• Marcinkowski, Manja
• Schmidt, Richard
• Körner, Stefan
• Partsch, Uwe

Abstract

Fraunhofer IKTS is developing customized thick film pastes for different ceramic substrate materials, but especially thick film resistor (TFR) pastes for Aluminum Nitride (AlN) ceramics. AlN exhibits an excellent thermal conductivity paired with a high dielectric strength and a low thermal coefficient of expansion close to that of silicon, making it ideally suitable as substrate material for power electronics. Nevertheless, thick film pastes for this reactive ceramic need an adapted composition. All chemical components are required to show a controllable or no reactive behavior with the substrate material. Whitin this regard, the key role plays specifically adapted glass phases which prevent or slow down the kinetics of possible reactions. Furthermore, up to now the chemically stable Ruthenium dioxide (RuO2) is the only available conducting phase material usable in resistor pastes for AlN until now. This limits the sheet resistivity of the pastes to approximately 10 kOhm/sq in maximum. Bismuth and Lead Ruthenates (Bi2Ru2O7 / Pb2Ru2O6.5) with higher specific resistivities than Ruthenium dioxide are not applicable on AlN, because they decompose to RuO2 and the respective base metal Bi2O3/PbO while firing of the films at 850 °C in air. The latter will react spontaneously with the substrate material to form Al2O3 and gaseous Nitrogen, which inevitably forms bubbles in the films making its properties uncontrollable. The thin evolving Alumina interface also creates a thermal barrier, reducing the desired thermal conduction abilities of the substrate and leading to poor adhesion of the TFR. For TFR pastes > 10 kOhm/sq on AlN, there is a prior need for a chemically stable conducting phase under firing conditions paired with a higher specific resistivity than Ruthenium dioxide. Studies of Rane [1] showed, that CaRuO3 can be used for lead and bismuth free resistor pastes on Alumina. The work of Abe [2] is interesting due to that a mixture of WO3·RuO2 was used as functional phase on Alumina. In the work, a lead borosilicate containing glass phase was used, which is not applicable on AlN and thus was exchanged. The WO3·RuO2 material composition was examined because the exchange of a RuO2 with WO3 promises higher sheet resistivities of the pastes due to that the specific resistivity of WO3 is several potencies higher than that of RuO2. Paired with a high chemical stability the material is also a promising candidate to be reviewed. Additionally, both materials mentioned were not reported in the literature so far to be used as functional phases in TFR pastes on AlN ceramics. The presentation will cover the synthesis of the functional phases by solid state reaction, the characterization using X-ray diffraction as well as paste manufacturing. The pastes were screen printed, fired in air at 850 °C and electrically characterized for sheet resistivity and Temperature Coefficient of resistance as well as resistance change after artificial aging at elevated Temperature and humidity and assessed in comparison to film properties of standard RuO2 containing TFR pastes.

Topics

• impedance spectroscopy
• resistivity
• phase
• x-ray diffraction
• aluminium
• glass
• reactive
• glass
• Nitrogen
• nitride
• strength
• chemical stability
• Silicon
• aging
• thermal conductivity
• aging
• Bismuth
• dielectric strength