Synthesis and characterization of aluminum carboxylate gels

Article abstract of DOI:10.1016/S0040-6031(03)00265-X

Aluminum formate and basic acetate gels obtained from commercial pseudoboehmite were synthesized in this work. Phase evolution after successive thermal treatments was studied on the raw material and both gels. In both carboxylates the influence of aging t

Full text of DOI:10.1016/S0040-6031(03)00265-X

Thermochimica Acta 407 (2003) 33–40
Synthesis and characterization of aluminum carboxylate gels
C. Clara,b,∗, A.N. Sciana,c, E.F. Agliettia,c
Centro de Tecnolog´ıa de Recursos Minerales y Cerámica, CETMIC, CIC-CONICET-UNLP,
a
Cno. Centenario y 506-C.C. 49, B1897ZCA, M.B. Gonnet-Bs. As., La Plata, Argentina
Comisión de Investigaciones Cient´ıficas de la Provincia de Buenos Aires, CICBA, La Plata, Argentina
b
c
Consejo Nacional de Investigaciones Cient´ıficas y Técnicas, CONICET, La Plata, Argentina
Received 30 January 2003; received in revised form 10 April 2003; accepted 2 May 2003
Abstract
Aluminum formate and basic acetate gels obtained from commercial pseudoboehmite were synthesized in this work.
Phase evolution after successive thermal treatments was studied on the raw material and both gels. In both carboxylates the
influence of aging time on the composition, the crystalline structure and the phase transition temperatures was studied by X-ray
diffraction (XRD) and differential thermal analysis (DTA) and thermogravimetric analysis (TGA). Two different stages were
observed in the thermal decomposition of both carboxylates: the first, corresponding to the decarboxilation and dehydration
of the gels, was different for both cases because of the different type of carboxilic acids used; and the second, corresponding
to the formation of transition aluminas, was the same whatever the precursor used. However, whenever the aging time was
changed on any of the carboxylates, the mechanisms of the thermal transformations were affected.
© 2003 Elsevier B.V. All rights reserved.
Keywords: Pseudoboehmite; Aluminum formate; Aluminum basic acetate; Alumina; Gels
1. Introduction
precursor particles can be prepared by different tech-
niques [1–6].
The boehmite (AlOOH) is the most common
alumina precursor, which undergoes a sequence of
transformations with temperature from the transition
aluminas (␥, ␦, ␪) to the ␣-stable form. Formation
of transition alumina phases is a consequence of the
relative surface energies associated with the phases.
Usually these transformations are simply written as
␥ → ␣-Al2O3, as these phases are the predominant
ones, which are also of technical interest. Many au-
thors succeed in getting the ␣-Al2O3 transition at
lower temperatures than 1200 ◦C [7–9].
Alumina is an important oxide for the production of
technical ceramics and is extensively used in catalysis,
microelectronics, refractories and structural applica-
tions. Alumina ceramic membranes are new materials
technically important in separation processes associ-
ated with low requirements of energy, high selectivity
and capacity to operate at low and high temperatures.
Dip-coating on a porous support is one of the methods
used to make microporous membranes. In this method,
depending on the membrane pore size required, the
The membrane synthesis by the dip-coating method
using boehmite sols was discussed in many works
[9–12]. The preparation of alumina ceramic mem-
branes from aluminum hydrates or salts as precursors
∗
Corresponding author. Tel.: +54-2214840247;
fax: +54-2214710075.
E-mail address: cetmic@netverk.com.ar (C. Clar).
0040-6031/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0040-6031(03)00265-X

