Cascade [l,3]-sigmatropic rearrangements of ketene O,O-acetals: Kinetic and DFT level mechanistic studies

Article abstract of DOI:10.1021/jo902520z

Chemical Equation Presented The regioisomeric a-cyano ketene-O,O-dialkyl acetals 2a-e and 4a-e, sequential intermediates in the diazomethane induced conversion of indole α-cyano-γ-lactones la-e to 2-indolyl cyanomalonates 5a-e, were isolated and characterized. Formation, of the steady-state intermediate cycloprop[b]indoles 3a-e was evidenced by means of NMR and confirmed by the X-ray structure of 3c, demonstrating that the formation of 5a-e from 2a-e proceeds through two consecutive and one parallel unimolecular steps, with intermediates 3a-e formed, in reversible processes. Evidence that the reversible reactions proceed via [l,3]-rearrangements is presented. The steady-state kinetic approach applied to intermediate 3 allowed a minimal two consecutive step 2 → 4→ 5 kinetic model, in which the steric bulkiness of the alkyl substituent affects strongly the associated rate constants, k₁ and k₂, inverting the rate-determining step. The solvation effects enhanced the feasibility of these skeletal, rearrangements as they stabilized the transition states to a great extent. The experimental determined, thermodynamic parameters and DFT calculations suggest, that these cascade rearrangements occur through [l,3]-sigmatropic mechanisms, in which asynchronous bond reorganization processes via four membered pseudopericyclic transition states are highly favorable.

Full text of DOI:10.1021/jo902520z

pubs.acs.org/joc
Cascade [1,3]-Sigmatropic Rearrangements of Ketene O,O-Acetals:
Kinetic and DFT Level Mechanistic Studies
†,‡
Perla Y. Lopez-Camacho, Pedro Joseph-Nathan, Barbara Gordillo-Roman,
†
†
ꢀ
ꢀ
Oscar R. Suarez-Castillo, and Martha S. Morales-Rı
ꢀ
§
´
os*,†
ꢀ
†
‡
Departamento de Quımica, and Departamento de Farmacologıa, Centro de Investigacion y de Estudios
´
Avanzados del Instituto Politeꢀcnico Nacional, Apartado 14-740, Meꢀxico, D. F., 07000 Mexico, and Centro
´
ꢀ
§
´
de Investigaciones Quımicas, Universidad Autonoma del Estado de Hidalgo, Apartado 1-622, Pachuca,
ꢀ
Hidalgo, 42001 Mexico
smorales@cinvestav.mx
Received December 7, 2009
The regioisomeric R-cyano ketene-O,O-dialkyl acetals 2a-e and 4a-e, sequential intermediates in
the diazomethane induced conversion of indole R-cyano-γ-lactones 1a-e to 2-indolyl cyanomalo-
nates 5a-e, were isolated and characterized. Formation of the steady-state intermediate cycloprop-
[b]indoles 3a-e was evidenced by means of NMR and confirmed by the X-ray structure of 3c,
demonstrating that the formation of 5a-e from 2a-e proceeds through two consecutive and one
parallel unimolecular steps, with intermediates 3a-e formed in reversible processes. Evidence that
the reversible reactions proceed via [1,3]-rearrangements is presented. The steady-state kinetic
approach applied to intermediate 3 allowed a minimal two consecutive step 2 f 4f 5 kinetic model,
in which the steric bulkiness of the alkyl substituent affects strongly the associated rate constants, k1
and k2, inverting the rate-determining step. The solvation effects enhanced the feasibility of these
skeletal rearrangements as they stabilized the transition states to a great extent. The experimental
determined thermodynamic parameters and DFT calculations suggest that these cascade rearrange-
ments occur through [1,3]-sigmatropic mechanisms, in which asynchronous bond reorganization
processes via four membered pseudopericyclic transition states are highly favorable.
Introduction
product synthesis.1 Among ketene acetals, those derived
from esters are the most common and best investigated
reactive precursors in the Claisen and Ireland-Claisen sigma-
tropic rearrangements.2,3
Involvement of an unexpected skeletal reorganization has
attracted the attention of organic chemists not only for the
search for synthetic applications but also to perform me-
chanistic studies. In this respect, experimental and theore-
tical studies concerning ketene acetal rearrangements are an
active area of chemical research, since they have found
considerable applications in the natural and unnatural
(1) For a review of applications of ketene acetal rearrangement in
synthesis, see: Tadano, K. In Studies in Natural Products Chemistry;
Rahman, A.-U., Ed.; Elsevier: Amsterdam, 1992; pp 405-455.
(2) For reviews of asymmetric Claisen rearrangement, see: (a) Enders, D.;
Knopp, M.; Schiffers, R. Tetrahedron: Asymmetry 1996, 7, 1847–1882.
(b) Ito, H.; Taguchi, T. Chem. Soc. Rev. 1999, 28, 43–50.
(3) For a review on Ireland and related Claisen rearrangement, see: Chai,
Y.; Hong, S.-p.; Lindsay, H. A.; McFarland, C.; McIntosh, M. C. Tetra-
hedron 2002, 58, 2905–2928.
*To whom correspondence should be addressed. Tel: þ(52 55)5747 7112.
Fax: þ(52 55)5747 7137.
1898 J. Org. Chem. 2010, 75, 1898–1910
Published on Web 02/24/2010
DOI: 10.1021/jo902520z
r
2010 American Chemical Society

ꢀ
Lopez-Camacho et al.
JOCArticle
However, contrary to the widely recognized [3,3]-sigma-
tropic rearrangement of ketene acetals, the [1,3] process is a
rare event witnessed previously in a few cases of acyclic and
semicyclic enolates derived from benzyl esters, in which the
intervention of the typical [3,3] process is energetically un-
favorable.4 In these cases, the 3,3-shift can be markedly
affected by substantial loss of the aromatic π-electron de-
localization of the benzyl group upon migration, which is a
requirement for this type of reaction to occur. In contrast,
the benzyl group presumably could stabilize either radical
intermediates or charge build-up in the transition state for
the concerted 1,3-shift. In addition, there exist only two
isolated reports in the literature involving [1,3]-sigmatropic
rearrangements of silyl lactone enolates to provide con-
tracted carbocyclic ring systems with high diastereoselecti-
vity.5 As a consequence, little or no information exists to date
about those electronic and steric factors on rates of [1,3]-
sigmatropic rearrangements of ketene acetals and the nature
of the transition states for such processes.
FIGURE 1. Structural formulas for R-cyano-γ-lactones 1a-e.
substrate 1c, appearing in CDCl3 solution as an equilibrated
racemic mixture of cis-fused 11:2 endo/exo diastereoisomers,9
was prepared in one pot from Z-1-carbomethoxy-2-hydroxy-
3-indolinylidenecyanoacetate10 with benzylmagnesium bro-
mide.
Cascade Reactions with Diazomethane. The R-cyano-γ-
lactones of type 1 revealed in solution and at room tempera-
ture a dynamic stereochemistry governed by the steric pre-
ferences of the cyano group at the C3 stereocenter.11 Since
the interconversion of the endo/exo isomers of 1 is expected
to involves passage through an enol tautomer, reaction of
1a-e with diazomethane could give the desired ketene O,O-
acetals.12 Compounds 1a-e were treated with an excess of
freshly prepared diazomethane13 in ether for 5 min at room
temperature. The GC-mass spectral analysis of the reaction
crude exhibited in all cases a clean molecular ion peak [M]þ
shifted by 14 Da with respect to that of starting material,
which could be ascribed, at first glance, to the expected
ketene O,O-acetals 2a-e. NMR analysis of the crude reac-
tion mixture conducted in acetone-d6 however, evidenced the
inherent instability of the ketene O,O-acetals 2.14 Thus,
whereas the 1H NMR spectra of 2d and 2e gave a single set
of signals, which agreed very well with the proposed struc-
tures, those owing to 2a-c allowed us to detect one or two
additional isomeric species which, although not identified,
had 1H NMR spectra resembling those of 2. To our surprise,
standing neat 2a under ambient conditions for one week
spontaneous and quantitatively rearranged to the thermo-
dynamically more stable 2-indolylcyanomalonate 5a, revel-
ing one of the two previously detected isomeric species.
In addition, when 2c was standing neat for several weeks, it
furnished a mixture of starting ketene O,O-acetal 2c, the
expected 2-indolylcyanomalonate 5c, and a compound
whose 13C NMR analysis showed the presence of signals
most likely attributable to a regioisomer of 2c, the corre-
sponding inverted ring-fusion furo[3,2-b]indole 4c, than to
the ring-contracted cyclopropyl scaffold structure 3c.15
Studies in the field of furoindolines chemistry has been of
great interest in our laboratory in the past few years.6 Herein,
we report the first systematic study of the steric effects on the
cascade [1,3]-sigmatropic rearrangement of β-alkyl-substi-
tuted ketene-O,O-acetals and evidence on the nature of the
transition states for such processes is discussed.
Results and Discussion
Synthesis. The studied R-cyano-γ-lactones 1a-e (Figure 1),
varying over a significant range in terms of the size of the alkyl
group (R=Me, Et, Bn, i-Pr, t-Bu) on the C3a bridgehead
carbon, were conveniently synthesized from a 2-hydroxyin-
dolenine via chelation-controlled 1,4-conjugated addition
of Grignard reagents7 and subsequent lactonization in
yields ranging from 74% for R = i-Pr to 21% for R = t-
Bu.7b The low yield of 1e is due to a greater steric hindrance
at the newly formed quaternary carbon center.8 Starting
(4) (a) Arnold, R. T.; Kulenovic, S. T. J. Org. Chem. 1980, 45, 891–894.
(b) Shiina, I.; Nagasue, H. Tetrahedron Lett. 2002, 43, 5837–5840. (c) Burger,
K.; Gaa, K.; Geith, K.; Schierlinger, C. Synthesis 1989, 850–855. (d)
Shishido, K.; Shitara, E.; Fukumoto, K. J. Am. Chem. Soc. 1985, 107,
5810–5812. (e) Susuki, T; Inui, M.; Hosokawa, S.; Kobayashi, S. Tetrahedron
Lett. 2003, 44, 3713–3716. For a review which includes the [1,3]-sigmatropic
rearrangement of ketene acetals, see: Nasveschuk, C. G.; Rovis, T. Org.
Biomol. Chem. 2008, 6, 242–244.
(5) (a) Danishefsky, S.; Funk, R. L.; Kerwin, J. F. J. Am. Chem. Soc.
1980, 102, 6891–6893. (b) Kahn, K. M.; Knight, D. W. J. Chem. Soc., Chem.
Commun. 1991, 1699–1701.
ꢀ
(6) (a) Morales-Rı
´
os, M. S.; Suarez-Castillo, O. R.; Garcı
´
´
a-Martınez, C.;
os, M. S.;
Joseph-Nathan, P. Synthesis 1998, 1755–1759. (b) Morales-Rı
´
Suarez-Castillo, O. R.; Alvarez-Cisneros, C.; Joseph-Nathan, P. Can. J.
ꢀ
Chem. 1999, 77, 130–137. (c) Morales-Rı
´
Joseph-Nathan, P. J. Org. Chem. 1999, 64, 1086–1087. (d) Morales-Rı
os, M. S.; Suarez-Castillo, O. R.;
(9) Based on the integration of the CH2 AB proton signals: 1H NMR (300
MHz, CDCl3) δ 3.10 (endo), 3.35 (exo) ppm.
ꢀ
´
(10) Morales-Rıos, M. S.; Bucio-Vasquez, M. A.; Joseph-Nathan, P.
J. Heterocycl. Chem. 1993, 30, 953–956.
ꢀ
´
M. S.; Suarez-Castillo, O. R.; Trujillo-Serrato, J. J.; Joseph-Nathan, P.
os,
ꢀ
ꢀ
J. Org. Chem. 2001, 66, 1186–1192. (e) Morales-Rı
´
os, M. S.; Santos-Sanchez,
os,
M. S.; Suarez-Castillo, O. R.; Joseph-Nathan, P. Tetrahedron 2002, 58, 1479–
N. F.; Joseph-Nathan, P. J. Nat. Prod. 2002, 65, 136–141. (f) Morales-Rı
´
(11) In these types of compounds, the equilibrium isomer distribu-
ꢀ
´
os, M. S.; Suarez-Castillo, O. R.;
ꢀ
tion is solvent dependent; see: Morales-Rı
Joseph-Nathan, P. J. Chem. Soc., Perkin Trans. 2 2000, 769–775.
ꢀ ꢀ
1484. (g) Morales-Rıos, M. S.; Santos-Sanchez, N. F.; Fragoso-Vazquez, M.
´
ꢀ
ꢀ ꢀ
(12) Morales-Rıos, M. S.; Lopez-Camacho, P. Y.; Suarez-Castillo, O. R.;
´
J.; Alagille, D.; Villagomez-Ibarra, J. R.; Joseph-Nathan, P. Tetrahedron
2003, 59, 2843–2853. (h) Morales-Rıos, M. S.; Rivera-Becerril, E.; Joseph-
NathanP. Tetrahedron: Asymmetry 2005, 16, 2493–2499. (i) Rivera-Becerril, E.;
´
Joseph-Nathan, P. Tetrahedron Lett. 2007, 48, 2245–2249.
(13) Arndt, F. Organic Syntheses; Wiley: New York, 1943; Vol. 2, pp 165-167.
(14) The solvent choice for NMR measurements was acetone-d6 rather
than CDCl3 because the acidic character of the later resulted in reconverted
cyano lactones 1a-e in minor amounts, together with decomposition
products.
ꢀ
ꢀ
Joseph-Nathan, P.; Perez-Alvarez, V. M.; Morales-Rı
2008, 51, 5271–5284.
(7) (a) Morales-Rı
´
os, M. S. J. Med. Chem.
´
os, M. S.; Bucio, M. A.; Joseph-Nathan, P. Tetra-
hedron 1996, 52, 5339–5348. (b) Morales-Rı
Bucio, M. A.; Joseph-Nathan, P. Monatsh. Chem. 1996, 127, 691–699.
(c) Morales-Rıos, M. S.; Bucio, M. A.; Garcıa-Martınez, C.; Joseph-Nathan,
P. Tetrahedron Lett. 1994, 35, 6087–6088. (d) Morales-Rıos, M. S.; Bucio,
M. A.; Joseph-Nathan, P. Tetrahedron Lett. 1994, 35, 881–882.
(8) Rueping, M; Nachtsheim, B. J.; Moreth, S. A.; Bolte, M. Angew.
Chem., Int. Ed. 2008, 47, 593–596.
os, M. S.; Garcıa-Martınez, C.;
´ ´ ´
(15) (a) For a 13C NMR example of a furo[3,2-b]indole, see: Tsai, A.-I.;
Lin, C.-H.; Chuang, C.-P. Heterocycles 2005, 65, 2381–2394. (b) For 13C
NMR examples of cycloprop[b]indoles, see: Wenkert, E.; Alonso, M. E.;
Gottlieb, H. E.; Sanchez, E. L. J. Org. Chem. 1977, 42, 3945–3949. (c) Ikeda,
M.; Matsugashita, S.; Yukawa, C.; Yakura, T. Heterocycles 1998, 49,
121–126.
´
´
´
´
J. Org. Chem. Vol. 75, No. 6, 2010 1899

