Gold-catalyzed waste-free generation and reaction of azomethine ylides: Internal redox/dipolar cycloaddition cascade

Article abstract of DOI:10.1002/anie.200802802

(Chemical Equation Presented) High-octane synthesis: Azomethine ylides can be generated from an internal redox reaction of a nitrone-tethered alkyne under electrophilic metal catalysis (see scheme; M=metal). This novel and atom-economical generation of an azomethine ylide does not involve potentially explosive diazo derivatives. The azomethine ylide can participate in a dipolar cycloaddition cascade to provide an azabicyclo[3.2.1]octane.

Full text of DOI:10.1002/anie.200802802

Communications
DOI: 10.1002/anie.200802802
Gold Catalysis
Gold-Catalyzed Waste-Free Generation and Reaction of Azomethine
Ylides: Internal Redox/Dipolar Cycloaddition Cascade**
Hyun-Suk Yeom, Ji-Eun Lee, and Seunghoon Shin*
Dedicated to Professor Junghun Suh on the occasion of his 60th birthday
1,3-Dipolar cycloaddition of ylidic species, such as azome-
thine ylides with p bonds, is a powerful method for the
construction of complex N heterocycles.[1] Its utility has been
proven in numerous natural product syntheses, providing
expedient routes to complex polycyclic skeletons.[2] Among
various approaches for the generation of azomethine ylide,
metal-catalyzed decomposition of a-diazo carbonyl com-
pounds,[3] and subsequent carbene addition and cycloaddition
has been one of the most popular and effective approaches.[4]
However, diazo derivatives (or their precursors) can be
explosive, which limits their use in large-scale applications. In
addition, their syntheses from carbonyl compounds involve
stoichiometric amounts of a base and the generation of waste
from the diazo-transfer agents. To solve this problem, we
envisioned using a nitrone as an oxidant for the alkyne under
Scheme 1. a) Generation of an azomethine ylide by a metal-catalyzed
internal redox reaction. b) Internal redox/dipolar cycloaddition (IR-DC)
electrophilic metal catalysis; that is, addition of the nitrone to
cascade.
ꢀ
a metal-activated alkyne with subsequent N O cleavage by
back bonding would lead to an a-carbonyl carbenoid by an
internal redox (IR) reaction.[5,6] Addition of an imine would
then generate an azomethine ylide equivalent in both a safe
and atom-economical fashion (Scheme 1a).[7,8] Furthermore,
the inter-/intramolecular dipolar cycloaddition (DC) cascade
with a catalyst turnover would provide an expedient route to
azabicyclo[3.2.1]octane systems (Scheme 1b) from a simple
precursor.
Table 1: Catalyst screening by using 1 as the substrate.
Entry[a] Cat. (mol%)
t [h] T [8C] Yield 2 [%][b] Yield 3 [%]
To probe such reactivity, nitrone 1 was easily prepared
from the condensation of ortho-1,6-enynyl benzaldehyde and
N-benzyl hydroxyamine, and then treated with electrophilic
metal salts (Table 1). When 1 was treated with 5 mol% of
PtCl2 in 1,2-dichloroethane (DCE) at 1008C, the expected
azabicyclo[3.2.1]octane 2 was indeed obtained and isolated in
a 73% yield as a single diastereomer (Table 1, entry 1).[9] Its
tetracyclic structure was unambiguously confirmed by X-ray
crystallographic analysis (Figure 1).[10] Among the metals we
1
2
3
4
5
6
7
8
9
PtCl2 (5)
PtBr2 (5)
AuCl3 (5)
AuCl3 (1)
1
1
2
1
100
100
RT
70
100
70
50
70
70
73[c]
70
13
13
<5
0
81
61[c,g]
44
[Au(IMes)]SbF6 (5) 20
6
[Au(PPh3)]OTf (5)
[Au(L)]OTf (5)[e]
AgSbF6 (5)
12
4
1
n.r.[d]
trace
45
–
66[c]
0[f]
0
AuCl3 (2)
1
88(82[c,g]
)
[a] [1]=0.1m in 1,2-dichloroethane. [b] Yield based on NMR spectra of
the crude reaction mixture unless otherwise noted. [c] Yield of isolated
product. [d] No reaction. [e] L=tBu2P(2-biphenyl). [f] 43% 1 remained.
