..

Joumalo!Tflemral :!11alrsis and Ca/orillle/1:1'. Vol. 61 (2000) 985 IJij./

KINETICS OF THER1\1AL DEC0l\1POSITION OF

COPPER(II) ACETATE MONOHYDRATE

A. Y. Obaid. A. 0. Alvoubi,
- A. A. Samarkandv,
.. S. A. AI-Thahaiti.
.
S. S. Al-Juaid, A. A. EI-Bellihi and El- II. M. De�jclllah
Che mistry Department, f aculty of Science. K ing /\bdula7i7 University, kddah, P.O. Box 902H
Saudi Arabia

(Recc) vcd January 20. 1999; in revised form September 5. I Y99)

Abstract

The thermal dccomposilion of copper(II) acc.:tatc nlOilllhydralL' wns stud icd in air and nitrogen atnw­
srhcres by means of OTA-TG and SEM mcnsuremcnts. The kinetics of the thermal decomposition
steps in air was studied by using isothermal nnd non-i�othcrrnal thermogravimdric techniques. The
results arc discussed in terms of various reaction interf:H.:c models and different techniqtJcs of com­
putationnl annlysis ofnon-isothcrnwl data. The activation rararnctcrs. cnh:ulatcd by using a compos­
ite method of integral analysis ofnon-isothcrn�al daw. rcwalcd not only their independence from the
heating rate and fracliunal reaction, but also a bcllcr com·lation and agreement with the results ob­
tained llndcr isothermal conditions.

Ke)·words: c opper accratc, deco mpo sition , ESM, DT1\. Hi

Introduction

Several studies have been performed on the thermal decomposition behaviour of metal
acetates [I, 5], but less information is available on the kinetics of the reactions [4, 5]. On
decomposition, metal acetates often yield acetone and acetic acid, and there may be in­
comple te melting and some hydrolysis of hydrated salts [I]. Leicester and Red man [2]
studied the thennal decompositions of nickel and cobalt acetates and found that t he y pro·
ceeded via an initial cleavage to metal oxide and acetic anhydride, and thence to various
pyrolysis products of the anhydride. Clough eta/. f3] suggeste d that the in itial step in the
thermal decomposition of uranyl acetate could be t he fission of the U02-0 bond, produc­
ing acetate radicals which react to give various pyrolysis products.
In the thermal decomposition ofCu(II) carboxylates, Clt salts have been identified
as intennediates, and the reduction of Cu(ll) proceeds in two steps (Cu 2+�cu' 4Cu ),
0
with a possibly appreciable overlap of the consec utive rate processes [6J.

•
To whom all corrcspomtcncc should be addrcs);cd .

14/8-1874!2000! $5.00 Akucli:miui Kiaclo. Buclat'''sl

('> 1000 Alwdemiai Kiado, Budapest 1\f111n:r Acud�tmic Puhlislwrs. !Jorclrecllt
9X6 OBAID cl al.: COPI,ER(!J) ACJ:'fAH MONOHYDRATE

All methods of kinetic analysis of the thennal decomposition of solids yield results
of a relative character, depending markedly on procedural variables [7], including the
sample mass, the heating rate, the pat1icle size and the partial pressure of the ambient at­
rnosphere. There are a variety of computational methods and the analysis of kinetic data
may not allow the choice of a single model to explain the results, which makes the estab­
lishment of a dependable correlation more difficult. However, applicable reaction models
identify the chemical change occurring at the reaction interfaces, which progressively ad­
vance into the unchanged reactant [6j. The imp011ance of characterizing the chemistry of
the interface reactions and of elucidating the chemical steps through which the reactants
are conve11ed into products has been emphasized.
In the present investigation, the thermal decomposition behaviour of copper( II)
acetate monohydrate has been studied by using DTA-TG and SEM techniques. The
ki)Ietics of the isothermal and non-isothermal dehydration and decomposition of cop­
per(ll) acetate monohydrate was studied in air. The composite method of kinetic inte­
gr al analysis ofnon:.isothcrmal data \vas applied to identify the mechanistic equation
describing the geometry of advance of the reaction interface and to calculate the acti­
vation parameters. The results were compared v,:ith those obtained by using the iso­
thctm al tcchni4uc and other integral methods of analysis of non-isothermal data.

