Fully glazed offices are being questioned due to issues with heat gain, view obstruction from shading, and glare. Energy-efficient facades should allow daylight penetration while preventing excessive solar heat. These fully glazed offices also have a monotonous appearance and, therefore, lack design freedom.

This thesis proposes a novel hybrid panel combining a transparent (glass) and an opaque (concrete) material in a single prefabricated unit without adhesives to create a new design language.

Since concrete and glass lack strong natural adhesion, this research focuses on optimizing the interface between these materials to enhance adhesion and structural integrity. Key parameters investigated include material composition, interlocking geometry, and surface roughness. Experimental testing revealed that combining interlocking forms with surface roughness significantly improves adhesion strength. Specifically, surface roughness combined with an interlocking form characterized by a repetitive series of peaks yielded the best results for robust interfacial adhesion.

Design guidelines were established to minimize stress concentrations in facade panels, ensuring sustainable solutions by enabling disassembly and recycling. This innovative hybrid panel represents a significant advancement in facade technology, blending the opacity of concrete with the transparency of glass in a free-form manner without the use of additives.

Glass is a material that is used in many different industries nowadays, including digitalisation, telecommunications, transportation, and architecture. Despite being widely used, the manufacture of flat glass presents serious environmental issues because to its energy-intensive methods, such as the float method.

Glass recycling and reuse have been explored to mitigate these environmental impacts. Incorporating cullet, into manufacturing processes has shown substantial reductions in energy consumption and CO2 emissions. However, challenges such as financial, infrastructural, and technical limitations hinder the increased use of cullet, especially from post-consumer sources.

Creative thinking is essential to overcome these challenges. In line with the EU's zero C&D waste goals, researchers aim to maximise the use of recycled glass in building construction. Various approaches to reusing waste glass in construction are under investigation by research organisations like Delft University of Technology. Methods such as casting allow for volumetric designs and can accommodate higher levels of impurities and contaminants. Since float glass typically breaks due to surface imperfections and flaws, enhancing surface quality is crucial while the bulk can tolerate lower quality. This concept has led to the development of composite panels, where the surface contains higher purity cullet and the bulk contains lower purity cullet.

Despite these advancements, there is still a significant gap in understanding the optimal ratio between the surface and bulk layers to achieve the best structural performance while maximising recyclability. Furthermore, the specific material compositions for both the surface and bulk layers remain unknown. Current research aims to address this gap by investigating the impact of various factors, including glass material composition and the thickness of both surface and bulk layers, on the performance of composite glass panels.

Research on glass casting and recycling is essential for sustainable development since it presents viable answers to environmental problems facing the building industry. This study focuses on experimental methods that are essential to the advancement of casting and glass recycling.

The research aims to increase glass recycling operations through investigations into experimental variables using cast glass recycled beams. To better understand structural behaviour, the experimental methodology includes testing both homogeneous and composite glass beams. In order to maximise surface-to-bulk ratios, various material compositions of surface and bulk and layering techniques are investigated. Techniques for both mechanical and microscopic validation are used to assess beam performance and to understand variables affecting structural integrity.

Recycled Composite Cast Glass Panels made of C&D waste are an example of a new solution for sustainable building applications that can be produced by improving beam configurations and understanding material behaviour. This helps the building construction industry make the shift to a circular economy.

Glass, a transparent and durable material with significant structural potential, is extensively used for architectural and structural applications in buildings. Existing approaches mostly use float glass, which is confined by planar designs, resulting in two-dimensional facades. Several examples of curved glass applications exist which, due to the limitations of current fabrication techniques, result in repetitive panels. Emerging architectural trends demand more fluid, three-dimensional freeform shapes, which present glass shaping processes cannot provide without incurring excessive costs and material waste in the fabrication process.

Research in other materials shows that flexible moulds are able to produce complex geometrical results. More recently, knitted fabric formworks have proven to be very successful in creating concrete shells of high complexity and low material usage and waste.

This thesis aims to address the gap by investigating the viability of using knitted basalt moulds for glass slumping, a novel fabrication approach that might allow for the manufacture of customizable, freeform curved glass components. The research investigates the possibility of this technology to accomplish extreme geometries while remaining simple and cost-effective in mould manufacture. The methodology of this thesis consists of a thorough literature research, experimental testing and a design application. The literature study investigates and compares present glass curving technologies and also the moulds used for each method to visual, geometrical, structural, and sustainability criteria. Flexible mould applications in other materials are also researched and compared to the ones used for glass.
The second part of the thesis is an experimental exploration. Basalt yarn is chosen to create hand-woven and CNC-knitted moulds to test with glass slumping. The experimental phase involves testing several knitted basalt moulds and combinations with coatings to determine ideal material combinations in order to improve surface quality and geometric precision. After reflecting on the experiment results, experimental data is used to create a final prototype attempting to achieve geometrical control and repeatability to prove the potential of the proposed
fabrication technique.

