Palladium-catalyzed synthesis of symmetrical urea derivatives by oxidative carbonylation of primary amines in carbon dioxide medium

Article abstract of DOI:10.1016/j.jcat.2011.06.003

An efficient palladium-catalyzed synthesis of symmetrically disubstituted ureas via oxidative carbonylation of primary amines is described. The reactions are carried out in the presence of a large excess of carbon dioxide as reaction medium or under solvent-free conditions. The adopted catalyst such as potassium tetraiodopalladate, stable and easy to prepare, allows the use of air as a cheap oxidizing agent. The reactions yield urea and water as the only by-product and proceed with high efficiency with aliphatic and aromatic amines as well. While with primary aliphatic amines, no significant improvement on reactivity is observed when carbon dioxide is used as a solvent, in comparison with the conventional ones, a remarkable high efficiency is obtained with aromatic amines, which shows a dramatic increase in the performance of the catalyst, in terms of turnover number (TON), the highest known so far for this kind of process. Reactions take place in two-phase systems consisting of a homogeneous liquid phase formed by the CO₂ expanded amine solution containing the catalyst and a supercritical phase of CO₂, CO, O₂, and N₂.

Full text of DOI:10.1016/j.jcat.2011.06.003

Journal of Catalysis 282 (2011) 120–127
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Palladium-catalyzed synthesis of symmetrical urea derivatives by oxidative
carbonylation of primary amines in carbon dioxide medium
Nicola Della Ca’ a, Paolo Bottarelli a, Angela Dibenedetto b, Michele Aresta b, Bartolo Gabriele c, Giuseppe
Salerno d, Mirco Costa a,
⇑
a Dipartimento di Chimica Organica e Industriale, Università di Parma, Parco Area delle Scienze, 17/A, 43124 Parma, Italy
b Dipartimento di Chimica and CIRCC, Università di Bari, Campus Universitario, 70126 Bari, Italy
c Dipartimento di Scienze Farmaceutiche, Università della Calabria, 87036 Arcavacata di Rende, Cosenza, Italy
d Dipartimento di Chimica, Università della Calabria, 87036 Arcavacata di Rende, Cosenza, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient palladium-catalyzed synthesis of symmetrically disubstituted ureas via oxidative carbonyl-
ation of primary amines is described. The reactions are carried out in the presence of a large excess of
carbon dioxide as reaction medium or under solvent-free conditions. The adopted catalyst such as potas-
sium tetraiodopalladate, stable and easy to prepare, allows the use of air as a cheap oxidizing agent. The
reactions yield urea and water as the only by-product and proceed with high efficiency with aliphatic and
aromatic amines as well. While with primary aliphatic amines, no significant improvement on reactivity
is observed when carbon dioxide is used as a solvent, in comparison with the conventional ones, a
remarkable high efficiency is obtained with aromatic amines, which shows a dramatic increase in the
performance of the catalyst, in terms of turnover number (TON), the highest known so far for this kind
of process. Reactions take place in two-phase systems consisting of a homogeneous liquid phase formed
by the CO2 expanded amine solution containing the catalyst and a supercritical phase of CO2, CO, O2, and
N2.
Received 10 January 2011
Revised 1 June 2011
Accepted 2 June 2011
Available online 7 July 2011
Keywords:
Ureas
Carbonylation
Homogeneous catalysis
Palladium
Carbon dioxide
Ó 2011 Elsevier Inc. All rights reserved.
RNH2 þ R0 NH þ CO þ 1=2O2 Pd cat: RNHCONR0 þ H O
ƒƒ!
ð2Þ
1. Introduction
2
2
2
Carbonylations of primary aliphatic amines [Eq. (1)] were car-
ried out in 1,2-dimethoxyethane (DME) at 100 °C under 60 bar of
a mixture of CO, air, and CO2 in the presence of a catalytic system
consisting of PdI2 in conjunction with KI [6]. The beneficial effect of
CO2 on carbonylation of primary aliphatic amines was related to
the fact that its presence led to the formation of carbamate inter-
mediates, thus buffering the substrate basicity, which would other-
wise hinder the palladium reoxidation [5]. In fact, working in the
absence of CO2 less satisfactory results were obtained. Primary aro-
matic amines were generally less reactive than aliphatic ones,
however. The conversion rate and product distribution turned
out to be dependent on the nature of the solvent. Thus, while no
conversion to urea derivatives was observed in the oxidative car-
bonylation of n-butylamine in a protic solvent such as MeOH, high-
er conversions were obtained in aprotic polar solvents [5b].
Apparently, the polarity of the aprotic solvent played a very impor-
tant role on the selectivity of the process: the monocarbonylation
The development of new, efficient, selective, and environmen-
tally friendly protocols for the production of ureas has recently at-
tracted great interest in view of their many important applications
[1]. The conventional syntheses of ureas have been based on the
use of dangerous reagents such as phosgene or isocyanates (mainly
prepared in their turn from phosgene itself). In recent years, how-
ever, alternative routes have been developed through the use of a
variety of carbonyl derivatives, CO2 or CO [2]. Particularly attrac-
tive, also from the standpoint of atom economy [3], is the oxidative
carbonylation methodology [4], which employs amines, carbon
monoxide, and oxygen as starting materials and produces only
water as coproduct.
Same years ago, we developed a new and efficient methodology
for performing the Pd-catalyzed oxidative carbonylation of pri-
mary amines leading selectively to ureas [Eqs. (1) and (2)], under
relatively mild reaction conditions and with unprecedented
catalytic efficiencies for this kind of reactions [5].
was strongly favored in low-polar dioxane (
mainly, for its higher coordination ability, in DME (
25 °C, 6.09 at 80 °C) [7], while double carbonylation was preferred
in the high polar N,N-dimethylacetamide (DMA, = 37.78 at 25 °C)
or N-methylpirrolidone (NMP, = 32.2 at 25 °C). The higher
e
= 2.21 at 25 °C) and
Pd cat:
2RNH2 þCO þ 1=2O2
RNHCOHNR þH O
ð1Þ
ƒƒ!
2
e
= 7.54 at
2
1
e
⇑
Corresponding author. Fax: +39 0521 905 472.
E-mail address: mirco.costa@unipr.it (M. Costa).
e
0021-9517/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2011.06.003

N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
121
nucleophilicity of primary amines toward CO in the more polar
DMA or NMP [8], favoring the formation of the Pd(CONHR)2 species
from which oxamide is generated by reductive elimination [9], was
responsible for this effect.
