New syntheses and ring expansion reactions of cyclobutenimines

Article abstract of DOI:10.1071/CH10318

Two routes are reported for the synthesis of iminocyclobutenones having N-(het)aryl substitution: an addition/substitution sequence starting with cyclobutenediones and an aza-Wittig method. A new synthetic route to N-alkyl derivatives is also presented. This involves O-alkylation of 3-alkylamino-1,2-cyclobutenediones using Meerwein's reagent and subsequent deprotonation under non-hydrolytic conditions. Lithium organyls were found to add to the remaining carbonyl group. The resulting tertiary alcohols undergo ring enlargement on heating in xylene to give 4-aminophenols, 4-amino-1-naphthols, or cyclopenta-annulated quinolines from 4-vinyl, 4-aryl, and 4-alkynyl derivatives, respectively. CSIRO 2010.

Full text of DOI:10.1071/CH10318

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2010, 63, 1656–1664
www.publish.csiro.au/journals/ajc
New Syntheses and Ring Expansion Reactions
of Cyclobutenimines
A C
,
A
B
Ernst Schaumann,
Gerrit Oppermann, Michael Stranberg,
B
and Harold W. Moore
A
Institute of Organic Chemistry, Clausthal University of Technology, Leibnizstrasse 6,
D-38678 Clausthal-Zellerfeld, Germany.
B
Department of Chemistry, University of California, Irvine, CA 92717, USA.
C
Corresponding author. Email: ernst.schaumann@tu-clausthal.de
Two routes are reported for the synthesis of iminocyclobutenones having N-(het)aryl substitution: an addition/substitution
sequence starting with cyclobutenediones and an aza-Wittig method. A new synthetic route to N-alkyl derivatives is also
presented. This involves O-alkylation of 3-alkylamino-1,2-cyclobutenediones using Meerwein’s reagent and subsequent
deprotonation under non-hydrolytic conditions. Lithium organyls were found to add to the remaining carbonyl group. The
resulting tertiary alcohols undergo ring enlargement on heating in xylene to give 4-aminophenols, 4-amino-1-naphthols, or
cyclopenta-annulated quinolines from 4-vinyl, 4-aryl, and 4-alkynyl derivatives, respectively.
Manuscript received: 30 August 2010.
Manuscript accepted: 1 November 2010.
Introduction
compounds.[2–4] Moreover, the cyclobutenone unit is found in
important derivatives such as cyclobutene-1,2-diones 2,[4–7]
semisquaric acid 3,[8] squaric acid 4,[6,9] and its aromatic
squarate 5 dianion (Scheme 1).[9,10]
Cyclobutenones 1, containing an enone unit in a four-membered
ring, combine ring strain with the well known enone reactiv-
ity.[1] Syntheses of these highly substituted derivatives are
available, and these undergo facile electrocyclic ring opening
to give conjugated ketene intermediates. Subsequent ring
closures provide useful synthetic routes to numerous other
Much less is known about heteroanalogues of cyclobute-
nones. Selected examples include monothionation of semisqua-
ric 3 to give the corresponding 4-thioxocyclobut-2-enones.[11]
Also 1,3-dithiolane derivatives of cyclobutenones have been
synthesized,[12,13] which can be considered as protected cyclo-
butenethiones.[14–16] There are also scattered examples of
cyclobutenimine derivatives in the literature. For example,
cyclobuta[e]triazines 6 and cyclobuta[c]quinoxalinones 7
were obtained when cyclobutenediones 2 were treated with
the corresponding nitrogen biselectrophiles such as amidra-
zones[17–19] or 1,2-diaminobenzenes, respectively.[18–21] Semi-
squaric acid derivatives yield cyclobutenone hydrazones 8 upon
treatment with arylhydrazones[19] and cyclobutenone oximes
R1
R2
O
R1
R2
O
O
O
O
O
O
HO
R1
O
O
HO
HO
O
O
R4
R3
1
2
3
4
5
Scheme 1. Cyclobutenone structures.
H
N
H
N
H
R3
Ph
O
Ph
Tos
OH
Ph
Ph
R1
O
N
R3
R1
Cl
R12N
NR24
N
N
Ar
N
NH
Ar
HO
H
O
N
H
Ar
6
7
8
9
10
R1
R2
O
R2O
R12N
O
Ph
R1
Ph
X
NR3
NR3
NR12
R2
11
12
13
Scheme 2. Compounds with a cyclobutenimine unit.
