The problem the researchers set out to solve is one of the most common triggers of geological disaster worldwide. When sandstone slopes develop vertical fissures, rainwater penetrates deep into the cracks and dramatically reduces the shear strength of the rock mass, setting the stage for catastrophic sliding. The standard remedy, filling cracks with cement grout or chemical slurries, works only temporarily and comes with heavy costs: high energy consumption, substantial carbon emissions, and the risk of secondary pollution. Cement-based fillers are also poorly suited to shallow, low-stress surface layers, where their expense and rigidity offer little ecological or economic advantage.

The team’s insight was to look at what the region already had in abundance. Oyster shells, composed mainly of calcium carbonate, are piled along the shores of the Beibu Gulf, where open stacking and landfilling consume land and threaten coastal environments. Expansive soil, meanwhile, blankets much of Guangxi and is notorious in geotechnical circles for its swelling-shrinkage behavior under alternating wet and dry conditions, which causes foundation deformation and shallow slope collapse. All three sampling sites, the cracked sandstone slope, the expansive soil deposit, and the shell-strewn coastline, lie within 30.7 kilometers of one another, meaning the raw materials for crack repair can be sourced locally, cutting transport costs and improving sustainability.

The recipe is deceptively simple. Crushed oyster shell powder and sieved expansive soil were mixed with octadecylamine, or ODA, an alkylamine compound with a long hydrophobic carbon chain, and heated to 70 degrees Celsius for an hour, just above ODA’s melting point. In its liquid state, the molecule spreads easily over soil particles. The researchers tested ODA dosages of 0.8, 1.6, 2.4, and 3.2 percent by mass of dry soil, and oyster shell powder contents of 10, 20, and 30 percent, screening each combination through standardized water-repellency and swelling tests.

Water repellency was assessed with two complementary measures. The water droplet penetration time, or WDPT, records how long a standard droplet sits on the soil surface before soaking in, while the contact angle measures the angle between the droplet and the solid surface. At the lowest ODA dosage of 0.8 percent, performance was erratic: contact angles swung from 86.7 degrees down to 24.4 degrees over twelve drying-wetting cycles, and penetration times fluctuated between 91 and 409 seconds. At 2.4 percent ODA, however, the material achieved durable superhydrophobicity, maintaining contact angles above 90 degrees and WDPT values exceeding 3600 seconds, the threshold for extremely water-repellent soil, throughout all twelve accelerated wet-dry cycles. Because 3.2 percent ODA performed no better, 2.4 percent was selected as the optimal, most economical dosage.

The oyster shell powder plays a different but equally important role. As an inert, coarser-grained filler, it dilutes the swelling montmorillonite clay minerals and occupies pore space within the soil. The free swelling rate of the expansive soil dropped from 92 percent in its raw state to 48 percent with 30 percent shell powder added. Shell powder alone also reduced the surface crack ratio after drying, from 25.4 percent for untreated soil to 19.3 percent. But the real breakthrough came from the synergy between the two additives: when 2.4 percent ODA was combined with 30 percent shell powder, the desiccation crack ratio plummeted to 5.47 percent, a 78.5 percent reduction compared with untreated expansive soil, far outperforming either additive on its own.

To understand why the composite works so well, the team turned to a battery of microscopic techniques. Scanning electron microscopy revealed that the sharp-edged, porous lamellar structure of raw expansive soil became coated with a continuous amorphous film after ODA treatment, with particle edges softened and micropores blocked. X-ray diffraction showed the montmorillonite basal diffraction peak shifting to a lower angle, indicating that ODA’s long alkyl chains had intercalated into the clay’s interlayer spaces, expanding the basal spacing by about 9.5 percent, from roughly 1.47 to 1.61 nanometers. This nanoscale intercalation, confirmed by infrared spectroscopy through new CH2 absorption peaks and weakened interlayer water bands, fundamentally converts the hydrophilic clay mineral into a water-repellent one. Thermogravimetric analysis added further evidence, showing reduced water loss in modified samples and a characteristic decomposition peak of ODA’s organic chains near 320 degrees Celsius.

Mechanical testing reinforced the practical value of the modification. In unconsolidated undrained triaxial tests after twelve drying-wetting cycles, raw expansive soil lost cohesion steadily, falling from 20.4 kilopascals to 14.3 kilopascals, while the composite material retained far higher strength values, with differences between the two reaching as much as 220.5 kilopascals at the eighth cycle. Because water barely infiltrates the hydrophobic composite, the wetting and drying that normally degrades clay strength simply cannot reach the material’s interior, preserving its shear parameters over time. Soil-water characteristic curves also showed that at low water contents, below 20 percent, the modified soil sustains significantly higher matric suction, since only its large pores participate in water migration while the small hydrophobic pores resist saturation.

The decisive test came from a 134-day field-scale model experiment. The team built two artificial model cracks, each 150 centimeters long, 20 centimeters wide, and 100 centimeters deep, filled one with untreated expansive soil and the other with the composite material, and instrumented both with moisture sensors at depths of 5, 25, 45, 65, and 85 centimeters, recording data every two hours through both rainy and dry seasons totaling 1055 millimeters of rainfall. The contrast was stark. In the untreated crack, a 210-millimeter-per-day storm raised saturation at 25 centimeters depth from 25.4 to 77.2 percent within 13 hours, and within roughly 60 hours water had percolated the full 85-centimeter depth, eventually saturating the bottom layer completely. Saturation fluctuations across depths ranged from 34.2 to 75.1 percentage points. In the composite-filled crack, by comparison, fluctuations never exceeded 7.2 percentage points at any depth, and only 2.2 points at 85 centimeters, demonstrating that rainfall had essentially no influence on the material’s interior.

The researchers describe the approach as treating hazards with waste: a surface barrier of hydrophobic composite material placed at the top of a crack intercepts rainfall recharge before it can penetrate, converting two problematic local waste streams into a value-added geotechnical product. Beyond the engineering performance, the environmental accounting is compelling, since the method avoids the carbon emissions of cement grout production while reducing the methane emissions and land occupation associated with shell dumping. With raw materials that are abundant, cheap, and locally available, and with the 134-day field observation validating the laboratory’s accelerated wet-dry cycling as a realistic proxy for natural conditions, the technique offers a genuinely green route to shallow landslide prevention, one that could be replicated wherever coastal shell waste and expansive soils coexist.

Subject of Research: Development of a hydrophobic crack-filling material from waste oyster shells and expansive soil for slope stabilization

Article Title: From waste to a watershed-scale solution: Transforming oyster shell and expansive soil into a hydrophobic barrier for intercepting rainfall recharge and stabilizing slopes

Article References: From waste to a watershed-scale solution: Transforming oyster shell and expansive soil into a hydrophobic barrier for intercepting rainfall recharge and stabilizing slopes. (n.d.). https://doi.org/10.1007/s12665-026-13149-z

Image Credits: AI Generated

DOI: 10.1007/s12665-026-13149-z

Keywords: oyster shell waste, expansive soil, octadecylamine, superhydrophobic soil, slope stability, landslide prevention, rainfall infiltration, montmorillonite, geotechnical engineering, solid waste recycling, water repellency, crack grouting

The urgency of this work is rooted in global urbanization trends. According to United Nations projections cited in the study, roughly 45 percent of the world’s population currently lives in urban areas, a figure expected to climb to about 60 percent by 2050. As cities expand vertically and laterally into topographically complex terrain, engineers must build on geologically difficult ground where slope failure threatens not only infrastructure but also environmental goals. Landslides degrade land, strip away vegetated cover, disrupt drainage systems, and expose urban communities to escalating geohazard risk, directly undermining the United Nations Sustainable Development Goals on sustainable cities and life on land.

The study’s setting is the Almaty Basin in Kazakhstan, a region flanked by the Zaili Alatau Mountains and blanketed in thick Quaternary sedimentary deposits. These soils, formed from the weathering of deeply buried Paleozoic bedrock, consist of heterogeneous mixtures of pebbles, sands, sandy loams, and clay-rich loams. Laboratory testing on soil sampled from Almaty at a depth of four meters classified the material as low-plasticity silty clay under the Unified Soil Classification System, with an effective friction angle of just 19 degrees and moderate cohesion of 33 kilonewtons per square meter. These weak shear parameters mean that failure surfaces tend to develop as shallow mechanisms within the weathered soil layer rather than in the intact rock below, particularly when rising pore water pressure accompanies rainfall infiltration.

To confront these conditions, the researchers turned to the Geobarrier System, or GBS, an integrated slope protection technology that evolved from the capillary barrier concept. A capillary barrier exploits the hydraulic contrast between an overlying fine-grained soil layer and an underlying coarse-grained layer: water infiltrating the fine layer is retained and diverted laterally by capillary forces rather than percolating downward into the slope. While effective, conventional capillary barriers are limited to slope inclinations of up to 45 degrees. The Geobarrier System overcomes this by adding retaining elements, drainage provisions, and geosynthetic reinforcement, allowing stable designs on slopes as steep as 70 degrees. Its facing is built from geobags filled with an Approved Soil Mixture that supports vegetation, turning the structure into a green, nature-based solution rather than a bare concrete wall.

In standard GBS practice, biaxial geogrids are connected to the geobags and extend into the soil mass behind the structure, mobilizing tensile resistance through frictional interaction with the compacted backfill. The customary minimum embedment length is 0.7 times the height of the structure, a figure consistent with guidance from the Federal Highway Administration and AASHTO for geosynthetic reinforced soil systems. Shortening the geogrid saves material and cost, but it risks insufficient anchorage: if the reinforcement does not extend beyond the potential failure surface, tensile stresses cannot be transferred into the stable region of the slope, and deformation escalates. The central question of the new study was whether steel J-pin anchors attached to the free ends of shortened geogrids could restore the lost stability.

Answering that question required a numerical model capable of simulating three distinct interactions simultaneously: soil against geogrid, geogrid against J-pin, and soil against J-pin. The team selected the general-purpose finite element software ABAQUS over the geotechnical package PLAXIS because ABAQUS offers the flexibility to model structure-to-structure contacts directly, something PLAXIS handles only through predefined interface elements. The researchers used ABAQUS/Explicit with central difference time integration to handle the severe geometric nonlinearity and complex contact conditions, while carefully monitoring the ratio of kinetic to internal energy, which stayed below 0.8 percent throughout the analyses, well under the 5 percent limit accepted for quasi-static behavior. Factor of safety values were computed with the strength reduction method, progressively scaling down the soil’s cohesion and friction angle until failure occurred, automated through a Python-based scripting framework.

The model represented a 12-meter-high, 25-meter-wide block of Almaty soil supporting a 4-meter-high GBS with eight reinforcement layers, a 70-degree face inclination, and a 10 kilonewton per square meter surcharge on the crest. The geogrid properties matched those of the Miragrid GX-80/30 product, providing 80 kilonewtons per meter of longitudinal strength, while the J-pins were modeled as 12-millimeter-diameter S275 steel rods, 0.5 meters long, with bilinear elastoplastic behavior. Before adding pins, the team validated their ABAQUS setup against PLAXIS 2D for the baseline 0.7H configuration and shortened variants of 0.6H, 0.5H, and 0.4H. Deformation differences between the two platforms ranged from 0.3 to 6.63 percent and factor of safety differences from 0.97 to 3.2 percent, comfortably within the 10 percent tolerance considered acceptable when comparing different computational platforms.

The validation runs alone revealed a stark pattern. Cutting the geogrid from 0.7H to 0.6H raised maximum deformation only modestly, from 6.98 to 7.37 millimeters, but shortening to 0.5H pushed deformation up by roughly 52 percent to 10.58 millimeters, and 0.4H produced a 140 percent increase to 16.74 millimeters. The factor of safety fell in parallel, from 1.34 at 0.7H to 1.24, 1.15, and finally 1.09 at 0.4H. When J-pins were then added one layer at a time, starting from the lowest geogrid, the results split cleanly. For the 0.6H model, deformation decreased steadily up to five pins before a slight rebound, and the factor of safety peaked at the same five-pin configuration. The 0.5H model improved up to six pins. The 0.4H model, however, showed only marginal gains, with its factor of safety hovering near unity regardless of how many pins were attached.

Force distribution data explained why. With eight pins installed, the J-pins carried 17.2 percent of the total reinforcement force in the 0.6H model and 19.3 percent in the 0.5H model, evidence of effective composite load transfer between geogrid and anchors. Individual pin contributions were non-uniform: middle layers, which experience the greatest driving forces, mobilized the most tensile resistance, while the topmost and bottommost pins contributed the least. In the 0.4H model, by contrast, the pins carried only 13.3 percent of the force and their contributions were evenly spread, a signature of a geometry-controlled failure mechanism in which the shortened geogrid simply cannot mobilize meaningful tensile resistance beyond the slip surface. The conclusion is unambiguous: anchorage density can compensate for moderate geogrid shortening, but below roughly 0.5H no number of pins can rescue the design in weak colluvial soil.

The team put their findings to the test in a pilot-scale field construction on a 3.5-meter slope, using the 0.5H design with five J-pins attached to the lowest five geogrid layers. Recycled concrete aggregates served as both the fine and coarse capillary barrier layers, underscoring the sustainability credentials of the approach, and young plants were established in vegetation pockets on the structure’s face. Numerical analysis of the as-built design predicted a maximum deformation of 6.5 millimeters and a factor of safety of 1.424, exceeding the long-term stability minimum required by British Standard BS 6031. One month after planting, the vegetation had grown into a substantial green cover, and the structure has shown no significant cracking or movement since completion. The authors caution that their model assumes idealized pin anchorage without progressive soil yielding around the shaft, so field-scale monitoring remains important, but the demonstrated threshold of 0.5H with five pins offers designers a practical, material-efficient rule for deploying geobarrier systems on the landslide-prone clay slopes found not only in Almaty but across comparable mountainous terrain worldwide.

Subject of Research: Hybrid geogrid and J-pin reinforced geobarrier systems for landslide prevention on clay-rich colluvial slopes

Article Title: Landslide preventive approach incorporating hybrid geogrid-J-pin reinforced geobarrier system

Article References: Landslide preventive approach incorporating hybrid geogrid-J-pin reinforced geobarrier system. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: landslide prevention, geobarrier system, geogrid reinforcement, J-pin anchorage, slope stability, capillary barrier, finite element analysis, Almaty Kazakhstan, colluvial soils, strength reduction method, geosynthetics, nature-based solution