REVIEW OF STUDY ON STRUCTURAL GLASS AND STRUCTURES

Tomas Serafinaviius1, Audronis Kazimieras Kvedaras2

Vilnius Gediminas Technical University, Saultekio ave. 11, LT-10223 Vilnius, Lithuania. E-mail: 1 tomas.serafinavicius@vgtu.lt; 2 audronis-kazimieras.kvedaras@vgtu.lt
Abstract. Lithuania has a long tradition of glass. Findings of archaeological research show that people of the Grand Duchy of Lithuania (XIIXIII century) already applied technology of glass melting. This is considered to be the oldest Lithuanian industry that was papered in historical sources. Currently Lithuania and the world felt like a surge of activity and excitement, and interest in the new glass and light influx. One of the main characteristics of the glass, giving mystique to this substance, is its transparency for light. That is why this material is attractive for designing. The study on the glass structure resistance for deformations on the structure stability are carried out in order to develop the load bearing glass structures or use the glass in industry. This paper presents a literature review and analysis of scientific articles in the field of structural glass by applying testing methods. The findings of analysis are given in the end of the paper, also the less investigated areas from the structural glass field are revealed. Keywords: structural glass; overview; analysis; transparency; fragility; behaviour of structure.

Introduction

Examples of glass bearing structures

Over the past 25 years there already are some is imThe main challenges for the construction of glass are transparency. Of course, the glass can be an excellent plemented glass buildings in the World. One of the most material to build responsible structures. In these days use famous examples of structures built of glass is the Apple of structural glass has become increasingly important as store (Fig 1.), which is built in New York (O'Callaghan et more of it is used not only in door or window manufac- al. 2007). ture, but more widely. Properly designed glass structure can withstand considerable loads and act as the main load bearing element of buildings, often as columns and roof beams, stairs, canopy rack, etc. Among a number of articles written on the structural glass structures, the article that describes in details the tests carried out should be mentioned (Beer 2005). This article reveals the tests that were carried out in one object BBC Broadcasting House Development in London. So it is a narrow focus on only one object, an overview. The narrow focus on only one object in this article illustrates the need for more research review on glass structures in order to determine which areas are intensively investigated and those that are weak, and what problems might come to use glass in modern or renovated construction. Fig 1. Apple Computer retail store on Fifth Avenue in In this paper the brief overview of glass structural New York City. elements (beams, columns and slabs) is given; their connection problems and design techniques, including design The aim of the building designers was to make all codes characteristics are overlooked. Conclusions include the load bearing structures from glass in order to make the the strongest and the least investigated areas associated building more transparent both from outside and inside. with glass structures, a number of proposals on what to First of all an internal stairway and bridge from structural focus while choosing the areas for further study. 787

glass were designed, constructed and built in this building. These elements demanded the most attention and effort regarding the characteristics of the connecting elements. Staircase steps with a typical load of 4.5 kN/m2 and a length of 2.2 m and all staircase design is made of glass, including bridge-boards. The walls of the building are made of bonded 3 glass sheets, and its roof beams are made of 5 sheets of glass. Special metal fitting details at the beam supports between the glass plates of beam webs were glued together. The implementation of this building from designing to construction took 5 years (2001-2006). The authors mention further researches in order to find more improved glass bearing elements of the connecting nodes, the special focus giving on cleaning of the surfaces of structures during their maintenance. There were also designed a fully transparent glass pavilions in full scale in Aachen (Germany) (Wellershoff et al. 2003) and in Delft (The Netherlands) with a scale 1:8 using glass beams with steel reinforcement (Bos et al. 2005). One of the most important goals of these glass pavilions make a scientific research in order to investigate behaviour of the loaded glass elements and behaviour of structure as a whole as far as possible under natural conditions. Most of the research carried out has been performed with individual and separate elements. Therefore, the focus of researchers was more concentrated on the behaviour of whole structure and the columns connecting nodes, as well as beams with steel reinforcement. Beams Glass beams usually are obtained by gluing-up a few (46) glass sheets of 6-18mm thickness (Figs 2, 3).