Finally, the third part of the thesis is a design application. Casa da Musica in Porto is chosen as a case study to recreate its glass façade in a freeform shape. Connection details between the glass panels are designed for the proposed facade design and the proposed fabrication of each panel is showcased.

Current research conducted at TU Delft focused on utilizing structural Topology Optimization (TO) for designing large monolithic cast glass structures with maximum stiffness and minimum mass. These structures demonstrated improved manufacturability in terms of time, energy, and cost efficiency due to their mass efficiency, which results in shorter annealing times. However, the complex geometries and customization of these forms posed challenges in terms of fabrication.

The manufacturability of intricate glass structures is explored by analysing and comparing three possible fabrication methods for three-dimensional glass structures with complex and customized geometries. The methods examined are:

(i) Kiln casting in disposable moulds,
(ii) Waterjet cutting and lamination of float glass panes,
(iii) Additive manufacturing of glass.

The assessment of these methods is based on a set of criteria related to structural performance, visual quality, fabrication limitations, and sustainability. This comparative study act as a guide to the design of a case study and the selection of the preferred fabrication method. An all-glass topologically optimized bridge observatory in Vikos Gorge, Greece, is chosen as the design case. Based on the comparative study and a set of soft criteria, casting in disposable moulds is selected as the preferred fabrication method.

However, glass casting currently faces major drawbacks that restrict its potential. The two main drawbacks that this thesis tries to address are:

Rough and opaque surface quality (main focus): This issue is tackled through laboratory experimentation with the aim of achieving good surface quality immediately after demoulding. The experiments involve the use of different types of disposable 3D printed sand moulds (3DPSM) and application of refractory coatings and coating combinations at various maximum firing temperatures.

Lack of redundancy: Redundancy is explored though research by design and the implementation of design strategies (segmentation, zoning, fabrication methods combination) to ensure the feasibility of the structure.

The end result of this thesis is a comprehensive study on how an all-glass structure with complex and customized shape can be realized. The experimental part of the research yielded improved results, indicating that the combination of refractory coatings and 3DPSM has the potential to bring such glass components to life, reduce the need for post-processing, and simplify the fabrication process.

Given the limitations of time and knowledge within a master’s thesis, further research is suggested to validate and evaluate the results obtained.

Dense urbanization increases the demand on the building sector who is responsible, on the one hand for producing large amounts of landfilled glass waste, and on the other for consuming the earth’s natural resources and release additional CO2 emissions, for the manufacture of new architectural glass. These unfavorable conditions together with the already apparent effects of climate change, demand resourceful solutions for adaptive and resilient cities, that utilize current waste and require low maintenance and low costs, to be implemented universally.

This, can only be ensured by letting nature to take over. Unexploited urban facades have a key role to provide this new ground that can be colonized by microorganisms, creating a microclimate by forming an outer green layer growing on its own. Aiming this, materials covering our urban facades need to be transformed to porous hosts by retaining rainwater and providing the right environment for bio-growth to take place.

This research aims to discover new ways that unutilized glass waste can be upcycled into bioreceptive applications, that form a promising set of criteria. By shedding light firstly on the specific material properties needed, the method of glass foaming is chosen to be investigated, as the means to provide an open porous network that can incorporate large amounts of waste into its recipe.

An experimental approach has been designed to explore the parameters related to the mixture, manufacturing process affecting the glass foam’s microstructure and potential biofilm formation by producing a total of 22 samples in the Glass lab in Stevin lab, TU Delft. These specimens were not foamed at once, but gradually being tested first of all, microscopically, to reveal the porosity network and secondly, with a series of quick tests related to their hydraulic performance, providing feedback for the next batch of samples regarding the most promising recipes to be further explored. All the samples were tested for their water absorption, evaporation rate and frosting resistance, while only the higher-scored specimens were put under test for moss-growth and compressive strength.

Apart from the experimental analysis, the next steps towards a product development were also explored by setting the bioreceptive design principles for manufacturing the meso-scale surface. Limitations in adjusting these guidelines to the way glass-foam is produced are addressed with possible solutions as suggestions for further experimentation. In addition, an application catalogue was composed as schematic recommendations to showcase the potential of bio-host glass. Combining this idea to the material science, these applications were also approached from an engineering view to analyze the material properties’ specific demands per product. This tool, can prove to be beneficial both in the hands of the future designer and material researcher to have a starting point on what needs to be further developed, depending on the chosen product out of bioreceptive glass-foam.

Taking the most immediately feasible example of a façade tile, based on the findings of the aforementioned analysis, during the last part of this project, a methodology for designing and implementing the new material into the urban environment is proposed. By informing the design of the macro-scale with weather data, precipitation levels can be exploited, in order to provide the maximum water content for the benefit of bioreceptivity.
Therefore, the novelty of this thesis, aspires by providing a holistic approach, based on the knowledge obtained both in the literature review and the conducted experiments, to stir not only bio-growth on our cities, but also innovative thinking and exploration on ways to combat the increasing landfill waste.

