The high ratio of the Sic grain-boundary energy to the Si(1) / S i c interfacial energy is used to explain two microstructural observations: (1) the penetration of dense Sic by Si(l), and (2) the morphology of the Sic formed when glassy carbon is reacted with Si(1). Comments are made concerning the mechanism of the Si +C reaction and the effect of sintering aids on the Sic grain-boundary energy. 12345HE penetration of molten Si along grain Tboundaries in polycrystalline S i c is shown on Fig. 1. Figure 2 illustrates the noticeable tendency to avoid the formation of S i c grain boundaries in the reaction product layer formed on carbon exposed to molten Si. These phenomena are explained on the basis of the interfacial energies involved in the Si-Sic system. Si-Sic interfacial energiesThe interfacial energy between S i c and molten Si, ys.L, and the S i c grainboundary energy, ys.s, can be estimated from data in the literature on the surface tension of liquid silicon, yL.v, and the surface energy of Sic, ys.v.At 15OO0C, the surface energy of liquid Si is estimated to be 850 mNIm.' Sessile drop experiments with Si on poly-CONTRIBUTING EDITOR-W. J. SMOTHERS Received March 12, 1981; revised copy received Octotler 13, 1981. Member, the American Ceramic Society. crystalline S i c show a contact angle of =30".' The classic Young-Dupre equation relates surface forces under equilibrium conditions to the contact angle:(1) Bruce3 estimates yS.,, for S i c to be -2000 d i m . This fixes ys.L at =I260 mNim.The S i c grain-boundary energy can also be estimated from yS.,, and from knowledge of the sintering behavior of Sic. Without additives or very high pressures, Sic will not sinter, presumably due to unfavorable energetic^.^.' Therefore, the dihedral angle for a pore at a grain boundary in S i c is <60° and: ys-v = ys-L + YL-VCOS 8 ~s.sl2ys.v cos 30"(2) This implies that a conservative estimate of ys.s is -3460 mN/m. Si penetration of sinterable S i cBoron doped sinterable S i c powder was hot-pressed at 1850°C for =1 h at 34.5 MPa to yield a high-density S i c body with an average grain size of <0.3 p m . Since sintering can occur, the grain-boundary Fig. 1. as-polished (left) and Si-affected areas. ( B ) Higher magnification of penetrated area. Penetration of hot-pressed S i c by molten Si (Si removed by acid attack). (A) Comparison of Fig. 2. at 1450°C for lo4 s. Si removed by acid etch. Scanning electron micrographs of S i c layer formed on glassy carbon by reaction with Si(l)must have been reduced to a value energy' < yv-3yS.,. However, if the vapor phase is rep4aced by a liquid phase, the S i c grain boundaries will be replaced by two liquid-solid interfaces if the dihedral angle equals zero or ys.s>2ys.L. Approximately 50 mg of silicon* was melted on a diamond-polished surface of the hot-pressed S i c and held at 1500°C for 15 min. Since the solubility of S i c in molten Si is so small at these temperatures6 (-lo-* at.%), the amount of SIC lost to dissolution would be imperceptible metallographically. After cooling, th...
IN SOL.II)S, the transport of thermal energy by phonon conduction is strongly affected by the degree of lattice perfection. Individual lattice site imperfections, e.g. impurity atoms or specifically alloyed solid solutions, as well as microstructural imperfections, e.g. porosity:' and microcracks,' have been reported to decrease thermal diffusivity . Other lattice defects may also affect significantly the rate of heat transfer via phonon processes. This hypothesis is supported by the present data for thermal diffusivity in nonstoichiometric TiO,.Polycrystalline TiO, disks 4 in. in diameter and ~0 . 5 in. thick were prepared by hot-pressing submicron anatase powder* in graphite dies at 1050°C and 3000 psi for i= 1.5 h. These disks were cut into many smaller specimens whose stoichiometry was adjusted by annealing i n CO/CO, gas mixtures"as described in Ref. 6. These nonstoichiometric specimens were given a homogenization anneal in sealed 96%-silica glass.; tubes for 4 days at 1200°C. As an additional check on the degree of nonstoichiometry, specimens of each batch were reoxidized in air at 1200°C and the stoichiometry verified from the weight gain. Disks for thermal diffusivity measurements were diamond ground to -0.5 in. in diameter and 0. l in. thick.Thermal diffusivities were measured on these small disks by the laser-flash technique.' Temperature was varied using a carbon resistance furnace with an Ar atmosphere.$ The front face of the specimen was coated uniformly with carbon on all sides and subjected to a single flash from a glass-Nd laser through one of the quartz furnace windows. At >200"C, an ir detector measured the transient temperature response of the back face of the specimen; detector output was recorded on an oscilloscope equipped with signal storage. Thermal diffusivity ( a ) was calculated from the specimen thickness ( L ) and the time (tliz) required for the back surface of the specimen to reach half its final value, using a= 1. 38L"l.rr2tl,2 (1)Measurements were made during both heating and cooling between 200" and 80O"C, the upper limit having been chosen to maintain the defect stoichiometry and structure. Figure 1 depicts the variation in a with temperature for the stoichiometries investigated. The trend is that expected on the basis of theory and other experimental results, i.e. a decrease with increasing temperature that is particularly evident at low temperatures and a decrease with increasing defect concentration. Over the range of O/Ti ratios from 2.000 to 1.963, adecreases by a factor of -2, clearly revealing the profound effect of lattice defects on heat transfer in reduced TiO,. A cross-plot of data from Fig. 1 (not shown here) reveals a linear increase in a between O/Ti ratios of 1.963 and 1.995; the increase between 1.995 and 2.000 is more pronounced. This behavior appears to correlate with the defect structures reported for titanium "dioxide," which progress from point defects to ordered planar defects with increasing reduction over the range of stoichiometries investigated. Ande...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.