value of 2500 N, but the node has not yet been finally disintegrated. It is also known the (Louter et al. 2007) natural experiments with steel reinforced glass beams, in which the pilot models which scales are 1:8, 1:4 and 1:2 were used. It was planned to manufacture and test the real 18 m long beams. While testing such main problems were encountered: the stability of the glass beam out-of-plane and support consolidation . The findings of these tests show that the stability in-plane is best ensured when the glass sheets are glued in asymmetric manner. There were tested several different ways of support consolidation: maximum stress of the glass beams till brakes was ~ 30 MPa. Several studies with U-shaped beams have been carried out in Germany, results of that provide a real opportunity to use such profiles in construction (Hof 2007). Also the glass cantilever with two supporting glass beams was investigated (Louter et al. 2005). Their findings concluded that such a structure can be effectively implemented, because the calculations showed a lower value of the load bearing capacity than structure was able to withstand; the laminated and temper glass ensures the safety.

Fig 3. Glass roof construction Fig 2. Connection of steel column and laminated glass beam

Columns Usually the glass columns are made to be as transparent and the load withstanding element. Most of the studies are carried out with the circular glass columns . (Nieuwenhuijzen et al. 2005) and (Veer et al. 1999) tested the columns of several lengths (1.2 m and 1.5 m), which outer diameter was 110 mm and 120 mm, the thickness of glass tube wall t=5 mm. The results of these tests

(Veer et al. 2007) made a glass bridge design research, while the tensile strength of glass beams was increased using the steel reinforcement. It was also investigated the zones of beam reinforcement anchorage, in the glass node, and the test results showed that first cracks in the glass appeared when the tensile force reached the 1500 N, the test was stopped while the force obtained the

788

described in both above mentioned articles show that the first cracks of 1.2 m long glass column were observed at the value of compression forse equal to 73 kN (29 MPa) and a full their collapse occurred at the value of such force equal to 146 kN (58 MPa). During the tests with 1.5 m long columns the first signs of collapse were observed at force equal to 61 kN (18 MPa) and full column collapse at force equal to 196 kN (61.7 MPa). It was concluded that lamination of columns gives the full transparency and it ensures the same safety requirements as the steel columns. (Blaauwendraad 2008) provided several formulas to calculate the stability of glass columns. (Luible et al. 2004) provided the test details on the compression strength of the glass plated elements. The glass plated columns with a thickness of plates 8-10 mm, different length 350-1600 mm and width of 200 mm (Fig 4.) were tested. The monolithic glass plate, the second- panels of laminated safety glass. The test was executed with the glass plate columns with a thickness of 810 mm and different length (350-1600 mm), and with a width of 200 mm (Fig 4.). First column type the monolithic glass plate, the second panels of laminated safe glass. The results show that the integral (t = 810 mm) monolithic glass plates compressive load capacity is ~10kN. Laminated (t = 10 mm) glass load capacity is ~ 60kN (L = 800 mm) and ~ 160 kN (L = 600 mm). During the tests the temperature of ~ 18.500C was fixed. The conclusions of this article reveal that compressive strength of flat glass element mainly depends on the initial deformation, the glass thickness, the shear stiffness and the PVB interlayer characteristics. The temperature and loading duration also have a significant impact. Higher load bearing capacity results were received at a higher temperature.

Plates Glass plates are used very widely: in elevation, as well as guard rails, stairs or balcony flooring (Fig 5.), roof construction (Fig 3.). The use of laminated glass plates (Figs 6, 7.) ensure security requirements in the event of collapse, the fragments of glass does not spread to the environment and does not damage people.

Fig 5. The glass staircase

Usually panels are consolidated by the full perimeter or partially affixed with silicone or aluminum profile (Fig 6). Glass panels can be fitted using a spot mode: gluing metal glass holders or drilling holes in glass. Recent investigations are mainly found. Differences of opinion and controversy about the drill holes are noticed in these articles. Some authors argue that holes drilling is not the best method of fixing. Numerical simulations are carried out using finite element modeling programs, considering the whole frame and plate deflections (Jacob et al. 2007). The findings argue that the behaviour of glass design on the operation of side impact has not been fully explored and there is a need for additional researches.

