Sublimation kinetics of scandium β-diketonates

Article abstract of DOI:10.1007/s10973-009-0167-4

The volatile scandium-β-diketonates are synthesised using acetylacetone (acac) and tetramethylheptanedione (tmhd) as coordinative ligands for CVD application. The X-ray powder patterns are indexed and analysed and found to be orthorhombic and monoclinic p

Full text of DOI:10.1007/s10973-009-0167-4

J Therm Anal Calorim (2010) 100:155–161
DOI 10.1007/s10973-009-0167-4
Sublimation kinetics of scandium b-diketonates
J. Selvakumar Æ V. S. Raghunathan Æ
K. S. Nagaraja
Received: 16 February 2009 / Accepted: 22 April 2009 / Published online: 19 June 2009
´
´
ꢀ Akademiai Kiado, Budapest, Hungary 2009
Abstract The volatile scandium-b-diketonates are
synthesised using acetylacetone (acac) and tetramethyl-
heptanedione (tmhd) as coordinative ligands for CVD
application. The X-ray powder patterns are indexed and
analysed and found to be orthorhombic and monoclinic
phases for Sc(acac)3(1) and Sc(tmhd)3(2) respectively. The
sublimation and evaporation kinetics have been analyzed
using three calculating techniques. The non-isothermal
based activation energy values are found to be 38 2 and
73 2 kJ mol-1 by Flynn–Wall technique for (1) and (2)
respectively. The measured Ea values are close to the value
obtained using Kissinger method.
dielectric, protective layers and catalysis. The precursors
need to be designed in such a way that the compounds are
volatile and thermally stable at growth temperatures [1, 2].
The volatility and thermal stability of b-diketonate of metals
in different oxidation states have been reported [3]. How-
ever, the studies on sublimation kinetics of Sc-b-diketonate
have not been reported. TG has been used to obtain the
thermal decomposition parameters of solid chemicals [4].
These approaches are based on monitoring weight change as
an indicator of solid-state conversion and may be performed
under isothermal, non-isothermal, or modulated conditions
[5, 6]. The data generated from these experiments can be
analyzed and manipulated by either model [7] or model-free
[8] approaches to obtain Arrhenius kinetic parameters such
as activation energy (Ea) and pre-exponential factor [9].
The present study was carried out for the sublimation
kinetics [10] of volatile tris(2,4-pentanedionato)scan-
dium(III) (Sc(acac)3) and tris(2,2,6,6-tetramethyl-3,5-hep-
tanedionato)scandium(III) (Sc(tmhd)3) by means of
thermal analysis (TG and DTG).
Keywords Powder X-ray diffraction ꢀ
Scandium-b-diketonate ꢀ Sublimation kinetics
Introduction
Metal b-diketonate complexes that are of particular rele-
vance to this paper have been used as highly volatile pre-
cursors for chemical vapor deposition of films with novel
multifunctionalities and a wide variety of applications in
modern high technology, including semiconductor micro-
electronics, damage resistance, antireflection, and high-
reflection multilayer coatings in light emitting diodes and
high power pulsed solid state UV-lasers, ferroelectric,
Experimental section
Materials
Crystalline scandium-b-diketonates are obtained from the
reaction of scandium nitrate with corresponding equiva-
lents of 2,4-pentanedione or 2,2,6,6-tetramethyl-3,5-hep-
tanedione in basic medium [11].
J. Selvakumar ꢀ K. S. Nagaraja (&)
Department of Chemistry, Loyola Institute of Frontier Energy
Loyola College, Chennai 600 034, India
e-mail: dr.ksnagaraja@gmail.com
Instrumentation
V. S. Raghunathan
Non Ferrous Material Technology Development Centre,
Hyderabad 500 050, India
1
The H and 13C NMR spectra of the scandium-b-diketo-
nates solutions were recorded with a Bruker spectrometer.