Although glass is a known material for centuries, extensive usage and today’s technology –with modern construction methods– made glass one of the most valuable materials of the present and future. Witnessing an immense progression and a broad innovative boost in the construction field, creating the perfect link between indoor and outdoor environments.

Given the growth of the world’s population and its constant need to improve the way of living, the demand for new flat glass production, and the replacement of the older ones are rising. However, what is happening at the end-of-life of such architectural glass waste coming from the construction and demolition sector? Α significant amount of post-consumer flat glass waste is generated, which is either down-cycled into low-value applications or discarded at the end of the life cycle, aggravating the existing problem of glass recycling whilst resulting in a significant impact on the ecological footprint.

Glass is a readily recyclable material, that can be remelted and reformed indefinitely into articles with the same quality and properties as the original one, nominating it as a perfect candidate in the transition efforts for a low-carbon environment. Even though the glass packaging industry is a successful recycling market, the same is not true for the floating glass industry, which occupies a more complex position. The vast majority of the remaining waste is rarely recycled into the same product, due to the lack of an organized recycling scheme, the incompatibility of different glass recipes, and the contamination rates of glazed products that could lead to risks in the total production line.

Among the down-recycled or landfilled glass, there is a great proportion of high-value float glass, pointing out the large potential in upgrading the glass recycling process from a linear chain to a circular one. The main scientific contribution of this research is the development and experimental verification with the aim of casting technique, not only new design concepts and engineering their fabrication following the design criteria, but also an ideal recycling collection system for cullets coming from the C&D sector while utilizing tools to check the product’s performance and assessing its life cycle impact to the environment.

The casting method, as a more flexible process, that easily shifts between different recipes, and is less energy-intensive than float line, proves feasible for the fabrication of 100% recycled panels out of post-consumer flat glass waste. These findings are introduced by means of closed-loop alternatives that extend their service life, establishing at the same time the foundation for a circular life-cycle of architectural glass, while can assist as guidelines for further exploration. These components are monolithic elements with higher thickness, compared to the conventional thin-walled glass to tolerate a higher contamination rate, intended for building envelope applications. Their format is adopted and explored through a strategic approach of a translucent composite product, which is made out of specific glass waste layering arrangement between different purity grades of cullet in such a manner to form an advantage tensile strength on the recycled panel, which is transferred on the surfaces and reinforces the object.

Grasping the boundaries of glass recycling, which has been continuously stretching, in the last years, only some primary attempts have been made by TU Delft. These are referred on the fabrication of 3-dimensional strong glass components or glass panels emphasizing the aesthetic result always with the method of casting, unveiling the ambitious potential of glass upcycling for the building sector.

Identifying the scientific gap, the focus of this research is the proper mapping of the current situation of glass recycling, and the identification of the most promising glass C&D waste through experimental research by contacting melting and tools that have not been utilized before. The experimental analysis of the new concept concentrates on the development of the composite panel and the thermal shock testing in uneven temperatures, as the strength of recycled glass has already been investigated previously. Additionally, this concept will be assessed for further analysis and understanding of the whole effectiveness of the proposed supply chain of this recycling method, which is carried out for the first time.

The development of the new composite panel and its experimental evaluation displayed strong findings for further exploration. This thesis set-up could be a promising strategy for float glass recyclability but also in this product’s performance, for transcending the main barriers of flat glass refusal owing to either impurities or different glass synthesis, while it is aligned with the 2050 Dutch policy of zero waste in the construction field and the European guidelines linked to the sustainable movement. Recovering discarded glass in cast glass products opens new paths to reintroduce such waste back to the supply chain, as a closed-loop approach.

Glass is one of the oldest building materials used in the building envelope. Its attractiveness to architects, designers and users is mainly due to its fascinating ability to combine two otherwise contradictory functions; to separate and protect the interior from the exterior environment while at the same time allow for openness by letting daylight penetrate the space.

Apart from the apparent and basic function of transparency, glass, if properly designed, can fulfill almost any role in a building structure, from insulating a facade, to taking building loads besides its own weight and the imposed wind forces. Especially in terms of a building’s thermal performance, the assembly of single glass panes into insulated units is a colossal advancement for the building sector towards more sustainable structures since it allows the creation of fully glazed facades with minimised thermal losses.
However, the use of insulated glass units while improves greatly the energy conservation of glass buildings, it neglects to account for the impact of this assembly procedure on the waste streams the materials used create once their service life is over. Currently, there is no provision for extending the life-time of IGUs by the means of reuse, refurbishment or recycling and consequently they end up being discarded in landfills after a limited time of use.

For this reason, it is crucial to design an alternative IGU that abides by the laws of circularity, meaning the use of design approaches with an end-of-life plan for new products used as construction elements. Such a circular approach could greatly benefit the transition from a linear to a circular economy which is indispensable for the building sector since it accounts for a large amount of construction and demolition waste that end in landfills every year. The pollution from these wastes could be significantly reduced if the products applied were initially designed to be used in a circular way instead of having a single-life character.