34
C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
implies a competition between the obtention of a
controlled porosity system with a suitable mechanical
strength and the material densification as a conse-
quence of the remaining stable phases after thermal
treatment.
Callender et al. [13] made a complete study of
the reaction between pseudoboehmite with sev-
eral carboxylic acids. The synthesis of carboxylate-
alumoxanes without aqueous media was studied by
Landry et al. [14].
crucibles with a heating rate of 10 ◦C/min up to
1200 ◦C, in air atmosphere and using tabular ␣-Al2O3
as reference. Weight losses were calculated from the
TGA.
Carboxylates were obtained by stirring mixture of
the pseudoboehmite with of each of the acids men-
tioned. The reaction took place at room temperature.
3. Results
Metal carboxylates are potential precursors for de-
position and the following formation of oxides at low
temperatures [15]. Other authors study the densifica-
tion of the gels obtained from aluminum basic acetate
with seeding agents [16]. Rajendran and Bhattacharya
[15] study the formation of ␣-Al2O3 obtained from
commercial carboxylates, observing the transforma-
tion sequence and the intermediate phases.
The characteristics and behavior of the aluminum
formate and basic acetate gels used to produce mem-
branes by dip-coating were analyzed in this work. Gels
were obtained by reaction between pseudoboehmite
and formic and acetic acid. The sequence of the dif-
ferent phases which were developed during thermal
decomposition could be observed. The aging time in-
fluence on the behavior of these gels was also studied.
3.1. Pseudoboehmite characterization
The starting pseudoboehmite was characterized by
XRD in its original state and as function of thermal
treatments. The ␥-AlOOH was dried at 100 ◦C and
then calcined at 800, 1000 and 1200 ◦C successively,
during 1 min with a heating rate of 10 ◦C/min. The
XRD patterns obtained are shown in Fig. 1.
The diagram of the sample treated at 100 ◦C
shows peaks that are in accordance with those of
the ␥-AlOOH (card no.: 21-1307). This material as
slightly crystalline; hence it was identified as pseu-
doboehmite, as reported by Mackenzie [17]. The
transformations of the transition aluminas, ␥-Al2O3
phase (card no.: 10-0425), ␦-Al2O3 phase (card no.:
16-0394), ␪-Al2O3 phase (card no.: 35-0121), were
observed in the diffractograms of the samples cal-
cined from 800 to 1200 ◦C. The first traces of the
␣-phase (card no.: 42-1468) could be appreciated
when the samples reached the 1200 ◦C. It was ob-
served that after a soaking time of 2 h at 1200 ◦C the
transformation to the ␣-phase was almost complete,
in agreement with Rajendran and Bhattacharya [15].
The TGA and DTA curves obtained for the pseu-
doboehmite previously dried at 100 ◦C are shown in
Fig. 2. Two stages of weight losses were observed up
to 600 ◦C in the TGA curve. The first one extended
from 20 to 230 ◦C and a loss of 14.1% took place,
corresponding to the removal of physically bonded
water molecules that are present in the pores of the
␥-AlOOH. This stage had an endothermic effect that
appeared by DTA at 147 ◦C. The second lost occurred
from 230 to 600 ◦C and corresponded to the pseu-
doboehmite transformation into ␥-Al2O3, with an en-
dothermic peak at 492 ◦C and a weight loss of 14.7%.
The exothermic effects corresponding to the ␥ →
␦, ␦ → ␪ and ␪ → ␣-Al2O3 transitions were not
2. Materials and methods
2.1. Raw materials, methods and equipment used
A commercial pseudoboehmite (␥-AlOOH), whose
characteristics will be described in Section 3, was
used as starting material for the synthesis of aluminum
carboxylate gels. Formic acid, HCOOH proanalysis
98% (w/w) d = 1.218 g/ml and glacial acetic acid,
COOHCH3 99.77% (w/w) d = 1.05 g/ml, were also
used.
Crystalline phases were determined by X-ray
diffraction (XRD) with a Philips Pw3710 equipment,
Cu K␣, Ni filter, in the 2θ = 5–80◦ region. The
cards of the identified compounds correspond to the
International Centre for Diffraction Data (ICDD).
Simultaneous differential thermal analysis (DTA)
and thermogravimetric analysis (TGA) were carried
out using a Netzsch STA 409 equipment. Tests were
performed with 200 mg of sample on platinum–iridium

C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
35
Fig. 1. XRD of ␥-AlOOH at different temperatures.
detected in this diagram; however, the existence of
3.2. Aluminum carboxylates
3.2.1. Synthesis
An amount of 59 ml of formic acid was added to
a 30 g of pseudoboehmite sample to obtain aluminum
these processes was confirmed by XRD.
This sequence of phases as function of temperature
was in agreement with what was observed by other
authors [15,17].
Fig. 2. TGA and DTA of ␥-AlOOH.

