Crystallization Studies on Se-Te-Cd Chalcogenide Glasses
M. A. Abdel Rahim
*1, A. El-Korashy
*2and S. Al-Ariki
Physics Department, Faculty of Science, Assiut University, Assiut, Egypt
The thermal crystallization behavior of bulk Se90xTe10Cdx(x¼0, 3, 9, and 15 at%) glasses was studied by differential thermal analysis, (DTA), under non-isothermal conditions. The glass transition temperature, (Tg), the onset crystallization temperature, (Tc), and the peak
temperature of crystallization, (Tp), were found to be dependent on both the composition and the heating rates. From the dependence on the
heating rates () of (Tg) and (Tp), the activation energy for the glass transition, (Et), and the activation energy for crystallization, (Ec), were
calculated and their composition dependence is discussed. The crystalline phases resulting from DTA and scanning electron microscopy (SEM) have been identified using X-ray diffraction. The results indicate one dimensional growth from the surface to the inside for all the studied compositions. The kinetic parameters determined have made it possible to discuss the glass forming ability.
[doi:10.2320/matertrans.MC200928]
(Received July 31, 2009; Accepted November 4, 2009; Published January 25, 2010)
Keywords: selenium-tellurium-cadmium system, chalcogenide glasses, crystallization kinetics, thermal analysis
1. Introduction
Chalcogenide glasses exhibit many useful properties and have recently drawn great attention because of their use in various solid state devices. The electrical properties are influenced by the structural effects associated with the thermal effects and can be related to thermal induced transitions.1)Common experimental techniques used to study the structure of chalcogenide glasses are electron micros-copy, X-ray diffraction, electrical resistivity and thermal Analysis.2–5)
The study of the crystallization kinetics of a glass upon heating can be undertaken in several different ways. In calorimetric measurements two techniques have been em-ployed for the study of the crystallization behavior upon heating, namely isothermal and non-isothermal crystalliza-tion analyses.6,7) In the isothermal method, the sample is
brought quickly to a temperature above the glass transition temperature,Tg, and the heat evolved during the
crystalliza-tion process at a constant temperature is recorded as a function of time. In the other method, the sample is heated from room temperature, at a fixed heating rate (), and the heat evolved is again recorded as a function of temperature. The present work is concerned with the study of the crystallization kinetics and the evaluation of the crystalliza-tion parameters for Se90xTe10Cdx(x¼0, 3, 9 and 15 at%)
chalcogenide glasses by using the non-isothermal method. The effect of composition on the crystallization mechanism has been reported.
2. Experimental
The bulk materials of Se90xTe10Cdx (x¼0, 3, 9 and
15 at%) were prepared by the usual melt quench technique; the highly pure materials (99.999%) were weighted accord-ing to their percentages and sealed in evacuated silica tubes which were heated at 900C for 20 h. During the melt process
the tube was frequently rocked to intermix the constituents and to increase homogenization of the liquid. This treatment was followed by quenching in an ice/water mixture. The amorphous nature of the sample was proved by X-ray diffraction. Differential Thermal Analyses (DTA) was carried out by using a DTA Perkin-Elmer DTG-60 under non-isothermal conditions.
The glass transition temperatures (Tg), the onset
crystal-lization temperatures (Tc), and the peak crystallization
temperatures (Tp), were determined with an accuracy of
1K using the microprocessor of a thermal analyzer. The crystallized fraction () was calculated using the partial area analysis. For scanning electron microscopy (SEM) examina-tion, sample surfaces were etched using a mixed solution of sodium hydroxide and hydrogen peroxide. The X-ray diffractometer, Philips type 1710, was used to identify the crystalline phases in the annealed glassy samples.
3. Results and Discussions
Differential thermal analysis (DTA) experiments were performed at different heating rates ranging from 5 to 25 K/min to investigate the crystallization kinetics of Se90xTe10Cdx chalcogenide glasses. Figure 1 shows the
thermograms of as-prepared Se90xTe10Cdxglasses (x¼0, 3,
9 and 15 at%) at a heating rate of 10 K/min. Three character-istic phenomena are clear in the studied temperature region. The first one, Tg, corresponds to the glass transition
tem-perature. The second one, Tc, corresponds to the onset
crystallization temperature. The last characteristic temper-ature,Tp, identified the crystallization temperature.
