Catalytic asymmetric syntheses of quinolizidines by dirhodium-catalyzed dearomatization of isoquinolinium/pyridinium methylides-the role of catalyst and carbene source

Article abstract of DOI:10.1021/ja406482q

Convenient access to highly enantioenriched substituted quinolizidines has been achieved by chiral dirhodium(II) carboxylate-catalyzed dearomatizing formal [3 + 3]-cycloaddition of isoquinolinium/pyridinium methylides and enol diazoacetates. Coordination of Lewis basic methylides to dirhodium(II) prompts the rearrangement of the enol-carbene that is bound to dirhodium to produce a donor-acceptor cyclopropene. The donor-acceptor cyclopropene is in equilibrium with the dirhodium-bound enol-carbene and undergoes both enantioselective [3 + 3]-cycloaddition from the dirhodium-bound enol-carbene and diastereoselective [3 + 2]-cycloaddition by uncatalyzed reaction of the cyclopropene with isoquinolinium or pyridinium methylides. Increasing the mol % of catalyst loading suppresses the [3 + 2]-cycloaddition pathway.

Full text of DOI:10.1021/ja406482q

Article
pubs.acs.org/JACS
Catalytic Asymmetric Syntheses of Quinolizidines by Dirhodium-
Catalyzed Dearomatization of Isoquinolinium/Pyridinium
Methylides−The Role of Catalyst and Carbene Source
Xichen Xu, Peter Y. Zavalij, and Michael P. Doyle*
Department of Chemistry and Biochemistry, University of Maryland, College Park, 20742, United States
S
* Supporting Information
ABSTRACT: Convenient access to highly enantioenriched
substituted quinolizidines has been achieved by chiral
dirhodium(II) carboxylate-catalyzed dearomatizing formal [3
+ 3]-cycloaddition of isoquinolinium/pyridinium methylides
and enol diazoacetates. Coordination of Lewis basic
methylides to dirhodium(II) prompts the rearrangement of
the enol-carbene that is bound to dirhodium to produce a
donor−acceptor cyclopropene. The donor−acceptor cyclo-
propene is in equilibrium with the dirhodium-bound enol-
carbene and undergoes both enantioselective [3 + 3]-cycloaddition from the dirhodium-bound enol-carbene and
diastereoselective [3 + 2]-cycloaddition by uncatalyzed reaction of the cyclopropene with isoquinolinium or pyridinium
methylides. Increasing the mol % of catalyst loading suppresses the [3 + 2]-cycloaddition pathway.
Scheme 1. Reaction Pathways to Enantioselective Synthesis
of Substituted Quinolizidines
INTRODUCTION
■
Quinolizidines are isolated from myriad sources in nature, and
they display diverse biological activities.1 Among nitrogen-
containing heterocyclic compounds, substituted quinolizidine
alkaloids are exceptionally prominent,1a and some of them are
lead compounds for the development of anticancer, anti-
inflammatory, and cardiovascular drugs.1b,c,d Despite long-
standing biological and synthetic interest in quinolizidines,
methodologies for the synthesis of these valuable compounds
have been limited. Asymmetric approaches to these systems
have relied on either using reactants from the chiral pool,1c,d
introducing chirality through the use of chiral auxiliaries,2 or
catalytic enantioselective approaches.3−5 However, catalytic
methods have been limited, and rhodium-catalyzed asymmetric
[2 + 2 + 2]-cycloadditions of isocyanates,3 catalytic asymmetric
formal aza-hetero-Diels−Alder reactions4 and one report of an
organocatalytic enantioselective dearomatization of N-alkyl
isoquinolinium salts (Scheme 1)5 constitute the only current
examples.
enol diazoacetates can be visualized as chiral metallo-1,3-dipole
equivalents with their vinylogous position electrophilic and
their metal carbene center nucleophilic, and this hypothesis has
been recently verified in the highly enantioselective formal [3 +
3]-cycloaddition reaction of enol diazoacetates with aryl
nitrones catalyzed by chiral dirhodium carboxylates.9b Intrigued
by the potential uses of abundant and easily accessible nitrogen-
containing heterocyclic rings as dipole acceptors, we reasoned
that the formal [3 + 3]-cycloaddition strategy could be
extended to stable and readily available isoquinolinium/
pyridinium methylides with asymmetry introduced in the
ring-closing dearomatization stage (Scheme 2).12 Successful
development of the asymmetric variant of this cycloaddition
Although aromatic frameworks are capable of participating in
reactions as electrophiles6 or nucleophiles,7 the development of
catalytic asymmetric transformations directly engaging the
aromatic π system has been achieved only recently.8 We have
been intrigued by asymmetric transformations involving
catalytically generated metal carbene intermediates9 that react
with aromatic and heteroaromatic rings and furnish products
ranging from electrophilic aromatic substitution (Friedel−
Crafts reaction),7 cyclopropanation and the subsequent Cope
rearrangement (Buchner reaction),10 and stable ylide forming
̈
reactions.11 In these transformations, the electrophilic nature of
metal carbenes dominates. We hypothesized that the catalyti-
cally generated chiral dirhodium carbene intermediates from
Received: June 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 2. Formal Enantioselective [3 + 3]-Cycloaddition of
Metallo-1,3-dipoles with Isoquinolinium/Pyridinium
Methylides
Table 1. Optimization of Reaction Conditions for
Enantioselective Formal [3 + 3]-Cycloaddition of
Isoquinolinium Dicyanomethylide 1a and Enol Diazoacetate
2a
reaction would offer direct access to enantioenriched highly
substituted quinolizidines that are amendable to further
functionalization.13 However, would the coordination of these
Lewis basic methylides with the reaction catalyst inhibit the [3
+ 3]-cycloaddition pathway and induce an alternative trans-
formation?
a
b,
c
d
e
entry
catalyst (y mol %)
3a:4a
yield (%) ee (%) of 3a
Investigations of these systems not only revealed a highly
enantioselective dearomatizing formal [3 + 3]-cycloaddition
transformation of isoquinolinium/pyridinium methylides with
enol diazacetates catalyzed by chiral dirhodium carboxylates in
up to 96% ee, but they also provided evidence for competitive
coordination-induced displacement of the dirhodium-bound
enol-carbene as a donor−acceptor cyclopropene that either
reforms the enol-carbene of dirhodium or undergos [3 + 2]-
cycloaddition to form densely functionalized indolizidines with
complete regioselectivity and diastereoselectivity.
1
2
3
4
5
6
Rh2(Oct)4 (1.0 mol %)
2.0:1
1:4.2
3.7:1
1:1.7
10:1
85
Rh2(S-DOSP)4 (1.0 mol %)
Rh2(S-PTA)4 (1.0 mol %)
Rh2(S-PTV)4 (1.0 mol %)
Rh2(S-PTTL)4 (1.0 mol %)
Rh2(S-PTIL)4 (1.0 mol %)
Rh2(S-PTIL)4 (1.0 mol %)
Rh2(S-PTIL)4 (1.0 mol %)
83
81
79
69
83
49
70
49
−20
49
90
89
93
94
90
nd
1.6:1
1.9:1
1:2.1
f
7
g
8
h
9
Rh2(S-TCPTTL)4
(1.0 mol %)
1:>20
h
10
Rh2(tfa)4 (1.0 mol %)
1:>20
1:1.5
4.3:1
15:1
71
69
82
85
89
90
88
RESULTS AND DISCUSSION
11
12
13
14
15
16
Rh2(S-PTIL)4 (0.5 mol %)
Rh2(S-PTIL)4 (1.5 mol %)
Rh2(S-PTIL)4 (2.0 mol %)
Rh2(S-PTIL)4 (2.5 mol %)
Rh2(S-PTIL)4 (3.0 mol %)
Rh2(S-PTV)4 (3.0 mol %)
93
93
93
93
93
90
■
Optimization of reaction conditions aimed at maximizing the
efficiency and selectivities for [3 + 3]-cycloaddition was
initiated between isoquinolinium dicyanomethylide 1a and
enol diazoacetate 2a. Performed in toluene by slowly adding a
solution of enol diazoacetate 2a to a mixture of partially soluble
isoquinolinium dicyanomethylide 1a and dirhodium catalyst,
reactions occurred at room temperature (Table 1). When
conducting the title reaction (eq 1) with excess 2a in the
presence of 1.0 mol % Rh2(Oct)4, complete consumption of the
dicyanomethylide 1a was achieved after 3 h. Two products
were obtained as a 2.0:1 mixture in high isolated yield, and their
structures were determined by spectroscopic analysis to be
those from the anticipated (Scheme 2) [3 + 3]-cycloaddition
(3a) and an unexpected diastereoselective [3 + 2]-cyclo-
addition of 1a14 with the donor−acceptor cyclopropene formed
from 2a by catalytic dinitrogen extrusion reaction (4a).15 Use
of Rh2(S-DOSP)4 as the catalyst16 formed 3a and 4a in a 1:4.2
product ratio (entry 2), but enantioselectivity for 3a was very
poor and for 4a was not evident. Switching to Hashimoto’s
Rh2(S-PTA)4 catalyst17 resulted in a significant increase in
enantioselectivity for the [3 + 3]-cycloaddition product 3a
(entry 3). The alkyl group of the Hashimoto’s dirhodium
catalysts impacted the enantiomeric excess for 3a (entries 3−
6), and optimal enantioselectivity for 3a was achieved with
previously unreported Rh2(S-PTIL)4 whose ligand incorporates
an additional chiral center. Lowering the reaction temperature
to 0 °C slightly enhanced both the chemo- and enantiose-
lectivity for 4a but with low reaction yield (entry 7); and
alternatively performing the reaction at 60 °C provided only
>20:1
>20:1
>20:1
a
Reactions were performed at room temperature with 0.1 mmol of 1a
(1.0 equiv). An excess of 2a (1.8 equiv) in 1.0 mL toluene was added
to the reaction mixture via syringe pump over 1 h with continued
b
stirring for another 2 h. Ratios were determined by 1H NMR analysis
of reaction mixtures. Duplicate reactions show remarkable consistency
in the 3a:4a ratio. The stereochemistry of 4a was determined by H
NOE experiments. See SI for detail. Yields reported are combined
isolated yields of 3a and 4a. Enantiomeric excesses were determined
by chiral HPLC analysis. See SI for experimental detail. Reaction
performed at 0 °C. Reaction performed at 60 °C. Reaction was
continued for 23 h after the completion of adding 2a.
c
1
d
e
f
g
h
inferior selectivities (entry 8). Use of the more Lewis acidic
Rh2(S-TCPTTL)4,18 however, reversed chemoselectivity with
[3 + 2]-cycloaddition product 4a as the sole reaction outcome
(entry 9), and this reversal in chemoselectivity was mirrored in
results from the use of Lewis acidic achiral Rh2(tfa)4 under
otherwise identical conditions (entry 10).
Surprisingly, catalyst loading has a pronounced effect on
chemoselectivity (3a:4a). For instance, reactions catalyzed by
0.5 mol % (entry 11) and 1.0 mol % (entry 6) of Rh2(S-PTIL)4
at room temperature showed low selectivity for the formation
of the [3 + 3]-cycloaddition product 3a, and only a slight
increase in the 3a:4a ratio occurred when the temperature was
lowered to 0 °C. However, significant increases in the ratio
B
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
were observed with incremental increases in mol % catalyst so
that at 2.0 mol % catalyst the ratio was 15.4 0.4 (entry 13),
and a further increase in catalyst loading to 2.5 then to 3.0 mol
% led to the formation of 3a as the sole reaction product
[entries 14, 15, and 16 with Rh2(S-PTV)4]. The excellent
enantiomeric excesses obtained for 3a were not at all affected
by catalyst loading, which suggested that the formation of 3a
and 4a was independent. The [3 + 2]-cycloaddition product 4a
was only obtained as a racemate.
Table 3. Substrate Scope for Enantioselective Dearomatizing
Formal [3 + 3]-Cycloaddition
Solvent strongly influenced both chemoselectivity and
enantioselectivity. As is evident from the results reported in
Table 2, reactions performed in chlorinated hydrocarbons
Table 2. Solvent Screening for the Enantioselective Catalytic
Formal [3 + 3]-Cycloaddition Reaction of Isoquinolinium
Dicyanomethylide 1a and Enol Diazoacetate 2a
a
b
c
d
entry
solvent
yield (%)
3a:4a
ee(%) of 3a
1
2
3
4
5
6
7
8
CH2Cl2
(CH2Cl)2
CHCl3
PhCl
71
59
65
79
77
80
74
90
2.4:1
4.8:1
72
80
88
90
91
80
90
93
1.2:1
1.6:1
o-xylene
p-xylene
PhF
>20:1
>20:1
>20:1
>20:1
toluene
a
Reactions were performed on 0.1 mmol of 1a (1.0 equiv), and an
excess of 2a (1.8 equiv) in 1.0 mL solvent was added to the reaction
mixture via syringe pump over 1 h, and then reaction was continued
b
with stirring for 2 h. Ratios were determined by 1H NMR analysis of
c
reaction mixtures. Reported yields are combined isolated yields of 3a
d
and 4a. Enantiomeric excesses were determined by chiral HPLC
analysis.
