The solid state structure and reactivity of NbCl5·(N, N′-dicyclohexylurea) in solution: Evidence for co-ordinated urea dehydration to the relevant carbodiimide

Article abstract of DOI:10.1039/c001669a

NbCl₅·(N,N′-dicyclohexylurea) 1a owns a distorted octahedral structure due to intramolecular NH...Cl bonding. The unit cell contains four units which are intermolecularly NH...Cl and NH...N bonded. An extended intramolecular network of H-bonding (N-H...Cl, CH...Cl, CH...N) causes the 3D self assembling of the units. Upon addition of base, the HCl release from 1a is observed with the transfer to Nb of the O-atom of the carbonylic function of the starting urea which is converted into the relevant carbodiimide CyNCNCy 4. The latter is quantitatively released by adding an excess of NEt₃ at 308 K (py and DBU are less efficient) with formation of the known NbOCl₃(NEt₃)₂, isolated in quantitative yield. Increasing the temperature leads to a loss in selectivity as the formed DCC undergoes further reactions. At 350 K, the isocyanate CyNCO has been isolated in 60% yield besides a mixture of Nb-complexes. DFT calculations have been coupled to IR and NMR experiments for characterizing possible reaction intermediates and the behaviour of 1a. Several other MCl_x species (ScCl₃,YCl₃,LaCl ₃,TiCl₄, TaCl₅, AlCl₃, SnCl ₄) have been shown to be able to co-ordinate DCU but not all of them promote the conversion of urea into DCC. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c001669a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
The solid state structure and reactivity of NbCl5·(N,N¢-dicyclohexylurea) in
solution: evidence for co-ordinated urea dehydration to the relevant
carbodiimide†‡
Michele Aresta,*a Angela Dibenedetto,a Paolo Stufano,a Brunella Maria Aresta,b Sabino Maggi,b Imre Pa´pai,c
Tibor Andra´s Rokobc and Bartolo Gabrieled
Received 28th January 2010, Accepted 7th May 2010
First published as an Advance Article on the web 14th June 2010
DOI: 10.1039/c001669a
NbCl5·(N,N¢-dicyclohexylurea) 1a owns a distorted octahedral structure due to intramolecular
NH ◊ ◊ ◊ Cl bonding. The unit cell contains four units which are intermolecularly NH ◊ ◊ ◊ Cl and NH ◊ ◊ ◊ N
bonded. An extended intramolecular network of H-bonding (N–H ◊ ◊ ◊ Cl, CH ◊ ◊ ◊ Cl, CH ◊ ◊ ◊ N) causes
the 3D self assembling of the units. Upon addition of base, the HCl release from 1a is observed with the
transfer to Nb of the O-atom of the carbonylic function of the starting urea which is converted into the
= =
relevant carbodiimide CyN C NCy 4. The latter is quantitatively released by adding an excess of
NEt3 at 308 K (py and DBU are less efficient) with formation of the known NbOCl3(NEt3)2, isolated in
quantitative yield. Increasing the temperature leads to a loss in selectivity as the formed DCC
= =
undergoes further reactions. At 350 K, the isocyanate CyN C O has been isolated in 60% yield
besides a mixture of Nb-complexes. DFT calculations have been coupled to IR and NMR experiments
for characterizing possible reaction intermediates and the behaviour of 1a. Several other MClx species
(ScCl3,YCl3,LaCl3,TiCl4, TaCl5, AlCl3, SnCl4) have been shown to be able to co-ordinate DCU but not
all of them promote the conversion of urea into DCC.
Introduction
We have recently used DCC 4 as promoter (eqn (1)) of the direct
carboxylation of alcohols to afford dialkylcarbonates.1
= =
2ROH + CO2 + CyN C NCy → (RO)2CO
+ CyHN–C(O)–NHCy
(1)
R = Me, Et, allyl
Such reaction has high selectivity (>98%) and occurs under
milder conditions (330 K and 0.2 MPa of CO2) when compared
to the transition metal catalyzed carboxylation of alcohols2-5 that
requires 410–470 K and 1–30 MPa of CO2.
Fig. 1 shows that if DCU is converted back into DCC it
would be possible to run a cyclic two-step catalytic carboxylation
of alcohols using DCC, a technique useful for the production
of either niche carbonates or carbonates bearing temperature
sensitive substituents.
Fig. 1 DCC as promoter of the direct carboxylation of alcohols and the
recycling of the formed DCU.
Removal of water from syn-dialkyl-ureas is reported in the
literature to occur with the use of toxic reagents such as phosgene6-9
or PCl57 or SOCl28 (eqn (2a)–(c)) with low selectivity and a complex
work-up of the reacting mixture.
aDepartment of Chemistry, University of Bari, Campus Universitario,
Bari and CIRCC, via Celso Ulpiani 27, Bari, Italy. E-mail: m.aresta@
chimica.uniba.it; Fax: (+)390805443606
bCNR-IC Institute of Crystallography, Via Amendola 122/O, 70126, Bari,
IT
CyHN–C(O)–NHCy + COCl2 → CO2 + 2HCl
= =
+ CyN C NCy
(2a)
(2b)
(2c)
cChemical Research Center, Hungarian Academy of Sciences, H-1525,
Budapest, P.O.B. 17, Hungary
dDepartment of Pharmaceutical Sciences, University of Calabria, 87036,
Arcavacata di Rende (CS), Italy
CyHN–C(O)–NHCy + PCl5 → POCl3 + 2HCl
= =
+ CyN C NCy
† Electronic supplementary information (ESI) available: The asymmetric
cell unit of 1a (Fig. S1); representation of the unit cell of 1a (Fig. S2);
CyHN–C(O)–NHCy + SOCl2 → SO2Cl2 + 2HCl
˚
3D architecture of 1a (Fig. S3); D–H ◊ ◊ ◊ A distances (A) in the complex
= =
+ CyN C NCy
NbCl5·DCU 1a and the angles of the set of atoms (Table S1). CCDC
reference number 743003. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c001669a
In order to find more efficient, cleaner and safer alternative
routes, we decided to investigate whether coordination of DCU
‡ Dedicated to Prof. Antonio Tiripicchio.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6985–6992 | 6985
©

View Article Online
Table 1 Infrared n(CO) frequency and 13C NMR chemical shift for C
O
=
to (transition) metal systems would promote its dehydration to
DCC.
of various 1 : 1 metal-halides adducts with DCU
Ureas may act as ligands to transition metal systems10 through
1
1
2
the O- or the N-atoms: the h -O, h -N and h -N,N¢ modes
of bonding11 have been demonstrated. “Hard” acids prefer a
coordination at the oxygen atom, while more “soft” acids prefer N
as a donor. Metal complexes containing 2, 4, 6 molecules of urea
itself,10,12 either O- or N-bonded have been described and a rare
Product of the reaction
of MClx·DCU with
MClx
vC= cm-1 dC=
excess triethylamine
O
O
Free DCU 1627
155.5
ScCl3
YCl3
1610
1614
1601
1603
1615
1620
1614
1599
158.2 Free DCU + DCC traces
158.1 Free DCU
157.4 Free DCU
158.9 DCC
158.5 DCC
158.4 DCC
2
example of a chelate h -N,N¢ mode of bonding has recently been
LaCl3
TiCl4
NbCl5
TaCl5
AlCl3
SnCl4
described.13 Metal complexes in which urea bridges two metal
centres have also been characterized14 and substituted ureas or
thioureas have intensively been studied.15-17 The renovated interest
in metal complexes (MnII, CoII, NiII, CuII, ZnII, lanthanides) with
N,N¢-dimethylurea-DMU, O-bonded to the metal, was due to their
ability to form supramolecular 2D structures.18
157.8 Free DCU + DCC traces
157.6 Free DCU
In the adduct of NbCl5 with DCU, the IR spectrum of solid 1
Despite such abundance of cases, very rare studies on the
reactivity of co-ordinated ureas are available in the literature:
Ti-urea complexes have been thermally decomposed to afford
nanoparticles of TiO2,19 while NiII-urea complexes have been
reported to show an urease-like action.20 Reactions of the co-
ordinated ureas selectively involving the carbonylic function and
the N–H bonds have never been described.
Very recently, the coordination of tetramethylureas (TMUs) to
Group 4 (Ti) and 5 (Nb, Ta) metals in their highest oxidation
state was for the first time investigated and the relevant complexes
characterized by XRD.21 TMU is reported to be O-bonded to
the metal centers with no distortion around the Nb–O–C bond
affording a symmetric molecule. Intra- or intermolecular H-bonds
were not evident. Noteworthy, N,N¢-dicyclohexylurea has never
been used so far as a ligand.
-1
in Nujol mull shows a n(C O) at 1615 cm shifted by 12 cm-1
=
towards lower energy with respect to the free urea: this supports
the O-coordination to Nb. The n(N–H) region shows two bands
at 3325 and 3290 cm-1, which infer an involvement of the aminic
hydrogen in H-bonding. This feature was confirmed by DFT
calculations, which indicate that the most stable form of the
NbCl5(DCU) complex is characterized by intramolecular N–
H ◊ ◊ ◊ Cl type H-bonding (see 1a in Fig. 2). The metal–ligand
interaction in NbCl5(DCU) is rather strong as the calculations
predict DE = -30.3 kcal mol-1 for the gas-phase binding free
energy and DGsolv = -12.1 kcal mol-1 for the solvent-corrected
Gibbs free energy of complex formation. The stabilizing effect of
the N–H ◊ ◊ ◊ Cl bonding is clearly borne out by a comparison with
a structural isomer without intramolecular H-bonding (see 1b in
Fig. 2), which is found to be 3.0 kcal mol-1 less stable than 1a.
In this paper we report the synthesis, characterization and reac-
tivity of several metal-halide complexes with DCU. Among them,
we detail the NbCl5(N,N¢-dicyclohexylurea) complex, describe its
structure, discuss the unique mode of bonding of urea to Nb and
show the presence of H-bonding in solid state which play a key
role in: (i) determining the primary structure of the complex; (ii)
promoting the association of four molecules in the unit cell; (iii)
driving the supramolecular organization of the complex into an
extended 3D architecture that influences the physical properties
of the solid. We also show that among the metal complexes
investigated in this work, Group 4 (Ti) and 5 (Nb, Ta) metal
halides22,23 in their highest oxidation state, perform quite easily
the dehydration reaction of DCU to afford DCC under very mild
conditions (300 K). In particular, NbCl5 shows high conversion
yield (> 98%) and selectivity (100%).
Fig. 2 Equilibrium structures of two NbCl5(DCU) isomers obtained from
DFT calculations. Selected bond lengths are given in A.
˚
˚
The calculated H ◊ ◊ ◊ Cl distance (2.42 A) is well within the value
of “strong or classic” H-bonding.24 Structure 1a makes the two
1
cyclohexyl rings non-equivalent. Instead, the H and 13C NMR
spectra in CD2Cl2 show their equivalence, that can be explained
invoking the fluxionality of 1 in solution at 298 K.
Results and discussion
Already such features make complex 1 quite unique and differ-
ent from NbCl5·[(Me2N)2CO]12 in which the urea is symmetrically
bonded to the metal so that its molecular axis corresponds to
the axis of the whole complex. A peculiarity of solid 1a is that
once crystallized it is scarcely soluble in the mother solution, that
suggests a supramolecular architecture, either 2D or 3D. The single
crystal XRD (Fig. 3) confirms that in the solid state, the DCU
molecule is asymmetrically bonded to Nb.
Transition metal halides with the metal (either transition metal or
not) in the highest oxidation state afford solid 1 : 1 adducts with
DCU. As it is shown in Table 1, the highest shift (28 cm-1) of
the n(CO) upon coordination is observed with SnCl4, the lowest
(7 cm-1) with TaCl5. Such shift is not diagnostic of the readiness of
the metal-complexes to undergo further reactivity of the adducts,
as we shall discuss later-on. A detailed analysis of the IR spectra
has shown in all cases the insurgence of multiple bands in the
region characteristic of N–H stretching, supporting a H-bonding
formation.