Ó CSIRO 2010
10.1071/CH10318
0004-9425/10/121656

RESEARCH FRONT
New Syntheses and Ring Expansion Reactions of Cyclobutenimines
1657
are available by aza-Wittig monoimination of cyclobuten-
diones.[22] The reaction of squaric or semisquaric acid
derivatives with aromatic amines gives zwitterionic com-
pounds,[9,23,24] e.g., 9[23] (Scheme 2). Cyclobutenimmonium
structures were also obtained by [2þ2] cycloaddition of
ynamines and the related a-chloroenamines to give chlorides
10[25,26] or by alkylation of squaric acid amides to give products
11 (Scheme 2).[9,27] In another approach, cycloprope-
nones[28–31] or their all-carbon analogues, triafulvenes, react
with isonitriles to give cyclobutenimines of type 12 and 13,
respectively (Scheme 2).[32]
Recently, the conversion of 2,2,4,4-tetrachlorocyclobutani-
mines 14 to cyclobutenimine acetals 15 by the action of alkoxide
in a mixed elimination/substitution process was reported
(Scheme 3).[33]
Comprehensive studies of cyclobutenimine chemistry
demanded a more general synthesis since the above methods
suffer a lack of synthetic scope. 4-Iminocyclobutenones were
particularly attractive goals since the remaining carbonyl group
provides a site for subsequent synthetic manipulation. Progress
towards developing general routes to such compounds is now
outlined below.
Synthesis of 4-Iminocyclobutenones
A particularly convenient route to cyclobutenones and cyclo-
butenediones employs squaric acid 4 as the basic starting
material. This is esterified to give squarates 16 via azeotropic
removal of water or by the action of the corresponding ortho-
formate (Scheme 4). For a multigram synthesis of the methyl
ester (16, R1 ¼ Me), the orthoester route is most convenient.[34]
However, for the present work we mainly used diisopropyl
squarate (16, R1 ¼ iPr),[35] which was easily obtained as a
crystalline white solid by the azeotropic esterification method
(Scheme 4). Diisopropyl squarate has advantages over other
analogues. For example, it is less easily hydrolyzed than the
dimethyl derivative and is safer to use since the dimethyl ana-
logue was shown to be a potent skin allergen.[36,37]
OR2
Cl
NR1
Cl
NR1
OR2
4 eq. NaOR2
R2OH
R2O
Ar
Cl
Ar
Cl
14
15
Ph
O
O
Ph
O
Ph3PϭN–R2
Scheme 3. Cyclobutenimines from tetrachlorocyclobutanimines.
ϪPh3PO
68–89%
R1O
R1O
NR2
R1OH
C6H6, reflux
R1OH
18
21
R1O
R1O
O
O
R1O
R1O
O
O
HC(OR1)3
R1 ϭ Me, Et, i Pr
4
reflux
R2 ϭ Ph,
,
N
16
16
N
Scheme 4. Synthesis of dialkyl squarates.
Scheme 7. Aza-Wittig synthesis of 4-iminocyclobutenones.
O
O
H2O
R1O
R2
1. R2M
2. TFAA
R1O
R1O
O
R1O
R1O
O
O
18
O
OCCF3
ϪF3CCOO
R2
17
OR1
O
R1O
R1O
16
R1OH
R2
R1 ϭ alk
R2 ϭ alk, unsat. alk, Ar, SiR3
19
Scheme 5. Squarate modification by an addition/substitution approach.
NHR
i PrO
i PrO
O
O
RN
O
i PrO
i PrO
RNH2
OCOCF3
ϪiPrOH
44–91%
ϪHOOCCF3
Ar
Ar
iPrO
Ar
17
(R1 ϭ i Pr)
20
21
Ar ϭ Ph, 2-MeOPh, 4-MeOPh, 2-pyrrolyl, 2-furyl
R = Ph, (MeO)nC6H4Ϫn, 3-MeC6H4, 4-CF3C6H4,
, 2,4,6-Me3C6H2, tBu
N
Scheme 6. The organyl addition/substitution route to cyclobutenimines.

RESEARCH FRONT
1658
E. Schaumann et al.
The alkoxy groups in 16 can be replaced by alkyl, alkenyl, or
aryl groups by the action of the corresponding lithium organyl.
For example, treatment of the squarate with an organometallic
reagent followed by the addition of trifluoroacetic anhydride
gave the trifluoroacetate 17 which was not isolated but treated
directly with oxygen nucleophiles. Thus, treatment of 17 with
water gave 4-substituted cyclobutenediones (semisquarates) 18
and with alcohols the corresponding acetals 19 were realized
(Scheme 5).[38,39] This method also allowed use of silyl
anions as nucleophiles to give silyl-substituted cyclobutenones
18, 19.[40]
In still another variant, dimethyl squarate 16 (R1 ¼ Me) was
treated with methyl lithium (2.5 equivalents) thus introducing
two methyl groups simultaneously to give 3,4-dimethyl-
cyclobut-3-ene-1,2-dione 2 (R1 ¼ R2 ¼ Me) in excellent yield
(95%).[41]
The efficient reaction of intermediate 17 with oxygen
nucleophiles (Scheme 5) suggested that 17 might be analo-
gously trapped by amines to give iminocyclobutenones.[42] In
fact, trifluoroacetate 17, formed from ester 16 and an (het)aryl
lithium compound (Scheme 5), reacted smoothly with various
anilines or 3-aminopyridine to yield the O,N-acetals 20. These
then undergo spontaneous elimination of 2-propanol to give the
desired compounds 21 in good to excellent yields (Scheme 6).
