Crystallization Front Research Articles

A Rayleigh model of cesium fractionation in granite-pegmatite systems

Abstract The K/Cs ratios of K-feldspars from granitic pegmatites are compared to models derived from the Rayleigh equation. The K/Cs and K/Rb ratios of K-feldspars and micas exhibit decreasing values when plotted against their Cs or Rb contents across cogenetic suites of granites and pegmatites and from margin to core of individual bodies. The trends in elemental ratios conform to Rayleigh fractionation for the crystallization of feldspars and micas from a silicate melt. Within two individual pegmatite bodies, the K/Cs ratio of K-feldspar initially falls more rapidly than the Rayleigh model predicts. That might reflect a local increase in the concentration of Cs relative to K due to the pile-up of incompatible elements in a boundary layer of melt adjacent to the crystal growth front. The addition of an aqueous solution to the Rayleigh model (i.e., the simultaneous crystallization of K-feldspars from melt and from aqueous solution) predicts high and increasing K/Cs ratios of K-feldspars that are not observed in natural rock suites, except when K-feldspars crystallize in miarolitic cavities or when primary K-feldspar recrystallizes to microcline perthite in an open hydrothermal system. In those cases, the Cs content of K-feldspars falls to nil because of the high solubility of Cs in aqueous solution and low compatibility of Cs in K-feldspar. Otherwise, the observed patterns of K/Rb or K/Cs in K-feldspar and micas in pegmatites conform to crystal-melt fractionation in which an aqueous solution plays no part. From the viewpoint of the geochemistry of Cs in pegmatites, these observations give support to the model proposed by Cameron et al. (1949) and endorsed by Jahns (1953a, 1953b).

Investigation of Chemical Heterogeneity of Cast Metal Ingots of Different Geometries and Assessment of their Applicability for Producing Round Rolled Products

Based on the study results, it is shown that the selected ingot geometry makes it possible to obtain a solidified metal structure with a favorable distribution of physical and chemical heterogeneities of the cast metal. The temperature profile calculations have shown that 1 h 45 min after the onset of solidification of a big-end-down ingot, the advancing crystallization front closes, and an extended “closed” zone of liquid metal is formed, the height of which reaches 37% of the height of the ingot body. The provided data correlated well with the actual shrinkage defects, the length of which constituted 36% of the height of the ingot body. The comparison of the temperature distribution profiles with the assessment of steel contamination with non-metallic inclusions and specifics of segregating impurity distribution has revealed a more uniform distribution of impurities and nonmetallic inclusions due to intensive advancement of the solid phase under the hot-top part of the ingot. In this case, the optimized thermal performance of the hot top helps move segregating impurities and tertiary inclusions to the ingot head. The assessment of applicability of the big-end-up and big-end-down ingots has demonstrated the feasibility of using the former to produce round rolled products with a diameter in excess of 250 mm. The big-end-down ingot weighing 5.9 ton is feasible to use for the production of rolled products having a diameter not exceeding 200 mm due to the formation of a closed secondary pipe cavity. The use of a low hot-top height big-end-down ingot allows increasing the yield of the usable metal, while improving the quality of the rolled products.

Analysis of hydrogen behavior by crystallization of secondary aluminum alloy

Methodology. There was conducted the process modeling of hydrogen by crystallization of secondary aluminum alloy. Findings. There was conducted an analysis of hydrogen behavior by indurating secondary aluminum alloy castings. There were obtained expressions for calculation of the current value in the process of secondary aluminum alloy crystallization of the hydrogen density in the boundary diffused layer, on the crystallization front, in the volume of the remaining melt, and the effective hydrogen distribution coefficient. Originality. For the first time, there was obtained an arithmetic model for the forecasting of hydrogen behavior by indurating secondary aluminum alloy castings. Practical value. The results of the study can be implemented with the purpose of the flow optimization process and the more efficient use of expensive equipment for getting high-quality secondary aluminum alloy castings. Keywords: aluminum, secondary aluminum alloys, indurating, hydrogen, crystallization front, distribution coefficient, boundary diffusion layer.

Textural and Geochemical Evidence for Multiple, Sheet-like Magma Pulses in the Limeira Intrusion, Paraná Magmatic Province, Brazil

Abstract Quantitative petrographic, structural, and textural parameters are integrated with geological, geochemical, and Sr-isotope data to examine the emplacement, growth processes, and magmatic evolution of the high-Ti tholeiitic Limeira Intrusion, in the Paraná Magmatic Province of SE Brazil. Our data strongly support a multiple-stage evolution, owing to the nested emplacement of distinct crystal-bearing magma pulses that probably evolved independently, except at their boundaries. A stage of cooling and crystallization between magma injections results in a stepwise T–t path, leading to variations in the plagioclase residence times and effective growth rates inwards, also occasioning sudden changes in crystal shape and size at the boundaries of each magma pulse. The time delay between pulses allows preservation of internal ‘chilled margins’ and the development of near-rigid surfaces at their contacts, increasing the alignment and clustering of crystals during magma replenishment. Isotopic and textural data demonstrate a complex assembly history, in which the appearance of mixed plagioclase populations between magma pulses coincides with the onset of initial Sr isotope ratio increase, which can be attributed to a locally enhanced cooling rate, and the extraction of residual melts from the previous crystallizing batches and mixing with the younger pulses. Typical C- and S-shaped MgO (wt%) compositional profiles within individual pulses indicate that the first one probably evolved by in situ fractional crystallization followed by melt migration inward, whereas the younger ones have contributions from both compaction of the lowermost crystallization front and compositional convection. Mafic globular structures are found at the boundaries of magma pulses and constituting the mafic-rich layers in layered rocks. They are interpreted as evidence for chemical disequilibrium, arguably associated with the trigger of silicate liquid immiscibility. The upwards compositional convection of the silica-rich residual liquid and the accumulation of the Fe–Ti–P-rich crystalbearing end-member in the bottom of the latest magma pulses might represent the most significant mechanism of differentiation in the Limeira Intrusion.

Zones of PGE–Chromite Mineralization in Relation to Crystallization of the Pados-Tundra Ultramafic Complex, Serpentinite Belt, Kola Peninsula, Russia

The lopolithic Pados-Tundra layered complex, the largest member of the Serpentinite belt–Tulppio belt (SB–TB) megastructure in the Fennoscandian Shield, is characterized by (1) highly magnesian compositions of comagmatic dunite–harzburgite–orthopyroxenite, with primitive levels of high-field-strength elements; (2) maximum values of Mg# in olivine (Ol, 93.3) and chromian spinel (Chr, 57.0) in the Dunite block (DB), which exceed those in Ol (91.7) and Chr (42.5) in the sills at Chapesvara, and (3) the presence of major contact-style chromite–IPGE-enriched zones hosted by the DB. A single batch of primitive, Al-undepleted komatiitic magma crystallized normally as dunite close to the outer contact, then toward the center. A similar magma gave rise to Chapesvara and other suites of the SB–TB megastructure. Crystallization proceeded from the early Ol + Chr cumulates to the later Ol–Opx and Opx cumulates with accessory Chr in the Orthopyroxenite zone. The accumulation of Chr resulted from efficient cooling along boundaries of the Dunite block. The inferred front of crystallization advanced along a path traced by vectors of Ol and Chr compositions. Grains and aggregates of Chr were mainly deposited early after the massive crystallization of olivine. Chromium, Al, Zn and H2O, all incompatible in Ol, accumulated to produce podiform segregations or veins of chromitites. This occurred episodically along the moving front of crystallization. Crystallization occurred rapidly owing to heat loss at the contact and to a shallow level of emplacement. The Chr layers are not continuous but rather heterogeneously distributed pods or veins of Chr–Ol–clinochlore segregations. Isolated portions of melt enriched in H2O and ore constituents accumulated during crystallization of Ol. Levels of fO2 in the melt and, consequently, the content of ferric iron in Chr, increased progressively, as in other intrusions of the SB–TB megastructure. The komatiitic magma vesiculated intensely, which led to a progressive loss of H2 and buildup in fO2. In turn, this led to the appearance of anomalous Chr–Ilm parageneses. Diffuse rims of Chr grains, abundant in the DB, contain elevated levels of Fe3+ and enrichments in Ni and Mn. In contrast, Zn is preferentially partitioned into the core, leading to a decoupling of Zn from Mn, also known at Chapesvara. The sulfide species display a pronounced Ni-(Co) enrichment in assemblages of cobaltiferous pentlandite, millerite (and heazlewoodite at Khanlauta), deposited at ≤630 °C. The oxidizing conditions have promoted the formation of sulfoselenide phases of Ru in the chromitites. The attainment of high degrees of oxidation during crystallization of a primitive parental komatiitic magma accounts for the key characteristics of Pados-Tundra and related suites of the SB–TB megastructure.

Modeling polymer crystallisation induced by a moving heat sink.

Recent experimental work has shown that polymer crystallisation can be used to "move" and organize nanoparticles (NP). As a first effort at modeling this situation, we consider the classical Stefan problem but with the modification that polymer crystallisation does not occur at a single temperature. Rather, the rate of crystallisation is proportional to its subcooling, and here we employ a form inspired by the classical Avrami model to describe this functional form. Our results for the movement of the polymer crystallisation front, as defined as the point where the crystallinity is 50%, closely track the results of the classical Stefan problem. Thus, at this level of approximation, the crystallisation kinetics of the polymer do not cause qualitative changes to the physics of this situation. Inspired by this fact we study the more interesting situation where the directional recrystallisation of a polymer melt is considered, e.g., through the application of a moving heat sink over an initially molten polymer, reminiscent of a processing technique termed zone annealing. The polymer crystallisation shows that a steady state exists for a range of sink velocities. The solid-melt interface moves slightly ahead of the sink but at the same velocity. The steady-state distance between the sink and the interface decreases with increasing sink velocity - this is a consequence of the excess cooling provided by the sink over what is required to crystallise the melt. The most interesting new result is that the temperature of the crystal-melt interface decreases with increasing sink velocity. This is in line with the ansatz of Lovinger and Gryte who suggested that larger zone annealing velocities correspond to progressively larger effective undercoolings at which polymer crystallisation occurs.

Boundary layer description of directional polymer crystallisation.

Nearly fifty years ago Lovinger and Gryte suggested that the directional crystallization of a polymer was analogous to the quiescent isothermal crystallization experiment but at a supercooling where the crystal growth velocity was equal to the velocity of the moving front. Our experiments showed that this equivalence holds in a detailed manner at low directional velocities. To understand the underlying physics of these situations, we modeled the motion of a crystallization front in a liquid where the left side boundary is suddenly lowered below the melting point (Stefan's problem) but with the modification that the crystallization kinetics follow a version of the Avrami model. Our numerical results surprisingly showed that the results of the polymer analog track with the Stefan results which were derived for a simple liquid that crystallizes completely at its melting point; in particular, the position of the crystal growth-front evolved with time exactly as in the Stefan problem. The numerical solution also showed that the temperature in the immediate vicinity of the growth-front decreased with increasing front velocity, which is in line with Lovinger and Gryte's ansatz. To provide a clear theoretical understanding of these numerical results we derive a boundary layer solution to the governing coupled differential equations of the polymer problem. The analytical results are in agreement with our observations from experiments and numerical computations but show that this equivalence between the small molecule and polymer analog only holds in the limit where the crystallization enthalpy is much larger than the rate at which heat is conducted away in the polymer. In particular, in the context of the temperature profile, the enthalpy generated by the crystallisation process which is spread out over a narrow spatial region can be approximated as a point source whose location and temperature correspond to the Lovinger-Gryte ansatz.

Расчет динамики границы аморфная фаза-кристалл при твердофазной взрывной кристаллизации

The nonlinear differential equation described a dynamics of solid-phase explosive crystallization front in a much larger parameters domain in comparison with the theoretical results available in literature was obtained. The features of the self-oscillating mode transition of the front motion to the mode of its self-propagation with a constant velocity was numerically studied in detail.

Generating liquid metal flow in modeling the metallurgical processes

In this work, methods of generating a flow of liquid metal using electromagnetic forces are studied, as well as the flows and the accompanying processes. The processes under consideration arise when modeling metallurgical technologies using setup developed for these purposes. The dependence of the shape of the crystallization front on the type of the traveling magnetic field in a rectangular cell is obtained. The dependence of the pressure drop on the flow rate of the liquid metal created by the conduction mechanism in a rectangular channel are found. The dependence of the productivity of a traveling field pump with a coaxial cylindrical channel on the frequency of the supply current are obtained.

STUDY OF PESO - PHO Y OF PESO - PHOTORESISTIVE CH ORESISTIVE CHARACTERISTICS OF TERISTICS OF TlInSe2 SINGLE CRYSTALS

Objective. The aim of the study is to establish the patterns and mechanisms of the effect of various impurities on the tensoresistive characteristics of TlInSe2 single crystals, as well as in solid solutions based on it. When studying the piezophotoresistive characteristics of TlInSe2 single crystals, it will determine the nature of the effect of uniaxial elastic deformation on the electrical, photoelectric and tensoresistive properties of TlInSe2 crystals. Methods. Crystals synthesized by fusing the components in accordance with stoichiometry in evacuated (~ 10-4 mm Hg) and sealed quartz ampoules were used for the study. Highly pure elements thallium (Tl - 000), indium (In - 000) and selenium (Se - high purity grade - 17 - 4) were used as initial components for the synthesis. Single crystals were grown by the improved Bridgman method, the crystallization front velocity varied from 0.5 to 0.9 mm / h. The samples required for the study were prepared by cleaving the grown single crystals from an ingot by the simplest indentation with a sharp knife, the blade thickness of which is ≤ 0.01 mm, along two mutually perpendicular planes of natural cleavage.

Analytical Evaluation of the Dendritic Structure Parameters and Crystallization Rate of Laser-Deposited Cu-Fe Functionally Graded Materials.

The paper is devoted to the direct energy deposition (DED) of functionally graded materials (FGMs) created from stainless steel and aluminum bronze with 10% content of Al and 1% of Fe. The results of the microstructure analysis using scanning electronic microscopy (SEM) demonstrate the existence of a dendritic structure in the specimens. The crystallization rate of the gradient binary Cu-Fe system structures was investigated and calculated using the model of a fast-moving concentrated source with an ellipsoid crystallization front. The width of the secondary elements of the dendrites in the crystallized slab was numerically estimated as 0.2 nm at the center point of the circle heat spot, and the two types of dendrites were predicted in the specimen: the dendrites from 0.2 to approximately 50 nm and from approximately 0.1 to 0.3 μm in width of the secondary elements. The results were found to be in good accordance with the measured experimental values of the dendritic structure geometry parameters.

Lithium Planar Deposition vs Whisker Growth: Crucial Role of Surface Diffusion.

Lithium plating-one of the critical processes in the desired high-energy lithium metal batteries-is accompanied by lithium whisker growth, which causes several problems that prevent the employment of metallic lithium anodes in rechargeable systems. They include low coulombic efficiency, electrolyte consumption, and the risk of short circuits, which can lead to thermal runaway of the battery. In recent years several strategies were suggested to mitigate whisker growth. The mechanism of this process, however, still lacks understanding. Here, we reveal the importance of surface diffusion along grain boundaries in solid lithium. We show that, at first, the plating of lithium onto a lithium substrate is possible as bulk crystal growth with a planar crystallization front for the Li grains with oblique (nonperpendicular to the surface) grain boundaries. Further, the developed compressive stress makes lithium diffusion to the grain base unfavorable, and new grains nucleate at the surface. The latter are the cause of the whisker growth from their roots. These findings indicate that the control of grain-boundary diffusion and grain size and structure paves the way to overcome the nonuniform morphology of plated lithium.

Хімічна і фізична неоднорідності та гази у великому сталевому зливку

For creation of the high-tech equipment that is used in energy, heavy engineering, chemistry and transport, the unique large-sized steel products are required. In the manufacture of such products, large forging ingots in the mass to 600 tons are used. However, an increase in the mass of the ingots leads to the formation of chemical and physical heterogeneity, enlargement and unfavorable distribution of non-metallic inclusions, of the development of segregation defects in them, which reduce the strength and exploitation characteristics of the metal. In this connection, the quality forgings and finished parts are not always meet the producing demands and the loss of metal, in the form of technological waste and rejects are reaching significant values. It is known that eccentric zonal segregation, especially it’s the most dangerous variety - cords, significantly reduce the quality and properties of products from large steel ingots. In connection with the continuous expansion of the production of large ingots, the problem of creating optimal technologies for their formation, which reduce or exclude the possibility of the formation of chemical heterogeneity and cords in steel during crystallization, it is currently important and relevant. In this paper it are presented the results of studies of the structure, gas distribution, physical and chemical heterogeneities in the cross section and height of an ingot in the mass of 140 tons, which was casted in vacuum from steel 25KHN3MFA. It is shown that depending on the temperature and time conditions of ingot hardening, among which the crystallization interval (due to the chemical composition of steels), cooling intensity in different volumes in height and cross section of ingot, temperature gradient before the crystallization front, solubility of alloying elements and gas content in the melt, etc. Based on this, when developing technology for large ingots to ensure their quality, optimal structure and properties should take into account not only their dimensions, but also the combination of these thermokinetic parameters on the crystallization process, dendritic structure formation, manifestations of liquation in different ingot volumes. Keywords: ingot, segregation strip and inclusions, dendrites, structure, oxygen, oxides, sulfides.

