Towards data-efficient mechanical design of bicontinuous composites usinggenerative AI

The distribution of material phases is crucial to determine the composite’s mechanical property.While the full structure-mechanics relationship of highly ordered material distributions can be studied with finite number of cases,this relationship is difficult to be revealed for complex irregular distributions,preventing design of such material structures to meet certain mechanical requirements.The noticeable developments of artificial intelligence(AI)algorithms in material design enables to detect the hidden structure-mechanics correlations which is essential for designing composite of complex structures.It is intriguing how these tools can assist composite design.Here,we focus on the rapid generation of bicontinuous composite structures together with the stress distribution in loading.We find that generative AI,enabled through fine-tuned Low Rank Adaptation models,can be trained with a few inputs to generate both synthetic composite structures and the corresponding von Mises stress distribution.The results show that this technique is convenient in generating massive composites designs with useful mechanical information that dictate stiffness,fracture and robustness of the material with one model,and such has to be done by several different experimental or simulation tests.This research offers valuable insights for the improvement of composite design with the goal of expanding the design space and automatic screening of composite designs for improved mechanical functions.

PHASE STRUCTURE FACTOR AND INTERFACE CONJUNCTION FACTOR OF CAST IRON AND ITS COMPOSITION DESIGN

Theempiricalelectrontheory of solidsand molecules( EET) and theimproved TFDtheory wereapplied tocalculatethe phasestructurefactorsand interfaceconjunction factorsofcom mon alloying elementsincastiron. Akind of Si- Mo- Cu ductileiron with rareearth Mg asnodularizer was designed accordingtothese valenceelectron structure parameters. Actual applicationtestsshow thatthelongevity of thisiron is 1.5 timesof thatof high manganesesteel. This accordance of theoretical results and actual effectsshows the composition design methodcan beused in othercastiron research.

Cluster-based composition rules for ternary alloy systems

The cluster-based composition rule in ternary alloy systems including quasicrystals, bulk metallic glasses, crystalline phases and Lave phases-related body-centered cubic (BCC) solid solution forming systems was summarized. The so-called cluster line in a ternary phase diagram refers to a straight composition line linking a specific binary cluster to the third element. The compo- sition ranges of quasicrystals and bulk metallic glasses can be determined by the direct use of cluster lines, where two cluster lines intersect at the optimum phase forming composition. Furthermore, the alloys on the cluster line in Laves phase-related BCC solid solution alloy systems have larger hydrogen storage capacities.

Erratum to: New high-strength Ti–Al–V–Mo alloy: from high-throughput composition design to mechanical properties

Erratum to:International Journal of Minerals, Metallurgy and Materials Volume 26, Number 9, September 2019, Page 1151https://doi.org/10.1007/s12613-019-1854-1The original version of this article unfortunately contained a mistake. The presentation of Fig. 11 was incorrect. The correct version is given below:

Application of Empirical Electron Theory of Solids and Molecules to Composition Design of Multi-Component Medium-Low-Alloy Steels

For austenitic octahedral segregation structure units, their pure mathematics statistic distribu!ive probability is calculated by the empirical electron theory （EET） of solids and molecules and K-B formula. The practical distributive probability can be obtained only if the statistic distribution of austenitic octahedral segregation structure units and the interaction of the alloying elements in steel are considered. Based on 8 groups of experimental data of original steels, three empirical formulas revealing relationships between material macromechanics factor （Sm） and tensile strength （ab）, or impact energy （AK）, or hardness （HRC） of multi-component medium-low-alloy steels were established, respectively. Through the three empirical formulas, new supersaturated carburizing steel has been successfully designed and developed. The other 2 groups of the original experimental steels are used as the standard steel for testing the percentage error of the new steel. The results show that the calculated values are well consistent with those of measured ones and the new supersaturated carburized steel can meet the requirements of the die assembly of cold-drawn seamless stainless steel tube of Taiyuan Iron ＆ Steel （Group） Company LTD.