The asymmetric cell unit is depicted in Fig. S1 (see ESI†) in
which intramolecular H-bonding is represented with continuous
black lines, while intermolecular H-bonding is represented by red
6986 | Dalton Trans., 2010, 39, 6985–6992
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
In a “D–H-–A” system (D-donor, A-acceptor), “classic or
strong” H-bonds are formed when both D and A have a high
electronegativity, such as in N–H ◊ ◊ ◊ X or O–H ◊ ◊ ◊ X where X
is N, O, F, Cl.24a On such basis, the N(2A)–H(2A1) ◊ ◊ ◊ Cl(4A)
(Table 3) is a “classic or strong” H-bonding. Besides such
interactions, we have observed other H-bonding in which C–H
bonds are involved that are characterized by a scarce polarization,
called “weak or non-classic” H-bonding interactions.24b,25,26 Such
weak interactions, for which the most likely distance D–A is
usually calculated on the basis of a statistic analysis of data
contained in Cambridge Structural Database-CSD,27 often play a
key role in the stabilization of some peculiar molecular geometries
or in the building-up of extended supramolecular 2D or 3D
architectures.28,29 In NbCl5·DCU they determine (Table 3) the
3D architecture (Fig. S3 in ESI†) and the physical properties:
as mentioned above, the solid complex is very scarcely soluble in
the solvent from which it is crystallized.
◦
˚
Table 2 Selected bond lengths (A) and angles ( ) for the A molecule
NbCl5·DCU in the unit cell
Bond lengths
Nb/A–Cl(1A)
Nb/A–Cl(3A)
Nb/A–Cl(5A)
O/A–C(1A)
2.333(3)
2.342(3)
2.280(3)
1.321(9)
1.294(10)
Nb/A–Cl(2A)
Nb/A–Cl(4A)
Nb/A–O(A)
C(1A)–N(1A)
2.351(3)
2.372(3)
2.007(5)
1.327(10)
C(1A)–N(2A)
Bond angles
Nb/A–O(A)–C(1A)
OA–C(1A)–N(2A)
139.96(5)
119.07(8)
O/A–C(1A)–N(1A)
N(1A)–C(1A)–N(2A)
118.56(7)
122.36(8)
Table S1 (ESI†) reports the D–A distance and the D–H ◊ ◊ ◊ A
angles which confirm the existence of weak H-bonding.27b,c In fact,
the directional properties of the C–H ◊ ◊ ◊ X set of atoms is a key
parameter in distinguishing a van der Waals interaction (such as
in R3C–H ◊ ◊ ◊ H–CR3, which is linear without any preferential di-
rection) from a weak hydrogen bonding which may have a marked
direction.27a The “metal-assisted hydrogen bonding” effect27a,30,31
may play a key role in the assembling of the supramolecular
structure27b and the Nb–Cl system has, indeed, good acceptor
properties.32 Noteworthy, four out of the five chlorine atoms
bonded to Nb are involved in H-bonding. The fifth (notably,
Cl5 in molecule 1a in Fig. 3) is not involved and shows the
˚
shortest of the Nb–Cl distances (2.28 A, compared to 2.33–2.37
Fig. 3 Two molecules of 1a in the asymmetric unit. The thermal ellipsoids
have a 30% probability level; H atoms are represented by circles of arbitrary
radius. Intramolecular H-bonding is represented with black lines.
of the other four Nb–Cl bonds). It is worth pointing out that
the extended 3D association of 1a units is undoubtedly driven by
H-bonding and not by dipole-coupling forces. In fact, although
the structure of NbCl5(DCU) is predicted by DFT calculations
to have a relatively large dipole moment (10.7 D) aligned with
the Cl–Nb ◊ ◊ ◊ O molecular axis, the calculations carried out for a
1a ◊ ◊ ◊ 1a dimer cut out from the crystal structure and involving
an antiparallel arrangement of the monomers indicate only a very
small interaction energy between the two monomers (-2.5 kcal
mol-1).
NbCl5·DCU represents, thus, a unique case, as in other urea-
metal-complexes the strong N–H ◊ ◊ ◊ Cl interactions were respon-
sible for building at maximum a 2D structure9 while in the recently
described Nb-tetralkyl-urea complexes12 H-bonding implying C–
H ◊ ◊ ◊ N bonds were not identified at all. The strong intramolecular
H-bonding may also play an important role in determining the
unusual observed reactivity of 1a. In fact, when 1a was treated with
dotted lines. Fig. S2 (see ESI†) shows the unit cell of 1a which
crystallizes in the P21 spatial group with both cyclohexyl moieties
having a “chair” conformation.
Table 2 summarizes the most important structural data. Crys-
tallographic data are reported in Table 4 (Experimental).
The coordination of the carbonylic oxygen to Nb is coherent
with the electron density at the oxygen with a NbOC angle of
140◦: oxygen is using one of its lone-pair for coordination. The
structure also shows the six-atom –C–O-–Nb–Cl-–H–N– cyclic
structure forced by the intramolecular NH ◊ ◊ ◊ Cl bond. The fine
analysis of crystal data has permitted the identification of two
more intramolecular H-bonding in 1a. In particular, N–H ◊ ◊ ◊ Cl
and CH ◊ ◊ ◊ N bonds of two different intensities are clearly evident
(Table 3).
◦
˚
Table 3 Intramolecular H-bonding (A) and D–H ◊ ◊ ◊ A angles ( ) in 1b. Weaker CH ◊ ◊ ◊ N bonds are in italics
Donor-H ◊ ◊ ◊ Acceptor
D–H
H ◊ ◊ ◊ Acceptor
Donor ◊ ◊ ◊ Acceptor
Donor-H ◊ ◊ ◊ Acceptor
N(2A)–H(2A1) ◊ ◊ ◊ Cl(4A)(0)
C(8A)–H(8A) ◊ ◊ ◊ N(1A)(0)
C(9A)–H(9A2) ◊ ◊ ◊ N(1A)(0)
N(2B)–H(2B1) ◊ ◊ ◊ Cl(4B)(0)
C(8B)–H(8B) ◊ ◊ ◊ N(1B)(0)
C(13B)–H(13C) ◊ ◊ ◊ N(1B)(0)
0.860(9)
0.980(7)
0.970(11)
0.860(8)
0.980(6)
0.970(.13)
2.473(3)
2.734(6)
2.956(.7)
2.570(3)
2.677(7)
2.961(.8)
3.260(9)
152.58(45)
91.66(41)
112.07(79)
149.43(29)
92.59(38)
110.90(80)
2.931(11)
3.440(140)
3.339(5)
2.892(11)
3.429(170)
Symmetry code: (0) x, y, z.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6985–6992 | 6987
©

View Article Online
an excess of base at 300 K, we observed the immediate formation
of the dehydrated form of the urea, i.e. dicyclohexylcarbodiimide,
and of NbOCl3(base)2 (Scheme 1).
