Kinetic study of the thermal decomposition on bis(dialkyldithiocarbamate) Cd(II) complexes by isothermal and non-isothermal thermogravimetry

Article abstract of DOI:10.1016/s0040-6031(98)00643-1

The thermal decomposition kinetics of solid Cd(S₂CNR₂)₂ complexes, where R=C₂H₅, n-C₃H₇, n-C₄H₉ and iso-C₄H₉, has been studied using i

Full text of DOI:10.1016/s0040-6031(98)00643-1

Thermochimica Acta 328 (1999) 201±207
Kinetic study of the thermal decomposition on bis(dialkyldithiocarbamate)
Cd(II) complexes by isothermal and non-isothermal thermogravimetry
Maria C. Nobrega Machadoa, Liliane MagalhaÄes Nunesb, Carlos Davidson Pinheiroc,
d
Jose Caetano Machado , Antonio Gouveia Souza
b,*
Â
a
  Â
Departamento de Quõmica, CCT, Universidade Estadual da Paraõba, Campina Grande, Paraõba, Brazil
  Â
Departamento de Quõmica, CCEN, Universidade Federal da Paraõba, 58059-900, JoaÄo Pessoa, Paraõba, Brazil
b
c
à  Â
Departamento de Ciencias Exatas e da Natureza, CFP, Universidade Federal da Paraõba, 58900-000, Cajazeiras, Paraõba, Brazil
d
Â
Departamento de Quõmica, ICEx, Universidade Federal de Minas Gerais, 31270-901, Belo Horizonte, Minas Gerais, Brazil
Accepted 17 November 1998
Abstract
The thermal decomposition kinetics of solid Cd(S2CNR2)2 complexes, where RˆC2H5, n-C3H7, n-C4H9 and iso-C4H9, has
been studied using isothermal and non-isothermal thermogravimetry. Superimposed TG/DTG/DSC curves show that thermal
decomposition reactions occur in the liquid phase. The kinetic model that best adjusted the experimental isothermal TG data
was the one-dimensional phase-boundary reaction controlling process R1. The thermal analysis data suggest the thermal
stability sequence CdꢀS2CNBun† > CdꢀS2CNPrn† > CdꢀS2CNBui † > CdꢀS2CNEt † , according to the order of stability
2
2
2
2
2
2
2 2
of apparent activation energy. # 1999 Elsevier Science B.V. All rights reserved.
Keywords: Thermogravimetry; Cadmium; Thermal decomposition; Dialkyldithiocarbamate; Kinetics
1. Introduction
liquefaction occurring in a con®ned and restricted
zone, as usually observed.
The interest in the chemistry of dithiocarbamates
lies in the analytical purpose of these compounds, as
well as in their industrial applicability [1±5]. The
thermochemistry of metal dithiocarbamates has been
widely studied in the last years [6±10] but little is
known about their thermal decomposition kinetics. In
addition, these solid metal complexes were found to
decompose in the liquid phase in a range of tempera-
ture far away from their respective melting points, and
as a consequence, it is expected that the overall
reactions should not be complicated by the partial
In the present work we propose to study the kinetics
of the thermal decomposition of bis(dialkyldithiocar-
bamates) of cadmium(II), Cd(S2CNR2)2, where Rˆ
C2H5 (Et), n-C3H7 (Prn), n-C4H9 (Bun) and iso-C4H9
(Bui) using isothermal and non-isothermal heating
methods.
2. Experimental
All operations involved in the preparations and
puri®cations were carried out either in vacuo or in a
dry-box in an atmosphere of dried nitrogen for air-
sensitive compounds.
*Corresponding author. Fax: +55-83-216-7441; e-mail:
gouveia@quimica.ufpb.br
0040-6031/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved.
P I I : S 0 0 4 0 - 6 0 3 1 ( 9 8 ) 0 0 6 4 3 - 1

