Influence of composition on the structure and

Materials Science-Poland, 33(4), 2015, pp. 862-866 http://www.materialsscience.pwr.wroc.pl/ DOI: 10.1515/msp-2015-0117

Influence of composition on the structure and properties of SrO–Sb2O3–P2O5 low-melting sealing glasses H AINING N IE , J INGJING Z HANG , YAJUN Q I , Z HIQIANG WANG∗ , H AI L IN , S HUWEN J IANG School of Textile and Materials Engineering, Dalian Polytechnic University, Dalian 116034, China SrO–Sb2 O3 –P2 O5 glass system was prepared by high temperature melting method. The effects of Sb2 O3 and P2 O5 content on the structure, thermal behavior and chemical durability of the glasses were studied by infrared spectrometer, thermal dilatometer, differential thermal analyzer and constant temperature water bath heating. It can be concluded that the characteristic temperatures of the glasses increased firstly and then decreased with the increasing of Sb2 O3 content, whereas the tendency of the coefficient of thermal expansion (CTE) varied inversely. The crystallization ability of the glasses was significantly increased and the water resistance was reduced for Sb2 O3 content of 35 mol % and 40 mol %. The glasses with 20 mol %, 25 mol % and 30 mol % Sb2 O3 showed better performance in every respect than the others and the glasses containing 25 mol % Sb2 O3 , characterized by the best performance, can be chosen as host glasses for further research. Keywords: thermal behavior; chemical durability; coefficient of thermal expansion; water durability © Wroclaw University of Technology.

1.

Introduction

Low-melting glasses have been widely used for sealing as a kind of solder in automotive industry, aerospace, electronics and coatings. There are many advantages of sealing glasses, such as good moisture resistance, air tightness, high temperature resistance, small expansion coefficient, etc. [1]. In the past several years, most commercial sealing glasses were lead-based with large amount of PbO, which is deleterious to health and environment. Restrictive environmental norms which limit the applicability of PbO systems due to their polluting emissions have been issued. Therefore, the measures to find feasible alternatives has been undertaken [2, 3]. Phosphate glasses with unique physical properties, such as low viscosity, good fluidity, low transition temperature and wide range of CTE, which make them superior to silicate and borosilicate glasses for low-melting applications, seem to fulfill such requirements [4]. However, the problem faced by such systems is that they usually suffer from poor chemical durability due to the ∗ E-mail:

[email protected]

Fig. 1. Glass-forming region of the SrO–Sb2 O3 –P2 O5 system.

existence of easily hydrated phosphate chains [5]. Several studies have shown that the addition of various oxides can modify the durability of phosphate glasses [6–8]. SrO is one of the oxides that can enhance the glass forming ability and improve the chemical stability [9]. To decrease the transition temperature (Tg ), Sb2 O3 are usually selected. The SrO–Sb2 O3 –P2 O5 based system of glasses

Unauthenticated Download Date | 1/15/16 10:51 PM

863

Influence of composition on the structure and properties of SrO–Sb2 O3 –P2 O5 . . .

became a novel ternary system for low-melting applications. Up to now no experiment has been conducted to evaluate the influence of composition on the structure and properties of SrO–Sb2 O3 –P2 O5 glasses. In this paper, the SrO–Sb2 O3 –P2 O5 glasses with different compositions were fabricated and the basic properties of the glasses were tested by infrared spectrophotometer, horizontal dilatometer and differential thermal analyzer. The effect of Sb2 O3 and P2 O5 content on the glass properties was analyzed. Furthermore, an optimized composition of the ternary system was given.

2.

Experimental

characteristic temperatures (Tg and Tf ) and CTE, using a horizontal dilatometer (PCY, Xiangtan Instrument, China). Besides, thermal stability was tested by a differential thermal analyzer (WCR-2D, Beijing Photics, China) under a heating rate of 15 °C/min. The infrared spectra of the glasses were obtained by an infrared spectrophotometer (PE, model spectrum One-B). Rectangular (5 mm × 5 mm × 20 mm) samples were also prepared for water durability testing. The specimens were polished and then immersed in deionized water at 50 °C for 12 h to evaluate the weight loss of a unit surface area (mg/cm2 ). An average weight loss of the glass was calculated from 5 samples. Table 1. The compositions of SrO–Sb2 O3 –P2 O5 system glasses.

The glass-forming region of the SrO–Sb2 O3 –P2 O5 system, illustrated in Fig. 1, was studied by numerous previous experiments. The open circles represent the transparent glass samples and the close circles denote crystallized samples, while the half-open circles denote opaque samples with phase separation. It can be seen that the glass-forming region crosses the entire ternary system, from 45 mol % to 70 mol % P2 O5 . According to the glassforming region, 10SrO(20+x)Sb2 O3 (70–x)P2 O5 system was designed to research the effect of Sb2 O3 content on the glass structure and properties (Table 1), as drawn in Fig. 1 by the line. Five glasses of nominal compositions (mol %) 10SrO(20+x)Sb2 O3 (70–x)P2 O5 , where x = 0, 5, 10, 15, 20, were studied (Table 1) here. Fabrication of the glass samples was carried out by a melt-quenching process using chemically pure (NH4 )2 HPO4 , SrCO3 and Sb2 O3 as raw materials. The well-mixed batch materials were pre-treated at 300 °C for 3 h to avoid the evaporation of P2 O5 at high temperature. After that, the mixture was melted in a high purity alumina crucible at 1100 °C for 2 h. Quenched in a graphite mold, the formed glasses were annealed in a muffle furnace at 400 °C for 0.5 h and cooled down slowly to room temperature.