Fig 6. Laminated glass plate during the test

Fig 4. Glass facade column structure (the new Opera House in Oslo)

Probabilistic methods for design of glass panels were developed in order to ensure the safety requirements (Lamela-Rey et al. 2007). The results obtained by this program propose to use the widest possible range of cal789

culation methods at the same time to estimate reliability of the glass plate (application of the theoretical load capacity calculations, 3D digital simulations by finite element method software, natural structures tests using the actual size or scale). It is possible to exercise more detailed exploration adjusting the values of safety factors. Also the cyclic fatigue of glass plates was investigated (Boxheimer 2007) and the findings suggest that Young's modulus is highly dependent on loading time. A load holding panels were studied as a cut element (Huveners et. al. 2007) (Fig 7.). There were carried out three different natural studies with different panels supporting fortifications.

(Siebert 2007), (Bernard 2008) showed that the connections with the drilled holes and steel rivets can withstand force up to 143 kN. It is possible to use a special glue for joints (Weller et. al. 2008). The main problem of the tempered glass joint is the residual stresses arising after the complex thermal process and general behaviour of node assessing the unilateral friction and its losses between the glass and metal connector (Dong To et al. 2007). Test results on different fastening methods of glass plate elements to the building bracings (Mocibob et al. 2007 show that better stability is achieved using the steel bolts of larger diameter, but it does not depend on the fastening conditions.

Fig 9. Example of drilling holes for metal insertion into the structural glass Fig 7. Laminated glass plate after test

Calculation and design limits While designing the glass structures difficulties concerning their design standards arise. As this is a relatively new field, it is currently governed by only a few standards: DIN 18008 in Germany, Austrian OORM 3716, Canadian CAN-CGSB-12.20-M89, the United States code ASTM-E1300-97, Australian- AS 1288-94, and the European projects of prEN 13474-1, prEN 13474-2 (Siebert 2007, Beer 2005, Calderon 2007). However, these codes include only a small field of glass adaptation in design and construction. The existing coges include: the basic rules and concepts, perimeter glazing and fixation by the spot, additional requirements for handrails and flooring, and special structures. Japanese scientists published the first normative documents in 1971 (Kikuta et al. 2001), and they are constantly improving them assessing the changes in glass production technology, a greater thickness of the glass and a more accurate wind loads by using corrections of safety factors. The increasing popularity of public buildings from glass and their effects on the environment and human security made a concern in Russia too. In 2007 the Russian Duma has considered the technical description of the rules about the safety of the glass and its products used in buildings and construction. The main purpose of this document is to establish such regulations and conditions which would reduce the life and health risks, to create a certification procedures for designers of glass buildings

Findings lead to further research in order to discover the ways to achieve minimum stresses on the main plate. There are fulfilled experiments with hybrid design (glass and steel) (Englhardt et al. 2007) which opens the way to develop a hybrid of glass and steel trusses. Connections One of the most sensitive places of glass structures is areas of the installation of the connections. (Fig 8.). Fixing of glass members in supporting parts usually has several types: circumferential and spot (Fig 9.).

Fig 8. Connection of glass beams

790

and structures, to develop the use of modern glass and its products in structures (Chesnokov et al. 2007). China takes the first steps of standardization in glass construction.(Xinyong et al. 2007). Ordinary design of glass structures is carried out by modeling of static loads until convergence to the allowable stresses. All structures of building are modeled as a assembly with a 3D finite element programs, as well as individual elements or complicated connections. Temperature changes are evaluated in calculations too. The most of authors in their publications are mentioning that the application of hybrid structural glass and steel structures/connections based only on calculations are not entirely reliable and the tests must be carried out (Beer, 2005). The majority of the glass structural design standards have questioned because of the simplicity of calculation, based on crude strength calculation force acting on the cross-sectional area. Or on the contrary a very complicated calculation methods, based on complex empirical expressions, which can lead to serious mistakes or inaccuracies of calculated results, are used. Therefore, the structural glass building designers generally have to carry out the design calculations parallelly with performing the natural testing in laboratories. But testing in laboratories cannot provide a real construction site conditions and the builders could make inaccuracies and errors (Beer, 2005). Conclusions and Summary

