close

Вход

Забыли?

вход по аккаунту

?

j.jnoncrysol.2018.08.012

код для вставкиСкачать
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids
journal homepage: www.elsevier.com/locate/jnoncrysol
Prediction of glass forming regions in mixed-anion phosphate glasses
Y.B. Xiao, W.C. Wang, X.L. Yang, J.L. Liu, B. Zhou, Q.Y. Zhang
⁎
State Key Laboratory of Luminescent Materials and Devices and Institute of Optical Communication Materials, South China University of Technology, Guangzhou 510640,
PR China
A R T I C LE I N FO
A B S T R A C T
Keywords:
Fluoro-sulfo-phosphate glass
Glass-forming region
Thermodynamic method
Fluoro-sulfo-phosphate glasses with rich ligand situations and excellent thermal stability are promising host
materials for optically active species, injection moulding and low-temperature sealing, which have received a
significant amount of interest in recent years. Herein, the glass-forming regions of such inverted glass systems
are predicted via the thermodynamic method. Based on this, the actual glass formation areas are further determined by the experiments. The liquidus and eutectic points of the binary systems are calculated and the
deviations in composition between the calculated results and available phase diagrams are determined with a
small deviation (< 6 mol%). Meanwhile, the glass-forming regions are derived quantitatively, which are relatively in agreement with the experimental results. These results demonstrate that it is practical to predict the
formation regions of fluoro-sulfo-phosphate glass systems based on thermodynamic calculation, providing an
effective and predictive method for the development of new glass systems.
1. Introduction
Fluoro-phosphate glasses are promising candidate materials for
high-performance optics and laser devices owing to their low linear and
nonlinear refractive index, large transmission range, high rare-earth
(RE) ions solubility, and tailorable spectroscopic properties [1–4].
Previous research has already demonstrated that the introduction of
sulfate anions can significantly improve the thermal stability of phosphate glasses in terms of both rheology and chemical properties [5–9].
Additionally, sulfate group was found to be responsible for increasing
the radiative properties of RE-doped phosphate glass matrix [10], tailoring the local structure [11, 12] and enhancing the durability of
phosphate glass in corrosion medium [13]. Inspired by these ideas, a
new type of fluoro-sulfo-phosphate glass with rich ligand situations and
excellent thermal stability was developed through simultaneously incorporating the sulfate and fluoride into the phosphate glass, demonstrating an interesting route for tailoring the structural dynamics and
hosting the optically active cation species in the field of optical glass
and fiber lasers [14, 15]. Besides, the fluoro-sulfo-phosphate glasses are
potential materials in organic/inorganic conforming processes, injection moulding and low-temperature sealing [16, 17].
On the other hand, it is necessary to ensure that the glass composition locates at the stable glass-forming region in the phase diagram
when developing novel glass compositions with special properties. The
traditional way of determining glass-forming region was directly
⁎
performed by a large number of experiments, which costs a lot on
human and material resources. Our previous work has established a
simple, rapid, and predictable research method for the prediction of
glass-forming region using the thermodynamic theory [18–20]. The
deviations between the computational and experimental composition of
eutectics in binary silicate and borosilicate systems are < 7 mol% [20].
Furthermore, the optimized glass-forming regions in ternary systems
can be predicted by identifying the eutectic points of corresponding
binary systems.
In the present work, the glass-forming regions in ternary phosphate
(R2O–Al2O3–P2O5 (R = Li, Na, K), MO–Al2O3–P2O5 (M = Mg, Ca, Ba)),
fluoro-phosphate
(MgF2–AlF3–Ba(PO3)2,
ZnF2–AlF3–Ba(PO3)2,
ZnF2–AlF3–Zn(PO)3) systems were calculated via the thermodynamic
method and compared with the available experimental results [4, 21].
Moreover, the formation regions in fluoro-sulfo-phosphate
(AlF3–R2SO4–RPO3) glass systems were predicted quantitatively using
this method and determined experimentally.
2. The thermodynamic method
When two types of compounds are mixed without forming a new
compound, the free energy of the mixed solutions (GML) and solids
(GMS) can be described as follows [22]:
GML = RT (xA ln xA + xB ln xB )
Corresponding author.
E-mail address: qyzhang@scut.edu.cn (Q.Y. Zhang).
https://doi.org/10.1016/j.jnoncrysol.2018.08.012
Received 19 June 2018; Received in revised form 1 August 2018; Accepted 9 August 2018
0022-3093/ © 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Xiao, Y.B., Journal of Non-Crystalline Solids (2018), https://doi.org/10.1016/j.jnoncrysol.2018.08.012
(1)
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
where xA and xB are the molar fraction of components A and B, respectively; TA and TB are the melting temperatures of compounds A and
B, respectively; ΔHf,A and ΔHf,B are the fusion heats of compounds A and
B, respectively; R is the gas content.
In order to reduce the computation complexity, Eq. (1) is simplified
to a parabolic equation as follows [20]:
Table 1
The melting points and P2O5 content of the eutectic points in binary systems.