The current master’s thesis focuses on the exploration, through detailed design, of the different ways in which an insulated glass unit can become more circular without compromising its transparency. The term circularity here refers to a design that tackles the end of life of the unit by facilitating its remanufacture in case its thermal performance starts decreasing; the reuse of parts of it or of the whole unit, and the recycling of the glass panes when neither reuse nor remanufacture are an option. Moreover, the aim is to create an IGU that is as transparent a possible by minimising the optical result of the current IGU edge seal connections.

In order to create a circular IGU, firstly, contaminating factors on the glass surfaces that could hinder their recylcability should be avoided. Secondly, a reversible way of connecting the different glass panes that comprise the IGU is opted for. Currently used IGUs have edge seals that consist of an element called a spacer bar and sealing elements that ensure the structural stability of the unit and prevent the air and moisture diffusion to the cavity. The element that connects the glass with the spacer bar in the current IGUs is a strong structural adhesive of black colour that cannot be removed without contaminating the glass. This is precisely the element that this thesis aims at replacing through an alternative connection design that enables the reversibility of the unit and the reduction of the optical impact of the edge seal.

The design exploration phase led to a final design that achieves the initial goal using an innovative alternative connection that replaces the spacer bar and secondary seal of a typical IGU with two cast glass elements each fused to the edge of the float glass panes. The connection of these cast glass elements occurs with the use of a simple mechanical clamping element, a metal spring clip that can be easily applied and just as easily removed. The small dimensions of this clip and its application internally between the float glass panes ensures the reversibility of the edge seal as well as its improved optical result.

The final design remains in a research stage proposal and further testing would be necessary regarding its thermal performance, its structural rigidity and the assurance of its good weather sealing. However, this research through design exploration concluded to a feasible, in theory, design that achieves the initial goal.

Glass has been widely used in the construction industry since the early nineteenth century. It was most commonly used in the form of a window, which is a flat glass panel. Glass has natural optical properties and is resistant to weather, making it ideal for use as a facade material. Glass’s novel applications in the built environment have led us to go beyond its 2D application of flat glass panels to a solid, 3D component, as seen in recent projects such as Maison de Hermes in Tokyo and Crystal House in Amsterdam.

Currently, there are two approaches in glass blocks: load-bearing solid glass blocks with poor thermal properties and hollow glass blocks with optimal thermal properties but no structural performance. Can we combine these approaches and develop a glass block with good load-bearing and thermal properties? Research on Hybrid glass blocks is one such topic that investigates the potential of this unique concept. A hybrid glass block is the combination of solid and hollow glass blocks designed as a single product with good thermal performance and load-bearing capacity.

The existing research on this topic is promising and noteworthy for further development as it conceptualises the rudimentary design guidelines for the system. The output of the existing research is oriented towards the design development of the novel hybrid glass blocks, ideation of the production methodologies, and validation of their thermal performance. Thus the scope for further research and development is actual prototyping of the design concepts, validation of the structural performance, and qualitative analysis. It is observed that the shape of the hybrid glass blocks impacts the structural-thermal performance, production methodology, and desirability of standardisation. Thus, to develop this product a thorough investigation and exploration are necessary.

The research first focuses on developing design guidelines for the innovative hybrid glass system. A product development methodology is formulated to guide and aid the design process of the hybrid glass block systems. Various design concepts are explored in-depth for each design problem based on their relationship with the structural-thermal performance, manufacturing, and assembly process and are then assessed through a set of design criteria to develop a final design concept. The final design concept is further refined based on manufacturing standards and challenges. The final design is then detailed and implemented in the Academy of Arts, Maastricht (case study), and various assembly strategies for dry stacked cast glass systems are formulated.

To validate the design a prototype manufacturing is done by using a kiln casting method. The prototyping challenges, observations, and decisions aid to assess the practicality of the design. A hybrid method of numerical and experimental validation is considered to evaluate the design based on its feasibility for application in the building industry. This research proves to be a guide for developing hybrid glass block systems and provides them with a design development, manufacturing, and evaluation framework to design, develop and assess their concepts.

The wide application of glass in buildings is due to its innate transparency and durability. This quality has led to the development of hollow glass blocks during the industrial revolution. These blocks are durable, fire-resistant, exhibit heat resistance, and sound deadening properties. In recent years, new explorations have begun to uncover the structural potential of glass. It is no longer just a cladding material but is also being used for load-bearing applications due to its high compressive strength. One of the most significant drawbacks of cast glass bricks’ current systems is the unsatisfactory thermal performance due to the absence of a cavity and the thick cross-section, which acts as one thick single glazed unit. On the other hand, hollow glass blocks are non-load-bearing due to their thin cross-section of the inner wall which results in buckling under a load and thus, the system is susceptible to failure. A promising solution to these problems is to develop a block that can exhibit structural strength and thermal insulating properties. To do this, design guidelines were developed by carefully studying both the systems followed by different design options which were analysed for their thermal performance. It was observed that the incorporation of cavity, inert gas and coatings greatly influences the thermal property. Also, the presence of continuous glass cross-section is important for structural integrity which generates thermal bridges that negatively impacts the thermal performance of the system. This analysis resulted in two different design options; the fusion block and the lattice block. To fabricate these blocks, careful consideration was taken in the design of the moulds and the chosen glass type. For the two designs, two separate connection systems are employed; one an embedded connection and the other an interlocking pattern. Both connections generate a dry assembly system which is reversible, easy to assemble and easy to maintain. To understand the feasibility of the proposed design solutions, a case study of Ports 1961 store in Shanghai was considered. The blocks were applied on the façade of the building and this was then analysed based on the developed design criterions. The proposed blocks have better thermal performance values, optical and aesthetical qualities. The installation process is much simpler and reversible. The fabrication process is however complex but that is due to the absence of standardized manufacturing system for cast glass bricks. The present research does not conclude in a single suitable design option but rather two concepts. The exploration of different concepts for thermal performance, it’s fabrication and installation results in a general understanding of the parameters that affect the development of this technology. To realize the proposed system, structural verification, fire safety and acoustics still need to be carefully considered and additionally this need to be validated experimentally to derive statistical data for its safe application. Nonetheless the performance values indicate a great potential in the technology.

This text is an investigation of process. It outlines the design development of a historic concrete slab retrofit using glass. By working through iterations that build off each other, it chronicles the steady unfolding of the concept coming into being. Materiality, statics, construction, and aesthetics are the parameters that are mediated in a case study intervention. The divide between preservation and progress, between the currently clear-cut choices of conservation or structural design, is a powerful space for transformation. The research looks in depth into the existing historic structure, recalculating it using modern methods. In turn, this informs geometry and configuration, which are iteratively redesigned to be minimal, respectful, and surgical. Foundationally, this project is a love letter to the ghosts in the hall, who are tasked with handing down their teachings to the future generations, so they better understand their responsibility to the universe and that all things past and present are equal parts of the whole.

It has been only a few decades since glass has started revealing its full potential and emerged in the engineering world in three-dimensional repetitive cast glass units with the shape and size of typical masonry bricks. Cast glass stands as the most competitive and novel answer to the relentless eagerness for transparency in the built environment, expanding in this way the shape and size barriers imposed by float glass. At present, cast glass has already proved its dynamism in few realized projects since it is the foundation of three structural systems which employ cast glass units. The additional supportive substructure, the adhesively bonded units and the third which is still on research stage, the
interlocking cast glass elements. The aforementioned systems present great potentiality for more transparent structures, yet some of the technical aspects could be further developed. The present master thesis explores the potential of embedding connections in cast glass units as a fourth structural system, supplementary to the existent ones. This system stands as an attempt to address the intensive assembly, the compromised transparency or the irreversibility, some of the key difficulties that are introduced by the precedent systems.

Glass is a material which can be seen as a double-edged sword. Its monolithic, smooth, and transparent look transmits a sense of elegance and simplicity. However, due to its brittle mechanical property, it can create complex engineering challenges when intended for a structural load-bearing application. Many forget glass has a comparable compression strength to steel and excels when compared to the mechanical properties of unreinforced concrete. (Oikonomopoulou, 2019; Ashby, 2006) Until now, there have been 4 primary discovered manufacturing techniques for glass. Float glass has a 25mm thickness limitation. (AGC, 2019) Extruded and float glass cannot form complex 3D geometries. (Roeder, 1971) 3D printed glass is not structurally validated. (Klein, 2015) Cast glass can offer a solution to all of these limitations. (Oikonomopoulou, 2019)
This thesis answers the following question: To what extent can topological optimisation and new fabrication methods be employed to create compressive free-form glass structures? To answer this this thesis will delve into glass, topology optimisation and shell structures. Throughout the thesis it will notice how topology optimisation and shell structures offer solutions to glass related challenges.
Compression Shells are ideal for Glass. Why? Because glass is brittle. This means that it fractures without warning when tension is applied to a crack. Glass has a low tensile strength but a very high compressive strength. That is why shell structures are ideal for glass as they experience compression only loads. (Adriaenssens et al., 2014) Shells best utilize the strengths of glass and avoid testing its weakest limitations.
Annealing time is one of the most challenging aspects of cast glass. As the volume/mass ratio increases the annealing time increases exponentially. (Oikonomopoulou, 2019) This means careful consideration should be attributed to decreasing this time. This thesis used topology optimisation as a tool to decrease the thickness and thus the annealing time. This thesis explains structural optimisation and evaluates different topological optimisation approaches and software. At the end a 43% mass reduction was achieved.
The goal in this thesis was to design an 6x6m booth from cast glass for the martial district exhibition. After the shell was topologically optimised and structurally validated by means of finite element analyses, the booth was designed for manufacturing, transportation and assembly. The booth was subdivided to fit in trucks and be assembled using spider cranes. The manufacturing procedure was developed using additively manufactured sand mould designs. Demountable and reusable connections, and foundation designs were also developed for the shell.

This thesis is about reconstructing historical vaults with cast glass. The vision is to design a complete dry-assembly cast glass arch vault where the Notre Dame de Paris is the chosen case study. The main question is: To what extent can a historical masonry arch vault be reconstructed by using cast glass components? The method is based on literature research and FEA simulations. The thesis contains three main phases of the research. First, the background, problem statement and research objectives/ questions are discussed in the introduction. Then, the complete literature research where heritage, arch vault structures, structural glass and cast glass in structures will be discussed. The third part describes the complete design phase including the criteria and final design. The thesis finishes with a conclusion and discussion chapter. The main results of this thesis are that a historical arch vault can be reconstructed from cast glass, but this will bring high risks and costs. Therefore, the suggestion is to use cast glass for the structural arches as this meets the complex geometry but use traditional materials for complex nodes and float glass where single, abstract curves can be made. The connections between different components are dry thanks to interlayers, but the cast glass elements are adhesively bonded. From safety, assembly, and labor perspective, these connections are chosen. This way, the blurry effect of glass in a heritage is reached, but the method is simplified. This principle can also be applied as roof structures for future buildings, but also for small pedestrian bridges as the geometry possibilities are unlimited.

The current thesis aims to explore the potential of producing large-scale structural cast glass elements with the help of 3d printed sand moulds. Until now, mostly small blocks from structural cast glass have been generated in the building industry, revealing a research gap regarding massive components. Apart from the size, the unlimited shape potentials of cast glass have hardly been explored so far. This is because glass casting can be a very challenging process when it comes to the solidification of the material which needs to be completed under specific program settings. Except from the brittleness of the body during the annealing process, it can take months, even years for a large-scale element to be completed, because of the full controllable conditions under which the annealing process takes place. In addition, the mould types that are currently used for cast glass components present several restrictions such as the manufacturing costs, the accuracy or the sizing. An existing glass slab in the Acropolis museum, in Athens, was chosen to embody the research goals. The slab is located on top of archeological discoveries making the need for transparency necessary. The current slab consists of float glass panes with a concrete substructure which severally restrains the viewing. The thesis aim is to develop a free-form monolithic slab exclusively from cast glass without the need of external substructure. In order to address the above demands, the topology optimization method was introduced. Several algorithms of topology optimization have been developed. A compliance-based optimization was demonstrated as the most appropriate for the specific project. Obtained from the literature framework, the main principles that the design follows gravitate towards the structural, cast glass and manufacturing requirements. The external applied load, in addition to the allowable structural values were calculated. The importance for a flat upper surface, where a safety float glass layer of a permitted deformation should be laminated, was established. Regarding the cast glass demands, it is necessary to acquire a design with smooth surfaces, round edges and uniform distributed material throughout the entire model. For the current project, the optimized slab is decided to mainly consist of borosilicate glass.. A slab separation is crucial for transportation reasons. After the definition of the main design principles that the final product should respect, the design development was conducted with the combination of three methods: topology optimization, manual design and structural validation. The final result of the optimized slab successfully fulfills the preset design criteria, as the obtained structure presents 55.2% less mass than the solid version that was initially designed to replace the existing slab in the museum. In addition, a more efficient material distribution in the optimized model renders the slab more compatible for the cast glass annealing process. The result of the topology optimization technique was a quite complex geometry which is not that effective to be produced in the conventional moulds that are currently available for cast glass applications. Therefore, a new technique needed to be explored, which follows the 3d printing evolution in the building industry. The main restriction of the printed moulds is the limited available sizing that the commercial printers provide. As a solution to this problem, the separation of the large-scale mould into smaller segments was studied, as well as the development of the connecting system of these parts. The entire fabrication technique, from the moulds printing method, to the assembly components and glass melting was investigated. Finally, the transportation and integration of the slab into the building was explored, concluding with structural solutions that respect the museum and the assembly order that the slab can be attached on it.

The building industry is a significant waste contributor. Especially at the end-of-life of the building it causes an enormous impact on the environment. Construction solid waste has caused serious environmental problems. Reuse, recycling and reduction of construction materials have been advocated for many years, and various methods have been investigated. However, the effectiveness of its applications seems limited. Waste will be reduced by extending the lifetime of the materials giving the second life.This research aims to the design of a construction component made of reused train windows of the old VIRM trains of NS. The continuous flow of the train windows due to the renovation of the trains demands a multiadaptable solution that offers different design scenarios.

