Thermal study of [Pd(2-Phpy)Cl(L)] complexes (L=pyridines and amines)

Article abstract of DOI:10.1023/A:1013118929427

The complex [Pd(2-Phpy)(μ-Cl)]₂ reacts with pyridines (L=pyridine, α-picoline and γ-picoline), amines (L=isopropylamine, tert-butylamine) and ammonia to form the corresponding ortho-palladated derivatives [Pd(2-Phpy)CIL]. The compounds have bee

Full text of DOI:10.1023/A:1013118929427

Journal of Thermal Analysis and Calorimetry, Vol. 66 (2001) 361–370
THERMAL STUDY OF [Pd(2-Phpy)Cl(L)] COMPLEXES
(L=PYRIDINES AND AMINES)
1
1
1
2
1
J. Pérez , G. Sánchez , J. García , J. L. Serrano and G. López
1
Departamento de Química Inorgánica, Campus Universitario de Espinardo, Universidad de
Murcia, 30071 Murcia, Spain
Departamento de Ingeniería Minera, Geológica y Cartográfica, Universidad Politécnica de
2
Cartagena, 30203, Cartagena, Spain
(
Received November 13, 2000; in revised form April 25, 2001)
Abstract
The complex [Pd(2-Phpy)(µ-Cl)] reacts with pyridines (L=pyridine, α-picoline and γ-picoline),
2
amines (L=isopropylamine, tert-butylamine) and ammonia to form the corresponding ortho-palladated
derivatives [Pd(2-Phpy)ClL]. The compounds have been characterized by C, H and N analyses and
1
13
spectroscopic methods (IR and H and C NMR). TG, DTG and DSC studies of the complexes were
carried out in dynamic nitrogen atmosphere. From DSC analyses the heats of decomposition were cal-
culated. The kinetics of the first step of thermal decomposition were evaluated from TG data by iso-
thermal methods for L=pyridine and isopropylamine. The activation energies obtained are in the range
–
1
9
0–100 kJ mol . The best fitting for data was observed for R2 and A1.5 kinetic models.
Keywords: kinetic analysis, ortho-palladated complexes, thermal behaviour
Introduction
Heteroaromatic ligands such as 2-phenylpyridine can easily be ortho-metalated by
2
Pd(II) salts via C(sp )–H bond cleavage [1] usually to give the corresponding acetate
or halide-bridged dimers. These dinuclear complexes have been thoroughly studied
[
2] and employed as precursors of mononuclear cyclometalates of general formula
M(C×N)LX (M=Pd; C×N=ortho-metalated ligand; L=neutral monodentate ligand
such as phosphines; X=halide) [3–5]. Thermal studies on organometallic compounds
of platinum metals group have been reported. In this sense, the thermal behaviour of
pentamethyl cyclopentadienylrhodium(III) [6–7], areneruthenium(II) [8] and ortho-
palladated [9] derivatives has shown the relatively high thermal stability of the
chloro-bridged dinuclear precursors.
Here we describe the thermal behaviour of ortho-palladated complexes of the
type [Pd(C×N)ClL] (C×N=2-phenylpyridine; L=pyridines and amines).
1
418–2874/2001/ $ 5.00
2001 Akadémiai Kiadó, Budapest
Akadémiai Kiadó, Budapest
©
Kluwer Academic Publishers, Dordrecht

