Pd-catalysed oxidative carbonylation of amino alcohols to N,N′-bis(hydroxyalkyl)ureas under mild conditions using molecular oxygen as the oxidant

Article abstract of DOI:10.1016/j.apcata.2009.12.022

A very simple method has been developed for the selective synthesis of symmetrical N,N′-bis(hydroxyalkyl)ureas, OC[NH-(CH₂)_x-OH]₂ (x = 3-6), by oxidative carbonylation of α,ω-amino alcohols [3-aminopropanol (3-AP), 4-aminobutanol (4-AB), 5-aminopentanol (5-APe), 6-aminohexanol (6-AH)] with CO/O₂ mixtures (O₂ = 5 mol%) in the presence of Pd(II)/ligand/NEt₃·HI catalytic systems. The catalytic process takes place under very mild conditions (p(CO/O₂) = 0.1 MPa; 303-333 K). The target products can be isolated in high yield through a very simple and straightforward procedure. The catalytic system can be easily recovered and recycled for several times. The influence of a few reaction parameters (nature of ancillary ligand and iodide co-catalyst, I/Pd molar ratio, etc.) on the catalytic activity has also been investigated and the main mechanistic features of the catalytic process fully elucidated.

Full text of DOI:10.1016/j.apcata.2009.12.022

Applied Catalysis A: General 375 (2010) 78–84
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Pd-catalysed oxidative carbonylation of amino alcohols to
N,N0-bis(hydroxyalkyl)ureas under mild conditions using molecular oxygen
as the oxidant
b
a
a
Potenzo Giannoccaro a, , Carla Ferragina , Michele Gargano , Eugenio Quaranta
*
a
`
Dipartimento di Chimica, Universita di Bari, Campus Universitario, via Orabona 4, 70126 Bari, Italy
b CNR-IMIP, via Salaria Km 29.300, 00016 Monterotondo, Roma, Italy
A R T I C L E I N F O
A B S T R A C T
very simple method has been developed for the selective synthesis of symmetrical N,N0-
bis(hydroxyalkyl)ureas, OC[NH-(CH2)x-OH]2 (x = 3–6), by oxidative carbonylation of -amino alcohols
Article history:
A
Received 23 September 2009
Received in revised form 12 December 2009
Accepted 14 December 2009
Available online 22 December 2009
a,v
[3-aminopropanol (3-AP), 4-aminobutanol (4-AB), 5-aminopentanol (5-APe), 6-aminohexanol (6-AH)]
with CO/O2 mixtures (O2 = 5 mol%) in the presence of Pd(II)/ligand/NEt3ꢀHI catalytic systems. The
catalytic process takes place under very mild conditions (p(CO/O2) = 0.1 MPa; 303–333 K). The target
products can be isolated in high yield through a very simple and straightforward procedure. The catalytic
system can be easily recovered and recycled for several times.
Keywords:
Oxidative carbonylation
Amino alcohols
N,N0-bis(hydroxyalkyl)ureas
The influence of a few reaction parameters (nature of ancillary ligand and iodide co-catalyst, I/Pd
molar ratio, etc.) on the catalytic activity has also been investigated and the main mechanistic features of
the catalytic process fully elucidated.
Pd-catalysts
ß 2010 Elsevier B.V. All rights reserved.
1. Introduction
are also available, but they often involve multistep procedures
aimed at protecting the OH functional group in order to avoid the
Oxidative carbonylation of
b
-amino alcohols to afford oxazo-
side-formation of carbamates and carbonates (Scheme 1).
lidin-2-ones is a topic of current interest due to the remarkable
biological activity exhibited by five-membered cyclic carbamates
[1]. Both homogeneous [2] and heterogeneous [3] catalytic
systems have been described for this process, which, depending
on the nature of the catalyst used, can be accomplished under
either mild or drastic conditions. In comparison, oxidative
carbonylation of non-vicinal amino alcohols has received much
poorer attention, despite its potential for the direct synthesis of
N,N0-bis(hydroxyalkyl)ureas rather than cyclic carbamates.
N,N0-bis(hydroxyalkyl)ureas are useful compounds which find
practical application in several fields [4]. Most of N,N0-bis(hydrox-
yalkyl)ureas can be prepared in a poorly selective way by direct
reaction of non-vicinal amino alcohols with phosgene (Scheme 1)
[4a,b], a toxic hazardous reagent whose utilization in chemical
industry finds larger and larger constraints due to governmental
policies for environmental protection. Environmentally more
friendly synthetic protocols, based on the reaction of amino alcohols
with less toxic phosgene-equivalents (1,1-carbonyldiimidazole,
dimethyl ditiocarbonate [5], dimethyl- or diethyl carbonate [6]),
Only recently N,N0-bis(hydroxyalkyl)ureas have been prepared
by oxidative carbonylation of amino alcohols. Diaz et al. used
W(CO)6 as catalyst and stoichiometric amounts of I2 as the oxidant,
under a CO pressure as high as 8 MPa [7]: co-generation of large
amounts of iodide salts, as well as tedious work-up of reaction
crude, was a serious drawback of the process.
As a part of our studies on the reactivity of key intermediates in
the oxidative carbonylation of alcohols [8,9], diols [10] and amines
[11], we have recently found that a few LnPdCl2 complexes can
promote the carbonylation of amino alcohols (NH2-(CH2)x-OH) to
cyclic carbamates and/or N,N0-bis(hydroxyalkyl)ureas and demon-
strated that these reactions proceed through the intermediate
formation of carbamoyl complexes LnPdCl[C(O)NH-(CH2)x-OH]
[12]. In general, these species exhibit very modest stability, but the
complexes with triphenylphosphine ligand are fairly stable, and
have been isolated, characterized and studied for their reactivity.
Scheme 2 shows that the complexes PdCl(PPh3)2[C(O)NH-(CH2)x-
OH] are thermally unstable and can decompose (T > 303 K) by
generating a Pd(0)-intermediate, ‘‘Pd(PPh3)2’’, and a very reactive
chloroformamide, ClC(O)NH-(CH2)x-OH, which, depending on the
reaction conditions and the amino alcohol used, can afford either
cyclic carbamates or N,N0-bis(hydroxyalkyl)ureas. According to
whether the decomposition was carried out under N2 or CO
* Corresponding author. Tel.: +39 080 5442093; fax: +39 080 5442129.
E-mail address: giannoccaro@chimica.uniba.it (P. Giannoccaro).