ꢀ
Lopez-Camacho et al.
JOCArticle
TABLE 1. Product Ratios of Ketene O,O-Acetals 2a-e
head CH proton and a signal around δ 67-70 ppm (1JCH ca.
160 Hz) for the CH carbon in the 1H and 13C NMR spectra,
respectively. Also characteristic is the 13C NMR signal at δ
ca. 95-102 ppm ascribed to the bridgehead quaternary
carbon. On the other hand, indolylcyanomalonates 5a-e
could also be characterized by means of indicative NMR
peaks. In particular, the methine group shows signals around
δ 6.0-6.3 (singlet) and 38 ppm (1JCH ca. 127 Hz) in the 1H
and 13C NMR spectra, respectively. The sp2 hybridation of
the alkyl substituted C3 carbon is indicated by its absorbance
at ca. δ 125-133 ppm in the 13C NMR spectrum.20 In this
yieldb (%)
of
2 þ 4 þ 5
>99
>99
>99
>99
>99
productsa
2/4/5
1
way, the H NMR spectrum provides a simpler means to
entry
substrate
R
distinguish between isomeric products 2, 4, and 5 by compar-
ing the chemical shift of the CH methine proton singlet signal
present in all the rearranged products (Table 1).
Mechanistic Rationale for Indolylcyanomalonation by Diazo-
1
methane. From the H results, and assuming that furo[3,2-
1
2
4
3
5
1a
1b
1c
1d
1e
Me
Et
Bn
i-Pr
t-Bu
2:1:traces
7:1:0
9:1:0
1:0:0
1:0:0
aDetermined by 1H NMR (300 MHz) in acetone-d6. bOverall yield
based on starting 1.
b]indoles 4 are formed through rearrangement of a kine-
tically significant intermediate, a possible stepwise mecha-
nism can be derived for indolylcyanomalonation by diazo-
methane, under thermodynamic conditions. The essential
elements are shown in Scheme 1. In the first step, O-methyla-
tion of the enolate anion of the R-cyano-γ-lactones 1a-e
occurs, followed by a 1,3-shift of C-8a with the consequent
ring cleavage by breaking the endocyclic C-O bond, to
generate the intermediates 3a-e for which the structure of
a ring-contracted cyclopropane is suggested. Note that the
rearrangement proceeds regioselectively; that is, the corre-
sponding lactone arising by breaking the exocyclic C-O
bond is not detected. A higher migrating aptitude of carba-
mate substituted methine group versus methyl group and
conformational grounds can be more favorable for the
methine 1,3-shift to be performed. In the second step, the
driving force is probably the release of ring strain of the
cyclopropane ring21 via parallel processes, reversible ring-
expansion rearrangements or ring-opening reaction, allow-
ing either equilibration between the regioisomeric R-cyano
ketene O,O-acetals 2a-e and 4a-e or affording the aroma-
tized 2-indolylcyanomalonates 5a-e by cleavage of the more
substituted cyclopropyl bond. This latter process is therefore
expected to be exothermic and irreversible, allowing the
termination product in the cascade reactions.
Although at this point the formation of the key intermediate 3
could not be detected by NMR, an unambiguous proof of
its structure was subsequently obtained by an X-ray cry-
stallographic analysis of the corresponding cyclopropane 3c
(Figure 2). The structure of 3c indicated the exo stereochemistry
of the ester group. Unfortunately, all attempts to reproduce the
conditions under this kinetically significant intermediate accu-
mulates and has much longer rearrangement lifetime were
unrewarding.21 The presence of two geminally placed electron
withdrawing groups at the cyclopropane ring are likely to
contribute to its vulnerability in solution. It should be noted
that the endocyclic cyclopropane C-C bonds of 3c show a high
level of substituent-induced bond-length asymmetry22 with the
Ketene O,O-acetal 2e, carrying a sterically bulky t-Bu group,
was shelf-stable for at least several weeks, evidencing that
this dramatic difference in the stability of the ketene O,O-
acetals 2a-e is tuned by the bulkiness of the alkyl substituent
at the C3a bridgehead carbon, with 2a and 2b being much less
stable entities than 2c-e. Therefore, the previously described
molecular ion peak [M]þ shifted by 14 Da in the GC/MS
spectra is attributed to the corresponding 2-indolylcyano-
malonates 5a-e arising from the rearrangement of ketene O,
O-acetals 2a-e in the column.16 Table 1 summarizes the 2/4/
1
5 product ratios arising from H NMR measurements run
immediately after evaporation of the reaction solvent at
room temperature under atmospheric pressure with a stream
of argon.
Structural Characterization of Isomeric Scaffolds 2, 4, and
1
5. In addition to standard H and 13C{1H} NMR spectra
(Figure S1-S25 in the Supporting Information), the struc-
tures of the three isomeric scaffolds 2, 4, and 5 were confir-
med using HMBC and HSQC NMR experiments.17 Final
confirmation of the structure of 2e was obtained from single-
crystal X-ray analysis (Figure S28, Supporting Information).18
The characteristic signals in the NMR spectra of 2a-e show
the bridgehead methine proton neighboring to the hetero-
atoms as a sharp singlet at ca. δ 6.6 ppm and a signal around
δ 96-101 ppm (1JCH = 180 Hz) for the corresponding
1
bridgehead CH carbon in the H and 13C NMR spectra,
respectively. In addition, the 13C NMR spectra of 2a-e show
the alkyl-substituted quaternary carbon at ca. δ 55-66 ppm.
A remarkable characteristic in this series, is the large 13C
chemical shift difference between the two ethylenic carbon
atoms C2dC3, within a range of 106 ( 4 ppm, reflecting
a significant zwitterionic character arising from their π-
conjugated push-pull nature.12,19
The regioisomeric ketene O,O-acetals 4a-e contain a
typical sharp singlet around δ 5.5-5.7 ppm for the bridge-
(16) The GC/MS of pure compounds 2e, 4e, and 5e provided identical
retention times and mass spectral patterns.
(17) Compounds 2a, 2b, 4a, and 4b, according to their instability, could
not be isolated at the pure state.
(18) Supporting Information; see the paragraph at the end of this paper
regarding availability.
(19) The substantial lower stability of R-CN ketene acetals compared
with those R-CO2Me substituted is consistent with their stronger push-pull
effect. For R-CO2Me ketene acetals Δδ 13CC2=C3 is 84-86 ppm; see ref 12.
~
(20) Morales-Rıos, M. S.; Espineira, J.; Joseph-Nathan, P. Magn. Reson.
´
Chem. 1987, 25, 377–395.
(21) Parker, V. D. Pure Appl. Chem. 2005, 77, 1823–1833.
(22) (a) Danishefsky, S. Acc. Chem. Res. 1979, 12, 66–72. (b) Gibe, R.;
Kerr, M. A. J. Org. Chem. 2002, 67, 6247–6249. (c) Ikeda, M.; Matsugashita,
S.; Tamura, Y. J. Chem. Soc. Perkin I 1977, 1770–1772. (d) Alonso, M. E.;
ꢀ
Gomez, M.; de Sierraalta, S. P.; Jano, S. P. J. Heterocycl. Chem. 1982, 19,
369–371.
1900 J. Org. Chem. Vol. 75, No. 6, 2010