36
C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
formate; on the other hand, 95.5 ml of acetic acid were
added to another 30 g of ␥-AlOOH to obtain aluminum
basic acetate. Both samples were made at room tem-
perature, manually stirring the reactives for half an
hour. In both cases white color gels were obtained.
They were stored in closed containers and set aside at
room temperature.
and peaks corresponding to the not reacted ␥-AlOOH
could be identified. The diffractogram for the gel
aged 200 days shows the peaks corresponding to
non-hydrated Al(CHO2)3 (card no.: 38-0583) which
is the predominant phase, as well as peaks of lower
intensity of Al(CHO2)3·3H2O and of ␥-AlOOH that
did not react.
By means of XRD and DTA–TGA, those gels were
characterized as regards their evolution under succes-
sive thermal treatments. Using these techniques the
changes and their behavior in relation with aging were
studied.
The DTA obtained for both samples of this gel are
shown in Fig. 4. As regards the aged 30 days sample,
three endothermic effects were observed. The first one,
at 259 ◦C, was due to the loss of adsorbed H2O by the
trihydrate formate. The second one, at 382 ◦C, was due
to the decomposition of aluminum formate to produce
␥-Al2O3; and the third one, at 485 ◦C, caused by the
residual ␥-AlOOH decomposition.
In the aged 200 days sample an important en-
dothermic effect was centered at 135 ◦C, due to the
adsorbed water loss. The shoulder at approximately
260 ◦C hardly suggest the trihydrate formate presence.
At 365 ◦C, the effect of the salt decomposition and at
476 ◦C, the pseudoboehmite conversion to ␥-Al2O3,
could be observed. These last two effects took place at
a temperature slightly lower than that of the aged 30
days sample. In this aged gel two exothermic effects
appeared: the first one, at 887 ◦C, that corresponds to
the ␦ → ␪-Al2O3 transition and the second one, at
Landry et al. [14] studied the same reaction be-
tween the pseudoboehmite and different carboxylic
acids (R = n-Cn H2n +1 with nꢀ = 1–3 and 13) but
under different conditions, that is to say, the authors
employed reflux at temperature during 4 days and
they obtained in all cases stable compounds known as
carboxylate-alumoxanes.
ꢀ
ꢀ
3.2.2. Aluminum formate characterization
XRD patterns for aluminum formate gel samples
taken at different stages of aging are shown in Fig. 3.
The diffractogram for the gel aged 30 days can be
observed in that figure, where well defined peaks that
correspond to Al(CHO2)3·3H2O (card no.: 38-0655)
Fig. 3. XRD of aluminum formate gel at different aging times.

C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
37
Fig. 4. DTA of aluminum formate gel at different aging times.
1110 ◦C, due to the ␪ → ␣-Al2O3 transformation. In
the aged 30 days gel these exothermic effects did not
appear.
Both gels were calcined at three temperatures, 460,
800 and 1200 ◦C, under the same conditions used for
the DTA, and the different crystalline phases were
studied by XRD. In theory, the transformations that
occurred in the aluminum carboxylates after the salt
decomposition were the same as those for the pseu-
doboehmite.
In these two samples, aged 30 and 200 days, it
was observed that those transformations occurred at
considerably different temperatures. In both cases,
the peaks corresponding to the ␥-AlOOH that had
not reacted and to the ␥-Al2O3 that started to develop
were observed at 460 ◦C. At 800 ◦C the ␦-Al2O3 was
identified as the predominant phase and some ␥- and
␪-Al2O3 traces were observed. The diffractograms of
the samples calcined at 1200 ◦C are shown in Fig. 5.
␪-Al2O3 phase and only ␣-Al2O3 traces were identi-
fied in the sample aged 30 days. The peaks observed
were wide and of low intensity, indicating a scarce
crystalline development of the phases present. In the
sample aged 200 days the ␣-Al2O3 was clearly iden-
tified as the predominant phase, with some ␪-Al2O3
traces. In this case the peaks were sharp and intense,
indicating high crystallinity. These observations were
in agreement with the DTA analysis; in the sample
aged 200 days, only the exothermic effects corre-
sponding to the alumina transformations were evident.
The results of the TGA analysis for the aluminum
formate gel samples after 30 days of aging (FAl 30D)
and those after 200 days of aging (FAl 200D) are
shown in Table 1. In all cases, samples were dried at
100 ◦C before being tested, therefore, the free water
was not included in the total mass loss indicated. The
stage 1, within the temperature range between 14
and 176 ◦C, corresponds to the elimination of water
molecules physically adsorbed by the gels; and the
stage 2, from 176 to 1000 ◦C, corresponds to the
different transformations they underwent.
Table 1
TGA of the studied gels
Sample
Stage
Mass loss (%)
FAl 30D
1
2
2.66
62.68
FAl 200D
AAl 30D
AAl 200D
1
2
10.67
53.26
1
2
5.5
48.55
1
2
2.09
59.9

