Determining the scope of the lanthanide mediated, sequential hydroamination/C-C cyclization reaction: Formation of tricyclic and tetracyclic aromatic nitrogen heterocycles

Article abstract of DOI:10.1016/j.tet.2003.08.071

The scope of the lanthanide mediated, sequential hydroamination/C-C cyclization reaction was determined for the formation of tricyclic and tetracyclic aromatic nitrogen heterocycles. An array of ring sizes was explored to determine the diastereoselectivity. The electronic characteristics of the aromatic ring was also varied to determine how it affected the cascade reaction. It was found that the benzo[a]quinolizine and the pyrido[2,1,a]isoindolizine ring systems formed with the highest diastereoselectivity (>20:1), regardless of the electronic characteristics of the aromatic ring. Additionally, a tetracyclic indole nitrogen heterocycle was formed with a 2.3:1 diastereomeric ratio. A novel procedure for substrate preparation is also presented.

Full text of DOI:10.1016/j.tet.2003.08.071

Tetrahedron 59 (2003) 10581–10591
Determining the scope of the lanthanide mediated,
sequential hydroamination/C–C cyclization reaction: formation
of tricyclic and tetracyclic aromatic nitrogen heterocycles
*
Gary A. Molander and Shawn K. Pack
Roy and Diana Vagelos Laboratories, Department of Chemistry, 231 South 34th Street, University of Pennsylvania,
Philadelphia, PA 19104-6323, USA
Received 25 June 2003; accepted 7 August 2003
Abstract—The scope of the lanthanide mediated, sequential hydroamination/C–C cyclization reaction was determined for the formation of
tricyclic and tetracyclic aromatic nitrogen heterocycles. An array of ring sizes was explored to determine the diastereoselectivity. The
electronic characteristics of the aromatic ring was also varied to determine how it affected the cascade reaction. It was found that the
benzo[a]quinolizine and the pyrido[2,1,a]isoindolizine ring systems formed with the highest diastereoselectivity (.20:1), regardless of the
electronic characteristics of the aromatic ring. Additionally, a tetracyclic indole nitrogen heterocycle was formed with a 2.3:1 diastereomeric
ratio. A novel procedure for substrate preparation is also presented.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
several issues must be addressed (Fig. 1). First, it was
uncertain if this method could be applied to the formation
of quinolizidine ring systems. Although the formation of
dihydroindolizine via the hydroamination reaction is
known, whether the sequential reaction would tolerate the
various aromatic ring systems such as those shown for
proteomitine and the indole alkaloids was unknown.2
Lastly, the diastereoselectivity of the formation of various
Reactions involving the formation of several bonds and ring
systems in a single synthetic operation have proven to be the
most powerful in the arsenal of the synthetic organic
chemist. However, the utility of these reactions cannot be
fully realized until the scope of the reaction is determined.
Li and Marks first presented the novel lanthanide-mediated,
sequential intramolecular hydroamination/C–C bond-
forming cyclization reaction (Eq. (1)).1
ð1Þ
In a thorough study, they have shown that, unlike other
metal catalyzed intramolecular, hydroamination reactions,
the organolanthanide formed after the initial hydro-
amination can insert into another p-system to form saturated
and partially saturated indolizines and pyrrolizines
(Scheme 1).
One of the primary goals of our laboratory is the
development and application of various methods toward
the synthesis of natural products. To apply the powerful
method outlined by Li and Marks to the synthesis of
complex natural products such as proteomitine, the indole
alkaloids, the necine nucleus and the pumiliotoxin core,
Keywords: hydroamination; lanthanide; sequential reaction; aromatic
nitrogen heterocycle; diastereoselectivity.
*
Corresponding author. Tel.: þ1-215-573-8604; fax: þ1-215-573-7165;
ð2Þ
e-mail: gmolandr@sas.upenn.edu
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.08.071

10582
G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
Scheme 1.
ring systems such as the necine nucleus and the pumilio-
toxin core, as well as the quinolizidine ring system, has not
been determined.
Herein we describe our efforts to address these questions as
well as our efforts to simplify the process of substrate
purification.
ð3Þ
ð4Þ
2. Results and discussion
To resolve some of the issues raised above, substrates 1–10
were prepared (Eqs. (2)–(4)).
Substrates 9 and 10 were synthesized from the nitrile
213 based on a modified procedure outlined by Brussee
(Scheme 2).4
Because of the acidic protons alpha to the nitrile and the
aromatic group, the following modifications were required:
1.2 equiv. of DIBALH were used rather than 2.5 equiv., the
HCl salt of the amine was utilized to quench the iminium
solution rather than NH4Cl, and NaCNBH3 was employed
rather than NaBH4 for the reduction of the secondary imine.
Model studies using benzyl nitrile indicated clean formation
of the primary imine (Scheme 3). However, formation of
the secondary imine followed by reduction resulted in
only 70% crude GC purity of product with the major
impurity being the primary amine followed by some late
eluting impurities. In comparison, the crude GC purity of
the aliphatic counterpart is .90%.
Figure 1.

G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
10583
Scheme 2. Key: (a) (i) KCN, EtOH, reflux; (ii) MeI, EtOAc/CH2Cl2 (91%). (b) (i) 10 wt% aq. NaOH, MeOH/H2O, rt to 808C; (ii) Bu4NHSO4, BnCl, CH2Cl2,
50 wt% aq. NaOH (73%). (c) (i) 1 M DIBALH in toluene, Et2O, 2208C to rt; (ii) allylamine or homoallylamine, NaCNBH3, MeOH, 2208C to rt (33–41%).
Scheme 3.
Increasing the amount of amine to 5 equiv., increasing or
decreasing the pH of the transimination reaction, and
changing the reducing agent of the second reduction had no
effect on the outcome of the reaction. However, decreasing
the number of equivalents of DIBALH to 1.5 and using
NaCNBH3 for the second reduction resulted in complete
consumption of the primary imine. Unfortunately, this
procedure also generates 5–20% of some late eluting
impurities. Consequently, incorporating this procedure only
resulted in a modest increase in yield, from an average of
25 to 41%.
toward the synthesis of other aromatic nitriles proved to be
difficult (Scheme 4).
Efforts to open the isoquinoline ring with a nitirile source
resulted in an elimination product (23 or 24) rather than the
desired aromatic amino nitrile. Unfortunately, this product
could not undergo further transformation without decompo-
sition. NMR experiments indicated consumption of starting
material and loss of the vinyl protons at 150–1808C in
deuterated DMSO. It is believed that the displacement of the
quaternary amine occurred at these elevated temperatures.
However, the resulting product was unstable under these
extreme conditions. Various nitrile sources were used
Attempts to apply the method outlined by Sapi3 (Scheme 2)
Scheme 4.

10584
G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
manipulated further because of the electron rich aromatic
ring. Attempts to make the corresponding tosylate, bromide,
or chloride resulted in decomposition and unpredictable
yields (10–70%). Consequently, an alternative approach
was taken to prepare the precursor 33 to substrates 5 and 6
(Eq. (6)).
Scheme 5. Key: (a) 10% Pd(PPh3)4, 1.1 equiv. tributylvinylstannane,
toluene, sealed 100–1108C (80%). (b) 5.0 equiv. KOH, 1.3 equiv. TosCl,
THF, 2108C to rt (99%). (c) 3.5 equiv. KCN, MeCN, rt (95%).
ð6Þ
including NaCN, KCN, LiCN, CuCN, and NaCu(CN)2. A
more conventional approach using a Stille coupling was
used to install the vinyl moiety (Schemes 5 and 6) and
obtain the desired nitrile derivatives 28 and 32.5
The tertiary amine 23, prepared according to the procedure
outlined above, was treated with methylchloroformate to
provide the chloride 33.7 It should be noted that this
procedure could also be used to make the chloro analog of
27 from the corresponding amine 24.
