Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Raman and infrared spectroscopy study on structure and microstructure of glass–ceramic materials from SiO2–Al2O3–Na2O–K2O–CaO system modified by variable molar ratio of SiO2/Al2O3 Janusz Partyka a,⇑, Magdalena Les´niak b a b
AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Ceramics and Refractories, Poland AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Silicate Chemistry and Macromolecular Compounds, Poland
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Article is focused on effect of
SiO2/Al2O3 ratio on microstructure of glass–ceramic. Glass–ceramic materials from oxide system of SiO2–Al2O3–CaO–MgO–Na2 O–K2O. Glass–ceramic within SiO2/Al2O3 molar ratio in range from 3.23 to 11.7. SiO2/Al2O3 ratio has a decisive influence on formation of NBO and BO bonds. SiO2/Al2O3 ratio has an influence on formation of crystalline phase.
a r t i c l e
i n f o
Article history: Received 28 April 2015 Received in revised form 1 July 2015 Accepted 7 July 2015 Available online 8 July 2015 Keywords: Glass–ceramic Molar ratio Microstructure Structure Glass lattice
a b s t r a c t This paper is focused on the effect of the molar ratio of SiO2/Al2O3 on the microstructure and structure of the internal aluminium–silicon-oxide lattice of the glass–ceramic materials from the SiO2–Al2O3–Na2O– K2O–CaO system. In order to examine the real composition of the obtained samples, a chemical analysis was performed. Following the heat-treatment procedure, pseudowollastonite, anorthite and the vitreous phase were identified. In order to determine the microstructure, research using the scanning electron microscope (SEM) with EDS was done. For the inner structural study, X-ray diffraction (XRD), Raman spectroscopy as well as MIR and FIR spectroscopy were performed. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Glass–ceramic materials are prepared through a process of controlled crystallisation of melted raw base materials yielding a fine
⇑ Corresponding author at: AGH University of Science and Technology, Faculty of Material Science and Ceramics, Department of Ceramics and Refractories, Al. Mickiewicza 30/B6, 32-830 Cracow, Poland. E-mail address: [email protected] (J. Partyka). http://dx.doi.org/10.1016/j.saa.2015.07.045 1386-1425/Ó 2015 Elsevier B.V. All rights reserved.
crystalline structure in the glassy phase matrix. The obtained material is characterised by a fully dense glassy phase and a given percentage of crystallized materials [1,2]. The percentage of the crystalline phase may vary between 0.5% and 98%; most commonly, it ranges from 30% to 80%. The initial oxide composition and the thermal treatment condition determine the type and quantity of the crystalline phase produced, as well as the chemical composition of the glassy phase bonding the crystallite grains [1–4]. Glass–ceramic materials are characterised by a range of properties
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
that include a mechanical strength, chemical durability and resistance to thermal shock. Additionally, other specific features relevant to the application of the material are important, such surface smoothness, wetting, gloss/matt, thermal expansion coefficient, fracture toughness and dielectric constant, dielectric loss and optical properties. Currently glass–ceramic materials have a variety of applications. They are used to produce surface coatings on a range of everyday wares including ceramic tiles, tableware and sanitary ware, materials for thermal shock resistant applications, biomaterials, glass–ceramic construction materials, abrasives and cutting materials, materials used in electronics, optics. Glassy phase in the glass–ceramic materials, bonding crystallites, is important for their final properties [2–4]. These are three types of ions in the structure of silicate glasses: network forming, intermediate and modifying [5–7]. Ions (mainly Si4+) which form glass network are called glass-forming ions or network-forming ions because of their ability to built up three-dimensional network (the formation of silicon–oxygen bridges). Aluminium is the most important of an intermediate ions. The aluminium ion can be four or six co-ordinated with oxygen giving rise to tetrahedral AlO4 or octahedral AlO6 groups. For tetrahedral coordination, aluminium ions together with silica ions take part in the formation of glass structure, i.e. they play the role of structure-forming ions – the formation of silicon–aluminium–oxygen bridges. Aluminium ions have the opposite effect in octahedral coordination. In this case, they break the silicon-aluminium structure, i.e. they fulfil the function of structure modifiers, which break silicon–oxygen bridges. Hence, the coordination of aluminium ions is of the upmost importance for both performances of glass as well as glass–ceramic materials. The effect of introducing modifying ions (mainly K+, Na+) has been to produce a broken bridges (SiAO ). The alkaline earth ions such as Mg2+, Ca2+ also act a modifying oxides. These cations occupying interstitial positions in the structure of glass. In the case of the divalent cations (Ca2+, Mg2+) one cation will be present for each pair non-bridging oxygen ions [8]. The alkaline earth such as CaO is active flux in simple glass only above 1100 °C. This temperature may be lower than 900 °C in the presence of other alkali oxides. It is available and contributes desirable properties to glass–ceramic. If too much CaO is used a matte surface will result from the crystallization of anorthite. Wollastonite, pseudowollastonite and anorthite are reported as desired crystalline phase in glass–ceramic materials [9–13]. According to Kingery [14] anorthite is the most common crystal type, but wollastonite and mullite are also frequently observed. According to Taylor [15], Bull [16] and Partyka [17] the presence of crystals in the glass matrix is influenced by many factors, especially by oxides molar composition and heating treatments A mainly role is played by the molar ratio of SiO2/Al2O3. The present researches mainly focus on behaviour Ca2+ ions in raw glass–ceramics materials by variable molar ratio of the SiO2/Al2O3. 2. Experimental procedure This research focused on glass–crystalline materials designed for a firing temperature of 1220–1250 °C from a multicomponent system of SiO2–Al2O3–K2O–Na2O–CaO. The input composition was designed based on the multicomponent oxygen systems
83
SiO2–Al2O3–CaO; SiO2–Al2O3–Na2O; SiO2–Al2O3–K2O, assuming a variable ratio of SiO2 to Al2O3 and the constant percentage of other oxides. The planned SiO2/Al2O3 ratio is presented in Table 1. Silica powder (sourced from the SKSM Sobótka mine), aluminium oxide (supplied by Helmut Kreutz GmbH), sodium carbonate (Na2CO3), potassium carbonate (K2CO3) (supplied by Avantor Polska) and chalk (supplied by ZPSChiM PIOTROWICE Sp. z o.o.) were used as raw materials for introducing individual oxides. The preparation of samples to be investigated is started by weighing all ingredients in accordance with the formula. Batches of raw ingredients were ground in a planetary ball mill for 30 min until 0.1% was retained on a 0.063 mm sieve. Dried batches were placed in porcelain crucibles with a capacity of approx. 90 cm3 and fired together in an electric laboratory furnace at 1230 °C for 14 h. The glass–crystalline materials obtained in this way were used for the preparation of samples for further research. For the purpose of SEM and Raman microscopy, cuboid samples (4 mm 15 mm 15 mm) were cut, where one of the large surfaces was additionally polished. The size of the remaining material was reduced to less than 0.063 mm. These powders were used for the determination of chemical and phase compositions as well as for structural tests in mid-infrared (MIR) and far-infrared (FIR). To obtain information on the chemical composition of fired glass–crystalline materials, wavelength dispersive X-ray fluorescence (WDXRF) was used to identify and determine the concentration of chemical elements within a broad measuring range from ppm concentrations to 100% of the weight. The chemical composition was analysed using a WDXRF Axios mAX spectrometer supplied by PANalytical. The phase composition obtained from the materials was determined using X-ray diffraction with an X-ray diffractometer X’Pert Pro supplied by PANalytical. The range of diffractogram recording was 5–90° (2h) with a recording speed of 0.05° 2h/2 s. Additional information on the vitreous silicon–aluminium–oxygen structure was provided by spectroscopy within mid-infrared (MIR) and far infra-red (FIR) and Raman spectroscopy. The MIR and FIR tests were performed using a Fourier spectrometer (Bruker Optics-Vertex70V). The samples were prepared using the tablet method in KBr (MIR) and polyethylene (FIR). The absorption spectra were recorded in 128 scans and the resolution of 4 cm 1, whereas Raman spectra were obtained using a confocal Raman microscope manufactured by JOBIN YVON (model LabRam HR) using the excitation wavelength of 514 nm. The diffraction grating was 1800 lines/mm. Additionally, the microstructure of all glass–crystalline materials was examined. Observations were made on microsections using a scanning electron microscope SEM–FEI Nova 200 NanoSEM with an analyser of the element composition in micro-areas (EDS-EDAX).