38
C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
Fig. 5. XRD of aluminum formate gel at different aging times and calcined at 1200 ◦C.
In the case of the aluminum formate gel it was ob-
crystalline structure of the material calcined and the
temperatures at which the different transformations
take place.
served that the Al(CHO2)3·3H2O underwent signifi-
cant changes during aging, because when time passed,
the water molecules that were chemically bonded to
the compound, went on being part of the water ad-
sorbed by the gel, generating non-hydrated formate.
This loss of the three water molecules of hydration,
verified by all the techniques mentioned, implies that
this gel is not a stable compound.
Due to the exact coincidence of the XRD peaks
with the cards mentioned and to the different re-
sults obtained from the DTA diagrams for the com-
pounds studied in this work compared with those
synthesized by Landry et al. [14], we consider that
the changes mentioned about the synthesis condi-
tions made the product obtained chemically different
and consequently having a different behavior. In the
carboxylate-alumoxanes the authors mentioned be-
fore did not observed changes in the composition,
nor in the solubility after several months at room
temperature. On the contrary, in the carboxylates ob-
tained by the route explained in this work, focussing
on the aluminum formate, changes in the chemical
composition were observed.
3.2.3. Aluminum basic acetate characterization
The XRD patterns of the aluminum acetate gel sam-
ples removed at different aging times can be observed
in Fig. 6. Peaks corresponding to the aluminum basic
acetate, Al(C2O2H3)2(OH) (card no.: 13-0833), and
peaks corresponding to the ␥-AlOOH residual were
identified in all cases. It could be observed in this fig-
ure that the Al(C2O2H3)2(OH) presented a peak in-
tensity that continuously increased during the first 150
days, indicating that a crystallization process existed.
By contrast, in the samples removed after that pe-
riod of time, there was not an important variation of
crystallinity with the passing of time. It was also ob-
served that the intensity of the peaks corresponding to
␥-AlOOH decreased with the passing of time.
The DTA diagrams for the aluminum basic acetate
gel samples aged 30 and 200 days are shown in Fig. 7.
Both samples presented peaks in similar positions.
The endothermic effects at 125 and 103 ◦C corre-
sponded to the loss of water adsorbed, the salt decom-
position occurred at 363 and 368 ◦C and the residual
pseudoboehmite decomposition took place at 475 and
476 ◦C, respectively. A final slight exothermic effect
As a result, we conclude that the aging time on
the aluminum formate gel, is an important parameter
that determines the composition of this material, the

C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
39
Fig. 6. XRD of aluminum basic acetate gel at different aging times.
that corresponded to the ␪ → ␣-Al2O3 transition
at 1186 ◦C for the aged 30 days gel and at 1197 ◦C
for the aged 200 days gel was also observed. The
exothermic effect shift to higher temperatures indi-
cates that the crystallinity increases the temperature
of ␪ → ␣-Al2O3 transformation.
To evaluate the behavior of this gel under succes-
sive thermal treatments, a 30 days sample was cal-
cined at four temperatures and then analyzed by XRD.
At 570 and 800 ◦C the ␥- and ␦-Al2O3 presence was
detected. At 1000 ◦C the ␪-Al2O3 phase was identi-
fied with ␦- and ␣-Al2O3 traces, whereas at 1200 ◦C
Fig. 7. DTA of aluminum basic acetate gel at different aging times.