Several advantages were realized by isolating the HCl salts
of the substrates. First, handling solids are much easier than
handling liquids, especially in milligram quantities. Second,
oxidation of the amines was greatly inhibited, allowing for
easy, prolonged storage. Third, loss of substrate owing to
the volatility of the free amine is not an issue. Lastly, the
requisite purity for catalytic reactions could be achieved on
milligram scale. In comparison, gram quantities of material
are desired to obtain comparably pure material by
distillation.
Scheme 6. Key: (a) 1.3 equiv. PPh3, 1.35 equiv. imidazole, 1.4 equiv. NCS,
CH2Cl2, 08C to rt (73%). (b) 2.5 equiv. KCN, DMF (82%). (c) 10%
Pd(PPh3)4, 1.1 equiv. tributylvinylstannane, toluene, sealed 100–1108C
(48%).
Trace amounts of Pd result in isomerization of the allylic
or homoallylic double bond (ca. 3–20%). Further, the
presence of Pd in the substrate kills the catalytic cycle.
Hence, the Stille coupling was incorporated as early in the
synthesis as possible. In addition, multiple purifications by
silica gel chromatography were required.
Unfortunately, the HCl salt of the amine destroys the
catalyst. Consequently, a simple free base protocol was
required. An anhydrous, easily filtered, polymer base that is
active enough to neutralize the HCl salt in a tri-phasic
system (HCl salt, benzene, and polymer) would be ideal.
However, to our knowledge, such a polymer does not exist.
Several anhydrous bases such as the metal hydrides and
metal carbonates were unsuccessful or the requisite process
too complicated. Eventually, the free amine was liberated
with aqueous NaOH and extracted with benzene. Drying the
benzene solution of the amine over activated molecular
sieves proved, to our delight, adequate enough to remove
the residual moisture without absorbing significant amounts
of the free amine. The benzene solution was then degassed
and taken into a glove box where it was treated with the
catalyst.
The instability of the dimethoxy analogs of 26 and 27
required the installation of the vinyl moiety of 32 to be last
(Scheme 6). Again, multiple purifications by silica gel
chromatography were required to obtain the desired purity.
This purification, coupled with the inherent instability of the
dimethoxy styrene core, resulted in a lower yield of nitrile
32. Using the above mentioned transimination reaction,
substrates 3, 4, 7, and 8 were prepared from nitriles 28 and
32 (39–51%). The HCl salts of 3 and 4 were isolated.
Treatment of the dimethoxy derivatives with HCl resulted in
chlorination of the vinyl group, hence substrates 5–8 were
isolated as the free base.
Substrates 1, 2, 5, and 6 were prepared from the
corresponding tosylate or chloride according to a modified
procedure outlined by Fu¨rstner (Eq. (5)).6 Again, the HCl
salts of 1 and 2 and the free base of 5 and 6 were isolated.
With the required substrates in hand, the cyclization studies
were begun. Previous results in our laboratory indicated that
the pre-catalyst Cpp2SmCH(TMS)2 gave the best diastereo-
selectivity for the five-membered hydroamination reaction.
Further, the pre-catalyst Cpp2NdCH(TMS)2 gave the best
diastereoselectivity for the six-membered hydroamination
reaction. The subtle difference in atomic radii could explain
this result where the smaller metal, Sm, stereoelectronically
favors the formation of the smaller five-membered ring. The
formation of the polycyclic nitrogen heterocycles proceeded
smoothly in all but one case, substrate 10 (Table 1). This is
an anomaly that cannot be readily explained at this point
of time. It is somewhat surprising that the dimethoxy
derivatives did not deactivate the catalytic cycle by
complexation to the metal center, given the oxophilic nature
of the lanthanide catalysts. Substrates 4 and 8 gave the
ð5Þ
Although the dimethoxyvinylbenzyl alcohol (the dimethoxy
analog of 26) could be synthesized, it could not be

G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
10585
Table 1. Sequential intramolecular/cyclization of aromatic amino dienes
highest diastereoselectivities. Upon silica gel chroma-
tography, the ratio of diastereomers could be enhanced to
.50:1. X-ray diffraction analysis of the HCl salt of 14 and
18 show the expected 1,3-cis relationship between the two
stereocenters (Fig. 2).8
ring might act to stiffen the transition state of the second
C–C cyclization, resulting in greater selectivity. The larger
and more flexible ring system of 13 and 17 resulted in
significantly lower diastereoselectivities (,1.5:1).
All of the products are prone to air oxidation. Unfortunately,
products 12 and 16 proved unusually unstable and could not
be isolated. Although the HCl salt of 16 was obtained in
the presence of BHT in 59% yield, it readily oxidized once
the salt was in solution. Based on previous results, the
relationship between the stereocenters is expected to be cis.
Substrates 1 and 5 show higher diastereoselectivity (.4:1)
than their non-aromatic counterpart (,1:1).1 The aromatic
Using the standard Cp2pNdCH(TMS)2 precatalyst, no
reaction appeared to be taking place with substrate 10,
even at 1208C. The root cause of why substrate 10 did not
react to form the product under these conditions is unclear
at this point, in particular because substrate 9 cyclized
Figure 2.

10586
G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
for neutralization. GC conditions: 508C for 5 min, 58C/min
ramp, max. temp.¼2508C; HP Ultra 2, cross-linked 5%
phenylmethyl silicone, 25 m£0.32 mm.
4.1.1. 2-Ethenyl-benzenemethanol (26). Representative
procedure for the Stille coupling. Adapted from a
procedure outlined by Wipf.5 A 100 mL sealable pressure
tube, fitted with an argon line and a magnetic stirrer was
charged with 2.8 g (2.4 mmol) of Pd(PPh3)4, 60 mL of
toluene, 6.00 g (25.6 mmol) of iodide 25, and 8.0 mL (8.6 g,
28 mmol) of tributylvinyltin. The mixture was cooled to
2788C and placed under vacuum for ca. 10 s prior to
refilling with argon. This step was repeated four times and
the mixture was allowed to warm to rt. The pressure tube
was sealed and placed in a 110–1208C oil bath until the
resulting solution turned black (typically 8–24 h). The
pressure tube was cooled to rt. The black slurry was filtered
through Celite and washed with MeOH. The solvents were
removed under vacuum and the red residue was purified two
times by silica gel chromatography using EtOAc/toluene as
eluent to yield 2.74 g (20.4 mmol, 80%) of the title
compound as a pale yellow oil. The spectral features
matched the reported spectral data.11
Figure 3.
smoothly to form 19 in a 2.3:1 diastereomeric ratio. The use
of a catalyst with a more open coordination sphere such as
the ansa bridged Me2SiCp200NdCH(TMS)2 or the dimer
[CpT2MSNdCH3]2 resulted in mainly the hydroaminated
product 34. This might be attributable to a second amine
coordinating to the more accessible metal center,2 prevent-
ing the requisite dative interaction between the metal center
and the alkene. The amine complex would serve to interrupt
the second cyclization by a rapid protonolysis of the
carbon–metal bond (Fig. 3).
4.1.2. 2-Ethenyl-benzeneacetonitrile (28). Preparation of
the tosylate 27 was adapted from a procedure outlined by
Nicolaou.12 A 100 mL round bottom flask equipped with a
nitrogen line, a magnetic stirrer, and an acetone/ice bath,
was charged with 50 mL of THF, 5.7 g (100 mmol) of
powdered KOH, 2.74 g (20.4 mmol) of alcohol 26, and 4.9 g
(26 mmol) of p-toluenesulfonyl chloride. The slurry was
allowed to warm to rt and stirred for ca. 4 h. The slurry was
filtered through Celite and the volume was reduced to ca.
