Oxidative carbonylation of primary amines to ureas using tungsten carbonyl catalysts

Article abstract of DOI:10.1021/om980389n

The iodo-bridged tungsten(IV) dimer [(CO)₂ W(NPh)I₂]₂ (1) has been demonstrated to be a catalyst for the oxidative carbonylation of primary amines to 1,3-disubstituted ureas. The amine complexes (CO)₂I₂W(NPh)(NH₂R) [R = n-Bu (2a), R = n-Pr (2b), R = t-Bu (2c)] and [(CO)₂IW(NPh)(NH₂R)₂]⁺ [R = n-Bu (4a), R = n-Pr (4b)] have been isolated from reactions of 1 with primary amines. Complexes 2a-c and 4a-b have been shown to lie along the pathway to the catalytically active species. In addition, it has been determined that primary amines can be oxidatively carbonylated to 1,3-disubstituted ureas using W(CO)₆ as catalyst and I₂ as the oxidant.

Full text of DOI:10.1021/om980389n

Organometallics 1998, 17, 4037-4041
4037
Oxid a tive Ca r bon yla tion of P r im a r y Am in es to Ur ea s
Usin g Tu n gsten Ca r bon yl Ca ta lysts1
J ennifer E. McCusker, J ennifer Logan, and Lisa McElwee-White*
Department of Chemistry, University of Florida, Gainesville, Florida 32611
Received May 18, 1998
The iodo-bridged tungsten(IV) dimer [(CO)2W(NPh)I2]2 (1) has been demonstrated to be a
catalyst for the oxidative carbonylation of primary amines to 1,3-disubstituted ureas. The
amine complexes (CO)2I2W(NPh)(NH2R) [R ) n-Bu (2a ), R ) n-Pr (2b), R ) t-Bu (2c)] and
[(CO)2IW(NPh)(NH2R)2]+ [R ) n-Bu (4a ), R ) n-Pr (4b)] have been isolated from reactions
of 1 with primary amines. Complexes 2a -c and 4a -b have been shown to lie along the
pathway to the catalytically active species. In addition, it has been determined that primary
amines can be oxidatively carbonylated to 1,3-disubstituted ureas using W(CO)6 as catalyst
and I2 as the oxidant.
In tr od u ction
report catalytic oxidative carbonylation of primary
amines using 1 as the precatalyst. In addition, further
mechanistic information on the process is provided.
During the course of these studies, we also determined
that primary amines could be carbonylated to 1,3-
disubstituted ureas using W(CO)6 as precatalyst and I2
as the oxidizing agent.
Although transition metal-catalyzed carbonylation
reactions have been investigated extensively over many
years,2 the topic remains of interest. Oxidative conver-
sion of primary amines into ureas3 and the related
conversion of primary amines into carbamates4 have
been the subject of multiple reports. Catalysts involving
Ni,3b Co,3d,4f Rh,4g,h Mn,3c,5,6 Ru,3a,4g,h and, most com-
monly, Pd3e,4 have been demonstrated to carry out this
reaction. However, although many transition metal
systems have been examined, carbonylation reactions
involving group 6 metals7,8 still remain rare.
We recently reported the carbonylation of primary
amines to ureas using the iodo-bridged dimer [(CO)2W-
(NPh)I2]2 (1) as a stoichiometric reagent (eq 1).9 We now
Resu lts a n d Discu ssion
Mech a n istic Stu d ies. To obtain mechanistic infor-
mation about the carbonylation of amines by 1, the
necessary equivalents of amine were added in a stepwise
manner. As shown in eq 2, reaction of 1 with 2 equiv
of primary amine yielded the complexes (CO)2I2W(NPh)-
(NH2R) (2), in which the iodo-bridges of 1 have been
replaced by amine ligands. Similar conversions of 1 to
(1) Dedicated to Professor Warren Roper on the occasion of his 60th
birthday.
(2) For monographs on the topic, see: (a) Colquhoun, H. M.;
Thompson, D. J .; Twigg, M. V. Carbonylation: Direct Synthesis of
Carbonyl Compounds; Plenum: New York, 1991. (b) Sheldon, R. A.
Chemicals from Synthesis Gas; Dordrecht: Boston, 1983. (c) Wender,
I.; Pino, P. Organic Syntheses via Metal Carbonyls; Wiley-Inter-
science: New York, 1968.
(3) (a) Mulla, S. A. R.; Rode, C. V.; Kelkar, A. A.; Gupte, S. P. J .
