Welcome to ICME-Welding

The research activities in the Group for Integrated Computational Materials Engineering for Welding (ICME-W) led by Dr. Wei Zhang encompass:

  • Additive manufacturing of metals (powder bed, blown-powder, binder jetting and sintering)
  • Light-metal and dissimilar-metal joining for transportation (automotive, shipbuilding etc.)
  • Creep-resistance steels and alloys for power generation
  • Inertia and linear friction welding for aerospace engine shafts and blisks
  • Resistance spot welding and mash seam welding of sheet metals
  • Modeling of welding and additive manufacturing processes and materials (Abaqus, Sysweld, Flow-3D, DEFORM, LS-Dyna, Simufact, and Thermo-Calc)
  • Specialized thermal-mechanical-metallurgical testing in Gleeble with high-temperature digital image correlation (DIC)

Dr. Zhang's google scholar site is here. For additional information, please contact him at " ".

Research Highlights

Pressureless sintering of stainless steel part printed by Binder Jetting Additive Manufacturing

Binder Jetting-Metal Additive Manufacturing (BJ-MAM) is a powder bed-based additive manufacturing technology which deposits liquid binder droplets to join powder particles to form complex shaped structures (i.e., green parts). A main issue for BJ-MAM is the part shrinkage and distortion during high-temperature sintering. In a recent paper published in Additive Manufacturing (free access through November 14, 2021 courtesy of Elsevier), a finite element model was developed
incorporating an elastic-viscoplastic constitutive equation for computing both uniaxial equivalent creep strain and volumetric swelling strain. Material property data used in the constitutive equation such as viscosity and creep parameters were collected from the literature, critically reviewed, and then inputted into the model. Other salient features of the model included thermal-mechanical property data that were dependent on both relative density and temperature as well as frictional contact between the part surface and the furnace wall under gravitational load. The calculated quantities such as shrinkage, final relative density, and deformed shapes were compared with the respective experimental data across different part geometries.

Sintering simulation of binder jetting printed 316L parts

Predictive modeling of deflection after paint-bake due to coefficient of thermal expansion differences in dissimilar metal structures

Automotive structures are increasingly utilizing dissimilar metals such as advanced high strength steels and high-strength aluminum alloys to reduce weight. As steels and aluminum alloys have different coefficients of thermal expansion, structures made of these two materials are susceptible to deflection during paint-bake. In a project sponsored by DOE Vehicle Technologies Program, Prof. Wei Zhang's group developed a predictive model for thermal-mechanical response during paint-baking of dissimilar metal structures. Picture below shows that the deflection profile at peak temperature (197 degC) predicted by the model is very consistent with that measured experimentally. The model can be used design the part geometry as well as pitch distance between joints to reduce the thermally-induced deflection.  More information can be found in this report.

Comparison of predicted and measured deflection at paint-bake peak temperature (*Experimental data courtesy of Oak Ridge National Laboratory)

Decreasing pitch distance significantly reduces thermally-induced deflection during paint-bake.

Non-spherical particles in powder bed fusion additive manufacturing

In laser-powder bed fusion (L-PBF) additive manufacturing, the powder particles commonly produced by gas atomization have some non-sphericities. It is known that non-spherical particles affect the particle packing process, however, there is a limited understanding of their effect on molten pool dynamics resulted from laser powder interaction. In their recent paper published in Computational Materials Science, Gao et al. studied the non-sphericities in L-PBF using a computational framework consisting of a particle packing model based on discrete element method (DEM) and a meso-scale molten pool model based on heat transfer and free surface flow. The results suggest some amount of non-sphericities could be tolerated without markedly affecting the melt track dimensions. Moreover, the types of non-sphericities (e.g., joined vs. satellite) could play an important role on the build quality.

Molten pool dynamics in laser powder bed consisted of non-spherical particlesMolten pool dynamics in laser powder bed consisted of non-spherical particles

Four journal papers, two patent applications, and one novel process for cost-effective dissimilar metal joining

Prof. Zhang and his colleagues just made a breakthrough in cost-effective dissimilar metal joining for fuel-efficient vehicles. Press hardened boron steels with ultrahigh strength (above 1500 MPa) are widely used in crash-sensitive safety components. Joining such steels to aluminum alloys is challenging due to various factors including the steel’s tenacious Al-Si coating. Prof. Zhang's team successfully applied ultrasonic plus resistance spot welding (abbreviated as U+RSW) to join 1.2-mm-thick AA6022-T4 to 1.4-mm-thick Al-Si coated press hardened boron steel, Usibor®-AS 1500. The joint had a high lap shear strength of 5 kN with a button pull-out fracture from the AA6022 sheet. The results of that study are published in Metallurgical and Materials Transactions A.

The U+RSW process was developed initially on joining aluminum alloy to uncoated low carbon steels. In their papers published in Materials & Design and Welding Journal, U+RSW was used to join 1-mm-thick AA6061-T6 to 0.9-mm-thick AISI 1008 steel with 0.4-mm-thick AA6061-T6 or 0.3-mm-thick AA3003 as the insert. The thickness of intermetallic layer formed at the Al insert/steel interface was less than 1.5 µm, resulting in a sound joint strength of 3.2 kN and a nugget pull-out failure mode.

In collaboration with Prof. Carolin Fink, they further studied the application of U+RSW to join 1.2-mm-thick AA6022-T4 to 1-mm-thick Zn-coated (galvannealed) Dual Phase steel DP980. In tensile (or lap) shear testing, the peak strength was 3.7 kN with a button pull-out fracture from the AA6022 sheet. The results are published in Science and Technology of Welding and Joining (access courtesy of Taylor & Francis Online).



Scientific, technological and economic challenges for metal AM

3D printing is now widely used in aerospace, healthcare, energy, automotive and other industries, and metal printing is the fastest growing sector. A commentary article published in Nature Materials, which was led by Prof. T. DebRoy of the Pennsylvania State University and coauthored by Prof. W. Zhang and others, discusses in detail the scientific, technological and economic challenges that must be understood and addressed for further development of metal printing.

*Images courtesy of Nature Materials

Structural support in additive manufacturing

For additive manufacturing of complex geometries, “sacrificial” structures, generally referred to as supports, are commonly deposited between the main part and the substrate to anchor overhanging features and prevent distortion induced by thermal-stress. In their recent paper published in the journal Additive Manufacturing, Prof. Zhang and his colleagues discussed the effect of such structural support on microstructure of additively manufacturing Nickel Alloy 718. Analytical equations, taking into account various laser processing parameters, material properties and support geometries, were developed to calculate the heat build-up and cooling conditions during laser-powder bed fusion.


Micro-resistance spot welding of dissimilar platinum to niobium wires

Thermo-electro-mechanical simulation was used to model the temperature and deformation behaviors during micro-resistance spot welding of dissimilar platinum to niobium wires. The results are published in a recent paper, Optimization of A Dissimilar Platinum to Niobium Micro-Resistance Weld: A Structure-Processing-Property Study, in Journal of Materials Science.


Softening of heat-affected zone

Prof. Wei Zhang's group recently developed a 3D fully coupled electro-thermo-mechanical model for resistance spot welding of aluminum silicon coated hot-stamped boron steel. A non-isothermal Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation is coupled with the process model to accurately predict local softening of subcritical heat-affected zone (SCHAZ). This model is essential for computer-aided engineering (CAE) based design of light-weight and impact resistant structures. Detailed description of their results can be found in their recent paper published in Materials & Design.

RSWSubcritical heat affected zone softening in hot-stamped boron steel during resistance spot welding


High-Temperature Mechanical Testing

High-temperature stress-strain curves are essential input to numerical simulation of manufacturing processes such as welding, additive manufacturing and forging. A group of researchers, comprising Mr. Alexey Kuprienko (MS student), Dr. Ying Lu, Dr. Daniel Tung (currently at Sandia National Laboratories) and Prof. Wei Zhang, all with Dept. of Materials Science and Engineering, recently developed a unique high-temperature testing procedure based on Gleeble® and digital image correlation (DIC). The testing capabilities include:

  • Rapid heating (up to 10,000 K/s) to testing temperature and quenching after hot deformation to preserve deformed microstructure
  • Use of complex thermal and mechanical loading cycles such as heating, holding, and tensioning during cooling
  • Measurement of local strain using DIC making it suitable for testing specimens with spatially non-uniform property such as a weld joint
  • High-sensitive load cell suitable for sheet metals such as advanced high-strength steels
  • Slow to high strain rates (up to ~10 s-1)
  • Temperature up to 1473 K (1200 C or 2192 F)

The animation in the first figure below shows the local strain map generated using DIC during testing at 1373 K (1100 C or 2012 F). The corresponding true stress-strain curve is shown in the second figure below.

DICFigure - Accurate measurement of local strain using DIC during high-temperature testing.

Stress-strain-1100Figure 2: True stress-strain curve for a low alloy steel tested at 1373 K (1100 C or 2012 F).


Simulation of powder recoating dynamics

Powder packing structure is a critical parameter for powder bed based additive manufacturing. Prof. Zhang's group has been developing a dynamic model based on discrete element method (DEM). Detailed description of the model is available in their paper just published in The International Journal of Advanced Manufacturing Technology Integration of the powder model into the molten pool simulation was presented in the 2015 SFF paper and the 2016 AM paper.



Brief bio of Dr. Wei Zhang

Dr. Zhang is a Professor at the Department of Materials Science and Engineering - Welding Engineering Program in The Ohio State University. He is also an affiliated faculty member of The Simulation Innovation and Modeling Center (SIMCenter). Prior to coming to OSU in January 2013, he worked as a Senior Researcher at the Oak Ridge National Laboratory (ORNL) from 2008-2012 and as an Engineer Team Leader at the Edison Welding Institute (EWI) from 2004-2008. He earned his Ph.D. in Materials Science and Engineering from the Pennsylvania State University in Spring 2004. His B.S. and M.S. in Materials Science and Engineering were both earned at the Huazhong University of Science and Technology in China.

About Welding Engineering

Welding Engineering is much more than "Arcs and Sparks!" It is a multi-disciplinary engineering field that involves materials science, metallurgy and mechanical engineering. Computational modeling is an essential science and math skill of welding engineers to develop new products, solve problems, and ensure that welded structures are safe and a benefit to society.

Useful links: