Controlled rate thermal analysis and differential scanning calorimetry of sepiolites and palygorskites

Article abstract of DOI:10.1016/S0040-6031(02)00228-9

A series of sepiolites, palygorskites and Rocky Mountain Leather (RML) clay minerals have been analysed by controlled rate thermal analysis and differential scanning calorimetry (DSC). Eight weight loss steps are observed and are structure and composition dependent. Three dehydration steps and five dehydroxylation steps are observed. The mass spectrometric curve mimicked the differential thermogravimetric (DTGA) curve enabling the detailed determination of the dehydration and dehydroxylation steps.

Full text of DOI:10.1016/S0040-6031(02)00228-9

Thermochimica Acta 397 (2003) 119–128
Controlled rate thermal analysis and differential scanning
calorimetry of sepiolites and palygorskites
∗
R.L. Frost , Z. Ding
Centre for Instrumental and Developmental Chemistry, Queensland University of Technology, 2 George Street,
Brisbane, G.P.O. Box 2434, Queensland 4001, Australia
Received 26 November 2001; received in revised form 11 April 2002; accepted 11 April 2002
Abstract
A series of sepiolites, palygorskites and “Rocky Mountain Leather” (RML) clay minerals have been analysed by controlled
rate thermal analysis and differential scanning calorimetry (DSC). Eight weight loss steps are observed and are structure
and composition dependent. Three dehydration steps and five dehydroxylation steps are observed. The mass spectrometric
curve mimicked the differential thermogravimetric (DTGA) curve enabling the detailed determination of the dehydration and
dehydroxylation steps.
©
2002 Published by Elsevier Science B.V.
Keywords: Attapulgite; Palygorskite; Sepiolite; Rocky Mountain Leather; Dehydration; Dehydroxylation; Controlled rate thermal analysis;
Differential thermogravimetric analysis
1
. Introduction
[(OH2Mg8(OH)·4Si12O30]·8H2O. The formulas are
written as such to indicate the two types of water
present, magnesium co-ordinated water and adsorbed
water [11,12]. Both sepiolite and attapulgite are chain
minerals: attapulgite consists of two double chains
Sepiolites and palygorskites form an important
group of clay minerals, which have many indus-
trial, catalytic and environmental applications [1–3].
Because of their structural morphology, sepiolites
and attapulgites have received considerable atten-
tion with regard to the adsorption of organics on the
clay surfaces [4–6] and to their use as support for
catalysts [7–10]. Sepiolites, attapulgites and various
forms of “Rocky Mountain Leather” (RML) (also
known as rocky mountain cork, mountain leather
or mountain wood) all have a fibrous like morphol-
ogy with a distinctive layered appearance. Atta-
pulgite has the structural formula [(OH2)4(Mg, Al,
2
−
of the pyroxene-type (SiO3)
like amphibole
(Si4O11)6 running parallel to the fibre axis. Unlike
the amphiboles, the attapulgite units are connected
to one another by shared oxygen atoms and not
cations such as Ca or Mg. Sepiolite consists of three
pyroxene-type chains instead of two as in attapulgite;
the oxygens linking the chains together are doubly
linked as opposed to the singly linked attapulgite oxy-
gen atoms. The structure of these two minerals results
in zeolite-like channels, which are approximately
3.7 Å × 6.0 Å and 5.6 Å × 11.0 Å wide, respectively.
These channels may be filled with water or organic
molecules. The water is partly arranged (ordered) in
these channels and water molecules are also bound to
the magnesium cations of the Mg (Al, Fe) brucite-like
−
Fe,) (OH)·2Si8O20]·4H2O and sepiolite the formula
5
∗
Corresponding author. Tel.: +61-7-38642407;
fax: +61-7-38641804.
E-mail address: r.frost@qut.edu.au (R.L. Frost).
0
040-6031/02/$ – see front matter © 2002 Published by Elsevier Science B.V.
PII: S0040-6031(02)00228-9

