Synthesis and characterization of a palladium(0) complex with cyclophosphazene bearing two diphenylphosphine ligands and application in Suzuki-Miyaura cross-coupling

Article abstract of DOI:10.1016/j.ica.2018.05.016

This paper describes the synthesis of a cyclophosphazene-based diphenylphosphine ligand and a new Pd(0) complex. Infrared spectroscopy (ATR-IR), ESI+-MS, ³¹P, ¹H and ¹³C NMR, Raman, WD-XRF, ICP-OES and TGA analysis show the coordination of two palladium atoms per unit of cyclophosphazene. A semiempirical calculation method was employed to find the lowest energy structure among the possible ones and Density Functional Theory (DFT) was used to optimize the found structure and obtain its bond angles, dihedral angles, bond lengths, atomic distances, and to calculate the vibrational spectrum (PBE/def2-TZVP(-f)). The new Pd complex showed activity in Suzuki-Miyaura cross-coupling reactions with halobenzenes and phenylboronic acid, tolerating different functional groups.

Full text of DOI:10.1016/j.ica.2018.05.016

Accepted Manuscript
Research paper
Synthesis and characterization of a palladium(0) complex with cyclophospha-
zene bearing two diphenylphosphine ligands and application in Suzuki-Miyaura
cross-coupling
Maria das G. de O. e Silva, Carolina Galuppo, Bárbara C. Tudisco, Arnaldo G.
de Oliveira Júnior, Manoel.V.F. Barrionuevo, Camilla Abbehausen, Regina
Buffon
PII:
S0020-1693(17)31611-0
https://doi.org/10.1016/j.ica.2018.05.016
ICA 18267
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
19 October 2017
12 April 2018
14 May 2018
Please cite this article as: M.d.G. de O. e Silva, C. Galuppo, B.C. Tudisco, A.G. de Oliveira Júnior, Manoel.V.F.
Barrionuevo, C. Abbehausen, R. Buffon, Synthesis and characterization of a palladium(0) complex with
cyclophosphazene bearing two diphenylphosphine ligands and application in Suzuki-Miyaura cross-coupling,
Inorganica Chimica Acta (2018), doi: https://doi.org/10.1016/j.ica.2018.05.016
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

TITLE: Synthesis and characterization of a palladium(0) complex with cyclophosphazene
bearing two diphenylphosphine ligands and application in Suzuki-Miyaura cross-coupling
AUTHOR NAMES: Maria das G. de O. e Silvaa,b*, Carolina Galuppoa, Bárbara C. Tudiscoa,
Arnaldo G. de Oliveira Júniora, Manoel V. F. Barrionuevoa, Camilla Abbehausena, Regina
Buffona,*
a
Institute of Chemistry, University of Campinas. Unicamp, PO Box 6154, 13083-970
Campinas, SP, Brazil.
b
Federal Institute of Education, Science and Technology of Maranhão. IFMA, 65700-000
Bacabal, MA, Brazil.
*
Corresponding authors. E-mail addresses: rbuffon@iqm.unicamp.br (R. Buffon) and
maria.silva@ifma.edu.br (M. G. O. Silva).
Abstract
This paper describes the synthesis of a cyclophosphazene-based diphenylphosphine ligand
and a new Pd(0) complex. Infrared spectroscopy (ATR-IR), ESI+-MS, 31P, 1H and 13C NMR,
Raman, WD-XRF, ICP-OES and TGA analysis show the coordination of two palladium
atoms per unit of cyclophosphazene. A semiempirical calculation method was employed to
find the lowest energy structure among the possible ones and Density Functional Theory
(DFT) was used to optimize the found structure and obtain its bond angles, dihedral angles,
bond lengths, atomic distances, and to calculate the vibrational spectrum (PBE/def2-TZVP(-
f)). The new Pd complex showed activity in Suzuki-Miyaura cross-coupling reactions with
halobenzenes and phenylboronic acid, tolerating different functional groups.
KEYWORDS: Pd(0) complexes; cyclophosphazene-based ligands; Suzuki-Miyaura cross-
coupling.
1. Introduction
Palladium-catalyzed cross-coupling reactions are among the most important synthetic
tools for organic chemistry, providing facile access to carbon-carbon (C-C) bond
construction. In particular, the Suzuki-Miyaura C-C coupling experienced a notorious growth
in both publications and patents over the past decades when compared to other classes of C-C
coupling [1,2]. This approach became the most attractive due to several advantages such as
mild reaction conditions and tolerance to many functional groups. The improvement of

methods allowed its application in several areas such as natural product synthesis, material
science, medicinal, biological and supramolecular chemistry, as well as catalysis and
coordination chemistry [3,4].
Cyclophosphazene-based ligands have attracted interest due to the versatility
presented by these structures: coordination can occur not only by nitrogen and phosphorus
atoms of the cycle, but also by exocyclic groups. In the latter case, examples were reported
with cyclophosphazenes bearing such side groups as pyrazolyl, hydrazides, pyridyl, amino,
and phosphine bonded to several transition metals [5–7]. A more ambitious application of
these inorganic rings as scaffolds to more complex structures was proposed by Chandrasekhar
et al. who anchored a copper complex in a polymeric matrix and employed it in catalytic
hydrolysis of phosphate esters without loss of activity even after catalyst recycling [8].
Since we have been interested in the immobilization of transition metal complexes by
the
sol-gel
method
[9–11],
we
decided
to
synthesize
the
ligand
[N3P3(O2C12H8)2(OC6H4PPh2)2] (3) and its complex with palladium in order to establish a
structural characterization, and to evaluate its activity in Suzuki-Miyaura coupling reactions.
Therefore, our objective was to determine the feasibility of employing such structures in a
future immobilization of palladium complexes.
2. Experimental
All reactions were carried out under a dried argon atmosphere by using standard
Schlenk techniques. Solvents were treated according to standard procedures [12] and were
degassed prior to use. Potassium carbonate (Nuclear) and cesium carbonate (Aldrich) were
oven dried at 140 °C for 24 h. Ammonium chloride (Vetec) was dried in a desiccator with
phosphorus pentoxide. Potassium tert-butoxide was dried under vacuum (tBuOK, Aldrich).
Deca-hydronaphthalene (Aldrich) was distilled and stored over molecular sieves.
Hexachlorocyclotriphosphazene (N3P3Cl6, Aldrich), 2,2'-biphenol (Aldrich), 4-bromophenol
(Aldrich),
n-butyllithium
(n-BuLi,
2.5
mol
L-1
in
hexane,
Aldrich),
tris(dibenzylideneacetone)dipalladium(0) (Pd2(dba)3, Aldrich), chlorodiphenylphosphine
(Aldrich), (4-hydroxyphenyl)diphenylphosphine (Aldrich), aryl bromides and chlorides as
well as arylboronic acids (Aldrich) were used without further treatment.
Proton-decoupled 31P and 13C NMR spectra, and 1H NMR spectra were recorded either
with a Bruker-Biospin Avance-III 500 MHz or an Avance DPX 250 MHz spectrometer at 25
ºC. Attenuated total reflectance infrared (ATR-IR) spectra were recorded with an Agilent
Technologies Cary 630 FTIR spectrometer, in the 4000-400 cm-1 range, 64 scans, with a