The glass as a material is not expensive and have good mechanical characteristics, therefore, keeping the use of elements in building construction are quite rational. It was observed that less studies of the fatigue were exercised. The most of tests are carried out with the individual and the separate elements. It was observed only the few articles considering all the construction work and monitoring the behaviour of structural glass. Structural glass is not fully studied subject. Too few test results do not warrant believing the behaviour of glass structures. Further research could be carried out in: assembly methods of glass elements; frictional joints; developed elements of I shape or square / rectangular shapes of glass elements; adhesive connections; behaviour of lateral pressure; fatigue of glass elements or whole structure. During the review there are carried out more and more testing and research in the World structural glass field. So this review is a very temporary thing and requires to keep a consistent interest and knowledge in addition. References
Beer, B. 2005. Structural Glass Engineering- a Review of Project Specific Testing, in The Leading International Architectual and Automotive Glass Conference Glass Processing Days, Tampere, Finland, June 1720, 2005. Tampere, 710 p. Bennison, J. S.; Smith, C. A.; Duser, A. V.; Jagota, A. 2001. Strength of Laminated Safety Glass, in The 7th International Conference on Architectural and Automotive Glass Glass Processing Days 2001, Tampere, Finland, June 18-21, 2001. Tampere, 368371. Bernard, F. 2008. Optimization of Bolted Connections in Loadbearing Glass Elements, in Challenging Glass, Conference on Architectural and Structural Applications of Glass: Selected papers. Ed. by Bos, Louter, Veer. May 2008, Delft, Netherlands. TUDelft, 2008. 660 p. Blaauwendraad, J. 2008. Buckling of Laminated Glass Columns. Comparison and Replacement, in Challenging Glass, Conference on Architectural and Structural Applications of Glass: Selected papers. Ed. by Bos, Louter, Veer. May 2008, Delft, Netherlands. TUDelft, 660 p. Bos, F.; Veer, F.; Romein, T.; Nijsse, R. 2005. The Evaluation of the All Transparent Pavilion Project, in The Leading International Architectual and Automotive Glass Conference Glass Processing Days, Tampere, Finland, June 1720, 2005. Tampere, 710 p. Boxheimer, K. 2007. Cyclic fatique of glass panels. The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 452454. Emelianova, O.A.; Chesnokov, A.G. 2007. Project of Russian law: technical order About safety of glass and its products used in buildings and constructions, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 455457.

From the performed review of study on structural glass and its structures the following conclutions are presented: The design of glass bearing structures is still big challenge for engineers. The most of authors argue about using glass structures practically, but not clearly disclose the reasons of that. One of the difficulties in the construction of the glass structures is installation of connections. The development of Eurocode for structural glass structures design has already started. There were not found any scientific article relating to the structural glass investigation or tests, as well as design codes in Lithuania. Only a few studies are carried out in relation with the heat conservation in buildings through the glass structures. The most of authors are mentioning that the calculations of hybrid structural glass and steel structures/connections are not entirely reliable and tests parallelly must be carried out. Quite low volume of scientific articles investigating columns as structural elements of the square, triangular or I profile is found. The most of articles present results of investigations on the circular columns. It was observed higher content of the articles on the exploration of plates under natural experiments and numerical experiments using the finite element method. 791