Expcrhncntal

The copper(ll) acetate monohydrate crystalline powder, a BOll product of reagent

grade, was used v>ithout further purification.
Simultaneous DTA-TU experiments were performed with a Shimadzu DT 30
thermal analyser. The experi ments were performed under isothermal conditions or at
different specified heating rates in air at a flow rate of 3.0 I h 1• The sample mass in
the PI cell of the the rmal analyser was kept at about 8 mg in all experiments, in order
10 ensure a linear heating rate and accurate temperature measurements.
Samples of copper acetate monohydrate were inserted into an electric oven at room
temperature in air and the temperature was rais�d to the desired value, which was main­
tained for 30 min before the sample was removed and cooled to room temperature in air
in a desiccator. The changes taking place in morphology and texture during the thermal
decomposition of the salt were investigated by using a JEOL T 300 scanning electron mi­
croscope. Samples of the parent compound and the products fonned in the different
stages during the decomposition were examined.

Kinetic consideration

In general, the rate constant, k of a solid-state reaction is given by the formal kinetic
equation:

da
=k{(a
. )=Ae 1.1<1 ({a) (I)
dt .

. til. 200()
J Tllerm. A11al. Cal.
 OBAID ct al.: t'OI'I'UWI) ACFI/\'11: MONOHYDRATE 9X7

where a is the fractional rea ctio n ; 1. is time: A is the pre e x ponential t�1ctor;
· E is the
activati on energy� R is the gas constant; Tis temperature in Kelvin. and/( a) is the ki·
netic function, which takes di ffe rent forms de pen di n g on the particular reactio n rat�
equation [7-9]. In isothermal kinetic studies, the rate equation used to calculate the
rate constant has the fonn

g(ct.)=kt (2)

where o·(a )::::J"� is the in tegrated f(mll of/(a)

. [7-91. However. non·isothenn al
.
t:t uf(a)
methods arc becoming more widely used because they are more convenient than the
classi cal isothermal methods [I 0).
In non-isothermal kinetics. the time d� pcn d �nc c on the lcfl side of h.J. (I) is
elimil'latcd by using a constant hea tin g rate J�=d Tldt. so that T=7�.+J3, wh ere T11 is the
starting te1nperature and t is the time of heating. With the usc of integral methods of
analysis, Eq. (I) may be written as

A I I
f HI
�(u )=- e · dt
.._

r� r.

The reaction rate is neg l igihlc allow temperatures. Hence:

AII I Rl
g(n )=- e d1 ( 4)
·
r� ��

[ ' J' " l

I nt egr ation of this equation leads to Doyle s equation Ill]: '

A F- e . c. AF
g(u )=- -- --du =-' P(.r) (5)
RP ·" .. " Rr�
where w=EIRT and x is the <:orrcsponding value of u at which a fraction of material
has decomposed. This eq uation has he en refnrmulated [ 12 J as

·1E
lng(u ) - ln P (x) = ln-'-· = B
Rr�
where B is a constant for a part i cular reaction at a constant heating rate r�. The integral
function P(x) is not definite; it may he written in an expanded form and estimated by
u sing a procedure of trial·and-error type involving iteration [ 13].
In th e present study, an alysi s of non·isothermal data was performed by using the
approximate computational approaches due to Coats and Redfern [ 14], Ozawa [ 15].
Mahusudanan [ 16] and Diefallah [ 17]. Integral methods are often more reliable and
generally preferred than differential methods of kinetic analysis .

ART2 [t- ]
In the Coats-Redfern method f 14]. the fun cti on g(a) is approximated to the form

2RT h'RT
(a) e (7)
� BE ��·

J. Them/. ;tnal. Cal.. M. 2000

 OBAID cl al.: COPPER( Ill ACITATE MONOIIYDRATL::

]
The equation has been written in the form:

[g(a) -
T1
AR( 2RT) E
-In -- --In-t
JlE E
+
RT
--- - (8)

The quantity lnl(AR)/UH:')]( I -2RT!l:') is reasonably constant for most values of

E and in the temperature range over which most reactions occur. llmvever, both E
and A could vary with the experimental heating rate.
In the Ozawa method f151, a master curve has been derived from the TG data ob­
tained at different heating rates q�) by using Doyle's equation and assuming that
lAEII.�Rjp(EIRT) is a constant for a given function of material decomposed. The func­
tion p(EIRT) was approximated by the equation:

logJ
. {
E
. RT
)
,
-=-2.3 15-{}.456
{F)-'
RT .
(9)

so that

-logJl=i1.456 { :� }'constant (I 0)

Hence, the activation energy is calculatt:d from the thermogravimetric data ob­
tained at different heating rates. The frequency f�u.:tor is calculated from the equation

r l
E E
lo1!. :l=-logg(u )-lo� -J>- ( I l)
.. .. ... I�R RT

Allhough the calculated activation energy is independent of the reaction model,

and the frequency factor depends on the determined form ofg(a), both E and A could
vary considerably with the fractional reaction PI-

[R(a)]
In the Madhusudanan method 1161. the equation used has the form

'IR F
-In < ·-+3.772050 l-l.921503lnE--·
L'l�l'-ol :::; -In- ( 12)
T �/:· RT
..

where the symbols have their usual signiticance. The activation energy E and the fre­
quency factor A are obtained from the slope and intercept of the linear fit of the plot of
-ln[g(a)/T
�215113]
vs. I IT.