202
M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
2.1. Chemicals
2.4. Thermal measurements
The solvents used in all preparations were distilled
and stored over Linde 4A molecular sieves. Diethy-
lamine, di-n-propylamine, di-n-butylamine and di-iso-
butylamine (Merck) and carbon disulphide (Merck)
were distilled, respectively. The cadmium(II) chloride
was puri®ed and dried in vacuo.
Thermogravimetric curves were obtained using a
Shimadzu model TGA-50 thermobalance, under nitro-
gen atmosphere, by isothermal and non-isothermal
heating techniques. The carrier gas ¯ow was
always 3.33 cm3 s
1
, the sample masses were
5.0Æ0.5 mg, the temperature in which the thermogra-
vimetric isothermal curves were made are between the
interval of 533±578 K for the complex Cd(S2CNEt2)2,
533±583 K for CdꢀS2CNPrn2†2, 538±588 K for
2.2. Preparations
The compounds Cd(S2CNR2)2, where (Rˆethyl,
n-propyl, n-butyl and iso-butyl) were prepared by
slowly adding a solution of dialkylammonium dia-
lkyldithiocarbamate bH2NR2cbS2CNR2c salt in acet-
one to a stirred solution of the CdCl2 in acetone in 1:2
molar proportions. After ®ltration, the crystals formed
were washed with petroleum ether (303±333 K),
which were recrystallized from acetone and dried in
vacuo, with yields in the range 83±87%. The com-
plexes obtained were stored in a desiccator over
calcium chloride.
CdꢀS2CNBun2†2, and 528±578 K for CdꢀS2CNBui2†
2
and the heating rates were 5 and 10 K min 1 for non-
isothermal experiments.
The differential scanning calorimetric curves were
obtained in nitrogen atmosphere, using a differential
scanning calorimeter Shimadzu model DSC-50, with a
1
heating rate of 10 K min
.
The kinetic parameters in the dynamic heating
method were determined according the Coats±Red-
fern [11] equation, using the thermal decomposition
model suggested by the data obtained in the isothermal
heating experiments.
2.3. Physical measurements
The melting temperatures for all compounds were
Â
3. Results and discussion
determined by means of a Microquõmica model
MQAPF-301 apparatus. Microanalysis for carbon,
hydrogen and nitrogen were obtained by Perkin-
Elmer, elemental analyser, model PE-2400. The ana-
lysis of cadmium was made in GBC atomic absorption
spectrophotometer, model 908-AA. Infrared spectra
were recorded as KBr pellets, in the region 4000±
200 cm 1 using a Bomem, model MB-102 series FT-
IR spectrophotometer, mass spectra of the ligand and
chelates on a Hewlett-Packard model 5988A spectro-
meter with an ionization energy of 70 eV, at 523 K.
The microanalysis for nitrogen, carbon, hydrogen
and cadmium (atomic absorption) of the complexes,
[Cd(S2CNR2)2], with RˆC2H5, n-C3H7, n-C4H9, and
i-C4H9, are in good agreement with the expected
values for compounds as shown in Table 1. The
melting temperatures observed are well de®ned, as
indicated in Table 1. The mode of coordination of the
dialkyldithocarbamate and the de®nite melting points
suggest the existence of very stable compounds. The
ꢀCN band assumes a double-band character which is
Table 1
Results of melting points (TM), mass percentage analysis of C, H, N (%) and atomic absorption of Cd (%) and C±N and C±S stretching wave
1
numbers (cm
)
Complexes
TM (K)
Calculated
C
Experimental
C=N
C=S
H
N
Cd
C
H
N
Cd
Cd(S2CNEt2)2
CdꢀS2CNPrn2†
523
424
410
437
29.37
36.15
41.48
41.48
4.93
6.06
6.95
6.95
6.85
6.02
5.37
5.37
27.48
24.17
21.56
21.56
29.02
36.20
41.51
41.76
4.87
6.10
6.82
6.74
6.84
5.85
5.61
5.42
25.89
23.98
20.85
20.93
1496.59
1495.73
1495.78
1486.66
1073.31
1100.13
1093.17
1090.84
2
CdꢀS2CNBun2†
CdꢀS2CNBui2†
2
2