Mol. Catal. A 1997, 122, 103-109. (b) Giannocaro, P.; Nobile, C. F.;
Mastrorilli, P.; Ravasio, N. J . Organomet. Chem. 1991, 419, 251-258.
(c) Srivastava, S. C.; Shrimal, A. K.; Srivastava, A. J . Organomet.
Chem. 1991, 414, 65-69. (d) Bassoli, A.; Rindone, B.; Tollari, S.;
Chioccara, F. J . Mol. Catal. 1990, 60, 41-48. (e) Pri-Bar, I.; Alper, H.
Can. J . Chem. 1990, 68, 1544-1547.
(4) (a) Valli, V. L. K.; Alper, H. Organometallics 1995, 14, 80-82.
(b) Pri-Bar, I.; Schwartz, J . J . Org. Chem. 1995, 60, 8124-8125. (c)
Hartstock, F. W.; Herrington, D. G.; McMahon, L. B. Tetrahedron Lett.
1994, 35, 8761-8764. (d) Kanagasabapathy, S.; Gupte, S. P.; Chaudhari,
R. V. Ind. Eng. Chem. Res. 1994, 33, 1-6. (e) Giannocaro, P. J .
Organomet. Chem. 1994, 470, 249-252. (f) Leung, T. W.; Dombek, B.
D. J . Chem. Soc., Chem. Commun. 1992, 205-206. (g) Mulla, S. A. R.;
Gupte, S. P.; Chaudhari, R. V. J . Mol. Catal. 1991, 67, L7-L10. (h)
Fukuoka, S.; Chono, M.; Kohno, M. J . Org. Chem. 1984, 49, 1458-
1460. (i) Fukuoka, S.; Chono, M.; Kohno, M. J . Chem. Soc., Chem.
Commun. 1984, 399-400. (j) Alper, H.; Hartstock, F. W. J . Chem. Soc.,
Chem. Commun. 1985, 1141-1142.
mononuclear complexes by addition of two-electron
donor ligands have been described previously.10 Pri-
mary amine complexes derived from n-butylamine (2a ),
n-propylamine (2b), and tert-butylamine (2c) were
prepared, establishing that this is a general reaction.
In addition, compounds 2a , 2b, and 2c can also be
generated by reacting the previously reported acetoni-
trile complex (CO)2I2W(NPh)(NCCH3) (3) with 1 equiv
of amine (eq 3).10
(5) Dombek, B. D.; Angelici, R. J . J . Organomet. Chem. 1977, 134,
203-217.
(6) Calderazzo, F. Inorg. Chem. 1965, 4, 293-296.
(7) J etz, W.; Angelici, R. J . J . Am. Chem. Soc. 1972, 94, 3799-3802.
(8) Zoeller, J . R.; Blakely, E. M.; Moncier, R. M.; Dickson, T. J . Catal.
Today 1997, 36, 227-241.
(10) (a) McGowan, P. C.; Massey, S. T.; Abboud, K. A.; McElwee-
White, L. J . Am Chem. Soc. 1994, 116, 7419-7420. (b) Barnett, N. D.
R.; Massey, S. T.; McGowan, P. C.; Wild, J . J .; McElwee-White, L.
Organometallics 1996, 15, 424-428.
(9) McCusker, J . E.; Abboud, K. A.; McElwee-White, L. Organome-
tallics 1997, 16, 3863-3866.
S0276-7333(98)00389-6 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 08/13/1998

4038 Organometallics, Vol. 17, No. 18, 1998
McCusker et al.
Sch em e 1
Amine complexes 2a -c have been characterized by
comparison of their spectral data to those of similar
species.9,10 The 1H NMR spectra of 2a -c reflect the
chirality at the metal center. The diastereotopic NH2
protons are seen as two broad singlets at 4.33 and 4.02
ppm for 2a and 4.35 and 4.03 ppm for 2b, while for 2c
they appear as two sets of doublets at 4.83 and 4.07
ppm. The 13C spectra of these complexes all exhibit two
inequivalent carbonyl signals, which allow assignment
of the carbonyl ligands as cis. Additional evidence for
the stereochemistry at the metal can be found in the
two IR stretches for the cis-carbonyl ligands.