3. Results 3.1. Chemical composition of glass–crystalline materials The analysis of chemical composition after firing the materials being investigated is presented in Table 2. Based on the data Table 2 Oxide composition of glass–crystalline materials after firing. Glass–ceramic
SiO2/Al2O3
K2O
Na2O
CaO
3.05 2.76 3.33 3.28 2.71 3.19
4.73 4.39 4.91 4.82 4.09 4.76
13.72 12.59 13.84 13.93 14.46 14.27
(wt%) Table 1 The assumed weight ratio of SiO2 to Al2O3 in experimental glass–crystalline materials. SiO2/Al2O3 ratio SiO2/Al2O3
A(Ca)
B(Ca)
C(Ca)
D(Ca)
E(Ca)
F(Ca)
11.7
8.07
6.06
4.77
3.88
3.23
A(Ca) B(Ca) C(Ca) D(Ca) E(Ca) F(Ca)
11.69 8.08 6.06 4.77 3.88 3.23
84
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
obtained, it was found that the value of SiO2 decreased in subsequent samples as assumed, while the percentage of A2O3 increased and the percentage of other oxides differed only slightly. For modifying oxides (K2O, Na2O and CaO), the standard deviation in individual samples did not exceed ±1%.
3.2. Phase composition of glass–crystalline materials The X-ray tests conducted on the obtained materials (Fig. 1) concluded that only the material marked as E(Ca) is a practically amorphous material, which is confirmed by no clear reflexes in the diffractogram and the occurrence of raised background. For other materials, apart from the amorphous phase (raised background within the range of 2h 5–30°), there is a range of reflexes, indicating the presence of crystalline phases. The performed analysis concluded that there was calcium silicate pseudowollastonite from the cyclosilicate group (Ca3[SiO3]3) (Fig. 2) in the A(Ca), B(Ca), C(Ca) and D(Ca) materials, while in the F(Ca) calcium aluminium silicate, there was anorthite from the tecto-aluminium silicate group (Ca[Al2Si2O8]) (Fig. 3). A change of the phase composition, disappearance of anorthite, transition through the amorphous material and the appearance of a new phase (pseudowollastonite) results from the change of the SiO2/Al2O2 ratio. A gradual decrease in the silica content with an increase in the percentage of aluminium oxide causes a shift in the composition towards areas of pseudowollastonite crystallisation.
Fig. 2. Diffractograms of glass–crystalline materials A(Ca), B(Ca), C(Ca), D(Ca) and pseudowollastonite.
3.3. Scanning electron microscopy (SEM)
Fig. 1. Diffractograms of fired glass–crystalline materials.
Scanning electron microscopy (SEM) allowed the determination of the microstructure of the tested samples (Figs. 4–9), mostly the size and shape of crystallites. An analysis of SEM images confirmed the results of the phase analysis; all tested materials are glass– crystalline materials, except for E(Ca). In the presented photographs (Figs. 4–9), apart from the amorphous phase, crystals of various sizes and shapes are visible. The performed analysis of chemical composition within the micro-areas (EDX) (Figs. 4–9) showed that, for the A(Ca)–D(Ca) materials, the visible crystals were mostly pure calcium silicate. For the F(E) material, it is calcium aluminium silicate. Considering the intensity ratios of individual peaks in EDX spectra and the X-ray tests, it can be concluded that the visible crystals are pseudowollastonite and anorthite in the A(Ca)–D(Ca) materials in the F(Ca) sample, respectively. As for the determination of the structure of the tested materials, it is of the utmost importance to determine the chemical composition of the remaining glass matrix. As can be seen in the EDX spectra (Figs. 4–7), for materials containing pseudowollastonite (materials A(Ca)–D(Ca)), the glass matrix is an aluminium– silicate phase with ions of alkaline metals (K+, Na+), completely deprived of Ca2+ ions. As far as the completely amorphous E(Ca) and the anorthite-containing F(Ca), the glass matrix is an aluminium–silicate phase containing both ions of alkaline metals and Ca2+ ions.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Fig. 3. Diffractogram of the glass–crystalline material F(Ca) and anorthite.
3.4. Spectroscopic tests An additional objective of the presented study is an attempt to define the structure of the vitreous phase in the obtained glass– crystalline materials, i.e. the phase binding all ingredients to form a solid material. This is of the upmost importance as both the type of crystalline phases and the structure of the vitreous matrix determine the final performance. Due to the considerable percentage of the amorphous phase, it is necessary to use methods which would
make it possible to define the so-called near- and medium-range orders. From this point of view, spectroscopic methods are the most appropriate ones, in particular for mid-infrared (MIR) and far-infrared (FIR) as well as Raman spectroscopy. Fig. 10 presents FIR spectra of the tested glass–crystalline materials. The lack of bands in the FIR spectrum of the E(Ca) material shows that it is clearly fully amorphous. However, in the spectra of the A(Ca) to D(Ca) and F(Ca) materials, bands can be seen, which clearly indicates the presence of crystalline phases. The A(Ca)–D(Ca) materials are characterised by the highest degree of crystallinity (the presence of an entire range of relatively intense bands). In this way, the far-infrared tests confirm the conclusions relating to the nature of the tested materials, drawn on the basis of XRD and SEM tests. Glass–ceramic materials designed for firing temperatures above 1200 °C are usually aluminium–silicate materials. Bands connected with vibrations of silico- and alumino-oxygen bonds appear within the range of 1400–400 cm 1 in MIR and Raman spectra of aluminosilicate materials. Fig. 11 presents MIR spectra of the tested materials together with reference spectra (Figs. 12 and 13) of the phases found on the basis of XRD tests, i.e. pseudowollastonite and anorthite. All of the presented spectra are characterised by the presence of three groups of bands to be located within the ranges 1350– 950 cm 1, 850–650 cm 1 and 500–400 cm 1. The high full width at half maximum of the bands confirms a high percentage of the amorphous phase of the tested materials. This mostly applies to samples with the lowest SiO2/Al2O3, i.e. E(Ca) and F(Ca). Based on the XRD research, it is known that the E(Ca) sample is fully amorphous, while pseudowollastonite is present in materials with the lowest content of Al2O3 (A(Ca)–D(Ca)) and anorthite can be found in the materials with the highest content of Al2O3. Any detailed interpretation of MIR spectra of amorphous and glass–crystalline materials is a very difficult due to the mentioned high full width at half maximum (overlapping bands) and the presence of bands characteristic of individual crystalline phases. To facilitate the interpretation of the spectra of such materials, the decomposition of MIR spectra for three selected materials was performed: the material containing pseudowollastonite – D(Ca) (Fig. 14), the completely amorphous material – E(Ca) (Fig. 15) and the material containing anorthite F(Ca) (Fig. 16). The interpretation of the MIR E(Ca) spectrum decomposition is the most unambiguous one due to the lack of crystalline phases. This is typical alumino-silicate glass, in which the structure is modified by the ions of sodium, potassium and calcium (EDX tests). Therefore, the band at 1180 cm 1 is associated with stretching
Analysis of EDS-A(Ca), point 3
Analysis of EDS-A(Ca), point 1
85
A(Ca)
Fig. 4. SEM image of the A(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
86
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Analysis of EDS-B(Ca), point 3
AnalysisB(Ca) of EDS-B(Ca), point 2
B(Ca)
Fig. 5. SEM image of the B(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
Analysis of EDS-C(Ca), point 1
Analysis of EDS-C(Ca), point 3
C(Ca)
Fig. 6. SEM image of the C(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
Analysis of EDS-D(Ca), point 2
Analysis of EDS-D(Ca), point 3
D(Ca)
Fig. 7. SEM image of the D(Ca) material together with the analysed chemical composition of the crystalline phase (point 2) and the vitreous phase (point 3).
vibrations of Si@O (defects) [18], the band at 1035 cm 1 with stretching vibrations of SiAO(Si,Al) – silico- and alumino-silicooxygen bridges (bridging bonds), bands at 953 and 903 cm 1 with
stretching vibrations of SiAO and AlAO – broken silico- and alumina-silico-oxygen bridges (terminal bonds), bands at 790 and 747 cm 1 with bending vibrations of SiAOASi and SiAOAAl
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
87
Analysis of EDS-E(Ca), point 3
Analysis of EDS-E(Ca), point 2
E(Ca)
Fig. 8. SEM image of the E(Ca) material together with the analysed chemical composition of the vitreous phase (points 2 and 3).
Analysis of EDS-F(Ca), point 2
Analysis of EDS-F(Ca), point 3
F(Ca)
Fig. 9. SEM image of the F(Ca) material together with the analysed chemical composition of the crystalline phase (point 2) and the vitreous phase (point 3).