(entries 1−4) gave significantly lower product control (3a:4a)
and enantioselectivities for [3 + 3]-cycloaddition; aromatic
solvents generally provided good chemoselectvities and high
enantiocontrol (entries 5−7) except when chlorobenzene was
used as solvent (entry 4). Toluene stood out as the optimal
choice (entry 8).
The substrate scope of the enantioselective [3 + 3]-
cycloaddition reaction was examined using the optimal
conditions obtained with 1a and 2a (Table 1 entry 15), and
these results are presented in Table 3. Product yields and
enantioselectivities appear to be independent of the ester
substituent of enol diazoacetate 2 (entries 1 and 2). The
absolute configuration of 3b, and others in this series by
analogy, was unambiguously determined to be S through X-ray
single crystal analysis (Figure 1). Enol diazoacetate 2c with a
methyl group instead of a hydrogen attached at the vinylogous
position underwent the [3 + 3]-cycloaddition reaction with
complete diastereocontrol (entry 3). Electronically disparate
substituents on the isoquinolinium ring are well-tolerated
(entries 4−7); consistently high reaction yields and high
enantiomeric excesses were achieved. Furthermore, the more
challenging19 pyridinium dicyanomethylides (entries 8−10)
participated in the [3 + 3]-cycloaddition transformation
providing high yields and high enantiomeric excesses of
quinolizidines that were comparable to those from the
a
Reactions were performed with 0.1 mmol of 1 (1.0 equiv) and an
excess of 2 (1.8 equiv) in 1.0 mL toluene was added to the reaction
b
mixture over 1 h, and then reacted for another 2 h. Ratios were
c
determined by 1H NMR analyses of reaction mixtures. Yields
d
reported are combined isolated yields of 3 and 4. Enantiomeric
excesses were determined by HPLC analyses on a chiral stationary
e
phase. See SI for experimental details. The stereochemistry of 3c was
f
determined by 1H NOE experiments. See SI for details. Reaction
performed at 60 °C.
C
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
influence of the vinyl substituent R4 is substantial in its effect
on reaction pathway.
The results from these experiments show a gradation in
reactivity of the intermediate metal carbene toward either
intermolecular [3 + 3]-cycloaddition with methylide 1a to form
3 or isomerization to afford donor−acceptor cyclopropene 515
that is susceptible to [3 + 2]-cycloaddition with 1a. Since only
racemic product was obtained with the diverse array of chiral
catalysts that was employed (Table 1), the diastereoselective [3
+ 2]-cycloaddition of 5 and 1 can be regarded to be a catalyst-
free process. In sharp contrast, the high enantiocontrol in the [3
+ 3]-cycloaddition process demonstrates its direct dependence
on the dirhodium catalyst. In a separate experiment, treating
the preformed and catalyst-free cyclopropene 5a with
isoquinolinium dicyanomethylide 1a produced the [3 + 2]-
cycloaddition product 4a exclusively at room temperature (eq
5).
Figure 1. ORTEP view of benzyl (S)-2-(tert-butyldimethylsilyl)oxy-
4,4-dicyano-4,11b-dihydro-3H-pyrido-[2,1-a]isoquinoline-1-carboxy-
late (3b). Ellipsoids are shown at 30% probability. CCDC 946885
contains supplementary crystallographic data for 3b.
isoquinolinium systems. Interestingly, only a single regioisomer
was obtained for 3-picolinium dicyanomethylide 1g, although
there were two potential reaction sites (entry 9), and the
product obtained was the one from addition to the less
sterically encumbered 6-position rather than to the 2-position.
In contrast to the corresponding isoquinolium dicyanome-
thylides (1, EWG = CN), the dicarbomethoxy isoquinolinium
methylide (1i, EWG = COOMe) was much less reactive toward
dirhodium-catalyzed reaction with enol diazoacetate 2a;
reactions performed at room temperature reached only 30%
completion in the normal 3 h reaction time. However, at 60 °C
over the same time period the reaction reached full
consumption of 1i, and 3k was isolated in moderate yield
although with significantly lower enantiomeric excess (Table 2,
entry 11) than from reaction with the corresponding
dicyanomethylide 1a. Overall, the dearomatizing [3 + 3]-
cycloaddition methodology represents a general approach for
the catalytic asymmetric functionalization of the dicyanome-
thylides of isoquinoline and pyridine. Although the dicarbo-
methoxy isoquinolinium methylide 1i was much less reactive
than 1a, other heterocyclic ylides14a,20 may have enhanced
reactivity and selectivity for these cycloaddition reactions.
To probe the influence from alternate substituents (R4) to
the OTBS group on the vinyl group of 2, combinations of vinyl
diazoacetates 2 and methylide 1a were subjected to Rh2(Oct)4
catalysis. Although such determinations have been conducted
to compare product yields for individual transformations,21
competitive reactions with 2 have not been reported. Whereas
the reaction of 2a (R4 = OTBS) with 1a gave a mixture of 3a
and 4a (Table 1 entry 1); with R4 = Ph the Rh2(Oct)4-catalyzed
reaction afforded only the [3 + 2]-cycloaddition product 4b (eq
4) and vinyldiazoacetate 2d (R4 = H) gave exclusive formation
However, the mechanistic explanation in which the metal
carbene formed from dirhodium carboxylate and the enol
diazoacetate either reacts with methylide 1a to produce the
product from [3 + 3]-cycloaddition or dissociates the bound
carbene with rearrangement in the form of the donor−acceptor
cyclopropene 5 (Scheme 3) does not explain the dramatic
Scheme 3. Formation of Both [3 + 3]- and [3 + 2]-
Cycloaddition Products are Dependent on Metal Carbene
Intermediate 6
increase in the ratio of 3:4 with increasing mol % of catalyst
loading (Table 1, entry 4,6, 11−16). The formation of both 3
and 4 have the same catalyst dependence in this scheme; both
emanate from the metal carbene intermediate, and product
formation for both involves reaction with 1. Thus dependence
of catalyst on product distribution suggests a more complex
role for dirhodium in these transformations, and perhaps one
that involves its coordination with methylide Lewis bases.
The coordination of the dirhodium carboxylate with an
isoquinolinium methylide was assessed in toluene at room
temperature. Unlike most of the methylides used in this study,
5-(tert-butyldimethylsilyloxy)isoquinolium dicyanomethylide
1e displayed good solubility in most common organic solvents,
and this feature enabled us to accurately determine the first
coordination constant (K1) of 1e and Rh2(S-PTIL)4.