123

156
J. Selvakumar et al.
Melting points were obtained in a sealed capillary using
optical microscopy. To determine the thermal behavior of
the complexes besides the congruent nature of vaporization
to nil residues under reduced pressure (2 mbar), vapor or
mass transport study under vacuum was carried out. The
experimental conditions are the same as described earlier
[12]. Mass spectra of the samples were recorded with a
Micromass LCT time-of-flight mass spectrometer operat-
ing in electron spray mode. The X-ray diffraction analyses
of the complexes were carried out using an upgraded
Philips PW-1730 X-ray diffraction system interfaced with
VisXRD software. Horizontal Perkin-Elmer Pyris-Dia-
mond TG-DTA thermal analyzer was adopted for non-
isothermal and isothermal analyses. The measurement was
carried out in platinum crucible under high pure nitrogen,
which was dried by passing through molecular sieves
(Linde 4A) with a flow rate of 12 dm3 h-1 at a linear
heating rate of 10 K min-1 in ambient pressure. Non-iso-
thermal TG measurements are run at 4, 6, 8 and
10 K min-1 with a sample size of 1–6 mg to provide a
control set of values for thermal sublimation parameters.
The experimental conditions are the same as described
earlier [13, 14]. Prior to the isothermal measurements,
blank runs were carried out with a flow rate of 20 dm3 h-1
of dry nitrogen gas, from room temperature to 1173 K at a
heating rate of 10 K min-1 keeping the reference and
sample pans empty. The system was then cooled to
ambient temperature under the same conditions. The
powdered sample was spread out on a platinum crucible
and was flushed with nitrogen at a rate of 6 dm3 h-1 at
ambient temperature. The initial heating to *360 K, is
rather rapid (10 K min-1), and after temperature stabili-
sation, the sample is subsequently heated (2 K min-1) in
steps of 10 K to isothermal temperatures in the range
376–407 K and 385–475 K for (1) and (2), respectively.
where E is the activation energy, A is the preexponential
factor, R the gas constant and T the absolute temperature.
The plot of ln k versus 1/T is linear and from the slope, the
activation energy (Ea) for the sublimation of the complex
was calculated.
The proposed kinetic analysis, based on dynamic model-
free methods using data obtained at different fixed heating
rates, b, seems to be the most reliable approach. The
thermal sublimation kinetics is examined using Kissinger
Technique [17–19]:
ꢀ
0
1
nꢁ1
AR nð1 ꢁ aÞm
Ea
b
Tm2
Ea
@
A
ln
¼ ln
ꢁ
ð3Þ
RTm
where Tm is the DTG peak temperature at a given b value,
A the pre-exponential factor, Ea the activation energy and R
the gas constant. The Ea value can be calculated from the
slope of ln (b/T2m) as a function of 1/Tm.
Kinetic study of sublimation steps was also performed
using the isoconversional method of Flynn–Wall Tech-
nique [20]. The isoconversional expression is,
ꢁ
ꢂ
AE
R
Ea
lnb ¼ ln
ꢁ ln½FðaÞꢂ ꢁ
ð4Þ
RT
The activation energy, Ea can be calculated from a plot of
ln b versus 1/T at a fixed weight loss since the slope of such
a line is given by Ea/R (gas constant 8.314 J mol-1 K-1).
The ln A is calculated from the intercept value of the line
and the derived Ea value.
For isothermal TG measurements at several isothermal
temperatures [21]:
ꢁ
ꢂ
da
ln
m ¼ lnA þ n lnð1 ꢁ amÞ ꢁ Ea=RT
ð5Þ
dt
The Ea value of isothermal degradation was calculated
from the slope of the plot of the maximum weight-loss rate,
ln(da/dt)m versus the reciprocals of the several isothermal
temperatures.