Numer-ical values of these temperatures are given in Table 1. The morphology of the studied samples after annealing to crystallization temperature using the same heating and cooling rate as used in DTA measurements were examined by SEM. The sample was fractured and gold coated before SEM examination to study both the internal and surface morphology. The scanning micrographs of specimens of different compositions, annealed at 398 K for 1 h, are shown in Fig. 2(a)–(d). The microstructure of the annealed sample for the composition Se90Te10 is shown in Fig. 2(a). The *1Corresponding author, E-mail: maabdelrahim@yahoo.com
*2Present address: Basic Science Department, Faculty of Engineering,
British University, Cairo, Egypt
photomicrograph shows the existence of a polycrystalline structure consisting of crystallites embedded in amorphous phases. The crystalline morphology is not distinct due to the fact that the crystallization is only in its initial stages. On the other hand, the area covered by the crystalline particles increased by increasing the Cd content in the Se90xTe10Cdx
chalcogenide glasses. The crystalline morphology of the annealed sample for the composition Se87Te10Cd3 is shown
in Fig. 2(b). The surface morphology showed the appearance of a homogeneously distributed dendrite crystalline phase embedded in the glass matrix. Further increase of the Cd content in Se-Te-Cd system reveals that the amount of the
(a) (b)
(c) (d)
Fig. 2 The SEM photograph for the compositions annealed at 398 K for 1 h: (a) Se90Te10, (b) Se87Te10Cd3, (c) Se81Te10Cd9 and (d) Se75Te10Cd15.
350 400 450 500 550
Se75Te10Cd15
Se81Te10Cd9
Se87Te10Cd3
Se90Te10
Temperature, T / K
Exo
Endo
Tg Tc
∆
Q
Tp
Fig. 1 The DTA Thermograms for the as-prepared compositions (Se90Te10), (Se87Te10Cd3), (Se81Te10Cd9) and (Se75Te10Cd15).
10 20 30 40 50 60 70 80 90
Intensity
,
I
/
arb. unit
∆
∆ ∆
∗ Ο
(d)
(c)
(b)
(a)
Diffraction angle, 2θ / °
Ο SeTeCd CdSe
∗ CdTe Se ∆ Te
[image:2.595.74.265.71.225.2]Fig. 3 X-ray diffraction for compositions annealed at 398 K for 1 h. (a) Se90Te10, (b) Se87Te10Cd3, (c) Se81Te10Cd9and (d) Se75Te10Cd15.
Table 1 Tg,Tc,Tp,Tc-Tg, (at 10 K/min) B,Et andHrof the Se-Te-Cd
glasses.
Parameters Se90Te10 Se87Te10Cd3 Se81Te10Cd9 Se75Te10Cd15
Z 2 2.12 2.36 2.6
Tg(K) 335 336.1 337 338.2
Tc(K) 374.9 370.6 368.2 365.6
Tp(K) 414.7 407.4 402.9 382.4
Tc-Tg(K) 39.8 34.5 31.2 27.5
Et(kCal/mol) 37.70 38.52 39.56 40.70
[image:2.595.329.520.73.222.2] [image:2.595.130.465.273.597.2] [image:2.595.46.291.674.787.2]transformed crystalline phase increased and the crystallized particles decrease in size. The photomicrograph in Fig. 2(c) shows the surface microstructure of the annealed Se81Te10Cd9 composition. A polycrystalline structure
con-sisting of different crystalline phases embedded in an amorphous matrix is observed. Some of these crystallized particles are interconnected and others are isolated. The scanning electron micrograph for the annealed Se75Te10Cd15
composition is shown in Fig. 2(d). The photomicrograph shows very fine structure with different crystalline phases and sizes. The crystallites are dispersed homogeneously and occupy most of the structure.
In order to identify the crystalline phases that appeared in the DTA and SEM examinations, the X-ray diffraction for specimens annealed at a temperature close to the crystal-lization temperature is shown in Fig. 3. Analysis of the X-ray diffraction pattern reveals that the composition, Se90Te10,
which was annealed at 398 K for 1 h, has two phases of Se and Te embedded in an amorphous matrix as shown in Fig. 3(a). Addition of Cd to the composition leads to the formation of new phases like CdTe, SeTeCd and CdSe as shown in Figs. 3(b), (c) and (d).
The variation of Tg with composition for Se90xTe10Cdx
(x¼0, 3, 9 and 15 at%) is shown in Fig. 4. It is observed that there is a small increase in Tg with increasing Cd
content. On the other hand, the average coordination number
Z increases with increasing the Cd content. The variation of Tg with the average coordination number follows the
relation:
lnTg¼aZþb ð1Þ
where Z denotes the average coordination number per atom which is calculated in terms of covalent bonding. TheTgof
a multi-component glass is known to depend on several independent parameters such as band gap, coordination number, effective molecular weight and the type and fraction of various structural units formed.8–11)The slight increase of
Tg with increasing Cd content in the present system may be
explained by considering the structural changes occurring due to the further addition of Cd content. The generally accepted structural model for amorphous Se includes3)two molecular species meandering chains which contain helical
chains of trigonal Se and Se8ring molecules of monoclinic
Se. In the present case, the addition of Cd makes bonds with Se and is probably dissolved in the Se chains. Hence, the number of Se8 rings decreases while the number of long
Se-Te polymeric chains and Se-Te mixed rings increases. It is known that the glass transition temperature,Tg, should
increase with increasing chain length and decreases with increasing ring concentration12) hence T
g increases with
increasing Cd content. Furthermore, we attribute the increase of Tg with increasing Cd content to the marginal increases
which occur in the coordination number (shown in Fig. 4) and the mean atomic masses of these glasses.
On the other hand, the glass transition activation energy (Et) deduced by Kissinger, formula:13)
lnð=Tg2Þ ¼ ðEt=RTgÞ þconst: ð2Þ
Where is the heating rate,Et is the activation energy for
the glass transition andRis the gas constant. Figure 5 shows
lnð=T2
g) versus 1000=Tg plots for Se90xTe10Cdx glasses,
from which the activation energy for the glass transition,Et,
was evaluated and listed in Table 1. TheEtvalues lie in the
range generally observed for chalcogenide glasses.14,15)It is
clear from Table 1, the activation energy for glass transition,
Et, increases with increasing Cd content. The increase ofEt
with increasing Cd content resulted in an apparent increase of Tg with increasing Cd content at all the heating rates.
These results are in good agreement with those obtained by others.5,15)
The values of the onset crystallization temperature,Tc, and Tc-Tg at heating rates 10 K/min are listed in Table 1. It is
clear from the table that both Tc and Tc-Tg decrease with
increasing Cd content. Both Tc and Tc-Tg represented the
thermal stability of the glass.14,16) The kinetic resistance to
crystallization is given by the difference betweenTcandTg.
The larger difference gives higher resistance to crystalliza-tion. The glass samples with lowerTc-Tgvalues are expected
to have higher electrical conductivity.17)These results are in
good agreement with other workers.5,18,19)Another parameter usually employed to estimate the glass forming tendency (GFT) is the one introduced by Hruby,20)and is defined as follows:
Hr¼ ðTcTgÞ=ðTmTcÞ ð3Þ
2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7
5.79 5.80 5.81 5.82 5.83 5.84
5.85 0 2 4 6 8 10 12 14 16
05 K/min. 10 K/min. 15 K/min. 20 K/min. 25 K/min.
ln(T
g
)
Coordination number, Z
Cd content (%)
Fig. 4 The variation of the glass transitionTgwith average coordination
number Z and Cd content for Se90xTe10Cdx (x¼0, 3, 9 and 15 at%) glasses.
2.92 2.94 2.96 2.98 3.00 3.02 -10.2
-10.0 -9.8 -9.6 -9.4 -9.2 -9.0 -8.8 -8.6 -8.4
Se90Te10 Se87Te10Cd3 Se81Te10Cd9 Se75Te10Cd15
ln(
α
/
Tg
2 )
1000 / Tg/ K-1
Fig. 5 Plots of lnð=T2
gÞ versus 1000=Tg for the compositions
[image:3.595.328.527.72.217.2] [image:3.595.74.262.74.225.2]whereTmis the peak melting temperature. The parameterHr
gives the probability of obtaining a glass which increases as
Tm-Tc decreases and Tc-Tg increases. It is usually difficult
to prepare glasses withHr0:1 while good glasses can be
formed if Hr0:4.21,22) The deduced values of Hr for the
studied compositions are listed in Table 1. The table reveals that the GFA decreases with increasing Cd content.
The activation energy, Ec, of amorphous crystalline
transformation were calculated using the equation derived by Kissinger13)
lnð=Tp2Þ ¼ ðEc=RTpÞ þconst: ð4Þ
Figure 6 shows the plot oflnð=T2
pÞversus1000=Tp for the
studied compositions. From the slopes of the straight lines, the activation energies,Ec, for the crystallization process are
deduced and listed in Table 2. The activation energy,Ec, and
the frequency factor, kp, were calculated using the method
proposed by Augis and Bennett.23)
lnð=TpÞ ¼ ðEc=RTpÞ þlnkp ð5Þ
The plots of lnð=TpÞ versus 1000=Tp of different heating
rates for the studied compositions are shown in Fig. 7. From the slopes of the fitted to the data we calculated the activation energy, Ec, for the crystallization process. In addition, the
origin ordinate of these curves gives the frequency factor,
kp, values which are given in Table 2 together with theEc
values.