Fig. 4 DFT calculated structures for intermediate 2 obtained upon single
deprotonation of complex 1a.
Scheme
1
Step-by-step conversion of CyHNC(O)NHCy into
=
=
CyN
C NCy promoted by NbCl5 and a base by progressive loss
of HCl and transfer of the O-atom from C– to –Nb. Compounds 2a was
identified by multinuclear NMR.
We have observed that the yield of the reaction was dependent
on the base used (Py < DBU < NEt3) while the selectivity was
governed by the temperature. In fact, above 325 K the reaction
afforded a mixture of Nb-compounds and cyclohexylisocyanate
(nNCO at 2260 cm-1) that was determined by GC-MS (eqn (3)).
Fig. 5 The effect of addition of NEt3 to a solution of 1a in CH2Cl2.
Black: IR spectrum of starting compound 1a: the band at 1615 cm-1 is
=
due to the n(C O) of the co-ordinated urea. Red: The sample above after
addition of two mol of base per mol NbCl5·DCU. The band at 1653 cm-1
is most probably due to the carbodiimide co-ordinated to NbOCl3. Blue:
The above sample after the addition of an excess of NEt3. The band at
= =
= =
NbOCl3 + CyN C NCy → Nb-species + CyN C O
(3)
Such reaction is reminiscent of the first step of the metathesis of
carbodiimides promoted by oxo-complexes of vanadium(V) which
have been proposed to co-ordinate the cumulene and form an
2117 cm-1 is due to free CyN
C NCy.
=
=
Intermediate 2a can be represented as the resonance hybrid
of three different possible structures (Scheme 3). The chemical
equivalence of the cyclohexyl groups in the 13C NMR spectra can
be, thus, explained by the fluxionality of the complex, with a rapid
exchange of the two N atoms which co-ordinate to Nb and bind
the proton. The addition of a second equivalent of base afforded
a further change of the IR spectrum (Fig. 5) with formation of
imido-V complex and isocyanate.33 The V NR moiety would then
=
= =
promote the metathesis with a second carbodiimide R¢N C NR¢
= =
=
to afford RN C NR¢ and the V NR¢ moiety (Scheme 2).
putative 3a (band at 1653 cm-1 due to the C N stretching).
=
Scheme 2 Proposed mechanism for carbodiimide metathesis.
So far, we were not able to isolate the imido complex
=
CyN NbCl3 as a pure compound as it was contaminated with
other Nb-species difficult to separate.
The dehydration of DCU 300 K was followed by IR (Fig. 5)
or NMR. Intermediate 2a (Scheme 1) was obtained by addition
of stoichiometric amounts (1 : 1) of the base to complex 1a with
mono-deprotonation of DCU. It was characterised in situ by IR
and NMR. The formation of compound 2a caused the insurgence
of an intense IR absorption at 1651 cm-1; the 13C NMR of
2a shows two equivalent cyclohexyl groups (see Experimental).
DFT calculations suggest that the structure of such intermediate
corresponds to a 4-membered ring involving the metal center and
Scheme 3 (a) Resonance structures; and (b) fluxionality equilibrium for
intermediate 2a.
=
the NC O moiety of the deprotonated DCU (2a in Fig. 4).
After an excess of base was added, free DCC was formed (band
at 2117 cm-1) and isolated. The nature of 3a was inferred on the
basis of the fact that the same IR spectrum was observed when
an authentic sample of NbOCl3 was reacted with an equimolar
In 2a, Nb reaches a hexa-coordinate configuration, which is
more stable by 21.8 kcal mol-1 than the penta-coordinated form
2b (Fig. 4).
6988 | Dalton Trans., 2010, 39, 6985–6992
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
amount of DCC (Fig. 6). Most interestingly, the 13C NMR of
putative 3a in solution shows a net change of the resonance of the
most reactive Nb(V) > Ti(IV), Ta(V) species is some how parallel
to the low-field shift of the 13C-resonance.
= =
original N C N carbon which is shifted from 140.0 ppm in the
free DCC to 158.8 in the co-ordinated DCC. Such value recalls
very close the resonance of the urea indicating a C–O interaction.
Experimental
General
All starting reagents and solvents were commercial products
purchased by Aldrich. Solvents were dried, distilled,34 and stored
under dinitrogen. All manipulations were carried out under a dry
dinitrogen atmosphere, due to the high sensitivity of NbCl5 to
moisture.
Nuclear magnetic resonance experiments were carried out with
a 600 MHz Bruker Avance II Ultrashield Plus instrument using
deuterated solvents. Infrared spectra were recorded with a FTIR
Shimadzu Prestige 21 instrument. GC-MS analyses were carried
out with a gas chromatograph Shimadzu 17 A (capillary column:
30 m; Agilent 19091J-433E HP5 5% phenyl methyl siloxane; Ø
0.25 mm, 0.25 mm film) coupled to a Shimadzu QP5050 A mass
spectrometer.
Quantitative determinations on the reaction solutions were
recorded using a Hewlett-Packard 6850 GC-FID (capillary col-
umn: 30 m; Phenomenex ZB-Wax; Ø 0.25 mm, 0.25 mm film).
Elemental analyses on Nb were conducted colorimetrically35
using a Shimadzu UV-1601 spectrophotometer. Other elements
were determined using Atomic Absorption Spectrophotometer
Shimadzu AA-6200.
Fig. 6 Comparison of the IR spectra of putative compound 3a obtained
from: (black) NbCl5·DCU + 2NEt3; (red) the direct reaction of an authentic
=
=
sample of NbOCl3 with CyN
C NCy.
Attempts to grow a mono-crystal of 3a for its XRD characteri-
zation (that would allow the definition of the mode of bonding of
DCC to NbOCl3) failed so far as a fast precipitation produced
nanocrystals not suited for XRD determination while a long
permanence of 3a in solution caused its conversion into a mixture
of products. This was demonstrated by the IR spectrum of the
isolated solid which shows the appearance of several new bands
below 1650 cm-1. Also, the powder-method did not give a good
resolution of the structure.