0926-860X/$ – see front matter ß 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2009.12.022

P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
79
adding the acid to the amine solution, the organic layer was
separated and the salt, NEt3ꢀHI, was extracted from the aqueous
layer with CH2Cl2. The CH2Cl2 extracts were collected together
and from the resulting solution, concentrated to a small volume,
white-cream crystal needles of NEt3ꢀHI were isolated upon
addition of Et2O.
IR spectra were recorded on
a Shimadzu IR-Prestige-21
Scheme 1. Carbonylation of amino alcohols with phosgene or phosgene-
spectrophotometer. GC separations were performed by using a
Varian Cromopack CP3800 GC equipped with a CP Sil 8 CB capillary
column (30 m, 0.53 mm ID), connected to a FID detector. GC–MS
analyses were carried out with a Shimadzu GC 17-A linked to a
Shimadzu GC/MS QP5050A selective mass detector (capillary
column: HP-5 MS, 30 m). NMR spectra were run on a Bruker AM
equivalents.
atmosphere, the palladium(0)-intermediate, ‘‘Pd(PPh3)2’’, may
further react in the reaction mixture by converting, into Pd(0)-
carbonyl complexes and/or palladium metal, respectively. Re-
markably, in the presence of a suitable oxidant, the latter species
can be easily reoxidized to Pd(II), which in the presence of free
amino alcohol and CO can regenerate the starting carbamoyl
complex, LnPdCl[C(O)NH-(CH2)x-OH], allowing, thus, the overall
process to proceed in a catalytic way [12] (Scheme 2).
500 instrument or
a Varian Inova 400 spectrometer. Gas-
chromatographic analyses of gas mixtures were carried out on a
Porapak Q column (3.5 m), using a Carlo Erba Fractovap GC
equipped with a thermal conductivity detector.
Herein, we report the results obtained with a few tetra-
coordinated LnPdCl2 complexes [(L = 2,20-dipyridine (dipy) (n = 1);
2.2. Pd(II)-catalysed oxidative carbonylation of amino alcohols:
general procedure
2-(b-diphenylphosphine)ethylpyridine) (PN) (n = 1); PPh3 (n = 2);
CH3CN (n = 2)] which, in the presence of NEt3ꢀHI as co-catalyst and
a mixture of CO/O2 (p(CO/O2) = 0.1 MPa; O2 = 5 mol%), act as
environmentally benign, highly efficient and stable catalysts for
the conversion
eas.
All reactions were carried out in a ꢁ100 mL glass reactor
equipped with a Sovirel cap and a Thorion stopcock. Herein, as an
example of typical procedure, the oxidative carbonylation of 4-
aminobutanol (4-AB), catalysed by PdCl2(PN)/NEt3ꢀHI, is described
in detail.
a,v
-amino alcohols to N,N0-bis(hydroxyalkyl)ur-
CH3CN (10 mL), NEt3 (0.5 mL), 4-AB (4.00 mmol), PdCl2(PN)
(0.021 mmol) and NEt3ꢀHI (0.17 mmol; I/Pd = 8 mol/mol) were
introduced into the reactor under a N2 stream. After N2 removal,
the reactor was connected to a flask filled with a CO/O2 mixture
(O2 = 5 mol%) at ambient pressure (761 mmHg) and temperature
(300 K), introduced into an electrical oven equipped with a
thermostat and a magnetic stirrer, and allowed to react at 333 K.
After 7 h, the volume of consumed gas mixture was measured by
means of a gas burette and the solution was analyzed for reactants
and products.
2. Experimental
2.1. General
All manipulations were carried out by using standard vacuum-
line techniques. Solvents, reactants (amino alcohols, NEt3, KI), as
well as triphenylphosphine and 2,20-bipyridine (dipy) ligands,
were Aldrich or Fluka products and were used as received. PN (2-
(b-diphenylphosphino)ethylpyridine) ligand and the Pd-complex
The reaction mixture was reacted with pure O2 at 333 K for
20 min (to oxidize minor amounts of Pd-black formed in the
reaction system) and then cooled to room temperature. A white
precipitate was obtained which was filtered off, washed with a 2:1
(v/v) Et2O/CH3CN mixture, dried in vacuo and identified as N,N0-
bis(4-hydroxybutyl)urea by means of elemental analysis and
PdCl2(PN) were synthesized as described in the literature [13]. The
other Pd(II)-complexes, PdCl2(PPh3)2 [14], PdCl2(dipy) [15] and
PdCl2(CH3CN)2 [16], were obtained according to the literature by
reacting, at 343 K, a suspension of PdCl2 in CH3CN with the
appropriate ligand. NEt3ꢀHI was prepared at 273 K, by reacting
aqueous HI with an excess of NEt3 dissolved in CH2Cl2. After
Scheme 2. Thermal decomposition of Pd-carbamoyl complexes from amino alcohols and formation of products under oxidative conditions.

80
P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
Table 1
Oxidative carbonylation of
a,v
-amino alcohols to N,N0-bis(hydroxyalkyl)ureasa.
b
Entry
Pd-complex
Substrate
Reaction time (h)
VCO=O (mL)
Substrate consumedc (mmol)
Uread (%)
TOFe (hꢂ1
)
2
1
2
PdCl2(dipy)
PdCl2(PN)
3-AP
3-AP
3-AP
3-AP
4-AB
4-AB
4-AB
4-AB
5-APe
5-APe
5-APe
5-APe
6-AH
6-AH
6-AH
6-AH
4-AB
7
7
7
7
7
7
7
7
8
8
8
8
9
9
9
9
7
65 (97)
64 (95)
62 (93)
55 (82)
64 (95)
62 (92)
80 (89)
54 (81)
64 (95)
65 (97)
62 (92)
56 (83)
65 (99)
64 (95)
60 (90)
50 (75)
56 (81)
3.80 (95)
3.76 (94)
3.72 (93)
3.24 (81)
3.76 (94)
3.68 (90)
3.52 (88)
3.16 (79)
3.76 (94)
3.84 (96)
3.60 (90)
3.28 (82)
3.80 (95)
3.76 (94)
3.52 (88)
2.96 (74)
3.24 (81)
78 (82)
78 (83)
78 (84)
65 (80)
87 (93)
84 (94)
84 (95)
72 (91)
85 (90)
88 (92)
81 (90)
76 (93)
90 (95)
88 (94)
79 (90)
67 (90)
75 (93)
25.9
25.6
25.3
22.0
25.6
25.0
23.9
21.5
22.4
22.9
21.4
19.5
20.1
19.9
18.6
15.7
22.0
3
PdCl2(PPh3)2
4
PdCl2(CH3CN)2
PdCl2(dipy)
PdCl2(PN)
5
6
7
PdCl2(PPh3)2
PdCl2(CH3CN)2
PdCl2(dipy)
PdCl2(PN)
8
9
10
11
12
13
14
15
16
18
PdCl2(PPh3)2
PdCl2(CH3CN)2
PdCl2(dipy)
PdCl2(PN)
PdCl2(PPh3)2
PdCl2(CH3CN)2
PdCl2(CH3CN)2 + Ph3PO
a
Experimental conditions. Pd-complex: 0.021 mmol; substrate (amino alcohol): 4.00 mmol, NEt3: 0.5 mL; NEt3ꢀHI: 0.17 mmol (I/Pd = 8 mol/mol); solvent (CH3CN): 10 mL;
CO/O2 (19 mol/mol): 0.1 MPa; T = 333 K.