[18–20], the group of bands at 695, 640 and 577 cm 1 with vibrations of 3, 4 and 6-membered silico- and alumino-silico-oxygen rings [21–24], respectively, and bands at 469 and 419 cm 1 with bending vibrations of OASiAO and OAAlAO [18–20]. The decomposition of the MIR spectrum for the D(Ca) material is also relatively clear, which results from the percentage of the crystalline phase in the tested material (XRD, FIR and SEM). Based on the XRD and SEM tests, it is known that pseudowollastonite is the crystalline phase present in the tested material; pseudowollastonite is a three-membered cyclosilicate [25,26]. In the MIR spectrum of each of the 3-membered cyclosilicate, there is a characteristic, very intense band within the range of 750 to 700 cm 1, associated with the presence of isolated 3-membered silico-oxygen rings [27]. As it can be easily noticed in the MIR spectra of the tested A(Ca)–D(Ca) materials (Fig. 11) and decomposition of the MIR spectrum of the D(Ca) material (Fig. 14), a band is present at approx. 716 cm 1, which confirms the presence of pseudowollastonite. In addition, by comparison with the analysed decomposition (Fig. 14) and the MIR spectrum of pure pseudowollastonite (Fig. 12), bands at 981 cm 1 (stretching vibrations of SiAO(Si)), 942 cm 1 (stretching vibrations of SiAO ) and 565 cm 1 (bending vibrations of SiAOASi) can be unambiguously assigned to pseudowollastonite [27]. The other bands are mostly associated with the remaining amorphous phase. It results from the EDX spectra (Fig. 8a and b)
that the remaining amorphous phase is an alumino-silicate phase, in which only cations of alkaline metals are the modifiers. Therefore, similarly to the E(Ca) material, the band at 1189 cm 1 should be assigned to stretching vibrations of Si@O [18] and the band at 1052 cm 1 to stretching vibrations of SiAO(Si,Al) [18,19]. It must be noted that the intensity of the latter band is also influenced by the presence of pseudowollastonite. The bands at 914 and 890 cm 1 are derived from stretching vibrations of SiAO and AlAO , while the two bands at 795 and 762 cm 1 are associated with bending vibrations of SiAOASi and SiAOAAl [18–20]. The band at 684 cm 1 is associated with the vibrations of 4 and/or 6-membered silico- and alumino-silico-oxygen rings present in the glass structure [28], whereas the bands at 469 and 430 cm 1 with bending vibrations of OASiAO and OAAlAO [18–20]. The identification of bands characteristic of anorthite in the spectrum of the F(Ca) material is much more difficult, the reason of which mostly results from the low anorthite content in the tested material. Anorthite is a framework alumino-silicate e.g. [29,30], and consequently the bands associated with vibrations of silico- and alumino-silico-oxygen bonds corresponds to the vibrations of such bonds present in the amorphous phase. As a result, anorthite does not have an intensive identification band in the MIR spectrum (Fig. 11) as in the case of pseudowollastonite. Based on the comparison of the decomposition of the MIR spectrum of the F(Ca)
88
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
amorphous phase, regardless of crystalline phases. This problem can be partially solved by using IR and Raman microspectroscopy. On the grounds of the size of the pseudowollastonite and anorthite crystals (see SEM images – Figs. 4–9) in the tested materials, IR microspectroscopy is useless due to the size of the surface area, from which the spectrum is collected (approx. 50 50 lm). This means that it is impossible to obtain separate MIR spectra for crystalline phases and the remaining glass matrix. Raman spectroscopy provides here many more possibilities, i.e. approx. 1 1 lm. Fig. 17 presents Raman spectra of D(Ca), E(Ca) and F(Ca) materials and reference spectra of crystalline phases found on the basis of XRD tests (pseudowollastonite and anorthite). For each sample, measurements were taken at two points to obtain, to the fullest extent possible, an independent spectrum of the crystalline and amorphous phases of the matrix. As can be seen in Fig. 17, for glass–crystalline materials (D(Ca) and F(Ca)), it was possible to perform point measurements of crystalline phases. Raman spectra of the measured crystalline phases correspond exactly to the model spectra of pseudowollastonite and anorthite. The tests unambiguously confirm previous conclusions drawn from XRD and MIR tests. They are yet another piece of evidence for the presence of pseudowollastonite and anorthite in these materials. As well, it was possible to obtain an amorphous phase spectrum for the F(Ca) sample. As can be seen in Fig. 17, the Raman spectrum of the amorphous phase for F(Ca) is virtually identical with the Raman spectrum of the amorphous phase obtained for the E(Ca) sample.
Fig. 10. FIR spectra of the tested glass–crystalline materials.
glaze (Fig. 13) with the MIR spectrum of anorthite (Fig. 13), it can only be assumed that anorthite would be associated with the band at 994 cm 1 (stretching vibrations of SiAO(Si,Al)) and the group of bands within the range of 630–540 cm 1 (bending vibrations of SiAOASi and SiAOAAl) [18–20]. Similarly to the E(Ca) material, the remaining amorphous phase is an alumino-silicate phase, in which alkaline metal ions and calcium ions are the modifiers of the structure. Therefore, the band at 1180 cm 1 should be assigned to stretching vibrations of Si@O [18] and the band at 1088 cm 1 to stretching vibrations of SiAO(Si,Al) [19,20]. The intensity of the latter band is also influenced by the presence of anorthite. The bands at 931 and 891 cm 1 results from stretching vibrations of SiAO and AlAO , the bands at 774 and 722 cm 1 from bending vibrations of SiAOASi and SiAOAAl, and the bands at 468 and 427 cm 1 from bending vibrations of OASiAO and OAAlAO [18–20]. By comparing and analysing the decomposition of spectra of selected materials, it can be concluded that the glass matrix in each case is an alumino-silicate phase with a similar structure. The only difference, to which attention must be paid, is a considerably higher integral intensity (the surface area under the curve) of the band associated with the silico- and alumino-silico-oxygen bridges (at 1052 cm 1) in the D(Ca) spectrum, compared to the analogous band in the spectrum for E(Ca) – 1035 cm 1, and F(Ca) – 1088 cm 1. This probably results from the mentioned differences in the type of structure modifiers (Na+, K+ and Ca2+). The presented analysis of the MIR spectra of glass–crystalline materials D(Ca) and F(Ca) is not fully accurate due to the averaging nature of MIR spectra, which prevents a direct determination of the structure of the
Fig. 11. MIR spectra of the tested glass–crystalline materials.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Fig. 12. MIR spectra pseudowollastonite.
of
the
D(Ca)
glass–crystalline
material
and
The most intense broad band for wavenumbers ranging from 1200 to 900 cm 1 is associated with asymmetrical vibrations of SiAO in silicon–oxygen tetrahedra ([SiO4]4 ). This range of wavenumbers depends on the number of non-bridging oxygen atoms (NBOs), which form a tetrahedron [31]. The data to be found in the literature [31–33] show that in the Raman spectra of silicate glass, bands associated with the Q4–Q0 groups (where Q donates the number of bridging bonds) appear at approx. 1150 cm 1 – Q4, 1050 cm 1 – Q3, 1000 – 950 cm 1 – Q2, 880 cm 1 – Q1 and 850 cm 1 – Q0, respectively. Therefore, E(Ca) and F(Ca) glass matrices are dominated by tetrahedra with one broken oxygen bridge (Q3), i.e. for a band at approx. 1040 cm 1. However, slight bending at approx. 1150 and 1000 cm 1 indicate the presence of the Q4 and Q2 groups, respectively. No bands within the range of 950 to 850 cm 1 unambiguously indicates the lack of Q1 and Q0 groups. The band at 580 cm 1 is associated with the vibrations of three-membered silico- and aluminosilico-oxygen rings. The band at approx. 485 cm 1 is mostly associated with bending vibrations of SiAOASi and SiAOAAl [31]. The intensity of this band is also influenced by the presence of 6-membered silico- and alumino-silico-xygen rings [38]. Rings featuring a variable number of members, ranging from 3 to 8, and a variable degree of deformation are an element of the lattice of silico- and alumino-silico-oxygen glasses [18,24]. The band at approx. 790 cm 1 is associated with the vibrations of aluminooxygen tetrahedra with three bridging bonds [34,35]. No band at approx. 710 cm 1 unambiguously indicates that there are no
89
Fig. 13. MIR spectra of the F(Ca) glass–crystalline material and anorthite.
Fig. 14. Decomposition of the MIR spectrum of the D(Ca) material.
alumino-oxygen octahedra [36,37], and thus all aluminium ions are arranged in a tetrahedral coordination, meaning that they take part in the formation of the glass structure. This is confirmed by the previous statement that the tested amorphous phases are silico-alumino-oxygen glasses.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
90
Fig. 15. Decomposition of the MIR spectrum of the E(Ca) material.