10 mL. The concentrated solution was cooled to rt and
charged with 100 mL of hexanes. The resulting slurry was
solvent exchanged into hexanes while maintaining a bath
temperature less than 358C. The slurry was filtered and
washed with hexanes to yield 5.9 g (20 mmol, 100%) of the
tosylate 27 as a white crystalline solid. The solid was used
immediately in the next step. It may be stored cold and
under argon for only a short period of time.
3. Conclusions
The application of the lanthanide mediated sequential
hydroamination/C–C cyclization has been applied to the
synthesis of polycyclic aromatic nitrogen heterocycles,
including the quinolizine ring system. The benzo[a]quino-
lizine (14 and 18) and the pyrido[2,1,a]isoindolizine (12 and
16) ring systems gave the greatest diastereoselectivities
(.20:1), regardless of the electronic characteristics of the
aromatic ring. In addition, the pyrrolo[2,1-a]isoindole
system was formed with a surprisingly high diastereo-
selectivity (.4:1) compared to its aliphatic counterpart.
Lastly, a novel protocol for substrate preparation involving
the use of the HCl salts of the amines enabled the rapid
synthesis of high purity substrates.
4. Experimental
A 250 mL round bottom flask, equipped with a nitrogen line
and magnetic stirrer was charged with 50 mL of acetonitrile,
2.5 g (38.4 mmol) of KCN and 3.0 g (10.4 mmol) of the
tosylate 27. After allowing the slurry to stir at rt overnight,
50 mL of MTBE was charged. The slurry was filtered and
the solvents were removed under vacuum. The resulting
yellow mixture was separated by silica gel chromatography
using HPLC grade EtOAc/hexanes as eluent. An additional
purification by silica gel chromatography was required to
remove trace amounts of palladium to yield 1.41 g
(9.85 mmol, 95%) of the title compound as a clear colorless
oil: 1H NMR (500 MHz, CDCl3) d 7.53–7.51 (d, J¼7.5 Hz,
1H), 7.41–7.40 (d, J¼7.2 Hz, 1H), 7.36–7.30 (m, 2H),
6.89–6.84 (dd, J¼10.8, 17.2 Hz, 1H), 5.72–5.69 (dd, J¼
0.8, 17.2 Hz, 1H), 5.47–5.45 (dd, J¼0.8, 10.8 Hz, 1H), 3.76
(s, 2H); 13C NMR (125 MHz, CDCl3) d 137.13, 133.21,
129.04, 128.79, 128.60, 127.36, 126.94, 118.45, 117.71,
21.87; IR (CDCl3) 3068.8, 2359.1, 2341.2, 2253.8, 1486.6,
1452.9, 1417.3 cm21; HRMS calcd for C10H10N (MþH;
CIþ): 144.0813, found: 144.0814; LRMS (CIþ) m/z 115
(100), 116 (85), 117 (40), 124 (23).
4.1. General procedures
Et2O, THF and C6D6 were distilled from sodium benzo-
phenone ketyl. C6D6 was degassed prior to use. The
catalysts were prepared according to a literature procedure
and handled in a nitrogen filled glove box.9 The homo-
allylamine was prepared according to a literature pro-
cedure.10 All other reagents were used as received except
where noted.
The substrates for catalysis were prepared by neutralizing
with 10 wt% aqueous NaOH and extracted with C6D6. The
free amine solution was dried over activated molecular
sieves (cooled under high vacuum) and freeze/pump/thawed
prior to transferring to the glove box. A 50 mg sample was
then acidified with 2 M HCl in diethyl ether, and the
solvents were removed to obtain the weight percent of
amine in solution, ca. 3%. The yields of the small scale
reactions were based upon this quantitation. The larger scale
yields were based on the actual amount of the HCl salt used

G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
10587
4.1.3. 1-Bromo-2-(chloromethyl)-4,5-dimethoxy-benzene
(30). Adapted from a procedure outlined by Kita.13
with 34.8 g (1.74 g, 43.5 mmol) of a 5 wt% solution of
NaOH in MeOH. The solution was allowed to warm to rt
and stirred for ca. 1 h. The solution was dried with 5 g of
K2CO3 and the slurry was filtered. The methanol solution
was solvent exchanged into EtOAc with a final target
volume of ca. 150 mL. The clear, colorless EtOAc solution
was cooled to 08C and slowly charged with 4.0 mL (9.1 g,
64.3 mmol) of MeI. The resulting slurry was allowed to
warm to rt over ca. 1 h and stirred for an additional hour.
The crystalline solids were filtered, washed with EtOAc,
and dried under vacuum. The off white solids were charged
into a 250 mL round bottom flask followed by 100 mL of
water and 1.8 g (45 mmol) of NaOH. The solution was
allowed to stir overnight at rt. The aqueous solution was
solvent exchanged into isopropanol with a final target
volume of ca. 100 mL. The resulting slurry was filtered and
washed with cold isopropanol to yield 7.2 g (19 mmol,
44%) of the MeI salt 22 and 3.2 g (16.6 mmol, 38%) of the
free base of 22.
A
250 mL round bottom flask, equipped with a nitrogen line
and a magnetic stirrer was charged with 120 mL of
methylene chloride, 6.96 g (26.5 mmol) of PPh3, 1.83 g
(26.9 mmol) of imidazole and 3.70 g (27.7 mmol) of NCS.
The resulting green/brown, hazy solution was allowed to stir
for ca. 30 min. The mixture was then charged with 5.0 g
(20.2 mmol) of the alcohol 29. The mixture was allowed to
stir at rt for ca. 8 h and charged with 80 g of AlO3 and
80 mL of hexanes. The resulting slurry was stirred for ca.
30 min prior to filtration. The volume was reduced under
vacuum to ca. 100 mL and charged with an additional 20 g
of AlO3. The slurry was filtered and the resulting solution
was solvent exchanged into hexanes with a target volume of
ca. 200 mL. The resulting slurry was cooled to 08C, filtered,
and washed with cold hexanes. The solvent was removed
to yield 3.93 g (14.8 mmol, 73%) of the title compound.
The spectral features matched the reported spectral data.14
The material was used in the next step without further
purification.
A 100 mL sealable pressure tube was charged with 4.88 g
(14 mmol) of the MeI salt 22, 730 mg (18.2 mmol) of
NaOH, 40 mL of absolute EtOH, and 2 mL of DMSO. The
mixture was heated to 608C under nitrogen and sealed.
The temperature was increased to ca. 908C and stirred
overnight. After cooling, the clear solution was charged
into a mixture of 50 mL of water and 50 mL of diethyl ether.
The organic layer was separated and the aqueous layer
was washed with an additional 2£50 mL of diethyl ether.
The combined organic layers were washed with 2£50 mL
of water and 50 mL of saturated brine. The organic
layer was dried over K2CO3, filtered, and the solvents
were removed under vacuum to yield 2.4 g (10.8 mmol,
78%) of the title compound as a clear colorless oil. The
spectral features matched the reported spectral data.16 The
material was used in the next step without further
purification.
4.1.4. 2-Bromo-4,5-dimethoxy-benzeneacetonitrile (31).
