Secondary Dendrite Arm Research Articles

Effect of Molten Pool Spatial Arrangement on Texture Evolution in Pulsed Laser Additive Manufacturing of Inconel 718.

The epitaxial growth of dendrites, which often results in a strong texture, is the most common phenomenon during the laser additive manufacturing process. In this study, the epitaxial growth of dendrites and texture evolution in three directions were studied by changing the z-increment, pulse period, and track offset, respectively. The influence of the molten pool interface on the growth and competition of dendrites is analyzed. Both green grains (<110> // BD) with rotated cube texture in the molten pool overlapping zones and red grains (<100> // BD) with fiber texture in the molten pool center zones coexist for different z-increment samples, forming the typical sandwich texture feature. In a short pulse period, the dendrites can grow directly epitaxially and form the strong fiber texture due to gentle interface and short distance. With the decrease of the track offset, the molten pool morphology changes from flat to narrow and deep. When θ is close to 90°, dendrites grow along the secondary dendrite arms at the overlapping zone, forming V-shape grains. This work also provides a promising method for texture customization for laser additive manufacturing.

A Probabilistic Fatigue Strength Assessment in AlSi-Cast Material by a Layer-Based Approach

An advanced lightweight design in cast aluminium alloys features complexly shaped geometries with strongly varying local casting process conditions. This affects the local microstructure in terms of porosity grade and secondary dendrite arm spacing distribution. Moreover, complex service loads imply changing local load stress vectors within these components, evoking a wide range of highly stressed volumes within different microstructural properties per load sequence. To superimpose the effects of bulk and surface fatigue strength in relation to the operating load sequence for the aluminium alloy EN AC 46200, a layer-based fatigue assessment concept is applied in this paper considering a non-homogeneous distribution of defects within the investigated samples. The bulk fatigue property is now obtained by a probabilistic evaluation of computed tomography results per investigated layer. Moreover, the effect of clustering defects of computed tomography is studied according to recommendations from the literature, leading to a significant impact in sponge-like porosity layers. The highly stressed volume fatigue model is applied to computed tomography results. The validation procedure leads to a scattering of mean fatigue life from −2.6% to 12.9% for the investigated layers, inheriting strongly varying local casting process conditions.

Microstructure evolution and age-hardening response in Mg-Sn-Sm alloys under a wide range of Sm/Sn ratio

The microstructure evolution and age-hardening response for different Sm/Sn ratios (0–2.55, in wt.%) of Mg-Sn-Sm alloys were investigated. The second phase formation in as-cast alloys and the Mg3Sm precipitates formed in aged alloys were characterized using XRD, FESEM and HAADF-STEM with EDS techniques. Results indicate that the Sm/Sn ratio has a great influence on the phase constitution, α-Mg grain size and age-hardening response. With the increment of Sm/Sn ratio, Mg41Sm5 and thermally stable MgSnSm phases precipitate. When the Sm/Sn ratio is about 1.19, the secondary dendrite arm spacing of α-Mg grains significantly decreases. Furthermore, the alloy with Sm/Sn ratio up to 2.55 exhibits the highest age-hardening response, the hardness value increases from 52 HB at solution-treated condition to 74 HB at peak-aged condition (ageing at 220 °C for a short time of 4 h). This is attributed to the large volume fraction of needle-like Mg3Sm precipitates formed in the α-Mg matrix during ageing treatment, which results in a significant precipitation strengthening effect.

Influence of heat input on bead profile and microstructure characteristics in laser and laser-hybrid welding of Inconel 617 alloy

Bead profile and microstructure analysis are greatly influenced by heat input in any welding process. The present work provides a comprehensive comparative investigation on these aspects during Autogenous Laser welding (ALW) and Laser-MIG Hybrid welding (LHW) processes in Inconel 617 superalloy. In case of LHW, weld bead profile formed was a typical wine-cup shaped one, whereas, in case of ALW, bead profile observed was Y-type shape on account of differences in number of heat sources and intensity profiles. Weld bead width and penetration depth were found to increase with heat input in both ALW and LHW processes. The fusion zone (FZ) area of LHW was found to increase by 3 times as compared to ALW as a result of high heat input in LHW than ALW. Indeed, the addition of MIG source to laser in LHW produced wide upper bead width (trebled) as compared to ALW. Radius of curvature (ROC) at neck zone, an important bead profile characteristic for assessing liquation cracking susceptibility, was found to increase with heat input and was significantly higher in LHW than ALW due to heat stagnation at neck zone. In comparison with ALW, LHW showed a coarser columnar/cellular dendritic microstructure in FZ with segregation of inter-dendritic carbides relatively richer in Mo and Cr. Secondary dendritic arm spacing (SDAS), cooling rates and microhardness will also be discussed.

Comparative investigation on ingot evolution and product quality under different arc distributions during vacuum arc remelting process

The arc characteristics especially the arc distribution of the vacuum arc remelting (VAR) process are closely related to the ingot evolution and consequently the final product quality. Taking segregation defects and solidification quality as the indexes to evaluate the final product quality under different arc distributions, a three-dimensional full-scale transient model for predicting segregation defects is developed in this paper. Local solidification time (LST) and secondary dendrite arm spacing (SDAS) are used to characterize the solidification quality. This model is adopted to predict channel segregation (freckle) and macrosegregation defects during VAR process of Ni-5.8 wt.%Al-15.2 wt.%Ta alloy under several typical arc distributions. The results show that the diffuse arc has the shallowest molten pool among several arc distributions. At the early stage of solidification, the solute enrichment channels formed under the Gaussian and diffuse arc are slender and concentrated, while they are relatively sparse and uneven in thickness under the rotating arc. The value of volume-averaged segregation extent is always the smallest under the diffuse arc and the largest under the Gaussian arc, with final values of 67.53% and 69.66%, respectively. The volume-averaged values of LST and SDAS are relatively the largest under the Gaussian arc, about 1599 s and 209 μm, respectively; and they are the smallest under the diffuse arc, about 1520 s and 206 μm, respectively. In the same process conditions, the product quality of VAR ingots prepared under the diffuse arc is the best, followed by the rotating arc, and the worst under the Gaussian arc.

Towards the Prediction of Tensile Properties in Automotive Cast Parts Manufactured by LPDC with the A356.2 Alloy

Aluminum-silicon-magnesium alloys are commonly used in the automotive industry to produce structural components. Among usual quality controls of produced castings, microstructure characterization and determination of mechanical properties are the most critical aspects. However, important problems can be found when measuring mechanical properties in those areas of castings with geometrical limitations. In this investigation, a set of A356 alloys have been prepared and then used to manufacture test castings and automotive castings in a laboratory and in industrial conditions, respectively, using Low Pressure Die Casting (LPDC) technology. Test castings were used to predict secondary dendritic arm spacing (SDAS) by using thermal parameters obtained from experimental cooling curves. The results have been then compared to the ones found in the literature and improved methods for estimating SDAS from cooling curves have been developed. In a subsequent step, these methodologies have been checked with different industrial castings by using simulated cooling curves and experimentally measured SDAS values. Finally, the calculated SDAS values together with the Mg contents present in A356 alloys and the temperature and aging time data have been used to develop new models so as to predict the tensile properties in different areas of a given casting prototype. These developed models allow casters and designers predicting tensile properties in selected areas of a given prototype casting even during design and simulation steps and considering the processing variables expected in a given foundry plant. The structures of these new models have been described and experimentally validated using different processing conditions.

Effects of Ni and Co on the Corrosion Resistance of Al-Si-Cu-Zn-Fe Alloys in NaCl Solution

The corrosion behavior of Fe-containing directionally solidified (DS) and centrifugally cast (CC) Al-Si-Cu-Zn alloys with either Co or Ni additions has been investigated. Electrochemical and immersion corrosion methods were used to investigate the corrosion behavior in 0.6 M NaCl after short (1-h) and long (30-day) exposure periods. The employed solidification methods allowed the production of samples with a wide range of secondary dendrite arm spacing (SDAS) while preserving Si and Fe-containing phases. The 0.5 wt.% Ni and Co additions led to the growth of the AlFeSi(Ni) and AlFeSi(Co) phases, but no binary AlNi nor AlCo intermetallic particles have been generated. Potentiodynamic polarization studies at early exposure revealed an increase in the corrosion potential as the Ni was added for either fast or slow solidified samples. The electrochemical impedance spectroscopy at early exposure demonstrated that the Ni-modified alloy, on the other hand, was associated with smaller charge transfer resistances, indicating a reduction in the corrosion resistance after a short elapsed time into the electrolyte. However, the 30-day immersion tests revealed much lower corrosion rate of the Ni-modified alloy than the other alloys, while the corrosion rates of the Co-modified and non-modified alloys were similar. In the Ni-containing alloy, a decreased corrosion rate under a long-term corrosion process was attributed to the formation of a thick and dense alumina layer, effectively protecting the surface under such conditions. This work contributes to better knowledge of the corrosion behavior of Ni- and Co-corrected Al industrial scrap compositions.

Effect of Shielding Gas on the Microstructure and Properties of Laser-MAG Hybrid Welded Joint for Nickel-Saving Stainless Steel

Laser-MAG (metal active gas) hybrid welding of nickel-saving 08Cr19Mn6Ni3Cu2N stainless steel was carried out by using 98%Ar + 2%N2 and 95%Ar + 5%CO2 as shielding gases. The effect of different shielding gases on the microstructure and properties of the welded joints was investigated. The results showed that arc shrinkage was significant with the addition of nitrogen, weld spatter increased with the expansion of arc volume, and arc stability deteriorated. The ferrite content in the weld decreased by about 60%, the ferrite dendrite also gradually became finer, and the secondary dendrite arm was shorter. Only a small amount of δ and γ phases existed in the weld, and no precipitation of the σ phase and nitride was found. Observing four crystal planes, we found that size of the austenite grains decreased with the addition of nitrogen. The average tensile strength of the welded joints decreased from 712 MPa to 704 MPa, but with improved corrosion resistance, the pitting corrosion rate increased from 19.45 g·m2/h to 18.72 g·m2/h, and the hardness of weld was slightly reduced.

Interactive effects of interfacial energy anisotropy and solute transport on solidification patterns of Al-Cu alloys

Combined effects of the cooling rate, alloy composition, and crystal-melt (CM) interfacial anisotropy on solidification of Al-Cu alloys are studied by integrating molecular dynamics and phase-field simulations. Capillary fluctuation method is used to determine the CM interfacial energy properties by molecular dynamics simulations for alloys ranging from 3 to 11 at% Cu. While the average CM interfacial energy decreases with increasing Cu content, its anisotropy does not present a clear trend with composition change. Primary and secondary dendrite arm spacings as well as θ-phase fraction are calculated by phase-field simulations, and validated against experimental measurements and analytical solutions at cooling rates ranging from 1 to 1250 K/s. Results show that the θ-phase fraction decreases with increasing the cooling rate, and this reduction is more drastic in alloys with a higher Cu content. Also, the microstructure features are influenced by the growth dynamics, where seaweed structure formation results in a more homogenous distribution of θ-phase and a finer microstructure. The effects of temperature gradient, Cu concentration gradient, and interfacial energy properties on the dendritic growth morphology of Al-Cu alloys are summarized by a map of supercooling versus the CM interfacial anisotropy to predict pattern formation. The results show that, irrespective of Cu content and cooling rate, the seaweed structure formation is halted at CM interfacial anisotropies larger than 0.005. As the anisotropy decreases, different seaweed structures can form regarding the constitutional supercooling. At low anisotropies (Al-3 and Al-8.4 at% Cu) and low supercooling (Al-3 at% Cu) fractal or degenerate seaweed is dominant while at high supercooling (Al-8.4 at% Cu) compact seaweed forms. This difference in supercooling stems from different solute atom transport rates.

Influence of Solidification Parameters on the Amount of Eutectic and Secondary Arm Spacing of Al–7wt% Si Alloy Solidified under Microgravity

During the solidification of hypoeutectic Al–7% Si alloy, density differences develop in the melt due to variations in concentration and temperature. On Earth, melt flow can occur due to gravity, which then affects the solidification process. The microgravity environment strongly eliminates convection in the melt and allows investigation of the solidification process in purely diffusive circumstances. In this study, four solidification experiments were performed on grain-refined and non-grain-refined Al–7 wt% Si alloy on-board the International Space Station (ISS) in the Materials Science Lab (MSL) to study the effect of solidification parameters (solid/liquid front velocity (v) and temperature gradient (G)) on the grain structure and dendritic microstructure. The grain structure has been analyzed in detail in some earlier studies. The aim of this work was to carry out detailed analysis of the macrosegregation caused by the diffusion of Si from the initial mushy zone during the homogenization step and the subsequent solidification phase of the experiments as well as the correlated distribution of eutectic along the solidification direction. The secondary dendrite arm spacing (SDAS) for different process conditions was also studied. For these two issues, microgravity experimental results were compared to simulation results. The macrosegregation was calculated by the finite difference method. Because the steady-state solidification conditions were never reached, the solidification process was characterized by the average front velocity and temperature gradient. Considering the actual liquidus temperature (TL) caused by macrosegregation, the SDAS was calculated as a function of the average processing parameters and the actual liquidus temperature with the classical Kirkwood’s equation. As a result, good agreement was obtained between the calculated and measured SDAS.

Electrochemical Properties of Diluted Al-Mg Alloys With Columnar-To-Equiaxed Transition

The objective of the present research is to study the corrosion susceptibility of two Al-Mg diluted alloys (Al-0.5wt.%Mg and Al-2wt.%Mg) with different grains structures obtained by directional solidification (columnar, equiaxed and columnar-to-equiaxed transition, CET) in 0.5% NaCl solution, at room temperature. The corrosion resistance is analyzed by potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) techniques in both longitudinal and transversal sections of the samples. The columnar grain zone presents higher corrosion resistance than the equiaxed grain zone. In addition, the transversal section shows higher corrosion resistance than the longitudinal section of the samples. Then, the Al-0.5wt.% Mg alloy displays higher corrosion resistance than the Al-2wt.% Mg alloy. The values of the polarization resistance are used as a basic criterion for the evaluation of the corrosion resistance of both alloys. In this way, when the polarization resistance decreases with the increasing in the distance from the base, the grain size, secondary dendritic arm spacings and hardness increases. In addition, when the polarization resistance increases, the critical temperature gradient decreases.

SCC behaviour of laser and hybrid laser welded stainless steel weld joints

In this study, stress corrosion cracking (SCC) resistance was evaluated for type 316LN stainless steel weld joints fabricated by advanced welding techniques such as Laser, Hybrid Laser – Tungsten inert gas (HLT) and Hybrid Laser – Metal inert gas (HLM) welding. 316 LN stainless steel base metal exhibited significantly better resistance to SCC. The solidification mode, secondary dendritic arm spacing (SDAS) and tensile residual stress distribution in the weld metal was found to strongly influence SCC resistance of 316LN stainless steel weld joints. The weld joint processed by HLM process exhibited the highest time to failure during SCC testing and hence better SCC resistance due to primary ferritic mode of solidification, finer SDAS and lower width of tensile residual stress distribution.

Enhanced strength and ductility in Ti46Al4Nb1Mo alloys via boron addition

To improve the strength and ductility of TiAl alloys by second phase, Ti46Al4Nb1Mo alloys doped with different B content (0.4%, 0.8%, 1.2%, 1.6% and 2.0%, atomic percent, hereafter in at.%, referred to as TNM-xB) were prepared. Macro/microstructure evolution, mechanical properties and deformation mechanisms of the alloys were studied systematically. Results showed that the microstructure of TNM-0.4B and TNM-0.8B alloy remained columnar dendrites, and the secondary dendritic arms of columnar grains were more obvious. When the content of B is 1.2%, the columnar dendrites transformed to equiaxed grains, and the α2/γ lamellae colony size was further refined in TNM-1.6B and TNM-2.0B alloy. The morphologies and kinds of the borides were changed with increasing B content, XRD results showed that TiB phase appeared with 1.6%B content, and both TiB and TiB2 phase formed in TNM-2.0B alloy. There were straight and curved TiB phases located around grain boundaries in TNM-0.4B and TNM-0.8B alloy, and when the content of B increased to 1.2%, the curved TiB phases were reduced, while the tiny and straight TiB phases increased. With further increasing B content to 1.6% and 2.0%, the tiny and straight TiB phases were coarser. Compressive testing results showed that the mechanical properties of the TNM alloy were enhanced with increasing B content. The maximum strength and strain of TNM alloy were 2339MPa and 33.7% with 1.6% B addition. The compressive strength and strain were mainly enhanced via refinement of lamellar colony and formation of TiB, and it is found that pile-up of dislocations and deformed twins promoted by TiB are predominant in improving the mechanical properties of TNM alloys with higher strength and strain.

The effect of melt flow induced by RMF on the meso- and micro-structure of unidirectionally solidified Al–7wt.% Si alloy Benchmark experiment under magnetic stirring

This work deals with the effect of melt flow on the solidified structure by RMF. The microstructure of the first non-stirred part of the Al-7Si alloy is columnar. After switching on stirring first a transient zone was formed with eutectic “cupola”. After the first transient zone, the “Christmas Tree Like” microstructure was formed. During the solidification of stirred part, a “tulip-like” interface evolves. During the last two decades, many algorithms were developed to simulate the solidification process for different casting methods like ingots, continuous casting of steel and the direct chill cast of aluminum. Experiments performed under exactly known conditions and with the detailed knowledge of meso- and micro-structures are required for validating these simulations. The aim of this paper is to give a data set to validate these simulations. Unidirectional solidification experiments were performed by using a rotating magnetic field (RMF) to study the effect of melt flow on the solidified micro- and meso-structure of the Al–7wt.% Si binary alloy. The first and the third 1/3 parts of samples were solidified without magnetic stirring, and the second (middle) 1/3 part was solidified by using magnetic stirring. The magnetic induction was 10 mT, the temperature gradient was ∼7 K/mm, and the sample movement velocity was 0.1 mm/s. On the longitudinal section of the sample, the columnar/equiaxed transition (CET), the equiaxed/columnar transition (ECT), the secondary dendrite arm spacing (SDAS), and the macrosegregation (concentration distribution and the amount of eutectic) were investigated. The primary dendrite arm spacing (PDAS) and the grain structure were studied on the cross-section after color etching. The effect of melt flow induced by RMF on the meso- and microstructure of unidirectionally solidified Al–7wt.% Si alloy Benchmark experiment under magnetic stirring .

Multi-phase field model simulation based on MPI+OpenMP parallel: Evolution of seaweed and dendritic structure in directional solidification

A multi-phase model was established to imitate the growth of algal and dendritic grains during directional solidification. We studied the effects of temperature on the growth of bi-crystals and quantitatively analyzed the influence of anisotropic strength, thermal gradient, and pulling velocity on the evolution of bi-crystals. The results show that both weaker anisotropy strength and smaller pulling velocity can maintain the formation of seaweed tissue. The increase in the pulling velocity can degenerate the seaweed grains into dendrites and improve the growth rate of the dendrites, which make grain B produce more spindles, thereby accelerating the elimination of grain A. The thermal gradient is inversely proportional to the average initial spacing of dendrites. When the thermal gradient is too small, dendritic dendrites produce developed secondary dendrite arms, which, in turn, develop into tertiary dendrite arms to occupy the grain boundary, accelerating the elimination of seaweed grains. In addition, the multi-phase field model is solved by using central processing unit serial computation, single MPI (message passing interface) parallel programming method calculation, and MPI+OpenMP hybrid parallel programming structure, and the relevant factors affecting the efficiency of program operation are analyzed and tested. By comparing the computational efficiency of the three methods, it can be seen that the MPI+OpenMP hybrid parallel programming technology can make full use of computing resources in the case of large computing scale, further optimize the MPI parallel model, and obtain a higher acceleration ratio.

NbB refining capability: Effects of slow and rapid solidification on dendritic spacings and grain sizes of a 6201 alloy

NbB inoculants are refiners that might be of interest to the Al industry. Grain size (GS) and secondary dendrite arm spacing (SDAS) are two important microstructural variables to consider when assessing a refiner's performance. Both of these attributes may have an impact on the alloy’s properties. The interplay of grain refiner and cooling rate on grain size, SDAS, and grain shape in the 6201 alloy is investigated in this article. It is demonstrated that GS associated with rapid solidification of the 6201 + NbB alloy was roughly 7 times smaller than that associated with slow solidification. Furthermore, regardless of the processing method (either slow or fast solidification), α-Al cells are generated for GS/SDAS ratio below and equal to 6.

Microsegregation of a new Ni-Co-based superalloy prepared by electron beam smelting layered solidification technology and its homogenization behavior

The electron beam smelting layered solidification technology (EBS-LST) is put forward to fabricate a new Ni-Co-based superalloy with less segregation. The microstructures and homogenization behavior of the prepared ingot were studied. The experimental results show that the cross-sectional microstructure of the EBS-LST alloy consists of four typical areas, namely melted zone (MZ), partially fusion zone (PFZ), heat affected zone (HAZ), and bottom layer (BL). During the EBS-LST solidification process, the cooling rate is calculated to be 22.3 k/s according to the well-established model associated with secondary dendrite arm spacing. It can be inferred that the EBS-LST will reduce the solidification time, and thus eliminating the macrosegregation, and weaken the microsegregation of the ingot effectively. The effect of homogenization treatment on the microsegregation and microstructure evolution of the alloy is also studied. The residual segregation index equation during homogenization is constructed for the diffusion behavior of the strongly segregated elements of Ti and W. The diffusion coefficients D W and D Ti are calculated as 3.9 × 10 −15 m 2 ·s −1 and 5.4 × 10 −15 m 2 ·s −1 , respectively. • A new electron beam smelting layered solidification technology is presented. • Macro-segregation elimination of the Ni-Co-based superalloy ingots. • Less homogenization time for the Ni-Co-based superalloy ingots prepared by EBS-LST.

Microstructural evolution and mechanical property of Al–Mg–Mn alloys with various solidification cooling rates

In the present work, the microstructural evolution and mechanical property of as-cast Al–6 Mg–0.8Mn alloys with respect to the solidification cooling rate was investigated. The results demonstrate that as the cooling rate increases from 0.5 K/s to 46.9 K/s, the average grain size decreases firstly and then increases slightly; the average secondary dendrite arm spacing, the average length of Al 6 Mn and Al 3 Mg 2 phases decreases. A high cooling rate (21.3 K/s and 46.9 K/s) causes the increase in the solute segregation/shrinkage porosities, and a low cooling rate (0.5 K/s) leads to the formation of casting defects, thus deteriorating the mechanical properties of alloys. The alloy with the cooling rate of 5.4 K/s shows an optimum mechanical property i.e., the tensile strength, yield strength, and elongation are 298.7 MPa, 133.8 MPa, and 30.5%, respectively. The enhanced strength-ductility synergy is mainly attributed to the refined secondary dendrite arm spacing and reduced casting defects. This work provides a reference for the fabrication of as-cast Al–Mg–Mn alloys with excellent mechanical performance. • The relationship between cooling rate and microstructure/property was established. • The variation in tensile property due to the microstructural evolution was discussed. • The microstructural evolution induced by cooling rate was analyzed. • The strength-ductility synergy associated with the cooling rate was revealed.

Influence of the lathe rotation and structural parameters in the machinability of an Al-3.0wt% si alloy / Influência da rotação do torno e parâmetros estruturais na usinabilidade de uma liga Al-3%si

The metal alloys structures resulting from the solidification process can influence the machinability of these materials, however, few studies have attempted to establish relationships between microstructure and machining properties. This study aimed to verify the influence of lathe rotation and secondary dendritic arm spacing, λ2, on the maximum flank wear (Vbmax) of an Al-3.0wt.%Si alloy solidified in a horizontal directional device under unsteady-state conditions. After the analysis of the results it can be observed that for the same rotation, the variation of Vbmax as a function of λ2 can be expressed through a power-type function, , where “a” and “b” are constants, denoting, therefore, that bigger secondary dendritic arm spacing contributes to greater tool wear. It was also been found that, for the same dendritic arm spacing, higher rotations contribute to greater tool flank wear.

Microstructure Evolution of Mg-Sn-Y Alloy Solidified under High Pressure and Temperature Gradient

The microstructures of Mg-1Sn-2.5Y (wt%) alloys solidified under high pressures were investigated. In addition, a mathematical model was established to analyze the effects of solidification pressure and cooling rate on the average grain size. The results show that the alloy was solidified under high pressure and temperature gradient using the cooling rate difference in the high pressure chamber, resulting in the formation of the outer equiaxed zone, the columnar zone, and the equiaxed zone in the sample. With an increase in the solidification pressure, the columnar-to-equiaxed transition was inhibited in Mg-1Sn-2.5Y alloy. In the outer fine equiaxed zone and the columnar zone, the solubility of Sn in the Mg matrix increased with an increase in solidification pressure. The average secondary dendrite arm spacing decreased from 14–17 μm under 1 GPa to 9–11 μm under 1.5 GPa. Increases in pressure and cooling rate resulted in a reduction in average grain size.