Reaction of 1 with 4 equiv of a primary amine
produces the [(CO)2IW(NPh)(NH2R)2]+ complex 4 (eq 4),
in which one of the iodide ligands of 2 has been replaced
by an amine. The n-butylamine (4a ) and n-propylamine
to the carbamoyl complex (η5-C5H5)W(CO)3(CONHCH3)
upon reaction with 2 equiv of CH3NH2. Formation of 5
from 4 would require an analogous nucleophilic attack
of the amine on a carbonyl ligand, followed by depro-
tonation with a second equivalent of the amine. The
carbonyl stretching frequency of 2066 cm-1 for 4 falls
within the range where nucleophilic attack is expected
to occur,11a rendering the conversion of 4 to 5 a reason-
able possibility.
Given that the carbamoyl IR stretches disappear upon
exposure of the reaction mixtures to air, the next step
in the mechanism appears to be oxidation. Upon
oxidation of the complex, the carbamoyl proton would
become more acidic. Deprotonation of 5 by the excess
amine would form an isocyanate complex such as 7.
Although isocyanates are likely intermediates in the
carbonylation of amines by 1 and have been isolated
from the reaction of Et3N with (η5-C5H5)W(CO)3-
(CONHCH3),7 no free or coordinated isocyanates have
been detected in the reaction mixtures of amines with
1. However, nucleophilic attack of an amine on either
cationic isocyanate complex 7 or the free isocyanate to
form the 1,3-disubstituted ureas would be expected to
be facile.
derivatives (4b) have been successfully synthesized in
this manner. These bis(amine) cations are also avail-
able via reaction of 2a ,b with 1 equiv of the correspond-
ing amine or by addition of 2 equiv of amine to 3. The
1H NMR spectra of these complexes exhibit diaste-
reotopic NH2 protons at 5.4 and 4.7 ppm for 4a and at
5.2 and 4.9 ppm for 4b. The 13C spectra of these
complexes contain one carbonyl signal each, as well as
only one set of alkyl signals. The stereochemical as-
signment of the carbonyl ligands as cis derives from the
two IR stretches at 2060 and 1989 cm-1 for both 4a and
4b.
The monomeric amine complexes were demonstrated
to be on the reaction pathway by reaction of amine
complexes 2a , 2b, and 2c with excess amine under
oxidative conditions. As expected, the corresponding
1,3-disubstituted ureas were produced. Similar experi-
ments involving the cationic bis(amine) complexes 4a
and 4b established that the reaction pathway was also
accessible from these species. Attempts to isolate
complexes further along the pathway by reaction of 1
with additional equivalents of amine proved to be
unsuccessful.
Although the remaining intermediates in the carbo-
nylation process were too reactive to be characterized,
literature precedent and the results of further experi-
ments (vide infra) suggest the mechanism shown in
Scheme 1. IR spectra of unoxidized reaction mixtures
A less well-precedented mechanistic option is depro-
tonation of one of the amine ligands of 4, followed by
migratory insertion of the carbonyl into the tungsten-
nitrogen bond. A similar sequence has been reported
by Wang for a dicarbonyl rhenium cation bearing an
amine tethered to a Cp ring.12 In this system, depro-
of 1 with n-butylamine exhibited a stretch at 1589 cm-1
,
consistent with the presence of carbamoyl ligands.11
Intermediacy of carbamoyl complex 5 (Scheme 1) or a
related species is supported by Angelici’s work on the
carbonylation of CH3NH2 by [(η5-C5H5)W(CO)4]PF6,7 for
which the first step is conversion of [(η5-C5H5)W(CO)4]+
(11) (a) Angelici, R. J . Acc. Chem. Res. 1972, 5, 335-341. (b)
Anderson, S.; Cook, D. J .; Hill, A. F. Organometallics 1997, 16, 5595-
5597.
(12) Wang, T.-F.; Hwu, C.-C.; Tsai, C.-W.; Wen, Y.-S. Organome-
tallics 1998, 17, 131-138.

Oxidative Carbonylation of Primary Amines to Ureas
Organometallics, Vol. 17, No. 18, 1998 4039
Ta ble 1. Ca ta lytic Ca r bon yla tion of n -Bu tyla m in e
metric reaction. Use of chlorinated solvents (CH2Cl2,
CHCl3, or CCl4) resulted in the highest yields of product.
Ureas were also formed in coordinating solvents such
as acetonitrile and THF. However the yields were quite
low, probably because these solvents compete with the
amines for coordination sites. Since the starting dimer
1 is insoluble in most other common solvents, the range
of usable solvents is limited.