1
20
R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
ribbon edges that border the channels running along
the length of the crystals.
and dried in a desiccator to remove adsorbed water
before being submitted for thermal analysis. The clay
minerals were ground to a fine powder of <0.5 ␮m
particle size for thermal analysis.
The dehydration and dehydroxylation [13–16] of
the palygorskite clays have been studied in detail
by thermogravimetric techniques [17–19]. Indeed
a review of the thermal analysis of these types of
clay minerals has been published [20]. Differential
thermal analysis (DTA) in combination with other
techniques such as X-ray diffraction and Fourier
transform-infrared spectroscopy (FT-IR) proved most
useful for the study of the dehydration process [11,12].
It was found that dehydration of sepiolite and atta-
pulgite take place in a series of four steps [19]. Up
1.2. Thermal analysis
Differential thermogravimetric analysis (DTA) was
®
conducted on TA Instruments Thermogravimetric
Analyser (TGA, Q500) equipped with evolved gas
analysis (EGA) furnace, which was connected to a
quadrupole mass spectrometer (PFEIFFER, QMS 200
Prisma) through a 1/8 in., in diameter, transfer line.
Nitrogen was used as the purging gas and the flow rate
was controlled precisely at 80 ml/min. For each run,
sample powders were loaded onto platinum sample
◦
to 200 C both hygroscopic and zeolitic water were
◦
lost. Between 250 and 450 C bound water was lost;
more strongly bound water (co-ordinated water) and
◦
the hydroxyl units were lost in the temperature range
pan and heated to 1000 C at high-resolution heating
◦
◦
4
50–610 C. All of the dehydration/dehydroxylation
rate of 2 C/min, where the heating rate was dynam-
steps were endothermic in DTA. For both minerals,
the partial dehydration of bound H2O in the ranges
ically and continuously modified in response to the
changes in the rate of sample’s weight loss, termed
as controlled rate thermal analysis (CRTA). DSC was
◦
2
50–610 and 210–550 C resulted in the formation
®
of sepiolite anhydride and palygorskite anhydride,
respectively. Dehydration of the bound H2O in two
steps was attributed to the difference in bonding
position of H2O in the structure of these minerals.
With the advances in thermal analysis technology,
and the invention of new and novel techniques such
as controlled rate thermal analysis, it was considered
worthwhile to remeasure the thermal characteristics
of a selection of sepiolites and palygorskites. This
research reports these analyses based upon the CRTA
of sepiolites. A comparison is made with the results
of differential scanning calorimetry (DSC).
performed on a TA Instrument DSC Q10 analyser.
Sample powders were loaded into sealed alumina pan
◦
◦
and heated to 500 C at heating rate of 2 C/min. The
empty alumina pan was used as reference and the
heat flow between the sample and reference pans was
recorded.
2. Results and discussion
2.1. Thermogravimetic analysis and differential
thermogravimetric analysis
1
.1. Clay minerals
The thermogravimetric and DTA curves of a se-
piolite from Spain are shown in Fig. 1. Four distinct
weight losses are observed. Two are ascribed to dehy-
dration and two to dehydroxylation. The first step oc-
The clay minerals used in this research are (a)
Source Clay Minerals Repository Standard Atta-
pulgite PFl-1, (b) sepiolite (Sepsp-1) from Vallecas,
Spain, (c) sepiolite from Marschaum and (d) RML
collected from the F7 field trip of the 11th Inter-
national Clay conference (Ottawa, June 1997) [1].
Samples of sepiolite and palygorskite were obtained
from the van Der Marel collection. The RML sample
was collected from the Nicholson Silica mine near
Golden, British Columbia, Canada, in the sandstones
of the Ordovician Mount Wilson formation. The clays
were analysed by X-ray diffraction for phase purity
◦
curs at 50 C and is attributed to the loss of adsorbed
◦
water (Fig. 2). The second step is observed at 246 C.
The third step is complex with two overlapping
◦
weight loss steps observed at 450 and 494 C. The
final step is also complex and two weight loss steps
◦
are observed 787 and 820 C. Nagata et al. proposed
a set of steps for the dehydration and dehydroxyla-
tion of a sepiolite [19]. These steps correspond to (a)
the loss of adsorbed water, (b) the loss of hydration
water, (c) the loss of co-ordination water, (d) the loss

R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
121
Fig. 1. Thermogravimetric and differential thermogravimetric curves for the controlled rate thermal analysis of sepiolite from Vallecas
Spain).
(
Fig. 2. Component analysis of the DTGA curves for sepiolite.