ꢀ
Lopez-Camacho et al.
JOCArticle
SCHEME 1. Proposed Mechanism for the Cascade [1,3]-Rearrangements of 2a-e
C3-C3a bond significantly elongated, as long as 1.584(3) A, in
contrast to 1.540(3) A) and 1.487(2) A) of the other two bonds.
Thus, the high instability of intermediates 3 could be attributed,
at least in part, to this long23 and therefore weak C3-C3a bond.
(vide infra). In general, a negligible effect on the NMR
chemical shift was observed for the diagnostic CH methine
signal on solvent change from acetone-d6 to H-bond accep-
tor DMSO-d6, except in the case of 5a-d in which inter-
molecular hydrogen bond interactions induce a methine
1
proton shift of ca. 0.3 ppm to higher frequency. Thus, H
NMR was used for kinetic studies, allowing direct registra-
tion of three species, a situation that is challenging. Reac-
tions were investigated at substrate concentrations varying
from 0.03 to 0.04 mmol/mL of solvent at normal probe
temperature (22 °C).
We found that the rearrangement of 2a (R = Me) in
DMSO-d6 affords 5a in quantitative yield after only ca. 30
min reaction time (Table 2, entries 1-6), whereas the NMR-
monitored experiment of the sterically more demanding 2b
(R=Et), immediately taken upon dissolution in DMSO-d6,
clearly indicated three sets of signals which represent starting
2b, the inverted ring-fusion furo[3,2-b]indole 4b, and the final
rearranged product 5b. The sharp singlet signals at δ 6.57,
5.54, and 6.28 ppm owing to the methine protons in 2b, 4b,
and 5b, correspondingly, had emerged in a ca. 8:1:1 ratio
(Table 2, entry 7). With time, the characteristic singlet signals
corresponding to 2b and 4b were diminishing and disap-
peared within 3-4 h, whereas the singlet signal assigned to 5b
(δ 6.28 ppm) steadily increased in magnitude over time and
no byproducts were detected. Similar behavior is observed
with the remaining ketene O,O-acetals 2c-e. Nonetheless,
on increasing the steric bulkiness of the alkyl substituent, the
FIGURE 2. X-ray crystal structure of 3c.
Kinetic Studies for 2a-e. Unlike the aforementioned
NMR experiments of ketene O,O-acetals 2a-e conducted
in acetone-d6, in which the reaction progress is roughly
comparable to that occurring when they are stored neat, in
the more polar DMSO-d6 solvent the reaction kinetics are
greatly enhanced, allowing the conversion of 2a-e to 2-
indolylcyanomalonates 5a-e in an interval from half an
hour until few weeks. The strong rate dependence for these
rearrangements on the polarity of the surrounding medium
may be attributed to transition states more polar than the
reactants and, as such, zwitterionic processes or asynchro-
nous bond reorganizations could conjectured to occur.24
Kinetic and DFT level computational studies were used to
discern between the aforementioned reaction mechanisms
1
rate for the rearrangements decrease significantly. The H
NMR (DMSO-d6) analysis at differing degrees of conversion
of 2a-e are summarized in Table 2. Typically, 30 min to 76 h
were required to achieve ca. 90% conversion of ketene O,O-
acetals 2a-e.
From a kinetic standpoint, a steady-state approach25
applied to kinetically significant intermediate 3 was used
for the treatment of the kinetic data given in Table 2 (eq 1)
k1
h
2
2a-e
4a-e k
s
f
5a-e
ð1Þ
k -1
where k1, k-1, and k2 are the corresponding kinetic con-
stants. However, since the reverse reaction of 4a-e to 2a-e
(23) Kiyohara, S.; Ishizuka, K.; Wakabayashi, H.; Miyamae, H.; Kana-
zumi, M.; Kato, T.; Kobayashi, K. Tetrahedron Lett. 2007, 48, 6877–6880.
(24) Reichardt, C. Solvent Effects in Organic Chemistry, 3rd ed.; Wiley-
VCH: Weinheim, Germany, 2003.
(25) Hammett, L. P. In Physical Organic Chemistry, 2nd ed.; McGraw-
Hill: New York, 1970; Chapter 5.
J. Org. Chem. Vol. 75, No. 6, 2010 1901

ꢀ
Lopez-Camacho et al.
JOCArticle
TABLE 2. Cascade Reactions Evolutiona,b of Ketene O,O-Acetals 2a-e: Influence of the Alkyl Substituent
entry
compd
R
time (h)
mol fraction 2:4:5
entry
compd
R
time (h)
mol fraction 2:4:5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
2a
2a
2a
2a
2a
2a
2b
2b
2b
2b
2b
2b
2c
2c
2c
Me
Me
Me
Me
Me
Me
Et
Et
Et
Et
Et
0.00
0.03
0.10
0.20
0.30
0.50
0.00
0.40
0.53
1.33
2.00
2.40
0.00
1.00
2.00
0.52:0.25:0.23
0.45:0.23:0.32
0.35:0.19:0.46
0.23:0.14:0.63
0.16:0.11:0.73
0.08:0.06:0.86
0.79:0.09:0.12
0.47:0.27:0.26
0.31:0.27:0.42
0.20:0.21:0.59
0.13:0.15:0.72
0.10:0.10:0.80
0.86:0.10:0.04
0.47:0.31:0.22
0.30:0.26:0.44
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
2c
2c
2c
2d
2d
2d
2d
2d
2d
2e
2e
2e
2e
2e
2e
Bn
Bn
Bn
3.00
4.00
5.00
0.00
3.00
6.00
9.00
12.00
15.00
0.00
10.00
20.00
40.00
60.00
76.00
0.20:0.19:0.61
0.14:0.14:0.72
0.11:0.11:0.78
0.94:0.06:0.00
0.56:0.26:0.18
0.34:0.25:0.41
0.20:0.16:0.64
0.11:0.11:0.78
0.06:0.07:0.87
1.00:0.00:0.00
0.69:0.27:0.04
0.49:0.45:0.06
0.25:0.58:0.17
0.15:0.57:0.28
0.11:0.52:0.37
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
Et
Bn
Bn
Bn
aMonitored by 1H NMR in DMSO-d6 solution at 22 °C. bQuantification of the product mixtures was followed via integration relative to residual
solvent peaks from DMSO-d6.
FIGURE 3. Time-dependent partial 1H NMR spectra (DMSO-d6) following the rearrangement of 2d to 5d at 22 °C. For the molar fraction, see
Table 2. The inset shows a logarithmic curve fit of the peak height at 6.56 ppm vs time.
(k-1) was inferred to be much slower than its rearrangement
2a-e to 4a-e (k1),26 the minimal kinetic model given in eq 2
may be considered and accounts for a first-order rate law
expression for the disappearance of 2a-e with time (eq 2).
TABLE 3. Kinetic Data (22 °C) for the Rearrangements of 2a-ea
rate constant
k1 x 104 s-1
half-life
t1/2/min
fit parameter
R2
compd
2a
2b
2c
2d
2e
10.30 ((0.3)
2.37 ((0.04)
1.31 ((0.003)
0.45 ((0.04)
0.09 ((0.001)
11.2
48.7
88.2
259.6
1283.3
0.998
0.990
0.995
0.997
0.995
k1
k2
2a-e
s
f
4a-e
s
f
5a-e
ð2Þ
Since the transformation of 2d or 2e occurred slowly or
very slowly (Table 2), one can assume that any of the
intermediates 4d or 4e can be isolated from the reaction
mixture, allowing the condition k1 . k-1 to be verified (vide
infra). First-order rate constants were measured in duplicate
runs by monitoring the disappearance of 2a-e at 22 °C. As
expected, plots of ln [2] versus time gave, in all cases, straight
lines with slop k1 (see the Supporting Information). The
kinetic data are summarized in Table 3.
aDetermined in the conversion range listed in Table 2.
the methine signals are shown in Figure 3. As one can see,
during the rearrangement of 2d the 1H NMR methine signal
corresponding to intermediate 4d initially grows in intensity
reaching the maximum level (28%) after 3 h and then
disappears gradually, with formation of indole 5d. The
intermediate cyclopropane 3d remains at first glance unde-
tectable during the reaction course, and no other species were
observed up to 3 half-lives. The steady-state kinetic approach
shows that the rate-limiting is the formation of 4d (k1),
followed by its faster interconversion to product 5d (k2) (eq
2: k2 > k1).
Typical series of NMR spectra following the rearrange-
ment of 2d (R=i-Pr) using the change in relative intensity of
(26) Because C3-C8a bond breaking in 3a-e is expected to be disfavored
in comparison to C3-C3a bond breaking, the ring-contraction 2a-e to 3a-e
should be essentially irreversible on enthalpic grounds.
1902 J. Org. Chem. Vol. 75, No. 6, 2010

