Application of DTA/DSC and dilatometry for optimization of Ba–Ce–Y–P–Si–O glass phase for composite protonic conductors based on BaCe0.9Y0.1O3−δ

  • Katarzyna Silarska
  • Marcin Środa
  • Paweł Pasierb


The aim of this work was to determine the influence of chemical composition on thermal and electrical properties of selected glass from the BaO–CeO2–Y2O3–P2O5–SiO2 system. Properties of the investigated glasses were optimized in terms of glass-BaCe0.9Y0.1O3 composite protonic conductor preparation. Three types of glasses with different P2O5/SiO2 ratio were prepared and investigated. The DSC and XRD measurements indicated limited tendency to crystallization of obtained glasses up to T = 1023 K. The dilatometry measurements led to determination of the thermal expansion coefficients (TEC) and the softening point. It was found that obtained glasses soften in temperature range of 903–923 K, depending on the composition which is advantageous from the viewpoint of further preparation and formation of the composite material. The Electrochemical Impedance Spectroscopy measurements done as a function of temperature and gas atmosphere allowed to determine the activation energy of the electrical conductivity. Observed fluctuations of the activation energy with gas atmosphere suggest the possible role of the protonic defects in total conductivity. Based on the obtained results it can be stated that proposed glass system may be considered as the modifying phase for improvement of BaCe0.9Y0.1O3-based materials.


Glass-ceramic composite Conducting glass Protonic conductors Barium cerate Electrochemical impedance spectroscopy (EIS) 


The compounds from the perovskite group with general formula ABO3 including barium cerate (BaCeO3), constitute important class of materials exhibiting proton conductivity at high temperatures [1, 2, 3, 4]. These materials have been the subject of intensive research on their basic properties such as the crystalline structure, the structure of defects, transport and electrical properties or chemical resistance in atmospheres with CO2. The proposed modifications were chosen to improve a number of properties for the potential application of materials, and in particular to increase the ionic conductivity (proton) and to improve chemical resistance. Research on the materials from the perovskite-type ABO3 group (including A = Ba, Sr; B = Ce, Zr) are primarily aimed at increasing the proton conductivity and at improving the chemical resistance, particularly in atmospheres with CO2. So far, the proposed methods of modifying comprised mainly inducing of acceptor dopants from lanthanides group, yttrium [5, 6], holmium and indium [7], the formation of solid solutions with BaZrO3 [8] or BaTiO3 [9, 10] or modification of the microstructure through the choice and optimization of synthesis method. Described methods to improve the properties of materials based on BaCeO3 have been virtually exhausted, because all possible dopants have been already studied. Thus, new directions, including composite materials are needed. The proposed previous approach was to introduce another solid phases in the form of barium phosphate Ba3(PO4)2 and tungstate BaWO4 to obtain materials with improved properties [11]. Currently, such systems as BaCe0.8Sm0.2O3−δ–Ce0.8Sm0.2O2−δ [12], BaCe0.8Gd0.2O3−δ–Ce0.8Gd0.2O2−δ [13], BaCe0.8Y0.2O3−δ–Ce0.8Gd0.2O2−δ [14], BaCe0.9Y0.1O3–Ba3(PO4)2 [11] and BaCe0.9Y0.1O3–BaWO4 [15] were also investigated. Based on presented results, conception of composite materials seems to be a promising way towards improving the functional properties like electrical conductivity and chemical resistance in the presence of CO2. Improving the properties of materials based on barium cerate is possible by modifying the properties of grain boundaries. The most actual results concerning the composite materials indicate clearly that proper selection of modifying phase seems to be the crucial in terms of composite formation. One of the possible approaches may be the use of the glassy phase which fills the intergranular parts in the material. As it was mentioned in barium cerate-based materials grain boundaries have a predominant role of total conductivity. Improvement of electrical properties seems to be possible by modification of grain boundaries properties. The introduction of grain boundary modifying phase with non-negligible protonic conductivity into the BaCeO3 material may be beneficial in terms of total conductivity improvement. Favourably, such glassy material should exhibit the softening point at relatively low temperatures, close to the working temperature of the BaCeO3-based material, in order to ensure high total electrical conductivity.

The aim of this work was to determine the influence of chemical composition on thermal and electrical properties of selected glasses in the system 0.1(BaO0.9CeO20.05Y2O3)–0.9 (xP2O5ySiO2), where x = 0,6; 0,7; 0,8; y = 0,4; 0,3; 0,2. The composition of glass components was chosen to fit the composition of host BaCe0.9Y0.1O3 material. Namely, the ratio of Ba/Ce/Y cations in glass was kept the same as in the barium cerate, while P2O5/SiO2 ratio of glass-forming elements was varied in order to tune the glass properties, such as low glass transition temperature Tg. Glass with similar composition were the subject of interest [16, 17, 18, 19] and the promising electrical properties were found.