1
22
R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
of water through dehydroxylation. Such a scheme is
represented by the following chemical equations:
is to measure the evolved water vapour from each of
the steps shown in the reactions in 3.1 using mass
spectrometry. Fig. 3 displays the mass spectrum of
evolved water together with the DTGA curve. Fun-
damentally this mass spectrum follows the DTGA
curve with precision. Hence, the percentage relative
area under the MS curve will provide a second set of
results for the weight changes. Table 1 also reports
these weight changes.
•
•
•
•
Step 1: Mg8Si12O30 (OH)4·(H2O)4·8H2O →
Mg8Si12O30 (OH)4·(H2O)4 + 8H2O
Step 2: Mg8Si12O30 (OH)4·(H2O)4·8H2O →
Mg8Si12O30 (OH)4·(H2O)2 + 2H2O
Step 3: Mg8Si12O30 (OH)4·(H2O)4·8H2O →
Mg8Si12O30 (OH)4 + 2H2O
Step 4: Mg8Si12O30 (OH)4·(H2O)4·8H2O →
The temperature for the first dehydration step was
8
MgSiO3 +4SiO2 + 2H2O
◦
found to be 50 C by DTGA and 48.3 by mass spec-
Such a scheme is an apparent oversimplification as
each of these steps may be subdivided into component
weight loss steps. If one uses the formula (Mg8Si12O30
trometry. However, as may be observed in Fig. 3, some
variation in determining the weight loss for this step
is experienced. Two overlapping weight loss steps are
observed at 222 and 246 by DTGA method and at 216
(
OH)4·(H2O)4·8H2O) for sepiolite then the theoretical
◦
weight loss for step 1 should be 11.0%. The % weight
loss observed for the TGA curve for sepiolite was
and 254 C by mass spectrometry. Two weight loss
◦
steps are observed at 450 and 494 C by the DTGA
◦
1
1%. The theoretical weight loss for step 2 is 2.76%
method and at 466 and 531 C by MS method. The
compared with the observed value of 3.3%. The third
step theoretical weight loss also should be 2.76% and
a value of 2.3% is observed. The final weight loss step
should be 5.21%. However, only 2.1% weight loss is
found. Thus, there is reasonable agreement with the
predicted and the observed weight loss values. The ob-
servation that the experimentally determined weight
loss step is less than that predicted is an indication
that some dehydroxylation has taken place in the pre-
vious steps with the evolution of the bound water. It
has been found that sepiolite folds to the anhydride
like form when about half the H2O of co-ordination is
total areas for these two steps are 32.7% by DTGA
and 32.2% by MS. Two further dehydroxylation steps
are observed at 787 and 820 by DTGA and at 785 and
823 C by MS methodology. The % relative weight
losses by the two methods are very comparable.
◦
2.3. DTGA and TGA of palygorskite
The principal difference between the formula of
sepiolite and palygorskite rests with the formula.
Sepiolite [Mg8Si12O30 (OH)4·(H2O)4·8H2O] is a
trioctahedral type mineral with three divalent mag-
nesium cations filling the three positions of the oc-
tahedron. On the other hand, palygorskite with the
◦
removed, at <200 C in vacuum or at approximately
◦
◦
3
00 C in air. Removal of the remaining H2O at 530 C
under reduced pressure produces little further struc-
tural change. Infrared evidence suggests that partially
dehydrated sepiolite is a folded structure with H2O of
co-ordination trapped in hexagonal holes. The remain-
ing H2O is lost, without significant structural change,
formula [(OH2)4(Mg, Al, Fe) (OH)·2Si8O20]·4H2O
5
has some of the magnesium cations replaced with
trivalent cations such as Al or Fe, thus, palygorskites
tend towards a dioctahedral structure. Fig. 4 displays
the TGA and DTGA curves for a palygorskite. Five
steps may be observed in the DTGA curve. The first
◦
at approximately 500 C under vacuum to give a true
anhydride [21–24].
◦
step occurs at 63.4 C and a % water loss of 12.7%
is observed. The second dehydration step occurs at
considerably lower temperatures than for sepiolites
and is observed at 168 C. A 16.3% relative weight
2
.2. Relationship between the mass spectrum of
◦
evolved water vapour and the DTGA curve
loss is observed for this step. Two weight loss steps
◦
The integral of the peaks in the DTGA curve in
Fig. 1 should be equivalent to the % weight loss for
each of the steps. The summary of these results for a
series of sepiolites and palygorskites are reported in
Table 1. A second method of estimating weight loss
are observed for the palygorskite at 357 and 415 C.
It is possible to speculate that these two tempera-
tures represent the dehydroxylation of the OH units
from the octahedral Fe and Al, respectively. However,
such a concept needs further proof using for example

1
24
R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
Fig. 3. DTGA curve and the mass spectral curve of evolved water from sepiolite from Vallecas (Spain).
Fig. 4. TGA and DTGA of palygorskite with high magnesium substitution.

