June 15 2008
Ceramics Bonding Using Solder Glass Frit
Ceramics bonding is becoming an important technology and has found wide applications in different engineering and electronic industries. In this paper, furnace bonding of ceramics using solder glass frit was investigated with emphasis on the effects of surface treatment and bonding conditions on bonding strength. Alumina (Al^sub 2^O^sub 3^) sheet and SCHOTT solder glass G017-393 were used as the base and brazing filler materials, respectively. Chemical surface treatments using various acids were tested. The results reveal the effects of spreading and voids on bonding strength. An optimum bonding strength can be produced by an appropriate combination of bonded glass-frit density and spreading area. Bonding strength is not only related to surface-contact angle but also surface roughness. The study shows that high-quality ceramics bonding or sealing can be achieved with the application of appropriate bonding conditions.Key words: Ceramics, bonding, surface treatment, solder glass frit, processing parameters, surface roughness, surface-contact angle
INTRODUCTION
Like some advanced ceramics, glass has also exerted a formative influence on the development of various industries, particularly in optics, electronics, chemistry, and pharmaceutics.1-7 It has been realized that glasses, especially those used to enhance densification during sintering, have potential to join ceramic components. One of the advantages of using these glass materials as fillers for joining is that chemical compatibility with the parent ceramics is generally assured. Other advantages are that the viscosity, flowability, and melting characteristics of glasses can be controlled over wide ranges, and adherence of the glasses to the ceramics is usually quite good. Another desirable feature of glasses is that many compositions can be crystallized to improve their mechanical and corrosive properties. Some successful cases, but not many, were reported, which mainly involved bonding alumina (Al^sub 2^O^sub 3^) to metals, alumina to glasses, zirconia (ZrO^sub 2^) to zirconia, and silicon nitride (Si^sub 3^N^sub 4^) to silicon nitride.8 However, their bonding temperatures ranged from 1,200°C to 1,700°C. This means that the glass fillers (or intermediates) used in these experiments served as brazing materials, based on the definition of brazing.
Solder glasses are specially designed for joining glass to other glasses, ceramics, or metals at lower temperatures, which is attributed to the lower softening point of the glasses. It is expected that the materials to be joined will not be thermally damaged during bonding.7,9 Therefore, it has many potential applications in electronics packaging. Although many efforts have been concentrated on diffusion, electrostatic (or anodic), and eutectic bonding of ceramics,8,10,11-16 the relevant literature focusing on low-temperature ceramics furnace bonding using solder glasses is rather limited.14,17 As a matter of fact, ceramics bonding using solder glasses can be a cost-effective process.5,8
Like other ceramics joining techniques, ceramics bonding using solder glasses is also a field where much research and development work is in progress. However, only a few examples have achieved widespread commercial applications.8 One of the important issues is that the so-called solder glasses are not truly solders because their melting points can (and frequently do) exceed the highest soldering temperature, 450°C, which was specified by the definition of soldering. Therefore, the bonding temperatures fall in the range of brazing temperature. Moreover, during bonding, the so-called solder glasses do not flow by capillary action,5,18 which implies that the mechanism of the ceramics bonding using the solder glasses is likely to be fundamentally different from that of typical brazing and soldering processes, and the mechanism needs to be investigated.
The objective of the present paper is to investigate the effect of bonding parameters on the bonding strength of the ceramic joints using solder glass frit. These parameters include substrate surface conditions (surface treatment media, roughness, and contact angle) and bonding conditions (temperature, time, and applied load).
MATERIALS AND EXPERIMENTAL PROCEDURES
An alumina (96% Of Al^sub 2^O^sub 3^, Rubalit 708S, CeramTec AG, Marktrendwitz, Germany) sheet of 1-mm thickness was used as the substrate material. Commercially available diamond scriber was used to cut the alumina sheet into 1 in. × 1 in. substrates.
To study the effect of surface treatment, various acids were tested, namely, hydrofluoric acid (HF), sulfuric acid (H^sub 2^SO^sub 4^), and phosphoric acid (H^sub 3^PO^sub 4^). Before acid treatment, all of the original substrates in the form of as-received were ultrasonically alkalinecleaned for 15 min by soaking the substrates in a solution containing commercially available detergent. Then, the substrates were completely rinsed by deionized water. After this, the substrates were ultrasonically treated in a solution of specified acid for 30 min. All the substrates were completely rinsed by deionized water after the acid treatment, then dried, and stored in a dry cabinet.