The majority of bridges in Amsterdam’s historic city centre are built on timber foundation piles, typically 12 to 15 meters in length, which function as end-bearing elements reaching the underlying stiff sand layer. Currently, many timber foundations have been in service for up to 300 years, raising concerns about their remaining load-bearing capacity and the overall safety of the bridges they support. The timber piles beneath the bridges remain fully submerged, where their outer cross-section is exposed to slow bacterial decay in anaerobic conditions, leading to a reduction in the sound load-bearing core of the piles. Despite decay, the piles maintain their structural capacity for many years, although their load-bearing capacity gradually decreases over time. The two primary risks are attributed to the applied loads exceeding the load-bearing capacity of the timber piles, and progressive damage accumulation over time due to sustained loading, ultimately leading to large settlements or failure. In this context, 201 pile segments were extracted from two bridges in Amsterdam and mechanically characterised, with respect to their amount of biological decay and service durations ranging from 100 to 300 years. Large-scale compression tests were carried out to determine the remaining saturated short-term compressive strength of the piles. Micro-drilling measurements were conducted to assess the amount of bacterial decay, validated with Computed Tomography (CT) scanning. On this basis, this study investigates the load history and current loads acting on the timber foundation of Bridge 30 (De Isa van Eeghenbrug) in Amsterdam, to assess the remaining load-bearing capacity of the historical piles, considering mechanical damage and decay as function of time.

This study investigates the fatigue behavior of European ash (Fraxinus excelsior) with density range of 580-620 kg/m3 under relatively short-period cyclic compressive loading. The aim is to understand its mechanical behavior, such as: strain development, recovery, and internal damage mechanisms. Mechanical testings at varying frequencies (0.1, 1.0, and 10 Hz) were performed and the samples were scanned with micro-computed tomography (Micro-CT). Multi-scale assessment of fatigue effects was performed. Strain behavior revealed progressive increases in both viscous and viscoelastic components, demonstrating time- and rate-dependent deformation. For the same loading time, higher loading frequencies resulted in consistently lower accumulated strain; however, no significant differences in strain recovery were observed between frequency groups. Strength and stiffness showed minimal change after up to 2,000 loading cycles at 78 % stress level, highlighting relatively high fatigue resistance of wood under compression. Micro-CT imaging detected internal microcracks in samples pre-loaded with fatigue, inferring that cyclic loading induces more microstructural damage than static conditions. These findings enhance the understanding of the fatigue mechanism in European ash and highlight Micro-CT as a valuable non-destructive tool for internal damage assessment under fatigue loading.

Phragmites australis L., a widespread vegetation in riparian zones such as rivers and canals, is extensively studied for its ecological benefits such as nutrient removal and hydraulic retention. However, its direct contribution to bank stability through root reinforcement, a key factor for its use in soil bioengineering techniques, has received limited attention. This study investigated the root reinforcement provided by P. australis and its root traits at a soil bioengineering test site on a canal bank in the Province of North-Holland in the Netherlands. Direct measurements of root-soil composite strength were performed using a corkscrew extraction technique at two distinct distances from the canal. Concurrently, root distribution parameters, including Root Area Ratio (RAR) and Root Length Density (RLD), were quantified from extracted soil plugs. Root reinforcement was also indirectly estimated using biomechanical models, incorporating measured root tensile strength and root distribution parameters as inputs. A total of 12 excavations, each 0.25 m2, were conducted for comprehensive root trait analysis at both locations. Direct measurements revealed substantial root reinforcement (max 36 kPa; avg 6–19 kPa). RAR showed effective stabilization values between 0.03 and 0.65 %, peaking at 0.65 % in the area close to canal. Root systems were dominated by fine roots (<0.5 mm diameter), comprising >80 % of total root length and creating dense reinforcing networks. Corkscrew measurements yielded conservative values. Modeled estimates significantly exceeded these field measurements, which is consistent with conventional shear testing. The extensive root surface area (>3.9 m2 m−2) further demonstrates the species' soil-binding capacity, with higher values occurring in hydrologically favorable zones. While the ecological implications of using this widespread species must be contextually considered, its pronounced mechanical reinforcement makes it a highly effective biotechnical tool, particularly in managed environments like canals.