ꢀ
Lopez-Camacho et al.
JOCArticle
FIGURE 4. Time-dependence curves for the cascade rearrangement of (a) 2d and (b) 2e.
FIGURE 5. Dependence of k1 for compounds 2a-e with steric parameters: (a) Es, (b) Esc.
Examination of the kinetic data for the rearrangement of
2e indicated a dramatic “slowing-down effect” for the bulk-
ier t-Bu group, with k1 decreasing more than 2 log units
compared to the consumption of 2a (Table 3). The time
dependence curves for the cascade rearrangement of 2e
showed that, after about 2 days, the starting material 2e
achieved 80% conversion (Table 2, entries 28/29), fitting a
first-order kinetics, while at the same time intermediate 4e
reached its maximum intensity value of 60%. These data
evidence a change in the rate-determining step with k1 > k2
(Figure 4).
Steric Effects of the R Group. One of the main goals of this
work was to quantify the steric effects of the group R on the
reactivity of ketene O,O-acetals 2a-e. Thus, the reaction
FIGURE 6. Eyring plots for the rearrangements of 2b, 2d, and 2e.
kinetics were analyzed as function of the substitution degree
on the carbon directly bounded to the ketene acetal group,
comprising primary R groups (Me and Et), secondary
(benzyl and isopropyl), as well as tertiary ones (tertbutyl).
Figure 6 shows the correlations found between the rate
constants (k1) of disappearance of substrates 2a-e with time
and the steric substituent parameter Es originated by Taft27
and its correction Esc28 (Table 4). We observed that the curve
for k1 as function of Es does not show a good correlation
(R2=0.844). However, a highly significant improvement in
fit was obtained when the steric parameter Esc was applied
(R2=0.932, slope=0.788), confirming that the steric bulki-
ness of the R group is the main factor determining the
observed kinetic trend in these rearrangements. Clearly, the
rate constant (k1) for 2e (R=t-Bu) exceeds in relation to the
corresponding steric parameter Esc (Figure 5b), as predicted
by a line passing through the points for the first four
compounds (R2=0.983, slope=1.226). The interpretation
of this fact in terms of the transition state is that the bulky
t-Bu group could cause a relative enthalpically based steric
acceleration by bond lengthening. Considering the fact that
the k1 rates for these rearrangements decrease linearly in
the order of increasing steric effect Esc (Table 4), then it can
be argued that bulky substituents at C3a destabilize the
transition states more than they destabilize the starting
materials.
(27) (a) Pavelich, W. A.; Taft, R. W. J. Am. Chem. Soc. 1957, 79, 4935–
4940. (b) Unger, S. H.; Hansch, C. In Progress in Physical Organic Chemistry;
Taft, R. W., Ed.; John Wiley and Sons: New York, 1976; Vol. 12, pp 91-118.
(28) Hancock, C. K.; Meyers, E. A.; Yager, B. J. J. Am. Chem. Soc. 1961,
83, 4211.
J. Org. Chem. Vol. 75, No. 6, 2010 1903

ꢀ
Lopez-Camacho et al.
JOCArticle
TABLE 4. Steric Substituent Parameters for R Groups
TABLE 6. Kinetic Data (22 °C) for the Rearrangements of 4d and 4e
28
R
Es27
Esc
rate constant
k2 x 104 s-1
half-life
t1/2/min
fit parameter
R2
compd
Me
Et
Bn
isopropyl
tertbutyl
0.00
-0.07
-0.38
-0.47
-1.54
0.00
-0.38
-0.69
-1.08
-2.46
4d
4e
0.74 ( 0.09
152
5391
0.996
0.996
0.02a
aSingle run.
TABLE 5. Activation Parameters for the Rearrangements of 2b, 2d,
and 2e
ΔHq
ΔSq
ΔGq
(298)
(kcal mol-1
)
compd
(kcal mol-1
)
(cal K-1 mol-1
)
2b
2d
2e
20.78
22.28
21.18
-17.66
-16.90
-23.03
22.22
23.21
24.14
Thermodynamic Parameters of the Transition State for 2b,
2d, and 2e. With the aim of determining the enthalpic (ΔHq)
and entropic (ΔSq) components of the steric effects for
transition states reached from reactants 2b, 2d, and 2e, the
effect of temperature on reaction rates was studied at three
different temperatures over the range of 22-37 °C for the
disappearance of 2b and at four different temperatures from
22 to 46 °C for the disappearance of 2d and 2e in DMSO-d6.
Because the precision of the NMR data and the narrow
temperature range in which these rearrangements could be
conveniently studied, the corresponding values ΔHq and ΔSq
are probably reliable to (1 kcal mol-1 and 2-3 eu, respec-
tively. The first-order rate of these reactions was determined
by repeating the reactions two times at each of the chosen
temperatures (see the Supporting Information), and the
activation parameter values calculated from Eyring plots
are displayed in Table 5. In all cases studied, good linear
correlations of ln (k/T) versus 1/T were obtained (Figure 6),
indicating that the calculated activation parameters depend
only on the nature of the rearrangement process. The low
activation energies and the strongly negative entropies of
activation, in the range associated with unimolecular pro-
cesses,29 provide evidence for a concerted [1,3]-sigmatropic
rearrangement via a four-membered pseudopericyclic tran-
sition state30 and, therefore, with loss in conformational
freedom. The large negative entropy values can also imply
that the polar environment stabilizes the transition states
(TS) more than the ground state (GS) as solvation is more
efficient, with a lower rearrangement energy barrier. The
observed enthalpic activation trend in the order 2d > 2e >
2b (Table 5) is consistent with our preceding proposal that
rearrangement of 2e proceeds with a relative enthalpically
based steric acceleration of the tertbutyl group. In addition,
we expected entropy loss upon forming the transition state to
be lower with bulkier substituents, because of hindered
motion in the GS, and in fact, this is the observed trend
between 2b and 2d. However, the entropy parameter for 2e is
5.4 or 6.1 cal K-1 mol-1 lower than that of 2b or 2d,
respectively. We attributed the increase in negative entropy
of activation for the rearrangement of 2e to destabilizing
effects of the sterically bulky t-Bu group in achieving a
FIGURE 7. Time-dependent partial 1H NMR spectra (DMSO-d6)
following the rearrangement of 4e.
specific conformation in which the four-membered transi-
tion state can occur.
Kinetic Studies for 4d and 4e. In light of the above results,
and to support the mechanistic details for the cascade
rearrangements of 2a-e (Scheme 1), we attempted to isolate
regioisomeric indolines 4. Representative 4d and 4e were
successfully isolated in 27% and 58% yield, respectively (see
the Experimental Section) and characterized on the basis of
their NMR spectral data. An unambiguous proof of the
structure of the inverted ring-fusion furo[3,2-b]indoles 4d
and 4e was established by their crystal structure analysis31
(Figures S30 and S31, Supporting Information).
When a DMSO-d6 solution of 4d was monitored by NMR
spectroscopy at 22 °C, a clean first-order rearrangement to
5d was observed as far as the reaction proceeds to a 90%
conversion (∼8.5 h) (Table 6). As expected from the curve-
fittings in Figure 4, the transformation of 4d to 5d (k2) took
place faster than that of 2d to 4d (k1). Thus, 2df4d was the
rate-determining step of the overall reaction.
The most fully defined outcome in these kinetic studies
was obtained by monitoring the rearrangement of the furo-
[3,2-b]indole 4e (Table 6) by taking the 1H NMR spectra over
a period of approximate 20 days at normal probe tempera-
ture (22 °C). Typical series of NMR spectra following the
rearrangement of 4e using the change in relative intensity of
the methine signals are shown in Figure 7. As can be seen, the
NMR spectrum of 4e obtained after 7.5 h reaction time
showed, in addition to the expected single signals attributa-
ble to each of the methine protons of starting furo[3,2-
b]indole 4e (δ 5.64) and the final rearranged product 5e (δ
6.28), two clean sharp singlets at δ 6.54 and δ 5.73 ppm whose
height remains less than 8% and almost constant through the
rearrangement process. The singlet signal at δ 6.54 ppm
corresponds to furo[2,3-b]indole 2e and, therefore, does
not exert a significant influence on the overall rate, setting
(29) (a) Bates, G. S.; Ramaswamy, S. Can. J. Chem. 1981, 59, 3120–3122.
(b) Frey, H. M.; Solly, R. K. Trans. Faraday Soc. 1968, 64, 1858.
(30) (a) Jones, G. O.; Li, X.; Hayden, A. E.; Houk, K. N.; Danishefsky,
(31) A significant shortening of the C3-C11 bond in 4d (1.407(3) A) and
4e (1.405(3) A) as compared to the C3-C11 bond in 3c (1.442(3) A) was
observed.
€
€
S. J. Org. Lett. 2008, 10, 4093–4096. (b) C-elebi-Olc-um, N.; Aviyente, V.;
Houk, K. N. J. Org. Chem. 2009, 74, 6944–6952.
1904 J. Org. Chem. Vol. 75, No. 6, 2010