Materials preparation and experimental methods

In this work three compositions of glass with different content of P2O5 and SiO2–0.1(BaO0.9CeO20.05Y2O3)–0.9(0.6P2O50.4SiO2) (G1), 0.1(BaO0.9CeO20.05Y2O3)–0.9(0.7P2O50.3SiO2) (G2) and 0.1(BaO0.9CeO20.05Y2O3)–0.9 (0.8P2O50.2SiO2) (G3) were prepared. Appropriate amounts of barium carbonate (99.9%), cerium(IV) oxide (99.9%), yttrium(III) oxide (99.9%), diammonium hydrogen phosphate (99.9%) and silica powders (all reagents were supplied by Sigma-Aldrich Chemical Company, Inc.) were mixed and melted in alumina crucible for 2 h at 1673 K in an electrically heated furnace with Kanthal Super heating elements and after that poured on a brass plate and annealed at 823 K for 12 h. After annealing process glass was cut and polished to required shape and size, depending on the experimental method. DTA/DSC and dilatometry measurements were carried out using Perkin Elmer DTA-7 working in the heat flux mode (DSC mode) and TMA-7 equipment, respectively. DSC measurement was taken from 323 to 1173 K at 10 K min−1 with accuracy of ± 5 °C. The heating rate was 10 K min−1 for dilatometry. XRD measurements were conducted using X’Pert Phillips apparatus within the 2θ range 10°–70°. The study of electrical properties of the materials, such as bulk or grain boundaries conductivity, were done using Electrochemical Impedance Spectroscopy (EIS). The measurements were performed as a function of the temperature in three different atmospheres (air, air 100% RH at 298 K, and 5% H2 in Ar). The measurements were carried out using FRA 1260 frequency response analyzer coupled with the dielectric DI 1296 interface (both Solartron Analytical Inc., USA) in the frequency range of 10−2–10−7 Hz with the sinusoidal amplitude voltage of 20 mV in the temperature range of 293–923 K. Fitting the impedance spectra to the equivalent circuit was done using ZView 2.0 (Scribner Asc., USA) software. Prior to electrical measurements samples were cut to required shape: G1 (6.0 × 6.0 × 4.2 mm), G2 (8.35 × 8.7 mm × 2.6 mm) and G3 (7.0 × 7.0 × 4.0 mm) and thin layers of porous Pt electrodes (Heraeus Pt LP 11-4493 paste) were applied at both sides of every pellet, dried at 363 K for 1 h and fired at 850 °C for 10 min.

Results and discussion

The DSC measurements (sample: powder, 60 mg) were performed in order to investigate the possibility of crystallization of obtained glass. Figure 1 shows the DSC curves (heating rate 10 K min−1) obtained for G1–G3 glasses. Based on these results it was found that above 950 K (for G2 and G3) or 1023 (for G1) the possibility of crystallization occurs (Fig. 1).
Fig. 1

DSC curves obtained for G1–G3 glasses

To examine if the crystallization occurs, the samples of glass heated in 1123 K for 12 h were studied by the X-ray diffraction method (XRD). Figure 2 shows the confirmation of the presence of amorphous phase in the G1–G3 samples (powders). Narrowing of the glassy halo on XRD patterns for heat-treated samples indicates only ordering of the glassy network without the traces of crystallization. The stability of glass structure and properties at high temperatures is essential in terms of its further utilization in composite materials.
Fig. 2

XRD results for G1–G3 samples after heat treatment at T = 1123 K for 12 h

To determine the softening temperature of the glass dilatometry measurements were performed (sintered bodies–rectangular plates). It was found that the G1 glass softens at a temperature of 903 K (Fig. 3). Additionally, the thermal expansion coefficient was found to be 6.07 × 10−6/°C (at 673 K).
Fig. 3

Dilatometry results for glass G1 with softening point at the temperature of 903.7 K

Compared to the first glass (G1), the softening temperature of glass G2 is noticeably higher and reaches 913 K (Fig. 4). Further change of the glass composition (P/Si ratio) in glass G3 leads to the increase of the softening temperature and reaches 923 K (Fig. 5). Additionally, the thermal expansion coefficients were also determined, as collected for all samples in Table 1. These parameters are important in terms of compatibility with the host BaCe0.9Y0.1O3 in composite material, and will be discussed later.
Fig. 4

Dilatometry results for G2 glass with softening point at temperature of 915 K

Fig. 5

Dilatometry results for G3 glass with softening point at temperature of 923 K

Table 1

Comparison of thermal expansion coefficients determined for G1–G3 samples and compared with the literature data for BaCeO3 material [20]


Thermal expansion coefficient TEC × 106/°C (at 673 K)



6.07 ± 0.15

This work


6.77 ± 0.15

This work


6.77 ± 0.15

This work




The determined values of Tg by dilatometry are different that derived from DTA/DSC measurements. The observed differences in Tg temperatures can be explained in terms of different form of the samples (samples used for DTA/DSC measurements were powders and for dilatometry measurements were sintered bodies) measured and apparent differences of the applied methods.