[g] In CH3NO2. IMes=1,3-bis(2,4,6-trimethylphenyl)imidazolin-2-yli-
dene.
[*] H.-S. Yeom, Prof. Dr. S. Shin
Department of Chemistry, Hanyang University
Seoul 133-791 (Korea)
Fax: (+82)2-2299-0762
E-mail: sshin@hanyang.ac.kr
tested, various PtII, AgI, and AuIII salts were found to be
effective for the IR-DC cascade reaction (Table 1, entries 1–4
and 8), whereas cationic AuI complexes gave inferior results
(Table 1, entries 5 and 6).[11] When bulky [tBu2(o-biphenyl)-
Au]OTf was used, a side product, whose NMR spectra are
consistent with isoindole 3, was isolated in 66% yield
(Table 1, entry 7; also see Eq. (2) and Scheme 2).[13] Among
the catalysts surveyed, AuCl3 (5 mol%) gave the best results,
J.-E. Lee
Department of Chemistry, Gyeongsang National University
Chinju 660-701 (Korea)
[**] This work was supported by the Korea Research Foundation Grant
funded by the Korean Government (MOEHRD, Basic Research
Promotion Fund, KRF-2007-313-C00408). We gratefully thank Prof.
Shim Sung Lee (GNU) for X-ray crystallographic analysis.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200802802.
7040
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 7040 –7043

Angewandte
Chemie
Table 2: Scope of IR-DC cascade reaction.
Entry[a] Substrate[b]
Conditions Product,
yield [%][c]
1
2
1 R1 =Bn, R2, R3, R4 =H
708C, 1 h 2, 82
4 R1 =Bn, R2, R3 =H, R4 =Me
708C,
1.5 h
708C,
0.5 h
708C,
0.5 h[e]
708C,
0.5 h
5, 59
3
4
5
6
6 R1 =Bn, R2/R3 =Me/H and H/Me
(E/Z=3:1), R4 =H
7, 80[d]
9, 67
8 R1 =Bn, R2, R3 =Me, R4 =H
Figure 1. Molecular structure of compound 2 (ORTEP view). Selected
bond lengths [Š] and angles [8]: O1–C12 1.2187(15), N1–C5
1.4726(15), N1–C13 1.4771(14); N1-C5-C6 111.14(10), N1-C5-C4
99.97(9), C6-C5-C4 112.52(10).
10 R1 =Bn, R2 =CO2Me, R3, R4 =H
12 R1 =Me, R2, R3, R4 =H
11, 83
13, 75
708C,
2.5 h
providing 2 (81%) in 2 hours at room temperature (Table 1,
entry 3). Significant catalytic activity was observed even when
the catalyst loading was lowered to 1 mol%, although higher
temperature was required (Table 1, entry 4). Additional
screening of the solvent led to AuCl3 (2 mol%) in CH3NO2
(708C) as our optimized conditions (Table 1, entry 9), and
then we examined the generality of this reaction.[11]
7
8
14 R=Me
16 R=Ph
708C, 1 h[e] 15, 76
708C,
17, 77
0.5 h[e]
Various substituents on the enyne skeleton were well-
tolerated, efficiently providing the desired azabicyclo-
[3.2.1]octanes (Table 2). For example, variously substituted
olefins (Table 2, entries 1–4), as well as enoates (Table 2,
entry 5) on the alkene were suitable. N-methyl nitrone 12, 1,6-
diynyl 14 and 16, as well as substrates having an NTs (Ts = p-
toluenesulfonyl) or C(CH2OBn)2 tether (e.g. 18 and 21) gave
satisfactory yields of the desired azabicycles (Table 2,
entries 6–11). However, the allyl propargyl ether 20 was not
a suitable substrate under the reaction conditions. Impor-
tantly, the skeletal variation allows departure from the o-
alkynyl benzaldehyde motif; substrates 21, 23, and 25, having
an alkene tether between the nitrone and the enyne, also
underwent an efficient IR-DC cascade with reasonable yields
(Table 2, entries 11–13). Notably, for the reaction of 21
(E/Z=1:1 mixture), only the Z isomer participated in the
reaction (Table 2, entry 11).