ꢀ
Lopez-Camacho et al.
JOCArticle
rearrangement of computationally less expensive substrate
2a in an attempt to increase our understanding on these
fundamental rearrangements and in fact to confirm or
invalidate the postulated mechanism based on an energe-
tically reasonable pathway.
An initial systematic conformational search was per-
formed for the reactant, intermediates, and product at the
B3LYP/6-31G(d) theoretical level32 using Spartan 0433 and
then optimized at the same level of theory using Gaussian 03
suite of quantum chemical programs. Geometry optimiza-
tions of transition states (TS), as well as concerted or
stepwise pathways for the 1,3-rearrangements were also
explored at the B3LYP/6-31G(d) theoretical level using
Gaussian 03.34 Structures at stationary points were charac-
terized by frequency analysis and reported energies include
zero-point energy (ZPE) corrections scaled by 0.9806.35
Solvent effects were taken into account via geometry opti-
mization and frequency calculations using a polarized con-
tinuum model (PCM)36 in DMSO as the solvent. The free
energy (G) of a given structure in the solvent was calculated
by eq 3.
FIGURE 8. Eyring plots for the rearrangements of 4d and 4e.
TABLE 7. Activation Parameters for the Rearrangements of 4d and 4e
ΔHq
ΔSq
ΔGq
(298)
(kcal mol-1
)
compd
(kcal mol-1
)
(cal K-1 mol-1
)
4d
4e
18.01
20.15
-27.56
-28.8
22.90
25.03
G ¼ Egas þ Ezpe þ thermal
ð3Þ
where Egas is the gas-phase electronic energy and Ezpeþthermal
is the sum of the zero-point energy and the thermal
free energy contributions to the gas-phase energy at
298.15 K.
k1 . k-1. The new singlet at δ 5.73 ppm was associated with
the methine group of the kinetically significant intermediate
3e. The peak at δ 5.64 ppm nearly disappeared over a period
of 20 days, indicating the decay of starting 4e. The height of
peak at δ 6.28 ppm increased over the same time period,
evidencing the formation of 5e. Hence, we conclude Scheme 1
shows the correct mechanism of cascade rearrangements.
The collected NMR data fitted to a first-order rate law over
three half-life periods supporting the validity of the steade-
state kinetic approach used for data interpretation (eq 2).
Comparison between k1 and k2 showed that the product rate
formation, 4e f 5e step (k2), is slower than the rate decay of
the substrate, 2e f 4e step (k1).Therefore, in this two-step
consecutive unimolecular model, the rate-determining step is
the formation of 5e.
Thermodynamic Parameters of the Transition State for 4d
and 4e. Figure 8 shows k2 values for the disappearance of
regioisomeric indolines 4d and 4e in DMSO-d6 at four
different temperatures over the range of 22-46 °C for 4d
and of 22-55 °C for 4e plotted according to the Eyring
equation (see the Supporting Information). The activation
parameters are listed in Table 7.
Temperature-dependent rate studies demonstrate sub-
stantial negative entropies of activation for the rearrange-
ments of 4d and 4e (Table 7), which may be indicative of
concerted unimolecular processes with highly ordered tran-
sition states. The entropic cost of 4d or 4e is balanced in its
contribution to the free energy of activation (ΔGq) by an
enthalpic benefit as compared to 2d or 2e (Table 5), respec-
tively. In agreement with the rate-determining step, k1 for
R=i-Pr or k2 for R=t-Bu (vide supra), it was found that the
difference in free energy between transition states reached
from 2d and 4d is lower for 4d (ΔΔGq=0.31 kcal mol-1; k2 >
k1), while this difference between 2e and 4e is lower for 2e
(ΔΔGq=0.89 kcal mol-1; k1 > k2).
The optimized geometries of substrate 2a and intermedi-
ates 3a and 4a evidence the remarkable rigidity of these
molecular structures, which show only two stable confor-
mers having the carbonyl carbamate group syn and anti to
the H7 benzene ring proton, respectively. The syn conforma-
tion is the most stable in the gas phase by 1.1-2.8 kcal mol-1
(Supporting Information) and was the only one considered
toward the construction of reaction profiles. These results
are summarized in Figure 9. It should be noted that the syn
conformation was observed in all of the X-ray crystal
structures of related compounds (2e, 3c, 4d, and 4e).
The next step in our study was to analyze both the stepwise
and concerted processes. In all of our attempts to find the
reaction pathway involving ionic intermediates, we observed
that the concerted shift occurred before a stable ionic
(32) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Becke, A. D.
J. Chem. Phys. 1993, 98, 1372–1377. (c) Lee, C.; Yang, W.; Parr, R. G. Phys.
Rev. B 1988, 37, 785–789. (d) Stephens, P. J.; Devlin, F. J.; Chabalowski,
C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623–11627.
(33) Spartan 04 Wavefunction, Irvine, CA, 2004.
(34) Gaussian03, (Revision C.02): Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.,
Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.;
Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Pettersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc., 2003.
Computational Considerations on the 1,3-Shifts. In con-
junction with kinetic studies, which provided evidence sug-
gesting that rearrangements of ketene O,O-acetals 2a-e
occur via a concerted mechanism, we have modeled the
(35) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502–16513.
(36) (a) Miertus, S.; Tomasi, J. Chem. Phys. 1982, 65, 239–245.
(b) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117–129.
J. Org. Chem. Vol. 75, No. 6, 2010 1905

ꢀ
Lopez-Camacho et al.
JOCArticle
FIGURE 9. Low energy conformers of 2a-4a (gas phase), relative energies (kcal mol-1), ΔGgas (ΔGDMSO), selected bond lengths (d, A), dipole
moments (μ, debyes), and the optimized structure of indole 5a (gas phase).
TABLE 8. Activation Free Energies ΔGqgas phase and ΔGqDMSO (in kcal
mol-1) for the [1,3] Rearrangements of 2a and 4aa
good agreement with the X-ray structure for structurally
related compound 3c.
ΔGq
ΔGq
As can be seen from the data in DMSO phase (Table 8), the
barrier for ring contraction of 2a (21.3 kcal mol-1) to yield 3a
is substantially lower than that for ring contraction of 4a
(24.7 kcal mol-1) to yield 3a. This energy difference is
essentially due to the relative stability of substrates 2a and
4a. In this connection, we noticed that the reverting back ring
contraction reaction of 4a to yield 3a is 0.5 kcal mol-1 lower
in energy than the ring expansion reaction of 3a to yield 4a.
Since the [1,3]-rearrangements are under thermodynamic
control, the TS2 presumably plays a decisive role in deter-
mining the rate-determining step. The high stability of indole
5a could be the driven force for the reactions to occur.
Inspection of transition-state geometries for TS1 and TS2
were performed, and their main geometric features are
shown in Figure 10. All discussion refers to DMSO solvent
phase. Despite being concerted, the 2a f 3a and 4a f 3a
rearrangements appears to be highly asynchronous as shown
by the TS1 and TS2 geometries, respectively. While d3
distances are relatively short (2.447 or 2.524 A) the d4
distances are very long (2.603 or 2.853). The largest structur-
al changes of TS1 with respect to fully formed bonds in 2a are
the pronounced decrease in the forming OdC bond length
(d1 TS1=1.243 A and d1 2a=1.345 A) concomitant with an
increase in the breaking O-C bond length (d4 TS1=2.603 A
and d4 2a = 1.460 A). The shortening of the N-C bond
length for TS1 as compared to 2a (d5 TS1=1.317 A and d5
2a=1.447 A) is suggesting that the opening of the ketene acetal
ring is assisted by the indole nitrogen lone pair. Similarly, for
TS2 the main structural differences with respect to 4a are the
forming O1dC2 bond length (d1 TS2=1.236 A and d1 4a=
1.332 A) and the breaking O-C bond length (d4 TS2=2.853 A
and d4 4a = 1.490 A). In this case, a delocalization of the
nitrogen lone pair through the adjacent benzene ring is oper-
ating, as clearly evidenced the shortening in the N-C and
C-C bond lengths as compared to 4a (d6 TS2=1.368 A and
gas phase
DMSO
2afTS1f3a
4afTS2f3a
aReverse reaction energies are given in parentheses.
28.1 (30.6)
31.1 (30.6)
21.3 (25.1)
24.7 (25.2)
transition structure was found.37 Thus, theoretical calcula-
tions indicated that regioisomers 2a and 4a rearranged in one
step processes to give ring contracted 3a through planar four-
membered transition states TS1 and TS2, respectively. As
can be seen from Table 8, the ΔGq values for TS1 and TS2 in
the gas phase are found to be higher in energy than those
calculated with the incorporation of the DMSO solvent
environment. Values of ΔGq in DMSO phase are in the range
of those determined experimentally, contributing to validate
our approach. Because of the geometrical restrictions, regio-
isomers 2a and 4a are expected to rearrange via suprafa-
cial shifts with inversion of configuration at the migrating
carbon center, leading both to a highly stereoselective for-
mation of the cyclopropane diastereomer 3a, in which the R
and cyano groups are trans to each other.38 The orbital
symmetry theory predicts that thermal, suprafacial 1,3-shifts
with retention of configuration are forbidden.39 However,
1,3-shifts become allowed if the migrating group possesses a
p-orbital within a pseudopericyclic orbital topology,40 as we
observed. The stereochemistry of computed 3a is in very
(37) This result could indicate that the ionic path has a transition state too
high in energy which crosses the surface of the sigmatropic shift, or simply
does not exist.
(38) The trans-cyclopropane 3a is more stable than its contrathermo-
dynamic cis counterpart by 12.5 kcal mol-1 in the gas phase.
(39) (a) Hoffmann, R.; Woodward, R. B. J. Am. Chem. Soc. 1965, 87,
4389–4390. (b) Woodward, R. B.; Hoffmann, R. J. Am. Chem. Soc. 1965, 87,
2511–2513. (c) Pearson, R. G. Symmetry Rules for Chemical Reactions, 1st ed.;
John Wiley and Sons: New York, 1976.
(40) (a) Ross, J. A.; Seiders, R. P.; Lemal, D. M. J. Am. Chem. Soc. 1976,
98, 4325–4327. (b) Birney, D. M.; Wagenseller, P. E. J. Am. Chem. Soc. 1994,
116, 6262–6270. (c) Birney, D. M. J. Am. Chem. Soc. 2000, 122, 10917–10925.
1906 J. Org. Chem. Vol. 75, No. 6, 2010