b
Volume (mL) of gas mixture consumed. The % volume of consumed gas mixture, with respect to the stoichiometric value, is reported in parentheses.
c
Values determined by GC. Substrate conversion (%) is reported in parentheses.
d
Isolated yields, based on the starting material. Selectivity to urea is reported in parentheses.
e
Mol of carbonylated substrate per mol of catalyst per hour. Based on GC yields. Error for TOF values is, in general, equal to ꢃ0.1 hꢂ1
.
Table 2
Oxidative carbonylation of 4-AB: influence of I/Pd molar ratio on the catalytic activity of PdCl2(dipy)a.
b
Entry
I/Pd molar ratio
VCO=O (mL)
4-AB consumedc (mmol)
Uread (%)
TOFe (hꢂ1
)
2
1
2f
3f
4f
5f
6g
7g
0,00
2
Traces
Traces
Traces
–
31 (47)
58 (87)
64 (95)
64 (95)
21 (31)
44 (65)
1.80 (45)
3.44 (86)
3.76 (94)
3.80 (95)
1.16 (29)
2.56 (64)
42 (93)
80 (93)
87 (93)
87 (92)
26 (91)
58 (90)
12.2
23.4
25.6
25.9
7.9
6
8
10
2
8
17.4
a
PdCl2(dipy): 0.021 mmol; 4-AB: 4.00 mmol, NEt3: 0.5 mL; solvent (CH3CN): 10 mL; CO/O2 (5 mol% O2): 0.1 MPa; T = 333 K; reaction time: 8 h.
Volume (mL) of gas mixture consumed. The % volume of consumed gas mixture, with respect to the stoichiometric value, is reported in parentheses.
Values determined by GC. Substrate conversion (%) is reported in parentheses.
b
c
d
e
f
Isolated yields, based on the starting material. Selectivity to urea is reported in parentheses.
Mol of carbonylated substrate per mol of catalyst per hour. Based on GC yields. Error for TOF values is, in general, equal to ꢃ0.1 hꢂ1
.
Iodide added as NEt3ꢀHI.
g
Iodide added as KI.
spectroscopic techniques (IR, 1H NMR, 13C NMR) [12a]. The filtered
mother solution, which can be directly reused for a new catalytic
run (see Section 2.3 and Table 3), was analyzed by IR spectroscopy.
The IR spectrum showed a weak band at 1715 cmꢂ1, assigned to
cyclic carbamate 1-oxa-3-aza-cycloheptan-2-one, whose presence
was also confirmed by means of GC–MS analysis of the solution
(M+ = 115 m/z) [12a].
For further details (volume of absorbed CO/O2 mixture, amount
of reacted amino alcohol, urea yield and selectivity) see Tables 1
and 2.
2.3. Catalyst recycling
Catalyst recycling experiments were carried out using 4-AB as
the substrate. After separating the urea from the reaction mixture
(see also Section 2.2), fresh 4-AB (4.00 mmol) was added to the
filtered mother solution. The resulting mixture was allowed to
According to the above procedure, 3-aminopropanol (3-AP), 5-
aminopentanol (5-APe), 6-aminohexanol (6-AH) were also con-
verted into the corresponding ureas, N,N0-bis(3-hydroxypropyl)urea,
N,N0-bis(5-hydroxypentyl)urea and N,N0-bis(6-hydroxyhexyl)urea,
respectively. The isolated ureas were solid compounds and were
identified by means of elemental analyses and spectroscopic data.
Their full characterization (IR, 1H and 13C NMR, MS) has been
reported elsewhere [12a]. In accordance with what was found for
oxidative carbonylation of 4-AB, the IR spectra of filtered reaction
Table 3
Catalyst recycling: TOFa (hꢂ1) values over six cyclesb.
Catalyst
Cycles
1st
2nd
3rd
4th
5th
6th
solutions display a weak absorption between 1722 and 1708 cmꢂ1
,
PdCl2(dipy)
PdCl2PN
25.6
25.0
23.9
21.5
25.2
24.5
23.7
16.3
24.6
24.2
23.5
8.5
24.0
23.9
23.3
4.9
23.7
23.5
23.1
23.5c
23.8c
23.4d
which, on the basis of our previous observations [12a] and GC–MS
spectra, was attributed to small amounts of cyclic carbamate (1-oxa-
3-aza-cyclohexan-2-one (1708 cmꢂ1; M+ = 101 m/z), 1-oxa-3-aza-
cycloctan-2-one (1722 cmꢂ1; M+ = 129 m/z) and 1-oxa-3-aza-cyclo-
nonan-2-one (1722 cmꢂ1; M+ = 143 m/z), respectively).
An analogous procedure was also followed when KI was used as
iodide source in place of NEt3ꢀHI (see, for instance, entries 6 and 7,
Table 2).
PdCl2(PPh3)2
PdCl2(CH3CN)2
a
Mol of carbonylated substrate per mol of catalyst per hour. Based on GC yields.
Error for TOF values is, in general, equal to ꢃ0.1 hꢂ1
.
b
4-AB (4.00 mmol) was used as the amino alcohol, in all the runs. See also Section
2.3.
c
0.021 mmol of ligand were added in this run.
0.042 mmol of ligand were added in this run.
d

P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
81
react with CO/O2 for 7 h at 333 K and then treated as described in
Section 2.2. The TOF values observed in each recycling experiment
have been reported in Table 3.
P, 8.78; Pd, 15.11. trans-PdCl2(PPh3)2 yield: 50–60%. The best yields
were obtained in the experiments using an excess of phosphine.
