Soluble and polymer-anchored rhodium catalyst for carbonylation reaction: Kinetics and mechanism of diphenylurea formation

Article abstract of DOI:10.1006/jcat.2002.3665

The catalytic conversion of PhNO₂ to PhNHCONHPh by soluble and polymer-supported Rh catalysts in mild coordinating solvents containing methanol was studied. The carbonylation of nitrobenzene at low temperature and atmospheric or elevated pressure using soluble and polymer supported Rh species was presented. N,N′-diphenylurea (DPU) was the main product under moderate CO pressure while methyl-N-phenyl carbamate was formed predominantly as P_CO and methanol concentration was increased. The highest conversion of PhNO₂ (100%) to DPU was achieved at P_CO = 60 atm and 80°C and methanol concentration of 6 M. A higher methanol concentration increased the yield of aniline at the cost of DPU. The in situ spectroscopic studies of the high-pressure catalysis established that the principal species observed in solution was [Rh(A)(PPh₃)(COOCH₃)(μ-OCH₃ ]₂ and that several intermediate steps were involved in the conversion of nitroarene to useful diarylurea in the presence of alcohol.

Full text of DOI:10.1006/jcat.2002.3665

Journal of Catalysis 210, 255–262 (2002)
doi:10.1006/jcat.2002.3665
Soluble and Polymer-Anchored Rhodium Catalyst for Carbonylation
Reaction: Kinetics and Mechanism of Diphenylurea Formation
,1
∗
D. K. Mukherjee† and C. R. Saha
∗
Department of Chemistry, Indian Institute of Technology, Kharagpur 721 302, India; and †Department of Chemistry,
Ramsaday College, Amta, Howrah-711 401, West-Bengal, India
Received November 5, 2001; revised April 30, 2002; accepted May 8, 2002
and water or aniline using a new catalytic system have also
been reported (19). The stoichiometry of this synthesis is
expressed by
The high catalytic activity of rhodium(I) complexes toward
the reductive carbonylation of nitroaromatics under high car-
bon monoxide pressure prompted investigation of the polymer-
bound complex of rhodium(I). The polymer-supported complex PhNO2 + PhNH2 + 3CO = (PhNH)2CO + 2CO2 . . . . [1]
[RhA(Ph2P–CH2–polystyrene)CO] having the same coordination
environment as [RhA(CO)PPh3] (HA; 2-aminobenzoic acid) was
Generally d8 metal complexes with π-acid ligands are
prepared and the catalytic behavior studied under various reac-
tion conditions and methanol concentrations. The carbonylation of
nitrobenzene at low temperature and atmospheric or elevated pres-
sure using both soluble and polymer-supported rhodium species
is reported. N,Nꢀ-Diphenylurea is the main product under moder-
ate carbon monoxide pressure while methyl-N-phenyl carbamate
is formed predominantly as Pco and methanol concentration is in-
creased. Spectroscopic and kinetic studies showed that the reaction
proceeds through the species [Rh(A)(PPh3)(COOCH3)(µ-OCH3)]2
and the isocyanate formed at an intermediate stage is immedi-
ately scavenged by excess amine to form N,Nꢀ-diphenylurea. A
tentative reaction mechanism based on the identification of reac-
tive intermediates has been proposed for the carbonylation process.
active for the carbonylation of nitroaromatics, although
detailed reaction mechanisms have been reported only
in few cases (20–24). Compared to other d8 metal com-
plexes, rhodium(I) compounds have been much less studied
for the carbonylation reaction. The high catalytic activity
of [RhA(CO)Py] (HA; anthranilic acid) toward reductive
carbonylation of nitroaromatics under moderate carbon
monoxide pressure (25) prompted examination of polymer-
bound rhodium(I) complexes of a similar nature. The
present paper reports the catalytic conversion of PhNO2
to PhNHCONHPh by soluble and polymer-supported
rhodium catalysts in mild coordinating solvents containing
methanol and the relationship this has to the mechanism of
the catalytic carbonylation process.
c
ꢁ 2002 Elsevier Science (USA)
INTRODUCTION
EXPERIMENTAL
The search for environmentally friendly processes to re- Instrument and Chemicals
place the use of phosgene for the production of isocyanates
Analytical-grade reagents and freshly distilled solvents
has focused attention on the homogeneous catalytic car-
bonylation of nitroarenes (1). Many industrially important
compounds, such as carbamates, ureas, azo and azoxyare-
nes, oximes, and several heterocycles, can be selectively
obtained in one step by either reductive carbonylation
of nitroaromatics (2–6) or oxidative carboxylation of ani-
lines (7–10). The nature and distribution of products de-
pend on the reaction parameters and the catalyst system.
