One pot synthesis of ureas and carbamates via oxidative carbonylation of aniline-type substrates by CO/O2 mixture catalyzed by Pd-complexes

Article abstract of DOI:10.1016/j.molcata.2015.06.027

Abstract Carbonylation of aromatic amines by direct insertion of carbon monoxide is catalyzed by PdCl₂(X_nPy)₂ complexes (where Py = pyridine, X = -CH_3, -Cl; n = 0-2) and gives, depending on the conditions, ethyl N-phenylcarbamates or N,N′-diphenylureas. For carbonylation of aniline, a proper choice of X_nPy ligands in PdCl₂(X_nPy)₂ catalyst and application of molecular oxygen instead of nitrobenzene (conventionally used oxidant for carbonylations) allow to carry out the process under mild conditions with high yield and selectivity. The best results (75% yield of the main product with selectivity of catalyst above 90%) were obtained for the process catalyzed by PdCl₂(2,4-Cl₂Py)₂ complex at 100°C and they were greatly improved in comparison to 41% yield and 68% selectivity obtained for CO/nitrobenzene used at 180°C.

Full text of DOI:10.1016/j.molcata.2015.06.027

Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
One pot synthesis of ureas and carbamates via oxidative
carbonylation of aniline-type substrates by CO/O2 mixture catalyzed
by Pd-complexes
Agnieszka Krogul∗, Grzegorz Litwinienko
Faculty of Chemistry, University of Warsaw, 02-093 Warsaw, Pasteura 1, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 April 2015
Received in revised form 14 June 2015
Accepted 23 June 2015
Available online 30 June 2015
Carbonylation of aromatic amines by direct insertion of carbon monoxide is catalyzed by PdCl2(XnPy)2
complexes (where Py = pyridine, X = −CH3, −Cl; n = 0–2) and gives, depending on the conditions, ethyl
N-phenylcarbamates or N,Nꢀ-diphenylureas. For carbonylation of aniline, a proper choice of XnPy ligands
in PdCl2(XnPy)2 catalyst and application of molecular oxygen instead of nitrobenzene (conventionally
used oxidant for carbonylations) allow to carry out the process under mild conditions with high yield
and selectivity. The best results (75% yield of the main product with selectivity of catalyst above 90%)
were obtained for the process catalyzed by PdCl2(2,4-Cl2Py)2 complex at 100 ◦C and they were greatly
improved in comparison to 41% yield and 68% selectivity obtained for CO/nitrobenzene used at 180 ◦C.
© 2015 Elsevier B.V. All rights reserved.
Keywords:
Palladium complexes
Carbonylation
Ureas and carbamates
Aromatic amines
Aniline
Mechanism
Kinetics
NHCOOR
+
NH2
1. Introduction
NH
HN
T
+ ROH
Carbonylation of nitrocompounds [1–5] and amines [6–10] is
particularly important in polymer industry – during production of
carbamates, isocyanates and ureas. Carbonylation of amines is effi-
ciently catalyzed by square planar palladium(II) complexes with
nitrogen donor ligands [11–21] and two directions of this process
are possible – the process carried out in alcohols leads to carba-
mates (Eq. (1)) whereas in aprotic solvents the ureas are formed
as the main product (Eq. (2)). Ureas can be also formed in alcohols,
but at temperatures below 150 ◦C, otherwise a thermal transforma-
tion of ureas to carbamates occurs with participation of an alcohol
molecule (ROH, Eq. (3)) [22].
O
(3)
In contrast to the reductive carbonylation of aromatic nitro-
compounds (ArNO2 → carbamates), the oxidative carbonylation of
anilines (ArNH2 → carbamates, ureas) requires the presence of an
oxidant [9,22–24]. During this process Pd(II) is reduced to Pd(0) and
its re-oxidation to the active form of the catalyst i.e., Pd(0) → Pd(II),
is a crucial step [9,23], facilitated by an oxidizing agent (for exam-
ple, molecular oxygen [24,25], aromatic nitrocompounds [25–27],
iodine [28], quinones and their derivatives [23,29], or unsaturated
compounds such as 1,4-dichlorobut-2-ene [23]).
NHCOOR
NH2
In our laboratory was successfully applied the catalytic sys-
tem PdCl2/Fe/I2/pyridine for carbonylation of aromatic amines
in the presence of nitrobenzene (NB) or molecular oxygen as
oxidizing agent to obtain ethyl N-phenylcarbamate (EPC) or N,N-
diphenylurea (DPU) [25,30–32]. No substituent effects (in the
aniline ring) on the yield of EPC or DPU in the presence of NB as
an oxidizing agent was observed [32]. Surprisingly, significant sub-
stituent effects on the reaction yield were reported when molecular
oxygen was employed [25]. Therefore, these results suggest two
different mechanisms of aniline carbonylation depending on the
type of oxidant applied. Previous experiments with carbonylation
cat.
+
CO + ROH + 1/2O2
+ H2O
(1)
(2)
NH2
NH
HN
cat.
2
+ CO + 1/2O2
+ H2O
O
∗
Corresponding author. Fax: +48 22 8225996.
http://dx.doi.org/10.1016/j.molcata.2015.06.027
1381-1169/© 2015 Elsevier B.V. All rights reserved.

A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
205
of aniline in the presence of NB (as oxidizing agent) catalyzed by
2.3. Synthesis of PdCl2(PhNH2)2
PdCl2(XnPy)2 complexes (X = Cl or CH3, n = 0–2) allowed us to pro-
pose the mechanism of this reaction. We showed that electron
transfer from palladium to nitro group of NB is a rate determin-
ing step (RDS). Aniline does not participate in RDS and, therefore,
no substituent effect in the aromatic ring of aniline on the reaction
rate is observed. However, as reported in our earlier work [25] the
application of O2 instead of NB leads to a significant substituent
effect (in the aniline ring) on the rate of reaction, indicating that
aniline molecules are directly involved in RDS during carbonylation
with O2 applied as an oxidant (this mechanism will be discussed in
Section 3).