ꢀ
Lopez-Camacho et al.
JOCArticle
FIGURE 10. Transition-state geometries (TS1 and TS2) in DMSO solvent phase, bond lengths (d, A), and dipole moments (μ, debyes) for the
[1,3]-rearrangements of 2a and 4a.
d6 4a=1.410 A; and d7 TS2=1.390 A and d7 4a=1.505 A).
Additionally, it can be envisaged that in TS1 and TS2 un-
favorable steric constraints developed between the OdC dou-
ble bond formation and the methyl group, explaining the
experimentally determined energetic penalty originated by
bulkier R groups. Because of the high degree of asynchronous
character of the transition states, polarization is found to be
higher than in the ground states. In the “hill climb” movement
from the reactants 2a and 4a to TS1 and TS2, respectively, the
dipole moment significantly increase, suggesting that consid-
erable charge separation is involved in transition states. The
dipole moments of compounds 2a and 4a and transition states
TS1 and TS2 are 3.91, 8.88, 12.97, and 12.81 D, respectively
(Figures 9 and 10). These computational studies forward to a
fundamental understanding of the experimentally observed
steric and solvent rate effects.
under argon atmosphere was added 6 (250 mg 0.87 mmol) in a
THF (10 mL) solution over a period of 30 min with stirring. The
temperature was maintained at 0 °C for 3 h during which the
mixture turned orange. The temperature was allowed to rise to
room temperature, and the reaction mixture was quenched with
saturated aqueous NH4Cl solution (5 mL) and diluted with
EtOAc (30 mL). The organic layer was decanted, washed with
saturated aqueous NH4Cl solution (2 ꢀ 10 mL), and dried
(Na2SO4). After evaporation of the solvent under vacuum, the
residue was purified by flash column chromatography (silica gel,
hexane/EtOAc 6:1) to give 195 mg of 1c (59%). Two recrystalli-
zations from hexane/AcOEt gave a colorless solid. Mp:
157-158 °C. IR (CH2Cl2, cm-1): νmax 3026, 2256, 1734 cm-1
.
1
Endo isomer (for 11:2 endo/exo mixture) H NMR (300 MHz,
CDCl3): δ 7.85 (br, 1H, H7), 7.68 (d, 1H, J=7.8 Hz, H4), 7.42 (t,
1H, J=7.3 Hz, H6), 7.34 (m, 3H, Ar), 7.22 (td, 1H, J=7.8, 1.0
Hz, H5), 7.04 (m, 2H, Ar), 6.38 (br, 1H, H8a), 4.15 (s, 1H,
interchangeable by deuterium, H3), 3.90 (s, 3H, COOMe), 3.13
and 3.08 (2d, 2H, JAB=14.2 Hz). 13C NMR (75 MHz, CDCl3) δ:
164.7 (s, C2), 155.1 (s, NCO), 140.2 (s, Ci), 133.2 (s, C7a), 131.0
(d, C6), 129.9 (d, CO), 129.3 (d, Cm), 128.3(d, Cp), 128.0 (d,
C3b), 125.7 (d, C4), 124.7 (d, C5), 115.9 (d, C7), 112.4 (s, CN),
93.7 (d, C8a), 55.2 (s, C3a), 53.7 (q, COOMe), 42.3 (t, CH2),
40.5(s, C3, Me). MS (EI), m/z: (relative inensity) 348 (Mþ, 100),
213 (79). HRMS (FAB), m/z: 349.1185 (Mþ, C20H16N2O4 þ H
requires 349.1188). Exo isomer: The 1H NMR spectrum of the
minor exo isomer cannot be fully characterized due to extensive
overlapped signals: δ 4.08 (s, 1H, interchangeable by deuterium,
H3), 3.79 (s, 3H, COOMe), 3.42 and 3.29 (2d, 2H, JAB=13.7
Hz).
Conclusions
The present study constitutes a significant advancement in
the mechanistic elucidation of the cascade rearrangements of
indol derived R-cyano ketene O,O-acetals to afford 2-indolyl
cyanomalonates in quantitative yields. On the basis of the
simplified kinetic equation used for data interpretation of the
kinetic results and computational studies, we propose that
the transformations involve the combination of four con-
certed [1,3]-sigmatropic shifts, and a vicinal hydrogen trans-
fer for the final aromatization step. A mechanistic highlight
resides in the pivotal role assigned to the steady-state cyclo-
propyl intermediate in directing the cascade rearrangements.
These processes are unique in the sense that, absent signifi-
cant polar substituent effects, the reaction rate of ketene O,
O-acetals decrease linearly and dramatically in the order of
increase the steric effect. The observed solvent rate effects are
attributable to a selective transition-state stabilization by
increased polarization stabilization because of the asynchro-
nous bond reorganization processes. Work is in progress to
profit these compounds as precursors to bioactive indolic
β-amino acids.
General Procedure for Reaction of Furoindolines 1a-e with
Diazomethane. To an Et2O (6 mL)/THF (2 mL) solution of the
corresponding 3a-alkyl-2-oxofuroindoline 1a-e (0.17 - 0.22
mmol) was added an excess of freshly prepared ethereal solution
of diazomethane13 (10 mL, ca. 5.3 mg CH2N2/mL, 1.3 mmol).
The reaction mixture was stirred at room temperature for 5 min
and evaporated at room temperature under atmospheric pres-
sure with a stream of argon which was bubbled into a solution
containing 5% acetic acid in ethanol.
Methyl 3-Cyano-2-methoxy-3a-methyl-3aH-furo[2,3-b]indole-
8(8aH)-carboxylate (2a), Methyl 3-Cyano-2-methoxy-8b-methyl-
3aH-furo[3,2-b]indole-4(8bH)-carboxylate (4a), and Methyl 2-(1-
Cyano-2-methoxy-2-oxoethyl)-3-methyl-1H-indole-1-carboxylate
(5a). Following the generalprocedure, reaction offuroindoline1a
with diazomethane gave a colorless oil mixture of isomeric
products 2a, and 4a in proportions of 1:1 in an overall >99%
yield. This material was standing at 22 °C in DMSO (1 mL) until
Experimental Section
3a-Benzyl 2-Oxofuroindoline (1c). To a cooled solution (0 °C)
of BnMgBr (683.7 mg, 3.5 mmol) in Et2O (13 mL)/THF (20 mL)
J. Org. Chem. Vol. 75, No. 6, 2010 1907