N,Nꢀ-Diphenylurea and methyl-N-phenyl carbamate are of
growing interest in the field of chemical technology, as is ev-
ident from numerous patents and papers in the literature
(11–18). Details of the synthesis of DPU from PhNO2, CO,
were used throughout the investigation. DMF was purified
by drying over CaH2 under N2 for 24 h followed by dis-
tillation under reduced pressure. PhNH2 was purified by
vacuum distillation in an inert atmosphere prior to use. CO
(99.5%) purchased from IOL, Mumbai, and RhCl3 · 3H2O
from Arora Matthey Ltd., India, were used as received.
Macroporous polystyrene beads crosslinked with 5% DVB
were supplied as hard insoluble 20- to 50-mesh spheres of
❛
average pore diameter 800 A by Aldrich Chemical Com-
pany, USA.
The polymer-supported complex [Rh(A)CO(Ph2P–
CH2–(P)], where (P) is the polystyrene moiety having the
same coordination environment as the soluble rhodium
complex [Rh(A)CO(PPh3)], was prepared and charac-
terized as described below. Vibrational, electronic, and
1 To whom correspondence should be addressed. E-mail:
debkumarmukherjee@rediffmail.com.
255
0021-9517/02 $35.00
c
ꢁ 2002 Elsevier Science (USA)
All rights reserved.

256
MUKHERJEE AND SAHA
PMR spectra were taken with Perkin–Elmer 883, Shimadzu
[Rh A(CO) (Ph2P–CH2–polystyrene)]. . . (2). Polysty-
MPC-3700, and Bruker 200-MHz instruments, respectively. rene beads with 5% crosslinking of divinylbenzene were
XPS studies and thermal analyses were made with VG- chloromethylatedbytheprocedureadoptedbyPepper etal.
scientific ESCA lab mark (E) and Shimadzu DT-400 in- (27). The chloromethylated polymer was treated with a 1 M
struments, respectively. Gas chromatographic analysis was tetrahydrofuran solution of lithiodiphenyl phosphine for
performed with a Chrompack CP-9000 with a 15% FFAP 24 h to replace 80% of the chlorines with diphenyl phos-
ss column and flame ionization detector, with temperature phine groups (28). These beads were then equilibrated with
programming from 110 to 240◦C at a rate of 10◦C/min. The a twofold excess of dicarbonylanthranilato rhodium(I) for
progress of the reaction was monitored by periodic analysis 2 weeks. At the conclusion of the equilibration period the
of the reaction mixture by GC. Diphenylurea was identified brown beads were washed with fresh oxygen-free benzene
and estimated by HPLC using a Bondapack column with several times and dried in a vacuum.
mobile phase as 62% methanol in aqueous sodium acetate
solution.
[RhA(PPh3)(COOCH3)(µ-OCH3)]2. . . (3).
A 50-ml
stainless-steel autoclave containing [Rh(A)CO(PPh3)]
(1.2 mmol) was purged several times with CO. A solution
of o-nitrotoluene (1.5 mmol) in 15 ml of toluene was
injected into the autoclave via an exhaust valve. After
5 min, 3 ml of methanol was injected in the same manner
and the system was pressurized with CO (60 atm) and
heated to 80◦C. After 24 h, the reactor was cooled, the
pressure was released, and the solution was transferred
to a two-necked 100-ml rb flask. Solvent was removed
under vacuum and the resulting brown oil was stirred and
washed three times with 20-ml portions of diethyl ether
until a light brown solid was obtained. This was filtered and
dissolved in a minimum quantity of pure benzene, and the
process of stirring and washing with ether was continued
until the brown compound separated out. The compound
thus isolated was subjected to 1H NMR and IR studies for
Carbonylation Procedure
In a typical experiment, a DMF solution (10 ml) of the
catalyst (10−4 M) and the substrate (1.0 M) was taken in
a 50-ml glass-lined stainless-steel autoclave provided with
an inlet and outlet and equipped with a magnetic element.
The reactor was first evacuated and flushed with nitrogen.
After nitrogen was pumped out, it was immersed in a ther-
mostated silicone oil bath preheated to the desired reaction
temperature. The whole arrangement was then placed on a
magnetic stirrer and the reaction mixture was subjected to
the required pressure of pure carbon monoxide, which was
maintained constant throughout the run. At the end, the
autoclave was rapidly cooled in an ice salt bath and blown
off. The products were identified and analyzed after subse-
quent workup by GC and HPLC using authentic samples
for comparison.