PdCl2Py2 complex (0.56 mmol), aniline (10.8 mmol), pyridine
(1.24 mmol) and ethanol (4 ml) were heated at 180 ◦C for 60 min. On
completion of the reaction golden-yellow flakes of PdCl2(PhNH2)2
complex were obtained. The precipitate was crystallized from
acetone. Elemental analysis of PdCl2(PhNH2)2% (exp./calc.): C
(39.98/39.66),
M.p. = 240–245 ◦C. IR (KBr): 3287, 3202, 3117, 1594, 1573, 1111,
756 cm−1
H (3.97/3.86), N (7.74/7.71), Cl (19.04/19.56).
.
2.4. The reaction of PdCl2(PhNH2)2 complex with carbon
monoxide
Recent experiments with carbonylation of nitrobenzene cat-
alyzed by PdCl2(XnPy)2 complexes (X = Cl or CH3, n = 0–2) showed
that the activity of the palladium catalyst (expressed as turnover
frequency, TOF) depended on the structure and basicity of XnPy
ligand—the electron density on N atom and steric hindrance around
Pd play a crucial role in the process. Mechanism of NB carbony-
lation proposed by Krogul and Skupinska [19,31,32] includes the
formation of aniline (AN) as the first step with the electron trans-
fer from palladium to nitro group of NB as a rate determining step.
In general, carbonylation of NB can be considered as a carbonyla-
tion of AN with nitrobenzene being an oxidant [19] but mechanism
of the carbonylation of AN catalyzed by PdCl2(XnPy)2 complexes
in the presence of oxidants other than NB still remains unsolved.
Therefore, this work is focused on the mechanistic aspects of
carbonylation of AN catalyzed by PdCl2(XnPy)2 complexes in the
presence of O2 (instead of NB). We applied PdCl2(XnPy)2 complexes
with the following ligands: pyridine (I), 2-methylpyridine (II);
3-methylpyridine (III); 4-methylpyridine (IV); 2,6-dimethyl-
pyridine (V); 2,4-dimethylpyridine (VI); 3,5-dimethylpyridine
(VII); 2-chloropyridine (VIII); 3-chloropyridine (IX); 2,6-dichloro-
pyridine (X); 2,4-dichloropyridine (XI); 3,5-dichloropyridine
(XII).
PdCl2(PhNH2)2 complex (0.7 mmol), in chlorobenzene (10 ml)
was kept at 150 ◦C under CO pressure of 4 MPa for 60 min and
white needles and palladium black precipitated. White needles
were analyzed by elemental analysis and IR. N,Nꢀ-diphenylurea
was identified on the basis of elemental analysis% (exp./calc.): C
(73.60/73.57), H (5.75/5.70), N (13,22/13.20); IR (KBr): 3328, 3198,
1649 cm−1. M.p. = 235–237 ◦C [25].
2.5. Carbonylation procedure and analysis of the products
The procedure was described elsewhere [19]. Briefly, the reac-
tion was carried out in 200-mL stainless-steel autoclave equipped
with a magnetic stirrer. Before experiment, the autoclave was
heated at 120 ◦C for 3 h and cooled down to room temperature. Sub-
sequently, PdCl2(XnPy)2 or PdCl2 catalyst (0.056 or 0.168 mmol)
and Fe powder (0–0.7 mmol) were placed in the autoclave, the
air was evacuated and the system was filled with purified argon.
Then, under an argon stream, other reagents and solvents were
added: I2 (0–0.39 mmol), Py or XnPy (6.2 mmol), aniline (54 mmol);
ethanol or toluene (another solvent, 20 ml). The cover was closed
and autoclave was directly filled with molecular oxygen (0.6 MPa)
and carbon monoxide (3.6 MPa), fixed, placed in a hot oil bath
and kept at 100–180 ◦C, for 60–360 min, depending on the reac-
tion. After the reaction was completed, the autoclave was cooled
in a water bath then vented. The solid phase separated by cen-
trifugation (20,000 rpm for 15 min) as a black precipitate was
washed several times and analyzed by elemental analysis, IR,
UV–vis and 1H NMR. Polyaniline was identified on the basis of
elemental analysis% (exp./calc.): C (79.82/78.26), H (5.48/6.52), N
2. Experimental
2.1. Materials
All operations were carried out using standard Schlenk tech-
niques under oxygen and water-free argon. Carbon monoxide
(99.9%), molecular oxygen (99.999%), PdCl2, iodine and iron pow-
der were used as received. Pyridine (Py), substituted pyridines
(2-MePy; 3-MePy; 4-MePy; 2,6-Me2Py; 2,4-Me2Py; 3,5-Me2Py;
2-ClPy; 3-ClPy; 2,6-Cl2Py; 2,4-Cl2Py; 3,5-Cl2Py), aniline, ethanol,
toluene were distilled (or fractionally distilled) over CaH2 and
stored under argon.
(15,06/15,22), IR: 3419 cm−1, 1448 cm−1, 1398 cm−1, 1261 cm−1
,
1091 cm−1—the bands characteristic for the polymer chain [34,35],
UV–vis: band (ꢀmax = 286 nm) corresponding to ꢁ → ꢁ* transi-
tion (there was no n → ꢁ * band observed and it means that
totally reduced form of polyaniline was obtained [34]), and 1H
NMR: characteristic for polyaniline multiplet in the range of
6.5–7.5 ppm, being consistent with the literature data [36,37]. Anal-
ysis of the liquid phase was performed by the use of gas and
liquid chromatography (GC–FID, GC–MS and HPLC). The compo-
sition of the gas phase (obtained by the carbonylation of AN in
the presence of CO/O2) was analyzed by the Warder’s method.
The gaseous sample is passed through the flask containing 0.1 M
NaOH under conditions ensuring absorption of CO2. The determi-
nation of sodium carbonate contents in 0.1 M NaOH solution was
performed by titration with 0.1 M HCl using phenolphthalein and
methyl orange as indicators. The samples of titrated solution were
cooled (in a mixture of ice and NaCl) in order to avoid the loss of
CO2.
