Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “INTERMETALLIC COMPOUNDS”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 847 records · Page 47

Cu Interlayer-Induced High-Strength Laser-Welded AISI 304 Steel-Niobium Joint

Joining niobium to austenitic stainless steel is necessary for the fabrication of particle accelerators. However, the brittle iron-niobium intermetallics formed in the joint significantly reduce the joint strength. Copper was used as an interlayer to prevent the brittle phase formation while joining niobium to AISI 304 steel. The laser beam was first focused on the Cu-steel faying surface and then on the Cu-Nb faying surface. The tensile strength of the joint is greater than that of the Nb. The weld microstructure consists of austenite with Cu-rich precipitates along with the austenitic grain boundaries and Cu-rich phase embedded with Nb-rich particles and dendrites. Solid solution strengthening of Cu by Fe and second-phase strengthening of Cu by Nb play a critical role in improving the mechanical properties of the joint. These results highlight new possibilities for the joining of niobium to steel.

36 MATERIALS SCIENCE↗

Corrosion Behavior of Fe-7Cr-1Mo-12Zr Metal Waste Form Alloy in Simulated Repository Environment

Metal waste forms (MWF) are solid metallic waste categorized as high-level radioactive waste generated during pyro-reprocessing of spent metallic fuels and geologically disposed off. This waste form generated requires an excellent corrosion resistance in a geological repository environment. In the present work, Fe-7Cr-1Mo-12Zr MWF alloy was developed by casting in an arc melting furnace. The x-ray diffraction analysis revealed the formation of stable phases of α-Fe solid solution and the intermetallic phases of Fe{sub 2}Zr and Fe{sub 23}Zr{sub 6}. The corrosion behavior of Fe-7Cr-1Mo-12Zr alloy was investigated using electrochemical techniques in aggressive (highly acidic with Cl{sup −}) and nominal repository pH environment. Cyclic potentiodynamic polarization studies indicated the shift in corrosion potential toward cathodic potential in highly acidic solution with Cl{sup −}. The corrosion mechanism of pitting is associated with acidified pH medium with more Cl{sup −}, whereas uniform corrosion or transpassive dissolution is resultant of mild acidity or mild alkaline medium. The passive region is wider in pH 5 and 8 when compared to pH 1 and 3. Electrochemical impedance spectroscopic (EIS) analysis revealed the lowering of passive film stability in the highly acidic medium in the presence of Cl{sup −}. The role of passive oxide layers on the corrosion resistance of the Fe-7Cr-1Mo-12Zr MWF alloy is discussed.

36 MATERIALS SCIENCE↗

On the Microstructure and Solidification Behavior of N-Bearing Ti-46Al-8Ta (at.%) Intermetallic Alloys

Ti-46Al-8Ta (at.%) intermetallic alloys exhibit acceptable strength at low and high temperatures with good oxidation resistance. Different alloying elements are usually added to TiAl-based alloys to develop required properties. In this work, effects of N addition up to 2 at.% on microstructural features and solidification behavior of the Ti-46Al-8Ta alloy is investigated. The achieved results showed that alloying with N causes dendrite refinement and reduction of the retained metastable B2 phase in the as-cast microstructure. The dendrite morphology is also changed from fourfold to fivefold and then sixfold symmetry by increasing N content. The effect of N on interlamellar spacing, amount of constituent phases, and lattice parameters was not uniform. Increasing N content resulted in an increase in both small punch maximum load and hardness up to 1.05 at.%, whereas further N alloying, despite increasing of hardness, led to a drop in the maximum load.