The kinetic parameters of our four compositions have also been evaluated through the technique used by Gao and Wang.24)Based on the Johnson-Mehl-Avrami equation and the Henderson equation,25)the authors obtained the following relation:
lnðd=dtÞp¼lnð0:37nkpÞ ðEc=RTpÞ ð6Þ
whereis the volume fraction of crystallization,ðd=dtÞpis
the maximum crystallization rate andkpis the crystallization
reaction rate constant. Figure 8 shows the plot oflnðd=dtÞp
versus1000=Tp for the studied compositions. TheEcvalues
for the studied compositions have been determined and are listed in Table 2.
Furthermore, another proposed method26)is used to deduce the order of crystallization reaction (n) for continuous heating crystallization at fixed temperature. The volume fraction of crystals precipitated in glass heated at a uniform ratecan be related to (n) by the expression
dfln½lnð1Þg=dln¼ n ð7Þ
On this bases, plotting ln½lnð1Þ versus lnðÞ where
is obtained at the same temperature from a number of crystallization exotherms taken at different heating rates. The plots are shown in Fig. 9 for the composition Se75Te10Cd15. The deducedns for all compositions are listed
in Table 2.
A considerable difference has been observed in the values of Ecevaluated by different formulations. This may be due
to different approximations that have been adopted while arriving at the final equation of the various formalisms. On the other hand, the simplified Kissinger equation is the
2.3 2.4 2.5 2.6 2.7 2.8
-10.4 -10.2 -10.0 -9.8 -9.6 -9.4 -9.2 -9.0 -8.8 -8.6
Se90Te10 Se87Te10Cd3 Se81Te10Cd9 Se75Te10Cd15
ln(
α
/
TP
2)
1000 / TP/ K-1
Fig. 6 Plots of lnð=T2
pÞ versus 1000=Tp for the compositions
[image:4.595.78.261.73.220.2]Se90xTe10Cdx(x¼0, 3, 9 and 15 at%).
Table 2 Calculated values ofEc,lnðkpÞand n for Se-Te-Cd composition
glasses.
Composition Ec(kCal/mol) lnðkp) n
eq. (4) eq. (5) eq. (6)
Se90Te10 16.71 17.54 15.28 17.58 1.26
Se87Te10Cd3 13.63 13.95 12.78 13.55 1.19
Se81Te10Cd9 11.55 12.36 10.27 11.75 1.24
Se75Te10Cd15 10.33 11.11 8.13 10.96 1.29
2.3 2.4 2.5 2.6 2.7 2.8
-4.6 -4.4 -4.2 -4.0 -3.8 -3.6 -3.4 -3.2 -3.0 -2.8 -2.6
Se90Te10 Se87Te10Cd3 Se81Te10Cd9 Se75Te10Cd15
ln(
α
/
TP
)
1000 / TP/ K-1
Fig. 7 Plots of lnð=TpÞ versus 1000=Tp for the compositions
Se90xTe10Cdx(x¼0, 3, 9 and 15 at%).
2.3 2.4 2.5 2.6 2.7 2.8
0.0 0.5 1.0 1.5
2.0 Se90Te10
Se87Te10Cd3 Se81Te10Cd9 Se75Te10Cd15
ln (d
χ
/
dt)
P
1000 / TP/ K-1
Fig. 8 Plots of lnðd=dtÞp versus 1000=Tp for the compositions
[image:4.595.340.517.75.222.2] [image:4.595.333.518.281.420.2] [image:4.595.46.293.296.379.2]conventional equation for evaluation of the activation energy for crystallization because of the convenience and accuracy of the measurements of heating rate. A similar considerable difference in the value of Ec calculated by different
procedures has also been observed in binary and ternary chalcogenide glass.27–29)
In general, there is a decrease in this parameter as the Cd concentration increases. These results indicate that the crystallization ability increases with increasing Cd content, which agrees with the conclusions reached by others.30–32)
This fact is also confirmed by theEc,Hr andkpvalues given
in Tables 1, 2. The observed decreases inEcwith increasing
Cd content indicated a decrease of the potential barrier height connected with the amorphous to crystalline phase transition or, in other words, a decrease of the preserving ability of the amorphous state in these alloys. Furthermore, the frequency factor, kp, (which measures the probability
of effective molecular collisions for the formation of the activated complexes in each case) decreases with increasing Cd concentration. This effect shows that the rate of crystallization is faster as the Cd content increases.