Glass is a material with high potentiality in a structural application. Brittleness has always been challenging for glass structure. Recyclability is also challenging for the current glass application since the chemical bonding is contaminating the recipe. Interlocking cast glass construction offers an interesting bypass to solve these two problems. To not rely on chemical bonding, the interlocking mechanism is used. Interlocking requires self-stabilising bulky units which are the perfect fit for the casting process, which is not conventional for the structural glass. Interlocking construction with glass blocks relies on soft interlayer to avoid peak stress and protect the glass. Also, this makes the interlocking cast glass construction rather challenging to analyse.

The thesis investigates the key aspects of the topological interlocking cast glass (referred as TICG) system with elastomer interlayers. Glass is a sensitive material; its structural behaviour needs to be investigated considering its geometry and boundary condition of the application. The complexity of the structural behaviour lies in three levels, namely, single block level, block-interlayer-block interaction level and interlocking assembly level.

The literature research was done as the first step to understanding the relevant aspect for the TICG system could be. It then concluded that the interlayer mechanical attributes, glass geometry, interlocking amplitude and lateral pre-compression could be the key aspects. A case of the Optical House in Hiroshima, Japan is chosen to apply this technology, for which the possible failure and risk is analysed according to the possible failure mode in the literature research. A strategy of using active façade with compression-release is determined for the key feature of the façade.

Given the complexity in the TICG composition, it is decided to do physical testing on the interlocking cast glass system and blocks, instead of doing numerical simulation. This part generates the requirement for façade engineering to be considered for the next section. The TICG structural behaviour complexity is investigated in three tests, namely, uniaxial test, three-block shear test and assembly test. The uniaxial test yielded the relevance of interlayer strengths and block geometry. The three-block shear test showed the relation between shear key geometry and shear load transfer. The final assembly test proved that the pre-compression is still relevant to the stiffness of the TICG façade and actively controlled releasing mechanism should be integrated.

In the last part, the façade is engineered and detailed considering the requirement previously investigated. Feasibility, detailing and assembly sequence is given. Comparison has been made between existing threaded cast glass solution and the TICG solution.

The thesis plays a part in the glass research group in the Building Engineering department of TU Delft and it is aimed to provide a step forward towards a more reliable and more recyclable glass structural solution.

Cast glass is a promising structural material that has seen increasing use in the build environment over the last year. It offers a great freedom of shape for designers and engineers, beyond the two-dimensional float glass that is still primarily used. Despite this, cast glass in practice has only been used for solid bricks, mimicking traditional ceramic brickwork. Little exploration has been made on the complex shapes that can be achieved in cast glass. One of the largest challenges of cast glass is its slow and meticulous cooling process required after casting, to prevent the occurrence of internal stresses. This costly annealing process is the main limitations for producing large scale cast glass elements in an affordable way. Experience in large scale solid glass mirror castings performed for NASA show that annealing times can be reduced through smart design.
Structural topology optimisation is a design tool for creating light-weight, material efficient elements. Its application has mostly been limited to industrial and aerospace design, though it is slowly being introduced in the built environment as the additive manufacturing required for these geometries becomes more advanced, and affordable. So far, topology optimisation of cast glass design has remained unexplored. The goal of this thesis is to investigate the potential of using topology optimisation as a design tool for complex structural cast-glass elements with a reduced annealing time. For this, a complex-shaped grid shell has been redesigned in glass, using optimised cast glass nodes. The entire fabrication process, from digital design, to physical fabrication using additive manufacturing has been explored.

Glass is a material that has a high compressive strength and a low tensile strength. While being mostly produced as flat sheets of float glass, it can also be casted into components that have a thicker geometry and thus a higher buckling resistance. These properties make cast glass components suitable for the construction of shell structures that are mainly subjected to compressive stresses. Shell structures often have the shape of surfaces with varying Gaussian curvature. When constructing such a shell structure out of cast glass components, components of varying geometries are needed. During this research, an adjustable mould was developed that can be used for the casting of glass voussoirs of varying geometries. The possible voussoirs geometries that can be cast in the adjustable mould are limited to voussoirs with planar, convex polygonal intrados and extrados. These voussoirs can be used to construct fully transparent shell structures. The voussoirs are dry-assembled with an interlayer in between that compensates for any production tolerances and avoids glass to glass contact. Furthermore, by using dry-assembly instead of an adhesive as a bonding method, the structure allows for disassembly and full recyclability. Tongue and groove shaped interfaces ensure an interlocking connection. The interlocking connection prevents any lateral movement and will guide the voussoir into the right position during assembly. By tessellating a shell structure, the shell structure is divided into a discrete number of voussoirs that can be cast in the adjustable mould. Several aspects have to be taken into account when optimizing the tessellation pattern including planarity, alignment to the flow of forces, face size, interior angle size and if the pattern is staggered or not. A comparison was made between three different regular tessellation patterns (triangular, quadrangular and hexagonal) with these aspects kept in mind. A shell structure was designed to cover a courtyard of the Armamentarium in Delft. This shell structure served as a case study used to demonstrate the design and production process developed for this thesis. The shape of the shell was generated through a form finding process and its structural performance was validated with a FEM analysis. The shell was tessellated into a triangular tessellation. An algorithm used for assigning either tongues or grooves to the interfaces of each voussoir was developed. A list was generated containing the following information for each voussoir: voussoir index number, interface types, edge lengths, voussoir joining angles and the adjacent voussoir index numbers. The data in this list serves as input data for the set-up of the adjustable mould. Furthermore, the connection between the shell structure and the existing structure was designed and an assembly method was developed.