The three bases used in this work were selected for their different
properties and showed a different activity, as expected (py <
DBU < NEt3). With NEt3, pure DCC was isolated from the
reacting mixture with a yield of 98–99% and a selectivity of 100%
at 300 K. DBU allowed to isolate DCC with a maximum yield of
70% and pyridine was the worst with a yield of only 50%. Such
different behaviour can be explained in terms of both the proton-
affinity and the coordination ability to NbOCl3 of the three bases.
Both such functions play a key role in the elimination of H–Cl and
in the fast de-coordination from the metal of the formed DCC that
prevents its further conversion. NEt3 combines the best capacity
in the two reactions, maximizing, thus, the yield.
X-Ray data collection, structure solution, and refinement†
The structure of 1a was determined by X-ray single-crystal
diffraction data from a crystal mounted on a glass fibre. Because
of the instability of the compound under ambient conditions, the
crystalline sample was covered with a thin layer of glaze, to protect
the compound for the time of experiments.
Single crystal X-ray data were collected on a Bruker-Nonius
KappaCCD diffractometer equipped with a CCD area detector,
a 0.35 mm capillary collimator optic (important to improve
intensity of Mo radiation and to reduce the beam divergence),
˚
using Mo radiation (lKa1 = 0.71073 A) and an adjustable graphite
monochromator.
The structure was solved by direct methods using SIR200836
and refined by full-matrix least-squares on F2 with SHELXL-97,37
first with isotropic thermal parameters and then with anisotropic
thermal parameters in the last cycles of refinement for all the non-
hydrogen atoms.
It is worth emphasizing that among the metal halides-DCU
complexes we have synthesized so far (Table 1), only the Nb
derivative promotes the quantitative conversion of DCU into DCC
(and NbOCl3) upon treatment with NEt3. The Ti(IV) and Ta(V)
derivatives show an interesting but lower yield (> 90%), while
the Al(III) and Sc(III) complexes afford only a limited conversion.
Other adducts undergo a ligand substitution reaction (eqn (4))
more than a dehydration of the co-ordinated DCU.
All H atoms were introduced into the geometrically calculated
positions and then treated as riding-model in idealized positions
on the corresponding parent atoms, with bond lengths C–H =
˚
˚
0.97 A for methylene H-atoms and N–H = 0.86 A, in spite
of several of them were observed in a difference Fourier map.
The H atoms of each group were assigned the same Uiso (H)
value, set at 1.2 Ueq (carrier), which was refined isotropically. The
absolute configuration was confirmed by the absolute structure
determination by X-ray data indicated by the Flack parameter of
x = 0.17(8).
MClx·DCU + 2 NEt3 → MClx·(NEt3)2 + DCU
(4)
We recall that there is no correlation between the low-energy
=
shift of the C O stretching and the reactivity of the adducts. The
variation of the chemical shift of the co-ordinated carbonyl is not
=
aligned with the low energy shift of the C O stretching. A careful
Crystallographic and experimental details of the structure are
summarized in Table 4.
analysis of data shows that, anyway, the trend of reactivity for the
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6985–6992 | 6989
©

View Article Online
IR: 1615 cm-1 (nCO); 1568 cm-1 (dN–H + nC=N).
C NMR (CD2Cl2, 150 MHz): 158.5 ppm (C O); 20–40 ppm
(C-cyclohexyl rings).
Table 4 Summary of crystallographic data for 1a
13
=
1a
Anal. Calcd for C13H24Cl5N2NbO: C 31.57, H 4.89, Cl 35.84, N
5.67, Nb 18.79%. Found: C 31.40, H 4.83, Cl 35.70, N 5.63, Nb
18.65%.
Formula
Fw
Crystal system
Space group
T/K
C13H24Cl5N2NbO
494.50
Monoclinic
P21 (No. 4)
298(2)
0.71073
10.623(4)
12.392(5)
15.881(4)
90
107.286(5)
90
1996.20(12)
4
Synthesis of other MClx·DCU complexes
˚
Mo-lKa1/A
˚
a/A
All the complexes described below were synthesized by mixing
equimolecular quantities of DCU and metal chloride and pro-
˚
b/A
˚
c/A
a/◦
b/◦
g /◦
=
cessing the mixture as reported above. In each case the n(C O)
and the carbonyl relevant 13C-chemical shift (see Table 1) indicate
the coordination of DCU to the metal center.
3
˚
V/A
Z
ScCl3·(N,N¢-dicyclohexylurea). IR: 1610 cm-1 (nCO); 13C NMR
m/mm-1
1.274
=
(CD2Cl2, 150 MHz): 158.2 ppm (C O). Anal. Calcd for
Crystal size/mm
max = 0.60, mid = 0.10, min = 0.06
rc/g cm-3
1.645
C13H24Cl3N2OSc (M 375.66): C 41.56, H 6.44, Cl 28.31, N 7.46,
Sc 11.97%. Found: C 41.62, H 6.60, Cl 28.28, N 7.48, Sc 11.93%.
S = GoF
1.022
5.01–24.40
q range/◦
YCl3·(N,N¢-dicyclohexylurea). IR: 1614 cm-1 (nCO); 13C NMR
h,k,l ranges
-12 → 12
-13 → 14
=
(CD2Cl2, 150 MHz): 158.1 ppm (C O). Anal. Calcd for
-18 → 18
C13H24Cl3N2OY (M 419.61): C 37.21, H 5.77, Cl 25.35, N 6.68,
Y 21.19%. Found: C 37.15, H 5.75, Cl 25.32, N 6.60, Y 21.25%.
N(hkl)measured
N(hkl)unique
18 159
6417 [Rint = 0.074]
N(hkl)used [Iobs2s (Iobs)]
4866
LaCl3·(N,N¢-dicyclohexylurea). IR: 1601 cm-1 (nCO); 13C NMR
N(param)refined
398
R[Iobs2s(Iobs)]
R1 = 0.0449; wR2 = 0.0992
R1 = 0.0688; wR2 = 0.1106
0.17(8)
=
(CD2Cl2, 150 MHz): 157.4 ppm (C O). Anal. Calcd for
R[all data]
C13H24Cl3LaN2O (M 469.61): C 33.25, H 5.15, Cl 22.65, La 29.58,
N 5.97%. Found: C 33.37, H 5.22, Cl 22.73, La 29.62, N 5.90%.