ꢀ
Lopez-Camacho et al.
JOCArticle
its complete transformation (1 h), and then concentrated to
dryness to afford the pure enough indole 5a as colorless solid in
>99% yield. Due to the reaction progress, the spectral NMR
data of 2a and 4a listed below stem from the 2:1 mixture for 1H
and from the 1:2 mixture for 13C.
137-138 °C. Rf: 0.42 (CH2Cl2). IR (CHCl3): νmax 3028, 2252,
1752, 1606 cm-1. 1H NMR (300 MHz, acetone-d6): δ 8.21 (dt,
1H, J=8.3, 0.8, H7), 7.74 (dd, 1H, J=7.8, 1.4 Hz, H4), 7.46 (td,
1H, J=7.2, 1.4 Hz, H6), 7.37 (td, 1H, J=7.7, 1.1 Hz, H5), 6.00 (s,
1H, H8), 4.10 (s, 3H, NCOOMe), 3.84 (s, 3H, COOMe), 2.92 (qi,
2H, J=2.2 Hz, CH2), 1.28 (t, 3H, J=7.4 Hz, Me). 13C NMR (75
MHz, acetone-d6): δ 166.5 (s, CCO), 153.1 (s, NCO), 137.5 (s,
C7a), 130.2 (s, C3a), 128.3 (s, C3), 127.4 (d, C6), 125.3 (s, C2),
125.0 (d, C5), 121.0 (d, C4), 117.3 (d, C7), 116.6 (s, CN), 55.1 (q,
NCOOMe), 54.6 (q, COOMe), 37.6 (d, C8), 18.3 (t, CH2), 15.9
(q, Me). MS (EI), m/z: (relative inensity) 300 (Mþ, 48), 268 (100),
241 (186). HRMS (FAB), m/z: 323.1011 (Mþ, C16H16N2O4 þ
Na requires 323.1008).
Methyl 3a-Benzyl-3-cyano-2-methoxy-3aH-furo[2,3-b]indole-
8(8aH)-carboxylate (2c), Methyl 8b-Benzyl-3-cyano-2-methoxy-
3aH-furo[3,2-b]indole-4(8bH)-carboxylate (4c), and Methyl 3-
Benzyl-2-(1-cyano-2-methoxy-2-oxoethyl)-1H-indole-1-carboxy-
late (5c). Following the general procedure, reaction of furoindo-
line 1c with diazomethane gave a colorless oil mixture of
isomeric products 2c and 4c in proportions of 9:1 in an overall
>99% yield. This material was standing at 22 °C in DMSO
(1 mL) until its complete transformation (8 h) and then con-
centrated to dryness to afford pure enough indole 5c as colorless
crystals in >99% yield.
1
2a. H NMR (300 MHz, acetone-d6): δ 7.83 (br s, 1H, H7),
7.37 (dd, J=7.4, 1.4 Hz, H4), 7.33 (td, 1H, J=7.4, 1.4 Hz, H6),
7.17 (td, J=7.7, 1.1 Hz, H5), 6.56 (s, 1H, H8a), 4.05 (s, 3H,
OMe), 3.93 (s, 3H, COOMe), 1.71 (s, 3H, Me). 13C NMR (75
MHz, acetone-d6): δ 170.7 (s, C2), 154.3 (s, N-CO), 141.5 (s,
C7a), 137.3 (s, C3b), 130.2 (d, C6), 125.7 (d, C5), 123.9 (d, C4),
116.6 (d, C7), 116.4 (s, CN), 101.4 (d, C8a), 65.3 (s, C3), 59.5 (q,
OMe), 55.1 (s, C3a), 54.4 (q, COOMe), 24.4 (q, Me).
1
4a. H NMR (300 MHz, acetone-d6): δ 7.94 (br s, 1H, H7),
7.57 (dd, 1H, J=7.7, 1.4 Hz, H4), 7.46 (td, J=7.4, 1.4 Hz, H6),
7.17 (td, J=7.7, 1.1 Hz, H5), 5.47 (s, 1H, H8a), 4.02 (s, 3H,
OMe), 3.85 (s, 3H, COOMe), 1.94 (s, 3H, Me). 13C NMR (75
MHz, acetone-d6): δ 173.4 (s, C2), 153.8 (s, N-CO), 143.7 (s,
C7a), 132.8 (d, C6), 131.2 (s, C3b), 125.9 (d, C4), 124.9 (d, C5),
117.1 (s, CN), 116.7 (d, C7), 95.3 (s, C3a), 72.5 (d, C8a), 60.0 (s,
C3), 59.1 (q, OMe), 53.5 (q, COOMe), 24.7 (q, Me).
5a. Pure sample was obtained by recrystallization from a
mixture of CH2Cl2/hexane to give a white solid. Mp: 138-
140 °C. Rf 0.39 (CH2Cl2). IR (CHCl3) νmax 3016, 2254, 1752,
1612 cm-1; 1H NMR (300 MHz, acetone-d6): δ 8.19 (dt, 1H, J=
8.2, 0.9, H7), 7.68 (dd, 1H, J=7.7, 1.5 Hz, H4), 7.46 (td, 1H, J=
7.2, 1.3 Hz, H6), 7.37 (td, 1H, J=7.7, 1.1 Hz, H5), 6.00 (s, 1H,
H8), 4.11 (s, 3H, NCOOMe), 3.84 (s, 3H, COOMe), 2.43 (s, 3H,
Me). 13C NMR (75 MHz, acetone-d6): δ 166.4 (s, CCO), 153.1 (s,
NCO), 137.2 (s, 7a), 131.2 (s, C3a), 127.5 (d, C6), 125.7 (s, C3),
125.0 (d, C5), 122.1 (s, C2), 121.1 (d, C4), 117.1 (d, C7), 116.5 (s,
CN), 55.1 (q, NCOOMe), 54.7 (q, COOMe), 37.7 (d, C8), 9.5 (q,
Me). MS (EI), m/z: (relative inensity) 286 (Mþ, 44), 254 (100),
227 (81). HRMS (FAB), m/z: 309.0859 (Mþ, C15H14N2O4 þ Na
requires 309.0851).
2c. Pure sample was obtained by recrystallization from a
mixture of ether/hexane to give colorless crystals. Mp: 115-
116 °C. 1H NMR (300 MHz, acetone-d6): δ 7.74 (br s, 1H, H7),
7.53 (m, 1H, H4), 7.35 (td, 1H, J=7.7, 1.4 Hz, H6), 7.31-7.27 (m,
5H, Ar), 7.21 (td, 1H, J=7.4, 1.1 Hz, H5), 6.63 (s, 1H, H8a), 3.91
(s, 3H, OMe), 3.82 (s, 3H, COOMe), 3.48 and 3.25 (2d, 2H, JAB
=
13.8 Hz, CH2). 13C NMR (75 MHz, acetone-d6): δ 171.1 (s, C2),
153.8 (s, NCO), 141.8 (s, C7a), 136.9 (s, Ci), 136.1 (s, C3b), 131.7
(d, Co), 130.0 (d, C6), 129.9 (d, Cp), 128.6 (d, Cm), 125.6 (d, C5),
124.2 (d, C4), 116.6 (s, CN), 116.6 (d, C7), 98.4 (d, C8a), 64.2 (s,
C3), 60.2 (s, C3a), 59.4 (q, OMe), 54.4 (q, COOMe), 42.1 (t, CH2).
4c. Due to the reaction progress, the spectral NMR data of 4c
listed below stem from the 1:1 mixture of 2c and 4c for 1H and
from the 1:2 mixture for 13C. 1H NMR (300 MHz, acetone-d6): δ
7.91 (br s, 1H, H7), 7.54 (dd, 1H, J=7.7, 1.4 Hz, H4), 7.45 (m,
1H, H6), 7.39-7.33 (m, 5H, Ar), 7.16 (td, 1H, J=7.4, 0.8 Hz,
H5), 5.58 (s, 1H, H8a), 3.97 (s, 3H, OMe), 3.77 (s, 3H, COOMe),
3.70 and 3.44 (2d, 2H, JAB=13.8 Hz, CH2). 13C NMR (75 MHz,
acetone-d6): δ 173.1 (s, C2), 153.7 (s, NCO), 143.8 (s, C7a), 135.4
(s, Ci), 132.9 (d, C6), 132.0 (d, Co), 130.5 (d, Cp), 130.3 (s, C3b),
129.0 (d, Cm), 126.4 (d, C4), 124.7 (d, C5), 116.7 (s, CN), 116.5
(d, C7), 96.9 (s, C3a), 69.8 (d, C8a), 60.4 (s, C3), 59.1 (q, OMe),
53.5 (q, COOMe), 43.5 (t, CH2).
Methyl 3-Cyano-3a-ethyl-2-methoxy-3aH-furo[2,3-b]indole-
8(8aH)-carboxylate (2b), Methyl 3-Cyano-8b-ethyl-2-methoxy-
3aH-furo[3,2-b]indole-4(8bH)-carboxylate (4b), and Methyl 2-(1-
Cyano-2-methoxy-2-oxoethyl)-3-ethyl-1H-indole-1-carboxylate
(5b). Following the general procedure, reaction of furoindoline
1b with diazomethane gave a colorless oil mixture of isomeric
products 2b and 4b in proportions of 7:1 in an overall >99%
yield. This material was standing at 22 °C in DMSO (1 mL) until
its complete transformation (4 h), and then concentrated to
dryness to afford pure enough indole 5b as colorless crystals in
>99% yield. Due to the reaction progress, the spectral NMR
data of 2b and 4b listed below stem from the 2:1 mixture for 1H
and from the 1:1 mixture for 13C.
5c. Pure sample was obtained by recrystallization from a
mixture of CH2Cl2/hexane to give colorless crystals. Mp:
144-145 °C. Rf: 0.39 (CH2Cl2). IR (CHCl3): νmax 3024, 2254,
1756, 1604 cm-1. 1H NMR (300 MHz, acetone-d6): δ 8.22 (dt,
1H, J=8.4, 0.9, H7), 7.56 (dd, 1H, J=7.8, 1.3 Hz, H4), 7.43 (td,
1H, J=7.2, 1.3 Hz, H6), 7.36-7.22 (m, 5H, Ar), 7.28 (td, 1H, J=
1
2b. H NMR (300 MHz, acetone-d6): δ 7.79 (br s, 1H, H7),
7.33 (dd, 1H, J=7.3, 1.4 Hz, H4), 7.32 (td, 1H, J=7.2, 1.6 Hz,
H6), 7.15 (td, 1H, J=7.5, 1.2 Hz, H5), 6.58 (s, 1H, H8a), 4.03 (s,
3H, OMe), 3.90 (s, 3H, COOMe), 2.10 and 1.99 (2dq, 2H, J=
14.3, 7.4 Hz, CH2), 0.92 (t, 3H, J=7.4 Hz, Me). 13C NMR (75
MHz, acetone-d6): δ 170.9 (s, C2), 154.2 (s, NCO), 141.8 (s,
C7a), 136.1 (s, C3b), 130.2 (d, C6), 125.6 (d, C5), 124.0 (d, C4),
116.9 (s, CN), 116.5 (d, C7), 98.9 (d, C8a), 63.6 (s, C3), 59.8 (s,
C3a) 59.5 (q, OMe), 54.4 (q, COOMe), 29.7 (t, CH2), 9.3 (q, Me).
7.2, 0.9 Hz, H5), 6.11 (s, 1H, H8), 4.36 and 4.29 (2d, 2H, JAB
=
C
16.2, CH2), 4.13 (s, 3H, NCO2Me), 3.85 (s, 3H, COOMe). 13
NMR (75 MHz, acetone-d6): 166.4 (s, CCO), 153.1 (s, NCO),
140.7 (s, Ci), 137.6 (s, C7a), 130.1 (s, C3a), 130.0 (d, Cm), 129.9
(d, Co), 127.9 (d, Cp), 127.5 (d, C6), 126.8 (s, C2), 125.1 (s, C3),
125.0 (d, C5), 121.7 (d, C4), 117.3 (d, C7), 116.4 (s, CN), 55.2 (q,
NCOOMe), 54.7 (q, COOMe), 37.9 (d, C8), 30.6 (t, CH2). MS
(EI), m/z: (relative inensity) 362 (Mþ, 55), 330 (47), 264 (100),
243 (49). HRMS (FAB), m/z: 385.1172 (Mþ, C21H18N2O4 þ Na
requires 385.1172).
1
4b. H NMR (300 MHz, acetone-d6): δ 7.92 (br s, 1H, H7),
7.50 (dd, 1H, J=7.6, 1.4 Hz, H4), 7.44 (td, 1H, J=7.5, 1.5 Hz,
H6), 7.14 (td, 1H, J=7.5, 1.0 Hz, H5), 5.55 (s, 1H, H8a), 4.00 (s,
3H, OMe), 3.82 (s, 3H, COOMe), 2.32 and 2.19 (2dq, 2H, J=
14.2, 7.3 Hz, CH2), 0.97 (t, 3H, Me). 13C NMR (75 MHz,
acetone-d6): δ 173.5 (s, C2), 153.8 (s, NCO), 143.8 (s, C7a),
132.9 (d, C6), 130.4 (s, C3b), 126.0 (d, C4), 124.9 (d, C5), 116.8
(d, C7), 116.4 (s, CN), 98.2 (s, C3a), 69.7 (d, C8a), 60.2 (s, C3),
59.1 (q, OMe), 53.5 (q, COOMe), 31.1 (t, CH2), 8.3 (q, Me).
5b. Pure sample was obtained by recrystallization from a
mixture of CH2Cl2/hexane to give colorless crystals. Mp:
Methyl 3-Cyano-3a-isopropyl-2-methoxy-3aH-furo[2,3-b]indole-
8(8aH)-carboxylate (2d), Methyl 3-Cyano-8b-isopropyl-2-meth-
oxy-3aH-furo[3,2-b]indole-4(8bH)-carboxylate (4d), and Methyl
2-(1-Cyano-2-methoxy-2-oxoethyl)-3-isopropyl-1H-indole-1-
carboxylate (5d). Following the general procedure, reaction of
1908 J. Org. Chem. Vol. 75, No. 6, 2010