Englhardt, O.; Bergmeister, K. 2007. Hybrid structural elements- an innovative high filigree glass-steel-system, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 134137. Hof, P. 2007. Investigations on U-shaped glass with a polyurethane lacquer for splinter shield. The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 314316. Huveners, E.M.P; Herwijnen, F.; Soetens, F.; Hofmeyer, H. 2007. In-plane loaded glass pane (shear wall), in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 362365. Jacob, L.; DCruz, J. 2007. Behaviour of laterally loaded glass panels with an emphasis on frame and panel deflections, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 419421. Kikuta, M.; Murakami, T.; Aratani, S. 2001. Conception of A New Notification of Building Glass Design in Japan, in The 7th International Conference on Architectural and Automotive Glass Glass Processing Days 2001, Tampere, Finland, June 1821, 2001. Tampere, 688691. Lamela-Rey, M.J.; Fernandez-Canteli, A.; Garcia-Prieto, M.A., Castillo-Ron, E. 2007. Probabilistic design model for glazing plates: a standard proposal for building, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 581583. Louter, C.; Belis, J.; Bos, F.; Veer, F.; Hobbelman, G. 2005. Reinforced Glass Cantilever Beams. The Leading International Architectual and Automotive Glass Conference Glass Processing Days, Tampere, Finland, June 1720, 2005. Tampere, 430434. Louter, C.; Veer, F. 2007. Experimental research on scale 1:4 models of an 18m reinforced glass beam, part I. The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 8792. Luible, A.; Crisinel, M. 2005. Buckling of glass elements under compression. Using the World-Wide Web to Deliver Complex Electronic Documents: [vieved on December 12, 2009]. Available on the Internet: <www.glassfiles.com>. Mocibob, D.; Crisinel, M. 2007. Glass panel under in-plane shear loading: Experimental investigation on structural glass panel point support, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 1518, 2007. Tampere, 380383. Nieuwenhuijzen, E.J.; Bos, F.P.; Veer, F.A. 2005. The Laminated Glass Column, in The Leading International Architectual and Automotive Glass Conference Glass Process-

ing Days, Tampere, Finland, June 1720, 2005. Tampere, 710 p. OCallaghan, J; Coult, G. 2007. An all glass cube in New York City, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 98101. Quy-Dong To; Qi-Chang He; Cossavella, M.; Morcant, K.; Panait. A.; Yvonnet, J. 2007. A coupled numerical and experimental approach for failure analysis of tempered glass stiffeners with pin joints, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 437441. Siebert, B. 2007. Safety aspects of point-fixed glass constructions, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 432436. Siebert, G. 2007. New German DIN standard DIN 18008 for design of glass elements, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 460462. Siebert, G. 2007. Design of glass structures based on fracture mechanics and its integration in national and European design codes, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 591593. Veer, F.A.; Bosma, M.; van Swieten, P.M.J.; Louter, P.C.; Romein, T. 2007. Walking on air, designing and engineering a glass bridge, in The leading conference on architectural and automotive glass Glass Performance days 2007, Tampere, Finland, June 15-18, 2007. Tampere, 244246. Veer, F.A.; Louter, P.C.; Bos, F.P.; Romein, T.; van Ginkel, H.; Riemslag, A.C. 2008. The strength of architectural glass, in Challenging Glass, Conference on Architectural and Structural Applications of Glass: Selected papers. Ed. by Bos, Louter, Veer. May 2008, Delft, Netherlands. TUDelft, 660 p. Veer, F.A.; Pastunink, J.R. 1999. Developing a Transparent Tubular Laminated Column, in International Conference on Architectural and Automotive Glass Glass Processing Days 1999, Tampere, Finland, June 13-16, 1999. Tampere, 277280. Wellershoff, F.; Sedlacek, G. 2003. Glass Pavilion Rheinbach Stability of Glass Columns, in The Leading International Glass Conference Glass Processing Days 2003, Tampere, Finland, June 15-18, 2003. Tampere, 316318. Weller, B.; Vogt, I. 2008. Calculating Bonded Joints with Acrylates, in Challenging Glass, Conference on Architectural and Structural Applications of Glass: Selected papers. Ed. by Bos, Louter, Veer. May 2008, Delft, Netherlands. TUDelft, 660 p.

792