1
characterization: H NMR (ppm, C6D6) 3.03 (s, OCH3),
3.77 (s, OCH3), 6.17 (m, Ar H), 7.41 (m, Ar H), 8.02 (m,
Ar H), IR(KBr, cm−1) 1628, 1616; Anal. (found: C, 56.38;
N, 2.28; H, 4.42%; calcd: C, 57.04; N, 2.37; H, 4.58%).
Preparation and Isolation of the Complexes
Characterization
[RhA(CO)PPh3]. . . (1). The rhodium(I) complex was
prepared according to the literature (26). A deoxygenated
The compounds [RhA(CO)PPh3] (1), [RhA(CO)Ph2P–
solution of RhCl3 · 3H2O in DMF was treated with a CH2–polystyrene] (2), and [RhA(PPh3)(COOCH3)(µ-
twofold excess of anthranilic acid and the mixture was OCH3)]2 (3) were characterized on the basis of their analyt-
heatedunderrefluxfor2h. Thesolutionwascooledtoabout ical and vibrational spectral data (Table 1) and electronic,
20◦C and diluted with a double volume of water when a yel- 1H NMR, and XPS data (Table 2).
low precipitate separated out. The solution with its precip-
Due to the insolubility of the polymer catalyst in organic
itate was cooled to 5◦C for 1 h and then filtered off, washed and inorganic solvents, its characterization was limited to
with H2O and Me2CO, and dried in a vacuum. The yellow chemical analysis, IR spectra, and XPS and DTA studies.
complex formed was [RhA(CO)2] and was highly soluble The IR peaks at 1580 cm−1 (νcoo, as) and 1360 cm−1 (νcoo,
in DMF. A solution of [RhA(CO)2] (1.02 mmol, 0.3 g) in s) and in the region 3000–3200 cm −1 (νNH2) are present in
DMF (10 ml) was treated with excess triphenyl phosphine the spectra of all the complexes, indicating the presence of
(1.5 mmol, 0.4 g) and the mixture was heated under reflux coordinated anthranilate in all of them. Weak asymmetric
for 2 h. The solution changed from yellow to brown within NH2 bending in the region 1660–1610 and symmetric bend-
15 min. The solution was cooled to 20◦C and diluted with ing at 1510–1490 cm−1 are also observed for the complexes.
a double volume of water (20 ml). During dilution of the The peaks of 1616 (w) and 1628 in complex 3 in the νco
reaction mixture a light brown precipitate formed. The so- region are characteristic of a methoxy carbonyl ligand (29).
lution with the precipitate was further cooled to 5◦C for 1 h The peak at 1088 cm−1 in complex 3 is common to both
–
and filtered and the residue washed with H2O and MeOH δPPh3 and ν(O CH3) of the bridging ligand.
and dried in a vacuum. The brown amorphous compound
was very much soluble in DMF and C6H6.
The rhodium content in each complex was determined
by refluxing them with concentrated HCl for 24 h and then

SOLUBLE AND POLYMER-ANCHORED RHODIUM CATALYST
257
TABLE 1
Analytical and Vibrational Data of the Compoundsa
Compounds
[RhA(CO)PPh3] (1)
νCO
νNH2
νOCO
νC(O)OCH3
δPPH3
C (%)
59.40
H (%)
4.40
N (%)
2.21
Rh (%)
20.7
1960
3175 (as)
3075 (s)
3175 (as)
3070 (s)
3172 (as)
3072 (s)
1575 (as)
1350 (s)
1575 (as)
1340 (s)
1585 (as)
1362 (s)
1085
[RhA(CO)Ph2P–CH2–(P)] (2)
1955
1085
1088
1.38
17.2
[RhA(PPh3)(COOCH3)(µ-OCH3)]2 (3)
1628
1616
56.38
4.42
2.28
a All ν values are given in cm−1. (P) = Polystyrene.
estimatingthemetalconcentrationinthesolutionbyatomic and 2 have binding energies of ∼308 and ∼312 eV, respec-
absorption spectrometry at a wavelength of 255.6 nm using tively. The energies are comparable with the +1 oxidation
anair–acetyleneflame. Analysisofcomplex2afterfivecata- state of rhodium (32). The binding energy for the isolated
lytic runs indicated the Rh content (1.19%) was reduced complex 3 was found to be 310.2 (3d5/2) and 314.2 eV
by 13% compared to that of the original catalyst (1.38%). (3d3/2), whichsuggeststhepresenceofRh(III)inthespecies
This modest loss does not appear to be metal elution, as (Fig. 2). To avoid possible X-ray-induced metal reduction in
the complex showed similar catalytic activity and the beads the polymer host, measurements were made with the power
showed no visible color change. The metal loss may be a of the X-ray source reduced to 10 kV, 10 mA. Survey spectra
consequence of changes in the bead structures, as is evident were recorded for all samples in the binding energy range
from a slight shift in the νCO absorption peaks (Fig. 1).
The electronic spectra of complex 1 in DMF medium
show three absorption bands, at 280 (sh), 363, and 380 nm.
The former was assigned to an intramolecular charge trans-
fer of a coordinated anthranilate anion (26), while that of
0–1400 eV at steps of 0.5 eV.
RESULTS AND DISCUSSION
The catalyst [RhA(CO)2] (HA; 2-amino benzoic acid)
363 nm may be due to π–π transition of the ligand. The was found to be highly efficient for the reductive N-
weak band at 380 nm has been attributed to d–d transition carbonylation of nitroaromatics in mild coordinating sol-
of the rhodium(I) complex. For complex 3, in addition to vents at high temperature (140◦C) and high Pco (80 atm)
the bands at 280 and 363 nm, the third band, at 398 nm in the presence of cocatalyst and cosolvent (33). With the
(w, sh), is due to the d–d transition of the rhodium(III) oc- idea that polymer-based catalysts can withstand more dras-
1
tahedral complex. The H NMR spectra (ppm, C6D6) of tic conditions and can influence product distribution due
complexes 1 and 3 exhibit multiple signals in the region to their steric factors, two groups of catalysts (1 and 2)
6.7–8.0 (phenyl protons) and a broad signal at 7.38 (NH2) were prepared and their catalytic properties examined.
(Table2). NewPMRsignalsat3.03and3.77ppmofcomplex The soluble catalyst [RhA(CO)PPh3] and its polymer ana-
3 are attributed to CH3 protons of C(O)OCH3 groups and loguewereusedforthereductivecarbonylationofnitroaro-
to the O–CH3 protons of the bridging ligands, respectively matics in DMF medium under a wide range of carbon
1
(30, 31). H NMR signals in the methoxide and aromatic monoxide pressures and reaction temperatures. In the ab-
methyl regions of complex 3 are masked at higher concen- sence of methanol, low conversion of nitrobenzene (20%)
trations of aromatic amine and CH3OH.
to aniline and diphenylurea was achieved with the sol-
XPS study of the compounds were carried out in the uble rhodium catalyst but the polymer-supported com-
range 290–320 eV using Al Kα (1486.3 eV) as target mate- plex showed no conversion of the nitroaromatics. Reac-
rial in order to determine the oxidation state of rhodium in tion at room temperature and low Pco (20 atm), in the
them. The 3d5/2 and 3d3/2 levels of the metal in complexes 1 presence of methanol, did not give satisfactory results.
TABLE 2
XPS and PMR Data of the Complexesa
ESCA Data (eV)
PMR spectral data (ppm, C6D6)
No.
Compounds
[RhA(CO)PPh3]
[RhA(CO)Ph2P–CH2–(P)]
[Rh(A)PPh3(COOCH3)(µ-OCH3)]2
3d5/2
3d3/2
Aromatic–H
–NH2
–COOCH3
1
2
3
308
308.6
310.2
312
312.8
314.2
6.7–7.8 (m)
7.38 (s)
6.9–8.02 (m)
7.36 (s)
3.03 (s)/3.77 (s)
a (P) = Polystyrene.

258
MUKHERJEE AND SAHA
TABLE 3
Influence of PhNO2 : PhNH2 Molar Ratio on the Reductive
a
Carbonylation of PhNO2 in the Presence of Methanol
Selectivity (%)
Molar ratio
Run PhNO2 : PhNH2
Conversionb
(%)
DPUc MPCd PhNH2 NPFe
1
2
3
4
5
6
1 : 0
0 : 1
1 : 0.5
1 : 1
1 : 2
1 : 4
100
40
100
100
100
100
59 (53) 16 (20) 9 (13) 10 (10)
74 (66) 4 (16)
—
17 (15)
69 (60) 9 (15) 7 (10) 12 (10)
70 (62) 9 (14) 8 (10) 13 (10)
71 (62) 13 (15) — (08) 14 (12)
70 (60) 13 (16) — (08) 14 (12)
Note. Values in parentheses are those for the soluble rhodium catalyst.
a Conditions: PhNO2 + PhNH2 = 0.5 mol, CH3OH = 2 ml, catalyst =
Rh, content = 2 × 10−4 mol, medium = DMF, T = 80◦C, t = 4 h, Pco
=
60 atm.
b Calculated to the conversion of PhNO2 (runs 1, 3–6) and PhNH2
(run 2).