The comparison of the softening point as a function of glass composition is shown in Fig. 6.
Fig. 6

Dependence of softening temperature on P2O5/SiO2 ratio

As it can be seen almost linear dependence of softening temperature on P2O5/SiO2 ratio can be observed. These results indicate the possibility of further lowering of the softening temperature during the optimization of glass composition. In terms of formation of composite material the first glass (G1) with the lowest value of softening point temperature seems to be the best candidate.

The comparison of Thermal Expansion Coefficients (TEC) determined for glass G1–G3 are listed in Table 1 and compared with the literature value of TEC for BaCeO3 material.

As can be noticed the TEC values of G1–G3 glass samples differ strongly from the TEC value of BaCeO3. The TEC values for G2 and G3 glasses are closer to the BaCeO3, and in terms of glass-BaCe0.9Y0.1O3 composite formation these compositions are better choice. It must be mentioned that the TEC value shown for BaCeO3 corresponds to the undoped, non-defected BaCeO3 material. Based on literature data, the formation of oxygen vacancies leads to the remarkable decrease of thermal expansion coefficient of BaCeO3-based material, as reported in [21], where lowering of TEC to the value of 10.6 × 10−6/°C were found. One can notice that this value is closer to TEC values determined for G2 and G3 glass samples, giving the chance of better compatibility of materials used for composite preparation. Additionally, doping of BaCeO3 (e.g. using acceptor dopants such as yttrium) can lead to further modification of TEC.

In order to determine the influence of glass composition on the electrical properties the G1, G2 and G3 samples were measured using the Electrochemical Impedance Spectroscopy (EIS) in three different gas atmospheres (air, wet air (100% RH at 298 K) and 5% H2 in Ar) as a function of temperature (293 K ≤ T ≤ 923 K). Such information is necessary for the optimization of composite material. All measured spectra were fitted using the equivalent circuit consisting of one Constant Phase Element (CPE) and Resistance (R) connected in parallel, using ZView 2.0 (Scribner Asc., USA) software. Figures 79 show the representative impedance spectra for the obtained samples (G1–G3) in 3 different atmospheres for temperature 873 K.
Fig. 7

EIS spectra measured for G1 glass in three different atmospheres: 5% H2 in Ar, air and wet air (100% RH at 298 K)

Fig. 8

EIS spectra measured for G2 glass in three different atmospheres: 5% H2 in Ar, air and wet air (100% RH at 298 K)

Fig. 9

EIS spectra measured for G3 glass in three different atmospheres: 5% H2 in Ar, air and wet air (100% RH at 298 K)

As can be seen, only one semicircle which can be attributed to the bulk properties is observed. Based on the determined conductivity values in function of temperature the activation energies of electrical conductivity were determined. Figures 1012 show the Arrhenius plots of ln(σT) versus 1/T for G1–G3 samples, for different gas atmospheres.
Fig. 10

Arrhenius plots for glass G1 and activation energies in different atmospheres: 5% H2 in Ar, air and wet air (100% RH at 298 K)

Fig. 11

Arrhenius plots for glass G2 and activation energies in different atmospheres: 5% H2 in Ar, dry and wet air (100% RH at 298 K)

Fig. 12

Arrhenius plots for glass G3 and activation energies in different atmospheres: 5% H2 in Ar, dry and wet air (100% RH at 298 K)

As it can be noticed, in the whole range of temperature presented dependences are linear. This suggests single electrical conduction mechanism in the whole range of temperatures investigated. Generally, almost no influence of glass composition on the absolute values of electrical conductivity can be observed. The presented data obtained for all glass composition and gas atmospheres were used to determine the activation energies using the Arrhenius equation:
$$ \sigma T = \sigma_{0} { \exp }\left( { - \frac{{E_{\text{a}} }}{kT}} \right) $$
where Ea is the activation energy for conduction, T is the absolute temperature, k is the Boltzmann’s constant and σ0 is a pre-exponential factor.
The determined activation energies are presented in Table 2.
Table 2

Activation energies Ea (eV) for samples G1–G3 in three different atmospheres


Dry air

Wet air (100% RH at T = 298 K)

H2 (5%) in Ar













The uncertainties of Ea determination does not exceed 0.01 for all determined values

Generally, the determined activation energy values are relatively high, which is in general agreement with the results obtained for similar glass, as reported in [19].