Cyclobutendiones 19 may be submitted to the above reaction
sequence to give 3,4-disubstituted-1,2-cyclobutenediones 2.[38]
Ph
O
RO
NTMS
21
Ph
O
O
Ph3PϭNSiMe3
C6H6, reflux
RO
18
Ph
O
Ph
O
PPh3
O
SiMe3
N
RO
SiMe3
RO
N
O
PPh3
22
23
Ph
O
From R ϭ Me: 90%
From R ϭ i Pr: 31%
O
N
P
Ph3
24
Scheme 8. Attempted synthesis of N-silylcyclobutenimines.
Ph
O
O
Ph
OMe
O
Ph
O
O
R2NH2
Me3OϩBFϪ4
54–99%
BF4
67–99%
R1O
R2HN
R2HN
18
25
26
Ph
OMe
O
NMe2 NMe2
CH2Cl2
R2N
72–98%
21
R1 ϭ i Pr
R2 ϭ Et, n-C5H11Bn
Scheme 9. Synthesis of 4-(alkylimino)cyclobutenones.
O
OH
OH
R1
R2
O
O
R1
R2
C
R1
R2
R3
R1
R2
R3
R3
⌬
R3
OH
OH
OH
Scheme 10. Hydroquinones by ring expansion of 4-alkenylcyclobutenones.

RESEARCH FRONT
New Syntheses and Ring Expansion Reactions of Cyclobutenimines
1659
The structure assignments of 21 are based upon both spectral and
chemical data. For example, the IR spectra of 21 show imino
stretch vibrations around 1690 cmꢀ1 and carbonyl stretch vibra-
tions at 1750 cmꢀ1. Their 13C NMR spectra reveal resonances
for the imino and carbonyl carbons at approximately d ¼ 165
and 188, respectively. Interestingly, in each case only one set
of NMR signals was observed even though inversion at the
imino nitrogen should be slow on the NMR time scale.[43] This
indicates that only one diastereomer is formed.
N-(Het)arylcyclobutenimines 21 are reasonably stable and
can be purified by chromatographic methods. Such stability is
consistent with previous reports showing that imine hydrolysis
of N-arylcyclobutenimines required heating with 2 M hydro-
chloric acid.[33] Attempts to expand the scope of the method for
the synthesis of N-alkyl-iminocyclobutenones resulted in the
isolation of the corresponding cyclobutenediones 2. Apparently,
unlike their N-(het)aryl counterparts, N-alkyl derivatives readily
hydrolyze during workup to give 2. The only exception was the
NR
NR
Ph
NR
O
Ph
Ph
Ph
Ph
RNC
NR
R2
O
O
R ϭ 2,6-Me2C6H3
ring opening of
Scheme
11. Light-induced
electrocyclic
cyclobutenimines.
R
R
OH
Ph
R
PhN
O
O
PhN
Dil HCl, THF
58–81%
Li
OH
46–82%
Ph
i PrO
i PrO
Ph
i PrO
21
27
28
O
O
O
Li
HO
R ϭ H, Me
75%
Ph
i PrO
Ph
i PrO
18
29
Scheme 12. Reaction of 2-phenyl-3-isopropoxy-4-iminocyclobutenones and of 3-phenyl-4-isopropoxy-1,2-cyclobutendione with 1-alkenyl lithium
compounds.[42]
NHR1
NPh
NR1
Ag2O
KHCO3, MeCN
NR1
Xylene
reflux
i PrO
iPrO
i PrO
i PrO
R2
42–88%
Ph
R2
Ph
R2
Ph
R2
Ph
OH
OH
OH
O
30
27
31
32
R1 ϭ C5H11, Ar
R2 ϭ H, Me
Scheme 13. Ring expansion of 4-alkenylcyclobutenimines.[42]
1. ArLi
2. H2O
or
Xylene
reflux
i PrO
i PrO
NPh
O
NPh
NPh
MeOTf
iPrO
Ph
Ph
OR2
Ph
R1
R1
OR2
34
21
33
(R2
ϭ
Ph)
R1
NHPh
PhHN
R1 ϭ H (81%), Me, OMe
R2 ϭ H, Me
i PrO
i PrO
ϩ
Ph
R1
Ph
OR2
OR2
35 (37–57%)
36 (11–18%)
Scheme 14. Synthesis and ring expansion of 4-arylcyclobutenimines.[42]

RESEARCH FRONT
1660
E. Schaumann et al.
N-tert-butyl derivative 21 (Ar ¼ Ph, R ¼ tBu), which was iso-
lated in 72% yield.
and 1670 cmꢀ1 for the analogous unit in 7 (Scheme 2) and in the
light of imine absorptions in N-phosphinyl imines in the 1680–
1690 cmꢀ1 range,[47] i.e. at higher wavenumber than ordinary
imines (1610–1635 cmꢀ1).[48] The 13C NMR spectrum shows
low-field resonances at d ¼ 200.5, 200.4, 195.7, and 191.7.
Formation of 24 is assumed to involve initial attack of the
iminophosporane on the more reactive carbonyl group of 18 to
give 22. This zwitterionic intermediate, having a choice of an
aza-Wittig[44,45] or aza-Peterson reaction,[49,50] eliminates tri-
methylsiloxide followed by dealkylation to give 24.