Fig. 17. Raman spectra of the D(Ca), E(Ca) and F(Ca) glass–crystalline materials and the model spectra of crystalline phases. Fig. 16. Decomposition of the MIR spectrum of the F(Ca) material.
4. Conclusions (1) Changing SiO2/Al2O3 ratio in the SiO2–Al2O3–K2O–Na2O–CaO system has a decisive influence on the type and amount of the crystalline phase in the glass–crystalline materials. At the highest SiO2/Al2O3 ratios, the tested materials are typical glass–crystalline materials, in which calcium silicatepseudowollastonite occurs, apart from the vitreous phase. (2) By lowering the SiO2/Al2O3 ratio gradually to 4.77, the quantity of the crystalline phase is decreased, and with SiO2/Al2O3 equal to 3.88, the material is completely amorphous (SEM, XRD). (3) With the SiO2/Al2O3 ratio equal to 3.23, the vitreous phase and calcium aluminium silicate – anorthite occurs in the material. (4) The results of spectroscopic research show that the change in the SiO2/Al2O3 ratio only slightly influences the chemical composition (the type of structure modifiers) and the internal structure of the vitreous phase-bonding crystallites to form a uniform material. This influence is exposed by the degree of polymerisation (the number of broken silicooxygen bridges).
(5) In the tested materials, independently of the SiO2/Al2O3 ratio, aluminium ions occur only in tetrahedral coordination, i.e. they form the amorphous phase structure.
Acknowledgements This research has been carried out thanks to financing under the framework of NCBiR (Polish National Research and Development Committee) programme No. N N508 477734 and PBS1/B5/17/2012.
References [1] W. Holand, G.H. Beall, Glass Ceramic Technology, second ed., Wiley, 2012, ISBN: 978-0-470-48787-7. [2] T. Manfredini, G.C. Pellacani, J.M. Rincon, Glass–Ceramic Materials: Fundamentals and Applications, Mucchi Editore, Modena, Italy, 1997, ISBN10: 887000287X. [3] D.R. Eppler, R.A. Eppler, Glazes and Glass Coatings, The American Ceramic Society, Westerville, Ohio, 2000. [4] R.K. Brow, Inorganic glasses and glass–ceramics, in: M.A. Stoneham, R.E. Loehman (Eds.), Characterization of Ceramics, Butterworth-Heinemann, 1993, pp. 103–118. [5] J.E. Shelby, Introduction to Glass Science and Technology, Athenaeum Press Ltd, UK, 1997.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91 [6] Li. Neng, W.Y. Ching, Structural, electronic and optical properties of a large random network model of amorphous SiO2 glass, J. Non-Cryst. Solids 383 (2014) 28–32. [7] J. Zhong-Hong, Z. Qin-Yuan, The structure of glass: a phase equilibrium diagram approach, Prog. Mater Sci. 61 (2014) 144–215. [8] Ch. Haiyan, H. Guosong, M. Hanfen, G. Fuxi, Structure and Raman spectra of glasses containing several glass-forming oxides and no glass-modifying oxide, J. Non-Cryst. Solids 80 (1986) 152–159. [9] K. Alptekin, F. Kara, Glass–ceramic glazes in the CaO–SiO2 system, Key Eng. Mater. 264–68 (2004) 1709–1712. [10] C.F. Binns, The development of the matt glaze, Trans. Am. Ceram. Soc. 5 (1903) 50–63. [11] A. Karamanov, M. Pelino, Induced crystallization porosity and properties of sintereds diopside and wollastonite glass-ceramics, J. Eur. Ceram. Soc. 28 (2008) 555–562. [12] I.A. Levitskii, N.V. Mazura, Opacified glazes produced by high-temperature firing for sanitary ceramicware, Glass Ceram. 62 (2002) 215–218. [13] E.F. Osborn, The system CaSiO3-diopside-anorthite, J. Am. Sci. 11 (1942) 751– 788. [14] W.D. Kingery, H.K. Browen, D.R. Uhlman, Introduction to Ceramics, second ed., John Wiley and Sons, New York, 1960. [15] J.R. Taylor, A.C. Bull, Ceramic Glaze Technology, Pergamon Press, Oxford, 1986. [16] R.T. Stull, Influence of variable silica and alumina on the porcelain glazes, Trans. Am. Ceram. Soc. XIV (1912) 62–72. [17] J. Partyka, M. Sitarz, M. Les´niak, K. Gasek, P. Jelen´, The effect of SiO2/Al2O3 ratio on the structure and microstructure of the glazes from SiO2–Al2O3–CaO–MgO– Na2O–K2O system, Spectrochim. Acta Part A 134 (2015) 621–630. [18] M. Sitarz, The structure of simple silicate glasses in the light of Middle Infrared spectroscopy studies, J. Non-Cryst. Solids 357 (2011) 1603–1608. [19] W. Mozgawa, M. Sitarz, M. Rokita, Spectroscopic study of different aluminosilicate structures, J. Mol. Struct. 511–512 (1999) 251–257. [20] W. Mozgawa, M. Sitarz, Vibrational spectra of aluminosilicate ring strictures, J. Mol. Struct. 614 (2002) 273–279. [21] M. Sitarz, W. Mozgawa, M. Handke, Vibrational spectra of complex ring silicate anions-method of recognition, J. Mol. Struct. 404 (1997) 193–197. [22] M. Handke, M. Sitarz, W. Mozgawa, Model of silicooxygen ring vibrations, J. Mol. Struct. 450 (1998) 229–238.
91
[23] M. Sitarz, M. Handke, W. Mozgawa, The calculations of silicooxygen rings vibrations, Spectrochim. Acta Part A 55 (1999) 2831–2837. [24] M. Sitarz, M. Handke, W. Mozgawa, Rings in the structure of silicate glasses, J. Mol. Struct. 511–512 (1999) 281–285. [25] J.W. Jeffery, L. Heller, Acta Cryst. 6 (1953) 807. [26] T. Yamanaka, H. Mori, Acta Cryst. 1981 (B37) (1981) 1010. [27] M. Sitarz, M. Handke, W. Mozgawa, E. Galuskin, I.O. Galuskina, The non-ring cations influence on silicooxygen rings vibrations, J. Mol. Struct. 555 (2000) 357–362. [28] M. Sitarz, M. Handke, W. Mozgawa, Identification of silicooxygen rings in SiO2 based on IR spectra, Spectrochim. Acta Part A 56 (2000) 1819–1823. [29] H.D. Megaw, C.J.E. Kempster, E.W. Radoslovich, The structure of anorthite, CaAl2Si2O8. II. Description and discussion, Acta Cryst. 15 (1962) 1017–1035. [30] D.W. Matson, S.K. Sharma, J.A. Philpotts, Raman spectra of some tectosilicates and of glasses along the orthoclase-anorthite and nepheline-anorthite joins, Am. Mineral. 71 (1986) 694–704. [31] S. Agathopoulos, D.U. Tulyaganov, J.M.G. Ventura, S. Kannan, A. Saranti, M.A. Karakassides, J.M.F. Ferreira, Structural analysis and devitrification of glasses based on the CaO–MgO–SiO2 system with B2O3, Na2O, CaF2 and P2O5 additives, J. Non-Cryst. Solids 352 (2006) 322–328. [32] O. Bjorn, J. Mysen, D. Frantz, Raman spectroscopy of silicate melts at magmatic temperatures, Chem. Geol. 96 (1992) 321–332. [33] M. Szumera, The structural role of manganese ions in soil active silicate– phosphate glasses, Spectrochim. Acta Part A 129 (2014) 601–608. [34] P.L. Higby, R.J. Ginther, I.D. Aggarwal, E.J. Friebele, Glass formation and thermal properties of low-silica calcium aluminosilicate glasses, J. Non-Cryst. Solids 126 (1990) 209–215. [35] P. MacMillan, B. Piriou, Raman spectroscopy of calcium aluminate glasses and crystals, J. Non-Cryst. Solids 55 (1983) 221–242. [36] C. Huang, E.C. Behrman, Structure and properties of calcium aluminosilicate glasses, J. Non-Cryst. Solids 128 (1991) 310–321. [37] K. Sharma, J.A. Philipotts, D.W. Matson, Ring distributions in alkali and alkaline earth aluminosilicate framework glasses-A Raman spectroscopic study, J. NonCryst. Solids 71 (1985) 403–410. [38] F.L. Galeener, R.A. Barrio, E. Martinez, R.J. Elliott, Vibrational decoupling of rings in amorphous solids, Phys. Rev. Lett. 53 (1984) 2429–2432.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Raman and infrared spectroscopy study on structure and microstructure of glass–ceramic materials from SiO2–Al2O3–Na2O–K2O–CaO system modified by variable molar ratio of SiO2/Al2O3 Janusz Partyka a,⇑, Magdalena Les´niak b a b
AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Ceramics and Refractories, Poland AGH University of Science and Technology, Faculty of Materials Science and Ceramics, Department of Silicate Chemistry and Macromolecular Compounds, Poland
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Article is focused on effect of
SiO2/Al2O3 ratio on microstructure of glass–ceramic. Glass–ceramic materials from oxide system of SiO2–Al2O3–CaO–MgO–Na2 O–K2O. Glass–ceramic within SiO2/Al2O3 molar ratio in range from 3.23 to 11.7. SiO2/Al2O3 ratio has a decisive influence on formation of NBO and BO bonds. SiO2/Al2O3 ratio has an influence on formation of crystalline phase.
a r t i c l e
i n f o
Article history: Received 28 April 2015 Received in revised form 1 July 2015 Accepted 7 July 2015 Available online 8 July 2015 Keywords: Glass–ceramic Molar ratio Microstructure Structure Glass lattice
a b s t r a c t This paper is focused on the effect of the molar ratio of SiO2/Al2O3 on the microstructure and structure of the internal aluminium–silicon-oxide lattice of the glass–ceramic materials from the SiO2–Al2O3–Na2O– K2O–CaO system. In order to examine the real composition of the obtained samples, a chemical analysis was performed. Following the heat-treatment procedure, pseudowollastonite, anorthite and the vitreous phase were identified. In order to determine the microstructure, research using the scanning electron microscope (SEM) with EDS was done. For the inner structural study, X-ray diffraction (XRD), Raman spectroscopy as well as MIR and FIR spectroscopy were performed. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Glass–ceramic materials are prepared through a process of controlled crystallisation of melted raw base materials yielding a fine
⇑ Corresponding author at: AGH University of Science and Technology, Faculty of Material Science and Ceramics, Department of Ceramics and Refractories, Al. Mickiewicza 30/B6, 32-830 Cracow, Poland. E-mail address: [email protected] (J. Partyka). http://dx.doi.org/10.1016/j.saa.2015.07.045 1386-1425/Ó 2015 Elsevier B.V. All rights reserved.
crystalline structure in the glassy phase matrix. The obtained material is characterised by a fully dense glassy phase and a given percentage of crystallized materials [1,2]. The percentage of the crystalline phase may vary between 0.5% and 98%; most commonly, it ranges from 30% to 80%. The initial oxide composition and the thermal treatment condition determine the type and quantity of the crystalline phase produced, as well as the chemical composition of the glassy phase bonding the crystallite grains [1–4]. Glass–ceramic materials are characterised by a range of properties
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
that include a mechanical strength, chemical durability and resistance to thermal shock. Additionally, other specific features relevant to the application of the material are important, such surface smoothness, wetting, gloss/matt, thermal expansion coefficient, fracture toughness and dielectric constant, dielectric loss and optical properties. Currently glass–ceramic materials have a variety of applications. They are used to produce surface coatings on a range of everyday wares including ceramic tiles, tableware and sanitary ware, materials for thermal shock resistant applications, biomaterials, glass–ceramic construction materials, abrasives and cutting materials, materials used in electronics, optics. Glassy phase in the glass–ceramic materials, bonding crystallites, is important for their final properties [2–4]. These are three types of ions in the structure of silicate glasses: network forming, intermediate and modifying [5–7]. Ions (mainly Si4+) which form glass network are called glass-forming ions or network-forming ions because of their ability to built up three-dimensional network (the formation of silicon–oxygen bridges). Aluminium is the most important of an intermediate ions. The aluminium ion can be four or six co-ordinated with oxygen giving rise to tetrahedral AlO4 or octahedral AlO6 groups. For tetrahedral coordination, aluminium ions together with silica ions take part in the formation of glass structure, i.e. they play the role of structure-forming ions – the formation of silicon–aluminium–oxygen bridges. Aluminium ions have the opposite effect in octahedral coordination. In this case, they break the silicon-aluminium structure, i.e. they fulfil the function of structure modifiers, which break silicon–oxygen bridges. Hence, the coordination of aluminium ions is of the upmost importance for both performances of glass as well as glass–ceramic materials. The effect of introducing modifying ions (mainly K+, Na+) has been to produce a broken bridges (SiAO ). The alkaline earth ions such as Mg2+, Ca2+ also act a modifying oxides. These cations occupying interstitial positions in the structure of glass. In the case of the divalent cations (Ca2+, Mg2+) one cation will be present for each pair non-bridging oxygen ions [8]. The alkaline earth such as CaO is active flux in simple glass only above 1100 °C. This temperature may be lower than 900 °C in the presence of other alkali oxides. It is available and contributes desirable properties to glass–ceramic. If too much CaO is used a matte surface will result from the crystallization of anorthite. Wollastonite, pseudowollastonite and anorthite are reported as desired crystalline phase in glass–ceramic materials [9–13]. According to Kingery [14] anorthite is the most common crystal type, but wollastonite and mullite are also frequently observed. According to Taylor [15], Bull [16] and Partyka [17] the presence of crystals in the glass matrix is influenced by many factors, especially by oxides molar composition and heating treatments A mainly role is played by the molar ratio of SiO2/Al2O3. The present researches mainly focus on behaviour Ca2+ ions in raw glass–ceramics materials by variable molar ratio of the SiO2/Al2O3. 2. Experimental procedure This research focused on glass–crystalline materials designed for a firing temperature of 1220–1250 °C from a multicomponent system of SiO2–Al2O3–K2O–Na2O–CaO. The input composition was designed based on the multicomponent oxygen systems
83
SiO2–Al2O3–CaO; SiO2–Al2O3–Na2O; SiO2–Al2O3–K2O, assuming a variable ratio of SiO2 to Al2O3 and the constant percentage of other oxides. The planned SiO2/Al2O3 ratio is presented in Table 1. Silica powder (sourced from the SKSM Sobótka mine), aluminium oxide (supplied by Helmut Kreutz GmbH), sodium carbonate (Na2CO3), potassium carbonate (K2CO3) (supplied by Avantor Polska) and chalk (supplied by ZPSChiM PIOTROWICE Sp. z o.o.) were used as raw materials for introducing individual oxides. The preparation of samples to be investigated is started by weighing all ingredients in accordance with the formula. Batches of raw ingredients were ground in a planetary ball mill for 30 min until 0.1% was retained on a 0.063 mm sieve. Dried batches were placed in porcelain crucibles with a capacity of approx. 90 cm3 and fired together in an electric laboratory furnace at 1230 °C for 14 h. The glass–crystalline materials obtained in this way were used for the preparation of samples for further research. For the purpose of SEM and Raman microscopy, cuboid samples (4 mm 15 mm 15 mm) were cut, where one of the large surfaces was additionally polished. The size of the remaining material was reduced to less than 0.063 mm. These powders were used for the determination of chemical and phase compositions as well as for structural tests in mid-infrared (MIR) and far-infrared (FIR). To obtain information on the chemical composition of fired glass–crystalline materials, wavelength dispersive X-ray fluorescence (WDXRF) was used to identify and determine the concentration of chemical elements within a broad measuring range from ppm concentrations to 100% of the weight. The chemical composition was analysed using a WDXRF Axios mAX spectrometer supplied by PANalytical. The phase composition obtained from the materials was determined using X-ray diffraction with an X-ray diffractometer X’Pert Pro supplied by PANalytical. The range of diffractogram recording was 5–90° (2h) with a recording speed of 0.05° 2h/2 s. Additional information on the vitreous silicon–aluminium–oxygen structure was provided by spectroscopy within mid-infrared (MIR) and far infra-red (FIR) and Raman spectroscopy. The MIR and FIR tests were performed using a Fourier spectrometer (Bruker Optics-Vertex70V). The samples were prepared using the tablet method in KBr (MIR) and polyethylene (FIR). The absorption spectra were recorded in 128 scans and the resolution of 4 cm 1, whereas Raman spectra were obtained using a confocal Raman microscope manufactured by JOBIN YVON (model LabRam HR) using the excitation wavelength of 514 nm. The diffraction grating was 1800 lines/mm. Additionally, the microstructure of all glass–crystalline materials was examined. Observations were made on microsections using a scanning electron microscope SEM–FEI Nova 200 NanoSEM with an analyser of the element composition in micro-areas (EDS-EDAX).