ꢀ
c
N,N -Diphenylurea.
d Methyl-N-phenyl carbamate.
e
N-Phenylformamide.
FIG. 1. IR spectra showing νCO peaks of (a) complex 1, (b) com-
plex 2, (c) used catalyst 2 after 3 runs, (d) used catalyst 2 after 5 runs, and
(e) complex 3.
At temperatures lower than 20◦C, the rate of conversion
was very low even after 24 h of a catalytic run (Fig. 3). A
high reaction temperature (T > 140◦C), however, produced
some unidentified brown resinous product (∼5%) but no
phenyl isocyanate was detected at the end. PhNCO was
expected to be formed by the thermal decomposition of
the carbamate at such high temperatures. Complexes 1 and
2 were stable and showed no signs of decomposition even
under such severe conditions (Fig. 4). TGA curve for the
soluble and the polymer-supported complex show 22 and
Effect of temperature and Pco on the carbonylation of ni-
trobenzene in the presence of an optimum concentration
of methanol was therefore studied. The highest conversion
of PhNO2(100%) to diphenylurea (DPU) was achieved at
Pco = 60 atm and temperature = 80◦C and at a methanol
concentration of 6 M. A higher methanol concentration in-
creased the yield of aniline at the cost of DPU. Comparable
results were obtained with both the soluble and polymer-
anchored rhodium complex (Table 3).
FIG. 2. XPS spectra of 3d5/2 and 3d3/2 levels of Rh in the cata-
lysts: (a) complex 1, (b) complex 2, (c) used catalyst 2 after 5 runs, and
(d) complex 3.
FIG. 3. Effect of temperature on reductive carbonylation of PhNO2
using 1 as catalyst. Pco = 60 atm, solvent = DMF, time = 4 h, ꢀ = PhNH2,
= DPU, and = PhNO2.

SOLUBLE AND POLYMER-ANCHORED RHODIUM CATALYST
259
Both catalysts 1 and 2 exhibited similar behavior, though
with slightly different rates. It is therefore likely that the
soluble and polymer-anchored rhodium complexes pass
through the common intermediate stages of the carbony-
lation process. With regard to the formation of aniline as
the by-product during reductive carbonylation of nitroben-
zene in the presence of methanol to carbamate (run 1,
Table 3), the influence of PhNO2 to PhNH2 molar ratio in
the reductive carbonylation process was also the subject of
study. It is clear from the results that though the molar ratio
(PhNO2 : PhNH2) decreased, the percentage conversion to
DPU and MPC remained almost unaltered. When no ani-
line was added to the system, conversion of PhNO2 to DPU
and PhNH2 (∼15%) suggested the role of methanol as hy-
drogen provider (the system was degassed to free it from air
and moisture). With PhNH2 as substrate, the product dis-
tribution changed substantially, with the highest recorded
FIG. 4. TGA curve for the soluble (C) and the polymer-supported yield of DPU (70%). The presence of moisture in the sys-
(P) catalysts.
tem greatly hinders the catalytic conversion of PhNO2 to
DPU and almost 90% aniline was recorded. High aniline
formation may be due to hydrolysis of any DPU formed
(reaction 2)
9% weight loss, respectively, at 200◦C. The corresponding
weight loss at 300◦C is 30 and 15%.
At a carbon monoxide pressure of 20 atm or below, no
2(PhNHCONHPh) + 2H2O → 2PhNH2 + 2CO2 . . . . [2]
carbonylation reaction takes place, as nitrobenzene con-
PhNCO + H2O → PhNH2 + CO2 . . . .
[3]
centration remained unchanged and no reaction products
were observed even after 16 h of a catalytic run (Fig. 5). As
the pressure of carbon monoxide was slowly raised from as
low as 20 to 100 atm, the percentage of diphenylurea in-
creased gradually and reached the optimum value of 70%
at Pco = 60 atm.
A high methanol content in the system greatly reduced the
extent of DPU formation while simultaneously increasing
the percentage of MPC and PhNH2. Reactions 4 and 5 seem
to be the most probable reason for low conversion to DPU:
PhNHCOPhNH + CH3OH
→PhNHCOOCH3 + PhNH2 . . . .
PhNCO + CH3OH → PhNHCOOCH3 . . . .
[4]
[5]
Formation of NPF in the system (detected in all six runs;
Table 3) is believed to be due to direct interaction of CO
and PhNH2:
PhNH2 + CO → PhNHCHO . . . .
[6]
An increase in aniline concentration however resulted in
no significant change in NPF formation (runs 3–6).
