Bull. Mater. Sci., Vol. 30, No. 4, August 2007, pp. 339–344. © Indian Academy of Sciences.

Thermal decomposition of natural dolomite

S GUNASEKARAN† and G ANBALAGAN*

PG and Research Department of Physics, Presidency College, Chennai 600 005, India
†
PG and Research Department of Physics, Pachaiyappa’s College, Chennai 600 030, India

MS received 3 November 2006; revised 22 May 2007

Abstract. Thermal decomposition behaviour of dolomite sample has been studied by thermogravimetric
(TG) measurements. Differential thermal analysis (DTA) curve of dolomite shows two peaks at 777⋅8°C and
834°C. The two endothermic peaks observed in dolomite are essentially due to decarbonation of dolomite and
calcite, respectively. The TG data of the decomposition steps have also been analysed using various differential,
difference-differential and integral methods, viz. Freeman–Carroll, Horowitz–Metzger, Coats–Redfern methods.
Values of activation entropy, Arrhenius factor, and order of reaction have been approximated and compared.
Measured activation energies vary between 97 and 147 kJ mol–1. The large fluctuation in activation energy is
attributed to the presence of impurities such as SiO2, Al2O3, Fe2O3, Cl– etc in the samples. FTIR and XRD
analyses confirm the decomposition reaction. SEM observation of the heat-treated samples at 950°C shows
cluster of grains, indicating the structural transformation.

Keywords. TGA–DTA; FTIR; X-ray diffraction; dolomite.

1. Introduction the environment. Many researchers have studied the kinetics

of thermal decomposition of carbonate minerals (Powell
Dolomite typically occurs as the major constituent of and Searcy 1980; Warne et al 1981; Iwafuchi et al 1983;
sedimentary formations in association with calcite. Borgwardt 1985; Yariv 1989; McInosh et al 1990;
Thermal analysis might offer the means of defining the Rubiera et al 1991; McCauley and Johnson 1991; Mulo-
fraction of each mineral lattice in such mixtures and the kozi and Lugwisha 1992; Ersoy-Merichoyu et al 1993;
concentration of each cation in each lattice. The correlation Shoval et al 1993; Xiao et al 1997). A general review of
of the thermal data with the structural pattern should pro- the literature on the decomposition of carbonates indi-
vide a broader understanding of these minerals in their cates that a great deal of variability exists in the reported
natural occurrence. Romero Salvador et al (1989) studied values of the decomposition temperatures, activation en-
the effects of experimental variables i.e. sample weight, ergies and rates of decomposition. Although the decom-
particle size, purge gas velocity and crystalline structure, position of NBS dolomite sample has been reported
on the kinetic parameters of calcium carbonate decompo- (Criado and Ortega 1991), there have been no systematic
sition. Criado and Ortega (1992) studied the influence of studies on dolomite samples having a small concentration
particle size on the thermal decomposition of calcium of naturally occurring alkali earth salts. In this paper, the
carbonate and found that the activation energy of thermal effect of chlorine ions on the decomposition kinetics of
decomposition of calcium carbonate smoothly increases dolomite at various temperatures studied by differential
as the particle size increases. This behaviour can be attri- thermal analysis, thermogravimetric analysis, Fourier
buted to the lower particle size and greater fraction of transform infrared spectroscopy and powder X-ray di-
CaCO3 molecules located on surface with regard to the ffraction, have been reported.
bulk. Therefore, the activation energy decreases because
of the extra calcium carbonate stored on the surface of
2. Experimental
the smaller particles. In general, it has been reported that
a diminition of the activation energy takes place when the
Natural dolomite sample collected from Neralakere mines,
sample size increases. This behaviour would be attributed
Bagalkot area of (latitude, 16°11′, longitude, 75°45′) Kar-
to the influence of heat and mass transfer effects because
nataka state was ground with agate mortar and pestle. The
of the poor control that is usually exerted on both decom-
powder sample with size, 1⋅2 nm, was subjected to X-ray
position rate (i.e. rate of heat evolution) and pressure of
diffraction (XRD), Fourier-transform infrared spectro-
scopy (FT–IR) and thermal analyses (differential thermal
*Author for correspondence ([email protected]) analysis (DTA) and thermogravimetric analysis (TGA)).

