Synthesis of highly substituted cyclobutane fused-ring systems from n-vinyl β-lactams through a one-pot domino process

Article abstract of DOI:10.1002/chem.200902748

In this contribution, aminocyclobutanes, as well as eight-membered enamide rings, have been made from N-vinyl β-lactams. The eightmembered products have been formed by a [3,3]-sigmatropic rearrangement, whereas the aminocyclobutanes have been derived from a domino [3,3]-rearrangement/6π- electrocyclisation process. The aminocyclobutanes have been obtained in a highly diastereoselective fashion. The cyclobutane ring system tolerates fusion even if adjacent quaternary centres are present. Systems containing up to four fused rings are readily accessible. The reaction profile has been investigated by using Gaussian 03. This study suggests that two reaction pathways for aminocyclobutane formation are possible. In one pathway the [3,3]-sigmatropic rearrangement is the rate-limiting step and in the second pathway the electrocyclisation is rate limiting. Taken together, these reactions should facilitate the construction of fused heterocycles.

Full text of DOI:10.1002/chem.200902748

DOI: 10.1002/chem.200902748
Synthesis of Highly Substituted Cyclobutane Fused-Ring Systems from
N-Vinyl b-Lactams through a One-Pot Domino Process
Lawrence L. W. Cheung and Andrei K. Yudin*[a]
Abstract: In this contribution, amino-
cyclobutanes, as well as eight-mem-
bered enamide rings, have been made
from N-vinyl b-lactams. The eight-
membered products have been formed
by a [3,3]-sigmatropic rearrangement,
whereas the aminocyclobutanes have
been derived from a domino [3,3]-rear-
fashion. The cyclobutane ring system
tolerates fusion even if adjacent quater-
nary centres are present. Systems con-
taining up to four fused rings are readi-
ly accessible. The reaction profile has
been investigated by using Gaussian 03.
This study suggests that two reaction
pathways for aminocyclobutane forma-
tion are possible. In one pathway the
[3,3]-sigmatropic rearrangement is the
rate-limiting step and in the second
pathway the electrocyclisation is rate
limiting. Taken together, these reac-
tions should facilitate the construction
of fused heterocycles.
Keywords: lactams · cyclobutanes ·
rangement/6p-electrocyclisation
pro-
domino reactions
· electrocyclic
cess. The aminocyclobutanes have been
obtained in a highly diastereoselective
reactions · pericyclic reactions
Introduction
most common way to make cyclobutanes is by a photochem-
ical [2+2] cycloaddition between two olefins. This photo-
chemical pathway has inherent drawbacks, such as a lack of
chemo- and stereoselectivity. Thermal [2+2] cycloadditions
between olefins are known to take place, but they are much
more rare because the olefins that participate in the reaction
have to be electronically biased or must exhibit a great deal
of strain.[8]
We have been interested in the synthesis and transforma-
tions of b-lactam and cyclobutane ring systems and recently
reported a synthetic method to make aminocyclobutanes
from N-vinyl b-lactams in a highly stereoselective manner
through a pericyclic cascade sequence.[9] Herein, we present
an account of the synthetic utility of N-vinyl b-lactams,
which includes the synthesis of eight-membered enamide
rings through a sigmatropic ring expansion, the synthesis of
b-Lactams are important molecules because of their utility
as therapeutic agents and as intermediates in chemical syn-
thesis.[1] Numerous methods have been developed to prepare
these molecules in both academic and industrial settings.
The ester–enolate imine cycloaddition, olefin–isocyanate cy-
cloaddition, Kinugasa reaction and the Staudinger reaction
are just a few of the many methods used to make b-lac-
tams.[2] The pharmaceutical industry has significant demands
for b-lactams because of their biological activity.[3] Ever
since the discovery of penicillin,[4] great strides were made
towards synthesising various penicillin analogues to combat
the growing resistance of bacteria against therapeutics.[5]
Compared with b-lactams, cyclobutanes are much less
studied four-membered rings, but they are common motifs
in many natural products.[6] Cyclobutanes can also serve as
reactive handles for the synthesis of natural products.[7] The
aminocyclobutanes through
a
ring-strain transposition
domino sequence. While dwelling on mechanistic underpin-
nings of this reaction, we also performed a computational
study of the mechanism, which provides evidence for the in-
termediates postulated along the reaction pathway.