resolution of 4 cm-1. The thermogravimetric analyses (TGA) were carried out on a TGA Q600
(TA Instruments) operated with a constant air flow of 100 mL min-1, with a heating rate of 10
ºC min-1. The samples were heated to 100 °C, kept at this temperature for 15 min, and then
heated up to 900 °C, with this temperature maintained for 120 min. Electrospray mass
spectrometry (ESI+-MS) analysis was performed on a Q-Tof (Waters) mass spectrometer
within an m/z range of 100-2000, with a capillary voltage of 3 kV and a cone voltage of 50 V.
The nebulizer gas flow was 0.5 L h-1. The samples were injected by direct infusion in a flow
of 40 µL min-1. Raman spectra were recorded using a confocal Horiba Scientific T64000
spectrometer. The excitation source was a laser of 633 nm, focused with a 100× focal-lens
objective with a resolution of 2 cm-1. The recorded spectra were smoothed by using the
Savitsky-Golay [13] method, setting a window size of 7 points, and a polynomial fit of second
order. The baseline correction was performed by employing the Tan et al [14] method with λ
= 107 and p = 0.001. The degradation and decomposition of the ligand and complex,
respectively, were recorded on a PF1500 Farma (Gehaka). X-ray fluorescence spectrometry
was carried out with a Shimadzu XRF-1800 spectrometer, in which the wavelength dispersion
(WD-XRF) was employed using Rh radiation (3 kV and of 95 mA). The palladium
percentage in the complex was determined by inductively coupled plasma optical emission
spectrometry (ICP-OES), on a Perkin Elmer Optima 8300 spectrograph, with λ = 340.458 nm.
The sample was prepared by digestion in H2SO4:H2O2 (3:1). After that, the solution was
placed on an ultrasound bath, diluted in high purity water (resistance of 18.2 MΩ), filtered
and analyzed.
The catalytic experiments were monitored by gas phase chromatography using an
Agilent 7890 gas chromatograph equipped with an HP 5 capillary column and a flame
ionization detector (FID). Conversions of halobenzenes with different functional groups were
determined using calibration curves obtained with standard solutions. Mass spectra were
obtained on a gas chromatograph-mass spectrometer (GC-MS) Agilent 7890A using an HP-5
capillary column.
2.1. Synthesis of the ligand
[N3P3Cl2(O2C12H8)2] (1) and [N3P3(OC6H4Br)2(O2C12H8)2] (2) were prepared as
described by Carriedo et al. [15]. For details, see Supporting Information (Figs. S1-S8 and
Scheme S1).
2.1.1. [N3P3(O2C12H8)2(OC6H4PPh2)2] (3) – by lithiation
In an attempt to form (3), a procedure adapted from the literature [16] was carried out.
Several experiments were performed aiming to optimize the reaction conditions for the attack

on the electrophile by n-butyllithium, and for further reaction with chlorodiphenylphosphine,
see Supporting Information (Table S1 and Figs. S9-S12).
25 (m), 9.6 (m), -6.4 (s).
31P NMR (CDCl3,  ppm):
ESI+-MS: m/z calcd for
[N3P3(O2C12H8)2(OC6H4)2]+
690.1113,
found
690.0945;
m/z
calcd
for
[N3P3(O2C12H8)2(OC6H4PPh2)(OC6H5)] + O2 890.1504 [M+ + O2], found 890.1368; m/z calcd
for [N3P3(O2C12H8)2(OC6H4PPh2)2] + O2 1090.1895 [M+ + O2], found 1090.1940.
2.1.2. [N3P3(O2C12H8)2(OC6H4PPh2)2] (3) – conventional route
The synthesis of (3) was based on the methodology proposed by Carriedo et al. [17].
(1) (1.0 g, 1.74 mmol), (4-hydroxyphenyl)diphenylphosphine (0.96 g, 3.48 mmol) and
Cs2CO3 (2.27 g, 6.96 mmol) were dissolved in acetone (80 mL) and the solution was refluxed
for 1.5 h. The solvent was removed under vacuum and the resulting product was then
extracted with CH2Cl2 (8 × 20 mL). Yield: 1.43 g, 77.8%.
The degradation of the ligand (3) was carried out in the temperature range 86-92 °C in
oxidizing (air) atmosphere. 31P NMR (CDCl3,  ppm): 9.6 [dd, P(OC6H4-PPh2)2], 25.6 [d,
P(O2C12H8), AB2 system, JAB = 93 Hz], -6.4 (s, PPh2). 13C NMR (CDCl3,  ppm): 121.4,
121.9, 128.7, 129.6, 130.7, 148.0 (C12H8), 116.7, 128.6, 128.8, 133.6, 133.9, 135.2, 137.1,
1
151.4 (OC6H4PPh2). H NMR (CDCl3,  ppm): 7.1-7.5 (m). For details, see Supporting
Information (Figs. S13 and S14). ATR-IR (cm-1): 3060 νC-H(ar.), 1585 and 1497 νC=C(ar.), 1227
and 1165 νP=N, 971 νP-O-C. TGA: -2.7% (100 °C), -97.3% (100-890 °C). Residue at 900 °C:
0%. ESI+-MS: m/z calcd for [N3P3(O2C12H8)2(OC6H4PPh2)2] + O2 1090.1895 [M+ + O2],
found 1090.1730. Raman (cm-1): 1612, 1586 and 1501 νC=C(ar.), 1310 and 1274 νC-OP, 1251
and 1157 νP=N, 1094 and 1039 νC-H(ar.), 754 νP-N-P
.
2.2. Pd complex (4): Pd2[N3P3(O2C12H8)2(OC6H4PPh2)2(dba)2]
Ligand (3) (0.49 g, 0.04 mmol) and Pd2(dba)3 (0.21 g, 0.02 mmol) were dissolved in 4
mL of dry THF and the solution was stirred for 1 h at room temperature. After the reaction,
the solvent was evaporated under vacuum, followed by the addition of 5 mL of benzene. The
resulting red/brown solution was filtered, the solid dried under vacuum and washed with
ether/hexane (50% v/v).
The decomposition of the Pd complex (4) was carried out in the temperature range 76-
78 °C in oxidizing (air) atmosphere. 31P NMR (C6D6,  ppm): 10.2 [dd, P(OC6H4-PPh2)2],
25.5 [d, P(O2C12H8), AB2 system, JAB = 94 Hz], 24.7 (s, Pd-PPh2). 1H (C6D6,  ppm): 6.8-7.7
(m). ATR-IR (cm-1): 3060 νC-H(ar.), 1590 and 1475 νC=C(ar.), 1230 and 1164 νP=N, 970 νP-O-C.

Raman (cm-1): 190 νPd-P. WD-XRF: P, Pd, and Si. TGA: -86.8% (100-895 °C). Residue at
900 °C: 13.2%. ICP-OES: 11.6%.
2.3. Theoretical calculations of the ligand (3) and Pd complex (4)
As the complex contains two metal centers with one dibenzylideneacetone (dba)
ligand coordinated to each Pd(0), it is expected to find up to four possible conformations by
changing the way both dba ligands are coordinated to Pd(0). Since the lowest conformation
energy is of great interest due to its higher stability, we computationally employed a low level
of theory to investigate the energy of all possible conformations. Hence, our energy analysis
of all conformations was done using the semiempirical Geometry, Frequency, Non-covalent
Tight Binding method (GFN-xTB) [18] by Grimme's XTB software, which was set to perform
its highest level of optimization (optlv 2). Once the conformation with the lowest energy was
found, it was submitted to further computation using DFT.
All computations including hessian calculations were performed using DFT in the
Orca software package (version 4.0.1) [19] with tight criteria set to both optimization and
self-consistent field (SCF). The employed functional was PBE [20] with basis set def2-
TZVP(-f) [21] and auxiliary basis def2/J [22] with resolution of identity (RI) [23–29]
approximation to speed up calculations. Also, dispersion correction was computed by using
D3BJ [30,31]. The guess method used to predict the starting orbitals was PModel. Finally, in
order to overcome instabilities found in SCF the approximate second order in the SCF [32,33]
was chosen, and it was set to start with an orbital gradient value below 0.001.
2.4. General procedure for Suzuki-Miyaura coupling reactions
In a typical experiment, a mixture of aryl bromide (0.175 mmol), phenylboronic acid
(0.28 mmol), potassium tert-butoxide (0.42 mmol), a solution of the catalyst in THF (200 µL;
0.5 mol% with respect to palladium), 14 µL of deca-hydronaphthalene (as an internal standard
for gas chromatography) and dry toluene (4 mL) were placed in a Schlenk tube, which was
then sealed under argon and warmed at 110 °C for 4 h. Duplicate experiments were carried
out for each catalytic run in order to ensure reproducibility.
3. Results and discussion
Synthesis and characterization of species (1) and (2) are described in Supporting
Information. An attempt to synthesize (3) via lithiation of (2) followed by a nucleophilic
attack of a chlorodiphenylphosphine aimed to check its feasibility, and open access to a
variety of cyclophosphazene-based phosphines, as shown in Supporting Information (Scheme
S1). Such a route had already been described for total substitution of chlorine in

31
hexachlorocyclophosphazene [16], even by one of us [34]1. To our surprise, the P NMR
spectrum of a solution of the obtained solid shows peak splitting (Fig. S9), suggesting the
presence of more than one configuration in solution: a racemic mixture of (R,R) and (S,S)
enantiomers, as well as the meso (R,S) form [35]. Such results imply that the desired
bisphosphine was not obtained.
31
The P NMR spectra obtained in experiment 7 (after lithiation and without reaction
with chlorodiphenylphosphine) do not show peak splitting, depicting a homogeneous
chemical environment (Fig. S10), while no signal is observed around 118 ppm (Fig. S11) in
13
the C NMR region where a peak assigned to the carbon bonded to the halide would be
expected [15,36]. These results indicate the complete replacement of bromine.
According to Díaz et al. [36], the difficulty in achieving the desired product could be
due to the lack of reaction of the second phosphine group (Fig. S15). Despite n-butyllithium
excess and dilithiation completion, the bulkiness of the phosphine ligand would hamper its
access to the phenoxide carbon. Thus, even if (3) was formed in minor amounts, the reaction
with diphenylphosphine also favored the formation of an inseparable and undesirable mixture
of products as [N3P3(O2C12H8)2(OC6H5)2] and [N3P3(O2C12H8)2(OC6H4PPh2)(OC6H5)].
Therefore, this route to ligand (3) and eventual analogues appeared to be unsatisfactory.
Although the procedure is described in the literature by Allcock et al. [16,37], it is
important to mention that this experimental approach seems to have been set aside. Thus, the
subsequent experimental procedures were carried out according to the conventional route,
available only for the diphenylphosphine derivative.
When (1) was allowed to react with (4-hydroxyphenyl)diphenylphosphine, the 31P
NMR spectrum of the resulting product (Fig. 1) exhibits an AB2 spin pattern with three
resonance signals: a doublet at 25.6 ppm; a doublet of doublets at 9.6 ppm, corresponding to
the phosphorus atom of phosphazene-bearing two (O2C12H8) and (OC6H4-PPh2)2 groups,
respectively; and a singlet at -6.4 ppm, relative to the phosphorus atom of PPh2 groups. This
spectrum clearly shows that the synthesis of (3) was successful.
1 Although peak integration in 31P NMR always suggested six phosphine groups per cyclophosphazene
unit, a misinterpretation of the experimental results and the presence of more than one species cannot
be ruled out.