System
Subsystem
Melting point
(°C)
P2O5 content (mol
%)
Li2O–P2O5
Li2O·P2O5–2Li2O·P2O5
2Li2O·P2O5–3Li2O·P2O5
Na2O·P2O5–2Na2O·P2O5
2Na2O·P2O5–3Na2O·P2O5
K2O·P2O5–2K2O·P2O5
2K2O·P2O5–3K2O·P2O5
MgO·P2O5–2MgO·P2O5
2MgO·P2O5–3MgO·P2O5
P2O5–CaO·2P2O5
CaO·2P2O5–CaO·P2O5
CaO·P2O5–2CaO·P2O5
2CaO·P2O5–3CaO·P2O5
3CaO·P2O5–CaO
BaO·P2O5–2BaO·P2O5
2BaO·P2O5–3BaO·P2O5
3BaO·P2O5–10BaO·3P2O5
10BaO·3P2O5–BaO
Al2O3·3P2O5–Al2O3·P2O5
Al2O3·P2O5–3Al2O3·P2O5
3Al2O3·P2O5–Al2O3
Li2O–Li2O·Al2O3
Li2O·Al2O3–Li2O·5Al2O3
Li2O·5Al2O3–Al2O3
Na2O–Al2O3
K2O·Al2O3–Al2O3
MgO–Al2O3
CaO–Al2O3
BaO–3BaO·Al2O3
3BaO·Al2O3–BaO·Al2O3
BaO·Al2O3–BaO·6Al2O3
BaO·6Al2O3–Al2O3
600 [35]
870 [35]
490 [36]
943 [36]
610 [37]
1025 [37]
1150 [38]
1282 [38]
488 [39]
740 [39]
980 [39]
1302 [39]
1577 [39]
870 [40]
1415 [40]
1570 [40]
1480 [40]
1212 [41]
1881 [42]
1847 [42]
1055 [43]
1652 [43]
1915 [43]
1540 [44]
1450 [45]
1996 [46]
1371 [47]
1425 [48]
1480 [48]
1620 [48]
1875 [48]
43.9
30.9
43.5
30.1
43.4
29.4
47.5
27.6
91.0
63.0
48.8
30.8
21.9
47.4
30.0
23.7
21.4
67.5
32.5
23.5
–
–
–
–
–
–
–
–
–
–
–
Na2O–P2O5
K2O–P2O5
MgO–P2O5
CaO–P2O5
BaO–P2O5
Al2O3–P2O5
Li2O–Al2O3
Na2O–Al2O3
K2O–Al2O3
MgO–Al2O3
CaO–Al2O3
BaO–Al2O3
[35]
[35]
[36]
[36]
[37]
[37]
[38]
[38]
[39]
[39]
[39]
[39]
[39]
[40]
[40]
[40]
[40]
[41]
[42]
[42]
GML = 2.3x (x − 1) RT − 0.1181RT
When the liquid phase and solid phase achieve an equilibrium,
namely, GML = GMS. The temperatures of both phases are equal and the
function T(x) can be obtained by simultaneously solving Eqs. (2) and
(5) in form of [20]:
2.3RxB2
+
(
∆Hf , A
TA
−
∆Hf , B
TB
)
− 2.3R xB − 0.1181R −
∆Hf , A
(6)
TA
Then the eutectic temperature TE and composition x of the binary
system can be calculated by solving the minimum of function T(x).
Generally, the thermodynamic parameters of the common compounds are available from related manuals. While the fusion heats of
the certain substances and congruently melting compounds might be
obtained from the phase diagrams. In this case, based on the phase
diagram and thermodynamic theory, the fusion heat can be estimated
using the freezing-point depression method by the following equation
[20]:
∆Hm =
(2)
T ⎞
∆Gf , A = ∆Hf , A ⎛1 −
TA ⎠
⎝
(3)
T⎞
∆Gf , B = ∆Hf , B ⎛1 −
TB ⎠
⎝
(4)
R (Tm )2
xB (xB < < 1)
∆Tm
(7)
where Tm is the melting point of component A and ΔHm is the molar
fusion heat of component A; R is the gas constant. Additionally, if the
melting point and fusion heat of some compounds cannot be evaluated
or queried, they can be obtained by measuring their thermodynamic
parameters by differential scanning calorimetry (DSC) in experiment.
Based on the aforementioned thermodynamic method, the glass
forming regions in ternary phosphate (R2O–Al2O3–P2O5 and
MO–Al2O3–P2O5)
and
fluoro-phosphate
(MgF2–AlF3–Ba(PO3)2,
ZnF2–AlF3–Ba(PO3)2 and ZnF2–AlF3–Zn(PO3)2) systems were calculated
and compared with the experimental results reported [4, 21]. Table 1
summarizes the melting temperature and the glass former (i.e. P2O5)
content of the eutectic points in binary systems. The binary subsystems
⎟
⎜
(∆Hf , A − ∆Hf , B ) xB − ∆Hf , A
T=
S
GM
= −(xA ∆Gf , A + xB ∆Gf . B )
⎜
(5)
⎟
Table 2
The thermodynamic parameters, the calculated and experimental eutectic compositions of the binary subsystems in phosphate glass systems.
