Catalytic oxidative carbonylation of primary and secondary diamines to cyclic ureas. Optimization and substituent studies

Article abstract of DOI:10.1021/jo0109319

W(CO)₆-catalyzed oxidative carbonylation of 1,3-propanediamine to the corresponding urea has been examined under a variety of conditions. Following optimization, the Thorpe-Ingold effect on ring closure was studied using 2,2-dialkyl-1,3-propanediamines. For the 2,2-dimethyl- and 2,2-dibutyl-1,3-propanediamines, the yields were increased significantly as compared to that of the unsubstituted case. The eight-membered cyclic urea 5-butyl-5-ethyl-1,3-diazepan-2-one (5f) was formed in 38% yield, while only trace amounts of the cyclic urea were produced from the parent 1,5-pentanediamine. In a study of secondary diamines, yields from the carbonylation of N,N′-dialkyl-2,2-dimethyl-1,3-propanediamines were lower than those obtained from the primary diamines. The main byproducts from secondary diamines were tetrahydropyrimidine derivatives formed from a competitive reaction of the substrate with the oxidant and base.

Full text of DOI:10.1021/jo0109319

4086
J . Org. Chem. 2002, 67, 4086-4092
Ca ta lytic Oxid a tive Ca r bon yla tion of P r im a r y a n d Secon d a r y
Dia m in es to Cyclic Ur ea s. Op tim iza tion a n d Su bstitu en t Stu d ies
Fang Qian, J ennifer E. McCusker, Yue Zhang, A. Denise Main, Mary Chlebowski,
Michiyo Kokka,† and Lisa McElwee-White*
Department of Chemistry and Center for Catalysis, University of Florida, Gainesville, Florida 32611-7200
lmwhite@chem.ufl.edu
Received September 18, 2001
W(CO)6-catalyzed oxidative carbonylation of 1,3-propanediamine to the corresponding urea has
been examined under a variety of conditions. Following optimization, the Thorpe-Ingold effect on
ring closure was studied using 2,2-dialkyl-1,3-propanediamines. For the 2,2-dimethyl- and 2,2-
dibutyl-1,3-propanediamines, the yields were increased significantly as compared to that of the
unsubstituted case. The eight-membered cyclic urea 5-butyl-5-ethyl-1,3-diazepan-2-one (5f) was
formed in 38% yield, while only trace amounts of the cyclic urea were produced from the parent
1,5-pentanediamine. In a study of secondary diamines, yields from the carbonylation of N,N′-dialkyl-
2,2-dimethyl-1,3-propanediamines were lower than those obtained from the primary diamines. The
main byproducts from secondary diamines were tetrahydropyrimidine derivatives formed from a
competitive reaction of the substrate with the oxidant and base.
In tr od u ction
ates,15,16 dialkyl dithiocarbonates,17 carbonyl sulfide,18
carbonyl selenide,19 carbon dioxide,20 and urea.2-4
The conversion of ethylenediamine to 2-imidazolidi-
none has served as a test case for many of these reagents.
Despite the generally facile closure of five-membered
rings, phosgene itself affords the cyclic urea from ethyl-
enediamine in only 13% yield.6 In comparison, S,S-
dimethyl dithiocarbonate17 or hexachloroacetone21 can be
used to convert ethylenediamine to the imidazolidinone
in 80-90% yield. Carbonate derivatives such as disuc-
cinimido carbonate (DSC)15 and benzyl succinimido car-
bonate22 also generate the product urea in 80-90% yield.
Urea itself can be transaminated in the presence of water
to produce the imidazolidinone from ethylenediamine in
about 75% yield.23
Cyclic ureas have recently attracted attention due to
their presence in biologically active molecules, such as
the HIV protease inhibitors DMP 323 and DMP 450.1
Furthermore, cyclic ureas have found use as chiral
auxiliaries for asymmetric synthesis.2-4 Among the re-
ported methods for synthesis of cyclic ureas,5 the earliest
involved the nucleophilic attack of primary diamines on
phosgene.6-9 The primary diamines can be silylated
before treatment with phosgene if polymerization is
problematic.10 More recently, phosgene itself has been
replaced by derivatives such as 1,1-carbonyldiimidazole
(CDI).11-14 Other carbonyl sources that have been used
in the preparation of cyclic ureas include dialkyl carbon-
Despite multiple synthetic routes to prepare cyclic
ureas directly from primary diamines, there is still room
for further development of methodology. The direct
conversion of secondary diamines to N,N′-disubstituted
cyclic ureas is plagued by the difficulty of preparing
tetrasubstituted ureas.24 As a result, N,N′-disubstituted
* To whom correspondence should be addressed. Fax: (352) 846-0296.
† Permanent address: Catalysis Research Center and Graduate
School of Pharmaceutical Sciences, Hokkaido University, Sapporo 060-
0811, J apan.
(1) Hodge, C. N.; Lam, P. Y. S.; Eyermann, C. J .; J adhav, P. K.; Ru,
Y.; Fernandez, C. H.; De Lucca, G. V.; Chang, C. H.; Kaltenbach, R.
F.; Holler, E. R.; Woerner, F.; Daneker, W. F.; Emmett, G.; Calabrese,
J . C.; Aldrich, P. E. J . Am. Chem. Soc. 1998, 120, 4570-4581.
(2) Sankhavasi, W.; Yamamoto, M.; Kohmoto, S.; Yamada, K. Bull.
Chem. Soc. J pn. 1991, 64, 1425-1427.
(13) DeLucca, G. V.; Liang, J .; Aldrich, P. E.; Calabrese, J .; Cordova,
B.; Klabe, R. M.; Rayner, M. M.; Chang, C. H. J . Med. Chem. 1997,
40, 1707-1719.
(3) Cardillo, G.; Damico, A.; Orena, M.; Sandri, S. J . Org. Chem.
1988, 53, 2354-2356.
(14) Rossano, L. T.; Lo, Y. S.; Anzalone, L.; Lee, Y. C.; Meloni, D.
J .; Moore, J . R.; Gale, T. M.; Arnett, J . F. Tetrahedron Lett. 1995, 36,
4967-4970.
(4) Davies, S. G.; Mortlock, A. A. Tetrahedron Lett. 1991, 32, 4791-
4794.