Runs (e) and (f): The colour of the reaction solution, initially
orange, turned red-brown very rapidly (30 min). In this case, pure
PdI2(PPh3)2, identified by means of IR spectroscopy and elemental
analysis, was isolated. Elemental analysis. Calcd for C36H30I2P2Pd
(%): I, 28.68; P, 7.00; Pd, 12.03. Found: I, 28.61; P, 6.94; Pd, 11.98;
Cl, traces. trans-PdI2(PPh3)2 yield: 90–95%. The best yields were
obtained in the experiments using an excess of phosphine.
In all the runs ((a)–(f)), the GC analysis of the reaction solution
showed the formation of variable amounts of triphenylphosphine
oxide.
2.4. Elucidation of the catalytic steps
2.4.1. Reductive step: stoichiometric reaction of PdCl2(PPh3)2 with 4-
AB and CO
To a suspension of PdCl2(PPh3)2 (0.600 g, 0.86 mmol) in CH3CN
(10 mL), both NEt3 (0.5 mL) and 4-AB (0.170 g, 1.89 mmol; 4-AB/
Pd = 2.2 mol/mol) were added under a N2 stream. N2 was pumped
off and CO was introduced at atmospheric pressure into the glass
reactor. The mixture was allowed to react at room temperature for
24 h. In the long run the suspension, initially yellow, turned white-
cream and, then, pink. The IR spectrum of the reaction solution
3. Results and discussion
(orange) showed bands at 1858 (w) and 1834 (w, br) cmꢂ1
,
3.1. Oxidative carbonylation of a,v-amino alcohols promoted by
respectively assigned to the three-nuclear carbonyl complexes
Pd3(CO)3(PPh3)3 and Pd3(CO)3(PPh3)4 [17]. The reaction mixture
was filtered and the pink residue was characterized as a mixture of
Pd3(CO)3(PPh3)3 and N,N0-bis(4-hydroxybutyl)urea. The separation
of the latter compounds was easily achieved by treating the
mixture with methanol, which readily dissolved the urea. The red
residue, insoluble in methanol, was identified as Pd3(CO)3(PPh3)3
LnPdCl2/NEt3ꢀHI systems: factors affecting the catalytic activity
Under safe and very mild conditions (303–333 K; p(CO/
O2 = 0.1 MPa; O2 = 5 mol%), LnPdCl2/NEt3ꢀHI systems [(L = 2,20-
dipyridine (dipy) (n = 1); 2-(b-diphenylphosphine)ethylpyridine)
(PN) (n = 1); PPh3 (n = 2); CH3CN (n = 2)] effectively promote the
oxidative carbonylation of -amino alcohols, as 3-AP, 4-AB, 5-
a,v
(0.238 g; yield based on palladium: 70%). IR(nujol):
(
nCO),
APe and 6-AH, to the corresponding N,N0-bis(hydroxyalkyl)ureas
(Eq. (1)) with high yield and selectivity (Tables 1 and 2). All the
urea products were isolated as crystalline solids and characterized
by means of IR and NMR spectroscopy [12a].
1860 cmꢂ1 [17]. Elemental analysis. Calcd for C57H45O3P3Pd3
(%): C, 57.52; H, 3.81; P, 7.81; Pd, 26.83. Found: C, 57.46; H, 3.97; P,
7.76; Pd, 26.74.
The methanol solution was evaporated in vacuo. From the
residue, after washing with toluene to remove residual traces of
Pd(0)-carbonyls, a white solid was isolated, whose IR spectrum
was identical to that of a pure sample of N,N0-bis(4-hydroxybu-
tyl)urea (0.52 mmol). More urea (0.15 mmol) was recovered from
the CH3CN mother solution (see above) by evaporating in vacuo the
solvent and washing the white residue with toluene and, then,
diethyl ether (urea overall yield: 78%).
2H2N-ðCH2Þx-OH þ CO þ 1=2O2
! OC½NH-ðCH2Þx-OHꢅ2 þ H2O ðx ¼ 3ꢂ6Þ
(1)
The catalytic process was monitored by means of GC and IR
analysis of reaction solution and by measuring the volume of gas
mixture consumed. As CO2 was never found among the reaction
products, the volume of gas consumed can be related directly with
the amount of amino alcohol carbonylated. Table 1 shows the
results obtained when the amino alcohols 3-AP, 4-AB, 5-APe and 6-
AH were reacted with CO/O2 mixtures (0.1 MPa; O2 = 5 mol%) in
the presence of the catalytic systems LnPdCl2/NEt3ꢀHI, at 333 K and
using a I/Pd molar ratio equal to 8. Under the working conditions
(Table 1), good to excellent substrate conversions (74–96%) and
urea yields (67–90%, isolated) were achieved within reasonable
times (7–9 h) with all the amino alcohols, whatever LnPdCl2
(L = = dipy, PN, PPh3, CH3CN) catalyst was used. Selectivity to urea
(based on isolated product) was excellent (ꢆ90%) for amino
alcohols 4-AB, 5-APe and 6-AH. A lower, but still satisfactory,
selectivity (80–84%) was found for the carbonylation of 3-AP to
N,N0-bis(3-hydroxypropyl)urea because of more significant side-
formation, in this case, of the six-membered cyclic carbamate 1-
oxa-3-aza-cyclohexan-2-one. The latter fact can find an easy
rationale in the shorter aliphatic chain, which makes the –OH and –
NH2 functional groups closer in 3-AP than in the other substrates, a
structural feature expected to facilitate, in the case of 3-AP, the
cyclization side-process affording the cyclic carbamate.
2.4.2. Oxidative step
Pd3(CO)3(PPh3)3, isolated as reported above (see Section 2.4.1),
was allowed to react with O2, for 6 h, under various conditions (see,
below, items (i)–(iii)).
(i) Pd3(CO)3(PPh3)3 was added to the filtered CH3CN solution
ensuing from the oxidative step (see Section 2.4.1) and reacted
with O2 either in the absence (run (a)) or in the presence of
added triphenylphosphine (run (b): added PPh3/Pd = 1 mol/
mol);
(ii) Pd3(CO)3(PPh3)3 was added to a CH3CN solution of NEt3ꢀHCl
(Cl/Pd = 2.3) and reacted with O2 in the presence of added PPh3
(run (c): added PPh3/Pd = 1 mol/mol; run (d): added PPh3/
Pd = 1.5 mol/mol);
(iii) Pd3(CO)3(PPh3)3 was added to a CH3CN solution of NEt3ꢀHI (I/
Pd = 2.3) and reacted with O2 in the presence of added PPh3
(run (e): added PPh3/Pd = 1 mol/mol; run (f): added PPh3/
Pd = 1.5 mol/mol).