The use of iodide complexes or the addition of iodine-
containing promoters3e,4b,d,f-i is common in the literature
on oxidative carbonylation, especially in Pd-catalyzed
reactions. Since the role of the iodide ligands in the
reactivity of 1 was unclear, a comparison of different
halide ligands was carried out using complexes 3, 9a ,
and 9b.15 When iodide complex 3 was reacted with 10
equiv of n-butylamine under the stoichiometric condi-
tions, 0.5 equiv of di-n-butylurea was produced per
equivalent of 1. In similar experiments, the bromide
complex 9a and the chloride complex 9b were reacted
with excess primary amines under the stoichiometric
oxidative conditions. However, neither 9a nor 9b suc-
cessfully formed ureas or any other detectable organic
products, thus indicating that iodide ligands are neces-
sary for the carbonylation reaction.
Usin g Dim er 1
catalyst
oxidant
TONa
1
1
1
1
1
1
1
1
1
I2
I2
PhIO
Me3NO
t-BuOOH
t-BuOO-t-Bu
Ce(IV)
CuCl2
15b
25c
12b
0.4b
1.2b
1.2b
6b
4b
air
1.2d
a
b
Turnover number per equivalent of dimer 1. Reaction condi-
tions: amine (22 mmol), oxidizing agent (22 mmol), CO (100 atm),
dimer 1 (0.22 mmol), room temperature, 24 h. c Reaction condi-
tions: amine (22 mmol), oxidizing agent (22 mmol), CO (100 atm),
K2CO3 (22 mol), dimer 1 (0.22 mmol), room temperature, 24 h.
d
Reaction conditions: amine (71 mmol), CO (800 psi), air (200
psi), dimer 1 (0.24 mmol), room temperature, 24 h.
tonation of the coordinated amine triggers CO insertion
to produce an η2-carbamoyl complex. Further addition
of a two-electron donor ligand converts the carbamoyl
to an η1 configuration. Application of this sequence to
amine complex 4 would produce η2-carbamoyl complex
6 (eq 5). Addition of another equivalent of amine to 6
would produce intermediate 5, at which point this route
would converge with Scheme 1.13
In efforts to determine the optimum conditions for
catalytic conversion of primary amines to ureas, several
oxidizing agents were examined (Table 1). Since iodide
ligands facilitate the carbonylation reaction, I2 was a
reasonable choice as an oxidizing agent.3e,4b,16 In fact,
I2 was the most effective oxidant examined, as measured
by the number of catalyst turnovers. The addition of
K2CO3 also proved to be important.3e,4b,16 In the absence
of added base, the turnovers are limited by the produc-
tion of 2 equiv of the amine hydroiodide (RNH3I) per
equivalent of urea. If the reaction is run in the presence
of K2CO3, the amine salt byproduct can be recycled.
Upon addition of K2CO3 to the I2-oxidized reaction, di-
n-butylurea was produced in an amount corresponding
to 25 turnovers per equivalent of dimer 1, or 52% yield
with respect to amine (Table 1).
Among the other oxidants that were examined, iodo-
sylbenzene gave the best results of the oxo transfer
reagents. Other oxo transfer reagents such as tri-
methylamine N-oxide also work but tend to be prob-
lematic when not extremely dry, since the carbonylation
reaction is moisture sensitive. In the presence of water,
yields of ureas are low. For this reason, hydrogen
peroxide is a poor choice as oxidizing agent. Di-tert-
butylperoxide and tert-butylhydroperoxide do induce
carbonylation of amines by 1, although multiple turn-
overs cannot be obtained and several unidentified side
products are also formed. Outer sphere oxidants, such
as Ce(IV) and CuCl2, rendered the reaction catalytic as
well. Since the original stoichiometric conditions uti-
Ca ta lytic Ca r bon yla tion of Am in es to Ur ea s.
The mechanism proposed in Scheme 1 raised the
interesting possibility that the reaction could be ren-
dered catalytic in the presence of excess CO. As shown
in Scheme 1, loss of isocyanate from 7 would yield the
unsaturated intermediate 8. Addition of CO to 8 would
regenerate bis(amine) complex 4, closing the catalytic
cycle.