Coordination between Rh2(S-PTIL)4 and dicyanomethylides
of the [3 + 3]-cycloaddition product 3l. Similar to reaction
outcomes with varying catalyst and reaction solvent, the
D
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1 was indicated by the intense color change of Rh2(S-PTIL)4
from light green to deep red when they were mixed. A plot of
the spectrum of the Rh2(S-PTIL)4-containing solution as
function of increasing amounts of methylide 1e with minimal
change in volume (Figure 2) shows a clear isosbestic point at
or demonstrate the influence of axial ligands on catalytic
reactivity or selectivity27 have not been successful.
In a recent attempt to determine if a bound ligand could
influence a catalytic reaction, Padwa and co-workers synthe-
sized a stable dirhodium carbene complex between an
Arduengo carbene and dirhodium pivalate,27a but all endeavors
to detect a unique reactivity or selectivity in cycloaddition or
insertion reactions from diazocarbonyl compounds for this
complex were unsuccessful. Identical catalytic reactivities and
selectivities were obtained with the parent dirhodium catalyst,
and the authors concluded that the ligated dirhodium carbene
complex underwent dissociation of carbene ligand to release the
active dirhodium catalyst (Scheme 5) that then participated in
Scheme 5. Stable Carbene on Dirhodium Undergoes
Dissociation Prior to Reaction of Dirhodium with the
Diazocarbonyl Compound
Figure 2. UV−vis titration curves and equilibrium constant K1 for
complex formation between 1e and Rh2(S-PTIL)4 in toluene at room
temperature. [Rh2(S-PTIL)4] = 2.0 × 10−3 M; 0.17 equiv of 1e
[Relative to Rh2(S-PTIL)4] was added in each increment. K1 = 545
14.
the catalytic metal carbene reactions. In contrast to this SN1-like
role for an axial carbene ligand, could the dependence of
catalyst on product distribution in reactions of 1 with 2 be
influenced by 1 as an axial ligand?
Recognizing that the catalyst in the reaction solution is in
equilibrium with methylide 1, addition of enol diazoacetate 2
could associate with the methylide-coordinated dirhodium
catalyst or the catalyst that is free of 1. The equilibrium
constant for association between 1e and Rh2(S-PTIL)4 gives
evidence of a complex, which is highly unlikely to associate the
weakly coordinating diazo compound 2. Instead, coordination
with the catalyst that is free of methylide is most likely, and
with dinitrogen extrusion this complex forms the metal carbene
intermediate (Scheme 6). However, previous reports of the
catalytic formation of donor−acceptor cyclopropenes 5 from
enol diazoacetate 2 did not discuss how this product was
formed.15
684 nm. The equilibrium constant for association between
Rh2(S-PTIL)4 and dicyanomethylides 1e was determined to be
545
14 by the methodology that we have previously
employed,22 which is a binding affinity comparable to that of
the same dirhodium compound and acetonitrile (K1 = 155
2). Although coordination to rhodium through either the
methanide carbon center of 1a or through one of its nitrile
nitrogens is possible, association through the nitrile nitrogen
offers the lesser steric resistance.
Pioneering work by Drago23 established that Lewis base
coordination with Lewis acidic dirhodium carboxylates occurs
at the axial positions to form 1:1 and 2:1 adducts with
acetonitrile and with pyridine, and that the second coordination
constant K2 was at least two to 3 orders of magnitude lower
than K1 (Scheme 4). From this and a vast array of related
investigations of the equilibrium processes of dirhodium
carboxylates, the influence of one coordinated ligand on the
association of a second ligand clearly established an inhibition
for association;24,25 but attempts to displace bound carbenes26
Two possible pathways exist for the displacement of the
rhodium-bound carbene of 6 as cyclopropene 5. One is the
SN1-like pathway that was demonstrated for the dirhodium
carbene complex between an Arduengo carbene and dirhodium
Scheme 6. Reaction Pathway for the Formation of
Cyclopropene 5
Scheme 4. Lewis Bases Occupy Axial Coordination Sites on
Dirhodium Complexes
E
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
pivalate (Scheme 5). However, in this case, increasing the mol
% of catalyst could not have influenced the ratio of [3 + 3]- to
[3 + 2]-cycloaddition products, as has been explained (Scheme
3). However, if methylide 1 serves to induce formation of
cyclopropene 5 in a through dirhodium SN2′-like displacement
reaction, the effect of increased dirhodium catalyst concen-
tration will be to increase the concentration of 6 and thereby
increase the rate for formation of [3 + 3]-cycloaddition product
3 relative to that of [3 + 2]-cycloaddition product 4 whose
formation occurs subsequent to the formation of 5. There are
several lines of evidence that support this interpretation: (1) If
the Lewis basic methylide 1 does assist the generation of
cyclopropene 5 with concomitant release of a Lewis acid−base
complex (Scheme 6), then this effect should be more
pronounced in reactions catalyzed by more Lewis acidic
dirhodium compounds (higher Keq, lower concentration of
ligand-free catalyst). As predicted, the [3 + 2]-cycloaddition
pathway completely overrides the competing [3 + 3]-
cycloaddition reaction when Rh2(S-TCPTTL)4 (Table 1,
entry 9) or Rh2(tfa)4 (Table 1, entry 10) is used. (2) External
Lewis base additives decrease the catalyst concentration and
increase the production of cyclopropene 5, thereby decreasing
chemoselectivity for [3 + 3]-cycloaddition obtained from the
reaction catalyzed by 3 mol % Rh2(S-PTIL)4 (Scheme 7). As
pathway, which is not observed, and no direct dependence on
[1] for the SN2′-like pathway.
Although the enantioselective [3 + 3]-cycloaddition and the
diastereoselective [3 + 2]-cycloaddition reactions are linked
through the intermediate rhodium carbene 6, they are
separated by the divergent outcomes from actions of methylide
1 on 6. Lewis base additives do not participate in the
enantioselective [3 + 3]-cycloaddition but they do induce
rearrangement of rhodium carbene to generate cyclopropene 5.
Indeed, it is possible that the influence of at least some of the
chlorinated solvents on reaction selectivity may arise from
Lewis base displacement of cyclopropene 5 from the dirhodium
carbene 6.28 However, when the bulky nitriles 3 or 4 were used
as additives in the catalytic reactions, minimal impact on the
product distribution was displayed.