System
A
Li2O–Al2O3–P2O5
Na2O–Al2O3–P2O5
K2O–Al2O3–P2O5
MgO–Al2O3–P2O5
CaO–Al2O3–P2O5
BaO–Al2O3–P2O5
a
b
c
d
2Li2O·P2O5
Al2O3·3P2O5
e(2Li2O·P2O5–Li2O·P2O5)c
2Na2O·P2O5
Al2O3·3P2O5
e(2Na2O·P2O5–Na2O·P2O5)
2K2O·P2O5
Al2O3·3P2O5
e(2K2O·P2O5–K2O·P2O5)
2MgO·P2O5
Al2O3·3P2O5
e(2MgO·P2O5–MgO·P2O5)
CaO·P2O5
Al2O3·3P2O5
e(CaO·P2O5–CaO·2P2O5)
2BaO·P2O5
Al2O3·3P2O5
e(2BaO·P2O5–BaO·P2O5)
B
Li2O·P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
Na2O·P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
K2O·P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
MgO·P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
CaO·2P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
BaO·P2O5
Al2O3·P2O5
e(Al2O3·3P2O5–Al2O3·P2O5)
Melting point Tm (K)
Fusion heat (J/mol)
Eutectic composition xB (mol%)
TA
ΔHf,A
ΔHf,B
Calculated
Measured
21,819 [49]
48307a
27,349 b
27,040 [49]
48307a
27349b
55,491 [49]
48307a
27349b
80,577 [49]
48307a
27349b
30662a
48307a
27349b
57,269 [49]
48307a
27349b
44.4
31.4
15.5
44.5
31.4
10.0
43.8
31.4
18.5
43.3
31.4
61.0
62.6
31.4
29.5
48.4
31.4
37.5
43.9
35.2
–
43.5
35.2
–
43.4
35.2
–
47.5
35.2
–
63.0
35.2
–
47.4
35.2
–
TB
1158 [35]
1762 [41]
873 [35]
1263 [36]
1762 [41]
813 [36]
1377 [37]
1762 [41]
883 [37]
1655 [38]
1762 [41]
1423 [38]
1263 [39]
1762 [41]
1013 [39]
1703 [40]
1762 [41]
1123 [40]
938 [35]
2273 [41]
1485 [41]
900 [36]
2273 [41]
1485 [41]
1096 [37]
2273 [41]
1485 [41]
1438 [38]
2273 [41]
1485 [41]
1073 [39]
2273 [41]
1485 [41]
1143 [40]
2273 [41]
1485 [41]
a
27118
20176a
23594b
21190a
20176a
25051b
20147a
20176a
20148b
55295a
20176a
70464b
57,166 [49]
20176a
37125b
54123a
20176a
41266b
Data calculated based on the freezing-point depression method.
Data evaluated by the summation of fusion heats of two corresponding compounds.
e(2Li2O·P2O5–Li2O·P2O5) means the deepest eutectics of binary system 2Li2O·P2O5–Li2O·P2O5, and so on.
Tg is referenced from [4] and TL is the eutectic temperature.
2
[35]
[41]
[36]
[41]
[37]
[41]
[38]
[41]
[39]
[41]
[40]
[41]
Tg/TLd
–
0.63
–
0.49
0.63
–
0.60
0.63
–
0.45
0.63
–
0.68
0.63
–
0.54
0.63
–
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
Table 3
The thermodynamic parameters, the calculated and experimental eutectic compositions of binary subsystems in fluoro-phosphate glass systems.
System
A
MgF2–AlF3–Ba(PO3)2
ZnF2–AlF3–Ba(PO3)2
ZnF2–AlF3–Zn(PO3)2
a
b
c
d
Ba(PO3)2
Ba(PO3)2
e(Ba(PO3)2–MgF2)
Ba(PO3)2
Ba(PO3)2
e(Ba(PO3)2–ZnF2)
Zn(PO3)2
Zn(PO3)2
e(Zn(PO3)2–ZnF2)
B
MgF2
AlF3
e(Ba(PO3)2–AlF3)
ZnF2
AlF3
e(Ba(PO3)2–AlF3)
ZnF2
AlF3
e(Zn(PO3)2–AlF3)
Melting point Tm (K)
Fusion heats (J/mol)
TA
ΔHf,A
1153 [49]
1153 [49]
1122c
1153 [49]
1153 [49]
1013c
1136 [53]
1136 [53]
951c
TB
1533 [50]
1271 [51]
1023c
1220 [52]
1271 [51]
1023c
1220 [52]
1271 [51]
962c
57,269 [49]
57,269 [49]
56103b
57,269 [49]
57,269 [49]
41197b
29396a
29396a
28897b
Eutectic composition xB (mol%)
ΔHf,B
a
51844
27551a
41518b
28311a
27551a
41518b
28311a
27551a
28584b
Calculated
Measured
21.5
53.0
65.5
55.5
53.0
49.0
46.0
44.0
49.5
–
–
–
–
–
–
–
–
–
Tg/TLd
0.54
0.61
–
0.62
0.61
–
0.59
0.58
–
Data calculated based on the freezing-point depression method.