Absolute structure: Flack (1983)
Flack parameter
Absorption correction: multi-scan Tmin = 0.515, Tmax = 0.928
TiCl4·(N,N¢-dicyclohexylurea). IR: 1603 cm-1 (nCO); 13C NMR
(SADABS; Sheldrick, 2002)
=
(CD2Cl2, 150 MHz): 158.9 ppm (C O). Anal. Calcd for
ꢀ
ꢀ
ꢀ
2
R1
=
(|Fo|
-
|Fc|)/ (|Fo|); wR2
=
{
[w(Fo
-
Fc2)2]/
ꢀ
C13H24Cl4N2OTi (M 414.02): C 37.71, H 5.84, Cl 34.25, N 6.77, Ti
11.56%. Found: C 37.65, H 5.82, Cl 34.29, N 6.68, Ti 11.58%.
1/2
[w(Fo2)2]}
.
TaCl5·(N,N¢-dicyclohexylurea). IR: 1620 cm-1 (nCO); 13C NMR
Computational details
=
(CD2Cl2, 150 MHz): 158.4 ppm (C O). Anal. Calcd for
The DFT calculations in the present work were carried out at
the B3LYP/SDDP level, where the B3LYP refers to the applied
exchange–correlation functional38-40 and SDDP denotes a basis set
including the Stuttgart-Dresden relativistic small core ECP basis
and the Dunning/Huzinaga DZ electron polarization basis set for
the lighter atoms.41-44 The structures of all investigated species have
been fully optimized. Vibrational frequencies have been calculated
to estimate the zero-point energy and thermal contributions to the
gas phase Gibbs free energy of the reaction components. The effect
of solvent (CH2Cl2) was included via the PSM methodology.45 The
thermal corrections were calculated for standard conditions (T =
298 K and P = 0.1 MPa). All these calculations were performed
using the Gaussian 03 package.46
C13H24Cl5N2OTa (M 582.55): C 26.80, H 4.15, Cl 30.43, N 4.81,
Ta 31.06%. Found: C 26.85, H 4.19, Cl 30.55, N 4.78, Ta 31.14%.
AlCl3·(N,N¢-dicyclohexylurea). IR: 1614 cm-1 (nCO); 13C NMR
=
(CD2Cl2, 150 MHz): 157.8 ppm (C O). Anal. Calcd for
C13H24AlCl3N2O (M 357.68): C 43.65, H 6.76, Al 7.54, Cl 29.74,
N 7.83%. Found: C 43.77, H 6.85, Al 7.46, Cl 29.66, N 7.90%.
SnCl4·(N,N¢-dicyclohexylurea). IR: 1599 cm-1 (nCO); 13C NMR
=
(CD2Cl2, 150 MHz): 157.6 ppm (C O). Anal. Calcd for
C13H24Cl4N2OSn (M 484.86): C 32.20, H 4.99, Cl 29.25, N 5.78,
Sn 24.48%. Found: C 32.12, H 4.87, Cl 29.31, N 5.63, Sn 24.55%.
Reaction of NbCl5·DCU with NEt3: formation of DCC
To a solution of NbCl5·DCU in dichloromethane a large excess
of triethylamine (8 : 1 molar ratio with respect to the starting
DCU) was added at 298 K. The addition of the amine caused
a change in color of the reaction mixture from yellow to red with
the production of white vapors and formation of some precipitate.
The reaction was followed by FTIR, GC and GC-MS analyses
showing the complete conversion of DCU after few minutes and
the formation of DCC (intense IR absorption at 2117 cm-1).
The solvent was evaporated under vacuum and the residue was
repeatedly extracted with pentane (5 ¥ 3 mL). The collected
pentane fractions were evaporated under vacuum and DCC was
Synthesis of NbCl5·DCU
DCU (0.31 g, 1.37 mmol) was added to a suspension of NbCl5
(0.36 g, 1.35 mmol) in dichloromethane (30 mL) with continuous
stirring at 298 K until the starting precipitate was completely
dissolved. FTIR and NMR analyses on the yellow solution showed
the coordination of DCU to niobium chloride. The insoluble
impurities were filtered off and the filtrate was layered with pentane
at 298 K until prismatic yellow crystals were formed. The solvent
was filtered off and the crystals were washed with pentane and
dried under vacuum.
6990 | Dalton Trans., 2010, 39, 6985–6992
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
obtained in high purity, as shown by FTIR and NMR analyses of
the isolated product compared with those of an authentic sample.
Reaction of NbCl5·DCU with pyridine and DBU
To a solution of NbCl5·DCU in dichloromethane was added a
large excess of either base (in 8 : 1 molar ratio respect to the starting
DCU). The color of the mixture turned to red and some precipitate
and white vapors were formed. The reaction mixture was analyzed
by FTIR and GC techniques. In both case DCC production was
observed. It was quantified by GC and shown to be 40–50% of the
theoretical amount with pyridine and 60–70% with DBU. In both
cases the starting compound was present in the reaction mixture.
The yield of conversion of DCU into DCC was 98%.
13
= =
C NMR (CD2Cl2, 150 MHz) for DCC: 140.0 ppm (N C N),
56.0, 35.4, 25.9 and 25.1 ppm (C-cyclohexyl rings).
Step-by-step deprotonation of 1a
Step 1. To a solution of NbCl5·DCU in CH2Cl2 a diluted
solution of triethylamine in CH2Cl2 (1 : 10 v/v) in a 1 : 1 molar ratio
respect to the starting complex was added dropwise. The resulting
mixture was analyzed by FTIR and NMR. The IR analysis showed
Conclusions
the disappearance of the band due to the coordinated DCU
In conclusion, N,N¢-dicyclohexylurea gives unique complexes with
Group 3, 4, 5, 13 and 14 metal halides.22,23 In NbCl5·DCU H-
bondings of different intensity dictate the molecular structure and
the 3D architecture. Also, the intramolecular H-bonding prompts
the easy elimination of HCl upon action of an external base and
the formation of NbOCl3 and free DCC.