We then attempted to extend this chemistry to an
intermolecular dipolar cycloaddition. To our delight, when
alkyne/nitrone 27 was reacted with 5 equivalents of diethyl
acetylenedicarboxylate (DEAD) in the presence of AuCl3,
dipolar cycloaddition of the azomethine ylide gave product 28
in 58% yield; notably, none of the competing nitrone
cycloaddition product was observed [Eq. (1)]. Intriguingly,
in the absence of DEAD, substrate 27 did not react at all to
9
10
18 X=NTs, R=H
20 X=O, R=Cy
708C, 1 h 19, 85
708C, 1 h dec.[f]
11
12
13
21 R1,R2 =H (E/Z=1:1),
X=C(CH2OBn)2
708C, 2 h 22, 89[g]
23 R1 =H, R2 =nBu, X=C(CO2Et)2
(Z isomer)
708C,
24, 55
0.5 h
708C, 1 h 26, 58
25 R1, R2 =(CH2)4, X=C(CO2Et)2
[a] 2 mol% of AuCl3 unless otherwise noted. [b] [Substrate]=0.1m in
CH3NO2. [c] Yield of products isolated after chromatography; all
products were obtained as single respective diastereomers, except
entry 3. [d] Product 7 in 3:1 d.r. in favor of the structure shown.
[e] 4 mol% of AuCl3 was used. [f] Decomposed. [g] Based on the
recovered starting E isomer (47%). Cy=cyclohexyl.
afford a free azomethine ylide under AuCl3 catalysis; it
remained intact under otherwise identical conditions.[12]
However, treating 27 with [Au(IPr)]OTf (IPr= N,N’-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene; 5 mol%) in CH2Cl2 in
the absence of DEAD, led instead to the isolation of
isoindole/aldehyde 29 in 55% yield [Eq. (2)]. Under similar
conditions 30 was transformed into known compound 31,
confirming our assignment of the isoindole structure.[13]
The generality and efficiency of the above IR-DC cascade
reaction strongly supports the intermediacy of an azomethine
ylide (C, Scheme 1b). The formation of isoindole 3 (as well as
Angew. Chem. Int. Ed. 2008, 47, 7040 –7043
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7041

Communications
strate 32 having a saturated bridge between the enyne and the
nitrone was prepared and treated with AuCl3 [Eq. (3)]. In this
case, desired 33 was isolated in 43% yield, proving that the
conjugation between the nitrone and the alkyne is not
29 and 31) could also be rationalized as a result of an internal
redox sequence shown in Scheme 2. In this case, the initial 7-
endo-dig attack of the O atom of the nitrone on the alkyne
ꢀ
and subsequent N O cleavage will lead to a-oxo Au-
carbenoid F. Subsequent addition of the imine to this
carbenoid leads to G, from which the catalyst is regenerated
to provide 3.
required for the internal redox; this led us to favor path B
as a possible IR mechanism for the formation of azomethine
ylide J/J’.[14]
Additionally, we reasoned that the subsequent dipolar
cycloaddition probably occurs at the stage of the Au-bound
J/J’ based on some observations: failure of 27 to undergo a
redox reaction in the absence of DEAD under AuCl3 catalysis
to form a free stabilized azomethine ylide suggests that the
catalyst turnover occurs only after the cycloaddition of Au-
bound azomethine ylide J/J’. In addition, the current dipolar
cycloaddition occurs at lower temperatures relative to those
of related azomethine ylides (even at RT, Table 1, entry 3),[2]
presumably because of the favorably modulated frontier
molecular orbital (FMO) of the dipole for a facile cyclo-
addition.