The above mentioned results, showing a considerable improve-
ment in efficiency of the Pd-catalyzed primary amine carbonyl-
ation reactions in the presence of aprotic solvents of low polarity
such as DME and gaseous CO2, prompted us to investigate the
possibility of carrying out the reactions in the absence of organic
solvents using CO2 as the only reaction medium, with particular
attention to the selectivity toward ureas rather than the undesired
by-products (oxamides).
In recent years, it has been clearly demonstrated that new sol-
vents, called gas expanded liquids (GXLs), have furnished effective
alternative media for performing chemical reactions. A GXL is a
mixed solvent composed of a compressible gas (such as CO2 or eth-
ane) dissolved in an organic solvent [10]. By varying the CO2
amount, the properties of the liquid media change and in addition,
these can be adjusted by tuning the operating pressure. Reaction
advantages include higher gas miscibility compared with organic
solvents, enhanced mass transport rate, due to the properties of
dense CO2 and increased reaction rates in comparison with the
ones in neat organic solvents or in scCO2 [11].
Our investigations is aimed to verify whether the beneficial
properties of CO2 expanded liquids can be fully or in part trans-
ferred from organic solvents to liquid reagents thus eliminating
the solvent in this particular urea derivative synthesis. The switch
from a conventional ‘‘wet chemistry’’ system to a ‘‘dry chemistry’’
system, besides the environmental advantages and the simplifica-
tion of the work-up procedures, could be particularly attractive for
this reaction, since it would provide the opportunity to completely
avoid explosion and combustion hazards connected with the use of
carbon monoxide/air mixtures [12]. Moreover, the use of CO2, due
to its particular features, could also, in some cases, led to an
improvement of the performance of the catalytic system leading
to unprecedented efficiencies.
thermocouple. Carbon dioxide was loaded as a liquid through a
dip tube cylinder and the amount was measured by weighing.
The temperatures reported in the Results and Discussion section
refer to the values measured by the internal thermocouple.
Elemental analyses were carried out with a Carlo Erba Elemen-
tal Analyzer Mod. 1106. 1H, and 13C NMR spectra were taken on a
Bruker AC300 (300 MHz) spectrometer. IR spectra were taken on a
Nicolet 5700 FT-IR spectrometer. Mass spectra were obtained
using a GC system HP6890 Series coupled with a HP 5973 Mass
Selective Detector at 70 eV ionization voltage. GC analyses were
performed on a HRGC Mega 2 series Fisons Instruments equipped
with a polymethylsilicone +5% phenylsilicone as a stationary phase
(HP-5) capillary column. Column chromatography was performed
on silica gel 60 (Merck, 70–230 mesh).
To monitor the phase behavior of the system consisting of CO2,
CO, O2, N2, and amine during the reaction, we used a stainless steel
high pressure reactor (Parr Instrument, 200 cm3 of volume)
equipped with two quartz windows placed at the bottom of the
reactor. Stirring was provided with a magnetic bar and heating
with an electric heater consisting of a fissured mantel in aluminum
well. Temperature control was achieved by means of a PID control-
ler connected to a thermocouple located in the reactor. Pressure in-
side the reactor was measured with a pressure gauge (220 bar end
scale).
2.2. Synthesis of potassium tetraiodopalladate (K2PdI4)
The catalytic system used was the easily prepared and stable
complex K2PdI4. 1 mmol of PdI2 (360 mg), 2 mmol of KI (332 mg),
and 15 ml of methanol were introduced in a round-bottom, sin-
gle-necked flask. The mixture was heated at reflux for one hour.
After evaporating the solvent under vacuum, the product was col-
lected as a deep brown powder (Yield: 98%).
2.3. General procedure for the carbonylation reactions
Recently, the synthesis of symmetrical and unsymmetrical ur-
eas has been achieved using amines, diamines, or some of their
derivatives in gaseous or supercritical carbon dioxide in the pres-
ence of suitable catalytic systems [13]. In particular, it has been re-
ported that dialkylamines react with CO2 (1 bar) at 60 °C giving
tetralkylureas in a single- or two-step procedure in the presence
of CCl4 and PPh3 or DMAN [1,8-bis(dimethylamino)naphthalene,
proton-sponge] in moderate to high yields [14]. Symmetrically
disubstituted ureas have also been obtained in modest to excellent
yields from primary amines working in ionic liquids in the
presence of CsOH at 170 °C and 60 bar of CO2 [15]. The synthesis
A 125 ml stainless steel reactor was dried in oven for 12 h, and
then cooled under argon to room temperature. The desired
amounts of amine and catalyst together with the eventual organic
solvent, as reported in the Results and Discussion section, were
loaded. Then the reactor was closed and purged three times with
purified dry air. The desired amount of liquid CO2 (99.5% purity)
was loaded as a liquid, weighing the reactor prior of and after
the charging. Then CO (10 bar) and air (15–17 bar) measured by
reading the pressure on the reactor manometer were added at
room temperature. The reaction was magnetically stirred and
heated in the range of 70–110 °C in a temperature-controlled oil
bath for 10–72 h. After the reaction, the reactor was cooled at room
temperature (0 °C in the case of the more volatile amines) and
carefully depressurized. The solid product was recovered by adding
CH2Cl2/petroleum ether (1 to 1 mixture, 10 ml, in the case of ali-
phatic urea derivatives) or acetone (10 ml, in the case of aromatic
urea derivatives) to the raw reaction material and by filtering to
separate the unconverted amine and the catalyst. The organic solu-
tion was analyzed by GC and GC–MS. Unless otherwise specified,
yields reported refer to isolated products. Crystallization from suit-
able solvents gives the pure products. Melting points, elemental
analyses, 1H, 13C NMR, and FT-IR spectra were consistent with
the reported literature (see supporting information section).
of
a family of O-silylcarbamates from silylamines has been
achieved by heating silylamines in scCO2. These O-silylcarbamates
were shown to be effective precursors for the synthesis of a range
of symmetrical and unsymmetrical ureas [16].
In the present work, the reactions of primary aliphatic and
aromatic amines with CO and O2 under high pressure of CO2 to give
symmetrical N,N0-disubstituted ureas, in the presence of K2PdI4 as
the catalyst, are described.
2. Experimental
2.1. General
All solvents were dried over activated molecular sieves for 24 h
or freshly distilled. All amines were freshly distilled. Purified air,
carbon monoxide, carbon dioxide, and all other reagents were pur-
chased from commercial sources and utilized as received.