In the composite method of analysis of dynamic data [7], the results obtained
(not only at different heating rates, but also at different a values) are superimposed on
one master curve. This was achieved by rewriting the approximate integral equations
due to different workers in a form such that the kinetic function g(a) and the linear
heating rate J3 lie on one side of the equation and liT on the other side. When use is
made of the modified Coats-Redfern equation [ 14], then in order to allow the com­

[ )] (
posite analysis the equation is written in form

J3g(a AR l E
In =In --- (13)
T2 E RT

.
./. Therm. Anal. Cal . 6 I. 2000
 OBAID et al.: COPI'LRtll) t\CLTATE t\IONOHYDRATI:

lienee, the dependence of lnH�K(a)iT!] calculated for different a values at their re­
spective r3 values on liT must give rise to a single master straight line for the con·ect form
of g(a ). and a single activation energy and frequency factor can readily be calculated.
When Doyle's approximate equation [Ill is used, the equation for composite

{ A£ ]
analysis has the form

n.
E
log[ pg(a)]- logR-2315 -().4567 T ( 14)
R

Use may also be made of the approximate equation f 16] of Madhusudanan, re­

[ ]
written as

pg(a) AR F
-In =-ln-+3.772050 1-1.921 I 503ln/:'--' ( 15)
t9�t�n.1
T r: RT
_

Again. the dependence of the kft side of Eqs ( 14) and { 15) on liT should gin.·
rise to a single master straight line for the correct form of g(a). and hence the activa­
tion energy and the frequency factor can readily be calculated.
In general. the use of the different approximate integral equations f()r the kinetic
analysis of non-isolhennal decomposition kinetic data according to the composite
method of analysis gave rise. within experimental error. to identical values of the ac­
tivation parameters and the correct f(.>rm ofg(o.). A computer program has been writ­
ten to perform the data analysis [ 17].

II

-- -­

•'
�-

'
II
II

�-,

r
I ---- t" '
- I
... -----­
.;

- , ,-

" ' I
'.,' I J :\ T
-�

II
..,

1
fnoo

60

70

100 zoo :[Link] •oo :;oo 1 00 200 300 �00 500

Fig. I DTA-TG curves for coppcr(ll) acetate monohydrate in A air and B · nitrogen
'
at a heating rate of I5°C min·

.J. 1lwn11. :111(1/. Cal.. 6/, 2000

l)l)f) OBAID l't al.: COPPI:l�(ll) ACITATE MONOHYDRATE

Results and discussion

Figure I shows the DTA-TG curves obtnined for Cu(CH1C00)2·H20 in air (A) and nitro­
gen (8), respectively. The TG curve contained two steps due to dehydration and decom­
position of the anhydrous salt, in general agreement with previous studies f4]. The first
step was endothcm1ic and took place in the temperature range 1 J0-150°C, involving the
loss of on� molecule of water with a mass loss of about l 0%. The second step started at
about 200°C and was complete by about 300°C; it v.'as endothermic in nitrogen, and exo­
thermic in air, due to oxidation. It corresponded to a mass loss of about 58%, due to de­
composition of the anhydrous salt to copper metal. ln air, above 3000C, a slight mass in­
crease was observed in the TG curve and a corresponding broad exothermic peak in the
OT A curve. due to the pattial oxidation of Cu to CuO.
SEM micrographs indicating the changes in morphology and texture that ac­

company the salt decomposition are shown in Fig. 2. Each sample was calcined for

Fig. 2 Scanning electron micrographs showing the changes in texture and morphology
that accompany the thermal [Link] of copper( II) acetate monohydrate in
air: a parent compound at room temperature, displaying relatively large crys­
··

tals of different sizes; b - parent compound after partial dehydration at l20°C;

c . sample calcined at 200°C after complete dehydration of the parent com·
pound to the anhydrous salt; d · sampk calcined at 300°C, showing partial de­
composition and fissioning of the anhydrous s alt; e- sample calcined at 500°C,
showi ng comp lete decomposition of the anhydrous salt into small crystallites
with a cubic crystal habit and some aggregation of particles; f sample calcined
··

at 800°C, showing coalescence of the small crystallites into Jarge aggregates ac­
companying the oxidation of Cu to CuO; (scale bar: a, band c I 00 pm; d, c
and f 10 pm)