In summary, we have described herein a novel gold-
catalyzed generation of an azomethine ylide, featuring an
internal redox reaction between a tethered nitrone and an
alkyne under electrophilic metal catalysis. The azomethine
ylide that is formed undergoes an efficient cycloaddition
cascade in a highly diastereoselective manner. The benefit of
atom economy (100%) as well as environmental safety is
apparent from this approach. Efforts are currently directed
toward an asymmetric version of this IR-DC cascade.
Scheme 2. Proposed mechanism for the formation of isoindole 3.
Interestingly, the use of a bulky and relatively electron-
rich AuI catalyst produced 3 selectively (Table 1, entry 7 and
Eq. (2)), for reasons that are not clear at this stage. On the
other hand, an AuIII complex completely diverts the reaction
Experimental Section
Representative procedure for an IR-DC cascade reaction: AuCl3
(0.5 mg, 0.0018 mmol, 2 mol%; for a lower catalyst loading, 0.02m
stock solutions in respective solvents were used) was added to a
solution of alkyne/nitrone 1 (40 mg, 0.089 mmol) in nitromethane
(0.8 mL). The resulting mixture was heated to 708C for an hour. (At
this point the mixture turned blue-green and the TLC indicated
complete conversion. However, there was no appreciable amount of
precipitate or metallic gold.) The solvent was removed under reduced
pressure and the residue was purified by silica gel chromatography
(EtOAc/hexanes = 1:2) to afford 32.8 mg (82%) of 2 as pale yellow
crystals.
Scheme 3. Proposed mechanism for the IR-DC cascade reaction.
pathway to 6-exo-dig process as delineated in Scheme 3.
ꢀ
Detailed mechanistic analysis of the N O bond cleavage
Received: June 13, 2008
(H/H’!I) in this redox cascade led to the identification of two
plausible pathways: 1) The 6-exo-dig attack of the nitrone on
the Au-activated alkyne would generate H, which has a
possible resonance structure depicted as H’. A retro-electro-
cyclization would then lead to a-carbonyl Au-carbenoid I
(path A) and subsequent carbene addition to J (or its O-
bound tautomer J’). 2) Alternatively, the same species (I)
could be generated without undergoing dearomatization
(path B). To differentiate between these possibilities, sub-
Published online: August 1, 2008
Keywords: azomethine ylides · carbenes · cyclization ·
.
homogeneous catalysis · redox chemistry
[1] For recent reviews on the cycloaddition of azomethine ylides:
a) I. Coldham, R. Hufton, Chem. Rev. 2005, 105, 2765 – 2809;
b) C. Nµjera, J. M. Sansano, Angew. Chem. 2005, 117, 6428 –
7042 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 7040 –7043

Angewandte
Chemie
6432; Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276; c) C. Nµjera,
J. M. Sansano, Curr. Org. Chem. 2003, 7, 1105 – 1150.
[2] For an elegant route to hetisine alkaloids: a) K. M. Peese, D. Y.
Gin, J. Am. Chem. Soc. 2006, 128, 8734 – 8735; b) K. M. Peese,
D. Y. Gin, Chem. Eur. J. 2008, 14, 1654; For a review: c) V. Nair,
T. D. Suja, Tetrahedron 2007, 63, 12247 – 12275.
[3] For reviews: a) H. M. L. Davies, R. E. J. Beckwith, Chem. Rev.
2003, 103, 2861 – 2903; b) M. P. Doyle, A. G. H. Wee, Modern
Catalytic Methods for Organic Synthesis with Diazo Compounds,
Wiley, New York, 1998.
[4] For the formation of azomethine ylides from the addition of an
imine to a Rh–carbenoid: a) C. V. Galliford, K. A. Scheidt, J.
Org. Chem. 2007, 72, 1811 – 1813, and references therein; For
nonmetal carbenes derived from CF2Br2: b) M. S. Novikov, A. F.
Khlebnikov, O. V. Besedina, R. R. Kostikov, Tetrahedron Lett.