339
340 S Gunasekaran and G Anbalagan

The particle size was determined using the Scherrer for- mineral are characterized by endothermic peaks at various
mula (Kunel et al 1997). temperatures caused by the evolution of carbon dioxide
XRD analysis was performed with a SEIFERT X-ray (figure 1). DTA curve of dolomite shows two endotherms
diffractometer with CuKα radiation (λ = 1⋅54 Å), Cu filter at 772⋅6°C and 834°C. The first one begins at 687°C,
on secondary optics, 45 kV power and 20 mA current. reaches a peak at 773°C and ends at 781°C and the second
The powder sample was mounted on a quartz support to one begins at 781°C, reaches a peak at 834°C and ends at
minimize background. 916°C. The lower temperature peak represents the de-
Perkin-Elmer spectrum one FT–IR spectrometer was composition of the dolomite structure, releasing carbon
used, and the samples were analysed in KBr pellets. dioxide from the carbonate ion associated with magne-
Spectra were traced in the range 4000–400 cm–1, and the sium part of the structure accompanied by the formation
band intensities were expressed in transmittance (%). The of calcite and magnesium oxide. The higher temperature
infrared analysis helped in the identification of the main peak represents the decomposition of calcite with the
groups in the carbonate samples. evolution of carbon dioxide (McInosh et al 1990). Mc
The powder sample was heated in a muffle furnace Cauley and Johnson (1999) observed the peak tempera-
under air atmosphere for 6 h in each temperature. The tures at 790°C and 845°C for +16 mesh dolomite samples.
spectral recordings were carried out at room temperature. However, Li and Messing (1983) reported the correspond-
Thermal analysis was performed in a simultaneous ing peak temperatures for CaCl2 doped dolomite sample
TG–DTA (Netzsch STA 409). The experimental condi- at 750°C and 830°C. This result suggests that the pre-
tions were: (i) continuous heating from room temperature sence of salt (Cl–) enhances the decomposition of dolomite
to 1000°C at a heating rate of 10 K/min, (ii) N2-gas dyna- in the present study. The decomposition process is initiated
mic atmosphere (90 cm3 min–1), (iii) alumina, as reference at lower temperatures than observed for the pure dolomite.
material and (iv) sample: 53⋅37 mg of the sample having The salts promote the formation of MgO and CaCO3 dur-
grain size 1⋅2 nm without pressing. The temperature was ing the early stage of decomposition. The ratio of the
detected with a Pt–Pt 13% Rh thermocouple fixed in a peak area (~5⋅2) and the high characteristic temperature
position near the sample pan. TG and DTA curves were (773°C) indicates that the dolomite is in well-ordered
obtained. The following data was obtained by thermal crystalline structure (Garn 1965), which is also confirmed
analysis: (i) reaction peak temperature and main effect through X-ray diffraction analysis. According to Barcina
(endothermic or exothermic) and (ii) content of bound et al (1997), smaller size of magnesium with respect to
water, which is the weight loss in the temperature range calcium facilitates the magnesium mobility and thus the
200–600°C and content of CO2 released during the de- formation of carbon dioxide associated to magnesium
composition of carbonate phases. oxide is kinetically favoured against the formation of CO2
associated to calcium oxide. The first endothermic peak
of dolomite, however, caused by the reaction of more
3. Results and discussion
complicated mechanism, has inverted symmetry/shape index
of 1⋅45 (Garn 1965). After calcinations, the resultant
Table 1 gives composition of the sample, which was deter-
oxides have lower molar volumes, larger surface areas,
mined by the standard analysis (Vogel 1951). The main
and greater porosities than the carbonates. The calcina-
undesirable impurities in the carbonate rocks are silica,
tion of a carbonate entails the formation of an oxide hav-
K2O, Na2O, Cl– and alumina. These impurities combine
ing a pseudo-lattice of the carbonate and subsequent re-
with calcium oxide at elevated temperatures to form a slag,
crystallization to the normal cubic lattice of the oxide.
which reduce the pore volume and the amount of availa-
If the temperature is high enough, sintering of oxide
ble active lime (Chan et al 1970). Such impurities either
takes place (Glasson 1958). The results indicate that the
occurred in the matrix or came from the material in the
effect of Fe2O3 and Al2O3 on peak temperature is maxi-
crevices and other strata excavated along with limestone.
mum when these oxides are present in low concentra-
Samples were chosen from quarries in which these impuri-
tions.
ties were low.

3.1 Differential thermal analysis 3.2 Thermogravimetric analysis

The typical DTA curve of dolomite sample is presented The typical TG curve of dolomite sample is presented in
in figure 1. The thermal curves representing the carbonate figure 1. The observed weight loss was 1⋅33% below

Table 1. Results of the chemical analysis of samples (%).