Brittleheart, also known as compression failure, is a widespread phenomenon observed in numerous tropical wood species, significantly diminishing their strength properties. According to strength grading standards such as BS 5756, NEN 5493 and EN 16,737, timber exhibiting brittleheart characteristics must be rejected. Oftentimes brittleheart remains undetectable on the outer surface and cross-section of sawn timber. This study focuses on qualitatively characterizing compression failures in tropical hardwood and its mechanical properties. In this context, various non-destructive detection methods were explored. Five grades of compression failures were characterized based on the deformation and displacement of wood tissue. Results demonstrate that CT-scanning shows promising as a technique for detecting these five defined grades. Quantitative assessments of brittleheart on the mechanical properties were conducted to determine bending strength (f_m) and modulus of elasticity (E_m). Multiple regression models were developed to predict the bending strength with a highest coefficient of determination (R²) of 0.778 and a relatively high SEE of 17 N/mm².

By using minerals instead of bioresources in construction, greenhouse gas emissions are nearly doubled. It is vital to transition from this traditional paradigm to a low-carbon model in which woods and bamboos are essential components. To proliferate bio-urbanization, challenges must be overcome in forestry and construction. Our study is a necessary starting point for more sophisticated studies and policies to support the valorization and utilization of bioresources, especially wood and bamboo, in greener construction solutions for a sustainable urbanization. Our main results elucidate examples and benefits of biobased cities and buildings, raise issues and current challenges, and suggest opportune actions to be globally addressed in collaborative proposals. Assertive codes, well-managed resources, resolution of challenges, and clarification campaigns on decarbonization are priority targets in future environmental and societal commitments.

Strength grading is essential to ensure standardized and reliable design processes in timber construction. Current visual strength grading (VSG) standards, such as DIN 4074-1, focus on regulating natural features and grading new sawn timber. Sorting criteria for man-made defects resulting from prior use are lacking. This study examines the potential of VSG for battens processed from salvaged rafters for structural applications and proposes modifications to DIN 4074-1:2012 to make it applicable to battens processed from recovered wood. Battens were visually graded based on natural features, and fastener holes were characterized. Bending strength was tested against EN 338 classes, and the impact of fastener holes was assessed. The results indicate that grades S10 and S13 together account for a total yield of 57%. Clusters of fastener holes reduced bending strength in S10 battens, whereas larger holes affected S13 battens. Modification options for DIN 4074-1:2012 were identified, and their effectiveness was validated. As a result, a new sorting criterion, SRC (strength reducing criteria), was introduced to address the interaction between natural features and fastener holes. Battens processed from recovered wood can serve as an alternative to new timber if classified under the proposed grade W10+.

The lack of strength values for wooden foundation piles in the design standards for timber (Eurocode 5) hinders their proper engineering design and assessment. In order to fill this gap, an extensive experimental campaign was conducted to characterize the mechanical properties of large-scale, water-submerged spruce (Picea abies L.) and pine (Pinus sylvestris L.) piles. This was achieved through the execution of axial compression tests on 253 full-scale pile segments. Wet compressive strength and stiffness values were derived for both spruce and pine piles, applicable to the whole pile and/or its parts: head, middle-part, and tip. The quality variables that most influenced the wet compressive strength of the piles were density, knot ratio (KR), number of annual rings (age), and growth rate. Based on this, characteristic strength values were derived for piles with the following grading limits: KR < 0.5, age between 20 and 100 years, and a growth rate <5 mm/year. These variables were used as key parameters to develop prediction models for the wet compressive strength of spruce and pine piles. The saturated compressive strength values and grading boundaries presented in this study contribute to the engineering design of timber piles and support the integration of reliable design values into future versions of Eurocode 5.

Fatigue failures pose significant challenges across various engineering disciplines. Wood, due to its low carbon emissions and high strength-to-weight ratio, has been gaining attention in engineering applications. The fatigue behavior of wood is complex due to its heterogeneous, anisotropic, and viscoelastic nature. This research explores essential insights into the fatigue behavior of wood, with a focus on S–N curves, stress–strain behavior, and failure mechanisms. Due to often varying failure criteria and test settings, direct comparison of S–N curves across different studies can be challenging and inconclusive. A closer look shows that wood in fatigue shows both irreversible and recoverable strain components that are delayed. However, there have been conflicting reports about residual stiffness changes under fatigue loading. Theoretical fatigue life models based on S–N curves or duration of load theory have shown limited applicability. Efforts to develop progressive damage model based on stress–strain behaviors have been challenging and largely unsuccessful due to the lack or inconsistency of data. Understanding the microstructural failure mechanism is crucial in order to build a more trustworthy fatigue modeling technique. Further work is suggested to monitor the microstructural deterioration during high-cycle fatigue loading.

The lack of strength values for timber foundation piles in the current Eurocode 5 hinders their appropriate engineering design and assessment. Timber piles, often submerged for their entire service life, endure high moisture levels, highlighting the need to define strength parameters of round wood under fully saturated conditions. To ensure reliable material properties, a large-scale study was conducted on 70 European softwood piles, determining strength and stiffness through axial compression tests on saturated segments extracted along the pile. Mean and characteristic wet compressive strength and stiffness values were derived, applicable to the whole pile and/or its parts. The mechanical properties of the piles were analysed in relation to grading parameters that may influence the saturated compressive strength, leading to the classification of three strength classes for visual grading. Additionally, two regression models were developed-one based on the most influencing visually graded parameters, and the other on the dynamic modulus of elasticity. The saturated compressive strength values and grading boundaries presented in this study contribute to the engineering design of European softwood foundation piles in the context of a new circular construction ecosystem, and support the integration of reliable design values into future versions of Eurocode 5.

Enhancing urban tree stability is critical for public safety and infrastructure protection. This study evaluates a nature-based method for improving tree stability using inosculations to form interconnected tree systems. These systems establish biomechanical connections through inosculation, offering both biological and mechanical support. The research focused on lime trees (Tilia Cordata Mill.), comparing parallel and cross connected tree systems with the single tree to evaluate their mechanical performance. The mechanical performance of the interconnected tree systems was evaluated by pulling tests in different directions to simulate wind loads. The study spanned a two-year growth period to investigate the effects of growth on mechanical behavior, with the analysis supported by finite element modeling. The results showed that growth-induced changes increased the overall rigidity of the tree systems and reduced deformation, rotation, and local elongation. Cross connected trees exhibited notable bracing effects in the connected plane, which improved lateral resistance. In a parallel connected tree system, the basal stiffness increased due to the connection between the lower region. Compared to the single tree, interconnecting tree systems can provide additional support and reduce deformation caused by lateral loads, making it a promising strategy to improve tree stability under horizontal loads.

There is an increasing need for using nature based solutions in protecting canal and stream embankments in the Netherlands and delta areas in general. Vegetation provides additional reinforcement and forms an integral part of many nature-based solutions. However, quantifying this reinforcement in-situ is challenging. This study aims to quantify the root reinforcement of three species prevalent along canal embankments – Salix fragilis L. (SF), Salix purpurea L. (SP), and Crataegus laevigata DC. (CL) – using the corkscrew extraction technique. Furthermore, canal bank stability was analyzed under different bank conditions regarding protection (unprotected, protected by vegetation), bank geometry, and hydraulic conditions.

Quantity of roots and Root Area Ratio (RAR) generally decreased with depth for all species. While root breakage was observed in most samples, all species exhibited increased ductility with higher root densities, except for CL at two depths. SF showed higher root reinforcement at shallower depths (≤ 250 mm), while SP demonstrated greater reinforcement at deeper depths. Results demonstrate that the corkscrew extraction technique is a quick and minimally invasive method for measuring root reinforcement in riparian environments.

Bank stability simulations revealed that vegetation significantly increases the stability of canal banks. Notably, when considering measured root reinforcement, the factor of safety improved dramatically from 1.08 to 2.46, even under analyzed worst case conditions. However, the analysis suggests a limiting root reinforcement beyond which further increases in root reinforcement have minimal impact on stability. Monitoring using the corkscrew apparatus and future design approaches could aim to achieve this minimum reinforcement.

The strength degradation resulting from duration-of-load (DOL) effect and bacterial decay poses significant challenges to historical timber piles. Many historical European cities still heavily rely on the infrastructure supported by their original timber foundations. A reliable modelling approach on the structural performance of timber piles is needed to avoid the economic loss caused by closing down infrastructure. In this work, we consider a simplified bacterial decay model and develop a numerical framework to integrate the decay model into a standard DOL model. Two approaches are proposed and compared: one considering the homogenised effect of bacterial decay over the entire cross section, and the other taking into account the localised failure accelerated by bacterial decay and applying stiffness reduction to allow stress redistribution. Although the homogenised failure criterion is found to potentially underestimate the effect of bacterial decay, both approaches are able to capture the designated decay pattern. Ultimately, there is a potential for future extension to more intricate loading conditions and decay patterns.