36 MATERIALS SCIENCE↗

Processing and Thermoelectric Properties of TiNiSn Materials: A Review

The half-Heusler alloy TiNiSn is a promising material for high-temperature thermoelectric applications due to its good thermal stability and semiconductor-like electrical properties. Numerous synthesis techniques have been utilized to make TiNiSn, but a comparative study on its thermoelectric properties with respect to processing parameters has not been reported. Based on published experimental data, primary melting methods, such as arc melting or induction levitation melting, do not have a noticeable effect on the final thermoelectric properties, although choice of densification technique and annealing parameters correlate with each thermoelectric property. Thermal conductivity (2.47-6.08 W/m K), governed by lattice scattering effects, was maximized with high density from the spark plasma sintering (SPS) technique as well with low inclusions of metallic impurities and interstitial Ni defects from low-temperature, non-densified processes. Electrical resistivity (4.75-30 µΩ m), inversely related to thermal conductivity, is minimized when density is maximized and chemical defects are favorable from using SPS. All reported TiNiSn alloys contain intermetallic impurities, imparting a global trend of lowered electrical resistivity. Seebeck coefficient (− 273.20 to − 50.71 µV/K), proportional to electrical resistivity, is maximized when charge carrier concentration is minimized by eliminating phase impurities and Ni defects. Porosity has been observed to scatter low-energy electrons, which can increase Seebeck coefficient and electrical resistivity, although no statistical significance is found. An ideal TiNiSn alloy should have intermediate values of all interconnected thermoelectric properties to maximize ZT. Future research work should strive to include quantitative phase analysis to better characterize the contributions of Ni defects and impurity phases toward thermoelectric properties.

36 MATERIALS SCIENCE↗

Microstructure Evolution and Mechanical Properties of the AZ91 Magnesium Alloy with Sr and Ti Additions in the As-Cast and As-Aged Conditions

Effect of different amounts of Sr and Ti additions with T6 heat treatment on microstructure and mechanical properties of the AZ91 alloy was investigated. The microstructure of the AZ91 alloy is consisted of the α-Mg phase surrounded by the semicontinuous network of β-Mg{sub 17}Al{sub 12} intermetallics. The average grain size and volume fraction of β phase decreased up to 1 wt.% Sr and 0.1 wt.% Ti additions. Exceeding Ti additions led to increase in both grain size and volume fraction. It was found that the grain refinement mechanism of AZ91 alloys with Sr additions was grain restriction factor. Also, 0.1 wt.% Ti addition could refine the grain structure of the AZ91 alloy by the formation of Al{sub 3}Ti particles that act as nucleation sites for the α-Mg phase. Furthermore, adding 1 wt.% Sr and 0.1 wt.% Ti enhanced the hardness values for both cast and aged AZ91 alloy. Moreover, ultimate tensile strength and tensile elongation values reached their maximum amounts with the 0.5 wt.% Sr and 0.1 wt.% Ti additions and more Sr and Ti additions deteriorated the mechanical properties of cast and aged AZ91 alloys. The fracture surface characterizations had good relations with the tensile properties. The fracture modes were cleavage and quasi-cleavage in the cast and aged AZ91 alloys. Furthermore, the optimum levels of Sr and Ti additions altered the fracture modes to more quasi-cleavage fractures.

36 MATERIALS SCIENCE↗

Improved High-Temperature Aluminum Alloys Containing Cerium

A common rare earth (cerium) when added to aluminum in compositions up to the eutectic compositions of around 10 wt.% improves the high-temperature performance of aluminum alloys. In the early 1980s, some promising research and development efforts focused on powder metallurgy revealed that aluminum alloys containing 4 wt.% cerium exhibit high-temperature mechanical properties exceeding those of the best commercial aluminum casting alloys then in production. Those compositions, which also included high levels of iron, were difficult to process. Recently, magnesium replaces iron to reduce density and improve processing. Cerium oxide is an abundant rare earth oxide that is often discarded during the refining of more valuable rare earths such as Nd and Dy. Therefore, the economics are compelling for cerium as an alloy additive. In this review, we report results obtained during an investigation of the processing and properties of aluminum-cerium alloys produced via casting, extrusion and additive manufacturing. The results show mechanical properties are retained at higher temperatures than other aluminum alloys and show complete recovery of mechanical properties at room temperature when exposed to elevated temperatures as high as 500 °C for 1000 h. Alloys containing cerium also have superior corrosion properties when compared to most aluminum alloys.