J Therm. Anal. Cal . 61. 2000

.
9<J2 OBAID ct al.: COPPER( II) ACETATE MONOHYDJ{ATF

hydrous salt. after partial decomposition, had undergone extensive degradation with a
reduction in particle size from more than about l 00 pm to less than about 4 pm. At
500°C, after complete decomposition of the anhydrous salt to Cu0 metal, the particle
size was further reduced to less than about 2 pm, with a cubic crystal habit. At 800°C,
the small crystallites coalesced into aggregates of irregular shape and size, accompa­
nying the oxidation of copper to Cu(ll) oxide.
Figure 3 shows plots of isothermal fractional reaction t•s. time curves, whereas
Fig. 4 depicts the tl·actional reaction remaining as a function of temperature, for the
dehydration and decomposition steps. The curves in Fig. 3 exhibited a deceleratory

I inw:min

Fig. 3 lsothcrma I all curn�s for the decomposition of copp.:r(ll) acetate monohydrate
in air: a tkhydration step; h <k�compusition step

II UI
1.0

0.�

0.6

0.4

0.2

100 ](oil I��� :wu 220 240 260

•·ig. 4Dynamic measurements for the thermal decomposition of coppcr(II) acetate

monohydrate in air: a- dehydration step; b - decomposition step. Heating rate:
1 1
curve A- 2°C min- ; curve B -·soc min 1 ; curve C- l0°C min- ;
-

1
curveD I5°C min 1 and curve E 20°C min

.!. Tllerm. A 11ol. Col. 6/. 2000

 OBt\ID ct al.: COPPER( Ill 1\CETi\TE MONOHYDR/\ TE 991

about 30 min at the specified temperature. The results reveal that the particle shape
and size change throughout the decomposition. The regular shape of the parent reac-
tant crystallites (Fig. 2(a)), underwent some superficial roughening, after partial de-
hydration at I20°C, (Fig. 2(b)), whereas complete dehydration at 200°C resulted in
more roughening and cracking of the crystallite surface (Fig. 2(c)). At J00°C, the an-

Table I Activation parameters of the thermal dehydration and decomposition of Cu{CI hCOO)�·
H!O, calculat\!d according to the R� motkl

[kh:z:dration stc� fk�·omp�sition step

Method of analysis . I
Eik.l mol 1 Ig.4 imm I f:'/k.l mol 1 lg_4imin
Isothe rmal I07.H±-1.3 13.1±0.5 71.::!±3. 7 6. 0±0.3
Non'-isnthcrrnaI

I\
2 1!\fU 23. 0 n. 7 R.2
5 129.3 16.0 95.0 �.4
a) Coats-Redfern 10 11 !\.5 14.6 95.9 i\.(1
15 11�.x 1 3.7 I 06.0 9.7

20 R7.X I0.7 II 7.7 10.3

1\verag.� 125.5�2.1.4 15.6�3.1 I 00.3±6.9 9.0±0.R

f!
2 I R0.6 23.0 93.1 R.2
5 1 34. R 16.8 95.3 X.5
b} Madhusudanan 10 IIR.H 14.6 96.2 X.(}
15 112.0 13.7 106.3 9.7
20 XX.O l 0. (J 112.0 I0.3
/\vcrngc 126.X:::.24.7 [Link] 100.6..:.6.9 9. I =.O.H

I (l

0. I 1 82.7 23.8 92.3 R.2

0.2 145.8 IR.5 93.7 RJ

0.3 129.7 16.2 98.7 8.9

0.4 123.2 15.3 1 07.9 9.9

c) Ozawa 0.5 115.7 14.3 117.2 10 .9

0.6 112.0 13.7 124.5 11.7

0.7 105.7 12.9 129.6 12.2

O.R 99.2 12.0 142.0 13.5

0.9 RR.9 I0.6 141t0 1 4.2
Averag e l22.5i20.2 15.3+2.8 117.1±16.8 10.9±1.8

d�ComQusitc 117.7±4.2 14.5+1.3 I00.2+.1.9 9.1 ±0 .5

J. 71u·nn. Anal. Cal .. ti I. 2000

 OBAIDct al.: COPPER!IIJ ACETATF.I\!ONOHYDRJ\TE 993

_,
A

'----'---..1.--..J---'----'�-·-

� -1

3
(t0 fT) I K-1

Fi_g. 5 Composite analysis o f the dynamic T(i data based on the modified
Coats-Kctlfcrn equation(;\) and Doyk's cqmtinn (B). assuming the contracting
surface (R2) model. I dehydration step; 2 decomposition step
-