3. Results 3.1. Chemical composition of glass–crystalline materials The analysis of chemical composition after firing the materials being investigated is presented in Table 2. Based on the data Table 2 Oxide composition of glass–crystalline materials after firing. Glass–ceramic
SiO2/Al2O3
K2O
Na2O
CaO
3.05 2.76 3.33 3.28 2.71 3.19
4.73 4.39 4.91 4.82 4.09 4.76
13.72 12.59 13.84 13.93 14.46 14.27
(wt%) Table 1 The assumed weight ratio of SiO2 to Al2O3 in experimental glass–crystalline materials. SiO2/Al2O3 ratio SiO2/Al2O3
A(Ca)
B(Ca)
C(Ca)
D(Ca)
E(Ca)
F(Ca)
11.7
8.07
6.06
4.77
3.88
3.23
A(Ca) B(Ca) C(Ca) D(Ca) E(Ca) F(Ca)
11.69 8.08 6.06 4.77 3.88 3.23
84
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
obtained, it was found that the value of SiO2 decreased in subsequent samples as assumed, while the percentage of A2O3 increased and the percentage of other oxides differed only slightly. For modifying oxides (K2O, Na2O and CaO), the standard deviation in individual samples did not exceed ±1%.
3.2. Phase composition of glass–crystalline materials The X-ray tests conducted on the obtained materials (Fig. 1) concluded that only the material marked as E(Ca) is a practically amorphous material, which is confirmed by no clear reflexes in the diffractogram and the occurrence of raised background. For other materials, apart from the amorphous phase (raised background within the range of 2h 5–30°), there is a range of reflexes, indicating the presence of crystalline phases. The performed analysis concluded that there was calcium silicate pseudowollastonite from the cyclosilicate group (Ca3[SiO3]3) (Fig. 2) in the A(Ca), B(Ca), C(Ca) and D(Ca) materials, while in the F(Ca) calcium aluminium silicate, there was anorthite from the tecto-aluminium silicate group (Ca[Al2Si2O8]) (Fig. 3). A change of the phase composition, disappearance of anorthite, transition through the amorphous material and the appearance of a new phase (pseudowollastonite) results from the change of the SiO2/Al2O2 ratio. A gradual decrease in the silica content with an increase in the percentage of aluminium oxide causes a shift in the composition towards areas of pseudowollastonite crystallisation.
Fig. 2. Diffractograms of glass–crystalline materials A(Ca), B(Ca), C(Ca), D(Ca) and pseudowollastonite.
3.3. Scanning electron microscopy (SEM)
Fig. 1. Diffractograms of fired glass–crystalline materials.
Scanning electron microscopy (SEM) allowed the determination of the microstructure of the tested samples (Figs. 4–9), mostly the size and shape of crystallites. An analysis of SEM images confirmed the results of the phase analysis; all tested materials are glass– crystalline materials, except for E(Ca). In the presented photographs (Figs. 4–9), apart from the amorphous phase, crystals of various sizes and shapes are visible. The performed analysis of chemical composition within the micro-areas (EDX) (Figs. 4–9) showed that, for the A(Ca)–D(Ca) materials, the visible crystals were mostly pure calcium silicate. For the F(E) material, it is calcium aluminium silicate. Considering the intensity ratios of individual peaks in EDX spectra and the X-ray tests, it can be concluded that the visible crystals are pseudowollastonite and anorthite in the A(Ca)–D(Ca) materials in the F(Ca) sample, respectively. As for the determination of the structure of the tested materials, it is of the utmost importance to determine the chemical composition of the remaining glass matrix. As can be seen in the EDX spectra (Figs. 4–7), for materials containing pseudowollastonite (materials A(Ca)–D(Ca)), the glass matrix is an aluminium– silicate phase with ions of alkaline metals (K+, Na+), completely deprived of Ca2+ ions. As far as the completely amorphous E(Ca) and the anorthite-containing F(Ca), the glass matrix is an aluminium–silicate phase containing both ions of alkaline metals and Ca2+ ions.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Fig. 3. Diffractogram of the glass–crystalline material F(Ca) and anorthite.
3.4. Spectroscopic tests An additional objective of the presented study is an attempt to define the structure of the vitreous phase in the obtained glass– crystalline materials, i.e. the phase binding all ingredients to form a solid material. This is of the upmost importance as both the type of crystalline phases and the structure of the vitreous matrix determine the final performance. Due to the considerable percentage of the amorphous phase, it is necessary to use methods which would
make it possible to define the so-called near- and medium-range orders. From this point of view, spectroscopic methods are the most appropriate ones, in particular for mid-infrared (MIR) and far-infrared (FIR) as well as Raman spectroscopy. Fig. 10 presents FIR spectra of the tested glass–crystalline materials. The lack of bands in the FIR spectrum of the E(Ca) material shows that it is clearly fully amorphous. However, in the spectra of the A(Ca) to D(Ca) and F(Ca) materials, bands can be seen, which clearly indicates the presence of crystalline phases. The A(Ca)–D(Ca) materials are characterised by the highest degree of crystallinity (the presence of an entire range of relatively intense bands). In this way, the far-infrared tests confirm the conclusions relating to the nature of the tested materials, drawn on the basis of XRD and SEM tests. Glass–ceramic materials designed for firing temperatures above 1200 °C are usually aluminium–silicate materials. Bands connected with vibrations of silico- and alumino-oxygen bonds appear within the range of 1400–400 cm 1 in MIR and Raman spectra of aluminosilicate materials. Fig. 11 presents MIR spectra of the tested materials together with reference spectra (Figs. 12 and 13) of the phases found on the basis of XRD tests, i.e. pseudowollastonite and anorthite. All of the presented spectra are characterised by the presence of three groups of bands to be located within the ranges 1350– 950 cm 1, 850–650 cm 1 and 500–400 cm 1. The high full width at half maximum of the bands confirms a high percentage of the amorphous phase of the tested materials. This mostly applies to samples with the lowest SiO2/Al2O3, i.e. E(Ca) and F(Ca). Based on the XRD research, it is known that the E(Ca) sample is fully amorphous, while pseudowollastonite is present in materials with the lowest content of Al2O3 (A(Ca)–D(Ca)) and anorthite can be found in the materials with the highest content of Al2O3. Any detailed interpretation of MIR spectra of amorphous and glass–crystalline materials is a very difficult due to the mentioned high full width at half maximum (overlapping bands) and the presence of bands characteristic of individual crystalline phases. To facilitate the interpretation of the spectra of such materials, the decomposition of MIR spectra for three selected materials was performed: the material containing pseudowollastonite – D(Ca) (Fig. 14), the completely amorphous material – E(Ca) (Fig. 15) and the material containing anorthite F(Ca) (Fig. 16). The interpretation of the MIR E(Ca) spectrum decomposition is the most unambiguous one due to the lack of crystalline phases. This is typical alumino-silicate glass, in which the structure is modified by the ions of sodium, potassium and calcium (EDX tests). Therefore, the band at 1180 cm 1 is associated with stretching
Analysis of EDS-A(Ca), point 3
Analysis of EDS-A(Ca), point 1
85
A(Ca)
Fig. 4. SEM image of the A(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
86
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Analysis of EDS-B(Ca), point 3
AnalysisB(Ca) of EDS-B(Ca), point 2
B(Ca)
Fig. 5. SEM image of the B(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
Analysis of EDS-C(Ca), point 1
Analysis of EDS-C(Ca), point 3
C(Ca)
Fig. 6. SEM image of the C(Ca) material together with the analysed chemical composition of the crystalline phase (point 1) and the vitreous phase (point 3).
Analysis of EDS-D(Ca), point 2
Analysis of EDS-D(Ca), point 3
D(Ca)
Fig. 7. SEM image of the D(Ca) material together with the analysed chemical composition of the crystalline phase (point 2) and the vitreous phase (point 3).
vibrations of Si@O (defects) [18], the band at 1035 cm 1 with stretching vibrations of SiAO(Si,Al) – silico- and alumino-silicooxygen bridges (bridging bonds), bands at 953 and 903 cm 1 with
stretching vibrations of SiAO and AlAO – broken silico- and alumina-silico-oxygen bridges (terminal bonds), bands at 790 and 747 cm 1 with bending vibrations of SiAOASi and SiAOAAl
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
87
Analysis of EDS-E(Ca), point 3
Analysis of EDS-E(Ca), point 2
E(Ca)
Fig. 8. SEM image of the E(Ca) material together with the analysed chemical composition of the vitreous phase (points 2 and 3).
Analysis of EDS-F(Ca), point 2
Analysis of EDS-F(Ca), point 3
F(Ca)
Fig. 9. SEM image of the F(Ca) material together with the analysed chemical composition of the crystalline phase (point 2) and the vitreous phase (point 3).