Sample code CaO MgO SiO2 Fe2O3 Al2O3 K2O Na2O Cl– LOI

D01 30⋅24 21⋅33 0⋅18 0⋅63 0⋅25 0⋅03 0⋅23 0⋅19 46⋅04
 Thermal decomposition of natural dolomite 341

Table 2. Kinetic parameters for the thermal decomposition of dolomite in N2 atmosphere for different calculation methods.
Freeman–Carroll Coats–Redfern Horowitz–Metzger

Sample E Log A S E Log A S E Log A S

code (kJ mol–1) (s–1) (JK mol–1) (kJ mol–1) (s–1) (JK mol–1) n r (kJ mol–1) (s–1) (JK mol–1)

D01 113⋅56 3⋅246 –133⋅586 123⋅684 3⋅743 –183⋅851 0⋅15 0⋅991 147⋅58 4⋅346 –172⋅55

Figure 1. TG–DTA curves of dolomite.

600°C and between 600°C and 850°C, it was 46⋅6%. The

weight loss detected in the temperature range 100–120°C,
was followed by a weight loss attributed to the decompo-
sition of carbonates. The weight loss in this temperature
range can be attributed to the chemically bound water.
The kinetics of decomposition processes were analysed
by means of the three popular methods (Freeman and
Carroll 1958; Coats and Redfern 1964; Horowitz and
Metzger 1963). Measured activation energies are given in
table 2 and are in agreement with that reported by Criado
and Ortega (1991) for pure dolomite sample. The mea-
sured activation energy indicates that Cl– do not activate
the process by lowering the thermal requirements for de-
composition. The slight variation in activation energy
may be attributed to the difference in particle size and
mineral origin in the samples. Lower the particle size, Figure 2. XRD pattern of dolomite at different temperatures.
greater the fraction of molecules located on the surface
with regard to the bulk. The wide dispersion of the avai-
observed kinetic parameters were found to be strongly
lable data is in relation to the influence of physical pro-
affected by small amounts of impurities; however, the
cesses, viz. inter- and intra-particle diffusion, heat transfer
kinetic model is not affected. Comparison of the values of
resistance, sintering etc. Garcia Calvo et al (1990) studied
E and A in the present study with those values of pure
the influence of macro-kinetic parameters on the value of
calcite reported by Garcia Calvo et al (1990) indicates
the activation energy over a wide range of experimental
that the presence of impurities is a cause of variation of
conditions and concluded that the influence of macro-
kinetic parameters obtained. The impurities could func-
kinetic parameter is low within this range. The calculated
tion as catalysts owing to their influence in the crystalline
activation energy, E, increases with increasing concentra-
structure.
tion of the decomposition product (CO2), which is similar
to the observation of Ersoy-Merichoyu et al (1993). This
change is attributed to the reversible nature of the de- 3.3 X-ray diffraction analysis
composition reaction. The increase in E is balanced by a
corresponding increase in A which is due to the compen- Figure 2 depicts the powder X-ray diffraction pattern of
sation behaviour. If higher activation energies are ob- natural dolomite sample at different temperatures. The
tained, the pre-exponential factors (A) are higher too. The room temperature XRD pattern of the sample displays
342 S Gunasekaran and G Anbalagan

Table 3. Indexed powder XRD pattern for natural dolomite.

2θ (degree) 2θ (degree) Diff. 2θ
Sl. no. h k l d (Å) obs d (Å) cal Diff. D (Å) obs cal (degree)

1 0 1 2 3⋅6917 3⋅6906 0⋅0011 24⋅087 24⋅094 –0⋅007

2 1 0 4 2⋅8884 2⋅8887 –0⋅0003 30⋅935 30⋅931 0⋅004
3 0 0 6 2⋅6720 2⋅6724 –0⋅0004 33⋅511 33⋅506 0⋅005
4 0 1 5 2⋅5417 2⋅5408 0⋅0009 35⋅283 35⋅296 –0⋅013
5 1 1 0 2⋅4100 2⋅4104 –0⋅0004 37⋅281 37⋅275 0⋅006
6 1 1 3 2⋅1974 2⋅1975 –0⋅0001 41⋅042 41⋅040 0⋅002
7 0 2 1 2⋅0209 2⋅0214 –0⋅0005 44⋅812 44⋅801 0⋅011
8 0 2 4 1⋅8519 1⋅8522 –0⋅0003 49⋅158 49⋅150 0⋅008
9 0 1 8 1⋅8095 1⋅8098 –0⋅0003 50⋅390 50⋅380 0⋅010
10 1 1 6 1⋅7920 1⋅7919 0⋅0001 50⋅918 50⋅919 –0⋅001
11 2 1 1 1⋅5722 1⋅5726 –0⋅0004 58⋅676 58⋅659 0⋅017
12 1 2 2 1⋅5493 1⋅5495 –0⋅0002 59⋅628 59⋅620 0⋅008
13 2 1 4 1⋅4698 1⋅4698 0⋅0000 63⋅214 63⋅214 0⋅000
14 0 3 0 1⋅3933 1⋅3934 –0⋅00008 67⋅126 67⋅121 0⋅005

