturated aqueous NH4Cl. The
aqueous layer was extracted with EtOAc (2 × 30 mL), and
the combined organic layers were washed with brine, dried
over MgSO4, filtered, and concentrated to a yellow oil. Puri-
fication via flash chromatography (100 g SiO2, 3:1 hexane/ethyl
acetate eluent) provided 9b (inseparable 4:1 mixture of rota-
mers) as a fluffy yellow solid (1.90 g, 69%). Because of the
limited success of this substrate in cyclopentannulation ex-
periments, no further attempts were made to separate the
rotamers and the mixture of rotamers was characterized by
1H NMR only. The chemical signals for the major and minor
rotamers were assigned on the basis of the integral values.
Chemical signals for the minor rotamer are reported as
normalized (×4) values. Minor rotamer: 1H NMR (CDCl3) δ
solid (5.46 g, 57%): IR (neat) 2064, 1929 cm-1 1H NMR
;
(CDCl3) δ 6.85 (m, 2H), 6.61 (d, 1H, J ) 8.1 Hz), 5.97 (s, 2H),
4.48 (s, 3H); 13C NMR (CDCl3) δ 312.0, 224.5, 215.8, 147.8,
138.0, 135.1, 121.8, 116.2, 109.4, 101.4, 66.3; HRMS (EI) [M]+
m/z calcd for C14H8O8Cr 355.9624, found 355.9634.
Aminocarbene Complex 8. To a solution of alkoxycarbene
complex 7 (3.95 g, 11.1 mmol) in 75 mL of Et2O at 0 °C was
added a solution of amine 5 (4.41 g, 12.2 mmol) in 60 mL of
THF via cannula. The resulting pale orange solution was
stirred at 0 °C for 1 h and then concentrated to a yellow oil.
Purification of the yellow oil via flash chromatography (120 g
of SiO2, 3:1 hexane/ethyl acetate) provided 8 (inseparable 4:1
6228 J. Org. Chem., Vol. 70, No. 16, 2005

Total Synthesis of Bulgaramine
6.87 (m, 3H), 6.65 (m, 1H), 6.28 (d, J ) 8.1 Hz, 1H), 5.96 (s,
1H), 5.83 (s, 1H), 4.47 (m, 2H), 3.86 (s, 6H), 3.31 (m, 2H), 3.25
(s, 1H), 3.14 (s, 3H). Major rotamer: 1H NMR (CDCl3) δ 6.89
(m, 2H), 6.65 (m, 1H), 6.22 (s, 1H), 6.17 (d, J ) 8.1 Hz, 1H),
6.01 (s, 1H), 5.81 (s, 1H), 3.96 (s, 3H), 3.81 (s, 3H), 3.74 (s,
3H), 3.64 (m, 2H), 3.16 (s, 1H), 2.97 (m, 2H).
evacuation and backfilling with argon (4×). To this solution
was added PBu3 (360 µL, 1.43 mmol), and the mixture was
heated to reflux for 18 h during which time the color changed
from yellow to dark orange. The mixture was concentrated,
redissolved in 30 mL of THF, and anhydrous HCl (1.0 M in
Et2O, 7.25 mL, 7.25 mmol) was added dropwise via syringe.
The mixture was stirred for 15 min during which time a
precipitate formed. The precipitate was collected by filtration,
dissolved in CHCl3, and neutralized with saturated aqueous
NaHCO3. The organic layaer was then washed with brine,
dried over MgSO4, filtered, and concentrated to give a yellow
solid after trituration with Et2O. Recrystallization from boiling
EtOH at 0 °C afforded cyclopentadiene 10a as a yellow
crystalline solid (479 mg, 68%): 1H NMR (CD2Cl2) δ 6.96 (d, J
) 8.1 Hz, 1H), 6.93 (s, 1H), 6.68 (d, J ) 8.1 Hz, 1H), 6.66 (s,
1H), 5.98 (d, J ) 4.4 Hz, 2H), 4.31 (s, 1H), 3.84 (s, 3H), 3.83
(s, 3H), 3.62 (m, 1H), 3.14 (m, 1H), 2.86 (m, 2H), 2.88 (s, 3H),
0.88 (m, 21H); 13C NMR (CD2Cl2) δ 147.7, 147.1, 145.9, 141.7,
140.0, 139.2, 135.1, 129.4, 126.8, 126.6, 116.4, 114.8, 112.9,
104.7, 100.8, 57.1, 56.6, 56.2, 44.3, 40.7, 33.9, 19.4, 19.1, 12.4;
HRMS (ESI) [M + H]+ m/z calcd for C30H42NO4Si 508.2883,
found 508.2887.
Bulgaramine 11. A solution of cyclopentadiene 10a (789
mg, 1.55 mmol) in 75 mL of DCM was cooled to 0 °C and 7.5
mL of 30% HBr in HOAc was added via syringe. The ice bath
was removed, and the solution was allowed to stir at room
temperature for 6 h during which time the color changed from
yellow to blue-green. The solution was diluted with CHCl3 and
neutralized with saturated aqueous K2CO3. The aqueous layer
was extracted with additional portions of CHCl3 (3 × 30 mL),
and the combined organic layers were washed with brine, dried
over MgSO4, filtered, and concentrated to give an orange solid.
This solid was suspended in 70 mL of 10% KOH in a 1:1
mixture of H2O/EtOH, stirred under reflux for 5 min, and
sealed under argon and placed in a -10 °C freezer overnight.