R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
125
Fig. 5. Mass spectrum of evolved water vapour and DTGA curve of palygorskite with high magnesium substitution.
infrared spectroscopy as the cations are randomly dis-
tributed in the octahedral layer. Another more likely
consideration is that these two temperatures represent
the irreversible dehydration of residual bound water
matted intergrowth of asbestiform crystals, shows that
41 are either sepiolite or palygorskite (attapulgite) or
some mixture of the two clay minerals [15]. Fig. 6 dis-
plays the TGA and DTGA curves for a “RML”. The
thermal analysis pattern at first appears complex, how-
ever, the pattern appears as a combination of the two
minerals sepiolite and palygorskite. The RML sample
◦
and the folding of the tunnels (at 357 C) and the de-
◦
hydroxylation (at 415 C). A minor dehydroxylation
◦
step is also observed at 864 C and is ascribed to the
◦
dehydroxylation of OH units from the remnant mag-
nesium.
shows a strong loss of water of hydration at ∼50 C
both by MS and by DTGA. The difficulty of determin-
ing the amount is attributed to the sloping baseline in
both the DTGA and MS curves. MS shows a dehydra-
The mass spectrum of evolved water together with
the DTGA pattern of the palygorskite is illustrated in
Fig. 5. Again the MS curve follows with precision the
DTGA curve. The first weight loss step according to
◦
tion step at 216 and 237 C in which 50% of the total
water is lost. Two dehydroxylation steps are observed
◦
◦
◦
the MS pattern is observed at 65.8 C, which is slightly
at 364 and 428 C (DTGA) and 353 and 433 C (MS).
The weight loss by DTGA is 6.6 and 16.6% for these
two steps and 9.2 and 13.7% by MS. A further step
higher than the value from the DTGA result. The mass
◦
spectrum shows two weight losses at 131 and 171 C.
◦
◦
Two further weight losses corresponding to steps 3 and
is observed at 573 C (DTGA) and 594 C (MS) in
which 21.8% weight loss is observed.
◦
4
in Fig. 5, are observed at 348 and 407 C. Because
of the wide variation in the mass spectrum ion current,
step 5 could not be observed in the mass spectrum.
2.5. Comparison with other sepiolites
2
.4. DTGA and TGA of “Rocky Mountain Leather”
In this work, some four sepiolites and three paly-
gorskites together with an example of a “RML” were
studied using thermal analysis techniques. Consider-
able variation in the temperatures of dehydration and
A powder X-ray diffraction study of 52 samples
having the mountain leather texture of a fibrous,

1
26
R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
Fig. 6. TGA and DTGA of “Rocky Mountain Leather”.
dehydroxylation were obtained (Table 1). Differences
are observed between the temperature of dehydration
of the sepiolites and the palygorskites. This difference
is attributed to the different structure of the two clay
minerals. Sepiolites tend towards a trioctahedral struc-
ture whereas palygorskites tend towards a dioctahedral
structure. These conclusions are in agreement with the
findings of Shuali et al. who by deuteration studies of
sepiolite and palygorskites showed that the dehydrox-
ylation and crystallisation of the two minerals were
different [25]. The temperatures of dehydroxylation of
sepiolites and palygorskites are similar except for the
last dehydroxylation step, which is higher for the sepi-
olites. The variation may be attributed to the presence
of high amounts of Mg in the sepiolites.
2.6. Differential scanning calorimetry
The differential scanning calorimetric results for a
selection of sepiolites and palygorskites are shown in
Fig. 7. The results of the heat flow analyses are re-
ported in Table 2. There are six endotherms observed
◦
for sepiolites in the 50–500 C region. The principal
endotherm consists of two overlapping endotherms
centred upon approximately 83–89 and 100–122 C.
◦
The two endotherms are attributed to adsorbed and ze-
olitic or ‘channel’ water. These two endotherms make
up some 80% of the heat required to remove water
from the sepiolites. The DSC results show that the tem-
peratures for the removal of water for palygorskites are
somewhat lower than for sepiolites. The temperatures
Table 2
Results of the differential scanning calorimetry
◦
◦
◦
◦
◦
Sepiolites 1, C (%)
Sepiolites 2, C (%)
Sepiolites 3, C (%)
Palygorskite 1, C (%)
RML 1, C (%)
Heat flow (step 1)
Heat flow (step 2)
Heat flow (step 3)
Heat flow (step 4)
Heat flow (step 5)
Heat flow (step 6)
83 (41.7)
122.5 (38.0)
89.2 (34.7)
125.9 (24.1)
233.1 (7.0)
257 (3.8)
391.7 (13.9)
430.9 (16.5)
83.0 (44.3)
114 (33.8)
54.7 (33.3)
99.0 (53.1)
67.2 (40.4)
99.7 (41.7)
282 (5.9)
332.5 (14.4)
269 (6.6)
335 (13.7)
452 (1.6)
300.2 (6.6)
325 (2.9)
416 (4.0)
271.9 (4.6)
316 (10.0)
418 (3.2)

