Geometry-dependent divergence in the gold-catalyzed redox cascade cyclization of o-alkynylaryl ketoximes and nitrones leading to isoindoles

Article abstract of DOI:10.1039/b910757f

We report geometry-dependent cyclizations of o-alkynylaryl ketoximes and nitrones catalyzed by gold complexes. (E)-Ketoximes undergo N-attack to give isoquinoline-N-oxides. In sharp contrast, (Z)-ketoximes undergo unprecedented O-nucleophilic attack, foll

Full text of DOI:10.1039/b910757f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Geometry-dependent divergence in the gold-catalyzed redox cascade
cyclization of o-alkynylaryl ketoximes and nitrones leading to isoindoles†
Hyun-Suk Yeom,a Youngun Lee,a Ji-Eun Leeb and Seunghoon Shin*a
Received 2nd June 2009, Accepted 23rd July 2009
First published as an Advance Article on the web 1st September 2009
DOI: 10.1039/b910757f
We report geometry-dependent cyclizations of o-alkynylaryl ketoximes and nitrones catalyzed by gold
complexes. (E)-Ketoximes undergo N-attack to give isoquinoline-N-oxides. In sharp contrast,
(Z)-ketoximes undergo unprecedented O-nucleophilic attack, followed by a redox cascade leading to a
novel catalytic entry to isoindoles of diverse scope. The structure of an isoindole was unambiguously
supported by X-ray crystallography. We demonstrated the generality of the isoindole synthesis from
either (Z)-oximes or nitrones, and presented a mechanistic model of this redox cascade based on the
reaction profiles of various substrates.
reported a series of interesting cyclization reactions of closely
1. Introduction
related enynyl and enallenyl nitrone substrates.6
Oximes and nitrones often function as stable precursors of amino
Despite the relative scarcity of precedents, there are several
compounds through the reductive cleavage of the N–O bond.1
desirable characteristics in the use of oximes and nitrones as
Besides, the cleavage of the N–O bond mediates a variety of
oxidants: (1) they are readily available, hydrolytically stable, non-
redox processes, including conversions of oximes into amides,2a–c
toxic and non-explosive; (2) their redox chemistry can often be
nitriles2d,e and enamides,2f N-substitutions of O-acyl (or sulfonyl)
performed under a mild reaction condition with the aid of metal
oximes,2g–i and [3,3]-sigmatropic rearrangements of oximes and
catalysis, and furthermore, (3) the imine, formed as a putative
nitrones.2j,k Redox reactions involving nitrones have scarcely
intermediate of a redox reaction of the oxime or nitrone, can
been precedented, particularly when compared to their structural
be re-incorporated into the reaction in a cascade fashion, thus
isomers, oxaziridines,3 presumably due to the thermodynamic
stability of the former. A conspicuous exception in this perspective
is the Kinugasa reaction between a nitrone and a terminal
acetylene, which was proposed to proceed through an initial
dipolar cycloaddition of a nitrone onto a Cu-acetylide (Scheme 1).4
Another relevant formal redox process is the rearrangement
of a 2,3-dihydroisoxazole, formed from a dipolar cycloaddition
of a nitrone with an alkyne, into an acyl aziridine reported
by Takahashi and Kano,5a and Huisgen et al.,5b of which the
mechanism was clarified by Baldwin and Pudussery via isolation
of acyl aziridine intermediates.5c In addition, Eberbach et al. have
rendering the overall process completely atom-economical.7 In
this regard, we recently reported a gold-catalyzed generation
of the azomethine ylide from the intramolecular redox reaction
between alkynes and nitrones based on a concept that combines
the exceptional carbophilic Lewis acidity of gold complexes8 and
the potential ability of a nitrone as an oxidant (Scheme 2).9
More recently, Nakamura et al. reported a remarkable cascade
cyclization through N–O bond cleavage of oximes catalyzed by
a Cu(I) complex, leading to b-lactams.10 To further showcase its
generality, and to extend the mechanistic insights and synthetic
implications of this gold-catalyzed N–O cleaving redox cascade,
we wish to describe an intramolecular redox cyclization of
o-alkynylaryl ketoximes and nitrones, leading to the general
synthesis of isoindoles (Scheme 3). Isoindoles can be potential
precursors of porphyrin analogs or pyrroles with extended con-
jugation and therefore, should find important applications in
material science.11 However, they are rather unstable and their
preparatory methods are still limited, especially in a catalytic
fashion.11h–j Here, we disclose that the mechanism of this Au-
catalyzed redox is strictly geometry-dependent and thus, both
(Z)-oximes as well as nitrones readily undergo redox cyclization
into isoindoles, as demonstrated by the synthesis of a variety of
isoindoles under relatively mild conditions.
Scheme 1 Intramolecular redox reactions involving a nitrone and alkynes.
aDepartment of Chemistry, Hanyang University, 133-791, Seoul, Korea.
E-mail: sshin@han yang.ac.kr; Fax: +82-2-2299-0762; Tel: +82-2-2220-
0948
bDepartment of Chemistry, Gyeongsang National University, Chinju, 660-
701, Korea
† Electronic supplementary information (ESI) available: 1H and 13C NMR
spectra of starting materials and products, X-ray crystallographic data of
3e, as well as Table S1 and S2. CCDC 738725. For ESI and crystallographic
data in CIF or other electronic format, see DOI: 10.1039/b910757f
Scheme 2 Intramolecular redox–dipolar cycloaddition (IR-DC) cascade.
4744 | Org. Biomol. Chem., 2009, 7, 4744–4752
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Attempted in situ isomerization–cyclization of (E)-1a
Entry Additivea
Conditionsb,c Conv. (%)d 2a (%)d
1
2
3
4
5
6
7
8
9
None
70 ◦C, 1 h
rt, 1.5 h
>95
>95
36
83
TfOHe
TfOH (10%)e,f
TfOH (2 equiv.)e
Cu(OTf)·C6H6
Cu(OTf)2
70 ◦C, 1 h
rt, 2 h
20 (10 : 1)g 15
>95
>95
>95
Dec.h
Dec.h
60
94
63
63
—
—
52
Scheme 3 Isoindole synthesis via redox cyclization of oximes and
nitrones.
70 ◦C, 0.5 h
70 ◦C, 1.5 h
Cu(OTf)·C6H6 (1 equiv.) rt, 1.5 h
2. Results and discussion
Cu(OTf)2 (1 equiv.)
rt, 1.5 h
n-Bu4NOHi
70 ◦C, 1.5 h
2.1. Geometry-dependent reaction profiles and attempted in situ
E–Z isomerization
a 5 mol% of additives, unless otherwise noted. b Reactions were conducted
at 70 ◦C in a sealed vial. c The reaction was quenched by addition of Et3N.
d The conversion and the yield of 2a were based on the crude NMR. e A
stock solution (1.0 M in CH2Cl2) was used. f In the absence of Au catalyst.
g E : Z ratio of the recovered 1a in parenthesis. h Decomposed. i 3.0 M
solution in MeOH.
In our previous studies on the cyclization of o-alkynylaryl
aldoximes catalyzed by electrophilic metal complexes,12 we have
shown that they undergo an efficient cyclization into isoquinoline-
N-oxides with either Au(I) or Ag(I) catalysts via 6-endo-dig N-
attack of the (E)-oxime. We projected that the absence of O-attack
that we observed in the redox reaction of a nitrone (Scheme 2)
is due to the geometrical constraints of the (E)-oxime. Therefore,
our initial efforts were focused on the cyclization of ketoximes, in
which (Z)-geometry is more available than that of aldoxime. As
expected, when the carefully separated (E)-1a (formed as a major
isomer of E : Z = 10 : 1 mixture) was treated with Au(IMes)OTf
(5%) in CH2Cl2 (70 ◦C, sealed vial), 36% of isoquinoline-N-oxide
2a was isolated as the only identifiable product (Eqn 1). With
5 mol% of TfOH as a co-catalyst, the yield further increased to
83% (vide infra, Table 1). In sharp contrast, (Z)-1a transformed
into isoindole 3a in 66% yield under identical conditions (Eqn 2).
In both cases, the formation of 3a from (E)-1a (or 2a from the
reaction of (Z)-1a) could not be observed even, in trace amount.
dramatically improved yield of 2a (83%) under a much milder
condition (rt) was observed when the reaction was conducted with
5 mol% of TfOH as a co-catalyst along with 5 mol% of Au(I)
catalyst (entries 1 and 2). This is presumably due to the facilitation
of the catalyst turnover by the acid additive.13 In the presence
of 10% of TfOH alone without Au(I) catalyst, the cyclization
into 2a did not occur effectively (15%), but the recovered starting
(E)-1a isomerized into its equilibrium ratio (E : Z = 10 : 1) (entry
3). However, even with 2 equiv. of TfOH and Au(I) catalyst, the
formation of 3a was not observed (entry 4), indicating a much
=
higher rotational barrier around C N bond compared to that
of cyclization into isoquinoline-N-oxide 2a.14 Other Lewis acids
(entries 5–8) or base (entry 9) similarly failed to isomerize the
C N bond in situ.
=
2.2. Optimization and scope of the isoindole formation from
(Z)-ketoximes
(1)
We screened various metals as catalysts for the optimization of
isoindole formation from (Z)-1e (Table 2). At rt, little or no
conversion was observed with any of the catalysts we examined.