A 250 mL round bottom flask equipped with a nitrogen line
and a magnetic stirrer was charged with 3.41 g (13.0 mmol)
of the chloride 30, 63 mL of DMF, and 1.46 g (22.4 mmol)
of KCN. The mixture was allowed to stir at rt for ca. 6 h and
cooled to 08C. The mixture was then charged with 30 mL
of diethyl ether and 30 mL of water. The mixture was
separated and the aqueous layer was washed with an
additional 2£30 mL of diethyl ether. The combined organic
layers were washed with 30 mL of water and 30 mL of
saturated brine. The organic layer was dried over K2CO3
and the solvents were removed under vacuum. The residue
was purified by silica gel chromatography using EtOAc/
hexanes as eluent to yield 2.73 g (10.7 mmol, 82%) of the
title compound. The spectral features matched the reported
spectral data.15
4.1.7. 1-(Chloromethyl)-2-ethenyl-4,5-dimethoxy-benz-
ene (33). Adapted from a procedure outlined by Mariano.7
A 100 mL round bottom flask equipped with a nitrogen line,
magnetic stirrer, and an acetone/Dry Ice bath, was charged
with 2.34 g (10.6 mmol) of amine 23, 10 mL of THF,
and 2.1 g (15 mmol) of K2CO3. After the slurry cooled to
2788C, 1.2 mL (1.5 g, 16 mmol) of methyl chloroformate
was added dropwise with vigorous agitation. The thick,
white paste was allowed to warm to rt and stirred for ca.
10 min. The slurry was cooled to 2208C and charged with
20 mL of water and 30 mL of Et2O. The organic layer
was separated and the aqueous layer was washed with an
additional 2£30 mL of Et2O. The combined organic layer
was cooled to 2208C and washed with 20 mL of water. The
organic layers were separated and dried over K2CO3. The
slurry was filtered and the solvents were removed to yield
2.20 g (10.3 mmol, 98%) of the title compound as a pale
4.1.5. 2-Ethenyl-4,5-dimethoxy-benzeneacetonitrile (32).
Prepared from bromide 31 following the Stille coupling
procedure outlined above. After filtration, the solvents were
removed under vacuum and the red residue was purified
three times by silica gel chromatography using EtOAc/
toluene as eluent to yield 1.34 g (6.60 mmol, 48%) of the
1
title compound as a pale yellow oil: H NMR (500 MHz,
CDCl3) d 6.96 (s, 1H), 6.79 (s, 1H), 6.75–6.70 (dd, J¼11.0,
17.2 Hz, 1H), 5.57–5.54 (d, J¼17.2 Hz, 1H), 5.31–5.28 (d,
J¼11.0 Hz, 1H), 3.85 (s, 3H), 3.84 (s, 3H), 3.64 (s, 2H); 13C
NMR (125 MHz, CDCl3) d 149.16, 149.00, 132.56, 129.24,
119.58, 117.88, 116.05, 111.89, 109.33, 56.01, 55.95, 20.97;
IR (CDCl3) 3090.1, 3010.4, 2964.0, 2255.0, 1607.2, 1514.7,
1465.1, 1419.0, 1343.2, 1295.9, 1269.7, 1221.8, 1202.1,
1183.4 cm21; HRMS calcd for C12H13NO2 (Mþ; CIþ):
203.0946, found: 203.0939; LRMS (CIþ) m/z 177 (67), 183
(67), 197 (90), 203 (100).
1
yellow oil: H NMR (500 MHz, CDCl3) d 7.04–6.98 (dd,
J¼10.8, 17.2 Hz, 1H), 7.03 (s, 1H), 6.82 (s, 1H), 5.66–5.63
(d, J¼17.2 Hz, 1H), 5.35–5.33 (d, J¼10.8 Hz, 1H), 4.64 (s,
2H), 3.91 (s, 3H), 3.90 (s, 3H); 13C NMR (125 MHz,
CDCl3) d 149.74, 149.06, 133.18, 130.28, 127.19, 115.50,
113.14, 109.10, 56.19, 56.13, 44.52; IR (CDCl3) 3089.3,
3010.2, 2964.3, 2254.3, 1605.9, 1514.5, 1465.0, 1420.0,
1344.1, 1270.4, 1220.1, 1102.4 cm21; HRMS calcd for
4.1.6. 2-Ethenyl-4,5-dimethoxy-N,N-dimethyl-benzene-
methanamine (22). A 250 mL round bottom flask,
equipped with a magnetic stirrer and an ice bath, was
charged with 10.0 g (43.5 mmol) of the HCl salt of
6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline and 50 mL
of MeOH. The solution was cooled to 08C and charged

10588
G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
C11H13O2Cl (M; CIþ): 212.0604, found: 212.0598; LRMS
cooled to ca. 2158C and charged with 4.2 mL (4.2 mmol) of
a 1 M solution of DIBALH in toluene while maintaining an
internal temperature less than 08C. The hazy solution was
allowed to warm to rt and stirred for ca. 30 min. The mixture
was then cooled to ca. 2158C and charged with 393 mg
(4.2 mmol) of the HCl salt of allylamine and 341 mL
(260 mg, 4.5 mmol) of allylamine. The resulting slurry
was sealed and allowed to warm to rt. After stirring at rt
overnight, the mixture was cooled to ca. 2158C and charged
with 400 mg (6.4 mmol) of NaCNBH3. The slurry was
sealed and held at 215 to 08C for ca. 6 h prior to warming to
rt. After stirring at rt for ca. 1 h, the slurry was charged
with 50 mL of 10 wt% aqueous NaOH. The mixture was
separated and the aqueous layer was washed with 2£20 mL
of diethyl ether. The combined organic layer was dried over
K2CO3. The slurry was filtered and the resulting solution
was cooled to ca. 2158C. The solution was acidified with
5 mL (10 mmol) of 2 M HCl in diethyl ether. The solvents
were removed under vacuum and the resulting mixture was
separated by silica gel chromatography using isopropanol/
EtOAc as eluent. Crystallization was achieved by solvent
exchanging the combined fractions into heptane. The
white crystalline solids were filtered, providing 310 mg
(1.39 mmol, 40%) of the title compound: mp 114–1208C;
1H NMR (500 MHz, CDCl3) d 10.01 (s, 2H), 7.50–7.49 (d,
J¼7.2 Hz, 1H), 7.26–7.21 (m, 3H), 7.11–7.06 (dd, J¼11.0,
17.2 Hz, 1H), 6.15–6.10 (ddt, J¼6.8, 10.4, 17.4 Hz, 1H),
5.68–5.65 (d, J¼17.2 Hz, 1H), 5.51–5.47 (d, J¼17.4 Hz,
1H), 5.46–5.44 (d, J¼10.4 Hz, 1H), 5.39–5.36 (d, J¼
11.0 Hz, 1H), 3.64–3.62 (d, J¼7.0 Hz, 2H), 3.38–3.35 (m,
2H), 3.10–3.07 (m, 2H); 13C NMR (125 MHz, CDCl3) d
137.25, 134.07, 134.03, 130.23, 128.44, 127.99, 127.89,
126.48, 124.31, 117.32, 49.76, 47.10, 30.13; IR (CDCl3)
2946.1, 2760.2, 2225.9, 1586.0, 1451.2 cm21; HRMS calcd
for C13H18N (M2Cl; ESIþ): 188.1439, found: 188.1445;
LRMS (ESIþ) m/z 117 (19), 131 (37) 188 (100).
(CIþ) m/z 197 (5), 212 (100).
