First practical synthesis of formamidine ureas and derivatives

Article abstract of DOI:10.1021/ol034287r

(Matrix presented) Isonitriles and ureas undergo a condensation reaction in the presence of acid chlorides to give formamidine ureas, for which no general synthetic routes currently exist. A mechanism is proposed in which the key intermediate is an electrophilic adduct of isonitrile and acid chloride. The process is tolerant of moderate variability in the nature of the components, and access to formamidine ureas of varying substitution patterns is further enhanced by a facile exchange reaction with amines.

Full text of DOI:10.1021/ol034287r

ORGANIC
LETTERS
2003
Vol. 5, No. 9
1531-1533
First Practical Synthesis of Formamidine
Ureas and Derivatives
Amy S. Ripka,* David D. D´ıaz, K. Barry Sharpless, and M. G. Finn*
Department of Chemistry and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 N. Torrey Pines Rd., La Jolla, California 92037
mgfinn@scripps.edu
Received February 18, 2003
ABSTRACT
Isonitriles and ureas undergo a condensation reaction in the presence of acid chlorides to give formamidine ureas, for which no general
synthetic routes currently exist. A mechanism is proposed in which the key intermediate is an electrophilic adduct of isonitrile and acid
chloride. The process is tolerant of moderate variability in the nature of the components, and access to formamidine ureas of varying substitution
patterns is further enhanced by a facile exchange reaction with amines.
Ureas are important components of biologically active
molecules, having greater hydrogen bonding potential than
amides while being less acidic than sulfonamides.1 They have
found use as amide bond surrogates,2 allowing for a â-sheet-
like display of functionality.3 The development of new urea
derivatives is therefore of practical interest. Here we report
a novel and convenient condensation reaction that assembles
formamidine ureas from readily available precursors.
Figure 1 outlines the process, in which the addition of a
substituted urea to a mixture of isocyanide and acid chloride
gives formamidine urea salts 1 in pure form. The structure
of 1a was established by X-ray crystallography (Figure 1)
as well as appropriate spectroscopic data. The mother liquor
contains N-acylureas 2, identified by comparison to authentic
compounds, in amounts approximately equimolar to the
formamidine precipitates.4
most cases the solid formamidine hydrochloride salts can
be isolated by filtration and washing to remove excess urea
The yields of 1 are maximized by the use of 3 equiv of
urea, and THF provides the best results in general (except
when the urea is not soluble, in which case acetonitrile is
often preferred).4 The reaction is easy to perform, and in
(1) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
(2) (a) Burgess, K.; Linthicum, D. S.; Shin, H. Angew. Chem., Int. Ed.
Engl. 1995, 34, 907-909. (b) Burgess, K.; Ibarzo, J.; Linthicum, D. S.;
Russell, D. H.; Shin, H.; Shitangkoon, A.; Totani, R.; Zhang, A. J. J. Am.
Chem. Soc. 1997, 119, 1556-1564
Figure 1. (Top) Formamidine and N-acyl urea formation. (Middle)
X-ray crystal structure of 1a. (Bottom) Formamidine derived from
a hydrazide instead of urea.
(3) Nowick, J. S. Acc. Chem. Res. 1999, 32, 287-296.
(4) See Supporting Information.
10.1021/ol034287r CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/29/2003