4. Conclusions
Kinetic studies made on various glassy alloys in the Se-Te-Cd system indicate that all compositions have single glass transitions and single stage crystallization. The values of Tg and Et were found to increase with increasing Cd
content. This is explained in terms of a change in chemical bonding at higher concentrations of Cd. The crystallization parameters namely Tc,Tp, kp, Hr and Ec were found to be
composition dependent. The calculated activation energies of crystallization and the frequency factor indicate that the glass forming ability (GFA) decreases with increasing Cd content.
In general, the results show that the crystallization mechanism in Se90xTe10Cdx occurs in one dimensional
growth from the surface to the inside according to the Avrami exponentn.
REFERENCES
1) G. B. Thomas, A. F. Frat and J. R. Bosnell: Philos. Mag.26(1972) 617–635.
2) M. A. Abdel-Rahim: J. Non-Crystall. Solids241(1998) 121–127. 3) S. A. Khan, M. Zulfequar and M. Husain: Solid Stat. Commun.123
(2002) 463–468.
4) P. Kumar and R. Thangaraj: J. Non-Crystall. Solids352(2006) 2288– 2291.
5) M. A. Abdel-Rahim, M. M. Hafiz and A. M. Shamekh: Physica B369 (2005) 143–154.
6) H. Yinnon and D. R. Uhlmann: J. Non-Cryst. Solids54(1983) 253– 275.
7) M. A. Abdel-Rahim, A. Y. Abdel-Latif, A. El-Korashy and G. A. Mohamed: J. Mater. Sci.30(1995) 5737–5742.
8) M. K. Rabinal, K. S. Sangunmi and E. S. R. Copal: J. Non-Cryst. Solids 188(1995) 98–106.
9) K. Tanka: Phys. Rev. B39(1989) 1270–1279.
10) A. Giridhar and S. Mahauevan: J. Non-Cryst. Solids151(1992) 245– 252.
11) M. A. Abdel-Rahim: J. Mater. Sci.27(1992) 1757–1761. 12) G. Lucovsky: J. Non-Crystal. Solids97(1987) 155–158. 13) H. E. Kissinger and J. Res: Nat. Bur. Stan.57(1956) 217–221. 14) N. Afify, M. A. Abdel-Rahim, A. S. Abdel-Halim and M. M. Hafiz:
J. Non-Cryst. Solids128(1991) 269–278.
15) S. Mahadevan and A. Giridhar: J. Non-Cryst. Solids197(1996) 219– 227.
16) A. A. Othman, H. H. Amer, M. A. Osman and A. Dahshan: J. Non-Cryst. Solids351(2005) 130–135.
17) A. Giridhar and S. Mahadevan: J. Non-Cryst. Solids197(1996) 228– 234.
18) S. Mahadevan and A. Giridhar: J. Mater. Sci.29(1994) 3837–3842. 19) S. A. Khan, M. Zulfequar and M. Husain: Solid Stat. Commun.123
(2002) 463–468.
20) A. Hurby: Czech. J. Phys. B22(1972) 1187–1193. 21) D. D. Thornburg: Mater. Res. Bull.9(1974) 1481–1485. 22) A. Y. Abdel-Latif: Physica B311(2002) 348–355.
23) J. A. Augis and Bennet: J. Thermal Anal.13(1978) 283–292. 24) Y. Q. Gao and W. Wang: J. Non-Cryst. Solids81(1986) 129–134. 25) D. W. Henderson: J. Non-Cryst. Solids30(1979) 301–315. 26) K. Matusita, T. Konatsu and R. Yokota: J. Mater. Sci.19(1984) 291–
296.
27) M. A. Abdel-Rahim, A. Y. Abdel-Latif, A. S. Soltan and M. Abuel-Oyoun: Physica B322(2002) 252–261.
28) K. Ramesh, S. Asokan, K. S. Sangunni and E. S. R. Gopal: J. Phys. Condens. Matter8(1996) 2755–2762.
29) M. A. Imran, D. Bhandari and N. S. Saxena: Mat. Sci. Eng. A292 (2000) 56–65.
30) R. A. Ligero, J. Vazuez, P. Villares and R. Jimenez-Garay: Mater. Lett. 8(1989) 6–11.
31) J. Vasguez, R. A. Ligero, P. Villares and R. Jimenez-Garay: J. Thermochim. Acta157(1990) 181–191.
32) M. A. Abdel-Rahim: Physica B239(1997) 238–244.
1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
ln(α)
ln(-ln(1-χ
))
Se75Te10Cd15 T = 408.2 K T = 404.9 K T = 399.2 K
[image:5.595.85.254.73.198.2]