Structures made from cast glass components are transparent and make full use of the high compressive strength of glass. As they have more freedom in shapes and can therefore be designed with a large cross-section, it is more suitable as a
structural element than fl oat glass. However, most cast glass structures that have been completed suffered from compromised transparency, or could not be recycled, and required meticulous construction. This is caused by the building system, that requires adhesive bonding of the elements, or a metal substructure.
These issues can be resolved by the use of interlocking geometries, where the glass components are designed to directly provide structural stability, and the structure can be disassembled. This allows re-using the glass. This thesis investigates the design of an interlocking structure and compares this with an adhesively bonded structure in terms of its structural behaviour and assembly. To this end, a brick geometry for an irregular curved surface is designed, different connector shapes for the interlocking elements are evaluated based on their ability to withstand lateral forces, found through experimental testing. Furthermore, the interlayer between different brick layers is chosen based on literature review, a building system is designed and its structural behaviour is validated. It is found that interlocking structures have a lower stiffness than adhesively bonded structures. But because of the lower stiffness, it has a higher flexibility. In case of an impact, it has the tendency to move back in its original location when the impact force is removed and gives a warning before it fails where an adhesively bonded structure would suddenly break. Next to the structural differences, the assembly also changes. Due to the use of an interlayer there is more space for tolerances in the surface of the bricks and the bricks can be dry-stacked. Although during assembly it is important for both structures to prevent eccentricity.

Currently, tons of high-quality borosilicate glass are discarded to landfills each year, taking up space on valuable land. Within the existing soda-lime glass recycling industry it is not possible to use recycled borosilicate glass due to its higher melting temperatures and a difference in chemical composition. In fact, borosilicate glass disturbs the recycling process of soda-lime glass, resulting in more unused waste. Borosilicate glass is a type of glass that has favourable properties for application in the built environment, such as high optical qualities and a low thermal expansion coefficient. However, due to its high melting temperature and high energy consumption it is much more expensive to produce compared to the more common soda-lime glass. Nevertheless, in general it is possible to recycle glass waste almost 1:1, therefore, it is a waste to dump such high-quality glass on landfills. Cost and energy consumption issues related to the fabrication of borosilicate glass can be decreased by using recycled crushed glass (cullet) to the melt. Cullet reduces the required melting temperature, resulting in a lower energy consumption and a cheaper production process. However, a fully closed borosilicate glass recycling loop is not possible. The current borosilicate glass product industry has very strict quality control. Recycling glass introduces contaminations in the glass melt, which could generate flaws and cracks in the end product. This means it is not feasible to use recycled glass in these industries. This research proposes a promising approach to tackle the borosilicate glass waste problem and the recycling issues related to contamination: using the collected borosilicate glass waste to produce cast glass components. The bulkiness of such a component allows for a higher amount of impurities without risk of failure and a decrease in quality. In addition, cast glass allows for self-supporting facade systems in many configurations due to their increased cross-section and high-compressive strength capability. To allow disassembly of such a cast glass facade system a dry-interlocking system is proposed. Such a system can ‘close’ the recycling loop of borosilicate glass by feeding in waste glass from traditional borosilicate products into a new cast glass recycling loop. The challenges and opportunities regarding recycling borosilicate glass were researched by conducting experiments. Several mixtures of borosilicate glass with different chemical compositions were cast into small specimens to assess the mixability and corresponding mechanical properties of recycled borosilicate glass. In terms of mixability, good results were observed when the cullet is ground to powder, although there is a risk of contamination during grinding and thus degrading quality. Using fine cullet is also promising, but there is an increased risk of internal stresses. Specimens with a homogeneous chemical composition showed that it is possible to create borosilicate glass components with a lower firing temperature. In addition, these homogeneous specimens illustrated through a three-point bending test that recycled borosilicate glass has comparable mechanical properties to non-recycled borosilicate glass. This indicates that recycled borosilicate glass is suitable for application in cast glass components for use in the built environment. Due to time constraints and the preliminary nature of this research the number of specimens was relatively small. This thesis presents a dry-interlocking cast glass component system applied in a case study. The case study concerning the existing Casa da Música is meant to illustrate the possibilities of applying high-quality glass in an aesthetically pleasing facade system and to showcase the possibilities of recycling borosilicate glass. At the end-of-life of this design the facade can be disassembled. The components can either be directly reused in another building or recycled again. The proposed solution can reduce the growing amount of glass waste.