-1
=
(nC= at 1615 cm ) and the appearance of the C N stretching
O
at 1651 cm-1. The formed NEt3·HCl was isolated and monitored
by 13C NMR (signals at 47.0 and 8.6 ppm due to the two C-atoms
of the ethyl group, identical with those of an authentic samples).
The solution was monitored by 13C NMR that showed the
equivalence of the two cyclohexyl rings (four sharp signals at 53
partially overlapping with the signal of the solvent, 34.9, 25.4,
24.8 ppm).
Acknowledgements
The authors thank the EU Project TOPCOMBI and MiUR PRIN
2008 for financial support and Mr. G. Chita (CNR-IC, Bari) for
his precious technical assistance in collecting XRD data. Research
fund from OTKA (grant K-60549) is also acknowledged.
Step 2. The filtered solution was added with a further equiva-
lent of NEt3 which caused a slight change of the IR spectrum.
The formation of putative 3a is proposed on the basis of the
fact that the IR and NMR spectra are identical with those of
a compound isolated from the reaction of NbOCl3 with DCC (see
next paragraph).
Notes and references
1 M. Aresta, A. Dibenedetto, E. Fracchiolla, P. Giannoccaro, C. Pastore,
I. Pa`pai and G. Schubert, J. Org. Chem., 2005, 70, 6177.
2 (a) K. Taek Jung and A. T. Bell, J. Catal., 2001, 204, 339; (b) K. T. J.
Alexis and T. Bell, Top. Catal., 2002, 20, 97.
3 (a) M. Aresta, A. Dibenedetto and C. Pastore, Inorg. Chem., 2003, 42,
3256; (b) M. Aresta, A. Dibenedetto and C. Pastore, Catal. Today,
2006, 115, 88.
4 M. Aresta, A. Dibenedetto, F. Nocito and C. Pastore, Inorg. Chim.
Acta, 2008, 361, 3215.
5 A. Dibenedetto, M. Aresta, F. Nocito and C. Pastore, J. Mol. Catal.
A: Chem., 2006, 257, 149.
6 (a) US Patent 5 648 537, 1997; (b) US Patent 3 972 933, 1976.
7 (a) T. Schlama, V. Governeur and C. Mioskowski, Tetrahedron Lett.,
1996, 37, 7047; (b) A. Williams and I. T. Ibrahim, Chem. Rev., 1981,
81, 589; (c) G. Amiard, R. Heymes and L. Velluz, Chem. abstr., 1958,
52, 426. CD2Cl2, 600 MHz; (d) H. Eilingsfeld, M. Seefelder and H.
Weidinger, Angew. Chem., 1960, 72, 836.
8 (a) H. Ulrich and A. A. R. Sayigh, J. Org. Chem., 1965, 30, 2779; (b) H.
Ulrich and A. A. R. Sayigh, Angew. Chem., Int. Ed. Engl., 1966, 5, 8.
9 C. L. Stevens, G. H. Singhal and A. B. Ash, J. Org. Chem., 1967, 32,
2895.
10 T. Theophanides and P. D. Harvey, Coord. Chem. Rev., 1987, 76, 237.
11 P. Vaughan and J. Donohue, Acta Crystallogr., 1952, 5, 530.
12 R. B. Penland, S. Mizushima, C. Curran and J. V. Quagliano, J. Am.
Chem. Soc., 1957, 79, 1575.
13 A. S. Gaballa, S. M. Teleb and El-M. Nour, J. Kor. Chem. Soc., 2007,
51, 339.
Step 3. The addition of an excess of NEt3 caused further
change of the IR spectrum with the appearance of the typical
band of free DCC (Fig. 5), quantitatively dosed by GC-MS, and
the formation of NbOCl3(NEt3)2 which was separated from the
mother solution and was filtered off. DCC remained in solution
and was separated by fractional crystallization. NbOCl3(NEt3)2
was identified by comparison with an authentic sample47 prepared
from NbOCl3 and NEt3.
Reaction of NbOCl3 and DCC: formation of NbOCl3·DCC 3a
48
To
a
suspension of NbOCl3
(0.193 g, 0.90 mmol) in
dichloromethane (15 mL) at room temperature a solution of DCC
(0.185 g, 0.90 mmol) in the same solvent (5 mL) was added.
The mixture was stirred at 300 K for 30 min with the slow
dissolution of the starting solid. The filtered yellow solution was
= =
analyzed by FTIR, showing the absence of the N C N stretching
=
of free carbodiimide and the presence of a C N stretching at
1653 cm-1. The solvent was evaporated under vacuum and the
formed solid was crystallized as nanocrystals not suited for XRD
determination. A slow crystallization by pentane layering caused
the conversion of the compound into several other species.
14 P. S. Gentile, J. White and S. Haddad, Inorg. Chim. Acta, 1974, 8, 97.
15 C. Airoldi, A. P. Chagas and F. P. Assuncao, Can. J. Chem., 1982, 60,
2132.
16 R. A. Bailey, I. R. Feins and T. R. Peterson, Can. J. Chem., 1969, 47,
171.
IR: 1652 cm-1 (N C N).
C NMR (CD2Cl2, 150 MHz): 158.8 ppm (N C N); 51.3,
= =
17 P. Askalani and R. A. Bailey, Can. J. Chem., 1969, 47, 2275.
18 (a) G. S. Papaefstathiou, R. Keuleers, C. J. Milios, C. P. Raptopoulou, A.
Terzis, H. O. Desseyn and S. P. Perlepes, Z. Naturforsch., 2003, 58b, 74;
(b) R. Keuleers, H. O. Desseyn, G. S. Papaefstathiou, L. Drakopoulou,
S. P. Perlepes, C. P. Raptopoulou and A. Terzis, Transition Met. Chem.,
2003, 28, 548; (c) E. Diamantopoulou, G. S. Papaefstathiou, A. Terzis,
C. P. Raptopoulou, H. O. Desseyn and S. P. Perlepes, Polyhedron, 2003,
13
= =
33.2, 25.5, 21.1 ppm (C-cyclohexyl rings).