Fig. 1. 31P NMR spectrum (CDCl3, 500 MHz) of (3).
In addition, the spectrum shows a singlet at 25.7 ppm, corresponding to the
replacement of all chlorine atoms by 2,2'-dioxybiphenyl groups, [N3P3(O2C12H8)3].
Nevertheless, this impurity does not affect the synthesis of (3) nor the synthesis of the
palladium complex (4), since it is not reactive during the process [15].
Ligand (3) was also characterized by infrared spectroscopy. Its ATR-IR spectrum is
shown in Fig. 2 and can be compared to the spectra of (1) and (2), shown in Supporting
Information, Figs. S4 and S8.
Fig. 2. ATR-IR spectrum of (3).

The ATR-IR spectrum of (3) confirms the formation of the ligand as the characteristic
stretching peaks in the spectra of the phosphazenes νP=N are clearly present. The band at 3060
cm-1 (see Supporting Information, Fig. S16) can be assigned to stretching vibrations of the C-
H bond while the aromatic ring νC=C(ar.) has symmetrical stretching frequencies at 1585 and
1497 cm-1 [38–40]. The spectrum also shows bands corresponding to the phosphazene moiety,
νP=N stretching vibrations between 1227 and 1165 cm-1, and a vibration band around 971 cm-1,
ascribed to νP-OC [35,38,41].
ESI+-MS analysis of a methanol solution of the ligand (3) showed a main peak at m/z
= 1090.1730, which presents the isotopic patterns corresponding to the oxidized species, Fig.
3.
Fig. 3. ESI+-MS of (3) (a) simulated isotopic pattern and (b) experimental isotopic pattern for
[N3P3(O2C12H8)2(OC6H4(P=O)Ph2)2].
The calculated IR spectrum of the optimized structure of the Pd complex (4) shows
good agreement with the experimental spectrum (see Supporting Information, Fig. S17), and
it was in accordance with the obtained results for the previously proposed structure, as the
vibrations bands of the spectra of Pd complex (4) and ligand (3) were similar.
Raman spectra of ligand (3) and of the corresponding Pd complex (4) are shown in
Fig. 4 and Fig. S18. The vibrational bands observed at 1612, 1586 and 1501 cm-1 are
attributed to the νC=C(ar.) stretching. The medium intense bands near 1310 and 1274 cm-1
correspond to the νC-OP. In addition, the bands near 1251 and 1157 cm-1 can be assigned to
νP=N. The medium intense bands near 1094 and 1039 cm-1, and 754 cm-1 can be assigned to νC-

H(ar.) and νP-N-P stretching modes, respectively. The strong band near 1000 cm-1 corresponds to
the interaction between the aromatic ring vibration, and the stretching of the P-C bonds
[42,43]. The Raman spectrum of (4) shows bands of medium intensity because the laser
power was regulated by a filter to prevent sample heating and degradation, a new band at 190
cm-1, assigned to the Pd-P stretching, confirms the coordination of the phosphine ligand to the
palladium center [44].
Fig. 4. Raman spectra of the Pd complex (4) and of ligand (3).
The 31P NMR spectrum of (4) is shown in Fig. 5. The signal assigned to the
coordinated phosphine is shifted downfield (24.7 ppm) compared to the free ligand (-6.4 ppm)
[45].

Fig. 5. 31P NMR spectrum (C6D6, 500 MHz) of the Pd complex (4).
The 1H NMR and 13C NMR spectra corroborate these results and suggest the presence
of coordinated dba in the metal complex, see Supporting Information (Figs. S19 and S20).
In addition, the Raman spectrum of Pd2(dba)3 was also recorded, (see Supporting
Information, Fig. S21). It shows that the commercial complex is fluorescence active, most
probably due to its structural characteristics, presenting conjugated double bonds, quite
different from Pd complex (4). X-Ray fluorescence analysis detected P, Pd, and Si. The
presence of Si is associated with the starch impurity used in the analysis.
A thermal decomposition analysis was performed, and the TGA curves of (3) and (4)
are shown in Figs. S22 and S23, respectively. The curve for the complex shows a series of
weight loss events, the first loss (5.4 wt %) around 100 °C, corresponding to the volatiles. The
Pd complex (4) also exhibits losses between 100 and 895 °C, corresponding to the
decomposition of the ligands (86.8% observed weight loss, 87.7% calculated). It was not
possible to attribute those weight losses of the ligands, since they have high molar mass and
complex structures, with many possibilities of bond-breaking and consequent formation of
several decomposition products. The Pd complex (4) was completely degraded to metal Pd
with a mass loss of 13.2% (calcd: 12.3%), even after a period of 2 h at 900 °C. The ICP-OES
analysis was used to quantify the amount of palladium present in solution. The obtained value
of 11.6% is in agreement with the proposed structure with two palladium atoms per unit of

X-ray average
value (CSD)
Entry
Molecular modelling
cyclophosphazene, Pd2[N3P3(O2C12H8)2(OC6H4PPh2)2(dba)2] (Fig. 6). The difference between
the experimental and the calculated value can be attributed to the presence of the
[N3P3(O2C12H8)3]. Therefore, two geminal phosphines are not involved in coordination to
only one palladium center. It is worthwhile to mention that by using a cyclophosphazene
bearing five phosphine groups, Chandrasekhar et al. [45] concluded that there was only one
Pd atom per cyclophosphazene unit, and the metal was bonded to two nongeminal
phosphines.
The unusual 14-electron structure [PdL2] of complex (4) can be rationalized in terms
of the sterical demands of bulky ligands [46–49], which play an essential role in geometry and
coordination of the proposed structure.
As the structure determination was not possible by XRD of single crystals, theoretical
calculations for both the ligand (3) and the Pd complex (4) were performed. Since transition
metal complexes bearing more than 180 atoms, with possible non-covalent interactions and
with different possible conformations, highly increases the computational cost, the GFN-xTB
[18] semiempirical method was employed to screen among the four possible molecular
structures of Pd complexes. The lowest energy conformation among the four starting
structures was chosen as the first guess for the most stable compound (see Supporting
Information, Figs. S24-S27). Subsequently, the molecular structures of both the ligand (3) and
the chosen Pd complex (4) were optimized using DFT with the functional PBE/def2-TZVP(-f)
[20,21].
The structure of the ligand (3) before and after coordination to the metal center is
depicted in Fig. 6. They were studied by observing the following structural descriptors: bond
angles, dihedral angles, atomic distances and bond lengths, which were compared to transition
metal complexes containing cyclophosphazene-based ligands, as well as for Pd complexes
with dibenzylideneacetone (dba) and phosphines ligands. The structures to which we
compared our calculated structural descriptors are reported in the Cambridge Crystallographic
Database CSD (Table 1). Although the number of structures is restricted, the geometry
optimization at DFT level showed a good agreement with the average values of experimental
structural descriptors observed herein.
Table 1.
Ligand (3) and Pd complex (4) - bond lengths, bond angles, dihedral angles and atomic
distances calculated using DFT with the functional PBE/def2-TZVP(-f) compared to values
found from crystal structures from the CSD