Several runs were conducted in a completely dry alcohol-
free medium with and without aniline to study the nature
of product distribution and to monitor how much PhNCO,
if any, was formed at the intermediate stage. PhNCO could
not however be detected at any intermediate stage of these
catalytic runs. Monitoring the IR and 1H NMR spectra of
the solutions during a catalytic run did not show the forma-
tion of intermediate metal hydride species. Carbonylation
of nitrobenzene was conducted in DMF containing either
1-hexene or phenylacetylene under optimum reaction con-
ditions. The alkene and the alkyne remained unchanged
during the entire carbonylation process. Formation of metal
FIG. 5. Effect of Pco on reductive carbonylation of PhNO2.
Temperature = 80◦C, solvent = DMF, time = 4 h, = PhNH2, = DPU,
and ꢀ = PhNO2.

260
MUKHERJEE AND SAHA
hydride species has therefore been ruled out, as its forma- mation of intermediate compound 3 requires CO inser-
tion should reduce alkene or alkyne to the corresponding tion in one of the stages, which explains the requirement of
saturated products.
moderate carbon monoxide pressure for the carbonylation
Several authors resorted to nitrene mechanism in ra- process. Low PCO (<20 atm) prevents the formation and
tionalizing the products, e.g., isocyanates, carbamates, and isolation of the intermediate and therefore no carbonyla-
urea, formed during reductive carbonylation of nitroben- tion products were observed under such conditions even
zene (33, 34). Attempts were therefore made to trap at high methanol concentration. The monomer species at
any phenylnitrene formed at the intermediate stage using initial stages of the reaction is readily transformed during
diphenylacetylene in the reaction medium. The phenylni- isolation into the dimer and this isolated complex 3 was sub-
trene derivatives could not be detected under the optimum jected to carbonylation reaction under reaction conditions
reaction conditions using either the soluble or the polymer- (Table 3) identical to that maintained for the starting cata-
supported complex. PhN coordinated to Rh(I)/Rh(III) may lyst, 1. GC analysis showed >50% DPU formation within
not allow its reaction with diphenylacetylene and the un- 1 h of the catalytic run. Several runs were conducted in
stable species in solution may quickly be transformed into DMF using PhNO as substrate and the similar product dis-
thedimer-bridgedcomplex(Scheme1). Spectroscopicstud- tribution suggests that PhNO2 was deoxygenated to PhNO
ies however showed that the reaction proceeds through at some earlier stage of the catalytic cycle (Scheme 1).
species [RhA(PPh3)(COOCH3)(µ-OCH3)]2. . . (3) and the
1H NMR spectra of a CH3OH/C6D6 solution of 3 and p-
isocyanate formed at an intermediate stage is immediately toluidineweremonitoredforaperiodof24h. After12h, the
scavenged by excess amine to N,Nꢀ-diphenylurea (30). For- spectrum collected revealed resonances from 3, CH3OH,
SCHEME 1

SOLUBLE AND POLYMER-ANCHORED RHODIUM CATALYST
261
TABLE 4
p-toluidine, and three new signals. There was a singlet at
10.72 ppm, similar in chemical shift to the hydrogen-bonded
amide proton (35), a singlet at 2.55 ppm, characteristic of
a methoxy carbonyl proton (30), and a singlet at 2.04 ppm,
a shift similar to methyl resonance from the p-tolyl moiety.
The integration of these three signals was 1 : 3 : 3, respec-
tively. 1H NMR signals are masked at higher concentration
of aromatic amine and methanol. The result describes the
formation of the methoxy carbonyl–carbamoyl species 4
in the reaction medium although all attempts to isolate the
solid sample were unsuccessful. During the experiments de-
scribedabove, theconcentrationof4increasedtoaconstant
ratio with 3, suggesting that an equilibrium was achieved.
The reaction quotient [4]/[3] decreased in a linear fashion
as CH3OH concentration was increased from 1.0 to 8.0 M
at a constant concentration of p-toluidine. Formation of the
intermediate species 4 by the reaction of p-toluidine with
3 and simultaneous liberation of one equivalent of methyl
alcohol, as predicted in the catalytic cycle, is more probable.
An effort was also made to identify the reaction path-
way by treating a solution of p-toluidine and methanol in a
septum-sealed vial with p-tolylisocyanate. Visible amounts
of N,Nꢀ-ditolylurea were noticed within 5 min at ambient
temperature. After 30 min the solution in the vial was ana-
lyzed by gas chromatography. The belief that the isocyanate
formed at an intermediate stage is immediately taken up
by the amine resulting in the formation of urea is therefore
substantial and no trace of carbamate could be detected at
any stage of the reaction. That reaction (4) led to the forma-
tion of carbamate seemed unlikely due to CH3OH having
a basicity lower than PhNH2.