The synthesis of iminophosphoranes as outlined here
requires potentially dangerous organic azides which are often
not readily available. Also, theaza-Wittig routeto 21 (Scheme 7)
does not allow access to N-alkylimino derivatives. These short-
comings were overcome by employing semisquaric acid amides
25, which are readily obtained from the corresponding semi-
squarates 18.[9,11,24,51,52] Specifically, O-alkylation of 25 using
Meerwein reagent in refluxing dichloromethane generated
cyclobutenimmonium salts 26, which were unstable when
exposed to moist air and regenerated the starting amides 25.
However, they can be deprotonated under non-hydrolytic con-
ditions using a bulky base with proton sponge (1,8-bis-dimethyl-
amino-naphthalene)[53] at room temperature giving the best
results (Scheme 9). Precipitation of the salt is enhanced by
cooling to ꢀ108C. The spectroscopic data of the N-alkyl
derivatives resemble those of the N-aryl compounds 21, i.e.
they show a carbonyl stretch in their IR spectra at 1756 cmꢀ1
and a vinylogous imidate stretch at 1699 cmꢀ1. In analogy to
the N-aryl derivatives, their NMR spectra show only one set
of signals implying again a fixed configuration of the imino
unit.[46]
An alternative method for the synthesis of N-arylimino-
cyclobutenones 21 was developed and is outlined as follows.[42]
It is well established that the reaction of iminophosphoranes
with carbonyl compounds constitutes an excellent method for
the construction of imino groups (‘aza-Wittig’ reaction).[44,45]
To test this in the cyclobutenone series, 18 (R1 ¼ iPr; R2 ¼ Ph)
was treated with (N-phenylimino)triphenylphosphorane and this
gave the corresponding 21 in 89% yield. This compound was
identical to the iminocyclobutenone obtained in 73% by the
method outlined in Scheme 6. As expected, the enone carbonyl
rather than the vinylogous ester in 18 reacts preferentially with
the aza-Wittig reagent. It is noteworthy that N-(2- or 4-pyridyl)
derivatives could be obtained by this method (Scheme 7) but not
by the route outlined in Scheme 6.
In attempts to expand the synthetic scope of the aza-Wittig
method, 3-alkoxy-4-phenyl-1,2-cyclobutenediones 18 were
treated with N-(trimethylsilyl)imino-triphenylphosphorane.
Surprisingly, the corresponding 4-(trimethylsilylimino)cyclo-
butenone 21 was not found (Scheme 8). Instead, the unusual
bicyclic product 24 was isolated (Scheme 8).[46] The tentative
structure assignment is based on spectral and analytical data.
That is, the IR spectrum shows absorptions for the strained
carbonyl and imine at 1758 and 1721 cmꢀ1, respectively, which
may seem reasonable when compared with absorptions at 1760
Xylene
reflux
iPrO
O
O
i PrO
Ph
Ph
R1
OMe
R1
OMe
Ring Expansion Chemistry of Cyclobutenimines
37
38
Significant synthetic utility of cyclobutenones 1 rests on their
thermally induced 4p-electrocyclic ring opening to reactive
alkenylketene[3] intermediates coupled with the subsequent
6p-electrocyclic ring closure involving appended unsaturated
side chains to give ring expanded products.[2,4,54] For example, a
unique synthesis of hydroquinones involves the thermolysis of
4-alkenylcyclobutenones to give initially dienylketenes, which
provide the hydroquinones upon 6p-electrocyclic ring closure
(Scheme 10).
OH R1
OH
R1 ϭ H (86%), Me, OMe
iPrO
Ph
i PrO
ϩ
R1
Ph
OMe
OMe
39 (25–45%)
40 (19–23%)
Scheme 15. Synthesis and ring expansion of 4-arylcyclobutenones.[42]
Xylene
OH
or
PhCl
⌬
OH
Ph
Ph
O
Ph
R
Li
R
R
59–83%
iPrO
NPh
i PrO
NPh
i PrO
N
Ph
21
(R2 ϭ Ph)
41
42
R
O
R
OH
O
R
[O]
Ph
Ph
Ph
10–49%
N
N
H
iPrO
N
i PrO
i PrO
44
45
43
R ϭ C3H7, CH2SiMe3, Ph
Scheme 16. Synthesis and ring expansion of 4-alkynylcyclobutenimines.

RESEARCH FRONT
New Syntheses and Ring Expansion Reactions of Cyclobutenimines
1661
OH
Ph
R
OH
N
i PrO
R
Ph
43
i PrO
N–Ph
NPh
O
NHPh
NPh
OH
42
iPrO
R
iPrO
i PrO
R
R
Ph
Ph
Ph
OH
48
46
R ϭ C3H7, CH2SiMe3
47
Scheme 17. Alternate reaction pathways in the cyclization of ketenimine 42.[42]
Ph
Ph
O
O
OH
LiC≡ Ph
Xylene
⌬
Ph
O
Ph
N
THF
Ph
ϩ
Ph
Ph
N
80%
N
N
iPrO
N
N
i PrO
9
i PrO
i PrO
N
N
21
(R2 ϭ 3-pyridyl)
49
50 (7%)
51 (12%)
Ph
N
O
Ph
O
N
N
Ph
iPrO
22%
iPrO
N
21
(R2 ϭ 4-pyridyl)
52
Scheme 18. Synthesis and ring expansion of 4-alkynyl-N-pyridyl-cyclobutenimines.[42]
Although it was an open question whether cyclobutenimines
would behave analogously, anticipation of such was suggested
based upon previous work showing that photolysis of an
iminocyclobutenone of type 21 leads to a products apparently
derived from an a,b-unsaturated alkenyl ketenimine derivative
(Scheme 11).[28,29,55]
The availability of iminocyclobutenone 21 as outlined above
allowed a study of their reactions with lithium organyls in
anticipation of establishing a viable route to the required
cyclobutenimines having unsaturated substituents at position-4.