However, upon heating to 70 ◦C, various metals transformed
(Z)-ketoximes into isoindole 3e with varying efficiencies (entries
1–8). Among those, Au(IMes)OTf was the most effective, pro-
viding 3e in 73% yield (entry 3). It is noteworthy that AuCl3
gave only a modest yield of 3e from a messy mixture (entry
2). The structural assignment of 3e was supported by X-ray
crystallography (Fig. 1). Further efforts were made to optimize
the yield of 3e: (1) by changing the gold counter-anion (entries
9–12), (2) by changing the solvents (Table S1, ESI†), or (3) by
changing the ligand (Table S2, ESI†). Somewhat inferior yields
were obtained from these trials, but Au(IMes)SbF6, Au(IPr)OTf,
and [t-Bu2P(o-biphenyl)Au]OTf in CH2Cl2 or Au(IMes)OTf in
CH3NO2 gave comparably good yields. Importantly, there was
no isoquinoline-N-oxide observed in these trials, even in a trace
amount. The use of TfOH (5%) co-catalyst along with Au(I), also
(2)
We initially hoped to find a suitable condition that induces
facile in situ E–Z isomerization of ketoximes under the reaction
conditions, so that the cyclization of the (E)-oxime might give the
corresponding isoindole. Thus, various additives (5 mol%), includ-
ing Brønsted acid (TfOH), base (n-Bu4NOH), Cu(OTf)·C6H6 and
Cu(OTf)2, were examined along with Au(IMes)OTf (5 mol%) as a
catalyst in CH2Cl2 (Table 1). In all trials, the formation of isoindole
3a from (E)-1a was not observed (Table 1). Notably, however, a
=
did not produce any isoquinoline-N-oxide 2e, confirming the C N
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4744–4752 | 4745
©

View Article Online
Table 2 Screening of metal catalysts for the conversion of (Z)-1e into 3e
Table 3 Cyclization of (E)- and (Z)-ketoximesa
Entry
Oxime
R1, R2
Time/h
2/3/4 (%)b
Entry
Catalyst
Conv. (%)a
3e (%)a
1
2
3
4
5
6
7
8
9
10
11
12
13
(E)-1a
(Z)-1a
(E)-1b
(Z)-1b
(E)-1c
(Z)-1c
(E)-1d
(Z)-1d
(Z)-1e
(Z)-1f
(Z)-1g
(Z)-1h
(Z)-1i
Me, Me
Me, n-Bu
Me, Ph
n-Pr, Me
2
1
2
0.5
2
2a, 83
3a, 66
1
2
3
4
5
6
7
8
9
10
11
12
13
AuCl
AuCl3
[Au(IMes)]OTf
PtCl2
In(OTf)3
Zn(OTf)2
Cu(OTf)2
AgOTf
[Au(IMes)]NTf2
[Au(IMes)]ClO4
[Au(IMes)]BF4
[Au(IMes)]SbF6
[Au(IMes)]OTf/TfOHc
17
75
>95
18
18
8
67
46
>95
>95
>95
>95
>95
15
44
73
8
11
8
2b, 91
3b, 42
2c, 46
12
1
3c, 30/4c, 41
2d, 94
1.5c
1
3d, 49
Dec.b
32
65
65
65
69
59
Ph, n-Bu
Ph, Me
Ph, 1-cyclohexenyl
Me, 1-cyclohexenyl
Ph, Ph
3e, 62
0.5
5
3f, 65
3g, 63
1.5c
10c
3h, 73
3i, 36/4i, 41
a Conditions: (for (E)-oximes): TfOH (5%) and Au(IMes)OTf (5%) in
CH2Cl2 at rt; (for (Z)-oximes): Au(IMes)OTf (5%) in CH2Cl2 at 70 ◦C
(sealed vessel). b Isolated yields of 2–4 after chromatography. c 10 mol% of
catalyst was used.
a Yields were based on the crude NMR spectra. b Decomposed. c 5 mol%
of TfOH was used as an additive.
This represents an interesting example of in situ isomerization of
(E)-oximes before it can cyclize by O-nucleophilic attack.
(3)
2.3. Optimization and scope of the nitrone cyclization
One apparent drawback in the isoindole synthesis from
(Z)-ketoximes is the low availability of the (Z)-ketoxime, which
often forms as a minor component in the equilibrium, especially
when R1 is an alkyl group. From the chemo-divergence in Table 1
and 2, a nitrone would be a logical starting material, because the
(Z)-isomer is the exclusive form in nitrones. Thus, we screened a
variety of reaction conditions for the cyclization of nitrones into
isoindoles (Table 4). The efforts to find the optimized condition
was less successful than in the case of (Z)-oximes. Although
reasonable conversions and rates were achieved with some metal
salts (entries 3, 4, 6 and 8), the reaction was typically plagued
by extensive decomposition, and only a low yield of isoindole
6d was obtained in dichloromethane as the solvent. The facile
decomposition of the isoindoles of type 6, as compared to type
3, is not surprising, because the absence of a substituent at the
3-position of the isoindole would render 6 very susceptible to
decomposition via undesired conjugate additions. Upon changing
solvents into nitromethane, however, up to 58% of isoindole
6d was obtained with [t-Bu2P(o-biphenyl)Au]OTf catalyst. Al-
though the yield of nitrone 6d was modest, the overall yield
from the carbonyl precursor significantly improved compared to
(Z)-oximes, and isoindoles without the C3-substituent could be
readily synthesized.
Fig. 1 A view of one of the four independent molecules of the isoindole
in the crystal structure of isoindole 3e. Ellipsoids are drawn at the 30%
probability level.
rotational barrier is much higher than the cyclization to isoindole
3e (entry 13). When the reaction was conducted at 150 ◦C without
any catalyst as a control experiment, no reaction took place and an
extensive decomposition occurred over 24 h, supporting that this
redox cyclization is a metal-catalyzed process and not a thermal
isomerization of 2,3-dihydroisoxazoles (vide infra in Scheme 5).5,6b
We next examined the generality of this geometry-dependent
divergence with the following examples in Table 3. The ketoximes
used in this study were carefully separated by chromatography
and were separately subjected to the reaction conditions shown
in Table 3. In all cases examined, (E)-oximes gave isoquinoline-
N-oxides 2, whereas (Z)-oximes gave isoindoles 3 uneventfully
(entries 1–8). When the alkynyl terminus is substituted with an
alkenyl group, good yields of the corresponding isoindoles were
obtained (entries 11 and 12). Interestingly, (Z)-oximes 1c and 1h
with a phenyl acetylene unit (entries 6 and 13), produced yet more
products, which were assigned as 4-hydroxyisoquinolines 4c and
4i, respectively.15 The structural assignment of these products was
also supported by the following related transformation of (E)-1j
bearing a (Z)-acrylate unit (Eqn 3). The obtained product 4j is
consistent with the structure shown, based on the absence of an
ethoxy group, non-polar behaviour in TLC (Rf = 0.65, EtOAc–
Hex = 1 : 1 on silica gel), and IR, 1H, 13C and COSY-NMR spectra.
We next investigated the generality of this cyclization of nitrones
5 into isoindoles 6, using substrates with different substituents
at the aryl core and alkynyl terminus, and as N-substituents
(Table 5).16 Substrates with an N-alkyl group, and an alkyl group
4746 | Org. Biomol. Chem., 2009, 7, 4744–4752
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 5 The scope of the isoindole formation from the nitronea
Table 4 Screening of metal catalyst for the cyclization of nitrone 5da
Conditions
Time
Entry Substrate
1
6 (%)b
5a A, 70 ◦
C
6a 55%
6b 53%
6c 57%
6d 52%
6e 43%
Entry Catalyst
Solvent
Time
Conv. (%) Yield (%)b
1 h
1
2
3
Zn(OTf)2
Cu(OTf)2
PtCl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
3 h
3 h
1 h
1 h
3 h
1.5 h
3 h
0.5 h
3 h
0
52
—
11
32
28
16
29
13
24
28
58
48
42
38
2
3
4
5
5b A, 70 ◦
C
>95
>95
60
1 h
4
5
6
7
AuCl3
Au(PPh3)OTf
[Au(L)] OTfc
Au(IMes)OTf
Au(IPr)OTf
>95
54
5c B, rt
0.3 h
8
>95
87
9
[P(C6F5)3Au]OTf CH2Cl2
10
11
12
13
[Au(L)] OTfc
Au(IPr)OTf
PtCl2
CH3NO2 0.5 h
CH3NO2 0.5 h
CH3NO2 2 h
CH3NO2 0.5 h
>95
>95
74
5d B, 70 ◦
C
C
0.5 h
AuCl3
>95
a Conditions: catalyst (5%) at 70 ◦C (sealed vessel), unless otherwise noted.
b Crude NMR yield. c L = t-Bu2P(o-biphenyl).