Pd complex (4)
Ligand (3)
Bond lengths (Å)
1
2
3
4
5
6
7
8
Pd117-C159
Pd117-C158
P0-Pd117
2.172
2.176
2.269
2.168
2.194
2.259
1.409
1.350
1.404
1.350
1.594
1.604
1.596
1.598
1.591
1.591
1.632
1.618
1.623
1.621
1.626
1.624
2.176 a, 2.172 b
2.149 a, 2.154 b
2.343 b
Pd118-C130
Pd118-C131
P85-Pd118
C158-C159
C162-C163
C131-C130
C126-C127
P33-N42
P33-N45
P36-N42
P36-N39
P37-N39
P37-N45
P33-O34
P33-O35
P36-O46
P36-O49
P37-O40
P37-O43
2.154 b
2.172 b
2.331 b
1.411 a, 1.395 b
1.292 b
9
-
-
10
11
12
13
14
15
16
17
18
19
20
21
22
1.595
1.601
1.596
1.596
1.592
1.590
1.631
1.622
1.628
1.617
1.624
1.624
1.581 c, 1.581 d
1.585 c, 1.583 d
1.581 c, 1.579 d
1.573 c, 1.589 d
1.576 c, 1.574 d
1.573 c, 1.578 d
-
-
1.593 c
1.587 c
1.591 c
1.586 c
Bond angles (°)
N45-P33-N42
N39-P36-N42
N39-P37-N45
O34-P33-O35
O46-P36-O49
O40-P37-O43
P36-O46-C57
P36-O49-C63
P37-O43-C38
P37-O40-C51
Dihedral angles (°)
C51-C56-C50-C38
C57-C62-C64-C63
Atomic distances (Å)
P0-P85
23
24
25
26
27
28
29
30
31
32
118.624
120.060
119.589
98.757
102.326
101.945
120.901
118.276
121.598
119.809
118.641
120.401
119.277
98.371
102.588
102.157
121.542
119.112
120.982
119.457
117.044 c, 117.719 d
118.088 c, 118.189 d
118.328 c, 117.571 d
-
102.762 c
103.121 c
119.997 c
120.295 c
119.473 c
120.704 c
33
34
-42.215
-42.292
-42.187
-43.006
-42.833 c
-43.105 c
35
36
37
11.062
6.495
6.880
9.751
6.411
6.768
-
-
-
P0-P33
P85-P33
The bond lengths, atomic distance, bond angles and dihedral angles represented are for the
a
crystal structures reported in the October 2017 version of CSD. Averages calculated
b
including all structures containing Pd complexes with dba and phosphines ligands. [Pd(η2-
c
dba)(PPh3)2] [50]. Averages calculated including structures containing cyclophosphazene
d
backbone blocked by the biphenoxide groups. Averages calculated including all structures
containing a cyclophosphazene backbone.

For the Pd complex (4) only one of the double bonds (C158-C159 with 1.409 Å and
C131-C130 with 1.404 Å) of each dba-ligand is coordinated to Pd centers (Table 1, entries 7
and 9). As described by Majchrzak et al. [50], this causes a decrease in the bond lengths of the
double bonds (C162-C163 and C126-C127 with 1.350 Å) that are not involved with the metal
interaction (Table 1, entries 8 and 10).
Looking at Table 1 it is clear that besides a slight variation found in both bond lengths
and angles, for the cyclophosphazene backbone of the Pd complex (4), and the ligand (3)
(Table 1, entries 11-34) those values remained almost unchanged even after coordination to
the metal center, showing that the cyclophosphazene backbone does not have a significant
interaction with Pd. Comparing entries 35-37, however, we observe an increase in the atomic
distances P0-P85 (from 9.751 Å to 11.062 Å), P0-P33 (from 6.411 Å to 6.495 Å) and P85-
P33 (from 6.768 Å to 6.880 Å), which is attributed to the hindrance caused after coordination
to the Pd center.
Fig. 6. Theoretical structure for both (a) ligand (3) and (b) Pd complex (4).
The Pd complex (4) was tested in Suzuki-Miyaura coupling reactions since this is the
most studied method for the C-C bond formation from simple molecules [51,52]. The catalyst
was inactive for aryl chlorides with neutral or electron-donating groups due to the binding
strength of Ar-Cl bonds (Table 2, entries 1-3), which could be potentially retarding the
oxidative addition, the first step of the catalytic cycle, hampering the formation of ArPd(II)X
[53]. Moreover, homocoupling products of the phenylboronic acid and/or the halide were
observed and identified by GC-MS.