Kinetic procedure
Solid state decomposition kinetics is usually described by
the following basic equation:
Results and discussion
da
¼ kðTÞfðaÞ
ð1Þ
dt
The scandium-b-diketonates are systematically character-
1
1
ised by H NMR and Mass Spectral studies. H and 13C
where a is the degree of reaction defined as a = (wi - wT)/
(wi - wf) (being wi and wf the initial and final mass loss,
while wT is the mass loss at temperature T), f(a) is the model
function, which assumes different mathematical forms
depending on the reaction mechanism [15] and k(T) is the
specific rate constant, whose temperature dependence is
commonly described by the Arrhenius equation [16]:
NMR (298 K, 300 MHz, CDCl3, dH, ppm) of Sc(acac)3(1):
•
1.997 (s, 6H, CH3), 5.505–5.552 (s, 1H, =CH); 13C NMR
(298 K, 75 MHz, CDCl3, d, ppm): 26.84–27.12 (•CH3),
103.32 (=CH) and 191.82 (•C=O). ES-MS: m/z = 342 and
243 in the spectrum are due to the fragmentation of
[Sc(acac)3]? and [Sc(acac)2]? from the molecular mass
1
(343). Sc(tmhd)3(2) demonstrates the following: H NMR
lnkðTÞ ¼ ln A ꢁ Ea=RT
ð2Þ
(298 K, 300 MHz, CDCl3, dH, ppm): 0.915, 1.206, 1.337
123

Sublimation kinetics of scandium b-diketonates
157
(s, 6H, CH3), 5.800 (s, 1H, =CH); 13C NMR (298 K,
•
100
80
60
40
20
0
75 MHz, CDCl3, d, ppm): 28.07 (•CH3), 40.38(C–CH3),
92.04 (=CH), and 200.89 (•C=O). NMR chemical shifts
[22] are close to those observed for olefinic protons, and
Sc(acac)3 (1)
•
they are independent of the size, charge, and -bonding
ability of the metal ion. Holm and Cotton [23] had earlier
examined the –CH = resonances of several neutral metal
acetylacetonates in carbon tetrachloride solution. The
melting and sublimation/vaporization to nil residues for
Sc(acac)3 and Sc(tmhd)3 under reduced pressure were
carried out to demonstrate the clean volatilisation behav-
iour. The observed melting points from this study (at
*2 mbar) were 453 and 417 K for Sc(acac)3 and Sc(tmhd)3,
respectively. The sublimed product was found to have the
same XRD and IR as the initial product taken. ES-MS: m/z
values 594, 537, 411, 185 and 127 corresponds to
[Sc(tmhd)3]?, [Sc(tmhd)3-C4H9]?, [Sc(tmhd)2]?, (tmhd)?
and [(tmhd)–C4H8]?. The powder X-ray diffraction patterns
of Sc(acac)3 and Sc(tmhd)3 were recorded in 2h range of 5ꢁ to
70ꢁ, measured in steps and speed of 0.02ꢁ and 2ꢁ min-1
respectively. The powder diffraction pattern is indexed by
the programme DICVOL04 technique in FullProf package
[24]. The full pattern fitting, peak decomposition and space
groups are performed using the program chekcell in
CRYSFIRE package. The profile is fitted using the pseudo-
Voigt function [25]. The X-ray data analysis of crystalline
β
= 4, 6, 8 and 10 K min-1
-20
-40
300
350
400
450
500
550
T/K
100
80
60
40
20
0
Sc(tmhd)3 (2)
β
= 4, 6, 8 and 10 K min-1
-20
-40
˚
powder (1) gave: a = 16.79 A, b = 15.47 A; c = 13.71 A,
˚
˚
3
˚
a = b = c = 90ꢁ, V = 3560.42 A , space group = Pbca,
300
350
400
450
500
550
crystal system: orthorhombic which are in agreement with
the single crystal data [26]. The results revealed the crys-
talline nature and phase purity of the compound. The powder
T/K
Fig. 1 The effect of the heating rate on the TG curves of the
Sc(acac)3 (1) and Sc(tmhd)3 (2)
˚
˚
pattern of (2) gave: a = 24.44 A, b = 18.34 A, c = 17.59 A,
˚
3
˚
a = c = 90ꢁ, b = 96.34ꢁ, V = 7839.76 A , space
group = P21, crystal system = monoclinic which agreed
with the single crystal data [27].