5e B, 70 ◦
0.5 h
at the alkynyl position, generally underwent a smooth reaction
(entries 1–4). When substrates with different electron demands at
the aryl core were tested, trifluoromethyl-substituted 5e smoothly
cyclized, albeit in a modest yield (entry 5). However, more electron-
rich 5f underwent a sluggish reaction and under forced conditions
(70 ◦C, 3 h) decomposed slowly into an intractable mixture (entry
6). Generally, the reaction of substrates with an aryl group at
the alkynyl terminus was difficult due to a facile decomposition
of the isoindole products. For example, in entries 7 and 8, we
managed to isolate the corresponding isoindoles 6g and 6h in low
yields, but these materials decomposed in hours upon storage at
6
7
5f B, 70 ◦
C
C
Dec.
3 h
5g B, 70 ◦
6g 32%
6h 30%
6i 35%
6j 40%
6k 55%
3 h
8
9
5h B, rt
0.6 h
5i A, 70 ◦
C
C
4
◦C. Interestingly, pyridinyl substrates 5i and 5j underwent a
0.3 h
smooth cyclization leading to 6H-pyrrolo[3,4-b]pyridine deriva-
tives (entries 9 and 10). Substrate 5k bearing an alkenyl tether
between the nitrone and the alkyne functionality also smoothly
participated in the cyclization to afford the corresponding pyrrole
derivative (entry 11).
10
11
5j A, 70 ◦
4 h
5k B, 70 ◦
C
2.4. Mechanistic scenarios
0.5 h
The possible mechanism for the current redox cyclization into
isoindoles from (Z)-ketoximes can be envisioned as in Scheme 4,
following a corollary of the N–O bond redox in the generation of
a Unless otherwise mentioned, the following conditions were used.
Condition A: Au(IPr)OTf (5%) in CH3NO2; condition B: [t-Bu2P(o-
biphenyl)Au]OTf (5%) in CH3NO2. b Isolated yields after chromatography.
azomethine ylides from nitrones.9a A notable feature in this mecha-
nism is the divergence into isoindole 3 and 4-hydroxyisoquinoline 4
for (Z)-oximes bearing a phenyl substituent at the alkynyl position
(R2). For (Z)-1c and (Z)-1i (R2 = Ph), a mixture of 3 and 4 were
obtained (Table 3, entries 6 and 13), while all substrates with R2 =
alkyl gave single isomeric products 3. This observation can be
rationalized based on the following. Intermediate C, formed via
the faster 6-exo-dig cyclization and the following redox, represents
a Au-carbenoid, which could be envisioned as a gold-stabilized
carbenium ion.17,18 Therefore, the presence of a phenyl group
at R2 would stabilize C to afford 4-hydroxyisoquinoline 4 as a
Scheme 4 Possible mechanism of isoindole formation: 6-exo-dig or
7-endo-dig cyclization followed by N–O bond redox.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4744–4752 | 4747
©

View Article Online
mixture with 3. In the absence of such a cation-stabilizing group,
the slower 7-endo-dig cyclization and the following redox become
the dominant pathway, leading to isoindole 3. A similar 7-endo
cyclization mechanism was also invoked by Hashmi et al.19a in
the intramolecular oxygen transfer reaction between epoxides and
alkynes.19
4. Experimental
General details
Anhydrous solvents (dichloromethane and nitromethane) were
distilled from calcium hydride. All commercially available reagents
were used as received without further purification. Ketoximes were
prepared either by stirring a mixture of a ketone, NH2OH·HCl (2
equiv.) and pyridine (5 equiv.) in ethanol at room temperature, or
by treating a mixture of ketone and NH2OH·HCl (1.5 equiv.) with
NaOAc (2 equiv.) in CH2Cl2. A similar procedure was used for
the preparation of nitrones except, in this case, that BnNH2OH
or MeNH2OH·HCl was used instead of hydroxylamine salt. Flash
column chromatography was performed with 230-400 mesh silica
gel. 1H and 13C NMR spectra were recorded on a 400 MHz
spectrometer with TMS as an internal standard.
A similar mechanism could be envisioned for the isoindole
formation from nitrones. However, given the complexity of the
N–O bond redox processes,4–6 one of the alternative mechanisms
for the reaction of nitrone–alkyne substrates could be envisioned
as that in Scheme 5. Initial intramolecular [3 + 2] dipolar
cycloaddition between the nitrone and the alkyne, followed by
retro-[2 + 2] electrocyclization, would provide a strained allene
H,6 which then rearrange into acyl aziridine I.5 Instead, direct
thermal rearrangement of G into acyl aziridine I could also
be considered and the following retro-[2 + 2] electrocyclization
would then furnish the desired isoindole 6. However, we disfavor
this alternative mechanism, because they involve rather strained
intermediates G–I. In addition, a control experiment for the
conversion of nitrone 5d in the absence of any catalyst in CH3NO2
Characterization of substrates
(E)-1a. dH (CDCl3, 400 MHz): 9.60 (1H, broad s), 7.60-7.40
(1H, m), 7.40-7.30 (3H, m), 2.32 (3H, s), 2.05 (3H, s); dC (CDCl3,
100 MHz): 158.3, 140.4, 133.7, 129.1, 128.8, 128.3, 123.1, 91.2,
79.0, 16.0, 5.2.
◦
at 150 C led to gradual decomposition of the starting material,
with no noticeable isoindole formation. In sharp contrast, the
Au(I)-catalyzed conversion of nitrone 5h into 6h occurs even at
room temperature, with a significant conversion (Table 5, entry 8).
Thus, a thermal rearrangement of 4-isoxazoline G, in this case,
is a less likely option. Therefore, the isoindole formation from
nitrones most likely follows a gold-catalyzed redox pathway that
was depicted in Scheme 4 for (Z)-oximes.
(Z)-1a. dH (CDCl3, 400 MHz): 8.23 (1H, broad s), 7.57-7.40
(1H, H), 7.40-7.22 (2H, m), 7.22-7.12 (1H, m), 2.20 (3H, s), 2.04
(3H, s); dC (CDCl3, 100 MHz): 157.1, 138.7, 133.2, 128.9, 128.2,
127.5, 121.9, 90.2, 78.2, 21.8, 5.2.
(E)-1b. dH (CDCl3, 400 MHz): 9.68 (1H, broad s), 7.49-7.39
(1H, m), 7.37-7.30 (1H, m), 7.30-7.22 (2H, m), 2.41 (2H, t, J 7.3),
2.32 (3H, s), 1.58 (2H, quintet, J 7.3), 1.45 (2H, sextet, J 7.7), 0.93
(3H, t, J 7.3); dC (CDCl3, 100 MHz): 158.3, 140.4, 133.7, 129.0,
128.7, 128.2, 123.2, 95.8, 79.8, 31.2, 22.7, 19.9, 16.0, 14.2.
(Z)-1b. dH (CDCl3, 400 MHz): 7.54-7.40 (1H, m), 7.38-7.24
(2H, m), 7.22-7.09 (1H, m), 2.40 (2H, t, J 7.0), 2.21 (3H, s), 1.57
(2H, quintet, J 7.0), 1.46 (2H, sextet, J 7.3), 0.93 (3H, t, J 7.0); dC
(CDCl3, 100 MHz): 157.2, 138.8, 133.1, 128.9, 128.1, 127.4, 122.0,
94.8, 79.0, 31.3, 22.6, 21.8, 19.8, 14.3.
Scheme 5 Alternative mechanism: initial [3 + 2] cycloaddition and
(E)-1c. dH (CDCl3, 400 MHz): 9.40 (1H, broad s), 7.68-
7.54(1H, m), 7.54-7.44 (2H, m), 7.44-7.10 (6H, m), 2.40 (3H, s);
dC (CDCl3, 100 MHz): 158.2, 140.6, 133.7, 132.2, 129.2, 129.12,
129.08, 129.01, 128.96, 123.8, 122.4, 94.4, 88.7, 16.1.
thermal rearrangements.
3. Conclusions
(Z)-1c. dH (CDCl3, 400 MHz): 8.20 (1H, broad s), 7.71-7.55
(1H, m), 7.55-7.45 (2H, m),7.45-7.35 (2H, m), 7.28-7.35 (3H, m),
7.17-7.35 (1H, m), 2.25 (3H, s); dC (CDCl3, 100 MHz): 156.7,
139.0, 133.0, 132.3, 129.1, 129.0, 128.9, 127.7, 123.7, 121.2, 93.5,
88.0, 21.9.
Herein, we have demonstrated a novel isoindole synthesis starting
from (Z)-ketoximes. The key mechanistic feature of this redox
cyclization is that the mechanistic divergence into isoquinoline-
N-oxides or isoquinoline is strictly dictated by the geometry
of the starting ketoxime. A reaction profile of a range of
(Z)-ketoximes and nitrones provides further mechanistic support
for the proposed N–O bond redox mechanism.21 In addition to
the formation of isoindoles, we have observed the formation of
4-hydroxyisoquinoline derivatives (4c, 4i and 4j), which also fits
into the currently proposed mechanistic model and reveals another
controlling element in the redox reaction of nitrones. Currently,
work aimed at trapping and utilizing the imine intermediate in a
more synthetically useful fashion is under way in our laboratory,
which should provide more direct evidence of the involvement of
a-oxo carbenoids and imines as reactive species.
(E)-1d. dH (CDCl3, 400 MHz): 9.70 (1H, broad s), 7.54-7.32
(1H, m), 7.34-7.16 (3H, m), 2.85 (2H, t, J 7.7), 2.04 (3H, s), 1.52
(2H, sextet, J 7.4), 0.93 (3H, t, J 7.4); dC (CDCl3, 100 MHz): 161.8,
139.6, 133.6, 129.8, 128.9, 128.2, 123.4, 90.8, 78.9, 30.9, 19.8, 14.9,
5.2.