Data evaluated by the summation of fusion heats of two corresponding compounds.
Data calculated by the thermodynamic equation.
Tg is referenced from [4] and TL is the eutectic temperature.
Fig. 1. The calculated and experimental liquidus of binary phosphate systems.
compositions in binary phosphate systems (e.g. R2O–P2O5, MO–P2O5
and Al2O3–P2O5) and binary “quasi-compound” systems (see Fig. 2).
Fig. 3 compares the calculated and experimental glass forming regions
in ternary fluoro-phosphate glass systems.
The calculated liquidus in phosphate glass systems coincide with the
experimental values (see Fig. 1). The composition deviations between
the calculated and experimental eutectics are generally < 5 mol%. The
deviations between theoretical calculation and experimental results
with higher P2O5 content and lower melting point are selected for a
quantitative calculation. Tables 2 and Table 3 list the thermodynamic
parameters, the calculated and experimental eutectic compositions of
the binary subsystems in phosphate and fluoro-phosphate glasses.
Based on these fundamental data, the theoretical computing and actual
testing liquidus of the related subsystems are compared and shown in
Fig. 1. The glass-forming regions of ternary phosphate systems are
quantitatively determined according to the computational eutectic
3
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
Fig. 2. The calculated (determined based on the data in Table 2) and experimental glass-forming regions in ternary phosphate glass systems (mol%).
might originate from several factors. Firstly, the calculation of liquids
and eutectics are based on the thermodynamic theory, however the
experimental results are depending on both thermodynamic and kinetic
factors. Secondly, glass melt is not regular solution and the experimentally determined activity α should have been used instead of the
concentration x in Eqs. If so the value x cannot be predicted and additional experiments must be performed to determine the value of α.
Thirdly, partial basic data are from the phase diagram database and
literatures while the others are approximately calculated, which may
cause errors. Last but not at least, the experimental liquids queried form
the references are influenced by the experiment conditions, purity and
volatility of raw materials etc. Besides, the temperatures deviations are
much larger mainly because the equations used to calculate the eutectic
point parameters are directly derived from thermodynamic theory. Yet
glass melt is composed of mixed chemical bonds and vertex-angle
connections with large viscosity, which easily leads to a supercooling or
superheating phenomenon when it is cooled down or heated. The liquidus is significantly influenced by both kinetic and thermodynamic
factors. Additionally, the good agreement between theoretical and experimental results illustrated in Fig. 1 fairly proved the accuracy of the
computational method and reasonableness of fusion heats, laying the
foundation for further calculating the eutectic parameters in binary
“quasi-compound” systems.
Based on the calculation of the eutectic points and liquids of binary
systems, the glass forming regions in ternary phosphate glass systems
are derived as follows. According to Zachariasen's network hypothesis,
P2O5 is considered as a random network glass former [23]. In order to
Fig. 3. The calculated (determined based on the data in Table 3) and experimental glass-forming regions in ternary fluoro-phosphate glass systems (mol%).
tailor the physicochemical properties such as chemical durability and
transition temperature, the intermediate (Al2O3) and modifiers (R2O/
MO) are introduced into phosphate glass [24–26]. It is widely accepted
that the glass-forming regions are most likely to be situated near the
eutectic areas with higher glass former content [18, 20, 27–29]. As illustrated in Table 1, the melting temperatures of the eutectic points in
the binary systems containing P2O5 (R2O–P2O5, MO–P2O5, and
Al2O3–P2O5) decrease with increasing P2O5 content. Our previous work
has demonstrated that the glass-forming ability is proportional to the
viscosity at melting point [19]. It is therefore reasonable to infer that
the viscosity increases with increasing P2O5 content, which leads to a
stronger tendency of glass transition. In contrast, the melting points of
the eutectics in the binary systems (R2O–Al2O3 and MO–Al2O3) without
glass formers (e.g. P2O5) are much higher, resulting in a poor glassforming ability. Consequently, the glass formation regions in ternary
phosphate systems should be located in the related ternary component
diagrams that have higher P2O5 content. Additionally, the network
former content of the eutectics in binary systems R2O/MO–P2O5 are
comparable with that in Al2O3–P2O5 system while the melting points of
the eutectics in the R2O/MO–P2O5 systems are much lower. Typically,
the viscosity decreases sharply with increasing the temperature of the
4
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
Table 4
The thermodynamic parameters, the calculated and experimental eutectic compositions of binary subsystems in fluoro-sulfo-phosphate glass systems.
