DEVELOPMENT OF HIGH TEMPERATURE JOINING AND THERMOMECHANICAL CHARACTERIZATION APPROACHES FOR SiC/SiC COMPOSITES

Michael C. Halbig1, Mrityunjay Singh2, and Jerry Lang1

1 NASA Glenn Research Center, Cleveland, OH

2 Ohio Aerospace Institute, Cleveland, OH

ABSTRACT

Advanced joining technologies are enabling for the fabrication of large and complex shaped silicon carbide-based ceramic and ceramic matrix composite components to be utilized in high temperature extreme environment applications. Many joining approaches are being proposed and developed. New standardized tests are needed to fully characterize joint properties and capabilities. One such test ISO-13124, was used for mechanical testing in this work. This test configuration allows for testing of joined crossbars in either a tensile or a shear stress state. The REA Bond joining approach using Si-8.5%Hf eutectic phas ealloy was used to join ceramic matrix composite and monolithic silicon carbide materials. In mechanical testing, low strengths were obtained with failures occurring in the joined substrates. Finite element analysis of the stress states revealed stresses concentrations at the edges of up to 30 times higher than the 2 MPa nominal stress for the tensile state. For the shear state, out of plane displace ments occurred.

INTRODUCTION

Silicon carbide fiber reinforced / silicon carbide matrix composites (SiC/SiC) are a class of ceramic matrix composite (CMC) materials are being developed for turbine engine applications for such components as combustor liners, shrouds, vanes, and blades1-4. These CMC components can operate at higher temperatures, require less cooling, and are lighter weight than metal components. The use of CMCs in such applications contributes to increased turbine engine fuel efficiencies, reduced emissions, and long term durability. As interest in fiber reinforced SiC-based composite materials continues to grow due to advancements in their properties, new integration technologies and testing capabilities will be critically needed.

In order to evaluate the mechanical properties of joints, standardized tests and testing capabilities are needed. One such standardized test5, BS-ISO-13124:2011, "Fine ceramics (advanced ceramics, advanced technical ceramics): Test method for interfacial bond strength of ceramic materials," was applied for evaluation of mechanical properties of monolithic SiC and SiC/SiC materials joined to themselves. In this test, two long rectangular substrates are bonded across one another at their midsection to form an "X" shaped crossbar to provide samples for testing either in a tensile stress state or a shear stress state. Due to the need for multiple crossbars for testing and because of the uniques hape, a simple joining approach was needed for processing the joints. The authors had previously reported a diffusion bonding approach for joining SiC based materials using titanium interlayers6-7. However, such an approach needs relatively smooth surfaces and requires high applied loads from a hot press to aid in bond formation. Another joining approach, Refractory Eutectic Assisted BONDing (REABOND) was used for evaluating joints ac cording to ISO-13124. REABond uses Si-8.5Hf eutectic phase allow powder in a green tape for the joining interlayer. During joint processing, no load is needed and the eutectic phase melts to flow over the substrate surface and solidifies during cooling. REBOND has been demonstrated on the joining of SiC/SiC composites resulting dense, crack free joints that filled the contour of the rough CMC surface8.

Joining of SiC/SiC substrate s and monolithic SiC was conducted to support the mechanical test method development. Micros tructural analysis was conducted using optical microscopy and scanning electron microscopy (SEM) coupled with energy dispersive spectroscopy (EDS) to evaluate bond quality. Mechanical tests were conducted at room temperature according to ISO-13124 for testing joints under tensile or shear stress conditions.

In correl ation to the experimental tests, the test standard was evaluated in a finite element model investigation. The purpose of the investigation was to determine the reliability of the interfacial bond strength test for two types of test methods for the ISO-13124, by characterizing the stress state within the bond region by finite element methods. The objective was to determine the stress concentration within the bond region so that a more accurate stress me asure ment can be determined from the experimental test results of the tensile and shear specimens. Knowing this stress state would better characte rize the strength of the ceramic bond and would he lp determine if any modifications were needed to the specimen and/or test setup while still remaining compliant to the ISO-13124 standard for fine and advance ceramics interfacial bond strength test method.