(5) Hegarty, A. F.; Drennan, L. J . In Comprehensive Organic
Functional Group Transformations; Katritzky, A. R., Meth-Cohn, O.,
Rees, C. W., Eds.; Pergamon: Oxford, 1995; Vol. 6, pp 499-526.
(6) Boon, W. R. J . Chem. Soc. 1947, 307-318.
(7) Hayward, R. J .; Meth-Cohn, O. J . Chem. Soc., Perkin Trans. 1
1975, 212-219.
(15) Takeda, K.; Ogura, H. Synth. Commun. 1982, 12, 213-217.
(16) Skinner, G. S.; Hall, R. H.; Susi, P. V. J . Am. Chem. Soc. 1957,
79, 3786-3788.
(17) Leung, M. K.; Lai, J . L.; Lau, K. H.; Yu, H. H.; Hsiao, H. J . J .
Org. Chem. 1996, 61, 4175-4179.
(18) Ulrich, H.; Tucker, B.; Richter, R. J . Org. Chem. 1978, 43,
1544-1546.
(8) Maclaren, J . A. Aust. J . Chem. 1977, 30, 455-457.
(9) Crego, M.; Marugan, J . J .; Raposo, C.; Sanz, J .; Alcazar, V.;
Caballero, C.; Moran, J . R. Tetrahedron Lett. 1991, 32, 4185-4188.
(10) Birkofer, L.; Kuhlthau, H. P.; Ritter, A. Chem. Ber./ Recl. 1960,
93, 2810-2813.
(19) Kondo, K.; Yokoyama, S.; Miyoshi, N.; Murai, S.; Sonoda, N.
Angew. Chem., Int. Ed. Engl. 1979, 18, 692.
(20) Barker, B. J .; Rosenfarb, J .; Caruso, J . A. Angew. Chem., Int.
Ed. Engl. 1979, 18, 503-507.
(21) Rezende, M. C.; Marques, C. A.; Dall’Oglio, E. L.; Zucco, C.
Liebigs Ann./ Recl. 1997, 5, 925-929.
(11) Kaiser, A.; Balbi, M. Tetrahedron: Asymmetry 1999, 10, 1001-
1014.
(12) Patel, M.; Kaltenbach, R. F.; Nugiel, D. A.; McHugh, R. J .;
J adhav, P. K.; Bacheler, L. T.; Cordova, B. C.; Klabe, R. M.; Erickson-
Viitanen, S.; Garber, S.; Reid, C.; Seitz, S. P. Bioorg. Med. Chem. Lett.
1998, 8, 1077-1082.
(22) Wakamiya, T.; Kamata, M.; Kusumoto, S.; Kobayashi, H.; Sai,
Y.; Tamai, I.; Tsuji, A. Bull. Chem. Soc. J pn. 1998, 71, 699-709.
(23) Butler, A. R.; Hussain, I. J . Chem. Soc., Perkin Trans. 2 1981,
317-319.
10.1021/jo0109319 CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/10/2002

Oxidative Carbonylation of Diamines to Cyclic Ureas
J . Org. Chem., Vol. 67, No. 12, 2002 4087
cyclic ureas are generally prepared either by reaction of
phosgene6,25,26 or urea27 with the secondary diamine or
by N-alkylation of the unsubstituted ring structures.28-31
Also, there are problems with the use of phosgene and
its derivatives.32 Phosgene is highly toxic and corrosive,
and phosgene derivatives can be expensive to use on a
large scale. The direct metal-catalyzed conversion of
diamines and CO to cyclic ureas could provide an
attractive alternative to the use of phosgene and its
derivatives.
ally yielded cyclic ureas only as minor products. Carbo-
nylation of the diamines H2N(CH2)nNH2 (n ) 2-4, 6)
catalyzed by Mn2(CO)10 produces no cyclic products when
n ) 2, 4, or 6 and only 6% of the six-membered urea when
n ) 3.42 Carbonylation of ethylenediamine catalyzed by
Ni(CO)4 affords 2-imidazolidinone as a minor product in
10% yield.50 In addition to the catalytic processes using
CO as the carbonyl source, there is a related conversion
of both primary and secondary diamines to cyclic ureas
with CO2 using Ph3SbO/P4S10 as catalyst.51
Disadvantages exist for both the main-group- and tran-
sition-metal-catalyzed amine carbonylation reactions that
have been previously reported. For the main-group cata-
lysts, selenium-catalyzed carbonylation generates hydro-
gen selenide as a byproduct and certain substrates re-
quire stoichiometric or excess selenium.46-48 Aromatic
amines are also problematic in the selenium-catalyzed
reactions,19 although a switch from elemental selenium
to K2SeO3 has been reported to produce high yields of
1,3-diarylureas.52 The transition-metal-catalyzed reac-
tions usually require stringent reaction conditions such
as high temperatures and pressures. Moreover, yields
from aliphatic amines tend to be lower than those from
aromatic substrates when transition-metal catalysts are
used.
Although catalytic carbonylation has been investigated
for some time,33-35 the topic remains of interest. Other
groups have reported catalytic conversion of primary
amines and CO to acyclic ureas using complexes of
several transition metals including Ni,36 Pd,37-39 Ru,40
Mn,41,42 and Co.43 Main-group elements such as sulfur44,45
and selenium46-49 have also been reported to catalyze this
reaction.
Reports of catalytic carbonylation of diamines to cyclic
ureas are rarer. The selenium-catalyzed reaction can
produce high yields, but certain substrates such as 2-(2-
aminoethyl)aniline afford the cyclic urea only in the pre-
sence of stoichiometric or excess selenium.46-48 Transi-
tion-metal-catalyzed carbonylation of diamines has gener-
In an effort to improve upon the transition-metal-
catalyzed amine carbonylation methods, we have ex-
plored the use of tungsten catalysts.53-57 We previously
reported preliminary studies on the catalyzed oxidative
carbonylation of R,ω-diamines to cyclic ureas using
W(CO)6 as catalyst and I2 as the oxidant.55 Those
preliminary results established that both primary and
secondary diamines are substrates for the reaction. We
have now expanded the range of substrates and carried
out optimization studies. The results of these further
investigations are reported here.
(24) Katritzky, A. R.; Pleynet, D. P. M.; Yang, B. Z. J . Org. Chem.
1997, 62, 4155-4158.
(25) Lumbroso, H.; Lie´geois, C.; Devillanova, F. A.; Verani, G. J .