The influence of glass composition on activation energy is almost negligible, but a minimum can be noticed for G2 glass for all gas atmospheres. When considering the influence of gas atmosphere, the relatively higher activation energies are present for wet air atmosphere, for all glass G1–G3 compositions. The fluctuations of activation energy with gas atmosphere suggest the possibility of slight modification of electrical conduction mechanism where the protonic defects are also involved.

Summary and conclusions

This paper presents results concerning the thermal and electrical properties of three barium–cerium–yttrium–phosphate–silica glass with different P2O5 and SiO2 content. The glass compositions selected for tests in this work are based on the same elements as BaCe0.9Y0.1O3 and the cation ratios in glass are kept the same (Ba: Ce: Y = 1: 0.9: 0.1). It is assumed that such glass may be successfully used for the preparation of composites based on BaCe0.9Y0.1O3 ceramic protonic conductors. The modifying glass phase should have a positive impact on the properties of composite BaCeO3–glass materials.

In terms of composite material formation the modifying glass phase should exhibit some desired properties. The lack of mutual reactivity with host BaCe0.9Y0.1O3 material, high long-term thermal stability, similar thermal expansion coefficient of both materials and high electrical conductivity are the most important ones. In this work the selected thermal, structural and electrical properties were investigated using Differential Scanning Calorimetry (DSC), dilatometry (DIL), X-ray diffraction (XRD) and Electrochemical Impedance Spectroscopy (EIS) techniques.

The DSC and XRD measurements showed that the crystallization of obtained glass is not present, even at the temperatures as high as 1023 K. The dilatometry measurements allowed to determine the thermal expansion coefficients (TEC) and softening points for all glass compositions. It was found that these parameters are glass composition dependent. Namely, the increase of P2O5/SiO2 ratio led to the almost linear increase of softening point and abrupt change of thermal expansion coefficient (TEC) between G1 and G2 samples. The electrical conductivities of glass as a function of gas atmosphere and temperature were determined basing on the EIS measurements. The results plotted in Arrhenius coordinates (Figs. 1012) led to the determination of activation energies of electrical conductivity, for all glass compositions and gas atmospheres. Almost no influence of glass composition on the absolute values of electrical conductivity and activation energy was observed, although in the case of G2 glass the minimum of activation energy can be noticed for all gas atmospheres. Also, the observed fluctuations of activation energy with gas atmosphere suggest the possible role of the protonic defects in the total conductivity. Presented results indicate that proposed glass systems may be considered as modifying phase for improvement of BaCe0.9Y0.1O3-based materials. The lack of crystallization at higher temperatures, similar thermal expansion coefficients of glass and BaCe0.9Y0.1O3 with the possibility of further modification are the most important parameters, especially in terms of mechanical properties. The modification of grain boundaries by introduction of glass phase in the form of very thin layer may lead to the improvement of electrical conductivity of this region, thus leading to the improvement of total electrical conductivity of composite material. The selection and optimization of the method of BaCe0.9Y0.1O3–glass composite formation and the investigation of the properties of obtained composites is under way, and the results will be published later.



The financial support of Statutory Project for Science (Grant for Young Scientists No., 2017) and the Statutory Project for Science (Ministry of Science and Higher Education) No at the Faculty of Materials Science and Ceramics AGH UST are acknowledged.