2.2. Synthesis of PdCl2(XnPy)2 (compounds I–XII)
Palladium chloride complexes with pyridines were prepared
under argon. A known amount (1.128 mmol) of PdCl2 was placed
in 10 ml flask equipped with magnetic stirrer and 2.26 mmol of Py
or substituted XnPy in 10 ml acetonitrile were added. Reaction was
carried out at room temperature for 24 h. After the reaction was
completed, yellow precipitate was crystallized from acetone.
Elemental analysis of PdCl2(Py)2 (I); PdCl2(2-MePy)2 (II);
PdCl2(2,6-Me2Py)2 (V); PdCl2(3-ClPy)2 (IX); PdCl2(2,4-Cl2Py)2 (XI);
PdCl2(3,5-Cl2Py)2 (XII) complexes was carried out by a stan-
dard method (see Table S1 in Supporting Material). Single yellow
crystals of PdCl2(3-MePy)2 (III), PdCl2(4-MePy)2 (IV), PdCl2(2,4-
Me2Py)2 (VI), PdCl2(3,5-Me2Py)2 (VII), PdCl2(2-ClPy)2 (VIII) and
PdCl2(2,6-Cl2Py)2 (X), obtained by slow evaporation of their
acetone solution, were characterized by an X-ray diffraction mea-
surements [20,33].
3. Results and discussion
As we demonstrated previously, the process of carbonylation of
NB contains a step of oxidative carbonylation of aniline (interme-

206
A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
Table 1
Parameters obtained for the carbonylationa of aniline by mixture of CO/O2, catalyzed by PdCl2(XnPy)2 complexes: conversion (CAN) and yield (YEPC), of aniline, selectivity
(SEPC) and turnover frequency (TOF) of PdCl2(XnPy)2 catalyst.a
b
d
e
No.
Complex
Symbol
pKa
CAN
YEPC
SEPC
TOFEPC
[%]
[%]
[%]
1.
2.
3.
4.
5.
6.
7.
9.
PdCl2(2,6-Me2Py)2
PdCl2(2,4-Me2Py)2
PdCl2(4-MePy)2
PdCl2(2-MePy)2
PdCl2(3-MePy)2
PdCl2Py2
PdCl2(3-ClPy)2
PdCl2(2-ClPy)2
PdCl2(2,6-Cl2Py)2
PdCl2(2,4-Cl2Py)2
PdCl2(3,5-Cl2Py)2
V
6.2c
36
40
34
44
37
50
55
56
51
60
57
26
28
22
30
24
27
31
39
36
41
31
72
70
65
68
65
54
56
70
71
68
54
125
135
106
145
116
130
150
188
174
196
150
VI
IV
II
III
I
IX
VIII
X
XI
XII
6.1c
5.43
5.38
5.17
4.09
2.6c
2.0c
10
11.
12.
0.97c
1.4c
1.22c
a
Reaction conditions: PdCl2(XnPy)2/Fe/I2/XnPy = 0.056/2.68/0.12/6.2 mmol; AN = 54 mmol; ethanol = 20 ml; 180 ◦C; 3.4 MPa CO; 0.6 MPa O2; 120 min. Py = pyridine
(X = −CH3,−Cl; n = 0–2); AN = aniline; EPC = ethyl N-phenylcarbamate.
b
pKa values (with error 0.05 unit) for ligands measured in methanol/water (50/50), see Ref. [19].
c
For these ligands pKa values are determined with error 0.10.
d
Selectivity of EPC expressed as mmol EPC × (mmol converted AN)−1 [%].
e
Turnover frequency expressed in [mmol EPC × (mmol Pd)−1 × h−1].
diate) in the presence of nitrobenzene as an oxidant (NB is reduced
to AN) [19,20,32]. PdCl2(XnPy)2 complexes are effective catalysts
for this process [19]. Since the application of molecular oxygen has
economical and environmental advantages over the application of
NB as the oxidizing agent, in this work we decided to investigate
the catalytic activity of PdCl2(XnPy)2 complexes in the presence of
molecular oxygen (other reaction conditions were the same as in
the presence of NB). When molecular oxygen is used as oxidant
(instead of NB) the higher yields of carbamate or urea and higher
selectivity of catalyst are achieved.
3.1. Substituent effect in the XnPy on the catalytic activity of
PdCl2(XnPy)2 complexes
Palladium chloride complexes with XnPy ligands (where:
Py = pyridine, X = CH3, Cl; n = 0–2) were used as catalysts for car-
bonylation of AN to form ethyl N-phenylcarbamate (EPC) in the
presence of molecular oxygen as oxidizing agent, Eq. (1). Reac-
tion conditions were the same as for carbonylation of AN in the
presence of NB as oxidant, see footnote a in Table 1 [19]. Col-
lected parameters demonstrate increasing conversion of AN and
increasing yield of EPC when passing from methylpyridines to
unsubtituted pyridine and chloropyridines. This tendency is in
agreement with decreasing basicities of XnPy ligands (i.e., the
lower pKa of protonated ligand the higher conversion is obtained).
Among all investigated complexes the compound XI (with 2,4-
dichloropyridine) was the most active catalyst (TOF = 196) giving
60% conversion of AN, 41% yield of ethyl N-phenylcarbamate (EPC),
with selectivity 68%. For complexes with electron donating XnPy
ligands (methylpyridines) the conversions and yields were almost
half of the values obtained for XI, see Table 1. In our previous stud-
ies for the system employing NB as oxidant we noticed increasing
conversions and yields with increasing basicity of XnPy ligand [19].
Surprisingly, when NB is replaced with O2 the opposite tendency
(decreasing conversion and yield) is observed. Another difference
is the effect of a steric hindrance: an ortho-group in XnPy ring
causes a decrease of catalytic activity of the Pd(II) complexes
when NB is an oxidant, whereas no effect of steric hindrance on
the catalyst activity is observed when O2 was applied instead of
NB.