In all the runs ((a)–(f)), to minimize the decomposition of the
Pd(0)-complex to Pd-black, reaction temperature was gradually
increased to 333 K during the first 2 h, and then maintained
constant at this value for the remaining time.
Runs (a)–(d): The colour of the reaction solution, initially
orange, turned yellow very slowly (ꢄ3–4 h). trans-PdCl2(PPh3)2,
impure of Pd-metal, was isolated by filtration. Pure PdCl2(PPh3)2
was obtained by extraction with hot CH2Cl2 and identified by
means of elemental analysis and the IR spectrum, which were
similar to those of an authentic sample. Elemental analysis. Calcd
for C36H30Cl2P2Pd (%): Cl, 10.10; P, 8.82; Pd, 15.16. Found: Cl, 10.06;
A greater insight into the catalytic process comes from the
inspection of the TOF (turnover frequency) values reported in
Table 1, which allow to compare not only the efficiency of a given
catalytic system in the carbonylation of the different substrates,
but also their different activities in the ureidization of a given
amino alcohol. The TOF values (Table 1) show that, whatever
substrate (3-AP, 4-AB, 5-APe and 6-AH) was used, the catalytic
activity was affected by the nature of the ancillary ligand
coordinated to palladium and decreased in the order:
dipy ꢁ PN > PPh3. In the absence of any ligand, a less efficient
catalytic system was obtained (see Table 1, entries 4, 8, 12, 16).

82
P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
Moreover, whatever LnPdCl2/NEt3ꢀHI [(L = 2,20-dipyridine (dipy)
LnMþðnꢂ2Þ þ 2Hþ þ 1=2O2 ! LnMðþnÞ þ H2O
(3)
(n = 1); 2-(b-diphenylphosphine)ethylpyridine) (PN) (n = 1); PPh3
(n = 2); CH3CN (n = 2)] system was utilized, the TOF values
decreased in the order 3-AP > 4-AB > 5-APe > 6-AH (Table 1),
showing that the longer the amino alcohol aliphatic chain, the
lower the rate of carbonylation process.
We were also able to ascertain that the carbonylation of
amino alcohol to urea (Eq. (1)) promoted by palladium(II)-
complexes may proceed through such a mechanism. Herein, we
discuss a few major mechanistic features of the overall catalytic
process.
The influence of other factors, such as the I/Pd molar ratio, as
well as the nature of iodide source, was also investigated, using 4-
AB as reference substrate and PdCl2(dipy) as the Pd-catalyst
(Table 2). The results reported in Table 2 show unambiguously that
(a) iodide has a beneficial effect on the catalytic activity (see also
Section 3.2.2) and (b) the catalytic efficiency (TOFs) markedly
depends on the I/Pd molar ratio. In the absence of any iodide
source, substrate conversion was negligible and only trace
amounts of urea were formed (entry 1, Table 2). Using NEt3ꢀHI
as iodide source, a marked increase of TOF was observed for I/Pd
values rising between 0 and 8 mol/mol (entries 2–4, Table 2). I/Pd
molar ratios higher than 8 bring about only a very modest increase
of TOF (entry 5, Table 2). A different iodide promoter, such as KI,
was also effective, but not as efficient as NEt3ꢀHI (entries 6 and 7,
Table 2). KI/Pd molar ratios up to 10 have been employed in Pd-
catalysed oxidative carbonylation of vicinal amino alcohols under
relatively mild conditions [2c].
The catalytic systems investigated are stable under the working
conditions and can be easily recovered and reused for successive
runs. Recovery of the catalytic system can be accomplished in a
direct simple way. At the end of the catalytic run the reaction
mixture was allowed to react for about 20 min with pure O2, to
avoid loss of Pd-catalyst as Pd-metal. Urea was separated by
filtration and the filtered solution was reusable for a new catalytic
run (see Section 2.3). Catalyst recycling experiments (Table 3) were
carried out using, in each run, fresh 4-AB as substrate, which was
reacted with a CO/O2 mixture (0.1 MPa; O2 = 5 mol%) at 333 K for
7 h. The TOF (hꢂ1) values measured for the different Pd-catalysts
over several (up to six) recycling runs are reported in Table 3. The
catalytic activity decayed steadily when PdCl2(CH3CN)2 was used
as catalyst, a behaviour which finds full rationalization in the
reaction mechanism discussed in the next section (see Section
3.2.2). When the catalytic activity of metal centre (Pd) was assisted
by ancillary ligands as dipy, PN or PPh3, TOF diminished very
slowly after each run. In these cases the slight decrease of catalytic
efficiency over successive cycles can be ascribed to unavoidable
minor losses of catalyst during the recovery procedure and, with
the catalytic systems stabilized by PPh3 or PN, to the oxidation of
the ligands.1 Accordingly, the TOF values measured for the sixth
recycling (Table 3) show that, upon adding free ligand to the
catalyst solutions deriving from the fifth cycle, the catalytic activity
increased only for the catalytic systems stabilized by P- or P,N-
donor ligands as PPh3 or PN.
3.2.1. The reductive step
To get a greater insight into the reductive step, PdCl2(PPh3)2 and
4-AB were reacted stoichiometrically under a CO atmosphere
(0.1 MPa) in the presence of an excess of tertiary amine (NEt3), at
room temperature (Eq. (4)). The reaction afforded the correspond-
ing urea, N,N0-bis(4-hydroxybutyl)urea, together with the polynu-
clear ‘‘Pd(PPh3)-carbonyl’’ complexes Pd3(CO)3(PPh3)3 and
Pd3(CO)3(PPh3)4, detected in the reaction solution by means of
IR analysis (bands at 1858 and 1838 cmꢂ1, respectively) [17]. Only
N,N0-bis(4-hydroxybuthyl)urea and Pd3(CO)3(PPh3) precipitated
from the reaction medium and were isolated as pure compounds in
78% and 70% yields, respectively.
PdCl2ðPPh3Þ2 þ CO þ 2H2N-ðCH2Þ4-OH þ 2NEt3
! OC½NH-ðCH2Þ4-OHꢅ2þ‘‘PdðPPh3Þ-carbonyl’’ þ 2NEt3ꢀHCl
(4)
The hydroxyalkylurea formed also when PdCl2(CH3CN)2 was
used in place of PdCl2(PPh3)2, showing that the nature of the
ancillary ligand does not affect markedly the course of the
carbonylation step. Nevertheless, the data in Table 3 demonstrate
that the ancillary ligand can play a crucial role in stabilizing the
catalytically active species and prolonging the lifetime of the
catalytic system.