Although the uncatalyzed reaction of methylide 1a with
donor−acceptor cyclopropene 5a, formed by catalytic
dinitrogen extrusion from enol diazoacetate 2a, underwent [3
+ 2]-cycloaddition exclusively, and this process adequately
accounts for the formation of 4a, the effect of increasing mol %
catalyst on the relative yield of [3 + 3]-cycloaddition product
3a (Table 1) coupled with the Lewis base promoted
cyclopropene formation (Schemes 6 and 7) suggested that
the role of the cyclopropene intermediate is more complex than
what has been portrayed. Could cyclopropene 5a form metal
carbene 6 in what would be a reversal of the reaction that
formed 5a? Although cyclopropenes are known to be
stoichiometric precursors to metal carbenes in selected
cases,29 only recently have reports emerged of catalytic
reactions of cyclopropenes in metal carbene transformations.30
To test this hypothesis, we replaced enol diazoacetate 2a in
eq 1 with cyclopropene 5a, generated with rhodium(II) acetate
but separated from the catalyst, and performed reactions in the
presence of variable mol % of Rh2(S-PTIL)4 under otherwise
identical conditions. As anticipated, high enantiocontrol for the
formation of the [3 + 3]-cycloaddition product 3a was
achieved; and this transformation and its enantioselectivity
demonstrated the direct involvement of the chiral catalyst in
the bond-forming steps. Chemoselectivities (3a:4a) directly
correlated with the mol % of Rh2(S-PTIL)4 used, and the
enantioselective [3 + 3]-cycloaddition product 3a was obtained
with greater than 95% selectivity when the catalyst loading was
increased to 2 mol % (Scheme 8). Thus the outcomes of these
reactions, including the enantioselectivity of 3a and chemo-
selectivity for 3a:4a, were identical and independent of the
source of the carbene-forming reactant.
Scheme 7. Effect of Lewis Bases on Chemoselectivity
Considering the boomerang interconversion between the
enol-TBS substituted chiral dirhodium carbene 6 and the
donor−acceptor cyclopropene 5, the relationship between the
amount of catalyst and the ratio of products from [3 + 3]- and
[3 + 2]-cycloaddition reactions now resembles the experimental
observation (Scheme 9). If the interconversion between (5 +
Rh2L4) and 6 is rapid relative to cycloadditions, then the ratio
3:4 is directly related to the concentration of catalyst (Scheme
9). However, a plot of the ratio 3a:4a versus mol % Rh2(S-
PTIL)4 suggests a more complex exponential relationship with
the catalyst that awaits further mechanistic definition.
Taking these findings into account, we now have a more
complete rationale for the events that occur during the highly
enantioselective dearomatizing [3 + 3]-cycloaddition reaction
of enol diazoacetates and isoquinolinium or pyridinium
methylides (Scheme 10). The central outcome is stereo-
selective cycloaddition that results from the vinylogous reaction
anticipated, more of the [3 + 2]-cycloaddition product 4a is
formed in reactions where CH3CN or the more strongly
coordinating Lewis base (Et3N or pyridine) is present. That
these additives do not influence the [3 + 3]-cycloaddition
pathway is indicated by the observation that enantiomeric
excesses of 3a are not influenced by the presence of CH3CN.
Use of the stronger σ-donorsEt3N or pyridinecompletely
shuts down the [3 + 3]-cycloaddition pathway. (3) The steady
state approximation for formation of cyclopropene 5 in the
competitive processes described in Scheme 3 suggests a direct
relationship in the [3]/[4] ratio with [1] for the SN1-like
F
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
TBS-substituted dirhodium carbene 6 forms donor−acceptor
cyclopropene 5 that is proposed to occur by a through-
rhodium−rhodium bond displacement by a Lewis base that
includes isoquinolinium or pyridinium methylides. The
dirhodium catalyst is in equilibrium with reactant isoquinoli-
nium or pyridinium methylides that has the effect of lowering
the turnover rate for [3 + 3]-cycloaddition. The decreased
amount of catalyst allows direct [3 + 2]-cycloaddition of the
donor−acceptor cyclopropene 5 with isoquinolinium or
pyridinium methylides to form 4. As with the isoquinolinium
or pyridinium methylides that have multiple reaction pathways
for product formation, so also does cyclopropene 5 which
undergoes either [3 + 2]-cycloaddition with 1 or forms
dirhodium carbene 6. The net result is that the catalyst has
ultimate control on the reaction pathway and reaction
stereoselectivity.
Scheme 8. Dependence of Product Distribution on Catalyst
Loading with Cyclopropene 5a as the Metal Carbene
Precursor
The new processes uncovered in this studySN2′ induced
formation of donor−acceptor cyclopropene 5 and its
boomerang equilibrium with the axial ligand-free dirhodium
catalystare fundamental to understanding metal carbene
reaction chemistry. The dirhodium carbene derived from enol
diazoacetate 2 is uniquely capable of rearrangement to
cyclopropene 5 with release of the dirhodium catalyst, and
this reaction occurs in competition with vinylogous addition of
methylides. Isoquinolinium and pyridinium methylides are
relatively strong Lewis bases compared to acetonitrile which is a
much stronger Lewis base than the reactant diazo compound,
and their Lewis basicity prompts competing association with
the dirhodium catalyst, vinylogous addition to the enol-TBS
substituted dirhodium carbene 6, and displacement of cyclo-
propene 5 from the enol-TBS substituted dirhodium carbene 6.
The further implications of these processes are under
investigation.
Scheme 9. Relationship Between the Cycloaddition Product
Ratio (3:4) and the Dirhodium Catalyst
In summary, we have disclosed a highly enantioselective
dearomatizing [3 + 3]-cycloaddition reaction. A [3 + 2]-
cycloaddition of isoquinolinium or pyridinium methylides with
the cyclopropene derived from rhodium carbene competes with
the [3 + 3]-cycloaddition reaction but can be turned on or off
with a higher mol % of catalyst or increasing amount of Lewis
base.
EXPERIMENTAL SECTION
■
Sample Procedure for the Preparation of Isoquinolinium/
Pyridinium Dicyanomethylides. To a THF solution (2.0 mL) of 4-
bromoisoquinoline (416 mg, 2.0 mmol) cooled to 0 °C, tetracyano-
ethylene oxide (298 mg, 2.1 mmol) dissolved in 2.0 mL THF was
added dropwise over 10 min. The reaction solution was stirred at 0 °C
for 12 h, during which time a yellow precipitate formed. After warming
to room temperature, the reaction mixture was diluted with diethyl
ether (10 mL), and the precipitate was filtered and washed with
diethyl ether (5.0 mL). The resulting yellow solid was collected then
dissolved in CH2Cl2. Diatomaceous earth (3.0 g) was added to the
solution, and then solvent was removed under reduced pressure. The
solid residue was loaded onto a silica gel column (with CH2Cl2 and
ethyl acetate as eluents) to isolate 4-bromoisoquinolinium dicyano-
methylide 1b (234 mg, 0.86 mmol, 43% yield).