The solid which precipitated was collected by filtration and
partitioned between H2O and CH2Cl2. The aqueous layer was
extracted with additional portions of CH2Cl2 (3 × 30 mL), and
the combined organic layers were dried over MgSO4, filtered,
and concentrated to give an orange solid. This residue was
then suspended in 30 mL of Et2O, heated to reflux for 5 min,
and collected via filtration to afford 11 as a pale orange powder
(436 mg, 80%): 1H NMR (CHCl3) δ 7.00 (s, 1H), 6.88 (d, J )
7.4 Hz, 1H), 6.69 (d, J ) 7.4 Hz, 1H), 6.69 (s, 1H), 6.01 (s,
2H), 3.92 (s, 3H), 3.88 (s, 3H), 3.81 (s, 2H), 3.21 (m, 2H), 2.97
(m, 2H), 2.91 (s, 3H); 13C NMR (CHCl3) δ 147.0, 146.8, 146.4,
144.1, 138.6, 136.1, 133.4, 127.8, 126.9, 121.0, 116.2, 113.0,
110.9, 105.4, 100.5, 56.1, 55.9, 53.6, 43.1, 39.7, 33.7; HRMS
(ESI) [M + H]+ m/z calcd for C21H22NO4 352.1549, found
352.1546.
Aminocarbene Complex 9c. To a solution of aminocar-
bene complex 9b (799 mg, 1.47 mmol) in 25 mL of THF at
-78 °C was added n-BuLi (1.6 M in hexanes, 964 µL, 1.54
mmol) via syringe. After maintaining the solution at -78 °C
for 1 h, we added TMSCl (559 µL, 4.41 mmol) via syringe. The
resultant solution was allowed to warm to room temperature
with stirring overnight. The mixture was then partitioned
between saturated aqueous NaHCO3 and Et2O, and the
aqueous layer was extracted with additional portions of Et2O
(3 × 30 mL). The combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated to a yellow
oil. Purification via flash chromatography (40 g of SiO2, 3:1
hexane/ether eluent) provided 9c (inseparable 4:1 mixture of
rotamers) as a fluffy yellow solid (538 mg, 59%). Because of
the limited success of this substrate in cyclopentannulation
experiments, no further attempts were made to separate the
rotamers and the mixture of rotamers was characterized by
1H NMR only. The chemical signals for the major and minor
rotamers were assigned on the basis of the integral values.
Chemical signals for the minor rotamer are reported as
normalized (×4) values. Minor rotamer: 1H NMR (CDCl3) δ
6.92 (m, 4H), 6.66 (m, 1H), 6.26 (m, 1H), 6.12 (s, 1H), 5.83 (s,
1H), 4.44 (m, 2H), 3.87 (s, 3H), 3.86 (s, 3H), 3.29 (m, 2H), 3.17
(s, 3H), 0.21 (s, 9H). Major rotamer: 1H NMR (CDCl3) δ 6.86
(m, 2H), 6.66 (m, 1H), 6.17 (m, 1H), 6.14 (s, 1H), 6.01 (s, 1H),
5.80 (s, 1H), 3.95 (s, 3H), 3.82 (s, 3H), 3.72 (s, 3H), 3.61 (m,
2H), 2.96 (m, 2H), 0.25 (s, 9H).
Aminocarbene Complex 9d. To a solution of aminocar-
bene complex 9b (799 mg, 1.47 mmol) in 25 mL of THF at
-78 °C was added n-BuLi (1.6 M in hexanes, 964 µL, 1.54
mmol) via syringe. After maintaining the solution at -78 °C
for 1 h, we added TESCl (271 mL, 1.62 mmol) via syringe. The
resultant solution was allowed to warm to room temperature
with stirring overnight. The mixture was then partitioned
between saturated aqueous NaHCO3 and Et2O, and the
aqueous layer was extracted with additional portions of Et2O
(3 × 30 mL). The combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated to a yellow
oil. Purification via flash chromatography (40 g of SiO2, 3:1
hexane/ether eluent) provided 9d (inseparable 4:1 mixture of
rotamers) as a fluffy yellow solid (541 mg, 59%). Because of
the limited success of this substrate in cyclopentannulation
experiments, no further attempts were made to separate the
rotamers and the mixture of rotamers was characterized by
1H NMR only. The chemical signals for the major and minor
rotamers were assigned on the basis of the integral values.
Chemical signals for the minor rotamer are reported as
normalized (×4) values. Minor rotamer: 1H NMR (CDCl3) δ
6.88 (m, 3H), 6.67 (m, 1H), 6.28 (m, 1H), 6.01 (s, 1H), 5.83 (s,
1H), 4.45 (m, 2H), 3.87 (s, 3H), 3.86 (s, 3H), 3.32 (m, 2H), 3.12
(s, 3H), 0.65 (m, 15H). Major rotamer: 1H NMR (CDCl3) δ 6.85
(m, 2H), 6.65 (m, 1H), 6.16 (m, 1H), 6.15 (s, 1H), 6.01 (s, 1H),
5.81 (s, 1H), 3.92 (s, 3H), 3.82 (s, 3H), 3.73 (s, 3H), 3.62 (m,
2H), 2.97 (m, 2H), 0.99 (m, 15H).
Acknowledgment. Financial support provided by
the American Chemical Society Petroleum Research
Fund (38338-AC1), and the Indiana University-Purdue
University Indianapolis School of Science is gratefully
acknowledged.
Supporting Information Available: Copies of NMR
spectra for all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
Cyclopentadiene 10a. A solution of aminocarbene complex
9a (1.00 g, 1.43 mmol) in 180 mL of THF was degassed by
JO050631H
J. Org. Chem, Vol. 70, No. 16, 2005 6229