When different halobenzenes, Ar-I (Table 2, entry 4) or Ar-Br (Table 2, entry 5) were
employed, the expected product was obtained, with yields following the order of Ar-I > Ar-Br
> Ar-Cl. However, in the presence of non-activated aromatic rings and/or sterically hindered
substrates, lower yields (Table 2, entries 6-8) were obtained. An ortho methoxy group
(moderate electron-donating group) also led to lower yield, probably due to steric problems
[54].
Activated aryl iodides (Table 2, entries 9 and 10), and slightly electron-deficient
bromophenol (Table 2, entry 11) led to excellent yields. A moderate yield (Table 3, entry 1)
was obtained when a deactivated phenylboronic acid was employed. Arylboronic acids with
an electron-donating group led to good yields of the coupling reaction products, even in the
case of a hindered acid (Table 3, entry 3). Finally, all catalytic results obtained with Pd
complex (4) are in agreement with those reported in the literature [51,53,54].
Table 2.
Results for Suzuki-Miyaura coupling reactions of aryl halides and phenylboronic acid in the
presence of the Pd complex (4). a
Entry
Aryl halide
Product
Yield (%)b
1
0
2
3
4
5
0
0
49
40
6
7
37
30

8
31
9
>99
>99
95
10
11
a
t
Reaction conditions: aryl halide (0.175 mmol); phenylboronic acid (0.28 mmol); BuOK
b
(0.42 mmol); Pd Complex (4) (0.5 mol%); T = 110 °C; Toluene/THF (v/v = 20:1); 4 h.
Yield was determined by GC using deca-hydronaphthalene as an internal standard.
Table 3.
Pd-catalyzed Suzuki-Miyaura reactions of aryl bromides with arylboronic acids. a
Entry
Aryl bromides
Arylboronic acid
Yield (%)b
1
35
2
3
56
79
4
5
61
65

a
t
Reaction conditions: aryl bromide (0.175 mmol); arylboronic acid (0.28 mmol); BuOK
b
(0.42 mmol); Pd Complex (4) (0.5 mol%); T = 110 °C; Toluene/THF (v/v = 20:1); 4 h.
Yields were determined by GC using deca-hydronaphthalene as an internal standard.
4. Conclusions
The attempted synthesis of a cyclophosphazene-based diphenylphosphine by lithiation
of two para-bromophenoxy groups was proposed. However, a systematic study showed that
this strategy led to a major product containing only one phosphine attached to the
cyclophosphazene ring, probably due to steric factors associated with this ligand.
Nevertheless, we have successfully prepared a new Pd(0) complex anchored to a
cyclophosphazene bearing two phosphine ligands. The Pd complex (4) was characterized as
Pd2[N3P3(O2C12H8)2(OC6H4PPh2)2(dba)2]. The geometry optimization, performed by DFT
with the functional PBE/def2-TZVP(-f), for both the ligand (3) and the Pd complex (4) were
successful, showing the best conformation of dba groups in the proposed structure among the
four possible structures. Our data are in a very good agreement with those found in the
Cambridge Crystallographic Database. This complex showed a good catalytic activity in
Suzuki-Miyaura coupling reactions. As expected, aryl halides containing electron-
withdrawing groups, and arylboronic acids with electron-donating substituents favored the
Suzuki-Miyaura coupling reactions.
Acknowledgments
C. G. thanks the CNPq for a fellowship (Grant number 134654/2015-3), as well as
FAPESP for financial support (Grant number 2017/12719-0). Running of TGA curves by Dr.
Daniela Zanchet’s research group is gratefully acknowledged. We also thank Eduardo
Guimarães Ratier de Arruda for fruitfull discussions.
References
[1] T.J. Colacot, The 2010 Nobel Prize in Chemistry: Palladium-Catalysed Cross-Coupling,
Platin. Met. Rev. 55 (2011) 84–90. doi:10.1595/147106711X558301.
[2] C.C.C. Johansson Seechurn, M.O. Kitching, T.J. Colacot, V. Snieckus, Palladium-
Catalyzed Cross-Coupling: A Historical Contextual Perspective to the 2010 Nobel Prize,
Angew. Chemie Int. Ed. 51 (2012) 5062–5085. doi:10.1002/anie.201107017.
[3] R. Jana, T.P. Pathak, M.S. Sigman, Advances in Transition Metal (Pd,Ni,Fe)-Catalyzed
Cross-Coupling Reactions Using Alkyl-organometallics as Reaction Partners, Chem.

Rev. 111 (2011) 1417–1492. doi:10.1021/cr100327p.
[4] F.-S. Han, Transition-metal-catalyzed Suzuki–Miyaura cross-coupling reactions: a
remarkable advance from palladium to nickel catalysts, Chem. Soc. Rev. 42 (2013) 5270.
doi:10.1039/c3cs35521g.
[5] V. Chandrasekhar, P. Thilagar, B. Murugesa Pandian, Cyclophosphazene-based multi-
site coordination ligands, Coord. Chem. Rev. 251 (2007) 1045–1074.
doi:10.1016/j.ccr.2006.07.005.
[6] V. Chandrasekhar, K.R.J. Thomas, Coordination and organometallic chemistry of
cyclophosphazenes and polyphosphazenes, Appl. Organomet. Chem. 7 (1993) 1–31.
doi:10.1002/aoc.590070102.
[7] V. Chandrasekhar, R.S. Narayanan, Pyridyloxy Cyclophosphazenes and
Carbophosphazenes: Inorganic Ring-Supported Coordination Platforms, Chim. Int. J.
Chem. 67 (2013) 64–70. doi:10.2533/chimia.2013.64.
[8] V. Chandrasekhar, A. Athimoolam, S.G. Srivatsan, P.S. Sundaram, S. Verma, A. Steiner,
S. Zacchini, R. Butcher, Pyrazolylcyclotriphosphazene Containing Pendant Polymers:
Synthesis, Characterization, and Phosphate Ester Hydrolysis Using a Cu(II)-Metalated
Cross-Linked Polymeric Catalyst, Inorg. Chem. 41 (2002) 5162–5173.
doi:10.1021/ic011159v.
[9] R.B. Pellegrino, R. Buffon, Encapsulation of vanadium complexes in inorganic or hybrid
matrices via the sol-gel method: application to the epoxidation of allylic alcohols, J. Braz.
Chem. Soc. 15 (2004) 527–531. doi:10.1590/S0103-50532004000400014.
[10]J.D. Ribeiro de Campos, R. Buffon, Entrapment of rhodium complexes in inorganic or
hybrid matrices via the sol–gel method, New J. Chem. 27 (2003) 446.
doi:10.1039/b207522a.
[11]S. Teixeira, K. Dallmann, U. Schuchardt, R. Buffon, Molybdenum-based epoxidation
catalysts heterogenized in silica matrixes via the sol–gel method, J. Mol. Catal. A Chem.
182–183 (2002) 167–173. doi:10.1016/S1381-1169(01)00459-9.
[12]W.L.F. Armarego, C.L.L. Chai, Purification of Laboratory Chemicals, 6a, John Wiley &
Sons, Ltd, Chichester, UK, 2009. http://doi.wiley.com/10.1002/047084289X.
[13]A. Savitzky, M.J.E. Golay, Smoothing and Differentiation of Data by Simplified Least
Squares Procedures., Anal. Chem. 36 (1964) 1627–1639. doi:10.1021/ac60214a047.
[14]J. Peng, S. Peng, A. Jiang, J. Wei, C. Li, J. Tan, Asymmetric least squares for multiple
spectra baseline correction, Anal.
doi:10.1016/j.aca.2010.08.033.
Chim.
Acta. 683 (2010) 63–68.

