Jelke Dijkstra

Recently, calcined clays have emerged as a viable Supplementary Cementitious Material (SCM) in structural concrete. In contrast, less knowledge exists on the use of clay-based SCM as a binder for stabilisation of natural clays. This study investigates the partial replacement of conventional binders, i.e. Ordinary Portland Cement (CEM I) and lime, with calcined kaolinites for stabilising a natural sensitive clay, aiming to achieve mechanical strength and microstructural evolution comparable to a traditional lime–cement mix (50:50). A holistic approach is adopted to investigate properties across length scales using Electrochemical Impedance Spectroscopy (EIS), miniature Cone Penetration Tests (mCPT), and synchrotron-based X-ray phase-contrast nano-holo-tomography (nano-CT). The results indicate that mixes with up to 20% replacement of conventional binders with calcined clay exhibit comparable strength and material properties to traditional lime–cement stabilised samples, while offering potential reductions in CO2 emissions. The findings further indicate that pore water availability is a key factor influencing hydration and the resulting mechanical properties in binder-mixed clays. This depends on the proportions of lime, cement, and calcined clay relative to the natural water content, with cement having the greatest influence on water demand. Consequently, the binder composition in stabilisation of soft clays can be optimised to achieve enhanced performance by considering the water retention characteristics and reactivity of the components in the mix, ultimately leading to improved mechanical strength.

Low-kaolinite clays represent a globally abundant resource for reducing clinker content, yet their durability under carbonation remains poorly understood. This study examines the carbonation behavior of a 2:1 mixed-layer clay blended binder (30% replacement of OPC), focusing on changes in hydrate assemblage, C–A–S–H structure, and pore network. To isolate carbonation effects, drying was decoupled from CO₂ exposure, and the influence of curing duration was assessed to account for the slow pozzolanic reaction. Results show that carbonation follows the typical sequence of CH depletion, AFm/AFt destabilization, and C–A–S–H decalcification, but its severity depends strongly on curing. Extended curing promoted AFm formation, increased Al incorporation into C–A–S–H, and densified the microstructure, reducing decalcification (Ld ≈ 34% vs. 92% in short-cured samples) and limiting pore coarsening. These findings highlight that carbonation resistance in low-kaolinite systems is governed primarily by hydrate chemistry and microstructural maturity rather than portlandite content, underscoring the importance of curing protocols for durable implementation

Piled embankments are traditionally designed by using either guidelines based on simplified limit-equilibrium theories or advanced finite-element (FE) numerical analyses. Both methods have limitations: the former do not allow the assessment of settlements at the top of the embankment, whereas the latter easily become overly complex, hence limiting practical applications. This paper introduces a new mathematical model capable of reproducing, with minimal computational effort, the mechanical response of piled embankments modelled by means of FEs. The model is based on a set of fundamental principles, assumptions and phenomenological equations obtained from a deep understanding of the mechanics behind the FE problem. The model, evaluating average and differential settlements at the top of the embankment during the consolidation of the soft soil, is validated against full-scale test data and benchmarked against independent numerical results. The results are compared with existing formulas to evaluate the critical height of the embankment, demonstrating the great potential of the new model for engineering practice (giving nearly instantaneous displacement-based solutions for the design of piled embankments in a preliminary stage).

Slope stability analysis in 2D ranges from the classical and conventional limit equilibrium method to the robust and computationally demanding finite element (FE) analysis. Discontinuity layout optimisation (DLO) is an interesting intermediate method that applies an upper bound limit analysis with the assumption of rigid-perfectly plastic soil behaviour. Here, the whole soil mass is discretized using a set of potential slip-lines and optimisation is used to identify the critical mechanism that can be formed from a subset of these lines that dissipates the least energy. This method has only been used for isotropic soil models, except for rare studies that included an anisotropic model. This paper introduces the use of an anisotropic failure criterion in DLO, based on the total stress-based NGI-ADP model. The performance of DLO with this simplified NGI-ADP model is compared with respect to failure mechanism and safety factor determined by corresponding FE analysis. The results show good agreement between the two methods and highlight the use of DLO as a powerful method with straightforward input parameters and low computational time for slope stability assessment.