(Z)-1d. dH (CDCl3, 400 MHz): 8.40 (1H, broad s), 7.46 (1H,
d, J 7.7), 7.39-7.20 (2H, m), 7.13 (1H, d, J 7.0), 2.53 (2H, t, J
7.7), 2.03 (3H, s), 1.49 (2H, sextet, J 7.4), 0.94 (3H, t, J 7.4); dC
(CDCl3, 100 MHz): 160.2, 138.0, 133.1, 128.8, 128.0, 122.3, 90.1,
78.7, 37.7, 20.2, 14.4, 5.2.
4748 | Org. Biomol. Chem., 2009, 7, 4744–4752
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
(Z)-1e. dH (CDCl3, 400 MHz): 9.00 (1H, s), 7.62-7.48 (1H,
m), 7.48-7.42 (2H, m), 7.42-7.36 (2H, m), 7.36-7.28 (4H, m), 2.18
(2H, t, J 7.0), 1.46-1.04 (4H, m), 0.76 (3H, t, J 7.3); dC (CDCl3,
100 MHz): 157.8, 136.2, 135.5, 132.5, 129.5, 128.8, 128.7, 128.4,
127.6, 127.2, 123.3, 94.8, 78.7, 30.5, 21.8, 19.2, 13.7.
131.8, 130.5, 130.3, 129.8, 129.7, 129.5, 129.4, 129.0, 128.3, 123.2,
123.1, 96.0, 87.0, 72.5.
5h.
(CDCl3, 400 MHz): 9.33 (1H, d, J 11.7), 8.05 (1H, s),
H
7.54-7.49 (3H, m), 7.41-7.32 (7H, m), 6.91 (2H, J d, 8.8), 5.12 (2H,
s), 3.87 (3H, s); dC (CDCl3, 100 MHz,): 160.6, 133.7, 133.1, 132.7,
131.6, 130.5, 130.2, 130.0, 129.1, 128.2, 123.5, 115.3, 114.7, 96.1,
85.8, 72.4, 56.0.
(Z)-1f. dH (CDCl3, 400 MHz): 8.14 (1H, broad s), 7.58-7.50
(1H, m), 7.50-7.43 (2H, m), 7.42-7.36 (2H, m), 7.36-7.28 (3H, m),
7.28-7.22 (1H, m), 1.85 (3H, s); dC (CDCl3, 100 MHz,): 158.3,
136.7, 136.1, 133.0, 130.0, 129.2, 129.1, 128.9, 128.1, 127.7, 123.8,
90.6, 78.3, 4.9.
5i. dH (CDCl3, 400 MHz): 9.54 (1H, d, J 8.0), 8.46 (1H, d, J
4.4), 7.97 (1H, s), 7.55-7.47 (2H, m), 7.47-7.36 (3H, m), 7.22 (1H,
dd, J 4.8, 8.1), 5.10 (2H, s), 2.38 (2H, t, J 6.6), 1.56 (2H, quint,
J 7.4), 1.46 (2H, sextet, J 7.7), 0.95 (3H, t, J 7.3); dC (CDCl3,
100 MHz): 150.5, 142.7, 134.6, 133.1, 131.3, 130.0, 129.5, 129.4,
128.3, 122.8, 97.4, 78.3, 72.2, 30.6, 22.5, 19.4, 14.0.
(Z)-1g. dH (CDCl3, 400 MHz): 8.80 (1H, broad s), 7.62-7.52
(1H, m), 7.52-7.44 (2H, m), 7.44-7.38 (2H, m). 7.36-7.26 (4H,
m), 5.91 (1H, m), 2.16-1.98 (2H, m), 1.98-1.82 (2H, m), 1.64-1.35
(4H, m); dC (CDCl3, 100 MHz,): 158.2, 136.5, 136.1, 136.0, 132.7,
130.0, 129.4, 129.3, 129.0, 128.2, 127.8, 123.6, 121.2, 95.9, 85.5,
29.3, 26.3, 22.8, 22.1.
5j. dH (CDCl3, 400 MHz): 9.61 (1H, dd, J 1.1, 8.4), 8.55 (1H,
dd, J 1.5, 4.8), 8.03 (1H, s), 7.52-7.47 (2H, m), 7.47-7.41 (3H,
m), 7.41-7.34 (5H, m), 7.30 (1H, dd, J 8.1, 4.8), 5.13 (2H, s); dC
(CDCl3, 100 MHz): 150.9, 142.3, 135.0, 133.1, 132.6, 131.1, 130.3,
130.1, 129.9, 129.8, 129.0, 128.8, 123.6, 122.1, 95.3, 86.5, 72.7.
(Z)-1h. dH (CDCl3, 400 MHz): 7.91 (1H, broad s), 7.60-7.40
(1H, m), 7.40-7.26 (2H, m), 7.24-7.12 (1H, m), 6.18 (1H, m), 2.21
(3H, s), 2.32-2.17 (2H, m), 2.17-2.04 (2H, m), 1.78-1.50 (4H, m);
dC (CDCl3, 100 MHz,): 156.5, 138.3, 135.8, 132.3, 128.4, 127.9,
127.1, 121.3, 120.9, 95.0, 84.9, 29.2, 26.0, 22.5, 21.7, 21.3.
5k. dH (CDCl3, 400 MHz): 8.32-8.10 (1H, m), 7.46-7.30 (5H,
m), 4.85 (2H, s), 2.85 (2H, m), 2.28 (2H, t, J 6.6), 2.22 (2H, s), 1.57
(4H, t, J 2.9), 1.44-1.30 (4H, m), 0.92 (3H, t, J 7.3); dC (CDCl3,
100 MHz): 136.1, 134.1, 133.9, 130.0, 129.5, 129.3, 127.0, 99.3,
80.4, 71.6, 32.3, 31.4, 27.2, 22.6, 22.2, 19.9, 14.2.
(Z)-1i. dH (CDCl3, 400 MHz): 8.20 (1H, broad s), 7.76-7.60
(1H, m), 7.58-7.50 (2H, m), 7.50-7.41 (2H, m), 7.41-7.28 (4H, m),
7.28-7.18 (5H, m); dC (CDCl3, 100 MHz): 158.2, 136.8, 136.0,
132.9, 132.2, 130.1, 129.6, 129.4, 129.1, 128.9, 128.82, 128.78,
127.8, 123.6, 123.0, 93.9, 88.1.
Representative cyclization of (E)-ketoximes into isoquinoline-N-
oxides. To solution of (E)-1a (40 mg, 0.231 mmol) in CH2Cl2
(2.0 mL) was added Au(IMes)Cl (6.2 mg, 0.0115 mmol), AgOTf
(3.0 mg, 0.0115 mmol) and TfOH (0.12 mL of 0.10 M solution
in CH2Cl2, 0.0115 mmol), and the mixture was stirred for 1 h at
rt. The reaction was terminated by addition of Et3N (5 drops),
the mixture was concentrated under vacuum, and the residue was
purified by silica gel chromatography (MeOH–CH2Cl2 = 1 : 30)
to give 33.2 mg (83%) of 2a as a pale brown solid (Rf = 0.4,
MeOH–CH2Cl2 = 1 : 10).
Representative cyclization of (Z)-ketoximes into isoindoles. To
solution of (Z)-1a (40 mg, 0.231 mmol) in CH2Cl2 (2.0 mL) was
added Au(IMes)Cl (6.2 mg, 0.0115 mmol) and AgOTf (3.0 mg,
0.0115 mmol). The resulting mixture was heated at 70 ◦C for 1 h,
then cooled to rt and the reaction was quenched by the addition
of Et3N (5 drops). The mixture was concentrated under vacuum
and the residue was purified by silica gel chromatography (EtOAc–
Hex = 1 : 10) to give 26.4 mg (66%) of 3a as white solid (Rf = 0.3,
EtOAc–Hex = 1 : 4).
(E-oxime, Z-alkene) 1j. dH (CDCl3, 400 MHz): 8.86 (1H, s),
7.97 (1H, broad s), 7.92-7.81 (1H, broad s), 7.63-7.48 (1H, m),
7.41-7.30 (2H, m), 6.39 (1H, d, J 11.4), 6.18 (1H, d, J 11.3), 4.31
(2H, q, J 7.0), 1.33 (3H, t, J 7.0); dC (CDCl3, 100 MHz): 165.4,
149.5, 134.6, 133.6, 130.1, 129.4, 125.7, 122.99, 122.96, 98.4, 92.0,
61.3, 30.3, 14.9.
5c. dH (CDCl3, 400 MHz): 9.25 (1H, d, J 7.7), 7.94 (1H, s),
7.42 (1H, d, J 7.3), 7.36-7.24 (2H, m), 3.90 (3H, s), 2.11 (3H, s); dC
(CDCl3, 100 MHz): 134.0, 132.9, 131.7, 130.4, 128.6, 127.9, 123.8,
92.8, 77.8, 55.6, 5.2.
