Latticed Steel Joints Give Prefabricated Buildings a Major Seismic Boost

Engineers in China have unveiled a new type of beam-column joint for prefabricated concrete buildings that dramatically outperforms conventional cast-in-place construction under simulated earthquake loading. The novel connection, described in Results in Engineering, replaces bulky solid steel connectors with a lightweight latticed steel cage, and laboratory tests show it can carry more than 40 percent higher peak loads while dissipating far more seismic energy than a traditional monolithic joint. The finding could reshape how multi-story precast buildings are designed in earthquake-prone regions, where the connections between beams and columns have long been the weakest link in the chain of structural safety.

Precast concrete construction has become a cornerstone of building industrialization because it offers rapid assembly, easier quality control, and a smaller environmental footprint than conventional cast-in-place methods. Yet the beam-column joints that tie these structures together remain a persistent engineering challenge. These regions experience some of the most complex stress states in an entire frame, transferring and distributing internal forces while holding the structure together during violent ground shaking. In traditional precast joints, columns are typically spliced with grouted sleeves or grouted lap splices, techniques that demand exceptional workmanship on site. Incomplete grouting is a common defect, and the story-by-story assembly process requires extensive temporary shoring that consumes space, prolongs construction, and accumulates installation errors.

To overcome these shortcomings, the research team led by Xiao-Ran Wang and He-Tao Hou of Shandong University developed what they call a latticed steel-concrete composite, or LSCC, beam-column joint. The design builds on multi-story precast column technology, in which columns spanning two or three stories are cast as a single unit in the factory and hoisted into place, leaving the joint regions open for wet connections. The new joint fills these open regions with a latticed steel column connector fabricated from chords, batten plates, through-diaphragms, and welded cross stiffeners. Structurally, the chords behave like the longitudinal reinforcement of a conventional column, while the batten plates act as stirrups. Before the joint concrete is poured, the chords provide a continuous load-transfer path across the open joint zone, and the cross stiffeners restrain local distortion of the central latticed segment, giving the column temporary stiffness, integrity, and stability during transport and erection.

The connection to the beams is equally deliberate. Each precast beam end carries an embedded H-section steel connector, with the beam’s longitudinal reinforcing bars welded to the outer flange surfaces. During assembly, the H-section web is bolted to a connecting plate on the latticed column while the flanges are welded to the through-diaphragms. By substituting a latticed configuration for the solid-web steel components used in many earlier composite joints, the design cuts steel consumption substantially. The researchers calculated that an equivalent joint built with a solid-web steel tube would require roughly 103.7 kilograms of steel, whereas their latticed specimens needed only 84.6 and 79.8 kilograms, savings of 18.4 and 23.0 percent respectively.

To test the concept, the team fabricated three full-scale exterior joint specimens: a monolithic cast-in-place reference specimen and two precast LSCC specimens differing only in the width of the batten plates. The columns measured 450 by 350 millimeters in cross-section and stood 3,040 millimeters tall, while the beams were 350 by 250 millimeters and 1,780 millimeters long. All specimens used C30 concrete, Q355B steel plates, and HRB400 reinforcing bars, with weld lengths deliberately exceeding code minimums to improve reliability under cyclic loading. Each specimen was subjected to a constant axial load of about 1,050 kilonewtons on the column, representing a realistic design axial load ratio drawn from an actual multi-story precast frame project, while two hydraulic actuators pushed and pulled the beam tips in a low-cycle reversed cyclic protocol following Chinese seismic testing standards.

The differences in behavior were striking. The cast-in-place specimen developed dense diagonal cracks in the joint core, concrete spalling at the beam-column interface, and a pinched, inverse S-shaped hysteretic response caused by joint shear failure and bond-slip of the reinforcement. The two LSCC specimens, by contrast, showed minimal cracking in the joint core and plump, spindle-shaped hysteresis loops with negligible strength degradation. Their damage concentrated in the beam span, roughly 300 to 800 millimeters from the column face, rather than at the joint itself. In quantitative terms, the composite joints achieved peak loads 45.1 and 43.1 percent higher than the monolithic reference, ultimate displacements 56.6 and 50.8 percent greater, and cumulative energy dissipation at their ultimate states 135.8 and 115.6 percent higher. At a beam-tip displacement of 70 millimeters, their equivalent viscous damping coefficients exceeded the cast-in-place specimen’s by more than 80 percent.

Strain measurements and curvature analysis revealed the mechanism behind this superior performance. The embedded H-section steel stiffens the beam end so effectively that it relocates the plastic hinge, the zone of controlled flexural yielding where a structure is designed to absorb earthquake energy, away from the vulnerable joint region toward the beam span. In the cast-in-place specimen, the near-joint segment dominated flexural deformation, contributing up to 89 percent of the total displacement at large amplitudes. In the composite joints, once displacement exceeded 30 millimeters, the outer beam segments took over, with one segment alone contributing 54 percent of the deformation in a representative case. This more uniform curvature distribution allows a larger volume of material to participate in energy dissipation, which is precisely what seismic designers strive to achieve.

The team also built high-fidelity finite element models in ABAQUS, using the concrete damaged plasticity model and a reinforcement hysteretic constitutive law that accounts for bond-slip effects. The simulations reproduced the observed failure modes and damage evolution with peak-load errors below 7.1 percent, validating the modeling approach. A parametric study then explored design trade-offs. Thinning the latticed connector from 10 to 8 millimeters, a 7.4 percent steel saving, barely affected capacity, but reducing it to 4 millimeters cut peak load by 13.8 percent. Lengthening the embedded H-section steel increased capacity steadily, adding 6.9 to 13.5 kilonewtons of peak load for every 75 millimeters of additional length. Upgrading reinforcement from HRB400 to HRB500 or HRB600 raised peak capacity by 12.7 and 24.9 percent respectively, though the higher-strength bars showed slightly faster post-peak strength degradation due to their narrower plastic deformation margin.

The authors caution that their experimental conclusions rest on a single specimen per configuration, without repeated tests, so the reported improvements apply primarily to the dimensions, materials, and loading conditions investigated. Still, the implications are considerable. The study demonstrates that rational detailing, such as narrowing the batten plates in the joint core, can trim steel consumption by more than 5 percent with almost no loss of structural performance, and that a latticed connector can simultaneously solve the temporary-stability problems of multi-story precast columns and deliver seismic behavior that surpasses monolithic construction. If subsequent testing confirms these trends across a wider range of parameters, latticed steel-concrete composite joints could become an attractive option for resilient, rapidly assembled buildings in seismic zones worldwide.

Subject of Research: Seismic performance of novel prefabricated latticed steel-concrete composite beam-column joints

Article Title: Seismic performance of novel prefabricated latticed steel-concrete composite beam-column joints

Article References: Wang, X.-R., Hou, H.-T., Xia, H.-J., Zhang, B., Wang, D.-Y., & Liu, S.-Q. (2026). Seismic performance of novel prefabricated latticed steel-concrete composite beam-column joints. Results in Engineering, 32, Article 113355. https://doi.org/10.1016/j.rineng.2026.113355

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113355

Keywords: prefabricated concrete, beam-column joints, seismic performance, latticed steel connector, cyclic loading test, plastic hinge relocation, energy dissipation, finite element analysis, precast construction, steel-concrete composite, ductility, earthquake engineering