Combinatorial and High-Throughput  Alloy Design, considering processing and bulk microstructure 

High-Throughput Bulk Synthesis and Prototyping Platforms

We have established Rapid Alloy Prototyping (RAP) as a primary experimental methodology to circumvent the efficiency bottleneck of traditional structural alloy development. While conventional workflows require sequential, time-consuming loops of casting individual charges, hot rolling, heat treatment, and mechanical testing, RAP compresses the evaluation of complex compositional and thermo-mechanical spaces from weeks to hours.

Our bulk RAP setup utilizes a vacuum induction melting furnace equipped with a linear stage carrying multiple water-cooled copper molds. This design allows us to cast up to five different alloys in a single operation by systematically adjusting the chemistry of the remaining melt through an air-lock feeding system.

Following casting, the blocks are hot-rolled at high temperatures (typically 1100 degrees Celsius) to eliminate casting defects, water-quenched, and sectioned parallel to the transverse-normal plane into segments. These segments are subjected to localized heat-treatment matrices and then stacked and cut simultaneously via spark erosion to rapidly produce tensile and metallographic specimens.

Through this comprehensively adapted processing chain, we successfully evaluated 45 distinct metallurgical conditions of a Fe-30Mn-1.2C-xAl Triplex steel system—comprising five aluminum concentrations (0 to 8 wt.%) across nine distinct aging treatments—within only 35 hours, demonstrating a six-to-tenfold acceleration over conventional metallurgical methods.

To further expand our high-throughput toolkit, we developed a combinatorial approach using laser additive manufacturing (LAM), specifically laser metal deposition (LMD). By injecting systematically varied ratios of metallic powders into the laser-generated melt pool, we fabricated bulk, piece-wise compositionally graded alloy specimens.

We validated this by blending a Cr-Mo-V hot-working tool steel and a nickel-based maraging steel across six distinct graded layers. Following hot thickness reduction of 50 percent at 1100 degrees Celsius, the graded material exhibited excellent microstructural and chemical homogeneity within each 500-micrometer layer. This allowed us to map a wide strength-ductility profile from 800 to 1650 MPa in tensile strength and 15 to 25 percent in elongation from a single bulk sample.

Microstructural Engineering of Low-Density Steels and High Modulus Composites

Our research in the quaternary Fe-Mn-Al-C system highlights the complex interplay between processing, bulk microstructure, and mechanical performance. These alloys are categorized into two primary structural classes:

• SIMPLEX Steels: High-manganese, single-phase austenitic steels containing low aluminum fractions (25 to 30 wt.% Mn, less than 4 to 5 wt.% Al, or up to 8 wt.% Al under natural aging conditions). These alloys are primarily strengthened via dislocation cell formation followed by massive mechanical twinning (the TWIP effect) under mechanical load, achieving tensile strengths of approximately 830 MPa and excellent total elongations of up to 77 percent.
• kappa-Carbide Strengthened Steels: Alloys containing higher aluminum fractions (above 8 wt.%) where aging treatments at 500 to 600 degrees Celsius trigger the precipitation of nanosized L12-ordered perovskite kappa-carbides within the austenitic matrix. When aluminum exceeds 10 wt.%, these carbides can precipitate directly during quenching.

Our atom probe tomography and transmission electron microscopy analyses reveal that these kappa-carbides form cuboidal nanoprecipitates with a lattice mismatch of less than 3 percent relative to the fcc austenite matrix. During deformation, dislocations bypass these coherent precipitates via Orowan looping, moving through the widest channels between carbide rods.

At high stress levels, dislocations pile up and shear the carbide interfaces. Conversely, in aluminum-free alloys, thermal aging leads to the formation of coarse, pearlitic ferrite-carbide particles along the grain boundaries, which causes severe embrittlement.

For weight-critical applications requiring exceptional stiffness-to-density ratios, we designed metal matrix composite high modulus steels (HMS) using ternary Fe-10B-5X systems (where X represents transition metals like Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, and W). In these systems, standard liquid metallurgy and hot swaging at 1100 degrees Celsius facilitate the in-situ precipitation of boride phases:

• Zirconium-Alloyed Composites: Swaging promotes the alignment of sharp-edged, spiky zirconium diboride particles along the deformation axis. This texturing increases the Young's modulus to 246 GPa, yielding the highest specific modulus of our group at 32.5 GPa g-1 cm3. However, the spiky morphology concentrates local stresses, restricting the tensile elongation to 13 percent.
• Chromium-Alloyed Composites: These alloys form spherical M2B-type borides that do not degrade ductility. They achieve a yield strength of 450 MPa and an elongation of 20 percent. This property profile matches the performance of our conventional titanium diboride benchmark composite but at a much lower raw material cost.

Phase Stability and Sluggish Kinetics in Multi-Component Systems

We also utilized combinatorial and ab initio techniques to study phase stability and elastic behavior in multi-component systems, specifically challenging the classical thermodynamic constraints of high entropy alloys (HEAs).

Conventional HEA design focuses strictly on equiatomic compositions to maximize configurational entropy. We investigated non-equiatomic CoCrFeMnNi alloys, specifically the Fe(64-x)MnxNi27.7Co5.6Cr2.3 system with manganese concentrations of 21, 24, 27, 34, and 38 at.%.

Despite a 17 to 20 percent reduction in configurational entropy compared to equiatomic HEAs, these alloys successfully form single-phase fcc solid solutions that remain stable up to their melting temperatures. While thermodynamic Calphad calculations predict that a bcc phase should co-precipitate at 495 degrees Celsius, our kinetic DICTRA simulations demonstrate that sluggish diffusion kinetics freeze the high-temperature fcc phase.

At 495 degrees Celsius, bcc nuclei require more than 100 days to initiate growth and cannot exceed 3.2 nanometers in size even after 3.2 years, rendering the bcc volume fraction negligible.

These non-equiatomic solid solutions deform through planar slip at low strains (under 2 percent), which transitions to highly dense dislocation walls and cell structures at higher strains. By cold-working and subsequently annealing the 27 at.% manganese variant at 900 degrees Celsius for 10 minutes, we achieved a refined grain size of 12 micrometers. This grain refinement triggered a strong Hall-Petch effect, increasing the yield strength by 150 percent (to 240 MPa) and the ultimate tensile strength by 72 percent (to 645 MPa) while preserving ductile plastic flow.

Elasticity and Localized Electronic Structure in fcc Fe-Mn-C Alloys

To validate our experimental findings, we used density functional theory to calculate the elastic properties of fcc Fe-Mn-C alloys, comparing combinatorial sputtered thin films with bulk alloys. The studied composition range spanned 1.5 to 3.0 at.% carbon, 28.0 to 37.5 at.% manganese, and 60.6 to 69.8 at.% iron.

Our experimental lattice parameters vary minimally between 3.597 and 3.614 Angstroms across the entire carbon range. Our theoretical VASP calculations (utilizing disordered local moment and antiferromagnetic 3Q configurations) underestimate these lattice parameters by only 1 to 2 percent.

   Fe-Mn-C Thin Film & Bulk Elastic Properties:

   ============================================

   Lattice Parameter:     3.597 - 3.614 Angstroms

   Young's Modulus:       185 - 251 GPa

   Carbon Addition:       Marginal effect on elastic modulus

Both bulk samples and polished thin films exhibit similar Young's modulus values ranging from 185 to 251 GPa, confirming that combinatorial thin films accurately replicate bulk elastic behavior.

Our ab initio calculations indicate that adding up to 3 at.% carbon increases the Young's modulus by only 4 percent, a variation too small to resolve experimentally.

We explained this marginal influence by studying the electron density distribution in the (004) crystallographic plane. Carbon occupies interstitial sites, sharing electrons with neighboring manganese and iron atoms to form highly localized, strong covalent Fe-C and Mn-C bonds.

Because these covalent bonds represent only a tiny fraction of the overall atomic matrix (at 3 at.% carbon), the dominant metallic bonding network remains undisturbed. Consequently, the addition of carbon does not alter the overall electronic structure or the bulk Young's modulus of high-manganese steels within this composition range.

Environment-Assisted Degradation and Hydrogen Trapping

As structural alloys approach and exceed tensile strengths of 1 GPa, managing hydrogen embrittlement becomes critical. We analyzed hydrogen-assisted fracture in a j-carbide-strengthened Fe-26Mn-11Al-1.2C alloy under in-situ hydrogen charging using electron backscatter diffraction and electron channeling contrast imaging:

• Damage Initiation: Fracture occurs through intergranular cracking at grain boundary triple junctions and the gradual formation of microvoid chains along grain boundaries.
• Deformation Localization: The presence of ordered kappa-carbides promotes intense, localized planar slip. This slip localization concentrates plastic strain at grain boundaries, accelerating void coalescence.
• Hydrogen Trapping: Thermal desorption spectroscopy and Redlich-Kister analyses show that the coherent interfaces of kappa-carbides act as hydrogen trap sites with a high activation energy of 76 to 80 kJ/mol.

While these precipitates serve as beneficial traps that immobilize solute hydrogen, the associated strain localization of the carbide-strengthened structure accelerates void formation along the grain boundaries under mechanical load.