Effects of organic additives on viscosity of aqueous tape-casting aluminum nitride slurries

Abstract

Tape casting has been widely used to produce substrates, solid oxide fuel cells (SOFC), capacitors, and heat exchangers [1]. Recently, aqueous tape casting has received special attention because of safety, environmental, and cost considerations [2]. In order to assure the stability of the slurry and the strength and flexibility of green sheets, it is necessary to introduce organic additives to the aqueous slurry. The organic additives can have significant effects on the viscosities of aqueous slurries, and these need to be well understood [3]. However, little information is available. In this work, the effects of organic additives on the viscosities of aqueous tape-casting AlN slurries were studied. Aluminum nitride (AlN) was chosen as the raw material for this work because it is of interest in the area of microelectronic packaging [4]. The non-aqueous tape casting of AlN has been reported by Streicher et al. [5, 6]. However, there are few literature reports of aqueous processing of AlN [7]. The AlN powder used was synthesized by selfpropagation high temperature synthesis and antihydrolysis treated as described previously [8]. The average diameter of the powder is 0.5 μm. The oxygen and nitrogen contents are 1.44 wt% and 32.31 wt% respectively. DP270 (polyacrylate ester, provided by Rhodia Co. of France) and PVA124 (Kuraray Chemical Co., Japan, 10 wt% emulsion) were used as dispersant and binder respectively. PEG (polyethylene glycol) and DBP (di-butyl phthalate) were chosen as plasticizers. The contents of the organic components were expressed in wt% with respect to the AlN powder. Deionized water (PH 5.60–6.0) was used as solvent. All suspensions had 75 wt% solid loading before PVA was added and were prepared by ball-milling with AlN balls. A rotary viscometer was used (Model NDJ-7, Shanghai Balance Instrument Plant) to measure the apparent viscosity of the suspensions. Shear dependent behavior was tested with a Model SR5 viscometer (Rheometric Scientific, Inc., Piscataway, NJ). Fig. 1a shows the apparent viscosity of slurries (75 wt%) without PVA addition as a function of the DP270 dispersant content. It can be seen that high viscosity slurries were obtained at low and high dispersant content. At 0.4 wt% content of the dispersant DP270, the viscosity of the slurry was a minimum, 16 mPa · s. Fig. 1b shows the apparent viscosity of slurries with 5 wt% active PVA124 as a function of the amount of DP270 dispersant, indicating that the PVA124 binder tended to bridge the AlN particles. (5 wt% active PVA124 was chosen as it was the minimum content to ensure sufficient AlN green sheet strength.) At any level of DP270 addition, the apparent viscosities are much higher than for the corresponding slurry without PVA124. However, the minimum viscosity was reached at approximately the same content. This indicates that the binder PVA124 could be compatible with the DP270 dispersant and support stabilization of the slurry against flocculation. The apparent viscosity of slurries (plasticizers/ binder = 1.5, 0.4 wt% dispersant) as a function of the content of active PVA124 is shown in Fig. 2. As the amount of PVA124 increased, the viscosity of slurries gradually increased, implying that PVA124 could impede the relative motion of AlN particles. However,