We have ascertained that the reductive carbonylation step
involves, as the key intermediate,
a relatively unstable
carbamoyl complex LnPdCl[C(O)NH-(CH2)x-OH], which forms
according to reaction (5). Nevertheless, the PPh3-complexes are
fairly stable and have been isolated recently by us, fully
characterized by means of IR and NMR (1H, 13C, 31P)
spectroscopy and studied for their reactivity [12a]. Upon
heating, LnPdCl[C(O)NH-(CH2)x-OH] complexes can decompose
by generating the corresponding cyclic carbamate. However,
when heated in the presence of an excess of the amino alcohol
(and, therefore, under conditions closer to those used in the
catalytic process) the corresponding urea OC[NH-(CH2)x-OH]2
was formed (Scheme 2).
LnPdCl2 þ CO þ 2H2N-ðCH2Þx-OH þ NEt3
! LnPdCl½CðOÞNH-ðCH2Þx-OHꢅ þ NEt3ꢀHCl
(5)
3.2. Reaction mechanism and catalytic cycle
3.2.2. The oxidative step
It is well established [18] that transition metal promoted
oxidative carbonylation of alcohols or amines proceeds generally
through two basic steps. In the first step, carbonylation of substrate
is accompanied by reduction of the metal centre, whose oxidation
state decreases by two units (Eq. (2)). In the second step, the metal
centre is reoxidized (Eq. (3)) by the used oxidant. The starting
catalytic species is thus regenerated and can initiate a new
catalytic cycle.
To achieve a better understanding of the oxidative step and the
factors affecting it, the complex Pd3(CO)3(PPh3)3 was separated
from the urea and reacted with O2 under various conditions (see
Section 2.4.2).
In a few experiments, Pd3(CO)3(PPh3)3 was re-combined with
the mother liquor (see Section 2.4.2) and the resulting mixture was
reacted with O2 either in the absence or in the presence of added
phosphine (reaction (6); see Section 2.4.2, runs (a) and (b)). In other
related experiments, Pd3(CO)3(PPh3)3 was suspended in a CH3CN
solution of NEt3ꢀHCl and reacted with O2, in the presence of PPh3
(reaction (7); added PPh3/Pd ꢆ 1 mol/mol; see also Section 2.4.2,
runs (c) and (d)). In all the cases, reoxidation of Pd(0) (reactions (6)
and (7)) was very slow at room temperature. At higher
temperatures (313–333 K) it was relatively faster, but, after 6 h,
LnMðþnÞ þ CO þ 2HY ! COY2 þ LnMþðnꢂ2Þ þ 2Hþ
ðY ¼ OR; NHRÞ
(2)
1
Table 1 shows clearly that the catalytic system based on phosphine oxide OPPh3
is also effective (entry 18), but less efficient than the system based on PPh3 (entry 7).

P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
83
afforded PdCl2(PPh3)2 in moderate
mother liquorðPPh3þNEt3HClÞ;O2
Pd3ðCOÞ3ðPPh3Þ3
þ products
ꢂ!
PdCl2ðPPh3Þ2
(6)
O2
Pd3ðCOÞ3ðPPh3Þ3 þ 6NEt3HCl þ 3PPh3ꢂ!3PdCl2ðPPh3Þ2
þ products
(7)
yield (max 60%, isolated) as the complex Pd3(CO)3(PPh3)3 partially
decomposed to Pd-metal, which was hardly oxidized to Pd(II)
under the working conditions.
Prompt reoxidation of Pd(0)-species to Pd(II) is a factor of
crucial importance for the lifetime of the catalytic system and
efficiency (TOF) of the catalytic process and can be facilitated by
using
a suitable oxidation promoter able to speed up the
Scheme 3. Catalytic cycle in the absence of iodide promoters.
reoxidation reaction. To this end, stoichiometric amounts of
expensive oxidants as Cu(OAc)2 have been used [2e], which has
been shown to be effective under relatively mild conditions. Other
authors focused on the use of iodide as co-catalyst, but the catalytic
systems investigated so far were effective only under relatively
more drastic conditions [2c,19]. In the present study we have
shown that the use of NEt3ꢀHI as iodide source co-catalyst
promotes easy reoxidation of Pd(0) to Pd(II) by dioxygen and
allows the overall process to proceed effectively, under very mild
conditions. Accordingly, Pd3(CO)3(PPh3)3 readily converted into
PdI2(PPh3)2 (90–95% isolated yield) when reacted with O2 in the
presence of NEt3ꢀHI and phosphine (see Section 2.4.2 runs (e) and
(f)). Also reoxidation of Pd-metal (Eq. (8)) took place more easily in
the presence of the iodide salt and ligand. Reoxidation of Pd-black,
which can occur also in the absence of any ligand [2c,19] is of
particular relevance to catalysis, as it allows to individuate an
additional pathway through which, under catalytic conditions
(313–333 K; p(CO/O2) = 0.1 MPa), the catalycally active species can
be regenerated.
Scheme 4. Catalytic cycle in the presence of NEt3.HI.
to prevent aggregation of Pd particles and deactivation of catalytic
system.
In general, all those factors able to stabilize Pd(0)-species
prolong the lifetime of catalytic system and prevent its deactiva-
tion to Pd-metal. In this regard, we note that a substantial
improvement of efficiency of the overall catalytic process results
from employing N- or P-donor ancillary ligands (dipy, PN, PPh3),
which, by stabilizing low-valent Pd(0)-complexes, can restrain
their deactivation to less easily reoxidable Pd-metal or, also, assist
the reoxidation of Pd nanoparticles to Pd(II), as described in Eq. (8).
This may explain why the systems based on LnPdCl2 complexes
(L = dipy, PN, PPh3) were more effective than those based on
PdCl2(CH3CN)2 (Tables 1 and 3).
As is well documented in the literature [2c,8,9,19,21], in the
presence of iodide, dioxygen is not directly involved in reoxidation
of palladium, but the latter step implies the direct reaction of
Pd(0)-complexes or black palladium with molecular iodine
(Eqs. (9) and (10)), which generates in situ by oxidation of iodide
with O2 (Eq. (11)). Reaction (11) requires a source of protons. The
influence of acidity on oxidation of iodide provides an easy
rationale for the fact that, under our conditions, NEt3ꢀHI is a more
efficient co-catalyst than KI (compare, in Table 2, entries 2 and 4
with entries 6 and 7, respectively). In fact, differently from KI,
which can behave only as iodide source, NEt3ꢀHI may also act as
proton donor, making the protons required for iodide oxidation
more easily available in the reaction medium.