Mol. Struct. 1981, 77, 239-251.
(26) Tattersall, J .; Taylor, J . B. J . Chem. Soc. C 1970, 931-933.
(27) Martell, A. E.; Frost, A. E. J . Am. Chem. Soc. 1950, 72, 1032-
1033.
(28) Nugiel, D. A.; J acobs, K.; Worley, T.; Patel, M.; Kaltenbach, R.
F.; Meyer, D. T.; J adhav, P. K.; DeLucca, G. V.; Smyser, T. E.; Klabe,
R. M.; Bacheler, L. T.; Rayner, M. M.; Seitz, S. P. J . Med. Chem. 1996,
39, 2156-2169.
(29) Lam, P. Y. S.; Ru, Y.; J adhav, P. K.; Aldrich, P. E.; DeLucca,
G. V.; Eyermann, C. J .; Chang, C. H.; Emmett, G.; Holler, E. R.;
Daneker, W. F.; Li, L. Z.; Confalone, P. N.; McHugh, R. J .; Han, Q.;
Li, R. H.; Markwalder, J . A.; Seitz, S. P.; Sharpe, T. R.; Bacheler, L.
T.; Rayner, M. M.; Klabe, R. M.; Shum, L. Y.; Winslow, D. L.;
Kornhauser, D. M.; J ackson, D. A.; EricksonViitanen, S.; Hodge, C.
N. J . Med. Chem. 1996, 39, 3514-3525.
Resu lts a n d Discu ssion
Op tim iza tion of Rea ction Con d ition s. In the initial
paper on W(CO)6-catalyzed oxidative carbonylation of
primary and secondary R,ω-diamines to cyclic ureas (eq
1),55 we reported conversion of 1,3-propanediamine to the
corresponding urea in 52% yield. Ring closure of this
substrate under varying conditions has subsequently
been examined.
(30) Dehmlow, E. V.; Rao, Y. R. Synth. Commun. 1988, 18, 487-
494.
(31) Li, C. D.; Mella, S. L.; Sartorelli, A. C. J . Med. Chem. 1981, 24,
1089-1092.
(32) Bigi, F.; Maggi, R.; Sartori, G. Green Chem. 2000, 2, 140-148.
(33) Colquhoun, H. M.; Thompson, D. J .; Twigg, M. V. Carbonyla-
tion: Direct Synthesis of Carbonyl Compounds; Plenum: New York,
1991.
(34) Sheldon, R. A. Chemicals from Synthesis Gas; Dordrecht:
Boston, 1983.
(35) Wender, I.; Pino, P. Organic Syntheses via Metal Carbonyls;
Wiley-Interscience: New York, 1968.
(36) Giannoccaro, P.; Nobile, C. F.; Mastrorilli, P.; Ravasio, N. J .
Organomet. Chem. 1991, 419, 251-258.
(37) Pri-Bar, I.; Alper, H. Can. J . Chem. (Rev. Can. Chim.) 1990,
68, 1544-1547.
(38) Gupte, S. P.; Chaudhari, R. V. J . Catal. 1988, 114, 246-258.
(39) Dahlen, G. M.; Sen, A. Macromolecules 1993, 26, 1784-1786.
(40) Mulla, S. A. R.; Rode, C. V.; Kelkar, A. A.; Gupte, S. P. J . Mol.
Catal. A: Chem. 1997, 122, 103-109.
(41) Calderazzo, F. Inorg. Chem. 1965, 4, 293-296.
(42) Dombek, B. D.; Angelici, R. J . J . Organomet. Chem. 1977, 134,
203-217.
As seen in Table 1, yields of the cyclic urea were
(43) Bassoli, A.; Rindone, B.; Tollari, S.; Chioccara, F. J . Mol. Catal.
1990, 60, 41-48.
(50) Martin, W. E.; Farona, M. F. J . Organomet. Chem. 1981, 206,
393-397.
(44) Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F. J . Org.
Chem. 1961, 26, 3306-3308.
(51) Nomura, R.; Hasegawa, Y.; Ishimoto, M.; Toyosaki, T.; Matsuda,
H. J . Org. Chem. 1992, 57, 7339-7342.
(45) Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F.; Bolze,
C. J . Org. Chem. 1961, 26, 3309-3312.
(52) Kim, H. S.; Kim, Y. J .; Lee, H.; Lee, S. D.; Chin, C. S. J . Catal.
1999, 184, 526-534.
(46) Sonoda, N.; Yasuhara, T.; Kondo, K.; Ikeda, T.; Tsutsumi, S.
J . Am. Chem. Soc. 1971, 93, 6344.
(53) McCusker, J . E.; Abboud, K. A.; McElwee-White, L. Organo-
metallics 1997, 16, 3863-3866.
(47) Yoshida, T.; Kambe, N.; Murai, S.; Sonoda, N. Bull. Chem. Soc.
J pn. 1987, 60, 1793-1799.
(54) McCusker, J . E.; Logan, J .; McElwee-White, L. Organometallics
1998, 17, 4037-4041.
(48) Yoshida, T.; Kambe, N.; Murai, S.; Sonoda, N. Tetrahedron Lett.
1986, 27, 3037-3040.
(55) McCusker, J . E.; Grasso, C. A.; Main, A. D.; McElwee-White,
L. Org. Lett. 1999, 1, 961-964.
(49) de Souza, W. F.; Kambe, N.; J in, Z. J .; Kanehisa, N.; Kai, Y.;
Sonoda, N. J . Org. Chem. 1995, 60, 7058-7062.
(56) McCusker, J . E.; Qian, F.; McElwee-White, L. J . Mol. Catal. A:
Chem. 2000, 159, 11-17.

4088 J . Org. Chem., Vol. 67, No. 12, 2002
Qian et al.