Sc(acac)3(1)
Sc(tmhd)3(2)
560
540
520
500
480
460
440
420
Tf
Tp
520
500
480
460
440
420
Non-isothermal kinetics
A strong and regular effect of the heating rate on the non-
isothermal TG curves of the Sc(acac)3 and Sc(tmhd)3 are
shown in Fig. 1. The volatilization processes are a single-
step process in the defined temperatures ranges. The sub-
limation temperature rises with the increasing of the
heating rate. Figure 2 shows that the heating rates effect on
the temperatures at the maximum weight-loss rate (Tp), at
the beginning (To) and at the end (Tf) of the weight-loss of
Sc(acac)3 and Sc(tmhd)3 TG, respectively. It is found that
the values of Tp, To and Tf increases of heating rates.
Linear fittings of To, Tp and Tf versus b/K min-1 gives
Eq. 6–8, respectively.
Tf
Tp
To
To
4
5
6
7
8
9
10
Heating Rate (
β
/K min-1)
Fig. 2 Plots of the sublimation temperatures versus heating rates
123

158
J. Selvakumar et al.
Table 1 The sublimation kinetics of the Sc(acac)3 and Sc(tmhd)3 by various methods
Non-isothermal process
Isothermal process
Compound
1
Kissinger
Ea/kJ mol-1
Flynn–Wall
Arrhenius Plot
1
a
Ea/kJ mol-1
b/Jm in-
Ea/kJ mol-1
Ea/kJ mol-1
38
2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
42
40
38
37
37
37
36
36
38
71
78
76
74
73
72
73
72
73
3
2
2
2
2
1
2
2
2
3
5
6
7
6
6
6
8
6
4
6
114a
82b
117a
81b
112a
78b
83b
81 2 (375–450 K)a
8
10
38
67
2
5
2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
4
6
80b
79b
79b
78b
95 3 (356–414 K)a
75 3 (434–465 K)b
8
10
67
5
a conversion; b heating rate
a
Sublimation process
b
vaporisation process
ScðacacÞ3ð1Þ ScðtmhdÞ3ð2Þ
-10.2
-10.4
-10.6
-10.8
To ¼ 5:9b=K minꢁ1 þ 398 K;
ð6Þ
ð7Þ
ð8Þ
To ¼ 3:1b=K minꢁ1 þ 428 K
Tp ¼ 7:7b=K minꢁ1 þ 424K;
Tp ¼ 4:1b=K minꢁ1 þ 466K
(1)
Tf ¼ 8:3b=K minꢁ1 þ 431 K;
Tf ¼ 4:2b=K minꢁ1 þ 478 K
(2)
-11.0
She equilibrium sublimation temperatures are gained by
extrapolation of To, Tp and Tf to b = 0: Too = 398, 428 K,
Top = 424, 466 K and Tof = 431, 478 K for (1) and (2),
respectively.