Fig. 1. Hammett type plot for carbonylation of aniline by CO/O2 in the presence of
PdCl2(XnPy)2: for nonsterically hindered pyridines (marked as squares) and sterically
hindered pyridines (marked as open circles). The straight line was constructed for
nonsterically hindered pyridines only. For reaction conditions see footnotes a in
Table 1.
expressed as TOF, with the Hammett ꢂ constants of X in XnPy
ligands:
log(TOF) = ꢃꢂ + const
(4)
We determined the electronic effects of substituents in the XnPy
ring for four representative XPy’s. We exclude derivatives with
substituents in position 2- because of strong steric effects dom-
inating over the inductive effects and for these derivatives the
straight line Hammett correlation turns into a curve. However,
even for substituents in 2- position the same tendency (increasing
conversion and yield with increasing electron-withdrawing prop-
erties) is observed for the whole series of XnPy. For CO/O2/AN
reaction catalyzed by PdCl2(XnPy)2/Fe/I2 /XnPy system the param-
eters ꢃ = 0.2643 and R2 = 0.9138 were obtained (Fig. 1). A positive
ꢃ value indicates that carbonylation is more efficiently accelerated
by PdCl2(XnPy)2 with lower electron density on Pd center. For the
carbonylation of AN in the presence of NB the opposite result (neg-
ative ꢃ) was obtained—perhaps the change of the oxidizing agent
changes the rate determining step.
Direct application of CO/O2 mixture is simpler and potentially
more profitable than application of other oxidizing agents. Activ-
ity and selectivity of PdCl2(XnPy)2 catalysts depend on the type of
oxidant but the results obtained for carbonylation of AN by CO/O2
at 180 ◦C are unsatisfactory (see Table 1). The highest selectiv-
The Hammett type equations can be applied to study the impact
of electron donating/withdrawing groups [38] and for the series of
experiments carried out under the same conditions we observed
a reasonable correlation of the catalytic activity of PdCl2(XnPy)2,

A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
207
Fig. 2. Effect of: amount of iodine (A), amount of iron (B), reaction time (C) and temperature (D) on the conversion (CAN) and yield (YEPC) of carbonylation of aniline and
selectivity (SEPC) of the catalyst. Plots were constructed on the basis of the results collected in Table 2S (A and B), Table 3S (C) and Table 4S (D) in Supporting material. For the
reaction conditions see footnotes a in Table 2S,3S and 4S, respectively.
ity is 72% and, taken together with relatively high conversion of
AN, suggests that significant amounts of by-products are formed.
Therefore, we decided to identify all products in order to explain the
mechanism of the carbonylation of AN by CO/O2 and to achieve the
highest yield and selectivity. The post-reaction mixture obtained
after carbonylation of AN by CO/O2 contained both, liquid and
solid phases. Careful analysis of the liquid phase indicated ethyl
N-phenylcarbamate (as the main product) and two side prod-
ucts: N-ethylaniline (yield 10–15%) and 2-methylquinoline (yield
5–10%). In the solid phase we identified polyaniline (yield 10–20%).
2-methylquinoline is probably formed during condensation of ani-
line with croton aldehyde (Eqs. (5) and (6)) [9,39], in agreement
We investigated the effects of various parameters on the con-
version of AN and yield of EPC and we optimized the reaction
conditions in order to minimize the formation of side-products and
to increase the selectivity of the catalyst.
3.2. Kinetic studies (the results of the optimization of reaction
conditions)
We investigated the effects of three factors: components of the
catalytic system, reaction time and reaction temperature on the
conversion of AN, EPC yield and selectivity of the catalyst.
with our previous reports [31,32].
3.2.1. The effect of the components of the catalytic system on the
rate of aniline carbonylation
NH2
Results shown in Fig. 2A and B (see also Table 2S in the Support-
ing material) indicate that reaction carried out in the presence of
PdCl2 (0.056 mmol) without addition of Fe and I2 leads to relatively
high conversion of AN (61%), however, the low yield of EPC and
low selectivity of the catalyst are observed (entry 1, Table 2S). Next
series of reactions was performed with PdCl2 and different amounts
(0–0.39 mmol) of iodine were applied (Fig. 2A and entries 2–4, Table
2S). Initially, addition of iodine to the reaction mixture increases the
conversion of AN and the yield of EPC, however, further increase of
added I2 reduced the conversion, yield, and selectivity of the cata-
lyst. The highest conversion of AN was observed for 0.04 mmol of
iodine (entry 2, Table 2S). In agreement with some earlier reports
[41,42], iodine appears as a substantial for recovering the catalytic
system by oxidation of the iron powder Fe(0) → Fe2+ (the role of
Fe2+ species will be discussed further). However, large excess of
iodide ions is undesirable because insoluble anilinium iodide is
formed and the amount of free aniline is decreased [28,43]. Another
side-effect is a coordination of I− to palladium(II) and this process
can be regarded as catalyst poisoning.
+
CH3CH=CHCHO
N
(5)
(6)
2CH3CHO → CH3CH = CHCHO + H2O
Aniline can also react with ethanol to produce another side prod-
uct, N-ethylaniline (Eq. (7)) and, in the presence of I2, polyaniline
might be also formed (Eq. (8) [40]).
NH2
HN
+
CH3CH2OH
NH
+ H2O
(7)
(8)
NHI
n
NHI
n
The effect of amount of iron powder on the rate of carbonyla-
tion was also examined and found to be moderate, see Fig. 2B (and
entries 5–8 in Table 2S). Even in the absence of iron powder (i.e.,

208
A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
Table 2
Conversion of oxygen (CO2) during carbonylation of aniline by mixture of CO/O2 in the presence of PdCl2 and turnover frequency (TOF) depending on the composition of
catalytic system (amounts of Fe, I2, Py, AN) and type of solvent.a
b
c
c
No.
nFe
[mmol]
nI2
[mmol]
nPy
[mmol]
nAN
[mmol]
CO2
[%]
TOFAN
TOFEPC
1.