4.1.8. 2-Ethenyl-N-2-propenyl-benzenemethanamine,
hydrochloride (1). Representative procedure for the
preparation of secondary amines from their correspond-
ing tosylate or chloride. Adapted from a procedure
outlined by Fu¨rstner.6 A 10 mL sealable flask, equipped
with a stir bar, was charged with 750 mg (2.60 mmol) of
tosylate 27, 3 mL of THF, and 500 mL (6.7 mmol) of
allylamine. The flask was sealed, heated to 60–708C, and
allowed to stir at that temperature overnight. The yellow
solution was charged into 10 mL of Et2O and allowed to stir
for ca. 30 min. The resulting slurry was filtered [NMR
indicates the solids are the tosylate salt of allylamine
(0.99 equiv.)], cooled to 2208C, and acidified with 3.4 mL
of 2 M HCl in diethyl ether. The solvents were removed
under vacuum and the resulting mixture was separated
by silica gel chromatography using isopropanol/EtOAc as
eluent. Crystallization was achieved by solvent exchanging
the combined fractions into heptane. The white crystalline
solids were filtered, providing 460 mg (2.19 mmol, 84%) of
the title compound. Recrystallization can be achieved using
EtOAc as solvent: mp 86–888C; 1H NMR (500 MHz,
CDCl3) d 10.03 (s, 2H), 7.72–7.71 (m, 1H), 7.48–7.46
(m, 1H), 7.34–7.32 (m, 2H), 7.06–7.01 (dd, J¼10.8,
17.1 Hz, 1H), 6.09–6.03 (ddt, J¼6.8, 10.4, 17.2 Hz, 1H),
5.66–5.62 (d, J¼17.1 Hz, 1H), 5.47–5.44 (d, J¼10.8 Hz,
1H), 5.42–5.40 (d, J¼10.4 Hz, 1H), 5.41–5.38 (d, J¼
17.2 Hz, 1H), 4.11 (s, 2H), 3.42–3.41 (d, J¼6.8 Hz, 2H);
13C NMR (125 MHz, CDCl3) d 138.94, 133.88, 131.34,
129.90, 128.71, 128.10, 127.65, 127.34, 124.19, 119.26,
48.21, 45.86; IR (CDCl3) 2935.7, 2744.7, 2225.3, 1583.0,
1450.8, 1216.2 cm21; HRMS calcd for C12H16N (M2Cl;
ESIþ): 174.1283, found: 174.1286; LRMS (ESIþ) m/z 115
(33), 174 (100).
4.1.9. 2-Ethenyl-N-3-butenyl-benzenemethanamine,
hydrochloride (2). Prepared from tosylate 27 and homo-
allylamine as outlined for the preparation of secondary
amines from the corresponding tosylate (1.97 mmol, 76%):
mp 102–1068C; 1H NMR (500 MHz, CDCl3) d 9.92 (s, 2H),
7.76–7.74 (m, 1H), 7.47–7.46 (m, 1H), 7.35–7.33 (m, 2H),
7.08–7.05 (dd, J¼11.0, 17.1 Hz, 1H), 5.67–5.63 (ddt, J¼
6.6, 10.5, 16.9 Hz, 1H), 5.67–5.63 (dd, J¼1.0, 17.1 Hz,
1H), 5.48–5.46 (dd, J¼1.0, 11.0 Hz, 1H), 5.12–5.08 (dd,
J¼1.5, 16.9 Hz, 1H), 5.10–5.08 (dd, J¼1.5, 10.5 Hz, 1H),
4.17 (s, 2H), 2.84–2.81 (m, 2H), 2.61–2.57 (dt, J¼7.4,
7.9 Hz, 2H); 13C NMR (125 MHz, CDCl3) d 139.06,
133.95, 132.80, 131.50, 130.01, 128.80, 127.58, 127.34,
119.38, 118.65, 47.05, 45.49, 30.20; IR (CDCl3) 2950.6,
2750.9, 2224.9, 1584.0, 1461.7 cm21; HRMS calcd for
C13H18N (M2Cl; ESIþ): 188.1439, found: 188.1441;
LRMS (ESIþ) m/z 115 (55), 117 (63) 188 (100).
4.1.11. 2-Ethenyl-N-3-butenyl-benzeneethanamine,
hydrochloride (4). Prepared from 28 and both the HCl
salt and free base of homoallylamine as outlined for the
preparation of secondary amines from their corresponding
nitriles (1.35 mmol, 39%): mp 146–1488C; 1H NMR
(500 MHz, CDCl3) d 9.94 (s, 2H), 7.49–7.48 (d, J¼
7.3 Hz, 1H), 7.24–7.20 (m, 3H), 7.09–7.07 (dd, J¼11.0,
17.2 Hz, 1H), 5.82–5.73 (ddt, J¼6.6, 10.3, 17.0 Hz, 1H),
5.67–5.64 (d, J¼17.2 Hz, 1H), 5.37–5.35 (d, J¼11.0 Hz,
1H), 5.18–5.14 (dd, J¼1.3, 17.0 Hz, 1H), 5.14–5.11 (dd,
J¼1.3, 10.3 Hz, 1H), 3.41–3.38 (m, 2H), 3.13–3.12 (m,
2H), 3.05 (m, 2H), 2.73–2.69 (dt, J¼6.6, 7.9 Hz, 2H); 13C
NMR (125 MHz, CDCl3) d 137.21, 133.99, 133.98, 132.60,
130.16, 128.43, 127.87, 126.46, 118.84, 117.32, 48.43, 47.14,
30.27, 30.04; IR (CDCl3) 2948.6, 2764.0, 2449.5, 2223.1,
1643.8, 1626.4, 1587.9, 1485.5, 1467.4 cm21; HRMS calcd
for C14H20N (M2Cl; ESIþ): 202.1596, found: 202.1604;
LRMS (ESIþ) m/z 129 (12), 160 (8) 202 (100).
4.1.10. 2-Ethenyl-N-2-propenyl-benzeneethanamine,
hydrochloride (3). Representative procedure for the
formation of secondary amines from their correspond-
ing nitriles. Adapted from Brusse’s procedure.4 A 100 mL
round bottom flask fitted with a nitrogen line, magnetic
stirrer and an acetone/ice bath, was charged with 1.33 g
(500 mg, 3.49 mmol) of a 37.6 wt% solution of nitrile 28 in
diethyl ether and 30 mL of diethyl ether. The solution was
4.1.12. 2-Ethenyl-4,5-dimethoxy-N-2-propenyl-benzene-
methanamine (5). Prepared from chloride 33 and allyl-
amine as outlined for the preparation of secondary amines
from the corresponding chloride. After the reaction was
complete, the acidic mixture was neutralized with 5 mL of
10 wt% aqueous NaOH and extracted with 3£5 mL of
diethyl ether. The combined organic layers were dried over

G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
10589
K2CO3 and the slurry was filtered to obtain a clear colorless
solution. The solvents were removed under vacuum and the
resulting residue was separated by silica gel chroma-
tography using isopropanol/EtOAc as eluent to provide
the title compound as a clear colorless oil (1.80 mmol,
and free base of homoallylamine as outlined for the
preparation of secondary amines from their corresponding
nitriles (1.25 mmol, 51%): H NMR (500 MHz, CDCl3) d
1
7.03 (s, 1H), 6.98–6.92 (dd, J¼10.8, 17.2 Hz, 1H), 6.68 (s,
1H), 5.79–5.75 (ddt, J¼6.8, 10.3, 17.2 Hz, 1H), 5.57–5.53
(dd, J¼1.1, 17.2 Hz, 1H), 5.23–5.20 (dd, J¼1.1, 10.8 Hz,
1H), 5.08–5.04 (dq, J¼1.6, 17.2 Hz, 1H), 5.03–5.00 (d,
J¼10.3 Hz, 1H), 3.90 (s, 3H), 3.87 (s, 3H), 2.83–2.80 (m,
4H), 2.72–2.69 (t, J¼6.1 Hz, 2H), 2.27–2.24 (q, J¼6.8 Hz,
2H), 1.15 (s, 1H); 13C NMR (125 MHz, CDCl3) d 149.09,
147.92, 136.61, 134.24, 130.54, 129.16, 116.49, 113.66,
113.23, 108.97, 56.18, 56.16, 51.12, 49.06, 34.57, 33.68; IR
(CDCl3) 3082.4, 3005.5, 2936.8, 2832.1, 2257.2, 2183.3,
1639.8, 1605.5, 1573.9, 1511.6, 1464.4, 1418.8, 1339.0,
1308.8, 1264.8, 1217.3, 1197.5, 1101.9 cm21; HRMS calcd
for C16H23NO2Na (MþNa; ESIþ): 284.1626, found:
284.1628; LRMS (ESIþ) m/z 160 (30), 179 (28), 191 (40),
262 (100).