[15]G.A. Carriedo, L. Fernández-Catuxo, F.J. García Alonso, P. Gómez-Elipe, P.A.
González, Preparation of a New Type of Phosphazene High Polymers Containing 2,2‘-
Dioxybiphenyl
Groups,
Macromolecules.
29
(1996)
5320–5325.
doi:10.1021/ma951830d.
[16]H.R. Allcock, T.L. Evans, T.J. Fuller, Small-molecule cyclic models for the synthesis of
new polyphosphazenes: side-group construction via lithiophenoxy derivatives, Inorg.
Chem. 19 (1980) 1026–1030. doi:10.1021/ic50206a047.
[17]G Carriedo, Garc a lonso, P.A. González, P. Gómez-Elipe, Direct synthesis of
cyclic and polymeric phosphazenes bearing diphenylphosphine groups and their
complexes with [W(CO)5] fragments, Polyhedron. 18 (1999) 2853–2859.
doi:10.1016/S0277-5387(99)00194-1.
[18]S. Grimme, C. Bannwarth, P. Shushkov, A Robust and Accurate Tight-Binding Quantum
Chemical Method for Structures, Vibrational Frequencies, and Noncovalent Interactions
of Large Molecular Systems Parametrized for All spd-Block Elements ( Z = 1–86), J.
Chem. Theory Comput. 13 (2017) 1989–2009. doi:10.1021/acs.jctc.7b00118.
[19]F. Neese, The ORCA program system, Wiley Interdiscip. Rev. Comput. Mol. Sci. 2
(2012) 73–78. doi:10.1002/wcms.81.
[20]F. Weigend, R. Ahlrichs, Balanced basis sets of split valence, triple zeta valence and
quadruple zeta valence quality for H to Rn: Design and assessment of accuracy, Phys.
Chem. Chem. Phys. 7 (2005) 3297. doi:10.1039/b508541a.
[21]J.P. Perdew, K. Burke, M. Ernzerhof, Generalized Gradient Approximation Made
Simple, Phys. Rev. Lett. 77 (1996) 3865–3868. doi:10.1103/PhysRevLett.77.3865.
[22]F. Weigend, Accurate Coulomb-fitting basis sets for H to Rn, Phys. Chem. Chem. Phys.
8 (2006) 1057. doi:10.1039/b515623h.
[23]E.J. Baerends, D.E. Ellis, P. Ros, Self-consistent molecular Hartree—Fock—Slater
calculations I. The computational procedure, Chem. Phys. 2 (1973) 41–51.
doi:10.1016/0301-0104(73)80059-X.
[24]B.I. Dunlap, J.W.D. Connolly, J.R. Sabin, On first-row diatomic molecules and local
density models, J. Chem. Phys. 71 (1979) 4993. doi:10.1063/1.438313.
[25]C. Van Alsenoy, Ab initio calculations on large molecules: The multiplicative integral
approximation, J. Comput. Chem. 9 (1988) 620–626. doi:10.1002/jcc.540090607.
[26]R.A. Kendall, H.A. Früchtl, The impact of the resolution of the identity approximate
integral method on modern ab initio algorithm development, Theor. Chem. Accounts
Theory, Comput. Model. (Theoretica Chim. Acta). 97 (1997) 158–163.

doi:10.1007/s002140050249.
[27]K. Eichkorn, M. Htiser, R. Ahlrichs, K. Eichkorn, O. Treutler, H. Marco, R. Ahlrichs,
Auxiliary basis sets to approximate Coulomb potentials ( Chem . Phys . Letters 240
(1995)283)* Auxiliary basis sets to approximate Coulomb potentials, Chem. Phys. Lett.
240 (1995) 283–290. doi:http://dx.doi.org/10.1016/0009-2614(95)00838-U.
[28]R. Ahlrichs, S.D. Elliott, U. Huniar, Quantum chemistry: Large molecules - small
computers, Berichte Der Bunsengesellschaft Für Phys. Chemie. 102 (1998) 795–804.
doi:10.1002/bbpc.19981020602.
[29]J.L. Whitten, Coulombic potential energy integrals and approximations, J. Chem. Phys.
58 (1973) 4496–4501. doi:10.1063/1.1679012.
[30]S. Grimme, S. Ehrlich, L. Goerigk, Effect of the damping function in dispersion corrected
density functional theory, J. Comput. Chem. 32 (2011) 1456–1465.
doi:10.1002/jcc.21759.
[31]S. Grimme, J. Antony, S. Ehrlich, H. Krieg, A consistent and accurate ab initio
parametrization of density functional dispersion correction (DFT-D) for the 94 elements
H-Pu, J. Chem. Phys. 132 (2010) 154104. doi:10.1063/1.3382344.
[32]T.H. Fischer, J. Almlof, General methods for geometry and wave function optimization,
J. Phys. Chem. 96 (1992) 9768–9774. doi:10.1021/j100203a036.
[33]F. Neese, Approximate second-order SCF convergence for spin unrestricted
wavefunctions, Chem. Phys. Lett. 325 (2000) 93–98. doi:10.1016/S0009-2614(00)00662-
X.
[34]V.I. de Paula, C.A. Sato, R. Buffon, Pd complexes based on phosphine-linked
cyclophosphazenes: synthesis, characterization and application in Suzuki coupling
reactions, J. Braz. Chem. Soc. 23 (2012) 258–266. doi:10.1590/S0103-
50532012000200010.
[35]I. Dez, J. Levalois-Mitjaville, H. Grützmacher, V. Gramlich, R. de Jaeger, Syntheses of
Chiral Cyclotriphosphazenes and Their Use in Cyclolinear Polymers, Eur. J. Inorg.
Chem.
1999
(1999)
1673–1684.
doi:10.1002/(SICI)1099-
0682(199910)1999:10<1673::AID-EJIC1673>3.0.CO;2-D.
[36]C. Díaz, M.L. Valenzuela, S. Ushak, Synthesis and Characterization of
N3P3(O2C12H8)2(OC6H4Si(CH3)3)(OC6H4Br) and Its Conversion to Nanostructured Si
Material, J. Clust. Sci. 19 (2008) 471–479. doi:10.1007/s10876-008-0189-2.
[37]T.L. Evans, T.J. Fuller, H.R. Allcock, Preparation of [NP(p-OC6H4Li)2]3 by metal-
halogen exchange, and its reactions with electrophiles, J. Am. Chem. Soc. 101 (1979)