0
0
0
0
2.7
0
0
0
0
0.12
0.12
0
0
0
0
0
0
6.2
0
54
54
54
54
0
18
8
13
51
47
36
0
500
295
268
285
0
0
125
62
125
143f
0
2.d
3.
4.
5.e
6.g
a
Reaction conditions: PdCl2 = 0.056 mmol; AN = 54 mmol (5 ml) or without aniline (entry 1 and 6); ethanol = 20 ml; 180 ◦C; 3.6 MPa CO; 0.6 MPa O2; 120 min. Py = pyridine;
AN = aniline; EPC = ethyl N-phenylcarbamate; DPU = N,N’-diphenylurea.
b
Conversion of O2 was calculated on the basis of the amount of CO2 (determined using Warder’s method) presented in the gaseous phase after the reaction and values
shown in this Table are related only to undesirable converion O2 → CO2 (the amount of O2 which reacts with aniline is not included).
c
Turnover frequency expressed in [mmol AN × (mmol Pd)−1 × h−1] or [mmol EPC × (mmol Pd)−1 × h−1].
d
Time = 60 min.
e
Solvent = toluene (20 ml).
f
TOF calculated for N,Nꢀ-diphenylurea.
g
Catalyst = PdCl2(Py)2 (0.056 mmol).
iron is present as the component of the stainless-steel autoclave
walls) the yield is 26% and this can be reasoned as participation of
very low amounts (traces) of iron cations as oxidants for Pd(0) in
presence of carbon monoxide [44,45]:
3.2.3. Identification of gaseous products
The yield of ethyl N-phenylcarbamate hardly depends on the
molar ratio of components of the catalytic system (Fig. 2A and B
and Table 2S), on reaction time (Fig. 2C and Table 3S) and on tem-
perature (Fig. 2D and Table 4S). We supposed that carbonylation of
AN to EPC was stopped by the lack of aniline, CO or O2. However,
the GC–MS analysis of post-reaction liquid showed that AN is still
present and no products containing oxygen or carbon monoxide
(apart from EPC or DPU) were identified among liquids and solids.
Therefore, we analyzed the composition of the gas phase and we
detected CO2. In order to check whether CO2 was formed from
water (the Water–Gas Shift Reaction), the experiment with 2 ml of
water introduced to the system1 was carried out but it gave nega-
tive result (no CO2 detected). It means that CO2 comes from O2, and
indeed, the conversions of O2 to CO2 determined from the amounts
of CO2, range from 8 to 51%, see Table 2. A possible explanation is
that oxidation of CO is promoted by palladium salt – a similar pro-
cess was described for the oxidation of CO by O2 in the presence of
Pd(0) [28,41,44]. The participation of Pd(II) catalyst in the oxidation
of CO to CO2 may be the reason of the dramatic decrease of both:
TOFAN and TOFEPC with two-fold increase of the amount of palla-
dium chloride. We conclude that reaction O2 → CO2 is responsible
for diminishing amount of accessible O2 in the system, and for the
observed decrease of total yield of carbonylation. We studied the
effect of several parameters on this unfavorable process.
Fe2+ + Pd0 + 5CO → Fe(CO)5 + Pd2+
(9)
Similar results were also obtained for the process carried out in
the reactor with a teflon insert (but with stainless-steel cover of
the autoclave), indicating that traces of iron are sufficient to be a
co-catalyst. If the system is overloaded with iron powder, the yield
and selectivity are lower, Fig. 2B. Perhaps, iron reacts also with O2
and the amount of oxidizing agent (O2) is decreased. Both reactions
of metallic iron, with iodine and with O2, occur easily [44].
Surprisingly, the impact of the third component (i.e. PdCl2) is
more significant and two-fold increase of the amount of palladium
chloride causes a dramatic decrease of both: TOFAN and TOFEPC
(entry 9, Table 2S).
3.2.2. The effect of the reaction time and temperature
Despite the modification of the quantities of components of co-
catalyst, the yield and selectivity were not satisfactory, see Table
2S. Therefore, we determined the impact of the reaction time
on the conversion, yield, selectivity, and TOF values for the car-
bonylation of aniline in the presence of CO/O2. The results are
shown in Fig. 2C. The prolongation of the reaction time from 60
to 360 min (at 180 ◦C) caused rather small increase of the AN con-
version (from 54 to 63%) with the yield of EPC decreasing from 33%
to 26% and selectivity decreasing from 62% to 41%. EPC might react
with water (by-product of carbonylation of AN) and we experi-
mentally confirmed this hypothesis: after two hours of reaction of
ethyl N-phenylcarbamate with water under the same conditions
as carbonylation reactions, about 2 mmol (below 10% of starting
amount) of carbamate was decomposed to aniline, further react-
ing to other products: quinoline, 2 methylquinoline, N-ethylaniline,
and polyaniline.
3.2.4. Effect of catalyst and of temperature on CO2 production
Results shown in Table 2 (see also Table 5S) indicate that oxida-
tion of CO to CO2 occurs even in the absence of Fe and I2. Addition
of iodine, iron or pyridine caused an increase of CO2 amount. Iodine
alone as well as iodine together with iron caused the consumption
of 50% of the molecular oxygen – entries 4 and 5, Table 2 – (see also
Table 5S in Supporting material). When complex PdCl2Py2 was used
in presence of 0.5 ml of pyridine (no other components of catalytic
system were used in the reaction) 36% of O2 reacted with CO to give
CO2. In general, excess of each component accelerates the reaction
of O2 with CO and negatively affects the yield of the main product
(see Table 2 and S5).
The effect of temperature on the amount of CO2 (the conversion
of O2 to CO2) was also investigated in the presence of PdCl2 (with-
out addition of co-catalyst). At temperatures below 140 ◦C neither
carbonylation of aniline nor oxidation of CO to CO2 occurred (see
Table 6S). At 140 ◦C the conversion of AN was higher, but selectivity
For the series of experiments at temperature range 100–180 ◦C
(with fixed composition of the catalytic system) the process dura-
tion was always 60 min. The yield of diphenylurea (DPU) was
monitored for temperatures 100–140 ◦C, whereas, at 160–180 ◦C
we measured the yield of EPC because above 150 ◦C urea reacts with
ethanol to form carbamates. The yield of EPC (or DPU) remained
constant at 30% for the whole temperature range (from 100 ◦C to
180 ◦C), results are presented in Fig. 2D and in Table 4S.
