INFLUENCE OF MECHANOACTIVATION

ON THE B2O3-SiO2-Na2O GLASS THERMAL STABILITY

A.S. Bykov, T.I. Filinkova, R.I. Gulyaeva, E.A. Pastukhov,

S.A. Petrova, R.G. Zakharov

Institute of Metallurgy, UD RAS, Ekaterinburg, Russia

Abstract

The influence of mechanoactivation in a planetary ball mill of the initial mixtures of the B2O3-SiO2-Na2O glasses on their thermal stability has been studied. It was concluded the growth of duration of pretreatment of an initial charge for glass preparation in planetary-type mill FRITSCH from 1 hour up to 6 hours yields a positive effect.

Introduction

Glasses containing fine (nanometric) particles of semi-conducting chemical elements or chemical compounds are of major scientific interest during last 25 years. These advanced materials, owing to the special optical properties (mainly to optical absorption), are rather promising as a base for making such nonlinear optical devices with the fast response, as high-speed optical commutators, nonlinear waveguides, flat display panels, light-emitting materials, etc. for an electronics, computational hardware, a fiber optics [1-3].

One of the principal ways of making these new materials [1-3] is joint melting of premixed substance of a base (so-called base glass) and a semiconductor additive with formation of a high-temperature solution and the subsequent quenching of the melt with reception of a metastable supersaturated solid state. Then the non-equilibrium material undergoes special heat treatment resulting in precipitation of semiconducting nanocrystals in the bulk of base glass. Of course, the material of a base should demonstrate high thermodynamic stability during such heat treatment. In other words the base glass should not crystallize at treatment temperatures.

Base glass is usually presented by well-known kinds of optical silicate glasses, including one of 54 mol% B2O3–33 mol% SiO2–13 mol% Na2O system [2]. In the present work this one (named sample N) together with the glass of 72mol% SiO2– 9mol%B2O3 – 19mol% K2O (sample K) were investigated.

Experiment

Initial mixtures were prepared from the SiO2, HBO3 and Na2CO3 or K2CO3 reagents, correspondingly (fig.1).

Fig.1 XRD patterns of the initial reagents: 1 - Na2CO3, 2- B2O3, 3-K2CO3, 4-SiO2.

For mechanoactivation of the initial oxides and mixtures the FRITSCH planetary ball mill and the AGO2 activator were used. XRD of the treated samples revealed mechanochemical activation of the mixture N (33mol% SiO2 – 54mol%B2O3 – 13mol% Na2O) leads to the formation therein of the boron based phases, whereas in the mixture K metastable phases based on potassium carbonate appeared after treatment (fig.2). The initial silicon oxide was XRD-amorphous and the grinding led to some short range changes (fig.3). The components were mixed together and melted in air at temperatures ranged between 1200 and 1650C. The time and temperature of melting depended on the glass composition. Final glasses are presented in fig.4.

Fig.2 XRD patterns of the mixture K differently treated in the planetary ball mills: 1, 5, 6 – Fritsch, t=1h, 6h, 14h; 2-4- AGO2, t=30s, 60s, 90s
Fig.3. XRD patterns of the amorphous silicon oxide before (1) and after (2) grinding.

Phase transitions in glass samples of the given composition depending on a method of initial charge materials preparation were studied using thermal analyzer NETZSCH STA 449C Jupiter. This instrument was equipped with quadrupole mass-spectrometer QMS 403 СА Aeolos for the examination of gaseous products of reactions in samples. The so-called method of a synchronous thermal analysis (STA), which is a combination of differential scanning calorimetry (DSC) and thermogravimetry (TG), was used in the work. The instrument was preliminary calibrated by heats of phase transitions of a standard set of the reference samples supplied by NETZSCH to make it possible the quantitative determination of thermal effects.

Fig.4 XRD-patterns of the glasses: 1, 2 – sample K; 35 – sample N. Treatment duration for the initial mixtures: 1,3 – 1h, 5 – 6h (Fritsch); 2,4 – 30s (AGO).

In all the experiments the samples under study were heated in platinum analytical crucibles with lids as fast as 10 K/min in air. Procedure of samples preparation for STA was the same for better comparability of observed results: each batch of glass was pounded manually in an agate mortar during 1.5 min before placing into a measuring cell of the thermal analyzer. The masses of samples were about 40-50 mg. We studied two kinds of glass (No 1 and No 2) of the same composition 54 mol% B2O3–33 mol% SiO2–13 mol% Na2O. The only difference was duration of initial charge processing in planetary-type mill FRITSCH before glass smelting. It was 1 hr in the first case and it was increased up to 6 hr in the second one.

Results and Discussion

Measurement results are shown in fig.5 and fig.6.

The thermal behaviour of the glass in fig. 5 completely meets conventional expectations.

Fig. 5. DSC (solid line) and TG (dotted line) of heating of glass No 1 prepared from charge after mechanoactivation for 1 hour.

At an early stage of heating of a sample No 1 (starting with 272ºС) small exothermic peak equal to -11,29 J/gm is observed in DSC curve (fig. 5). Temperature of an onset of transformation here and after was determined according to a routine practice of a thermal analysis as an intersection point of tangents to a basic line and to lateral peak generating line. This peak matches to minor decrease of a sample mass in TG curve. Mass-spectrometry data indicate a presence of carbon dioxide in an atmosphere above a sample at the mentioned temperature. Considering that the source of Na2O in an initial charge for glass melting was sodium carbonate, which dissociates when heating, one can suppose, that the observed low-temperature exothermic reaction may be caused by a desorption of the residualCO2 from a surface of sample’s particles even through small heating.

The further heating of the sample results in a second order phase transition, which is not accompanied by a discontinuous change of an enthalpy. It is a process of devitrification. Figure 5 gives glass transition temperatureTg = 593 ºС.

At 841 ºС one can observe "cold" crystallization of the glass, which is structural ordering of the sample’s material. According to DSC data crystallization heat is equal to -91.32 J/gm.

And finally at 1038ºС a melting of the sample occurs. Thermogravimetry results indicate a beginning of gradual decrease of sample mass due to melt evaporation.

The charge for the glass No 2 from fig. 6 was exposed to more vigorous mechanoactivation.

Fig. 6. DSC (solid line) and TG (dotted line) of heating of glass No 2 prepared from charge after mechanoactivation for 6 hours.

In this case the same sequence of transformations is observed, as in fig. 5. Nearly all values of temperatures and heats of observed phase transitions are close to previous ones. The only principle difference from the curves in fig. 5 is the shift of an onset of crystallization to higher temperatures approximately by 60 ºС. It is visible from experimental data that the sample No 2 has some superposition of crystallization and melting peaks against each other, that makes difficulties for an estimation of melting temperature. That is why its value was found in this case as an inflection point of thermo analytic curve using the plot of the 1-st derivative of DSC signal.

Comparison of experimental results in. fig. 5 and fig. 6 makes it possible to conclude that the growth of duration of pretreatment of an initial charge for glass preparation in planetary-type mill FRITSCH from 1 hour up to 6 hours yields a positive effect since it results in increase of thermal stability of the studied glass.

The work was supported by RBRF (project 07-03-00308).

References

  1. V. Sukumar, R.H. Doremus. Growth Kinetics of CdS Nanocrystallites in Glass through Quantum-Size Effects. Phys. Stat. Sol. (b), 1993, v 179, p. 307-314.
  2. V. Sukumar, S-C Kao, Pratima G.N. Rao et all. Metal Coating on Semiconductor Particles in Glass. J. Mater. Res. 1993, v. 8, No. 10, p. 2686-2693.
  3. R. Doremus, S-C Kao, R. Garcia. Optical Absorption of Small Copper Particles and the Optical Properties of Copper. Applied Optics. 1992, v. 31, No. 27, p. 5773-5778.

1