36 MATERIALS SCIENCE↗

A Nanoindentation Study of the Plastic Deformation and Fracture Mechanisms in Single-Crystalline CaFe 2 As 2

In this paper, the plastic deformation and fracture mechanisms in single-crystalline CaFe 2 As 2 has been studied using nanoindentation and density functional theory simulations. CaFe 2 As 2 single crystals were grown in a Sn-flux, resulting in homogeneous and nearly defect-free crystals. Nanoindentation along the [001] direction produces strain bursts, radial cracking, and lateral cracking. Ideal cleavage simulations along the [001] and [100] directions using density functional theory calculations revealed that cleavage along the [001] direction requires a much lower stress than cleavage along the [100] direction. This strong anisotropy of cleavage strength implies that CaFe 2 As 2 has an atomic-scale layered structure, which typically exhibits lateral cracking during nanoindentation. This special layered structure results from weak atomic bonding between the (001) Ca and Fe 2 As 2 layers.

36 MATERIALS SCIENCE↗

Combining direct ink writing with reactive melt infiltration to create architectured thermoelectric legs

Here, we present a new additive-reactive synthesis method where inks – cast into molds or 3D-additively extruded into architectured shapes – are reacted into intermetallic thermoelectric compounds. The new method, as demonstrated for equiatomic TiNiSn, combines: (i) extrusion printing (or casting) of inks containing Ni and Ti powders, (ii) debinding and reactive sintering to form a porous NiTi network, (iii) network infiltration with liquid Sn and subsequent reaction to synthesize the TiNiSn phase. Thin plates, created through this method, show high phase purity and low residual porosity. A thermoelectric figure of merit zT = 0.47 ± 0.05 is achieved at 800 K, within the broad range of values (zT= 0.3–1.0) reported for TiNiSn created via traditional powder hot-pressing methods. Layered TiNiSn microlattices – after 3D ink-extrusion printing and infiltration – exhibit high relative densities and minimal undesirable secondary phase content, with a Seebeck coefficient on par with the ink-cast TiNiSn plates, demonstrating that thermoelectric legs with far-ranging architectural freedom can be created additively with this novel reactive manufacturing method. Numerous other thermoelectric (and other) compounds, currently limited to basic geometries due to brittleness, are amenable to this new method based on a sequence of reactive sintering/infiltration of cast or ink-extruded precursors.

36 MATERIALS SCIENCE↗

Effect of Nb and W substitutions on the stability of the A15 Mo3Si phase

A combination of theoretical and experimental studies on the stability of Mo3Si with Nb and W substitutions has been presented in this paper. The first principles calculations suggest that the formation enthalpy of the A15 phase favors Nb substitution up to 37.5 atom %. In case of W substitution, the formation enthalpies increase monotonically. Experimental studies on the other hand, showed that the A15 phase is destabilized when approximately 27.5 atom % Nb is added while 7.5–10 atom % W substitution was sufficient to destabilize A15 phase. It was further determined that upon destabilization of the A15 structure, Nb partitions preferentially to the T1 phase, whereas W partitions preferentially to a solid solution with Mo.

36 MATERIALS SCIENCE↗

Sessile dislocations by reactions in NiAl severely deformed at room temperature

B2 ordered NiAl is known for its poor room temperature (RT) ductility; failure occurs in a brittle like manner even in ductile single crystals deforming by single slip. In the present study NiAl was severely deformed at RT using the method of high pressure torsion (HPT) enabling the hitherto impossible investigation of multiple slip deformation. Methods of transmission electron microscopy were used to analyze the dislocations formed by the plastic deformation showing that as expected dislocations with Burgers vector a(100) carry the plasticity during HPT deformation at RT. In addition, we observe that they often form a(110) dislocations by dislocation reactions; the a(110) dislocations are considered to be sessile based on calculations found in the literature. It is therefore concluded that the frequently encountered 3D dislocation networks containing sessile a(110) dislocations are pinned and lead to deformation-induced embrittlement. In spite of the severe deformation, the chemical order remains unchanged.

36 MATERIALS SCIENCE↗

The Heusler phases LiRh{sub 2}Si and LiRh{sub 2}Ge: Synthesis, structure and properties

The isostructural Heusler phases LiRh{sub 2}Si and LiRh{sub 2}Ge have been synthesized from the elements and an excess of lithium at 1000 deg. C. Both materials adopt the CuMn{sub 2}Al crystal structure, space group Fm-3m (No. 225) with the room temperature lattice parameter a=5.747(1) A [Vol=189.866(1) A{sup 3}] and a=5.847(1) A [Vol=199.88(6) A{sup 3}] for LiRh{sub 2}Si and LiRh{sub 2}Ge, respectively. X-ray analyses suggest mixed site occupancy of the form Li{sub 1-x}Rh{sub 2}Si{sub 1+x} (x<0.4), but not for LiRh{sub 2}Ge. Both materials are diamagnetic, {chi}{sub mol}(LiRh{sub 2}Si)=-6x10{sup -5} cm{sup 3}(mole){sup -1} and {chi}{sub mol}(LiRh{sub 2}Ge)=-10x10{sup -5} cm{sup 3}(mole){sup -1} and metallic with room temperature resistivities of approximately 19 and 32 {mu}{omega} cm, respectively. These properties are consistent with the calculated electronic structure. - Graphical abstract: Rh, Li and either Si or Ge form an ordered coloring pattern of the bcc lattice in the Heusler phases LiRh{sub 2}Si and LiRh{sub 2}Ge. Crystals grown from a reactive Li flux are diamagnetic metals that exhibit a T-linear resistivity above 60 K.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Crystal structure and magnetic properties of Ce{sub 7}Ni{sub 5{+-}x}Ge{sub 3{+-}x}In{sub 6} and Pr{sub 7}Ni{sub 5{+-}x}Ge{sub 3{+-}x}In{sub 6}

The indides Ce{sub 7}Ni{sub 5{+-}x}Ge{sub 3{+-}x}In{sub 6} and Pr{sub 7}Ni{sub 5{+-}x}Ge{sub 3{+-}x}In{sub 6} were synthesized from the elements by arc-melting of the components. Single crystals were grown via special annealing sequences. Both structures were solved from X-ray single crystal diffraction data: new structure type, P6/m, Z=1, a=11.385(2), c=4.212(1) A, wR{sub 2}=0.0640, 634F{sup 2} values, 25 variables for Ce{sub 7}Ni{sub 4.73}Ge{sub 3.27}In{sub 6} and a=11.355(6), c=4.183(2) A, wR{sub 2}=0.0539, 563F{sup 2} values, 25 variables for Pr{sub 7}Ni{sub 4.96}Ge{sub 3.04}In{sub 6}. Both indides show homogeneity ranges through Ni/Ge mixing (M sites). This new structure type can be derived from the AlB{sub 2} structure type by a substitution of the Al and B atoms by CeM{sub 12} and NiIn{sub 6}Ce{sub 3} polyhedra (tricapped trigonal prism). Magnetic susceptibility measurements on a polycrystalline sample of Ce{sub 7}Ni{sub 5}Ge{sub 3}In{sub 6} indicated Curie-Weiss like paramagnetic behavior down to 1.71 K with the effective magnetic moment slightly reduced in relation to the value expected for trivalent cerium ions. No magnetic ordering is evident. -- Graphical Abstract: Packing of the polyhedra in the Ce{sub 7}Ni{sub 4.73}Ge{sub 3.27}In{sub 6} structure. Display Omitted

36 MATERIALS SCIENCE↗

{sup 45}Sc Solid State NMR studies of the silicides ScTSi (T=Co, Ni, Cu, Ru, Rh, Pd, Ir, Pt)

The silicides ScTSi (T=Fe, Co, Ni, Cu, Ru, Rh, Pd, Ir, Pt) were synthesized by arc-melting and characterized by X-ray powder diffraction. The structures of ScCoSi, ScRuSi, ScPdSi, and ScIrSi were refined from single crystal diffractometer data. These silicides crystallize with the TiNiSi type, space group Pnma. No systematic influences of the {sup 45}Sc isotropic magnetic shift and nuclear electric quadrupolar coupling parameters on various structural distortion parameters calculated from the crystal structure data can be detected. {sup 45}Sc MAS-NMR data suggest systematic trends in the local electronic structure probed by the scandium atoms: both the electric field gradients and the isotropic magnetic shifts relative to a 0.2 M aqueous Sc(NO{sub 3}){sub 3} solution decrease with increasing valence electron concentration and within each T group the isotropic magnetic shift decreases monotonically with increasing atomic number. The {sup 45}Sc nuclear electric quadrupolar coupling constants are generally well reproduced by quantum mechanical electric field gradient calculations using the WIEN2k code. Highlights: Black-Right-Pointing-Pointer Arc-melting synthesis of silicides ScTSi. Black-Right-Pointing-Pointer Single crystal X-ray data of ScCoSi, ScRuSi, ScPdSi, and ScIrSi. Black-Right-Pointing-Pointer {sup 45}Sc solid state NMR of silicides ScTSi.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Intermetallic solid solution Fe{sub 1-x}Co{sub x}Ga{sub 3}: Synthesis, structure, NQR study and electronic band structure calculations

Unlimited solid solution Fe{sub 1-x}Co{sub x}Ga{sub 3} was prepared from Ga flux. Its crystal structure was refined for Fe{sub 0.5}Co{sub 0.5}Ga{sub 3} (P4{sub 2}/mnm, a=6.2436(9), c=6.4654(13), Z=4) and showed no ordering of the metal atoms. A combination of the electronic band structure calculations within the density functional theory (DFT) approach and {sup 69,71}Ga nuclear quadrupole resonance (NQR) spectroscopy clearly shows that the Fe-Fe and Co-Co dumbbells are preferred to the Fe-Co dumbbells in the crystals structure. The band structure features a band gap of about 0.4 eV, with the Fermi level crossing peaks of a substantial density of electronic states above the gap for x>0. The solid solution is metallic for x>0.025. The study of the nuclear spin-lattice relaxation shows that the rate of the relaxation, 1/T{sub 1}, is very sensitive to the Co concentration and correlates well with the square of the density of states at the Fermi level, N{sup 2}(E{sub F}). - Graphical abstract: Rate of the nuclear spin-lattice relaxation, 1/T{sub 1}, observed in the {sup 69}Ga NQR experiments for the intermetallic solid solution Fe{sub 1-x}Co{sub x}Ga{sub 3} is the highest for x=0.25 with the highest calculated density of electronic states at the Fermi level, N(E{sub F}); in general, 1/T{sub 1} correlates with N{sup 2}(E{sub F}). Highlights: Black-Right-Pointing-Pointer Fe{sub 1-x}Co{sub x}Ga{sub 3} solid solution is prepared in single crystalline form from Ga flux. Black-Right-Pointing-Pointer In the crystal structure Fe-Fe and Co-Co dumbbells are preferred to Fe-Co dumbbells. Black-Right-Pointing-Pointer Metal-to-semiconductor transition occurs at 0<x<0.025. Black-Right-Pointing-Pointer {sup 69,71}Ga NQR spectra efficiently probe local environment of two independent Ga atoms. Black-Right-Pointing-Pointer Rate of nuclear spin-lattice relaxation follows squared DOS at the Fermi level.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Magnetic structures in RNi₄B (R=Nd, Tb, Ho, Er)

Neutron diffraction has been performed on RNi₄B (R=Nd, Tb, Ho, Er) polycrystals. The orthorhombic structure for NdNi₄B and the CeCo₄B structure type (hexagonal) for TbNi₄B and HoNi₄B are confirmed. Our data also show that this last structure is currently the best approximant for ErNi₄B. The RNi₄B (R=Nd, Tb, Ho, Er) order ferromagnetically at respectively 11.0, 18.1, 6.2 and 10.0 K. The crystal electric field (CEF) interaction controls the magnetic anisotropy in this series leading to an easy axis ~30 deg above the basal plane in RNi₄B (R=Nd, Tb, Ho) and parallel to the c-axis in ErNi₄B at 1.6 K. The RNi₄B (R=Nd, Tb, Ho) display a spin re-orientation below T c which arises from a competition between the second order term and the higher order terms of the CEF hamiltonian. - Graphical abstract: Simplified magnetic structure in NdNi₄B and full magnetic structure in RNi₄B (R=Tb, Ho). Variation with temperature of the easy magnetization axis direction (angle with c-axis) showing a spin re-orientation. Highlights: • RNi₄B (R=Nd, Tb, Ho, Er) polycrystals were inductively melted. • Neutron diffraction confirms they all order ferromagnetically. • The magnetization easy axis is parallel to the c-axis in ErNi₄B. • The RNi₄B (R=Nd, Tb, Ho) display a spin re-orientation below T c . • Their easy axis tilts from the c-axis at T c to ~30 deg above the basal plane at 1.5 K.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Pronounced matrix effect in YbMo{sub 2}Al{sub 4}-type Ca(Au{sub x}Zn{sub 2−x})Au{sub 4} (x=0.09–0.89)

Electron-poor polar intermetallics Ca(Au{sub x}Zn{sub 2−x})Au{sub 4} have been synthesized through fusion of stoichiometric metals in sealed tantalum tubes at 800 °C and annealing at 400 °C for one week. Single crystal X-ray diffraction analyses reveal that this phase belongs to the YbMo{sub 2}Al{sub 4}-type structure (I4/mmm, Pearson symbol tI14), a≈6.943–7.017 Å, c≈5.278–5.286 Å, Z=2, with homogeneous composition range of x=0.09(1)–0.89(1). The structure exhibits a three-dimensional framework of (Au{sub 8}){sub 1/2} featuring square and octagonal channels extending in c, in which Ca and the infinite linear chains of [(Au,Zn){sub 2}]{sub 1/2} are located, respectively. Mulliken population analyses demonstrate that Zn prefers to form the linear chains in the whole homogeneous composition range, consistent with experimental observations. Crystal orbital Hamilton population (COHP) analyses reveal that the channel-to-chain Au–Zn contact has strong bonding interactions regardless of its large interatomic distance (∼2.85 Å), a signature of pronounced matrix effect. The last mentioned effect in YbMo{sub 2}Al{sub 4}-type structures is expected in case the linear chains are defined by small size atoms. - Graphical abstract: Pronounced matrix effect incurred by Au–Au bonding within the gold substructure in Ca(Au{sub x}Zn{sub 2−x})Au{sub 4} results in an elongation of the channel-to-chain Au–Zn interatomic distance without weakening bonding interactions. - Highlights: • The complete solid solution of Ca(Au{sub x}Zn{sub 2−x})Au{sub 4} was accomplished. • Site preference was explained by “coloring” analyses. • Abnormally large Au–Zn distance but with strong bonding was observed. • Pronounced matrix effect incurred by Au–Au bonding in gold substructure was found.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Polycrystalline Mg{sub 2}Si thin films: A theoretical investigation of their electronic transport properties

The electronic structures and thermoelectric properties of a polycrystalline Mg{sub 2}Si thin film have been investigated by first-principle density-functional theory (DFT) and Boltzmann transport theory calculations within the constant-relaxation time approximation. The polycrystalline thin film has been simulated by assembling three types of slabs each having the orientation (001), (110) or (111) with a thickness of about 18 Å. The effect of applying the relaxation procedure to the thin film induces disorder in the structure that has been ascertained by calculating radial distribution functions. For the calculations of the thermoelectric properties, the energy gap has been fixed at the experimental value of 0.74 eV. The thermoelectric properties, namely the Seebeck coefficient, the electrical conductivity and the power factor, have been determined at three temperatures of 350 K, 600 K and 900 K with respect to both the energy levels and the p-type and n-type doping levels. The best Seebeck coefficient is obtained at 350 K: the S{sub yy} component of the tensor amounts to about ±1000 μV K{sup −1}, depending on the type of charge carriers. However, the electrical conductivity is much too small which results in low values of the figure of merit ZT. Structure–property relationship correlations based on directional radial distribution functions allow us to tentatively draw some explanations regarding the anisotropy of the electrical conductivity. Finally, the low ZT values obtained for the polycrystalline Mg{sub 2}Si thin film are paralleled with those recently reported in the literature for bulk chalcogenide glasses. - Graphical abstract: Structure of the polycrystalline thin film of Mg{sub 2}Si. - Author-Highlights: • Polycrystalline Mg{sub 2}Si film has been modelled by DFT approach. • Thermoelectric properties have been evaluated by semi-classical Boltzmann theory. • The structure was found to be slightly disordered after relaxation. • The highest value of Seebeck coefficient reaches −1000 μV/K at 350 K. • The ZT value is estimated between 0.0175 at 350 K and 0.2 at 900 K.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