5d. dH (CDCl3, 400 MHz): 9.26 (1H, d, J 7.4), 7.98 (1H, s),
7.54-7.46 (2H, m), 7.46-7.38 (4H, m), 7.37-7.26 (2H, m), 5.10 (2H,
s), 2.36 (2H, t, J 7.0), 1.62-1.36 (4H, m), 0.97 (3H, t, J 7.4); dC
(CDCl3, 100 MHz): 133.9, 133.4, 132.6, 131.6, 130.4, 130.1, 129.5,
128.5, 128.0, 124.1, 97.4, 72.2, 31.2, 22.6, 19.7, 14.2.
Representative cyclization of nitrones into isoindoles. To solu-
tion of 5d (40 mg, 0.137 mmol) in CH3NO2 (1.4 mL) was added
[t-Bu2P(o-biphenyl)Au]Cl (3.6 mg, 0.00685 mmol) and AgOTf
(1.8 mg, 0.00685 mmol). The resulting mixture was heated at
70 ◦C for 30 min, then cooled to rt, and was quenched by the
addition of Et3N (5 drops). The mixture was concentrated under
vacuum and the residue was purified by silica gel chromatography
(EtOAc–Hex = 1 : 4) to afford 20.8 mg (52%) of 6d as a white
solid.
5e. dH (CDCl3, 400 MHz): 9.64 (1H, s), 8.00 (1H, s), 7.58-7.46
(4H, m), 7.46-7.34 (3H, m), 5.10 (2H, s), 2.39 (2H, t, J 7.0), 1.62-
1.30 (4H, m), 0.97 (3H, t, J 7.0); dC (CDCl3, 100 MHz): 133.4,
133.0, 132.2, 132.1, 130.4 (q, JC-F 32.7), 130.2, 129.8, 129.7, 127.3,
126.7 (q, JC–F 3.8), 124.8 (q, JC–F 3.8), 124.3 (q, JC–F 266), 100.3,
77.7, 72.6, 31.3, 22.7, 19.8, 14.2.
5f. dH (CDCl3, 400 MHz): 8.98 (1H, s), 8.21 (1H, s), 7.97 (1H,
s), 7.48-7.57 (3H, m), 7.39-7.28 (4H, m), 6.94 (1H, s), 6.03 (2H, s),
5.08 (2H, s)
Characterization of products
5g. dH (CDCl3, 400 MHz): 9.38 (1H, d, J 7.7), 8.04 (1H, s),
7.55 (1H, dd, 7.4, 0.7), 7.52 (2H, d, J 23.9), 7.42-7.33 (10H, m),
5.13(2H, s); dC (CDCl3, 100 MHz): 133.6, 133.0, 132.9, 132.2,
2a. Pale brown solid (mp 61-62 ◦C);20a nmax (film)/cm-1: 3052,
2975, 2923, 1736 w, 1556 m, 1503 s, 1333 s, 1227 s; dH (CDCl3,
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4744–4752 | 4749
©

View Article Online
400 MHz): 7.91 (1H, d, J 8.0), 7.70 (1H, d, J 8.2), 7.64-7.44 (3H,
m), 2.92 (3H, s), 2.67 (3H, s); dC (CDCl3, 100 MHz): 146.2, 146.1,
129.4, 128.7, 128.5, 128.3, 127.2, 124.5, 121.5, 19.1, 14.1; HRMS
(EI+) Calc. (M+) 173.0841 found 173.0839.
dC (CDCl3, 100 MHz): 149.6, 146.1, 129.4, 128.5, 128.4, 128.1,
127.2, 124.4, 121.6, 29.4, 20.6, 19.1, 15.2; HRMS (EI+) Calc. (M+)
201.1154 found 201.1155.
3d. Brown crystal (mp 135-136 ◦C); nmax (film)/cm-1: 3399,
3174, 3048, 2956, 2926, 2866, 1724 m, 1597 s, 1503 s, 1429 m, 1354
w, 1289 w, 1265 w, 1219 w; dH (CDCl3, 400 MHz): 11.48 (1H, broad
s), 7.85 (1H, d, J 8.4), 7.69 (1H, d, J 8.5), 7.35 (1H, t, J 7.0), 7.10
(1H, t, J 8.1), 3.05 (2H, t, J 7.3), 2.69 (3H, s), 1.82 (2H, sextet,
J 7.4), 0.99 (3H, t, J 7.3); dC (CDCl3, 100 MHz): 184.7, 140.6,
136.1, 135.4, 130.2, 128.9, 127.5, 125.5, 122.9, 121.9, 121.7, 120.3,
28.9, 28.1, 23.6, 14.6 (additional peaks were observed, presumably
due to the intramolecular hydrogen bonding); HRMS (EI+) Calc.
(M+) 205.1154 found 205.1154.
3a. Brown solid (mp 199-202 ◦C); nmax (film)/cm-1: 3151, 3045,
2920, 1733 w, 1599 s, 1458 s, 1326 m, 1290 m, 1233 m; dH (CDCl3,
400 MHz): 11.70 (1H, broad s), 7.84 (1H, d, J 8.4), 7.67 (1H, d, J
8.4), 7.36 (1H, t, J 7.3), 7.11 (1H, t, J 7.3), 2.70 (3H, s), 2.69 (3H,
s); dC (CDCl3, 100 MHz): 184.6, 131.6, 129.0, 127.6, 125.9, 121.9,
121.8, 121.0, 120.3, 28.1, 12.1; HRMS (EI+) Calc. (M+) 173.0841
found 173.0840.
2b. Pale brown solid (mp 77-79 ◦C); nmax (film)/cm-1: 3057,
2958, 2933, 2869, 1731 w, 1558 m, 1503 m, 1354 s, 1231 s, 1210 s;
dH (CDCl3, 400 MHz): 7.90 (1H, d, J 8.1), 7.71 (1H, d, J 8.3), 7.64-
7.50 (2H, m), 7.48 (1H, s), 3.06 (2H, t, J 7.4), 2.91 (3H, s), 1.80
(2H, quintet, J 7.3), 1.50 (2H, sextet, J 7.3), 1.00 (3H, t, J 7.3); dC
(CDCl3, 100 MHz): 149.7, 146.1, 129.3, 128.6, 128.4, 128.1, 127.4,
124.4, 120.4, 31.4, 29.8, 23.3, 14.6, 14.1; HRMS (EI+) Calc. (M+)
215.1310 found 215.1310.
3e. Yellow crystal (mp 125-127 ◦C); nmax (film)/cm-1: 3259,
2958, 2930, 2871, 1703 s, 1612 m, 1453 s, 1410 w, 1346 w, 1315 w,
1194 w; dH (CDCl3, 400 MHz): 11.00 (1H, broad s), 8.00 (1H, d, J
8.4), 7.91 (1H, d, J 8.4), 7.77 (2H, d, J 7.7), 7.53 (2H, t, J 7.3), 7.43
(1H, t, J 7.4), 7.40 (1H, t, J 7.3), 7.20 (1H, t, J 7.7), 3.05 (2H, t,
J 7.7), 1.81 (2H, quintet, J 7.7), 1.50 (2H, sextet, J 7.4), 0.99 (3H,
t, J 7.4); dC (CDCl3, 100 MHz): 189.4, 131.6, 130.0, 129.2, 128.2,
127.4, 124.7, 123.4, 122.5, 120.7, 40.6, 27.7, 23.4, 14.7; HRMS
(EI+) Calc. (M+) 277.1467 found 277.1466.
3b. Dark brown solid (mp 133-135 ◦C); nmax (film)/cm-1: 3182,
3062, 2949, 2927, 2868, 1706 m, 1587 s, 1596 m, 1460 s, 1227 m; dH
(CDCl3, 400 MHz): 11.85 (1H, broad s), 7.83 (1H, d, J 8.4), 7.67
(1H, d, J 8.4), 7.35 (1H, t, J 7.7), 7.10 (1H, t, J 7.7), 3.02 (2H, t,
J 7.3), 2.72 (3H, s), 1.85 (2H, quintet, J 7.3), 1.51 (2H, sextet, J
7.3), 0.99 (3H, t, J 7.3); dC (CDCl3, 100 MHz): 188.1, 131.5, 128.4,
127.5, 125.9, 121.9, 121.6, 120.8, 120.5, 40.0, 28.0, 23.5, 14.7, 12.7;
HRMS (EI+) Calc. (M+) 215.1310 found 215.1311.
3f. Yellow solid (mp 61-63 ◦C);20d max (film)/cm-1: 3196, 3056,
n
2921, 2852, 1703, 1609 s, 1504 w, 1459 s, 1348 m, 1232 s; dH (CDCl3,
400 MHz): 11.15 (1H, broad, s), 8.0 (1H, d, J 8.4), 7.93 (1H, d,
J 8.4), 7.78 (2H, d, J 8.0), 7.53 (2H, t, J 7.4 Hz), 7.48-7.36 (2H,
m), 7.21 (1H, t, J 7.7). 2.74 (3H, s); dC (CDCl3, 100 MHz): 185.3,
132.1, 131.6, 129.9, 129.3, 128.9, 128.3, 127.5, 124.7, 123.4, 122.8,
122.5, 120.6, 28.4; HRMS HRMS (EI+) Calc. (M+) 235.0997 found
235.0996.
