PNNL Researchers Win Award for Metal-to-Metal Riveting

One of the key challenges facing today’s automobile industry is to develop a strong, corrosion-resistant, and cost-efficient method for joining dissimilar metals—a critical step in manufacturing multi-material car body structures. 

Traditional fusion-based joining techniques, which rely on melting materials at high temperatures, often lead to untoward changes in material metallurgical properties, as well as liquid-metal embrittlement. Such techniques also increase the risk of spot corrosion, high-residual stress and component distortion, and difficulties in joining materials with very different melting points.

In March 2025, seven researchers at Pacific Northwest National Laboratory (PNNL) received the Light Metals Subject Award-Aluminum Alloys, bestowed by the Materials, Metals & Materials Society (TMS). 

The TMS award recognizes the individual excellence of a paper that simulates and experimentally validates a promising solid-phase technique that is emerging in the world of joining lightweight materials: friction riveting.

Fric-riveting, as it is called, is the latest addition to a family of solid-phase joining techniques already well established at PNNL, including friction stir welding and friction stir processing. 

All three employ a spinning tool to generate heat, plasticize the joined materials, and enhance solid-state properties. Fric-riveting, however, does not use the traversing movement of friction stir welding and friction stir processing to create a weld. Instead, it deploys tool rotation and plunging to affix rivets at different joining spots.

Simulating a process

The winning paper, led by PNNL Computational Scientist Lei Li, presents a comprehensive process model that demonstrates how fric-riveting works. The study also shows how the model can accurately predict key process conditions, such as temperature and stress, within specific process parameters.

Notably, the study is the first in the light metals field to offer a robust simulation of the entire fric-riveting process.

In fric-riveting, a rotating rivet generates frictional heat but softens the surrounding material without reaching the material’s melting point. As a result, this technique effectively mitigates or eliminates the drawbacks associated with traditional fusion-based methods.

Fric-riveting is also energy-efficient and requires no surface preparation, pre-drilled through-holes, or external heat sources. Moreover, this solid-phase process promotes material mixing and grain refinement, which enhances both joint strength and corrosion resistance.

Modeling challenges

Process modeling fric-riveting is difficult because of the numerical challenges required to simulate complex contact conditions, large material deformations and mixing, and extreme thermomechanical conditions.

To get around these challenges, Li and his PNNL team employed a computational framework based on smoothed particle hydrodynamics (SPH)──a method originally developed for astrophysical simulations.

Traditional mesh, or grid-based, methods rely on simulated fixed meshes to represent material geometry and deformation. Instead, SPH is a mesh-free approach. It uses particles to represent the material and track its flow throughout the fric-riveting process. This approach makes the model well-suited for simulating large deformations since each particle carries time-dependent physical properties. Collectively, the particles simulate (with high accuracy) material behavior, deformation patterns, and temperature distributions.

Using SPH, the PNNL researchers accurately predicted the severe plastic deformation zone created by the rotating and plunging steel rivet. Under severe friction and pressure, the rivet-metal interface is highly plasticized, which releases heat that softens the material and can cause material deformations.

Accurate predictions and model validation

Li and his colleagues validated their simulation model with a carefully designed physical experiment that confirmed the accuracy of the model’s predictions.

The fric-riveting experiment employed two aluminum alloys sheets (AA5052-H32 and AA6061-T6) with thermocouples attached to monitor temperature changes. The experiment measured the plunging force of a spinning M42 threaded steel rivet, driven at varying speeds by a motorized rotational system.

The experiment’s results revealed that a uniform metallurgical bonding area successfully forms between the dissimilar aluminum alloys and the rivet.

“We got a very nice, curved interface between the metal sheets with material mixing and a bonding zone free of defects,” said Li.

He pointed out that such thorough metal mixing resulted in a gap-free joint─an important feature for both preventing corrosion and increasing joint strength.

Most importantly, Li said, the experimental data validated their computational approach.

“It was important to see how well our model predicted conditions,” he said. “The model accurately summarized the entire material flow of the spinning, plunging rivet and its ability to efficiently join two dissimilar metal layers.”

Accelerating process development

In the end, Lei Li and his coauthors concluded that their process model accurately predicted key phenomena observed during experiments. It’s an achievement that he said lays the groundwork for developing future applications that have enhanced efficiency. 

The PNNL team’s fric-riveting modeling work is a significant step toward the technique’s industrial-scale implementation and robotic automation.

Looking ahead, the researchers will use their validated model to accelerate the identification of optimal parameters for rotational and plunging speeds. The goal is to refine rivet geometry and to maximize joint strength, material mixing, and lap shear strength─a measure of how hard it is to pull bonded surfaces apart.

The team also intends to use their model to explore ways of joining other metal-to-metal combinations, including the challenging interface of aluminum to steel.

Such an advancement, Li said, “could have far-reaching implications for lightweight, multi-material structures in automotive and aerospace manufacturing.”

This research was funded by the Department of Energy’s Vehicle Technologies Office. Authors of the publication include Lei Li, Mayur Pole, Hrishikesh Das, Sridhar Niverty, Md Reza-E-Rabby, Jorge F. Dos Santos, and Ayoub Soulami.