Ta ble 1. Effect of Solven t a n d Tem p er a tu r e Va r ia tion
on th e Ca ta lytic Ca r bon yla tion of 1,3-P r op a n ed ia m in e
Sch em e 1
T
(°C)
yielda,b
(%)
T
(°C)
yielda,b
(%)
solvent
solvent
THF
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
CHCl3
CHCl3
EtOAc
EtOAc
0d
25
50
90
0d
25
50
25
50
23
52
33
24
21
43
29
32
21
25
50
25
50
25
25
25
25
28
22
33
23
24
13
16
12
THF
CH2Cl2/H2Oc
CH2Cl2/H2Oc
Et2O
Et2O/H2O
PhCF3
PhCF3/H2O
of the carbonylation of several substituted diamines was
undertaken.
a
Isolated yield of tetrahydro-2-pyrimidinone calculated per
The 2,2-dialkyl-1,3-propanediamines 4a -c were cho-
sen as the first set of substrates because in the initial
study, the carbonylation of 1,3-propanediamine gave the
highest yield of cyclic urea. In addition, diamine 4a is
commercially available. The other requisite diamines
(4b,c) were synthesized by the double alkylation of
malononitrile to afford the corresponding disubstituted
dinitrile compound followed by reduction with lithium
aluminum hydride to yield the diamine59 (Scheme 1).
Acceptable yields (60-70%) were obtained for each step
where R ) benzyl and butyl.
b
equivalent of diamine. Reaction conditions: 1,3-propanediamine
(4.27 mmol), W(CO)6 (0.0425 mmol), I2 (3.84 mmol), K2CO3 (8.9
mmol), solvent (90 mL), 80 atm of CO, 24 h. c CH2Cl2, 80 mL; H2O,
d
10 mL. The reaction time was increased to 72 h.
lowered upon raising the temperature. Instead, reaction
mixtures from higher temperature reactions contained
larger amounts of intractable material that we have
attributed to oligomers. Lowering the temperature to 0
°C and increasing the reaction time to 72 h also resulted
in decreased yields.
The 2,2-dialkyl-1,3-propanediamine substrates 4a -c
were subjected to the oxidative carbonylation conditions
as described in Table 2. For the methyl and butyl cases,
the yields were significantly higher than for the unsub-
stituted case. 2,2-Dimethyl-1,3-propanediamine (4a ) pro-
duced its respective N,N′-disubstituted urea 5a in 80%
yield, while the butyl derivative 4b afforded the cyclic
urea in 70% yield. Yields from the benzyl substrate 4c
were somewhat lower.
Different solvents were also explored at room temper-
ature and 50 °C. As seen in Table 1, of the solvents
examined, yields were highest in CH2Cl2, with CHCl3
producing comparable but somewhat lower yields. Al-
though the addition of water to produce a biphasic CH2-
Cl2/H2O solvent system resulted in dramatically in-
creased yields of ureas from substituted benzylamines,57
similar improvements are not observed using 1,3-pro-
panediamine.
Further studies on diamines 4d ,e probed the effects
of methyl substitution on the formation of five-membered
rings. In contrast to the results from substrates 4a ,b, the
yields of cyclic ureas 5d ,e were comparable to those of
the unsubstituted cases. This can be attributed to
competing effects of the methyl groups. On one hand, the
Thorpe-Ingold effect would be expected to facilitate ring
closure by bringing the amino groups closer on average.
However, the methyl groups also introduce steric hin-
drance around the amino groups, a factor that has been
found to lower the yields in oxidative carbonylation of
alkylamines to dialkylureas.57
A more promising result was obtained in the oxidative
carbonylation of pentanediamine 4f. Although the parent
1,5-pentanediamine produced the eight-membered cyclic
urea in only trace amounts,55 the disubstituted eight-
membered ring urea 5f was formed in 38% yield under
similar reaction conditions. Apparently, the alkyl sub-
stituents promote ring closure. It is thus possible that
other substituent patterns and/or conformational con-
straints could render this method more general for the
formation of eight-membered cyclic ureas.
Effect of Su bstitu tion in th e Lin k er of Secon d a r y
Dia m in es. On the basis of the promising results obtained
for carbonylation of substituted primary diamines, we
carried out a similar study on the carbonylation of
secondary diamines to N,N′-disubstituted cyclic ureas.
The substrates for this investigation were derived from
2,2-dimethylpropane-1,3-diamine because 4a could be
converted to 5a in 80% yield. The secondary diamine
substrates 4g and 4h were synthesized by CsOH-
One problem with 1,3-propanediamine as a substrate
arises from the solubility of the product in water. Our
inability to use the standard aqueous Na2SO3 wash and
extraction to remove residual I2 and inorganic salts from
the reaction mixtures led us to examine different workup
procedures. A “dry wash” method, in which Na2S2O4
slurry was stirred vigorously with the reaction mixture
for 30 min, was more successful. Although sublimation
of the product was required to obtain analytically pure
material, this method made the workup procedure more
straightforward.
Oxidative carbonylation of 1,2-ethylenediamine under
the optimized conditions produced 2-imidazolidinone (1)
in 40% yield. The seven-membered cyclic urea 1,3-
diazepan-2-one (3) was prepared in 38% yield under the
same conditions. Individual optimizations for different
substrates were not carried out.
Effect of Su bstitu tion in th e Lin k er of P r im a r y
Dia m in es. Previous studies on oxidative carbonylation
of diamines have established that preparation of the
parent five-, six-, and seven-membered cyclic ureas can
be achieved in moderate yields.55 However, these are
intrinsically difficult substrates in two regards. First,
their solubility in water leads to difficulties with the
workup (vide supra). In addition, ring closure of the
unsubstituted diamines cannot take advantage of the
Thorpe-Ingold effect58 to facilitate ring closure. To
explore the influence of substitution in the linker, a study
(57) McCusker, J . E.; Main, A. D.; J ohnson, K. S.; Grasso, C. A.;
McElwee-White, L. J . Org. Chem. 2000, 65, 5216-5222.
(58) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Organic Compounds; Wiley-Interscience: New York, 1994; pp 682-
684.
(59) Briggs, C. B. A.; Newington, I. M.; Pitt, A. R. J . Chem. Soc.,
Chem. Commun. 1995, 379-380.

Oxidative Carbonylation of Diamines to Cyclic Ureas
J . Org. Chem., Vol. 67, No. 12, 2002 4089
Ta ble 2. Oxid a tive Ca r bon yla tion of Su bstitu ted
P r im a r y Dia m in es
Ta ble 3. Oxid a tive Ca r bon yla tion of Su bstitu ted
Secon d a r y Dia m in es
a
Isolated yield of urea calculated per equivalent of amine.
b
Reaction conditions: amine (7.07 mmol), W(CO)6 (0.28 mmol),
I2 (7.1 mmol), K2CO3 (14.4 mmol), CH2Cl2 (40 mL), room temper-
ature, 80 atm of CO, 24 h. c Reaction conditions as in footnote b
d
except the solvent was CH2Cl2 (35 mL) plus H2O (5 mL). This
work. e Isolated as the hydroiodide salt. f Reference 55.
Sch em e 2
a
Isolated yield of urea calculated per equivalent of amine.
b
Reaction conditions: amine (7.07 mmol), W(CO)6 (0.28 mmol),
I2 (7.1 mmol), K2CO3 (14.4 mmol), CH2Cl2 (40 mL), 90 °C, 80 atm
of CO, 24 h. c Reaction conditions as in footnote b except 25 °C.
promoted N-alkylation,60 which produced the requisite
substituted secondary diamines in moderate to good
yields (eq 2). Due to difficulties in preparing 4i by direct
alkylation, an alternative preparation via reduction of
the diformamide61 was utilized for that particular sub-
strate (eq 3).
properties and Thorpe-Ingold effect combined to raise
the urea yields with respect to that of the parent
propanediamine, gem-dialkyl substitution was detrimen-
tal in the carbonylation of secondary diamines. As shown
in Table 3, yields from the carbonylation of substituted
secondary diamines are lower than those previously
obtained from N,N′-dimethylpropane-1,3-diamine (4j).55
Diamines 4g and 4h produced the corresponding ureas
5g and 5h in low yields, and concomitant formation of
tetrahydropyrimidine derivatives occurred in both cases.
In an attempt to make urea formation faster than the
accumulation of byproducts, the reaction was carried out
at 90 °C. Unfortunately, there was no improvement in
the yield of the urea. Subsequent control experiments
established that the tetrahydropyrimidine formation did
not require the presence of the catalyst or CO (Scheme
2). In the experiment without CO, the benzyl-substituted
diamine afforded an 80% yield of the tetrahydropyrimi-
dine 6h under the same conditions. Furthermore, in the
absence of both CO and the catalyst, 6h was still formed
in 33% yield.
In contrast to the behavior of gem-dialkyl-substituted
primary diamines 4a -c, in which the improved solubility
The formation of tetrahydropyrimidines 6g and 6h
from the secondary diamines 4g and 4h was unexpected
since we had previously carbonylated several secondary
diamines to the corresponding cyclic ureas55 (e.g., the
(60) Salvatore, R. N.; Nagle, A. S.; Schmidt, S. E.; J ung, K. W. Org.
Lett. 1999, 1, 1893-1896.
(61) Betschart, C.; Schmidt, B.; Seebach, D. Helv. Chim. Acta 1988,
71, 1999-2021.

4090 J . Org. Chem., Vol. 67, No. 12, 2002
Qian et al.
Gen er a l P r oced u r e for th e Ca ta lytic Ca r bon yla tion
of Un su bstitu ted P r im a r y Dia m in es w ith W(CO)6. The
following procedure is typical for the parent primary diamines.
To a stirred solution of W(CO)6 (15 mg, 0.042 mmol) in 90 mL
of CH2Cl2 in the glass liner of a 300 mL Parr high-pressure
vessel were added 1,3-diaminopropane (0.35 mL, 4.27 mmol),
K2CO3 (1.2 g, 8.9 mmol), and I2 (0.97 g, 3.8 mmol). The vessel
was then charged with 80 atm of CO, and the solution was
stirred under pressure at room temperature for 24 h. The
pressure was released, and the maroon solution was filtered
away from a red solid. The CH2Cl2 was removed by evaporation
and the residue washed with methanol, while the initial red
solid was rinsed with methanol. The methanol layer was
treated with 3 g of Na2S2O4 that had been slurried with 2 mL
of deionized water. The solution was then stirred vigorously
for 30 min. The Na2S2O4 was removed by filtration, and the
solution was concentrated to obtain a pale yellow powder. The
yellow powder was then sublimed to obtain a white solid (0.23
g, 52% yield). The solid was identified as tetrahydro-2-
pyrimidinone (2) by comparison with an authentic sample.
2-Im id a zolid in on e (1). The general procedure afforded the
product in 40% yield. The product was identified as 2-imida-
zolidinone by comparison with an authentic sample.
1,3-Dia zep a n -2-on e (3). The general procedure afforded
the product in 38% yield. The product was identified as 1,3-
diazepan-2-one by comparison of its spectroscopic data with
literature values.62,63
Gen er a l P r oced u r e for th e Syn th esis of 2,2-Dia lk yl-
m a lon on itr iles. The following procedure is typical. To a two-
necked flask provided with a reflux condenser were added
malononitrile (1.65 g, 25 mmol), n-butyl bromide (7.5 g, 55
mmol), and tetrabutylammonium bromide (0.52 g, 1.6 mmol).
After the reaction mixture had been stirred for 30 min at room
temperature, potassium tert-butoxide (6.17 g, 55 mmol) was
added little by little at 0 °C, and the stirring was continued
for 18 h. The reaction mixture was washed with CH2Cl2, and
the organic fractions were combined and concentrated in vacuo.
The crude product mixture was further purified by flash
chromatography on silica gel (CH2Cl2/hexanes, 1:1) to afford
pure 2,2-dibutylmalononitrile in 67% yield. The product was
identified by comparison with literature data.64
conversion of the parent N,N′-dimethylpropanediamine
4j to urea 5j). Although the yield of 5j was moderate, no
major byproduct was detected in the reaction mixtures.
The difference in the behavior of 4g,h vs 4j led us to
prepare 4i to determine whether the tetrahydropyrimi-
dine formation was due to conformational effects from
the gem-dimethyl group in the backbone of 4g,h or the
presence of the larger N-alkyl groups. As expected from
the reactivity of 4j, N,N′-dimethyldiamine 4i produced
urea 5i in 37% yield. Attempts to detect tetrahydropy-
rimidine 6i among the reaction products by 1H NMR
spectroscopy and GC-MS were unsuccessful. The mech-
anism for the formation of tetrahydropyrimidines is not
clear yet. However, for the diaminopropane substrates,
the critical feature for competitive tetrahydropyrimidine
formation appears to be the presence of an alkyl sub-
stituent larger than a methyl group. In light of these
results, it is also interesting to note that the five-
membered cyclic urea can be obtained from N,N′-dieth-
ylethylenediamine in 46% yield and N,N′-dibenzylethyl-
enediamine can also be converted to the corresponding
urea.55 Clearly, the sizes of the ring, backbone substit-
uents, and N-alkyl substituent are involved in a complex
interplay which determines the ratio of urea to alterna-
tive heterocyclic products.
Another possible cause of the modest yields of ureas
may involve the difficulties encountered in deprotonating
the byproduct amine salts of secondary substrates.56
Amine salts are produced during the oxidative carbonyl-
ation, and in previous experiments with secondary
amines, the amine salts have been detected in significant
quantities in the reaction mixtures despite attempts to
scavenge the protons with base.
Con clu sion
In conclusion, optimization studies on the carbonyla-
tion of propane-1,3-diamine have been carried out. Yields
of cyclic ureas are significantly higher for the 2,2-dialkyl-
1,3-propanediamines as a result of the Thorpe-Ingold
effect and improved solubility in organic solvents during
workup. The carbonylation of N,N′-dialkyl-2,2-dimeth-
ylpropane-1,3-diamines afforded tetrasubstituted ureas;
however, the products were obtained in modest yields,
and tetrahydropyrimidine derivatives were formed in
significant amounts when the substrates bore N-alkyl
substituents larger than methyl. Comparison of these
results with previously reported carbonylations of sec-
ondary diamines to form five-membered cyclic ureas
suggests that the effects of ring size and N-substituent
size on the carbonylation reaction are complex. On the
basis of these results, primary diamines with substituted
linkers are promising substrates for this reaction, making
utilization of catalytic carbonylation in more complex
systems a possibility.
2,2-Diben zylm a lon on itr ile. From 2.64 g of malononitrile
and 15.05 g of benzyl bromide, the general procedure afforded
the product in 58% yield. The product was identified by
comparison with literature data.64
Gen er a l P r oced u r e for th e Syn th esis of 2,2-Dia lk yl-
p r op a n e-1,3-d ia m in es. The following procedure is typical.
To a suspension of lithium aluminum hydride (0.9 g, 24 mmol)
in 50 mL of dry diethyl ether at 0 °C was added a solution of
2,2-dibutylmalononitrile (1.7 g, 9.5 mmol) in 30 mL of diethyl
ether. After the addition was complete, the mixture was heated
under reflux for 1 h and then stirred at room temperature for
another 24 h. Water (5 mL) was then added to hydrolyze the
remaining LiAlH4. The granular solid was isolated by vacuum
filtration and washed with diethyl ether. The ether layer was
dried with MgSO4 and filtered. Removal of the solvent afforded
pure 2,2-dibutylpropane-1,3-diamine (4b) as a colorless oil in
70% yield. 1H NMR (CDCl3): δ 3.54 (br, 4H), 2.74 (s, 4H), 1.30
(m, 8H), 1.13 (t, J ) 3.6 Hz, 4H), 0.88 (t, J ) 7.2 Hz, 6H). 13C
NMR (CDCl3): δ 45.9, 40.2, 32.1, 25.0, 23.6, 14.1. HRMS
(FAB): m/z calcd for C11H26N2 (M + H+) 187.2174, found
187.2173.
2,2-Diben zylp r op a n e-1,3-d ia m in e (4c). The general pro-
1
cedure afforded the product as a colorless oil in 65% yield. H
Exp er im en ta l Section
NMR (CDCl3): δ 7.15 (m, 10H), 2.60 (s, 4H), 2.47 (s, 4H), 0.99
(br s, 4H). 13C NMR (CDCl3): δ 138.5, 130.3, 129.0, 128.0, 45.1,
40.4, 38.2. HRMS (FAB): m/z calcd for C17H22N2 (M + H+)
255.1861, found 255.1863.
Gen er a l In for m a tion . Diethyl ether and tetrahydrofuran
were distilled from sodium/benzophenone. Methylene chloride
and chloroform were distilled over calcium hydride. N,N-
Dimethylformamide and ethyl acetate were dried with 4 Å
molecular sieves. All other chemicals were purchased in
reagent grade and used with no further purification unless
stated otherwise. Diamines were purchased unless specified
below. W(CO)6 was purified by chromatography on alumina
using hexane as eluent.
(62) Irsch, G.; Rademacher, P. J . Mol. Struct. 1990, 222, 265-273.
(63) Lien, E. J .; Guadauska, G.; Chou, J . T. Spectrosc. Lett. 1972,
5, 293-305.
(64) Diez-Barra, E.; De la Hoz, A.; Moreno, A.; Sa´nchez-Verdu´, P.
J . Chem. Soc., Perkin Trans. 1 1991, 2589-2592.

Oxidative Carbonylation of Diamines to Cyclic Ureas
J . Org. Chem., Vol. 67, No. 12, 2002 4091
Gen er a l P r oced u r e for th e Ca ta lytic Ca r bon yla tion
of Su bstitu ted P r im a r y Dia m in es w ith W(CO)6. P r oce-
d u r e A. The following procedure is typical. To a stirred
solution of W(CO)6 (100 mg, 0.284 mmol) in a 40 mL solution
of CH2Cl2 in the glass liner of a 300 mL Parr high-pressure
vessel were added 2,2-dimethyl-1,3-propanediamine (0.85 mL,
7.1 mmol), K2CO3 (1.95 g, 14.1 mmol), and iodine (1.8 g, 7.1
mmol). The vessel was then charged with 80 atm of CO, and
the solution was stirred under pressure at room temperature
for 24 h. The pressure was released, and the maroon solution
was filtered away from a yellow solid. The CH2Cl2 layer was
treated with 3 g of Na2S2O4 that had been slurried with 2 mL
of deionized water, and then the solution was stirred vigorously
for 30 min. The NaS2O4 was removed by filtration, and the
solution was concentrated in vacuo. The resulting pale yellow
oil was recrystallized from CH2Cl2/ether to afford a white solid
(0.78 g, 50% yield).
P r oced u r e B. Procedure B is identical to procedure A
except that the CH2Cl2 layer was washed with a 1.0 M HCl
solution followed by a saturated Na2S2O4 solution. The result-
ing pale yellow solution was then dried with MgSO4 and
filtered. The solution was concentrated in vacuo. The resulting
pale yellow oil was recrystallized from CH2Cl2/ether to afford
a white solid.
up in 1 N NaOH and extracted with EtOAc. The solvent was
removed, and the crude product was chromatographed on
neutral alumina using EtOAc/MeOH as eluent to afford clean
N,N′-dibutyl-2,2-dimethyl-1,3-propanediamine (4g) as a color-
1
less oil in 67% yield. H NMR (CDCl3): δ 2.52 (t, J ) 6.6 Hz,
4H) 2.38 (s, 4H), 1.41 (m, 4H), 1.30 (m, 4H), 0.87 (s, 6H), 0.83
(t, J ) 7.5 Hz, 6H). 13C NMR (CDCl3): δ 60.7, 50.6, 34.6, 32.1,
24.8, 20.4, 14.0. Anal. Calcd for C13H30N2: C, 72.94; N, 13.32;
H, 14.46. Found: C, 72.08; N, 13.06; H, 14.10.
N,N′-Diben zyl-2,2-d im eth ylp r op a n e-1,3-d ia m in e (4h ).
The general procedure afforded the product as a colorless oil
in 43% yield from 1.63 g of 2,2-dimethyl-1,3-propanediamine.
1H NMR (CDCl3): δ 7.21 (m, 10H), 3.68 (s, 4H), 2.37 (s, 4H),
0.84 (s, 6H). 13C NMR (CDCl3): 140.7, 128.2, 127.9, 126.6, 59.0,
54.5, 34.7, 24.8. HRMS (FAB): m/z calcd for C19H26N2 (M +
H+) 283.2174, found 283.2178.
N,N′-Difor m yl-2,2-d im eth ylp r op a n e-1,3-d ia m in e. The
diamine was synthesized according to a literature procedure
and identified by comparison with published data.65
N,N′-Dim eth yl-2,2-d im eth ylp r op a n e-1,3-d ia m in e (4i).
To a mixture of 250 mL of Et2O and 10 g of LiAlH4 (95%) at 0
°C was added 9.88 g (62.5 mmol) N,N′-diformyl-2,2-dimethyl-
propane-1,3-diamine in small portions. After the addition was
finished, the reaction mixture was stirred at room temperature
for 20 h. The mixture was then hydrolyzed with Et2O and H2O,
and the white solid was removed by filtration through Celite.
After the filtrate was dried over MgSO4, removal of the solvent
afforded 4i as 6.1 g (75% yield) of a colorless oil. The product
was identified by comparison with literature data.61
5,5-Dim eth yltetr a h yd r o-2(1H)-p yr im id in on e (5a ). Pro-
cedure A at room temperature afforded the product in 50%
yield. Procedure A at 90 °C afforded the product in 80% yield.
The solid was identified as 5,5-dimethyltetrahydro-2(1H)-
pyrimidinone by comparison with literature data.17
Gen er a l P r oced u r e for th e Ca ta lytic Ca r bon yla tion
of Secon d a r y Dia m in es. The following procedure is typical.
To a stirred solution of W(CO)6 (50 mg, 0.14 mmol) in 40 mL
of CH2Cl2 in the glass liner of a 300 mL Parr high-pressure
vessel were added N,N′-dibutyl-2,2-dimethylpropane-1,3-di-
amine (0.85 mL, 3.6 mmol), K2CO3 (0.98 g, 7.1 mmol), and
iodine (0.90 g, 3.6 mmol). The vessel was then charged with
80 atm of CO, and the solution was stirred under pressure at
room temperature for 24 h. After the pressure was released,
a yellow solid was removed from the maroon solution by
filtration. The CH2Cl2 layer was washed first with a 1.0 M
HCl solution and then with saturated Na2S2O4 solution. The
resulting pale yellow solution was then dried with MgSO4 and
filtered. The solution was concentrated in vacuo. The crude
mixture was chromatographed on neutral alumina (EtOAc/
MeOH) to afford pure 1,3-dibutyl-5,5-dimethyltetrahydro-
2(1H)-pyrimidinone (5g) as a yellow oil in 10% yield. To isolate
1-butyl-2-propyl-5,5-dimethyl-1,4,5,6-tetrahydropyrimidine (6g),
the crude mixture was washed with EtOAc and MeOH instead
of chromatography to afford the hydroiodide salt 6g‚HI as a
white solid in 17% yield. The following are the data for 5g. IR
5,5-Dibu tyltetr a h yd r o-2(1H)-p yr im id in on e (5b). From
0.35 g of 2,2-dibutylpropane-1,3-diamine, procedure B afforded
the product as a yellow solid in 63% yield. IR (CH2Cl2): νCO
1678 cm-1. 1H NMR (CDCl3): δ 5.15 (br, 2H), 2.95 (s, 4H) 1.28
(m, 12H), 0.87 (t, J ) 7 Hz, 6H). 13C NMR (CDCl3): δ 157.0
(CdO), 48.9, 32.7, 32.6, 25.2, 23.2, 14.0. HRMS (FAB): m/z
calcd for C12H24N2O (M + H+) 213.1966 found 213.1970.
5,5-Diben zyltetr a h yd r o-2(1H)-p yr im id in on e (5c). Pro-
cedure B afforded the product as a yellow solid in 48% yield.
IR (CH2Cl2): νCO 1671 cm-1. 1H NMR (CDCl3): δ 7.33 (m, 10H),
4.12 (br s, 2H), 3.08 (s, 4H), 2.38 (s, 4H). 13C NMR (CDCl3): δ
164.5 (CdO), 136.4, 130.6, 128.1, 126.6, 46.5, 40.3, 34.7. HRMS
(FAB): m/z calcd for C18H20N2O (M + H+) 281.1654, found
281.1658.
4,4-Dim eth yltetr a h yd r o-2H-im id a zol-2-on e (5d ). Pro-
cedure A at room temperature afforded the product in 50%
yield. Procedure A at 90 °C afforded the product in 30% yield.
The product was identified by comparison with literature
data.17
4-Meth yltetr a h yd r o-2H-im id a zol-2-on e (5e). Procedure
A at room temperature afforded the product in 20% yield.