0.00190
0.00195
0.00200
K/T
0.00205
0.00210
Fig. 3 Plots of ln[(b/T2)/min-1] versus 1/T at different heating rates
according to Kissinger method
According to the equations of Kissinger and Flynn–Wall
techniques the activation energy of sublimation for (1) and
(2) are measured (Table 1). Using Kissinger method, Eq. 3,
the activation energy can be calculated from the plot of
ln(b/T2m) versus 1000/Tm (Fig. 3). The obtained activation
energy values are 38 2 and 67 5 kJ mol-1 for (1) and
(2), respectively. The activation energy can also be deter-
mined using the method Flynn–Wall technique, from the
linear fitting of ln b versus 1/T at different conversions
values, which is from 0.1 to 0.8 (Fig. 4). Figure 3 shows
that the lines are nearly parallel, indicating the applicability
of this method to our metallorganic complexes system in
the conversion range studied. This fact suggests that a
single reaction mechanism is operative. Mean values of
38 2 kJ mol-1 and 73 6 kJ mol-1 are found for (1)
and (2), respectively. Table 1 shows that the activation
energy is very close to the value obtained using Kissinger
method. Compared to other methods these two methods
123

Sublimation kinetics of scandium b-diketonates
159
2.6
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
(1)
(2)
α
= 0.1 to 0.8
α
= 0.1 to 0.8
2.4
2.2
2.0
1.8
1.6
1.4
1.2
0.00190 0.00195 0.00200 0.00205 0.00210 0.00215 0.00220
0.0019
0.0020
0.0021
K/T
0.0022
0.0023
K/T
Fig. 4 Plots of ln [b/J min-1] against 1/T with weight loss from 0.1 to 0.8 in steps of 0.1 according to Flynn–Wall method (b = 4, 6, 8 and
10 K min-1) for (1) and (2)
(1) 4.0
(2) 4.0
3.5
3.5
melting
3.0
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Vapourization
sublimation
vapourization
0.00192 0.00198 0.00204 0.00210 0.00216 0.00222 0.00228
0.0019
0.0020
0.0021
0.0022
0.0023
0.0024
K/T
K/T
Fig. 5 Plots of ln [k/min-1] against 1/T with weight loss from 5% to 90% according to Arrhenius plots (b = 4, 6, 8 and 10) for (1) and (2)
present the advantage that they do not require previous
knowledge of the reaction mechanism for determining the
activation energy [28, 29].
point is due to Sc(acac)3(1) evaporation process. Thus, the
kinetic parameters were evaluated for different degrees of
conversions, taking in mind that data points for the tem-
perature region below melting point corresponding to
sublimation, while data points for the temperature region
above the melting point corresponding to evaporation
(Table 1). Over this range of conversion, Arrhenius plots
of ln k versus 1/T (Fig. 5) is a linear plot from which
vaporisation activation energy (Ea) can be calculated as
82, 81, 78 and 83 kJ mol-1 for the heating rate of 4, 6, 8
and 10 K min-1, respectively, for Sc(acac)3. The activa-
tion energy values with respect the heating rates of 4, 6, 8
and 10 K min-1 are found to be 80, 79, 79 and
78 kJ mol-1, respectively for evaporation process of
Sc(tmhd)3.
Further, the activation energy values of Sc precursors
were derived from the plot of ln k versus 1/T (Arrhenius
plot). The expression for k is given by k = da/dt, where
da/dt is the derivative of the fraction sublimation with
respective to time and k the rate of sublimation. It should
be noted from TG curves given in Fig. 1 that sample mass
loss at the temperature region below the melting point
(i.e. below 457 K) is due to Sc(acac)3(1) sublimation (e.g.
sample mass loss at the melting point equals *40% at
4 K min-1 heating rate, while at 10 K min-1 heating rate
sample mass loss at the melting points equals *5%),
while mass at the temperature region above the melting
123

160
J. Selvakumar et al.
0
-1
-2
-3
-4
-5
-6
-7
-8
Conclusion
0
Sc(acac)3 (1)
Sc(tmhd)3 (2)
The sublimation and evaporation of precursors (1) and (2)
were studied applying the isothermal and non-isothermal
thermogravimetry measurements. The isothermal subli-
mation and vaporization activation energy values are found
to be 81 2, 95 3 kJ mol-1(Ea(sub)) for (1) and (2),
75 3 kJ mol-1 (Ea(vap)) for (2), respectively.
-2
-4
-6
Acknowledgement The authors thank Department of Science and
Technology (DST), Government of India (No. SR/S3/ME/03/2005-
SERC) for the financial assistance.
-8
0.0020
0.0022
0.0024
0.0026
0.0028
0.0030
References
K/T
1. Lim BS, Rahtu A, Park JS, Gordon RG. Synthesis and charac-
terization of volatile, thermally stable, reactive transition metal
amidinates. Inorg Chem. 2003;42:7951–8.