Thus, iminocyclobutenones 21 were treated with vinyllithium or
2-propenyl lithium giving good to excellent yields of the 1,2-
adducts 27 (Scheme 12). The IR spectra of these adducts lack the
presence of a carbonyl absorption but do show the presence of an
imino absorption at approximately 1690 cmꢀ1. Similarly, their
13C NMR spectra show no carbonyl resonance, but they do show
the absorption for the imino carbon at approximately d 165.
Chemical evidence for their assigned structures rests on their
observed acid-catalyzed hydrolysis to give the vinylogous
hemiacetal 28 (Scheme 12).[42]
been exploited in natural product synthesis by the Moore[56]
and Trost[57] groups to secure the required regiochemistry in the
target compounds.
Iminocyclobutenes 27 are stable at room temperature, but
upon heating undergo ring expansion to give the aminophenols
31 (Scheme 13). This transformation apparently involves electro-
cyclic ring opening of 27 to the corresponding dienylketenimine
intermediates 30 followed by ring closure to ultimately give
the observed products. Thus, these iminocyclobutenes follow
the same general reaction pathway as that previously reported
for the analogous cyclobutenones (Scheme 10).
Once again, since 28 and 29 are regioisomers, their synthesis
and the above ring expansion provides selective control of
the substitution pattern of the corresponding aminophenols.
Furthermore, since the aminophenols are precursors to the
corresponding quinones upon oxidation with silver(I) oxide,[58]
they can also control the quinone substitution pattern.
Following the successful synthesis and ring expansion of 4-
(1-alkenyl)-cyclobutenimines 27, our study was then extended
to include the synthesis and ring expansion of 4-aryl analogues
(Scheme 14). Again regiospecific 1,2-addition to the carbonyl
group in 27 is observed to give 33 when the intermediate
alkoxide was quenched with water (R2 ¼ H) or methyl triflate
(R2 ¼ OSO2CF3). By analogy to the ring expansion of 4-
(1-alkenyl) derivatives 27 (Scheme 13), heating 33 in xylene
In passing it is noted that 28 and 29 are regioisomeric, the
former arising from the iminocyclobutenone 21 and the latter
from cyclobutendione 18. As a result, 21 is effectively a
protected form of 18 and together these provide a potentially
useful route to regioisomeric cyclobutenones. This aspect has

RESEARCH FRONT
1662
E. Schaumann et al.
OSiMe3
Ph
OH
OSiMe3
NPh
Me3SiCl
NEt3
Xylene
reflux
Ph
Ph
Ph
Ph
Ph
41%
i PrO
N
NPh
i PrO
iPrO
41
53
54A
Ph
OSiMe3
N
Me3SiO
Ph
H
Ph
Ph
i PrO
19%
N
i PrO
H
54B
55
Scheme 19. Synthesis and ring expansion of a 4-alkynyl-4-trimethylsiloxycyclobutenimine.
O
OH
Ph
Ph
O
N
Ph
Xylene
LiC≡ CPh
86%
Ph
⌬
Ph
53%
Mes
i PrO
N
Mes
i PrO
N
Mes
i PrO
21
(R2 ϭ Mes)
56
57
Me
Mes ϭ
Me
Me
Scheme 20. Synthesis and ring expansion of a 4-alkynyl-N-mesitylcyclobutenimine.[42]
provides aminonaphthols or naphthyl ethers (Scheme 14).
The naphthols tend to be oxidized by air while the ethers are
reasonably stable compounds. The two products 35, 36 arise
from the iminoketene 34 bearing a m-substituted aryl group
(R1 ¼ H). Thus two possible ring closure pathways are available.
Here, the position para to substituent R1 is sterically less
shielded and may also be electronically preferred, thus influen-
cing the preference for product 35. The same competition of two
substitution pathways was observed earlier for the cyclization
of the ketene congeners 37 of ketenimines 34, in which the
cyclization to 39/40 was found to be less selective, demonstrat-
ing the higher electrophilicity and thus lower selectivity for
ketene 38 attack as compared with attack by the ketenimine 34
(Scheme 15).[59] Also the smaller carbonyl group in ketenes
relative to the imino unit in ketenimines may further reduce
selectivity.