Rate Constant for the Reaction of 3 with p-Toluidine
Temp (◦C)
[ArNH2] (M)
[Rh] (mM)
k (M−1 min−1
)
80
80
80
80
80
80
0.20
0.44
0.56
0.15
0.15
0.15
30.2
30.2
30.2
30.2
14.4
44.2
0.24
0.21
0.23
0.23
0.20
0.21
Theratewasfollowedbythestudyofthe3.03and3.77sig-
nals characteristic of CH3 protons of the C(O)OCH3 group
of 3 and the peak areas were evaluated with the integra-
tion for all species normalized to the initial concentration
of 3. The reaction was complete within 90 min at 80◦C, and
upon standing for an additional 24 h, crystals of N,Nꢀdi-p-
tolylurea were visible in the reaction tube and 1 was the
major metal species present in the reaction solution. It was
imperative that the kinetic reaction be strictly performed
under dry conditions, as the smallest measureable concen-
tration of water caused an erroneous result due to the hy-
drolysis of complex 3 with liberation of the starting catalyst
1, CO2, and two equivalents of CH3OH:
[3] + H2O + CO → [1] + CO2 + 2CH3OH . . . . [8]
The kinetics outlined in the paper were conducted in dif-
ferent temperature regimes. Although the final products
were identical, the kinetic work in the low-temperature
region was complicated by the appearance of peaks
Kinetic Analysis of the Reaction of
[RhA(PPh3) (COOCH3)(µ-OCH3)]2
with p-Toluidine at 80◦C
These reactions were carried out in a high-pressure cell
designed for spectroscopic studies under high gas pressure
(36) in order to simulate the conditions of the actual cata-
lytic process. Various volumes of the isolated complex 3 and
p-toluidineintoluene-d8 andCD3ODweretakenina5-mm
NMR tube to give a total volume of 0.5 mL. The initial
concentrations of 3 and p-toluidine in these solutions are
listed in Table 4. The valve of the high-pressure tube was
connected to the pressure line and the cell was pressurized
with carbon monoxide (60 atm) The solution in the cell was
shaken to insure that a reasonable amount of CO dissolved
in solution. The pressure was released and the tube was
placed in an NMR spectrometer with the probe warmed
to the desired temperature. The organic product observed
from the reaction of 3 and p-toluidine under all conditions
was N,Nꢀ-di-p-tolylurea:
[3] + 2ArNH2 + CO → 2CH3OH
+ ArNHCONHAr + [1] . . . . [7]
FIG. 6. A first-order rate dependence is shown w.r.t the concentration
of p-toluidine from the reaction of 3 with p-toluidine at 80◦C.

262
MUKHERJEE AND SAHA
support by the University Grants Commission (Eastern region), India, is
gratefully acknowledged.
REFERENCES
1. Cenini, S., Pizzoti, M., and Crotti, C., in “Aspects of Homogeneous
Catalysis” (R. Ugo, Ed.), Vol. 6. Reidel, Dordrecht, 1998.
2. Cenini, S., Pizzoti, M., Crotti, C., and Porta, F., J. Org. Chem. 53, 1243
(1988).
3. Cenini, S., Pizzoti, M., Crotti, C., Raigaini, F., and Porta, F., J. Mol.
Catal. 49, 59 (1988).
4. Raigaini, F., and Cenini, S., J. Mol. Catal. 109, 1 (1996).
5. Laconte, P., Metz, F., Morterux, A., Osborn, J. A., Paul, F., Petit, F.,
and Pillot, A., J. Chem. Soc. Chem. Commun. 1616 (1990).
6. Kunin, A. J., Noirot, M. D., and Gladfelter, W. L., J. Am. Chem. Soc.
111, 2739 (1989).
7. Falluoka, S., Chono, M., and Khonos, M., J. Chem. Soc. Chem. Com-
mun. 399 (1984).
8. Valli, V. L. K., and Alper, H., Organometallics 14, 80 (1995).
9. Li, K. T., and Peng, Y. J., J. Catal. 143, 631 (1993).
10. Maddinelli, F., Nali, M., Rindone, B., Tollari, S., Cenini, S., Monica,
G. La., and Porta, F., J. Mol. Catal. 39, 71 (1987).
11. Grate, J. H., Hamm, D. R., and Valentine, D. H., U.S. Patent 4,629,804
(1986).