Composition formula of Al-TMs high-entropy alloys derived by cluster-plus-glue-atoms model and its experimental verification

High-entropy alloys(HEAs)are based on solid solutions which characterized by chemical short-range ordering(CSRO),but there is no accurate structural tool to address CSRO characteristic,which obstacles precise composition design for HEAs.In this study,based on the cluster-plus-glue-atoms model,the composition law of Al-TMs(TMs,transition metals)HEAs with BCC and FCC structures was revealed.In BCC structure,with five elements in equi-atomic ratio,the composition formula of Al-TMs HEAs can be expressed by a cluster formula[Al-M_(14)]Al_(3),and in non equi-atomic ratio can be expressed by[Al-M_(14)]Al,where M is the average atom of the TMs.But in FCC structure,both Al-TMs HEAs with five elements in equiatomic and non equi-atomic ratios can be expressed by a cluster formula[Al-M_(12)]Al.To confirm the effectiveness of cluster formula for Al-TMs,two alloys were designed,[Al-Ti_(4)V_(3)Nb_(4)Mo_(3)]Mo and[Al-Ti_(3)V_(3)Nb_(4)Mo_(4)]Al.Results show that the[Al-Ti_(3)V_(3)Nb_(4)Mo_(4)]Al alloy has both higher strength and higher plastic deformation at room temperature.Besides,[Al-Ti_(3)V_(3)Nb_(4)Mo_(4)]Al alloy shows slower soften effect at 800℃,contributed to its higher strength.By substituting Ta for some of Mo,the strength of[Al-Ti_(3)V_(3)Nb_(4)Mo_(3)Ta]Al at room temperature and high temperature drastically decreases,suggesting that Ta element deteriorates the properties of Al-TMs alloys.

COMPOSITION DESIGN FOR A BEARING STEEL WITH LIMITED HARDENABILITY

The composition of a bearing steel was designed for limited hardenability by use of Grossmann's method. A medium frequency induction Process was applied to heat bearings to ensure penetrant heating and suitable solving of carbon and other elements in the matrix. The hardened depth measured from the end quenching test samples and actual bearings matches well with the designed one.

High-strength Ti-Al-V-Zr cast alloys designed using α and β cluster formulas

Ti-Al-V-Zr quaternary titanium alloys were designed followingα-{[Al-Ti12](AlTi2)}17-n+β-{[Al-Ti12Zr2](V3)}n,where n=1-7(the number ofβunits),on the basis of the dual-cluster formula of popular Ti-6Al-4V alloy.Such an alloying strategy aims at strengthening the alloy via Zr and V co-alloying in theβ-Ti unit,based on the originalβformula[Al-Ti14](V2Ti)of Ti-6Al-4V alloy.The microstructures of the as-cast alloys by copper-mold suction-casting change from pureα(n=1)toα+α’martensite(n=7).When n is 6,Ti-5.6Al-6.8V-8.1Zr alloy reaches the highest ultimate tensile strength of 1,293 MPa and yield strength of 1,097 MPa,at the expense of a low elongation of 2%,mainly due to the presence of a large amount of acicularα’martensite.Its specific strength far exceeds that of Ti-6Al-4V alloy by 35%.

A novel Ti-5.55Al-6.70Zr-1.50V-0.70Mo-3.41Nb-0.21Si alloy designed using cluster-plus-glue-atom model for laser additive manufacturing

A novel Ti-5.55Al-6.70Zr-1.50V-0.70Mo-3.41Nb-0.21Si alloy was designed using the cluster formula approach(cluster-plus-glue-atom model)and prepared by laser melting deposition(LMD).Its composition formula 12[Al-Ti_(12)](AlTi_(2))+5[Al_(0.8)Si_(0.2)-Ti_(12)Zr_(2)](V_(0.8)Mo_(0.2)Nb_(1)Ti)features an enhancedβ-Ti via co-alloying of Zr,V,Mo,Nb and Si.The experimental results show that the cluster formula ofαandβphases in the novel alloy are respectivelyα-[Al-Ti_(11.5)Zr_(0.5)](Al_(1)Ti_(2))andβ-[Al_(0.8)Si_(0.2)-Ti_(13.2)Zr_(0.8)](V_(1)Mo_(0.4)Nb_(1.6)),both containing Zr elements.The fitted composition via the α andβphase cluster formulas has little difference with the actual alloy composition,suggesting that the validity of cluster-plus-glue-atom model in the alloy composition design.After hot isostatic pressing(HIP),both the Ti-6Al-4V and the novel alloy by LMD are characterized by prior-βcolumnar grains,while the typical<100>texture disappears.Compared with Ti-6Al-4V,Ti-5.55Al-6.70Zr-1.50V-0.70Mo-3.41Nb-0.21Si alloy exhibits a combination of higher strength(1,056 MPa)and higher ductility(14%)at room temperature and higher strength(580 MPa)at 550℃ after HIP,and can potentially serves as LMD materials.

A bionic controllable strain membrane for cell stretching at air–liquid interface inspired by papercutting

Lung diseases associated with alveoli,such as acute respiratory distress syndrome,have posed a long-term threat to human health.However,an in vitro model capable of simulating different deformations of the alveoli and a suitable material for mimicking basement membrane are currently lacking.Here,we present an innovative biomimetic controllable strain membrane(BCSM)at an air–liquid interface(ALI)to reconstruct alveolar respiration.The BCSM consists of a high-precision three-dimensional printing melt-electrowritten polycaprolactone(PCL)mesh,coated with a hydrogel substrate—to simulate the important functions(such as stiffness,porosity,wettability,and ALI)of alveolar microenvironments,and seeded pulmonary epithelial cells and vascular endothelial cells on either side,respectively.Inspired by papercutting,the BCSM was fabricated in the plane while it operated in three dimensions.A series of the topological structure of the BCSM was designed to control various local-area strain,mimicking alveolar varied deformation.Lopinavir/ritonavir could reduce Lamin A expression under over-stretch condition,which might be effective in preventing ventilator-induced lung injury.The biomimetic lung-unit model with BCSM has broader application prospects in alveoli-related research in the future,such as in drug toxicology and metabolism.

Effect of Variable Selection on Multidisciplinary Design Optimization:a Flight Vehicle Example

Different multidisciplinary design optimization （MDO） problems are formulated and compared. Two MDO formulations are applied to a sounding rocket in order to optimize the performance of the rocket. In the MDO of the referred vehicle, three disciplines have been considered, which are trajectory, propulsion and aerodynamics. A special design structure matrix is developed to assist data exchange between disciplines. This design process uses response surface method （RSM） for multidisciplinary optimization of the rocket. The RSM is applied to the design in two categories： the propulsion model and the system level. In the propulsion model, RSM determines an approximate mathematical model of the engine output parameters as a function of design variables. In the system level, RSM fits a surface of objective function versus design variables. In the first MDO problem formulation, two design variables are selected to form propulsion discipline. In the second one, three new design variables from geometry are added and finally, an optimization method is applied to the response surface in the system level in order to find the best result. Application of the first developed multidisciplinary design optimization procedure increased accessible altitude （performance index） of the referred sounding rocket by twenty five percents and the second one twenty nine.

Design for Ti-Al-V-Mo-Nb alloys for laser additive manufacturing based on a cluster model and on their microstructure and properties

In this study,α+βTi-Al-V-Mo-Nb alloys with the addition of multiple elements that are suitable for laser additive manufacturing(LAM)were designed according to a Ti-6Al-4V cluster formula.This formula can be expressed as 12[Al-Ti12](AlTi2)+5[Al-Ti14]((Mo,V,Nb)2Ti),in which Mo and Nb were added into the alloys partially instead of V to give alloys with nominal compositions of Ti-6.01Al-3.13V-1.43Nb,Ti-5.97Al-2.33V-2.93Mo,and Ti-5.97Al-2.33V-2.20Mo-0.71Nb(wt.%).The microstructures and mechanical properties of the as-deposited and heat-treated samples prepared via LAM were examined.The sizes of theβcolumnar grains andαlaths in the Nb-containing samples are found to be larger than those of the Ti-6Al-4V alloy,whereas Mo-or Mo/Nb-added alloys contain finer grains.It indicates that Nb gives rise to coarsenedβcolumnar grains andαlaths,while Mo significantly refines them.Furthermore,the single addition of Nb improves the elongation,whereas the single addition of Mo enhances the strength of the alloys.The simultaneous addition of Mo/Nb significantly improves the comprehensive mechanical properties of the alloys,leading to the best properties with an ultimate tensile strength of 1,070 MPa,a yield strength of 1,004 MPa,an elongation of 9%,and micro-hardness of 355 HV.The fracture modes of all the alloys are ductile-brittle mixed fracture.

Optimization of low-temperature alkaline smelting process of crushed metal enrichment originated from waste printed circuit boards

A novel low-temperature alkaline smelting process is proposed to convert and separate amphoteric metals in crushed metal enrichment originated from waste printed circuit boards. The central composite design was used to optimize the operating parameters,in which mass ratio of Na OH-to-CME, smelting temperature and smelting time were chosen as the variables, and the conversions of amphoteric metals tin, lead, aluminum and zinc were response parameters. Second-order polynomial models of high significance and3 D response surface plots were constructed to show the relationship between the responses and the variables. Optimum area of80%-85% Pb conversion and over 95% Sn conversion was obtained by the overlaid contours at mass ratio of Na OH-to-CME of4.5-5.0, smelting temperature of 653-723 K, smelting time of 90-120 min. The models were validated experimentally in the optimum area, and the results demonstrate that these models are reliable and accurate in predicting the smelting process.

Effect of Planting Density and Fertilizer Application on Fiber Yield of Ramie (Boehmeria nivea)

As the most important cultural practices for ramie （Boehmeria nivea） production, the single effects of plant density and nitrogen （N）, phosphorus （P）, potassium （K） fertilization on yield are well documented. To achieve the high yield and quality of ramie fiber, it is principal to quantify the fertilizer dosage for ramie sustainable production. A central composite design （CCD） was adopted with three replications, for a two-year field experiment （2008-2009） in the Yangtze River Valley of China. The aim was to evaluate both the individual and combined effects of plant density and N, P, and K fertilization on yield and fiber quality. The effects of the four factors on yield and quality of ramie cultivar Huazhu 4 were tested. Then mathematical models of the relationship among the four factors affecting ramie production and quality were established and analyzed to optimize the four factors, and used to establish optimum cultivation methods for the elite cultivar suitable to this area of the Yangtze River Valley. Supplements of N （X2）, P 0（3） and K 0（4）, and the density （X0 with N （X2） interaction significantly influenced strength in tests of mature ramie. The influence on fiber yield in the two year tests by the four factors ranked as follows： X3〉X4〉X~〉X2. We established the optimization technique attaining yield of 2 600 kg ha-~ in the tests as following： density of 28 350-31 650 plants ha-1, and supplements of N, P and K of 363-387, 98.58-105.48 and 280.20- 319.8 kg ha-1, respectively. It was concluded that nitrogen mostly improved plant growth and fiber yield while potassium had discernible effects on fiber quality.

A time-released osmotic pump fabricated by compression-coated method: Formulation screen, mechanism research and pharmacokinetic study

In this investigation,time-released monolithic osmotic pump(TMOP)tablets containing diltiazem hydrochloride(DIL)were prepared on the basis of osmotic pumping mechanism.The developed dosage forms were coated by Kollidon®SR-Polyethylene Glycol(PEG)mixtures via compression-coated technology instead of spray-coating method to form the outer membrane.For more efficient formulation screening,a three-factor five-level central composite design(CCD)was introduced to explore the optimal TMOP formulation during the experiments.The in vitro tests showed that the optimized formulation of DIL-loaded TMOP had a lag time of 4 h and a following 20-h drug release at an approximate zero-order rate.Moreover,the releasemechanismwas proven based on osmotic pressure and its profile could be well simulated by a dynamic equation.After oral administration by beagle dogs,the comparison of parameters with the TMOP tablets and reference preparations show no significant differences for C_(max)(111.56±20.42,128.38±29.46 ng/ml)and AUC_(0-48 h)(1654.97±283.77,1625.10±313.58 ng h/ml)but show significant differences for T_(max)(13.00±1.16,4.00±0.82 h).These pharmacokinetic parameters were consistent with the dissolution tests that the TMOP tablets had turned out to prolong the lag time of DIL release.

Microstructure and mechanical properties of laser additive manufactured novel titanium alloy after heat treatment

A novel α+β titanium alloy with multi-alloying addition was designed based on the cluster formula 12[Al-Ti_(12)](AlTi_(2))+5[Al-Ti_(14)](AlV_(1.2)Mo_(0.6)Nb_(0.2))which was derived from Ti-6Al-4V.The nominal composition of this novel alloy was determined as Ti-6.83Al-2.28V-2.14Mo-0.69Nb-6.79Zr.In this study,the novel alloy and Ti-6Al-4V alloy samples were prepared by laser additive manufacturing.The microstructure,micro-hardness,room/high temperature tensile properties of the as-deposited samples were investigated.Compared to Ti-6Al-4V,the novel alloy has much higher room and high temperature(600℃)tensile strengths,which are 1,427.5 MPa and 642.2 MPa,respectively;however,it has a much lower elongation(3.2%)at room temperature because of the finer microstructure.To improve the elongation of the novel alloy,heat treatment was used.After solution at 960℃ or 970℃ for 1 h followed by air cooling and aging at 550℃ for 4 h followed by air cooling,a unique bi-modal microstructure which contains crab-like primaryαand residual β phase is obtained,improving the compression elongation by 80.9% compared to the as-deposited samples.The novel alloy can be used as a high-temperature and high-strength candidate for laser additive manufacturing.

Optimization of Process Parameters for in High-Energy Ball Milling of CNTs/Al2024 Composites Through Response Surface Methodology

The mathematical models are developed to evaluate the ultimate tensile strength( UTS) and hardness of CNTs / Al2024 composites fabricated by high-energy ball milling. The effects of the preparation variables which are milling time,rotational speed,mass fraction of CNTs and ball to powder ratio on UST and hardness of CNTs / Al2024 composites are investigated. Based on the central composite design( CCD),a quadratic model is developed to correlate the fabrication variables to the UST and hardness. From the analysis of variance( ANOVA),the most influential factor on each experimental design response is identified. The optimum conditions for preparing CNTs / Al2024 composites are found as follows: 1. 53 h milling time,900 r / min rotational speed,mass fraction of CNTs 2. 87% and Ball to powder ratio 25 ∶ 1. The predicted maximum UST and hardness are 273.30 MPa and 261.36 HV,respectively. And the experimental values are 283.25 MPa and256.8 HV,respectively. It is indicated that the predicted UST and hardness after process optimization are found to agree satisfactory with the experimental values.

Computer-Aided Design of Some Advanced Steels and Cemented Carbides

Thermodynamic and kinetic study on TRIP （transformation induced plasticity） steels, cemented carbides and mold steel for plastics were carried out in order to design modern advanced materials. With the sublattice model, equilibrium compositions of ferrite and austenite phases in TRIP steels, as well as volume fraction of austenite at inter-critical temperatures for different time were calculated. Concentration profiles of carbon, manganese, aluminum and silicon in the steels were also estimated in the lattice fixed frame of reference. The effect of Si and Mn on TRIP was discussed according to thermodynamic and kinetic analyses. In order to understand and produce the graded nanophase structure of cemented carbides, miscellaneous phases in the M-Co-C （M= Ti, Ta, Nh） systems and Co-V-C system were modeled. Solution parameters and thermodynamic： properties were listed in detail. The improvement of machining behavior of prehardened mould steel for plastics was obtained by computer-aided composition design. The results showed that the matrix composition of large-section prehardened mould steel for plastic markedly influences the precipitation of non-metallic inclusion and the composition control by the aid of Thermo-Calc software package minimizes the amount of detrimental oxide inclusion. In addition, the modification of calcium was optimized in composition design.

Machine learning technique for prediction of magnetocaloric effect in La(Fe,Si/Al)_(13)-based materials

Data-mining techniques using machine learning are powerful and efficient for materials design, possessing great potential for discovering new materials with good characteristics. Here, this technique has been used on composition design for La（Fe,Si/Al）（13）-based materials, which are regarded as one of the most promising magnetic refrigerants in practice. Three prediction models are built by using a machine learning algorithm called gradient boosting regression tree（GBRT） to essentially find the correlation between the Curie temperature（TC）, maximum value of magnetic entropy change（（？SM）（max））,and chemical composition, all of which yield high accuracy in the prediction of TC and（？SM）（max）. The performance metric coefficient scores of determination（R^2） for the three models are 0.96, 0.87, and 0.91. These results suggest that all of the models are well-developed predictive models on the challenging issue of generalization ability for untrained data, which can not only provide us with suggestions for real experiments but also help us gain physical insights to find proper composition for further magnetic refrigeration applications.