2c. White solid (mp 143-144 ◦C); nmax (film)/cm-1: 3406 m br,
3055, 2923, 2853, 1732 w, 1595 w, 1499 s, 1352 s, 1399 s, 1290 m,
1222 s; dH (CDCl3, 400 MHz): 7.94 (1H, d, J 8.4), 7.86-7.72 (3H,
m), 7.67 (1H, s), 7.62 (1H, t, J 7.4), 7.55 (1H, t, J 7.4), 7.53-7.38
(3H, m), 2.97 (3H, s); dC (CDCl3, 100 MHz): 147.4, 146.3, 134.4,
130.5, 129.7, 129.35, 129.27, 129.0, 128.8, 128.7, 128.0, 124.5,
123.2, 14.3; HRMS (EI+) Calc. (M+) 235.0997 found 235.0996;
3g. Yellow solid (mp 139-142 ◦C); nmax (film)/cm-1: 3177, 3028,
2970, 2929, 2856, 1738 s, 1561 s, 1532 s, 1445 s, 1365 m, 1227 s;
dH (CDCl3, 400 MHz): 11.28 (1H, broad s), 7.97 (1H, d, J 8.4),
7.92 (1H, d, J 8.8), 7.79 (2H, d, J 7.3), 7.51 (2H, t, J 7.3), 7.41
(1H, t, J 7.3), 7.30 (1H, t, J 7.7), 7.17 (1H, t, J 7.7), 6.57 (1H,
m), 2.62-2.42 (2H, m), 2.42-2.24 (2H, m), 1.96-1.50 (4H, m); dC
(CDCl3, 100 MHz): 186.7, 140.1, 136.3, 132.4, 131.8, 129.8, 129.1,
128.4, 126.9, 124.9, 123.5, 122.0, 121.9, 121.5, 26.4, 25.4, 23.0,
22.6; HRMS (EI+) Calc. (M+) 301.1467 found 301.1467;
3c. Brown crystal (mp 189-190 ◦C); nmax (film)/cm-1: 3180,
3150, 3054, 2921, 2851, 1721 w, 1621 m, 1566 s, 1531 m, 1457 s,
1437 m, 1330 s, 1234 s; dH (CDCl3, 400 MHz): 7.90-7.76 (2H, m),
7.65 (1H, d, J 8.11), 7.63-7.46 (3H, m), 7.15 (1H, t, J 7.5), 7.11-7.04
(2H, m), 6.99 (1H, s), 2.78 (3H, s); dC (CDCl3, 100 MHz): 183.3,
141.1, 140.5, 135.4, 131.6, 130.2, 129.2, 129.1, 127.5, 126.2, 122.8,
122.3, 121.5, 120.9, 120.5, 18.4 (additional peaks were observed,
presumably due to intramolecular hydrogen bonding); HRMS
(EI+) Calc. (M+) 235.0997 found 235.0998.
3h. Dark brown solid (mp 137-140 ◦C); nmax (film)/cm-1: 3182,
3150, 3052, 2929, 2856, 1713 w, 1557 s, 1532 m, 1446 s, 1374 m,
1326 m, 1228 m; dH (CDCl3, 400 MHz): 11.67 (1H, broad s), 7.83
(1H, d, J 8.8), 7.63 (1H, d, J 8.0), 7.26 (1H, t, J 8.2), 7.08 (1H, t, J
7.7), 6.49 (1H, m), 2.70 (3H, s), 2.62-2.44 (2H, m), 2.40-2.24 (2H,
m), 1.94-1.72 (4H, m); dC (CDCl3, 100 MHz): 185.7, 140.5, 139.9,
135.4, 135.0, 132.0, 130.2, 129.8, 127.1, 126.1, 122.8, 121.8, 121.4,
121.2, 120.0, 26.3, 25.6, 23.1, 22.7, 18.4 (additional peaks were
observed, presumably due to hydrogen bonding); HRMS (EI+)
Calc. (M+) 239.1310 found 239.1310.
4c. Yellow oil;20b nmax (film)/cm-1: 3409 (3780~2800, s br),
3048, 2921, 2854, 1665 m, 1604 m, 1418 s, 1398 s, 1289 s, 1189 s;
dH (CDCl3, 400 MHz): 8.26 (1H, d, J 8.4), 8.07 (1H, d, J 8.5), 7.74
(2H, d, J 7.7), 7.71 (1H, t, J 7.7), 7.62 (1H, t, J 7.3), 7.54 (2H,
t, J 7.3), 7.42 (1H, t, J 7.3), 5.78 (1H, broad s), 2.92 (3H, s); dC
(CDCl3, 100 MHz): 151.0, 143.1, 137.9, 135.1, 130.2, 130.0, 129.6,
129.0, 128.6, 128.4, 127.8, 126.4, 122.7, 22.6; HRMS (EI+) Calc.
(M+) 235.0997 found 235.0998.
3i. Yellow solid (mp 143-145 ◦C); nmax (film)/cm-1: 3195, 2923,
2851, 1731 m, 1586 m, 1455 s, 1303, 1235; dH (CDCl3, 400 MHz):
11.60 (1H, broad s), 8.08-7.93 (1H, m), 7.93-7.74 (4H, m), 7.67-
7.57 (1H, m), 7.57-7.48 (4H, m), 7.48-7.40(1H, m), 7.40-7.30(1H,
m), 7.24-7.10(2H, m); dC (CDCl3, 100 MHz): 184.5, 140.8, 133.8,
131.9, 131.5, 130.2, 130.0, 129.4, 129.1, 128.4, 127.3, 125.1, 123.8,
2d. Brown oil; nmax (film)/cm-1: 3054, 2961, 2929, 2871, 1543
m, 1500 s, 1461 m, 1356 s, 1324 s, 1235 s; dH (CDCl3, 400 MHz):
7.89 (1H, d, J 8.4), 7.69 (1H, d, J 7.3), 7.63-7.44 (3H, m), 3.41 (2H,
t, J 7.7), 2.66 (3H, s), 1.81 (2H, sextet, J 7.4), 1.12 (3H, t, J 7.4);
4750 | Org. Biomol. Chem., 2009, 7, 4744–4752
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
122.3, 122.2, 121.3; HRMS (EI+) Calc. (M+) 297.1155 found
297.1157.
J 4.0, 1.4), 7.94 (1H, dd, J 8.0, 1.4), 7.37 (1H, s), 7.34-7.20 (3H,
m), 7.16 (2H, d, J 6.6), 7.03 (1H, dd, J 8.5, 4.1), 5.98 (2H, s), 3.50
(2H, t, J 7.3), 1.71 (2H, quint, J 7.4), 1.39 (2H, sextet, J 7.7), 0.92
(3H, t, J 7.4 Hz); dC (CDCl3, 100 MHz): 192.8, 150.9, 144.9, 137.6,
129.5, 129.4, 128.6, 128.3, 121.4, 121.2, 118.3, 117.5, 55.6, 42.0,
28.1, 23.2, 14.8.
4i. Yellow oil; nmax (film)/cm-1: 3391(3700-2800, broad s),
3048, 2909, 2843, 1695 m, 1598 w, 1518 m, 1395 s, 1304 s; dH
(CDCl3, 400 MHz): 8.33 (1H, d, J 8.5), 8.07 (1H, d, J 8.5), 7.82
(2H, d, J 7.3), 7.78-7.66 (3H, m), 7.64-7.36 (7H, m), 5.90 (1H,
broad s); dC (CDCl3, 100 MHz): 153.6, 143.7, 140.4, 137.8, 135.7,
130.8, 130.2, 129.9, 129.7, 129.2, 128.9, 128.8, 127.9, 127.8, 122.5;
HRMS (EI+) Calcd for (M+) 297.1154 found 297.1155.
6j.
n
max (film)/cm-1: 3059, 3030, 2853, 1724, 1611 s, 1575, 1449,
1431, 1329, 1284, 1219, 1169; dH (CDCl3, 400 MHz): 8. 47 (1H,
dd, J 4.0, 1.4, Hz), 7.92 (1H, dd, J 8.4, 1.5 Hz), 7.88 (2H, broad d,
J 6.9), 7.55 (1H, t, J 7.3), 7.48 (1H, s), 7.43 (2H, t, J 7.7), 7.35-7.26
(3H, m), 7.26-7.22 (2H, m), 6.96 (1H, dd, J 8.5, 4.4), 5.96 (2H,
s); dC (CDCl3, 100 MHz): 186.6, 151.2, 144.2, 140.5, 137.6, 132.7,
131.1, 129.5, 129.3, 128.8, 128.4, 121.8, 121.6, 118.5, 117.3, 55.1.
4j.
n
max (film)/cm-1: 3062, 2923, 2854, 1721 s, 1602, 1564, 1488,
1369, 1329, 1283, 1250; dH (CDCl3, 400 MHz): 9.14 (1H, s), 8.52
(1H, d, J 8.1), 8.11 (1H, d, J 9.6), 8.07 (1H, d, J 8.5), 7.91 (1H,
t, J 7.3), 7.81 (1H, t, J 7.3), 6.66 (1H, d, J 9.5); dC (CDCl3,
100 MHz): 160.7, 150.1, 146.6, 132.2, 130.6, 130.5, 130.3, 130.1,
128.3, 126.3, 122.1, 118.5; HRMS (EI+) Calcd for (M+) 197.0477
found 197.0476.