In situ visualization of crystal nucleation and growth behaviors of linear and long chain branched polypropylene under shear and CO2 pressure

Crystallization and foaming behaviors of a semi-crystalline polymer in conditions comparable to those found in polymer processing, where the polymer melt experiences shear under elevated pressures, are key for modeling polymer processes and predicting the final structure and mechanical properties of polymer products. We investigate the crystallization behaviors of two different high-melt-strength polypropylene (PP) resins using a novel high-pressure visualization system that was developed to capture the crystallization process of a plastic specimen under high pressure and controlled shear. In situ visualization results under isothermal conditions demonstrate that both linear and long chain branched (LCB) PP exhibit crystal growth rates and nucleation densities that increase with higher applied shear stress but reach an optimum value with increasing CO2 pressure. By selectively choosing processing conditions, we can enhance the formation of observable critical size nuclei and the exclusion of CO2 from the crystal growth front. Therefore, overall crystallization kinetics can be easily controlled through the effect of induced-shear stress and the presence of pressurized CO2.

Atomistic simulation of the generation of vacancies in rapid crystallization of metals

The generation of vacancies at a crystal-liquid interface propagating under conditions of undercooling below the equilibrium melting temperature is investigated in molecular dynamics simulations performed under well-controlled temperature and pressure conditions for two representative metals, bcc Cr and fcc Ni. The results of the simulations reveal that, for both metals, vacancy concentrations produced in the course of the steady-state propagation of crystallization fronts can exceed the equilibrium values at the corresponding temperatures by orders of magnitude. Different trends in the temperature dependences of vacancy concentration observed for the two metals, namely, the continuous increase with increasing undercooling in Ni and a nonmonotonous temperature dependence in Cr, are related to the qualitatively different temperature dependences of the crystallization front velocity predicted for the two metals. The general character of the computational predictions is confirmed in simulations performed for Ni with four different interatomic potentials and for both metals with (001), (011), and (111) interface orientations. A detailed analysis of atomic rearrangements at the crystallization front suggests that the level of vacancy supersaturation is largely defined by the ability of atoms to migrate within the interfacial region and to fill the numerous vacant sites produced through the simultaneous construction of several atomic crystal planes within the interfacial region. While the majority of the transient vacancies generated during the construction of the crystal planes are annihilated by the atomic flux coming from the liquid side of the interface, a small fraction of the vacancies are trapped behind the crystallization front. Under conditions of strong undercooling, the fast movement of the interface and decreased vacancy mobility prevent the equilibration of vacancy concentration in the newly built crystalline region, thus creating a strong vacancy supersaturation. Analysis of the temperature dependence of the atomic rearrangements occurring at the crystallization front suggests incomplete relaxation of the disordered phase at and in front of the crystal-liquid interface rapidly advancing under conditions of strong undercooling.

Electron beam-induced crystallization of Al2O3 gate layer on β-Ga2O3 MOS capacitors

Electron irradiation was observed to induce crystallization of amorphous Al2O3 films grown by atomic layer deposition on β-Ga2O3 substrates. Growth of large, strongly oriented crystalline γ-Al2O3 regions was induced using conventional-mode transmission electron microscopy (TEM) and observed to propagate outward from the interface as well as from the previously crystallized Al2O3. A few nm of epitaxial Al2O3 was already visible at the beginning of the crystallization front propagation. The phenomenon is not explained by electron beam-induced heating, which amounted to less than 1 K at all times. Direct measurement of the beam current permitted quantitative correlation between electron dose rates and crystallization rates. Enlarging the electron beam to reduce current density was found to slow the propagation of the crystallization front. Furthermore, a factor of 4 smaller electron dose was required for a given rate using 100 keV electrons as compared to 200 keV, indicating that crystallization is driven by ionization-induced atomic rearrangement within the gate layer. Lattice spacing between the oxygen sub-lattices of β-Ga2O3 and γ-Al2O3 are favorable for the nucleation of crystallites at the interface. Multivariate statistical analysis of electron energy loss spectroscopy (EELS) data also showed evidence of diffusion between Al and Ga in the substrates and gate oxides, respectively. These structural transformations at the semiconductor–insulator interface are expected to influence the device electrical behavior and are relevant to the continued refinement of β-Ga2O3 device technology.

Computer modeling of distribution of dispersed particles by the cross-section of cylindrical dispersion-strengthened metal materials

The article presents the process of creating a computer model for predicting the distribution of particles during centrifugal casting using the ANSYS FLUENT 16.0 software module. To predict the distribution of particles by volume in the world at the moment there are several mathematical models. Most of them are based on the steady state assumption: models describing the criteria for dropping particles by a growing crystallization front and models calculating critical particle absorption rates by growing dendritic crystals. Some models attempt to describe the dynamic state of the system or to determine the criterion for capturing non-metallic inclusions by the solidification front during centrifugal casting of metal. The process of creating the new model, its scheme and geometry are described. Its preprocessor takes into account such phenomena as two-phase flow, energy equation, lamellar flow, introduction of discrete phases (strengthening particles), melting/crystallization. The model considers account of interaction of two liquid phases: air and steel melt; interfacial interaction is described by the equation of surface tension. As the materials used, the authors used steel grade 12Kh18N10T as the base metal, carbides of tungsten, boron and yttrium oxide as input particles. During simulation, the physicochemical parameters of these substances were taken into account. The process of modeling the distribution of particles during centrifugal casting using the Skif-Ural computing cluster, included in the TOP-500 of the world’s most powerful computers, is presented. As a result of the simulation, in addition to graphical display, data arrays were obtained that describe the coordinates of each particle at each moment in time in increments of 0.00001 seconds, which allows us to predict the exact location of each particle at each moment of casting. The results of the work indicate that centrifugal casting technology with the introduction of dispersed particles during the casting process allows obtaining dispersion-strengthened metal materials with predicting the distribution of refractory particles.

Origin of the biphase nature and surface roughness of biogenic calcite secreted by the giant barnacle Austromegabalanus psittacus

The calcite grains forming the wall plates of the giant barnacle Austramegabalanus psittacus have a distinctive surface roughness made of variously sized crystalline nanoprotrusions covered by extremely thin amorphous pellicles. This biphase (crystalline-amorphous) structure also penetrates through the crystal’s interiors, forming a web-like structure. Nanoprotrusions very frequently elongate following directions related to the crystallographic structure of calcite, in particular, the <− 441> directions, which are the strongest periodic bond chains (PBCs) in calcite. We propose that the formation of elongated nanoprotrusions happens during the crystallization of calcite from a precursor amorphous calcium carbonate (ACC). This is because biomolecules integrated within the ACC are expelled from such PBCs due to the force of crystallization, with the consequent formation of uninterrupted crystalline nanorods. Expelled biomolecules accumulate in adjacent regions, thereby stabilizing small pellicle-like volumes of ACC. With growth, such pellicles become occluded within the crystal. In summary, the surface roughness of the biomineral surface reflects the complex shape of the crystallization front, and the biphase structure provides evidence for crystallization from an amorphous precursor. The surface roughness is generally explained as resulting from the attachment of ACC particles to the crystal surface, which later crystallised in concordance with the crystal lattice. If this was the case, the nanoprotrusions do not reflect the size and shape of any precursor particle. Accordingly, the particle attachment model for biomineral formation should seek new evidence.

Evolution of glass structure during femtosecond laser assisted crystallization of LaBGeO5 in glass

Single crystal LaBGeO5 architecture formed in La2O3-B2O3-2GeO2 glass serves as a model system for understanding of the transformation from glass to single crystal under fs laser irradiation. This transformation is highly sensitive to glass composition, raising the need to understand how the glass structure and composition evolve during crystallization. In this work, micro X-ray absorption spectroscopy and X-ray fluorescence mapping were used to study the glass structure and composition preceding LaBGeO5 crystallization in the region surrounding the single crystal growth front. The results show precipitation of germanium nanoparticles, as well as anisotropic segregation of Ge and La prior to LaBGeO5 crystallization, which are explained assuming electromigration in high fields within the plasma formed by the fs laser irradiation. However, these nanoparticles do not serve as sites for heterogeneous nucleation. The similarity of local structure around Ge in the starting glass and the ultimate crystal appears to enhance the crystallization process.

Edge-Directed LTPS Via Flash Lamp Annealing Using a Cr Adhesion Layer for Improved Wettability

Flash lamp annealing (FLA) has been explored as an alternative to excimer laser annealed low-temperature polycrystalline silicon (LTPS) for display applications and has shown potential to streamline and decrease the cost of fabrication while maintaining electrical performance [1], [2]. In this method, amorphous silicon is exposed to a single pulse of light emitted from a xenon flash lamp, causing melting and recrystallization in an area of cm2 rather than nm2 in less than a millisecond. FLA has shown value in crystallizing silicon and activating dopants in both PFET and NFET thin-film transistors (TFTs) [3]. However, FLA suffers from inconsistent material uniformity, a consequence of large areas of silicon melting simultaneously. This can cause the molten silicon to de-wet from the oxide or nitride underlayer, leading to the formation of numerous randomized voids. Though previous studies have shown encouraging results when using this void-associated LTPS morphology, improved structure control is necessary to reduce the variance of FLA-LTPS device operation.Edge-directed LTPS has a distinctly different macroscopic morphology than the void associated material, with potential advantages in uniformity and electrical performance. Differences between these morphologies have been previously discussed [4], however in that work the realization of the edge-directed material was hardly intentional and the formation mechanism was not clearly determined. A repeatable and controllable method of producing edge-directed LTPS would effectively solve a key obstacle towards industrial integration of FLA, that of randomly-dispersed irregularities and material nonuniformity.In this work, a replicable morphology of FLA-LTPS is produced and analyzed by using a thin chromium underlayer to promote adhesion without rendering the silicon entirely conductive. With thickness reduced to < 10 nm, Cr does not absorb damaging amounts of energy from the FLA pulse, as opposed to other metal and ceramic layers, which could improve wettability. Chrome has been applied as a silicon underlayer in thin-film photovoltaic applications [5]. The result of this Cr adhesion layer beneath an a-Si mesa during FLA is a characteristic ridge-like pattern of grain formation, as shown in Figure 1A. Large, elongated domains oriented perpendicular to the edges of the silicon mesas are clearly visible. An abrupt ridge is seen equidistant between the long edges of this mesa, with further diagonal ridges present at a 45° angle from the corners. This structure suggests a propagation of crystallization that begins at the edge of the mesas and travels inwards, terminating when it reaches another advancing crystallization front. This structure is only realized when the Cr layer is a continuous film rather than discrete microparticles, suggesting that the effect is due to adhesion rather than metal-induced crystallization. Without this adhesion layer, the randomly-irregular structure seen in Figure 1B is produced. Figure 1C demonstrates atomic force microscopy on this ridge morphology. In this structure, a maximum height differential of around 55 nm between the tops of ridges and the lowest point on the mesa is demonstrated (ignoring scanning artifacts). This is significant, as the original thickness of amorphous silicon as deposited was only 60 nm, suggesting large-scale mass transfer during crystallization. However, dewetting behavior is minimized. The elongated structure, lack of voids, and edge-inward pattern of this morphology is hypothetically consistent with explosive crystallization [6], versus a quasi-equilibrium melt-phase which results in randomly-irregular material.Similar morphology has been demonstrated using excimer laser and continuous-wave laser annealing, in which only a small area of silicon is in liquid phase at any given moment [7], [8]. However, its formation in FLA, in which large areas are melted simultaneously, is an important step towards the commercial viability of flash lamp annealing. This polycrystalline material is being explored as a viable channel material for PFET TFTs. The apparent presence of micrometer-scale grains is expected to improve channel mobility [9], while providing a uniform and predictable structure, thus minimizing variance in device operation. Material properties and electrical characteristics will be presented. Figure 1