Unveiling the third dimension of glass

Abstract

Glass as a material has always fascinated architects. Its inherent transparency has given us the ability to create diaphanous barriers between the interior and the exterior that allow for space and light continuity. Yet, we are just starting to understand the full potential, properties and characteristics of glass as a material. Only in the last decades did we discover the structural potential of glass and started to use it, besides as a cladding material, also for load-bearing applications thanks to its high compressive strength. Indeed, at present the structural applications of glass in architecture are continuously increasing, yet they are dominated by a considerable geometrical limitation: the essentially 2-dimensionality imposed by the prevailing float glass industry. Although glass panels can stretch more than 20 m in length, the maximum monolithic thickness by this manufacturing method remains a mere 25 mm. As a result glass structures are currently dominated by virtually 2-dimensional, planar elements and confined to the limited shapes that can be achieved by those. This research focuses on the exploration of cast glass as a promising, 3-dimensional construction material in architecture. The main aim of this research is therefore to investigate the potential, as well as the constraints, of cast glass components for the engineering of transparent, 3-dimensional glass structures in architecture. By pouring molten glass into moulds, solid 3-dimensional glass components of virtually any shape and cross-section can be made. Owing to their monolithic nature, such components can form repetitive units for the construction of freeform, full-glass structures that are not sensitive to buckling. Such structures can take full advantage of the high compressive strength of glass, sparing the necessity of additional supporting elements. To achieve cast glass structures, it is essential to use an intermediate material between the individual glass components that contributes to the structure’s stiffness, ensures a homogeneous load distribution and prevents early failure due to concentrated stresses triggered by glass-to-glass contact. To maximize transparency, this intermedium should be colourless and any additional substructure should be minimized. Accordingly, the main scientific contribution of this research work is the design, development and experimental investigation of two distinct systems for selfsupporting envelopes of maximized transparency: An adhesively bonded glass block system, using a colourless adhesive as an intermedium and a dry-assembly, interlocking cast glass block system, employing a colourless dry interlayer. Although, in this work, both systems have been developed for self-supporting envelopes, the results can be used as a guideline for further structural applications of cast glass components in compressive elements, such as columns, arches and bending elements, such as beams and fins. At present, the load-bearing function of cast glass in architecture remains an uncharted field. Discouraging factors such as the lengthy annealing process required, the to-date non-standardized production and the corresponding high manufacturing costs, have limited cast glass to only a few realized architectural applications. As a result, there is a lack of engineering data and a general unawareness of the potential and risks of employing cast glass structurally. Hence, in order to accomplish the research goal, all pertinent aspects of a cast glass structure should be tackled, ranging from cast glass’s production method to practical implications when building with cast glass. These distinct aspects are addressed through the formulation of the research sub-questions, which in turn define the different chapters of this dissertation. Accordingly, the presented work is divided in four parts. Part I provides the Introduction to the Research, and aims at giving a brief summary of the involved challenges, identify the research gap and introduce the research questions and the research methodology.  Part II focuses on the Theoretical Framework of the Research. It lays the foundations for this dissertation and contributes to the scientific field of structural glass by providing the first comprehensive literature review and state-of-the art overview of cast glass structural applications. Initially, the material compositions and production methods for solid cast glass components are explored. Then, to address both possibilities and limitations in the size and form of cast glass components, an overview and critical assessment of the largest produced monolithic pieces of cast glass is made. Given the limited published scientific output on this specific field, an extensive field research was conducted in order to derive the relevant data. The discussed examples, although coming from different fields of science and art, provide great insight into the practical implications involved in casting as a manufacturing method. Subsequently, a separate chapter gives an overview of the state-ofthe- art in cast glass structural applications in architecture. Aiming on providing the reader with an holistic overview of the structural potential of cast glass in architectural applications, this chapter includes the synopsis, feasibility assessment and comparison of not only the realized structural design systems but also of the adhesively-bonded and dry-assembly interlocking systems developed in this dissertation. Special attention is given to the advantages and disadvantages of the connection method of each -existing and developed in this dissertation- structural design system with solid glass blocks. Following the findings of the literature review and field research, Part III, consisting of four chapters, presents the design and experimental investigation of two distinct, novel structural systems out of cast glass components, developed for selfsupporting envelopes. Part III can be considered the main scientific outcome of this dissertation. Firstly, the research, development and experimental validation of an adhesively bonded system utilizing solid cast glass blocks is presented. Numerous full-scale prototypes are made and tested in order to comprehend the structural behaviour of the adhesively bonded glass assembly. A separate chapter explores the main challenges and innovations and defines the construction requirements necessary for the realization of the investigated system at the Crystal Houses Façade in Amsterdam. An important conclusion is that such an adhesively bonded system requires an extremely high dimensional accuracy both in the fabrication of the glass blocks and in the entire construction, and has an irreversible nature, which in turn results in a meticulous and unsustainable construction. Based on the aforementioned challenges, a new concept for glass structures out of dry-assembled interlocking cast glass components is developed that tackles the integral limitations of the adhesively-bonded system. An entire chapter is dedicated to the principles, the establishment of design criteria and to the preliminary exploration and assessment of different interlocking cast glass shapes that can yield an interlocking cast glass system of satisfactory structural performance. Following, the last chapter of this part concerns the experimental and numerical investigation of this second system. The effect of various parameters in the structural behaviour of the system is explored through the production of scaled prototypes and their experimental validation. A numerical model further explores the correlation of the various geometrical parameters of the interlocking geometry to the structural behaviour of the system. Finally, Part IV presents an integrated discussion of the research results, summarizing and discussing the main outcomes of the dissertation. Initially, responses to the research questions are given in order to assess the particular findings. Based on the conclusions, further recommendations are made, firstly for overcoming the limitations of the presented research, following by general suggestions on a wider range of the aspects of cast glass that can be explored and contribute to its structural applicability. The findings of this dissertation prove the feasibility of the discussed systems and can serve as solid guidelines for further applications. The research presented in this work has been positively received by the international architectural and engineering community. In specific, the presented adhesivelybonded cast block system, which was realized at the Crystal Houses façade, received numerous awards by the structural engineering community, including the Outstanding Innovation Award 2016 by the Society of Façade Engineers and the Glass Innovation Award 2016 from the Bouwend Nederland association. Still, the Crystal Houses façade is but the first real-scale prototype of the developed adhesively bonded system. The actual construction of the façade provided invaluable feedback on the engineering challenges and construction requirements involved in such a system, giving room for new suggestions. This triggered the development of the second presented system with interlocking glass blocks as a reversible, easily assembled solution. The interlocking cast glass block system, initiated within TU Delft and funded partially by a 4TU.bouw grant is yet to be applied in practice. Prototypes of this research, using recycled cast glass components, have been exhibited in international fairs such as the Venice Design 2018, the Dutch Design Week 2018 and Salone del Mobile 2019 and are currently displayed at the material collection of the Vitra Design Museum at the Vitra Schaudepot. The project was also nominated for the New Material Award 2018 under the title Re3 Glass. Even though cast glass has, so far, been rarely applied in structural applications, the development of new building systems and their experimental validation presented in this work provide a strong basis for further developments and applications in a range of compressive structures. At present, the most considerable drawbacks hindering the marketability of cast glass components are (a) the cost barriers imposed by their customized production and application and (b) the lack of standardized strength data and building guidelines. Thus, even if cast glass elements have proved to be suitable structural components, several economic aspects and logistics need to be tackled, and performance issues need to be further explored, in order to make cast glass a competitive manufacturing method to float production for structural components.