Analogously to their alkenyl and aryl congeners, alkynyl
lithium reagents were found to react with 4-iminocyclobute-
nones 21 to give 4-alkynylcyclobutenimines 41 by 1,2-addition
(Scheme 16).[42] Thermolysis of these did not result in facile
ring expansion to iminoquinones (Scheme 14) as did the
corresponding 4-alkenylcyclobutenone ring expansion to qui-
nones (see Scheme 15). Low yields of these were realized
(Scheme 17) but the major product was the cyclopenta-
annulated quinolines 45, isolated in 10–49% yield.[42] In
analogy to the 4-alkynylcyclobutenone series,[60,61] the diradi-
cal 43 is presumably involved which allows delocalization of an
unpaired electron into the N-aryl substituent. Diradical ring
closure at an o-position of the N-phenyl substituent occurs.
The moderate to low yields may be due to the fact that an
oxidation step from the primary product 44 to 45 is involved, but
the nature of the oxidant is not known. It might be oxygen from
air during workup, but a possible oxidizing agent might also be
an iminoquinone 47, which in some cases was isolated in low
yield (,20%) from the complex reaction mixture along with
the aminophenol 48 (,18%). Products 47, 48 are presumably
formed from the enynylketenimine 42 via radical attack at the
terminal alkyne carbon (Scheme 17), a transformation in ana-
logy to the cyclobutenone/quinone reaction (Scheme 10).
The interesting heterocycles formed on heating the N-
(3-pyridyl)- and the N-(4-pyridyl)-cyclobutenimines 21 appear
to stem from the same pathway as the 4-aryl derivatives
(Scheme 18), i.e. 50/51 and 52 respectively. Two cyclization
modes are possible starting from 21 (R2 ¼ 3-pyridyl) and, in
fact, two products 50 and 51 are isolated (Scheme 17, upper
reaction). Structure assignment is based on analogy and on the
spectroscopic data. Specifically, a characteristic feature of 50
is a broad singlet in the H NMR spectrum at d 9.44 for H-9
(Scheme 17). This proton gives resonance at d 9.00 in product
52.[42]
Further mechanistic insight for the 4-alkynylcyclobuteni-
mine rearrangements was obtained from a study of the thermo-
lysis of 4-alkynyl-4-trimethylsiloxycyclobutenimine 53. Here,
the product was the quinoline 55, isolated in 19% yield
(Scheme 19). Thus, the trimethylsiloxy group in 51 prevents
facile oxidation as was noted for 44 giving further evidence for
1

RESEARCH FRONT
New Syntheses and Ring Expansion Reactions of Cyclobutenimines
1663
the mechanistic details outlined above. Evidence for the sug-
gested structure of 55 comes from its 1H NMR spectrum
showing the cyclopentyl methine hydrogen at d 5.53 and from
NOE experiments where irradiation of the cyclopentyl methine
resonance caused a 7% enhancement of the absorption of the
methine hydrogen of the isopropoxy group at d 4.29. Likewise,
irradiation of the absorption at d 4.29 caused a 10% enhance-
ment of the cyclopentyl hydrogen absorption at d 5.53.[42]
A prerequisite for the formation of products 45 and 55 is
the presence of an unsubstituted o-position in the N-phenyl
substituent. Therefore, a methyl group in this position should
prevent ring closure on the N-aryl group. This was, in fact,
observed. Specifically, thermolysis of the N-mesityl derivative
56 in refluxing xylene gave the cyclopentene 57 in 53% isolated
yield (Scheme 20). The formation of 57 is assumed to arise via
hydrogen abstraction from the proximal hydroxy group in a
diradical intermediate analogous to 43. The structure 57 was
proven by a single-crystal X-ray investigation.[42] It should be
noted that the observed E configuration of the exocyclic C¼C
unit in 57 speaks against an intramolecular hydrogen transfer
from the hydroxy group in 43.
[11] M. Mu¨ller, M. J. Heileman, H. W. Moore, E. Schaumann, G. Adiwidjaja,
Synthesis 1997, 50. doi:10.1055/S-1997-1496
[12] W. Regenhardt, E. Schaumann, H. W. Moore, Synthesis 2001, 1076.
doi:10.1055/S-2001-14559
[13] K. Benda, W. Regenhardt, E. Schaumann, G. Adiwidjaja, Eur. J. Org.
Chem. 2009, 1016. doi:10.1002/EJOC.200801101
[14] E. Schaumann, Bull. Soc. Chim. Belg. 1986, 95, 995. doi:10.1002/
BSCB.19860951106
[15] E. Schaumann, Perspectives in the Organic Chemistry of Sulfur 1987,
p. 251 (Eds B. Zwanenburg, A. J. H. Klunder) (Elsevier: Amsterdam).
[16] E. Schaumann, S. Winter-Extra, G. Ru¨hter, J. Org. Chem. 1990, 55,
4200. doi:10.1021/JO00300A045
[17] W. Ried, P. Schomann, Liebigs Ann. Chem. 1968, 714, 140.
doi:10.1002/JLAC.19687140114
[18] (a) W. Ried, A. H. Schmidt, Angew. Chem. 1972, 84, 1048.
doi:10.1002/ANIE.197209971
(b) W. Ried, A. H. Schmidt, Angew. Chem. Int. Ed. Engl. 1972, 11,
997. doi:10.1002/ANIE.197209971
[19] H. Knorr, W. Ried, Synthesis 1978, 649. doi:10.1055/S-1978-24841
[20] (a) W. Ried, W. Kunstmann, Angew. Chem. 1968, 80, 121.
doi:10.1002/ANIE.196801351
(b) W. Ried, W. Kunstmann, Angew. Int. Ed. Engl. 1968, 7, 135.
doi:10.1002/ANIE.196801351
[21] W. Ried, W. Kunstmann, Chem. Ber. 1969, 102, 1422. doi:10.1002/
CBER.19691020438
Conclusions
[22] G. Oppermann, Ph.D. Thesis: Zur Chemie von exo-Methylen-oder
Stickstoff-modifizierten Cyclobutendionen, Clausthal 2007.
[23] W. Ried, W. Kunkel, G. Isenbruck, Chem. Ber. 1969, 102, 2688.
doi:10.1002/CBER.19691020825
[24] G. Seitz, P. Imming, Chem. Rev. 1992, 92, 1227. doi:10.1021/
CR00014A004
[25] (a) L. Ghosez, B. Haveaux, H. G. Viehe, Angew. Chem. 1969, 81, 468.
doi:10.1002/ANIE.196904541
(b) L. Ghosez, B. Haveaux, H. G. Viehe, Angew. Int. Ed. Engl. 1969, 8,
454. doi:10.1002/ANIE.196904541
[26] (a) H. G. Viehe, Angew. Chem. 1967, 79, 744. doi:10.1002/ANIE.
196707671
Three routes to 4-iminocyclobutenones 21 were developed. In
all cases, lithium organyls bearing unsaturated organic residues
undergo selective 1,2-addition to the enone carbonyl group as
opposed to the imine group. This provides compounds for a
broad-based study of the thermolyses of substituted imino-
cyclobutenes. These reactions show similarity to those of the
analogous cyclobutenones. That is, they are initiated by
4p-electrocyclic ring opening, followed by 6p-electrocyclic
ring closure to give aminophenols or aminonaphthols. It was
demonstrated that the products show reversed regiochemistry
as compared with the corresponding carbonyl-derived products.
Thus, the imino group in iminocyclobutenones can be viewed
as a protective group for the more reactive carbonyl group in
alkoxy-substituted cyclobutendiones.
(b) H. G. Viehe, Angew. Int. Ed. Engl. 1967, 6, 767. doi:10.1002/ANIE.
196707671
[27] B. Singer, G. Maas, Chem. Ber. 1987, 120, 485. doi:10.1002/CBER.
19871200407
[28] N. Obata, T. Takizawa, Tetrahedron Lett. 1970, 2231. doi:10.1016/
S0040-4039(01)98196-1
[29] N. Obata, T. Takizawa, Bull. Chem. Soc. Jpn. 1977, 50, 2017.
doi:10.1246/BCSJ.50.2017
[30] J. S. Chickos, J. Org. Chem. 1973, 38, 3642. doi:10.1021/
JO00960A051
[31] D. Moderhack, Synthesis 1985, 1083. doi:10.1055/S-1985-31437
[32] T. Eicher, U. Stapperfenne, Synthesis 1987, 619. doi:10.1055/S-1987-
28024
A mechanistically more complex reaction was observed for
4-alkynyl derivatives 41. Here a diradical intermediate is
proposed which undergoes ring closure involving the N-aryl
group to ultimately give cyclopenta[b]quinolines as the primary
products. Although the yields of these products have yet to be
optimized, the unusual structures give evidence of the versatility
of cyclobutenimine chemistry.
References
[33] G. Verniest, J. Colpaert, K. W. To¨rnroos, N. De Kimpe, J. Org. Chem.
[1] S. P. Marsden, Sci. Synth. 2005, 26, 1045.
2005, 70, 4549. doi:10.1021/JO0502044
[2] (a) H. W. Moore, B. R. Yerxa, Chemtracts 1992, 5, 273.
(b) H. W. Moore, B. R. Yerxa, Adv. Strain Org. Chem. 1995, 4, 81.
[3] L. Danheiser, G. B. Dudley, W. F. Austin, Sci. Synth. 2006, 23, 493.
[4] L. S. Liebeskind, Tetrahedron 1989, 45, 3053. doi:10.1016/S0040-
4020(01)80131-9
[5] S. J. Collier, S. K. Singh, Sci. Synth. 2009, 45a, 49.
[6] (a) D. Bellusˇ, B. Ernst, Angew. Chem. 1988, 100, 820.
(b) D. Bellusˇ, B. Ernst, Angew. Chem. Int. Ed. Engl. 1988, 27, 797.
[7] A. Mukkanti, M. Periasamy, Arkivoc 2005, xi, 48.
[8] H. Frauenrath Houben-Weyl, Methods of Organic Chemistry 1993,
Vol. E15b, p. 1383 (Eds H. Kropf, E. Schaumann) (Thieme: Stuttgart).
[9] H. Frauenrath Houben-Weyl, Methods of Organic Chemistry 1993,
Vol. E15b, p. 1468 (Eds H. Kropf, E. Schaumann) (Thieme: Stuttgart).