1
2 mL of water, 20 ml of ethanol, 0.056 mmol of PdCl2, 0.12 mmol I2, 2.68 mmol
of Fe was heated in the autoclave at 180 ◦C for120 min under 4 MPa CO.

A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
209
Table 3
Conversion (CAN), yield (YDPU) and selectivity (SDPU) in carbonylation of aniline by mixture of CO/O2 and turnover frequency (TOF) depending on the amount of pyridine.a
b
c
d
No.
Catalyst
Py
[ml]
CAN
[%]
YDPU
[%]
SDPU
[%]
TOFAN
TOFDPU
1.
2.
3.
4.e
5.
6.
PdCl2
PdCl2
PdCl2(Py)2
PdCl2(Py)2
PdCl2(2-ClPy)2
PdCl2(2,4-Cl2Py)2
0
39
52
65
83
76
78
37
44
63
74
71
75
95
86
97
89
93
96
375
500
627
201
733
752
357
429
608
178
685
723
0.5
0.5
0.5
0.5
0.5
aReaction conditions: PdCl2 or PdCl2(Py)2/Fe/I2 = 0.056/1,2/0,04 mmol; AN = 54 mmol; ethanol = 20 ml; 100 ◦C; 3.6 MPa CO; 0.6 MPa O2; 60 min. Py = pyridine; AN = aniline;
DPU = N,N’-diphenylurea.
bSelectivity explained in footnoted in Table 1.
cConversion of O2 explained in footnote b in Table 2.
dTurnover frequency calculated analogously to TOFEPC explained in footnotes c in Table 1.
eTime = 120 min.
was very low (approximately 3%, see entry 2 in Table 6S). Further
increase of temperature caused the increase of the AN conversion
(up to 52%) and the yield of EPC (max. 22%). The production of CO2
was observed at 160 ◦C and 180 ◦C.
3.2.5. The effect of amount of pyridine on the reaction rate
Finally, the effect of pyridine on the reaction rate was inves-
tigated under optimized reaction conditions, i.e., at 100 ◦C (the
highest selectivity of the catalyst towards DPU was observed at
this temperature), the PdCl2:Fe:I2 ratio was 0.056:1.2:0.04 mmol,
respectively. Addition of pyridine increased the conversion and
yield but the selectivity of the catalyst was lower (entries 1 and
2, Table 3). When PdCl2(Py)2 complex with addition of 0.5 ml of
pyridine (entry 3, Table 3) was applied, all three parameters (con-
version, yield and selectivity) increased. When pyridine derivatives
(2-ClPy and 2,4-Cl2Py) were applied instead of pyridine, the higher
values of conversion and yield were obtained comparing to the pro-
cess carried out in the presence of Py. For the process carried out
with PdCl2(Py)2 and 0.5 ml of pyridine with extended time from 60
to 120 min the conversion (AN) reached 83% and the yield (DPU)
increased to 74%, but the selectivity was smaller (entry 4, Table 3).
In general, pyridine causes a significant increase of the conversion
and yield.
Complexes of PdCl2 with mono- or dichloropyridines applied
by us for carbonylation of aromatic amines under mild conditions
are more active catalysts than other described in the literature. For
catalytic system of Pd(OAc)2 with various N-heterocyclic water-
soluble ligands (67 atm of CO and O2, 150 ◦C) the highest TOF
obtained by Gupte et al. was 221 h−1 (for time = 2 h, yield = 88,6%)
or 543 h−1 (for time = 0.5 h, yield = 57%) [46]. For the complex of
Co(II) with Schiff base (150 ◦C, 7 h, 40 atm of CO and O2) the yield
Scheme 1. Proposed mechanism of the carbonylation of aniline by CO/O2 catalyzed
79% (i.e. TOF = 132 h−1) was obtained by Chen et al. [47]. Lower
by PdCl2(XnPy)2complexes.
TOF values (<100 h−1) were obtained by Chuang et al. for three cat-
alytic systems: PdCl2/NaI, CuCl2/NaI, and PdCl2/CuCl2/NaI (165 ◦C,
two molecules of AN are coordinated to Pd center instead of XnPy
4 h, 13 amt of CO and O2) [6]. TOF = 493 h−1 was obtained by Deng
ligands, Eq. (10) and Scheme 1.
et al. for PdCl2/ZrO2–SO4 (135 ◦C, 1 h, 40 amt of CO and O2) [48].
2−
Mulla et al. [39] obtained yields 59% and 34% for NBu4Ru(CO)3I3 and
PdCl2(PPh3)2 catalysts, respectively, but TOF values were smaller
than 2 h−1 (100 ◦C, 5 h, atmospheric pressure, CO:O2 = 2:1). Chaud-
hari et al. [49] for PdI2/NaI (170 ◦C, 2 h, 52 atm CO and O2) reached
TOF = 972 h−1 (calculation based on the literature data).
PdCl2(XnPy2)2 + 2PhNH2 → PdCl2(PhNH2)2 + 2XnPy
(10)
The complex PdCl2(PhNH2)2 was detected and characterized by
us during the reaction carried out without CO at 180 ◦C (see Section
2). The aniline ligands react with CO to form urea (Reaction (11))
and in this step palladium is reduced from Pd2+ to Pd0.
PdCl2(PhNH2)2 + CO + 2XnPy
3.3. Mechanism of the process
→ (PhNH)2CO + Pd0(XnPy)2 + 2HCl
(11)
Our previous studies on the carbonylation of AN in the pres-
ence of NB as oxidant [19] taken together with the results of
this work allow us to propose a mechanism of the carbonylation
of AN in the presence of molecular oxygen and catalytic system
PdCl2(XnPy)2/Fe/I2/XnPy (see Scheme 1). The process begins when
Partial precipitation of inactive Pdblack reported by Ragaini [50]
is one of the proofs of the Pd0 presence in the system (see also
Section 2). New ligand (aniline → urea) is weakly bonded to Pd0

210
A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
and is immediately replaced by XnPy forming Pd0(XnPy)m complex
(1a). Inactive Pdblack is recovered to Pd2+ ion by iron (II) cations (Eq.