and the soluble byproduct 2. With a standard set of reaction
conditions, the process was found to be tolerant of substantial
variations in the isocyanide and urea components, as shown
in Table 1.
benzoyl chloride). Most interestingly, triphenylmethyl (trityl)
chloride promotes the reaction, albeit in modest yield (37%).
A proposed mechanism is shown in Scheme 1. In the
chemistry of Livinghouse and co-workers, isocyanide-acyl
Scheme 1
Table 1. Isolated Yields of Formamidine Urea Adducts
Formed under a Standard Set of Reaction Conditionsa
%
entry
R1
R2
Me
Me
R3 R4 solvent product yield
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
CH2Ph
Me
Me
Me
Me
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
THF
THF
THF
THF
THF
THF
MeCN
MeCN
MeCN
MeCN
MeCN
THF
THF
THF
MeCN
MeCN
1a
1b
1c
1d
1e
1fb
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
73
71
68
71
79
52
77
38c
27c
37c
23c
77
68
66
72g
49g
t-Bu
cyclohexyl Me
CH2CO2Me Me
n-Bu
Me
Me
-CH2CH2-
Me
Add
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2SO2Are Me
-(CH2)6-f Me
2,6-Me2Ph Me
H
H
H
H
Me
Me
Me
allyl
H
Ph
CH2Ph
CH2Ph
CH2Ph
allyl
allyl
a Isonitrile (0.8 M); 1.1 equiv of acetyl chloride, 3.0 equiv of urea. b 1,3-
Dimethylthiourea was used. c Only 1.1 equiv of urea was used due to poor
solubility, which probably accounts for the relatively low yield. d Ad )
1-adamantyl. e Ar ) p-tolyl. f 1,6-Diisocyanohexane; 4 equiv of urea and
2 equiv of acetyl chloride used per equivalent of diisocyanide. g Product
does not precipitate and was isolated by column chromatography.
chloride adducts of the type 4 are treated with Ag+ to
generate highly electrophilic acylnitrilium ions, which are
trapped intramolecularly to afford cyclization products in
high yield.5 We suggest that 4 is sufficiently electrophilic to
undergo substitution by urea at the chloroiminium carbon
to give 5, analogous to Vilsmeier-Haack-Arnold chemis-
try.6 (Note, however, that activation of isocyanide with excess
HCl, giving the imidoyl chloride species R1NdCHCl and,
presumably, its conjugate acid, is apparently not sufficient,7
even though the Vilsmeier reagent [Me2NdCHCl]+ is
reactive with ureas.8) Another equivalent of urea is then
proposed to attack the electrophilic carbonyl center of 5 to
release acylurea 2 and intermediate 6. The latter species, an
ylide (or a stabilized carbene resonance form, not shown),
should undergo rapid proton transfer to give the formamidine
7. An equivalent of HCl is extracted by precipitation of the
hydrochloride salt 1. Thus, when only 1 equiv of urea is
used, a maximum of 50% yield can be expected, as is indeed
observed.4 Trityl chloride can presumably activate the
isonitrile by formation of an analogous electrophilic adduct,
although the nature of such a species is not yet clear. The
efficiency and convenience of the reaction is governed by
both the generation of the reactive adducts and the precipita-
tion of the final product. Thus, the most effective solvents
Yields were in the 50-80% range, except for cases in
which the poor solubility of the urea forced its use in only
stoichiometric amounts (entries 8-11). The process strongly
favors monosubstituted urea nitrogen atoms in favor of
unsubstituted (NH2) or doubly substituted (NR2) centers.
Thus, monosubstituted ureas react at the substituted position
exclusively (entries 8-11 and 16), even in the hindered
adamantyl case, and N,N-dimethylurea is completely unre-
active in both MeCN and THF. Other notable reactions
include the facile formation of a bis(formamidine urea) (entry
13), the successful use of hindered isocyanides (entries 2
and 14), and the incorporation of allyl-substituted ureas
(entries 15-16). Amides such as N-methylacetamide are
unreactive, but p-tolylhydrazide was found to be converted
to the corresponding formamidine adduct (3, Figure 1),
although in low yield (39%) due to the relatively poor
solubility of the hydrazide nucleophile. The preparation of
1b has been scaled up to 0.1 mol with no difficulty.
An extensive survey of electrophiles4 established that the
reaction is restricted almost exclusively to acid chlorides:
little or no yield was obtained with acyl bromides, oxalyl
and sulfuryl chloride, sulfonyl chlorides, an acyl fluoride,
several activated alkyl chlorides and bromides, trimethylsilyl
chloride, and protic acids. In contrast, yields are independent
of the nature of the acid chloride until steric hindrance
becomes too great (e.g., pivaloyl chloride or 2,6-dimethoxy-
(5) (a) Westling, M.; Smith, R.; Livinghouse, T. J. Org. Chem. 1986,
51, 1159-1165. (b) Westling, M.; Livinghouse, T. Tetrahedron Lett. 1985,
26, 5389-5392. (c) Westling, M.; Livinghouse, T. J. Am. Chem. Soc. 1987,
109, 590-592. (d) Lee, C. H.; Westling, M.; Livinghouse, T.; Williams,
A. C. J. Am. Chem. Soc. 1992, 114, 4089-4095.
(6) Jutz, C. AdV. Org. Chem. 1976, 9, 225-342.
(7) Hegarty, A. F.; Chandler, A. Tetrahedron Lett. 1980, 21, 885-888.
(8) (a) Jentzsch, W.; Seefelder, M. Chem. Ber. 1965, 98, 274-279. (b)
Bitter, I.; Ka´rpa´ti-AÄ da´m, E.; To´ke, L. Acta Chim. Acad. Sci. Hung. 1979,
102, 235-246. (c) Csuros, Z.; Soos, R.; Bitter, I.; Karpati-Adam, E. Acta
Chim. (Budapest) 1972, 73, 239-246.
1532
Org. Lett., Vol. 5, No. 9, 2003

(THF, MeCN) are those that can support charged or polar
intermediates but also allow the formamidine urea salts to
crystallize. Solvents either more (DMF) or less (Et2O,
toluene) polar are not as effective.
Although several N-arylformamidine compounds have
been identified as insecticides,9 formamidine ureas are little-
known species. They may be regarded as a subset of
acylamidinium compounds (8, Scheme 2), which are avail-
centers.11 However, compounds 1 are somewhat stable in
alcohol and water at neutral pH and ambient temperature,
showing substantial hydrolysis only after 12-24 h. The salts
are soluble and stable in polar organic solvents (DMF, THF,
and CH3CN), and are also unchanged upon heating in
nonpolar solvents, showing no tendency to undergo the
reported thermal fragmentation of amidine ureas to isocy-
anate and amidinium species.11c,d
We have also discovered that formamidine ureas in the
unprotonated form undergo a facile exchange reaction with
amine nucleophiles as shown in Scheme 3. The position and
Scheme 2
Scheme 3
rate of the equilibrium depends on both steric and electronic
factors, being driven to the right by large groups at R1 and
electron-rich groups at R4. This exchange route offers very
convenient access to different formamidine ureas from a
single isonitrile precursor. Further explorations of the
synthesis, reactivity, and biological activity of these and
related compounds are in progress.
able by capture of nitrilium ions with amide nucleophiles.10
In such reactions, the acylamidine nucleus is formed by a
transfer of oxygen from the amide component (Scheme 2),
a difficult process enabled by the highly reactive nature of
the electrophilic nitrilium carbon. In contrast, our reaction
accomplishes a formal H-atom transfer from a urea to the
isonitrile carbon (Scheme 2). The overall transformation is
made possible by the ability of the acid chloride to change
the nature of the isonitrile carbon from nucleophile to
electrophile. Such an umpoulung can also be accomplished
with transition metals,4 but apparently such routes have not
employed ureas as the capturing nucleophiles.
Complete hydrolysis of 1a or 1b with aqueous NaHCO3
in the presence of CH2Cl2 at 40 °C provides 1,3-dimethylurea
and benzylamine or tert-butylamine, consistent with the
reported reactivity of acylamidinium compounds with nu-
cleophiles (including water) at both sp2-hybridized carbon
Acknowledgment. We thank The Skaggs Institute for
Chemical Biology for support of this work and Dr. Michal
Sabat (University of Virginia) for X-ray crystallography of
compound 1a. D.D. thanks the Spanish MECD (Ministerio
de Educacio´n, Cultura y Deportes; Secretaria de Estado de
Educacio´n y Universidades) for a postdoctoral fellowship
cofinanced by the Fondo Social Europeo.
Supporting Information Available: Expanded discus-
sion, experimental details, characterization of new com-
pounds, and X-ray crystallographic details. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL034287R
(9) (a) Orr, G. L.; Orr, N.; Cornfield, L.; Gole, J. W. D.; Downer, R. G.
H. Pestic. Sci. 1990, 30, 285-294. (b) Ismail, S. M. M.; Matsumura, F.
Insect Biochem. Mol. Biol. 1992, 22, 713-720. (c) Longtine, C. A.; Zehnder,
G. W.; Radcliffe, E. B. J. Entomol. Sci. 1996, 31, 89-101.
(10) (a) Glocker, M. O.; Shrestha-Davadi, P. B.; Ku¨chler-Krischun, J.;
Hofmann, J.; Fischer, H.; Jochims, J. C. Chem. Ber. 1993, 126, 1859-
1865. (b) Jochims, J. C.; Glocker, M. O. Chem. Ber. 1990, 123, 1537-
1544
(11) (a) Glocker, M. O.; Shrestha-Davadi, P. B.; Ku¨chler-Krischun, J.;
Hofmann, J.; Fischer, H.; Jochims, J. C. Chem. Ber. 1993, 126, 1859-
1865. (b) Jochims, J. C.; Glocker, M. O. Chem. Ber. 1990, 123, 1537-
1544. (c) Bitter, I.; Ka´rpa´ti-AÄ da´m, E.; To´ke, L. Acta Chim. Acad. Sci. Hung.
1979, 102, 235-246. (d) Csuros, Z.; Soos, R.; Bitter, I.; Karpati-Adam, E.
Acta Chim. (Budapest) 1972, 73, 239-246.
Org. Lett., Vol. 5, No. 9, 2003
1533