Scheme 10. Detailed General Mechanism for the Competing
[3 + 3]- and [3 + 2]-Cycloaddition Reactions
Preparation of Dirhodium(II) Tetrakis[N-Phthaloyl-(S,S)-iso-
leucinate] Bis(ethyl acetate) Adduct [Rh2(S-PTIL)4(EtOAc)2]. To
a flame-dried, 50-mL, single-necked, round-bottomed flask equipped
with a magnetic stirring bar, Rh2(OAc)4 (221 mg, 0.50 mmol), N-
phthaloyl-(S,S)-iso-leucine (653 mg, 2.50 mmol) and chlorobenzene
(25 mL) were added sequentially under a nitrogen atmosphere. The
flask was fitted with a 10-mL Soxhlet extraction apparatus into which
was placed a thimble containing 5 g of an oven-dried mixture of 2 parts
sodium carbonate and 1 part of sand. The mixture was heated to reflux
of enol-TBS substituted chiral dirhodium carbene 6 with
isoquinolinium or pyridinium methylides that is accompanied
by ring closure to cycloaddition product 3 with displacement of
the chiral catalyst. In competition with this process, the enol-
G
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
for 3 h and then cooled to room temperature. Solvent was removed
under reduced pressure, and the residue was dissolved in ethyl acetate
(30 mL). The resulting solution was washed with saturated aqueous
NaHCO3 (2 × 20 mL) and brine (20 mL) and then dried over
anhydrous Na2SO4. Filtration and subsequent solvent removal under
reduced pressure furnished a green solid that was purified by column
chromatography on silica gel (hexanes/EtOAc). The resulting green
solid was dissolved in 5 mL ethyl acetate and 50 mL hexanes was then
added to the solution. Green solids that formed after standing
overnight at room temperature were collected by filtration, washed
with hexanes (2 × 5 mL) and dried under high vacuum (0.1 Torr) at
room temperature for 3 h to provide dirhodium(II) tetrakis[N-
phthaloyl-(S,S)-iso-leucinate] bis(ethyl acetate) adduct [Rh2(S-PTI-
L)4(EtOAc)2] (590 mg, 0.42 mmol, 83% yield).
methylene protons [δ (ppm): 1.88 (s, 2H)] from cyclopropene 5a.
The solution was filtered through a short pad (∼1 cm) of
BAKERBOND-CN silica (40 μm Prep LC packing) to remove the
dirhodium catalyst and the silica pad was washed with 0.25 mL
toluene. The combined filtrates containing methyl 2-(tert-
butyldimethylsilyl)oxy-cycloprop-1-enecarboxylate 5a were used di-
rectly in the subsequent reactions.
Sample Procedure for the Diastereoselective [3 + 2]-
Cycloaddition of Isoquinolinium/Pyridinium Methylides 1
and cyclopropene 5. To the prepared toluene solution of methyl
2-(tert-butyldimethylsilyl)oxy-cycloprop-1-enecarboxylate 5a at room
temperature, isoquinolinium dicyanomethylide 1a (20 mg, 0.10 mmol)
was added as a solid under a nitrogen atmosphere. The resulting
suspension was stirred at room temperature for 3 h during which time
the mixture gradually became a homogeneous solution. The solution
was then concentrated under reduced pressure and directly loaded
onto a silica gel column (with 1:1 of CH2Cl2:hexanes as eluents) to
isolate methyl (8aSR,9aSR,9bSR)-8a-(tert-butyldimethylsilyl)oxy-8,8-
dicyano-8a,9,9a,9b-tetrahydro-8H-cyclopropa[3,4]pyrrolo[2,1-a]-
isoquinoline-9a-carboxylate 4a (32 mg, 0.076 mmol, 76% yield).
Sample Procedure for the Enantioselective [3 + 3]-Cyclo-
addition of Isoquinolinium/Pyridinium Methylides 1 and
cyclopropene 5. To a 10-mL flame-dried Schlenk flask containing
a magnetic stirring bar, [Rh2(S-PTIL)4(EtOAc)2] (2.8 mg, 0.0020
mmol), isoquinolinium dicyanomethylide 1a (20 mg, 0.10 mmol) and
1.0 mL of toluene were added sequentially under a nitrogen
atmosphere. Then the flask was capped by a rubber septum and the
resulting solution was stirred at room temperature. The prepared
toluene solution of methyl 2-(tert-butyldimethylsilyl)oxy-cycloprop-1-
enecarboxylate 5a was added to the stirring mixture of [Rh2(S-
PTIL)4(EtOAc)2] and isoquinoline dicyanomethylide 1a over 1 h via
syringe pump. Stirring was continued at room temperature for 2 h
after the completion of adding 2a. Then the reaction solution was
concentrated under reduced pressure and directly loaded onto a silica
gel column (with 1:1 of CH2Cl2:hexanes as eluents) to isolate methyl
(S)-2-(tert-butyldimethylsilyl)oxy-4,4-dicyano-4,11b-dihydro-3H-
pyrido[2,1-a]isoquinoline-1-carboxylate 3a (32 mg, 0.075 mmol, 75%
yield). HPLC analysis of the enantioselective [3 + 3]-cycloaddition
product 3a on chiral stationary phase showed an enantiomeric excess
of 93%.
Sample Procedure for the Enantioselective [3 + 3]-Cyclo-
addition of Isoquinolinium/Pyridinium Methylides 1 and Enol
Diazoacetate 2. To a 10-mL flame-dried Schlenk flask containing a
magnetic stirring bar, [Rh2(S-PTIL)4(EtOAc)2] (4.2 mg, 0.0030
mmol), isoquinolinium dicyanomethylide 1a (20 mg, 0.10 mmol) and
1.0 mL of toluene were added sequentially under a nitrogen
atmosphere. Then the flask was capped by a rubber septum and the
resulting solution was stirred at room temperature for 5 min before
methyl 3-(tert-butyldimethylsilyl)oxy-2-diazobut-3-enoate 2a (46 mg,
0.18 mmol) dissolved in 1.0 mL of toluene was added via syringe
pump over 1 h. Stirring was continued at room temperature for 2 h
after the completion of adding 2a. Then the reaction solution was
concentrated under reduced pressure and directly loaded onto a silica
gel column (with 1:1 of CH2Cl2:hexanes as eluents) to isolate methyl
(S)-2-(tert-butyldimethylsilyl)oxy-4,4-dicyano-4,11b-dihydro-3H-
pyrido[2,1-a]isoquinoline-1-carboxylate 3a (38 mg, 0.090 mmol, 90%
yield). HPLC analysis of the enantioselective [3 + 3]-cycloaddition
product 3a on a chiral stationary phase indicated an enantiomeric
excess of 93% [Chiralpak OD-H; flow rate: 1.0 mL/min; hexanes/i-
PrOH: 95:5; 254 nm; tR (minor) = 5.5 min; tR (major) = 6.2 min].
