Elsevier

Physica B: Condensed Matter

Volume 389, Issue 2, 15 February 2007, Pages 275-280
Physica B: Condensed Matter

Mean-coordination number dependence of the fragility in Ge–Se–In glass-forming liquids

https://doi.org/10.1016/j.physb.2006.06.163Get rights and content

Abstract

Differential scanning calorimetry measurements have been performed on elemental Se as well as on GexSe94−xIn6 (x=4, 8, and 11 at%) and on GeySe88−yIn12 (y=5, 7, and 9 at%) chalcogenide glasses. From the cooling rate dependence of the fictive temperature, the apparent activation energies, Δh*, and the fragility indices, m, as defined in the strong–fragile glass-forming liquid concept, are determined. It is found that, in Ge–Se–In system, there is an evolution from strong (m=67) to fragile (m=116) glass-forming liquids. The dependence of ‘m’ on the mean-coordination number, Z, is also obtained. This dependence is rationalized by assuming that, in this glassy alloy system, there is a tendency for the formation of In2Se3 clusters.

Introduction

The study of structural relaxation in the glass transition region of glass-forming liquids is important from academic and technological points of view. Relaxation processes in glasses occur at temperatures lower than their glass transition temperatures, Tg. The glass transition temperature can be defined as the temperature at which the equilibrium liquid viscosity is of the order of 1012 Pa s or as the temperature at which the average relaxation time in the equilibrium liquid is about 100 s [1]. From the variations of the viscosity, η, or relaxation time, τ, with the normalized reduced Tg/T quantity, a classification scheme of glass-forming liquids has been proposed. This scheme is well known as the ‘fragility concept’ or as strong–fragile glass-forming liquid concept [2], [3] and allows the definition of a fragility index, m [4]. This ‘m fragility’ or ‘steepness’ index is a measure of the rate at which the relaxation time τ (or related properties) decreases with increasing temperature around Tg and is given by [4]m=dlog10(τ)d(TgT)|T=Tg.In this classification scheme, glass-forming liquids which exhibit an approximately Arrhenius temperature dependence of their τ's are defined as strong and are characterized with a low value of m (m16 [5]). Those whose τ's are fitted with a Vogel–Fulcher–Tammann (VFT) equation are referred to as fragile and are characterized by a high value of m (m200 [6]). It is well recognized that oxide glass formers such as GeO2 and SiO2 with well-formed tetrahedral network rigid structures and directional bonds belong to the category of strong forming liquids [3], [7], [8], [9], [10]. On the other hand, linear polymeric materials are fragile forming liquids [11]. Thus, the concept of fragility in supercooled liquids has often been used as a basis for organizing data on amorphous materials: inorganic glasses [12], thermoplastics [9], polymer networks [13], and liquid-crystalline polymers [11].

However, despite the extensive efforts expended in research on this ‘fragility’ concept, the factors that determine m for a given liquid or polymer remain poorly understood [14]. As the unpredictability of this property is a problem, because many aspects of the viscous liquid, such as aging behaviour and non-exponentiality of relaxation, are closely related to m, it is important to obtain new data on any liquid that can be vitrified to get a better understanding of the strong–fragile glass-forming liquid concept.

Because the addition of a third element as an impurity has a pronounced effect on the structural relaxation process, we propose in this paper to study the role played by metallic indium on the relaxation processes of covalently bonded Ge–Se glasses. Therefore, the purpose of this work is to obtain new m-data on Ge–Se–In glassy alloys, correlate it with their mean-coordination number which could be changed in a controlled way, compare the newly obtained results with those already existing on GexSe100−x vitreous system, and discuss them in terms of the strong and fragile behaviours.

Section snippets

Method

The determination of the fragility index, m, requires the choice of an expression for the relaxation time, τ. For a glass, τ is known to be dependent on the temperature and on some order parameters that define the glass structure [15], [16], [17], [18], [19], [20]. In the present work and from among the different expressions in the literature for τ, the Tool–Narayanaswamy–Moynihan (TNM) expression [17], [18], [21] has been chosen which is given byτ=τ0exp(xΔh*RT)exp((1-x)Δh*RTf),where x (0x1)

Experimental

The samples were prepared from a mixture of high-purity Ge, Se, and In according to the melt-quench technique. The mixture was introduced into cylindrical glass silica tubes (inner diameter=8 mm) and sealed under vacuum of 10−5 Torr. Then, the tubes were placed in an electric furnace and annealed at 450 °C for 24 h. Afterwards, the temperature of the furnace was raised to 850 °C for 48 h. At this temperature, the tubes were frequently shaken to homogenize the melt. The quenching was done in an

Results and discussion

Fig. 1, Fig. 2 show the DSC curves obtained at the constant heating rate of 20 Kmin−1 on, respectively, Ge4Se90In6 and Ge5Se83In12 glassy alloys, cooled with different rates in the range 20–0.5 K min−1. Similar DSC curves were obtained on the other investigated glass compositions and are not shown here to avoid repetition. The measured glass transition temperatures, Tg, are given in Table 1. These values are in good agreement with the reported ones for this system [24], [25], [26] and as one

Conclusions

By adding In to the Ge–Se binary system, we change the mean-coordination number of the alloys by changing the number ratio of Ge/Se atoms. This affects the relaxation behaviour in the glass transition region by changing the value of the fragility index. In the Z range from approximately 2.20 to 2.26, we find that the addition of In to Ge–Se system destroys the bi-dimensional network of Ge–Se and a linear-like behaviour is restored. This is because the size of the CRR available for the

Acknowledgements

One of us (G.S) would like to thank the Arab Fund for Economic and Social Development, The Arab Fund Fellowship Program, Kuwait, for their financial support during the course of this work.

References (44)

  • C.A. Angell

    J. Non-Cryst. Solids

    (1991)
  • P. Cortes et al.

    J. Non-Cryst. Solids

    (1998)
  • J.M. Saiter et al.

    Physica B

    (2000)
  • M.E. Godard et al.

    J. Non-Cryst. Solids

    (1998)
  • A. Saiter et al.

    Polymer

    (2002)
  • K. Chebli et al.

    Physica B

    (2001)
  • A. Saiter et al.

    J. Non-Cryst. Solids

    (2002)
  • M.K. Rabinal et al.

    J. Non-Cryst. Solids

    (1995)
  • A. Giridhar et al.

    J. Non-Cryst. Solids

    (1992)
  • J.M. Hutchinson et al.

    J. Non-Cryst. Solids

    (1991)
  • J.M. Saiter et al.

    Mater. Lett.

    (1996)
  • J. Ledru et al.

    J. Non-Cryst. Solids

    (1998)
  • J.C. Phillips

    J. Non-Cryst. Solids

    (1979)
  • J.M. Saiter et al.

    Physica B

    (1998)
  • J.M. Saiter et al.

    J. Non-Cryst. Solids

    (1991)
  • J.M. Saiter et al.

    Physica B

    (2005)
  • C.A. Solunov

    Eur. Polym. J.

    (1999)
  • A. Saiter et al.

    Eur. Polym. J.

    (2006)
  • W.T. Laughlin et al.

    J. Phys. Chem.

    (1972)
  • C.A. Angell
  • K.L. Ngai et al.

    The Physics of Non-Crystalline Solids

    (1992)
  • T.A. Vilgis

    Phys. Rev. B

    (1993)
  • Cited by (0)

    View full text