R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
127
well with the weight loss step observed by both DTGA
◦
and MS in the 225–250 C temperature range. An-
◦
other endotherm is observed in the 316–335 C region.
This endotherm makes up some 10–14.4% of the heat
flow. Again this endotherm is in harmony with the
◦
weight loss step observed in the 339–357 C temper-
ature range. A higher temperature endotherm is ob-
◦
served in the 416–452 C temperature region. This is
a small endotherm and is ascribed to the removal of
hydroxyls from the octahedral cation. This endotherm
corresponds to the weight loss step 5 (Table 1). The
DSC experiment is temperature limited and any higher
endotherms are not observed using this technique.
The variation in the DSC results between the differ-
ent sepiolites and palygorskites may be attributed to
a number of factors. Firstly, the variation in chemical
composition will effect the position of the endotherm.
DSC pattern (a) is typical of a high magnesium sepio-
lite. The endotherms are more temperature defined for
the ‘pure’ sepiolite. If some of the magnesium is re-
placed with other cations for example Fe and Al, then
a DSC pattern as observed in (b) is obtained. Such
replacement shifts the structure of the sepiolite from
trioctahedral towards dioctahedral. This structural
change will then effect the position of the endotherms.
For palygorskites and “RML” the endotherms are
broad to very broad as is observed in Fig. 7d and e.
Fig. 7. Differential scanning calorimetry of sepiolite from Val-
lecas (Spain) (a), sepiolite (Tanzia), (b) sepiolite, (Turkey) (c),
palygorskites (Texas, USA), and (d) Rocky mountain leather.
3
. Conclusions
of the endotherms are higher than that observed for
the thermogravimetric analyses. The first weight loss
A number of conclusions may be drawn:
◦
step was observed in the 50–60 C region and the sec-
◦
ond above 100 C. This difference may be attributed to
(a) The number of steps in the thermal analysis of
sepiolites and palygorskites are greater com-
pared with previous published results when using
high-resolution DTGA techniques.
the rates of heating of the two experiments. The TGA
experiment runs isothermally and at a heating rate of
◦
◦
2
C. The DSC experiment runs at 2 C/min. So it
could be expected that the results of the slower exper-
iment might be observed at lower temperatures. There
is also a fundamental difference between the way the
two experiments are conducted. The DSC is a con-
stantly heating experiment whereas the TGA experi-
ment is a high-resolution experiment with the weight
loss changes occurring under isothermal conditions.
(b) The mass spectrum of water for sepiolites, paly-
gorskites and “RML” follows with precision the
DTGA curves in the CRTA experiment.
(c) The CRTA experiment enables the separation of
the weight losses for the dehydration steps of se-
piolite clay minerals.
(d) Three dehydration steps are observed. These oc-
◦
◦
A third endotherm is observed in the 257–300 C
cur around 50–60 and 150–220 C.
region. The endotherm appears broad and makes up
between 3.8 and 6.6% of the total heat flow. This en-
dotherm is assigned to the heat required to remove
chemically bound water. The endotherm corresponds
(e) Two dehydroxylation steps are observed in the
CRTA experiment. A low temperature weight loss
◦
at 350–400 C and a higher temperature weight
◦
loss >500 C.