2001, 42, 533 – 535.
Iwasawa, Org. Lett. 2006, 8, 289 – 292; c) H. Kusama, J. Takaya,
N. Iwasawa, J. Am. Chem. Soc. 2002, 124, 11592 – 11593; d) N.
Iwasawa, M. Shindo, H. Kusama, J. Am. Chem. Soc. 2001, 123,
5814 – 5815; e) S. Shin, A. K. Gupta, C. Yun, C. H. Oh, Chem.
Commun. 2005, 4429 – 4431.
[9] Dipolar cycloadditions between the nitrone and the olefin in an
intra- or intermolecular fashion was not observed. A. AdØ, E.
Cerrada, M. Contel, M. Laguna, P. Merino, T. Tejero, J.
Organomet. Chem. 2004, 689, 1788 – 1795.
[10] See the Supporting Information (SI) for details. CCDC 680836
(2) contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.
uk/data_request/cif.
[11] See Table S1 in the Supporting Information for full details of
catalyst screening.
ꢀ
[5] For a metal-catalyzed cleavage of the N O bond: a) H. S. Yeom,
[12] As suggested by the reviewers, the reaction was also run with
1.0 equivof AuCl 3 in an attempt to detect the azomethine ylide
intermediate. In this case, however, immediate decomposition of
the starting material (< 1 min at RT as observed by TLC) and
very messy NMR spectra resulted. Addition of dipolarophile
(DEAD) to this mixture did not produce any observable amount
of 28.
[13] The structure of isoindole 3 was assigned based on the similarity
of spectral characteristics to those of 31 reported in the literature
(see the Supporting Information for details) and also on the
synthesis of the known compound 31 by internal redox [Eq. (2)].
D. SolØ, L. Vallverdffl, X. Solans, M. Font-Bardꢀa, J. Bonjoch, J.
Am. Chem. Soc. 2003, 125, 1587 – 1594.
[14] A possible [1,3]-sigmatropic rearrangement at the N atom of H
to give I was excluded because of poor orbital overlap. For a
formal [1,3]-sigmatropic rearrangement of related vinyl–Au or
vinyl–Pt species: I. Nakamura, U. Yamagishi, D. Song, S. Konta,
Y. Yamamoto, Angew. Chem. 2007, 119, 2334 – 2337; Angew.
Chem. Int. Ed. 2007, 46, 2284 – 2287, and references therein.
E. S. Lee, S. Shin, Synlett 2007, 2292 – 2294; b) B. M. Trost, Y. H.
Rhee, J. Am. Chem. Soc. 2002, 124, 2528 – 2533.
[6] For pioneering works on gold-catalyzed intramolecular redox
reactions between alkynes and tethered sulfoxides: a) N. D.
Shapiro, F. D. Toste, J. Am. Chem. Soc. 2007, 129, 4160 – 4161;
b) G. Li, L. Zhang, Angew. Chem. 2007, 119, 5248 – 5251; Angew.
Chem. Int. Ed. 2007, 46, 5156 – 5159; For gold-catalyzed diazo-
transfer reactions: c) C. A. Witham, P. Mauleón, N. D. Shapiro,
B. D. Sherry, F. D. Toste, J. Am. Chem. Soc. 2007, 129, 5838 –
5839.
[7] For recent reviews on gold catalysis: a) D. J. Gorin, F. D. Toste,
Nature 2007, 446, 395 – 403; b) A. S. K. Hashmi, Chem. Rev.
2007, 107, 3180 – 3211; c) A. Fürstner, P. W. Davies, Angew.
Chem. 2007, 119, 3478 – 3519; Angew. Chem. Int. Ed. 2007, 46,
3410 – 3449.
[8] For other approaches of formal generation of metal carbenoids
for cycloadditions: a) S. Su, J. A. Porco, J. Am. Chem. Soc. 2007,
129, 7744 – 7745; b) H. Kusama, Y. Miyashita, J. Takaya, N.
Angew. Chem. Int. Ed. 2008, 47, 7040 –7043
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7043