shape, and kinetic analyses of the two steps were perf(mned with reference to the dif­
ferent models of heterogeneous solid-state reactions [7-9]. The method of least
square regression analysis was performed for each of the tested functions, and it was
found that the two-dimensional phase boundary and the first-order reaction models
gave the best fits for both the dehydration and the decomposition steps. Analyses of
the non-isothermal data were perfonned according to the integral methods of
Coats-Rcdfern, Ozawa, Madhusudanan and Diefallah, and the activation parameters
were calculated. They are listed in Table I, in comparison with those obtained from
the analysis of the isothermal data.
Figure 5 shows typical composite plots for the dehydration and decomposition
steps according to the contracting surface (R1) equation, which gives better regression
factors and standard deviations than those for the other models.
The results in Table 1 demonstrate that the integral composite analysis of the dy­
namic data gave better agreement with the isothermal results, with a higher correla�
tion and less deviation in the calculated experimental parameters. The activation en­
ergy for the dehydration step is in agreement with that reported by Mu and Perlmutter
[5] and indicates that the water of hydration is co-ordinated water. The activation en�

./_ 1/u•rm. A11al. Cal

.. 61. 2000
994 OBt\ID cl al.: COPPI;R(II) ACTl/\TE MONOHYDRATE

crgy for the decomposition of the anhydrous salt is lower than the values of many
thermal decomposition reactions, which may indicate the catalytic action of the
tine-grained copper metal particles formed in the early stages of the decomposition
step. This supports the view that charge transfer is the rate-controlling step in the
mechanism of carboxylate decompositions [I].
* * *

The authors thank Mr. M. Abdcl Fattah (Ch..:m. Dept., KAU) for the OTA-TG analysis, and Mr. M.
Sid'(Production Eng. DepL KAlJ) for performing the SEM experimems.

J�efcrences

I, A. K. Galwey, in MTP International Review of Science, Inorganic Chemistry, Series 2, Slllid

State Chemistry, Vol. I 0, H. J. Emckus (ed. ), Buttcrworths, London I 97 5, p. 14 7.
1 J. Leicester and M. II. Redmon. J. Appl. Chem., 12 ( 1lJ(l2) 357.
3 P. S. Clough, D. Dollimorc and P. Grundy. J. lnorg. Nucl. Chcm., ( 1969) 36 1.
4 A. Afzil1, P. K. Butt and A. Ahmad, J. Thermal Anal., 37 ( 1991) 1015.
5 J. Mu and D. D. Perlmutter, Thcrmochim. Acta, 4<J (I <JX I ) 207.
6 A. K. (lalwey, J. Thermal Anal., 41 ( 1994) 2(J7.
7 El-11. M. Oidallah, Thcrmochim. Acta, 202 ( 1992) I.
H M. E. Brown, D. Dollimorc and A. K. (ialwey, Culllpn;hcnsivc Chemical Kinetics, Vol. 22,
Elsevier. Amsterdam 19XO.
<J M. E. Brown, lntroductinn to Thermal Analysis, Chapman and Jlall, 1988, Chapter 13.
I 0 D. Chen, X. Gao and D. Dollimore, Analytical lnstnunenliltion, 20 ( 1992} 137.
II C. D. Doyle. Nature, 207 ( 1%5) 290.
12 J. Zsuko, J. Thcrmul Anal., 2 ( 1970) I45.
13 0. Carrol and E. P. Manch�. Thcnno�.:him. Acta, 3 ( 1972) 449.
14 A. W. Coats and J. P. Redfern, N uturc, 201 ( 1964) 68.
15 T. Ozawa, Bull. Chcm. Soc. Jpn.. 38 ( 1965) 1881; J. Thermal Anal., 2 ( 1970) 301.
16 P. M. Madhusudanan. Thcrmo�:him .Act<.�, 97 (I n(i) 189.
17 Fl-H. M. Did'a11uh. A. Y. Obnid, 1\. II. Qusti, A. t\. E1-fkllihi and M. Abdcl W<thab,
Thermochim. /\eta, 27-t ( 19%) I 72.

J. Tham. Anal. Ctt!.. 61. 2000