(9) in the text) generated from co-catalyst containing Fe (powder)
and I2, see step 9 on Scheme 1, with Y = I [45].
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2015.06.
027
Pd0(XnPy)2 (1a) formed in Reaction (11) reacts with molecular
oxygen and Pd0 is oxidized to Pd2+ in 1b that immediately reacts
with two molecules of AN to give complex 1c. At the same moment a
molecule of water is released. Alternatively, Pd(0) in 1a may be oxi-
dized during the oxidative addition of I2 to Pd(0), according to the
equation: Pd(0) + I2 → PdI2. Then HI (instead of water) is released
and this HI is oxidized by molecular oxygen: 2HI + 0.5 O2 → I2 + H2O.
The further steps are as shown in Scheme 1. Two molecules of AN
coordinated to Pd2+ react with carbon monoxide and during this
step the complex 1d (with carbamoyl and amide ligand) is formed.
Positive ꢃ value obtained from Hammett equation indicates that
the RDS is accelerated when electron density in the reaction cen-
tre is decreasing. Moreover, significant substituent effects (in the
aniline ring) on the reaction rate were observed during the exper-
iments described previously [25] indicating that aniline molecules
are involved in RDS when carbonylation of AN occurs in the pres-
ence of O2 as oxidant. On the basis of both substituent effects — in
the pyridine and in the aniline ring — on the reaction rate we pro-
pose RDS: perhaps, the reaction rate is limited by the nucleophilic
attack of the amide ligand on CO ligand (both coordinated to Pd2+
in complex 1d). Diphenylurea is formed as leaving molecule and
the complex Pd0(XnPy)2 is recovered to be ready for the next cycle.
For process carried out at temperature above 150 ◦C diphenylurea
reacts with ethanol to form ethyl N-phenylcarbamate.
Supplementary data associated with this article contain: ele-
mental analysis of PdCl2(XnPy)2 complexes; conversion, yield and
selectivity in carbonylation of aniline by mixture of CO/O2 and
turnover frequency (TOF) depending on the composition of cat-
alytic system, time and temperature.
References
[1] F. Ragaini, S. Cenini, E. Gallo, A. Caselli, S. Fantauzzi, Curr. Org. Chem. 10
(2006) 1479.
[2] F. Ragaini, S. Cenini, D. Brignoli, M. Gasperini, E. Gallo, J. Org. Chem. 68 (2003)
460–466.
[3] M. Gasperini, F. Ragaini, S. Cenini, E. Gallo, J. Mol. Catal. A: Chem. 204 (2003)
107–114.
[4] T.L. Scott, B.C.G. Soderberg, Tetahedron 59 (2003) 6323–6332.
[5] R. Santi, A.M. Romano, F. Panella, G. Mestroni, O. Sessanti, A. Santi, J. Mol.
Catal. A: Chem. 144 (1999) 41–45.
[6] B. Chen, S.S.C. Chuang, J. Mol. Catal. A: Chem. 195 (2003) 37–45.
[7] F. Ragaini, M. Gasperini, S. Cenini, L. Arnera, A. Caselli, P. Macchi, N. Casati,
Chem. Eur. J. 15 (2009) 8064–8077.
[8] P.J. Giannoccaro, J. Organomet. Chem. 470 (1994) 249–252.
[9] D.J. Diaz, A.K. Darko, L. McElwee-White, Eur. J. Org. Chem. 27 (2007)
4453–4465.
[10] I. Chiarotto, M. Feroci, J. Org. Chem. 68 (2003) 7137–7139.
[11] N. Kumari, M. Sharma, P. Das, D.K. Dutta, Appl. Organometal. Chem. 16 (2002)
258–264.
[12] L. Pazderski, E. Szłyk, J. Sitkowski, B. Kamien´ ski, L. Kozerski, J. Tousek, R.
Marek, Magn. Reson. Chem. 44 (2006) 163–170.
[13] L. Lapidus, C. Pirozkhov, A.R. Tumanowa, A.W. Dolidze, A.M. Yuhimenko, Izv.
Akad. Nauk SSSR 8 (1991) 1885–1887.
[14] P. Wehman, V.E. Kaasjager, W.G.J. de Lange, F. Hartl, P.C.J. Kamer, P.W.N.M.
van Leeuwen, Organometallics 14 (1995) 3751–3761.
[15] T.M. Klapötke, K. Polborn, T.Z. Schütt, Anorg. Allg. Chem. 626 (2000)
1444–1447.
[16] P. Wehman, L. Borst, P.C.J. Kamer, P.W.N.M. van Leeuwen, J. Mol. Catal. A:
Chem. 112 (1996) 23–36.
[17] B.A. Steinhoff, I.A. Guzei, S.S. Stahl, J. Am. Chem. Soc. 126 (2004) 11268–11278.
[18] M. Gasperini, F. Ragaini, C. Cazzaniga, S. Cenini, Adv. Synth. Catal. 347 (2005)
105–120.
[19] A. Krogul, J. Skupin´ ska, G. Litwinienko, J. Mol. Catal. A: Chem. 337 (2011) 9–16.
[20] A. Krogul, J. Cedrowski, J. Wiktorska, W.P. Ozimin´ ski, J. Skupin´ ska, G.
Litwinienko, Dalton Trans. 41 (2012) 658–666.
[21] A. Krogul, J. Cedrowski, K. Wiktorska, W.P. Oziminski, J. Skupinska, G.
Litwinienko, Bioorg. Med. Chem. Lett. 23 (2013) 2765–2768.
[22] F. Ragaini, Dalton Trans. (2009) 6251–6266.