System
Li2SO4–AlF3–LiPO3
Na2SO4–AlF3–NaPO3
K2SO4–AlF3–KPO3
a
b
c
Melting point Tm (K)
Fusion heat (J/mol)
Eutectic composition xB (mol%)
A
B
TA
TB
ΔHf,A
ΔHf,B
Calculated
Measured
LiPO3
LiPO3
LiPO3
AlF3
NaPO3
NaPO3
NaPO3
AlF3
KPO3
KPO3
KPO3
AlF3
Li2SO4
AlF3
e(LiPO3–Li2SO4)
e(LiPO3–Li2SO4)
Na2SO4
AlF3
e(NaPO3–Na2SO4)
e(NaPO3–Na2SO4)
K2SO4
AlF3
e(KPO3–K2SO4)
e(KPO3–K2SO4)
907 [55]
907 [55]
907 [55]
1271 [51]
900 [36]
900 [36]
900 [36]
1271 [51]
1071 [57]
1071 [57]
1071 [57]
1271 [51]
1133 [55]
1271 [51]
781 [55]
781 [55]
1158 [36]
1271 [51]
855 [36]
855 [36]
1342 [57]
1271 [51]
987 [57]
987 [57]
21,819 [49]
21,819 [49]
21,819 [49]
27551a
27,040 [49]
27,040 [49]
27,040 [49]
27551a
20,299 [49]
20,299 [49]
20,299 [49]
27551a
13,807 [56]
27551a
18053b
18053b
23,012 [56]
27551a
25570b
25570b
36,819 [57]
27551a
24759b
24759b
47.0
26.0
62.0
87.5
36.5
26.5
55.0
78.0
27.0
36.0
51.5
67.0
52.5 [55]
–
–
–
31.5 [36]
–
–
–
21.5 [57]
–
–
–
Tg/TLc
0.54–0.67
0.49–0.64
0.57–0.68
Data calculated based on the freezing-point depression method.
Data evaluated by the summation of fusion heats of two corresponding compounds.
Tg/TL is measured by differential scanning calorimetry (DSC, 10 K/min).
subsystems with the lowest eutectic temperature in R2O/MO–P2O5 and
Al2O3–P2O5 systems are selected to calculate the liquidus and eutectic
parameters. The eutectic points (e.g. e(2Li2O·P2O5–Li2O·P2O5) and e
(Al2O3·3P2O5–Al2O3·P2O5)) are fixed in the phase diagram, which
means that their compositions are definite. And the eutectic points can
be regarded as “quasi-compounds” to form a new binary system [18].
The eutectics of the “quasi-compound” systems (noted as e) are computed and the fusion heats of “quasi-compound” can be approximately
evaluated by the summation of fusion heats of two corresponding
components [20]. Consequently, the triangular range consisting of e, e
(2Li2O·P2O5–Li2O·P2O5), and pure component P2O5 becomes the optimal glass-forming region in ternary phosphate systems
Li2O–Al2O3–P2O5 (Fig. 2a) and so on.
In ternary phosphate systems, the computational glass-forming regions are comparable to the experimental ones and both of them are
located in the areas that are near the binary systems P2O5–R2O/MO
with higher glass formers (see Fig. 2). The glass-forming region in the
phosphate glass system is unique because no phase separation occurs in
the region with high P2O5 content and a flat region with only slight
temperature variations. Therefore, the glass-forming regions extend up
to approximately 50 mol% P2O5, which are in agreement with the
thermodynamically calculated results. It is widely accepted that the
phosphate glasses are composed of PO4 tetrahedrons with different
bridging oxygens. There might be four kinds of PO4 tetrahedrons in
phosphate glasses dependent on the ratio of modifiers to P2O5. These
structure species can be noted as Qi=0, 1, 2, 3 groups where i donates the
number of bridging oxygen per phosphate tetrahedron [30]. The addition of network modifiers promotes the formation of non-bridging
oxygen species by depolymerizing the phosphate network. The
branching units (Q3) that contribute a lot to the network structure react
with the modifiers to form middle units (Q2). As the content of R2O and
MO increases up to 50 mol%, the ideal structure predominantly consists
of long chains based on Q2 units [31]. Both network structure and long
chain structure provide a high viscosity that tends to form glass under
fast cooling conditions. However, when the content of modifiers are
higher than 50 mol%, the end phosphate units (Q1) and isolated phosphate units (Q0) might form at the expense of middle units (Q2) [31].
This kind of melt with lower viscosity provides a higher tendency of
crystallization. The addition of intermediate (Al2O3) reconnects the
broken network, thereby the glass-forming region towards the Al2O3
component. In ternary phosphate glass systems, the mixture of two
alkali species results in a non-linear variation of several properties, such
as glass transition temperature, internal friction, and volumetric relaxation, which is regarded as “modifier interaction” and “mixed-alkali
effect” [32–34]. The presence of mixed modifiers results in asymmetry
Fig. 4. The calculated and experimental liquidus of binary phosphate-sulfate
systems.
glass melt. When the high-temperature glass melt cools down, it may
cause the phase segregation, making it difficult to form stable and
homogeneous glass. Thus, the calculated glass forming regions in
ternary phosphate systems should be located in the regions near the
R2O/MO–P2O5 systems with higher P2O5 content. Therefore, the
5
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
Fig. 5. The calculated (determined based on the data in Table 4) and experimental glass-forming regions in ternary fluoro-sulfo-phosphate glass systems
(mol%) and two photos of the glass samples.