Anal. Calcd for C13H22Cl3N2NbO: C 37.02, H 5.26, Cl 25.26, N
6.64, Nb 22.03%. Found: C 36.90, H 5.23, Cl 25.17, N 6.63, Nb
21.85%.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6985–6992 | 6991
©

View Article Online
22, 825; (d) L. Drakopoulou, C. Papatriantafyllopoulou, A. Terzis, S. P.
Perlepes, E. Manessi-Zoupa and G. S. Papaefstathiou, Bioinorg. Chem.
Appl., 2007, 51567.
33 K. R. Birdwhistell, J. Lanza and J. Pasos, J. Organomet. Chem., 1999,
584, 200.
34 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Purification of
Laboratory Chemicals, Pergamon, Oxford, England, 1986.
35 G. Charlot, Colorimetric Determination Of Elements, principles and
methods, Elsevier Pub. Co. Amsterdam, New York, 1964.
36 M. C. Burla, R. Calandro, M. Camalli, B. Carrozzini, G. L. Cascarano,
L. De Caro, C. Giacovazzo, G. Polidori, D. Siliqi and R. Spagna,
SIR2008. J. Appl. Cryst., 2007, 40, 609.
37 G. M. Sheldrick, SHELXL97. A short history of SHELX. Acta
Crystallogr., 2008, A64, 112.
38 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
39 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988,
37, 785.
40 P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys.
Chem., 1994, 98, 11623.
41 M. Dolg, H. Stoll, H. Preuss and R. M. Pitzer, J. Phys. Chem., 1993,
97, 5852.
42 T. H. Jr. Dunning, J. Chem. Phys., 1970, 53, 2823.
19 J.-G. Li, X. Yang and T. Ishigaki, J. Phys. Chem. B, 2006, 110, 14611.
20 H. Carlsson, M. Haukka, A. Bousseksou, J.-M. Latour and E.
Nordlander, Inorg. Chem., 2004, 43, 8252.
21 F. Marchetti, G. Pampaloni and S. Zacchini, Eur. J. Inorg. Chem., 2008,
453.
22 M. Aresta, A. Dibenedetto, and P. Stufano, Patent application
MI2009A001221, 2009.
23 P. Stufano, Ms. Sc. Thesis, University of Bari, 2009.
24 (a) T. Steiner, Angew. Chem., Int. Ed., 2002, 41, 48; (b) G. R. Desiraju,
Acc. Chem. Res., 2002, 35, 565.
25 (a) T. Zeegers-Huyskens, J. Mol. Struct., 2008, 887, 2; (b) E. S.
Kryachko and T. Zeegers-Huyskens, J. Phys. Chem. A, 2002, 106, 6832.
26 R. Taylor and O. Kennard, J. Am. Chem. Soc., 1982, 104, 5063.
27 (a) J.-A. van den Berg and K. R. Seddon, Cryst. Growth Des., 2003,
3, 643; (b) L. Brammer, E. A. Bruton and P. Sherwood, Cryst. Growth
Des., 2001, 1, 277; (c) P. K. Thallapally and A. Nangia, CrystEngComm,
2001, 27, 1; (d) C. B. Aakero¨y, T. A. Evans, K. R. Seddon and J. Pa`linko`,
New J. Chem., 1999, 23, 145; (e) G. Aullo`n, D. Bellamy, L. Brammer,
E. A. Bruton and A. G. Orpen, Chem. Commun., 1998, 653; (f) M.
Mascal, Chem. Commun., 1998, 303.
28 (a) V. Balamurugan, J. Mukherjee, M. S. Hundal and R. Mukherjee,
Struct. Chem., 2007, 18, 133; (b) J. G. Planas, F. Teixidor, C. Vin˜as,
M. E. Light and M. B. Hurthouse, Chem.–Eur. J., 2007, 13, 2493;
(c) K.-T. Youm, J. Ko and M.-J. Jun, Polyhedron, 2006, 25, 2717; (d) V.
Balamurugan, M. S. Hundal and R. Mukherjee, Chem.–Eur. J., 2004,
10, 1683; (e) C. Janaik and T. G. Scharmann, Polyhedron, 2003, 22,
1123; (f) L. D. Gibb, E. D. Stevens and B. C. Gibb, J. Am. Chem. Soc.,
2001, 123, 5849; (g) M. Freytag and P. G. Jones, Chem. Commun., 2000,
277.
29 (a) L. Song and T. Iyoda, J. Inorg. Organomet. Polym. Mater., 2009,
19, 124; (b) B. K. Park, S. H. Lee, E. Y. Lee, H. Kwak, Y. M. Lee, Y. J.
Lee, J. Y. Jun, C. Kim, S.-J. Kim and Y. Kim, J. Mol. Struct., 2008,
890, 123; (c) A. Angeloni, P. C. Crawford, A. G. Orpen, T. J. Podesta
and B. J. Shore, Chem.–Eur. J., 2004, 10, 3783; (d) R. K. O. Sigel, E.
Freisinger, S. Metzger and B. Lippert, J. Am. Chem. Soc., 1998, 120,
12000; (e) O. Navon, J. Bernstein and V. Khodorkovsky, Angew. Chem.,
Int. Ed. Engl., 1997, 36, 601.
30 A. Kova`cs and Z. Varga, Coord. Chem. Rev., 2006, 250, 710.
31 F. Neve and A. Crispini, Cryst. Growth Des., 2001, 1, 387.
32 T. Spaniel, H. Go¨rls and J. Scholz, Angew. Chem., Int. Ed., 1998, 37,
1862.
43 T. H. Jr. Dunning and P. J. Hay, in Methods of Electronic Structure
Theory, ed. H. F. III Schaefer, Plenum Press, 1977, vol. 3.
44 H. F. III Schaefer, J. Chem. Phys., 1985, 83, 5721.
45 J. Tomasi, B. Mennucci and E. Cance`s, THEOCHEM, 1999, 464,
211.
46 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz , Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
47 S. M. Sinitsyna, W. I. Sinyagin and Yu. A. Buslaev, Inorg. Mater., 1969,
5, 514.
48 V. C. Gibson, T. P. Kee and A. Shaw, Polyhedron, 1988, 7, 2217.
6992 | Dalton Trans., 2010, 39, 6985–6992
This journal is
The Royal Society of Chemistry 2010
©