Fig. 6 Isothermal temperature based plots of ln (da/dt)m versus 1/T
for Sc(acac)3 and Sc(tmhd)3
2. Li Z, Barry ST, Gordon RG. Synthesis and characterization of
copper(I) amidinates as precursors for atomic layer deposition
(ALD) of copper metal. Inorg Chem. 2005;44:1728–35.
3. Mehrotra RC, Bohra R, Gaur DP. Metal b-diketonates and allied
derivatives. London Ltd.: Academic Press Inc.; 1978.
4. Huang Y, Cheng Y, Alexander K, Dollimore D. The thermal
analysis study of the drug captopril. Thermochim Acta. 2001;
367–368:43–58.
5. Keuleers RR, Janssens JF, Desseyn HO. Comparison of some
methods for activation energy determination of thermal decom-
position reactions by thermogravimetry. Thermochim Acta. 2002;
385:127–42.
6. Vyazovkin S, Wight CA. Isothermal and nonisothermal reaction
kinetics in solid: in search of ways toward consensus. J Phys
Chem. 1997;101:8279–84.
7. Chrissafis K. Kinetics of thermal degradation of polymers. J
Therm Anal Calorim. 2009;95:273–83.
8. Vyazovkin S. Model-free kinetics. J Therm Anal Calorim. 2007;
83:45–51.
9. Vyazovkin S, Wight CA. Model-free and model-fitting approa-
ches to kinetic analysis of isothermal and nonisothermal data.
Thermochim Acta. 1999;340–341:53–68.
10. Jingyan S, Jie L, Yun D, Ling H, Xi Y, Zhiyong W, et al.
Investigation of thermal behavior of nicotinic acid. J Therm Anal
Calorim. 2008;93:403–9.
11. Hammond GS, Nouhebel DS, Wu CHS. Preparation and prop-
erties of chelates containing sterieally hindered ligands. Inorg
Chem. 1963;2:73–6.
12. Arockiasamy S, Johnson MG, Mallika C, Sreedharan OM, Na-
garaja KS. Spectral characterisation of five volatile bis(N-R-sal-
icylaldimine)nickel(II) (where n-R = methyl to pentyl) complexes
and single crystal study on methyl analogue. Mater Chem Phys.
2008;114:456–61.
13. Antony Premkumar P, Pankajavalli R, Sreedharan OM, Raghu-
nathan VS, Nagaraja KS. Determination of vapour pressure and
standard enthalpies of sublimation and vapourisation of N,N0-
ethylenebis(2,4-pentanedion-iminoato)nickel(II) by a TG-based
transpiration method. Mater Lett. 2004;58:2256–60.
14. Antony Premkumar P, Nagaraja KS, Pankajavalli R, Mallika C,
Sreedharan OM. DTA studies on the liquidus temperatures of Cr
complex with the addition of an anhydrous Ni complex. Mater
Lett. 2004;58:474–7.
Isothermal kinetics
According to the Eq. 5 the activation energy of sublimation
for (1) and (2) is determined and summarized in Table 1.
The activation energy for both compounds is derived from
the plot of ln(da/dt)m versus 1/T (Fig. 6). The plot of ln(da/
dt)m versus 1/T is fit for (2) with two temperature spans,
which is separated by melting temperature and is repre-
sented with sublimation and vaporization kinetics of the
compound. Multiplying the slope of the isothermal plots by
universal gas constant (8.314 J K-1 mol-1), values of
81 2 kJ mol-1 (375–450 K) and 95 3 kJ mol-1
(356–414 K) could be derived for the activation energy of
isothermal sublimation for (1) and (2), respectively. The
activation energy of isothermal vaporization was found to
be 75 3 kJ mol-1 (434–465 K) for 2.
Table 1 show that the two kind of non-isothermal tech-
niques, Flynn–Wall and Kissinger, afford similar activation
energies of sublimation for both scandium complexes.
However, it is interesting that the activation energy values
of of isothermal vaporisation (75 3 kJ mol-1 (434–
465 K)) and non-isothermal process, Flynn–Wall (73
2 kJ mol-1
) ) techniques
and Friedman (79 kJ mol-1
afforded, almost the same as the activation energy derived
by Kissinger (67 5 kJ mol-1) technique for precursor
(2). Nevertheless, the Ea values derived for precursor (1)
shows dissimilarity with the values from non-isothermal
and isothermal techniques. The Ea values of precursors (1)
and (2) calculated by Flynn–Wall technique, the constant Ea
values over the different conversions (0.1–0.8), revealed
that the non-isothermal TG processes are diffusion con-
trolled. The above results imply that the sublimation char-
acteristics could vary with the heating method, sample size
and temperature range.
15. Gao X, Chen D, Dollimore D. The correlation between the value
of a at the maximum reaction rate and the reaction mechanisms: a
theoretical study. Thermochim Acta. 1993;223:75–82.
123

Sublimation kinetics of scandium b-diketonates
161
16. Burnham L, Dollimore D, Alexander K. Calculation of the vapor
pressure–temperature relationship using thermogravimetry for the
drug allopurinol. Thermochim Acta. 2001;367–368:15–22.
17. Kissinger HE. Reaction kinetics in differential thermal analysis.
Anal Chem. 1957;29:1702–6.
24. Boultif A, Louer D. Powder pattern indexing with the dichotomy
method. J Appl Crystallogr. 2004;37:724–31.
25. Thompson P, Cox DE, Hostings JB. Rietveld refinement of
Debye-Scherrer synchrotron X-ray data from Al2O3. J Appl
Crystallogr. 1987;20:79–83.
18. Budrugeac P, Segal E. Applicability of the Kissinger equation in
thermal analysis. J Therm Anal Calorim. 2007;88:703–7.
19. Jankovic B, Mantus S. Model-fitting and model-free analysis of
thermal decomposition of palladium acetylacetonate [Pd(acac)2].
J Therm Anal Calorim. 2008;94:395–403.
20. Flynn JH, Wall LA. A quick, direct method for the determination
of activation energy from thermogravimetric data. J Polym Sci B.
1966;4:323–8.
21. Li XG, Huang MR. Thermal decomposition kinetics of thermo-
tropic poly(oxybenzoate-co-oxynaphthoate) Vectra copolyester.
Polym Degrad Stab. 1999;64:81–90.
22. Smith JAS, Thwaites JD. High-resolution nuclear magnetic res-
onance study of substitution in b-diketone chelate complexes.
Discuss Faraday Soc. 1962;34:143–6.
26. Anderson TJ, Neuman MA, Melson GA. Coordination chemistry
of scandium. V. Crystal and molecular structure of tris (acetyl-
acetonato)scandium(III). Inorg Chem. 1973;12:927–30.
27. Lisoivan VI, Gromilov SA. X-ray diffraction study of scani-
um(III) and indium(III) tris(dipivaloylmethanates). Russ Chem
Bull. 1987;36:1950–1.
28. Wang H, Yang J, Long S, Wang X, Yang Z, Li G. Studies on the
thermal degradation of poly (phenylene sulfide sulfone). Polym
Degrad Stab. 2003;83:229–35.
29. Nunez L, Castro A, Villanueva M, Nunez MR, Rial B. Ther-
mogravimetric study of degradation process of diglycidyl ether of
bisphenol A-1,2-diaminocyclohexane/calcium carbonate system.
J Appl Polym Sci. 2001;83:1528–35.
23. Holm RH, Cotton FA. Spectral investigations of metal complexes
of b-diketones. I. Nuclear magnetic resonance and ultraviolet
spectra of acetylacetonates. J Am Chem Soc. 1958;80:5658–63.
123