1
75%): H NMR (500 MHz, CDCl3) d 7.04 (s, 1H), 7.01–
6.95 (dd, J¼11.0, 17.4 Hz, 1H), 6.85 (s, 1H), 5.95–5.91
(ddt, J¼5.9, 10.6, 17.0 Hz, 1H), 5.60–5.56 (dd, J¼1.1,
17.4 Hz, 1H), 5.25–5.23 (dd, J¼1.1, 11.0 Hz, 1H), 5.22–
5.19 (dd, J¼1.3, 17.0 Hz, 1H), 5.13–5.11 (dd, J¼1.3,
10.6 Hz, 1H), 3.90 (s, 3H), 3.89 (s, 3H), 3.78 (s, 2H), 3.30–
3.29 (d, J¼5.9 Hz, 2H), 1.22 (s, 1H); 13C NMR (125 MHz,
CDCl3) d 149.01, 148.40, 137.12, 133.91, 130.54, 129.37,
116.19, 114.07, 112.81, 108.96, 56.16, 56.14, 52.25, 50.58;
IR (CDCl3) 3084.4, 3007.6, 2960.9, 2937.1, 2912.0, 2833.6,
2256.6, 1605.3, 1511.8, 1464.6, 1418.4, 1338.5, 1309.3,
1266.4, 1219.7, 1187.9, 1106.8 cm21; HRMS calcd for
C14H20NO2 (MþH; ESIþ): 234.1494, found: 234.1495;
LRMS (ESIþ) m/z 131 (17), 146 (30), 177 (100), 234 (23).
4.1.16. 2-Ethenyl-1-(phenylmethyl)-N-2-propenyl-1H-
indole-3-ethanamine (9). Prepared from 21 and both the
HCl salt and free base of allylamine as outlined for the
preparation of secondary amines from their corresponding
4.1.13. 2-Ethenyl-4,5-dimethoxy-N-3-butenyl-benzene-
methanamine (6). Prepared from chloride 33 and homo-
allylamine as outlined for the preparation of secondary
amines from the corresponding chloride. The free base was
isolated according to the procedure outlined above
(0.93 mmol, 47%): 1H NMR (500 MHz, CDCl3) d 7.03
(s, 1H), 7.00–6.94 (dd, J¼11.0, 17.4 Hz, 1H), 6.85 (s, 1H),
5.83–5.77 (ddt, J¼6.8, 10.3, 17.0 Hz, 1H), 5.59–5.56 (d,
J¼17.4 Hz, 1H), 5.25–5.23 (d, J¼11.0 Hz, 1H), 5.11–5.07
(dd, J¼1.3, 17.0 Hz, 1H), 5.04–5.02 (dd, J¼1.3, 10.3 Hz,
1H), 3.90 (s, 3H), 3.89 (s, 3H), 3.78 (s, 2H), 2.74–2.72
(t, J¼7.0 Hz, 2H), 2.30–2.26 (dt, J¼6.8, 7.0 Hz, 2H), 1.22
(s, 1H); 13C NMR (125 MHz, CDCl3) d 148.93, 148.29,
136.67, 133.87, 130.60, 129.25, 116.46, 114.11, 112.62,
108.80, 56.12, 51.19, 48.80, 34.43; IR (CDCl3) 3081.9,
3005.4, 2936.6, 2911.6, 2833.3, 2256.5, 1639.9, 1605.3,
1574.3, 1511.5, 1464.3, 1418.4, 1338.6, 1266.4, 1216.0,
1187.1, 1109.3 cm21; HRMS calcd for C15H21NO2Na
(MþNa; ESIþ): 270.1470, found: 270.1467; LRMS
(ESIþ) m/z 146 (43), 177 (100), 248 (43).
1
nitriles (0.49 mmol, 33%): H NMR (500 MHz, CDCl3) d
7.71–7.69 (d, J¼7.7 Hz, 1H), 7.30–7.15 (m, 6H), 7.07–
7.05 (d, J¼7.7 Hz, 2H), 6.74–6.72 (dd, J¼11.9, 18.0 Hz,
1H), 5.98–5.90 (ddt, J¼5.9, 10.3, 17.2 Hz, 1H), 5.60–5.57
(dd, J¼1.1, 18.0 Hz, 1H), 5.47–5.44 (dd, J¼1.1, 11.9 Hz,
1H), 5.41 (s, 2H), 5.20–5.16 (dq, J¼1.6, 17.2 Hz, 1H),
5.09–5.07 (dq, J¼1.3, 10.3 Hz, 1H), 3.34–3.32 (d, J¼
5.9 Hz, 2H), 3.16–3.13 (t, J¼7.2 Hz, 2H), 3.02–2.99 (t,
J¼7.3 Hz, 2H), 1.20 (s, 1H); 13C NMR (125 MHz, CDCl3) d
138.24, 137.61, 137.21, 135.06, 128.93, 128.35, 127.40,
126.15, 126.03, 122.66, 119.75, 119.30, 118.76, 115.72,
112.89, 109.79, 52.60, 50.27, 47.56, 25.85; IR (CDCl3)
3063.2, 3030.2, 2923.9, 2822.1, 2245.1, 1626.3, 1605.7,
1495.9, 1464.6, 1453.4, 1354.9, 1336.6, 1300.8, 1179.5,
1108.9, 1028.4 cm21; HRMS calcd for C22H25N2 (MþH;
ESIþ): 317.2018, found: 317.2012; LRMS (ESIþ) m/z 317
(100).