The expectation that the catalytic cycle could be closed
was confirmed by providing excess CO to the reaction
mixture. For example, when 1, 100 equiv of n-butyl-
amine, and 100 equiv of the oxidizing agent iodosylben-
zene were placed in a 125 mL Parr high-pressure vessel
and pressurized with 100 atm CO, di-n-butylurea was
produced in an amount corresponding to 12 turnovers
per equivalent of dimer 1 (Table 1).14
Before catalytic reactions were studied further, effort
was devoted to optimizing conditions for the stoichio-
(13) A reviewer has suggested a third mechanistic option in which
oxidation of 5 leads to the preferential deprotonation of an amine ligand
instead of the carbamoyl ligand. Reductive elimination from the
resulting carbamoyl-amido complex [(CO)IW(NPh)(NH2R)(NHR)-
(CONHR)]+ would then yield the urea. We cannot rule out this
alternative. However, we still favor the mechanism shown in Scheme
1 due to the literature precedent in ref 6 and our observation of small
amounts of isocyanates in some of the reaction mixtures by GC. The
GC evidence should be viewed with some skepticism since the observed
isocyanates could arise from cracking of labile species in the injection
port. However, their presence does provide some support for Scheme
1.
(14) Note that since dimer 1 is cleaved to mononuclear complexes
in the presence of amines, each tungsten is turning over six times.
However, since the tungsten was added to the reaction mixtures in
the form of 1, the yields are reported on this basis.
(15) He, Y.-X., McGowan, P. C.; Abboud, K. A.; McElwee-White, L.
J . Chem. Soc., Dalton Trans., in press.
(16) Dahlen, G. M.; Sen, A. Macromolecules 1993, 26, 1784-1786.

4040 Organometallics, Vol. 17, No. 18, 1998
McCusker et al.
Ta ble 2. Ca ta lytic Ca r bon yla tion of n -Bu tyla m in e
was not significant with respect to the quantities of urea
obtained in the presence of the catalysts.
Usin g W(CO)6
No mechanistic studies were performed on the W(CO)6
system, but it is known that oxidation of W(CO)6 with
I2 results in the formation of iodocarbonyl complexes
that readily undergo ligand exchange with nucleo-
philes.17,18 The mechanism is thus likely to follow a
cycle similar to that proposed in Scheme 1 for the
chemistry of dimer 1.
equiv
n-BuNH2
equiv
K2CO3
catalyst
equiv I2
time (h) TONa,b
W(CO)6
W(CO)6
W(CO)6
W(CO)6
W(CO)6
W(CO)6
50
50
50
50
50
24
0.5
0
3
25
25
25
25
50
3
24
24
24
9
9
15
39
50
100
100
Su m m a r y. We have developed mild methods to
convert primary amines to 1,3-disubstituted ureas using
dimer 1 as precatalyst and I2 as the oxidant. A possible
pathway involving carbamoyl and isocyanate complexes
has been proposed. Based on the reactions of 1, it was
established that W(CO)6 could also be used as a carbo-
nylation catalyst in the presence of I2 as the oxidant.
These reactions are rare examples of catalytic carbony-
lation using a group 6 metal. Further work on related
carbonylation reactions is in progress.
a
b
Turnover number per equivalent of W(CO)6. Reaction condi-
tions: CO (100 atm), W(CO)6 (0.21 mmol), room temperature.
lized air as the oxidant, efforts were made to explore
O2 as the oxidant in the catalytic reaction. However,
using a mixture of 800 psi CO and 200 psi air did not
induce multiple catalyst turnovers, presumably because
an intermediate species in the catalytic cycle is oxygen
sensitive. For example, if the unsaturated complex 8
were to react with oxygen, the stoichiometric reaction
would run in the presence of air as observed, but loss
of 8 would prevent the reaction from turning over in
the presence of O2.
Exp er im en ta l Section
Gen er a l. Standard inert atmosphere Schlenk, cannula,
and glovebox techniques and freshly distilled solvents were
used in all experiments unless stated otherwise. Diethyl ether
was distilled from sodium/benzophenone. Methylene chloride
and hexane were distilled over calcium hydride. All NMR
solvents were degassed by three freeze-pump-thaw cycles,
then stored in an inert-atmosphere glovebox over 3 Å molec-
ular sieves. All other chemicals were purchased in reagent
grade and used with no further purification unless stated
otherwise. Authentic samples of the ureas were prepared from
the corresponding amines and isocyanates.19
1H and 13C NMR spectra were recorded on a Varian Gemini
300 spectrometer. IR spectra were recorded on a Perkin-Elmer
1600 FTIR. GC was performed on an HP5890 chromatograph
containing a 30 m × 0.75 mm column of SPB-20 on fused silica.
High-resolution mass spectrometry was performed by the
University of Florida analytical service. Elemental analysis
was performed by Robertson Microlit Laboratories.
