EELS Study of Oxygen Diffusion in Aluminum Nitride

The oxygen/nitrogen interdiffusion in AIN ceramics was investigated by electron energy-loss spectroscopy (EELS) in combination with transmission electron microscopy (TEM). AL₂O₃/AIN diffusion couples were prepared by oxidation of an AIN ceramic. The samples were annealed for different times in the temperature range from 1500° to 1900°C. The diffusion couples were encapsulated in a tantalum ampoule to guarantee an inert atmosphere. Oxygen concentration profiles across the oxide/nitride interface were measured by EELS. The oxygen/nitrogen interdiffusion coefficient in AIN is given by D = 1.5_–1.2^+6.0× 10^–8 exp[–240 ± 40 (kJ/mol)/( RT )] (cm²/s). The magnitude and temperature dependence of the N/O interdiffusion is comparable to diffusion coefficients reported in the literature for other non-oxide ceramic materials. At standard AIN sintering conditions the O/N interdiffusion is so slow that it cannot provide an effective means for oxygen removal from the AIN grains.     

Measurement of the Oxygen and Impurity Distribution in Polycrystalline Aluminum Nitride with Secondary Ion Mass Spectrometry

The lattice oxygen content in sintered polycrystalline aluminum nitride substrates was measured via secondary ion mass spectrometry (SIMS). This was achieved by quantitative analysis of spatial and depth-resolved ion images of polished specimens using the solid-state standard addition method. The thermal conductivity of the polycrystalline material, measured with the laser flash technique, was found to be strongly correlated with the oxygen content in the AIN grains. This dependence is similar to that observed in single-crystal studies and is consistent with the phonon scattering model of AIN thermal conductivity. Scanning electron microscopy and SIMS images of a variety of other species (C, F, Cl, Y, Si, and Ca) were also obtained. In general, impurities were localized within second-phase regions although calcium was also found to be distributed uniformly along AIN grain boundaries. Other impurity constituents are also discussed.     

稀土氧化物对氮化铝瓷介电性能的影响(英文)

研究了添加Nd2O3和Er2O3对氮化铝陶瓷烧结性能、介电性能和显微结构的影响。结果表明:在氮化铝瓷中添加Nd2O3和Er2O3,有利于降低氮化铝陶瓷烧结温度,提高致密性,并且介电性能能够得到显著改善。添加3%(质量分数)Er2O3的AlN陶瓷的相对密度达98.8%,介电损耗为1.3×10-4,是纯AlN陶瓷的5%。其显微结构分析表明,氮化铝晶粒尺寸更均匀,并且其晶格参数更接近理论值,晶界相较少,从而使得其介电性能得到较大改善。     

Composition and Microstructure of Chemically Vapor-Deposited Boron Nitride, Aluminum Nitride, and Boron Nitride + Aluminum Nitride Composites

The composition and microstructure of dispersed-phase ceramic composites containing BN and AIN as well as BN and AIN single-phase ceramics prepared by chemical vapor deposition have been characterized using X-ray diffraction, scanning electron microscopy, electron microprobe, and transmission electron microscopy techniques. Under certain processing conditions, the codeposited coating microstructure consists of small single-crystal AIN fibers (whiskers) surrounded by a turbostratic BN matrix. Other processing conditions resulted in single-phase films of AIN with a fibrous structure. The compositions of the codeposits range from 2 to 50 mol% BN, 50 to 80 mol% AIN with 7% to 25% oxygen impurity as determined by electron microprobe analysis.     

Composition and Properties of Hot-Pressed SiC-AIN Solid Solutions

High-density SiC-AIN compositions were fabricated from powder mixtures by hot-pressing in the 1700° to 2300°C temperature range. At 2100°C, a 2H solid solution was found from =35 to 100 wt% AlN. The single-phase solid solution samples had steep composition gradients of >10%/μm within the grains. Lattice parameters closely followed Vegard's law. For compositions with <35% AIN, multiphase assemblages were found. Increasing grain size was observed for increasing firing temperature for SiC and AIN. Grain size of the solid solutions was significantly smaller than for SiC or AIN fired to the same temperature. Microhardness values decreased linearly in the solid solution region with increasing AIN content. Flexural strengths of SiC and AIN decreased with increasing firing temperature and increasing grain size. The strengths of SiC, AIN, and the solid solutions were low for materials hot-pressed at 2100°C.     