4.1.14. 2-Ethenyl-4,5-dimethoxy-N-2-propenyl-benzene-
ethanamine (7). Prepared from 32 and both the HCl salt and
free base of allylamine as outlined for the preparation of
secondary amines from their corresponding nitriles
4.1.17. 2-Ethenyl-1-(phenylmethyl)-N-3-butenyl-1H-
indole-3-ethanamine (10). Prepared from 21 and both the
HCl salt and free base of homoallylamine as outlined for the
preparation of secondary amines from their corresponding
1
1
(0.81 mmol, 41%): H NMR (500 MHz, CDCl3) d 7.03 (s,
nitriles (0.73 mmol, 43%): H NMR (500 MHz, CDCl3) d
1H), 6.98–6.93 (dd, J¼11.0, 17.4 Hz, 1H), 6.68 (s, 1H),
5.90–5.89 (ddt, J¼6.1, 10.3, 17.2 Hz, 1H), 5.58–5.54 (dd,
J¼1.3, 17.4 Hz, 1H), 5.23–5.21 (dd, J¼1.3, 11.0 Hz, 1H),
5.17–5.14 (dq, J¼1.6, 17.2 Hz, 1H), 5.09–5.07 (dq, J¼1.3,
10.3 Hz, 1H), 3.90 (s, 3H), 3.88 (s, 3H), 3.28–3.26 (dt, J¼
1.3, 6.1 Hz, 2H), 2.85–2.81 (m, 4H), 1.20 (s, 1H); 13C NMR
(125 MHz, CDCl3) d 149.00, 147.85, 136.99, 134.17,
130.38, 129.08, 115.99, 113.67, 113.12, 108.77, 56.13,
52.59, 50.63, 33.62; IR (CDCl3) 3085.0, 3008.0, 2937.5,
2832.6, 2256.5, 1605.5, 1511.6, 1464.5, 1418.7, 1339.7,
7.67–7.66 (d, J¼7.5 Hz, 1H), 7.29–7.13 (m, 6H), 7.05–
7.04 (d, J¼7.0 Hz, 2H), 6.75–6.72 (dd, J¼11.7, 17.6 Hz,
1H), 5.82–5.75 (ddt, J¼7.0, 10.3, 17.0 Hz, 1H), 5.57–5.53
(dd, J¼1.1, 17.6 Hz, 1H), 5.44–5.42 (dd, J¼1.1, 11.7 Hz,
1H), 5.40 (s, 2H), 5.06–5.02 (dq, J¼1.5, 17.0 Hz, 1H),
5.00–4.98 (dd, J¼0.9, 10.3 Hz, 1H), 3.12–3.09 (t, J¼
7.3 Hz, 2H), 2.99–2.96 (t, J¼7.5 Hz, 2H), 2.76–2.73 (t,
J¼6.8 Hz, 2H), 2.28–2.24 (dt, J¼7.0, 13.8 Hz, 2H), 1.20 (s,
1H); 13C NMR (125 MHz, CDCl3) d 138.26, 137.58,
136.64, 135.02, 128.96, 128.30, 127.43, 126.16, 126.00,
122.66, 119.73, 119.29, 118.75, 116.49, 112.90, 109.79,
50.72, 49.10, 47.56, 34.62, 25.84; IR (CDCl3) 3062.8,
3030.2, 2929.6, 2827.9, 2245.0, 1626.4, 1605.7,
1495.9, 1464.7, 1453.4, 1355.1, 1300.8, 1179.5, 1117.3,
1028.5 cm21; HRMS calcd for C23H27N2 (MþH; ESIþ):
331.2174, found: 331.2184; LRMS (ESIþ) m/z 248 (47),
260 (100), 331 (90).
1309.5, 1264.7, 1217.8, 1197.5, 1183.4, 1102.2 cm21
;
HRMS calcd for C15H22NO2 (MþH; ESIþ): 248.1650,
found: 248.1647; LRMS (ESIþ) m/z 145 (14), 160 (35), 191
(47), 248 (100).
4.1.15. 2-Ethenyl-4,5-dimethoxy-N-3-butenyl-benzene-
ethanamine (8). Prepared from 32 and both the HCl salt

10590
G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
4.1.18. 2-Methyl-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a]-
isoindole (11). Representative procedure for the sequen-
tial hydroamination/C–C cyclization reaction. A seal-
able NMR tube was charged with 551 mg (5.51 mg,
9.53 mmol) of a 1 wt% solution of Cpp2SmCH(TMS)2 in
C6D6. The catalyst solution was then charged with
an additional 400 mg of C6D6 and 730 mg (16.5 mg,
95.2 mmol) of a 2.26 wt% solution of the free amine of 1
in C6D6. The NMR tube was sealed and allowed to sit at rt.
Upon completion of the reaction completion, ca. 22 h, the
green solution was charged with 3 mL of heptane and
allowed to oxidize for ca. 2 h. The yellow slurry was filtered
through Celite and the clear, pale yellow solution was
analyzed by GC (5.5:1 dr). The solvents were removed
under vacuum and the resulting residue was separated by
silica gel chromatography using EtOAc/hexanes and
isopropanol/EtOAc as eluent to yield 13.1 mg (75.6 mmol,
79%) of the title compound as a 5.5:1 mixture of dia-
stereomers: 13C NMR (125 MHz, CDCl3) d 145.31,
144.70, 140.93, 139.99, 127.46, 127.37, 127.19, 127.08,
123.18, 122.92, 122.60, 122.54, 70.76, 70.46, 64.44, 63.98,
61.78, 59.79, 41.00,. 40.74, 35.60, 32.84, 18.32, 17.51; IR
(CDCl3) 2958.6, 2929.6, 2872.3, 2359.7, 2251.8, 2171.5,
4.1.21. 7,8-Dimethoxy-2-methyl-2,3,5,9b-tetrahydro-1H-
pyrrolo[2,1-a]isoindole (15). The general procedure for the
sequential hydroamination/C–C cyclization reaction was
followed with the noted exceptions (54 mmol, 84%): 13C
NMR (125 MHz, CDCl3) d 149.25, 149.21, 149.07, 148.97,
136.97, 136.25, 132.23, 131.22, 106.24, 105.96, 105.73,
105.69, 70.94, 70.64, 64.56, 64.14, 62.02, 59.97, 56.35,
56.32, 41.02, 40.85, 35.69, 32.78, 18.20, 17.50; IR (CDCl3)
2958.4, 2871.3, 2835.2, 2253.7, 2167.4, 1504.5, 1465.4,
1378.4, 1339.9, 1291.4, 1279.6, 1250.1, 1217.7, 1190.1,
1175.6, 1104.5, 1012.4 cm21; HRMS calcd for C14H20NO2
(MþH; ESIþ): 234.1494, found: 234.1504; LRMS (ESIþ)
m/z 202 (19), 234 (100).
4.1.22. 8,9-Dimethoxy-1,2,3,5,6,10b-hexahydro-2-
methyl-pyrrolo[2,1-a]isoquinoline (17). The general pro-
cedure for the sequential hydroamination/C–C cyclization
reaction was followed with the noted exceptions
(49.7 mmol, 82%): 13C NMR (125 MHz, CDCl3) d
147.64, 147.51, 147.48, 147.47, 131.94, 131.36, 126.64,
126.61, 111.80, 111.77, 109.28, 109.07, 62.90, 62.68, 62.22,
60.61, 56.25, 56.15, 48.82, 48.51, 41.10, 39.46, 31.63,
30.50, 28.36, 27.21, 20.80, 20.70; IR (CDCl3) 2957.8,
2936.4, 2869.5, 2835.8, 2798.3, 2253.8, 2181.7, 1610.4,
1511.2, 1464.9, 1374.1, 1359.3, 1323.1, 1262.3, 1250.2,
1477.4, 1456.1, 1378.4, 1340.8, 1091.5, 1042.9 cm21
;
HRMS calcd for C12H14N (M2H; CIþ): 172.1126,
found: 172.1129; LRMS (ESIþ) m/z 118 (25), 145 (18),
174 (100).
1232.7, 1214.5, 1163.0, 1140.7, 1100.3, 1014.1 cm21
;
HRMS calcd for C15H22NO2 (MþH; ESIþ): 248.1650,
found: 248.1649; LRMS (ESIþ) m/z 248 (100).
4.1.19. 1,2,3,5,6,10b-Hexahydro-2-methyl-pyrrolo[2,1-a]-
isoquinoline (13). The general procedure for the sequential
hydroamination/C–C cyclization reaction was followed
with the noted exceptions (68 mmol, 76%): 13C NMR
(125 MHz, CDCl3) d 139.55, 139.32, 134.60, 134.54,
128.69, 128.66, 126.17, 126.03, 126.00, 125.91, 125.85,
125.61, 63.42, 63.12, 62.49., 60.93, 48.98, 48.71, 40.83,
39.18, 31.56, 30.49, 28.91, 27.94, 20.92, 20.71; IR (CDCl3)
2957.4, 2929.1, 2870.4, 2793.4, 2738.5, 2248.8, 2182.3,
1492.9, 1473.0, 1452.8, 1433.0, 1374.5, 1354.1, 1310.3,
1282.3, 1248.7, 1217.8, 1161.9, 1135.2, 1116.3,
1040.0 cm21; HRMS calcd for C13H18N (MþH; ESIþ):
188.1439, found: 188.1448; LRMS (ESIþ) m/z 132 (15),
188 (100).