Fig. 7. DSC patterns of several samples within the experimental glass-forming
region of AlF3-K2SO4-KPO3 system.
3. Experiments
The experimental glass-forming regions in fluoro-sulfo-phosphate
systems have been determined according to the previously calculated
results in our group. High purity raw materials of LiPO3 (99%,
Macklin), (NaPO3)3 (99%, J&K Chemical), KPO3 (≥ 99%, Macklin),
Li2SO4 (99.9%, Aladdin), Na2SO4 (99.99%, Aladdin), K2SO4 (99.99%,
Aladdin), AlF3 (99.9%, Aladdin) were used for preparing
AlF3–R2SO4–RPO3 glasses. The nominal compositions of 20 g batches
were melted in Al2O3 crucibles for 40 min at 1000–1200 °C and the
melts were cast into preheated graphite moulds and annealed for 2 h at
250–360 °C. Then the samples were cooled down to room temperature
at a rate of about 8–10 K/h. All the samples were detected visually to
determine whether forming transparent amorphous glasses. Several
transparent samples within the experimental glass-forming regions
were analyzed by differential scanning calorimetry (DSC; STA449C
Jupiter, Netzsch, German) and X-ray diffraction (XRD; X'Pert PRO X,
PANalytical, Holland). The mass loss of metaphosphate, fluoride and
sulfate were measured separately under the melting conditions by loss
on ignition, which were < 1 wt%, 10 wt% and 10 wt%, respectively.
Besides, X-ray fluorescence (XRF; Zetium, PANalytical, Holland) analysis reflected that the fluoride and sulfate evaporated < 5 wt%
and < 10 wt%, respectively. Additional tiny amounts of Al2O3 (< 1 wt
%) dissolved from the crucible, which has no obviously effect on the
experimental glass-forming regions. The actual compositions of the
glass samples are closely related to the content of the unstable compounds and experiment conditions (e.g. melting temperature and time).
All the data on compositions in our work were stated by nominal
concentrations.
Fig. 6. XRD patterns of several samples within the experimental glass-forming
region of AlF3-K2SO4-KPO3 system.
field that might favour the glass formation. The mixed-alkali effect is
not considered when the glass forming regions are predicted.
In fluoro-phosphate glass systems, the formation regions are calculated and predicted and the calculated glass forming regions are also
located nearby the experimental ranges (see Fig. 3), demonstrating the
feasibility of the glass-forming prediction in these systems through
thermodynamic method. In addition, the regions of the glass formation
in fluoro-phosphate glass systems are fairly large, which mean that the
glass compositions can be adjusted in broad ranges for obtaining the
tunable chemical and physical properties.
6
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
4. Results and discussion
experimental results. These results have illustrated that the thermodynamic method is useful and practical for preliminary forecasting and
investigating the glass-forming regions of the new fluoro-sulfo-phosphate glass systems, which provides a scientific reference for further
exploitation on novel glass.
Based on the calculation results of the glass forming regions in
ternary phosphate and fluoro-phosphate systems, the glass-forming
regions in fluoro-sulfo-phosphate systems (AlF3–R2SO4–RPO3) have
been predicted quantitatively and determined experimentally. Table 4
lists the thermodynamic parameters, the calculated and experimental
eutectic compositions of the binary subsystems in ternary fluoro-sulfophosphate glass systems. The calculated and experimental liquidus
values of the corresponding R2SO4–RPO3 binary systems are presented
in Fig. 4, which presents that the deviations in the compositions of the
eutectic points are < 6.0 mol%. The experimental and calculated glass
forming regions are illustrated in Fig. 5 and partial of the experimental
data (marked as colored circles) in Fig. 5(b) is from Ref. [14]. All the
glass samples within the experimental glass-forming regions are transparent and colorless. As shown in Fig. 5 (d) and (e), the fluoro-sulfophosphate glass can be easily prepared with large size and high transparency. The fluorides and sulfates might crystallize into minority
species [54]. However, the X-ray differential patterns are dominated by
one broad and amorphous hump that illustrates the glassy state (see
Fig. 6). The calculated glass-forming regions in AlF3–R2SO4–RPO3 systems are similarly consistent with the experimental results reported,
demonstrating that the thermodynamic method is of significance for the
investigation of the glass forming regions in fluoro-sulfo-phosphate
glasses.