Reactions under pressure were performed in stainless steel
125 ml Parr autoclaves, equipped with manometer and
2.4. Synthesis of 1,3-di-n-butylurea
As an example (entry 2, Table 1), n-butylamine (1.46 g,
20 mmol) and K2PdI4 (0.014 g, 0.02 mmol) were charged in the
reactor. Air was first flushed at low pressure and then CO2
(43.0 g, 1.0 mol) loaded as a liquid, weighing the reactor prior of

122
N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
Table 1
K2PdI4-catalyzed oxidative carbonylation of n-butylamine to N,N0-di-n-butylurea in CO2 under different conditions.a
Entry
mmol of
BuNH2
BuNH2/K2PdI4
molar ratio
CO2 loaded (g)
Reaction
temperature (°C)
Reactor
Reaction
time (h)
Urea
TONd
pressureb (bar)
yield (%)c
1
2
3
4
5
6
7e
8
9
20
20
20
20
20
20
20
30
30
1000
1000
1000
1000
1000
2000
2000
4000
10,000
33
43
53
61
74
50
–
70
70
70
70
70
80
80
90
100
111
136
161
180
211
166
18
18
18
18
18
24
24
48
48
75
85
87
80
70
91
97
93
42f
375
425
435
400
350
910
970
1860
2100
49
45
175
186
a
Reactor capacity = 125 ml. Unless otherwise noted, the P(CO) charged was +10 bar with respect to the pressure reached by the system at room temperature after the
loading of CO2; the P(air) charged was +15 bar with respect to the pressure reached by the system at room temperature after the loading of CO2 and CO.
b
Measured at the reaction temperature.
Isolated yield.
TON (turnover number) = mmol of urea obtained per mmol of K2PdI4 employed.
The reaction was carried out under solvent-free conditions, under a 4:1 mixture of CO–air (total pressure = 50 bar at room temperature).
c
d
e
f
Substrate conversion was 45%.
and after the charging. Then CO (10 bar) and air (15 bar) measured
by reading the pressure on the reactor manometer were added at
room temperature. The reaction was magnetically stirred and
H
N
H
N
R
R
(a)
(b)
heated at 70 °C in an oil bath for 18 h. After the reaction, the reac-
tor was cooled at room temperature and carefully depressurized.
Ten milliliter of CH2Cl2/petroleum ether (1:1) were added to the
raw reaction material, and the insoluble product was recovered
as a white solid (1.46 g, 85%) by filtration on a Buchner funnel sep-
arating the unconverted amine and the catalyst. The organic solu-
tion was analyzed by GC and GC-MS. Recrystallization from diethyl
ether/petroleum ether (1:1) gave pure 1,3-di-n-butylurea as color-
less crystals (m.p. 71–72 °C). All the spectroscopic analyses were
consistent with the reported literature (see supporting information
section).
O
O
n = 1
n = 2
PdI2/KI
H2O
2 R NH
2 + n CO + (1/2) O2
O
H
N
R
R
N
H
Scheme 1. PdI2/KI-catalyzed oxidative carbonylation of primary amines leading to
ureas (a) or oxamides (b).
Carbon dioxide can behave, under appropriate conditions, like
an aprotic solvent of low polarity, so we reasoned that it could
be a suitable reaction medium for the selective catalytic synthesis
of ureas, also because it is completely miscible with CO and O2 at
temperature higher than 304 °K (31 °C) [17]. We first examined
the reaction of n-butylamine with CO and air in the presence of a
catalytic amount of K2PdI4 and determined the yield of the corre-
sponding urea under different pressures of CO2 (Table 1). To this
aim, five experiments were initially carried out loading different
amounts of liquid CO2 in a stainless steel high pressure reactor
(125 ml capacity). Thus, the autoclave was charged in the presence
of air with K2PdI4 (0.02 mmol) and n-butylamine (20 mmol). After
introduction of the fixed and weighed amount of liquid CO2, CO
(+10 bar with respect to the reached pressure, P25 mmol CO)
and air (+15 bar with respect to the reached pressure, P10 mmol
O2) were charged at room temperature. The autoclave was im-
mersed into a thermostated oil bath, and the heating was adjusted
to have a temperature of 70 °C, measured with a thermocouple at
the bottom of the reactor. The reaction mixture was eventually
stirred for 18 h. Under these conditions, N,N0-di-n-butylurea was
selectively obtained, with yields ranging from 70% to 87%, depend-
ing on the amount of CO2 loaded (Table 1, entries 1–5). The higher
yields were observed in correspondence of about 40–50 g of loaded
CO2 (130–160 bar of total pressure at 70 °C, Table 1, entries 2 and
3). It should be noted, however, that the amount of CO2 necessary
for obtaining the best results may vary depending on the substrate
structure and temperatures.
2.5. Procedure for the carbonylation reaction of aniline as a function of
the reaction time (Fig. 1)
Aniline (1.862 g, 20 mmol) and K2PdI4 (1.4 mg, 0.002 mmol)
were charged in a 125 ml stainless steel reactor. Air was first
flushed at low pressure and then CO2 (99.5% purity, 40.0 g,
0.909 mol) loaded as a liquid. Then, air (+15 bar) measured by
reading the pressure on the reactor manometer was added at room
temperature. The reactor was immersed into an oil bath, magnet-
ically stirred and heated at 95 °C. CO (+10 bar) was added and this
was considered the starting reaction time. Total pressure was
170 bar. After established time, the reaction was stopped and the
residual liquid was examined by GLC for quantitative determina-
tion of aniline and phenylisocianate, whereas diphenyl urea was
separated and the yield determined as isolated product.
3. Results and discussion
3.1. Oxidative carbonylation of primary aliphatic amines
As mentioned in the Introduction, the PdI2/KI-catalyzed oxida-
tive carbonylation of primary amines to the corresponding sym-
metrical N,N0-disubstituted ureas has been shown to proceed
efficiently better in an aprotic, low polar, and weakly coordinating
solvent, such as DME. In this case, monocarbonylation [path (a),
Scheme 1] is preferred over dicarbonylation [path (b), Scheme 1],
so ureas are selectively formed.
The same substrate was then tested at different molar ratios
with the catalyst and at different temperatures (Table 1, entries

N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
123
BuNHCOHNBu
+
2KI +Pd(0) + 2HI
(3)
(4)
(5)
2BuNH2
+
CO
(1/2) O2
I2
+
K2PdI4
I2
+
H2O
K2PdI4
2HI
+
Pd(0)
+
+
2KI
Scheme 2. Catalytic cycle: production of urea and water and regeneration of the catalyst.