FIG. 7. Initial rate of urea formation from the reaction of 3 with
12. Grate, J. H., Hamm, D. R., and Valentine, D. H., U.S. Patent 4,705,883
(1987).
p-toluidine at 80◦C.
13. Drent, E., and Vanleeuwen, P., Eur. Patent 0,086,281 (1993).
14. Rosenthol, R., and Zajcek, J. G., U.S. Patent 3,919,280 (1983); Chem.
Abstr. 84,60177g.
15. Rosenthol, R., and Zajcek, J. G., U.S. Patent 3,963,202 (1976); Chem.
Abstr. 85,177053a.
corresponding to intermediate 4. The effects of the vary-
ing concentrations of 3 and p-toluidine on the initial rate of
urea formation are shown in Figs. 6 and 7, respectively. The
observed rate constants shown in Table 4 were determined
by the rate law
16. Maho, V., and Hudec, J., Czech Patent 162,837 (1976); Chem. Abstr.
85,142876u.
17. Lee, S. M., Cho, N. S., Kim, K. D., Oh, J. S., Lee, Ch. W., and Lee, J. S.,
J. Mol. Catal. 73, 43 (1992).
18. Lee, Ch. W., Cho, N. S., Kim, K. D., Oh, J. S., Lee, S. M., and Lee, J. S.,
J. Mol. Catal. 81, 17 (1993).
19. Macho, V., Kralic, M., and Halmo, F., J. Mol. Catal. A 109, 119 (1996).
20. Gargulak, J. D., Noirot, M. D., and Gladfelter, W. L., J. Am. Chem.
Soc. 113, 1054 (1991).
21. Skoog, S. J., and Gladfelter, W. L., J. Am. Chem. Soc. 113, 11049
(1994).
22. Skoog, S. J., Campell, J. P., and Gladfelter, W. L., Organometallics 13,
4137 (1994).
23. Han, S. H., Geffroy, G. L., and Rheingold, A. L., Inorg. Chem. 26, 3426
(1987).
24. Mukherjee, D. K., Palit, B. K., and Saha, C. R., J. Mol. Catal. 91, 19
(1994).
25. Mukherjee, D. K., Indian J. Chem. 40A, 1176 (2001).
26. Kwaskowska, C. E., and Zioikowski, J. J., Transition Met. Chem. 8, 103
(1983).
27. Pepper, K. W., Paisley, H. M., and Young, M. A., J. Chem. Soc. 4097
(1953).
28. Grubbs, R. H., and Kroll, L. C., J. Am. Chem. Soc. 93, 3062 (1971).
29. Frod, P. C., and Fokicki, A., Adv. Organomet. Chem. 28, 139
(1988).
30. Gargulak, G. D., and Gladfelter, W. L ., J. Am. Chem. Soc. 116, 3792
(1994).
31. Yoshida, T., Okano, T., Ueda, Y., and Otsuka, S., J. Am. Chem. Soc.
103, 3411 (1981).
Rate = k[ArNH2][Rh(A)(PPh3)(COOCH3)(µ-OCH3)]2.
Based on the above observations and identification of the
reactive intermediates, the mechanism shown in Scheme 1
has been proposed for the carbonylation process using both
the soluble and the polymer-supported rhodium catalyst.
The in situ spectroscopic studies of the high-pressure cata-
lysis established that the principal species observed in so-
lution was [Rh(A)(PPh3)(COOCH3)(µ-OCH3)]2 and that
several intermediate steps were involved in the conversion
of nitroarene to useful diarylurea in the presence of alcohol.
CONCLUSION
The reaction conducted under catalytic conditions pro-
duced either diarylurea or the carbamate depending upon
the conditions. Even in those runs that produced carba-
mates, urea was still proposed to be the intermediate. In-
tramolecular elimination of aryl isocyanate is the mecha-
nism which releases the organic product from the catalyst,
and in all of the stoichiometric reactions, the isocyanate is
immediately trapped by ArNH2 to form the observed prod-
uct, diarylurea.
32. Holy, N. L., J. Org. Chem. 44, 239 (1979).
33. Islam, S. M., Palit, B. K., Bose, A., and Saha, C. R., J. Mol. Catal. A
142, 169 (1999).
ACKNOWLEDGMENTS
34. Iqbal, A. F. M., CHEMTECH 566 (1974).
35. Gargulak, G. D., and Gladfelter, W. L., Inorg. Chem. 33, 253 (1994).
36. Koe, D. C., J. Magn. Reson. 63, 388 (1985).
One of the authors, D.K.M , thanks the department of Chemistry, Indian
Institute of Technology, for providing the instrumental support. Financial