Pdð0Þ þ 2PPh3 þ 1=2O2 þ 2NEt3HCl ! PdCl2ðPPh3Þ2 þ H2O þ 2NEt3
(8)
For instance, when reaction (4) was carried out at 333 K, palladium
black formed during the reductive step (Eq. (4)) was rapidly
oxidized in quantitative yield only if NEt3ꢀHI was added to the
system before the reaction mixture reacted with O2. In the absence
of NEt3ꢀHI the reoxidation reaction was slower and incomplete
after a reaction time of 6 h. Moreover, if the mixture resulting from
the reductive step (Eq. (4)) was stirred for about 2 h under N2
before being reacted with O2, Pd-black formed during the process
became much more reluctant to oxidation, even in the presence of
NEt3ꢀHI. This fact may be ascribed to the tendency of colloidal
palladium formed during the reaction to undergo a passivation
process, which likely involves aggregation of nanoparticles of the
metal. Such a phenomenon is well described in the literature, and it
is known that some palladium based catalytic systems, in which
the active species are colloidal metal particles, exhibit enhanced
activity in the presence of protective agents, as polyvinylalcohols
or surfactants, able to protect palladium metal particles against
aggregation [20].
Scheme 3 summarizes the behaviour of the reacting system in
the absence of the iodide promoter. Under the latter conditions,
catalysis is poor as decomposition of Pd(0)-complexes takes place
faster than the reoxidation process, which is slow in the absence of
NEt3ꢀHI. The consequent increase of concentration of metal
nanoparticles in the reaction medium facilitates their aggregation
and causes the fast deactivation of catalytic system. In the presence
of NEt3ꢀHI (Scheme 4) the oxidative step is faster. This fact
maintains the concentration of metal nanoparticles at low levels,
LnPd þ I2 ! LnPdðIIÞ þ 2Iꢂ
(9)
(10)
(11)
Pdð0Þ þ Ln þ I2 ! LnPdðIIÞ þ 2Iꢂ
2Iꢂ þ 2Hþ þ 1=2O2 ! I2 þ H2O

84
P. Giannoccaro et al. / Applied Catalysis A: General 375 (2010) 78–84
[2] (a) F.W. Li, X.G. Peng, C.G. Xia, B. Hu, Chin. J. Chem. 23 (2005) 643–645;
Although the data reported above individuate in the discussed
(b) I. Chiarotto, M. Feroci, Tetrahedron Lett. 42 (2001) 3451–3453;
(c) B. Gabriele, R. Mancuso, G. Salerno, M. Costa, J. Org. Chem. 68 (2003) 601–604;
(d) B. Gabriele, G. Salerno, D. Brindisi, M. Costa, G.P. Chiusoli, Org. Lett. 2 (2000)
625–627;
(e) 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–
5958.
mechanism a very plausible reaction pathway, we also note that
iodine formed in the reaction mixture may open an additional way
to the formation of the target product and the regeneration of the
Pd(II)-catalyst (Scheme 4 routes 2–3). Elsewhere [11,12], in fact,
we have demonstrated that iodine can easily react with Pd-
carbamoyl complexes. This reaction can regenerate the starting
Pd-catalyst and afford a reactive carbamoyl iodide (Eq. (12)), which
can be converted into urea by reaction in situ with the substrate
(Eq. (13)).
[3] (a) F.W. Li, C.G. Xia, J. Catal. 227 (2004) 542–546;
(b) J.M. Liu, X.G. Peng, J.H. Liu, S.Z. Zheng, W. Sun, C.G. Xia, Tetrahedron Lett. 48
(2007) 929–932.
[4] (a) N.A. Puschin, R.V. Mitic, Justus Liebigs Ann. Chem. 532 (1937) 300–303;
(b) A.F. Hegarty, L.J. Drenman, in: A.R. Katritzky, O. Meth-Cohn, C.W. Rees (Eds.),
Comprehensive Organic Functional Group Transformation, vol. 6, Pergamon,
Oxford, UK, 1998, pp. 499–526;
(c) T.P. Vishnyavova, I.A. Golubeva, E.V. Glebova, Russ. Chem. Rev. (Engl. Transl.)
54 (1985) 249–261;
(d) J.E. McClusker, A.D. Main, K.S. Johnson, C.A. Grasso, L. McElwee-White, J. Org.
Chem. 65 (2000) 5216–5222.
PdðPPh3Þ Cl½CðOÞNH-ðCH2Þx-OHꢅ þ I2
! Pdð2PPh Þ ClðIÞ þ ICðOÞNH-ðCH Þx-OH
(12)
(13)
3
2
2
[5] M.K. Leung, J.L. Lai, K.H. Lau, H.H. Yu, H.J. Hsiao, J. Org. Chem. 61 (1996) 4175–
4179.
[6] (a) J.R. Gage, D.A. Evan, Org. Synth. 68 (1990) 77–82;
(b) D.J. Ager, I. Prakash, D.R. Schaad, Chem. Rev. 96 (1996) 835–875 (and
references therein);
ICðOÞNH-ðCH2Þx-OH þ H2N-ðCH2Þx-OH
! OCðNH-ðCH2Þx-OHÞ2 þ HI
(c) M. Feroci, A. Gennaro, A. Inesi, M. Orsini, L. Palombi, Tetrahedron Lett. 43
(2002) 5863–5865.
[7] D.J. Diaz, K.G. Hylton, L. McElwee-White, J. Org. Chem. 71 (2006) 734–738.
[8] P. Giannoccaro, C.F. Nobile, M. Latronico, Inorg. Chim. Acta 175 (1990) 133–139.
[9] (a) P. Giannoccaro, N. Ravasio, M. Aresta, J. Organomet. Chem. 451 (1993) 243–
248;
(b) P. Giannoccaro, J. Organomet. Chem. 470 (1994) 249–252.
[10] P. Giannoccaro, D. Cornacchia, S. Doronzo, E. Mesto, E. Quaranta, M. Aresta,
Organometallics 25 (2006) 2872–2879.
[11] (a) P. Giannoccaro, I. Tommasi, M. Aresta, J. Organomet. Chem. 476 (1994) 13–18;
(b) M. Aresta, P. Giannoccaro, I. Tommasi, Educ. Adv. Chem. 3 (1996) 249–255;
(c) M. Aresta, P. Giannoccaro, I. Tommasi, A. Dibenedetto, A.M. Manotti, F.