1
28
R.L. Frost, Z. Ding / Thermochimica Acta 397 (2003) 119–128
(
f) Significant differences in the results as deter- [3] M.F. Brigatti, G.C. Franchini, L. Medici, L. Poppi, A. Stewart,
Ann. Chim. (Rome) 88 (1998) 461.
mined by the dynamic and CRTA experiment are
observed.
[
4] S. Akyuz, T. Akyuz, J.E.D. Davies, J. Mol. Struct. 293 (1993)
279.
(g) The results of the DSC experiment harmonise
[5] S. Akyuz, T. Akyuz, J.E.D. Davies, K. Esmer, E. Ozel, J.
Raman Spectrosc. 26 (1995) 883.
with the CRTA TGA/DTGA experiment.
(
h) Six endotherms are observed in the DSC temper-
[6] S. Akyuez, T. Akyuez, J.E.D. Davies, J. Mol. Struct. 349
1995) 61.
◦
(
ature range from 50 to 500 C.
[7] Q. Jiang, Q. Lin, Z. Huang, G. Deng, R. Chen, Cuihua
Xuebao 18 (1997) 243.
(
i) Three endotherms are observed at around 80, 120
◦
and 270 C and are attributed to the removal of
[8] M.L. Occelli, R.S. Maxwell, H. Eckert, J. Catal. 137 (1992)
water from the sepiolite clay minerals.
36.
(
j) Two higher temperature endotherms are observed
[9] L. Yang, J. Wang, Z. Meng, Shiyou Huagong 20 (1991) 750.
10] A. Corma, A. Mifsud, J. Perez Pariente, Ind. Eng. Chem.
Res. 27 (1988) 2044.
◦
[
[
[
[
[
at around 330 and >400 C and are assigned to
the dehydroxylation of the sepiolites.
11] R.L. Frost, J.T. Kloprogge, G. Cash, Vibr. Spectrosc. 16
(k) This step results in the folding of the sepiolite.
(1998) 173.
(
l) Both the DSC and DTGA patterns of sepiolites
12] R.L. Frost, O.B. Locos, H. Ruan, J.T. Kloprogge, Vibr.
Spectrosc. 27 (2001) 1.
13] R. Otsuka, H. Hayashi, N. Imai, Waseda daigaku rikogaku
kenkyusho hokoku 47 (1970) 56.
14] R. Otsuka, T Mariko, T Sakamoto, Mem. Sch. Sci. Eng.
Waseda Univ. 37 (1973) 43;
differ from those of palygorskites.
(
m) RML appears to be a combination of sepiolite
and palygorskite.
R. Otsuka, N. Imai, M. Nishikawa, Niigata Prefecture Kogyo
Kagaku Zasshi 69 (1966) 1677.
15] D. Alvarez-Estrada, C. Sanchez, Bol. Soc. Espan. Ceram. 6
Acknowledgements
[
[
[
[
[
[
(
1967) 717.
16] G.E. VanScoyoc, C.J. Serna, J.L. Ahlrichs, Am. Min. 64
1979) 215.
The Centre for Instrumental and Developmental
Chemistry, of the Queensland University of Tech-
nology is gratefully acknowledged for financial, and
infra-structural support for this project. The Aus-
tralian Research Council (ARC) is thanked for the
funding of the Queensland Thermal Analysis Facil-
ity including the differential scanning calorimeter
and the high-resolution thermogravimetric analysis
instrument coupled to the Pfeiffer (Balzers) mass
spectrometer.
(
17] J. Khorami, A. Lemieux, Thermochim. Acta 138 (1989)
97.
18] T. Kiyohiro, R. Otsuka, Thermochim. Acta 147 (1989)
1
27.
19] H. Nagata, S. Shimoda, T. Sudo, Clays Clay Min. 22 (1974)
85.
2
20] J.L. Martin-Vivaldi, P. Fenoll Hach Ali, in: R.C. Mackenzie
(Ed.), Differential Thermal Analysis: Applications, Vol. 2,
Academic Press, London, New York, 1972 (Chapter 20).
21] H. Hayashi, R. Otsuka, N. Imai, Am. Min. 54 (1969) 1613.
22] J.L. Ahldrich, C.J. Serna, J.M. Serratosa, Clays Clay Min.
[
[
2
3 (1975) 119.
23] C.J. Serna, J.L. Ahldrich, J.M. Serratosa, Clays Clay Min.
3 (1975) 452.
References
[
2
[1] K. Takahashi, M. Fujii, T. Kodama, Y. Kitayama, Mater.
Integr. 19 (1999) 71.
[24] C.J. Serna, G.E. van Scoyoc, J.L. Ahldrich, Am. Min. 62
(1977) 784.
[2] H.T. Ma, G.C. Deng, L. Xu, M.O. Li, R.T. Chen, S. Han,
G.H. Xu, J. Nat. Gas Chem. 8 (1999) 93.
[25] U. Shuali, S. Yariv, M. Steinberg, M. Mueller-Vonmoos, G.
Kahr, A. Rub, Thermochim. Acta 135 (1988) 291.