4.1.23. 9,10-Dimethoxy-1,3,4,6,7,11b-hexahydro-2-
methyl-2H-benzo[a]quinolizine (18). The general pro-
cedure for the sequential hydroamination/C–C cyclization
reaction was followed with the noted exceptions (264 mmol,
69%). The HCl salt of the free amine was generated as
outlined above. Recrystallization using acetone provided
1
suitable crystals for X-ray analysis: H NMR (500 MHz,
CDCl3) d 6.70 (s, 1H), 6.57 (s, 1H), 3.85 (s, 3H), 3.83 (s,
3H), 3.14–3.02 (m, 2H), 2.99–2.95 (m, 2H), 2.67–2.58 (dd,
J¼3.1, 15.3 Hz, 1H), 2.49–2.47 (dt, J¼4.0, 11.6 Hz, 1H),
2.36–2.29 (dt, J¼2.2, 12.1 Hz, 1H), 2.23–2.19 (dq, J¼2.8,
12.7 Hz, 1H), 1.70–1.58 (m, 2H), 1.42–1.36 (dq, J¼4.1,
12.1 Hz, 1H), 1.19–1.07 (dt, J¼11.4, 12.7 Hz, 1H), 1.02–
1.01 (d, J¼6.4 Hz, 3H); 13C NMR (125 MHz, CDCl3) d
147.58, 147.35, 130.68, 126.96, 111.77, 108.47, 62.88,
56.83, 56.22, 56.02, 52.70, 40.28, 34.09, 31.94, 29.47,
22.42; IR (CDCl3) 3004.2, 2950.2, 2836.2, 2808.2, 2751.9,
2254.0, 2182.4, 1611.6, 1510.5, 1464.9, 1409.9, 1388.6,
1358.9, 1327.1, 1303.8, 1287.5, 1264.0, 1249.1, 1228.1,
1209.8, 1192.0, 1170.5, 1135.8, 1121.4, 1106.6, 1088.7,
1026.7, 1001.6 cm21; HRMS calcd for C16H22NO2 (M2H;
CIþ): 260.1650, found: 260.1638; LRMS (ESIþ) m/z 165
(52), 262 (100).
4.1.20. 1,3,4,6,7,11b-Hexahydro-2-methyl-2H-benzo[a]-
quinolizine (14). The general procedure for the sequential
hydroamination/C–C cyclization reaction was followed
with the noted exceptions (323 mmol, 73%). The HCl salt of
the free amine was generated as outlined above. Recrys-
tallization using acetone provided suitable crystals for X-ray
analysis: 1H NMR (500 MHz, CDCl3) d 7.24–7.09 (m, 4H),
3.21–3.13 (m, 2H), 3.02–2.99 (m, 2H), 2.76–2.70 (dd,
J¼3.1, 16.1 Hz, 1H), 2.54–2.52 (dt, J¼4.0, 11.6 Hz, 1H),
2.36–2.33 (m, 2H), 1.73–1.62 (m, 2H), 1.48–1.36 (dq,
J¼4.2, 12.3 Hz, 1H), 1.20–1.14 (dt, J¼11.4, 12.8 Hz, 1H),
1.04–1.03 (d, J¼6.4 Hz, 3H); 13C NMR (125 MHz, CDCl3)
d 138.69, 134.75, 129.05, 126.07, 125.82, 124.92, 63.25,
56.94, 52.58, 40.15, 34.14, 31.98, 29.93, 22.47; IR
(CDCl3) 3020.5, 2947.4, 2868.9, 2800.2, 2746.4, 2285.7,
2183.8, 1492.9, 1453.1, 1390.0, 1356.1, 1307.4, 1294.3,
1250.0, 1220.5, 1198.6, 1168.1, 1122.7, 1108.3, 1090.0,
1043.5 cm21; HRMS calcd for C14H19N (M; CIþ):
201.1518, found: 201.1514; LRMS (ESIþ) m/z 117 (10),
202 (100).
4.1.24. 2,3,5,6,11,11b-Hexahydro-2-methyl-11-(phenyl-
methyl)-1H-indolizino[8,7-b]indole (19). The general
procedure for the sequential hydroamination/C–C cycli-
zation reaction was followed with the noted exceptions
(39 mmol, 83%): 13C NMR (125 MHz, CDCl3) d 138.32,
138.13, 138.04, 137.16, 128.91, 127.46, 127.44, 127.37,
126.26, 126.13, 121.42, 119.43, 118.36, 118.34, 109.69,
108.41, 107.47, 60.14, 57.88, 57.31, 56.15, 47.48, 47.45,
46.60, 40.46, 38.87, 32.90, 31.37, 20.38, 20.20, 18.84,
17.94; IR (CDCl3) 2957.8, 2928.1, 2871.0, 2251.9, 2181.2,

G. A. Molander, S. K. Pack / Tetrahedron 59 (2003) 10581–10591
10591
4. Zandbergen, P.; van den Nieuwendijk, A.; Brussee, J.; van der
Gen, A.; Kruse, C. C. Tetrahedron 1992, 48, 3977.
5. Wipf, P.; Weiner, W. S. J. Org. Chem. 1999, 64, 5321.
6. Fu¨rstner, A.; Langemann, K. Synthesis 1997, 7, 792.
7. Dai-Ho, G.; Mariano, P. S. J. Org. Chem. 1988, 53, 5113.
8. Molander, G. A.; Dowdy, E. D.; Pack, S. K. J. Org. Chem.
2001, 66, 4344.
1495.8, 1465.8, 1453.8, 1376.0, 1352.1, 1328.4, 1309.4,
1179.3, 1134.0, 1095.3, 1028.6, 1016.6 cm21; HRMS calcd
for C22H25N2 (MþH; ESIþ): 317.2018, found: 317.2003;
LRMS (ESIþ) m/z 222 (28), 234 (20), 317 (60), 333 (100).
Acknowledgements
9. Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.;
Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107,
8091.
We would like to thank Dr Pat Carroll for the X-ray
structure determination of 14 and 18. We gratefully
acknowledge the National Institutes of Health (Grant
GM48580) for their generous support of this research.
S. K. P. thanks the Process R&D group of Bristol–Myers
Squibb for their generous financial support and study leave.
10. Sato, T.; Nakamura, N.; Ikeda, K.; Okada, M.; Ishibashi, H.;
Ikeda, M. J. Chem. Soc., Perkin Trans. 1 1992, 18, 2399.
11. Rao, A. V. R.; Rao, B. V.; Reddy, D. R. Indian J. Chem. 1991,
30, 723.
12. Marron, B. E.; Spanevello, R. A.; Elisseou, M. E.; Serhan,
C. N.; Nicolaou, K. C. J. Org. Chem. 1989, 54, 5522.
13. Anilkumar, G.; Nambu, H.; Kita, Y. Org. Process Res. Dev.
2002, 6, 190.
References
´
14. Olivera, R.; San Martin, R.; Domınguez, E.; Solans, X.;
1. Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1757.
´
2. Gagne, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc.
1992, 114, 275.
Urtiaga, M. K.; Arriortua, M. I. J. Org. Chem. 2000, 65, 6398.
15. Kyong, D. H.; Sladkov, V. I.; Suvorov, N. N. J. Org. Chem.
USSR (Engl. Transl.) 1987, 23, 1732.
´ ´
3. Sapi, J.; Grebille, Y.; Laronze, J.; Levy, J. Synthesis 1992, 4,
383.
16. Sato, Y.; Shirai, N.; Machida, Y.; Ito, E.; Yasui, T. J. Org.
Chem. 1992, 57, 6711.