[18–20], the group of bands at 695, 640 and 577 cm 1 with vibrations of 3, 4 and 6-membered silico- and alumino-silico-oxygen rings [21–24], respectively, and bands at 469 and 419 cm 1 with bending vibrations of OASiAO and OAAlAO [18–20]. The decomposition of the MIR spectrum for the D(Ca) material is also relatively clear, which results from the percentage of the crystalline phase in the tested material (XRD, FIR and SEM). Based on the XRD and SEM tests, it is known that pseudowollastonite is the crystalline phase present in the tested material; pseudowollastonite is a three-membered cyclosilicate [25,26]. In the MIR spectrum of each of the 3-membered cyclosilicate, there is a characteristic, very intense band within the range of 750 to 700 cm 1, associated with the presence of isolated 3-membered silico-oxygen rings [27]. As it can be easily noticed in the MIR spectra of the tested A(Ca)–D(Ca) materials (Fig. 11) and decomposition of the MIR spectrum of the D(Ca) material (Fig. 14), a band is present at approx. 716 cm 1, which confirms the presence of pseudowollastonite. In addition, by comparison with the analysed decomposition (Fig. 14) and the MIR spectrum of pure pseudowollastonite (Fig. 12), bands at 981 cm 1 (stretching vibrations of SiAO(Si)), 942 cm 1 (stretching vibrations of SiAO ) and 565 cm 1 (bending vibrations of SiAOASi) can be unambiguously assigned to pseudowollastonite [27]. The other bands are mostly associated with the remaining amorphous phase. It results from the EDX spectra (Fig. 8a and b)
that the remaining amorphous phase is an alumino-silicate phase, in which only cations of alkaline metals are the modifiers. Therefore, similarly to the E(Ca) material, the band at 1189 cm 1 should be assigned to stretching vibrations of Si@O [18] and the band at 1052 cm 1 to stretching vibrations of SiAO(Si,Al) [18,19]. It must be noted that the intensity of the latter band is also influenced by the presence of pseudowollastonite. The bands at 914 and 890 cm 1 are derived from stretching vibrations of SiAO and AlAO , while the two bands at 795 and 762 cm 1 are associated with bending vibrations of SiAOASi and SiAOAAl [18–20]. The band at 684 cm 1 is associated with the vibrations of 4 and/or 6-membered silico- and alumino-silico-oxygen rings present in the glass structure [28], whereas the bands at 469 and 430 cm 1 with bending vibrations of OASiAO and OAAlAO [18–20]. The identification of bands characteristic of anorthite in the spectrum of the F(Ca) material is much more difficult, the reason of which mostly results from the low anorthite content in the tested material. Anorthite is a framework alumino-silicate e.g. [29,30], and consequently the bands associated with vibrations of silico- and alumino-silico-oxygen bonds corresponds to the vibrations of such bonds present in the amorphous phase. As a result, anorthite does not have an intensive identification band in the MIR spectrum (Fig. 11) as in the case of pseudowollastonite. Based on the comparison of the decomposition of the MIR spectrum of the F(Ca)
88
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
amorphous phase, regardless of crystalline phases. This problem can be partially solved by using IR and Raman microspectroscopy. On the grounds of the size of the pseudowollastonite and anorthite crystals (see SEM images – Figs. 4–9) in the tested materials, IR microspectroscopy is useless due to the size of the surface area, from which the spectrum is collected (approx. 50 50 lm). This means that it is impossible to obtain separate MIR spectra for crystalline phases and the remaining glass matrix. Raman spectroscopy provides here many more possibilities, i.e. approx. 1 1 lm. Fig. 17 presents Raman spectra of D(Ca), E(Ca) and F(Ca) materials and reference spectra of crystalline phases found on the basis of XRD tests (pseudowollastonite and anorthite). For each sample, measurements were taken at two points to obtain, to the fullest extent possible, an independent spectrum of the crystalline and amorphous phases of the matrix. As can be seen in Fig. 17, for glass–crystalline materials (D(Ca) and F(Ca)), it was possible to perform point measurements of crystalline phases. Raman spectra of the measured crystalline phases correspond exactly to the model spectra of pseudowollastonite and anorthite. The tests unambiguously confirm previous conclusions drawn from XRD and MIR tests. They are yet another piece of evidence for the presence of pseudowollastonite and anorthite in these materials. As well, it was possible to obtain an amorphous phase spectrum for the F(Ca) sample. As can be seen in Fig. 17, the Raman spectrum of the amorphous phase for F(Ca) is virtually identical with the Raman spectrum of the amorphous phase obtained for the E(Ca) sample.
Fig. 10. FIR spectra of the tested glass–crystalline materials.
glaze (Fig. 13) with the MIR spectrum of anorthite (Fig. 13), it can only be assumed that anorthite would be associated with the band at 994 cm 1 (stretching vibrations of SiAO(Si,Al)) and the group of bands within the range of 630–540 cm 1 (bending vibrations of SiAOASi and SiAOAAl) [18–20]. Similarly to the E(Ca) material, the remaining amorphous phase is an alumino-silicate phase, in which alkaline metal ions and calcium ions are the modifiers of the structure. Therefore, the band at 1180 cm 1 should be assigned to stretching vibrations of Si@O [18] and the band at 1088 cm 1 to stretching vibrations of SiAO(Si,Al) [19,20]. The intensity of the latter band is also influenced by the presence of anorthite. The bands at 931 and 891 cm 1 results from stretching vibrations of SiAO and AlAO , the bands at 774 and 722 cm 1 from bending vibrations of SiAOASi and SiAOAAl, and the bands at 468 and 427 cm 1 from bending vibrations of OASiAO and OAAlAO [18–20]. By comparing and analysing the decomposition of spectra of selected materials, it can be concluded that the glass matrix in each case is an alumino-silicate phase with a similar structure. The only difference, to which attention must be paid, is a considerably higher integral intensity (the surface area under the curve) of the band associated with the silico- and alumino-silico-oxygen bridges (at 1052 cm 1) in the D(Ca) spectrum, compared to the analogous band in the spectrum for E(Ca) – 1035 cm 1, and F(Ca) – 1088 cm 1. This probably results from the mentioned differences in the type of structure modifiers (Na+, K+ and Ca2+). The presented analysis of the MIR spectra of glass–crystalline materials D(Ca) and F(Ca) is not fully accurate due to the averaging nature of MIR spectra, which prevents a direct determination of the structure of the
Fig. 11. MIR spectra of the tested glass–crystalline materials.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
Fig. 12. MIR spectra pseudowollastonite.
of
the
D(Ca)
glass–crystalline
material
and
The most intense broad band for wavenumbers ranging from 1200 to 900 cm 1 is associated with asymmetrical vibrations of SiAO in silicon–oxygen tetrahedra ([SiO4]4 ). This range of wavenumbers depends on the number of non-bridging oxygen atoms (NBOs), which form a tetrahedron [31]. The data to be found in the literature [31–33] show that in the Raman spectra of silicate glass, bands associated with the Q4–Q0 groups (where Q donates the number of bridging bonds) appear at approx. 1150 cm 1 – Q4, 1050 cm 1 – Q3, 1000 – 950 cm 1 – Q2, 880 cm 1 – Q1 and 850 cm 1 – Q0, respectively. Therefore, E(Ca) and F(Ca) glass matrices are dominated by tetrahedra with one broken oxygen bridge (Q3), i.e. for a band at approx. 1040 cm 1. However, slight bending at approx. 1150 and 1000 cm 1 indicate the presence of the Q4 and Q2 groups, respectively. No bands within the range of 950 to 850 cm 1 unambiguously indicates the lack of Q1 and Q0 groups. The band at 580 cm 1 is associated with the vibrations of three-membered silico- and aluminosilico-oxygen rings. The band at approx. 485 cm 1 is mostly associated with bending vibrations of SiAOASi and SiAOAAl [31]. The intensity of this band is also influenced by the presence of 6-membered silico- and alumino-silico-xygen rings [38]. Rings featuring a variable number of members, ranging from 3 to 8, and a variable degree of deformation are an element of the lattice of silico- and alumino-silico-oxygen glasses [18,24]. The band at approx. 790 cm 1 is associated with the vibrations of aluminooxygen tetrahedra with three bridging bonds [34,35]. No band at approx. 710 cm 1 unambiguously indicates that there are no
89
Fig. 13. MIR spectra of the F(Ca) glass–crystalline material and anorthite.
Fig. 14. Decomposition of the MIR spectrum of the D(Ca) material.
alumino-oxygen octahedra [36,37], and thus all aluminium ions are arranged in a tetrahedral coordination, meaning that they take part in the formation of the glass structure. This is confirmed by the previous statement that the tested amorphous phases are silico-alumino-oxygen glasses.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91
90
Fig. 15. Decomposition of the MIR spectrum of the E(Ca) material.
Fig. 17. Raman spectra of the D(Ca), E(Ca) and F(Ca) glass–crystalline materials and the model spectra of crystalline phases. Fig. 16. Decomposition of the MIR spectrum of the F(Ca) material.
4. Conclusions (1) Changing SiO2/Al2O3 ratio in the SiO2–Al2O3–K2O–Na2O–CaO system has a decisive influence on the type and amount of the crystalline phase in the glass–crystalline materials. At the highest SiO2/Al2O3 ratios, the tested materials are typical glass–crystalline materials, in which calcium silicatepseudowollastonite occurs, apart from the vitreous phase. (2) By lowering the SiO2/Al2O3 ratio gradually to 4.77, the quantity of the crystalline phase is decreased, and with SiO2/Al2O3 equal to 3.88, the material is completely amorphous (SEM, XRD). (3) With the SiO2/Al2O3 ratio equal to 3.23, the vitreous phase and calcium aluminium silicate – anorthite occurs in the material. (4) The results of spectroscopic research show that the change in the SiO2/Al2O3 ratio only slightly influences the chemical composition (the type of structure modifiers) and the internal structure of the vitreous phase-bonding crystallites to form a uniform material. This influence is exposed by the degree of polymerisation (the number of broken silicooxygen bridges).
(5) In the tested materials, independently of the SiO2/Al2O3 ratio, aluminium ions occur only in tetrahedral coordination, i.e. they form the amorphous phase structure.
Acknowledgements This research has been carried out thanks to financing under the framework of NCBiR (Polish National Research and Development Committee) programme No. N N508 477734 and PBS1/B5/17/2012.
References [1] W. Holand, G.H. Beall, Glass Ceramic Technology, second ed., Wiley, 2012, ISBN: 978-0-470-48787-7. [2] T. Manfredini, G.C. Pellacani, J.M. Rincon, Glass–Ceramic Materials: Fundamentals and Applications, Mucchi Editore, Modena, Italy, 1997, ISBN10: 887000287X. [3] D.R. Eppler, R.A. Eppler, Glazes and Glass Coatings, The American Ceramic Society, Westerville, Ohio, 2000. [4] R.K. Brow, Inorganic glasses and glass–ceramics, in: M.A. Stoneham, R.E. Loehman (Eds.), Characterization of Ceramics, Butterworth-Heinemann, 1993, pp. 103–118. [5] J.E. Shelby, Introduction to Glass Science and Technology, Athenaeum Press Ltd, UK, 1997.
J. Partyka, M. Les´niak / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 152 (2016) 82–91 [6] Li. Neng, W.Y. Ching, Structural, electronic and optical properties of a large random network model of amorphous SiO2 glass, J. Non-Cryst. Solids 383 (2014) 28–32. [7] J. Zhong-Hong, Z. Qin-Yuan, The structure of glass: a phase equilibrium diagram approach, Prog. Mater Sci. 61 (2014) 144–215. [8] Ch. Haiyan, H. Guosong, M. Hanfen, G. Fuxi, Structure and Raman spectra of glasses containing several glass-forming oxides and no glass-modifying oxide, J. Non-Cryst. Solids 80 (1986) 152–159. [9] K. Alptekin, F. Kara, Glass–ceramic glazes in the CaO–SiO2 system, Key Eng. Mater. 264–68 (2004) 1709–1712. [10] C.F. Binns, The development of the matt glaze, Trans. Am. Ceram. Soc. 5 (1903) 50–63. [11] A. Karamanov, M. Pelino, Induced crystallization porosity and properties of sintereds diopside and wollastonite glass-ceramics, J. Eur. Ceram. Soc. 28 (2008) 555–562. [12] I.A. Levitskii, N.V. Mazura, Opacified glazes produced by high-temperature firing for sanitary ceramicware, Glass Ceram. 62 (2002) 215–218. [13] E.F. Osborn, The system CaSiO3-diopside-anorthite, J. Am. Sci. 11 (1942) 751– 788. [14] W.D. Kingery, H.K. Browen, D.R. Uhlman, Introduction to Ceramics, second ed., John Wiley and Sons, New York, 1960. [15] J.R. Taylor, A.C. Bull, Ceramic Glaze Technology, Pergamon Press, Oxford, 1986. [16] R.T. Stull, Influence of variable silica and alumina on the porcelain glazes, Trans. Am. Ceram. Soc. XIV (1912) 62–72. [17] J. Partyka, M. Sitarz, M. Les´niak, K. Gasek, P. Jelen´, The effect of SiO2/Al2O3 ratio on the structure and microstructure of the glazes from SiO2–Al2O3–CaO–MgO– Na2O–K2O system, Spectrochim. Acta Part A 134 (2015) 621–630. [18] M. Sitarz, The structure of simple silicate glasses in the light of Middle Infrared spectroscopy studies, J. Non-Cryst. Solids 357 (2011) 1603–1608. [19] W. Mozgawa, M. Sitarz, M. Rokita, Spectroscopic study of different aluminosilicate structures, J. Mol. Struct. 511–512 (1999) 251–257. [20] W. Mozgawa, M. Sitarz, Vibrational spectra of aluminosilicate ring strictures, J. Mol. Struct. 614 (2002) 273–279. [21] M. Sitarz, W. Mozgawa, M. Handke, Vibrational spectra of complex ring silicate anions-method of recognition, J. Mol. Struct. 404 (1997) 193–197. [22] M. Handke, M. Sitarz, W. Mozgawa, Model of silicooxygen ring vibrations, J. Mol. Struct. 450 (1998) 229–238.
91
[23] M. Sitarz, M. Handke, W. Mozgawa, The calculations of silicooxygen rings vibrations, Spectrochim. Acta Part A 55 (1999) 2831–2837. [24] M. Sitarz, M. Handke, W. Mozgawa, Rings in the structure of silicate glasses, J. Mol. Struct. 511–512 (1999) 281–285. [25] J.W. Jeffery, L. Heller, Acta Cryst. 6 (1953) 807. [26] T. Yamanaka, H. Mori, Acta Cryst. 1981 (B37) (1981) 1010. [27] M. Sitarz, M. Handke, W. Mozgawa, E. Galuskin, I.O. Galuskina, The non-ring cations influence on silicooxygen rings vibrations, J. Mol. Struct. 555 (2000) 357–362. [28] M. Sitarz, M. Handke, W. Mozgawa, Identification of silicooxygen rings in SiO2 based on IR spectra, Spectrochim. Acta Part A 56 (2000) 1819–1823. [29] H.D. Megaw, C.J.E. Kempster, E.W. Radoslovich, The structure of anorthite, CaAl2Si2O8. II. Description and discussion, Acta Cryst. 15 (1962) 1017–1035. [30] D.W. Matson, S.K. Sharma, J.A. Philpotts, Raman spectra of some tectosilicates and of glasses along the orthoclase-anorthite and nepheline-anorthite joins, Am. Mineral. 71 (1986) 694–704. [31] S. Agathopoulos, D.U. Tulyaganov, J.M.G. Ventura, S. Kannan, A. Saranti, M.A. Karakassides, J.M.F. Ferreira, Structural analysis and devitrification of glasses based on the CaO–MgO–SiO2 system with B2O3, Na2O, CaF2 and P2O5 additives, J. Non-Cryst. Solids 352 (2006) 322–328. [32] O. Bjorn, J. Mysen, D. Frantz, Raman spectroscopy of silicate melts at magmatic temperatures, Chem. Geol. 96 (1992) 321–332. [33] M. Szumera, The structural role of manganese ions in soil active silicate– phosphate glasses, Spectrochim. Acta Part A 129 (2014) 601–608. [34] P.L. Higby, R.J. Ginther, I.D. Aggarwal, E.J. Friebele, Glass formation and thermal properties of low-silica calcium aluminosilicate glasses, J. Non-Cryst. Solids 126 (1990) 209–215. [35] P. MacMillan, B. Piriou, Raman spectroscopy of calcium aluminate glasses and crystals, J. Non-Cryst. Solids 55 (1983) 221–242. [36] C. Huang, E.C. Behrman, Structure and properties of calcium aluminosilicate glasses, J. Non-Cryst. Solids 128 (1991) 310–321. [37] K. Sharma, J.A. Philipotts, D.W. Matson, Ring distributions in alkali and alkaline earth aluminosilicate framework glasses-A Raman spectroscopic study, J. NonCryst. Solids 71 (1985) 403–410. [38] F.L. Galeener, R.A. Barrio, E. Martinez, R.J. Elliott, Vibrational decoupling of rings in amorphous solids, Phys. Rev. Lett. 53 (1984) 2429–2432.
