The concept borrows from a classic of aerospace engineering: the sandwich composite. Two stiff outer skins are bonded to a lightweight core, and under bending the skins carry most of the tensile and compressive stress while the core resists shear and crushing. By holding the skins apart, the core dramatically increases the panel’s moment of inertia, meaning it resists bending far better than its weight would suggest. Honeycomb and foam cores have long served aircraft and ships, but the Brazilian team wanted a core that grows in a few years, sequesters carbon as it grows, and would otherwise be thrown away.

Bamboo fits that brief almost perfectly. The researchers worked with Phyllostachys aurea, a species harvested at the University of São Paulo’s Pirassununga campus, using the lower portions of roughly three-year-old culms. The waste problem they aimed to address is striking: bamboo shoot processing discards about 70 percent of the harvested material, and conventional culm harvesting leaves behind branches, leaves, and upper sections that account for a significant share of aboveground biomass. Upcycling this residue into engineered structural components could turn a disposal burden into a carbon-storing asset, since bamboo construction materials have been shown to yield lower life-cycle emissions than steel, concrete, and even some novel low-carbon cements.

Preparing the rings involved a preservation step borrowed from established bamboo engineering practice. The culms were immersed for seven days in an 8 percent aqueous solution of disodium octaborate tetrahedrate, a borate treatment that protects the lignocellulosic material against biological attack, then air-dried for another week. The natural circular cross-section of the culm was retained, producing rings about 30.5 millimeters in outer diameter with a mean wall thickness of 4.4 millimeters. That diameter was not arbitrary: earlier studies found 30-millimeter rings outperform both smaller 20-millimeter rings and larger 45-millimeter rings in flexural strength and stiffness, making them the sweet spot for sandwich cores.

The facings were commercial fiber cement boards with a bulk density of 1.28 grams per cubic centimeter, a flexural strength of 6.80 megapascals, and an elastic modulus of 2.44 gigapascals. Spectroscopic and mineralogical analysis using Fourier transform infrared spectroscopy, thermogravimetric analysis, and X-ray diffraction revealed a carbonated cementitious matrix containing portlandite, calcite, quartz, and residual alite. The bonding agent was a two-component polyurethane adhesive derived from castor oil, a bio-based resin with a 24-hour cure time that has become a favorite in sustainable composites research because it replaces petroleum-derived alternatives.

The experimental design was a rigorous full-factorial program. The team built 32 panels across eight configurations, varying three factors at two levels each: core height of 20 or 30 millimeters, compact versus gapped hexagonal ring arrangement, and adhesive bond-line thickness of 0.75 or 1.00 millimeters. Compact cores packed 23 rings tightly, leaving about 61 percent of the panel plan area as void, while gapped cores used 19 rings and pushed the void fraction to 73 percent. Panels were assembled under a gentle cold pressure of 2.3 kilopascals and tested in three-point bending according to the ASTM C393 standard, alongside measurements of density, thermal conductivity, and sound pressure levels.

The results delivered a genuinely counterintuitive lesson in structural mechanics. Making the core taller, from 20 to 30 millimeters, cut the equivalent density by 16.2 percent and boosted flexural rigidity by 86 percent in compact panels and 109 percent in gapped ones, exactly as sandwich theory predicts when skins move further from the neutral axis. Yet the taller panels were weaker in apparent flexural strength, dropping by 22 percent in compact and a dramatic 48 percent in gapped configurations. The gapped 30-millimeter panels also suffered a 33 percent reduction in nominal core shear stress, a consequence of fewer rings, smaller contact area between core and facing, reduced core continuity, and differences in specimen geometry. Lighter, it turns out, is not automatically better.

The standout performer was the 20-millimeter gapped panel with the thinner 0.75-millimeter adhesive line, which achieved the highest specific strength of 8.80 megapascal-centimeters-cubed per gram and the highest specific modulus among all configurations. Its low density more than compensated for a modest loss in absolute strength. Notably, thinning the adhesive line cut resin consumption by about 25 percent without significantly affecting flexural strength, and the adhesive content of 5.9 to 9.3 percent of panel mass compared favorably with bamboo boards that typically demand 10 to 20 percent resin. Failure analysis through microscopy showed that most panels failed by tensile cracking of the lower fiber cement facing, a sign that the adhesive bonds held firm, while the bamboo rings themselves never crushed under the loading nose, thanks to their impressive 64-megapascal compressive strength and 6-gigapascal modulus.

The thermal and acoustic results add practical appeal. Estimated thermal conductivity came in at 0.276 watts per meter-kelvin for compact panels and 0.260 for gapped ones, an order of magnitude below the 1.6 to 3.2 range typical of normal-weight concrete, thanks to the insulating air cavities threaded through the core. Acoustic testing at 125, 500, 1000, and 2000 hertz showed frequency-dependent sound pressure reductions, with significant differences between compact and gapped panels at 125 and 1000 hertz, likely reflecting internal scattering and viscous dissipation within the bamboo’s porous cellular architecture. The authors caution that these were comparative laboratory estimates rather than standardized absorption or transmission-loss measurements.

The researchers are candid about the road ahead before bamboo-ring panels reach real buildings. Future work must address concentrated-load resistance, screw-holding capacity, impact behavior, long-term durability under moisture and weathering, fire response, and full-scale wall-system testing. Still, the study demonstrates that a material once destined for the compost heap can be engineered into a panel that is lighter than conventional cementitious alternatives, thermally insulating, acoustically functional, and mechanically competitive, all while locking away biogenic carbon. As the construction industry scrambles to decarbonize, the answer may lie in a field of bamboo, sliced into rings and glued together with castor oil.

Subject of Research: Lightweight bamboo-ring core sandwich panels with fiber cement facings for sustainable construction

Article Title: Lightweight sandwich panels with bamboo-ring cores and fiber cement facings

Article References: Emadifard, A., Azevedo, A., Fioroni, C., Gholizadeh, P., Batista, F., Panzera, T., & Savastano, H., Jr. (2026). Lightweight sandwich panels with bamboo-ring cores and fiber cement facings. Results in Engineering, 32, Article 113327. https://doi.org/10.1016/j.rineng.2026.113327

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113327

Keywords: bamboo, sandwich panels, fiber cement, polyurethane adhesive, flexural strength, thermal conductivity, acoustics, sustainable construction, Phyllostachys aurea, core geometry, carbon emissions, building materials

The research matters because sandwich panels—two thin concrete wythes or steel sheets bonded around a lightweight insulating core—have become a dominant technology in energy-efficient construction. Their popularity stems from fast installation, excellent thermal performance, and reduced lifecycle costs, and building codes increasingly favor them as energy regulations tighten. But the insulation core is not merely filler. It transfers load between the outer layers, maintains composite action, and keeps the wall structurally stable. If heat from a fire passes through the concrete faces and softens the core, the panel’s entire load-transfer mechanism can be quietly undermined long before any visible failure occurs. Despite this, most prior research has focused on fire-resistance ratings and thermal conductivity rather than on how the mechanical properties of the core itself degrade.

The experimental program was unusually comprehensive. The team, led by Abhinav Anand with supervision from Vaibhav Singhal, cut specimens from commercially manufactured blocks of the three materials and subjected them to controlled heating followed by four types of mechanical testing at ambient conditions: compression, three-point flexure, double shear, and tension, all performed on a universal testing machine under displacement-controlled loading according to relevant international standards including EN 826, ASTM C203, EN 12090, and EN 1608. EPS specimens, with a density of 16 kilograms per cubic meter, were heated in a laboratory oven at temperatures from 50 to 80 degrees Celsius. The denser XPS, at 34 kilograms per cubic meter, was tested from 60 to 90 degrees Celsius. Rockwool, at 150 kilograms per cubic meter, was heated in a muffle furnace from 200 all the way to 800 degrees Celsius.

The first striking finding concerned geometry. Beyond 80 degrees Celsius for EPS and 90 degrees Celsius for XPS—temperatures associated with the glass transition of polystyrene, where polymer chains gain mobility and the material loses rigidity—the foams physically warped and bulged. EPS contracted dramatically, shrinking 28 percent in thickness, while XPS deformed anisotropically, expanding on some faces by more than 11 percent while contracting on others. At that point the materials could no longer sustain mechanical loading at all, making further testing impractical. Rockwool, by contrast, showed no melting or distortion even at 800 degrees Celsius, though specimens became noticeably softer, fluffier, and slightly discolored above 200 degrees Celsius—early signs that the organic binder gluing its basaltic fibers together was beginning to break down.

The compression results revealed a subtle hierarchy. EPS lost compressive strength progressively, dropping about 13 percent at 50 degrees Celsius and roughly 32 percent at 80 degrees Celsius, while its initial elastic modulus fell by nearly 48 percent. XPS proved remarkably resilient in compression, losing under 1 percent of compressive strength even at 70 degrees Celsius and only about 9 percent at 90 degrees—yet its elastic modulus still collapsed by 47 percent at the highest temperature, signaling irreversible damage to the closed-cell structure even when peak strength appeared intact. Rockwool told the most dramatic story: minor losses up to 400 degrees Celsius, then a catastrophic 78 percent strength reduction at 600 degrees and over 90 percent at 800 degrees, where the response was governed by loose fiber compaction rather than any true structural resistance, following near-complete burnout of the organic binder.

Flexural and shear tests exposed even greater vulnerabilities. EPS flexural strength fell by 20 percent at 80 degrees Celsius, with stiffness plummeting 55 percent. XPS lost about 22 percent of flexural strength and 15 percent of stiffness at 90 degrees. Rockwool, however, could barely be tested in bending after 400 degrees Celsius: the fibrous matrix delaminated into a loose, fluffy mass that could not support its own weight, with residual flexural strength estimated at a mere 1.57 kilopascals—a 99 percent loss. In shear, the pattern repeated. EPS shear strength dropped 38 percent and its shear modulus 50 percent at 80 degrees; XPS lost 47 percent of shear strength and a striking 73 percent of shear modulus at 90 degrees; and rockwool’s shear capacity fell by more than half at just 200 degrees, reaching an estimated 0.34 kilopascals at 400 degrees, when specimens could no longer even hold the steel loading plates of the test apparatus.

Tensile behavior followed suit, with an important nuance about failure modes. EPS failed brittlely through its core, with cracks propagating along weak inter-bead boundaries, losing 27 percent of tensile strength at 80 degrees Celsius. XPS showed a mixed-mode failure dominated by interfacial debonding near the steel plate boundary, losing about 23 percent of tensile strength at 90 degrees. Rockwool failed progressively through fiber pull-out, rupture, and interlayer delamination—a ductile, distributed damage process—but its tensile strength collapsed by more than 90 percent at 400 degrees, at which point a specimen failed immediately under the self-weight of the test assembly alone. Across all four loading modes, tensile and shear properties consistently deteriorated faster than compressive ones, indicating that heat attacks interfacial bonding and cellular connectivity before it destroys bulk load-bearing capacity.

The overall comparison upends conventional fire-safety intuition. Within their survivable temperature ranges, the polymeric foams kept strength losses below 50 percent, because they were tested below their glass transition points where softening occurs but cellular geometry survives. Rockwool, despite never distorting, suffered strength and stiffness reductions exceeding 90 percent in most loading modes because its mechanical integrity depends almost entirely on an organic binder that burns away between 300 and 400 degrees Celsius. In other words, the material that best resists fire in the visible sense—no melting, no dripping, no combustible gases—is also the one whose structural contribution vanishes most completely. EPS and XPS, meanwhile, released combustible gases and warped at relatively modest temperatures, yet retained a larger fraction of their mechanical function within the ranges they could tolerate.

The authors emphasize that these residual-property datasets fill a critical gap for engineers and modelers. The measurements can support post-fire performance assessment, inform the selection of insulation cores for precast concrete sandwich panels, and provide calibration data for numerical models predicting how panel structures behave under thermal loading. The team also notes an important limitation: only a single commercial density of each material was tested, and since density strongly influences stiffness, strength, and thermal response, broader parametric studies will be needed to develop generalized design equations. For now, the message for the construction industry is clear and slightly uncomfortable—fire resistance and residual structural performance are not the same property, and the non-combustible hero of building insulation may be the first to quietly stop doing its structural job when the heat rises.

Subject of Research: Residual mechanical properties of EPS, XPS, and rockwool insulation materials after elevated temperature exposure

Article Title: Residual mechanical properties of EPS, XPS, and rockwool insulation materials after elevated temperature exposure

Article References: Anand, A., & Singhal, V. (2026). Residual mechanical properties of EPS, XPS, and rockwool insulation materials after elevated temperature exposure. Results in Engineering, 32, Article 113053. https://doi.org/10.1016/j.rineng.2026.113053

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113053

Keywords: EPS, XPS, rockwool, insulation materials, sandwich panels, elevated temperature, fire safety, compressive strength, shear modulus, precast concrete, thermal degradation, structural engineering