Ca ta lytic Ca r bon yla tion of Am in es w ith W(CO)6
a s Ca ta lyst. Catalysis of amine carbonylation by 1 in
the presence of iodine suggested that other tungsten
carbonyl iodide complexes might also serve as catalysts.
The simplest choice as precatalyst was the readily
available, inexpensive, and air stable W(CO)6. Control
experiments for the catalytic carbonylation of n-butyl-
amine were carried out using W(CO)6 as catalyst, but
omitting the I2 oxidant. As expected, when W(CO)6 and
50 equiv of n-butylamine are placed in a 125 mL Parr
high-pressure vessel and pressurized with 100 atm CO,
no di-n-butylurea is produced (Table 2). This is consis-
tent with the conventional wisdom that group 6 metal
complexes are poor carbonylation catalysts.8
However, when W(CO)6, 50 equiv of n-butylamine,
and 25 equiv of iodine are placed in a 125 mL Parr high-
pressure vessel and pressurized with 100 atm CO, di-
n-butylurea is produced in an amount corresponding to
9 turnovers per equivalent of W(CO)6 (Table 2). As
discussed in conjunction with experiments involving
dimer 1 (vide supra), oxidative carbonylation of amines
to ureas produces 2 equiv of amine hydroiodide per
equivalent of urea. Once again, addition of K2CO3 to
release the free amine from the salt increases the
turnover number dramatically. Reaction of W(CO)6, 100
equiv of n-butylamine, 50 equiv of iodine, and 100 equiv
of K2CO3 in a 125 mL Parr high-pressure vessel pres-
surized with 100 atm CO produces di-n-butylurea in an
amount corresponding to 39 turnovers per equivalent
of W(CO)6, or 80% yield with respect to amine (Table
2).
Syn th esis of (CO)2I2W(NP h )(H2NCH2CH2CH2CH3) (2a ).
To a stirred solution of dimer 1 (77 mg, 0.066 mmol) in 20 mL
of CH2Cl2 was added 2 equiv of n-butylamine (0.013 mL, 0.13
mmol). A color change from red to purple occurred im-
mediately. After 1 h of stirring the CH2Cl2 was removed under
vacuum. The red oil was redissolved in 2 mL of CH2Cl2, and
5 mL of hexane was layered on top. The solution was placed
in a -40 °C freezer overnight. The red solution was cannu-
lated away from any precipitate and concentrated. The
resulting red oil was collected into 50 mL of ether and
cannulated away from any insoluble particles. The solvent
was removed under vacuum to yield 2a as a red oil (11 mg,
13% yield). IR (CH2Cl2): νCO 2066, 1988 cm-1
.
1H NMR
(CDCl3): δ 7.50-7.21 (m, 5H), 4.33 (s, 1H), 4.02 (s, 1H), 3.37
(m, 2H), 1.68 (m, 2H), 1.44 (m, 2H), 0.98 (t, 3H). 13C NMR
(CDCl3): δ 208.6, 208.5, 153.3, 129.5, 129.1, 124.9, 51.9, 34.6,
19.5, 13.4. HRMS (FAB): 629.886 (M - CO)+; calcd, 629.8824.
Lability of the compound precluded obtaining an elemental
analysis.
Additional control experiments included addition of
I2 in only a trace amount. Since the reaction was
starved for oxidant, only a trace amount of urea was
formed. The potential participation of the stainless steel
walls of the reactor was probed by control experiments
using a glass liner. The number of turnovers obtained
in the presence and absence of the glass liner did not
differ significantly with either W(CO)6 or dimer 1 as the
catalyst. Furthermore, the control experiment in which
no catalyst was added but the solution was allowed to
contact the stainless steel walls was also performed. A
negligible amount of urea was detected, but the amount
Syn th esis of (CO)2I2W(NP h )(H2NCH2CH2CH3) (2b). To
a stirred solution of dimer 1 (53 mg, 0.045 mmol) in 20 mL of
(17) (a) Colton, R.; Rix, C. J . Aust. J . Chem. 1969, 22, 305-310. (b)
Schmid, G.; Boese, R.; Welz, E. Chem. Ber. 1975, 108, 260-264. (c)
Umland, P.; Vahrenkamp, H. Chem. Ber. 1982, 115, 3565-3579. (d)
Schmid, G.; Boese, R. Chem. Ber. 1976, 109, 2148-2153.
(18) The reported reactions of I2 with W(CO)6 involve photolysis.
However, the more complex carbonylation reaction mixtures may
provide alternative pathways for oxidation.
(19) For a review on synthesis of ureas, see: Vishnyokova, T. P.;
Golubeva, I. A.; Glebova, E. V. Russ. Chem. Rev. 1985, 54, 249-261.

Oxidative Carbonylation of Primary Amines to Ureas
Organometallics, Vol. 17, No. 18, 1998 4041
CH2Cl2 was added 2 equiv of n-propylamine (0.0075 mL, 0.091
mmol). A color change from red to purple occurred im-
mediately. After 1 h of stirring the CH2Cl2 was removed under
vacuum. The red oil was redissolved in 2 mL of CH2Cl2, and
5 mL of hexane was layered on top. The solution was placed
in a -40 °C freezer overnight. The red solution was cannu-
lated away from any precipitate and concentrated. The
resulting red oil was taken up into 50 mL of ether and
cannulated away from any insoluble particles. The solvent
was removed under vacuum to yield 2b as a red oil (53 mg,
Ca ta lytic Ca r bon yla tion of P r im a r y Am in es w ith 1
(Solid Oxid a n ts). The following procedure is typical: To a
stirred solution of dimer 1 (260 mg, 0.22 mmol) in 30 mL of
CH2Cl2 was added 100 equiv of n-butylamine (2.2 mL, 22
mmol). A color change from red to deep purple occurred
immediately. The solution was stirred under N2 for 20 min,
then added to a Parr high-pressure vessel, followed by addition
of iodosylbenzene (4.9 g, 22 mmol). The vessel was then
charged with 100 atm of CO and left to stir under pressure
for 24 h. The pressure was released and the lemon yellow
solution was then filtered through Celite and concentrated.
The resulting yellow solid was dissolved in ethyl acetate and
chromatographed on silica with ethyl acetate as eluent to
obtain a white solid (616 mg, 3.59 mmol, 12 turnovers per
equivalent of dimer 1). The solid was identified as di-n-
butylurea by comparison with an authentic sample.
Ca ta lytic Ca r bon yla tion of P r im a r y Am in es w ith 1
(Air Oxid a n t). Caution! High pressure mixtures of CO and
O2 are explosive under certain conditions.20 To a stirred
solution of dimer 1 (278 mg, 0.24 mmol) in 50 mL of CH2Cl2
was added 300 equiv of n-butylamine (7.34 mL, 71 mmol). A
color change from red to deep purple occurred immediately.
The solution was stirred under N2 for 20 min, then added to
a Parr high-pressure vessel. The vessel was then charged with
800 psi of CO, followed by 200 psi of air, and left to stir under
pressure for 24 h. The pressure was released, and the pale
yellow solution was then filtered through Celite and concen-
trated. The resulting yellow oil was dissolved in ethyl acetate
and chromatographed on silica with ethyl acetate as eluent
to obtain a white solid (49 mg, 0.28 mmol, 1.2 turnovers per
equivalent of dimer 1). The solid was identified as di-n-
butylurea by comparison with an authentic sample.
Ca ta lytic Ca r bon yla tion of P r im a r y Am in es w ith
W(CO)6. The following procedure is typical: To a stirred
solution of W(CO)6 (75 mg, 0.21 mmol) in 25 mL of CH2Cl2 in
the glass liner of a Parr high-pressure vessel was added 100
equiv of n-butylamine (2 mL, 21 mmol), K2CO3 (3.52 g, 21
mmol), and 50 equiv of iodine (2.56 g, 10.5 mmol). The vessel
was then charged with 100 atm of CO and left to stir under
pressure for 24 h. The pressure was released, and the maroon
solution was filtered and then rinsed with a Na2SO3 solution.
The resulting pale yellow solution was then dried with MgSO4
and filtered. The solution was concentrated to yield a pale
yellow solid, which was then rinsed with 25 mL of hexane.
The solid was then taken up in 25 mL of ether and filtered to
remove any residual W(CO)6. The ether solution was concen-
trated to obtain a white solid (1.4 g, 8.13 mmol, 39 turnovers
per equivalent of tungsten). The solid was identified as di-n-
butylurea by comparison with an authentic sample.
63% yield). IR (CH2Cl2): νCO 2067, 1988 cm-1
.
1H NMR
(CDCl3): δ 7.50-7.26 (m, 5H), 4.35 (s, 1H), 4.03 (s, 1H), 3.34
(m, 2H), 1.75 (m, 2H), 1.04 (t, 3H). 13C NMR (CDCl3): δ 208.5,
208.4, 153.8, 129.4, 129.1, 124.9, 53.8, 25.8, 10.6. Lability of
the compound precluded obtaining an elemental analysis.
Syn th esis of (CO)2I2W(NP h )(H2NC(CH3)3) (2c). To a
stirred solution of dimer 1 (113 mg, 0.097 mmol) in 20 mL of
CH2Cl2 was added 2 equiv of tert-butylamine (0.020 mL, 0.195
mmol). A color change from red to purple occurred im-
mediately. After 2 h of stirring the CH2Cl2 was removed under
vacuum. The purple oil was redissolved in 2 mL of CH2Cl2,
and 5 mL of hexane was layered on top. The solution was
placed in a -40 °C freezer overnight. The red solution was
cannulated away from any precipitate and concentrated. The
resulting red solid was taken up into 50 mL of hexane and
cannulated away from any insoluble particles. The solvent
was removed under vacuum to yield 2c as a red solid (45 mg,
70% yield). IR (CH2Cl2): νCO 2067, 1987 cm-1
.
1H NMR
(CDCl3): δ 7.49-7.18 (m, 5H), 4.83 (d, 1H), 4.07 (d, 1H), 1.44
(t, 9H). 13C NMR (CDCl3): δ 210.3, 207.5, 153.3, 129.5, 129.2,
124.9, 55.8, 30.8. HRMS (FAB): 629.886 (M - CO)+; calcd,
629.8824. Lability of the compound precluded obtaining an
elemental analysis.
Syn t h esis of [(CO)2IW(NP h )(H2NCH2CH 2CH 2CH 3)2]I
(4a ). To a stirred solution of dimer 1 (100 mg, 0.086 mmol)
in 20 mL of CH2Cl2 was added 4 equiv of n-butylamine (0.034
mL, 0.344 mmol) at 0 °C. A color change from red to maroon
occurred immediately. After 2 h of stirring, the CH2Cl2 was
removed under vacuum. The red oil was redissolved in 2 mL
of CH2Cl2, and 25 mL of ether was added. The solution was
placed in a -40 °C freezer for 2 days. The resulting orange
powder (4a ) was isolated by filtration (43 mg, 40% yield). IR
(CH2Cl2): νCO, 2060, 1989 cm-1 1H NMR (CD2Cl2): δ 7.28 (m,
.
3H), 7.04 (d, 2H), 5.35 (s, 2H), 4.88 (s, 2H), 3.15 (m, 4H), 1.85
(m, 4H), 1.41 (m, 4H), 0.91 (t, 6H). 13C NMR (CD2Cl2): δ 214.7,
154.5, 129.3, 127.5, 127.3, 50.1, 34.3, 20.5, 13.8. Lability of
the compound precluded obtaining an elemental analysis.
Syn th esis of [(CO)2IW(NP h )(H2NCH2CH2CH3)2]I (4b).
To a stirred solution of dimer 1 (111 mg, 0.094 mmol) in 25
mL of CH2Cl2 was added 8 equiv of n-propylamine (0.062 mL,
0.752 mmol) at 0 °C. A color change from red to maroon
occurred immediately. After 1 h of stirring, the CH2Cl2 was
removed under vacuum. The red oil was redissolved in 2 mL
of CH2Cl2, and 25 mL of ether was added. The solution was
placed in a -40 °C freezer for 2 days. The resulting orange
powder (4b) was isolated by filtration (30 mg, 26% yield). IR
Ack n ow led gm en t. Support of this work was pro-
vided by the Office of Naval Research. Partial support
for J .L. was provided by the NSF-REU program at the
University of Florida (CHE-9424058). We thank Dr.
Yingxia He for providing samples of compounds 9a and
9b and Dr. J ames Pawlow for assistance with the high-
pressure CO/air mixture reaction.
(CH2Cl2): νCO, 2060, 1989 cm-1 1H NMR (CD2Cl2): δ 7.29 (m,
.
3H), 7.06 (d, 2H), 5.32 (s, 2H), 4.88 (s, 2H), 3.08 (m, 4H), 1.92
(m, 4H), 0.98 (t, 6H). 13C NMR (CD2Cl2): δ 214.7, 154.5, 129.3,
127.5, 127.4, 52.1, 25.8, 11.6. Lability of the compound
precluded obtaining an elemental analysis.
OM980389N
(20) Proper safety precautions against explosion hazards were
employed during this work.