Sample Procedure for the Enantioselective [3 + 3]-Cyclo-
addition of Isoquinolinium/Pyridinium Methylides 1 and Enol
Diazoacetate 2 in the Presence of a Lewis Base. To a 10-mL
flame-dried Schlenk flask containing a magnetic stirring bar, [Rh2(S-
PTIL)4(EtOAc)2] (4.2 mg, 0.0030 mmol), isoquinolinium dicyano-
methylide 1a (20 mg, 0.10 mmol), 1.0 mL of toluene and CH3CN (12
mg, 0.30 mmol) were added sequentially under a nitrogen atmosphere.
Then the flask was capped by a rubber septum and the resulting
solution was stirred at room temperature for 5 min before methyl 3-
(tert-butyldimethylsilyl)oxy-2-diazobut-3-enoate 2a (46 mg, 0.18
mmol) dissolved in 1.0 mL of toluene was added via syringe pump
over 1 h. Stirring was continued at room temperature for 2 h after the
completion of adding 2a. Then the reaction solution was concentrated
under reduced pressure. The residue was dissolved in CDCl3 to
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details, characterization data, chiral HPLC
analyses, and crystal structures for 3b, 4b. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
mdoyle3@umd.edu
1
■
determine the ratio of 3a:4a by H NMR spectroscopy. Then the
solution was directly loaded onto a silica gel column (with 1:1 of
CH2Cl2:hexanes as eluents) to isolate methyl (S)-2-(tert-
butyldimethylsilyl)oxy-4,4-dicyano-4,11b-dihydro-3H-pyrido[2,1-a]-
isoquinoline-1-carboxylate 3a (26 mg, 0.062 mmol, 62% yield). HPLC
analysis of 3a on chiral stationary phase showed an enantiomeric
excess of 93%.
Procedure for the Generation of Methyl 2-(tert-
Butyldimethylsilyl)oxy-cycloprop-1-enecarboxylate 5a in Tol-
uene from Enol Diazoacetate 2a by Rh2(OAc)4-catalyzed
Dinitrogen Extrusion Reaction. To a flame-dried vial equipped
with a magnetic stirring bar, Rh2(OAc)4 (0.9 mg, 0.0020 mmol) and
0.75 mL toluene were added sequentially under a nitrogen atmosphere
and then capped with a rubber septum. The solution was stirred at
room temperature while methyl 3-(tert-butyldimethylsilyl)oxy-2-
diazobut-3-enoate 2a (46 mg, 0.18 mmol) was added dropwise over
1 min. Rapid evolution of nitrogen occurred, and the yellow color of
2a disappeared within 5 min. Complete consumption of enol
diazoacetate 2a and the generation of donor−acceptor cyclopropene
5a were verified by 1H NMR spectroscopy by the disappearance of the
two vinyl protons [δ (ppm): 5.02 (d, J = 2.1 Hz, 1H), 4.27 (d, J = 2.1
Hz, 1H)] on enol diazoacetate 2a and the appearance of the
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Support for this research from the National Science Foundation
(CHE-1212446) is gratefully acknowledged.
■
REFERENCES
■
(1) (a) Civjan, N.; O’Connor, S. E. Alkaloids, Natural Products in
Chemical Biology; John Wiley & Sons, Inc., Hoboken, NJ, 2012.
(b) Scott, J. D.; Williams, R. M. Chem. Rev. 2002, 102, 1669.
(c) Bentley, K. W. Nat. Prod. Rep. 2005, 22, 249. (cc) Bentley, K. W.
Nat. Prod. Rep. 2006, 23, 444. (d) Michael, J. P. Nat. Prod. Rep. 2008,
25, 139.
(2) (a) Chrzanowska, M.; Rozwadowska, M. D. Chem. Rev. 2004,
104, 3341. (b) Stockigt, J.; Antonchick, A. P.; Wu, F.; Waldmann, H.
̈
Angew. Chem., Int. Ed. 2011, 50, 8538.
H
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(3) Perreault, S.; Rovis, T. Chem. Soc. Rev. 2009, 38, 3149.
(4) (a) Itoh, T.; Yokoya, M.; Miyauchi, K.; Nagata, K.; Ohsawa, A.
Org. Lett. 2006, 8, 1533. (b) Lalonde, M. P.; McGowan, M. A.;
Rajapaksa, N. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2013, 135, 1891.
(5) Frisch, K.; Landa, A.; Saaby, S.; Jørgensen, K. A. Angew. Chem.,
Int. Ed. 2005, 44, 6058.
(6) Catalytic enantioselective Reissert-type reaction: (a) Takamura,
M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2000,
122, 6327. (b) Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M.
J. Am. Chem. Soc. 2001, 123, 6801. (c) Funabashi, K.; Ratni, H.; Kanai,
M.; Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 10784. (d) Ichikawa, E.;
Suzuki, M.; Yabu, K.; Albert, M.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 2004, 126, 11808. (e) Taylor, M. S.; Tokunaga, N.; Jacobsen, E.
N. Angew. Chem., Int. Ed. 2005, 44, 6700. (f) Sun, Z.; Yu, S.; Ding, Z.;
Ma, D. J. Am. Chem. Soc. 2007, 129, 9300. (g) Yamaoka, Y.; Miyabe,
H.; Takemoto, Y. J. Am. Chem. Soc. 2007, 129, 6686.
(7) Catalytic enantioselecitve Friedel−Crafts reaction: (a) Bandini,
M.; Melloni, A.; Umani-Ronchi, A. Angew. Chem., Int. Ed. 2004, 43,
550. (b) Taylor, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2004, 126,
10558. (c) Uraguchi, D.; Sorimachi, K.; Terada, M. J. Am. Chem. Soc.
2004, 126, 11804. (d) Evans, D. A.; Fandrick, K. R.; Song, H. J. Am.
Chem. Soc. 2005, 127, 8942.
(8) Review of catalytic enantioselective hydrogenation of hetero-
arenes and arenes: Wang, D.; Chen, Q.; Lu, S.; Zhou, Y. Chem. Rev.
2012, 112, 2557.
(24) (a) Chifotides, H. T.; Dunbar, K. R. In Multiple Bonds Between
Metal Atoms, 3rd ed.; Cotton, F. A., Murillo, C. A., Walton, R. A., Eds.;
Springer: New York, 2005; Chapter 12. (b) Doyle, M. P.; Ren, T. Prog.
Inorg. Chem. 2001, 49, 113.
(25) (a) Trindade, A. F.; Coelho, J. A. S.; Afonso, C. A. M.; Veiros, L.
F.; Gois, P. M. P. ACS Catal. 2012, 2, 370. (b) Pirrung, M. C.; Liu, H.;
Morehead, A. T., Jr. J. Am. Chem. Soc. 2002, 124, 1014. (c) Wynne, D.
C.; Olmstead, M. M.; Jessop, P. G. J. Am. Chem. Soc. 2000, 122, 7638.