[23] K. Hiwatari, Y. Kayaki, K. Okita, T. Ukai, I. Shimazu, A. Yamamoto, Bull. Chem.
Soc. Jpn 77 (2004) 2237–2250.
[24] F.M. Mei, L.J. Chen, G.X. Li, Appl. Organomet. Chem. 24 (2010) 86–91.
[25] M. Karpin´ ska, P. Baran, J. Skupin´ ska, Pol. J. Chem. 82 (2008) 1105–1116.
[26] S.M. Lee, N.S. Cho, K.D. Kim, J.S. Oh, C.W. Lee, J.S. Lee, J. Mol. Catal. 73 (1992)
43–49.
4. Conclusions
Several PdCl2(XnPy)2 complexes (where: Py = pyridine,
X = −CH3, −Cl; n = 0–2) were synthesized and used as components
of the catalytic system: PdCl2(XnPy)2/Fe/I2/XnPy for carbonylation
of aniline to ethyl N-phenylcarbamate (or N,Nꢀ-diphenylurea) in
the presence of CO/O2. Conversion of aniline and yield of ethyl N-
phenylcarbamate (or N,Nꢀ-diphenylurea) were monitored in order
to investigate substituent effects in XnPy ligands on the catalytic
activity of PdCl2(XnPy)2 catalysts. The activity of PdCl2(XnPy)2
decreases with decreasing basicity of XnPy ligands (such effect was
not observed for the carbonylation of AN in the presence of NB
as oxidant). At 180 ◦C PdCl2(XnPy)2 complexes do not show high
selectivity towards carbamate during the carbonylation of aniline
by CO/O2 (in contrast to reaction carried out in the presence of NB)
and we propose that CO is consumed by side reactions. Analysis of
by-products confirms this explanation.
[27] F. Shi, Q. Zhang, Y. Gu, Y. Deng, Adv. Synth. Catal. 347 (2005) 225–230.
[28] I. Pri-Bar, J. Schwartz, J. Org. Chem. 60 (1995) 8124–8125.
[29] A. Vavasori, L. Toniolo, J. Mol. Catal. A: Chem. 151 (2000) 37–45.
[30] J. Skupin´ ska, M. Karpin´ ska, J. Mol. Catal. A: Chem. 161 (2000) 69–73.
[31] J. Skupin´ ska, M. Karpin´ ska, Appl. Catal. 267 (2004) 59–66.
[32] J. Skupin´ ska, M. Karpin´ ska, M. Ołówek, Appl. Catal. 284 (2005) 147–154.
[33] A. Krogul, J. Skupin´ ska, G. Litwinienko, J. Mol. Catal. A: Chem. 385 (2014)
141–148.
Optimization of reaction conditions lead to twice higher yield of
the main product and selectivity of catalyst (better than 90%) under
significantly milder conditions i.e., at lower temperature, shorter
time, for less amount of the catalyst. Comparing to the process
involving NB as an oxidant, the process carried out in the pres-
ence of molecular oxygen has lower impact on the environment.
PdCl2(XnPy)2 complexes are more efficient catalysts for this reac-
tion and under optimized conditions the catalytic activity of the
most effective complexes, PdCl2(2-ClPy)2 and PdCl2(2,4-Cl2Py)2,
increased about four times. According to our knowledge, these
complexes are more active than other catalysts described in the
literature for carbonylation of aromatic amines under mild condi-
tions.
[34] A. Drelinkiewicz, M. Hasik, M. Choczyn´ ski, Mater. Res. Bull. 33 (1998)
739–762.
[35] F. Yilmaz, Z. Kucukyavuz, E-Polymers (2009) 1–10.
[36] I. Mav, M. Zigon, Polym. Bull. 45 (2000) 61–68.
[37] J. Kriz, L. Starovoytova, M. Trchova, E.N. Konyushenko, J. Stejskal, J. Phys.
Chem. B 113 (2009) 6666–6673.
[38] J. Shorter, Correlation Analysis in Organic Chemistry: An Introduction to
Linear Free-Energy Relationships, Oxford University Press, 1973.
[39] S.A.R. Mulla, C.V. Rode, A.A. Kelkar, S.P. Gupte, J. Mol. Catal. A: Chem. 122
(1997).
[40] L.-J. Chen, F.-M. Ming, G.-X. Li, Y.-J. Xiang, Kinet. Catal. 51 (2010) 672–677.
[41] S. Fukuoka, M. Chono, M. Kohno, Chemtech (1984) 670–676.
[42] S.P. Gupte, R.V. Chaudhari, J. Catal. 114 (1988) 246–258.
[43] J.E. McCusker, F. Qian, L. McElwee-White, J. Mol. Catal. A: Chem. 159 (2000)
11–17.
Acknowledgements
This project was funded by the National Science Centre (research
project no. DEC-2011/01/N/ST5/02011)
[44] Gmelins Handbuch der Anorganischen Chemie, Springer–Verlag, New York,
59, 1979, 337.

A. Krogul, G. Litwinienko / Journal of Molecular Catalysis A: Chemical 407 (2015) 204–211
211
[45] J. Skupin´ ska, M. Karpin´ ska, M. Ołówek, T. Kasprzycka-Guttman, Centr. Eur. J.
[48] F. Shi, Y. Deng, T. SiMa, H. Yang, Tetrahedron Lett. 42 (2001)
2161–2163.
Chem. 3 (2005) 28–39.
[46] M.R. Didgikar, S.S. Joshi, S.P. Gupte, M.M. Diwakar, R.M. Deshpande, R.V.
Chaudhari, J. Mol. Catal. A: Chem. 334 (2011) 20–28.
[47] L.-J. Chen, J. Bao, F.-M. Ming, G.-X. Li, Catal. Commun. 9 (2008)
658–663.
[49] S. Kanagasabapathy, A. Thangaraj, S.P. Gupte, R.V. Chaudhari, Catal. Lett. 25
(1994) 361–364.
[50] F. Ragaini, S. Cenini, J. Mol. Catal. A: Chem. 109 (1996) 1–25.