3
62
PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
Experimental
Pyridines, amines and 2-phenylpyridine were obtained from commercial sources and
all the solvents were dried by conventional methods before use.
The complex [Pd(2-Phpy)(µ-Cl] was prepared according to previously pub-
2
lished methods [4].
Preparation of the complexes
The complexes [Pd(2-Phpy)ClL] (L=py, α-picoline, γ-picoline; NH , isopropyl-
amine, tert-butylamine) were obtained by reaction of the dinuclear complex
3
[
tion according to the following general method. An excess of neutral ligand
Pd(2-Phpy)(µ-Cl] with the corresponding pyridines and amines in (CH ) CO solu-
2
3 2
(
in acetone (25 cm ). When L=NH the ligand was added as a 20% aqueous solution.
0.65 mmol) was added to a solution of precursor [Pd(2-Phpy)(µ-Cl)] (0.162 mmol)
2
3
3
The mixture was then stirred at room temperature for 1 h and the resulting solution
was concentrated under reduced pressure to half original volume. The addition of di-
ethyl ether caused the formation of yellow solids, which were filtered off, washed
with diethyl ether and air dried.
Characterization
C, H and N analyses were performed with a Carlo Erba microanalyser. Conductivities
were measured with a Crison 525 conductimeter. IR spectra were recorded on a
Perkin Elmer 16F PC FT-IR spectrophotometer using Nujol mulls between polyeth-
1
13
ylene sheets. H NMR data were recorded on a Bruker AC 200E and C NMR data
were recorded on a Varian Unity 300 instrument.
[Pd(2-Phpy)Cl(py)]
Yield 70%. Analysis: found (%): C, 51.0; H, 3.3; N, 7.8; calcd.: C, 51.2; H, 3.5;
1
–
1
N, 7.5. Non-conductor in acetone. IR (cm ) (Nujol): 756 (py); 324 (ν PdCl). H NMR
(
δ) (solvent CDCl ; reference SiMe ): 9.40 (d, 1H), 8.90 (d, 2H), 7.79 (m, 2H), 7.60
3
4
13
(d, 1H), 7.41 (m, 3H), 7.10 (m, 2H), 6.87 (m, 1H), 6.14 (d, 1H). C NMR (δ) (solvent
CDCl ; reference SiMe ): 165.5, 154.8, 153.3 (py), 152.2, 145.7, 138.8 (py), 138.0,
3
4
1
32.5, 129.5, 125.5 (py), 124.8, 123.3, 122.1, 118.3.
[Pd(2-Phpy)(Cl(α-picoline)]
Yield 77%. Analysis: found (%): C, 52.2; H, 3.5; N, 7.0; calcd.: C, 52.5; H, 3.9;
1
–
N, 7.21. Non-conductor in acetone. IR (cm ) (Nujol): 774, 756 (α-picoline); 324
1
(
ν PdCl). H NMR (δ) (solvent CDCl ; reference SiMe ): 9.40 (d, 1H), 9.92 (d, 1H,
3
4
α-pic), 7.72 (m, 2H), 7.60 (d, 1H), 7.39 (m, 2H), 7.06 (m, 3H), 6.81 (t, 1H), 5.87 (d,
1
3
1
H), 3.05 (s, 3H, Me). C NMR (δ) (solvent CDCl ; reference SiMe ): 165.4, 160.3
3
4
(
1
α-pic), 153.5 (α-pic), 152.4, 151.9 (α-pic), 145.7, 138.7 (α-pic), 137.7, 132.3,
29.5, 126.3 (α-pic), 124.7, 123.3, 122.4, 122.1, 118.2, 27.8 (Me).
J. Therm. Anal. Cal., 66, 2001

PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
363
[Pd(2-Phpy)(Cl(γ-picoline)]
Yield 75%. Analysis: found (%): C, 52.2; H, 3.4; N, 7.1; calcd.: C, 52.5; H, 3.9;
–
1
N, 7.2. Non-conductor in acetone. IR (cm ) (Nujol): 818, 750 (γ-picoline); 326
1
(
ν PdCl). H NMR (δ) (solvent CDCl ; reference SiMe ): 9.39 (d, 1H), 8.72 (d, 2H,
3
4
γ-pic), 7.75 (t, 1H), 7.59 (d, 1H), 7.40 (d, 1H), 7.21 (d, 2H, γ-pic), 7.08 (m, 2H), 6.87
1
3
(
t, 1H), 6.18 (d, 1H), 2.40 (s, 3H, Me). C NMR (δ) (solvent CDCl ; reference
3
SiMe ): 165.5, 154.5, 152.5, 150.1 (γ-pic), 145.7, 138.7 (γ-pic), 132.6, 129.5, 126.4
4
(
γ-pic), 124.7, 123.3, 122.0, 118.2, 30.9 (Me).
[
Pd(2-Phpy)(Cl(NH )]
3
Yield 84%. Analysis: found (%): C, 42.2; H, 3.3; N, 8.6; calcd.: C, 42.0; H, 3.5;
–
1
1
N, 8.9. Non-conductor in acetone. IR (cm ) (Nujol): 3310, 3208 (NH ). H NMR (δ)
3
(
2
solvent CDCl ; reference SiMe ): 8.95 (2, 1H), 7.60 (m, 1H), 7.31 (m, 3H), 6.91 (m,
H), 6.72 (m, 1H), 2.82 (s, 3H, NH ).
3
3
4
[
Pd(2-Phpy)Cl(iPrNH )]
2
Yield 71%. Analysis: found (%): C, 47.4; H, 4.3; N, 7.4; calcd.: C, 47.2; H, 4.8;
–
1
N, 7.9. Non-conductor in acetone. IR (cm ) (Nujol): 3268, 3210, 3138 (iPrNH ).
2
1
H NMR (δ) (solvent CDCl ; reference SiMe ): 9.01 (d, 1H), 7.57 (t, 1H) 7.32 (d, 1H),
.14 (m, 1H), 6.91 (m, 3H), 6.77 (m, 1H), 3.36 (m, 3H), 1.37 (d, 6H). C NMR (δ)
solvent CDCl ; reference SiMe ): 164.5, 152.5, 151.5, 145.8, 138.0, 131.1, 129.0,
3
4
1
3
7
(
3
4
1
24.4, 123.2, 121.9, 117.8, 53.7 (CH-iPr), 32.5 (Me).
[
Pd(2-Phpy)Cl(tert-BuNH )]
2
Yield 72%. Analysis: found (%): C, 48.4; H, 4.8; N, 7.3; calcd.: C, 48.6; H, 5.1;
–
N, 7.6. Non-conductor in acetone. IR (cm ) (Nujol): 3270, 3211, (tert-BuNH ).
1
2
1
H NMR (δ) (solvent CDCl ; reference SiMe ): 9.24 (d, 1H), 7.68 (t, 1H), 7.47 (d,
1H), 7.29 (d, 1H), 6.94 (m, 4H), 3.34 (s, NH ), 1.44 (s, 9H, Me). C NMR (δ) (solvent
CDCl ; reference SiMe ): 165.0, 151.7, 151.3, 146.2, 138.4, 132.7, 128.4, 124.5,
3
4
1
3
2
3
4
1
23.5, 122.0, 118.0, 53.7 (C-tert-Bu), 32.5 (Me).
Thermal analysis
Thermoanalytical data were obtained from TG, DTG and DSC curves. TG and DTG
curves were recorded on a Mettler TA-3000 system provided with a Mettler TG-50
thermobalance and DSC curves were recorded on a DSC-20 Mettler instrument. The
3
–1
atmospheres used in different experiments were pure nitrogen (50 cm min ). The
sample mass range was of 4–7 mg.
J. Therm. Anal. Cal., 66, 2001

–
1
Table 1 TG and DTG data for the neutral palladium(II) complexes (under dynamic nitrogen atmosphere; heating rate 5°C min
Temperature
range/°C
DTGmax
/
Mass loss/%
found(calc.)
Enthalpy change/
Complex
Step
Assignment
–1
°C
kJ mol
1
2
85–145
286–437
>440
137
347
–
20.2 (21.1)
45.0 (50.5)
32.2 (28.4)
py
[
[
[
[
[
[
Pd(2-Phpy)Cl(py)]
Pd(2-Phpy)Cl(α-pic)]
Pd(2-Phpy)Cl(γ-pic)]
2-Phpy+Cl
Pd+C
28.7
69.7
70.3
25.4
88.4
62.6
residue
1
2
104–167
290–430
>450
160
351
–
23.3 (23.9)
44.7 (48.8)
29.4 (27.3)
α-pic
2-Phpy+Cl
Pd+C
residue
1
2
106–173
284–442
>450
165
347
–
22.9 (23.9)
44.0 (48.8)
31.7 (27.3)
γ-pic
2-Phpy+Cl
Pd+C
residue
1
2
90–165
291–445
>460
125
337
–
5.2 (5.4)
53.1 (60.6)
38.3 (34.0)
NH
3
2-Phpy+Cl
Pd+C
Pd(2-Phpy)Cl(NH
3
)]
residue
1
2
90–142
290–444
>450
135
335
–
16.0 (16.8)
47.1 (53.4)
33.7 (29.8)
iPr-NH
2
2-Phpy+Cl
Pd+C
Pd(2-Phpy)Cl(iPrNH
2
)]
residue
1
2
72–127
288–446
>460
118
347
–
18.7 (19.7)
46.1 (51.5)
32.1 (28.8)
terc-Bu-NH
2-Phpy+Cl
Pd+C
2
2
Pd(2-Phpy)Cl(terc-BuNH )]
residue

PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
365
Results and discussion
Thermal stability
The TG and DSC data of the complexes are summarized in Table 1. All the com-
plexes decompose on heating to give the binuclear complex [Pd(2-Phpy)µ-Cl)] , ac-
cording to Eq. (1).
2
[
Pd(2-Phpy)ClL](s)→1/2[Pd(2-Phpy)(µ-Cl)] (s)+L(g)
(1)
2
where L=pyridine, α-picoline, γ-picoline, NH , iPrNH , tert-BuNH
3
2
2
The chloro-bridged binuclear complex could be isolated in every case and identified
1
by IR and H-NMR spectroscopy. The enthalpy changes for the first step were measured
by integration of the endothermic peaks in the corresponding DSC curves (Table 1). The
binuclear intermediates decomposes irregularly between 286 and 446°C.
The complex [Pd(2-Phpy)Cl(py)] loses one pyridine molecule between 85 and
1
[
1
45°C while the complexes with substituted pyridines are more stable:
Pd(2-Phpy)Cl(α-picoline)] and [Pd(2-Phpy)Cl(γ-picoline)] decompose at 104 and
06°C respectively. The enthalpy change for the loss of α-pic and γ-pic is also larger
than for pyridine; this fact is in accordance with the stronger donor character of sub-
stituted pyridines. The thermal stability of the complex with iPrNH is larger than that
of the terc-BuNH derivative in accordance with the values of enthalpy found by
2
2
DSC (Table 1). In this case, the results suggest that the steric effects are the most im-
portant factors in the thermal stability.
Kinetic analysis
Different methods such as linear rising temperature, constant rate thermal analysis
[
10] or high resolution thermogravimetry [11] have been used in kinetic investiga-
tions but the most reliable method for determining kinetic parameters is care-
fully-controlled isothermal decomposition in which mass loss at a constant tempera-
ture is monitored as a function of time [11].
The isothermal TG curves for palladium complexes [Pd(2-Phpy)Cl(py)] and
Pd(2-Phpy)Cl(iPrNH )] at different temperatures are shown in Fig. 1. The delay t ,
[
2
0
preceding the onset of the main reaction usually include contributions of the time re-
quired for the sample to attain the reaction temperature. In order to minimize the un-
certainty introduced by that time we used the plot α vs. (t–t ) [12].
0
The differential kinetic equation can be expressed as
dα
−
x
=
Ae f (α )
(2)
dt
where A is a pre-exponential factor for the Arrhenius type rate constant and x the reduced
apparent activation energy (x=E /RT). The function f(α) in Eq. (2) is an analytical expres-
sion describing the kinetic model of the studied thermal decomposition process.
a
J. Therm. Anal. Cal., 66, 2001

3
66
PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
Fig. 1 α vs. time plots for the first step of isothermal decomposition of
a – [Pd(2-Phpy)Cl(py)] and b – [Pd(2-Phpy)Cl(iPrNH
2
)];
–
× –115°C; – Ÿ –118°C; — 121°C; – + – 124°C; – –127°C; – n –130°C
*
Isoconversional method
By integration of Eq. (2) in isothermal conditions the following equation is obtained:
–
x
g(α)=Ae t
(3)
The apparent activation energy of the decomposition process in isothermal con-
ditions can be calculated by isoconversional method which follows from logarithmic
form of Eq. (3):

lnt=ln

g(α ) Ea
 +
(4)
A  RT
The slope of lnt vs. 1/T for the same value of α gives the value of apparent acti-
vation energy. This procedure can be repeated for various values of α. The plots of Ea
Fig. 2 Plots of apparent activation energy calculated from isothermal TG data by the
isoconversional method as a function of α for a – [Pd(2-Phpy)Cl(py)] and
2
b – [Pd(2-Phpy)Cl(iPrNH )]
J. Therm. Anal. Cal., 66, 2001

PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
367
as a function of fractional conversion are shown in Fig. 2 within the certain error lim-
its (specified by bars). The average values of apparent activation energy determined
in the 0.3<α<0.9 range are 96±9 and 97±10 for the complexes [Pd(2-Phpy)Cl(py)]
and [Pd(2-Phpy)Cl(iPrNH )] respectively.
2
Plots of the integrated rate functions, g(α), vs. time
The linearity of the plots of the integrated rate functions, g(α), vs. time over the range
0
.3≤α≤0.9 was assessed using the correlation coefficient (r). It was difficult to distin-
guish between applicability of the Am and Rn models. Arrhenius plots constructed
using k values from the acceptable g(α) functions gave the apparent activation ener-
gies and the pre-exponential factors presented in Table 2. The values of E are almost
a
independent of the physical model proposed, a fact which is well established in the
literature [13]. Besides, the values of E are similar to that found by the
isoconversional method.
a
Table 2 Arrhenius parameters for the isothermal reactions
–1
–1
Models
A1.5
a
E /kJ mol
r
ln(A/s )
[
[
Pd(2-Phpy)Cl(py)]
97±6
94±9
0.9969
0.9989
22.8±1.8
23.5±2.1
Pd(2-Phpy)Cl(iPrNH
2
2
2
2
2
)]
)]
)]
)]
)]
A2
[
Pd(2-Phpy)Cl(py)]
Pd(2-Phpy)Cl(iPrNH
98±7
94±8
0.9946
0.9970
22.4±2.0
23.4±1.9
[
A3
[
Pd(2-Phpy)Cl(py)]
Pd(2-Phpy)Cl(iPrNH
99±9
95±8
0.9945
0.9981
22.0±1.6
22.9±2.1
[
R2
[
Pd(2-Phpy)Cl(py)]
Pd(2-Phpy)Cl(iPrNH
98±9
95±7
0.9961
0.9980
22.3±1.8
22.4±1.7
[
R3
[
Pd(2-Phpy)Cl(py)]
Pd(2-Phpy)Cl(iPrNH
98±8
94±7
0.9974
0.9971
22.0±2.5
22.3±1.9
[
Reduced-time plots
For an isokinetic process, if the times required to attain a set value of α (e.g., α=0.5)
at different (isothermal) temperatures are determined (t ), then plots of α vs. reduced
time (t =t/t ) can be prepared [14]. This method consists in comparing the experi-
0
.5
red
0.5
mental data (in the form of reduced time) with the well-known calculated data for the
models in Table 2. As can be seen from Fig. 3, the best fits of the experimental results
is for A1.5 and R2 models.
J. Therm. Anal. Cal., 66, 2001

3
68
PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
Fig. 3 Plots of reduced-time for a – [Pd(2-Phpy)Cl(py)] ⋅⋅⋅ R2; --- R3;
b – [Pd(2-Phpy)Cl(py)] --A2; A3; — A1.5; c – [Pd(2-Phpy)Cl(iPrNH
2
)])
⋅
⋅⋅ R2; --- R3; and d – [Pd(2-Phpy)Cl(iPrNH
2
)]) --A2; A3; — A1.5;
T= – × –115°C; – w –118°C; – ž – 121°C; – + –124°C; – * –127°C; – o –130°C
Method of Hancock–Sharp
In order to discriminate between models R2 and A1.5 we used the method of Han-
cock–Sharp [15] based in the following equation:
ln[–ln(1–α)]=nlnt+B
(5)
The plotting of ln[–ln(1–α)] vs. lnt for α values ranging from 0.05 to 0.65 allows
the n parameter to be determined. This parameter is characteristic of the reaction
mechanism [16], being n=1.09 and n=1.5 for R2 and A1.5 mechanisms respectively.
The results of the above plot are presented in Table 3, the found values for n pa-
rameter are closer to 1.09, so we can conclude that R2 is the most probable reaction
mechanism.
J. Therm. Anal. Cal., 66, 2001

PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
369
Table 3 Hancock–Sharp parameters for the isothermal reactions
[
Pd(2-Phpy)Cl(py]
2
[Pd(2-Phpy)Cl(iPrNH )]
T/°C
n
r
n
r
1.16±0.07
1.16±0.05
1.15±0.04
1.14±0.04
1.07±0.04
1.10±0.05
0.99376
0.99670
0.99750
0.99748
0.99674
0.99605
115
118
121
124
127
130
1.14±0.06
1.15±0.03
1.12±0.03
1.14±0.09
1.20±0.05
1.03±0.07
0.99699
0.99968
0.99945
0.99418
0.99926
0.99580
Conclusions
The results presented above show that the thermal stability of complexes
Pd(2-Phpy)Cl(L)] with substituted pyridines (L=α-picoline, γ-picoline) is higher
[
than that of the complex with pyridine. On the other hand, the complex with iPrNH2
shows a larger thermal stability than that with tert-BuNH2.
With regard to the kinetic of the first step of thermal decomposition we found for
the complexes with L=pyridine, iPrNH similar values of activation energies (in the
2
–
1
range 90–100 kJ mol ) and the models that best fit for data are R2 and A1.5 in both
cases. It can be established by the Hancock–Sharp method that R2 is the most proba-
ble mechanism.
The release of ligand yields the same product, [Pd(2-Phpy)(µ-Cl)] , in all the
cases. This compound shows a relatively high thermal stability, this fact probably is
2
related with the similar values of E found for the two complexes studied in the pres-
ent work.
a
References
1
2
A. Kasahara, Bull. Chem. Soc. Jpn., 41 (1968) 1272.
E. C. Constable, A. M. W. Cargill Thomson, T. A. Leese, D. G. F. Reese and D. A. Tocher,
Inorg. Chim. Acta, 182 (1991).
3
4
5
D. L. Weaver, Inorg. Chem., 9 (1970) 2250.
M. A. Gutierrez, G. R. Newkome and J. Selbin, J. Organomet. Chem., 202 (1980) 341.
J. M. Vila, M. Gayoso, M. Pereira, A. Romar, J. J. Fernandez and M. Thornton-Pett,
J. Organomet. Chem., 401 (1991) 385.
6
G. Sánchez, I. Solano, M. D. Santana, G. García, J. Gálvez and G. López, Thermochim. Acta,
2
11 (1992) 163.
7
8
G. Sánchez, J. García, J. Pérez, G. García and G. López, Thermochim. Acta, 307 (1997) 127.
G. Sánchez, J. García, J. Pérez, G. García, G. López and G. Villora, Thermochim. Acta, 293
(1997) 153.
9
L. Tusek-Bozic, M. Curic and P. Traldi, Inorg. Chim. Acta, 254 (1997) 49.
1
0 M. Reading, D. Dollimore, J. Rouquerol and F. Rouquerol, J. Thermal Anal., 29 (1984) 775.
J. Therm. Anal. Cal., 66, 2001

3
70
PÉREZ et al.: ORTHO-PALLADATED COMPLEXES
1
1
1 B. A. Howell and B. B. S. Sastry, Thermochim. Acta, 340–341 (1999) 311.
2 W. E. Brown, D. Dollimore and A. K. Galwey, Comprehensive Chemical Kinetics,
C. H. Bamford and C. F. H. Tipper (Ed.), Elsevier, Amsterdam 1980, Vol. 22, p. 80.
3 J. M. Criado, M. González and C. Real, J. Thermal Anal., 29 (1984) 243.
4 J. H. Sharp, G. W. Brindley and B. N. N. Achar, J. Amer. Ceram. Soc., 49 (1966) 379.
5 J. D. Hancock and J. H. Sharp, J. Am. Ceram. Soc., 55 (1972) 74.
1
1
1
1
6 J. M. Criado, Thermochim. Acta, 39 (1980) 361.
J. Therm. Anal. Cal., 66, 2001