6k. Colorless oil; nmax (film)/cm-1: 3087, 3063, 2932, 2860,
1708 s, 1638 m, 1410 s, 1348 w, 1312 w; dH (CDCl3, 400 MHz):
7.42-7.16 (3H, m), 7.08 (2H, d, J 7.0), 6.63 (1H, s), 5.50 (2H, s).
2.83 (2H, t, J 5.8 Hz, 2H), 2.63 (2H, t, J 7.4), 2.52 (2H, t, J 5.),
1.94-1.78 (2H, m), 1.78-1.68 (2H, m), 1.68-1.48 (2H, m), 1.29 (2H,
sextet, J 7.3), 0.88 (3H, t, J 7.3); dC (CDCl3, 100 MHz): 192.2,
139.7, 129.8, 129.1, 128.1, 127.7, 127.59, 127.56, 121.0, 53.6, 42.2,
27.3, 26.2, 24.5, 23.6, 23.2, 22.6, 14.7; HRMS (EI+) Calcd for (M+)
295.1936, found 295.1934.
6c.
n
max (film)/cm-1: 3093, 3056, 3000, 2948, 1769, 1714 s, 1627,
1416, 1335, 1216, 1171; dH (CDCl3, 400 MHz): 7.87 (1H, d, J 8.8),
7.64 (1H, d, J 8.4), 7.39 (1H, s), 7.31 (1H, app t, J 7.3), 7.12 (1H,
app t, J 7.4 z), 4.27 (3H, s), 2.72 (3H, s); dC (CDCl3, 100 MHz):
187.2, 129.9, 126.6, 125.4, 124.7, 122.4, 121.5, 120.5, 40.9, 31.1;
HRMS (EI+) Calcd for (M+) 173.0841 found 173.0842.
6d. White crystal (mp 90-92 ◦C); nmax (film)/cm-1: 3064, 3033,
2958, 2932, 2872, 1716, 1614, 1394 cm-1; dH (CDCl3, 400 MHz):
7.87 (1H, d, J 8.8), 7.64 (1H, d, J 8.5), 7.46 (1H, s), 7.40-7.19
(4H, m), 7.19-7.02 (3H, m), 5.94 (2H, s), 2.02 (2H, t, J 7.3), 1.74
(2H, quintet, J 7.7 Hz), 1.29 (2H, sextet, J 7.3 Hz), 0.94 (3H, t, J
7.3 Hz); dC (CDCl3, 100 MHz): 190.5, 138.1, 135.0, 129.5, 128.3,
128.0, 126.7, 124.9, 124.8, 122.5, 121.8, 121.3, 120.9, 55.5, 42.5,
27.5, 23.3, 14.7; HRMS (EI+) Calcd for (M+) 291.1623, found
291.1623.
Acknowledgements
The authors thank the support by the Korean Research Foun-
dation (KRF-2008-314-C00213), and Korea Science and Engi-
neering Foundation (KOSEF, R01-2006-11283). HSY and YL
thank BK21 program for support. We gratefully thank Prof. Shim
Sung Lee (GNU) for X-ray crystallographic analysis and Prof.
Y. H. Rhee for helpful discussions.
References
6e. White crystal (mp 81-83 ◦C); nmax (film)/cm-1: 3112, 3066,
2959, 2933, 2872, 1719 s, 1637 s, 1547 m, 1496 s, 1448 s, 1334 s,
1263 s, 1172 s; dH (CDCl3, 400 MHz): 7.98 (1H,s), 7.96 (1H, s),
7.58 (1H, s) 7.45 (1H, d, J 9.2), 7.28-7.20 (4H, m,), 7.12 (1H, d,
J 6.6), 5.96 (2H, s), 3.02 (2H, t, J 7.3 Hz), 1.74 (2H, quintet, J
7.3 Hz), 1.39 (2H, sextet, J 7.4), 0.94 (3H, t, J 7.3); dc (CDCl3,
100 MHz): 190.8, 137.4, 129.6, 129.5, 128.6, 128.1, 125.7, 125.2
1 (a) D. H. R. Barton, W. B. Motherwell, E. S. Simon and S. Z. Zard,
J. Chem. Soc., Perkin Trans. 1, 1986, 2243; (b) P. Ilankumaran and S.
Chandrasekaran, Tetrahedron Lett., 1995, 27, 4881.
2 (a) H. Fujiwara, Y. Ogasawara, K. Yamaguchi and N. Mizuno, Angew.
Chem., Int. Ed., 2007, 46, 5202; (b) C. Ramalingan and Y.-T. Park,
J. Org. Chem., 2007, 72, 4536; (c) N. A. Owston, A. J. Parker and M. J.
Williams, Org. Lett., 2007, 9, 3599; (d) S. H. Yang and S. Chang, Org.
Lett., 2001, 3, 4209; (e) K. Yamaguchi, H. Fujiwara, Y. Ogasawara, M.
Kotani and N. Mizuno, Angew. Chem., Int. Ed., 2007, 46, 3922; (f) H.
Zhao, C. P. Wandenbossche, S. G. Koenig, S. P. Singh and R. P. Bakale,
Org. Lett., 2008, 10, 505; (g) C. M. Counceller, C. C. Eichman, B. C.
Wray and J. P. Stambuli, Org. Lett., 2008, 10, 1021; (h) S. Liu, Y. Yu and
L. S. Liebeskind, Org. Lett., 2007, 9, 1947; (i) C. G. Savarin, C. Grise´,
J. A. Murry, R. A. Reamer and D. L. Hughes, Org. Lett., 2007, 9, 981;
(j) N. Takeda, O. Miyata and T. Naito, Eur. J. Org. Chem., 2007, 1491;
(k) R. M. Coates and C. H. Cummins, J. Org. Chem., 1986, 51, 1383.
3 (a) D. J. Michaelis, C. J. Shaffer and T. P. Yoon, J. Am. Chem. Soc.,
2007, 129, 1866; (b) K. M. Partridge, M. E. Anzovino and T. P. Yoon,
J. Am. Chem. Soc., 2008, 130, 2920; (c) J. Aube´, X. Peng, Y. Wang and
F. Takusagawa, J. Am. Chem. Soc., 1992, 114, 5466.
4 (a) M. Kinugasa and S. Hashimoto, J. Chem. Soc., Chem. Commun.,
1972, 466; (b) J. Marco-Contelles, Angew. Chem., Int. Ed., 2004, 43,
2198; for a catalytic version, see: (c) M. Miura, M. Enna, K. Okuro
and M. Nomura, J. Org. Chem., 1995, 60, 4999; for an asymmetric
version, see: (d) M. M.-C. Lo and G. C. Fu, J. Am. Chem. Soc., 2002,
124, 4572.
5 (a) S. Takahashi and H. Kano, J. Org. Chem., 1965, 30, 1118; (b) H.
Seidl, R. Huisgen and R. Knorr, Chem. Ber., 1969, 102, 904; (c) J. E.
Baldwin and R. G. Pudussery, J. Am. Chem. Soc., 1968, 90, 5325;
for a review on the rearrangement of 2,3-dihydroisoxazoles into acyl
aziridines, see: (d) J. P. Freeman, Chem. Rev., 1983, 83, 241.
(q, JC–F = 271), 124.6 (q, JC–F = 32 Hz), 123.2, 122.2 (q, JC–F
=
3.0), 121.9, 121.8, 120.1 (q, JC–F = 5.4), 55.8, 42.6, 27.3, 23.2, 14.7;
HRMS (EI+) Calcd for (M+) 359.1497, found 359.1502.
6g. dH (CDCl3, 400 MHz): 7.69 (1H, d, J 6.9), 7.62-7.52 (2H,
m), 7.45 (2H, t, J 7.7), 7.35-7.22 (7H, m), 7.04-7.94 (2H, m), 6.78
(1H, d, J 8.8), 5.97 (2H, s); 13C NMR spectrum could not be taken
due to its extreme instability (decomposed upon keeping at 4 ◦C
in hours).
6h. Pale yellow oil; dH (CDCl3, 400 MHz): 7.73 (2H, d, J 8.4),
7.58 (1H, d, J 6.6), 7.54 (1H, s), 7.54-7.48 (1H, m), 7.32-7.20
(4H, m), 7.05-7.06 (3H, m), 6.94 (2H, d, J 8.8), 5.92 (2H, s), 3.89
(3H, s); dC (CDCl3, 100 MHz): d 185.0, 162.7, 137.7, 133.9, 133.4,
131.8, 130.3, 128.9, 128.1, 127.7, 125.1, 124.4, 123.7, 122.2, 120.7,
122.1, 120.7, 120.6, 114.5, 113.7, 55.6, 54.0; Very unstable (slowly
decomposed upon keeping at 4 ◦C overnight).
6i.
1548, 1440, 1329, 1277, 1193; dH (CDCl3, 400 MHz): 8.69 (1H, dd,
n
max (film)/cm-1: 3090, 3031, 2955, 2926, 2869, 1730, 1626 s,
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 4744–4752 | 4751
©

View Article Online
6 (a) K. Knobloch and W. Eberbach, Org. Lett., 2000, 2, 1117; (b) W.
Eberbach and J. Roser, Tetrahedron Lett., 1987, 28, 2689; (c) W.
Friebolin and W. Eberbach, Helv. Chim. Acta, 2001, 84, 3822.