  1. 1.
    Katahira K, et al. Protonic conduction in Zr-substituted BaCeO3. Solid State Ion. 2000;138:91–8.CrossRefGoogle Scholar
  2. 2.
    Kreuer K. On the development of proton conducting materials for technological applications. Solid State Ion. 1997;97:1–15.CrossRefGoogle Scholar
  3. 3.
    Medvedev D, Murashkina A, Pikalova E, et al. BaCeO3: materials development, properties and application. Prog Mater Sci. 2014;60:72–129.CrossRefGoogle Scholar
  4. 4.
    Takahashi T, Iwahara H. Protonic conduction in perovskite type oxide solid-solution. Rev Chim Miner. 1980;17:243–53.Google Scholar
  5. 5.
    Amsif M. Influence of rare-earth doping on the microstructure and conductivity of BaCe0.9Ln0.1O3−δ proton conductors. J Power Sour. 2011;196:3461–9.CrossRefGoogle Scholar
  6. 6.
    Muccillo R, Muccillo ENS, Andrade TF, Oliveira OR. Thermal analyses of yttrium-doped barium zirconate with phosphor pentoxide, boron oxide and zinc oxide addition. J Therm Anal Calorim. 2017;130(3):1791–9.CrossRefGoogle Scholar
  7. 7.
    Matskevich NI, Wolf T, Le Tacon M, et al. Heat capacity on data of DSC calorimetry and thermodynamic functions of barium cerate doped by holmium and indium oxides in the temperature range of 200–700 K. J Therm Anal Calorim. 2017;130(2):1125–31.CrossRefGoogle Scholar
  8. 8.
    Ryu KH, Haile SM. Chemical stability and proton conductivity of doped BaCeO3–BaZrO3 solid solutions. Solid State Ion. 1999;125:355–67.CrossRefGoogle Scholar
  9. 9.
    Pasierb P, Drożdż-Cieśla E, Rękas M. Properties of BaCe1−xTixO3 materials for hydrogen electrochemical separators. J Power Sour. 2008;18:17–23.CrossRefGoogle Scholar
  10. 10.
    Pasierb P, Drożdż-Cieśla E, Gajerski R, Łabuś S, Komornicki S, Rękas M. Chemical stability of Ba(Ce1−xTix)1−yYyO3 proton-conducting solid electrolytes. J Therm Anal Calorim. 2009;96(2):475–80.CrossRefGoogle Scholar
  11. 11.
    Łacz A, Grzesik K, Pasierb P. Electrical properties of BaCeO3-based composite protonic conductors. J Power Sour. 2015;279:80–7.CrossRefGoogle Scholar
  12. 12.
    Medvedev D, Maragou V, Pikalova E, et al. Novel composite solid state electrolytes on the base of BaCeO3 and CeO2 for intermediate temperature electrochemical devices. J Power Sour. 2013;221:217–27.CrossRefGoogle Scholar
  13. 13.
    Khandelwal M, Venkatasubramanian A, Prasanna TRS, et al. Correlation between microstructure and electrical conductivity in composite electrolytes containing Gd-doped ceria and Gd-doped barium cerate. J Eur Ceram Soc. 2011;31(4):559–68.CrossRefGoogle Scholar
  14. 14.
    Lin D, Wang Q, Peng K, et al. Phase formation and properties of composite electrolyte BaCe0.8Y0.2O3−δ–Ce0.8Gd0.2O1.9 for intermediate temperature solid oxide fuel cells. J Power Sour. 2012;205:100–7.CrossRefGoogle Scholar
  15. 15.
    Łącz A, Pasierb P. Synthesis and properties of BaCe1−xYxO3−δ–BaWO4 composite protonic conductors. J Therm Anal Calorim. 2013;113:405–12.CrossRefGoogle Scholar
  16. 16.
    Kumar S, Vanatier P, Levasseur A, et al. Investigations of structure and transport in lithium and silver borophosphate glasses. J Solid State Chem. 2004;177:1723–37.CrossRefGoogle Scholar
  17. 17.
    Sene FF, Martinelli JR, Gomes L. Optical and structural characterization of rare earth doped niobium phosphate glasses. J Non-crystalline Solids. 2004;348:63–71.CrossRefGoogle Scholar
  18. 18.
    Dias AG, Skakle JMS, Gibson IR, et al. In situ thermal and structural characterization of bioactive calcium phosphate glass ceramics containing TiO2 and MgO oxides: high temperature- XRD studies. J Non-crystalline Solids. 2005;351:810–7.CrossRefGoogle Scholar
  19. 19.
    Anvari SF, Hogarth CA, Moridi GR. Electrical conductivity of zinc-barium phosphate glasses. J Mater Sci. 1991;26:3639–42.CrossRefGoogle Scholar
  20. 20.
    Yamanaka S, Fujikane M, Hamaguchi T, et al. Thermophysical properties of BaZrO3 and BaCeO3. J Alloys Compd. 2003;359(1–2):109–13.CrossRefGoogle Scholar
  21. 21.
    Løken A, Haugsrud R, Bjørheim TS. Unravelling the fundamentals of thermal and chemical expansion of BaCeO3 from first principles phonon calculations. Phys Chem Chem Phys. 2016;18(14):31296–303.CrossRefPubMedGoogle Scholar

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2018

Authors and Affiliations

  • Katarzyna Silarska
    • 1
  • Marcin Środa
    • 1
  • Paweł Pasierb
    • 1
  1. 1.Faculty of Materials Science and CeramicsAGH University of Science and TechnologyKrakówPoland

Personalised recommendations