Table 4. Unit cell parameters of natural dolomite at different Table 7. Vibrational frequency assignments (FT–IR).
temperatures.
ν1 ν2 ν3 ν4 ν1 + ν4 2ν2 + ν4
Temperature (°C) a (Å) c (Å) c/a (Å) Volume (Å)3 (cm–1) (cm–1) (cm–1) (cm–1) (cm–1) (cm–1)

RT 4⋅8247 15⋅9680 3⋅3135 322⋅28 – 881 1446 726 1881 2525

650 4⋅9680 17⋅1008 3⋅4422 365⋅5
750 5⋅0103 17⋅0146 3⋅3959 369⋅9
850 5⋅0466 16⋅9380 3⋅3563 373⋅6
950 5⋅0348 16⋅9558 3⋅3671 372⋅2

Table 5. FWHM of principal reflections of

dolomite for different heat treatments.
Dolomite

Temperature (°C) (104) (202)

650 0⋅16 0⋅24

750 0⋅16 0⋅24
850 0⋅16 0⋅12
950 0⋅16 0⋅16

Table 6. Observed X-ray data for the heat treated dolomite

sample (750°C) and reported data for calcite.
d (Å) for calcite

JCPDS
(h k l) Observed Reported Card 5-586 Intensity

(0 1 2) 3⋅8409 3⋅8520 3⋅80 12

(1 0 4) 3⋅0334 3⋅0300 3⋅035 100
(0 0 6) 2⋅8889 2⋅8340 2⋅845 3
(1 1 0) 2⋅4993 2⋅4950 2⋅495 14
(1 1 3) 2⋅2877 2⋅2840 2⋅285 18
(2 0 2) 2⋅1084 2⋅0940 2⋅095 18
(0 2 4) 1⋅9179 1⋅9260 – –
(1 1 6) 1⋅8837 1⋅8750 1⋅875 17
(2 1 5) 1⋅4968 1⋅4730 1⋅473 2
ASTM card no. (24-274). Figure 3. FTIR spectra of dolomite at different temperatures.
 Thermal decomposition of natural dolomite 343

3.4 FTIR spectral analysis

The typical transmittance FTIR spectra of the dolomite

mineral are shown in figure 3. In the room temperature
FTIR spectra of the samples, the out-of-plane bending
(ν2), the asymmetric stretching (ν3), and the in-plane
bending (ν4) modes of the carbonate group are found to
be active. Besides the internal modes, the ν2 + ν4 combi-
nation mode has also been observed (Adler and Kerr
1963; White 1974; Legodi et al 2001) and the observed
bands are compiled in table 7. Additionally peaks due to
silicates (1088 cm–1) and H-bonded water (at 3400 cm–1)
are visible (Keeling 1963). In addition, the weak band
due to quartz (465 cm–1) is also visible. The characteristic
dolomite bands are shifted to 713, 876 and 420 cm–1 in
the FTIR spectra of 750°C heat-treated dolomite sample
Figure 4. SEM photograph of dolomite at room temperature. (figure 3). This clearly indicates the structural transforma-
tion of dolomite to calcite. At this stage, a strong and broad
band at 540 cm–1 due to magnesium oxide is also visible.
At 950°C, the 1420 cm–1 band shifted to 1413 cm–1 and
the intensity of the band, 713 cm–1, very much decreased.
SEM microphotographs of the heat-treated and untreated
samples are shown in figures 4 and 5. SEM microphoto-
graph of the untreated samples shows distinct grains,
however, the heat-treated (950°C) samples exhibit clus-
ters of grains confirming the thermal decomposition.

4. Conclusions
DTA curve of dolomite shows two peaks at 777⋅8°C and
834°C. The two-stage decomposition reaction is con-
firmed by FTIR and XRD analysis. At 750°C, the dolo-
mite structure is changed into calcite, which is confirmed
by the presence of calcite characteristic peaks in the FTIR
spectra at 713, 875 and 1420 cm–1. The presence of cha-
Figure 5. SEM photograph of dolomite at 950°C. racteristic reflections at 3⋅0334 and 2⋅4993 Å for calcite
and MgO, respectively also confirm the first stage de-
composition of dolomite. At 950°C, the thermal decomposi-
sharp diffractions that can be attributed to dolomite tion reaction is completed which is confirmed by FTIR
(JCPDS Files card 11-78; 1999). Table 3 gives various and XRD analysis. The large fluctuation in the observed
Bragg reflections that are indexed using JCPDS Files activation energies is due to the presence of impurities in
card 11-78 and the calculated unit cell dimensions of the samples. The impurities could function as catalysts
natural dolomite at different temperatures are compiled in owing to their influence in the crystalline structure. The
table 4. In table 5, FWHM of (104) and (202) reflections results show that the clay with which they are heated
of dolomite are compared for different heat treatments. It which is reflected by the differences in the activation en-
can be noticed that the general pattern remains the same ergy, affects dolomite decomposition. These results could
and in addition, there is no change in the FWHM of the be important in site consideration of the dolomite sample.
(104) reflection. However, the (202) reflection shows
slight decrease in the FWHM value. The original reflec-
References
tion disappeared completely and new lines are developed
at the asymmetric positions of (104) and (202) reflec-
Adler H H and Kerr P F 1963 Am. Mineral. 48 839
tions. In table 6, the d spacing of the reflections observed Barcina L M, Espina A, Suarez M, Garcia J R and Rodriguez J
for 750°C heat treatment is compared with those reported 1997 Thermochim. Acta 290 181
in the literature for calcite indicating that the crystal Borgwardt R H 1985 AIchE 31 103
structure of dolomite has been transformed completely to Chan R K, Murthi K S and Harrison D 1970 Can. J. Chem. 48
calcite structure during heat treatment. 2972
344 S Gunasekaran and G Anbalagan

Coats A W and Redfern J P 1964 Nature (London) 201 68 Li Mao Qiang and Messing G L 1984 Thermochim Acta 78 9
Criado J M and Ortega A 1991 J. Thermal Anal. & Calorimetry Mc Cauley R A and Johnson L A 1991 Thermochim. Acta 185
37 2369 27
Criado J M and Ortega A 1992 Thermochim Acta 195 163 McInosh L M, Sharp J H and Willburn F W 1990 Thermochim.
Ersoy-Merichoyu A, Kuchkbayrak S and Durus B 1993 J. Acta 165 281
Thermal Anal. 39 707 Mulokozi A M and Lugwisha E 1992 Thermochim. Acta 194 375
Freeman E S and Carroll B 1958 J. Phys. Chem. 62 394 Powell E K and Searcy A W 1980 Metall. Trans. B11 427
Garcia Calvo E, Arranz M A and Leton P 1990 Thermochim. Romero Salvador A, Garcia Calvo E and Beneitz Aparico M
Acta 170 7 1989 Thermochim. Acta 143 339
Garn D D 1965 Thermoanalytical methods of investigation Rubiera F, Fuertes A B, Pis J J, Artos V and Marban G 1991
(New York: Academic Press) Thermochim. Acta 179 125
Glasson D R 1958 J. Appl. Chem. 8 793 Shoval S, Gaft M, Beck P and Krish V 1993 J. Thermal Anal.
Horowitz H H and Metzger G 1963 Anal. Chem. 35 1464 40 263
Iwafuchi K, Watanabe C and Otsuka R 1983 Thermochim. Acta Vogel A I 1951 Quantitative inorganic analysis (London:
66 105 Longmans) p. 582
JCPDS Files card 5-586 and 11-78 1999 Warne S S J, Morgan D J and Milondowshi A E 1981 Thermo-
Keeling P S 1963 Trans. Br. Ceram. Soc. (GB) 62 549 chim. Acta 51 105
Kunel D, Modi and Jadhav K M 1997 Indian J. Pure & Appl. White W B 1974 Min. Soc. Monogr. 4 227
Phys. 25 594 Xiao K, Sichen D U, Sohn H Y and Seetharaman S 1997 Met-
Legodi M A, deWaal D and Potgieter J H 2001 Appl. Spectrosc. all. and Mater. Trans. B28 1157
55 361 Yariv Shmuel 1989 Thermochim. Acta 148 421