M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
203
Fig. 1. Superimposed TG/DTG/DSC curves of complex CdꢀS2CNPrn2†2.
re¯ected in the appearance of a stretching band shifted
complexes
CdꢀS2CNR2†2ꢀcr† ! CdꢀS2CNR2†2ꢀl†
! S2CNR2ꢀg† ‡ SCNR2ꢀg† ‡ SCdꢀcr†
to a higher frequency. For all complexes this band
frequency is higher than that of the respective ligand.
On the other hand, a decrease in the ꢀCS stretching
1
band is observed and an isolated band near 1000 cm
indicates that the ligand is bonded in a bidentate
manner to the metal [7±10,12]. The principal infrared
bands observed are listed in Table 1. The mass spectra
of the complexes revealed the presence of the
parent ions with the following order of stability:
CdꢀS2CNEt2† > CdꢀS2CNPrn2† > CdꢀS2CNBu2n†
Table 2
Thermal decomposition results
Complexes
Experimental Stage Reaction
conditions
Mass-loss
internal (K) (%)
Cd(S2CNEt2)2 (1)a
1
1
498±613
>693
78.6
21.4
76.8
23.2
2
2
2
Residue
(2)b
> CdꢀS2CNBui2†2. The intensities of the peaks
observed are 13.43%, 10.82%, 6.65% and 5.00%,
respectively.
506±628
>693
Residue
CdꢀS2CNPrn2†
(1)a
1
1
505±618
>693
76.0
24.0
76.3
23.7
2
Fig. 1 shows the typical TG/DTG and DSC curves
for the thermal behaviour of Cd[S2CN(n-C3H7)2]2
compound. As shown by the DSC curve, the thermal
decomposition begins at temperatures signi®cantly
higher (ꢁ363 K) than the melting point of the com-
pound, and so in the liquid phase, in a single-stage
process as suggested by the TG/DTG curves. The
same behaviour was found for all the three other
complexes, as shown in Table 2. The thermal decom-
position data obtained in the dynamic TG experimen-
tal, listed in Table 2, together with those obtained
in DSC measurements are in good agreement with
the following scheme for the thermal decomposition
reactions for the bis(dialkyldithiocarbamate) Cd(II)
Residue
(2)b
520±634
>693
Residue
CdꢀS2CNBun2† (1)a
1
1
516±626
>693
75.1
24.9
77.0
23.0
2
Residue
(2)b
526±636
>693
Residue
CdꢀS2CNBui2† (1)a
1
1
503±616
>693
85.4
14.6
83.7
16.3
2
Residue
(2)b
509±633
>693
Residue
a(1) Sample mass around 5 mg, heating rate 5 K min 1, nitrogen
(33.3 cm3 s 1).
b(2) Sample mass around 5 mg, heating rate 10 K min 1, nitrogen
(33.3 cm3 s 1).

204
M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
Fig. 2. Isothermal decomposition of CdꢀS2CNPrn2†2.
The cadmium complexes decompose after melting
model as one of the equations that describes the
thermal decomposition processes in the liquid phase.
Table 3 shows the k values obtained according the R1
model for the cadmium chelates. The temperature
dependence of k was derived using the Arrhenius
plots, the adjustment can be seen in Fig. 4 for the
complex Cd[S2CN(n-C3H7)2]2. The calculated values
for the kinetic parameters are listed in Table 4.
The kinetics of the thermal decomposition of
Cd(S2CNR2)2 complexes was also studied using the
non-isothermal heating method at two different heating
rates, 5 and 10 K min 1, using the Coats±Redfern [11]
equation and the R1 model determined in the isothermal
heating experiments, as listed in Table 5, the adjust-
ments are presented in Figs. 5 and 6 for the complex
Cd[S2CN(n-C3H7)2]2. Table 5 shows that although the
same order in the activation apparent energy values for
the complexes also observed, as found in the thermo-
gravimetric isothermal experiments, the activation
apparent energies from non-isothermal experiments
are systematically higher than those obtained from the
isothermal method. A similar behaviour was also
observed by Ghosh and Jere [21]. These authors attrib-
utedthesedifferencestothekineticcompensationeffect
(KCE). In fact we observed a good linear correlation
between ln A and E, according to
in a single-stage process by partial loss of the ligands
leaving the cadmium sulphide as residue. The iso-
thermal mass-loss data were treated kinetically with
the aid of a computer program in terms of the most
commonly proposed models found in [13±18]. The
linearity of the kinetic plots was checked in terms of
the correlation coef®cients (r) and the standard devia-
tions (s) to determine the kinetic model and the
Arrhenius parameters.
Fig. 2 shows the typical fraction decomposition
reactions (ꢁ) versus time (t) curves at different tem-
peratures for the compound Cd[S2CN(n-C3H7)2]2. A
satisfactorily linear relationship is observed between
ꢁ and t. Kinetic obedience was determined by plotting
various kinetic functions G(ꢁ) versus t [13±18],
according to
Gꢀꢁ† ˆ kt;
(1)
where k is the apparent rate constant. The best ®tting
was found, as suggested by the shape of the ꢁ versus
t curves, with the one-dimensional phase-boundary-
controlled model (R1) within the range 0.15ꢂꢁꢂ0.95
as shown in Fig. 3 for the complex Cd[S2CN(n-
C3H7)2]2; this is in agreement with the work of
Tanaka±Koga [19] who studied the kinetics of thermal
decomposition of the melted composition of NH4NO3
and found this model to be more adapted to the
experimental data. These results are con®rmed by
Dollimore [20], who mentions the equation of this
ln A ˆ a ‡ bE; (2)
where A is the pre-exponential factor, E the activation
apparent energy and a and b are the constants to be