A1N陶瓷中的晶格缺陷

用透射电镜观察了不同热导率的Ａ１Ｎ陶瓷中存在的晶格缺陷，这类缺陷主要以位错线形式呈现，分布不均匀，大多集中在晶界处。一些晶粒中存在反相畴界。热导率不同的试样其缺陷密度明显不同，氧杂质进入Ａ１Ｎ晶格并形成铝空位是产生晶格缺陷的主要原因，也对晶格参数有显著影响。分析了抑制晶格缺陷形成、提高热导率的工艺措施。     

Characterization of the AIN–W Interface in a Cofired Multilayer AIN Substrate

The AIN–W Interfaces in a cofired multilayer AIN substrate were observed using an optical microscope, scanning electron microscope (SEM) and transmission electron microscope (TEM). Optical and SEM observations showed an intricate intricatelocking AIN-W grain structure at the interface. After the W pad was removed from the substrate with a NaOH etchant, the surface morphology of the W metal at the interface side was found to be very rough, with a small-grain microstructure compared with that at the free surface side. Electron microprobe analyses using SEM revealed that there was no diffusion of either W or Al at the interface at the order of a few micrometer's resolution. Bright-field images, dark-field images and selected area electron diffraction (SAD) patterns using TEM indicated there was no secondary phase between AIN and W. However, scanning transmission electron microscopy using an energy dispersive X-ray detector revealed that there was a 200-nm thick W diffusion layer from the interface into the AIN ceramics. It was concluded that the high adhesion strength between the W conductor and the AIN substrate (>20 MPa) was not due to any secondary phase but to mechanical interlocking of AIN and W during cofiring.     

Formation of β-SiAION as an Intermediate Oxidation Product of SiC–AIN Ceramics

The oxidation mechanisms in SiC-AlN solid solutions, exposed at 1370°C for 200 h in air, have been investigated by scanning electron microscopy (SEM), analytical and high-resolution electron microscopy (HREM), and X-ray diffraction techniques. Five hot-pressed compositions, including pure SiC and pure AIN, were examined. Assuming parabolic rate behavior, the values of the parabolic rate constants based on oxide layer thickness increased by over 2 orders of magnitude by increasing AIN contents from 0% to 62%. A continuous amorphous phase containing cristobalite and mullite crystallites was present at the oxide layer in 13% AIN and 27% AIN samples. Two reaction layers were found in the oxidized SiC–62% AIN solid solution. The inner layer consisted of a mixture of β-SiAION and graphite phases, while the outer layer consisted of mullite as a major phase and cristobalite as a minor phase. The oxidation proceeded in two stages: a partial oxidation of SiC–AIN to β-SiAION and carbon (graphite), and a final oxidation of β-SiAION to mullite + SiO₂.     

Effect of CaO on the Thermal Conductivity of Aluminum Nitride

The effect of CaO on the thermal conductivity of aluminum nitride pressureless sintered with 3 wt% Y₂O₃ as a sintering aid was investigated. Over the composition range of 0 to 2.0 wt% additions, CaO decreased the thermal conductivity of the sintered parts by 10%. CaO doping rendered the secondary oxide phases more wetting and thus with a greater tendency to penetrate along the grain boundaries. Furthermore, CaO segregation to the grain boundaries was observed even on those grain boundaries apparently free of secondary phases. These microstructural changes disrupted the connectivity of the high thermal conductivity AIN grains and were the main factors contributing to the decrease in the thermal conductivity of the ceramic parts. CaO additions to samples doped with SiO₂ had the opposite effect, increasing the thermal conductivity. CaO removed SiO₂ from the AIN grains and incorporated it into the oxide second phases, most likely through charge-compensating substitutions Ca²⁺+ Si⁴⁺ for Y³⁺ and/or Al³⁺. Thus, AIN samples containing both SiO₂ and CaO had higher thermal conductivity than those containing comparable amounts of SiO₂ alone.     

Auger Electron Spectroscopy of Aluminum Nitride Powder Surfaces

The chemical states of powder surfaces depend on the manufacturing processes of the powders. The surface chemistry of three different commercial AIN powders, which are processed by carbothermal nitridation of Al₂O₃, chemical vapor deposition (CVD), and direct nitridation of aluminum, were evaluated by using Auger electron spectroscopy (AES). In order to obtain reference AES spectra of aluminum compounds, α-, γ-, θ-Al₂O₃, γ-AIOOH, γ-AION, and sintered AIN were also examined. Line shapes of aluminum LVV ; Al( LVV ), nitrogen KLL ; N( KLL ) and oxygen KLL ; O( KLL ) are discussed for the AIN powders and all the other aluminum compounds. The differential Auger electron spectra, i.e., E dn /AE were obtained directly, where n is the number of Auger electrons, and E is the kinetic energy of the electron. Their integrated spectra, i.e., n ( E ) are also employed for analysis. The results confirm the conclusions of our previous temperature-programmed desorption work. The AES line shape analysis implies the presence of an oxide-like θ -Al₂O₃ containing AION phase on the carbothermal nitride AIN powder surfaces. The surfaces of CVD and direct-nitrided AIN powders are covered by an oxide–like γ-Al₂O₃ with an oxygen diffusion layer and does not have AION phase.     

Modulated Structures in SiC-AIN Ceramics

Single-phase solid solutions of2H crystal type in the SiC-AIN system containing 50 and 75 mol% AIN were fabricated by hot-pressing mixtures of SiC and AIN powders. Dense samples were subjected to thermal treatments over a range of temperatures between 1600° and 1900°C for a period of time up to 320 h. Transmission electron microscopy revealed the existence of modulated structures indicative of spinodal decomposition.     

Microstructural Effects on the Thermal Conductivity of Polycrystalline Aluminum Nitride

Polycrystalline AIN bodies were made with a range of porosities from various AIN powders by sintering or hot-pressing. Thermal conductivity data for material produced without sintering aids showed a gradual, yet definite, porosity dependence with scatter similar to other property-porosity studies. The thermal conductivity-porosity data for ALN with sintering aids showed the existence of two distinct regions: (1) a higher-porosity region (greater than approximately 2%) which was similar to the data for material without sintering aids, and (2) a low-porosity region where grain boundaries were seen to dominate thermal conductivity. Auger spectroscopy was used to investigate fracture surfaces for dense CaO-doped materials. Thermal conductivity was observed to correlate to effective grain-boundary thickness, which was calculated from quantitative analysis of the Auger data. A model consisting of cubic grains in a continuous grain-boundary phase accurately describes these data.     

Morphological Effect of Second Phase on the Thermal Conductivity of AIN Ceramics

The effect of the morphology of second phases in sintered microstructure on the thermal conductivity of AIN ceramics was investigated. When an Y₂O₃-doped AIN specimen was cooled down slowly at a rate of 3°C/min after sintering at 1850° or 1900°C, the second phases were concentrated in the corners of the AIN grains by an increase of dihedral angle during cooling. On the other hand, the fast-cooled specimen at a rate of 60°C/min showed a different structure of the second phases interconnected through the triple-grain junctions. The specimen with isolated second-phase morphology showed a higher thermal conductivity than those with interconnected second-phase morphology. The measured thermal conductivity of the specimens with different morphologies of the second phases agreed well with the calculated one derived from modeled microstructures. From the comparison of the measured and calculated thermal conductivity, it was shown that the thermal conductivity of the specimen with interconnected second-phase morphology decreased steeply with an increase of the amount of the second phases, assuming the content of lattice oxygen to be constant. However, the thermal conductivity of the specimen with isolated second-phase morphology was rather insensitive to an increase of the amount of the second phases.     

Investigation of Phase Stability in the System Sic-AlN

Dense samples of several compositions in the system SiC-AIN were fabricated by hot-pressing. The SiC-AIN powder was prepared by carbothermal reduction of an intimate mixture of alumina, silica, and carbon in a nitrogen atmosphere. X-ray diffraction and electron and optical microscopy were used to determine the chemical and microstructural characteristics of the hot-pressed specimens. Materials with bulk compositions between 15 and 75 wt% AIN were found to be nonhomo-geneous when hot-pressed below 2100°C. These materials were determined to be a mixture of SiC-AIN solid solutions with different compositions. The observed compositional variations were distinctly bimodal. The source of the in-homogeneity was the starting SiC-AIN powder. The powders, as well as the hot-pressed samples, consisted of a mixture of small crystals rich in SiC and large AIN-rich crystals. Compositions outside the 15 to 75 wt% AIN region were found to be single phase and to have the wurtzite structure. Hot-pressing SiC-AIN in the intermediate composition range at 2300°C produced an optically and chemically homogeneous material. The precipitation of an SiC-rich phase from a 75 wt% AIN solid solution and the precipitation of an AIN-rich phase from a 47 wt% AIN alloy when annealed at 1700°C are strong indications that a miscibility gas exists in the system SiC-AIN.     

TEM of Dislocations in AIN

Sintered AIN specimens were deformed by Vickers hardness (HV) indentations. Compared with Al₂O₃ the HV hardness values indicate a much higher plasticity of AIN at room temperature, but above 600°C a higher ductility for Al₂O₃. Deformed AIN specimens were examined by transmission electron microscopy. Basal and prismatic glide with the slip systems (0001) 〈1120〉 and {1 1 00}〈11 2 0〉 were frequently observed. This results in four linearly independent slip systems. The critical resolved shear stress for single prismatic slip seems to be even smaller than for basal slip. However, thermally activated dislocation reactions are frozen up to at least 1000°C. Thus, prismatic slip is suppressed as soon as more than one slip direction is activated.     

Inversion Domains in Aluminum Nitride

Flat and curved, basal and nonbasal planar defects are frequently observed in sintered AIN ceramics. To prove the existence of inversion domains, TEM techniques sensitive to crystal polarity were used. Dark-field imaging in certain multiple-beam situations with g = (0002) (where g is operating reflection) yielded strong domain contrast giving clear evidence of an inverted crystal structure within the domains. This was confirmed by the dynamic contrast in the disks of convergent-beam electron diffraction. All planar defects observed were found to be associated with inversion domains. The formation of inversion twins may be nucleated by accidental growth or by nucleation and growth within the bulk of a preexisting AIN grain. Both mechanisms are supposed to be associated with the accommodation of impurities.     

Stabilization of Cubic Zirconia by Aluminum Nitride

Cubic ZrO₂ is stabilized at room temperature by the addition of AIN. The percentage of c-ZrO₂ in the mixture of c-ZrO₂ and m-ZrO₂ increases linearly with the addition of up to 20 mol% AIN and decreases thereafter. Stabilization becomes complete at 50 mol% AIN. The lattice spacing of the cubic phase gradually expands up to 20 mol% AIN. A displacement reaction ZrO₂+ AIN → ZrN+Al₂O₃ occurs above 20 mol% AIN and is completed at 50 mol% AIN.     

Phase Equilibria and Phase Transformation in the Aluminum Nitride-Aluminum Oxycarbide Pseudobinary System

Samples in the pseudobinary AIN–AI₂OC were fabricated by hot-pressing mixtures of ANI, AI₄C₃, and AI₂O₃ powders in graphite dies in an atmosphere of nitrogen. The resulting dense samples were subjected to thermal treatments over a range of temperatures from 1550° to 1950°C. The hot-pressed as well as annealed samples were examined using optical microscopy, scanning transmission electron microscopy, and X-ray diffraction. An electron microprobe was used to determine composition. X-ray diffraction showed that AI₂OC dissolved in AIN up to 44 mol% at 1800°C. Thermal treatment at lower temperatures led to the decomposition of the solid solution into two isostructural phases. In samples containing ∼14 to 60 mol% AIN, the morphology of the precipitates was lenticular. Diffraction contrast analysis showed that the precipitates were rich in ANI. Lattice images showed that the (001) planes of the 2H structure were continuous between the matrix and the precipitates. These precipitates appear to be similar to Guinier-Preston zones observed in metallic alloys. When annealed for long periods of time, interface dislocations formed, signifying partial loss of coherency. In some compositions (∼61 to 64 mol% AIN), a lamellar microstructure developed similar to that observed in cellular phase separation. Also, in some of the compositions, an additional phase was observed whose composition and structure were not determined.     

Sintering Behavior of Fine Aluminum Nitride Powder Synthesized from Aluminum Polynuclear Complexes

The sintering behavior of fine AIN powder synthesized from an aluminum polynuclear complex was investigated. The focus of this work was to investigate the densification behavior of the AIN powder with different particle sizes (specific surface area: 3.2–22.8 m²/g). The AIN powder was synthesized from basic aluminum chloride and glucose mixed in a water solution. This powder was divided into two groups: one with 2 wt% Y₂O₃ added as the sintering aid and the other without such an additive. The AIN powder investigated possessed favorable densification potential. The density of the AIN powder with a surface area of 16.6 m²/g and without additives attained theoretical density at 1700°C. Adding Y₂O₃ further decreased the sintering temperature required for full densification to 1600°C. It is speculated that low-temperature sintering of our fine AIN powder with Y₂O₃ proceeds in two steps: in the initial stage, sintering proceeds predominantly through interdiffusion between yttrium aluminates formed on the AIN powder surface; in the second stage, the densification may occur by the interdiffusion between solid phases formed by a reaction between the yttrium aluminates and AIN. To investigate the effect of oxygen on sintering, the content of oxygen in AIN powder was varied while the particle size was kept constant. In this study, the difference in surface oxygen content scarcely affected the sintering behavior of fine AIN powder.     

Thermodynamic and Kinetic Effects of Oxygen Removal on the Thermal Conductivity of Aluminum Nitride

High thermal conductivity, low dielectric constant, high electrical resistivity, low density, and a thermal expansion coefficient that matches well with that of silicon are the principal attributes of AIN that have attracted much attention over the past decade. It is also now well established that oxygen as an impurity lowers the thermal conductivity of AIN. Processing techniques have been developed which not only facilitate pressureless densification of AIN but also enhance its thermal conductivity. The present work explores the thermodynamics and the kinetics of oxygen removal and the resultant enhancement of thermal conductivity. Polycrystalline AIN ceramics were fabricated with Y₂O₃, Dy₂O₃, Yb₂O₃, CaO, BaO, or MgO as additives. Samples were sinter/annealed at 1850°C for up to 1000 min. The AIN grain size of sintered samples ranged between 2 and 9 μm. The samples typically contained two or three phases with the predominant phase being AIN. Secondary phases in Y₂O₃-doped AIN consisted of yttrium aluminates which were along three grain junctions and along grain facets. The presence of Y₃Al₅O₁₂, YAIO₃, and Y₄Al₂O₉, as well as Y₂O₃, depending upon the Y₂O₃/Al₂O₃ ratio, was revealed by X-ray diffraction. Thermal conductivity increased with the amount of additive and annealing time. Thermal conductivity also depended on the type of additive. Samples with thermal conductivity up to 200 W/(m · K) were fabricated. The variation in thermal conductivity with the type and the amount of the additive is explained on the basis of the thermodynamics of oxygen removal. In particular, the higher thermal conductivity of CaO-doped, in comparison with MgO-doped, samples is rationalized on the basis that the free energy of formation, ΔG°, of CaAl₂O₄ is less than that of MgAl₂O₄. It is proposed that the higher the |ΔG°|, with ΔG° < 0, the higher is the resultant thermal conductivity. An increase in the thermal conductivity with annealing time is attributed to the kinetics of oxygen removal from AIN grains.