LOW TEMPERATURE AIR BRAZE PROCESS FOR JOINING SILICON CARBIDE COMPONENTS USED IN HEAT EXCHANGERS, FUSION AND FISSION REACTORS, AND OTHER ENERGY PRODUCTION AND CHEMICAL SYNTHESIS SYSTEMS

J.R. Fellowsa, C.A. Lewinsohna, Y. Katohb, T. Koyanagib

aCeramatec, Inc., Salt Lake City, UT 84119, USA

bOak Ridge National Laboratories, Oak Ridge, TN 37831, USA

Notice: This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

ABSTRACT

Fabrication of large, or complex, components from silicon carbide, or other technical ceramics, used in heat exchanger devices, energy production and chemical synthesis systems, and for components within fusion and fission reactors require robust joining processes. Ceramatec has developed a novel method for achieving bonds using an air brazing process. For silicon carbide joining, the braze acts under certain conditions to promote diffusion bonding. The resulting joined regions are thought to form by rapid interdiffusion of the diffusion-enhancing braze material and silicon and carbon species, resulting in a microstructure more similar to one formed by diffusion bonding than brazing. Processing of these joints is accomplished at relatively low temperatures, 900°C-1200°C in air, with minimal applied load. The brazed joint strength was found to be statistically equivalent to monolithic control samples at room temperature. Oxidation testing, using dry oxygen and saturated steam, was conducted at 1000°C for 1000 hours on joined specimens, resulting in further microstructural development of the joint, with subsequent shear testing showing no appreciable reduction in strength. Torsion tests on irradiated joined samples show that the joint's mechanical integrity is resistant to radiation degradation.

1. INTRODUCTION

The use of silicon carbide (SiC) ceramics within high efficiency heat exchanger systems and other energy related structures is increasingly prevalent due to SiC exhibiting high strength and corrosion resistance at elevated temperatures and pressures1. There is also a great deal of focus on SiC-SiC composites, due to the composite structure offering improved mechanical properties (compared to monolithic SiC), for accident tolerant fuel (ATF) cladding and fuel rods (assemblies that include tubular cladding sections and an endplug bonded together) that will survive a loss-of-coolant accident (LOCA), which is vital for the improved safety of light water reactors (LWR)2,3.

For the fabrication of ATF cladding and other structures used in industrial applications where silicon carbide-based ceramics are utilized, joining of individual components to produce larger structures is required where complex shapes, geometries, and often substrate morphological variations (such as gradients of structural porosity and volumetric alterations) cannot be fabricated as an individual component. Such is the case, for example, with heat exchanger stacks that utilize individual micro-channeled plates joined into larger modules4. Current efforts are being made to identify joining solutions to join a monolithic CVD-SiC endplug to SiC-SiC composite tube cladding for ATF application within light water reactors5. In all cases, the joint itself must meet certain criteria of strength, ability to obtain hermetic seals, resistance to corrosive environments such as oxidative damage, and also, especially in the case for use in LWR, the joint must be able to survive neutron irradiation and show stability in this environment.

The focus of this current research is to identify a candidate joining method with adequate properties that can be further evaluated for possible use in joining a dense silicon carbide endplug to a SiC-SiC composite tube used in light water reactor application, where the joint itself must also be suited to survive constant neutron irradiation and the possibility of a loss-of-coolant accident. In addition, it is desired that this joining solution will be applicable to other assemblies and applications, such as heat exchanger devices, electronic materials processing tools, metrology tools, satellite mirrors, modular structures, etc.

2. EXPERIMENTAL

2.1. Materials - Joint Initiator

Ceramatec has developed and patented6 a ceramic to ceramic brazing process that utilizes aluminum as a 'joint initiator'. As described in the patent, there are various processing and joining parameters to vary depending on the desired joint microstructure. All SiC joining described in this paper was accomplished according to methods discussed in the patent, with aluminum (purity greater than 99.5%) being applied to only one joining substrate surface (SiC surfaces ground using a 15 micron diamond grit wheel before aluminum application). The temperature used to join the samples described below was between 900°C and 1200°C. Nominal load was applied to the samples to maintain alignment during processing.

2.2. SiC Substrates used for Joining

This paper describes both qualitative and quantitative results of joints formed between direct sintered SC-30 SiC and CVD-SiC substrates provided by CoorsTek. It is acknowledged that direct sintered SiC is a candidate ceramic often used in micro-channeled heat exchanger devices, energy related SiC ceramic systems, and many other SiC ceramic applications, while CVD-SiC is representative of the CVD matrix and coating anticipated on the proposed ATF fuel rods. As will be discussed in this paper, the qualitative nature of the joint has been observed to be identical for either substrate type. Due to the experimental observations, it is reasonable at this time to state that resulting joints formed using direct sintered SiC substrates, with inherent sintering additives found in the SiC microstructure, are comparable to joints obtained using CVD-SiC substrates, which do not have sintering additives in the microstructure despite the minor differences between the materials in the presence of second phases and in grain morphology. This observation will be seen in testing data shown later in this report.

2.3 Joined Specimens for Shear, Tensile, and Torsion Strength Testing

2.3.1 Shear Testing: Double-lap Shear Samples (direct sintered SC-30 SiC substrates)

As a screening method to evaluate several joints being investigated by Ceramatec, a double-lap shear sample geometry was developed. Figure 1 shows FEA modeling that indicates that the notched samples produce the highest shear stress on the joint plane. The initial goal was to be able to test and rank various joints in double-lap shear and test the effects on processing variables. As such, this model is based on a monolithic structure, where (when the notches are placed and the structure is loaded as indicated by the arrows) maximum shear stresses develop at the region where the joints exist in actual joined samples. The dashed lines, as noted in this figure, indicate the location of the joints. Figure 1 also shows a typical sample that is finish ground (320 grit grinding parallel to the joint plane) to dimensions of 14.0 mm wide × 7.75 mm tall × 4.0 mm thick. While both Ferraris et al.7 and Ventrella et al.8 discuss more accurate shear testing methods, this initial test method was sufficient to identify which joint types used in the scoping trials were superior to others among those that were joined and evaluated.

Figure 1. FEA modeling showing that the maximum shear stress occurs on the joint plane.

Double-lap shear testing was completed at both room temperature (RT) and at elevated temperatures in air. RT shear testing was accomplished using an Instron Model 5566 test frame (utilizing a 10 kN static load cell). For elevated temperature testing, this same test frame was used with a box furnace placed within the frame as seen in Figure 2. An image of a single specimen placed within the furnace ready for testing is also seen in this same figure.

2.3.2 Tensile Testing: 4-pt Bend Samples (direct sintered SC-30 SiC substrates)

Joined samples were prepared and evaluated in 4-pt bend according to the ASTM C1161 standard (configuration B) with testing completed using the Instron Model 5566 test frame (utilizing a 10 kN static load cell).

2.3.3 Torsion Testing (CVD-SiC substrates)

CVD-SiC substrates were joined and ground to specific dimensions as discussed by Henager et al.9, and shown in Figure 3. Testing of torsion specimens was conducted at Oak Ridge National Laboratories (Oak Ridge, TN 37831, USA), with the torsion test fixture shown in Figure 4.

Figure 2. Shear testing on Instron universal testing machine Model 5566 (utilizing a 10 kN static load cell): a) Elevated temperature testing box furnace was placed within the frame, b) single shear sample, and c) sample placed inside the box furnace within the test frame.

RT torsion testing was accomplished on as-joined samples to establish baseline strength values. Additional samples were then subject to neutron irradiation (using the...