[a] L. L. W. Cheung, Prof. A. K. Yudin
Davenport Research Laboratories
Department of Chemistry
University of Toronto
80 St. George St., Toronto ON, M5S 3H6 (Canada)
Fax : (+1)416-946-7676
Results and Discussion
N-Vinyl b-lactam starting materials can be prepared on a
multigram scale in three steps that include a single purifica-
tion (Scheme 1). The synthesis begins with a formal [2+2]
E-mail: ayudin@chem.utoronto.ca
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902748.
4100
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4100 – 4109

FULL PAPER
Scheme 1. Synthesis of N-vinyl b-lactams.
cycloaddition between a diene, such as isoprene, and chloro-
sulfonyl isocyanate[10] to afford the N-chlorosulfonyl b-
lactam, which is cleanly reduced to the corresponding N-H
b-lactam with aqueous sodium sulfite.[11] The N-H b-lactam
is then cross coupled with a vinyl iodide[12] under copper-cat-
alysed coupling conditions[13] to yield the corresponding N-
vinyl b-lactam.
Scheme 3. Attempted one-pot-process synthesis of eight-membered en-
amides leading to aminocyclobutanes.
Upon thermal microwave heating of N-vinyl b-lactams,
we observed the formation of eight-membered enamide
rings[14] by a [3,3] sigmatropic rearrangement/tautomerisa-
tion sequence. As shown in Table 1, electron-rich, -poor and
neutral enamide substituents 1a–4a are tolerated, producing
compounds 1b–4b. Substrates containing various heterocy-
clic side chains, such as thiophene 7a, pyrrole 8a and furan
9a, also participate in this chemistry. We note that fused ar-
omatic heterocycles 8c and 9c are produced in addition to
the eight-membered enamide rings when furan or N-methyl
pyrrole side chains are used. The b-lactam component can
be modified to contain an allene in place of an olefin. Al-
lenes also partake in the [3,3] sigmatropic rearrangement
leading to cross-conjugated triene ring systems, such as 10b.
When the R group is an alkyl substituent, the initial product
is imine 6b, which tautomerises into enamide 6c.
The structure of fused aromatic heterocycle 8c (Table 1)
was unambiguously proven through X-ray crystallography.[15]
Formation of these fused aromatic heterocycles is likely to
proceed by a Friedel–Crafts addition onto the b-lactam
(Scheme 2), followed by extrusion of isoprene to yield the
heterocycle. None of the other electron neutral or electron-
rich substituted benzene rings followed this reaction path-
way, which is consistent with a requirement for a fairly elec-
tron-rich ring to be present.
Despite the low yield of this reaction outcome, it captured
our attention. A literature search of the fused cyclobutane
d-lactam fragment shows that the vast majority of proce-
dures involve a photochemical reaction between two olefins
to make the ring system.[16] Thus, our method appears to be
a rare thermal process to make fused cyclobutane d-lactams.
The optimisation of the reaction conditions for the aminocy-
clobutane synthesis was focused on using N-vinyl b-lactams
as starting materials rather than in situ synthesis of N-vinyl
b-lactams from N-H b-lactams and vinyl halides.
The one-pot reactions that lead to products through
either tandem, domino or cascade reactions are of substan-
tial value in organic synthesis. The syntheses of numerous
complex natural products incorporate these reactions at var-
ious stages.[17] These reactions also have countless benefits,
in that they can be atom, time and step economical.[18] In
particular, domino reactions are one-pot processes in which
all of the starting materials and reagents are present from
the very beginning. The functional group(s) produced in the
ensuing sequence of bond-forming and -breaking events
feed into the following step, much like a game of domi-
nos.[19]
Table 2 shows the results of optimising the reaction condi-
tions for the synthesis of the aminocyclobutane from N-
vinyl b-lactam 12a. Microwave heating of 12a in the temper-
ature range between 140 and 2008C without any additives
afforded the eight-membered enamide product exclusively
(Table 2, entry 1). As shown in entry 2, when caesium car-
bonate was added, we obtained a mixture of the 8-mem-
bered enamide and aminocyclobutane in a 4:6 ratio in
favour of the latter with a combined yield of 75%. Finally,
when both copper iodide and caesium carbonate were used,
the aminocyclobutane was produced exclusively.
Scheme 2. Proposed mechanism of fused aromatic heterocycle synthesis.
One of the notable features of the aminocyclobutanes is
the combined effect of a cyclobutane and an a,b-unsaturat-
ed carbonyl system on the 13C NMR chemical shift of the b
carbon. The 13C chemical shift of the b carbon ranges from
d=150 to 160 ppm in all of the aminocyclobutanes. The six-
membered a, b- unsaturated lactams typically display the
13C chemical shift of the b carbon around d=142.8 ppm
(CD3OD, TMS).[20] In our case, the combination of a cyclo-
butane and an a, b-carbonyl system dramatically increases
The next task was to try and make the eight-membered
enamide rings by generating the N-vinyl b-lactam in situ.
The latter was projected to undergo a [3,3] sigmatropic rear-
rangement/tautomerisation sequence to yield the eight-
membered enamide (Scheme 3).
In the event, there was substantial decomposition and
none of the eight-membered enamide ring was obtained. In-
stead, a d-lactam fused to the cyclobutane ring was isolated.
Chem. Eur. J. 2010, 16, 4100 – 4109
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4101

L. L. W. Cheung and A. K. Yudin
Table 1. Synthesis of eight-membered enamide rings from N-vinyl b-lactams.[a]
the chemical shift of the
b
carbon. This hints at the possi-
bility that the b carbon can be a
good electrophile for conjugate
additions.
Table 3 shows the substrate
scope of the aminocyclobutane
formation. The substrates ini-
tially tested were compounds
6a, 12a and 13a. The reaction
works well in each case with
moderate levels of diastereose-
lectivity and tolerates alkyl side
chains as well as a tethered
phenyl group. The geminally
disubstituted enamides were
then reacted to yield the amino-
cyclobutane products with a ter-
tiary stereocentre adjacent to a
quaternary stereocentre. The
transformations of substrates
equipped with alkyl substituents
or tethered phenyl rings
worked reasonably well to
afford products 14b through
17b. In many cases the crude
products did not require any
purification. The reaction selec-
tivity was not as high when
starting materials 19a–21a, con-
taining aryl groups were used
compared with starting materi-
als containing alkyl groups.
Both the aminocyclobutane and
the eight-membered enamide
ring product were isolated in
these instances. Carbocyclic
side chains, such as cyclopropyl,
cyclobutyl and cyclohexyl, at
the geminal position could also
be used to make compounds
22b, 23b and 24b in respecta-
ble yields. Lastly, the spirocyclic
compound 25b was made from
a tetrasubstituted olefin. Single
diastereoisomers were isolated
in all cases in which gem-disub-
stituted olefins were used.
b-Lactam
Product
Yield [%][b]
1
2
86
48
62
3
4
5
58
56
6
–
7
8
9
73
–
N-Vinyl b-lactams that con-
tained carbocyclic trisubstituted
olefins were also synthesised
(Table 4). Subjecting these sub-
strates to reaction conditions
produced two fused rings with
three contiguous stereocentres
in a single step. With the five-
membered carbocycle, the reac-
–
4102
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4100 – 4109

Highly Substituted Cyclobutane Fused-Ring Systems
FULL PAPER
cyclobutyl
carbocyclic
side
Table 1. (Continued)
chains gave compounds 34b
and 35b. The spirocyclic com-
pound 36b can be made from
the tetrasubstituted olefin 36a.
To push the scope further, b-
lactams derived from 2,3-disub-
stituted 1,3-butadienes and
chlorosulfonyl isocyanate were
employed. These N-vinyl b-
lactam systems can be trans-
b-Lactam
Product
Yield [%][b]
10
48
[a] Conditions: N-vinyl b-lactam (20.0 mg), DMF (500 mL), microwave heating 160–2008C, see the Supporting
Information.[9] [b] Isolated yield.
Table 2. Aminocyclobutane synthesis optimisation: additives screening.
Additive
T
[8C]
Result
Yield
[%][a]
1
2
3
–
140–200
180
180
100:0
40:60
0:100
50
75
98
Cs2CO3
CuI (5 mol%), Cs2CO3 (1.5 equiv)
Scheme 4. Synthesis of 2-cyclopentyl-1,3-butadiene and 4-cyclopentyl-4-
vinylazetidin-2-one. DBU=1,5-diazabicycloACTHNUTRGNE[NUG 5.4.0]undec-5-ene.
[a] Isolated yield.
tion worked smoothly to produce 26b in 75% yield. The
six-membered ring produced both the aminocyclobutane
and the 8-membered enamide ring with a combined yield of
95% in a 6:4 ratio favouring the former. The seven-mem-
bered carbocyclic substituent was also within the scope of
the process, delivering compound 28b in 83% yield. With
the eight-membered carbocycle, the product produced was
exclusively the aminocyclobutane, albeit as a mixture of dia-
stereoisomers (2.60:1.00).
A common thread among all substrates shown in Tables 3
and 4 is that the N-vinyl b-lactam starting materials were de-
rived from chlorosulfonyl isocyanate and isoprene
(Scheme 1). We were interested in extending the reaction to
N-vinyl b-lactams that were made from dienes other than
isoprene. 2-Substituted 1,3-butadienes can be readily ac-
cessed from Grignard reagents and (E)-1,4-dibromobut-2-
ene by using a two-step procedure reported recently by Sie-
burth et al.[21] The reaction delivered 2-cyclopentyl-1,3-buta-
diene from cyclopentyl magnesium bromide and (E)-1,4-di-
bromobut-2-ene in 88% yield over two steps. A [2+2] cy-
cloaddition between 2-cyclopentyl-1,3-butadiene and chloro-
sulfonyl isocyanate occurred in modest yields,[15] but was
scaleable such that gram scale quantities of the N-H b-
lactam were accessible (Scheme 4). Table 5 shows the amino
cyclobutanes accessible from the corresponding N-vinyl b-
lactams derived from 2-cyclopentyl-1,3-butadiene. Long and
short alkyl side chains can be used to make amino cyclobu-
tanes 30b and 31b in reasonable yields. Tethered phenyl
and cyano groups can both be tolerated to produce com-
pounds 32b and 33b in respectable yields. Cyclopropyl and
formed into an aminocyclobutane that would contain adja-
cent quaternary stereocentres. Table 6 shows the scope of
the reaction when 2,3-dimethylbuta-1,3-diene was used as
the diene partner to prepare the corresponding N-H b-
lactam. Alkyl side chains are well tolerated yielding amino-
cyclobutanes 37b–39b and the reaction worked fairly well
when five- and seven-membered carbocycles, 40a and 42a,
containing endocyclic trisubstituted olefins were used. Three
ring systems with three contiguous stereocentres were pro-
duced in these instances. For reasons that are not yet clear,
when the six-membered carbocycle 41a was used, the reac-
tion delivered eight-membered enamide ring 41b as the
major product.[15] Tethered phenyl and cyano groups are tol-
erated, as are vinyl carbocycles to provide aminocyclobu-
tanes 43b–46b. Finally, spirocycle 47b with three contiguous
quaternary carbons was made in 80% yield.
To further extend the reaction, spirocyclic b-lactams were
used as the starting materials. The N-H b-lactam was pro-
jected to emanate from a formal [2+2] addition between
chlorosulfonyl isocyanate and 1,2-dimethylenecyclohexane
as the diene partner (Scheme 5). The latter diene is readily
prepared in one step[22] and the corresponding spirocyclic N-
H b-lactam can be made by using standard procedures de-
scribed in Scheme 5.
The reaction of the spirocyclic N-vinyl b-lactam under the
developed rearrangement conditions fused cyclohexyl and
cyclobutyl groups to a common d-lactam in one step, giving
a tricyclic system with two adjacent quaternary stereocen-
tres. In most cases, the reaction was sluggish but still deliv-
ered the desired product in modest to good yields. Table 7
Chem. Eur. J. 2010, 16, 4100 – 4109
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4103

L. L. W. Cheung and A. K. Yudin
Table 3. Substrate scope of aminocyclobutane formation.[a]
b-Lactam Product
Yield [%][b]
b-Lactam
Product
Yield [%][b]
1
2
3
95
9
42
98
93
10
11
27
34
4
5
6
7
8
83
89
81
82
74
12
13
14
15
84
89
95
93
[a] Conditions: N-vinyl b-lactam (20.0 mg), CuI (10 mol%), Cs2CO3 (1.5 equiv), DMF (500 mL), microwave heating 30 min, 160 to 2008C, see the Sup-
porting Information.[9] [b] Isolated yield.
displays the scope of the process. The reaction works well
with simple alkyl substituents, a tethered nitrile, a cyclo-
propyl and a cyclobutyl substituent to yield aminocyclobu-
tanes 48b–51b. Four ring systems can be made in a single
step with three contiguous stereocentres as a single diaste-
reoisomer, as shown in the case of products 53b and 54b.
Several facets of the aminocyclobutane reaction deserve
further discussion. The proposed mechanism of the amino-
cyclobutane formation (Scheme 6) includes a pericyclic reac-
tion sequence beginning with a [3,3] sigmatropic rearrange-
ment followed by removal of the proton alpha to the ketone
in the eight-membered-ring intermediate. Subsequently, a
6p-electron intermediate undergoes a disrotatory electrocyc-
lisation.
It should be noted that deprotonation of the eight-mem-
bered ring is faster than tautomerisation of the imine to the
unreactive eight-membered enamide ring. In a separate ex-
periment we subjected eight-membered enamide ring 1b
(Table 1) to the aminocyclobutane reaction conditions to see
if the process was reversible once the eight-membered en-
amide was made. In the event, only the starting material
was recovered, which suggests that the eight-membered en-
4104
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4100 – 4109

Highly Substituted Cyclobutane Fused-Ring Systems
Table 4. Substrate scope using carbocyclic trisubstituted olefins.[a]
FULL PAPER
Table 5. Aminocyclobutane substrate scope with a cyclopentyl substi-
tuent.[a]
b-Lactam
Product
Yield [%][b]
1
2
3
4
5
6
7
40
b-Lactam (n)
Product (Yield[%])[b]
1
2
3
4
26a (1)
27a (2)
28a (3)
29a (4)
26b (75)
27b (52)[c]
28b (83)
29b (92)[c]
77
66
70
58
77
77
[a] Conditions: N-vinyl b-lactam (20.0 mg), CuI (10 mol%), Cs2CO3
(1.5 equiv), DMF (500 mL), microwave heating 30 min, 160 to 2008C, see
the Supporting Information. [b] Isolated yield. [c] See the Supporting In-
formation.
amide ring is stable and/or tautomerisation to the imine
form is energetically unfavourable.
Realising that there is a 6p-electron intermediate that
exists as an enol or enolate, we thought that the addition of
base would affect the equilibrium. The addition of caesium
carbonate (Table 2, entry 2) resulted in a 4:6 product distri-
bution in favour of the aminocyclobutane over the 8-mem-
bered enamide. The reaction conditions that prompted the
initial discovery of the aminocyclobutane (Scheme 3) con-
tained copper iodide. When we evaluated copper iodide to-
gether with caesium carbonate, the aminocyclobutane was
delivered as the exclusive product (Table 2, entry 3). It is
plausible that copper iodide acts as a Lewis acid, coordinat-
ing to the carbonyl oxygen of the eight-membered ring in-
termediate or forms a chelate between the oxygen and the
nitrogen (see Scheme 7). These events are expected to fur-
ther assist in deprotonation of the a proton, leading to a 6p
system.
The aminocyclobutane formation can present issues of
diastereoselectivity, depending on the substitution pattern of
the N-vinyl olefin. When the olefin is E substituted, a mix-
ture of diastereoisomers results, as exemplified with N-vinyl
b-lactams 6a, 12a and 13a and their respective aminocyclo-
butane products 11b–13b shown in Table 3. The other case
is when an N-vinyl b-lactam has a large carbocycle with an
endocyclic olefin. This is exactly the case with substrate 29a
of Table 4. When the carbocycle contains eight or more car-
bons, the ring size is large enough such that electrocyclisa-
tion is not controlled by the a stereocentre of the imine.
A computational study was performed on the possible re-
action intermediates and transition states by using Gaussi-
an 03[23] (Scheme 8). The N-vinyl b-lactam 5a was chosen as
the starting material with a relative energy of 0 kcalmolÀ1.
The pathway to aminocyclobutane starts off with a [3,3] sig-
matropic rearrangement, which was found to occur through
two possible pathways. The first transition state (A) is an
endo-boat transition, which is 63.5 kcalmolÀ1 greater in
energy than the starting material 5a. The second transition
state (A’), is an exo-boat transition state, which is only
34.0 kcalmolÀ1 greater in energy than the starting material
[a] Conditions: N-vinyl b-lactam (20.0 mg), CuI (10 mol%), Cs2CO3
(1.5 equiv), DMF (500 mL), microwave heating 30 min, 160 to 1808C, see
the Supporting Information. [b] Isolated yield.
5a. A vibrational frequency analysis of A revealed that
there was very little bond breaking or bond forming, which
is consistent with a non-concerted mechanism that proceeds
through a diradical intermediate. A similar analysis of A’ re-
vealed that there is a significant degree of bond breaking
and making, which is in agreement with a concerted aromat-
ic transition state and hence a lower activation barrier. Both
A and A’ have activation barriers that are in excellent
agreement with previously reported calculations performed
on the Cope reaction of 1,5-hexadiene.[24] For the purposes
of simplifying calculations, we choose to examine the energy
required for an electrocyclisation to occur from a 6p-enol
species rather than a 6p-enolate species. After transition
Chem. Eur. J. 2010, 16, 4100 – 4109
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4105

L. L. W. Cheung and A. K. Yudin
Table 6. Synthesis of aminocyclobutane with adjacent quaternary stereocentres.[a]
b-Lactam Product
Yield [%][b]
b-Lactam
Product
Yield [%][b]
1
2
3
94
7
8
9
56
69
50
68
94
4
5
81
10
11
77
80
65[c]
6
68
[a] Conditions: N-vinyl b-lactam (20.0 mg), CuI (10 mol%), Cs2CO3 (1.5 equiv), DMF (500 mL), microwave heating 30 min, 160 to 2008C, see the Sup-
porting Information. [b] Isolated yield. [c] See the Supporting Information.
arrangement occurring through a diradical transition state
due to the high-energy requirement.[24] In our case, this reac-
tion pathway appears to be quite possible. We have isolated
species B (Table 1, entry 5), which can only arise via an
endo-boat transition state. In addition, we have obtained no
theoretical support for the transformation of B’ into B, had
the former been the kinetic product. For the pathway where
the [3,3] sigmatropic rearrangement proceeds via an exo-
boat-type transition, the reaction is endothermic because
the eight-membered ring intermediate B’ contains an E
Scheme 5. Synthesis of 1,2-dimethylenecyclohexane and 5-methylidene-1-
azaspiroACHTUNGTRENNUNG[3.5]nonan-2-one.
olefin. Tautomerisation of the 8-membered ring B’ into enol
C’ followed by a rate-limiting 6p electrocyclisation would
occur to give the aminocyclobutane. In principle, the reac-
tion pathway that produces an eight-membered ring contain-
ing an E olefin is more energetically favourable, but the in-
termediate in less stable than the starting material and
cannot be isolated. It is further reacted in situ to give the
aminocyclobutane.
state A, the eight-membered imine B is more stable than
the starting material 5a by 16.4 kcalmolÀ1. The tautomerisa-
tion of B into enol C is followed by a 6p electrocyclisation.
There have been reports discounting a [3,3] sigmatropic re-
4106
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4100 – 4109

Highly Substituted Cyclobutane Fused-Ring Systems
FULL PAPER
Table 7. Synthesis of aminocyclobutane with adjacent quaternary stereocentres.[a]
b-Lactam
Product
Yield [%][b]
b-Lactam
Product
Yield [%][b]
1
2
43
5
6
78
42
28[c]
3
4
55
68
7
81[c]
[a] Conditions: N-vinyl b-lactam (20.0 mg), CuI (10 mol%), Cs2CO3 (1.5 equiv), DMF (500 mL), microwave heating 30 min, 160 to 2008C, see the Sup-
porting Information. [b] Isolated yield. [c] See the Supporting Information.
same plane. Novak and Chua[25] have shown that there is
roughly 24.5 kcalmolÀ1 of ring-strain energy in an unsubsti-
tuted b-lactam. Roughly 21.6 kcalmolÀ1 of “amide resonance
energy” is allocated to the nitrogen lone-pair donation into
the carbonyl carbon. In order for the [3,3] sigmatropic rear-
Scheme 6. Proposed mechanism of aminocyclobutane synthesis.
rangement to occur, the two reacting olefins must orientate
into a boat-type conformation. The olefin bonded to the ni-
trogen is in the plane of the b-lactam and cannot align with
the other olefin partner because it is out of the plane of the
b-lactam. A high temperature is thus needed to overcome
the 21.6 kcalmolÀ1 of amide resonance energy such that the
nitrogen can convert into a tetrahedral state with both ole-
fins aligned in a boat-type conformation.
Conclusion
Scheme 7. The role of CuI as a Lewis acid.
We have developed a method of broad applicability to make
various aminocyclobutanes from N-vinyl b-lactams. This ap-
proach relies on readily accessible starting materials and
allows for rapid assembly of multiple rings fused to an ami-
nocyclobutane. Several stereocentres can be created in a
single-pot domino reaction with high stereoselectivity. A
computational study suggests that the reaction proceeds
through an eight-membered ring intermediate that contains
an E olefin, not a common feature amongst small-ring-
strain-release [3,3]-sigmatropic rearrangements.[26] Future
After screening the reaction conditions, we uncovered
that the reaction requires a temperature of roughly 1608C
for 30 min to achieve full conversion. The relatively high re-
action temperature is believed to be needed on the basis
that the ground-state conformation of the b-lactam is flat,
based on computational optimisation, where the carbon, ni-
trogen and oxygen atoms of the b-lactam core lie in the
Chem. Eur. J. 2010, 16, 4100 – 4109
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4107

L. L. W. Cheung and A. K. Yudin
Scheme 8. Computational investigation of the reaction mechanism using Gaussian 03[23] with a calculation method of RB3LYP and a basis set of 6-
31G(d).
work will encompass the synthetic utility of aminocyclobu-
tanes and will be reported in due course.
[1] a) B. Alcaide, P. Almendros, Synlett 2002, 0381–0393; b) B. Alcaide,
P. Almendros, C. Aragoncillo, Chem. Rev. 2007, 107, 4437–4492;
c) B. Alcaide, P. Almendros, Curr. Med. Chem. 2004, 11, 1921–1949.
[2] a) H. Gilman, H. Speeter, J. Am. Chem. Soc. 1943, 65, 2255–2256;
b) D. J. Hart, D. C. Ha, Chem. Rev. 1989, 89, 1447–1465; c) R. Graf,
Justus Liebigs Ann. Chem. 1963, 661, 111–157; d) H. Hoffmann,
H. J. Diehr, Tetrahedron Lett. 1963, 4, 1875–1879; e) H. J. Friedrich,
Tetrahedron Lett. 1971, 12, 2981–2984; f) M. Kinugasa, S. Hashimo-
to, J. Chem. Soc. Chem. Commun. 1972, 466–467; g) R. Pal, S. C.
Ghosh, K. Chandra, A. Basak, Synlett 2007, 2321–2330; h) H. Stau-
dinger, Justus Liebigs Ann. Chem. 1907, 356, 51–123; i) C. Palomo,
J. M. Aizpurua, I. Ganboa, M. Oiarbide, Eur. J. Org. Chem. 1999,
3223–3235; j) C. Palomo, J. M. Aizpurua, I. Ganboa, M. Oiarbide,
Curr. Med. Chem. 2004, 11, 1837–1872; k) J. Marco-Contelles,
Angew. Chem. 2004, 116, 2248–2250; Angew. Chem. Int. Ed. 2004,
43, 2198–2200.
Experimental Section
Representative procedure for aminocyclobutane synthesis: CuI (10
mol%) and Cs2CO3 (1.5 equiv) were charged into a Biotage Microwave
vial with a capacity of 0.2–0.5 mL equipped with a stirrer bar. A solution
of an N-vinyl b-lactam (20 mg) in DMF (450 mL) was added to the vial
and the vial was capped and crimped. Microwave heating was performed
for 30 min with the following settings: 1 min pre-stirring, absorption level
of high and with fix hold time on. The temperature used is listed with the
characterisation of the compound. After the microwave heating was com-
plete, the contents of the vial were emptied into a test tube, the vial was
then rinsed with water and CH2Cl2. The organic layer was separated and
the aqueous layer was washed two more times with CH2Cl2. The com-
bined organic layers were dried with magnesium sulfate, filtered, concen-
trated and purified by column chromatography.
[3] A. Bruggink in Synthesis of b-Lactam Antibiotics, Kuwer Academic
Publishers, Amsterdam, 2001, pp. 13–54.
[4] a) A. Fleming, Br. J. Exp. Path. 1929, 10, 226–236; b) A. R. Katritz-
ky, C. A. Ramsden, E. F. V. Scriven, R. J. K. Taylor, Comprehensive
Heterocyclic Chemistry III, Vol. 5, Elsevier, Oxford, 2008, pp. 111–
171.
[5] a) A. Bruggink in Synthesis of b-Lactam Antibiotics, Kuwer Aca-
demic Publishers, Amsterdam, 2001, p. 20; b) A. R. Katritzky, C. A.
Ramsden, E. F. V. Scriven, R. J. K. Taylor, Comprehensive Heterocy-
clic Chemistry III, Vol. 5, Elsevier, Oxford, 2008, pp. 173–237.
[6] J. D. Winkler, C. M. Bowen, F. Liotta, Chem. Rev. 1995, 95, 2003–
2020.
[7] J. Iriondo-Alberdi, M. F. Greaney, Eur. J. Org. Chem. 2007, 4801–
4815.
[8] Z. Rappoport, J. F. Lieman in The Chemistry of Cyclobutanes,
Part 1, Wiley, New York, 2005, pp. 294–302.
Acknowledgements
We thank NSERC for financial support. We also thank Professor Robert
Batey and his group for generously sharing his microwave reactor, Dr.
Alan Lough is acknowledged for his contribution to X-ray analysis, and
Dr. Ryan Hili is acknowledged for his contribution to the computational
study.
4108
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4100 – 4109

Highly Substituted Cyclobutane Fused-Ring Systems
FULL PAPER
[9] L. L. W. Cheung, A. K. Yudin, Org. Lett. 2009, 11, 1281–1284.
[10] E. J. Moriconi, W. C. Meyer, Tetrahedron Lett. 1968, 9, 3823–3827.
[11] T. Durst, M. J. OꢁSullivan, J. Org. Chem. 1970, 35, 2043–2044.
[12] a) S. S. Harried, C. P. Lee, G. Yang, T. I. H. Lee, D. C. Myles, J. Org.
Chem. 2003, 68, 6646–6660; b) B. M. Trost, M. T. Rudd, Org. Lett.
2003, 5, 4599–4602; c) D. H. R. Barton, G. Bashiardes, J. L. Pourrey,
Tetrahedron Lett. 1983, 24, 1605–1608; d) M. Blaskovicova, A. Ga-
plovsky, J. Blasko, Molecules 2007, 12, 188–193; e) N. Kamiya, Y.
Chikami, Y. Ishii, Synlett 1990, 675–676.
[20] P. Baillargeon, S. Bernard, D. Gauthier, R. Skouta, Y. L. Dory,
Chem. Eur. J. 2007, 13, 9223–9235.
[21] S. Sen, S. Singh, S. McN. Sieburth, J. Org. Chem. 2009, 74, 2884–
2886.
[22] M. E. Garst, L. J. Dolby, S. Esfandiari, R. A. Okrent, A. A. Avey, J.
Org. Chem. 2006, 71, 553–556.
[23] Gaussian 03, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
AHCTUNGTREGeNNUN ry, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
[13] L. Jiang, G. E. Job, A. Klapers, S. L. Buchwald, Org. Lett. 2003, 5,
3667–3669.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[14] a) B. Alcaide, C. Rodrꢂguez-Ranera, A. Rodrꢂguez-Vicente, Tetrahe-
dron Lett. 2001, 42, 3081–3083; b) P. Almendros, C. Aragoncillo, G.
Cabrero, R. Callejo, R. Carrascosa, A. Luna, T. M. del Campo,
M. C. Pardo, M. T. Quirꢃs, M. C. Redondo, C. Rodrꢂguez-Ranera,
A. Rodrꢂguez-Vicente M. P. Ruiz, Arkivoc 2010, III, 74–92.
[15] See the Supporting Information.
[16] a) P. Chen, Y. Chen, P. J. Carroll, S. M. Sieburth, Org. Lett. 2006, 8,
3367–3370; b) D. L. Comins, X. Zheng, R.R Goehring, Org. Lett.
2002, 4, 1611–1613; c) T. Chiba, Y. Takada, C. Kaneko, F. Kiuchi, Y.
Tsuda, Chem. Pharm. Bull. 1990, 38, 3317–3325; d) T. Chiba, Y.
Takada, T. Naito, C. Kaneko, Chem. Pharm. Bull. 1990, 38, 2335–
2337; e) C. Kaneko, K. Ychiyama, M. Sato, N. Katagiri, Chem.
Pharm. Bull. 1986, 34, 3658–3671; f) T. Naito, C. Kaneko, Chem.
Pharm. Bull. 1985, 33, 5328–5331; g) C. Kaneko, N. Katagiri, K.
Uchiyama, T. Yamada, Chem. Pharm. Bull. 1985, 33, 4160–4166.
[17] K. C. Nicolaou, D. J. Edmonds, P. G. Bulger, Angew. Chem. 2006,
118, 7292–7344; Angew. Chem. Int. Ed. Angew. Chem. Int. Ed.
Engl. 2006, 45, 7134–7186.
[24] S. M. Bachrach in Computational Organic Chemistry, Wiley, New
York, 2007, pp. 133–134, and references therein.
[25] I. Novak, P. J. Chua, J. Phys. Chem. A. 2006, 110, 10521–10524.
[26] a) J. M. Brown, B. T. Golding, J.J Stofko, J. Chem. Soc. Chem.
Commun. 1973, 319–320; b) E. Vogel, Annalen 1958, 615, 1–14;
c) M. S. Baird, C. B. Reese, J. Chem. Soc. D 1970, 1519–1520.
[18] a) L. F. Tietze, U. Beifuss, Angew. Chem. 1993, 105, 137–170;
Angew. Chem. Int. Ed. Engl. 1993, 32, 131–163; b) L. F. Tietze,
Chem. Rev. 1996, 96, 115–136; c) L. F. Tietze, G. Brasche, K. M.
Gericke in Domino Reactions in Organic Synthesis, Wiley-VCH,
Weinheim, 2006.
Received: October 5, 2009
Revised: December 29, 2009
Published online: February 23, 2010
[19] P. A. Wender, V. A. Verma, T. J. Paxton, T. H. Pillow, Acc. Chem.
Res. 2008, 41, 40–49.
Chem. Eur. J. 2010, 16, 4100 – 4109
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4109