Samples S20 S25 S30 S35 S40

3. 3.1.

Compositions (mol %) P2 O5 SrO Sb2 O3 70 65 60 55 50

10 10 10 10 10

20 25 30 35 40

Results and discussion Structural analysis

Infrared spectra of the samples are shown in Fig. 2. The absorption peak at about 1284 cm−1 is attributed to the P=O bonds stretching vibration and the absorption peak at 1040 cm−1 is due to the stretching vibration of P–O tetrahedra [10]. In addition, the absorption peaks at 912 cm−1 and 775 cm−1 are ascribed to the symmetric and asymmetric stretching vibrations of P–O–P bonds [11–13], respectively. The band around 538 cm−1 is the absorption peak of bending vibration of P–O bonds, δ(P–O), of Q0 tetrahedra [10]. The absorption shoulder at 629 cm−1 is caused by the vibration of P–O–Sb linkages [1, 14, 15]. It is obvious that for S35 and S40 the intensities of the bands at The bulk glasses were machined into cylinders 1284 cm−1 and 775 cm−1 decreased markedly, bewith the dimension of Φ 5 × 50 mm to measure cause the excessive Sb2 O3 could not take part in the


864

H AINING N IE et al.

glass network due to inadequate free oxygen provided by SrO when the content of Sb2 O3 exceeded 30 mol %. As a network modifier, Sb2 O3 can provide free oxygen to glasses, which is able to break P=O bonds and decrease the vibrations of P–O–P bonds [15].

Sb2 O3 can participate in the S25 glass network in the largest extent. On the contrary, the glass samples of S35 and S40 have lower characteristic temperatures and higher CTE because in those glasses there is an excess of Sb3+ as the network modifier, which causes the glass-network weakening [1, 15]. Simultaneously, in this ternary system of phosphate glass, plenty of P2 O5 is replaced by Sb2 O3 with much lower melting point, leading to the worsening of its properties.

Fig. 2. Infrared spectra of the glass samples.

3.2.

Fig. 3. The characteristic temperatures and CTE of glass samples with different Sb2 O3 content.

Thermal behavior

The effect of Sb2 O3 content on the characteristic temperatures (Tg and Tf ) and CTE of the glasses is shown in Fig. 3. It can be seen that with increasing of Sb2 O3 , the CTE of the glasses initially decreases, and then increases with further increasing of Sb2 O3 . At the same time, the increasing of Sb2 O3 at first causes the increase of the characteristic temperatures but then the temperatures decrease. The characteristic temperatures and CTE of the glasses are primarily correlated with the glass composition and structure [16]. The integrity of the glass structure is associated with the increase in the characteristic temperatures but the drop of CTE. As Sb2 O3 plays an important role in the glass as an intermediate network oxide, in S20, S25 and S30, Sb3+ can participate in the glass-network because of the free oxygen provided by SrO, which can strengthen the network structure, so these glasses have high characteristic temperatures and low CTE. It is also obvious that S25 has the highest characteristic temperature and the lowest CTE. This illustrates that

Fig. 4 shows the DTA curves of the glasses. It is clear that the crystallization peaks have occurred only in S35 and S40 glasses. In S20, S25 and S30, Sb3+ takes part in the network structure as a glass network former. The enhancement of the network structure and dynamic viscosity impede the migration of Sr2+ , thus, the crystallization activation energy of the glasses improves and the crystallization ability decreases. In S35 and S40 there is an excess of Sb3+ in the glasses as a network modifier due to inadequate free oxygen provided by SrO, thus, plenty of Sb3+ tend to accumulate and the network of the glasses is weakened [17]. Correspondingly, the crystallization ability of the glasses increase.

3.3.

Water durability

The water durability of glasses with various Sb2 O3 content is shown in Fig. 5. It is clear that the glasses with Sb2 O3 less than 30 mol % have good water durability (S20, S25 and S30). However, the weight loss of the glasses increased rapidly when


Influence of composition on the structure and properties of SrO–Sb2 O3 –P2 O5 . . .

Fig. 4. DTA curves of the glasses with different Sb2 O3 content.