RN+H2CO2
-
RNH2 + CO2
RNH2 + RN+H2CO2
6–9). In one case (entry 7), the reaction was carried out under sol-
vent-free conditions, i.e., not adding CO2, and using a CO/air ratio of
4:1 in order to work outside the CO–air explosion range [18]. As
can be seen from Table 1, the results under solvent-free conditions
were slightly better than those obtained in presence of CO2. Yields
were still satisfactory for substrate-to-catalyst molar ratios as high
as 4000, although it was necessary to raise the reaction tempera-
ture and/or to extend the reaction time to achieve complete sub-
strate conversion (Table 1, entry 8). For a substrate-to-catalyst
molar ratio of 10,000, the reaction rate decreased significantly,
but a remarkable turnover number (TON) of 2100 mol of product
per mol of catalyst was achieved (Table 1, entry 9).
The results reported in Table 1 show that the rate of formation
of dibutylurea under CO2 is lower than that observed in DME
according to our previous works [5b]. As shown in Scheme 2, the
total catalytic process consists of the reactions 3–5, i.e., the sub-
strate carbonylation with reduction of PdI2 to Pd(0) and concomi-
tant formation of 2 mol of HI (Eq. (3)), the oxidation of the latter by
oxygen to give I2 (Eq. (4)), and the oxidative addition of I2 to Pd(0)
with regeneration of the catalyst (Eq. (5)) [5].
-
+
-
RNH3
RNHCO2
Scheme 3. Formation of alkylammonium carbamate derivatives.
under solvent-free conditions (entry 13, Table 2). This can be re-
lated to the high stability and low solubility of the carbamate
formed between tert-butylamine and carbon dioxide, which lowers
the reactivity of the system toward CO. This hypothesis was con-
firmed by loading tert-butylamine and CO2 in the reactor under
the usual reaction conditions; tert-butylammonium tert-butylcar-
bamate was isolated as an air-stable white solid (yield 86% work-
ing at 100 °C) [23]. Clearly, this salt could not be formed in the
absence of CO2, so the urea yields were higher under solvent-free
conditions (entries 13 and 14, Table 2). On the other hand, the
addition of an organic solvent (such as DME) along with CO2,
affecting the polarity of the reaction medium, could cause the for-
mation of an homogeneous CO2 expanded reaction mixture con-
sisting of the substrate, catalyst, DME, and CO2 [10a]. As a
consequence, ammonium carbamate dissolution and decomposi-
tion was achieved and the carbonylation reaction, carried out in
the presence of CO2 and DME (entry 12, Table 2), led to practically
the same yields obtained under solvent-free conditions (entry 13,
Table 2).
As we have already reported in the case of the reaction carried
out in DME, Pd(0) reoxidation can be favored by the presence of an
excess of added CO2, since the basicity of the amine (which would
‘‘block’’ HI, thus hindering Pd(0) reoxidation) is efficiently buffered
by CO2 [5].
These phase behaviors in the reaction course were observed
with the autoclave equipped with quartz windows. The corre-
sponding photographs are enclosed in the Supporting Information.
Francis has addressed the effect of basicity of organic com-
pounds on their solubility in CO2 [21a]. He has concluded that
weakly acidic CO2 has no noticeable affinity for moderately basic
aniline (Kb = 3.98 ꢀ 10ꢁ10) [20], pyridine (Kb = 2.30 ꢀ 10ꢁ10) [20],
and 2-picoline (Kb = 9.38 ꢀ 10ꢁ9) [20]. However, CO2 does form
In spite of the generally low polarity of primary aliphatic and
aromatic amines (n-BuNH2
e
= 5.3 at 21 °C, PhNH2
e = 6.9 at
20 °C, = 5.9 at 70 °C, PhCH2NH2
e
e
= 4.6 at 21 °C, = 4.3 at 50 °C)
e
[19], K2PdI4 was completely dissolved at room temperature in n-
BuNH2 and over 40 °C in PhNH2 under high CO2 pressure of the
reaction conditions. In addition, it is known that relatively strong
bases, such as primary and secondary aliphatic amines, may read-
ily react with CO2 to form alkylammonium carbamates (Scheme 3),
which in some cases can be rather stable under CO2 pressure and
insoluble in dense CO2, thus hindering the substrate reactivity
toward CO. As a consequence, these unfavorable factors can lower
the rate of the carbonylation reaction.
salts with stronger bases such as ammonia (Kb = 10ꢁ5
)
and
aliphatic amines (Kb = 10ꢁ4). He found p-phenetidine (Kb = 10ꢁ9
)
to be the borderline case, permitting observations on metastable
liquid–liquid solubilities before solid salt appears.
In the case of benzylamine (Kb = 9.35 ꢀ 10ꢁ5) [20], on the other
hand, the expected N,N0-dibenzylurea was obtained in high yields
both in CO2 and under solvent-free conditions, as shown in Table 3.
Actually n-BuNH2 (Kb = 4.37 ꢀ 10ꢁ4 [20]) formed a colorless
solid immediately on contact with dense CO2 [21]. By heating at
70 °C, it melted giving a colorless liquid that absorbed CO2 expand-
ing of 40–45% its volume compared with the one in the absence of
CO2 under the same conditions. At room temperature, the CO2
expanded IL (ionic liquid) [BuNHCO2- H3NBu+] mixture maintained
its fluidity under CO2 pressure [22]. After adding CO, the reaction
could start and the mixture turned to pale yellow its color. Precip-
itation of dibutylurea into the reaction mixture was observed when
the conversion of n-BuNH2 reached 25–30% at the temperature of
90 °C.
These factors were even more important with tert-butylamine
(Kb = 4.90 ꢀ 10ꢁ4) [20], which showed a lower reactivity with re-
spect to n-butylamine; 1,3-di-tert-butylurea being obtained in
33% yield working at 100 °C (Table 2, entry 10). tert-Butylamine
formed on contact with dense CO2 immediately a white solid so
much stable that even at 100 °C under CO2 pressure remained un-
changed. Under these conditions, CO could only react through a gas
solid reaction being the catalyst dispersed into the solid ammo-
nium salt. In order to obtain satisfactory yields, it was necessary
to add DME (entries 11 and 12, Table 2) or to carry out the reaction
3.2. Oxidative carbonylation of primary aromatic amines
As we previously reported, the yields and catalytic efficiencies
of the carbonylation reactions of aromatic amines carried out in
DME were lower than those observed with aliphatic amines [5b].