Sulfate groups are found to be responsible for enhancing the thermal
stability and glass-forming ability of the glass matrix. It is easier to
prepare transparent glass samples as the content of sulfate increases in
the glass-forming regions of AlF3–R2SO4–RPO3. As presented in Fig. 7,
the thermal stability parameter △T (△T = Tx − Tg) increases with
increasing sulfate content. Tx is the initial crystallization temperature
and Tg is the glass transition temperature. It is known to all that the
properties of the materials are depended on the compositions and
structures. The fluoro-sulfo-phosphate glasses contain multiple anions,
which leads to the increase of ionic part of the polar covalent bonds and
depolymerization of network structures and the decrease of viscosity of
the systems. Therefore, the glass transition in these systems is predominately dependent on the interaction of ion groups. As is known to
all, both sulfate anions and phosphate groups with non-bridge oxygens
are in the tetrahedron structures and the bond length of SeO and PeO
are 149 and 152 pm, respectively. However, SO42–has only two negative charges while phosphate group has three with a P]O double bond.
As a result, the field around sulfate anion is weaker and the degree of
bond localization is lower [14]. The coexisting multiple groups and ions
provide interfering asymmetric field, which reduces the crystallization
rate and promotes the glass transition tendency. However, the SO42−
groups have been introduced into the fluoro-phosphate glass by substituting RPO3 or AlF3 with R2SO4, which leads to an increase in the
nominal total molar fraction of alkali metal ions since the additional
alkali metal ions are introduced with SO42 – groups and the crystallization tendency increase with increasing the content of modifiers.
Therefore, certain amounts of sulfates can be incorporated into fluorophosphate glass matrix and the glass forming ability and thermal stability will be improved significantly, which is beneficial for fabricating
fluoro-sulfo-phosphate glasses.
Acknowledgements
This work was financially supported by the National Natural Science
Foundation of China (Grant Nos. U1601205 and 51472088) and
National Postdoctoral Program for Innovative Talents of China (No.
BX20180099).
References
[1] K. Linganna, C.S.D. Viswanath, R. Narro-Garcia, S. Ju, W.T. Han, C.K. Jayasankar,
V. Venkatramu, J. Lumin. 166 (2015) 328.
[2] G. Galleani, Y. Ledemi, E.S.D.L. Filho, S. Morency, G. Delaizir, S. Chenu,
J.R. Duclere, Y. Messaddeq, Opt. Mater. 64 (2017) 524.
[3] Y. Tian, T. Wei, F.Z. Chen, X.F. Jing, J.J. Zhang, S.Q. Xu, J. Quant. Spectrosc. Ra.
311 (2014) 133.
[4] D. Ehrt, Phys. Chem. Glasses 56 (2015) 217.
[5] N. Da, O. Grassmé, K.H. Nielsen, G. Peters, L. Wondraczek, J. Non-Cryst. Solids 357
(2011) 2202.
[6] N. Da, S. Krolikowski, K.H. Nielsen, J. Kaschta, L. Wondraczek, J. Am. Ceram. Soc.
93 (2010) 2171.
[7] P. Ebeling, D. Ehrt, M. Friedrich, Phosphorus Res. Bull. 10 (1999) 484–489.
[8] A. Thieme, D. Möncke, R. Limbach, S. Fuhrmann, E.I. Kamitsos, L. Wondraczek, J.
Non-Cryst. Solids 410 (2015) 142.
[9] D. Ehrt, T. Toepfer, Proc. SPIE 4102 (2000) 95.
[10] G.A. Kumar, A. Martinez, E. De La Rosa, J. Lumin. 99 (2002) 141.
[11] N. Da, M. Peng, S. Krolikowski, L. Wondraczek, Opt. Express 18 (2010) 2549.
[12] D. Möncke, S. Sirotkin, E. Stavrou, E.I. Kamitsos, L. Wondraczek, J. Chem. Phys.
141 (2014) 224509.
[13] S. Sirotkin, R. Meszaros, L. Wondraczek, Int. J. Appl. Glass Sci. 3 (2012) 44.
[14] Q.H. Le, T. Palenta, O. Benzine, K. Griebenow, R. Limbach, E.I. Kamitsos,
L. Wondraczek, J. Non-Cryst. Solids 477 (2017) 58.
[15] W.C. Wang, Q.H. Le, Q.Y. Zhang, L. Wondraczek, J. Mater. Chem. C 5 (2017) 7969.
[16] N. Da, L. Wondraczek, M. Schmidt, N. Granzow, P. St, J. Russell, J. Non-Cryst.
Solids 356 (2010) 1829.
[17] N. Da, A.A. Enany, N. Granzow, M.A. Schmidt, P. St, J. Russell, L. Wondraczek, J.
Non-Cryst. Solids 357 (2011) 1558.
[18] Z.H. Jiang, X.Y. Hu, X.S. Zhao, J. Non-Cryst. Solids 52 (1982) 235.
[19] Z.H. Jiang, Q.Y. Zhang, Prog. Mater. Sci. 61 (2014) 144.
[20] Z.H. Jiang, Q.Y. Zhang, Sci. China Mater. 58 (2015) 378.
[21] A. Kishioka, M. Hayashi, B. Chem. Soc. Jpn. 49 (1976) 3032.
[22] J.W. Gibbs, Yale Univ. Press, (1948).
[23] W.H. Zachariasen, J. Am. Chem. Soc. 54 (1932) 3841.