ꢀ
Lopez-Camacho et al.
JOCArticle
furoindoline 1d with diazomethane gave compound 2d as color-
less oil in >99% yield.
(br s, 1H, H7), 7.54 (dd, 1H, J=7.4, 1.1 Hz, H4), 7.38 (td, 1H, J=
7.4, 1.4 Hz, H6), 7.20 (td, 1H, J=7.7, 1.1 Hz, H5), 6.62 (s, 1H,
H8a), 4.01 (s, 3H, OMe), 3.95 (s, 3H, COOMe), 1.13 (s, 9H,
3Me). 13C NMR (75 MHz, acetone-d6): δ 172.5 (s, C2), 153.9 (s,
NCO), 142.2 (s, C7a), 134.8 (s, C3b), 130.2 (d, C6), 126.5 (d, C4),
125.3 (d, C5), 117.9 (s, CN), 116.9 (s, C7), 98.5 (d, C8a), 65.5 (s,
C3a), 62.4 (s, C3), 59.3 (q, OMe), 54.6 (q, COOMe), 37.8 (s, C),
26.5 (q, 3Me).
2d. Pure sample was obtained by silica gel chromatography
with CH2Cl2/hexane (1:1) to give a colorless oil. Rf: 0.40
(CH2Cl2). IR (CH2Cl2): νmax 3058, 2202, 1728, 1656, 1602
cm-1. 1H NMR (300 MHz, acetone-d6): δ 7.83 (br s, 1H, H7),
7.37 (dd, 1H, J=7.2, 1.4 Hz, H4), 7.36 (td, 1H, J=7.4, 1.4 Hz,
H6), 7.19 (td, 1H, J=7.4, 1.1 Hz, H5), 6.60 (s, 1H, H8a), 4.05 (s,
3H, OMe), 3.93 (s, 3H, COOMe), 2.47 (h, 1H, J=6.9 Hz, i-Pr),
Isolation of 4e. Compound 2e (0.08 mmol) was standing at
room temperature in DMSO (0.5 mL) until ca. 80% conversion
of 2e was achieved (48 h). The reaction mixture was diluted with
CH2Cl2 (10 mL), and the organic phase was washed with brine
(12 ꢀ 1 mL). Removal of the solvent under an argon stream led
to an oily residue which was chromatographed over a short path
of silica gel (10 cm, column 50 ꢀ 1.5 cm) eluted with hexane/
CH2Cl2 3:2. The fractions containing 4e were collected (25 mL)
and concentrated in vacuo at 40 °C to afford the pure compound
as a colorless oil in 58% yield. Pure sample was obtained by
recrystallization from a mixture of CH2Cl2/hexane to give
colorless crystals. Mp: 160-162 °C. Rf: 0.20 (CH2Cl2). IR
1.18 (d, 3H, J=6.9 Hz, i-Pr), 0.76 (d, 3H, J=6.9 Hz, i-Pr). 13
C
NMR (75 MHz, acetone-d6): δ 170.8 (s, C2), 154.1 (s, NCO),
141.9 (s, C7a), 135.9 (s, C3b), 130.3 (d, C6), 125.7 (d, C5), 123.9
(d, C4), 116.6 (d, C7), 116.5 (s, CN), 96.4 (d, C8a), 68.8 (s, C3),
63.9 (s, C3a), 59.5 (q, OMe), 54.5 (q, COOMe), 33.3 (d, CH),
18.4 (q, Me), 17.5 (q, Me).
Isolation of 4d. Compound 2d (0.08 mmol) was allowed to
stand at room temperature in DMSO (0.5 mL) until ca. 50%
conversion of 2d was achieved (3 h). The reaction mixture was
diluted with CH2Cl2 (10 mL), and the organic phase was washed
with brine (12 ꢀ 1 mL). Removal of the solvent under an argon
stream led to an oil residue which was chromatographed over a
short path of silica gel (10 cm, column 50 ꢀ 1.5 cm) eluted with
hexane/CH2Cl2 3:2. The fractions containing 4d were collected
(25 mL) and concentrated in vacuo at 40 °C to afford the pure
compound as a colorless oil in 27% yield. Pure sample was
obtained by recrystallization from a mixture of CH2Cl2/hexane
to give colorless crystals. Mp: 122-124 °C. Rf: 0.16 (CH2Cl2).
IR (CH2Cl2): νmax 3080, 2210, 1716, 1640, 1604 cm-1. 1H NMR
(300 MHz, acetone-d6): δ 7.93 (br s, 1H, H7), 7.49 (dd, 1H, J=
7.7, 1.5 Hz, H4), 7.45 (td, 1H, J=7.4, 1.5 Hz, H6), 7.15 (td, 1H,
J=7.5, 1.0 Hz, H5), 5.60 (s, 1H, H8a), 4.02 (s, 3H, OMe), 3.83 (s,
3H, COOMe), 2.67 (h, 1H, J=6.8 Hz, i-Pr), 1.09 (d, 3H, J=6.5
Hz, i-Pr), 0.82 (d, 3H, J=7.1 Hz, i-Pr). 13C NMR (75 MHz,
acetone-d6): δ 173.3 (s, C2), 153.7 (s, NCO), 143.7 (s, C7a), 132.9
(d, C6), 130.3 (s, C3b), 126.0 (d, C4), 125.0 (d, C5), 116.8 (d, C7),
116.7 (s, CN), 100.8 (s, C3a), 67.2 (d, C8a), 60.6 (s, C3), 59.2 (q,
OMe), 53.5 (q, COOMe), 35.4 (d, CH), 17.2 (q, Me), 16.5 (q,
Me).
1
(CH2Cl2): νmax 3070, 2208, 1716, 1644, 1602 cm-1. H NMR
(300 MHz, acetone-d6): δ 7.95 (br s, 1H, H7), 7.55 (dd, 1H, J=
7.8, 1.3 Hz, H4), 7.44 (td, 1H, J=7.5, 1.3 Hz, H6), 7.13 (td, 1H,
J=7.5, 1.1 Hz, H5), 5.71 (s, 1H, H8a), 4.04 (s, 3H, OMe), 3.83 (s,
3H, COOMe), 1.09 (s, 9H, 3 Me). 13C NMR (75 MHz, acetone-
d6): δ 173.0 (s, C2), 153.6 (s, NCO), 143.6 (s, C7a), 132.6 (d, C6),
129.7 (s, C3b), 127.9 (d, C4), 124.6 (d, C5), 116.9 (d, C7), 116.4
(s, CN), 101.8 (s, C3a), 68.7 (d, C8a), 60.3 (s, C3), 59.1 (q, OMe),
53.6 (q, COOMe), 38.9 (s, C), 24.8 (q, 3Me).
5e. Compound 4e was standing at 22 °C in DMSO (1 mL)
until its complete transformation (480 h) and then concentrated
to dryness to afford indole 5e as a colorless oil in >99% yield.
Pure sample was obtained by recrystallization from a mixture of
CH2Cl2/hexane to give white solid. Mp: 118-120 °C. Rf: 0.39
1
(CH2Cl2). IR (CHCl3) νmax 3030, 2252, 1752, 1586 cm-1. H
NMR (300 MHz, acetone-d6): δ 8.25 (ddd, 1H, J=8.5, 1.2, 0.8
Hz, H7), 8.07 (dd, 1H, J=8.2, 1.1, 0.6 Hz, H4), 7.42 (td, 1H, J=
7.1, 1.1 Hz, H6), 7.33 (td, 1H, J=7.4, 1.4 Hz, H5), 6.28 (s, 1H,
H8), 4.09 (s, 3H, NCO2Me), 3.89 (s, 3H, COOMe), 1.68 (s, 9H,
3Me). 13C NMR (75 MHz, acetone-d6): δ 166.5 (s, CCO), 153.2
(s, NCO), 137.9 (s, C7a), 133.0 (s, C3), 129.8 (s, C3a), 126.8 (d,
C6), 124.6 (d, C4), 124.3 (d, C5), 124.3 (s, C2), 117.1 (s, C7),
117.0 (d, CN), 55.2 (q, NCOOMe), 54.6 (q, CCOOMe), 38.9 (d,
C8), 35.5 (s, C), 32.7 (q, 3Me). MS (EI), m/z: (relative intensity)
328 (Mþ, 20), 282 (81), 241 (100). HRMS (FAB), m/z: 329.1497
(Mþ, C17H18N2O4 þ H requires 329.1501).
5d. Compound 4d was standing at 22 °C in DMSO (1 mL)
until its complete transformation (20 h) and then concentrated
to dryness to afford the pure enough indole 5d as colorless oil in
>99% yield. Pure sample was obtained by recrystallization
from a mixture of CH2Cl2/hexane to give colorless crystals. Mp:
163-165 °C. Rf: 0.38 (CH2Cl2). IR (CHCl3): νmax 3020, 2252,
1754, 1602 cm-1. 1H NMR (300 MHz, acetone-d6): δ 8.24 (dt,
1H, J=8.5, 1.1, H7), 7.92 (dd, 1H, J=8.0, 1.4 Hz, H4), 7.44 (td,
1H, J=7.1, 1.4 Hz, H6), 7.34 (td, 1H, J=7.7, 1.1 Hz, H5), 6.08 (s,
1H, H8), 4.10 (s, 3H, NCOOMe), 3.85 (s, 3H, COOMe), 3.45 (h,
1H, J=7.1 Hz, i-Pr), 1.50 (d, 3H, J=6.9 Hz, i-Pr), 1.49 (d, 3H,
J=7.1 Hz, i-Pr). 13C NMR (75 MHz, acetone-d6): δ 166.5 (s,
CCO), 153.1 (s, NCO), 138.0 (s, C7a), 131.5 (s, C3), 129.0 (s,
C3a), 127.1 (d, C6), 124.7 (d, C4), 124.6 (s, C2), 122.7 (d, C4),
117.4 (d, C7), 116.7 (s, CN), 55.1 (q, NCOOMe), 54.7 (q,
COOMe), 37.5 (d, C8), 27.5 (d, CH), 23.1 (q, Me), 22.9 (q,
Me). MS (EI), m/z: (relative inensity) 314 (Mþ, 72), 282 (67), 267
(100), 255 (62). HRMS (FAB) m/z: 337.1175 (Mþ, C17H18-
N2O4 þ Na requires 337.1164).
NMR Kinetic Experiments. 1H NMR studies (300 MHz) were
carried out in duplicated with a substrate concentrations of
0.03-0.04 mmol/mL of DMSO-d6 solvent. The time interval
between two successive experiments was dependent on the rate
of rearrangement in order to collect a number of data points
for accurate rate determination (12-25 registered spectra). All
reactions were monitored up to 80% conversion of starting
material (from 0.50 to 76 h). Analyses of 2a and 2c were carried
out at 295 K; for 2b at 295, 303, and 310 K; for 2d at 295, 303,
310, and 319 K; for 2e at 295, 303, 311, 319 K; for 4d at 295,
301, 311, and 319 K; and for 4e at 308, 318, and 328 K. After
the experiments in a kinetic run were finished, they were all
integrated serially. The singlet or singlets corresponding to the
methine proton or protons in the first experiment of the
series were normalized to 100% and the integrals of the singlet
methine peaks appearing in successive experiments were mea-
sured against the first experiment. The residual solvent
peaks from DMSO-d6 (at 2.56 ppm) served as an internal
standard for this analysis: the combined integrals for pro-
ducts and starting material methine singlets were constant
relative to the DMSO-d6 methyl groups over the course of
the reaction.
Methyl 3-Cyano-3a-tert-butyl-2-methoxy-3aH-furo[2,3-b]indole-
8(8aH)-carboxylate (2e), Methyl 3-Cyano-8b-tert-butyl-2-meth-
oxy-3aH-furo[3,2-b]indole-4(8bH)-carboxylate (4e), and Methyl
2-(1-Cyano-2-methoxy-2-oxoethyl)-3-tert-butyl-1H-indole-1-
carboxylate (5e). Following the general procedure, reaction of
furoindoline 1e with diazomethane gave compound 2e as color-
less oil in >99% yield.
2e. Pure sample was obtained by recrystallization from a
mixture of CH2Cl2/hexane to give colorless crystals. Mp:
130-132 °C. Rf 0.43 (CH2Cl2). IR (CH2Cl2): νmax 3058, 2202,
1728, 1646, 1602 cm-1. 1H NMR (300 MHz, acetone-d6): δ 7.85
J. Org. Chem. Vol. 75, No. 6, 2010 1909

ꢀ
Lopez-Camacho et al.
JOCArticle
Acknowledgment. Financial support by CONACYT
Grant No. 8180 is gratefully acknowledged. This work is
part of the doctoral thesis of P.Y.L.C.
assignment. X-ray crystallographic data for compounds
2e, 3c, 4d, and 4e. Kinetic data (rate constants of indivi-
dual kinetic runs). Cartesian coordinates, total energies,
thermochemistry, and variational PCM results (DMSO)
for calculated molecules and transition structures. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
Supporting Information Available: Copies of H and 13C
NMR spectra for all new compounds and tables of
all observed gHSQC and gHMBC interactions used for
1910 J. Org. Chem. Vol. 75, No. 6, 2010