(26) (a) Galan, B. R.; Gembicky, M.; Dominiak, P. M.; Keister, J. B.;
Diver, S. T. J. Am. Chem. Soc. 2005, 127, 15702. (b) Lop
Barrio, P.; Borge, J. Organometal. 2012, 31, 7844. (c) Baquero, E. A.;
Silbestri, G. F.; Gomez-Sal, P.; Flores, J. C.; de Jesus, E. Organometal.
2013, 32, 2814.
́
ez, L. A.;
́
́
(27) (a) Snyder, J. P.; Padwa, A.; Stengel, T.; Arduengo, A. J., III;
Jockisch, A.; Kim, H. J. Am. Chem. Soc. 2001, 123, 11318. (b) For the
use of a NHC-ligated dirhodium acetate for arylation of aldehydes with
arylboronates, see: Trindade, A. F.; Gois, P. M. P.; Veiros, L. F.; Andre,
́
V.; Duarte, M. T.; Afonso, C. A. M.; Caddick, S.; Cloke, F. G. N. J.
Org. Chem. 2008, 73, 4076.
(28) For a report of a strong solvent effect on using chiral cationic
dirhodium(II,III) carboxamidate as Lewis acid catalyst: Wang, X.;
Weigl, C.; Doyle, M. P. J. Am. Chem. Soc. 2011, 133, 9572.
(29) (a) Rubin, M.; Rubina, M.; Gevorgyan, V. Chem. Rev. 2007, 107,
3117. (b) Marek, I.; Simaan, S.; Masarwa, A. Angew. Chem., Int. Ed.
2007, 46, 7364. (c) Zhu, Z.; Wei, Y.; Shi, M. Chem. Soc. Rev. 2011, 40,
5534.
(9) (a) Xu, X.; Hu, W.; Zavalij, P. Y.; Doyle, M. P. Angew. Chem., Int.
Ed. 2011, 50, 11152. (b) Wang, X.; Xu, X.; Zavalij, P. Y.; Doyle, M. P.
J. Am. Chem. Soc. 2011, 133, 16402.
(10) Panne, P.; Fox, J. M. J. Am. Chem. Soc. 2007, 129, 22.
(11) Zhu, C.; Yoshimura, A.; Ji, L.; Wei, Y.; Nemykin, V. N.;
Zhdankin, V. V. Org. Lett. 2012, 14, 3170.
(30) (a) Muller, P.; Pautex, N.; Doyle, M. P.; Bagheri, V. Helv. Chim.
̈
Acta 1990, 73, 1233. (b) Seidel, G.; Mynott, R.; Furstner, A. Angew.
̈
Chem., Int. Ed. 2009, 48, 2510. (c) Miege, F.; Meyer, C.; Cossy, J. Org.
Lett. 2010, 12, 4144. (cc) Miege, F.; Meyer, C.; Cossy, J. Angew. Chem.,
Int. Ed. 2011, 50, 5932.
́
(12) Fernandez, N.; Carrillo, L.; Vicario, J. D.; Badía, D.; Reyes, E.
Chem. Commun. 2011, 47, 12313.
(13) Reviews of α-amino nitriles in organic synthesis: (a) Enders, D.;
Shilvock, J. P. Chem. Soc. Rev. 2000, 29, 359. (b) Wang, J.; Liu, X.;
Feng, X. Chem. Rev. 2011, 111, 6947.
(14) Examples of [3 + 2]-cycloaddition reactions of isoquinolinium
ylides with alkenes: (a) Kanemasa, S.; Takenaka, S.; Watanabe, H.;
Tsuge, O. J. Org. Chem. 1989, 54, 420. (b) Kostik, E. I.; Abiko, A.;
Oku, A. J. Org. Chem. 2001, 66, 1638. (c) Ruano, J. L. G.; Fraile, A.;
Martín, M. R.; Gonzal
Chem. 2011, 76, 3296.
́
ez, G.; Fajardo, C.; Martín-Castro, A. M. J. Org.
(15) Cyclopropene formation from enol diazoacetates: (a) Davies, H.
M. L.; Houser, J. H.; Thornley, C. J. Org. Chem. 1995, 60, 7529.
(b) Davies, H. M. L.; Ahmed, G.; Churchill, M. R. J. Am. Chem. Soc.
1996, 118, 10774. (c) Xu, X.; Shabashov, D.; Zavalij, P. Y.; Doyle, M.
P. Org. Lett. 2012, 14, 800. (d) Xu, X.; Shabashov, D.; Zavalij, P. Y.;
Doyle, M. P. J. Org. Chem. 2012, 77, 5313.
(16) (a) Davies, H. M. L.; Bruzinski, P. R.; Lake, D. H.; Kong, N.;
Fall, M. J. J. Am. Chem. Soc. 1996, 118, 6897. (b) Davies, H. M. L.;
Beckwith, R. E. J. Chem. Rev. 2003, 103, 2861.
(17) (a) Watanabe, N.; Anada, M.; Hashimoto, S.; Ikegami, S. Synlett
1994, 1031. (b) Anada, M.; Hashimoto, S. Tetrahedron Lett. 1998, 39,
79. (c) Doyle, M. P.; Liu, Y.; Ratnikov, M. Org. React. 2013, 80, 1.
(18) Shimada, N.; Hanari, T.; Kurosaki, Y.; Takeda, K.; Anada, M.;
Nambu, H.; Shiro, M.; Hashimoto, S. J. Org. Chem. 2010, 75, 6039.
(19) Barluenga, J.; Lonzi, G.; Riesgo, L.; Lop
́ ́
ez, L. A.; Tomas, M. J.
Am. Chem. Soc. 2010, 132, 13200.
(20) (a) Wu, L.; Sun, J.; Yan, C. Org. Biomol. Chem. 2012, 10, 9452.
(b) Liu, Y.; Zhang, Y.; Shen, Y.; Hu, H.; Xu, J. Org. Biomol. Chem.
2010, 8, 2449. (c) Katritzky, A. R.; Qiu, G.; Yang, B.; He, H. J. Org.
Chem. 1999, 64, 7618.
(21) Davies, H. M. L.; Stafford, D. G.; Doan, B. D.; Houser, J. H. J.
Am. Chem. Soc. 1998, 120, 3326.
(22) Xu, X.; Doyle, M. P. Inorg. Chem. 2011, 50, 7610.
(23) (a) Drago, R. S.; Long, J. R.; Cosmano, R. Inorg. Chem. 1981,
20, 2920. (b) Drago, R. S. Inorg. Chem. 1982, 21, 1697. (c) Drago, R.
S.; Long, J. R.; Cosmano, R. Inorg. Chem. 1982, 21, 2196.
I
dx.doi.org/10.1021/ja406482q | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX