Multicomponent reactions of cyclobutanones

Article abstract of DOI:10.1021/jo802170k

Cyclobutanones are essentially unknown as reactants in isonitrile-based multicomponent reactions. Ugi reactions of cyclobutanone and Passerini reactions of tetramethylcyclobutane-1,3-dione have been performed in this work. These reactions are significantly enhanced by being conducted in water, a subject of recent interest whose basis is still in question but whose effects are beyond doubt. The Ugi reaction of cyclobutanone has been used in a brief synthesis of an aspartame analogue.

Full text of DOI:10.1021/jo802170k

Multicomponent Reactions of Cyclobutanones
Michael C. Pirrung* and Jianmei Wang
Department of Chemistry, UniVersity of California, RiVerside, California 92521-0403
michael.pirrung@ucr.edu
ReceiVed September 28, 2008
Cyclobutanones are essentially unknown as reactants in isonitrile-based multicomponent reactions. Ugi
reactions of cyclobutanone and Passerini reactions of tetramethylcyclobutane-1,3-dione have been
performed in this work. These reactions are significantly enhanced by being conducted in water, a subject
of recent interest whose basis is still in question but whose effects are beyond doubt. The Ugi reaction
of cyclobutanone has been used in a brief synthesis of an aspartame analogue.
SCHEME 1
Introduction
Isonitrile-based multicomponent reactions have a long history
in organic synthesis.1 A wide variety of reactants participate in
the three-component Passerini reaction (acid, isonitrile, carbonyl)
and four-component Ugi reaction (acid, isonitrile, carbonyl,
amine), yet there is miniscule record of the use of cyclobu-
tanones in such reactions.2 There are two perspectives on the
potential reactivity of a cyclobutanone in such processes. As is
well-known,3 nucleophilic addition to cyclobutanones is faster
than for most other ketones. Such reactions benefit from relief
of strain in converting an sp2-hybridized carbon in a small ring
to a sp3-hybridized carbon. From a more modern perspective,
the LUMO energy of cyclobutanone is lower than that of most
other ketones (for example, 0.3 eV less than the cyclohexanone
LUMO4), increasing interactions between it and the isonitrile
HOMO in the transition state for the addition step. As the
protonated carbonyl (or imine) is the actual electrophile in
the multicomponent reactions of interest here, it seemed
worthwhile to consider the LUMO energies of those specific
reactive intermediates, comparing those derived from cyclobu-
tanone to those derived from a prototype reactant, cyclohex-
anone. These LUMO energies were calculated using the RHF/
6-31G* method in Spartan. Protonated cyclobutanone has a
LUMO energy (Scheme 1) 7.48 eV below that of cyclobutanone,
whereas protonated cyclohexanone has a LUMO energy 7.30
eV below that of cyclohexanone. Not only does cyclobutanone
have a lower LUMO energy to begin with, upon protonation
its LUMO energy decreases more than does cyclohexanone’s,
leading to a nearly 0.5 eV lower LUMO energy for protonated
cyclobutanone. These considerations project greater reactivity
of cyclobutanones in the Passerini reaction. The protonated
methyl imine of cyclobutanone has a LUMO energy (Scheme
1) 6.05 eV below that of cyclobutanone, whereas the protonated
methyl imine of cyclohexanone has a LUMO energy 5.98 eV
below that of cyclohexanone. Consideration of the FMO
interactions projects smaller differences in reactivity between
(1) (a) Do¨mling, A.; Ugi, I. Angew Chem., Int. Ed. 2000, 39, 3168. (b)
Do¨mling, A. Chem. ReV. 2006, 106, 17–89.
(2) Sorensen, B.; Rohde, J.; Wang, J.; Fung, S.; Monzon, K.; Chiou, W.;
Pan, L.; Deng, X.; Stolarik, D.; Frevert, E. U.; Jacobson, P.; Link, J. T. Bioorg.
Med. Chem. Lett. 2006, 16, 5958–5962.
(3) Brown, H. C.; Ichikawa, K. Tetrahedron 1957, 1, 221–230.
(4) Modelli, A.; Martin, H.-D. J. Phys. Chem. A 2002, 106, 7271–7275.
2958 J. Org. Chem. 2009, 74, 2958–2963
10.1021/jo802170k CCC: $40.75 2009 American Chemical Society
Published on Web 03/18/2009

Multicomponent Reactions of Cyclobutanones
cyclobutanone and cyclohexanone in the Ugi reaction. On the
other hand, cyclobutanone might be expected to react poorly
in the Ugi reaction owing to the need to generate a cyclobu-
tanone imine by elimination of water from a tetrahedral
intermediate. Converting an sp3-hybridized carbon to an sp2-
hybridized carbon in a small ring could hinder this step.
However, the sparse information available on the rate-limiting
step in the multistep mechanisms for Passerini and Ugi reactions
makes projection of the effectiveness of cyclobutanones in
multicomponent reactions difficult. Rate laws have been reported
for both the Passerini5 and Ugi6 reactions, but the quality of
the data and the kinetic mechanisms proposed in these studies
are still open to question.
We considered another potential reason for the lack of past
reports on isonitrile-based multicomponent reactions of cy-
clobutanones: that they follow an unusual reaction course that
did not allow the product(s) to be recognized. Shown in Scheme
1 is a hypothetical reaction that does not incorporate the acidic
reagent into the product. The addition of an isonitrile to a
protonated cyclobutanone carbonyl or imine would give nitri-
lium ion 1, as in traditional mechanisms. Since this intermediate
has an electrophilic center adjacent to a strained cyclobutane
bond, ring expansion (assisted by the unshared electron pair on
the heteroatom X, in a process similar to a pinacol rearrange-
ment) to 2 might be projected. Following deprotonation,
cyclopentane-1,2-dione monoimines or diimines could be
formed. Such compounds are not well-known, and their forma-
tion would represent a novel product type. However, contrary
to this hypothesis, we report here that two prototype cyclobu-
tanones participate effectively in Passerini and Ugi reactions.
The latter was used in the synthesis of a sweet dipeptide based
on 1-aminocyclobutanecarboxylic acid.
facilitate ring expansion. A similar ring expansion of this
compound is also known.9 However, reaction of tetramethyl-
cyclobutane-1,3-dione, 4-methoxybenzylamine, tert-butyl iso-
cyanide, and propionic acid in methanol or water gave neither
a ring-expansion product nor an Ugi product. Rather, the
Passerini product 4 is obtained in low yield (12%) in water,
and a complex reaction mixture is obtained in methanol. The
observation of Passerini side products in Ugi reactions is fairly
common and reflects a slow imine formation step that does not
compete with direct nucleophilic addition of the isonitrile to
the protonated carbonyl. When 4-methoxybenzylamine was
omitted and the reaction was performed as a straight Passerini
reaction in methanol, with 2 equiv of isonitrile and acid, 4 could
be obtained in low yield (12%), and reactants were still
detectable by GC after 2 d.
Because of our past experience in promoting multicomponent
reactions by conducting them in water, we aimed to make the
synthetically useless Passerini reaction of tetramethylcyclobu-
tane-1,3-dione with propionic acid more practical by performing
it in water and by using other acids. Some of these reactions
were also examined in methanol. Recent work from our
laboratory has suggested that reactions performed in water can
be affected by the hydrophobicity of the reactants.10 As a simple
and accessible measure of reactant hydrophobicity, we use the
log of the octanol/water partition coefficient or log P, which is
available for a wide variety of organic compounds, either from
compilation or calculation. We have observed that reactions that
experience a significant benefit from performing them in water
typically have reactant logP’s in the range of 1-2. Because
the log P of propionic acid is 0.35, we considered enhancing
the Passerini reaction of tetramethylcyclobutane-1,3-dione (log
P ) 1.69) by using more hydrophobic acid components. The
logP’s of the five carboxylic acids studied, cyclohexanecar-
boxylic, 3,3-dimethylacrylic, pivalic, adamantanecarboxylic, and
(adamantane)acetic, are given in Table 1 and range from 0.92
to 2.37. tert-Butyl isonitrile (log P ) 0.7) was used in all
reactions (2 equiv). We have also observed that reactions in
water are frequently dependent on the method of mixing,
attesting to their heterogeneous character.10 All reactions in this
study were performed with mixing by wrist-action shaking,
which we often find to be most efficient. These reactions were
performed at a ketone concentration of 0.1 M, at which all were
heterogeneous. For reactants with a low melting point, such as
pivalic or cyclohexanecarboxylic acid, the reaction mixture
appears cloudy, while for those with a high melting point, such
as the adamantanes, solid can be clearly observed. The outcomes
Results
Because the products of the putative ring-expansion process
do not incorporate an acid reactant, while an acid is required
for the Ugi and Passerini reactions, we thought that ring
expansion might be promoted by omitting acid completely.
When cyclobutanone, benzylamine, and tert-butyl isocyanide
are combined in methanol at room temperature, there is no
reaction. Thinking that acid might be necessary to catalyze initial
imine formation or promote addition of the isonitrile to the
imine, 10 mol % of propionic acid was added. However, the
only product observed (in 8% yield) is simply the Ugi product
3. When 1 equiv of propionic acid was used in this reaction, it
was complete after 24 h and 3 was obtained in 84% yield. When
the non-nucleophilic acid HBF4 was used, there was no reaction.
Past work in our laboratory has shown that multicomponent
reactions can be significantly accelerated by performing them
in water7 and that reactions can be observed when performed
in water that have no counterpart when conducted in organic
solvents.8 We therefore examined this reaction in water, but it
gives only Ugi product 3 in the same 84% yield as in methanol.
Another commercially available cyclobutanone, tetrameth-
ylcyclobutane-1,3-dione, was selected for study, reasoning that
its increased ring strain (two sp2-hybridized carbons in a small
ring) and migratory aptitude (tertiary migrating carbon) might
(5) Baker, R. H.; Stanonis, D. J. Am. Chem. Soc. 1951, 73, 699.
(6) Ugi, I.; Kaufhold, G. Liebigs Ann. Chem. 1967, 709, 11–28.
(7) (a) Pirrung, M. C.; Sarma, K. D. J. Am. Chem. Soc. 2004, 126, 444. (b)
Pirrung, M. C.; Sarma, K. D. Tetrahedron 2005, 61, 11456.
(8) Pirrung, M. C.; Das Sarma, K. Synlett 2004, 1425–1427.
(9) Mloston´, G.; Roman´ski, J.; Linden, A.; Heimgartner, H. HelV. Chim. Acta
1999, 82, 1302.
(10) Pirrung, M. C.; Das Sarma, K.; Wang, J. J. Org. Chem. 2008, 73, 8723.
J. Org. Chem. Vol. 74, No. 8, 2009 2959

Pirrung and Wang
TABLE 1. Passerini Reactions of Tetramethylcyclobutane-1,
be increased by using more acid, speeding addition to 11,
reactions with 5 and 10 equiv of pivalic acid were also
conducted in water. When 5 equiv of acid is used, the reaction
affords the two products in a similar ratio. When 10 equiv of
acid is used, the ratio of the Passerini product 8 to the hydrolysis
product 7 is slightly increased to around 1:1. Reaction with
adamantane-1-carboxylic acid, a more hydrophobic carboxylic
acid, is very slow in methanol and incomplete after 9 d. In water,
the reaction is complete after 5 d. The Passerini product 9 is
obtained in 60% yield, and the side product 7 is obtained in
12% yield. The reaction with (1-adamantane)acetic acid in
methanol is incomplete after 9 d. The reaction in water is
complete after 5 d, and the Passerini product 10 is obtained in
almost quantitative yield.
3-dione, tert-Butylisonitrile, and Carboxylic Acids RCO2H
acid
acid time
yield yield of
acid R group
cyclohexyl
cyclohexyl
2-methylprop-1-enyl 0.92 MeOH
2-methylprop-1-enyl 0.92 H2O
tert-butyl
tert-butyl
tert-butyl
adamantyl
log P solvent (equiv) (d) product (%) 7 (%)
1.67 MeOH
1.67 H2O
2
2
4
3
4
3
5
5
5
5
5
5
5
40
85
13
31
20
21
30
60
98
0
0
0
50
39
44
34
12
0
2
6
2
6
1.62 H2O
1.62 H2O
1.62 H2O
2.37 H2O
2.31 H2O
2
5
10
2
2
8
8
8
9
(adamantyl)methyl
10
CHART 1
As an example of the utility of the Ugi reaction of cyclobu-
tanone, we used this process as the key step in a synthesis of a
simple dipeptide. Since the discovery of aspartame as a peptide
sweetener,11 structure-taste studies have shown that the aspartic
acid residue is absolutely essential for sweetness.11 However,
the C-terminal residue can vary significantly. Studies have
shown that aspartic dipeptides derived from amino acids with
small (three- to five-membered) cycloalkanes are sweet, whereas
those with larger rings are not.12 Aspartyl-R-aminocyclobutan-
ecarboxylic acid methyl ester has been reported as a sweet
dipeptide by Goodman. His five-step synthesis uses cyclobu-
tanone as the starting material and pursues a classical approach
based on the Bucherer-Bergs reaction and traditional peptide
couplings, affording the target in 8% overall yield.
Recognizing the potential of the Ugi reaction of cyclobu-
tanone to assemble the full skeleton of this dipeptide, we
required an isonitrile whose Ugi product (an amide) could be
readily converted to the methyl ester. Many solutions to the
problem of so-called convertible isonitriles have been offered,13
including one from our own laboratory.14 Several of these were
tested in Ugi reactions with cyclobutanone, but most often their
reactivity was poor and the Ugi product was obtained in only
modest yield. We find wide variance in the reactivity of the
reported convertible isonitriles, with the most reactive in this
case being 1-isocyanocyclohexene. The relative nucleophilicity
of isonitriles has been recently investigated, with tert-butyl
isonitrile found to be the most nucleophilic among a limited
set of five isonitriles.15 We have also used water to promote
Ugi reactions,7 but here this strategy did not provide a benefit.
In further developing this synthetic plan, we realized a better
way to obtain 1-isocyanocyclohexene. Several previous prepara-
tions of this compound are known.16 Armstrong prepared it by
dehydration of N-cyclohex-1-enylformamide, but his synthesis
of all of these reactions are summarized in Table 1, and their
Passerini products are given in Chart 1.
The Passerini reaction of tetramethylcyclobutane-1,3-dione
and tert-butyl isonitrile with cyclohexanecarboxylic acid in
methanol gives the desired product 5 in 40% yield, with
reactants still being detected by TLC after 4 d. In water, the
reaction is complete after 3 d and the product is obtained in
85% yield. The theory that the Passerini reaction could be
enhanced with more hydrophobic acid reactants in water was
confirmed. Reaction with 3,3-dimethylacrylic acid in methanol
gives 6 in only low yield after 4 d. In water, the reaction is
complete after 3 d and the product 6 is obtained in modest yield.
The dominant product of this reaction is compound 7, which is
an abnormal Passerini reaction product. We envision this
compound arising in one of two ways (eq 3). Intermediate
nitrilium ion 11 could undergo direct reaction with water to
give the R-hydroxyamide 7. Alternatively, if 11 instead reacts
with the carboxylate to give the imidic anhydride 12, it might
still be hydrolyzed by water intermolecularly rather than
undergoing an intramolecular acyl-transfer reaction to give 6.
Data on other carboxylic acids discussed below may bear on
the pathway that is followed.
(11) Mazur, R. H.; Schlatter, J. M.; Goldkamp, A. H. J. Am. Chem. Soc.
1969, 91, 2684.
(12) Tsang, J. W.; Schmied, B.; Nyfeler, R.; Goodman, M. J. Med. Chem.
1984, 27, 1663.
(13) (a) Keating, T. A.; Armstrong, R. W. J. Am. Chem. Soc. 1995, 117,
7842. (b) Keating, T. A.; Armstrong, R. W. J. Am. Chem. Soc. 1996, 118, 2574.
(c) Gilley, C. B.; Buller, M. J.; Kobayashi, Y. Org. Lett. 2007, 9, 3631–4. (d)
Gilley, C. B.; Kobayashi, Y. J. Org. Chem. 2008, 73, 4198–4204. (e) Lindhorst,
T.; Bock, H.; Ugi, I. Tetrahedron 1999, 55, 7411–7420. (f) Linderman, R. J.;
Binet, S.; Petrich, S. R. J. Org. Chem. 1999, 64, 336–337. (g) Kennedy, A. L.;
Fryer, A. M.; Josey, J. A. Org. Lett. 2002, 4, 1167–70. (h) Rikimaru, K.;
Yanagisawa, A.; Kan, T.; Fukuyama, T. A. Synlett 2004, 41–43.
(14) Pirrung, M. C.; Ghorai, S. J. Am. Chem. Soc. 2006, 128, 11772.
(15) Tumanov, V. V.; Tishkov, A. A.; Mayr, H. Angew. Chem., Int. Ed. 2007,
46, 3563.
(16) (a) Ugi, I.; Rosendahl, K. Justus Liebigs Ann. Chem. 1963, 66, 65–7.
(b) Baldwin, J. E.; Yamaguchi, Y. Tetrahedron Lett. 1989, 30, 3335–8. (c)
Creedon, S. M.; Crowley, H. K.; McCarthy, D. G. J. Chem. Soc., Perkin Trans.
1 1998, 1015–1018. (d) Porcheddu, A.; Giacomelli, G.; Salaris, M. J. Org. Chem.
2005, 70, 2361–2363.
Reaction with pivalic acid is very slow in methanol and
incomplete after 9 d, whereas reaction in water is complete after
5 d. When standard conditions are used with 2 equiv of acid, 8
is still the minor product, and 7 is obtained in 39% yield. To
see if the ratio of the Passerini product to the side product could
2960 J. Org. Chem. Vol. 74, No. 8, 2009

Multicomponent Reactions of Cyclobutanones
SCHEME 2
SCHEME 3
CHART 2
of that compound is not the most efficient. The best method to
synthesize N-cyclohex-1-enylformamide is by acid-catalyzed
condensation of formamide and cyclohexanone, very inexpen-
sive starting materials,17 whereas in our hands Armstrong’s
method proved best for the dehydration. 1-Isocyanocyclohexene
was thereby conveniently synthesized in two steps in moderate
yield by combining two known procedures that had not
heretofore been used together.
Discussion
One interesting aspect of an Ugi reaction of cyclobutanone
(eq 1) is that it is accelerated when performed in water. We
have previously suggested that the log P of reactants may
indicate whether a particular reaction will be accelerated in
water.10 Polar, water-miscible compounds may not experience
the hydrophobic effect that plays a role in the rate acceleration.
The log P of cyclobutanone itself is only 0.59, likely too low
to benefit from the aqueous reaction medium because it is water-
miscible. However, we have previously shown that Ugi reactions
with compounds even as polar as amino acids can benefit from
being performed in water. We account for this observation based
on the initial formation of an imine intermediate that is
significantly more hydrophobic. The same consideration applies
to the condensation of cyclobutanone with benzylamine, giving
an imine with a log P of 2.78.
Passerini reactions of tetramethylcyclobutane-1,3-dione, tert-
butyl isocyanide, and hydrophobic carboxylic acids are acceler-
ated in water compared to methanol. However, the side product
7 is formed in some reactions. We envisioned that we might be
able to “drive” the acid to react with the nitrilium ion by using
more hydrophobic acids in the reactions in water. However,
the data in Table 1 do not support a simple picture in which
acid hydrophobicity determines the partitioning between Pas-
serini products and 7. The proportion of the Passerini product
could be increased slightly by the addition of a larger excess of
acid, but this ploy still does not surmount hydrolysis. That
increasing the rate of nucleophilic addition of the acid to
nitrilium ion 11 does not favor the formation of the Passerini
product at the expense of the R-hydroxyamide suggests that
water does not compete with the acid at this step to cause
formation of 7. That leaves the internal acyl-transfer reaction
in 12 as the step where hydrolysis competes with formation of
the desired product. The formation of 7 was suppressed by using
a less sterically hindered (but also more hydrophobic) carboxylic
acid, (adamantyl)acetic acid. Its hydrophobicity is comparable
to adamantanecarboxylic acid, whose acyl-transfer reaction may
be less facile because of the quaternary center flanking the
carbonyl group. Similar hydrophobicity is exhibited by cyclo-
hexanecarboxylic acid, which also gives a high yield. Pivalic
acid and adamantanecarboxylic acid are more hydrophobic so
as to encourage the addition step but too sterically hindered for
a facile acyl transfer reaction to give the Passerini product, so
The Ugi reaction uses 2,4-dimethoxybenzylamine as an easily
deprotectable ammonia equivalent. The other reactants are
cyclobutanone, Cbz-protected aspartic acid benzyl ester, and
1-isocyanocyclohexene. The Ugi reaction was performed in
methanol, and the desired product 13 is obtained in good yield.
In a previous incarnation of this synthesis, an Fmoc-protected
aspartic acid was used.18 Though this route failed owing to
selectivity issues in deprotection steps, it taught a novel method
to manipulate Ugi products similar to 13. We initially converted
the cyclohexenamide to the methyl ester (HCl/MeOH) and then
removed the DMB group, done as usual with TFA.19 To
telescope two steps into one, we treated the Ugi product with
TFA directly. Not only was the DMB group removed, but the
amide was also hydrolyzed and the carboxylic acid was obtained
upon quenching the reaction with water. When the reaction was
quenched with methanol instead, the methyl ester was obtained,
but in low yield. The addition of pyridine as a non-nucleophilic
base to neutralize the excess TFA in the system increased the
yield of methyl ester greatly. When this procedure was applied
to 13, the desired product 16 is obtained in acceptable yield,
but the product proved to be racemic. We understand this
reaction pathway as follows. Armstrong has shown that the
utility of Ugi products of 1-isocyanocyclohexene derives from
their ability to cyclize to oxazolones, which are fairly reactive
acylating agents, inter alia. In the case at hand, such intermedi-
ates would correspond to 14 or 15 (depending on whether the
DMB group was removed before or after the formation of the
oxazolone). These intermediates presumably labilize the adjacent
hydrogen for deprotonation, and when they are quenched with
water, the carboxylic acid is formed, whereas if they are
quenched with alcohol, the ester is formed.
Hydrogenolysis to remove the Cbz and benzyl groups
proceeds in excellent yield following Goodman’s procedure,
completing the synthesis. The spectroscopic and physical
properties match those reported by Goodman, establishing a brief
route to cyclobutane dipeptide amides.
(17) Maison, W.; Schlemminger, I.; Westerhoff, O.; Martens, J. Bioorg. Med.
Chem. 2000, 8, 1343.
(18) Wang, J. Ph.D. Thesis, Duke University, 2007.
(19) Besada, P.; Mamedova, L.; Thomas, C. J.; Costanzi, S.; Jacobson, K. A.
Org. Biomol. Chem. 2005, 3, 2016.
J. Org. Chem. Vol. 74, No. 8, 2009 2961

Pirrung and Wang
1774, 1740, 1684, 1513, 1452, 1368, 1314, 1288, 1247, 1218, 1155,
1125, 1066, 1030, 895, 830, 758 cm-1. HRMS (ESI): calcd for
C20H34NO4 [M + H]+ 352.2487, found 352.2488. Mp: 83 °C.
3-Methyl-but-2-enoic Acid 1-tert-Butylcarbamoyl-2,2,4,4-tet-
ramethyl-3-oxocyclobutyl Ester (6). Tetramethyl-1,3-cyclobu-
tanedione (14 mg, 0.10 mmol), tert-butyl isocyanide (23 µL, 0.20
mmol), and 3,3-dimethylacrylic acid (20 mg, 0.20 mmol) were
added to a glass vial (15 × 45 mm). H2O (1 mL) was then added.
The reaction mixture was mixed by shaking at room temperature
for 3 d and was extracted with CH2Cl2 (1 mL × 2). The organic
layer was washed with brine and dried over anhydrous Na2SO4.
The solvent was removed in vacuo, and the product was purified
by flash chromatography (ethyl acetate/hexanes 1:5) to give 6 as a
white solid (10 mg, 31%) and 7 as a white solid (12 mg, 50%). 6:
the imidic anhydride in these cases is vulnerable to hydrolysis
before it can rearrange.
It is surprising that Passerini reactions on tetramethylcy-
clobutane-1,3-dione proved possible. There are no other ex-
amples of Passerini reactions with ketones that are quaternary
at both R carbons. The success of these reactions can be
attributed to several factors, including the especially low LUMO
energy of tetramethylcyclobutane-1,3-dione, which is even 0.83
eV lower than cyclobutanone.4 Other factors favoring this
process include the use of water as the solvent and the near-
ideal ketone hydrophobicity (log P ) 1.69) to exploit the
hydrophobic effect in rate acceleration. It is also interesting that
double-addition products are never observed despite using an
excess of acid and isonitrile. This presumably reflects significant
resistance of the initial Passerini product to addition to the
remaining carbonyl, which must be sterically hindered by the
other three quaternary ring carbons.
1
Rf ) 0.36 (ethyl acetate/hexanes 1:5). H NMR (CDCl3): δ 5.83
(m, 1H), 5.48 (s, 1H), 2.21 (d, J ) 1.2 Hz, 3H), 1.98 (d, J ) 1.2
Hz, 3H), 1.43 (s, 6H), 1.31 (s, 9H), 1.29 (s, 6H). 13C NMR (CDCl3):
δ 217.7, 168.1, 165.4, 160.5, 114.3, 82.3, 63.9, 51.3, 28.7, 27.5,
21.3, 20.4, 19.8. IR (neat): 3407, 2974, 2922, 1784, 1721, 1668,
1646, 1514, 1453, 1380, 1364, 1287, 1219, 1135, 1067, 1036, 931,
843, 827, 763, 734 cm-1. HRMS (ESI): calcd for C18H30NO4 [M
+ H]+ 324.2174, found 324.2176. Mp: 88 °C. 7: Rf ) 0.49 (ethyl
In summary, cyclobutanones have proved viable reaction
partners in multicomponent reactions of isonitriles. The dipeptide
synthesis also provided a new protocol for converting the
products of Ugi reactions with 1-isocyanocyclohexene to other
acyl derivatives with the simultaneous removal of an N-
protecting group.
1
acetate/hexanes 1:2). H NMR (CDCl3): δ 6.78 (s, 1H), 2.57 (s,
1H), 1.37 (s, 9H), 1.32 (s, 6H), 1.31 (s, 6H). 13C NMR (CDCl3): δ
217.6, 169.8, 78.3, 63.5, 51.1, 28.8, 20.8, 20.1. IR (neat): 3403,
3364, 2971, 2925, 1773, 1651, 1525, 1457, 1364, 1328, 1249, 1227,
1203, 1167, 1122, 1055, 1024, 991, 979, 909, 831, 761, 725, 664
cm-1. HRMS (ESI): calcd for C13H24NO3 [M+H]+ 242.1756, found
242.1761. Mp: 146 °C.
Experimental Section
2,2-Dimethylpropionic Acid 1-tert-Butylcarbamoyl-2,2,4,4-
tetramethyl-3-oxocyclobutyl Ester (8). This compound was
synthesized as above using trimethylacetic acid. The product
was obtained as a white solid (7.0 mg, 20%). Side product 7 was
afforded as a white solid (10 mg, 39%). 8: Rf ) 0.40 (ethyl acetate/
hexanes 1:5). 1H NMR (CDCl3): δ 5.32 (s, 1H), 1.35 (s, 6H), 1.25
(s, 9H), 1.25 (s, 9H), 1.23 (s, 6H). 13C NMR (CDCl3): δ 217.3,
177.2, 167.8, 82.6, 64.2, 51.6, 39.4, 28.8, 27.3, 21.5, 19.9. IR (neat):
3418, 2972, 2932, 1770, 1745, 1721, 1680, 1519, 1451, 1379, 1365,
1281, 1231, 1220, 1187, 1125, 1072, 1030, 963, 877, 849, 830,
794, 771, 727 cm-1. HRMS (ESI): calcd for C18H32NO4 [M + H]+
326.2331, found 326.2320. Mp: 99 °C.
Adamantane-1-carboxylic Acid 1-tert-Butylcarbamoyl-2,2,4,4-
tetramethyl-3-oxocyclobutyl Ester (9). Compound 9 was synthe-
sized as a colorless oil (24 mg, 60%) as in example 6 starting with
1-adamantanecarboxylic acid. Rf ) 0.49 (ethyl acetate/hexanes 1:5).
1H NMR (CDCl3): δ 5.38 (s, 1H), 2.07 (brs, 3H), 1.96 (m, 6H),
1.75 (m, 6H), 1.40 (s, 6H), 1.31 (s, 9H), 1.28 (s, 6H). 13C NMR
(CDCl3): δ 217.4, 176.3, 167.7, 82.2, 64.0, 51.4, 41.2, 39.1, 36.3,
28.8, 27.8, 21.5, 19.7. IR (neat): 3448, 2969, 2907, 2853, 1781,
1738, 1684, 1506, 1453, 1365, 1216, 1182, 1103, 1067, 1032, 978,
889, 744, 686 cm-1. HRMS (ESI): calcd for C24H38NO4 [M + H]+
404.2800, found 404.2806. Side product 7 was afforded as a white
solid (3.0 mg, 12%).
1-(Benzylpropionylamino)cyclobutanecarboxylic Acid tert-
Butylamide (3). Cyclobutanone (7.4 µL, 0.10 mmol), benzylamine
(11 µL, 0.10 mmol), tert-butyl isocyanide (11 µL, 0.10 mmol), and
propionic acid (7.5 µL, 0.10 mmol) were added into a glass vial
(15 × 45 mm). Solvent (MeOH or H2O) (1 mL) was then added.
The reaction mixture was mixed by magnetic stirring or shaking at
room temperature overnight. Workup procedures: For reaction in
MeOH: The reaction mixture was concentrated, and the product
was purified by flash chromatography (ethyl acetate/hexanes 1:5)
to give 3 as a white solid (27 mg, 85%). For reaction in H2O: The
reaction mixture was extracted with CH2Cl2 (1 mL × 2). The
organic layer was washed with brine and dried over anhydrous
Na2SO4. The solvent was removed in vacuo, and the product was
purified by flash chromatography (ethyl acetate/hexanes 1:5) to give
3 as a white solid (25 mg, 80%). Rf ) 0.19 (ethyl acetate/hexanes
1
1:5). H NMR (CDCl3): δ 7.20-7.39 (m, 5H), 4.45 (s, 2H), 2.68
(m, 2H), 2.15-2.40 (m, 4H), 1.70 (m, 2H), 1.34 (s, 9H), 1.06 (t, J
) 6.9 Hz, 3H). 13C NMR (CDCl3): δ 175.9, 172.4, 138.1, 128.9,
127.2, 125.7, 66.6, 50.7, 49.1, 31.5, 28.7, 27.6, 14.6, 9.4. IR (neat):
3317, 2966, 2871, 1666, 1631, 1532, 1450, 1423, 1363, 1257, 1230,
1208, 1171, 1076, 1029, 972, 836, 815, 721, 695 cm-1. HRMS
(ESI): calcd for C19H29N2O2 [M + H]+ 317.2229, found 317.2217.
Mp: 91 °C (obtained with a Bu¨chi B-545 instrument using the
melting point method).
Adamantan-1-ylacetic Acid 1-tert-Butylcarbamoyl-2,2,4,4-
tetramethyl-3-oxocyclobutyl Ester (10). Compound 10 was
synthesized as a colorless oil (41 mg, 98%) as in example 6 starting
with 1-adamantaneacetic acid. Rf ) 0.45 (ethyl acetate/hexanes 1:5).
1H NMR (CDCl3): δ 5.47 (s, 1H), 2.23 (s, 2H), 2.01 (m, 3H),
1.61-1.78 (m, 12H), 1.42 (s, 6H), 1.34 (s, 9H), 1.29 (s, 6H). 13C
NMR (CDCl3): δ 217.3, 171.0, 167.9, 83.0, 64.0, 51.6, 48.0, 42.3,
36.6, 32.9, 28.8, 28.4, 21.5, 19.9. IR (neat): 3447, 3401, 2903, 2849,
1780, 1744, 1683, 1505, 1453, 1366, 1284, 1218, 1190, 1125, 1064,
1032, 980, 888, 829, 731 cm-1. HRMS (ESI): calcd for C25H40NO4
[M + H]+ 418.2957, found 418.2946.
3-Benzyloxycarbonylamino-N-[1-(cyclohex-1-enylcarbamoyl)-
cyclobutyl]-N-(2,4-dimethoxybenzyl)succinamic Acid Benzyl
Ester (13). Z-ꢀ-(Benzyl ester)-L-aspartic acid (36 mg, 0.10 mmol),
2,4-dimethoxybenzylamine (15 µL, 0.10 mmol), cyclobutanone (8.0
µL, 0.10 mmol), and 1-isocyanocyclohexene (0.10 mL of a 1 M
Cyclohexanecarboxylic Acid 1-tert-Butylcarbamoyl-2,2,4,4-
tetramethyl-3-oxocyclobutyl Ester (5). Tetramethyl-1,3-cyclobu-
tanedione (14 mg, 0.10 mmol), tert-butyl isocyanide (23 µL, 0.20
mmol), and cyclohexanecarboxylic acid (26 mg, 0.20 mmol) were
added into a glass vial (15 × 45 mm). H2O (1 mL) was then added.
The reaction mixture was mixed by shaking at room temperature
for 3 d and extracted with CH2Cl2 (1 mL × 2). The organic layer
was washed with brine and dried over anhydrous Na2SO4. The
solvent was removed in vacuo, and the product was purified by
flash chromatography (ethyl acetate/hexanes 1:5) to give 5 as a
white solid (30 mg, 85%). Rf ) 0.44 (ethyl acetate/hexanes 1:5).
1H NMR (CDCl3): δ 5.41 (s, 1H), 2.36 (tt, J ) 11.1, 3.6 Hz, 1H),
1.94-2.39 (m, 2H), 1.79-1.87 (m, 2H), 1.45-1.77 (m, 4H), 1.41
(s, 6H), 1.35-1.39 (m, 1H), 1.32 (s, 9H), 1.29 (s, 6H), 1.19-1.26
(m, 1H). 13C NMR (CDCl3): δ 217.3, 174.9, 167.7, 82.5, 64.0, 51.4,
43.4, 29.2, 28.7, 25.5, 25.3, 21.4, 19.8. IR (neat): 3410, 2932, 2854,
2962 J. Org. Chem. Vol. 74, No. 8, 2009

Multicomponent Reactions of Cyclobutanones
solution in pentane, 0.10 mmol) were dissolved in 1 mL of MeOH
in a 10 mL round-bottom flask with a stir bar. The reaction mixture
was stirred at room temperature overnight. The reaction mixture
was concentrated in vacuo. The product was purified by flash
chromatography (ethyl acetate/hexanes 1:2) to afford 13 as a white
solid (51 mg, 75%). Rf ) 0.37 (ethyl acetate/hexanes 1:2). 1HNMR
(CDCl3): δ 8.10 (s, 1H), 7.34 (m, 10H), 7.00 (d, J ) 8.4 Hz, 1H),
7.39 (dd, J ) 8.7, 1.8 Hz, 1H), 6.32 (d, J ) 1.8 Hz, 1H), 5.11-5.75
(m, 2H), 5.11 (s, 2H), 5.07 (s, 2H), 4.53 (d, J ) 15.9 Hz, 1H),
4.34 (d, J ) 16.5 Hz, 1H), 3.76 (s, 3H), 3.64 (s, 3H), 2.54-2.86
(m, 4H), 1.95-2.24 (m, 4H), 1.50-1.65 (m, 8H). 13C NMR
(CDCl3): δ 172.3, 170.7, 170.5, 161.1, 158.4, 155.5, 136.4, 135.6,
133.0, 130.0, 128.8, 128.7, 128.5, 128.5, 128.4, 128.2, 116.1, 111.7,
104.0, 98.6, 67.2, 67.0, 66.4, 55.5, 55.1, 49.4, 45.3, 38.1, 32.0, 30.8,
27.9, 24.1, 22.8, 22.3, 14.9. IR (neat): 3300, 2938, 2837, 1723,
1683, 1614, 1589, 1507, 1455, 1438, 1383, 1287, 1259, 1208, 1158,
1029 cm-1. HRMS (ESI): calcd for C39H46N3O8 [M + H]+
4.2 Hz, 1H), 2.73 (dd, J ) 16.8, 6.3 Hz, 1H), 2.55-2.66 (m, 2H),
2.27 (m, 2H), 1.90-2.10 (m, 2H). 13C NMR (CDCl3): δ 173.6,
171.8, 170.2, 156.2, 136.2, 135.5, 128.8, 128.8, 128.6, 128.5, 128.5,
128.3, 67.5, 67.1, 58.6, 52.7, 51.0, 36.4, 31.5, 31.5, 15.5. IR (neat):
3320, 2953, 1728, 1667, 1498, 1455, 1386, 1312, 1214, 1168, 1123,
1047, 1028, 984, 911, 844, 737, 696 cm-1. HRMS (ESI): calcd for
C25H29N2O7 [M + H]+ 469.1969, found 469.1961.
Aspartyl-r-aminocyclobutanecarboxylic Acid Methyl Ester.
Compound 16 (23 mg, 50 µmol) was dissolved in MeOH (1 mL).
Nitrogen was bubbled into the solution for 10 min. Palladium black
(10 wt % on carbon powder, 10 mg) was added at once. Nitrogen
was bubbled through for an additional 5 min, and hydrogen was
bubbled through the solution for 4 h. After filtration, the solvent
was removed in vacuo. The product was purified by recrystallization
from MeOH/ether to afford the title compound as a white solid
(11 mg, 90%). 1H NMR (acetone-d6): δ 4.05 (dd, J ) 8.7, 3.9 Hz,
1H), 3.71 (s, 3H), 2.49-2.78 (m, 4H), 2.15-2.34 (m, 2H),
1.98-2.10 (m, 2H). 13C NMR (acetone-d6): δ 176.3, 175.0, 170.1,
59.8, 53.1, 52.3, 38.4, 32.4, 32.3, 16.6. IR (neat): 3257, 2951, 1743,
1663, 1558, 1379, 1323, 1260, 1203, 1164, 1121, 1068, 993, 921,
802, 773 cm-1. HRMS (ESI): calcd for C10H17N2O5 [M + H]+
245.1137, found 245.1136. Mp: 115 °C dec (lit.12 mp 112-113
°C dec). This is a known compound, and these properties were
identical to those published.
684.3285, found 684.3318. Mp: 51 °C. [R]25 - 20.66 (c 0.75,
D
MeOH).
1-(3-Benzyloxycarbonyl-2-benzyloxycarbonylaminopropio-
nylamino)cyclobutanecarboxylic Acid Methyl Ester (16). TFA
(0.80 mL) was added to a solution of compound 13 (55 mg, 80
µmol) in methylene chloride (5 mL) in a 25 mL round-bottom flask.
The resulting mixture (purple solution) was stirred at room
temperature for 1 h. MeOH (4.0 mL) and anhydrous pyridine (0.90
mL) were added to the reaction mixture. The solution turned cloudy
yellow. The reaction mixture was stirred at room temperature for
1 h and concentrated in vacuo. Methylene chloride (20 mL) was
added to dissolve the residue. The organic solution was washed
with water (2 × 10 mL), dried over Na2SO4, and then concentrated
in vacuo. The product was purified by flash chromatography (ethyl
acetate/hexanes 1:2) to afford 16 as a colorless oil (24 mg, 65%).
This is a known compound.12 Rf ) 0.22 (ethyl acetate/hexanes 1:2).
1H NMR (CDCl3): δ 7.31-7.37 (m, 10H), 7.04 (s, 1H), 5.92 (m,
1H), 5.13 (m, 4H), 4.61 (m, 1H), 3.70 (s, 3H), 3.06 (dd, J ) 17.1,
Acknowledgment. We thank Prof. Richard Hooley for
providing a key reference and Dr. Tannya Ibarra for obtaining
some characterization data.
Supporting Information Available: 1H NMR and 13C NMR
spectra for 3, 5-10, 13, 16, and aspartyl-R-amino-cyclobutan-
ecarboxylic acid methyl ester. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO802170K
J. Org. Chem. Vol. 74, No. 8, 2009 2963