242–242. doi:10.1021/ja00495a052.
[38]B Çoşut, M Durmuş, K l ç, S Yeşilot, Synthesis, thermal and photophysical
properties of phenoxy-substituted dendrimeric cyclic phosphazenes, Inorganica Chim.
Acta. 366 (2011) 161–172. doi:10.1016/j.ica.2010.10.031.
[39]D. Kumar, A.D. Gupta, Aromatic Cyclolinear Phosphazene Polyimides Based on a Novel
Bis-Spiro-Substituted Cyclotriphosphazene Diamine, Macromolecules. 28 (1995) 6323–
6329. doi:10.1021/ma00122a045.
[40]M. Modesti, L. Zanella, A. Lorenzetti, R. Bertani, M. Gleria, Thermally stable hybrid
foams based on cyclophosphazenes and polyurethanes, Polym. Degrad. Stab. 87 (2005)
287–292. doi:10.1016/j.polymdegradstab.2004.07.023.
[41]C. DIAZ, Transition metal containing dendrimers based on cyclophosphazene units,
Polyhedron. 21 (2002) 909–915. doi:10.1016/S0277-5387(02)00852-5.
[42]G.A. Carriedo, F.J. García Alonso, P.A. González, J.R. Menéndez, Infrared and Raman
spectra of the phosphazene high polymer [NP(O2C12H8)]n, J. Raman Spectrosc. 29 (1998)
327–330. doi:10.1002/(SICI)1097-4555(199804)29:4<327::AID-JRS245>3.0.CO;2-5.
[43]J.R. Menendez, G.A. Carriedo, F.J. Garcia-Alonso, E. Clavijo, M. Nazri, R. Aroca,
Vibrational spectra of cyclic phosphazenes, J. Raman Spectrosc. 30 (1999) 1121–1125.
doi:10.1002/(SICI)1097-4555(199912)30:12<1121::AID-JRS512>3.0.CO;2-R.
[44]H.G.M. Edwards, I.R. Lewis, P.H. Turner, Raman and infrared spectroscopic studies of
tris triphenyl phosphine chloride complexes of transition metals, Inorganica Chim. Acta.
216 (1994) 191–199. doi:10.1016/0020-1693(93)03713-K.
[45]V Chandrasekhar, thimoolam, New Hybrid Inorganic−Organic Polymers as
Supports for Heterogeneous Catalysis: A Novel Pd(0) Metalated Cyclophosphazene-
Containing Polymer as an Efficient Heterogeneous Catalyst for the Heck Reaction, Org.
Lett. 4 (2002) 2113–2116. doi:10.1021/ol026098z.
[46]B.E. Mann, A. Musco, Phosphorus-31 nuclear magnetic resonance spectroscopic
characterisation of tertiary phosphine palladium(0) complexes: evidence for 14-electron
complexes in solution, J. Chem. Soc. Dalt. Trans. (1975) 1673.
doi:10.1039/dt9750001673.
[47]C. Dai, G.C. Fu, The First General Method for Palladium-Catalyzed Negishi Cross-
Coupling of Aryl and Vinyl Chlorides: Use of Commercially Available Pd(P(t-Bu)3)2 as a
Catalyst, J. Am. Chem. Soc. 123 (2001) 2719–2724. doi:10.1021/ja003954y.
[48]S. Otsuka, T. Yoshida, M. Matsumoto, K. Nakatsu, Bis(tertiary phosphine)palladium(0)
and -platinum(0) complexes: preparations and crystal and molecular structures, J. Am.

Chem. Soc. 98 (1976) 5850–5858. doi:10.1021/ja00435a017.
[49]A. Immirzi, A. Musco, Two-co-ordinate phosphine–palladium(0) complexes: X-ray
structure of the tricyclohexyl- and the di(t-butyl)phenyl-phosphine derivatives, J. Chem.
Soc., Chem. Commun. 68 (1974) 400–401. doi:10.1039/C39740000400.
[50]M. Majchrzak, S. Kostera, M. Kubicki, I. Kownacki, Synthesis of new styrylarenes via
Suzuki–Miyaura coupling catalysed by highly active, well-defined palladium catalysts,
Dalt. Trans. 42 (2013) 15535. doi:10.1039/c3dt52063c.
[51]P. Das, W. Linert, Schiff base-derived homogeneous and heterogeneous palladium
catalysts for the Suzuki–Miyaura reaction, Coord. Chem. Rev. 311 (2016) 1–23.
doi:10.1016/j.ccr.2015.11.010.
[52]A. Gniewek, Suzuki-Miyaura cross-coupling of phenylboronic acid with aryl halides
catalyzed by palladium and nickel species supported on alumina-based oxides, J.
Organomet. Chem. 823 (2016) 90–96.
[53]N. Miyaura, A. Suzuki, Palladium-Catalyzed Cross-Coupling Reactions of Organoboron
Compounds, Chem. Rev. 95 (1995) 2457–2483. doi:10.1021/cr00039a007.
[54]S. Ahammed, S. Nandi, D. Kundu, B.C. Ranu, One-pot Suzuki coupling of aromatic
amines via visible light photocatalyzed metal free borylation using t-BuONO at room
temperature, Tetrahedron Lett. 57 (2016) 1551–1554. doi:10.1016/j.tetlet.2016.02.097.

Synthesis and theoretical studies of a new Pd(0) complex with a cyclophosphazene-based
diphenylphosphine ligand. Activity in Suzuki-Miyaura cross-coupling reactions.