M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
205
Fig. 3. Decomposition like the R1 model for CdꢀS2CNPrn2†2.
Table 3
Constant rates according to R1 model
Complexes
Temperatures Parameter
(K)
1
kÂ104(s
)
r
sÂ103
Cd(S2CNEt2)2
533
543
533
563
573
578
1.30
1.95
3.08
4.94
8.96
10.8
0.9999
0.9997
0.9994
0.9997
3.5
5.7
7.9
5.8
0.9991 10.6
0.9990 11.6
CdꢀS2CNPrn2†
533
543
563
573
578
583
0.921
1.41
3.48
6.10
7.74
10.5
0.9999
0.9999
0.9998
3.5
2
2.22
4.22
0.9991 10.1
0.9979 15.3
0.9997
5.79
CdꢀS2CNBun2†
538
543
558
578
583
588
0.77
1.01
2.13
5.35
6.80
9.17
0.9996
0.9996
0.9998
6.58
6.7
Fig. 4. Plot of the equation of Arrhenius.
2
4.1
evaluated.Table 6showsthevaluesfora,bandthelinear
regression coef®cients (r) calculated using expression
(2).
0.9984 13.5
0.9992 9.6
0.9989 11.5
The values of the activation apparent energy for the
cadmium complexes suggest the same order of thermal
stability observed in the thermogravimetric curves,
CdꢀS2CNBui2†
528
543
553
563
573
578
0.661
1.36
2.15
3.51
6.02
7.50
0.9999
0.9999
0.9998
0.9997
0.9998
0.9998
2.24
2
1.52
4.32
5.31
4.16
4.17
CdꢀS2CNBun2† > CdꢀS2CNPrn2†
2
2
> CdꢀS2CNBui2† > CdꢀS2CNEt2†2;
in agreementwith Magee andHill [22], which proposed
2
rˆlinear correlation coefficient; sˆstandard deviation.

206
M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
Table 4
Calculated kinetic parameters using the Arrhenius equation kˆA exp( E/RT)
Parameters
Complexes
Cd(S2CNEt2)2
CdꢀS2CNPrn2†
124.8
CdꢀS2CNBun2†
127.8
CdꢀS2CNBui2†
123.9
2
2
2
1
)
E (kJ mol
1
123.4
A (s
r
)
1.5E‡08
0.9970
1.5E‡08
0.9974
1.95E‡08
0.9996
1.1E‡08
0.9990
Table 5
Kinetic parameters determined using dynamic thermogravimetric method of Coats±Redfern's equation
Parameters
Heating rate
1
Complexes
(K min
)
Cd(S2CNEt2)2
CdꢀS2CNPrn2†
142.5
CdꢀS2CNBun2†
148.6
CdꢀS2CNBui2†
139.2
2
2
2
1
E (kJ mol
1
)
)
5
138.6
A (s
)
4.28E‡09
0.9988
8.31E‡09
0.9996
1.51E‡10
0.9989
3.41E‡09
0.9980
r
1
E (kJ mol
1
10
142.7
152.5
163.1
144.3
A (s
)
1.14E‡10
0.9979
6.80E‡10
0.9995
4.31E‡11
0.9982
1.15E‡10
0.9997
r
Fig. 5. Plot of the equation of Coats±Redfern, with a heating rate
1
of 5 K min
.
Fig. 6. Plot of the equation of Coats±Redfern, with a heating rate
1
of 10 K min
.
a correlation between the thermal stability suggested by
the TG curves and the calculated activation apparent
energies.
atmosphere, occurs in a single stage, in the liquid
phase, after melting in temperatures signi®cantly
lower than the unstable thermal temperature ranges,
according to the scheme
4. Conclusions
CdꢀS2CNR2†2ꢀcr† ! CdꢀS2CNR2†2ꢀl†
! S2CNR2ꢀg† ‡ SCNR2ꢀg† ‡ SCdꢀcr†:
The thermal decomposition of the solid bis(dialk-
yldithiocarbamate) Cd(II) complexes, in nitrogen