This paper presents an experimental study on predicting the remaining short-term compressive strength of timber foundation piles in Amsterdam using micro-drilling. A large-scale investigation was conducted on 201 pile segments fromtwo bridges, dating back to 1727, 1886, and 1922.Microdrilling measurements, supported by a TU Delft-developed algorithm, successfully assessed decay by calculating the degraded portion of the cross-section—the soft shell. Piles from 1727 exhibited approximately 50% strength reduction due to significant, consistent decay along their length.More recent piles from 1922 and 1886 exhibited a 10-15% strength reduction despite lower decay levels. The correlation between decay and strength led to the development of an experimental prediction model for the in-situ short-term strength of the pile head, middle section, and tip. These findings aid the city of Amsterdam in estimating the remaining service life of its timber pile foundations.

Extending the service life of building components is essential for a circular economy. Wood, as a renewable raw material, plays due to its mechanical properties and ease of processing a crucial role in this process. Most studies focus on the reuse of building materials. However, it is essential to detect and investigate the use cases in which reuse is impossible due to changing dimensional requirements or damages. This study examines the bending properties of recovered wood, particularly battens with cross-sectional dimensions of 30x50 mm², which were processed from rafters originating from a roof truss deconstructed in southern Germany. The bending tests were performed and interpreted based on the damages of the prior use and the lumber pieces' background information. The visual observation resulted in many fastener holes, mainly derived for battens from the built-in upper layer of the rafters. Even though fastener holes contributed to or were the single cause for the failure of the battens, bending strength around the mean value or even higher was yet achieved for some battens. Developing unique sawing patterns for each rafter by taking into account the location of the pith and the arrangement of knots can enhance the yield. Additionally, introducing a third grade, S7, alongside the existing S10 and S13 grades - similar to the approach used for joists and boards in DIN 4074-1:2012 - could further optimize yield. Although it has been concluded that knots remain even for recovered wood the key sorting criteria, fastener holes, can additionally influence the mechanical properties and, therefore, need to be considered in a standardized strength grading.

Trees can adapt to external loads and form inosculations (self-growing connections), where stems or branches naturally fuse together. However, a limited understanding of biomechanical features of connections hinders their practical applications. This study used connections formed by Ficus benjamina L. to investigate their mechanical properties at different growth levels. Two parameters (fusion degree and interface curvature) were identified to describe growth levels. Customized tensile tests were designed to measure mechanical properties perpendicular to the interconnected surface. Growth levels of studied connections ranged from initial formation to almost fusion of piths, which provided a range of tensile strength of 0.23 to 1.38 MPa. Two primary failure modes (failure at the interface and failure across the stems) were found to be linked to growth levels. The fusion degree, at approximately 15%, contributed to distinguishing failure modes. The average diameter of a connection had the most significant effect on its tensile strength and stiffness. Moreover, the interface curvature correlated negatively with mechanical properties. Average diameter, interface curvature, and fusion degree were effective predictors of connections’ tensile strength. Regarding Ficus connections, dry connections were stronger than wet connections. These findings provide evidence for nature-based design using self-growing connections under different moisture conditions and growth levels.

Hundreds of kilometers of timber sheet piles protecting the banks of canals of the Netherlands are nearing the end of their service life and need to be replaced. Even though azobé wood is very widely used, little is known about the current state of the azobé timber sheet piles that have been in service for many decades. More information about the current strength is necessary for planning any intervention, maintenance or reuse of the recovered sheet pile boards. This study aims to characterize the current state of azobé sheet piles using destructive techniques such as four point bending tests, compression tests and non-destructive techniques such as micro-drilling and visual assessment. To achieve this objective sheet piles that were in service for over 57 years were pulled out and tested in the laboratory. The non-destructive tests results indicate that the deterioration of azobé sheet piles is concentrated on the superficial layers of the boards. Visual classification and micro drilling techniques did not yield results that support the findings from the destructive tests. The bending strength and modulus of elasticity of in service sheet piles used in current study was found to be lower by about 25 % and 30 % respectively when compared to new azobé sheet piles reported in literature. Based on average values of measured dimensions, density of the sheet piles in this study was in general, lower compared to new sheet piles. Thus, the lower strength could be due to deterioration, lower intrinsic quality of the recovered sheet piles, or simply fall within the natural scatter of the material. In addition an exercise to classify the samples to a strength class is shown for practicing engineers.