40
C. Clar et al. / Thermochimica Acta 407 (2003) 33–40
the ␣-Al2O3 was the predominant phase with ␪-Al2O3
traces.
the gel without aging, at the same temperature, the
␪-Al2O3 was the predominant phase.
From the TGA for the aluminum basic acetate gel
samples, aged 30 days (AAl 30D) and aged 200 days
(AAl 200D) (Table 1), it was observed that as time
passes the amount of water adsorbed by the gel is lower
and that the sample aged 200 days presents a greater
weight loss in the second stage of mass loss. This may
have been caused by the fact that the pseudoboehmite
was still reacting with the acetic acid present in the
gel and therefore the amount of Al(C2O2H3)2(OH)
formed increased as time passed. This can be also ex-
plained by the intensity decrease of the peaks corre-
sponding to ␥-AlOOH shown in the XRD diagrams of
the aged samples.
In relation with the aluminum basic acetate, due to
the coincidence of the XRD peaks with the cards men-
tioned and to the DTA curves that present different
peaks (related with different processes) compared with
the compounds obtained by the synthesis route stud-
ied by Landry et al. [14], we conclude that this work
was based on the aluminum carboxylate study while
those authors studied the carboxylate-alumoxanes.
In conclusion, the aging time of the aluminum ba-
sic acetate gel is an important parameter, but different
to that of the aluminum formate gel. In this case, 200
days after the gel synthesis, a crystallinity increase of
this material was the most important difference ob-
served, what means a greater temperature for the ␪ →
␣-Al2O3 transition.
On the other hand, in the aluminum basic acetate
gel, when aging, an increase of crystallinity was only
observed. In this case, a slight temperature increase of
the alumina phase transitions took place.
As a conclusion, the aging time affected the chem-
ical composition and properties of the aluminum for-
mate gel, whereas for the aluminum basic acetate this
parameter only slightly increased the temperature of
the Al2O3 phase transition.
Acknowledgements
The authors wish to thank Trad. M.E. Ghirimoldi
for her contribution to this work.
References
[1] H. Schmidt, Ceram. Int. 43 (1994) 90–93.
[2] S.S. Tarar, V. Günay, Ceram. Int. 44 (1995) 234–237.
[3] S.S. Tarar, V. Günay, Ceram. Int. 45 (1996) 254–260.
[4] Y. Kinemuchi, T. Suzuki, W. Jiang, K. Yatsui, J. Am. Ceram.
Soc. 84 (2001) 2144–2146.
[5] H.K. Kang, S.S. Park, M.M. Son, H.S. Lee, H.C. Park, Br.
Ceram. Trans. 99 (2000) 26–30.
[6] R.N. Das, P. Pramanik, Br. Ceram. Trans. 99 (2000) 153–158.
[7] M. Kumagai, G.L. Messing, J. Am. Ceram. Soc. 69 (1985)
500–505.
[8] M. Sathiyakumar, F.D. Gnanam, Br. Ceram. Trans. 98 (1999)
87–92.
[9] M. Kumagai, G.L. Messing, Am. Ceram. Soc. Bull. 73 (1994)
88–91.
[10] A.F.M. Leenaars, A.J. Burggraaf, J. Colloid Interface Sci.
4. Conclusions
105 (1985) 27–40.
[11] Y.S. Lin, A.J. Burggraaf, J. Am. Ceram. Soc. 74 (1999) 219–
224.
Two stages were observed in the thermal decom-
position of: pseudoboehmite, aluminum formate and
basic acetate. The first stage corresponds to the de-
composition of the compounds, which is different for
each of the three cases, due to the different type of
phases present. In the second stage, corresponding to
the formation of transition aluminas, the formation of
␥-Al2O3 followed by the ␦ → ␪ → ␣-Al2O3 transi-
tions was observed for the three materials.
The aging time affected both of the carboxylates
prepared, but in a different way.
When aging, the trihydrate aluminum formate gel
generated non-hydrated formate, which transforms at
1200 ◦C almost completely into ␣-Al2O3, while in
[12] A.C. Pierre, D.R. Uhlmann, J. Am. Ceram. Soc. 70 (1987)
28–32.
[13] R.L. Callender, C.J. Harlan, N.M. Shapiro, C.D. Jones, D.L.
Callahan, M.R. Wiesner, D.B. MacQueen, R. Cook, A.R.
Barron, Chem. Mater. 9 (1997) 2418–2433.
[14] C.C. Landry, N. Pappé, M.R. Mason, A.W. Apblett, A.N.
Tyler, A.N. Maclnnes, A.R. Barron, J. Mater. Chem. 5 (1995)
331–341.
[15] M. Rajendran, A.K. Bhattacharya, Mater. Lett. 39 (1999)
188–195.
[16] I.P. Kilbride, A.J. Barker, Br. Ceram. Trans. 93 (1994) 187–
191.
[17] R.C. Mackenzie, Differential thermal analysis, in: R.C.
Mackenzie (Ed.), Fundamental Aspects, Oxides Hydroxides
of Higher-Valency Elements, vol. 1, Academic Press, London,
1970, Chapter 9, pp. 279–286.