EXPERIMENTAL

The CMCs were two different types of silicon carbide fiber/silicon carbide matrix (SiC/SiC) CMCs. The first CMC was SiC/SiC HiPerCompTM Gen II by GE Energy (Newark, DE). The SiC fibers were Hi-Nicalon Type-S. The SiC matrix was manufactured by the prepreg melt infiltration method9. The second CMC was melt infiltrated (MI) SiC/SiC fabricated by Goodrich Corporation (CA) using Hi -Nicalon SiC fibers with a BN/SiC interface. Both Si C/Si C materials had a 0°/90° woven fiber tow architecture. Due to lower than expected mechanical strength results from the joined CMC materials, joining of monolithic SiC substrates was also conducted. The purpose was to eliminate the added complicati on of low interlaminar properties which are typical of CMCs.

REABond green tapes were prepare d for use as the interlayer for joining. Previously several eutectic phase alloys were evaluated and the Si-8.5Hf eutectic phase alloy was down-selected as giving the best results for joining CMCs8. For the current effort, powders of less than 70 microns in diameter of the Si-8.5Hf eutectic phase alloy were mixed with binders to prepare the green tapes by tape casting. The tapes had a solid loading of about 30-35% and were 0.21 mm thick. A second set of tapes were prepared with 5 wt.% of SiC nanofibers integrated in with the eutectic powders. The SiC nanofibers were approximately 0.15 microns in diameter and 10 microns in length. The SiC nanofibers were produced at NASA GRC10.

The substrates were rectangular bar shapes that had been machined from larger coupons. Joining of two bars at the crossover of their midsection as illustrated in Figure 1, forms the cross-bar shape for testing according to the ISO-13124 standard. The test standard recommends test bars that have dimensions of 12 mm × 4 mm × 4 mm. The test standard and the recommended sample size was developed for standardized testing of monolithic ceramics. However, since the standard is being applied here to the testing of CMCs, the small 4 mm × 4 mm crossover area was increased so that repeating unit cells of the fiber architecture could be present within the bond area to maximize the benefit of the fiber architecture. Therefor the bar size was doubled to 24 mm × 8 mm × 8 mm. However, actual dimensions of the CMC test bars varied due to the sizes of available CMC coupons. The dimensions in the length × width × height were roughly 24 mm × 6 mm × 2 mm for the GE SiC/SiC, 24 mm × 8 mm × 2.5 mm for the BFG SiC/SiC, and 33 mm × 6.4 mm × 3.2 mm for the monolithic SiC material.

Green tapes of each type, with nanotubes and without, were cut to match the mating surface area of the substrates being joined. Multiple layers of the green tape were used to achieve an interl ayer thickness sui table for filling in the surface voids of the paired substrates which arise due to surface roughness from the fiber architecture. Therefore, two green tapes were used to join the GE SiC/SiC and SiC monolithic materials since they were relatively smooth while three green tapes were stacked for use as the interlayer in joining the BFG SiC/SiC material which had a rougher surface. The fixture used to position the substrate s for joining is shown in Figure 1. Joint processing was conducted in a vacuum furnace at 1375°C with a 5 minute hold. A sl ow, stepped heating rate was used to burn-off the organic binders in the eutectic tape. The microstructures of polished cross-sections of the resulting joints were analyzed using an optical microscope and a field emission scanning electron microscope (FE SEM) coupled with energy dispersive spectroscopy (EDS) for elemental analysis of reaction formed phases in the joint.

Figure 1. Image of the processing fixture that was used to join two overlapping substrates to form the crossbar configuration needed for testing according to ISO-13124. Witness samples were also joined for conducting microscopy (sample in the upper right).

The joined crossbars were tested according to the ISO-13124 standard. Figure 2 shows illustrations of the sample loading and of the crossbar configuration in the fixtures during testing for the tensile stress state (top) and the shear stress state (bottom). Fractography using a scanning electron microscope was conducted on fracture surfaces of failed samples. In some cases, failure did not occur at the joint and the joint region remained intact. In these cases, macrographs were taken to capture the failure location.

Figure 2. Illustration of the sample loading (left) and the position of the samples in the test fixtures (right) for the tensile stress state (top) and for the shear stress state...