Laminated bamboo can be produced in sizes that are similar to glued laminated timber. Asaresult, large connections with multiple dowels and slotted-in steel plates are similarly possible with bamboo. MOSO bamboo was used in this study, withadensity of around 660 kg/m³, potentially creating connections having higher load carrying capacity than softwood. A large experimental campaign was set-up in order to determine the mechanical properties of connections with various ratios of dowel diameter to bamboo thicknesses and with single and double steel plates. Furthermore, influences of the density of the material, related to the embedding strength for fasteners, as well as the splitting sensitivity with multiple fasteners inarow are playing crucial roles with respect to the load carrying capacity. Therefore, multiple test series on large bamboo connections have been performed in order to study various possible failure modes, as dependent on embedding strength, steel grade, number of fasteners inarow, and the influence of multiple steel plates. The various failure modes have been analysed analytically with the Johansen equations, similar to the design equations proposed for the upcoming version of Eurocode 5 for multiple steel plate connections, confirming their applicability to bamboo and its similarity with wood.

Reliable finite element simulation of orthotropic-dependent failure mechanisms is crucial for understanding the mechanical behavior and optimizing engineered composites and fiber-based materials. Such materials behave brittle under tension and strongly depend on the orthotropic material orientation. Existing non-local models can reproduce brittle fracture for isotropic materials but, in most cases, they are based on the equivalent strain concept for damage initiation, which is unsuitable for orthotropic materials. This contribution introduces a stress-based non-local damage model enhanced with an implicit gradient formulation of the failure criteria. A localizing non-local length is assumed to avoid any pathological broadening of the damage band. The methodology introduces direction-dependent damage variables driven by non-local stress-based damage criteria and can thus distinguish different failure modes. The verification and validation are shown on numerical and experimental benchmark examples. The implicit gradient-based non-local damage approach allows mesh-independent results. Furthermore, it does not require a priori known crack paths and makes it possible to simulate complex failure modes. Perspectively, its effective implementation in the commercial software Abaqus and combination with other constitutive laws, e.g. to account for plasticity or moisture, make it an attractive tool for describing the mechanical material behavior of orthotropic materials, such as wood and fiber-composites.

This contribution aims to increase the understanding of the complex mechanical behavior of wood through a framework for simulating mixed-mode failure. Based on physical properties assessment, appropriate constitutive laws, and experimental validation, a generally applicable numerical strength prediction tool for wood from different species and with various natural imperfections is introduced. The 3D orthotropic elastic plastic non-local CDM model considers the local fiber orientation and is implemented as material subroutines in the commercial software Abaqus. Herein, orthotropic Hill-plasticity with exponential hardening represents the plastic behavior in compression. Separated stress-based gradient-enhanced transient non-local damage represents the brittle material behavior in tension and shear. The methodology is validated with experimental data on tensile veneer tests, shear- and compression tests. Moreover, the methodology is applied to four-point bending tests of boards with heterogeneities. The numerical results demonstrate that the proposed model is able to reproduce different crack patterns observed in the four-point bending tests. Detailed investigations of the impact on the strength of the boards can be performed with this method to optimize species-independent strength prediction and engineered wood products. Further combination with other material laws e.g. moisture is possible.

Traditional "hard" protection systems, such as hardwood timber sheet pile walls, are often used to protect banks of canals and streams, but the tropical hardwood they require is not always locally available. This has led to increasing interest in nature-based, bio-engineered solutions that combine locally sourced wood with vegetation to protect the soil. To assess the behaviour of locally available softwood timber sheet pile walls, a full-scale surcharge loading test was performed under realistic conditions. The test applied a 30 kPa surcharge load, representing the weight of a heavy agriculture machinery, while monitoring the wall's horizontal and vertical displacement, along with its rotation at the top, mid-height, and base of the retained soil. This resulted in a displacement of approximately 1.9% of the one meter retaining height. The potential onset of a failure wedge was observed after an extended loading period. Nonlinear tilt measurements showed peak curvature at mid-depth (0.66° top, 0.71° mid, 0.69° bottom), indicating dominant flexural bending. Additionally, the measured horizontal displacement exceeded the rotational contribution estimated from the tilt. The material properties of the softwood sheet piles were determined through four-point bending tests. A numerical model, calibrated with experimental data, was then developed to simulate the long-term performance (10 years) of decayed sheet piles with both bare and vegetated backfill. The results indicate that vegetated backfills significantly reduce displacement and the bending moment on the wooden sheet pile compared to bare soil.