M.C.N. Machado et al. / Thermochimica Acta 328 (1999) 201±207
207
Table 6
References
KCE constants calculated according to expression (2), using the
values of A and E obtained from the TG isothermal and dynamic
experiments
[1] D. Coucouvanis, Prog. Inorg. Chem. 11 (1970) 233.
[2] D. Coucouvanis, Prog. Inorg. Chem. 26 (1979) 301.
[3] G.J. Kerk Van Der, H.L. Klopping, Rev. Trav. Chem. 71
(1952) 1179.
Complexes
a
b
r
Cd(S2CNEt2)2
CdꢀS2CNPrn2†
8.84
8.90
8.79
9.34
0.224
0.222
0.218
0.225
0.9998
0.9997
0.9996
0.9999
[4] J.J. Staggerda, J.A. Cras, J. Willense, Rev. Trav. Chim. Pays-
Bas 100 (1981) 41.
2
CdꢀS2CNBun2†
2
2
[5] R. West, F.W. Sunderman, Am. J. Med. Soc. 50 (1928)
3106.
CdꢀS2CNBui2†
[6] M.A.V.R. Silva, A.M.M.V. Reis, R.I.M.C.P. Farias, J. Chem.
Thermodyn. 27 (1995) 1365.
The kinetic that better model was adjusted to the
experimental data, using the method isothermal was
one-dimensional phase-boundary controlled model
(R1) within the range 0.15ꢂꢁꢂ0.95.
[7] S.C. Dias, M.G.A. Brasilino, C.D. Pinheiro, A.G. Souza,
Thermochim. Acta 241 (1994) 25.
[8] M.A.P. Carvalho, C. Airoldi, A.G. Souza, J. Chem. Soc.,
Dalton Trans. (1992) 1235.
[9] A.G. Souza, J.H. Souza, F. Souza Neto, R.O. Macedo, J.B.L.
Oliveira, J. Thermal. Anal. 49 (1997) 679.
[10] A.G. Souza, C.D. Pinheiro, L.C.R. Santos, M.L.M. Melo,
Thermochim. Acta. 231 (1994) 31.
The thermal analysis data showed that all the four
Cd(S2CNR2)2chelatesdecomposeinasimilarwayinthe
temperature range 528±588 K, according to the same
kinetic model. This result is not surprising owing to the
factthatthestructuralmodi®cationwasintroducedinthe
site far removed from the coordination sphere.
[11] A.W. Coats, J.P. Redfern, Nature 201 (1964) 68.
[12] F. Bonati, R. Ugo, J. Organomet. Chem. 10 (1967) 257.
[13] H. Tanaka, S.C. Ohshima, S. Ichiba, H. Negita, Thermochim.
Acta 48 (1981) 137.
The thermal stability of Cd(S2CNR2)2 complexes,
according the TG curves follows the sequence:
CdꢀS2CNBun2† > CdꢀS2CNPrn2† > CdꢀS2CNBui2†
[14] J.D. Handcock, J.H.J. Sharp, J. Am. Ceramic Soc. 55(2)
(1972) 74.
[15] J.H. Sharp, G.W. Brindley, A.B.N. Narahari, J. Am. Ceramic
Soc. 49(1) (1966) 379.
2
2
2
> CdꢀS2CNEt2†2. The same order of thermal stability
is suggested by the calculated activation apparent
energies in both the isothermal and non-isothermal
experiments.
[16] A.C. Norris, M.I. Poppe, M. Selwood, Thermochim. Acta 41
(1980) 357.
[17] J.M. Criado, M. Gonzalez, A. Ortega, C. Real, J. Thermal.
Anal. 29(2) (1984) 243.
The values of the energy of activation apparently
estimated by dynamic method are larger than that
estimated by isothermal, however obeyed the effect
of the kinetic compensation.
[18] H. Tanaka, S.C. Ohshima, H. Negita, Thermochim. Acta 53
(1982) 387.
[19] N. Koga, H. Tanaka, Thermochim. Acta 209 (1992) 127.
[20] D. Dollimore, Thermochim. Acta 203 (1992) 7.
[21] J.K. Gosh, G.V. Jere, Thermochim. Acta 125 (1988) 51.
[22] J.R. Magee, O.J. Hill, Rev. Anal. Chem. (1985) 5.
Acknowledgements
The authors thank CNPq and CAPES for fellow-
ships and PADCT for ®nancial support.