Fig. 5. Weight loss Sb2 O3 content.

of

glasses

865

with different

the content of Sb2 O3 increased from 30 mol % behavior and water durability at different Sb2 O3 to 35 mol %, then, with the further increasing of contents were measured along with the analysis of the changes in the structure. The following Sb2 O3 , the weight decreased rapidly. conclusions can be drawn from our experimental It is well-known that the poor water durability results. of phosphate glasses is attributed to the breaking of phosphate chains and their dissolution in wa1. The structure of phosphate glasses has ter [18, 19]. All the Sb3+ may have joined the netchanged with the increase of Sb2 O3 content. work as the network modifier in S20, S25 and S30 The P=O bond was destroyed and the viglasses. The P–O–Sb bonds are corrosion resistant bration of P–O–P bond was weakened bebecause of their high polarizing power [15]. Howcause an excess of Sb2 O3 existed as the ever, when Sb2 O3 content increased from 30 mol % network modifier when Sb2 O3 content exto 35 mol %, there was inadequate free oxygen ceeded 30 mol %. provided by SrO for Sb2 O3 to take part in the 2. Phosphate glasses had high characteristic network, so the excess Sb2 O3 broke the glasstemperatures, low CTE and weak crystalnetwork as a network modifier. This caused the lization ability when the glass contained less water resistance of the glasses to decrease and than 30 mol % Sb2 O3 . On the contrary, the weight loss of the glasses increased markedly the content of Sb2 O3 above 30 mol % led (Fig. 5). At higher concentrations (from 35 mol % to a decrease in the characteristic temperato 40 mol %), Sb2 O3 as a replacement of water soltures but the CTE and crystallization ability uble P2 O5 is difficult to remove from the glass beincreased. cause of larger ionic radius. As a result, the chem3. Phosphate glasses with less than 30 mol % ical stability of S40 is better than that of S35 but Sb2 O3 had good water resistance, but when worse than S20, S25 and S30. the content of Sb2 O3 reached 35 mol %, the water resistance of the phosphate glasses decreased. With the further increasing of 4. Conclusions Sb2 O3 , the water resistance of phosphate Phosphate glasses based on the system glasses improved but it was worse than that 10SrO(20+x) Sb2 O3 (70–x)P2 O5 (x = 0, 5, 10, of the glasses with less than 30 mol % 15, 20) were fabricated. The structure, thermal Sb2 O3 .


866

H AINING N IE et al.

4. Phosphate glass with 25 mol % Sb2 O3 [9] WANG J.X., C HEN G.H., Key Eng. Mater., 633 (2014), 344. showed the best performance in comparison [10] M IERZEJEWSKI A., S AUNDERS G.A., S IDEK H.A.A., with the others and it was selected to proB RIDGE B., J. Non-Cryst. Solids, 104 (1988), 323. duce host glasses for further research. [11] S UBBALAKSHIMI P., S ASTRY P.S., V EERAIAH N., Acknowledgements Grateful acknowledgement is made to the support from the National Natural Science Foundation of China (61275057). It was also supported by the Ministry of Education, Liaoning Province of China under the Project L2014226.

References [1] Q I Y.J., WANG Z.Q., Z HAI S.R., J IANG S.W., L IN H., Mater. Sci.-Poland, 32 (2004), 414. [2] DIRECTIVE 2002/95/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 27 January 2003 on the restriction of the use of certain hazardous substances in electrical and electronic equipment. [3] H IROSHI U., YASUKO D., RYUICHI T., T SUNEO M., US Patent, (2002), 6355586B1. [4] Z AITSEV D.D., K AZIN P.E., T RET ’ YAKOV Y.D., M AKSIMOV Y.V., S UZDALEV I.P., JANSEN M., Inorg. Mater.+, 40 (2004), 1111. [5] Z HANG B., C HEN Q., S ONG L., L I H.P., H OU F.Z., Z HANG J.C., J. Non-Cryst. Solids, 354 (2008), 1948. [6] G U S.Y., WANG Z.Q., J IANG S.W., L IN H., Ceram. Int., 40 (2014), 7643. [7] H ONG J.H., J. Non-Cryst. Solids, 356 (2010), 1400. [8] H AFID M., J ERMOUMI T., N IEGISCH N., M ENNIG M., Mater. Res. Bull., 36 (2001), 2375.

Phys. Chem. Glasses-B, 42 (2001), 307. [12] S HIH P.Y., C HIN T.S., Mater. Chem. Phys., 60 (1999), 50. [13] DAYANAND C., B HIKSHAMAIAH G., S ALAGRAM M., M URTHY A.S.R.K., B HIKSHAMAIAH G., J. Mater. Sci., 31 (1996), 1945. [14] S UDARSAN V., K ULSHRESHTHA S.K., J. Non-Cryst. Solids, 286 (2001), 99. [15] Z HANG B., C HEN Q., S ONG L., L I H.P., H OU F.Z., J. Am. Ceram. Soc., 91 (2008), 2036. [16] R EFKA O.O., S AIDA K., J EAN J.V., I SMAIL K., A B DELAZIZ E.J., M OHAMED J., J. Non-Cryst. Solids, 390 (2014), 5. [17] NASU H., N INAGAWA S., I NOUE K., H ASHIMOTO T., I SHIHARA A., Bull. Chem. Soc. Jpn., 120 (2012), 436. [18] R EIS S.T., K ARABULUT M., DAY, D.E., J. Non-Cryst. Solids, 292 (2001), 150. [19] L IU H.S., C HIN T.S., Y UNG S.W., Mater. Chem. Phys., 5 (1997), 1.


Received 2015-03-08 Accepted 2015-10-06