7 Toste and Zhang have independently reported the use of sulfoxides as
oxidants for alkynes by Au(I)-catalysts. See: (a) N. D. Shapiro and F. D.
Toste, J. Am. Chem. Soc., 2007, 129, 4160; (b) G. Li and L. Zhang,
Angew. Chem., Int. Ed., 2007, 46, 5156.
12 (a) H.-S. Yeom and S. Shin, Synlett, 2008, 924; for the iodocyclizations
leading to isoquinoline-N-oxides, see: (b) Q. Ding and J. Wu, Adv.
Synth. Catal., 2008, 350, 1850; (c) Z. Huo, H. Tomeba and Y.
Yamamoto, Tetrahedron Lett., 2008, 49, 5531.
13 With stoichiometric amount of electrophilic metal, M-bound interme-
diate (Au-Csp2) was isolated, which proto-demetallates in the presence
of acid, see: (a) L.-P. Liu, M. S. Mashuta and G. B. Hammond, J. Am.
Chem. Soc., 2008, 130, 17642; for examples of the use of acid co-catalyst
in electrophilic metal catalysis, see: (b) S. Bhunia, K.-C. Wang and R.-S.
Liu, Angew. Chem., Int. Ed., 2008, 47, 5063; (c) C.-Y. Zhou, P. W. Hong
and C.-M. Che, Org. Lett., 2006, 8, 325; (d) W. Belting and N. Krause,
Org. Lett., 2006, 8, 4489; (e) L.-Z. Dai, M.-J. Qi, Y.-L. Shi and M. Shi,
Org. Lett., 2007, 9, 3191.
8 For selected recent reviews on gold catalysis, see: (a) A. Arcadi, Chem.
Rev., 2008, 108, 3266; (b) N. T. Patil and Y. Yamamoto, Chem. Rev.,
2008, 108, 3395; (c) D. J. Gorin, B. D. Sherry and F. D. Toste, Chem.
Rev., 2008, 108, 3351; (d) Z. Li, C. Brouwer and C. He, Chem. Rev., 2008,
108, 3239; (e) N. Marion and S. P. Nolan, Chem. Soc. Rev., 2008, 37,
1776; (f) A. S. K. Hashmi and M. Rudolph, Chem. Soc. Rev., 2008, 37,
1766; (g) H. C. Shen, Tetrahedron, 2008, 64, 3885; for a comprehensive
review on the redox aspects of Au-catalysis, see: (h) A. S. K. Hashmi
and G. J. Hutchings, Angew. Chem., Int. Ed., 2006, 45, 7896; for selected
recent reports not covered in these reviews, see: (i) M. Shi, L. Wu and
J.-M. Lu, J. Org. Chem., 2008, 73, 8344; (j) S. Bhunia, K.-C. Wang
and R.-S. Liu, Angew. Chem., Int. Ed., 2008, 47, 5063; (k) B. Baskar,
H. J. Bae, S. E. An, J. Y. Cheong, Y. H. Rhee, A. Duschenk and S. F.
Kirsch, Org. Lett., 2008, 10, 2605; (l) H. J. Bae, B. Basakr, S. E. An,
J. Y. Cheong, D. T. Thangadurai, I.-C. Hwang and Y. H. Rhee, Angew.
Chem., Int. Ed., 2008, 47, 2263; (m) S. G. Sethofer, S. T. Staven, O. Y.
Hung and F. D. Toste, Org. Lett., 2008, 10, 4315; (n) G. Zhang and L.
Zhang, J. Am. Chem. Soc., 2008, 130, 12598; (o) G. Li, X. Huang and
L. Zhang, J. Am. Chem. Soc., 2008, 130, 6944.
9 (a) H.-S. Yeom, J.-E. Lee and S. Shin, Angew. Chem., Int. Ed., 2008, 47,
7040; for an initial observation of N-O cleavage using gold catalysts in
our group, see: (b) H.-S. Yeom, E.-S. Lee and S. Shin, Synlett, 2007,
2292.
10 (a) I. Nakamura, T. Araki and M. Terada, J. Am. Chem. Soc., 2009,
131, 2804. Pt(II)-catalyzed N–O cleavage of alkoxyurea derivatives for
the synthesis of indoloquinazolidinone derivatives was also reported:
(b) I. Nakamura, Y. Sato and M. Terada, J. Am. Chem. Soc., 2009, 131,
4198.
11 For the synthesis and applications of isoindoles and annelated pyrroles
(selected examples): (a) E. Maekawa, Y. Suzuki and S. Sugiyama, Chem.
Ber., 1968, 101, 847; (b) Y. Matsuzawa, K. Ichimura and K. Kudo, Inorg.
Chim. Acta, 1998, 277, 151; (c) T. D. Lash, P. Chandrasekar, A. T.
Osuma, S. T. Chaney and J. D. Spence, J. Org. Chem., 1998, 63, 8455;
(d) T. D. Lash, T. M. Werner, M. L. Thompson and J. M. Manley, J. Org.
Chem., 2001, 66, 3152; (e) S. Duan, D. K. Sinha-Mahapatra and J. W.
Herndon, Org. Lett., 2008, 10, 1541; (f) R. Paolesse, Synlett, 2008, 2215;
(g) B. W.-Q. Hui and S. Chiba, Org. Lett., 2009, 11, 729; for catalytic
syntheses of isoindoles, see: (h) D. Kadzimirsz, D. Hildebrandt, K. Merz
and G. Dyker, Chem. Commun., 2006, 661; (i) Q. Ding, B. Wang and J.
Wu, Tetrahedron Lett., 2007, 48, 8599; (j) Q. Ding, Y. Ye, R. Fan and J.
Wu, J. Org. Chem., 2007, 72, 5439. In our preliminary communications
(ref. 9a), we reported two examples of isoindole formation from the
nitrone.
14 E–Z interconversion of dialkyl ketoximes seems to have much lower
barrier, see: (a) M. Kitamura, M. Yoshida, T. Kikuchi and K. Narasaka,
Synthesis, 2003, 2415; (b) X. Wan, G. Doridot and M. M. Joullie´, Org.
Lett., 2007, 9, 977.
15 See the ESI for the characterization.† 1H NMR of 4c showed some
deviation from the literature, possibly due to the acidic impurity, but
the 13C NMR showed a good match with the reported values; see (for
4c): ref. 20b,c.
16 Two preliminary examples (Table 5, entries 1 and 2) of the isoindole
formation from the nitrone have been reported in ref. 9a.
17 (a) P. De Fre´mont, N. M. Scott, E. D. Stevens and S. P. Nolan,
Organometallics, 2005, 24, 2411and references therein; (b) A. Fu¨rstner
and L. Morency, Angew. Chem., Int. Ed., 2008, 47, 5030; (c) G. Seidel,
R. Mynott and A. Fu¨rstner, Angew. Chem., Int. Ed., 2009, 48, 2510.
18 For C–H insertion of the relevant metal carbenoid, see: (a) A. S. K.
Hashmi, S. Scha¨fer, M. Wo¨lfe, C. D. Gil, P. Fischer, A. Laguna, M. C.
Blanco and M. C. Gimeno, Angew. Chem., Int. Ed., 2007, 46, 6184;
(b) C. H. Oh, J. H. Lee, S. J. Lee, J. I. Kim and C. S. Hong, Angew.
Chem., Int. Ed., 2008, 47, 7505; (c) C. H. Oh, J. H. Lee, S. M. Lee, H. J.
Yi and C. S. Hong, Chem.–Eur. J., 2009, 15, 71.
19 (a) A. S. K. Hashmi, M. Bu¨hrle, R. Salathe´ and J. W. Bats, Adv. Synth.
Catal., 2008, 350, 2059; (b) A. S. K. Hashmi and P. Sinha, Adv. Synth.
Catal., 2004, 346, 432; (c) X.-Z. Shu, X.-Y. Liu, K.-G. Ji, H.-Q. Xiao
and Y.-M. Liang, Chem.–Eur. J., 2008, 14, 5282; (d) K.-G. Ji, Y.-W.
Shen, X.-Z. Shu, H.-Q. Xiao, Y.-J. Bian and Y.-M. Liang, Adv. Synth.
Catal., 2008, 350, 1275.
20 For 2a: (a) K. C. Agrawal, R. J. Cushley, S. R. Lipsky, J. R. Wheaton and
A. C. Sartorelli, J. Med. Chem., 1972, 15, 192; for 4c: (b) H. Hussain,
E. Kianmehr and T. Durst, Tetrahedron Lett., 2001, 42, 2245; (c) T. L.
Gilchrist, G. E. Gymer and C. W. Rees, J. Chem. Soc., Perkin Trans. 1,
1975, 1; for 3f: (d) R. I. Fryer, B. Brust, J. V. Early and L. H. Sternbach,
J. Chem. Soc. C, 1967, 366.
21 Interestingly, Asao et al. have reported AuBr3-catalyzed cyclization of
o-alkynyl nitrobenzene into isatogens and anthranils, where the nitro
group might participate in an internal redox similar to this mechanism.
See: N. Asao, K. Sato and Y. Yamamoto, Tetrahedron Lett., 2003, 44,
5675.
4752 | Org. Biomol. Chem., 2009, 7, 4744–4752
This journal is
The Royal Society of Chemistry 2009
©