[24] E. Metwalli, R.K. Brow, J. Non-Cryst. Solids 289 (2001) 113.
[25] R.W. Mooney, Chem. Rev. 61 (1961) 433.
[26] R.K. Brow, J. Am. Ceram. Soc. 76 (1993) 913.
[27] K.H. Sun, J. Am. Ceram. Soc. 30 (2010) 277.
[28] H. Rawson, Proc. IV Intern. Congress on Glass, Impremenic Chaix, Paris, 1956.
[29] V.S. Minaev, Proc. Intern Conf. “Amorphous Semiconductors – 78”, AS ChSSR
Publishers, Pardubice- Prague, 1978.
[30] R.K. Brow, J. Non-Cryst. Solids 263 (2000) 1.
[31] R.K. Brow, R.J. Kirkpatrick, G.L. Turner, J. Non-Cryst. Solids 116 (1990) 39.
[32] B.P. Rodrigues, J. Deubener, L. Wondraczek, Front. Mater. Sci. 3 (2016) 25.
[33] K. Griebenow, E.I. Kamitsos, L. Wondraczek, J. Non-Cryst. Solids 468 (2017) 74.
[34] K. Griebenow, C.B. Bragatto, E.I. Kamitsos, L. Wondraczek, J. Non-Cryst. Solids 481
(2018) 447.
[35] J. Nakano, T. Yamada, S. Miyazawa, J. Am. Ceram. Soc. 62 (1979) 465.
[36] A.G. Bergman, V.A. Matrosova, Zh. Neorg Khim. 1669, Russ. J. Inorg. Chem. (Engl.
Transl.) 14 (1969) 875.
[37] G.W. Morey, J. Am. Ceram. Soc. 76 (1954) 4724.
[38] J. Berak, Rocz. Chem. 32 (1958) 17.
[39] E.R. Kreidler, F.A. Hummel, Inorg. Chem. 6 (1967) 884.
[40] R.A. McCauley, F.A. Hummel, Trans. Br. Ceram. Soc. 67 (1968) 619.
[41] P.E. Stone, E.P. Egan, J. Am. Ceram. Soc. 39 (1956) 89.
[42] I.V. Tananaev, E.V. Maksimchuk, Y.G. Bushuev, S.A. Shestov, Izv. Akad Nauk SSSR,
Neorg. Mater. 719, Inorg. Mater. (Engl. Transl.) 14 (1978) 562.
[43] L.P. Cook, E.R. Plante, Ceram. Trans. (Fabr. Prop. Lithium Ceram. 3) 27 (1992) 193.
[44] M. Aldén, Solid State Ionics 20 (1986) 17.
[45] J.S. Moya, E. Criado, S.D. Aza, J. Mater. Sci. 17 (1982) 2213.
[46] B. Hallstedt, J. Am. Ceram. Soc. 75 (1992) 1497.
[47] B. Hallstedl, J. Am. Ceram. Soc. 73 (1990) 15.
[48] W. Pies, A. Weiss, References for Vol. III/7, Springer Berlin Heidelberg, 1974.
[49] Z.H. Jiang, Y.H. Guo, Bull. Chin. Ceram. Soc. 1 (1983) 3.
[50] L.A. Olkhovaya, P.P. Fedorov, D.D. Ikrami, J. Therm. Anal. 15 (1979) 355.
[51] G. Fuseya, C. Sugihara, N. Nagao, J. Electrochem. Soc. Jpn. 18 (1950) 65.
[52] A. Dekozak, M. Samouel, J. Less Common Met. 40 (1975) 185.
5. Conclusions
In summary, the glass-forming regions in fluoro-sulfo-phosphate
glass systems have been quantitatively investigated through the thermodynamic method. The calculated values are in good agreement with
the experimental results for a series of glass systems, which confirm the
validity of this thermodynamic method. The deviations between the
calculated and measured eutectic compositions of the corresponding
binary subsystems are < 6 mol%. Moreover, the calculated glassforming regions of the ternary systems are comparable with the
7
Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx
Y.B. Xiao et al.
[56] D.L. Ye, Metallurgy Industry Press, Beijing (1981).
[57] A.G. Bergman, V.A. Matrosova, Zh. Neorg. Khim. 1354, Russ. J. Inorg. Chem. (Engl.
Transl.) 14 (1969) 710.
[53] M.T. Averbuch-Pouchot, C. Martin, M.A. Rakotomahanina-Rolaisoa, A. Durif, Bull.
Soc. Fr. Minéral. Cristallogr. 93 (1970) 282.
[54] S. Reibstein, N. Da, J.P. Simon, E. Spiecker, L. Wondraczek, Phys. Chem. Glasses B
53 (2012) 61.
[55] A.G. Bergman, M.L. Sholokhovich, Zh. Obshch. Khim. 23 (1953) 1075.
8
Документ
Категория
Без категории
Просмотров
0
Размер файла
2 259 Кб
Теги
jnoncrysol, 012, 2018
1/--страниц
Пожаловаться на содержимое документа