Surprisingly, however, the reaction of aromatic amines, conducted
in CO2 in the absence of any additional organic solvent, proceeded
with remarkably high efficiency, particularly in the case of aniline
to give N,N0-diphenylurea (DPU). Table 4 shows the results ob-
tained with aniline in CO2 under different reaction conditions.
Aniline was found to react efficiently with substrate-to-catalyst
molar ratios ranging from 2000 to 100,000. In this latter case, it
was sufficient to opportunely raise the reaction temperature and
extend the reaction time in order to obtain a DPU yield as high
as 87%, with an unprecedented turnover number of 43,500 (Table
4, entry 22). By comparison, the reaction in DME reached a maxi-
mum turnover number of 480, as reported in our previous work

124
N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
Table 2
K2PdI4-catalyzed oxidative carbonylation of tert-butylamine to 1,3-di-tert-butylurea in CO2 under different conditions.a
Entry mmol of t-
CO2 loaded
(g)
Reactor pressureb
(bar)
Reaction temperature
(°C)
Reaction time
(h)
Urea yield
(%)c
TONd
t-BuNH2 K2PdI4 molar
ratio
BuNH2
10
20
20
20
20
40
2000
2000
2000
2000
10,000
45
46
45
–
186
205
203
100
110
110
110
110
24
24
24
24
24
33
62
90
91
67
500
620
900
910
3350
11e
12f
13g
14g
–
a
Reactor capacity = 125 ml. Unless otherwise noted, the P(CO) charged was +10 bar with respect to the pressure reached by the system at room temperature after the
loading of CO2; the P(air) charged was +15 bar with respect to the pressure reached by the system at room temperature after the loading of CO2 and CO.
b
Measured at the reaction temperature.
Isolated yield.
TON (turnover number) = mmol of urea obtained per mmol of K2PdI4 employed.
3 ml of DME was added to the reaction mixture.
15 ml of DME was added to the reaction mixture.
The reaction was carried out under solvent-free conditions, under a 4:1 mixture of CO–air (total pressure = 50 bar at room temperature).
c
d
e
f
g
Table 3
K2PdI4-catalyzed oxidative carbonylation of benzylamine to N,N0-dibenzylurea in CO2 under different conditions.a
Entry mmol of
BnNH2
BnNH2/K2PdI4 molar
ratio
CO2 loaded
(g)
Reactor pressureb
(bar)
Reaction temperature
(°C)
Reaction time
(h)
Urea yield
(%)c
TONd
15
20
30
20
2000
4000
4000
40
–
40
142
166
80
100
100
24
36
24
99
96
96
990
1920
1920
16e
17
a
Reactor capacity = 125 ml. Unless otherwise noted, the P(CO) charged was +10 bar with respect to the pressure reached by the system at room temperature after the
loading of CO2; the P(air) charged was +15 bar with respect to the pressure reached by the system at room temperature after the loading of CO2 and CO.
b
Measured at the reaction temperature.
Isolated yield.
TON (turnover number) = mmol of urea obtained per mmol of K2PdI4 employed.
The reaction was carried out under solvent-free conditions, under a 4:1 mixture of CO–air (total pressure = 50 bar at room temperature).
c
d
e
Table 4
K2PdI4-catalyzed oxidative carbonylation of aniline to N,N0-diphenylurea (DPU) in CO2 under different conditions.a
Entry mmol of
PhNH2
PhNH2/K2PdI4 molar
ratio
CO2 loaded
(g)
Reactor pressureb
(bar)
Reaction temperature
(°C)
Reaction time
(h)
DPU yield
(%)c
TONd
18
19
20
21
22
23f
24f
30
40e
41e
40e
40e
20
4000
8000
20,000
60,000
100,000
2000
41
41
41
38
37
–
163
169
175
172
170
90
100
100
110
110
90
24
24
36
60
72
24
72
90
95
92
96
87
96
36g
1800
3800
9200
28,800
43,500
960
40
100,000
–
110
18,000
a
Reactor capacity = 125 ml. Unless otherwise noted, the P(CO) charged was +10 bar with respect to the pressure reached by the system at room temperature after the
loading of CO2; the P(air) charged was +15 bar with respect to the pressure reached by the system at room temperature after the loading of CO2 and CO.
b
Measured at the reaction temperature.
Isolated yield.
TON (turnover number) = mmol of urea obtained per mmol of K2PdI4 employed.
The P(air) charged was +17 bar with respect to the pressure reached by the system at room temperature after the loading of CO2 and CO.
The reaction was carried out under solvent-free conditions, under a 4:1 mixture of CO–air (total pressure = 50 bar at room temperature).
The substrate conversion was 41%.
c
d
e
f
g
[5b]. These results are unprecedented in literature and are about
two orders of magnitude higher than the best ones reported with
other catalytic systems [5b]. The difference in reactivity between
aniline and aliphatic amines may be ascribed mainly to two factors.
First, the formation of ammonium carbamates from aromatic
amines are prevented or owing to their low basicity (aniline

N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
125
Kb = 3.98 ꢀ 10ꢁ10) or to their very low stability compared with the
ones of aliphatic amines, so their conversion is more favorable,
leading to an increase in the carbonylation rate. An inspection
through the quartz windows of the mixture of aniline and CO2 at
25 °C into autoclave showed a 30–35% homogeneous expanded
aniline solution and overhanging liquid CO2. The aniline solution
containing the catalyst was homogeneous under CO2 pressure at
40 °C. Secondly, CO2 is able to increase the solubility of gases in
the mixture of the organic reagent and the catalyst. Recently, it
has been shown [24] that a high gas solubility can be achieved
by the addition of CO2 to a gas–liquid system; as a consequence,
mass transfer can be substantially enhanced and high reaction
rates can be reached in the system, which remains essentially bi-
phasic. In fact, while it is known that aniline is only slightly soluble
in scCO2 (3% at 32 °C and 175 bar [21b]), it is acknowledged that
gases with critical temperatures below 304 K, including CO and
oxygen, are miscible with scCO2 in all proportions [11e]. Thus,
the reaction apparently takes place in a biphasic system consisting
of a liquid phase formed by the expanded solution of the catalyst in
aniline at 90 °C (the catalyst has been found to be completely sol-
uble in aniline at the reaction temperatures under CO2 pressure)
and a supercritical phase formed by carbon dioxide, carbon mon-
oxide, nitrogen, and oxygen. Qualitative visual observations of
the biphasic system under the reaction conditions performed car-
rying out the experiments inside an autoclave equipped with two
quartz windows showed that no precipitation of diphenyl urea oc-
curred up to 40–45% of aniline conversion was reached at 90 °C. On
the other hand, in the absence of CO2, diphenyl urea is insoluble in
neat aniline at 90 °C. The mutual solubility of the two phases is
then efficiently enhanced by the presence of the supercritical med-
ium. This allows overcoming mass transport problems at the inter-
face of the phases, leading to a faster and more efficient reaction.
However, in absence of CO2, the gases reach the liquid phase
[11b] more difficultly. When the reaction of aniline was carried
out under solvent-free conditions, less satisfactory results were ob-
tained with respect to the experiment conducted in CO2: under the
same conditions of entry 23 (100,000:1 M ratio substrate: catalyst,
110 °C for 72 h, urea yield 87%), the DPU yield lowered to 36% at
41% substrate conversion (entry 24, Table 4), thus confirming the
effective role of CO2 in the reaction course.
Other aromatic amines were then examined, both in the
presence and in the absence of CO2. Tables 5 reports the results
obtained for several para- and meta-substituted anilines.
The results obtained deserve some comments. p-Methoxyani-
line showed a good reactivity even at 80 °C for 24 h under CO2
pressure (94% yield, Table 5, entry 27), whereas it showed no reac-
tivity at all under solvent-free conditions (entry 28). This behavior
may be connected to the melting point of the considered amine. In
fact, p-anisidine has a melting point of 57–60 °C, and under the
reaction conditions (80 °C) is probably in a very aggregated molten
Table 5
K2PdI4-catalyzed oxidative carbonylation of aromatic amines in CO2 under different conditions.a
Entry ArNH2 (Kb)b
ArNH2/K2PdI4 molar
ratio
CO2
(g)
T
(°C)
Reactor pressurec
(bar)
Time
(h)
Substrate conversion
(%)
Urea yield
(%)d
TONe
7200
25
p-CH3OC6H4NH2
20,000
41
100
169
24
75
72
(Kb = 2.29ꢂ10ꢁ9
)
26
27
28f
29
30
31f
32
33f
34
35
p-CH3OC6H4NH2
p-CH3OC6H4NH2
p-CH3OC6H4NH2
p-CF3OC6H4NH2
p-CF3OC6H4NH2
p-CF3OC6H4NH2
p-CH3C6H4NH2 (Kb = 1.20ꢂ10ꢁ9
p-CH3C6H4NH2
10,000
6000
6000
6000
6000
6000
6000
6000
15,000
6000
41
41
–
40
40
–
41
–
42
40
80
80
80
148
146
24
24
24
24
24
24
20
24
70
24
80
100
0
74
100
99
96
49
55
100
77
94
0
70
94
93
91
46
51
97
3850
2820
0
2100
2820
2790
2730
1380
3825
2910
80
150
165
100
100
90
)
159
85
p-CH3C6H4NH2
110
80
182
149
m-CH3C6H4NH2
(Kb = 5.12 ꢀ 10ꢁ10
m-CH3C6H4NH2
)
36f
37
38
39f
40
6000
6000
6000
6000
5000
–
38
39
–
80
80
80
80
80
24
24
24
24
24
79
88
59
60
45
75
85
56
57
41
2250
2550
1680
1710
1025
m-CF3C6H4NH2 (Kb = 10ꢁ12
)
140
142
g
p-FC6H4NH2 (Kb = 4.47 ꢀ 10ꢁ10
)
p-FC6H4NH2
p-ClC6H4NH2
39
143
140
(Kb = 9.55 ꢀ 10ꢁ11
)
)
41f
42
p-ClC6H4NH2
5000
5000
–
38
80
80
24
24
0
52
0
48
0
p-BrC6H4NH2
(Kb = 7.59 ꢀ 10ꢁ11
p-BrC6H4NH2
p-IC6H4NH2
1200
43f
44
5000
5000
5000
2000
–
38
–
80
80
80
80
24
24
24
48
0
51
0
0
32
0
0
800
0
141
140
45f
46
p-IC6H4NH2 (Kb = 6.55 ꢀ 10ꢁ11
p-NO2C6H4NH2
)
38
0
0
0
(Kb = 1.02 ꢀ 10ꢁ13
)
a
Reactor capacity = 125 ml. The amount of ArNH2 was 20 mmol in all experiments. Unless otherwise noted, the P(CO) charged was +10 bar with respect to the pressure
reached by the system at room temperature after the loading of CO2; the P(air) charged was +15 bar with respect to the pressure reached by the system at room temperature
after the loading of CO2 and CO.
b
Ref. [20].
c
Measured at the reaction temperature.
Isolated yield.
TON (turnover number) = mmol of urea obtained per mmol of K2PdI4 employed.
The reaction was carried out under solvent-free conditions, under a 4:1 mixture of CO–air (total pressure = 50 bar at room temperature).
d
e
f
g
Calculated according to the method reported in reference [25].

126
N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
phase, which hinders reactivity. In the presence of CO2, this phe-
nomenon is less important and the reactivity increases. In fact,
the presence of compressed CO2 is known to lower the melting
point of organic solids, as long as their temperature is not too far
above the critical temperature of CO2 and this is soluble in the mol-
ten organic [10c]. A similar behavior was observed for the reaction
of p-toluidine, whose melting point is a little higher than room
temperature (40–43 °C). Accordingly, under solvent-free condi-
tions, a reduced yield (46%, entry 33) was obtained in comparison
with 91% for the reaction carried out in CO2 (entry 32). The same
behavior was observed with p-haloanilines [with X = Cl,
(m.p. = 69–72 °C), Br, (m.p. = 62–64 °C), or I, (m.p. = 63–65 °C), en-
tries 40–45)]. These compounds have melting points close to or
higher than the reaction temperature; thus, no conversion at all
was observed in absence of solvent (entries 41, 43, and 45). On
the other hand, p-fluoroaniline and p-trifluoromethoxyaniline are
liquid at room temperature: thus, no positive effect of CO2 was ob-
served and the reactions proceeded smoothly also under solvent-
free conditions (see entries 30, 31, 38, and 39). In any case, the
reactivity of p-haloanilines was found to be lower in comparison
with p-methoxy and p-methylanilines, due to the deactivating
effect of the halogen substituent, while their relative reactivity
decreased in the order F > Cl ꢃ Br > I probably correlated with a
different capacity of absorption of CO2 into the p-haloamine. The
unusually low selectivity observed with p-iodoaniline (urea
yield = 32%, entry 44) can be due also to secondary reactions
between the palladium intermediates and the CAI bond. Finally,
p-nitroaniline showed no reactivity at all, due to the strong deacti-
vating effect of the nitro group and/or to the very high melting
point (mp 146–149 °C) (entry 46).
Fig. 1. Composition of the reacting mixture as a function of the reaction time during
the synthesis of diphenyl urea from aniline (20 mmol, PhNH2/K2PdI4 molar
ratio = 10,000), CO2 (40 g), air 15 bar, CO 10 bar; total pressure 170 bar at 95 °C.
RNH2
+
CO2
+
K2PdI4
CO
RNHCOOPdI
+ 2KI + HI
RNHCOOPd(CO)I
RNHCOOPdI
+
I
RNHCOOPd(CO)I
RNHCOPdI
RNHCOPdI
+
CO2
II
RN=C=O
+ HI +
Pd(0)
III
The reactivity of some m-substituted anilines was also tested in
order to better demonstrate the general applicability of this pro-
cess. The results obtained with m-toluidine (Kb = 5.12 ꢀ 10ꢁ10
[20]) and m-trifluoromethylaniline are reported in Table 5, entries
35–37. As expected, for m-toluidine (mp < room temperature), the
results obtained in the reaction carried out under solvent-free con-
ditions (entry 36) were similar to those observed in scCO2 (entry
35). As already noticed in the case of p-substitutes aryl amines,
the presence of the halogen atoms slightly decreased the reactivity
of the system, due to the deactivating effect on the aromatic ring
(entry 37).
RN=C=O
2HI 1/2O2
Pd(0) 2KI
+ RNH2
RNHCONHR
+ H2O
+
I2
K2PdI4
+
+ I2
Scheme 4. Proposed reaction pathway.
4. Conclusions
A green and efficient catalytic process has been developed for
the oxidative carbonylation of primary amines to N,N0-disubsti-
tuted ureas in the presence of CO2 as the reaction medium or under
solvent-free conditions. The reactions yield no other by-product
than water and proceed with high performance with aliphatic as
well as aromatic amines. The reactions of aniline and n-butylamine
take place in a biphasic systems consisting of an homogeneous
liquid phase formed by CO2 expanded aniline solution containing
the catalyst and a supercritical phase of CO2, CO, O2, and N2 and
in a biphasic system consisting of CO2 expanded IL of the salt BuN-
3.3. Proposed carbonylation pathway
A series of palladium-catalyzed carbonylation reactions of ani-
line (20 mmol.) were performed varying reaction times, and the
profile of the reaction was obtained (Fig. 1).
The catalyzed reaction exhibits an induction period of ca 2–3 h
during which a orange–red color appears. The production of DPU
proceeds in two steps. The time-conversion curves in Fig. 1 indi-
cate that the slow step of the Pd-catalyzed formation of PhNCO
is followed by the rapid condensation of aniline under reaction
conditions. The active catalyst species, which leads to PhNCO for-
mation, likely forms in the first hours after that there is a boost
in the urea production. Oxidation of Pd(0) does not appear the lim-
iting step since by varying the O2 pressure, the reaction rate does
not change appreciably.
Formation of ureas 2 can be rationalized as depicted in Scheme
4, involving the formation of a carbamoylpalladium complex II in
pre-equilibrium with starting materials or with a palladium carba-
mate intermediate complex I. Intermediate II then evolves to form
isocyanate III (by HI elimination from the PhHN–CO–Pd–I moiety),
from which ureas are obtained by nucleophilic attack by the amine
[5]. In agreement with this hypothesis, isocyanates were detected
in the reaction mixture in low conversion experiments.
-
HCO2 H3NBu+ containing the catalyst and a supercritical phase of
CO2, CO, O2, and N2, respectively. The use of CO2 as a reaction med-
ium allows several advantages with respect to the reactions carried
out in conventional solvents: it allows a greener process since no
organic solvent is needed; moreover, the safety of the process is
improved since the excess of carbon dioxide prevents any explo-
sion or combustion hazards. While with primary aliphatic amines,
no significant improvement on reactivity is observed when carbon
dioxide is used as a solvent, a remarkable high efficiency is ob-
tained with aromatic amines, which shows the highest perfor-
mances known so far for this kind of process. This is a further
proof of the fact that the switch to non-conventional reaction med-
ia can not only improve the environmental feasibility of a chemical
process, but in some cases can also lead to a significant improve-
ment on reactivity.

N.D. Ca’ et al. / Journal of Catalysis 282 (2011) 120–127
127
(2002) 2513;
Acknowledgments
(c) P.J. Jessop, B. Subramaniam, Chem. Rev. 107 (2007) 2666. and references
therein.
[11] (a) T. Clifford, Fundamental of Supercritical Fluids, Oxford University Press,
Oxford, New York, 1999;
Financial support from The Ministero dell’Università e della
Ricerca Scientifica e Tecnologica (Progetto d’Interesse Nazionale
PRIN 2008A7P7YJ-005) is acknowledged. The facilities of Centro
Interfacoltà di Misure (Università di Parma) were used for record-
ing NMR spectra. Thankfulness is expressed to Marco Dardari for
his technical assistance in the autoclave handling with carbon
dioxide and carbon momoxide.
(b) M. Poliakoff, N.J. Meehan, S.K. Ross, Chem. Ind. (1999) 750;
(c) P.G. Jessop, W. Leitner, in: P.G. Jessop, W. Leitner (Eds.), Chemical
Synthesis Using Supercritical Fluids, Wiley-VCH, Weinheim, Germany, 1999;
(d) J.M. De Simone, W. Tumas, in: J.M. De Simone, W. Tumas (Eds.), Green
Chemistry Using Liquid and Supercritical Carbon Dioxide, Oxford University
Press, Oxford, New York, 2003;
(e) E.J. Beckman, J. Supercrit. Fluids 28 (2004) 121;
(f) A. Beiker, Chem. Rev. 99 (1999) 453. and references therein;
(g) P.G. Jessop, Y. Hsiao, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 118 (1996) 344.
[12] J.O. Pande, J. Tonheim, Proc. Safety Progr. 20 (2001) 37.
[13] (a) B.M. Bhanage, S. Fujita, Y. Ikushima, M. Arai, Green Chem. 5 (2003) 340;
(b) A. Iou, V. Parvulescu, P. Jacobs, D. De Vos, Green Chem. 9 (2007) 158;
(c) T. Jiang, X. Ma, Y. Zhou, S. Liang, J. Zhang, B. Han, Green Chem. 10 (2008)
465.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2011.06.003.
[14] C.-C. Tai, M.J. Huck, E.P. McKoon, T. Woo, P.G. Jessop, J. Org. Chem. 67 (2002)
9070.
[15] F. Shi, Y. Deng, T. SiMa, J. Peng, Y. Gu, B. Qiao, Angew. Chem., Int. Edit. 42
(2003) 3257.
[16] M.J. Fuchter, C.J. Smith, M.W.S. Tsang, A. Boyer, S. Saubern, J.H. Ryan, A.B.
Holmes, Chem. Commun. (2008) 2152.
[17] M. Aresta, in: M. Aresta (Ed.), Carbon Dioxide as Chemical Feedstock, Wiley-
VCH, Weinheim, 2010.
[18] The explosion range for CO in air is 17–70% CO at 18–20 °C and atmospheric
pressure, 14.8–71.5% CO at 100 °C and atmospheric pressure; at higher total
pressure, the range of flammability decreases: for example, at 20 atm and
20 °C the range is ca. 20–60% CO). See C.M. Bartish, G.M. Drissel, in: M.
Grayson, D. Eckroth, G.J. Bushey, L. Campbell, A. Klingsberg, L. van Nes (Eds.),
Kirk-Othmer, Encyclopedia of Chemical Technology, 3rd ed., vol. 4, Wiley-
Interscience, New York, 1978, p. 774.
References
[1] (a) A.F. Hegarty, L.J. Drennan, in: A.R. Katritzky, O. Meth-Cohn, C.W. Rees
(Eds.), Comprehensive Organic Functional Group Transformations, Pergamon
Press, Oxford, 1995, p. 499;
(b) T.P. Vishnyakova, I.A. Golubeva, E.V. Glebova, Russ. Chem. Rev. (Engl.
Transl.) 54 (1985) 249.
[2] (a) F. Bigi, R. Maggi, G. Sartori, Green Chem. 2 (2000) 140;
(b) F. Shi, Y. Deng, T. SiMa, J. Peng, B. Qiao, Angew. Chem., Int. Edit. 42 (2003)
3257;
(c) J. Fournier, C. Bruneau, P.H. Dixneuf, S. Lecolier, J. Org. Chem. 56 (1991)
4456.
[3] B.M. Trost, Angew. Chem., Int. Edit. 34 (1995) 259.
[4] (a) For a general review on oxidative carbonylation of amines to ureas see: D.J.
Diaz, A.K. Darko, L. McElwee-White, Eur. J. Org. Chem. (2007) 4453. and
references therein;
[19] (a) A. Maryott, R. Smith, National Bureau of Standards, Circular 514 (1951)
(Issued August 10);
(b) A.J. Gordon, R.A. Ford,
A Handbook of Pratical Data, Technique and
(b) M.R. Didgikar, D. Roy, S.P. Gupte, S.S. Ioshi, R.V. Chaudhari, Ind. Eng. Chem.
Res. 49 (2010) 1027;
(c) K. Orito, M. Miyazawa, T. Nakamura, A. Horibata, H. Ushito, H. Nagasaki, M.
Yuguchi, S. Yamashita, T. Yamazaki, M. Tokuda, J. Org. Chem. 71 (2006) 5951;
(d) B. Gabriele, G. Salerno, M. Costa, Topics Organomet. Chem. 18 (2006) 239;
(e) F. Shi, Y. Deng, J. Catal. 211 (2002) 548.
References, Wiley, New York, 1972. p. 12.
[20] A.E. Martell, R.M. Smith, Critically Selected Stability Constants of Metal
Complexes, NIST Standard Reference Data, Gaitherburg, MD 20899 USA, 2003.
[21] (a) A.W. Francis, J. Phys. Chem. 58 (1954) 1099;
(b) D.K. Dange, J.P. Heller, K.V. Wilson, Ind. Eng. Chem. Prod. Res. Dev. 24
(1985) 162.
[22] J. Kumelan, A.P-S. Kamps, D. Tuma, G. Maurer, J. Chem. Thermodyn. 38 (2000)
1396.
[23] (a) M. Aresta, E. Quaranta, Tetrahedron 48 (1992) 1515;
(b) S. Inoue, N. Yamazaki, Organic and Bio-organic Chemistry of Carbon
Dioxide, Kodansha, Tokyo, 1981. p. 82;
(c) R. Nomura, Y. Hasegawa, M. Ishimoto, T. Toyosaki, H. Matsuda, J. Org.
Chem. 57 (1992) 7339.
[24] D. Chouchi, D. Gourgouillon, M. Courel, J. Vital, M.N. Da Ponte, Ind. Eng. Chem.
Res. 40 (2001) 2551.
[25] D.D. Perrin, B. Dempsey, E.P. Serjeant, Pka prediction for organic acids and
bases, Chapman and Hall, London, New York, 1981.
[5] (a) B. Gabriele, R. Mancuso, G. Salerno, M. Costa, Chem. Commun. (2003) 486;
(b) B. Gabriele, G. Salerno, R. Mancuso, M. Costa, J. Org. Chem. 69 (2004) 4741.
[6] B. Gabriele, M. Costa, G. Salerno, G.P. Chiusoli, J. Chem. Soc., Chem. Commun.
(1992) 1007.
[7] G. Goldoni, L. Marcheselli, G. Pistoni, L. Tassi, S. Fanali, J. Chem. Soc., Faraday
Trans. 88 (1992) 2003.
[8] For general solvent effects on nucleophilicity, see J. March, Advanced Organic
Chemistry, 4th ed., Wiley-Interscience, New York, 1992. p. 357.
[9] I. Pri-Bar, H. Alper, Can. J. Chem. 68 (1990) 1544.
[10] (a) G. Musie, M. Wei, B. Subramaniam, D.H. Busch, Coordin. Chem. Rev. 219
(2001) 789. and references therein;
(b) M. Wei, G.T. Musie, D.H. Bush, D. Subramaniam, J. Am. Chem. Soc. 124