Procedure A at 90 °C afforded the product in 30% yield. The
product was identified by comparison with literature data.65
5-Bu tyl-5-eth yl-1,3-d ia zep a n -2-on e (5f). Procedure B af-
forded the product as a white solid in 38% yield. Mp: 55-57
1
(CH2Cl2): νCO 1622 cm-1. H NMR (CDCl3): δ 3.23 (t, J ) 7.5
Hz, 4H), 2.87 (s, 4H), 1.42 (m, 4H), 1.27 (m, 4H), 0.98 (s, 4H),
0.89 (t, J ) 7.2 Hz, 6H). 13C NMR (CDCl3): δ 155.1 (CdO),
57.3, 47.0, 29.9, 28.3, 24.4, 20.0, 13.8. HRMS (FAB): m/z calcd
for C14H28N2O (M + H+) 241.2273, found 241.2241. The
1
°C. IR (CH2Cl2): νCO 1651 cm-1. H NMR (CDCl3): δ 5.59 (br,
1
following are the data for 6g‚HI. Mp: 106-107 °C. H NMR
1H), 5.32 (br, 1H), 3.31 (s, 2H), 3.00 (m, 2H), 1.66 (m, 2H),
1.42 (m, 2H), 1.11 (m, 8H), 0.83 (t, J ) 6.9 Hz, 3H), 0.70 (t,
3H, J ) 7.2 Hz). 13C NMR (CDCl3): δ 160.7 (CdO), 48.7, 41.4,
40.4, 33.8, 28.9, 27.0, 25.9, 24.9, 23.4, 14.1, 7.3. HRMS (FAB):
m/z calcd for C12H24N2O (M + H+) 213.1966, found 213.1971.
Gen er a l P r oced u r e for th e Syn th esis of N,N′-Dia lk yl-
2,2-d im eth ylp r op a n e-1,3-d ia m in es. The following proce-
dure is typical. To activated powdered 4 Å molecular sieves (2
g) in anhydrous N,N-dimethylformamide (15 mL) was added
cesium hydroxide monohydrate (5.6 g, 34 mmol). After the
white suspension was vigorously stirred for 10 min, 2,2-
dimethylpropane-1,3-diamine (2.02 mL, 16 mmol) was added.
Following an additional 30 min of stirring, 1-bromobutane (4.0
mL, 37 mmol) was added to the white suspension. The reaction
was stirred for 20 h, filtered to remove the molecular sieves
and undissolved inorganic salts, and rinsed with EtOAc. After
the filtrate was concentrated in vacuo, the residue was taken
(CDCl3): δ 9.91 (br s, 1H), 3.38 (t, J ) 7.5 Hz, 2H), 3.19 (s,
2H), 3.11 (s, 2H), 2.79 (t, J ) 7.8 Hz, 2H), 1.82 (m, 2H), 1.62
(m, 2H), 1.38 (m, 2H), 1.1 (t, J ) 7.5 Hz, 3H), 1.07 (s, 6H),
0.99 (t, J ) 7.2 Hz, 3H). 13C NMR (CDCl3): δ 163.0 (CdN),
57.5, 51.8, 49.6, 32.5, 30.3, 26.6, 24.2, 21.3, 19.9, 13.9, 13.8.
HRMS (FAB): m/z calcd for C13H27IN2 (M - I)+ 211.2174,
found 211.2172. Anal. Calcd for C13H27IN2: C, 46.16; H, 8.05;
N, 8.28. Found: C, 46.15; H, 8.18; N, 8.13.
1,3-Diben zyl-5,5-d im eth yltetr a h yd r o-2(1H)-p yr im id i-
n on e (5h ) a n d 1-Ben zyl-2-p h en yl-5,5-d im eth yl-1,4,5,6-
tetr a h yd r op yr im id in e (6h ). The general procedure afforded
5h as a yellow oil in 19% yield and the hydroiodide salt of 6h
as a white solid in 22% yield from 0.27 g of N,N′-dibenzyl-2,2-
dimethylpropane-1,3-diamine. The following are the data for
1
5h . IR (CH2Cl2): νCO 1627 cm-1. H NMR (CDCl3): δ 7.24 (m,
10H), 4.52 (s, 4H), 2.77 (s, 4H), 0.82 (s, 6H). 13C NMR
(CDCl3): δ 155.7 (CdO), 138.4, 128.3, 128.1, 127.0, 56.7, 51.6,
28.5, 24.4. HRMS (FAB): m/z calcd for C20H24N2O (M + H+)
309.1971, found 309.1970. The following are the data for 6h ‚
(65) Cortes, S.; Kohn, H. J . Org. Chem. 1983, 48, 2246-2254.

4092 J . Org. Chem., Vol. 67, No. 12, 2002
Qian et al.
HI. Mp: 198-201 °C. 1H NMR (CDCl3): δ 10.0 (br s, 1H),
7.14-7.80 (m, 10H), 4.58 (s, 2H), 3.39 (s, 2H), 3.30 (s, 2H),
1.05 (s, 6H). 13C NMR (CDCl3): δ 161.7 (CdN), 133.6, 132.7,
129.7, 129.5, 129.0, 128.7, 127.9, 56.5, 56.4, 50.4, 27.0, 24.5.
HRMS (FAB): m/z calcd for C19H23IN2 (M - I)+ 279.1861,
found 279.1853. Anal. Calcd for C19H23IN2: C, 56.17; H, 5.71;
N, 6.89. Found: C, 56.72; H, 6.10; N, 6.60.
Ack n ow led gm en t. We acknowledge the US/France
REU Site, funded by the National Science Foundation,
the CNRS, the French Ministry of Foreign Affairs, and
the French Ministry of National Education, Research,
and Technology. We also acknowledge partial funding
from the Aquitaine Regional Government. M.K. is a
visiting graduate student from Hokkaido University.
1,3,5,5-Tetr am eth yltetr ah ydr o-2(1H)-pyr im idin on e (5i).
The crude product from the general procedure was chromato-
graphed on silica using hexane/EtOAc as eluent to afford pure
5i as a colorless oil in 37% yield from 0.20 g of N,N′-dimethyl-
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 4b,c,h and 5b,c,f-i. This material is available free
of charge via the Internet at http://pubs.acs.org.
2,2-dimethylpropane-1,3-diamine. IR (Nujol): νCO 1644 cm-1
.
1H NMR (CDCl3): δ 2.86 (s, 10H), 0.97 (s, 6H). 13C NMR
(CDCl3): δ 156.2 (CdO), 60.0, 36.1, 29.1, 24.8. HRMS (FAB):
m/z calcd for C8H16N2O (M+) 156.1262, found 156.1248.
J O0109319