Ugozzoli, Organometallics 19 (2000) 3879–3889.
[12] (a) P. Giannoccaro, A. Dibenedetto, M. Gargano, E. Quaranta, M. Aresta, Organo-
metallics 27 (2008) 967–975;
(b) P. Giannoccaro, S. Doronzo, M. Gargano, P. Masiello, E. Quaranta, M. Aresta,
Europacat VIII, 26–31 August, 2007, Turku-Finland, pp. 5–30.
[13] E. Uhlig, Z. Keiser, Anorg. Chem. 406 (1974) 1–6.
[14] J. Chatt, L.H. Venanzi, J. Chem. Soc. (1957) 2351–2355.
[15] B.J. McCormick, E.N. Jaynes Jr., R.I. Caplan, H.C. Clarck, J.D. Ruddick, Inorg. Synth.
13 (1971) 216–218.
4. Conclusions
a,v-Amino alcohols have been effectively and selectively
carbonylated, in CH3CN, to symmetrical N,N0-bis(hydroxyalkyl)ur-
eas using Pd(II)/L/NEt3/HI (L = dipy, PN, PPh3, CH3CN) catalytic
systems and molecular oxygen as the oxidant. Using NEt3ꢀHI as co-
catalyst acting both as iodide source and proton donor formation of
urea proceeds efficiently under very mild conditions (303–333 K;
p(CO/O2) = 0.1 MPa). The iodide source, as well as the nature of
ancillary ligands, is of crucial importance for lifetime of catalytic
system and the efficiency of the catalytic process. The best catalytic
activities were observed using Pd-systems stabilized by mono- or
bidentate N- or P-donor ligands and employing an I/Pd molar ratio
close to 8.
The urea can be isolated in a very straightforward way and the
catalytic system can be easily recovered and recycled for several
times without showing any significant decrease in activity
(L = dipy, PN, PPh3).
A few mechanistic features of the process have also been
investigated. Formation of urea involves the intermediacy of
LnPdCl[C(O)NH-(CH2)x-OH] carbamoyl complexes. This step can
also generate Pd(0)-species, which are easily reoxidized to Pd(II) in
the presence of NEt3ꢀHI and O2.
[16] B.B. Wayland, R.F. Schramm, Inorg. Chem. 8 (1969) 971–976.
[17] M. Hidai, M. Kokura, Y. Ochida, J. Organomet. Chem. 52 (1973) 431–435.
[18] H.M. Colquhoun, D.J. Thompson, M.V. Twigg, Carbonylation: Direct Synthesis of
Carbonyl Compounds, Plenum Press, New York, 1991 (Chapter 2, p. 30).
[19] (a) B. Gabriele, M. Costa, G. Salerno, G. Chiusoli, J. Chem. Soc., Perkin Trans. 1
(1994) 83–87;
(b) B. Gabriele, G. Salerno, in: D. Crich (Ed.), PdI2 in e-EROS (Electronic Encyclo-
pedia of Reagents for Organic Synthesis), Wiley-Interscience, 2006, pp. 1–13;
(c) B. Gabriele, M. Costa, G. Salerno, G. Chiusoli, Curr. Org. Chem. 8 (2004) 919;
(d) B. Gabriele, M. Costa, G. Salerno, Synlett 14 (2004) 2468–2483;
(e) B. Gabriele, G. Salerno, Top. Organomet. Chem. 18 (2006) 239–272;
(f) B. Gabriele, G. Salerno, M. Costa, G.P. Chiusoli, J. Organomet. Chem. 687 (2003)
219–228;
Acknowledgments
(g) B. Gabriele, R. Mancuso, G. Salerno, M. Costa, Chem. Commun. 4 (2003) 486–
487;
(h) B. Gabriele, G. Salerno, R. Mancuso, M. Costa, J. Org. Chem. 69 (2004) 4741–
4750;
(i) D.J. Dı´az, A.K. Darko, L. McElwee-White, Eur. J. Org. Chem. 27 (2007) 4453–
4465.
The authors wish to thank MIUR (contract 2006031888) and
University of Bari for financial support.
[20] (a) X. Sun, X. Jiang, S. Dong, E. Wang, Macromol. Rapid Commun. 24 (2003) 1024–
1028;
Appendix A. Supplementary data
(b) T. Douglas, M. Young, Nature 393 (1998) 152–155;
(c) F.C. Meldrum, B.R. Heywood, S. Mann, Science 257 (1992) 522–523;
(d) S.-H. Yu, M. Antonietti, H. Co¨lfen, M. Giersig, Angew. Chem. Int. Ed. Engl. 41
(2002) 2356–2359;
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apcata.2009.12.022.
(e) L.I. Halaoui, Langmuir 17 (2001) 7130–7136;
(f) Q.F. Zhou, J.C. Bao, Z. Xu, J. Mater. Chem. 12 (2002) 384–387;
(g) L. Longenberger, G. Mills, J. Phys. Chem. 99 (1995) 475–478;
(h) S. Mann, J.P. Hannington, R.J.P. Williams, Nature 324 (1986) 565–567;
(i) M.P. Peleni, J. Tanori, A. Felakembo, J.C. Dedieu, T. Gulik-Krzywicki, Langmuir
14 (1998) 7359–7363.
References
[1] (a) M.R. Barbachyn, C.W. Ford, Angew. Chem. Int. Ed. 42 (2003) 2010–2013;
(b) C.M. Perry, B. Javis, Drugs 61 (2001) 525–551;
[21] (a) P. Giannoccaro, M. Aresta, S. Doronzo, C. Ferragina, Appl. Organomet. Chem.
14 (2000) 581–589;
(c) R. Norrby, Exp. Opin. Pharmacother. 2 (2001) 293–302;
(d) D. Clemmet, A. Markhan, Drugs 59 (2000) 815–827;
(e) T.A. Muktar, G.D. Wright, Chem. Rev. 105 (2005) 529–542;
(f) J.M. Cassady, K.K. Chan, E.G. Floss, E. Leistner, Chem. Pharm. Bull. 52 (2004)
1–26.
(b) P. Giannoccaro, S. Doronzo, C. Ferragina, in: H.U. Blaser, A. Baiker, R. Prins
(Eds.), Homogeneous Catalysis and Fine Chemicals IV, Elsevier Science B.V., 1997,
pp. 633–640.