[10] T. M. Krygowski, M. K. Cyran˜ski, Z. Czarnocki, G. Ha¨felinger, A. R.
Katritzky, Tetrahedron 2000, 56, 1783. doi:10.1016/S0040-4020(99)
00979-5
[34] H. Liu, C. S. Tomooka, H. W. Moore, Synth. Commun. 1997, 27, 2177.
doi:10.1080/00397919708006826
[35] L. S. Liebeskind, R. W. Fengl, K. R. Wirtz, T. T. Shawe, J. Org. Chem.
1988, 53, 2482. doi:10.1021/JO00246A017
[36] U. Noster, B. M. Hansen, B. Krische, K. H. Schulz, Contact Dermat.
1976, 2, 99. doi:10.1111/J.1600-0536.1976.TB02993.X
[37] J. Avalos, H. W. Moore, M. W. Reed, E. Rodriguez, Contact Dermat.
1989, 21, 341. doi:10.1111/J.1600-0536.1989.TB04757.X
[38] M. W. Reed, D. J. Pollart, S. T. Perri, L. D. Foland, H. W. Moore,
J. Org. Chem. 1988, 53, 2477. doi:10.1021/JO00246A016
[39] L. S. Liebeskind, K. R. Wirtz, J. Org. Chem. 1990, 55, 5350.
doi:10.1021/JO00306A011
[40] K. Benda, T. Knoth, R. L. Danheiser, E. Schaumann, Tetrahedron Lett.
2010, in press.
[41] G. Oppermann, Diploma Thesis: Ringerweiterungsreaktionen von
4-Iminocyclobutenonen, Clausthal 2003.

RESEARCH FRONT
1664
E. Schaumann et al.
[42] G. Oppermann, M. Stranberg, H. W. Moore, E. Schaumann,
G. Adiwidjaja, Synthesis 2010, 2027.
[53] (a) H. A. Staab, T. Saupe, Angew. Chem. 1988, 100, 895. doi:10.1002/
ANIE.198808653
[43] D. Wurmb-Gerlich, F. Vo¨gtle, A. Mannschreck, H. A. Staab, Liebigs
Ann. Chem. 1967, 708, 36. doi:10.1002/JLAC.19677080103
[44] K. Abbaspour Tehrani, N. De Kimpe, Sci. Synth. 2004, 27, 281.
[45] F. Palacios, C. Alonso, D. Aparicio, G. Rubiales, J. M. de los Santos,
Tetrahedron 2007, 63, 523. doi:10.1016/J.TET.2006.09.048
[46] M. Stranberg, Ph.D. Thesis: Synthesis of Iminocyclobutenones and the
Investigation of the Thermal Rearrangements of 4-Substituted Imino-
cyclobutenes, University of California, Irvine 1998.
(b) H. A. Staab, T. Saupe, Angew. Int. Ed. Engl. 1988, 27, 865.
doi:10.1002/ANIE.198808653
[54] M. Ohno, Y. Yamamoto, S. Eguchi, Synlett 1998, 1167. doi:10.1055/
S-1998-5893
[55] N. Obata, T. Takizawa, Chem. Commun. 1971, 587.
[56] M. P. Winters, M. Stranberg, H. W. Moore, J. Org. Chem. 1994, 59,
7572. doi:10.1021/JO00104A004
[57] M. Trost, O. R. Thiel, H.-C. Tsui, J. Am. Chem. Soc. 2002, 124, 11616.
doi:10.1021/JA0277834
[58] M. C. Carren˜o, M. Ribagorda, Sci. Synth. 2006, 28, 629.
[59] L. D. Foland, J. O. Karlsson, S. T. Perri, R. Schwabe, S. L. Xu, S. Patil,
H. W. Moore, J. Am. Chem. Soc. 1989, 111, 975. doi:10.1021/
JA00185A030
[47] H. Kinoshita, O. J. Ingham, W. W. Ong, A. B. Beeler, J. A. Porco, Jr,
J. Am. Chem. Soc. 2010, 132, 6412. doi:10.1021/JA100346W
[48] S. Pawlenko, Houben-Weyl, Methoden der Organischen Chemie
1990, Vol. 14b/1, p. 222 (Eds D. Klamann, H. Hagemann) (Thieme:
Stuttgart).
[49] G. Cainelli, D. Giacomini, P. Galletti, A. Gaiba, Synlett 1996, 657.
doi:10.1055/S-1996-5550
[50] D. Enders, Acros Org. Acta 1995, 1, 37.
[51] A. H. Schmidt, Synthesis 1980, 961. doi:10.1055/S-1980-29291
[52] A. H. Schmidt, M. Debo, B. Wehner, Synthesis 1990, 237.
doi:10.1055/S-1990-26842
[60] J. O. Karlsson, N. V. Nguyen, L. D. Foland, H. W. Moore, J. Am.
Chem. Soc. 1985, 107, 3392. doi:10.1021/JA00297A081
[61] H. W. Moore, S. T. Perri, J. Org. Chem. 1988, 53, 996. doi:10.1021/
JO00240A011